comprehensive biocompatibility testing of a new pmma-ha bone cement versus conventional pmma cement...
TRANSCRIPT
This article was downloaded by [California State University of Fresno]On 28 April 2013 At 0108Publisher Taylor amp FrancisInforma Ltd Registered in England and Wales Registered Number 1072954Registered office Mortimer House 37-41 Mortimer Street London W1T3JH UK
Journal of BiomaterialsScience Polymer EditionPublication details including instructions forauthors and subscription informationhttpwwwtandfonlinecomloitbsp20
Comprehensivebiocompatibility testing of anew PMMA-HA bone cementversus conventional PMMAcement in vitroM Jaumlger amp A WilkeVersion of record first published 02 Apr 2012
To cite this article M Jaumlger amp A Wilke (2003) Comprehensive biocompatibilitytesting of a new PMMA-HA bone cement versus conventional PMMA cement in vitroJournal of Biomaterials Science Polymer Edition 1411 1283-1298
To link to this article httpdxdoiorg101163156856203322553491
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use httpwwwtandfonlinecompageterms-and-conditions
This article may be used for research teaching and private studypurposes Any substantial or systematic reproduction redistributionreselling loan sub-licensing systematic supply or distribution in any formto anyone is expressly forbidden
The publisher does not give any warranty express or implied or make anyrepresentation that the contents will be complete or accurate or up todate The accuracy of any instructions formulae and drug doses shouldbe independently verified with primary sources The publisher shall notbe liable for any loss actions claims proceedings demand or costs or
damages whatsoever or howsoever caused arising directly or indirectly inconnection with or arising out of the use of this material
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
J Biomater Sci Polymer Edn Vol 14 No 11 pp 1283ndash1298 (2003)Oacute VSP 2003Also available online - wwwvsppubcom
Comprehensive biocompatibility testingof a new PMMA-HA bone cement versus conventionalPMMA cement in vitro
M JAumlGER 1curren and A WILKE 2
1 Orthopaedic Department Heinrich-Heine-UniversityMedical School Moorenstr 5D-40225 Duumlsseldorf Germany
2 Orthopaedic Department Philipps-UniversityMedical School Klinikum LahnbergeBaldinger-StrasseD-35043 Marburg Germany
Received 2 November 2002 accepted 15 July 2003
AbstractmdashFor more than 50 years PMMA bone cements have been used in orthopaedic surgery Inthis study attempts were made to show whether cultured human bone marrow cells (HBMC) showan osteogenetic response resulting in new bone formation production of extracellular matrix (ECM)and cell differentiationwhen they were cultured onto polymerized polymethylmethacrylate(PMMA)-hydroxyapatite (HA) conventional PMMA bone cement being taken as reference Biocompatibilityparameters were collagen-I and -III synthesis the detection of the osteoblast markers alkalinephosphatase (ALP) and osteocalcin the number of adherent cells and the cytodifferentiation ofimmunocompetent cells Cement surface structure HA stability in culture medium and chemicalelement analysis of specimens were considered Fresh marrow cells were obtained from thehuman femora during hip replacement Incubation time was up to ten weeks We used atomicforced microscopy (AFM) and scanning electron microscopy (SEM) for cement specimen analysisFluorescent activated cell sorter (FACS) immunohistochemical staining SEM and light microscopy(LM) served us to judge the cellular morphology Products of the extracellular matrix were analyzedby protein dot blot analysis SEM energy dispersive X-ray analysis (SEM-EDX) and Ca2C PO3iexcl
4detection HA particles increased the osteogenetic potential of PMMA bone cement regarding thecellular production of collagen alkaline phosphatase (AP) the number of osteoblasts and the cellulardifferentiation pattern in vitro Both tested cements showed good biocompatibility in a human long-term bone marrow cell-culture system
Key words Bone cement bone marrow cell culture biocompatibilityhydroxyapatite
currenTo whom correspondence should be addressed Phone (49-202) 942-2622 Fax (49-202) 942-2623 E-mail drjaegermarcusyahoode
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1284 M Jaumlger and A Wilke
INTRODUCTION
For over 50 years poly(methyl methacrylate) (PMMA) bone cements have beenemployed in orthopaedic and trauma surgery for xing arthroplasties Althoughthey are among the best investigated biomaterials aseptic loosening of cementedprostheses remains a frequent complication after total hip replacement Mechanicalfactors are considered to be principally responsible but biological reactions toimplanted PMMA bone cement may play an important role too [1 2] With regardto hip arthroplasties the histological characteristic structure at the bonendash cementinterface is the presence of a pseudomembranous brous structure predominatingin the proximity of femora [3] Willert and Semlitsch [4] investigated this kind offoreign body reaction in periprosthetic tissue specimens taken at revision surgeryfor failed cemented hip arthroplasties They proposed the hypothesis of the wearproduct transport equilibrium If the quantity of wear products like PMMAparticles exceeds the transport capacity wear products accumulate and lead tocellular reaction resulting in in ltrating granulomas periprosthetic bone resorptionand eventually prosthetic loosening On the other hand Jones [5] concluded that thebiomechanical properties of bone cements are mainly responsible for the lsquocementdiseasersquo phenomon Linder [6] suggested that the bone formation at the bone cementinterface in failed hip arthroplasties results from an attempt of tissue to stabilizethe prosthesis Though the exact mechanism and the role of bone cement in thepathogenesis is still unclear in literature includes PMMA among the factors thatare responsible for aseptic loosening of hip arthroplasties Santavirta et al [7] foundthat PMMA powder is an immunological inert material whereas Gil-Albarova [8]reported an activated immune response to PMMA Since many years hydroxyapatite(HA) ceramics are used as bone substitutes and bone graft materials includingcoatings on metal components for dental and orthopaedic applications [8] HApromotes the degradation of serum proteins and mineralization at its surface [9 10]These osteoconductive properties suggest direct physicochemical binding betweenthe HA material and surrounding bone resulting in a high mechanical stability [11]The tissue response to porous HA implants is different from that to dense HAwith regard to the opportunity for bone in-growth Because of its low surfacearea dense HA has demonstrated low rates of biodegradation [12] On the otherhand osteoconductive HA implants with a critical degree of porosity can undergo asigni cant degree of resorption For example the resorption rates of coral-derivedHA which has micropores of 1ndash5 sup1m in diameter ranged from 2 up to 5 peryear [12]
The aim of this study was to compare the biocompatibility of a PMMA bonecement to PMMA-HA cement on a long term human bone marrow cell culturesystem in vitro Our hypothesis was that the presence of HA particles can reduceforeign body reactions and stimulate osteoblastic differentiation Cell numbervitality and morphology chemical cement analysis cellular immuno-differentiationpro le on bone cements analysis of extracellular matrix components and numberof differentiated osteoblasts served as biocompatibility parameters
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1285
MATERIALS AND METHODS
Cement specimens
Vacuum mixed polymerized discoid bone cement probes (Palacos Rreg OsteopalHAreg) with 25 cm in diameter and a thickness of 2 mm were industrial compounds(Merck Germany) (Fig 1) The components were poly(methylmethacrylate)(PMMA) methylmethacrylate powder (MMA MW 10012 gmol melting pointiexcl48plusmnC boiling point C100plusmnC solubility in H2O20plusmnC 115 g l refraction indexn20plusmn=D 14140 density (d20=4plusmn) 0942ndash0944) and HA The particle size of HA(stoichiometry Ca10(PO4)6(OH)2) granules at sintered temperatures gt 1200plusmnC inPMMA-HA was lt10 sup1m the concentration in PMMA powder was 146 wt Thedensity of the HA-particles was 315 gcm3 the density of PMMA-HA cement was134 gcm3 The CaP ratio of the crystalline HA was 167 Zirconium was used asradiopaque agent
Cell culture
The bone marrow cell-culture system was developed by Dexter and Kalla andmodi ed by Wilke et al [13ndash18] A 7 pound 10 pound 20 mm spongial bone marrowblock was taken from the proximal femora during rst hip arthroplasty surgeryafter femoral neck osteotomy Patient criteria for inclusion 50ndash60-year-old whitefemales with osteoarthritis of the hip quali ed for autologous blood transfusion andbone donation Criteria for exclusion acute or chronic infections and neoplasms
Figure 1 PMMA and PMMA-HA cement probes in macroscopic geometry The discoid poly-merized specimens have an surface area of 42 cm2 and were industrial delivered as Palacos Rreg
(PMMA) and Osteopal HAreg (PMMA-HA) Sterilisation was done by radiation
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1286 M Jaumlger and A Wilke
in the medical history We analyzed the bone marrow of four probands (averageage 563 years) by Scanning Electron Microscopy (SEM) Light Microscopy(LM) dot blot Fluorescence Activated Cell Sorter (FACS) analysis cell vitalitytest (propidium bromide staining) osteocalcin staining and PO3iexcl
4 Ca2C detectionin culture supernatants The marrow cells were isolated by Percoll gradientcentrifugation and incubated with Iscoversquos modi ed Dulbeccorsquos Medium (IMDM)including 125 fetal calf serum 125 horse serum 24 ngml hydrocortisone03 NaHCO3 18 pound 106 cells were seeded onto a specimen surface of 491 cm2Growth promoters added after one day of cultivation were 10 ng IL-3 and 10 ngGM-CSF per ml IMDM medium All incubations were carried out at 37plusmnC in 5CO2 95 air Control groups cells without contact to bone cement cultivatedin Petri dishes For FACS analysis cells were seeded on (a) glass slides and (b)PMMAPMMA-HA discs in culture medium and PBS buffer solution without bonemarrow cells Medium was changed every fourth day
Scanning electron microscopy (SEM) SEM-EDX and atomic force microscopy(AFM)
The adherent cells were xed by ITO- xans (2 glutaraldehyde in 01 M cacodylatebuffer Serva Germany) pH 74 for 2 h dehydrated in graded ethanol and tert-butanol and vacuum dried for 36 h Afterwards cement specimens were carbonsputter coated (sputter coater S150B Edwards UK) Specimens were examinedby SEM Cam Scan 4 DV (Cam Scan UK) which was connected with the EDX-analyzer Voyager 2110 including an ultra-thin window (UTW Noran InstrumentsUSA) and photographed Surface roughness of uncoated cement probes wereinvestigated by AFM Nanoscope IIIa (Digital Instruments USA) in tapping modeHuman bone marrow cells from 10 donors were analyzed at 2 4 8 and 10 weeksculture time
FACS analysis
The mechanical removal of adherent marrow cells was supported enzymatically by005 trypsin (Sigma St Louis MO USA) and controlled by episcopic lightmicroscopy (Aristomet Leitz Germany) Afterwards antibody incubation againstthe CD-antigen of T-lymphocytes (CD-2 Behring code BMA 0111-FITC ProductNCPE1011 Behring Germany) mononuclear cells (CD-14 Behring code BMA0211-FITC Product NCPE 1011 Behring) granulocytes (CD-15 BMA 0200-FITC Behring) B-lymphocytes (CD-22 No F5546 Sigma) and hematopoieticstem cells (CD-34 Product 345801 BD Bioscience Germany) followed For theprimary antibody conditions were 1 h incubation time at 4plusmnC concentraion 1 10in TBS 2 BSA For the secondary antibody conditions were anti-mouse FITC-labeled (Sigma) concentration 1 10 TBS 2 BSA 1 h incubation at RT Propidiumbromide labelling served to identify dead cells before FACS analysis (FACSTARPlus Becton-Dickinson USA) started The culture period for FACS analysis was
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1287
14 days Control groups human bone marrow cells cultivated on (a) cover glassslides and (b) fresh human marrow cells 12 h after isolation
Dot blot analysis
Type-I and -III collagen and alkaline phosphatase (ALP) were detected by semi-quantitative dot blot analysis of culture medium supernants based on an anti-body system Primary antibody rabbit anti-human collagen I III (ChemiconUSA) secondary antibody anti-rabbit IgG biotin F0
ab2-fragment (Boehringer Ger-many) Streptavidine-peroxidase (Boehringer) and nitroblue-tetrazolium (SigmaUSA) served as staining chemicals The study period was 72 days
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis of collected culture medium was done by Hitachi Automatic
Analyzer (Boehringer) Study period was 23 days
Osteoblast staining
At 14 days of cultivation cement adherent marrow cells were washed in TBS bufferand incubated with a goat anti-osteocalcin antibody (Fitzgerald USA) over 24 h at4plusmnC After washing in TBS the second antibody (anti-goat biotin conjugate SigmaUSA) was added and incubated for 1 h RT Following another TBS washing thethird antibody (extraavidine-alkaline phosphatase Sigma Germany) was added for1 h For staining Neu-Fuchsin solution (Sigma USA) was used The number ofosteocalcin-positive cells per mm2 were measured by eye LM analysis
Statistics
Because of former experiments we could assume in this study a KolmogorovndashSmirnov-validated normal distribution of human marrow cells in regard to differ-entiation parameters (osteocalcin and CD-markers) collagen-I and -III productionand cell number of adherent cells To get statistical-based results we used theMannndashWhitney test for independent statistical groups P lt 001 was rated highlystatistical signi cant and P lt 005 statistically signi cant whereas P gt 005 wasconsidered not signi cant
RESULTS
Cement analysis before cellular contact (primary material analysis) showed nosigni cant impurities in cement specimens corresponding to the manufacturerrsquosstatement in SEM-EDX Surface roughness of probes was not well de ned andshowed high horizontal and vertical distance variations (nmndashsup1m) documented bySEM and AFM The SEM photo in Fig 2 shows the PMMA-HA morphology before
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1288 M Jaumlger and A Wilke
Figure 2 SEM analysis of PMMA-HA bone cement specimen without human marrow cells tippedover 60plusmn Big round ball-shaped PMMA polymers (arrows) are surrounded by HA-granules Therough surface enlarges the cell attachment area
we started cell culturing The corresponding data based on a SEM-EDX analysis areshown in Fig 3 The AFM analysis con rmed the rough irregular surface structureWe found signs of surface defects for PMMA and PMMA-HA (Fig 4)
Morphology and cell number
SEM analysis showed signi cantly much fewer cells adhered to PMMA-HA thanto PMMA slices at two weeks The analysis of ten specimens showed 268 lessadherent cells per area for PMMA-HA in comparison to PMMA (mean valuesPMMA 906 cells mm2 (SD 192) versus PMMA-HA 663 cells mm2 (SD 232))With increasing culture time the cells tend to build up clusters (cell aggregates)Cellular clusters were found for both cements from day 7 up to day 72 (end ofstudy) The SEM photo in Fig 5 shows at adherent marrow cells on PMMA-HAsurface Furthermore round cells can be seen (dead or detached) after 10 weeks ofincubation (Fig 5)
Cell differentiation and cell vitality
After detaching the cultivated marrow cells from cement probes the number of deadcells was lt10 in all experiments as detected by propidium bromide staining Atday 14 cell culture demonstrated a high number of T-lymphocytes and granulocytes
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1289
Fig
ure
3S
EM
-ED
Xan
alys
isba
sed
onth
ear
eaof
Fig
2
No
addi
tion
alel
emen
tsw
ere
dete
cted
and
the
empi
rica
lfor
mul
aC
a 5(P
O4)
3OH
2is
con
rmed
byqu
anti
tativ
ean
alys
is
O
oxyg
en
Cc
arbo
nC
aca
lciu
mP
pho
spho
rum
Z
irco
nium
(Zr)
serv
edas
X-r
ayco
ntra
stm
ediu
man
dis
notv
isib
lehe
rebe
caus
eof
the
over
lapp
ing
byP
-pea
k(L
-reg-p
eak)
A
ccel
erat
ing
ener
gy
10ke
Vl
ife
tim
e20
0s
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1290 M Jaumlger and A Wilke
Fig
ure
4A
FM
surf
ace
anal
ysis
ofa
PM
MA
spec
imen
inta
ppin
gm
ode
Ove
rvie
wof
aP
MM
Ace
men
tsur
face
area
B
esid
esth
eir
regu
lar
surf
ace
stru
ctur
ea
scra
tch
asa
sign
ofm
ater
iald
efec
tcan
bede
tect
ed(b
lack
arro
w)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1291
Figure 5 Adherent at human bone marrow cells after 10 weeks of incubation on PMMA-HABesides at adherent cells detached round cells are shown The surrounding extracellularmatrix wasrich of sulfur (SEM-EDX)
in case of PMMA in FACS analysis whereas the differentiation rate of monocytesshowed only few differences between both cements (Fig 6) Human bone marrowcells on PMMA-HA differentiated more B-cells than those which were cultivatedon PMMA but less than the glass control group
Immunohistochemical osteocalcin staining reported the superiority of the PMMA-HA cement with regard to the induction of an osteoblastic differentiation Thenumber of osteocalcin positive cells was 39-fold higher for PMMA-HA than formarrow cells which were seeded on PMMA P lt 0001 For PMMA we found inaverage 39 pound 103 cells mm2 (SD 178) compared to 151 pound 103 osteocalcin-positivecells mm2 (SD 191) in the case of PMMA-HA Parallel to the high amount ofosteocalcin positive cells on the PMMA-HA surface we detected a higher rate ofintercellular bridging
Extracellular matrix (ECM)
SEM-EDX analysis showed a sulfur-rich ECM which was located only pericyto-plasmatic In the case of PMMA sulfur was detected after 10 weeks incubationwhereas for the PMMA-HA cement sulfur already appeared after two weeks Fig-ure 7 shows the pericellular EDX analysis of human marrow cells after differentincubation times with PMMA or PMMA-HA cement The semi-quantatively ana-lyzed areas showed an increasing content of sulfur as a function of time for PMMA-
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1292 M Jaumlger and A Wilke
Figure 6 FACS analysis of bone cement adherent marrow cells 14 days after incubation with PMMAand PMMA-HA in comparison with control group Fresh marrow cells 12 h after isolation from theproximal femur (native) and after 14 incubation days onto glass slides The table below demonstratesthe values the bar charts in the diagram the standard deviation PMMA-HA shows a decreasing rateof T-lymhocytes in comparison to PMMA but the number of granulocytes (416) are signi cantlyhigher for PMMA (203) P lt 0005 There is nearly no difference of the number of mononuclearcells between both cements (PMMA 308 PMMA-HA 319) P gt 0005
HA We found an amorphous protein precipitate covering the cement surface in ex-periments with and without cells Sulfur was only detected by SEM-EDX in celladherent specimens
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis showed that HA-particles in PMMA-HA were stable in an
aquaeous medium during the 23-day study period The average concentration valuesfor PMMA cell culture medium were Ca2C 232 mmol l PO3iexcl
4 158 mmol l(SD 0041) and for PMMA-HA Ca2C 231 mmol l PO3iexcl
4 159 mmol l (SD 0026)Dot blot analysis of the collected culture medium supernatants was carried out
semiquantitative for collagen I III and alkaline phosphatase (ALP) over a studyperiod of 72 days The AP activity and collagen-III detection showed no differencesregarding the rythm of appearance for both cements In the case of PMMA-HAcollagen I was synthesized at day 6 and showed a signi cant staining reaction from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
damages whatsoever or howsoever caused arising directly or indirectly inconnection with or arising out of the use of this material
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
J Biomater Sci Polymer Edn Vol 14 No 11 pp 1283ndash1298 (2003)Oacute VSP 2003Also available online - wwwvsppubcom
Comprehensive biocompatibility testingof a new PMMA-HA bone cement versus conventionalPMMA cement in vitro
M JAumlGER 1curren and A WILKE 2
1 Orthopaedic Department Heinrich-Heine-UniversityMedical School Moorenstr 5D-40225 Duumlsseldorf Germany
2 Orthopaedic Department Philipps-UniversityMedical School Klinikum LahnbergeBaldinger-StrasseD-35043 Marburg Germany
Received 2 November 2002 accepted 15 July 2003
AbstractmdashFor more than 50 years PMMA bone cements have been used in orthopaedic surgery Inthis study attempts were made to show whether cultured human bone marrow cells (HBMC) showan osteogenetic response resulting in new bone formation production of extracellular matrix (ECM)and cell differentiationwhen they were cultured onto polymerized polymethylmethacrylate(PMMA)-hydroxyapatite (HA) conventional PMMA bone cement being taken as reference Biocompatibilityparameters were collagen-I and -III synthesis the detection of the osteoblast markers alkalinephosphatase (ALP) and osteocalcin the number of adherent cells and the cytodifferentiation ofimmunocompetent cells Cement surface structure HA stability in culture medium and chemicalelement analysis of specimens were considered Fresh marrow cells were obtained from thehuman femora during hip replacement Incubation time was up to ten weeks We used atomicforced microscopy (AFM) and scanning electron microscopy (SEM) for cement specimen analysisFluorescent activated cell sorter (FACS) immunohistochemical staining SEM and light microscopy(LM) served us to judge the cellular morphology Products of the extracellular matrix were analyzedby protein dot blot analysis SEM energy dispersive X-ray analysis (SEM-EDX) and Ca2C PO3iexcl
4detection HA particles increased the osteogenetic potential of PMMA bone cement regarding thecellular production of collagen alkaline phosphatase (AP) the number of osteoblasts and the cellulardifferentiation pattern in vitro Both tested cements showed good biocompatibility in a human long-term bone marrow cell-culture system
Key words Bone cement bone marrow cell culture biocompatibilityhydroxyapatite
currenTo whom correspondence should be addressed Phone (49-202) 942-2622 Fax (49-202) 942-2623 E-mail drjaegermarcusyahoode
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1284 M Jaumlger and A Wilke
INTRODUCTION
For over 50 years poly(methyl methacrylate) (PMMA) bone cements have beenemployed in orthopaedic and trauma surgery for xing arthroplasties Althoughthey are among the best investigated biomaterials aseptic loosening of cementedprostheses remains a frequent complication after total hip replacement Mechanicalfactors are considered to be principally responsible but biological reactions toimplanted PMMA bone cement may play an important role too [1 2] With regardto hip arthroplasties the histological characteristic structure at the bonendash cementinterface is the presence of a pseudomembranous brous structure predominatingin the proximity of femora [3] Willert and Semlitsch [4] investigated this kind offoreign body reaction in periprosthetic tissue specimens taken at revision surgeryfor failed cemented hip arthroplasties They proposed the hypothesis of the wearproduct transport equilibrium If the quantity of wear products like PMMAparticles exceeds the transport capacity wear products accumulate and lead tocellular reaction resulting in in ltrating granulomas periprosthetic bone resorptionand eventually prosthetic loosening On the other hand Jones [5] concluded that thebiomechanical properties of bone cements are mainly responsible for the lsquocementdiseasersquo phenomon Linder [6] suggested that the bone formation at the bone cementinterface in failed hip arthroplasties results from an attempt of tissue to stabilizethe prosthesis Though the exact mechanism and the role of bone cement in thepathogenesis is still unclear in literature includes PMMA among the factors thatare responsible for aseptic loosening of hip arthroplasties Santavirta et al [7] foundthat PMMA powder is an immunological inert material whereas Gil-Albarova [8]reported an activated immune response to PMMA Since many years hydroxyapatite(HA) ceramics are used as bone substitutes and bone graft materials includingcoatings on metal components for dental and orthopaedic applications [8] HApromotes the degradation of serum proteins and mineralization at its surface [9 10]These osteoconductive properties suggest direct physicochemical binding betweenthe HA material and surrounding bone resulting in a high mechanical stability [11]The tissue response to porous HA implants is different from that to dense HAwith regard to the opportunity for bone in-growth Because of its low surfacearea dense HA has demonstrated low rates of biodegradation [12] On the otherhand osteoconductive HA implants with a critical degree of porosity can undergo asigni cant degree of resorption For example the resorption rates of coral-derivedHA which has micropores of 1ndash5 sup1m in diameter ranged from 2 up to 5 peryear [12]
The aim of this study was to compare the biocompatibility of a PMMA bonecement to PMMA-HA cement on a long term human bone marrow cell culturesystem in vitro Our hypothesis was that the presence of HA particles can reduceforeign body reactions and stimulate osteoblastic differentiation Cell numbervitality and morphology chemical cement analysis cellular immuno-differentiationpro le on bone cements analysis of extracellular matrix components and numberof differentiated osteoblasts served as biocompatibility parameters
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1285
MATERIALS AND METHODS
Cement specimens
Vacuum mixed polymerized discoid bone cement probes (Palacos Rreg OsteopalHAreg) with 25 cm in diameter and a thickness of 2 mm were industrial compounds(Merck Germany) (Fig 1) The components were poly(methylmethacrylate)(PMMA) methylmethacrylate powder (MMA MW 10012 gmol melting pointiexcl48plusmnC boiling point C100plusmnC solubility in H2O20plusmnC 115 g l refraction indexn20plusmn=D 14140 density (d20=4plusmn) 0942ndash0944) and HA The particle size of HA(stoichiometry Ca10(PO4)6(OH)2) granules at sintered temperatures gt 1200plusmnC inPMMA-HA was lt10 sup1m the concentration in PMMA powder was 146 wt Thedensity of the HA-particles was 315 gcm3 the density of PMMA-HA cement was134 gcm3 The CaP ratio of the crystalline HA was 167 Zirconium was used asradiopaque agent
Cell culture
The bone marrow cell-culture system was developed by Dexter and Kalla andmodi ed by Wilke et al [13ndash18] A 7 pound 10 pound 20 mm spongial bone marrowblock was taken from the proximal femora during rst hip arthroplasty surgeryafter femoral neck osteotomy Patient criteria for inclusion 50ndash60-year-old whitefemales with osteoarthritis of the hip quali ed for autologous blood transfusion andbone donation Criteria for exclusion acute or chronic infections and neoplasms
Figure 1 PMMA and PMMA-HA cement probes in macroscopic geometry The discoid poly-merized specimens have an surface area of 42 cm2 and were industrial delivered as Palacos Rreg
(PMMA) and Osteopal HAreg (PMMA-HA) Sterilisation was done by radiation
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1286 M Jaumlger and A Wilke
in the medical history We analyzed the bone marrow of four probands (averageage 563 years) by Scanning Electron Microscopy (SEM) Light Microscopy(LM) dot blot Fluorescence Activated Cell Sorter (FACS) analysis cell vitalitytest (propidium bromide staining) osteocalcin staining and PO3iexcl
4 Ca2C detectionin culture supernatants The marrow cells were isolated by Percoll gradientcentrifugation and incubated with Iscoversquos modi ed Dulbeccorsquos Medium (IMDM)including 125 fetal calf serum 125 horse serum 24 ngml hydrocortisone03 NaHCO3 18 pound 106 cells were seeded onto a specimen surface of 491 cm2Growth promoters added after one day of cultivation were 10 ng IL-3 and 10 ngGM-CSF per ml IMDM medium All incubations were carried out at 37plusmnC in 5CO2 95 air Control groups cells without contact to bone cement cultivatedin Petri dishes For FACS analysis cells were seeded on (a) glass slides and (b)PMMAPMMA-HA discs in culture medium and PBS buffer solution without bonemarrow cells Medium was changed every fourth day
Scanning electron microscopy (SEM) SEM-EDX and atomic force microscopy(AFM)
The adherent cells were xed by ITO- xans (2 glutaraldehyde in 01 M cacodylatebuffer Serva Germany) pH 74 for 2 h dehydrated in graded ethanol and tert-butanol and vacuum dried for 36 h Afterwards cement specimens were carbonsputter coated (sputter coater S150B Edwards UK) Specimens were examinedby SEM Cam Scan 4 DV (Cam Scan UK) which was connected with the EDX-analyzer Voyager 2110 including an ultra-thin window (UTW Noran InstrumentsUSA) and photographed Surface roughness of uncoated cement probes wereinvestigated by AFM Nanoscope IIIa (Digital Instruments USA) in tapping modeHuman bone marrow cells from 10 donors were analyzed at 2 4 8 and 10 weeksculture time
FACS analysis
The mechanical removal of adherent marrow cells was supported enzymatically by005 trypsin (Sigma St Louis MO USA) and controlled by episcopic lightmicroscopy (Aristomet Leitz Germany) Afterwards antibody incubation againstthe CD-antigen of T-lymphocytes (CD-2 Behring code BMA 0111-FITC ProductNCPE1011 Behring Germany) mononuclear cells (CD-14 Behring code BMA0211-FITC Product NCPE 1011 Behring) granulocytes (CD-15 BMA 0200-FITC Behring) B-lymphocytes (CD-22 No F5546 Sigma) and hematopoieticstem cells (CD-34 Product 345801 BD Bioscience Germany) followed For theprimary antibody conditions were 1 h incubation time at 4plusmnC concentraion 1 10in TBS 2 BSA For the secondary antibody conditions were anti-mouse FITC-labeled (Sigma) concentration 1 10 TBS 2 BSA 1 h incubation at RT Propidiumbromide labelling served to identify dead cells before FACS analysis (FACSTARPlus Becton-Dickinson USA) started The culture period for FACS analysis was
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1287
14 days Control groups human bone marrow cells cultivated on (a) cover glassslides and (b) fresh human marrow cells 12 h after isolation
Dot blot analysis
Type-I and -III collagen and alkaline phosphatase (ALP) were detected by semi-quantitative dot blot analysis of culture medium supernants based on an anti-body system Primary antibody rabbit anti-human collagen I III (ChemiconUSA) secondary antibody anti-rabbit IgG biotin F0
ab2-fragment (Boehringer Ger-many) Streptavidine-peroxidase (Boehringer) and nitroblue-tetrazolium (SigmaUSA) served as staining chemicals The study period was 72 days
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis of collected culture medium was done by Hitachi Automatic
Analyzer (Boehringer) Study period was 23 days
Osteoblast staining
At 14 days of cultivation cement adherent marrow cells were washed in TBS bufferand incubated with a goat anti-osteocalcin antibody (Fitzgerald USA) over 24 h at4plusmnC After washing in TBS the second antibody (anti-goat biotin conjugate SigmaUSA) was added and incubated for 1 h RT Following another TBS washing thethird antibody (extraavidine-alkaline phosphatase Sigma Germany) was added for1 h For staining Neu-Fuchsin solution (Sigma USA) was used The number ofosteocalcin-positive cells per mm2 were measured by eye LM analysis
Statistics
Because of former experiments we could assume in this study a KolmogorovndashSmirnov-validated normal distribution of human marrow cells in regard to differ-entiation parameters (osteocalcin and CD-markers) collagen-I and -III productionand cell number of adherent cells To get statistical-based results we used theMannndashWhitney test for independent statistical groups P lt 001 was rated highlystatistical signi cant and P lt 005 statistically signi cant whereas P gt 005 wasconsidered not signi cant
RESULTS
Cement analysis before cellular contact (primary material analysis) showed nosigni cant impurities in cement specimens corresponding to the manufacturerrsquosstatement in SEM-EDX Surface roughness of probes was not well de ned andshowed high horizontal and vertical distance variations (nmndashsup1m) documented bySEM and AFM The SEM photo in Fig 2 shows the PMMA-HA morphology before
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1288 M Jaumlger and A Wilke
Figure 2 SEM analysis of PMMA-HA bone cement specimen without human marrow cells tippedover 60plusmn Big round ball-shaped PMMA polymers (arrows) are surrounded by HA-granules Therough surface enlarges the cell attachment area
we started cell culturing The corresponding data based on a SEM-EDX analysis areshown in Fig 3 The AFM analysis con rmed the rough irregular surface structureWe found signs of surface defects for PMMA and PMMA-HA (Fig 4)
Morphology and cell number
SEM analysis showed signi cantly much fewer cells adhered to PMMA-HA thanto PMMA slices at two weeks The analysis of ten specimens showed 268 lessadherent cells per area for PMMA-HA in comparison to PMMA (mean valuesPMMA 906 cells mm2 (SD 192) versus PMMA-HA 663 cells mm2 (SD 232))With increasing culture time the cells tend to build up clusters (cell aggregates)Cellular clusters were found for both cements from day 7 up to day 72 (end ofstudy) The SEM photo in Fig 5 shows at adherent marrow cells on PMMA-HAsurface Furthermore round cells can be seen (dead or detached) after 10 weeks ofincubation (Fig 5)
Cell differentiation and cell vitality
After detaching the cultivated marrow cells from cement probes the number of deadcells was lt10 in all experiments as detected by propidium bromide staining Atday 14 cell culture demonstrated a high number of T-lymphocytes and granulocytes
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1289
Fig
ure
3S
EM
-ED
Xan
alys
isba
sed
onth
ear
eaof
Fig
2
No
addi
tion
alel
emen
tsw
ere
dete
cted
and
the
empi
rica
lfor
mul
aC
a 5(P
O4)
3OH
2is
con
rmed
byqu
anti
tativ
ean
alys
is
O
oxyg
en
Cc
arbo
nC
aca
lciu
mP
pho
spho
rum
Z
irco
nium
(Zr)
serv
edas
X-r
ayco
ntra
stm
ediu
man
dis
notv
isib
lehe
rebe
caus
eof
the
over
lapp
ing
byP
-pea
k(L
-reg-p
eak)
A
ccel
erat
ing
ener
gy
10ke
Vl
ife
tim
e20
0s
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1290 M Jaumlger and A Wilke
Fig
ure
4A
FM
surf
ace
anal
ysis
ofa
PM
MA
spec
imen
inta
ppin
gm
ode
Ove
rvie
wof
aP
MM
Ace
men
tsur
face
area
B
esid
esth
eir
regu
lar
surf
ace
stru
ctur
ea
scra
tch
asa
sign
ofm
ater
iald
efec
tcan
bede
tect
ed(b
lack
arro
w)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1291
Figure 5 Adherent at human bone marrow cells after 10 weeks of incubation on PMMA-HABesides at adherent cells detached round cells are shown The surrounding extracellularmatrix wasrich of sulfur (SEM-EDX)
in case of PMMA in FACS analysis whereas the differentiation rate of monocytesshowed only few differences between both cements (Fig 6) Human bone marrowcells on PMMA-HA differentiated more B-cells than those which were cultivatedon PMMA but less than the glass control group
Immunohistochemical osteocalcin staining reported the superiority of the PMMA-HA cement with regard to the induction of an osteoblastic differentiation Thenumber of osteocalcin positive cells was 39-fold higher for PMMA-HA than formarrow cells which were seeded on PMMA P lt 0001 For PMMA we found inaverage 39 pound 103 cells mm2 (SD 178) compared to 151 pound 103 osteocalcin-positivecells mm2 (SD 191) in the case of PMMA-HA Parallel to the high amount ofosteocalcin positive cells on the PMMA-HA surface we detected a higher rate ofintercellular bridging
Extracellular matrix (ECM)
SEM-EDX analysis showed a sulfur-rich ECM which was located only pericyto-plasmatic In the case of PMMA sulfur was detected after 10 weeks incubationwhereas for the PMMA-HA cement sulfur already appeared after two weeks Fig-ure 7 shows the pericellular EDX analysis of human marrow cells after differentincubation times with PMMA or PMMA-HA cement The semi-quantatively ana-lyzed areas showed an increasing content of sulfur as a function of time for PMMA-
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1292 M Jaumlger and A Wilke
Figure 6 FACS analysis of bone cement adherent marrow cells 14 days after incubation with PMMAand PMMA-HA in comparison with control group Fresh marrow cells 12 h after isolation from theproximal femur (native) and after 14 incubation days onto glass slides The table below demonstratesthe values the bar charts in the diagram the standard deviation PMMA-HA shows a decreasing rateof T-lymhocytes in comparison to PMMA but the number of granulocytes (416) are signi cantlyhigher for PMMA (203) P lt 0005 There is nearly no difference of the number of mononuclearcells between both cements (PMMA 308 PMMA-HA 319) P gt 0005
HA We found an amorphous protein precipitate covering the cement surface in ex-periments with and without cells Sulfur was only detected by SEM-EDX in celladherent specimens
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis showed that HA-particles in PMMA-HA were stable in an
aquaeous medium during the 23-day study period The average concentration valuesfor PMMA cell culture medium were Ca2C 232 mmol l PO3iexcl
4 158 mmol l(SD 0041) and for PMMA-HA Ca2C 231 mmol l PO3iexcl
4 159 mmol l (SD 0026)Dot blot analysis of the collected culture medium supernatants was carried out
semiquantitative for collagen I III and alkaline phosphatase (ALP) over a studyperiod of 72 days The AP activity and collagen-III detection showed no differencesregarding the rythm of appearance for both cements In the case of PMMA-HAcollagen I was synthesized at day 6 and showed a signi cant staining reaction from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
J Biomater Sci Polymer Edn Vol 14 No 11 pp 1283ndash1298 (2003)Oacute VSP 2003Also available online - wwwvsppubcom
Comprehensive biocompatibility testingof a new PMMA-HA bone cement versus conventionalPMMA cement in vitro
M JAumlGER 1curren and A WILKE 2
1 Orthopaedic Department Heinrich-Heine-UniversityMedical School Moorenstr 5D-40225 Duumlsseldorf Germany
2 Orthopaedic Department Philipps-UniversityMedical School Klinikum LahnbergeBaldinger-StrasseD-35043 Marburg Germany
Received 2 November 2002 accepted 15 July 2003
AbstractmdashFor more than 50 years PMMA bone cements have been used in orthopaedic surgery Inthis study attempts were made to show whether cultured human bone marrow cells (HBMC) showan osteogenetic response resulting in new bone formation production of extracellular matrix (ECM)and cell differentiationwhen they were cultured onto polymerized polymethylmethacrylate(PMMA)-hydroxyapatite (HA) conventional PMMA bone cement being taken as reference Biocompatibilityparameters were collagen-I and -III synthesis the detection of the osteoblast markers alkalinephosphatase (ALP) and osteocalcin the number of adherent cells and the cytodifferentiation ofimmunocompetent cells Cement surface structure HA stability in culture medium and chemicalelement analysis of specimens were considered Fresh marrow cells were obtained from thehuman femora during hip replacement Incubation time was up to ten weeks We used atomicforced microscopy (AFM) and scanning electron microscopy (SEM) for cement specimen analysisFluorescent activated cell sorter (FACS) immunohistochemical staining SEM and light microscopy(LM) served us to judge the cellular morphology Products of the extracellular matrix were analyzedby protein dot blot analysis SEM energy dispersive X-ray analysis (SEM-EDX) and Ca2C PO3iexcl
4detection HA particles increased the osteogenetic potential of PMMA bone cement regarding thecellular production of collagen alkaline phosphatase (AP) the number of osteoblasts and the cellulardifferentiation pattern in vitro Both tested cements showed good biocompatibility in a human long-term bone marrow cell-culture system
Key words Bone cement bone marrow cell culture biocompatibilityhydroxyapatite
currenTo whom correspondence should be addressed Phone (49-202) 942-2622 Fax (49-202) 942-2623 E-mail drjaegermarcusyahoode
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1284 M Jaumlger and A Wilke
INTRODUCTION
For over 50 years poly(methyl methacrylate) (PMMA) bone cements have beenemployed in orthopaedic and trauma surgery for xing arthroplasties Althoughthey are among the best investigated biomaterials aseptic loosening of cementedprostheses remains a frequent complication after total hip replacement Mechanicalfactors are considered to be principally responsible but biological reactions toimplanted PMMA bone cement may play an important role too [1 2] With regardto hip arthroplasties the histological characteristic structure at the bonendash cementinterface is the presence of a pseudomembranous brous structure predominatingin the proximity of femora [3] Willert and Semlitsch [4] investigated this kind offoreign body reaction in periprosthetic tissue specimens taken at revision surgeryfor failed cemented hip arthroplasties They proposed the hypothesis of the wearproduct transport equilibrium If the quantity of wear products like PMMAparticles exceeds the transport capacity wear products accumulate and lead tocellular reaction resulting in in ltrating granulomas periprosthetic bone resorptionand eventually prosthetic loosening On the other hand Jones [5] concluded that thebiomechanical properties of bone cements are mainly responsible for the lsquocementdiseasersquo phenomon Linder [6] suggested that the bone formation at the bone cementinterface in failed hip arthroplasties results from an attempt of tissue to stabilizethe prosthesis Though the exact mechanism and the role of bone cement in thepathogenesis is still unclear in literature includes PMMA among the factors thatare responsible for aseptic loosening of hip arthroplasties Santavirta et al [7] foundthat PMMA powder is an immunological inert material whereas Gil-Albarova [8]reported an activated immune response to PMMA Since many years hydroxyapatite(HA) ceramics are used as bone substitutes and bone graft materials includingcoatings on metal components for dental and orthopaedic applications [8] HApromotes the degradation of serum proteins and mineralization at its surface [9 10]These osteoconductive properties suggest direct physicochemical binding betweenthe HA material and surrounding bone resulting in a high mechanical stability [11]The tissue response to porous HA implants is different from that to dense HAwith regard to the opportunity for bone in-growth Because of its low surfacearea dense HA has demonstrated low rates of biodegradation [12] On the otherhand osteoconductive HA implants with a critical degree of porosity can undergo asigni cant degree of resorption For example the resorption rates of coral-derivedHA which has micropores of 1ndash5 sup1m in diameter ranged from 2 up to 5 peryear [12]
The aim of this study was to compare the biocompatibility of a PMMA bonecement to PMMA-HA cement on a long term human bone marrow cell culturesystem in vitro Our hypothesis was that the presence of HA particles can reduceforeign body reactions and stimulate osteoblastic differentiation Cell numbervitality and morphology chemical cement analysis cellular immuno-differentiationpro le on bone cements analysis of extracellular matrix components and numberof differentiated osteoblasts served as biocompatibility parameters
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1285
MATERIALS AND METHODS
Cement specimens
Vacuum mixed polymerized discoid bone cement probes (Palacos Rreg OsteopalHAreg) with 25 cm in diameter and a thickness of 2 mm were industrial compounds(Merck Germany) (Fig 1) The components were poly(methylmethacrylate)(PMMA) methylmethacrylate powder (MMA MW 10012 gmol melting pointiexcl48plusmnC boiling point C100plusmnC solubility in H2O20plusmnC 115 g l refraction indexn20plusmn=D 14140 density (d20=4plusmn) 0942ndash0944) and HA The particle size of HA(stoichiometry Ca10(PO4)6(OH)2) granules at sintered temperatures gt 1200plusmnC inPMMA-HA was lt10 sup1m the concentration in PMMA powder was 146 wt Thedensity of the HA-particles was 315 gcm3 the density of PMMA-HA cement was134 gcm3 The CaP ratio of the crystalline HA was 167 Zirconium was used asradiopaque agent
Cell culture
The bone marrow cell-culture system was developed by Dexter and Kalla andmodi ed by Wilke et al [13ndash18] A 7 pound 10 pound 20 mm spongial bone marrowblock was taken from the proximal femora during rst hip arthroplasty surgeryafter femoral neck osteotomy Patient criteria for inclusion 50ndash60-year-old whitefemales with osteoarthritis of the hip quali ed for autologous blood transfusion andbone donation Criteria for exclusion acute or chronic infections and neoplasms
Figure 1 PMMA and PMMA-HA cement probes in macroscopic geometry The discoid poly-merized specimens have an surface area of 42 cm2 and were industrial delivered as Palacos Rreg
(PMMA) and Osteopal HAreg (PMMA-HA) Sterilisation was done by radiation
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1286 M Jaumlger and A Wilke
in the medical history We analyzed the bone marrow of four probands (averageage 563 years) by Scanning Electron Microscopy (SEM) Light Microscopy(LM) dot blot Fluorescence Activated Cell Sorter (FACS) analysis cell vitalitytest (propidium bromide staining) osteocalcin staining and PO3iexcl
4 Ca2C detectionin culture supernatants The marrow cells were isolated by Percoll gradientcentrifugation and incubated with Iscoversquos modi ed Dulbeccorsquos Medium (IMDM)including 125 fetal calf serum 125 horse serum 24 ngml hydrocortisone03 NaHCO3 18 pound 106 cells were seeded onto a specimen surface of 491 cm2Growth promoters added after one day of cultivation were 10 ng IL-3 and 10 ngGM-CSF per ml IMDM medium All incubations were carried out at 37plusmnC in 5CO2 95 air Control groups cells without contact to bone cement cultivatedin Petri dishes For FACS analysis cells were seeded on (a) glass slides and (b)PMMAPMMA-HA discs in culture medium and PBS buffer solution without bonemarrow cells Medium was changed every fourth day
Scanning electron microscopy (SEM) SEM-EDX and atomic force microscopy(AFM)
The adherent cells were xed by ITO- xans (2 glutaraldehyde in 01 M cacodylatebuffer Serva Germany) pH 74 for 2 h dehydrated in graded ethanol and tert-butanol and vacuum dried for 36 h Afterwards cement specimens were carbonsputter coated (sputter coater S150B Edwards UK) Specimens were examinedby SEM Cam Scan 4 DV (Cam Scan UK) which was connected with the EDX-analyzer Voyager 2110 including an ultra-thin window (UTW Noran InstrumentsUSA) and photographed Surface roughness of uncoated cement probes wereinvestigated by AFM Nanoscope IIIa (Digital Instruments USA) in tapping modeHuman bone marrow cells from 10 donors were analyzed at 2 4 8 and 10 weeksculture time
FACS analysis
The mechanical removal of adherent marrow cells was supported enzymatically by005 trypsin (Sigma St Louis MO USA) and controlled by episcopic lightmicroscopy (Aristomet Leitz Germany) Afterwards antibody incubation againstthe CD-antigen of T-lymphocytes (CD-2 Behring code BMA 0111-FITC ProductNCPE1011 Behring Germany) mononuclear cells (CD-14 Behring code BMA0211-FITC Product NCPE 1011 Behring) granulocytes (CD-15 BMA 0200-FITC Behring) B-lymphocytes (CD-22 No F5546 Sigma) and hematopoieticstem cells (CD-34 Product 345801 BD Bioscience Germany) followed For theprimary antibody conditions were 1 h incubation time at 4plusmnC concentraion 1 10in TBS 2 BSA For the secondary antibody conditions were anti-mouse FITC-labeled (Sigma) concentration 1 10 TBS 2 BSA 1 h incubation at RT Propidiumbromide labelling served to identify dead cells before FACS analysis (FACSTARPlus Becton-Dickinson USA) started The culture period for FACS analysis was
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1287
14 days Control groups human bone marrow cells cultivated on (a) cover glassslides and (b) fresh human marrow cells 12 h after isolation
Dot blot analysis
Type-I and -III collagen and alkaline phosphatase (ALP) were detected by semi-quantitative dot blot analysis of culture medium supernants based on an anti-body system Primary antibody rabbit anti-human collagen I III (ChemiconUSA) secondary antibody anti-rabbit IgG biotin F0
ab2-fragment (Boehringer Ger-many) Streptavidine-peroxidase (Boehringer) and nitroblue-tetrazolium (SigmaUSA) served as staining chemicals The study period was 72 days
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis of collected culture medium was done by Hitachi Automatic
Analyzer (Boehringer) Study period was 23 days
Osteoblast staining
At 14 days of cultivation cement adherent marrow cells were washed in TBS bufferand incubated with a goat anti-osteocalcin antibody (Fitzgerald USA) over 24 h at4plusmnC After washing in TBS the second antibody (anti-goat biotin conjugate SigmaUSA) was added and incubated for 1 h RT Following another TBS washing thethird antibody (extraavidine-alkaline phosphatase Sigma Germany) was added for1 h For staining Neu-Fuchsin solution (Sigma USA) was used The number ofosteocalcin-positive cells per mm2 were measured by eye LM analysis
Statistics
Because of former experiments we could assume in this study a KolmogorovndashSmirnov-validated normal distribution of human marrow cells in regard to differ-entiation parameters (osteocalcin and CD-markers) collagen-I and -III productionand cell number of adherent cells To get statistical-based results we used theMannndashWhitney test for independent statistical groups P lt 001 was rated highlystatistical signi cant and P lt 005 statistically signi cant whereas P gt 005 wasconsidered not signi cant
RESULTS
Cement analysis before cellular contact (primary material analysis) showed nosigni cant impurities in cement specimens corresponding to the manufacturerrsquosstatement in SEM-EDX Surface roughness of probes was not well de ned andshowed high horizontal and vertical distance variations (nmndashsup1m) documented bySEM and AFM The SEM photo in Fig 2 shows the PMMA-HA morphology before
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1288 M Jaumlger and A Wilke
Figure 2 SEM analysis of PMMA-HA bone cement specimen without human marrow cells tippedover 60plusmn Big round ball-shaped PMMA polymers (arrows) are surrounded by HA-granules Therough surface enlarges the cell attachment area
we started cell culturing The corresponding data based on a SEM-EDX analysis areshown in Fig 3 The AFM analysis con rmed the rough irregular surface structureWe found signs of surface defects for PMMA and PMMA-HA (Fig 4)
Morphology and cell number
SEM analysis showed signi cantly much fewer cells adhered to PMMA-HA thanto PMMA slices at two weeks The analysis of ten specimens showed 268 lessadherent cells per area for PMMA-HA in comparison to PMMA (mean valuesPMMA 906 cells mm2 (SD 192) versus PMMA-HA 663 cells mm2 (SD 232))With increasing culture time the cells tend to build up clusters (cell aggregates)Cellular clusters were found for both cements from day 7 up to day 72 (end ofstudy) The SEM photo in Fig 5 shows at adherent marrow cells on PMMA-HAsurface Furthermore round cells can be seen (dead or detached) after 10 weeks ofincubation (Fig 5)
Cell differentiation and cell vitality
After detaching the cultivated marrow cells from cement probes the number of deadcells was lt10 in all experiments as detected by propidium bromide staining Atday 14 cell culture demonstrated a high number of T-lymphocytes and granulocytes
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1289
Fig
ure
3S
EM
-ED
Xan
alys
isba
sed
onth
ear
eaof
Fig
2
No
addi
tion
alel
emen
tsw
ere
dete
cted
and
the
empi
rica
lfor
mul
aC
a 5(P
O4)
3OH
2is
con
rmed
byqu
anti
tativ
ean
alys
is
O
oxyg
en
Cc
arbo
nC
aca
lciu
mP
pho
spho
rum
Z
irco
nium
(Zr)
serv
edas
X-r
ayco
ntra
stm
ediu
man
dis
notv
isib
lehe
rebe
caus
eof
the
over
lapp
ing
byP
-pea
k(L
-reg-p
eak)
A
ccel
erat
ing
ener
gy
10ke
Vl
ife
tim
e20
0s
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1290 M Jaumlger and A Wilke
Fig
ure
4A
FM
surf
ace
anal
ysis
ofa
PM
MA
spec
imen
inta
ppin
gm
ode
Ove
rvie
wof
aP
MM
Ace
men
tsur
face
area
B
esid
esth
eir
regu
lar
surf
ace
stru
ctur
ea
scra
tch
asa
sign
ofm
ater
iald
efec
tcan
bede
tect
ed(b
lack
arro
w)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1291
Figure 5 Adherent at human bone marrow cells after 10 weeks of incubation on PMMA-HABesides at adherent cells detached round cells are shown The surrounding extracellularmatrix wasrich of sulfur (SEM-EDX)
in case of PMMA in FACS analysis whereas the differentiation rate of monocytesshowed only few differences between both cements (Fig 6) Human bone marrowcells on PMMA-HA differentiated more B-cells than those which were cultivatedon PMMA but less than the glass control group
Immunohistochemical osteocalcin staining reported the superiority of the PMMA-HA cement with regard to the induction of an osteoblastic differentiation Thenumber of osteocalcin positive cells was 39-fold higher for PMMA-HA than formarrow cells which were seeded on PMMA P lt 0001 For PMMA we found inaverage 39 pound 103 cells mm2 (SD 178) compared to 151 pound 103 osteocalcin-positivecells mm2 (SD 191) in the case of PMMA-HA Parallel to the high amount ofosteocalcin positive cells on the PMMA-HA surface we detected a higher rate ofintercellular bridging
Extracellular matrix (ECM)
SEM-EDX analysis showed a sulfur-rich ECM which was located only pericyto-plasmatic In the case of PMMA sulfur was detected after 10 weeks incubationwhereas for the PMMA-HA cement sulfur already appeared after two weeks Fig-ure 7 shows the pericellular EDX analysis of human marrow cells after differentincubation times with PMMA or PMMA-HA cement The semi-quantatively ana-lyzed areas showed an increasing content of sulfur as a function of time for PMMA-
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1292 M Jaumlger and A Wilke
Figure 6 FACS analysis of bone cement adherent marrow cells 14 days after incubation with PMMAand PMMA-HA in comparison with control group Fresh marrow cells 12 h after isolation from theproximal femur (native) and after 14 incubation days onto glass slides The table below demonstratesthe values the bar charts in the diagram the standard deviation PMMA-HA shows a decreasing rateof T-lymhocytes in comparison to PMMA but the number of granulocytes (416) are signi cantlyhigher for PMMA (203) P lt 0005 There is nearly no difference of the number of mononuclearcells between both cements (PMMA 308 PMMA-HA 319) P gt 0005
HA We found an amorphous protein precipitate covering the cement surface in ex-periments with and without cells Sulfur was only detected by SEM-EDX in celladherent specimens
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis showed that HA-particles in PMMA-HA were stable in an
aquaeous medium during the 23-day study period The average concentration valuesfor PMMA cell culture medium were Ca2C 232 mmol l PO3iexcl
4 158 mmol l(SD 0041) and for PMMA-HA Ca2C 231 mmol l PO3iexcl
4 159 mmol l (SD 0026)Dot blot analysis of the collected culture medium supernatants was carried out
semiquantitative for collagen I III and alkaline phosphatase (ALP) over a studyperiod of 72 days The AP activity and collagen-III detection showed no differencesregarding the rythm of appearance for both cements In the case of PMMA-HAcollagen I was synthesized at day 6 and showed a signi cant staining reaction from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1284 M Jaumlger and A Wilke
INTRODUCTION
For over 50 years poly(methyl methacrylate) (PMMA) bone cements have beenemployed in orthopaedic and trauma surgery for xing arthroplasties Althoughthey are among the best investigated biomaterials aseptic loosening of cementedprostheses remains a frequent complication after total hip replacement Mechanicalfactors are considered to be principally responsible but biological reactions toimplanted PMMA bone cement may play an important role too [1 2] With regardto hip arthroplasties the histological characteristic structure at the bonendash cementinterface is the presence of a pseudomembranous brous structure predominatingin the proximity of femora [3] Willert and Semlitsch [4] investigated this kind offoreign body reaction in periprosthetic tissue specimens taken at revision surgeryfor failed cemented hip arthroplasties They proposed the hypothesis of the wearproduct transport equilibrium If the quantity of wear products like PMMAparticles exceeds the transport capacity wear products accumulate and lead tocellular reaction resulting in in ltrating granulomas periprosthetic bone resorptionand eventually prosthetic loosening On the other hand Jones [5] concluded that thebiomechanical properties of bone cements are mainly responsible for the lsquocementdiseasersquo phenomon Linder [6] suggested that the bone formation at the bone cementinterface in failed hip arthroplasties results from an attempt of tissue to stabilizethe prosthesis Though the exact mechanism and the role of bone cement in thepathogenesis is still unclear in literature includes PMMA among the factors thatare responsible for aseptic loosening of hip arthroplasties Santavirta et al [7] foundthat PMMA powder is an immunological inert material whereas Gil-Albarova [8]reported an activated immune response to PMMA Since many years hydroxyapatite(HA) ceramics are used as bone substitutes and bone graft materials includingcoatings on metal components for dental and orthopaedic applications [8] HApromotes the degradation of serum proteins and mineralization at its surface [9 10]These osteoconductive properties suggest direct physicochemical binding betweenthe HA material and surrounding bone resulting in a high mechanical stability [11]The tissue response to porous HA implants is different from that to dense HAwith regard to the opportunity for bone in-growth Because of its low surfacearea dense HA has demonstrated low rates of biodegradation [12] On the otherhand osteoconductive HA implants with a critical degree of porosity can undergo asigni cant degree of resorption For example the resorption rates of coral-derivedHA which has micropores of 1ndash5 sup1m in diameter ranged from 2 up to 5 peryear [12]
The aim of this study was to compare the biocompatibility of a PMMA bonecement to PMMA-HA cement on a long term human bone marrow cell culturesystem in vitro Our hypothesis was that the presence of HA particles can reduceforeign body reactions and stimulate osteoblastic differentiation Cell numbervitality and morphology chemical cement analysis cellular immuno-differentiationpro le on bone cements analysis of extracellular matrix components and numberof differentiated osteoblasts served as biocompatibility parameters
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1285
MATERIALS AND METHODS
Cement specimens
Vacuum mixed polymerized discoid bone cement probes (Palacos Rreg OsteopalHAreg) with 25 cm in diameter and a thickness of 2 mm were industrial compounds(Merck Germany) (Fig 1) The components were poly(methylmethacrylate)(PMMA) methylmethacrylate powder (MMA MW 10012 gmol melting pointiexcl48plusmnC boiling point C100plusmnC solubility in H2O20plusmnC 115 g l refraction indexn20plusmn=D 14140 density (d20=4plusmn) 0942ndash0944) and HA The particle size of HA(stoichiometry Ca10(PO4)6(OH)2) granules at sintered temperatures gt 1200plusmnC inPMMA-HA was lt10 sup1m the concentration in PMMA powder was 146 wt Thedensity of the HA-particles was 315 gcm3 the density of PMMA-HA cement was134 gcm3 The CaP ratio of the crystalline HA was 167 Zirconium was used asradiopaque agent
Cell culture
The bone marrow cell-culture system was developed by Dexter and Kalla andmodi ed by Wilke et al [13ndash18] A 7 pound 10 pound 20 mm spongial bone marrowblock was taken from the proximal femora during rst hip arthroplasty surgeryafter femoral neck osteotomy Patient criteria for inclusion 50ndash60-year-old whitefemales with osteoarthritis of the hip quali ed for autologous blood transfusion andbone donation Criteria for exclusion acute or chronic infections and neoplasms
Figure 1 PMMA and PMMA-HA cement probes in macroscopic geometry The discoid poly-merized specimens have an surface area of 42 cm2 and were industrial delivered as Palacos Rreg
(PMMA) and Osteopal HAreg (PMMA-HA) Sterilisation was done by radiation
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1286 M Jaumlger and A Wilke
in the medical history We analyzed the bone marrow of four probands (averageage 563 years) by Scanning Electron Microscopy (SEM) Light Microscopy(LM) dot blot Fluorescence Activated Cell Sorter (FACS) analysis cell vitalitytest (propidium bromide staining) osteocalcin staining and PO3iexcl
4 Ca2C detectionin culture supernatants The marrow cells were isolated by Percoll gradientcentrifugation and incubated with Iscoversquos modi ed Dulbeccorsquos Medium (IMDM)including 125 fetal calf serum 125 horse serum 24 ngml hydrocortisone03 NaHCO3 18 pound 106 cells were seeded onto a specimen surface of 491 cm2Growth promoters added after one day of cultivation were 10 ng IL-3 and 10 ngGM-CSF per ml IMDM medium All incubations were carried out at 37plusmnC in 5CO2 95 air Control groups cells without contact to bone cement cultivatedin Petri dishes For FACS analysis cells were seeded on (a) glass slides and (b)PMMAPMMA-HA discs in culture medium and PBS buffer solution without bonemarrow cells Medium was changed every fourth day
Scanning electron microscopy (SEM) SEM-EDX and atomic force microscopy(AFM)
The adherent cells were xed by ITO- xans (2 glutaraldehyde in 01 M cacodylatebuffer Serva Germany) pH 74 for 2 h dehydrated in graded ethanol and tert-butanol and vacuum dried for 36 h Afterwards cement specimens were carbonsputter coated (sputter coater S150B Edwards UK) Specimens were examinedby SEM Cam Scan 4 DV (Cam Scan UK) which was connected with the EDX-analyzer Voyager 2110 including an ultra-thin window (UTW Noran InstrumentsUSA) and photographed Surface roughness of uncoated cement probes wereinvestigated by AFM Nanoscope IIIa (Digital Instruments USA) in tapping modeHuman bone marrow cells from 10 donors were analyzed at 2 4 8 and 10 weeksculture time
FACS analysis
The mechanical removal of adherent marrow cells was supported enzymatically by005 trypsin (Sigma St Louis MO USA) and controlled by episcopic lightmicroscopy (Aristomet Leitz Germany) Afterwards antibody incubation againstthe CD-antigen of T-lymphocytes (CD-2 Behring code BMA 0111-FITC ProductNCPE1011 Behring Germany) mononuclear cells (CD-14 Behring code BMA0211-FITC Product NCPE 1011 Behring) granulocytes (CD-15 BMA 0200-FITC Behring) B-lymphocytes (CD-22 No F5546 Sigma) and hematopoieticstem cells (CD-34 Product 345801 BD Bioscience Germany) followed For theprimary antibody conditions were 1 h incubation time at 4plusmnC concentraion 1 10in TBS 2 BSA For the secondary antibody conditions were anti-mouse FITC-labeled (Sigma) concentration 1 10 TBS 2 BSA 1 h incubation at RT Propidiumbromide labelling served to identify dead cells before FACS analysis (FACSTARPlus Becton-Dickinson USA) started The culture period for FACS analysis was
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1287
14 days Control groups human bone marrow cells cultivated on (a) cover glassslides and (b) fresh human marrow cells 12 h after isolation
Dot blot analysis
Type-I and -III collagen and alkaline phosphatase (ALP) were detected by semi-quantitative dot blot analysis of culture medium supernants based on an anti-body system Primary antibody rabbit anti-human collagen I III (ChemiconUSA) secondary antibody anti-rabbit IgG biotin F0
ab2-fragment (Boehringer Ger-many) Streptavidine-peroxidase (Boehringer) and nitroblue-tetrazolium (SigmaUSA) served as staining chemicals The study period was 72 days
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis of collected culture medium was done by Hitachi Automatic
Analyzer (Boehringer) Study period was 23 days
Osteoblast staining
At 14 days of cultivation cement adherent marrow cells were washed in TBS bufferand incubated with a goat anti-osteocalcin antibody (Fitzgerald USA) over 24 h at4plusmnC After washing in TBS the second antibody (anti-goat biotin conjugate SigmaUSA) was added and incubated for 1 h RT Following another TBS washing thethird antibody (extraavidine-alkaline phosphatase Sigma Germany) was added for1 h For staining Neu-Fuchsin solution (Sigma USA) was used The number ofosteocalcin-positive cells per mm2 were measured by eye LM analysis
Statistics
Because of former experiments we could assume in this study a KolmogorovndashSmirnov-validated normal distribution of human marrow cells in regard to differ-entiation parameters (osteocalcin and CD-markers) collagen-I and -III productionand cell number of adherent cells To get statistical-based results we used theMannndashWhitney test for independent statistical groups P lt 001 was rated highlystatistical signi cant and P lt 005 statistically signi cant whereas P gt 005 wasconsidered not signi cant
RESULTS
Cement analysis before cellular contact (primary material analysis) showed nosigni cant impurities in cement specimens corresponding to the manufacturerrsquosstatement in SEM-EDX Surface roughness of probes was not well de ned andshowed high horizontal and vertical distance variations (nmndashsup1m) documented bySEM and AFM The SEM photo in Fig 2 shows the PMMA-HA morphology before
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1288 M Jaumlger and A Wilke
Figure 2 SEM analysis of PMMA-HA bone cement specimen without human marrow cells tippedover 60plusmn Big round ball-shaped PMMA polymers (arrows) are surrounded by HA-granules Therough surface enlarges the cell attachment area
we started cell culturing The corresponding data based on a SEM-EDX analysis areshown in Fig 3 The AFM analysis con rmed the rough irregular surface structureWe found signs of surface defects for PMMA and PMMA-HA (Fig 4)
Morphology and cell number
SEM analysis showed signi cantly much fewer cells adhered to PMMA-HA thanto PMMA slices at two weeks The analysis of ten specimens showed 268 lessadherent cells per area for PMMA-HA in comparison to PMMA (mean valuesPMMA 906 cells mm2 (SD 192) versus PMMA-HA 663 cells mm2 (SD 232))With increasing culture time the cells tend to build up clusters (cell aggregates)Cellular clusters were found for both cements from day 7 up to day 72 (end ofstudy) The SEM photo in Fig 5 shows at adherent marrow cells on PMMA-HAsurface Furthermore round cells can be seen (dead or detached) after 10 weeks ofincubation (Fig 5)
Cell differentiation and cell vitality
After detaching the cultivated marrow cells from cement probes the number of deadcells was lt10 in all experiments as detected by propidium bromide staining Atday 14 cell culture demonstrated a high number of T-lymphocytes and granulocytes
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1289
Fig
ure
3S
EM
-ED
Xan
alys
isba
sed
onth
ear
eaof
Fig
2
No
addi
tion
alel
emen
tsw
ere
dete
cted
and
the
empi
rica
lfor
mul
aC
a 5(P
O4)
3OH
2is
con
rmed
byqu
anti
tativ
ean
alys
is
O
oxyg
en
Cc
arbo
nC
aca
lciu
mP
pho
spho
rum
Z
irco
nium
(Zr)
serv
edas
X-r
ayco
ntra
stm
ediu
man
dis
notv
isib
lehe
rebe
caus
eof
the
over
lapp
ing
byP
-pea
k(L
-reg-p
eak)
A
ccel
erat
ing
ener
gy
10ke
Vl
ife
tim
e20
0s
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1290 M Jaumlger and A Wilke
Fig
ure
4A
FM
surf
ace
anal
ysis
ofa
PM
MA
spec
imen
inta
ppin
gm
ode
Ove
rvie
wof
aP
MM
Ace
men
tsur
face
area
B
esid
esth
eir
regu
lar
surf
ace
stru
ctur
ea
scra
tch
asa
sign
ofm
ater
iald
efec
tcan
bede
tect
ed(b
lack
arro
w)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1291
Figure 5 Adherent at human bone marrow cells after 10 weeks of incubation on PMMA-HABesides at adherent cells detached round cells are shown The surrounding extracellularmatrix wasrich of sulfur (SEM-EDX)
in case of PMMA in FACS analysis whereas the differentiation rate of monocytesshowed only few differences between both cements (Fig 6) Human bone marrowcells on PMMA-HA differentiated more B-cells than those which were cultivatedon PMMA but less than the glass control group
Immunohistochemical osteocalcin staining reported the superiority of the PMMA-HA cement with regard to the induction of an osteoblastic differentiation Thenumber of osteocalcin positive cells was 39-fold higher for PMMA-HA than formarrow cells which were seeded on PMMA P lt 0001 For PMMA we found inaverage 39 pound 103 cells mm2 (SD 178) compared to 151 pound 103 osteocalcin-positivecells mm2 (SD 191) in the case of PMMA-HA Parallel to the high amount ofosteocalcin positive cells on the PMMA-HA surface we detected a higher rate ofintercellular bridging
Extracellular matrix (ECM)
SEM-EDX analysis showed a sulfur-rich ECM which was located only pericyto-plasmatic In the case of PMMA sulfur was detected after 10 weeks incubationwhereas for the PMMA-HA cement sulfur already appeared after two weeks Fig-ure 7 shows the pericellular EDX analysis of human marrow cells after differentincubation times with PMMA or PMMA-HA cement The semi-quantatively ana-lyzed areas showed an increasing content of sulfur as a function of time for PMMA-
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1292 M Jaumlger and A Wilke
Figure 6 FACS analysis of bone cement adherent marrow cells 14 days after incubation with PMMAand PMMA-HA in comparison with control group Fresh marrow cells 12 h after isolation from theproximal femur (native) and after 14 incubation days onto glass slides The table below demonstratesthe values the bar charts in the diagram the standard deviation PMMA-HA shows a decreasing rateof T-lymhocytes in comparison to PMMA but the number of granulocytes (416) are signi cantlyhigher for PMMA (203) P lt 0005 There is nearly no difference of the number of mononuclearcells between both cements (PMMA 308 PMMA-HA 319) P gt 0005
HA We found an amorphous protein precipitate covering the cement surface in ex-periments with and without cells Sulfur was only detected by SEM-EDX in celladherent specimens
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis showed that HA-particles in PMMA-HA were stable in an
aquaeous medium during the 23-day study period The average concentration valuesfor PMMA cell culture medium were Ca2C 232 mmol l PO3iexcl
4 158 mmol l(SD 0041) and for PMMA-HA Ca2C 231 mmol l PO3iexcl
4 159 mmol l (SD 0026)Dot blot analysis of the collected culture medium supernatants was carried out
semiquantitative for collagen I III and alkaline phosphatase (ALP) over a studyperiod of 72 days The AP activity and collagen-III detection showed no differencesregarding the rythm of appearance for both cements In the case of PMMA-HAcollagen I was synthesized at day 6 and showed a signi cant staining reaction from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1285
MATERIALS AND METHODS
Cement specimens
Vacuum mixed polymerized discoid bone cement probes (Palacos Rreg OsteopalHAreg) with 25 cm in diameter and a thickness of 2 mm were industrial compounds(Merck Germany) (Fig 1) The components were poly(methylmethacrylate)(PMMA) methylmethacrylate powder (MMA MW 10012 gmol melting pointiexcl48plusmnC boiling point C100plusmnC solubility in H2O20plusmnC 115 g l refraction indexn20plusmn=D 14140 density (d20=4plusmn) 0942ndash0944) and HA The particle size of HA(stoichiometry Ca10(PO4)6(OH)2) granules at sintered temperatures gt 1200plusmnC inPMMA-HA was lt10 sup1m the concentration in PMMA powder was 146 wt Thedensity of the HA-particles was 315 gcm3 the density of PMMA-HA cement was134 gcm3 The CaP ratio of the crystalline HA was 167 Zirconium was used asradiopaque agent
Cell culture
The bone marrow cell-culture system was developed by Dexter and Kalla andmodi ed by Wilke et al [13ndash18] A 7 pound 10 pound 20 mm spongial bone marrowblock was taken from the proximal femora during rst hip arthroplasty surgeryafter femoral neck osteotomy Patient criteria for inclusion 50ndash60-year-old whitefemales with osteoarthritis of the hip quali ed for autologous blood transfusion andbone donation Criteria for exclusion acute or chronic infections and neoplasms
Figure 1 PMMA and PMMA-HA cement probes in macroscopic geometry The discoid poly-merized specimens have an surface area of 42 cm2 and were industrial delivered as Palacos Rreg
(PMMA) and Osteopal HAreg (PMMA-HA) Sterilisation was done by radiation
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1286 M Jaumlger and A Wilke
in the medical history We analyzed the bone marrow of four probands (averageage 563 years) by Scanning Electron Microscopy (SEM) Light Microscopy(LM) dot blot Fluorescence Activated Cell Sorter (FACS) analysis cell vitalitytest (propidium bromide staining) osteocalcin staining and PO3iexcl
4 Ca2C detectionin culture supernatants The marrow cells were isolated by Percoll gradientcentrifugation and incubated with Iscoversquos modi ed Dulbeccorsquos Medium (IMDM)including 125 fetal calf serum 125 horse serum 24 ngml hydrocortisone03 NaHCO3 18 pound 106 cells were seeded onto a specimen surface of 491 cm2Growth promoters added after one day of cultivation were 10 ng IL-3 and 10 ngGM-CSF per ml IMDM medium All incubations were carried out at 37plusmnC in 5CO2 95 air Control groups cells without contact to bone cement cultivatedin Petri dishes For FACS analysis cells were seeded on (a) glass slides and (b)PMMAPMMA-HA discs in culture medium and PBS buffer solution without bonemarrow cells Medium was changed every fourth day
Scanning electron microscopy (SEM) SEM-EDX and atomic force microscopy(AFM)
The adherent cells were xed by ITO- xans (2 glutaraldehyde in 01 M cacodylatebuffer Serva Germany) pH 74 for 2 h dehydrated in graded ethanol and tert-butanol and vacuum dried for 36 h Afterwards cement specimens were carbonsputter coated (sputter coater S150B Edwards UK) Specimens were examinedby SEM Cam Scan 4 DV (Cam Scan UK) which was connected with the EDX-analyzer Voyager 2110 including an ultra-thin window (UTW Noran InstrumentsUSA) and photographed Surface roughness of uncoated cement probes wereinvestigated by AFM Nanoscope IIIa (Digital Instruments USA) in tapping modeHuman bone marrow cells from 10 donors were analyzed at 2 4 8 and 10 weeksculture time
FACS analysis
The mechanical removal of adherent marrow cells was supported enzymatically by005 trypsin (Sigma St Louis MO USA) and controlled by episcopic lightmicroscopy (Aristomet Leitz Germany) Afterwards antibody incubation againstthe CD-antigen of T-lymphocytes (CD-2 Behring code BMA 0111-FITC ProductNCPE1011 Behring Germany) mononuclear cells (CD-14 Behring code BMA0211-FITC Product NCPE 1011 Behring) granulocytes (CD-15 BMA 0200-FITC Behring) B-lymphocytes (CD-22 No F5546 Sigma) and hematopoieticstem cells (CD-34 Product 345801 BD Bioscience Germany) followed For theprimary antibody conditions were 1 h incubation time at 4plusmnC concentraion 1 10in TBS 2 BSA For the secondary antibody conditions were anti-mouse FITC-labeled (Sigma) concentration 1 10 TBS 2 BSA 1 h incubation at RT Propidiumbromide labelling served to identify dead cells before FACS analysis (FACSTARPlus Becton-Dickinson USA) started The culture period for FACS analysis was
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1287
14 days Control groups human bone marrow cells cultivated on (a) cover glassslides and (b) fresh human marrow cells 12 h after isolation
Dot blot analysis
Type-I and -III collagen and alkaline phosphatase (ALP) were detected by semi-quantitative dot blot analysis of culture medium supernants based on an anti-body system Primary antibody rabbit anti-human collagen I III (ChemiconUSA) secondary antibody anti-rabbit IgG biotin F0
ab2-fragment (Boehringer Ger-many) Streptavidine-peroxidase (Boehringer) and nitroblue-tetrazolium (SigmaUSA) served as staining chemicals The study period was 72 days
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis of collected culture medium was done by Hitachi Automatic
Analyzer (Boehringer) Study period was 23 days
Osteoblast staining
At 14 days of cultivation cement adherent marrow cells were washed in TBS bufferand incubated with a goat anti-osteocalcin antibody (Fitzgerald USA) over 24 h at4plusmnC After washing in TBS the second antibody (anti-goat biotin conjugate SigmaUSA) was added and incubated for 1 h RT Following another TBS washing thethird antibody (extraavidine-alkaline phosphatase Sigma Germany) was added for1 h For staining Neu-Fuchsin solution (Sigma USA) was used The number ofosteocalcin-positive cells per mm2 were measured by eye LM analysis
Statistics
Because of former experiments we could assume in this study a KolmogorovndashSmirnov-validated normal distribution of human marrow cells in regard to differ-entiation parameters (osteocalcin and CD-markers) collagen-I and -III productionand cell number of adherent cells To get statistical-based results we used theMannndashWhitney test for independent statistical groups P lt 001 was rated highlystatistical signi cant and P lt 005 statistically signi cant whereas P gt 005 wasconsidered not signi cant
RESULTS
Cement analysis before cellular contact (primary material analysis) showed nosigni cant impurities in cement specimens corresponding to the manufacturerrsquosstatement in SEM-EDX Surface roughness of probes was not well de ned andshowed high horizontal and vertical distance variations (nmndashsup1m) documented bySEM and AFM The SEM photo in Fig 2 shows the PMMA-HA morphology before
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1288 M Jaumlger and A Wilke
Figure 2 SEM analysis of PMMA-HA bone cement specimen without human marrow cells tippedover 60plusmn Big round ball-shaped PMMA polymers (arrows) are surrounded by HA-granules Therough surface enlarges the cell attachment area
we started cell culturing The corresponding data based on a SEM-EDX analysis areshown in Fig 3 The AFM analysis con rmed the rough irregular surface structureWe found signs of surface defects for PMMA and PMMA-HA (Fig 4)
Morphology and cell number
SEM analysis showed signi cantly much fewer cells adhered to PMMA-HA thanto PMMA slices at two weeks The analysis of ten specimens showed 268 lessadherent cells per area for PMMA-HA in comparison to PMMA (mean valuesPMMA 906 cells mm2 (SD 192) versus PMMA-HA 663 cells mm2 (SD 232))With increasing culture time the cells tend to build up clusters (cell aggregates)Cellular clusters were found for both cements from day 7 up to day 72 (end ofstudy) The SEM photo in Fig 5 shows at adherent marrow cells on PMMA-HAsurface Furthermore round cells can be seen (dead or detached) after 10 weeks ofincubation (Fig 5)
Cell differentiation and cell vitality
After detaching the cultivated marrow cells from cement probes the number of deadcells was lt10 in all experiments as detected by propidium bromide staining Atday 14 cell culture demonstrated a high number of T-lymphocytes and granulocytes
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1289
Fig
ure
3S
EM
-ED
Xan
alys
isba
sed
onth
ear
eaof
Fig
2
No
addi
tion
alel
emen
tsw
ere
dete
cted
and
the
empi
rica
lfor
mul
aC
a 5(P
O4)
3OH
2is
con
rmed
byqu
anti
tativ
ean
alys
is
O
oxyg
en
Cc
arbo
nC
aca
lciu
mP
pho
spho
rum
Z
irco
nium
(Zr)
serv
edas
X-r
ayco
ntra
stm
ediu
man
dis
notv
isib
lehe
rebe
caus
eof
the
over
lapp
ing
byP
-pea
k(L
-reg-p
eak)
A
ccel
erat
ing
ener
gy
10ke
Vl
ife
tim
e20
0s
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1290 M Jaumlger and A Wilke
Fig
ure
4A
FM
surf
ace
anal
ysis
ofa
PM
MA
spec
imen
inta
ppin
gm
ode
Ove
rvie
wof
aP
MM
Ace
men
tsur
face
area
B
esid
esth
eir
regu
lar
surf
ace
stru
ctur
ea
scra
tch
asa
sign
ofm
ater
iald
efec
tcan
bede
tect
ed(b
lack
arro
w)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1291
Figure 5 Adherent at human bone marrow cells after 10 weeks of incubation on PMMA-HABesides at adherent cells detached round cells are shown The surrounding extracellularmatrix wasrich of sulfur (SEM-EDX)
in case of PMMA in FACS analysis whereas the differentiation rate of monocytesshowed only few differences between both cements (Fig 6) Human bone marrowcells on PMMA-HA differentiated more B-cells than those which were cultivatedon PMMA but less than the glass control group
Immunohistochemical osteocalcin staining reported the superiority of the PMMA-HA cement with regard to the induction of an osteoblastic differentiation Thenumber of osteocalcin positive cells was 39-fold higher for PMMA-HA than formarrow cells which were seeded on PMMA P lt 0001 For PMMA we found inaverage 39 pound 103 cells mm2 (SD 178) compared to 151 pound 103 osteocalcin-positivecells mm2 (SD 191) in the case of PMMA-HA Parallel to the high amount ofosteocalcin positive cells on the PMMA-HA surface we detected a higher rate ofintercellular bridging
Extracellular matrix (ECM)
SEM-EDX analysis showed a sulfur-rich ECM which was located only pericyto-plasmatic In the case of PMMA sulfur was detected after 10 weeks incubationwhereas for the PMMA-HA cement sulfur already appeared after two weeks Fig-ure 7 shows the pericellular EDX analysis of human marrow cells after differentincubation times with PMMA or PMMA-HA cement The semi-quantatively ana-lyzed areas showed an increasing content of sulfur as a function of time for PMMA-
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1292 M Jaumlger and A Wilke
Figure 6 FACS analysis of bone cement adherent marrow cells 14 days after incubation with PMMAand PMMA-HA in comparison with control group Fresh marrow cells 12 h after isolation from theproximal femur (native) and after 14 incubation days onto glass slides The table below demonstratesthe values the bar charts in the diagram the standard deviation PMMA-HA shows a decreasing rateof T-lymhocytes in comparison to PMMA but the number of granulocytes (416) are signi cantlyhigher for PMMA (203) P lt 0005 There is nearly no difference of the number of mononuclearcells between both cements (PMMA 308 PMMA-HA 319) P gt 0005
HA We found an amorphous protein precipitate covering the cement surface in ex-periments with and without cells Sulfur was only detected by SEM-EDX in celladherent specimens
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis showed that HA-particles in PMMA-HA were stable in an
aquaeous medium during the 23-day study period The average concentration valuesfor PMMA cell culture medium were Ca2C 232 mmol l PO3iexcl
4 158 mmol l(SD 0041) and for PMMA-HA Ca2C 231 mmol l PO3iexcl
4 159 mmol l (SD 0026)Dot blot analysis of the collected culture medium supernatants was carried out
semiquantitative for collagen I III and alkaline phosphatase (ALP) over a studyperiod of 72 days The AP activity and collagen-III detection showed no differencesregarding the rythm of appearance for both cements In the case of PMMA-HAcollagen I was synthesized at day 6 and showed a signi cant staining reaction from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1286 M Jaumlger and A Wilke
in the medical history We analyzed the bone marrow of four probands (averageage 563 years) by Scanning Electron Microscopy (SEM) Light Microscopy(LM) dot blot Fluorescence Activated Cell Sorter (FACS) analysis cell vitalitytest (propidium bromide staining) osteocalcin staining and PO3iexcl
4 Ca2C detectionin culture supernatants The marrow cells were isolated by Percoll gradientcentrifugation and incubated with Iscoversquos modi ed Dulbeccorsquos Medium (IMDM)including 125 fetal calf serum 125 horse serum 24 ngml hydrocortisone03 NaHCO3 18 pound 106 cells were seeded onto a specimen surface of 491 cm2Growth promoters added after one day of cultivation were 10 ng IL-3 and 10 ngGM-CSF per ml IMDM medium All incubations were carried out at 37plusmnC in 5CO2 95 air Control groups cells without contact to bone cement cultivatedin Petri dishes For FACS analysis cells were seeded on (a) glass slides and (b)PMMAPMMA-HA discs in culture medium and PBS buffer solution without bonemarrow cells Medium was changed every fourth day
Scanning electron microscopy (SEM) SEM-EDX and atomic force microscopy(AFM)
The adherent cells were xed by ITO- xans (2 glutaraldehyde in 01 M cacodylatebuffer Serva Germany) pH 74 for 2 h dehydrated in graded ethanol and tert-butanol and vacuum dried for 36 h Afterwards cement specimens were carbonsputter coated (sputter coater S150B Edwards UK) Specimens were examinedby SEM Cam Scan 4 DV (Cam Scan UK) which was connected with the EDX-analyzer Voyager 2110 including an ultra-thin window (UTW Noran InstrumentsUSA) and photographed Surface roughness of uncoated cement probes wereinvestigated by AFM Nanoscope IIIa (Digital Instruments USA) in tapping modeHuman bone marrow cells from 10 donors were analyzed at 2 4 8 and 10 weeksculture time
FACS analysis
The mechanical removal of adherent marrow cells was supported enzymatically by005 trypsin (Sigma St Louis MO USA) and controlled by episcopic lightmicroscopy (Aristomet Leitz Germany) Afterwards antibody incubation againstthe CD-antigen of T-lymphocytes (CD-2 Behring code BMA 0111-FITC ProductNCPE1011 Behring Germany) mononuclear cells (CD-14 Behring code BMA0211-FITC Product NCPE 1011 Behring) granulocytes (CD-15 BMA 0200-FITC Behring) B-lymphocytes (CD-22 No F5546 Sigma) and hematopoieticstem cells (CD-34 Product 345801 BD Bioscience Germany) followed For theprimary antibody conditions were 1 h incubation time at 4plusmnC concentraion 1 10in TBS 2 BSA For the secondary antibody conditions were anti-mouse FITC-labeled (Sigma) concentration 1 10 TBS 2 BSA 1 h incubation at RT Propidiumbromide labelling served to identify dead cells before FACS analysis (FACSTARPlus Becton-Dickinson USA) started The culture period for FACS analysis was
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1287
14 days Control groups human bone marrow cells cultivated on (a) cover glassslides and (b) fresh human marrow cells 12 h after isolation
Dot blot analysis
Type-I and -III collagen and alkaline phosphatase (ALP) were detected by semi-quantitative dot blot analysis of culture medium supernants based on an anti-body system Primary antibody rabbit anti-human collagen I III (ChemiconUSA) secondary antibody anti-rabbit IgG biotin F0
ab2-fragment (Boehringer Ger-many) Streptavidine-peroxidase (Boehringer) and nitroblue-tetrazolium (SigmaUSA) served as staining chemicals The study period was 72 days
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis of collected culture medium was done by Hitachi Automatic
Analyzer (Boehringer) Study period was 23 days
Osteoblast staining
At 14 days of cultivation cement adherent marrow cells were washed in TBS bufferand incubated with a goat anti-osteocalcin antibody (Fitzgerald USA) over 24 h at4plusmnC After washing in TBS the second antibody (anti-goat biotin conjugate SigmaUSA) was added and incubated for 1 h RT Following another TBS washing thethird antibody (extraavidine-alkaline phosphatase Sigma Germany) was added for1 h For staining Neu-Fuchsin solution (Sigma USA) was used The number ofosteocalcin-positive cells per mm2 were measured by eye LM analysis
Statistics
Because of former experiments we could assume in this study a KolmogorovndashSmirnov-validated normal distribution of human marrow cells in regard to differ-entiation parameters (osteocalcin and CD-markers) collagen-I and -III productionand cell number of adherent cells To get statistical-based results we used theMannndashWhitney test for independent statistical groups P lt 001 was rated highlystatistical signi cant and P lt 005 statistically signi cant whereas P gt 005 wasconsidered not signi cant
RESULTS
Cement analysis before cellular contact (primary material analysis) showed nosigni cant impurities in cement specimens corresponding to the manufacturerrsquosstatement in SEM-EDX Surface roughness of probes was not well de ned andshowed high horizontal and vertical distance variations (nmndashsup1m) documented bySEM and AFM The SEM photo in Fig 2 shows the PMMA-HA morphology before
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1288 M Jaumlger and A Wilke
Figure 2 SEM analysis of PMMA-HA bone cement specimen without human marrow cells tippedover 60plusmn Big round ball-shaped PMMA polymers (arrows) are surrounded by HA-granules Therough surface enlarges the cell attachment area
we started cell culturing The corresponding data based on a SEM-EDX analysis areshown in Fig 3 The AFM analysis con rmed the rough irregular surface structureWe found signs of surface defects for PMMA and PMMA-HA (Fig 4)
Morphology and cell number
SEM analysis showed signi cantly much fewer cells adhered to PMMA-HA thanto PMMA slices at two weeks The analysis of ten specimens showed 268 lessadherent cells per area for PMMA-HA in comparison to PMMA (mean valuesPMMA 906 cells mm2 (SD 192) versus PMMA-HA 663 cells mm2 (SD 232))With increasing culture time the cells tend to build up clusters (cell aggregates)Cellular clusters were found for both cements from day 7 up to day 72 (end ofstudy) The SEM photo in Fig 5 shows at adherent marrow cells on PMMA-HAsurface Furthermore round cells can be seen (dead or detached) after 10 weeks ofincubation (Fig 5)
Cell differentiation and cell vitality
After detaching the cultivated marrow cells from cement probes the number of deadcells was lt10 in all experiments as detected by propidium bromide staining Atday 14 cell culture demonstrated a high number of T-lymphocytes and granulocytes
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1289
Fig
ure
3S
EM
-ED
Xan
alys
isba
sed
onth
ear
eaof
Fig
2
No
addi
tion
alel
emen
tsw
ere
dete
cted
and
the
empi
rica
lfor
mul
aC
a 5(P
O4)
3OH
2is
con
rmed
byqu
anti
tativ
ean
alys
is
O
oxyg
en
Cc
arbo
nC
aca
lciu
mP
pho
spho
rum
Z
irco
nium
(Zr)
serv
edas
X-r
ayco
ntra
stm
ediu
man
dis
notv
isib
lehe
rebe
caus
eof
the
over
lapp
ing
byP
-pea
k(L
-reg-p
eak)
A
ccel
erat
ing
ener
gy
10ke
Vl
ife
tim
e20
0s
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1290 M Jaumlger and A Wilke
Fig
ure
4A
FM
surf
ace
anal
ysis
ofa
PM
MA
spec
imen
inta
ppin
gm
ode
Ove
rvie
wof
aP
MM
Ace
men
tsur
face
area
B
esid
esth
eir
regu
lar
surf
ace
stru
ctur
ea
scra
tch
asa
sign
ofm
ater
iald
efec
tcan
bede
tect
ed(b
lack
arro
w)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1291
Figure 5 Adherent at human bone marrow cells after 10 weeks of incubation on PMMA-HABesides at adherent cells detached round cells are shown The surrounding extracellularmatrix wasrich of sulfur (SEM-EDX)
in case of PMMA in FACS analysis whereas the differentiation rate of monocytesshowed only few differences between both cements (Fig 6) Human bone marrowcells on PMMA-HA differentiated more B-cells than those which were cultivatedon PMMA but less than the glass control group
Immunohistochemical osteocalcin staining reported the superiority of the PMMA-HA cement with regard to the induction of an osteoblastic differentiation Thenumber of osteocalcin positive cells was 39-fold higher for PMMA-HA than formarrow cells which were seeded on PMMA P lt 0001 For PMMA we found inaverage 39 pound 103 cells mm2 (SD 178) compared to 151 pound 103 osteocalcin-positivecells mm2 (SD 191) in the case of PMMA-HA Parallel to the high amount ofosteocalcin positive cells on the PMMA-HA surface we detected a higher rate ofintercellular bridging
Extracellular matrix (ECM)
SEM-EDX analysis showed a sulfur-rich ECM which was located only pericyto-plasmatic In the case of PMMA sulfur was detected after 10 weeks incubationwhereas for the PMMA-HA cement sulfur already appeared after two weeks Fig-ure 7 shows the pericellular EDX analysis of human marrow cells after differentincubation times with PMMA or PMMA-HA cement The semi-quantatively ana-lyzed areas showed an increasing content of sulfur as a function of time for PMMA-
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1292 M Jaumlger and A Wilke
Figure 6 FACS analysis of bone cement adherent marrow cells 14 days after incubation with PMMAand PMMA-HA in comparison with control group Fresh marrow cells 12 h after isolation from theproximal femur (native) and after 14 incubation days onto glass slides The table below demonstratesthe values the bar charts in the diagram the standard deviation PMMA-HA shows a decreasing rateof T-lymhocytes in comparison to PMMA but the number of granulocytes (416) are signi cantlyhigher for PMMA (203) P lt 0005 There is nearly no difference of the number of mononuclearcells between both cements (PMMA 308 PMMA-HA 319) P gt 0005
HA We found an amorphous protein precipitate covering the cement surface in ex-periments with and without cells Sulfur was only detected by SEM-EDX in celladherent specimens
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis showed that HA-particles in PMMA-HA were stable in an
aquaeous medium during the 23-day study period The average concentration valuesfor PMMA cell culture medium were Ca2C 232 mmol l PO3iexcl
4 158 mmol l(SD 0041) and for PMMA-HA Ca2C 231 mmol l PO3iexcl
4 159 mmol l (SD 0026)Dot blot analysis of the collected culture medium supernatants was carried out
semiquantitative for collagen I III and alkaline phosphatase (ALP) over a studyperiod of 72 days The AP activity and collagen-III detection showed no differencesregarding the rythm of appearance for both cements In the case of PMMA-HAcollagen I was synthesized at day 6 and showed a signi cant staining reaction from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1287
14 days Control groups human bone marrow cells cultivated on (a) cover glassslides and (b) fresh human marrow cells 12 h after isolation
Dot blot analysis
Type-I and -III collagen and alkaline phosphatase (ALP) were detected by semi-quantitative dot blot analysis of culture medium supernants based on an anti-body system Primary antibody rabbit anti-human collagen I III (ChemiconUSA) secondary antibody anti-rabbit IgG biotin F0
ab2-fragment (Boehringer Ger-many) Streptavidine-peroxidase (Boehringer) and nitroblue-tetrazolium (SigmaUSA) served as staining chemicals The study period was 72 days
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis of collected culture medium was done by Hitachi Automatic
Analyzer (Boehringer) Study period was 23 days
Osteoblast staining
At 14 days of cultivation cement adherent marrow cells were washed in TBS bufferand incubated with a goat anti-osteocalcin antibody (Fitzgerald USA) over 24 h at4plusmnC After washing in TBS the second antibody (anti-goat biotin conjugate SigmaUSA) was added and incubated for 1 h RT Following another TBS washing thethird antibody (extraavidine-alkaline phosphatase Sigma Germany) was added for1 h For staining Neu-Fuchsin solution (Sigma USA) was used The number ofosteocalcin-positive cells per mm2 were measured by eye LM analysis
Statistics
Because of former experiments we could assume in this study a KolmogorovndashSmirnov-validated normal distribution of human marrow cells in regard to differ-entiation parameters (osteocalcin and CD-markers) collagen-I and -III productionand cell number of adherent cells To get statistical-based results we used theMannndashWhitney test for independent statistical groups P lt 001 was rated highlystatistical signi cant and P lt 005 statistically signi cant whereas P gt 005 wasconsidered not signi cant
RESULTS
Cement analysis before cellular contact (primary material analysis) showed nosigni cant impurities in cement specimens corresponding to the manufacturerrsquosstatement in SEM-EDX Surface roughness of probes was not well de ned andshowed high horizontal and vertical distance variations (nmndashsup1m) documented bySEM and AFM The SEM photo in Fig 2 shows the PMMA-HA morphology before
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1288 M Jaumlger and A Wilke
Figure 2 SEM analysis of PMMA-HA bone cement specimen without human marrow cells tippedover 60plusmn Big round ball-shaped PMMA polymers (arrows) are surrounded by HA-granules Therough surface enlarges the cell attachment area
we started cell culturing The corresponding data based on a SEM-EDX analysis areshown in Fig 3 The AFM analysis con rmed the rough irregular surface structureWe found signs of surface defects for PMMA and PMMA-HA (Fig 4)
Morphology and cell number
SEM analysis showed signi cantly much fewer cells adhered to PMMA-HA thanto PMMA slices at two weeks The analysis of ten specimens showed 268 lessadherent cells per area for PMMA-HA in comparison to PMMA (mean valuesPMMA 906 cells mm2 (SD 192) versus PMMA-HA 663 cells mm2 (SD 232))With increasing culture time the cells tend to build up clusters (cell aggregates)Cellular clusters were found for both cements from day 7 up to day 72 (end ofstudy) The SEM photo in Fig 5 shows at adherent marrow cells on PMMA-HAsurface Furthermore round cells can be seen (dead or detached) after 10 weeks ofincubation (Fig 5)
Cell differentiation and cell vitality
After detaching the cultivated marrow cells from cement probes the number of deadcells was lt10 in all experiments as detected by propidium bromide staining Atday 14 cell culture demonstrated a high number of T-lymphocytes and granulocytes
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1289
Fig
ure
3S
EM
-ED
Xan
alys
isba
sed
onth
ear
eaof
Fig
2
No
addi
tion
alel
emen
tsw
ere
dete
cted
and
the
empi
rica
lfor
mul
aC
a 5(P
O4)
3OH
2is
con
rmed
byqu
anti
tativ
ean
alys
is
O
oxyg
en
Cc
arbo
nC
aca
lciu
mP
pho
spho
rum
Z
irco
nium
(Zr)
serv
edas
X-r
ayco
ntra
stm
ediu
man
dis
notv
isib
lehe
rebe
caus
eof
the
over
lapp
ing
byP
-pea
k(L
-reg-p
eak)
A
ccel
erat
ing
ener
gy
10ke
Vl
ife
tim
e20
0s
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1290 M Jaumlger and A Wilke
Fig
ure
4A
FM
surf
ace
anal
ysis
ofa
PM
MA
spec
imen
inta
ppin
gm
ode
Ove
rvie
wof
aP
MM
Ace
men
tsur
face
area
B
esid
esth
eir
regu
lar
surf
ace
stru
ctur
ea
scra
tch
asa
sign
ofm
ater
iald
efec
tcan
bede
tect
ed(b
lack
arro
w)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1291
Figure 5 Adherent at human bone marrow cells after 10 weeks of incubation on PMMA-HABesides at adherent cells detached round cells are shown The surrounding extracellularmatrix wasrich of sulfur (SEM-EDX)
in case of PMMA in FACS analysis whereas the differentiation rate of monocytesshowed only few differences between both cements (Fig 6) Human bone marrowcells on PMMA-HA differentiated more B-cells than those which were cultivatedon PMMA but less than the glass control group
Immunohistochemical osteocalcin staining reported the superiority of the PMMA-HA cement with regard to the induction of an osteoblastic differentiation Thenumber of osteocalcin positive cells was 39-fold higher for PMMA-HA than formarrow cells which were seeded on PMMA P lt 0001 For PMMA we found inaverage 39 pound 103 cells mm2 (SD 178) compared to 151 pound 103 osteocalcin-positivecells mm2 (SD 191) in the case of PMMA-HA Parallel to the high amount ofosteocalcin positive cells on the PMMA-HA surface we detected a higher rate ofintercellular bridging
Extracellular matrix (ECM)
SEM-EDX analysis showed a sulfur-rich ECM which was located only pericyto-plasmatic In the case of PMMA sulfur was detected after 10 weeks incubationwhereas for the PMMA-HA cement sulfur already appeared after two weeks Fig-ure 7 shows the pericellular EDX analysis of human marrow cells after differentincubation times with PMMA or PMMA-HA cement The semi-quantatively ana-lyzed areas showed an increasing content of sulfur as a function of time for PMMA-
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1292 M Jaumlger and A Wilke
Figure 6 FACS analysis of bone cement adherent marrow cells 14 days after incubation with PMMAand PMMA-HA in comparison with control group Fresh marrow cells 12 h after isolation from theproximal femur (native) and after 14 incubation days onto glass slides The table below demonstratesthe values the bar charts in the diagram the standard deviation PMMA-HA shows a decreasing rateof T-lymhocytes in comparison to PMMA but the number of granulocytes (416) are signi cantlyhigher for PMMA (203) P lt 0005 There is nearly no difference of the number of mononuclearcells between both cements (PMMA 308 PMMA-HA 319) P gt 0005
HA We found an amorphous protein precipitate covering the cement surface in ex-periments with and without cells Sulfur was only detected by SEM-EDX in celladherent specimens
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis showed that HA-particles in PMMA-HA were stable in an
aquaeous medium during the 23-day study period The average concentration valuesfor PMMA cell culture medium were Ca2C 232 mmol l PO3iexcl
4 158 mmol l(SD 0041) and for PMMA-HA Ca2C 231 mmol l PO3iexcl
4 159 mmol l (SD 0026)Dot blot analysis of the collected culture medium supernatants was carried out
semiquantitative for collagen I III and alkaline phosphatase (ALP) over a studyperiod of 72 days The AP activity and collagen-III detection showed no differencesregarding the rythm of appearance for both cements In the case of PMMA-HAcollagen I was synthesized at day 6 and showed a signi cant staining reaction from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1288 M Jaumlger and A Wilke
Figure 2 SEM analysis of PMMA-HA bone cement specimen without human marrow cells tippedover 60plusmn Big round ball-shaped PMMA polymers (arrows) are surrounded by HA-granules Therough surface enlarges the cell attachment area
we started cell culturing The corresponding data based on a SEM-EDX analysis areshown in Fig 3 The AFM analysis con rmed the rough irregular surface structureWe found signs of surface defects for PMMA and PMMA-HA (Fig 4)
Morphology and cell number
SEM analysis showed signi cantly much fewer cells adhered to PMMA-HA thanto PMMA slices at two weeks The analysis of ten specimens showed 268 lessadherent cells per area for PMMA-HA in comparison to PMMA (mean valuesPMMA 906 cells mm2 (SD 192) versus PMMA-HA 663 cells mm2 (SD 232))With increasing culture time the cells tend to build up clusters (cell aggregates)Cellular clusters were found for both cements from day 7 up to day 72 (end ofstudy) The SEM photo in Fig 5 shows at adherent marrow cells on PMMA-HAsurface Furthermore round cells can be seen (dead or detached) after 10 weeks ofincubation (Fig 5)
Cell differentiation and cell vitality
After detaching the cultivated marrow cells from cement probes the number of deadcells was lt10 in all experiments as detected by propidium bromide staining Atday 14 cell culture demonstrated a high number of T-lymphocytes and granulocytes
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1289
Fig
ure
3S
EM
-ED
Xan
alys
isba
sed
onth
ear
eaof
Fig
2
No
addi
tion
alel
emen
tsw
ere
dete
cted
and
the
empi
rica
lfor
mul
aC
a 5(P
O4)
3OH
2is
con
rmed
byqu
anti
tativ
ean
alys
is
O
oxyg
en
Cc
arbo
nC
aca
lciu
mP
pho
spho
rum
Z
irco
nium
(Zr)
serv
edas
X-r
ayco
ntra
stm
ediu
man
dis
notv
isib
lehe
rebe
caus
eof
the
over
lapp
ing
byP
-pea
k(L
-reg-p
eak)
A
ccel
erat
ing
ener
gy
10ke
Vl
ife
tim
e20
0s
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1290 M Jaumlger and A Wilke
Fig
ure
4A
FM
surf
ace
anal
ysis
ofa
PM
MA
spec
imen
inta
ppin
gm
ode
Ove
rvie
wof
aP
MM
Ace
men
tsur
face
area
B
esid
esth
eir
regu
lar
surf
ace
stru
ctur
ea
scra
tch
asa
sign
ofm
ater
iald
efec
tcan
bede
tect
ed(b
lack
arro
w)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1291
Figure 5 Adherent at human bone marrow cells after 10 weeks of incubation on PMMA-HABesides at adherent cells detached round cells are shown The surrounding extracellularmatrix wasrich of sulfur (SEM-EDX)
in case of PMMA in FACS analysis whereas the differentiation rate of monocytesshowed only few differences between both cements (Fig 6) Human bone marrowcells on PMMA-HA differentiated more B-cells than those which were cultivatedon PMMA but less than the glass control group
Immunohistochemical osteocalcin staining reported the superiority of the PMMA-HA cement with regard to the induction of an osteoblastic differentiation Thenumber of osteocalcin positive cells was 39-fold higher for PMMA-HA than formarrow cells which were seeded on PMMA P lt 0001 For PMMA we found inaverage 39 pound 103 cells mm2 (SD 178) compared to 151 pound 103 osteocalcin-positivecells mm2 (SD 191) in the case of PMMA-HA Parallel to the high amount ofosteocalcin positive cells on the PMMA-HA surface we detected a higher rate ofintercellular bridging
Extracellular matrix (ECM)
SEM-EDX analysis showed a sulfur-rich ECM which was located only pericyto-plasmatic In the case of PMMA sulfur was detected after 10 weeks incubationwhereas for the PMMA-HA cement sulfur already appeared after two weeks Fig-ure 7 shows the pericellular EDX analysis of human marrow cells after differentincubation times with PMMA or PMMA-HA cement The semi-quantatively ana-lyzed areas showed an increasing content of sulfur as a function of time for PMMA-
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1292 M Jaumlger and A Wilke
Figure 6 FACS analysis of bone cement adherent marrow cells 14 days after incubation with PMMAand PMMA-HA in comparison with control group Fresh marrow cells 12 h after isolation from theproximal femur (native) and after 14 incubation days onto glass slides The table below demonstratesthe values the bar charts in the diagram the standard deviation PMMA-HA shows a decreasing rateof T-lymhocytes in comparison to PMMA but the number of granulocytes (416) are signi cantlyhigher for PMMA (203) P lt 0005 There is nearly no difference of the number of mononuclearcells between both cements (PMMA 308 PMMA-HA 319) P gt 0005
HA We found an amorphous protein precipitate covering the cement surface in ex-periments with and without cells Sulfur was only detected by SEM-EDX in celladherent specimens
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis showed that HA-particles in PMMA-HA were stable in an
aquaeous medium during the 23-day study period The average concentration valuesfor PMMA cell culture medium were Ca2C 232 mmol l PO3iexcl
4 158 mmol l(SD 0041) and for PMMA-HA Ca2C 231 mmol l PO3iexcl
4 159 mmol l (SD 0026)Dot blot analysis of the collected culture medium supernatants was carried out
semiquantitative for collagen I III and alkaline phosphatase (ALP) over a studyperiod of 72 days The AP activity and collagen-III detection showed no differencesregarding the rythm of appearance for both cements In the case of PMMA-HAcollagen I was synthesized at day 6 and showed a signi cant staining reaction from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1289
Fig
ure
3S
EM
-ED
Xan
alys
isba
sed
onth
ear
eaof
Fig
2
No
addi
tion
alel
emen
tsw
ere
dete
cted
and
the
empi
rica
lfor
mul
aC
a 5(P
O4)
3OH
2is
con
rmed
byqu
anti
tativ
ean
alys
is
O
oxyg
en
Cc
arbo
nC
aca
lciu
mP
pho
spho
rum
Z
irco
nium
(Zr)
serv
edas
X-r
ayco
ntra
stm
ediu
man
dis
notv
isib
lehe
rebe
caus
eof
the
over
lapp
ing
byP
-pea
k(L
-reg-p
eak)
A
ccel
erat
ing
ener
gy
10ke
Vl
ife
tim
e20
0s
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1290 M Jaumlger and A Wilke
Fig
ure
4A
FM
surf
ace
anal
ysis
ofa
PM
MA
spec
imen
inta
ppin
gm
ode
Ove
rvie
wof
aP
MM
Ace
men
tsur
face
area
B
esid
esth
eir
regu
lar
surf
ace
stru
ctur
ea
scra
tch
asa
sign
ofm
ater
iald
efec
tcan
bede
tect
ed(b
lack
arro
w)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1291
Figure 5 Adherent at human bone marrow cells after 10 weeks of incubation on PMMA-HABesides at adherent cells detached round cells are shown The surrounding extracellularmatrix wasrich of sulfur (SEM-EDX)
in case of PMMA in FACS analysis whereas the differentiation rate of monocytesshowed only few differences between both cements (Fig 6) Human bone marrowcells on PMMA-HA differentiated more B-cells than those which were cultivatedon PMMA but less than the glass control group
Immunohistochemical osteocalcin staining reported the superiority of the PMMA-HA cement with regard to the induction of an osteoblastic differentiation Thenumber of osteocalcin positive cells was 39-fold higher for PMMA-HA than formarrow cells which were seeded on PMMA P lt 0001 For PMMA we found inaverage 39 pound 103 cells mm2 (SD 178) compared to 151 pound 103 osteocalcin-positivecells mm2 (SD 191) in the case of PMMA-HA Parallel to the high amount ofosteocalcin positive cells on the PMMA-HA surface we detected a higher rate ofintercellular bridging
Extracellular matrix (ECM)
SEM-EDX analysis showed a sulfur-rich ECM which was located only pericyto-plasmatic In the case of PMMA sulfur was detected after 10 weeks incubationwhereas for the PMMA-HA cement sulfur already appeared after two weeks Fig-ure 7 shows the pericellular EDX analysis of human marrow cells after differentincubation times with PMMA or PMMA-HA cement The semi-quantatively ana-lyzed areas showed an increasing content of sulfur as a function of time for PMMA-
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1292 M Jaumlger and A Wilke
Figure 6 FACS analysis of bone cement adherent marrow cells 14 days after incubation with PMMAand PMMA-HA in comparison with control group Fresh marrow cells 12 h after isolation from theproximal femur (native) and after 14 incubation days onto glass slides The table below demonstratesthe values the bar charts in the diagram the standard deviation PMMA-HA shows a decreasing rateof T-lymhocytes in comparison to PMMA but the number of granulocytes (416) are signi cantlyhigher for PMMA (203) P lt 0005 There is nearly no difference of the number of mononuclearcells between both cements (PMMA 308 PMMA-HA 319) P gt 0005
HA We found an amorphous protein precipitate covering the cement surface in ex-periments with and without cells Sulfur was only detected by SEM-EDX in celladherent specimens
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis showed that HA-particles in PMMA-HA were stable in an
aquaeous medium during the 23-day study period The average concentration valuesfor PMMA cell culture medium were Ca2C 232 mmol l PO3iexcl
4 158 mmol l(SD 0041) and for PMMA-HA Ca2C 231 mmol l PO3iexcl
4 159 mmol l (SD 0026)Dot blot analysis of the collected culture medium supernatants was carried out
semiquantitative for collagen I III and alkaline phosphatase (ALP) over a studyperiod of 72 days The AP activity and collagen-III detection showed no differencesregarding the rythm of appearance for both cements In the case of PMMA-HAcollagen I was synthesized at day 6 and showed a signi cant staining reaction from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1290 M Jaumlger and A Wilke
Fig
ure
4A
FM
surf
ace
anal
ysis
ofa
PM
MA
spec
imen
inta
ppin
gm
ode
Ove
rvie
wof
aP
MM
Ace
men
tsur
face
area
B
esid
esth
eir
regu
lar
surf
ace
stru
ctur
ea
scra
tch
asa
sign
ofm
ater
iald
efec
tcan
bede
tect
ed(b
lack
arro
w)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1291
Figure 5 Adherent at human bone marrow cells after 10 weeks of incubation on PMMA-HABesides at adherent cells detached round cells are shown The surrounding extracellularmatrix wasrich of sulfur (SEM-EDX)
in case of PMMA in FACS analysis whereas the differentiation rate of monocytesshowed only few differences between both cements (Fig 6) Human bone marrowcells on PMMA-HA differentiated more B-cells than those which were cultivatedon PMMA but less than the glass control group
Immunohistochemical osteocalcin staining reported the superiority of the PMMA-HA cement with regard to the induction of an osteoblastic differentiation Thenumber of osteocalcin positive cells was 39-fold higher for PMMA-HA than formarrow cells which were seeded on PMMA P lt 0001 For PMMA we found inaverage 39 pound 103 cells mm2 (SD 178) compared to 151 pound 103 osteocalcin-positivecells mm2 (SD 191) in the case of PMMA-HA Parallel to the high amount ofosteocalcin positive cells on the PMMA-HA surface we detected a higher rate ofintercellular bridging
Extracellular matrix (ECM)
SEM-EDX analysis showed a sulfur-rich ECM which was located only pericyto-plasmatic In the case of PMMA sulfur was detected after 10 weeks incubationwhereas for the PMMA-HA cement sulfur already appeared after two weeks Fig-ure 7 shows the pericellular EDX analysis of human marrow cells after differentincubation times with PMMA or PMMA-HA cement The semi-quantatively ana-lyzed areas showed an increasing content of sulfur as a function of time for PMMA-
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1292 M Jaumlger and A Wilke
Figure 6 FACS analysis of bone cement adherent marrow cells 14 days after incubation with PMMAand PMMA-HA in comparison with control group Fresh marrow cells 12 h after isolation from theproximal femur (native) and after 14 incubation days onto glass slides The table below demonstratesthe values the bar charts in the diagram the standard deviation PMMA-HA shows a decreasing rateof T-lymhocytes in comparison to PMMA but the number of granulocytes (416) are signi cantlyhigher for PMMA (203) P lt 0005 There is nearly no difference of the number of mononuclearcells between both cements (PMMA 308 PMMA-HA 319) P gt 0005
HA We found an amorphous protein precipitate covering the cement surface in ex-periments with and without cells Sulfur was only detected by SEM-EDX in celladherent specimens
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis showed that HA-particles in PMMA-HA were stable in an
aquaeous medium during the 23-day study period The average concentration valuesfor PMMA cell culture medium were Ca2C 232 mmol l PO3iexcl
4 158 mmol l(SD 0041) and for PMMA-HA Ca2C 231 mmol l PO3iexcl
4 159 mmol l (SD 0026)Dot blot analysis of the collected culture medium supernatants was carried out
semiquantitative for collagen I III and alkaline phosphatase (ALP) over a studyperiod of 72 days The AP activity and collagen-III detection showed no differencesregarding the rythm of appearance for both cements In the case of PMMA-HAcollagen I was synthesized at day 6 and showed a signi cant staining reaction from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1291
Figure 5 Adherent at human bone marrow cells after 10 weeks of incubation on PMMA-HABesides at adherent cells detached round cells are shown The surrounding extracellularmatrix wasrich of sulfur (SEM-EDX)
in case of PMMA in FACS analysis whereas the differentiation rate of monocytesshowed only few differences between both cements (Fig 6) Human bone marrowcells on PMMA-HA differentiated more B-cells than those which were cultivatedon PMMA but less than the glass control group
Immunohistochemical osteocalcin staining reported the superiority of the PMMA-HA cement with regard to the induction of an osteoblastic differentiation Thenumber of osteocalcin positive cells was 39-fold higher for PMMA-HA than formarrow cells which were seeded on PMMA P lt 0001 For PMMA we found inaverage 39 pound 103 cells mm2 (SD 178) compared to 151 pound 103 osteocalcin-positivecells mm2 (SD 191) in the case of PMMA-HA Parallel to the high amount ofosteocalcin positive cells on the PMMA-HA surface we detected a higher rate ofintercellular bridging
Extracellular matrix (ECM)
SEM-EDX analysis showed a sulfur-rich ECM which was located only pericyto-plasmatic In the case of PMMA sulfur was detected after 10 weeks incubationwhereas for the PMMA-HA cement sulfur already appeared after two weeks Fig-ure 7 shows the pericellular EDX analysis of human marrow cells after differentincubation times with PMMA or PMMA-HA cement The semi-quantatively ana-lyzed areas showed an increasing content of sulfur as a function of time for PMMA-
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1292 M Jaumlger and A Wilke
Figure 6 FACS analysis of bone cement adherent marrow cells 14 days after incubation with PMMAand PMMA-HA in comparison with control group Fresh marrow cells 12 h after isolation from theproximal femur (native) and after 14 incubation days onto glass slides The table below demonstratesthe values the bar charts in the diagram the standard deviation PMMA-HA shows a decreasing rateof T-lymhocytes in comparison to PMMA but the number of granulocytes (416) are signi cantlyhigher for PMMA (203) P lt 0005 There is nearly no difference of the number of mononuclearcells between both cements (PMMA 308 PMMA-HA 319) P gt 0005
HA We found an amorphous protein precipitate covering the cement surface in ex-periments with and without cells Sulfur was only detected by SEM-EDX in celladherent specimens
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis showed that HA-particles in PMMA-HA were stable in an
aquaeous medium during the 23-day study period The average concentration valuesfor PMMA cell culture medium were Ca2C 232 mmol l PO3iexcl
4 158 mmol l(SD 0041) and for PMMA-HA Ca2C 231 mmol l PO3iexcl
4 159 mmol l (SD 0026)Dot blot analysis of the collected culture medium supernatants was carried out
semiquantitative for collagen I III and alkaline phosphatase (ALP) over a studyperiod of 72 days The AP activity and collagen-III detection showed no differencesregarding the rythm of appearance for both cements In the case of PMMA-HAcollagen I was synthesized at day 6 and showed a signi cant staining reaction from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1292 M Jaumlger and A Wilke
Figure 6 FACS analysis of bone cement adherent marrow cells 14 days after incubation with PMMAand PMMA-HA in comparison with control group Fresh marrow cells 12 h after isolation from theproximal femur (native) and after 14 incubation days onto glass slides The table below demonstratesthe values the bar charts in the diagram the standard deviation PMMA-HA shows a decreasing rateof T-lymhocytes in comparison to PMMA but the number of granulocytes (416) are signi cantlyhigher for PMMA (203) P lt 0005 There is nearly no difference of the number of mononuclearcells between both cements (PMMA 308 PMMA-HA 319) P gt 0005
HA We found an amorphous protein precipitate covering the cement surface in ex-periments with and without cells Sulfur was only detected by SEM-EDX in celladherent specimens
Ca2C PO3iexcl4 analysis
Ca2C PO3iexcl4 analysis showed that HA-particles in PMMA-HA were stable in an
aquaeous medium during the 23-day study period The average concentration valuesfor PMMA cell culture medium were Ca2C 232 mmol l PO3iexcl
4 158 mmol l(SD 0041) and for PMMA-HA Ca2C 231 mmol l PO3iexcl
4 159 mmol l (SD 0026)Dot blot analysis of the collected culture medium supernatants was carried out
semiquantitative for collagen I III and alkaline phosphatase (ALP) over a studyperiod of 72 days The AP activity and collagen-III detection showed no differencesregarding the rythm of appearance for both cements In the case of PMMA-HAcollagen I was synthesized at day 6 and showed a signi cant staining reaction from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1293
Fig
ure
7S
EM
-ED
Xan
alys
isof
peri
cell
ular
area
sof
hum
anbo
nem
arro
wce
lls
afte
r2
4an
d10
wee
ksin
cuba
tion
onto
PM
MA
PM
MA
-HA
bone
cem
ent
The
grap
hssh
owth
eap
pear
ance
ofsu
lfur
(S)
afte
rtw
ow
eeks
for
PM
MA
-HA
and
afte
rte
nw
eeks
for
PM
MA
cem
enti
nvi
tro
(arr
ows)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1294 M Jaumlger and A Wilke
Figure 8 Dot blot analysis of IMDM culture medium which was collected during incubationwith human bone marrow cells cultivated at PMMA-HA Study period was 72 days A deep darkreaction shows the presence of collagen type I The x-axis symbolizes the measurement points (dayof incubation) the y-axis demonstrates 1 10 dilution series of culture medium The negative controlwithout cells shows no reaction At day six a weak reaction can be seen (arrow) From day 16 to theend of the study a strong reaction for collagen type I is visible
day 16 up to the end of study Figure 8 illustrates an original dot blot againstcollagen I for PMMA-HA The bar charts in Fig 9 demonstrate the differences ofcollagen-I and -III detection between PMMA and PMMA-HA as a function of timeCells which were in contact with PMMA showed a delayed collagen-I productionThe collagen-I content of the culture medium decreases for PMMA at the end ofthe study period In contrast we found no difference between both bone cements fortype-III collagen synthesis levels
DISCUSSION
As shown by our experiments the presence of porous HA particles in a PMMAcement can in uence the cellular behavior of human bone marrow cells especiallymaterial adhesion proliferation differentiation and extracellular matrix (ECM)production Both bone cements showed a good biocompatibility measured by thenumber of adherent cells in comparison to other tested biomaterials in previousworks [16ndash18] It was demonstrated by Lampin et al [19] that the surface energyof apolar PMMA components signi cantly increases with roughness and that celladhesion enhancement was related to the degree of roughness and hydrophobicity
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1295
(a)
(b)
Figure 9 Collagen type-I (a) and -III (b) content of cell-culture medium during marrow cellincubationover the study period of 72 days Data based on a semiquantativedot blot antibodyanalysisA weak staining reaction was allocated to the value 05 whereas a strong reaction was symbolized bythe value 10 Cells on PMMA produced collagen type-I much later than in the case of PMMA-HA
In this study the total number of adherent marrow cells showed 268 less cellsadhered onto PMMA-HA although the roughness was increased by HA-granules(906 cells mm2 for PMMA versus 663 cells mm2 for PMMA-HA) The 14-dayculture showed much more osteocalcin-positive cells for PMMA-HA comparedto PMMA cement a result also obtained by De Santis et al [20] who analyzedthe effects of different HA roughnesses to osteocalcin-positive cells They found
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1296 M Jaumlger and A Wilke
out that rough (plasma sprayed) surfaces of titanium and HA induced a lowerproliferation rate a lower ALP activity but a higher osteocalcin synthesis on adultjaw bone cells compared to polished surfaces In this investigation we coulddemonstrate that HA particles of lt10 sup1m diameter promoted the differentiationof osteoblastic progenitors Dalby et al [21] investigated the biological response ofprimary osteoblast-like cells (HOBs) on PMMA discs with different HA contentsand showed a synergy between increasing focal contact formation cytoskeletalorganisation cell proliferation and expression of phenotype with increasing HAvolume
Furthermore the results of our study correspond to clinical data Oonishi etal [22] showed that there is a signi cant reduction of the incidence of radiolucentlines postoperatively after using PMMA cement with HA-granula
We detected a cell-independent amorphous matrix at the surface of study andcontrol groups This con rms the results of other investigators who described aprotein precipitate on hydrophobic materials [4 23] Protein layers on hydrophobicsurfaces like PMMA-polymers support cellular xation spreading and differentia-tion [9 24] A new nding was the presence of sulfur as an element of ECM Theculture time at which sulfur was detected as a new chemical pericellular elementwas material-dependent and never found in control groups PMMA-HA seems topromote the cellular production of sulfur-rich ECM Chen et al [25] postulated thatsulfated proteoglycans like chondroitin have a high binding af nity to HA Hunteret al [26] supplemented this hypothesis by in vitro experiments They showed thatpericellular localized sulfated proteoglycans bind Ca2C and support the calci ca-tion by a local enrichment of a supersaturated Ca2C solution This corresponds tothe work of Wight and Jackson et al [27 28] who emphasized that most adhesionpeptides interact with cell surface proteoglycans by a heparinsulfate rich binding do-main Especially the release of disul d isomerases by cells in an arti cial environ-ment for covalent S S-binding of ECM proteins is described in the literature [29]The HA granules in PMMA (diameter lt10 sup1m) showed stability in culture mediumover 23 days
With regard to the morphological cellular formation we found cluster formations(cell aggregates) increasing with cultivation period up to 72 days It is notclear if these cell aggregates derived from progenitor cells (colony forming units)or were induced by an integrin mediated cluster genesis as described by otherinvestigators [4 30]
The early detection of collagen I and III in the case of the PMMA-HA culturecorresponds to a physiological wound regeneration as described by McCarthy [31]On the other hand thick collagen-I bres which were localized in the connectivetissue after the rst weeks of wound healing could lead to micromovements ofthe implant [32] Other authors emphasized the necessity of collagen type I forthe mineralization process [9 24] In this study we could not detect direct signsof mineralization at bone cement surface For PMMA-HA it was not possible todiffer the preexisiting components phosphor and calcium in HA granules from
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
Biocompatibility testing of PMMA-HA cement vs conventional PMMA cement 1297
P and Ca as cellular products (sign of beginning calci cation of the extracellularmatrix by SEM-EDX analysis) Moreover after two weeks of incubation the cellcomposition of the analyzed lymphocytes monocytes granulocytes and stem cellshad similarity compared to cellular composition of physiological fracture healing invivo which was investigated by Andrew et al [33]
In summary with exception of the number of adherent cells PMMA-HA cementshowed a superiority regarding biocompatibility for a human bone marrow cellculture To avoid clinical failure biomaterials have been tested in the past priorto implantation mainly by animal experiments The problems of relating theresults of animal experiments to human reactions their high costs and increasingpublic disapproval have lead to develop various cell culture systems to assess thebiocompatibility of implant materials [32] The rst investigators used cell culturesof low differentiated embryonic or tumor cells One disadvantage of these testingsystems was the absence of a direct contact between the tested biomaterial andphysiological local cells In this study a human bone marrow cell culture systemwas used to test the biocompatibility of implant materials under in vitro conditionsThe human bone marrow cell-culture appears to have several advantages in therespect to the use of physiological cells cells of human origin a cell populationthat has primary contact with implanted prothesis material the presence immunecompetent B- and T-lymphocytes and inclusion of undifferentiated mesenchymalstem cells (osteoprogenitor cells) which are known to be capable to differentiateinto osteoblasts broblasts chondrocytes myoblasts and osteoclasts [34]
Animal experiments have to follow to con rm the in vitro screening Moreoverthe in uence of the HA-surface porosity on cell differentiation should be an aim offurther investigations
REFERENCES
1 R Pandey J Quinn C Joyner D W Murray J T Trif tt and N A Athanasou Ann RheumDis 55 388 (1996)
2 A Sabokbar R Pandey J M Quinn and N A Athanasou Arch Orthoped Trauma Surg 117136 (1998)
3 J H Hermann W G Sowder D Anderson A M Apfel and C N Hapson J Bone Joint Surg71-A 1530 (1989)
4 H G Willert and M Semlitsch J Biomed Mater Res 11 157 (1977)5 L C Jones and D S Hungerford Clin Orthoped 225 192 (1987)6 L Linder L Lindenberg and A Carlsson Clin Orthoped 175 93 (1982)7 S Santavirta Y T Konttinen M Gronblad M Kauppi M Sakaguchi M Hamalainen
J E Michelsson J M Polak and D Dahl Acta Orthoped Scand 62 29 (1991)8 J Gil-Albarova J Bone Joint Surg 74-B 825 (1992)9 K Anselme Biomaterials 21 667 (2000)
10 A El-Ghannam P Ducheyne and L M Shapiro J Orthoped Res 17 340 (1999)11 M Niki G Ito T Matsuda and M Ogino in The bone-biomaterial interface J M Davies
(Ed) p 350 University of Toronto Press Toronto (1990)12 R E Homes in Am Acad Orthoped Surg Symp p 355 (1984)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13
1298 M Jaumlger and A Wilke
13 T M Dexter E Spooncer P Simmons and T D Allen in Long Term Bone Marrow CultureAn Overview of Technique and Experience D G Wright and J S Greenberger (Eds) p 57Alan R Liss New York NY (1984)
14 P Kalla W Oster F Seiler E Ruumlde and D Krummwieh J Cell Clon 10 54 (1992)15 P Kalla P Hintz-ObertreisA Wilke P Griss F R Seiler E Rude and D Krumwieh Behring
Inst Mitt 90 93 (1991)16 A Wilke J Orth M Lomb R Fuhrmann H Kienapfel P Griss and R P Franke J Biomed
Mater Res 40 301 (1998)17 A Wilke M Jaumlger S Endres M Landgraff A Kiessling M Pfeiffer and P Griss Eur J
Trauma 5 279 (2002)18 F Stemberg and A Wilke J Biomater Sci Polymer Edn 12 171 (2001)19 M Lampin R Warocquier-Cleacuterout C Legris M Degrange and M F Sigot-Luizard J Biomed
Mater Res 36 99 (1997)20 D De Santis C Guerriero P F Nocini A Ungersbock G Richards P Gotte and U Armato
J Mater Sci Mater Med 7 21 (1996)21 M J Dalby L Di Silvio E J Harper and W Bon eld Biomaterials 23 569 (2002)22 H Oonishi Y Kadoya H Iwaki and N Kin J Biomed Mater Res 53 174 (2000)23 H Zreiqat P Evans and C R Howlett J Biomed Mater Res 44 389 (1999)24 K Anselme M Bigerelle B Noel E Dufresne D Judas A Iost and P Hardouin J Biomed
Mater Res 49 155 (2000)25 C C Chen and A L Bosekey Calcif Tissue Int 37 395 (1985)26 G K Hunter and S K Szigerty Matrix 12 362 (1992)27 R L Jackson S J Busch and A D Carcin Physiol Rev 71 481 (1991)28 T N Wight M G Kinsella and N N Qwarnstroumlm Curr Opin Cell Biol 4 793 (1992)29 H Birkdedall-HansenCurr Opin Cell Biol 7 728 (1995)30 D M Ferris G D Moodie P M Dimond C W D Gioranni M G Ehrlich and R F Valentini
Biomaterials 20 2323 (1999)31 J G Mc Carthy (Ed) Plastic Surgery Volume I (General Principles) WB Saunders Philadel-
phia PA (1990)32 G E Romanos S T K Holz R Beyer and J R Strub in Biological Matrices and Tissue
Reconstruction G B Stark R Horch and E Tanczos (Eds) p 13 Springer Berlin (1989)33 J G Andrew S M Anrew A J Freemont and D R Marsh Acta Orthoped Scand 65 462
(1994)34 M F Pittenger A M Mackay S C Beck R K Jaiswal R Douglas J D Mosca
M A Moorman D W Simoneti S Craig and D R Marshak Science 284 143 (1999)
Dow
nloa
ded
by [
Cal
ifor
nia
Stat
e U
nive
rsity
of
Fres
no]
at 0
108
28
Apr
il 20
13