cytotoxicity testing of wound dressings using normal human keratinocytes in culture
TRANSCRIPT
Cytotoxicity testing of wound dressings using normal human keratinocytes in culture
Martin Rosdy and Louis-Christian Clauss Biolaboratory, QA department, Dow Corning European Health Care France S.A. Route des cretes-B. P. 43 Sophia Antipolis-Les Bouillides 06561 Valbonne Cedex. France
Comparative cytotoxicity testing of 16 wound dressings of different composition show that normal human keratinocytes (NHK) growing on a fibroblastic feeder layer are as sensitive to toxic materials by direct contact as the confluent MRC5 fibroblasts used for standard cell culture cytotoxicity testing, and slightly more sen- sitive when extracts of the dressings were tested. After direct contact with each of the cell types, we found effects due to 12 dress- ing samples (75%), but the extracts of only 6 of them induced changes in cell shape or cell death on NHK, and 4 of them on MRC5 cells. In order to assess the compatibility of these dressings with a pure population
of epidermal cells, the cell type responsible for reepidermization of healing wounds, we then tested the sensitivity, both to dress- ing samples and extracts, of normal human keratinocytes (NHK) grown in chemically defined medium and without a feeder layer: The results show epidermal cytocompati- bility of 10 dressing extracts, while 6 others induced cytopathic effects. Three of these extracts specifically damaged epidermal cells and inhibited their proliferation. When comparing the sensitivities of NHK (in defined medium) and MRC5 cells, we observed complete correlation for 75% of the dressings by extract testing and in 94% of the cases after direct contact.
INTRODUCTION
During the past years, the biocompatibility testing of medical materials and devices has evolved toward the development and use of specific cyto- compatibility test systems. Cell or tissue culture techniques are now avail- a b P 6 for rapid and meaningful evaluation and control of the compatibility of medical devices with the cells or tissues with which they will be in con- tact. These developments are important because of the high costs of classical animal testing which, in addition, often gives results not easily transposable to the clinical situation in humans.
We report here a study of wound dressings supposed to permit reepider- mization of superficial wounds and which therefore should not interfere with the proliferation and migration of epidermal In order to assess their compatibility with epidermal cells, we exposed samples of these dress- ings and also their respective extraction medium to normal human kerati- nocytes in culture.
First we have studied 16 wound dressings of different compositions with regard to their cytotoxicity by direct contact to MRC5 embryonic fibroblastic
Journal of Biomedical Materials Research, Vol. 24, 363-377 (1990) 0 1990 John Wiley & Sons, Inc. CCC 0021-9304/90/030363-15$04.00
364 ROSDY AND CLAUSS
cells. Extraction media of the same dressings were also tested on MRC5 cells, using the B.S.I. standard procedure for cytotoxicity testing of medical devices.'
Second, equivalent dressing samples were extracted in epidermal cell cul- ture medium; both these extracts and 1-cm2 samples were placed in direct contact with normal human keratinocytes (NHK: the cell-type allowing reepidermization of the healing wound in vivo) cultured on a feeder layer of mouse 3T3 fibroblasts.10-" Sensitivity of these cells to toxic dressing material was comparable to the one of MRC5 cells.
But the similar sensitivities to toxic dressing material of confluent MRCS fibroblasts and colonies of keratinocytes grown on a feeder layer might have been due to an inhibition of the keratinocyte-growth-supporting prop- erties of the feeder-layer fibroblasts. To check if the normal human kerati- nocytes themselves are as sensitive to the tested material as the MRC5 fibroblasts, a pure keratinocyte culture system was needed. Therefore, a chemically defined media culture system developed by Ham and coworker^,^^-'^ was optimized for the growth of normal human adult kerati- nocytes directly on plastic culture dishes, without a feeder layer. Addition- ally, this allowed us to eliminate possible cross reactions with unknown factors of the fetal calf serum and, moreover, to simplify the culture method employed for this epidermal cytotoxicity test.
MATERIALS AND METHODS
Cell cultures
MRCS human embryonic lung fibroblasts (passage 28) were purchased at Biom&ieux, Lyon, France. Culture dishes 35 mm in diameter (Falcon) were inoculated with lo5 cells/cm2 and grown to confluency in BME medium con- taining 10% fetal calf serum and antibiotics (Boehringer, France).
3T3 mouse fibroblasts were treated 3 hours at confluency with 0.33 mg/mL Mitomycine C (Sigma Chemicals, France) and then inoculated in 35-mm- diameter culture dishes (Falcon, France) at a density of 2 X lo4 cells/cm2 to form a feeder layer for the epidermal cells. The normal human kerati- nocytes (NHK) were isolated from dermatologic surgery skin samples using the 0.25% trypsin dermis-epidermis separation method' and subcultured at a density of lo4 tells/cm2 on the 3T3 feeder layer in DME/F12 (1 : 1) medium containing 10% fetal calf serum, antibiotics, hydrocortisone (5 x M), choleratoxin (lo-'* M), insuIin (5 ,ug/mL) (Sigma, France) and EGF (10 ng/mL) (Collaborative Research, MA, USA).
Normal human keratinocytes were subcultured at lo4 cells/cm2 in a chemically defined medium (MCDB 1.53) which was entirely prepared in the laboratory, according to the procedure of Peehl and Ham? it was supple- mented with an amino-acid stock s o l u t i ~ n , ~ ~ - ' ~ insulin (5 pg/mL), hydrocor- tisone (5 X M), epidermal growth factor (10 ng/mL), ethanolamine
CYTOTOXICITY OF WOUND DRESSINGS 365
M), phosphoethanolamine M), antibiotics and, for stock cultures only, bovine pituitary extract (70 pg/mL). The calcium concentration was adjusted to lop4 M (= low calcium concentration).
All stock cultures were checked for absence of microplasma using the Hoechst stain method.
All chemicals and salts for MCDB 153 were purchased at Sigma Chemicals (France). Trace elements were from Fluka (Buchs, CH) and Merck (Darm- stadt, Germany).
Test procedures
Direcf contact
Sixteen wound dressings of different composition (see Table I) were cut aseptically in l-cm’ pieces, and deposited (one per dish) directly on the cells in the dishes containing 2 mL of fresh culture medium. When floating, they were charged with similar pieces of negative control material (Silicone M D X 4 4515) to be in direct contact with the cells. Pieces of natural latex from catheters were used as positive control material.
Extraction method
The quantity of any substances extracted from the dressings is related to the surface area with which the extractant is brought into contact both dur- ing exposure to a wound in vivo and during experimental extraction. There- fore, although the dressing materials have different weight/mass ratio, we extracted 6-cm2 samples of each dressing tested in 12 mL sterile culture me- dium at 37°C for 24 h. This corresponds to an extraction ratio of 0.5 cm2/mL extractant.g The extraction media were then checked for neutral pH and if necessary, adjusted using 10 mM HC1 or 10 mM NaOH solutions. Two milliliters of each extract was deposited on the cells in each of two dishes.
MRC5 cells were used for testing when confluent, and NHK were used on feeder layer or in defined medium when colonies had not fewer than 100 cells. Tests were performed in duplicate for each experiment. After expo- sure to the extraction liquids or directly to the dressing material, the shape of the cells in each dish was evaluated under the inverted microscope using the following scale by comparison with the control dishes:
0 = No change noted. 1 = Slight but clear change in shape without decrease of cell density. 2 = Considerable alterations in cell shape. Slight decrease of cell density. 3 = Loss of cell shape. More than 50% of cells detached. 4 = All cells are dead, lysed or detached.
TABL
E I
Lis
t of
Tes
ted
Dre
ssin
gs a
nd T
heir
Com
posi
tions
Com
mer
cial
nam
e N
o.
(Man
ufac
ture
r)
Lot N
o.
Com
posi
tion
1
Gra
nufle
x (S
quib
bs S
urgi
care
) 3U
166
Poly
uret
hane
foam
+ hy
droc
ollo
ld
parti
cles
t h
ydro
phob
ic p
olym
er
2 K
alto
stat
(Cai
r)
8614
.4.8
C
alci
um a
lgin
ate
fiber
s 100
%
nonw
oven
3 St
eris
eal S
orbs
an (N
I Med
ical
) 26
.06.
87
Alg
inat
e Fi
bers
4 A
dapt
ic (J
ohns
on +
John
son)
99
305
Vis
cose
100
% +
emul
sion
va
selin
e +
wat
er
5 O
psite
(Sm
ith +
Nep
hew
) 44
0484
Po
lyur
etha
ne f
oam
6 Sy
spur
derm
(H
artm
an)
0306
08 8
0 Po
lyur
etha
ne s
oft f
oam
7 Te
gade
rm (3
M)
8 Ly
ofoa
m 2
726
(Ultr
a La
bs)
9 C
orad
em (R
evlo
n H
ealth
)
10
Epig
ard
(Par
ke +
Dav
is)
11
Com
feel
3210
(Col
opla
st)
12
Tran
sige
n (S
mith
+ N
ephe
w)
0385
52
V04
9109
3
Poly
uret
hane
film
Visc
ose
Fibe
rs +
activ
ated
car
bon
+ po
lyur
etha
ne f
oam
Mod
ified
pol
yure
than
e fo
am
Poly
uret
hane
foa
m
+ po
lypr
opyl
ene
film
Car
boxy
met
hyl c
ellu
lose
+
poly
uret
hane
film
Poly
uret
hane
film
+ in
ert
nonw
oven
mat
eria
l + ad
hesi
ve
w
m
m P
TABLE
I.
(con
tinue
d)
13
Foam
She
etin
g Si
last
ic (D
ow C
orni
ng)
KO
5603
9 Po
lysi
loxa
ne fo
am
14
Gel
She
etin
g Si
last
ic (D
ow C
omin
g)
VL0
2601
3 Si
licon
e ge
l
15
Jelo
net (
Smith
+ N
ephe
w)
2932
3 Pa
raff
in g
auze
16
Om
ider
m (
Om
ikro
n Sc
ient
ific)
HS
KOOS
-164
Poly
uret
hane
film
17
Posi
tive
cont
rol
Nat
ural
late
x (c
athe
ter)
18
Neg
ativ
e co
ntro
l Si
licon
e MD
X4-
4515
(Dow
Cor
ning
)
368 ROSDY AND CLAUSS
Exposure to MRC5 cells was for a period of 24 h, while tests involving kera- tinocytes lasted 48 h to permit measurement of significant effects on cell proliferation.
Cell counting
After 48 h exposure to NHK in defined medium, the dressing samples were removed with forceps, the media or extracts aspirated, cells were rinsed twice in phosphate buffered saline (PBS) and detached by 1 mL trypsin 0.05%-EDTA 0.02%. Each cell suspension was added entirely to 10 mL isotonic solution and cells in 0.5 mL were counted twice by a Coulter Counter (Coultronics, France). The means of cell numbers in duplicate dishes were calculated.
RESULTS
Test on MRC5 fibroblastic cells
Twelve of the 16 dressing samples tested induced morphological changes when placed directly on MRC5 fibroblastic cells, but the corresponding ex- traction media induced cell alterations only in 4 cases (Kaltostat, Adaptic, Transigen, Silastic foam); the other 12 dressing extracts were not cytotoxic on MRC5 fibroblasts (Table 11).
Test on NHK with fibroblastic feeder-layer
Comparable results were obtained when samples of the same dressings were placed in direct contact with normal human keratinocytes growing on a fibroblastic feeder layer in serum containing medium. Six of the corre- sponding extracts induced alterations of the epidermal cells, including the 4 which were cytotoxic in the classical MRC5 test. The two additional dress- ings, extracts of which induced cytotoxicity only on epidermal cells, are both composed of modifjed polyurethane foams: Syspurderm and Cora- derm (Table 11).
To check whether this greater sensitivity was intrinsic to the epidermal cells or due to the longer exposure of the extracts to the cell layer, we ex- tracted samples of these two dressings during 48 h and exposed them 48 h to MRC5 cells: still no cytopathic effects were observable, suggesting spe- cific sensitivity of the epidermal cells to certain substances extractable from these dressings.
Test on NHK in defined medium
To test this hypothesis, samples of the same 16 dressings and their respec- tive extracts were placed in direct contact with colonies of growing normal
0
TABL
E I1
3 2 9
Cyt
otox
icity
Tes
t Res
ults
on
MR
C5
Cel
ls a
nd o
n N
orm
al H
uman
Ker
atin
ocyt
es (N
HK
) Gro
wn
on a
Fib
robl
astic
Fee
der
Laye
r 0
Dir
ect C
onta
ct
Ext
ract
Med
ium
E <
Te
st o
n M
RC
5 Te
st o
n N
HK
on
Test
on
MR
C5
Test
on
NH
K o
n N
o.
Com
mer
cial
Nam
e C
ells
Fe
eder
Lay
er
Cel
ls
Feed
er L
ayer
1 2 3 4 5 6 7 8 9 10
11
12
13
14
Gra
nufl
ex
Kal
tost
at
Ster
isea
l Sor
bsan
Ada
ptic
Ops
ite
Sysp
urde
rm
Teg
ader
m
Lyo
foam
Cor
ader
m
Epi
gard
Com
feel
Tra
nsig
en
Sila
stic
foam
Sila
stic
gel
2 4 1 4 3 1
0 0 4 0 2 0 2 0
2 3
3 2
2 1
4 4
2 3
1 2
0 0
1 3
4 4
1 1
2 2
1
0
2 1
0 0
4 0
2 4
1 0
4 2
4 0
2 0
0 0
2 0
3 0
1 0
2 0
1 2
1
1
0 0
0 0
4 3
0 0
2 4
0 0
0 2
0 0
0 0
0 4
0 0
0 0
2 4
1 1
0 0
0 3 0 4 0 2 0 0 4 0 0 4 1
0
TABL
E 11
. (c
ontin
ued)
15
Jelo
net
00
0
0
00
16
Om
ider
m
00
0
0
00
17
Posit
ive
cont
rol
33
3
3
33
18
Neg
ativ
e co
ntro
l 0
0
00
0
0
00
00
33
00
Note
. 0 =
No
chan
ge n
oted
; 1
= s
light
but
cle
ar c
hang
e in
sha
pe w
ithou
t dec
reas
e of
cel
l den
sity
; 2 =
Con
side
rabl
e al
tera
tions
in
cell
shap
e, s
light
dec
reas
e of
cell
dens
ity; 3
= lo
ss o
f ce
ll sh
ape,
mor
e th
an 5
0% o
f ce
lls d
etac
hed;
4 =
All
cells
are
dea
d, l
ysed
or
deta
ched
.
P
CYTOTOXICITY OF WOUND DRESSINGS 371
human adult keratinocytes: As described,*-** the chemically defined culture medium used suppresses growth of fibroblastic cells, and melanocytes are only detected in primary cultures; the test was performed on cultures at the third passage, containing a pure keratinocyte population of cells.
I) The sensitivity to direct dressing contact of NHK grown in chemically defined medium was comparable to that of confluent MRC5 cells or NHK cells growing on a fibroblastic feeder-layer in serum containing medium (see Tables II, III). Effects ranged from changes in cell shape and morphology, cytoplasmic granulations, to decrease in cell density and cell death (Fig. 1).
11) The extracts of six dressing samples induced clear effects on epidermal cells, indicating greater sensitivity of normal keratinocytes, compared to the MRC5 fibroblasts, to 3 of the 16 dressings tested: Lyofoam, Syspurderm, and Coraderm, and lesser sensitivity to Silastic foam dressing. To check if these effects correlated with growth arrest of the epidermal cells, the num- ber of cells remaining attached to the culture dish after rinsing with saline, before and after the 48-h incubation period (see Table 111) were counted: Cy- totoxicity induced by direct dressing contact or extracts (Adaptic, Syspur- derm, Coraderm) correlated with a significant decrease in the number of attached cells, whereas three cytotoxic extracts provoked just growth arrest of the keratinocytes (Kaltostat, Lyofoam, Transigen).
In presence of each of the noncytotoxic extracts or dressing samples the cell number increased between two to three fold during the 48 h incubation (Table 111).
DISCUSSION
In summary, the 16 dressings tested can be classified in four groups:
Group 1
Effects on Effects on MRC5 fibroblasts epidermal cells
direct extract direct extract Dressing
N" name contact medium contact medium
7 Tegaderm no no no no 14 Silastic gel no no no no 15 Jelonet no no no no 16 Omiderm no no no no
These dressings gave negative results in all tests performed; they can clearly be stated as non cytotoxic according to the test methods proposed by the BSI (9), ASTM, and AFNOR; moreover they are compatible with prolif- erating human normal epidermal cells in culture.
372 ROSDY AND CLAUSS
Figure 1. Shape of normal human keratinocytes cultured in defined me- dium after 48 h exposure to dressing extracts (original magnification x 100). A) Notation 0 (Silastic gel sheeting Q7-9119), B) Notation 2 (Coraderm), C) Notation 4 (Syspurderm).
CYTOTOXICITY OF WOUND DRESSINGS 373
Figure 1. (continued)
Group 2
Effects on Effects on MRC5 fibroblasts epidermal cells
direct extract direct extract N" name contact medium contact medium
Dressing
2 Kaltostat Yes Yes Yes Yes 4 Adaptic Yes Yes Yes Yes
12 Transigen Yes Yes Yes Yes 13 Silasticfoam Yes Yes Yes no
These dressings induced positive results on both cell types. They have to be considered as cytotoxic devices according to the standard test methods pro- posed by the BSI, ASTM, and AFNOR. Their cytotoxicity cannot be corre- lated with one or several specific biomaterials, because these four dressings are composed of very different materials (see Table I).
Group 3
Cytopathic effects Cytopathic effects on epidermal cells on MRC5 fibroblasts
Dressing direct extract direct extract
N" name contact medium contact medium
1 Granuflex Yes 3 Sorbsan Yes
10 Epigard Yes 11 Comfeel yes
5 Opsite yes
no no no no no
no no no no no
374 ROSDY AND CLAUSS
TABLE I11 Cytotoxicity Test Results on NHK Grown in Defined Medium
Direct contact Extract Medium
Cytotoxicity Cell NB/Dish Cytotoxicity Cell NB/Dish No. Commercial Name Notation 0-4 % of Control Notation 0-4 % of Control
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Granuflex
Kaltostat
Steriseal Sorbsan
Adaptic
Opsite
Syspurderm
Tegaderm
Lyofoam
Coraderm
Epigard
Comfeel
Transigen
Silastic foam
Silastic gel
Jelonet
Omiderm
Positive control
Negative control
3
3
1
4
3
2
0
1
1
0
1
1
1
0
0
0
3
0
- 3
2 3.5
1 57.0
4 5.3
4 7.6
3 22.4
0 76.0
1 66.7
1 47.6
0 47.6
1 52.4
1 52.4
1 47.6
0 133.0
0 81.0
0 90.0
3 9.5
0 100.0
Percentage of cells present before the 48 h incubation (mean of four dishes)
45.0
0
2
0
3
0
4
0
1
2
0
0
1
0
0
0
0
3
0
- 0
2 46.7
0 83.0
3 9.8
0 70.0
4 9.0
0
1 44.3
2 6.7
0 105.0
0 87.7
1 31.2
0 95.0
0 98.4
0 96.0
0 112.3
3 14.3
0 100.0
45.0
-
Note. 0 = No change noted; 1 = slight but clear change in shape without decrease of cell density; 2 = Considerable alterations in cell shape, slight decrease of cell density; 3 = Loss of cell shape, more than 50% of cells detached; 4 = All cells are dead, lysed or detached.
These five dressings induced cytopathic effects by direct contact, but their extraction media were not cytotoxic, thus passing the BSI standard test. Two dressing samples (Sorbsan/Opsite) significantly increased the pH of the extractant, therefore the relative acidity induced during direct contact
CYTOTOXICITY OF WOUND DRESSINGS 375
exposure could be responsible for the effects observed. By rectdying the pH of the extraction medium, we found that cytotoxicity by direct contact was due exclusively to excessive acidity.
Group 4
Cytopathic effects Cytopathic effects on epidermal cells on MRC5 fibroblasts
direct extract direct extract Dressing
medium contact medium N" name contact
6 Sy spurderm Yes no Yes Yes 8 Lyofoam Yes no Yes Yes 9 Coraderm Yes no Yes Yes
Three dressings induced cytopathic effects on both cell types by direct con- tact, but their respective extraction media gave positive results only on epi- dermal cells. This specific toxicity to epidermal cells could be due to their common constituant (polyurethane foam). In the direct contact method, 1 cm2 of dressing material is deposited on a monolayer of MRC5 cells or on colonies of proliferating epidermal cells in 2 mL of culture medium: the dressinghedium ratio is therefore the same as for the extraction method, thus the only difference remaining is the actual presence of the dressing samples in the culture dish. Therefore no molecu€e present in the extract medium is responsible for the cytotoxicity observed only after direct contact.
It is well known that growth of cells in culture is very sensitive to attach- ment molecules such as collagen and vitronectin present in the culture medium or secreted by the cells t h e m s e l v e ~ . ~ ~ ~ ~ Epidermal growth factor and related factors were recently reported to be active in epidermal cell mi- gration in vifro",26 and during wound healing in vivo. 272e It could be argued that the presence of a dressing sample in the culture dish interferes with the deposition of such factors on the surface of the dish: by releasing com- petitive molecules (which during extraction adhere to the walls of the ex- traction vials and are therefore not present in the extracts when tested next day), the dressing materials might disturb the attachmenthurface character- istics of the culture substrate.
Both phenomena, the possible interference with deposition of attachment molecules or the increased acidity, could disturb the migration and prolif- eration of epidermal cells in vivo during reepidermization of a healing or su- perficial wound; indeed, the same factors which are necessarily present or secreted in the culture dish to allow attachment and growth of epidermal cells, are also directly involved in the migration of epidermal cells in vivo during wound healing.24*29*30 Therefore, more specific parameters such as at- tachment factor deposition, detailed extracellular matrix composition, inter- ference with cell migration and dilution of acidity by wound fluids, for example, should be studied to assess epidermal cytocompatibility of dress- ing materials. We propose to use the described epidermal cell culture method (NHK in defined medium) for such cytocompatibility studies.
376 ROSDY AND CLAUSS
Only the four dressings of the second group can be assumed to be clearly cytotoxic, while the three dressings of the fourth group might just not be intended for reepidermization of healing or superficial wounds. Obviously, these indications have to be confirmed by in vivo reepidermization test data.
After screening 16 dressings regarding their compatibility with proliferat- ing normal human keratinocytes in culture, we suggest possible interfer- ence of certain dressings with the proliferation and attachment/migration mechanisms of epidermal cells during the reepidermization process.
Finally, comparing cell sensitivities of the NHK test method and of the standard MRC5 cytotoxicity test procedure,' we found complete correlation in 75% of the cases when tests are conducted on extracts, and for 94% of the dressings when tested by direct contact.
The authors wish to thank Nigel McWilliam for providing the dressing samples and Christine Ferrere for typing the manuscript.
References 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
C. Hecquet, X. Ronot, L. Le Naour, J. G. Josse, and M. Adolphe, "Use of cornea cell culture as screening models for the cytotoxicity of contact lenses," Eur. 1. Cell Biol., Suppl. 15, 42, 39 (1986). J. E. Davies and S. F. Tarrant, "Interaction between primary bone cell cultures and biomaterials (Part 1 & 2)," Proceedings of the European Con- gress on Biomaferials, Bologna 1986, Elsevier Science Publishers, New YorWAmsterdam, 1987. M.F. Harmand, L. Bordenave, R. Duphil, R. Jeandot, and D. Ducas- sou, "Human differentiated cell cultures: 'in-vitro' models for charac- terization of cell/biomaterial interface," Advances in Biomaterids, vol. VI, Elsevier, New York, 1986, pp. 361-366. J. A. Styles, "Tissue culture methods for evaluating biocompatibility of polymers," Fundamental Aspecfs of Biocompafibility, D. F. Williams (ed.), C.R.C. Press, Boca Raton, 1981, pp. 219-231. M. F. Sigot-Luizard, M. Lanfranchi, J. L. Duval, S. Benslimane, M. Sigot, R. G. Guidion, and M. W. King, "The cytocompatibility of compound polyester-protein surfaces using an in-vitro technique." I n vitro Cell. Dev. Biol., 22(5), 234-240 (1986). R. E. Wilsnack, et al., "Human cell culture toxicity testing of medical devices and correlation to animal tests," Biomat. Med. Dev. Art. Org., 3, 543-562 (1973). J. C. Lawrence, "Wound dressings, '' Pharmacy Update, 147-150 (April 1987). S. Thomas, P. Loveless, and N.P. Hay, "Comparative review of the properties of six semipermeable film dressings," Pharrnaceut. I . , 785- 788 (June 1988). B.S.I. cytotoxicity standard test for medical devices n"5736 part 10 (1988). J. G. Rheinwald and H. Green, "Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells," Cell, 6, 331-344 (1975). H. Green, 0. Kehinde, and J. Thomas, "Growth of cultured human epidermal cells into multiple epithelia suitable for grafting," Proc. Nufl . Acad. Sci. U.S.A. , 76(11), 5665-5668 (1979). J. G. Rheinwald and H. Green, "Epidermal growth factor and the mul- tiplication of cultured human epidermal keratinocytes," Nature, 265, 421-424 (1977).
CYTOTOXICITY OF WOUND DRESSINGS 377
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
D. M. Peehl and R. G. Ham, “Clonal growth of human keratinocytes with small amounts of dialysed serum,” In-Vitro, 16, 526-538 (1980). M. C. Tsao, B. J. Walthall, and R. G. Ham, ”Clonal growth of normal human epidermal keratinocytes in a defined medium,” J. Cell. Physiol., 110, 219-229 (1982). S. T. Boyce and R. G. Ham, ”Calcium-regulated differentiation of nor- mal human epidermal keratinocytes in chemically defined clonal cul- ture and serum-free serial culture,” J. Invest. Dermatol., 81, 33s-40s (1983). S. T. Boyce and R. G. Ham, ”Control of differentiation of human epi- dermal keratinocytes after frozen storage and serum-free subculture,” In-vitro Models for Cancer Research, M. M. Weber and L. Sekely (eds.), C.R.C. Press, Boca Raton, 1983. M. R. Pittelkow, J. J. Wille, and R. E. Scott, “Two functionally distinct classes of growth arrest states in human prokeratinocytes that regulate clonogenic potential,” J. Invest. Dermatol., 86, 410-417 (1986). M. R. Pittelkow and R. E. Scott, ”New techniques for the in-vitro cul- ture of human skin keratinocytes and perspectives on their use for grafting of patients with extensive burns,” Maya Clin. Proc., 61, 771- 777 (1986). F. Furukava, J. C. Huff, W. L. Weston, and D. A. Norris, ”Serum-free serial culture of adult human keratinocytes from suction-blister roof epidermis,” J. Invest. Dermatol., 89, 460-463 (1987). P. B. van Wachem, B. W. L. Mallens, A. Dekker, T. Beugeling, J. Feijen, A. Bantjes, J. P. Detmers, and W. G. van Aken, ”Adsorption of fi- bronectin derived from serum and from human endothelial cells onto tissue culture polystyrene,” J. Biorned. Muter. Res., 21, 1317-1327 (1988). M. J. Lydon and C . A. Foulger, ”Cell substratum interactions: serum spreading factor,” Proceedings of the European Congress on Biomaterials, Elsevier, New York, 1986. M. Kubo, M. Kan, M. Isemura, I. Yamane, and H. Tagami, ”Effects of extracellular matrices on human keratinocyte adhesion and growth and on its secretion and deposition of fibronectin in culture,” J. Invest. Der- matol., 88, 594-601 (1987). E. G. Hayman, M. D. Pierschbacher, and E. Ruoslahti, “Detachment of cells from culture substrate by soluble fibronectin peptides,” J. Cell. Biol., 100, 1948-1954 (1985). 0. Fyrand, “Studies on fibronectins in the skin,” Acta Derm. Venereol
R. J. Coffey et a]., ”Production and autoinduction of transforming growth factor-alpha in human keratinocytes,” Nature, 328, 817-820 (1987). Y. Barrandon and H. Green, ”Cell migration is essential for sustained growth of keratinocyte colonies: The roles of transforming factor-alpha and epidermal growth factor,” Cell, 50, 1131-1137 (1987). G. S. Schultz et al., ”Epithelial wound healing enhanced by transform- ing growth factor-alpha and vaccinia growth factor,” Science, 235, 350- 353 (1987). G. L. Brown, et al., ”Enhancement of epidermal regeneration by bio- synthetic epidermal growth factor,” J. Exp . Med., 163, 1319-1324 (1986). D. J. Donaldson and J. T. Mahan, ”Keratinocyte migration and the ex- tracellular matrix,” J. Invest. Dermatol., 90, 623-628 (1988). D. T. Woodley, E. J. O’Keefe, and M. Prunieras, ”Cutaneous wound healing: a model for cell matrix interactions,” J. A m . Acad. Dermatol., 12, 420-433 (1985).
(Stockh), 65, 379-384 (1985).
Received September 15, 1988 Accepted September 6, 1989