*Corresponding author. Tel.: 001-801-378-2589; fax: 001-801-378-7799.

1Current address: Department of Bioengineering, John HopkinsUniversity, USA.

2Current address: Amgen, Thousand Oaks, CA, USA.

Biomaterials 21 (2000) 31}36

Attachment of hyaluronic acid to polypropylene, polystyrene,and polytetra#uoroethylene

Mitchell Mason!, Koen P. Vercruysse", Kelly R. Kirker#, Ryon Frisch!,1,Dale M. Marecak",2, Glenn D. Prestwich", William G. Pitt!,*

!Chemical Engineering Department, Brigham Young University, 350 CB, Provo, UT 84602, USA"Department of Medicinal Chemistry, University of Utah, USA

#Department of Bioengineering, University of Utah, USA

Received 28 December 1998; accepted 18 June 1999

Abstract

Surfaces of polypropylene (PP), polystyrene (PS) and polytetra#uoroethylene (PTFE) were activated with radio frequency plasmasAr and NH

3to aminate the polymer surface and were subsequently reacted with hyaluronic acid (HA) in one of the three di!erent

attachment schemes. Results show that ammonia plasma treated polymers were more reactive toward HA attachment. The threechemistry schemes consisted of two distinct approaches: (1) direct attachment of the HA to the aminated surface, and (2) extending thereactive group away from the surface with succinic anhydride and then reacting the newly formed carboxylic acid group with anadipic dihydrazide modi"ed HA (HA-ADH). The latter scheme proved to be more e!ective, suggesting that steric e!ects were involvedwith the reactivity of the HA with surface groups. These HA-coated polymers are a candidate for cell attachment andgrowth. ( 1999 Elsevier Science Ltd. All rights reserved.

Keywords: Hyaluronic acid; RF plasma; Polypropylene; Polystyrene; PTFE

1. Introduction

Hyaluronic acid (HA) is a naturally occurring anionicpolysaccharide consisting of alternating 1,4-linked unitsof 1,3-linked glucoronic acid and N-acetylglucosamine.HA is one of the many glycosaminoglycans (GAGs),including heparin, chondroitin sulfate, and keratin, thatare widely distributed throughout the body as compo-nents of the extracellular matrix (ECM) of connectivetissue. The interactions of HA with HA-binding proteinsis important in cell adhesion, growth and migration,in#ammation, cancer metastasis, and wound healing[1}4]. The lack of immunogenicity makes HA an attract-ive building block for the design of novel biomaterials[5}8].

Polymer surfaces covalently coated with HA are candi-date substrates for cell attachment and growth. The ad-vantages of an HA-coated surface are (1) it can bemodi"ed to have the cell attachment polypeptides suchas RGD peptides (or other necessary ECM components)on the surface, or can be designed to selectively adsorbthese proteins or components [7,9]; (2) it is non-cytotoxic; (3) it has minimal adsorption of other cells,proteins or bacteria; (4) it is non-immunogenic and non-thrombogenic; (5) it has the mechanical properties tofunction as a support for cell growth [10,11].

There are many applications for such an HA-coatedbiomaterial. For example, this biomaterial could be usedto support osteoblast growth for prosthetic hard tissueapplications, to support endothelial cell growth for por-ous or non-porous vascular grafts, to support chon-drocyte growth for prosthetic cartilage, or to supporthippocampal cells for guided nerve regeneration [12].

In this communication we report the covalent attach-ment of HA to some polymeric biomaterials by usingradio frequency plasma modi"cation of the substratesurface. Plasma preparation of polymer substrates toenhance adhesion to the polymer surface is a proven

0142-9612/00/$ - see front matter ( 1999 Elsevier Science Ltd. All rights reserved.PII: S 0 1 4 2 - 9 6 1 2 ( 9 9 ) 0 0 1 2 9 - 5

Fig. 1. Chemistry schemes for attachment of HA to plasma treated polymers.

Table 1Plasma treatment parameters

Treatment Base pressure(Pa)

Processingpressure (Pa)

Gas #owrate (l/min)

Processtime (min)

Argon 6.67 66.66 0.0100 1.0Ammonia 2.67 50.00 0.0552 2.0

technology [13] that has been used to attach bio-molecules to substrate surfaces [14,15]. Plasma modi"ca-tion of the polymers is typically performed by placing thesubstrate in a partially evacuated environment. A powersource of radio frequency excites the gas molecules in thechamber, creating a mixture of photons, electrons, ions,radicals, and atoms that have the potential to react withthe substrate surfaces. In this research, plasma processingwas used to chemically activate the substrate surface.

2. Materials and methods

2.1. Materials

PP "lms were obtained from Exxon (Ex-29 "lm, Ex-xon, Pottsville, PA), PS samples were obtained fromsterile Fisherbrand petri dishes (cat 08-757-14), andPTFE sheets were purchased from Ain Plastics (GroveCity, OH). HA and adipic dihydrazide modi"ed HA

(HA-ADH) were prepared as previously described [16]with modi"cations [17]. Bis-tris HCl was obtained fromSigma (St. Louis, MO). All other chemicals were pur-chased from Aldrich (Milwaukee, WI).

2.2. Plasma treatment

PP, PS, and PTFE were cut into 10]10-cm couponsand cleaned by immersion in EtOH overnight. The sam-ples were then treated with argon, ammonia, or argonfollowed by ammonia (argonPammonia) plasma. The

32 M. Mason et al. / Biomaterials 21 (2000) 31}36

plasma treatments were performed by centering the cou-pons between parallel metal plates, 20 cm from eachplate, in a Plasma Science 0500 reactor (Plasma Science,Foster City, CA) and evacuating the chamber to the basepressure (see Table 1). The chamber was then "lled withargon or ammonia and raised to the processing pressure.Once the processing pressure was reached, the RF gener-ator was activated at 185 W for the set process time. Theparameters for the argonPammonia treatments aresimply a sequential combination of these two processesin which the argon treatment was performed "rst andthen the chamber was evacuated to the base pressure andthe ammonia process sequence was immediately ex-ecuted.

2.3. HA attachment

The argon, ammonia, and argonPammonia plasmatreated samples were cut into three smaller coupons.Along with a clean untreated sample, they were exposedto one of three chemistry schemes to attach the HA (seeFig. 1). In Scheme 1 the samples were placed in a beakercontaining 2 mg/ml HA in 50 mM bis-tris HCl solutionovernight. Scheme 2 was done in a solution of 1.2 mg/mlHA, 50 mM bis-tris HCl and 1mM 1-(3-dimethylamino-propyl)-3-ethyl-carbodiimide HCl (EDCI) solution over-night. Scheme 3 consisted of exposing samples to 10 mM

succinic anhydride in dry DMF solution (for PP andPTFE) or EtOH solution (for PS) for 10 h, followed bywashing for 30 min in 10 ml distilled water and thentreating under conditions similar to Scheme 2 except forthe use of HA-ADH instead of HA. The pH of all bis-trissolutions was 4}4.5 as indicated in Fig. 1. All reactionswere performed in a shaker at 90 rpm and 63C to preventbacterial growth. The samples were then washed over-night in doubly distilled water at a low #ow rate in orderto remove any physically adsorbed HA.

3. Analytical procedures

FTIR, water contact angle measurements, and XPSwere used to characterize sample surfaces. FTIR spectrawere obtained on a Nicolet 730 FTIR spectrometer usinga 453 Ge crystal and a SpectraTek variable angle ATRapparatus. Contact angles of sessile drops were measuredusing a goniometer. XPS analysis was performed ona Fisons multi-technique ESCALAB 220I-XL surfaceanalysis instrument. The XPS survey spectra and high-resolution spectra were collected using Al K

aX-rays at

constant analyzer energies of 100 and 20 eV, respectively.The FTIR spectra gave the most conclusive results forPP and PTFE, but was less informative for PS analysisdue to the poor contact between the rigid PS samples andthe Ge crystal.

4. Results and discussion

Radio frequency plasma etching is done by exposinga polymeric substrate to a non-polymerizable gas excitedin an RF electromagnetic "eld. The resulting cold plasmais a mixture of electrons, ionized gas, and molecularfragments of the gas. This mixture actively etches awaythe polymer surface, leaving polymer fragments contain-ing radicals that can subsequently react with other mol-ecules. Argon gas was postulated to clean the polymersurface and create free radicals that could then react withammonia to produce an aminated surface. Ammonia wasused to both etch the surface and attach amine groups tothe surface. This aminated surface was then reacted usingthe various chemistries to produce amide bonds linkingthe HA to the surface.

Most polymers treated with argon, ammonia or ar-gonPammonia plasma had water contact angles thatwere signi"cantly less (P(0.01, one-sided P-value) thanthe clean untreated samples (see Fig. 2). The only excep-tion is the PP treated with argon plasma (P"0.064,one-sided P-value). Reduction of contact angle is consis-tent with the oxidation and/or amination of the hydro-phobic polymers. Indeed, the XPS analysis shows anincrease in oxygen and nitrogen following plasma treat-ment (see Fig. 3).

FTIR spectra of the treated polymer surfaces revealedthat signi"cant amounts of HA were detected only in thecase of PP treated with Scheme 3, which is shown inFig. 4 along with a spectrum of pure HA. The PP has noabsorbance at the 1045 cm~1 (ether) or 1643 cm~1 (car-bonyl) and the HA has very little absorbance at the1458 cm~1 (CH bending). Comparison of these spectrashow no change in the PP spectra after ammonia plasmatreatment. There is a hint of HA adsorbed on the non-plasma treated PP following the derivatization proced-ure of Scheme 3. However, when the PP was plasmaactivated with argon, ammonia, or both, the spectra ofHA is superimposed on the PP spectrum, indicatingsigni"cant attachment. A semi-quantitative analysisof the amount of HA attachment was performed bytaking the ratio of the 1045 and 1643 cm~1 peaks ofHA to the 1458 cm~1 peak of PP (see Fig. 5). Theseratios represent the relative amount of HA boundon each sample. These data suggest that the ammoniaplasma treatment leads to the greatest amount ofHA attachment, although the argon and argonPammo-nia also caused better attachment than without plasmatreatment. It is intriguing that the argon plasmaresulted in a small amount of covalent attachment ofHA. The XPS analysis showed some oxidation of theplasma treated PP and it may be possible that the oxygenspecies reacted with the succinic anhydride, perhapsforming ester or anhydride bonds, although the latterbonds would be expected to hydrolyze during the sub-sequent washing.

M. Mason et al. / Biomaterials 21 (2000) 31}36 33

Fig. 2. Water contact angle data for the three polymers before and after plasma treatment.

Fig. 3. XPS data for plasma treated PS and PP.

Similar analyses were also performed on PP samplestreated with Schemes 1 and 2. Those spectra indicatedthat only trace amounts of HA were bound to the surface(data not shown).

The observation that the samples treated "rst withsuccinic anhydride resulted in a greater amount of HAattachment suggests that steric e!ects may be involved.For example, the carboxyl groups in an expanded and

34 M. Mason et al. / Biomaterials 21 (2000) 31}36

Fig. 4. FTIR spectra of polypropylene, HA and polypropylene "lms treated for HA attachment with Scheme 3: (1) untreated control, (2) am-monia plasma treated, (3) no plasma treatment, HA treated, (4) argon plasma treated, HA treated, (5) ammonia plasma treated, HA treated,(6) argonPammonia plasma treated, HA treated, and (7) pure HA.

Fig. 5. PP to HA peak ratios from FTIR spectra for PP treated according to Scheme 3. Solid bars represent the 1643 to the 1458 cm~1. Shaded barsrepresent the 1045 to the 1458 cm~1.

M. Mason et al. / Biomaterials 21 (2000) 31}36 35

highly hydrated HA molecule may not be able toapproach a solid surface su$ciently close to react withamines on the surface. Another explanation of theseresults is that the amine groups on the polymer surfacecould become protonated at 4.5 pH and therefore arerendered unreactive toward the carboxyl groups of theHA. We favor the former hypothesis that the succinicanhydride produces an extended arm to which the HAcan become covalently linked.

It was di$cult to obtain good IR spectra of the PSsamples because the rigidity of the samples preventedgood contact with the internal re#ection crystal. In thesenoisy spectra there was very little evidence of the attach-ment of HA. However, the XPS data of the PS showedthat it was oxidized by the argon plasma and aminatedby the ammonia plasma. The XPS data could not quan-tify the amount of HA attachment, but the ratios ofcarbon, nitrogen, and oxygen were similar to that ofnative HA, suggesting that there may have been someHA attached to the surface.

The FTIR spectra of PTFE showed only peaks at 1207and 1151 cm~1 (symmetric and asymmetric CF stretching)characteristic of pure PTFE. There was a slight decrease inthe water contact angle on the PTFE after plasma treat-ment, but there was no sign of subsequent HA attachmentas indicated by FTIR or water contact angle measurements.This is consistent with a previous attempt to plasma-ami-nate PTFE and subsequently attach biomolecules [18].

In summary, RF plasma treatment with argon andammonia derivatized PP and PS such that HA could becovalently attached. The best attachment chemistry in-volved derivatization of the surface with succinic anhy-dride, followed by carbodiimide coupling of HA-ADH.The PP appeared more reactive than PS, and the PTFEsurface was unreactive toward this chemistry. Thismethod of preparation may allow for the derivatizationof other types of polymer surfaces. This technology couldbe used to create a novel biomaterial for cell attachmentor growth with appropriate recognition factors, or alubricious surface with minimal bacterial adhesion prop-erties. The technology could also be extended to attachother biomolecules to polymer surfaces.

References

[1] Knudson CB, Knudson W, Hyaluronan-binding proteins indevelopment, tissue homeostasis, and disease. FASEB 1993;7:1233}41.

[2] Underhill CB. Interaction of HA with the cell surface. In: Sympo-sium CF, editor. The biology of hyaluronan. Chichester, UK:Wiley, 1989. p. 87}106.

[3] Turley EA. The role of a cell-associated hyaluronan-bindingprotein in "broblast behaviour. In: Symposium CF, editor.The biology of hyaluronan. Chichester, UK: Wiley, 1989.p. 121}37.

[4] Weigel PH, Forst SJ, LeBoeuf RD, McGary CT. The speci"cinteraction betweeen "brin(ogen) and hyaluronan: possible conse-quences in haemostasis, in#amation and wound healing. In: Sym-posium CF, editor. The biology of hyaluronan. Chichester, UK:Wiley, 1989. p. 248}64.

[5] Drobnik J. Hyaluronan in drug delivery. Adv Drug Delivery Res1994;7:295}308.

[6] Swann DA, Kuo JW. Biomaterials: novel materials from biolo-gical sources. In: Byrom, editor. New York, NY: Stockton Press,1991. p. 287}305.

[7] Vercruysse KP, Prestwich GD. Hyaluronate derivatives in drugdelivery. Critic Rev Therapeutic Drug Carrier Systems1998;15(5):513}55.

[8] Prestwich G, Vercruysse K. Therapeutic applications of hya-luronic acid and hyaluronan derivatives. Pharmaceut Sci TechnolToday 1998;1(1):42}3.

[9] Prestwich GD, Marecak DM, Marecek JF, Vercruysse KP,Ziebell MR. Chemical modi"cation of hyaluronic acid for drugdelivery, biomaterials and biochemical probes. In: Laurent TC,editor. The chemistry biology and medical applications of hya-luronan and its derivatives. London: Portland Press, 1998.p. 43}65.

[10] Band PA. Hyaluronan derivatives: chemistry and clinical applica-tions. In: Laurent TC, editor. The chemistry biology and medicalapplications of hyaluronan and its derivatives. London: PortlandPress, 1998. p. 33}42.

[11] Laurent TC, Fraser JR. Hyaluronan. FASEB 1992;6:2397}404.[12] Seckel BR, Jones D, Hekimian KJ, Wang KK, Chakalis DP,

Costas PD. Hyaluronic acid through a new injectable nerve guidedelivery system enhances peripheral nerve regeneration in the rat.J Neurosci Res 1995;40(3):318}24.

[13] Kaplan SL, Rose PW. Plasma surface treatment of plasticsto enhance adhesion. Int J Adhesion Adhesives 1991;11(2):109}13.

[14] Sipehia R, Chawla AS. Albuminated polymer surfaces for bio-medical applications. Biomater Med Dev Artif Organs 1982;10:229}46.

[15] Mutlu M. Matrix surface modi"cation by plasma polymer-ization for enzyme immobilization. J Mater Chem 1991;1(3):447}50.

[16] Pouyani T, Prestwich GD. Functionalized derivatives ofhyaluronic acid oligosaccharides: drug carriers and novel bio-materials. Bioconj Chem 1994;5:339}47.

[17] Vercruysse KP, Marecak DM, Marecek JF, Prestwich GD. Syn-thesis and in vitro degradation of new polyvalent hydrazidecross-linked hydrogels of hyaluronic acid. Bioconj Chem1997;8(5):686}94.

[18] Chu T-J, Caldwell KD, Weiss RB, Gesteland RF, Pitt WG. Low#uorescence background electroblotting membrane for DNA se-quencing. Electrophoresis 1992;13:105}14.

36 M. Mason et al. / Biomaterials 21 (2000) 31}36