J O U R N A L O F C H E M I S T R Y Materials Communication A comparative study of cell attachment to self assembled monolayers and plasma polymers Ros Daw,a Ian M. Brook,b A. Jane Devlin,b Robert D. Short,*a Elaine Cooperc and Graham J. Leggettd aDepartment of Engineering Materials, University of SheYeld, Sir Robert Hadfield Building, PO Box 600, SheYeld, UK S1 4DU. E-mail: r.short@shef.ac.uk bDepartment of Oral Maxillofacial Surgery, University of SheYeld, UK S1 4DU cDepartment of Materials Engineering and Materials Design, The University of Nottingham, University Park, Nottingham, UK NG7 2RD dDepartment of Chemistry, University of Manchester Institute of Science and Technology, PO Box 88, Manchester, UK M60 1QD Received 24th August 1998, Accepted 29th September 1998 The first comparative study of cell attachment to 4×10-2 mbar.Films were deposited on Thermanox tissue self-assembled monolayers (SAMs) and plasma-deposited culture plastic coverslips. films is reported. Osteoblast-like cells attached extensively SAMs were prepared according to well-established proto acid-terminated alkyl thiol SAMs and to a plasma co- cedures, by the immersion of gold-coated (50 nm) chromiumpolymer of acrylic acid and octa-1,7-diene (acid-PCP). primed (2 nm) glass microscope slides in 1 mM ethanolic However, they attached poorly to methyl-terminated solutions of mercaptoundecanoic acid (MUA), 3-mercapto- SAMs and a plasma polymer of octa-1,7-diene (OD-PP).propanoic acid (MPA), butanethiol (BT), octanethiol (OT) and dodecanethiol (DDT) for ca. 16 h. BT, OT and DDT were obtained from Fluka, UK, and MPA was obtained from Cell attachment to synthetic substrata is strongly influenced Sigma, UK, and were used as received. MUA was synthesised by the surface chemical structure, both directly (in serum-free following a procedure adapted from the method of Bain et al.7 conditions) and through the adsorption of proteins from Following removal of the SAMs from the thiol solution, they culture medium.SAMs have been used as ‘model’ surfaces to were rinsed with ethanol and dried in a stream of nitrogen. probe cellular responses to specific functional groups. ROS 17/2.8 cells were donated by G.A. Rodan of Merck, Carboxylic acid-terminated SAMs have been shown to pro- Sharp and Dohme. Cells were removed from liquid nitrogen, mote the attachment and spreading of osteoblasts,1 fibroblasts2 and fast thawed.They were suspended in serum supplemented and endothelial cells,3 whereas methyl-terminated SAMs have medium, seeded in tissue culture flasks and incubated for seven not promoted attachment. days. The cells were then trypsinised using 0.05% trypsin A number of attachment studies have been undertaken with containing 0.5 mM disodium ethylenediaminetetraacetic acid plasma-deposited films.4 Plasma polymers cannot be con- (Na2EDTA), before centrifuging at 405g.The cell pellet was sidered ‘model’ surfaces for studies of this type as the technique re-suspended in fresh medium. Three samples of each film leads to the incorporation of functional groups not present in were placed into 24 well trays.Cells were seeded at a density the original monomer unit. However, plasma deposits are of 1.1×105 cells ml-1 (with 1 ml per well ), and allowed to important in biomaterials science: a wide range of compounds attach in an incubator at 37 °C and 5% CO2 for 90 min. Each can be plasma polymerised and deposition is possible onto sample was then washed three times with warmed PBS to virtually any substrate.Deposits are produced in a sterile remove unattached cells. Cells were counted using a micro- environment free of initiator and solvents, and they can be scope graticule and at least four fields of view for each sample. readily used to coat medical devices and implants. The mean cells per mm2 and standard deviation were calcu- It has been shown that plasma co-polymerisation using low lated.Total DNA content of the samples was estimated using plasma powers may be used to control surface functional a Hoechst stain. The technique is described elsewhere.6 group concentration. The attachment of keratinocytes5 and Statistical significance was calculated using the Students’ t-test. osteoblast-like cells6 to plasma co-polymerised surfaces has Data were taken to be significant when a p value of 0.05 or been explored in previous studies.Films with oxygen-containless was obtained (showing a 95% confidence limit). ing functional groups encouraged cell attachment,5,6 with Fig. 1 shows the mean number of osteoblast cells per mm2 keratinocyte attachment correlating best with the carboxylic attached to each sample.Two distinct levels of attachment acid functional group.5 Optimum attachment was found to were seen. Almost equal numbers of cells attached to the acid- surfaces containing 3% carboxylic acid. Pure hydrocarbon terminated SAMs and the acid plasma co-polymer, with the films (OD-PP) produced poor attachment.5,6 possible suggestion of higher levels of attachment to the The purpose of this study was to compare osteoblast-like plasma co-polymer (acid-PCP).On the methyl-terminated cell attachment on these plasma-polymerised materials with SAMs and OD-PP, statistically similar numbers of cells were attachment to self-assembled monolayers of carboxylic acid found to have attached, although slightly higher mean levels and methyl-terminated adsorbates.of attachment were observed on the OD-PP and BT surfaces. The reactor vessel used for the plasma polymerisations is The numbers of cells on the acid-terminated SAMs and the described elsewhere.6 The plasma was sustained by a radioacid- PCP surfaces were significantly greater than the numbers frequency (13.56 MHz) signal generator and amplifier ‘inducof cells attaching to the methyl-terminated SAMs and the tively’ coupled to the reactor vessel.The base pressure of the OD-PP surfaces. reactor vessel was 8×10-3 mbar. Acrylic acid (40%) and octa- On the hydrocarbon surfaces, cells were only weakly 1,7-diene (60%), from Aldrich UK, were co-polymerised at a attached, clumped together and had a rounded morphology. plasma flow rate of 2 cm3 min-1 (STP), and power of 2 W.The cells more strongly adhered to the acid surfaces; they were Octa-1,7-diene was polymerised using the same plasma parameters. The pressure during polymerisation was typically better spread and less clumped. J. Mater. Chem., 1998, 8, 2583–2584 2583The result from the OD-PP and methyl-terminated SAMs is also interesting. Structural irregularity in the surface of the OD-PP was anticipated.The surface will present a number of diVerent carbon–hydrogen bonding arrangements (methyl, methylene, olefinic etc.). Order in SAMs is thought to be influenced by adsorbate chain length. Long chain SAMs, such as those formed from DDT, exhibit ordered, crystalline structures in which the alkyl chains pack relatively rigidly. Short chain SAMs, such as those formed from BT, are thought to exist in a two dimensional liquid state, in which the alkyl chains are relatively mobile.Consequently, while the surface of a long chain SAM is composed almost exclusively of methyl groups, the short chain SAMs are much more disordered and the alkyl chain methylene groups are exposed as well as terminal methyl groups. In view of this it is not surprising that the data for OD-PP resemble those for the short-chain Fig. 1 Cell attachment to SAMs, plasma-deposited surfaces, TCPS SAM, BT, rather than those formed from longer adsorbates and gold. Values are the average of 12 measurements (three samples (OT and DDT). ×four fields of view) except for BT.† The statistical significance ( p) In serum-containing media, cell attachment is influenced by between the acid surfaces and methyl surfaces was 0.002; between the OD-PP and methyl-terminated SAMs, p=0.1.the adsorbed protein layer. Identification of the factors that control protein adsorption and retention of activity (conformation) once adsorbed, has been the subject of considerable Fig. 2 shows the measured DNA concentrations per mm2 research eVort, with much emphasis put upon surface hydro- on these samples.The amounts of DNA measured on the philicity/hydrophobicity. However, adsorption may also be acid-PCP and the acid-terminated SAMs were almost equal. controlled by factors at the molecular level, such as the There appeared significantly more DNA on the OD-PP than availability of a specific functional group that acts at a specific both the DDT- and OT-SAMs, but a similar value was binding site for the protein.In this study, no attempt was recorded for the BT-SAM. The amounts of DNA on the acid- made to control the nature of the layer of adsorbed protein containing surfaces were significantly higher than on the at the sample surfaces. Undoubtedly, the nature of the hydrocarbon surfaces. adsorbed protein layer, which forms rapidly following exposure Also included in Fig. 1 are data for cells cultured on gold of the samples to the culture medium, plays a critical role in and tissue culture polystyrene (TCPS). The former is the determining the ultimate outcome of the cell–material intersubstrate on which the SAMs were prepared. It has been action. Future studies must address this problem.However, previously observed that cells attach well to this surface. TCPS the contrasting eVects of the diVerent chemistries, and the is the usual substratum material used in tissue culture, although similarity in the response of the plasma co-polymer and the it may be subsequently coated with various extracellular matrix acid terminated SAM, are clearly demonstrated. (ECM) components, prior to cell culture.Good attachment This study draws attention to the importance of having to TCPS was expected, but as described elsewhere, the surface well-defined surfaces on which to work. It illustrates the chemistry of TCPS can be variable and it has only limited importance of the acid functionality, but also questions value as a control.5 whether cell attachment requires high concentrations of this The degree of correlation observed in the cell counts and in functionality at the surface.The comparison of the OD-PP the DNA between the acid-terminated SAMs and the acid- with the methyl-terminated SAMs illustrates how SAMs may PCP is perhaps surprising. Although both contain carboxylic possibly be used to probe structural disorder in PPs.acid, the numbers of carboxylic acid groups per 100 carbons are very diVerent for these two types of surface. Furthermore, R.D.S. and R.D. thank the EPSRC for a studentship (for R.D.); G.J.L. and E.C. thank the Leverhulme Trust for the in the SAMs, the acid groups are known to be at the liquid provision of a research grant and G.J.L. thanks the NuYeld interface; this has not been established for the acid-PCP.In Foundation for provision of a Science Research Fellowship. aqueous media, the acid-PCP can hydrate, as we will demonstrate elsewhere.8 This behaviour is not expected of the acidterminated SAMs. Significant re-orientation of acid functional Notes and references groups within the surface of the acid-PCP, to present a higher † Cell counts comparable to those made on gold were made on two concentration of acid at the liquid interface, can be discounted of the BT SAMs, but these values were discounted on the grounds as the advancing contact angle measured on the acid-PCP (80° that the BT SAM is the least stable of the SAM surfaces studied.for distilled water) reflects the essentially hydrocarbon nature We strongly suspected that regions of the gold substratum had of this surface.become exposed. The anomaly was verified by the respective DNA values, which were also subsequently discarded. 1 C. A. Scotchford, E. Cooper, S. Downes and G. J. Leggett, J. Biomed. Mater. Res., 1998, 41, 431. 2 E. Cooper, W. Robin, D. A. Hunt, L. Parker, G. J. Leggett and T. L. Parker, J. Mater. Chem., 1997, 7, 435. 3 C. D. Tidwell, A. M. Belu, B. D. Ratner, B. Tarasevich B. S. Atre and D. L. Allara, Transactions Soc. Biomater, 23rd Annual Meeting, 1997, vol. XX, 203. 4 S. I. Ertel, A. Chilkoti, T. A. Horbett and B. D. Ratner, J. Biomater. Sci., Polym. Ed., 1991, 3, 163. 5 R.M. France, R. D Short, R. A. Dawson and S. MacNeil, J. Mater. Chem., 1998, 8, 1, 37. 6 R. Daw, A. J. Beck, R. D. Short, A. J. Devlin and I. M. Brook, Biomaterials, 1998, in the press. 7 C. D. Bain, E. B. Troughton, Y.-T. Tao, J. Evall, G. M. Whitesides and R. G. Nuzzo, J. Am. Chem. Soc., 1989, 111, 321. 8 R. M. France, R. J. Heaton, M. C. Davies, S. J. B. Tendler, C. J. Roberts, P. Williams, R. Daw and R. D. Short, unpublished Fig. 2 Total (mg mm-2) measured on the SAMs, PP, PCP, TCPS and work. gold. The statistical significance ( p) between the acid surfaces and methyl surfaces was 0.04; between the OD-PP and methylterminated SAMs, p 0.05. Communication 8/06612D 2584 J. Mater. Chem., 1998, 8, 2583–2584