improved cell adhesion to ion beam-irradiated polymer surfaces

ELSEVIER PII:SO142-9612(97)00090-2

Biomoterials 18 (1997)1461-1470 0 1997 Elsevier Science Limited

Printed in Great Britain. All rights reserved 014%9612/97/$17.00

Improved cell adhesion to ion beam- irradiated polymer surfaces

B. Pignataro, E. Conte, A. Scandurra* and G. Marletta Diparfinkto di Scienze Chimiche, University of Catania, Via/e A. Doria 6, 95125 Catania, Italy; *Surface and interface Laboratory, Consorzio Catania Ricerche, V. le A. Doria 6, 95125 Catania, ltaly

A very strong improvement of the cell adhesion, spreading and proliferation was observed for ion

beam irradiated surfaces of polyethersulphones and polyurethane. The improvement shows a

characteristic ion fluence-dependence with a threshold around 1 x 1015 ionscm-‘. We have compared

this improvement of surface cytocompatibility with the various ion-induced chemical and physical

modifications of the polymeric surfaces, taking into account their characteristic fluence-dependence.

The irradiation induced a severe compositional and chemical modification of the polymeric surfaces

as detected by X-ray photoelectron spectroscopy. Contact angle measurements showed that the

surface free energy was strongly modified by ion irradiation. The fluence-dependent formation of a

hydrogenated amorphous carbon phase was demonstrated by Raman spectroscopy. Our findings

indicate that neither the simple modification of the functional groups nor the mere elemental

composition nor the modification of the surface free energy can explain the observed fluence-

dependence of the cell adhesion enhancement. On the contrary, we show that this enhancement

correlates with the formation of a highly specific ion-induced ‘unsaturated’ a-C:H phase. According to

these findings, we suggest that the formation of a substantial amount of hydrogenated amorphous

carbon phases is the major factor promoting the cytocompatibility of ion irradiated polymer surfaces.

0 1997 Elsevier Science Limited. All rights reserved

Keywords: Cytocompatibility, cell adhesion, ion beams, polyethersulphone, polyurethane

Received 15 September 1996; accepted 5 May 1997

Ion irradiation is a very promising tool to modify the chemical structure and the physical properties of polymers19 ‘. In particular, it has been shown that not only the chemical composition of the irradiated polymer layers can be modified in a controlled way, but also the related physical properties can be selectively modified according to a quite easy-to- control parameter consisting of the ion fluence, i.e. the total number of particles bombarding a given surface3. Furthermore, it has been shown that the thickness of the polymeric modifiied layer can be easily controlled by choosing the suitable primary ion mass and energy4.

In a recent series of papers, Suzuki et a1.5*6 and Nakao7 showed that it is possible to greatly improve the cytocompatibility of several polymeric materials, including segmented polyurethane (SPU), polystyrene (PS) and silicone rubber, by using ion implantation (with reactive ions such as Na+, Oi, Nl and non- reactive ions like Krt) in the range of energy around 150 keV. In particular, they showed that the endothelial cell adhesion and spreading is greatly enhanced by ion implantation in SPU, while the effect is less important with PS. The effect was attributed to the formation of a peculiar structure of the carbonaceous phase formed under irradiation to around 1 x 1OJ7 ions cm-’ for both SPU and PS. Other authors related the cell adhesion and proliferation on the irradiated surfaces to the modification of the surface free energy’-” or to the

Correspondence to Professor G. Marietta.

chemical nature of the employed polymeric substrateslO.

The present paper was aimed to study the cytocompatibility of ion-irradiated surfaces of two different polymers, i.e. polyurethane (PU) and polyethersulphone (PES). The relevant parameters considered for the evaluation of the cytocompatibility of the various treated polymer surfaces were the adhesion, proliferation and spreading behaviour of the cells. Human astrocyte cell line was used to get a first screening test on the different surfaces. Primary human endothelial cells (HUVEC) were used to test the possibility of obtaining an innovative strategy for solving the problems owing to the thrombus formation in artificial vessels. Such a strategy is based on the development of fully blood compatible polymeric vascular grafts consisting of endothelial cell seeding onto ion-irradiated olymeric surfaces5-7.

Previous studied’ mostly dealt with very high fluence irradiation treatments, i.e. higher than 1 x 1016 ions cm-‘, where it is achieved the formation of a carbonized phase whose structure is still matter of debate11Z12. In the present paper we show that a significant improvement of cell adhesion, proliferation and spreading can be obtained also for much lower fluence treatments. Furthermore, as most of the chemical and physical properties of irradiated surfaces were modified above given fluence thresholds we try to correlate such critical thresholds with the improvement of the cytocompatibility.

1461 Biomaterials 1997, Vol. 18 No. 22

1462 Improved cell adhesion to ion beam-irradiated polymer surfaces: B. Pignataro et al.

MATERIALS AND METHODS

Sample preparation

Two different commercial materials were utilized: polyurethane and polyethersulphone, having respectively the formulas reported in Figure Z. The polyurethane was a Pellethane R 2363-80 AE (from Upjohn Polymers). The polymer has been purified by extraction in Soxhlet with a solution of CH3COCH3/ CH,OH (v/v; l/2.5) over a period of 3 days13. Poly-1,4- phenylene-ether-sulphone (PES) was purchased from Aldrich. The polymers were studied in the form of thin film (about 1 pm of thickness), obtained by spinning polymer solutions (3% in weight), respectively, in tetrahydrofurane (THF) for PU and in chloroform for PES, at 2000rpm for 30s onto 5-inch Si wafers. The complete removal of the solvent as well as the integrity of the polymer after the coating step was confirmed by XPS analysis. The wafers were cut into samples of about 4 cm’. The polymeric substrates were sterilized after the ion irradiation by further treatments with UV source (3, = 253.7 nm, power = 30 W, distance from the samples 60 cm) for 15 min. This treatment has been verified by XPS to not induce any significant modification of the polymer surface structure14.

Ion irradiation

Ion irradiation was performed with an ion implanter (High Voltage Inc.) operating between 30 and 400 keV. The polymer samples were irradiated by using 35 keV Ar+ to fluence of 1013, 5 x 1013, 1014, 5 x 1014, 10J5,

5 x 1Ol5 ioncm-’ for both polymers. In order to prevent thermal effects, the beam current was kept under 1 PA cm-‘. The pressure in the target chamber was maintained under 10m5 Pascal during irradiation.

Characterization techniques

The samples were analysed using X-ray photoelectron spectroscopy (XPS), Raman spectroscopy and contact angle measurements.

XPS analysis of the untreated and irradiated polymers were performed for at least six different samples using a Kratos AXIS HS spectrometer, equipped with a dual Al/Mg anode. The spectra were obtained using Mg Kcc,,, radiation (at 1253.6eV) with an intrinsic resolution of 0.8 eV. The analyses were performed in ultrahigh vacuum conditions (5 x lo-r’torr). The analysed area was 4mm’. The XPS spectra were analysed by using a peak fitting routine based onto Gaussian peaks and inelastic background subtraction.

The Raman spectra of the untreated and treated polymer thin films were obtained using a Jobin Yvon U 1000 spectrometer equipped with an Argon laser source (the wavelength of the exciting radiation was 514.5 nm).

The contact angle measurements were performed using drops (10~1) of pure deionized water (p = 17 MR cm) with a Kernco instrument at room temperature.

Cell culture

Human astrocytoma ADF cell line was maintained in RPM1 1640 medium supplemented with 10% fetal calf serum (FCS), penicillin (0.1 mgml-I), streptomycin (0.5 mg ml-‘), L-Glu 2 mM, at 70-80% confluence. On the day of the experiments the cells were placed on reference unirradiated samples as well as on the irradiated surfaces at a concentration of about 3 x lo4 cells cm-‘. Three samples for each irradiation fluence were prepared. After 2-7 days the cell cultures were observed using optical microscopy (Wild M-400 Heerbrugg) and at least five random visual field photographed with Agfa 100 ASA tungsten film.

Human umbilical vein endothelial cells (HUVEC) were extracted from umbilical cords as described in the literature15 and maintained in M-199 supplemented with 10% FCS, penicillin (0.1 mgml-‘), streptomycin (0.5 mg mll’), L-Glu 2 mM, heparin (0.1 mgmll’), endothelial cell growth factor (ECGF, 0.5 mgml-‘) on 2% porcine gelatine coated plates. On the day of the

/-\ +-+OCH,CH,CH,CH,~ O- . . -C- NH ,, \ ,fChi, ,f NH-i-..*.-O-GH,),-O--+,

0 0

a

b

Figure 1 Chemical structure of repeat units of: a, polyurethane; b, polyethersulphone.

Biomaterials 1997, Vol. 18 No. 22

Improved cell adhesion to ion beam-irradiated polymer surfaces: i3. Pignataro et al. 1463

experiments, cells were detached by trypsin, washed and plated on reference unirradiated samples as well as on the irradiated surfaces at a concentration of about 2.5 x lo4 cellsmll’. The cells were used at passage earlier than fourth. The cell cultures were observed after 2-5 days with the same procedure as for astrocytoma cells.

Table 1 Elemental atomic percentage obtained by X-ray photoelectron spectroscopy of polyethersulphone (PES) and polyurethane (PU) surfaces before ion treatment and after irradiation. The values for low temperature isotropic carbon (LTIC) are reported for comparison purposes”

Elemental atomic percentages

Ion influence (ions cmm2) C 0 S N Si Quantitative evaluation of adhered cells was

performed by computer analysis of digitalized images obtained with an optical microscope. The analysis was performed using the NIH Image software (version 1.2)

on a Macintosh Power PC. This software allowed to evaluate the total area covered by white spots (cells) on the black substrate in the photographs, evaluating the cell coverage in. terms of integrated density (ID = Nx[M-B], where I\J is the number of pixels in the selection, M is the average grey value of the pixels and B is the most common pixel value). The quantitative evaluation of adhered endothelial cell was performed by direct counting with optical microscopy. Statistical differences in cell adhesion coverage of irradiated surfaces were determined by ANOVA.

PES 10’3 5 x 10’3 10’4 5 x 10’4 10’5 5 x 10’5

PU 70’3 5 x 70’3 7o14

:OT510’4 5 x 70’5

LTIC

74.58 19.91 5.20 75.02 20.32 4.41 76.75 18.68 3.85 77.89 17.87 3.46 79.32 17.01 3.06 83.76 73.45 2.79 90.04 7.47 2.49

77.18 79.62 3.20 76.37 20.61 3.03 76.53 79.51 3.96 76.88 79.26 3.86 79.74 76.98 3.28 83.05 14.09 2.86 90.05 8.05 1.89

86.6 9.6 7.5 2.3

RESULTS

Surface characterization

XPS analysis The patterns of beam-induced modification for PES has been thoroughly investigated in previous papers from this laboratory1-3s16-18, while for PU the modification pattern has been less investigatedlg.

Two basic effects are induced by the ion irradiation in these polymers: the first one consists of the mere compositional modification of the irradiated surfaces. The second one consists in the modification of the concentration of the existing functional groups or in the creation of new groups.

Cm and ‘shake-up’ components in the C 1s peak, together with the decrease of the 01 and -SO1- components in the 0 1s and S 2p peaks, respectively. Simultaneously, new components attributed to C=O and COO- in C 1s peak are formed and a component owing to adsorbed Hz0 is observed in 0 1s peak. Finally, two reduced sulphur species are created at about 165.7eV (-SO- groups) and 163.5 eV (-S- linkages) in S 2p peaks.

The variation in the relative concentrations of these groups are reported in Figure 3 for the C 1s (Figure 3a), 0 1s (Figure 3b) and S 2p (Figure 3c) components.

The described modifications can be interpreted in terms of the following basic chemical modifications16-18.

The detailed elemental composition obtained by XPS analysis for pristine and Ar-irradiated PES and PU is reported in Table 1, together with the composition of a typical commercial biocompatible carbon surface (low temperature isotropic carbon, LTIC)“. One can observe that the basic: process consists in the depletion of the heteroatoms, i.e. S and 0 for PES and 0 and N for PU, with the simultaneous enrichment of the surfaces in carbon.

(1) the reduction of the sulphonyl groups (-SO,-_) to sulphidic-like groups (-S-);

(2) the elimination of sulphur either in the form of SO2 (as in the thermal case) or as sputtered -S- containing species;

(3) the formation-and-destruction of the ether or hydroxyl groups, owing to recoiling-induced or to transfer reactions of oxygen or to elimination of O- containing gaseous molecules;

Figure Za-c reports respectively the C Is, 0 1s and S 2p XPS peaks for PES samples before irradiation and at two different ion fl-uence.

(4) the destruction of the backbone aromatic rings. Let us consider now the behaviour of PU under

Before irradiation the C 1s peak is formed by three components respectively, owing to carbons in the aromatic moiety (C, component at about 285.0 eV), C-S

linkages Crr component at about 285.6eV), C-O-C groups (Car component at about 286.5eV) and to an ionization-excitation satellite owing to rc* + Tc transition (‘shake-up’ satellites), diagnostic of the presence of aromatic rings (at about 291.6 f 0.2 eV). 0 1s band before irradiation is formed by two components, respectively, owing to a sulphonyl group (component Or at about 532.2 eV) and to the ether-like group (component Ori at 533.6eV). Finally, the S 2p peak consists in a characteristic doublet (2p,,,-2p,,, components) centred at about 168.OeV owing to -SOZ- groups.

irradiation. Figure da-c reports the XPS spectra of C Is, 0 1s and N 1s peaks respectively, before and after Ar+ irradiation of PU samples. In this case the relevant functional groups for the unirradiated samples are the aromatic rings, the -CO-NH- and the ether groups. The C 1s band before ion irradiation is formed by three main components: C1 component at about 285.0eV owing to aliphatic hydrocarbons, Cir component at about 286.5eV owing to C-O and C-N linkages and Car component at about 288.2 eV owing to carbonyls. 0 1s band is formed by three components respectively, owing to carbdnyl oxygen (0, at about 532.2eV), to ether oxygen (Ori at about 533.4eV) and to adsorbed Hz0 (Ora at about 534.2eV). Finally, the N 1s band is formed by a single sharp component owing to -NH groups at about 400.0 eV.

After ion irradiation one can observe the decrease of Ion irradiation induces the decrease of the Crr


1464 Improved cell adhesion to ion beam-irradiated polymer surfaces: 8. Pionalaro et al.

Cls

Unirradiated

01s

a b Binding energy (eV)

Figure 2 XPS Tp;Qectra of_TES samples before (upper spectra), and after Ar + 35KeV ion irradiation at 1014 ion cm-* (middle spectra) and 10 Ions cm (lower spectra) for: a, C 1s; b, 0 Is; c, S 2p peaks.


improved cell adhesion to ion beam-irradiated polymer surfaces: 6. Pignataro et al. 1465

K

-5 t; \ G

25

20

15

10

5

0 r-

lo'* lOI

Fluence (ion/cm2)

6\” 4

0

i7-l \ ::

I nn I vu

60

60

R

b

1o12 1 o14 lOI

Fluence (ion/cm2)

C Fluence (ion/cm*)

Figure 3 Modification trend of the functional group percentage in PES versus Arf 35 KeV fluence, derived from: a, C 1s peak analysis (C,/&); b, 0’1s XPS analysis (0,/O,,,); c, S2p peak analysis (SJS,,,).

component, i.e. the C-O groups, and the relative increase of the C:rn component, owing to the carbonyl groups. Such a relative increase of the carbonyl groups is confirmed by the small increase of carbonylic component in the 0 1s band after irradiation (from I!5 to 25%). The effect can be explained in terms of the formation of new =C=O groups owing to reactions of recoiling oxygen atoms. Furthermore, the -NH band decreases, producing two new components respectively, assigned to iminic groups at about 399.0eV and tertiary amines or quaternary nitrogen groups at about 401.4 eV16r21.

The variation in the relative concentrations of these groups is reported in Figure 5a for the 0 1s and Figure 5b N 1s components.

Raman analysis Figure 6a shows the Raman spectra of PU samples irradiated at increasing fluence (1 x 1014-5 x 1Ol5 ions cm-2) in the spectra region between 1000 and 2000cm-I. At the fluence of 5 x 1014 ions cmm2 only a very small signal was observed. At the critical fluence of 1 x 1Ol5 ionscmW2 a prominent spectroscopic feature is developed for both polymers consisting of a wide asymmetric band. Very similar results were obtained for PES. According to the literature”, these spectra can be analysed in terms of two bands with Gaussian line shapes, respectively centred at about 1560 cm-’ and 1420 cm-‘. A typical example of deconvolution is reported in Figure 6b, showing the spectrum of PU sample irradiated with Ar+ 35 keV to 5 x 1Ol5 ions cm-‘. Such a two bands are diagnostic, as


1466 Improved cell adhesion to ion beam-irradiated polymer surfaces: B. Pignataro et al.

Intensity (a.u.) 01s

a

Unirradiated

Binding energy (eV) b Binding energy (eV)

Nls

404 403 402 401 400 399 398 397 346

C Binding energy (eV)

Figure 4 XPS spectra of PU samples before (upper spectra), and after Ar + 35KeV ion irradiation at lOi ion cm-* (lower spectra) for: a, C 1s; b, 0 1s; c, N 1s peaks.

a whole, of the formation of amorphous carbon phases22V23. In particular, the band around 1560cm-* is characteristic of localized graphitic structures with no long-range order”, i.e. more or less small clusters of sp2 carbon sites trapped into an amorphous phase. The second band at around 1420cm-’ is characteristic of the embedding amorphous carbon phase24. The

intensity ratio of the bands Z(14.20 cm~‘)/1(1560 cm-‘) is related to the structural features of the amorphous carbon phase. In fact, an intensity ratio of 0.9 is diagnostic of a very disordered structure with a high sp3 bonding fractionz5. The intensity ratios in our case range between 0.55 for PU and 0.77 for PES at lOI5 ions cme2 and 0.77 for PU and 0.85 for PES at 5 x 1Ol5


Improved cell adhesion to ion beam-irradiated polymer surfaces: i3. Pignataro et al. 1467

K -w 0

5 \ z

1 ol; 1 o14 1016

Fluence (ion/cm2)

80

60

1o12 1o14 lOI

F-hence (ion/cm2) Figure 5 Modification trend of the functional group percentage in PU versus Ar+ 35 KeV fluence, derived from: a, C 1s peak analysis (CJC,,,); b, N 1s peak analysis (NJN,,,).

ions cm-‘. These values suggest that PES is indeed more disordered than PU at the same total deposited energy. Finally, one has to point out that the intensity ratio in amorphous carbons is directly related to the

sPz crystallite size and numberz5, thus the data support the picture of a higher number of the sp’ crystallites in irradiated PES than in PU.

The present results are in close agreement with previous reports on the ion-induced effects in polymers above a critical fluence. mainlv consistine of the formation of hydrogenated amorphous carbo; phases (a-C:H)26*27.

Contact angle measurements One of the thermodynamic parameters playing an important role in the improvement of the cell

a

I 1 1 I I I I I I 1 I 1 I I I I _

1000 2000

Wavenumber (cm-‘)

A

rs

000

800

600

400

200

0 I I I I I I II 1 II 1’ 1000 1200 1400 1600 1800 2000

b Wavenumber (cm-‘)

Figure 6 a, Raman spectra of PU samples irradiated with ArC 35KeV at fluence ranging between 1014 and 5 x 1015 ionscm-‘. b, Gaussian peak fitting of the Raman spectrum of PU irradiated at 5 x 1015 ionscm-‘.

adhesion is claimed to be the increase of the surface free energy”. The contact angle measurement technique has been employed in the present work to evaluate the variation in surface free energy.

Figure 7 reports the water contact angles measured as a function of 35 keV Ar+ fluence for both PES and PU films. The contact angle in the unirradiated samples is about 72” for both polymers. No significant variation is observed up to a fluence of 1Ol4 ions cm-’ for PES and 5 x 1Ol3 ionscm-’ for PU), the contact angle is suddenly modified to about 60” for PU and 64” for PES. These values seem to saturate for higher fluence. Thus, the total variation is confined to about lo”-15”.


1468 Improved cell adhesion to ion beam-irradiated polymer surfaces: 6. Pignataro et al.

PES

J”

IdO 1o12 1o14 1o16

Fluence (ion/cm2>

Figure 7 Decrease of PES and PU water contact angle versus increasing Ar+ 35 KeV fluence. Values are averaged on at least 10 measurements.

Cell adhesion tests

Adhesion and proliferation of astrocytes on PES and PU Shown in Figure 8a-d are light micrographs illustrating the cell adhesion trend for astrocytes on PES surfaces irradiated by increasing ion fluence. Very similar images of PU samples were obtained (not shown). The increase of the cell coverage was quantitatively estimated by NIH computer image analysis (see Materials and Methods section). In Figure 9a the cell coverage ID mean values of adhered cells on both polymers versus the ion fluence are reported.

The cell coverage on irradiated PU surfaces grew slightly between 1 x 1013 and 5 x 1Ol4 ionscm-’ (i.e. there are no significant differences between irradiated and the reference unirradiated samples), while a drastic increase (about a factor 6 with respect to the reference samples) occurred at a fluence of 1 x 1Ol5 ions cm-‘.

A slightly different trend was observed for PES. The cell coverage of irradiated PES surfaces grew between 1 x lOI and 5 x 1014 ions cm-‘, up by about a factor of 3 compared with the reference unirradiated samples; however, a drastic increase of adhered cells (about a factor of 9 with respect to the reference sample) was observed, as for PU, at a fluence of 1 x 1015 ions cm-*.

Adhered cell morphology By direct optical microscopy observation, on the irradiated surfaces of both polymers, the cells appeared to be completely spreading, showing their characteristic morphology. On the contrary, on the unirradiated surfaces cells maintained a round shape.

Endothelial cells on PES and PU In Figure gb the endothelial cell adhesion trend for both polymers is reported. On the unirradiated PES and PU samples as well as on both polymer surfaces irradiated up to a fluence of 5 x 1Ol4 ions cm-‘, the cell adhesion was very poor and cells appeared not to be spreading.

Figure 8 Light micrographs of human astrocytoma ADF cells adhered on: a, untreated PES and b, c, cll PES irradiated with Ar+ 35KeV at 1013, 5 x 1014 and 10 Ions cme2, respectively.

Ion irradiation at a higher fluence, i.e. 1Ol5 ionscm-‘, was found to induce a significant improvement of cell adhesion on the surfaces of both polymers.

DISCUSSION AND CONCLUSIONS

In the present experiments a very strong improvement of cell adhesion, proliferation and spreading on both polymer surfaces irradiated above the fluence threshold of 1 x 1Ol5 ions cm-’ was observed. In order to correlate this improvement of surface cytocompatibility with the various ion-induced chemical and physical modifications of the polymeric surfaces, we have to take into account their characteristic fluence-dependence.

Let us consider first the compositional modifications of the irradiated polymer surfaces.

(1) For PES surfaces a characteristic fluence threshold for the -SO,- to -S- reduction process was observed. On the contrary, the loss of oxygen and sulphur occurred without any threshold up to highest studied fluence.


Improved cell adhesion to ion beam-irradiated polymer surfaces: B. Pignataro et a/. 1469

a Irradiation dose (ions/cm*)

b Irradiation dose (ions/cm*)

Figure 9 a, Astrocytic cell coverage on PU and PES surfaces irradiated with increasing fluence. ID values (see Materials and Methods) are average of at least five measurements of three samples for each treatment in two separate experiments. Statistical analysis was performed by ANOVA, l f < O.O!j, number cm-’

‘*P < 0.001. b, Endothelial cell for PES and PU. Values are average of at

least three counting on random fields of three samples for each treatment in two separate experiments. Statistical analysis as indicated earlier.

(2) For PU the production of various nitrogen- and oxygen-containing groups as well as their gradual loss without any fluence threshold was observed.

In any case, the ion-induced chemical processes induced on both polymer surfaces were completed below the fluence threshold for the observed strong enhancement of cell adhesion. Thus, for instance, the reduction of -SOZ-- was substantially completed at 3 x 1014 ions cm-‘, while the ether-like groups reached a maximum at about 5 x 1014 ions cm-‘. No chemical function was found to undergo abrupt modification at the threshold fluence for the enhanced cell adhesion, so that all such chemical modifications cannot be responsible for the major cytocompatibility improvement.

Another important finding was that while the surfaces of the two polymers irradiated to high fluence were chemically different (in terms of different concentration of residual 0 groups and different residual heteroatoms, S for PES and N and PU respectively) no significant variation was observed in the cell adhesion behaviour for both polymers.

Let us turn now to the surface free energy changes for the irradiated surfaces, which is known to be strictly related to the chemical modifications. In this case there was a fluence threshold for the sudden drop in contact angle at 1014 ionscm-’ for PES and 5 x 1013 ionscme2 for PU, respectively. Note that the fluence threshold for the modification of contact angle in PES may be related to the corresponding threshold observed in the -SOZ- reduction, while the modification in the contact angle of PU should be linked to a less defined modification, possibly consisting of the secondary structure modification of the PU chains. We suggest that this change in surface free energy may play a role in the small enhancement observed for cytocompatibility at low fluence. However, we stress the fact that the fluence threshold for the surface free energy change is not related to the dramatic increase of the number of adhered cells, i.e. 1 x 1015 ionscm-“.

Finally, let us consider the chemical and structural features deduced from Raman spectra. A substantial amount of complex hydrogenated amorphous carbon phases (a-C:H) was formed only above the threshold of 1 x 1OJ5 ionscm-’ for both polymers, even though a very low-intensity signal (related to a-C:H phases) can be found below this threshold. Thus, we suggest that the whole surface was globally transformed in a- C:H only around the threshold fluence. We underline that this threshold for the a-C:H formation is closely related to that observed for the strong enhancement of the cell adhesion.

According to these findings, we propose that the formation of a substantial amount of hydrogenated amorphous carbon phase is the major factor promoting the cytocompatibility of ion irradiated polymer surfaces. This hypothesis allows justification of the fact that very similar results for cell adhesion are obtained for polymers of different chemical natures such as PES and PU. Indeed, the similarity of the cell response for the two irradiated polymers may be justified in terms of the substantial similarity of the a-C:H phases formed in both cases, irrespec- tive of the two initial chemical structures, according to the well-established finding that irradiated polymers tend to lose the memory of the initial structure, evolving towards a complex tridimensional a-C:H networkzs3.

Furthermore, this hypothesis is in complete agreement with the results reported in the studies of Suzuki et a1.5-7, which related the improvement of biocompatibility with the formation of a-C:H in polystyrene, silicone rubber and segmented PU surfaces irradiated to 1016-1017 ions cm-‘.

Based on the above discussion, we suggest that the simple functional group modification or the mere elemental composition alone do not explain the strong enhancement of cell adhesion. On the contrary, we cannot exclude that the increase in the surface free energy for the irradiated surfaces plays a role in promoting cell adhesion, as one of the necessary but not sufficient factors, while the ion-induced ‘global’ chemical modification, i.e. the formation of a highly specific ion-induced ‘unsaturated’ a-C:H phase, seems to play the leading role.

Our results show that the use of ion irradiation to modify the chemical and physical properties of


1470 improved cell adhesion to ion beam-irradiated polymer surfaces: i3. Pignataro et al.

polymer surfaces are of interest as a way to improve the tissue integration of polymeric prostheses, by inducing the enhancement of tissue cell adhesion on their surfaces. However, the development of a well-established procedure to improve biocompatibility, still needs to develop the research over a wide field of problems. In particular, the basic biological mechanisms for the enhancement of cell adhesion on such modified surfaces should be clarified and the stability of the ion-induced modifications should be established. Work is in progress on these aspects.

ACKNOWLEDGEMENTS

Professor Roland0 Barbucci is acknowledged for many helpful discussions and for kindly providing polyurethane samples. Dr Marina Sironi (Istituto di Ricerche Farmacologiche ‘Mario Negri’, Milano, Italy) is gratefully acknowledged for kindly providing HUVEC. G.M. gratefully acknowledges the financial support of CNR (Rome) and Italian Ministry of University and Technological Research (MURST).

REFERENCES

1.

2.

3.

4.

5.

6.

7.

Marletta, G., Chemical reactions and physical property modifications induced by keV ion beam in polymers. Nucl. Instrum. Methods B, 1990, 46, 295-305. Marletta, G. and Iacona, F., Chemical and physical property modifications induced by ion irradiation in polymers. In Materials and Processes for Surface and Interface Engineering, Vol. 290, ed. Y. Pauleau. NATO- AS1 Series, Serie E, Applied Sciences, Kluwer, Dordrecht, The Netherlands, 1995, pp. 597-640. Pignataro, S. and Marletta, G., Ion beam induced chemical reactions at polymer surfaces. In Metallized Plastics 2 - Fundamentals and Applied Aspects, ed. K.L. Mittal. Plenum Press, New York, 1991, pp. 269-281. Venkatesan, T., Calcagno, L., Elman, B. S. and Foti, G., Ion beam effects in organic molecular solids and polymers. In Ion Beam Modification of Insulators, ed. P. Mazzoldi and G. W. Arnold. Elsevier, Amsterdam, 1987, pp. 301-379. Suzuki, Y., Kusakabe, M., Akiba, H., Kusakabe, K. and Iwaki, M., In vivo evaluation of antithrombogenicity for ion implanted silicone rubber using indium-Ill- tropolone platelets. Nucl. Instrum. Methods B, 1991, 59160, 698-704. Suzuki, Y., Kusakabe, M., Iwaki, M., Akiba, H. and Kusakabe, K., In vivo evaluation of antithrombogenicity and surface analysis of ion-implanted silicone rubber. Radiat. Phys. Chem., 1992, 39(6), 553-560. Nakao, A., Kaibara, M., Iwaki, M., Suzuki, Y. and Kusakabe, M., XPS and SERS studies of cell adhesion- controlled polymer modified by ion implantation. Surf. Interf. Anal., 1996, 24, 252-256.

8. Dejun, L. and Zhao, J., The structure and biomedical

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

behaviour of ion bombarded and plasma polymerized segmented polyurethane. Appl. Surf. Sci., 1994, 76, 195-200. Lhoest, J.B., Dewez, J.L. and Bertrand, P., PMMA surface modification under keV and MeV ion bombardment in relation to mammalian cell adhesion. Nucl. Instrum. Methods B, 1995, 105, 322-327. Svorcik, V., Rybka, V., Hnatowicz, V., Bacakova, L., Lisa, V. and Kocourek, F., Surface properties and biocompatibility of ion-implanted polymers. J. Mater. Chem., 1995,5, 27-31. Wang, Y. Q., Giedd, R. E. and Bridwell, L. B., Ion induced structures and electrical conduction in implanted polymer films. Nucl. Instrum. Methods B, 1993, 37136, 659-664. Davenas, J., Thevenard, P., Boiteux, G., Fallavier, M. and Xu, X. L., Hydrogenated carbon layers produced by ion beam irradiation of PMMA and polystyrene films. Nucl. Instrum. Methods B, 1990, 46, 317-323. Barbucci, R., Albanese, A., Magnani, A. and Tempesti, F., Coating of commercially available materials with a new heparinizable material. J. Biomed. Mater. Res., 1991, 25, 1259-1274. Pignataro, B. and Marletta, G., UV-irradiation induced chemical effects on polyethersulfones and polyur- ethanes. Polym. Degrad. Stabil. (submitted). Jaffe, E. A., Nachmann, R. I., Becker, C. G. and Minick, C. R., Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunological criteria. J. Clin. Invest., 1973, 52, 2745- 2756. Marletta, G., Catalano, S. M. and Pignataro, S., Chemical reactions induced in polymers by keV ions, electrons and photons. Surf. Interf. Anal., 1990, 16, 407-411. Marletta, G., Pignataro, S., Toth, A., Bertbti, I., Szekely, T., Keszler, B., X-ray, electron and ion beam induced modifications of poly(ethersulfone). Macromolecules, 1991,24,99-105. Marletta, G. and Iacona, F., Chemical selectivity and energy transfer mechanisms in the radiation-induced modification of polyethersulfone. NucJ. Instrum. Methods B, 1996,116,246-252. Dejun, L., Jie, Z., Hanquing, G., Mozhu, L., Fuqing, D. and Qiqing, Z., Surface modification of medical polyurethane by silicon ion bombardment. Nucl. Instrum. Methods B, 1993, 62, 57-62. Feng, L. and Andrade, J.D., Protein adsorption on low temperature isotropic carbon: V. How is it related to its blood compatibility? J. Biomater. Sci. Edn., 1995, T(5), 439-452. Iacona, F. and Marletta, G., Energy deposition mechanisms and radiation induced reactions in PMDA-ODA polyimide. NucJ. Instrum. Methods B, 1992, 65, 50-54. Yoshikawa, M., Katagiri, G., Ishida, H., Ishitani, A. and Akamatsu, T., Raman spectra of diamond-like amorphous carbon films. J. Appl. Phys., 1988, 64, 6464-6468. Knight, R. and White, A., Raman analysis of diamond- like films. J. Mater. Res., 1988, 4, 385-391. Iwaki, M., Yabe, K., Fukuda, A., Watanabe, H., Itoh, A. and Takeda, M., Characterization of Ag and W- implanted polyimide films. Nucl. Instrum. Methods B, 1993,80/81,1080-1083. Scheibe, H. J., Drescher, D. and Alers, P., Raman characterization of amorphous carbon films; Fresenius J. Anal. Chem., 1995,353,695-698.

Biomaterials 1997. Vol. 18 No. 22

improved cell adhesion to ion beam-irradiated polymer surfaces

Documents