8-1
CHAPTER 8.TOWARD NON-FOULING BRUSHES:POLYMERIC BETAINES
8-2
After Zwaal et al. demonstrated the non-thrombogenic nature of the phosphorylcholine (PC) moiety, an important component of cell membranes[4], various attempts to prepare low-fouling surfaces were carried out. The use of PC for modifying surfaces in order to improve their biocompatibility was soon recognised by Nakabayashi[5]
and Chapman[6, 7] preparing various non-thrombogenic materials.[8-10] At the start of this thesis some work had shown the resistance of polymers of methacryloyloxyethyl phosphorylcholine (phosphorycholine methacrylate, PCMA) and N-(3-sulfopropyl)-N-(methacryloxyethyl)-N,N-dimethylammonium (sulfobetaine methacrylate, SBMA) to the adsorption of single protein solutions.[11-14] Importantly, even if some authors claimed those surfaces as ‘super-low fouling’ no studies supporting the resistance to blood plasma fouling had been presented.[15]
The research presented in this chapter aims at screening the resistance from undiluted blood plasma on poly(betaine) brushes prepared by surface initiated polymerisations and to critically compare the achieved results with the state of the art in literature. The latter is fundamental in a growing field as is the case of antifouling materials, where typically abuses of terminology (ultra-low, super-low, non-fouling) are found.
8.1 Selection of monomers
Five zwitterionic betaines were utilised for preparation of homo- and block-copolymer brushes (Figure C8-1). Phosphorylcholine methacrylate (PCMA) and sulfobetaine methacrylate (SBMA) are commercially available monomers which were described as resistant to the fouling from single protein solutions. Carboxybetaines, however, were not commercially available and were synthesised as described in chapter 3.
Many different carboxybetaine monomers could be designed; however, for the work presented in this chapter, 3 types were selected as being the best candidates. A key parameter for the selection of the carboxybetaine monomers is the separation between the positive and negative charges. An ethylene spacer was selected for the all carboxybetaine monomers used in this study. It was previously demonstrated that that spacer rendered more hydrophilic brushes which were able to fully resist the fouling from fibrinogen.[16] A longer spacer, however, increased the hydrophobicity of the surface [17] and increased the pKa, therefore becoming more prone to protonation at low pH and fouling. A shorter spacer, methylene, was also shown to render brushes more hydrophobic[18] with a concomitant increase in the fouling.[16] Three types of polymerisable groups (methacrylate:
CBMA-2, methacrylamide: CBMAA-3 and acrylamide: CBAA-3) were combined with the carboxybetaine moiety and used as building block for polymer brushes. The different brushes prepared allowed to infer the influence of the backbone on the resistance to fouling.
8-3
Figure C8-1. Scheme of betaines utilised for the preparation of antifouling polymer brushes presented in this chapter.
8.2 Synthesis of poly(betaine) brushes
All the brushes utilised in the studies presented hereafter were polymerised via surface initiated ATRP. The polymerisation conditions are described in Appendices I, II and IV.
Polymerisations of PCMA and SBMA are straightforward as for most ATRP polymerisations of water soluble methacrylates in aqueous media.[11, 19, 20] ATRP polymerisation of CBAA-3 and CBMAA-3 is, however, a very challenging issue.
ATRP polymerisation of (meth)acrylamide monomers, especially in protic solvents has proved to be very difficult. The poor control for these monomers might be attributed to:
i) inactivation of catalyst by complexation with the polymer ii) strong bond between (meth)acrylamide and halogen
iii) nucleophilic displacement of the terminal halide by a cyclisation of the last to monomer units in the dormant chain
iv) complexation with some of the Cu(II) species leading to a poor electron transfer and termination by reaction with water.[21-23]
Therefore a careful selection of conditions is necessary if some polymerisation is to be achieved. The rationale for the selection of polymerisation conditions used in this work was:
Carboxybetaine methacrylamide (CBMAA-3)
NH N O
O O
NH N O
O O
Carboxybetaine acrylamide (CBAA-3)
O N O
O O
Carboxybetaine methacrylate (CBMA-2)
O O P
O
O O
O
N O N SO3
O
Phosphorylcholine methacrylate (PCMA) Sulfobetaine methacrylate (SBMA)
CHAPTER 8.TOWARD NON-FOULING BRUSHES:POLYMERIC BETAINES
8-4
a) solvent: highly polar due to solubility of monomers
b) the temperature is restricted by thiolate-gold bonds of initiator (maximum 50 ºC)
c) multidentate chelating agents, preferentially cyclic must be used to avoid competitive complexation of copper by the polymer
d) ligands resulting in highly electrochemically active complexes are necessary
Figure C8-2. Typical polymerisation kinetics of CBAA-3.
Over 400 combinations of solvents, ligands, halogens, halogen exchanged processes dilution of initiators, etc were assessed. Some of these conditions can be found in Appendix IV. The degree of control achieved was very poor in most cases even if the most advanced techniques were used. Figure C8-2 depicts a typical kinetics of CBAA-3 in methanol/water using as catalytic system 1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetradecane (Me4Cyclam), CuCl and CuBr2 and LiCl as an additional source of halogen to aid in achieving control.
Even in the optimised conditions, only poor control of the ATRP process was achieved. However, by varying the polymerisation conditions it was possible to prepare films with the targeted thickness (between 7 - 240 nm) with high reproducibility (Appendix IV). This allowed us to successfully prepare the required thickness for each application with minimal adjustments.
A note should be made about recent publications claiming control over surface initiated polymerisations of carboxybetaines.[24-27] None of them have shown any evidence of the control over the surface polymerisation. In Jiang et al. works the polymerisations were carried out using bipyridine and a ratio of monomer to Cu(I) of 6:1 while no deactivating species were added. Bipyridine has a very low complexation constant which
8-5
can hardly compete with complexation with the polymer which is one of the reasons of lack of control for acrylamides, in addition to the fact that is one of the least active catalysts for ATRP.[28, 29] The extremely large amount of copper is also not compatible with a controlled ATRP and probably led to disproportionation of Cu(I) generating some Cu(0) which can initiate a single electron transfer mechanism (in uncontrolled fashion).
Unsworth et al. work utilised nitroxide-mediated polymerisation with sacrificial free initiator. Although some degree of control was shown in solution, no evolution of the brush thickness with polymerisation time was reported.[24] In addition, the surfaces prepared by Unsworth et al. were not challenged with blood plasma.
The growth of poly(CBAA-3) brushes by SET polymerisation was also explored in this thesis. A Cu sheet was placed facing a gold coated chip with a SAM of ω-mercaptoundecyl bromoisobutyrate. The two plates were separated by a 300-μm-thick PDMS sealing placed in the lateral borders. This setup was placed in reactors (Figure C8-3) to which a degassed solution of tris[2-(dimethylamino)ethyl]amine (Me6TREN) and CBAA-3 was added. Polymerisation was allowed to proceed for 2 h at room temperature.
Figure C8-3. Scheme of the setup used for SET polymerisation of CBAA-3.
Samples collected at various polymerisation times yield homogeneous (AFM) brushes with thickness ranging between 100 and 200 nm. The high thickness obtained by this approach precludes its use for biosensor applications, however, the system is a very valuable tool to grow polymer brushes from the surface of materials meant for applications sensitive to traces of copper salts.
8.3 Characterisation of the brushes
Seven different zwitterionic brushes were prepared by surface initiated ATRP. The selected thickness was between 18 to 20 nm.
Figure C8-4 depicts the FTIR-GASR spectra of the zwitterionic polymer brushes.
Spectra (A) poly(SBMA) and (B) poly(CBMA-2) show peaks of ester carbonyl at 1725 cm-1 and (A) shows sulphur group peaks at 1213 and 1038 cm-1. Spectrum (B) also shows
Cu sheet Au + Initiator
PDMS separator
CHAPTER 8.TOWARD NON-FOULING BRUSHES:POLYMERIC BETAINES
8-6
a prominent band of ionised carboxyl at 1609 cm-1 (COO- asymmetric stretching) and its sym. stretching at 1365 cm-1. Spectrum (C), a block copolymer of the previous ones presents the characteristic features of both. The spectra of poly(CBAA-3) (D) and poly(CBMAA-3) (E) show amide bands at 1664 cm-1 and 1558 cm-1, the dominant band of ionised carboxyl at 1609 cm-1 and its symmetric stretching at 1370 cm-1. Spectrum of poly(HOEGMA-b-CBAA-3) shows similar features as poly(CBAA-3), vibrations of carbonyl ester at 1729 cm-1, a peak of C-O-C stretching at 1157 cm-1 and a group of bands between 1300 to 900 cm-1 typical of methacrylate polymers.
Figure C8-4. FTIR-GASR spectra of polymer brushes prepared (A) poly(SBMA), (B) poly(CBMA-2), (C) poly(SBMA-b-CBMA-2), (D) poly(CBMAA-3), (E) poly(CBAA-3) and (F) poly(HOEGMA-b-CBAA-3).
8-7
8.4 Fouling studies
Polymer brushes with thickness between 18 to 20 nm were used for all the fouling studies. Block copolymers were prepared by 2-step polymerisation taking advantage of the pseudo-living nature of ATRP (Appendix VII). The prepared surfaces were challenged with undiluted pooled human blood plasma and solutions of human serum albumin (HSA, 5 mg·mL-1), fibrinogen (Fbg, 1 mg·mL-1), and immunoglobulin G (IgG, 8 mg·mL-1). The fouling was monitored by surface plasmon resonance.
Table C8-1. Fouling from single protein solutions of zwitterionic polymer bruhes Water contact angle (º) Fouling (pg·mm-2)
Surface Advancing Receding IgG Fbg HSA Lys
Au 30.0 ± 2.7 9.0 ± 1.4 3123±100 3210±140 1265±82 1560±45
PCMA 18±0.4 7.0 ± 0.4 135±50 75±20 0 0
SBMA 20.0±2.0 5.2 ± 1.2 154±60 0 0 0
CBMA-2 34.1 ± 2.4 11.2 ± 4.1 0 0 0 0
SBMA-b-CBMA-2 41.9 ± 1.1 5.3 ± 2.6 135 ± 45 180±5 0 138± 43
CBMAA-3 30.2 ± 1.2 10.2 ± 2.1 0 0 0 0
CBAA-3 23.2±0.8 8.1±1.3 0 0 0 0
HOEGMA-b-CBAA-3 23.4 ± 1.2 8.0 ± 1.1 0 0 0 0
A value of zero is assigned to values below the limit of detection 0.3 pg·cm-2.
Table C8-1 presents the fouling from single protein solutions on the poly(betaine)s prepared. All the brushes fully prevented the fouling from HSA and importantly reduced the fouling from Fbg, IgG and Lys. The brushes based on homopolymers of carboxybetaines totally prevented the fouling from all the solutions.
Undiluted blood plasma posed a more serious challenge (Figure C8-5). Remarkably, being one of the most hydrophilic brushes, with water contact angle of θadv = 18 and θrec = 7,[12, 20] poly(PCMA) brushes presented a fouling from blood plasma similar to that of hydrophobic surfaces such as gold. Similar behaviour was observed on poly(SBMA) challenged with blood plasma. Interestingly, both polymer brushes have been termed as
‘non-fouling’ or ‘super-low fouling’ in previous publications.[15, 30] In Brash’s studies, a fouling of 50 pg·mm-2 of radiolabelled fibrinogen in blood plasma was used as proof of the non-fouling properties.[12] From our study, it is clear that the resistance to fibrinogen cannot be used as proof of resistance to fouling from blood plasma. Poly(SBMA) was shown to importantly decrease the fouling from blood plasma only for thickness exceeding 60 nm. Such thick brushes are not useful for SPR biosensors due to a very important decrease in sensitivity.
CHAPTER 8.TOWARD NON-FOULING BRUSHES:POLYMERIC BETAINES
8-8
Figure C8-5. Blood plasma fouling on poly(betaine)s. A circle indicates a value of fouling below the limit of detection of the SPR, 0.3 pg·mm-2.
Polymer brushes based on the carboxybetaine moiety showed the best resistance to fouling from blood plasma even though they were not as wettable as poly(SBMA) or poly(PCMA). Poly(CBMAA-3) reduced the fouling from blood plasma by over 95 % while no fouling was detected on poly(CBMA-2) and poly(CBAA-3). These results were confirmed by FTIR-GAATR and ellipsometry. Several attempts of various groups presenting poly(carboxybetaine)s as low fouling coatings can be found.[24, 31, 32] However, only poly(CBMA-2) and poly(CBAA-3) prepared in this work and by Jiang’s group have achieved full resistance to fouling from blood plasma.
Poly(HOEGMA-b-CBAA-3) exhibited surface properties similar to poly(CBAA-3) (Table C8-1) as well as full resistance to fouling from blood plasma. The first block of poly(HOEGMA) can be used to finely tune the thickness by means of its very well controlled ATRP polymerisation. A subsequent step growing poly(CBAA-3) ensured total resistance to blood plasma fouling.
It is important to note that no clear correlation between non-fouling properties and wettability was found. Although it seems clear that a surface must not be hydrophobic in order to rule out adsorption mediated hydrophobic interactions, a very high wettability does not necessary lead to superior resistance to fouling as suggested by Jiang’s group and others.[33, 34]
8-9
8.5 Summary
Poly(zwitterionic) brushes bearing side chains with phosphorycholine, sulfobetaine or carboxybetaine moiteies were preparared. All brushes markedly reduced or prevented the fouling from single protein solutions. Poly(carboxybetaine)s showed excellent reduction to blood plasma fouling. In particular poly(CBMA-2) and poly(CBAA-3) synthesised by the author and by Jiang’s group are the only surfaces fully resistant to fouling from blood plasma.
Large deposits from undiluted human blood plasma were observed on highly wettable poly(PCMA) and poly(SBMA) brushes, which is in clear contrast to the idea that wettability was the main force preventing the fouling. In particular, these surfaces were more fouling than all the other simpler and less wettable surfaces presented in previous chapters.
The lack of correlation between wettability of the surface and fouling resistance suggests that more complex mechanisms are at play and that the screening for non-fouling surfaces should be extended to new polymer brushes.
8.6 References
[1] Q. Shao, Y. He, A. D. White, S. Jiang, J. Phys. Chem. B 2010, 114, 16625.
[2] J. C. Hower, M. T. Bernards, S. Chen, H.-K. Tsao, Y.-J. Sheng, S. Jiang, The Journal of Physical Chemistry B 2008, 113, 197.
[3] S. Kudaibergenov, W. Jaeger, A. Laschewsky, "Polymeric betaines: Synthesis, characterization, and application", in Supramolecular Polymers Polymeric Betains Oligomers, Springer-Verlag Berlin, Berlin, 2006, p. 157.
[4] R. F. A. Zwaal, P. Comfurius, L. L. M. Vandeenen, Nature 1977, 268, 358.
[5] Y. Kadoma, N. Nakabayashi, M. E., J. Yamauchi, Kobunshi Ronbushu 1978, 35, 423.
[6] B. Hall, R. le E. Bird, M. Kojima, D. Chapman, Biomaterials 1989, 10, 219.
[7] J. A. Hayward, D. Chapman, Biomaterials 1984, 5, 135.
[8] K. Ishihara, T. Tsuji, T. Kurosaki, N. Nakabayashi, J. Biomed. Mater. Res. 1994, 28, 225.
[9] K. Ishihara, H. Oshida, Y. Endo, A. Watanabe, T. Ueda, N. Nakabayashi, J. Biomed.
Mater. Res. 1993, 27, 1309.
[10] A. L. Lewis, J. D. Furze, S. Small, J. D. Robertson, B. J. Higgins, S. Taylor, D. R.
Ricci, J. Biomed. Mater. Res. 2002, 63, 699.
[11] I. Y. Ma, E. J. Lobb, N. C. Billingham, S. P. Armes, A. L. Lewis, A. W. Lloyd, J.
Salvage, Macromolecules 2002, 35, 9306.
[12] W. Feng, X. Gao, G. McClung, S. Zhu, K. Ishihara, J. L. Brash, Acta Biomater. 2011, 7, 3692.
[13] H. Kitano, T. Mori, Y. Takeuchi, S. Tada, M. Gemmei-Ide, Y. Yokoyama, M. Tanaka, Macromol. Biosci. 2005, 5, 314.
[14] S. L. West, J. P. Salvage, E. J. Lobb, S. P. Armes, N. C. Billingham, A. L. Lewis, G.
W. Hanlon, A. W. Lloyd, Biomaterials, 25, 1195.
CHAPTER 8.TOWARD NON-FOULING BRUSHES:POLYMERIC BETAINES
8-10
[15] Z. Zhang, S. Chen, Y. Chang, S. Jiang, The Journal of Physical Chemistry B 2006, 110, 10799.
[16] Z. Zhang, H. Vaisocherová, G. Cheng, W. Yang, H. Xue, S. Jiang, Biomacromolecules 2008, 9, 2686.
[17] R. G. Laughlin, Langmuir 1991, 7, 842.
[18] J. G. Weers, J. F. Rathman, F. U. Axe, C. A. Crichlow, L. D. Foland, D. R. Scheuing, R. J. Wiersema, A. G. Zielske, Langmuir 1991, 7, 854.
[19] A. L. Lewis, Colloids Surf. B. Biointerfaces 2000, 18, 261.
[20] C. Rodriguez-Emmenegger, E. Brynda, T. Riedel, Z. Sedlakova, M. Houska, A. B.
Alles, Langmuir 2009, 25, 6328.
[21] S. K. Jewrajka, B. M. Mandal, Macromolecules 2003, 36, 311.
[22] J. T. Rademacher, M. Baum, M. E. Pallack, W. J. Brittain, W. J. Simonsick, Macromolecules 1999, 33, 284.
[23] M. Teodorescu, K. Matyjaszewski, Macromolecules 1999, 32, 4826.
[24] S. Abraham, L. D. Unsworth, J. Polym. Sci., Part A: Polym. Chem. 2011, 49, 1051.
[25] Z. Zhang, S. Chen, S. Jiang, Biomacromolecules 2006, 7, 3311.
[26] S. Chen, S. Jiang, Adv. Mater. 2008, 20, 335.
[27] W. Yang, H. Xue, W. Li, J. Zhang, S. Jiang, Langmuir 2009, 25, 11911.
[28] F. di Lena, K. Matyjaszewski, Prog. Polym. Sci. 2010, 35, 959.
[29] W. Tang, Y. Kwak, W. Braunecker, N. V. Tsarevsky, M. L. Coote, K. Matyjaszewski, J. Am. Chem. Soc. 2008, 130, 10702.
[30] W. Feng, S. P. Zhu, K. Ishihara, J. L. Brash, Biointerphases 2006, 1, 50.
[31] H. Kitano, H. Suzuki, T. Kondo, K. Sasaki, S. Iwanaga, M. Nakamura, K. Ohno, Y.
Saruwatari, Macromol. Biosci. 2011, n/a.
[32] K. Matsuura, K. Ohno, S. Kagaya, H. Kitano, Macromol. Chem. Phys. 2007, 208, 862.
[33] S. Jiang, Z. Cao, Adv. Mater. 2010, 22, 920.
[34] S. Chen, L. Li, C. Zhao, J. Zheng, Polymer 2010, 51, 5283.
9-1