Self-Assembled Monolayers of Alkanethiolates Presenting Tri(propylene sulfoxide) Groups Resist the Adsorption of Protein

This Communication demonstrates that self-assembled monolayers (SAMs) of alkanethiolates on gold that present tri(propylene sulfoxide) groups prevent the nonspecific adsorption of protein and subsequent attachment of cells. A common strategy used to passivate surfaces against protein adsorption is to coat them with poly(ethylene glycol) (PEG).1 It is not clear whether PEG is unique in its ability to confer resistance to adsorption, or if there are other materials with comparable (or superior) properties. SAMs of alkanethiolates on gold are a class of model organic surfaces that are well-suited to study the relationships between the structure of a surface and the adsorption of protein on the surface.2 We have previously demonstrated that SAMs presenting short oligomers of the ethylene glycol group ([-CH2CH2O-]n, n ) 2-7) effectively resist the nonspecific adsorption of protein.3,4 The goal of the present work was to design a new material that resists the adsorption of protein, but that has no counterpart in available biomaterials. In designing a new “inert” material, we sought to preserve three characteristics of the oligo(ethylene glycol) chains: (i) a hydrophilic repeating unit; (ii) a unit that can hydrogen bond with water, and that is well-solvated; (iii) an oligomer that is conformationally flexible. We chose oligomers of the propylene sulfoxide group [-CH2CH2CH2S(O)-] as candidates that shared these properties with the oligo(ethylene glycol) chains. Both materials are hydrogen bond acceptors, but not donors. SAMs presenting tri(propylene sulfoxide) groups5 are more hydrophilic than those presenting hexa(ethylene glycol) groups, as measured by advancing contact angles of water of 27° and 38°, respectively. A propylene linker was chosen in place of the ethylene linker because elimination reactions of the latter limit its stability. The biocompatibility of the propylene sulfoxide group is not known, but the parent functional groupsdimethyl sulfoxidesis more biocompatible than is ethylene glycol.6 We used surface plasmon resonance (SPR) spectroscopy4,7 to measure the adsorption of the proteins RNAse A and fibrinogen to mixed SAMs prepared from 1 and undecanethiol.8 In the SPR experiment, p-polarized light is incident on the backside of a glass slide coated with a thin layer of gold that is modified with a SAM.4,9,10 The angle of reflected light that shows a minimum in intensity (the resonance angle, θm) is related linearly to the amount of protein in the interfacial region. In the experiments described here, a buffer containing phosphate (10 mM) and sodium chloride (150 mM) (pH 7.4) was allowed to pass through the flow cell for 4 min, replaced with a solution of protein in the same buffer (1 mg/mL) for 6 min, and returned to buffer for 20 min. Both RNAse A and fibrinogen adsorbed quickly and irreversibly to hydrophobic SAMs presenting only methyl groups (o1 ) 0); a SAM prepared from only 1 (o1 ) 1.0) entirely resisted the adsorption of these two proteins (Figure 1). Even mixed SAMs containing as much as 60% undecanethiolate resisted the adsorption of both proteins; the amount of protein that adsorbed to SAMs having values of o1 < 0.4 increased with the mole fraction of undecanethiolate (Figure 1). In all cases, greater than 90% of the protein remained adsorbed irreversibly during the final wash with buffer. Figure 2 compares the adsorption of RNAse A and fibrinogen on mixed SAMs prepared from undecanethiol and either 1 or a hexa(ethylene glycol)-terminated alkanethiolate (-S(CH2)11(OCH2CH2)6OH). SAMs presenting only tri(propylene sulfoxide) groups or hexa(ethylene glycol) groups are indistinguishable in their behavior: they entirely resist the in situ adsorption of proteinseven “sticky” proteins such as fibrinogen.4 Comparing the effectiveness of these groups when mixed with methyl-terminated alkanethiolates in the SAM is complicated for four reasons: (i) the lengths of the monomeric polymethylene units are different (propyl and ethyl); (ii) the number of oligomeric units differs for the two alkanethiolates; (iii) there is an error of ∼5% in determining the value of o in the SAM; (iv) the structures of mixed SAMs comprising alkanethiolates terminated in methyl groups, and these two (1) Gombotz, W. R.; Guanhui, W.; Horbett, T. A.; Hoffman, A. S. J. Biomed. Mater. Res. 1991, 15, 1547-1562. Harris, J. M., Ed. Poly(ethylene glycol) chemistry: Biotechnical and biochemical applications; Plenum: New York, 1992. (2) Whitesides, G. M.; Gorman, C. B. In Handbook of Surface Imaging and Visualization; Hubbard, A. T., Ed.; CRC Press: Boca Raton, FL, 1995; pp 713-732. Dubois, L. H.; Nuzzo, R. G. Annu. ReV. Phys. Chem. 1992, 43, 437-463. (3) Prime, K. L.; Whitesides, G. M. Science 1991, 252, 1164-1167. Prime, K. L.; Whitesides, G. M. J. Am. Chem. Soc. 1993, 115, 1071410721. (4) Mrksich, M.; Sigal, G. B.; Whitesides, G. M. Langmuir 1995, 11, 4383-4385. (5) The synthesis of alkanethiol 1 is given in the supporting information. (6) Yu, Z. W.; Quinn, P. J. Biosci. Rep. 1994, 14 (6), 259-281. Chen, C. Y.; Chang, K. K.; Chow, N. H.; Leow, T. C.; Chou, T. C.; Lin, X. Z. Dig. Dis. Sci. 1995, 40 (2), 419-426. Igimi, H.; Asakawa, S.; Tamura, R.; Yamamoto, F.; Shimura, H. Hepato-Gastroenterology 1994, 4 (1), 65-69. (7) Raether, H. In Physics of Thin Films; Hass, G., Francombe, M., and Hoffman, R., Eds.; Academic Press: New York, 1977; pp 145-261. Liedberg, B.; Lundstrom, I; Stenberg, E. Sens. Actuators, B 1993, 11, 6372. (8) SAMs were prepared by immersing gold-coated glass slides in solutions of mixtures of the two alkanethiols in chloroform (2 mM total thiol) for 1 h. Experimental details are given in the supporting information. (9) Mrksich, M.; Grunwell, J. R.; Whitesides, G. M. J. Am. Chem. Soc. 1995, 117, 12009-12010. (10) Sigal, G. B.; Bamdad, C.; Barberis, A.; Strominger, J.; Whitesides, G. M. Anal. Chem. 1996, 68, 490-497. Figure 1. Data for the adsorption of RNase A to mixed SAMs comprising 1 and undecanethiolate. The relatiVe change in the resonance angle (θm) is plotted versus time for each mixed SAM. The region of time during which protein was present in the buffer is indicated above the plot; the rapid increase (and decrease) in signal is due to dissolved protein in the bulk solution. The curves are offset vertically for clarity. The mole fraction of 1 in each SAM (o1) was determined by XPS and is indicated on the right side of the plot. 5136 J. Am. Chem. Soc. 1996, 118, 5136-5137