13842
Langmuir 2008, 24, 13842-13849
Immuno-Atomic Force Microscopy Characterization of Adsorbed Fibronectin Jane W. C. Cheung and Gilbert C. Walker* Department of Chemistry, UniVersity of Toronto, Toronto, Ontario, Canada M5S 3H6 ReceiVed July 29, 2008. ReVised Manuscript ReceiVed October 30, 2008 The fibronectin (Fn) binding conformation on mica and ultraflat poly(D,L-lactide-co-glycolide) (UPLGA) was characterized using atomic force microscopy (AFM). AFM topographic images showed that Fn was in an extended form on mica and in a compact structure on UPLGA. With immuno-AFM, an antibody (Abhep) was used to characterize the Fn binding conformation. When Fn opens its binding site for an antibody upon adsorption to a surface, the resulting Fn-antibody complex creates an additional peak in the sample’s height distribution. Immuno-AFM uses this change to detect antigen-antibody binding. In this letter, height histograms (distributions) were generated using the mean true height of molecules, which was measured by examining the histogram for each individual molecule and subtracting the mica background. Mean true height values were obtained from the histograms and showed that Fn and Abhep formed complexes on mica, signifying that one of the heparin binding sites on Fn was open when Fn was adsorbed to mica. The mean true height of the Fn-antibody complex from the histogram is greater than expected, suggesting that the antibody had pulled the extended “arms” of Fn together and caused an Fn conformation change upon binding. The height histograms can illustrate the Fn binding conformation and other antigen-antibody binding.
Introduction Atherosclerosis is one of the major causes of mortality in the Western world.1 It is a chronic inflammatory disease that results in occlusion in the vasculature and leads to heart failure, stroke, and death. Atherosclerosis can be treated surgically using vascular bypass conduits or grafts that are either synthetic or natural.2,3 Unfortunately, small-diameter synthetic grafts, which are required for coronary arteries, have high failure rates because of immediate coverage by a film of proteins.4,5 The proteins on this film may undergo conformational change and trigger cellular responses to the implant, which often lead to complications such as inflammation and thrombogenesis.5,6 These adverse effects may be minimized and controlled by coating the artificial graft material with a layer of endothelial cells (EC). EC will thrive via cell attachment to the implant surface, which is mediated by a specific adhesion protein such as fibronectin (Fn). However, the availability of surface-bound Fn for EC interaction depends on its conformation on the surface, which relies on the characteristics of the solid surface.5,7 Hence it is important to understand how different surfaces will modify the Fn conformation and how this will enhance or diminish EC interaction with the surface. This letter investigates the Fn binding conformation on mica using antibodies, AFM, and immuno-AFM. The conformation of Fn on ultraflat poly(lactide-co-glycolide) (UPLGA), a biomaterial, is also examined using AFM. Fibronectin (Fn) is a key component of the extracellular matrix7-10 and blood plasma.11 It is a key regulator of blood vessel growth,12 and it is involved in many biological re* Corresponding authors. E-mail:
[email protected], gwalker@ chem.utoronto.ca. (1) World Health Organization. The World Health Report ; 2003; Chapter 6. (2) Barner, H. B. Ann. Thoracic Surgery 2008, 85, 1473–1482. (3) Anamelechi, C. C.; Truskey, G. A.; Reichert, W. M. Biomaterials 2005, 26, 6887–6896. (4) Wilson, C. J.; Clegg, R. E.; Leavesley, D. I.; Pearcy, M. J. Tissue Eng. 2005, 11, 1–18. (5) Antia, M.; Islas, L. D.; Boness, D. A.; Baneyx, G.; Vogel, V. Biomaterials 2006, 27, 679–690. (6) Bujan, J.; Garcia-Honduvilla, N.; Bellon, J. M. Biotechnol. Appl. Biochem. 2004, 39, 17–27. (7) Osaki, T.; Renner, L.; Herklotz, M.; Werner, C. J. Phys. Chem. B 2006, 110, 12119–12124.
sponses.7,11,13 Generally, it is a glycoprotein that is made of two similar polypeptide chains, which are linked by disulfide bonds near the carboxyl terminus (Figure 1).5,7,10,11,14,15 Fn has a molecular weight of 440 kDa7,15 and a backbone contour length of 120 nm.10,16 It contains three types of homologous repeats or modules that gather in groups to form functional domains that bind to various macromolecules such as heparin and collagen.7,8,14 Most importantly, Fn contains a cell binding domain that binds to the integrin receptor on the cell surface, resulting in cell adhesion.14,15 The cell binds to Fn via the RGD (Arg-Gly-Asp) sequence as well as the PHSRN (Pro-His-Ser-Arg-Asn) sequence.5,8,11-14,17 Another binding domain of Fn that should be noticed is the heparin binding domain near the carboxyl terminus or heparin binding domain II (Hep-II). It is believed to be the dominant interactive site for heparin,11,12 and studies have shown that it has a synergistic effect on cell proliferation8 as well as an angiogenic effect.18 The availability and potency of the aforementioned binding domains change depending on the conformations of Fn on the surface, which are in turn affected by surface wettability, roughness, surface charges, and other factors.10,19 Thus, it is important to study in detail how Fn changes conformation on various types of surfaces and the effects on cell adhesion. In his review, Horbett stated that, on a relative scale, Fn from a single (8) Kim, J.-H.; Park, S.-O.; Jang, H.-J.; Jang, J.-H. Biotechnol. Lett. 2006, 28, 1409–1413. (9) Smith, M.; Gourdon, D.; Little, W. C.; Kubow, K. E.; Eguiluz, R. A.; Luna-Morris, S.; Vogel, V. PLoS Biol. 2007, 5, 2243–2254. (10) Baujard-Lamotte, L.; Noinville, S.; Goubard, F.; Marque, P.; Pauthe, E. Colloids Surf., B 2008, 63, 129–137. (11) Lin, H.; Lal, R.; Clegg, D. O. Biochemistry 2000, 39, 3192–3196. (12) Wijelath, E. S.; Rahman, S.; Namekata, M.; Murray, J.; Nishimura, T.; Mostafavi-Pour, Z.; Patel, Y.; Suda, Y.; Humphries, M. J.; Sobel, M. Circ. Res. 2006, 99, 853–860. (13) Barkalow, F. J. B.; Schwarzbauer, J. E. J. Biol. Chem. 1991, 266, 7812– 7818. (14) Mosher, D. F. Fibronectin; Academic Press: San Diego, CA, 1989. (15) Potts, J. R.; Campbell, I. D. Curr. Opin. Cell Biol. 1994, 6, 648–655. (16) Bergkvist, M.; Carlsson, J.; Oscarsson, S. J. Biomed. Mater. Res. A 2003, 64, 349–356. (17) Hersel, U.; Dahmen, C.; Kessler, H. Biomaterials 2003, 24, 4385–4415. (18) Viji, R. I.; Sameer Kumar, V. B.; Kiran, M. S.; Sudhakaran, P. R. Int. J. Biochem. Cell Biol. 2008, 40, 215–226. (19) Horbett, T. A. Colloids Surf., B 1994, 2, 225–240.
10.1021/la802452v CCC: $40.75 2008 American Chemical Society Published on Web 11/12/2008
Letters
Langmuir, Vol. 24, No. 24, 2008 13843
Figure 1. Schematic illustration of one of the chains in Fn.15 Fn contains three modules or repeats: FnI (type I), FnII (type II), and FnIII (type III). One of the heparin binding domains is located near the carboxyl terminus of Fn, which is also near the cell binding domain.
protein solution had 0% adsorption on polyHEMA, 100% on tissue-cultured polystyrene, and 174% on collagen-coated glass.19 Despite the surface chemistry, the adsorption behavior also depends on the nature of the Fn mixture (i.e., single protein solution vs protein mixture). Horbett also stated that by adsorbing to the surfaces Fn was somehow “activated” to promote cell adhesion but on hydrophobic surfaces Fn did not mediate cell spreading. In studying the influence of substrate surface properties on adsorbed Fn conformation, using TOF-SIMS Lhoest et al. showed that Fn underwent significant denaturation on normal polystyrene, which is hydrophobic, but on oxidized polystyrene the denaturation of Fn was not as significant as that seen on untreated polystyrene.20 Hence, surface wettability is not the only factor that affects Fn conformation. This fact was also proven by Grinnell et al.21 and Garcia et al.22 in their studies of Fn adsorption and subsequent cell adhesion on hydrophobic and hydrophilic surfaces. Grinnell et al. showed that Fn had different conformations on hydrophilic tissue culture dishes and on hydrophobic bacteriological dishes. They stated that Fn adsorption and anti-Fn binding depended on serum concentration as well. In the study by Garcia et al., self-assembled monolayers (SAMs) of mixed surface chemistry were used, and they showed that even though the hydrophilicity of the substrate was increased the hydrophilic component (oligo(ethylene glycol)) of the SAMs decreased the adsorption strength of Fn. Bergkvist et al. found that Fn was extended and in a strandlike structure on hydrophilic silica, partially extended on mica, and in a compact structure on a hydrophobic methylated surface.16 They stated that Fn was more likely to be found in the extended form on mica, which might enhance cell adhesion. Yu et al. used fluorescently labeled antibodies to show that Fn had an increased exposure of its 30 kDa gelatin binding and 65 kDa cell binding domains on heparin-coated poly(vinyl chloride).23 BaujardLamotte et al. showed with FTIR-ATR spectroscopy that Fn molecules kept their native conformation on hydrophilic silica but underwent strong unfolding on hydrophobic polystyrene.10 In addition, Miller et al. used AFM to show that the Fn-adsorbed PLGA nanoscale patterned surface improved vascular cell adhesion.24 Among these studies, no groups have extensively analyzed the change in the sample surface height distribution induced by antibody binding to detect modulations in Fn binding conformation on different surfaces. In this study, immuno-AFM (antibodies and height distribution) was used to characterize the Fn binding conformation on mica (hydrophilic). Imaging individual molecules directly has potential (20) Lhoest, J.-B.; Detrait, E.; van den Bosch de Aguilar, P.; Bertrand, P. J. Biomed. Mater. Res. 1998, 41, 95–103. (21) Grinnell, F.; Feld, M. K. J. Biol. Chem. 1982, 257, 4888–4893. (22) Capadona, J. R.; Collard, D. M.; Garcia, A. J. Langmuir 2003, 19, 1847– 1852. (23) Yu, J.-L.; Johansson, S.; Ljungh, A. Biomaterials 1997, 18, 421–427. (24) Miller, D. C.; Haberstroh, K. M.; Webster, T. J. J. Biomed. Mater. Res. Part A 2007, 81, 678–684.
advantages over ensemble methods such as ToF-SIMS.20 The antibody that targets heparin binding domain II of Fn was used to detect Fn-antibody complexes. Fn-antibody complexes, such as those that are formed at Hep-II, were identified from the topography of a surface first exposed to Fn and then to antibody solutions.
Experimental Details General Procedures. All glassware was soaked overnight in a base bath containing 375 g of KOH (Sigma-Aldrich), 375 mL of tap water, and 2.5 L of isopropanol (Caledon Laboratories Ltd.). Unless specified, neat water (18 MΩ · cm resistivity) was used in all experiments, and it was filtered three times with sterile 0.2 µm cellulose acetate membrane (VWR) before use. All solutions were stored at 4 °C until use. Substrate Preparation. Circular disks of muscovite mica with 9.9 mm diameter (Ted Pella) were used. Mica was freshly cleaved before each experiment to obtain a clean, flat surface that allowed small, detailed features to be seen in an AFM topographic image. Poly(D,L-lactide-co-glycolide) (PLGA, 50% polylactide and 50% polyglycolide) was purchased from Sigma-Aldrich. For increased sensitivity to small features in AFM topographic images, ultraflat PLGA (UPLGA) was desired (rms roughness
