- Email: [email protected]
Mechanisms of hepatocyte attachment to keratin biomaterials
Biomaterials 32 (2011) 7555e7561
Contents lists available at ScienceDirect
Biomaterials journal homepage: www.elsevier.com/locate/biomaterials
Mechanisms of hepatocyte attachment to keratin biomaterials Jillian R. Richter, Roche C. de Guzman, Mark E. Van Dyke* Wake Forest Institute for Regenerative Medicine, Wake Forest School of Medicine, Winston-Salem, NC 27157, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 May 2011 Accepted 25 June 2011 Available online 23 July 2011
Keratin biomaterials support cellular adhesion, proliferation and migration, which have led to their exploitation in a variety of biomedical applications. The mechanism of cell adhesion to keratin biomaterials, however, is poorly understood. Therefore, the goal of this work was to investigate the mechanisms by which human hair keratin-based biomaterials facilitate cellular adhesion. Hepatocytes were used as a model cell type due to the abundance of published data on cell adhesion mechanisms and their relatively copious attachment to keratin substrates. The roles of b1- and b2-integrins and the hepatic asialoglycoprotein receptor (ASGPR) in hepatocyte adhesion to keratin substrates were studied using attachment assays with and without function blocking antibodies. Blocking of the hepatic integrin subunits did not decrease hepatocyte attachment to keratin. Furthermore, adhesion to keratin did not result in the formation of focal complexes or focal adhesions, nor did it produce an upregulation of phosphorylated-focal adhesion kinase. However, inhibition of hepatic ASGPR decreased the ability of hepatocytes to attach to keratin substrates, which is indicative of the role of this glycoprotein receptor in hepatocyte binding to keratin biomaterials. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: Keratin Hepatocyte Cell adhesion Integrin Glycoprotein receptor Focal adhesion
1. Introduction Cellular interactions with the extracellular matrix (ECM) play a critical role in cell survival and function. These interactions are facilitated through specific recognition of ECM binding domains by cell surface receptors and lead to a cascade of signaling events that act to mediate cell adhesion [1], migration [2], proliferation [3], differentiation [4] and apoptosis [5]. Recently, the understanding of such biological mechanisms has become of great interest for biomaterials researchers aiming to control cellular behavior by artificially recapitulating natural cellematrix interactions. Biomaterials derived from human hair keratins have previously been shown to support cellular attachment [6], proliferation [7e10] and migration [11], suggesting that regenerated keratin substrates may mimic the physical and biochemical roles of native ECM. Keratins contain peptide binding motifs such as leucine-aspartic acid-valine (LDV) that have been suggested by other researchers to facilitate cellular adhesion [7,8]. However, the specific mechanisms responsible for cellular adhesion and subsequent bioactivity have not been thoroughly investigated. In this work, primary rat hepatocytes were used as a model cell type to study the mechanisms by which keratins promote cellular adhesion.
* Corresponding author. Tel.: þ1 336 713 7266; fax: þ1 336 713 7290. E-mail address: [email protected] (M.E. Van Dyke). 0142-9612/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2011.06.061
Hepatocytes contain a variety of integrin and non-integrin adhesion molecules that aid in their binding to ECM [12]. In normal hepatic lobules, hepatocytes are in direct contact with the perisinusoidal space which contains an abundant amount of fibronectin and small discrete bundles of type I and type IV collagens. Laminin is also present in the hepatic basement membranes and surrounds the central vein of the heptic lobule [13]. Hepatocytes readily interact with the arginine-glycineaspartic acid (RGD) binding domains of these matrix components via integrin receptors, most notably the avb1 and a1b1 integrins. In addition, rat hepatocytes express a2b1, a3b1, a5b1, b2-integrins and ICAMs although their involvement in attachment and spreading on ECM molecules has been shown to be less important [14,15]. Hepatocytes also contain non-integrin receptors such as heparan sulfate proteoglycans (i.e. syndecans) that recognize matrix fibronectin and laminin via their heparan binding domains and serve as important co-receptors for facilitating hepatocyte interactions with these matrix proteins [12,16]. Lectins are also present on the surface of hepatocytes and are responsible for mediating their adhesion to glycoproteins bearing glycan chains that lack sialic acid, termed asialoglycoproteins. Physiologically, the hepatic asialoglycoprotein receptor (ASGPR; also known as the AshwelleMorell receptor) facilitates the capture and removal of a wide range of exogenously administered and potentially deleterious glycoproteins with galactose or N-acetylgalactosamine residues at their termini. In addition, hepatic ASGPR has been found to mitigate the lethal coagulopathy of sepsis following systemic Streptococcus
7556
J.R. Richter et al. / Biomaterials 32 (2011) 7555e7561
pneumoniae infection by binding and eliminating platelets desialylated by the bacterium’s neuraminidase [17,18]. Hepatocyte interaction with their surrounding matrix is required for cell survival. Lost interactions with the ECM leads to increased Fas (CD95) production and intracellular signaling that ultimately results in detachment-induced cell death, or anoikis [19]. Hepatocyte adhesion to natural ECM-derived molecules is mediated predominantly by integrins of the b1-subunit [14]. Cells can attach to substrates in a variety of specific (i.e. receptor mediated adhesion) and non-specific (i.e. electrostatic attachment) ways. Integrin-mediated adhesion of cells to their ECM first results in the formation of focal complexes, which mature to become focal adhesions [20,21]. These macromolecular assemblies serve as anchorage sites for cells and allow for the transmission of regulatory signals that mediate cell behavior (outside-in signaling) and ECM adaptations (inside-out signaling). Focal adhesion kinase (FAK) is a non-receptor protein tyrosine kinase associated with focal adhesions via the intracellular domain of the integrin bsubunits. Upon integrin-mediated interactions with the ECM, FAK becomes autophosphorylated and activated. Activated FAK serves as a docking site for other intracellular proteins and signaling molecules that regulate cellular adhesion, growth, motility and survival [22]. Such signaling is essential for cell survival on native ECM, and its absence can lead to the activation of downstream caspases that induce irreversible cell damage and apoptosis [14]. Unlike adhesion to natural ECM, the viability and function of hepatocytes on non-integrin-recognizable surfaces is not dependent on integrin-mediated signaling. For example, hepatocyte binding to biomimetic glycopolymer surfaces through ASGPR does not lead to focal adhesion formation or FAK activation, and adhered cells exhibit a spherical as opposed to flattened morphology. Anoikis of hepatocytes on non-integrin-recognizable matrices, however, is still suppressed by a reduction in Fas production and signaling [23], which allows these materials to maintain cell survival and functionality independent of integrin signaling. The exact regulation mechanisms by which non-integrin-recognizable matrices regulate cell behavior remain unclear. However, galactose-carrying synthetic ECMs have been shown to preserve the differentiation status of mature hepatocytes in culture and maintain their physiological functions such as albumin, urea and cytochrome P450 production better than substrates that mediate adhesion via integrin-binding [24e26]. In fact, the flattened and spread morphology of integrin-bound hepatocytes is now associated with dedifferentiated cells that lack liverspecific gene expression such as CYP2C11 which encodes for cytochrome P450 [27]. Therefore, galactose-carrying synthetic ECMs are capable of maintaining a suitable milieu for hepatocyte viability and function and show great promise for applications involving liver tissue engineering [28]. The goal of this work was to investigate the mechanisms by which human hair keratin-based biomaterials facilitate hepatocyte attachment. It has previously been suggested that cells adhere to keratin biomaterial substrates through integrin-mediated signaling, and that this is due to the innate capability of the LDV motif found in certain keratins to act as a ligand for this class of receptors [7,8]. In this study we extended our own and other group’s initial examinations of cell adhesion by investigating the role of b1 and b2 integrins, downstream integrin-mediated signaling, and the involvement of ASGPR in hepatocyte attachment to keratin substrates. 2. Experimental methods 2.1. Extraction and purification of hair proteins Keratins were extracted from commercially available Chinese hair using a modified protocol from Goddard and Michaelis [29]. Proteins within the hair fibers
were first solubilized through the reduction of cystine bonds by means of a 15 h treatment with 0.5 M thioglycolic acid (TGA) titrated to pH 11.0 using sodium hydroxide. The reduction solution was retained and additional proteins were extracted from the reduced hair fibers using a 2 h treatment with 100 mM tris base solution, followed by another 2 h extraction with deionized (DI) water. All extractions took place at 37 C while vigorously shaking, and two complete extraction cycles (i.e. TGA, tris, and DI water) were completed over a 48 h period. Following the extractions, all solutions were combined and dialysis was used to obtain more purified keratin fractions. The a-keratin fraction contained high molecular weight complexes of type I and type II keratins, and the g-keratin fraction was a heterogeneous mixture of protein fragments and low molecular weight keratin associated proteins (KAPs), which do not contain any known integrinbinding motifs and were not further tested for cell adhesion in the present study. To obtain the a-fraction, the crude keratin extract was dialyzed against DI water using a 100 kDa nominal low molecular weight cutoff, tangential flow, spiral wound cartridge (Millipore, Billerica, MA) connected to a gear pump operating at a flow rate of about 1.5 L/min and a back pressure of 10 psi. This process retained only the higher molecular weight complexes of keratins and removed the lower molecular weight and fragmented proteins. The protein solution was dialyzed until five complete system washes were achieved while monitoring pH and electrical conductivity. Following dialysis, the keratin solution was shell frozen in liquid nitrogen and then lyophilized. The lyophilized protein was ground into a fine powder and stored under dry conditions at 80 C until use. 2.2. Preparation of keratin substrates for cell culture Keratin coatings (5 mg/cm2) were produced by a 2 h, 37 C incubation of tissue culture plates with aqueous keratin solutions followed by removal of excess solution and two washes with phosphate buffered saline (PBS). This resulted in an adsorbed coating of keratin. Keratin films were formed by adding 3% (w/v) keratin solutions to cultureware (5 mg/cm2) and evaporating the excess water by exposure to ambient air for an 8e12 h period at 37 C. This resulted in a thin film that presented a more three-dimensional substrate to seeded cells. Prior to cell culture work, all substrates were sterilized using a 1 Mrad dose of g-irradiation. Coatings (two dimensional) were used only for the Western blot analyses while the thin gel films (three dimensional) were used for all other experiments. 2.3. Antibodies Function blocking monoclonal antibodies (mAb) against CD29 (b1-integrin subunit) and CD18 (b2-integrin subunit) were purchased from Pharmingen (Palo Alto, CA) and AbD Serotec (Raleigh, NC). An antibody against the hepatic lectin ASPGR and blocking peptide were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). For Western blotting, antibodies for FAK (Cell Signaling Technology, Danvers, MA) and phosphorylated-FAK (Millipore, Billerica, CA) were used. Total GAPDH protein was detected with a mAb from Cell Signaling Technology. Immunocytochemical stainings were performed using mAbs for the focal adhesion protein vinculin and the F-actin binding peptide phalloidin, conjugated to fluorescein (Millipore). 2.4. Hepatocyte attachment assay Cryopreserved, primary rat hepatocytes, serum-free seeding medium and precoated 2D hepatic culture plates (HCP) were purchased from SciKon Innovation, Inc. (Chapel Hill, NC). A hepatocyte suspension of 1 106 viable cells/ml was incubated with 40 mg/ml of the either anti-CD29 or anti-CD18 or 20 mg/ml of antiASGPR for 10 min at 37 C. As a control for the ASGPR antibody, a blocking peptide was used to inhibit antibody specificity for its ligand. An excess amount of peptide was mixed with the antibody and incubated at room temperature for 30 min while gently shaking. Keratin film substrates were created as described above and were equilibrated with seeding medium for 1 h prior to seeding. HCP were used as a positive control since these surfaces contain important ECM-derived components to facilitate hepatocyte adhesion [30]. Hepatocytes were seeded at a density of 50,000 cells/cm2 and allowed to attach for 5 h. Attachment to the substrates was determined using the CyQuantÒ cell assay (Invitrogen Corporation, Palo Alto, CA). The reported results were obtained from one representative experiment (three independent experiments were performed), and data was expressed as percent attachment corresponding to the number of attached cells relative to the number of cells initially seeded and normalized to the percent attachment of unblocked attachment to the HCP positive control. 2.5. Western blotting The ability of keratins to upregulate the activation of FAK via phosphorylation was assessed after 24 h of culture on keratin and control coatings. Coatings were used for these experiments (as opposed to films) so that total cell lysate could be obtained. Hepatocytes grown on films become embedded in this three dimensional matrix, making the harvest of lysate more challenging. Protein was harvested from total cell lysates using RIPA extraction buffer, quantified by a DC Protein assay, and analyzed by Western blot. Proteins were resolved on a 4e12% SDS-PAGE gel and
J.R. Richter et al. / Biomaterials 32 (2011) 7555e7561
7557
transferred to a nitrocellulose membrane. Membranes were blocked for 2 h with 5% non-fat milk protein dissolved in 1% Tween-20 in phosphate buffered saline (PBST). Membranes were probed with antibodies to FAK (1:1000) and phosphorylated-FAK (1:1000) by overnight incubation at 4 C while shaking. Following four 5 min washes in PBST buffer, membranes were incubated for 45 min with secondary antibody (1:5000) conjugated with horseradish-peroxidase (goat anti-rabbit IgG, Sigma, St. Louis, MO). Membranes were washed four times with PBST buffer, and detection was performed using the ECL plus Western blotting detection system (Amersham Biosciences, Piscataway, NJ). Relative changes in protein expression between treatment groups were quantified by densitometry measurements using ImageJ (NIH, Bethesda, MA). All FAK and phosphorylated-FAK expression was normalized to its corresponding GAPDH internal control and the levels of expression for unseeded hepatocytes were subtracted as background. 2.6. Immunocytochemistry Focal adhesion formation and cytoskeletal rearrangement were visualized with vinculin and actin staining, respectively, on keratin films and HCP. After 24 h, culture medium was removed, and the hepatocytes were washed with 1X PBS and fixed with 4% paraformaldehyde for 20 min at room temperature. Following fixation, cells were washed twice with wash buffer (0.05% Tween-20 in 1X PBS) and permeabilized with 0.1% Triton X-100 in 1X PBS. After blocking with 1% bovine serum albumin, hepatocytes were incubated with the anti-vinculin primary antibody (1:100) for 1 h at room temperature. Cells were then washed three times with wash buffer and incubated with an Alexa FluorÒ 594-conjugated secondary antibody (1:500, antimouse IgG, Invitrogen Corporation) and phalloidin-fluorescein (1:100) for 1 h at room temperature. Cells were then washed three times with wash buffer and mounted with ProLongÒ Gold anti-fade reagent with DAPI (Invitrogen Corporation). 2.7. Statistical evaluations All data was reported as mean standard error of the mean (SEM), and statistical analyses were performed using GraphPad Software, Prism 4.0 (La Jolla, CA). A single-factor analysis of variance (ANOVA) with a Tukey’s multiple comparison post-hoc test was used to determine statistical differences between the various experimental and control groups. For the comparisons within a single group, unpaired t-tests were used.
3. Results 3.1. Integrin-mediated attachment and signaling The involvement of b1- and b2-integrins in hepatocyte attachment to keratin and control substrates was assessed using function blocking antibodies to inhibit these surface receptors. Blockage of the b2-integrins decreased hepatocyte attachment to HCP positive control, but did not affect attachment to keratin films (Fig. 1). This data suggests a role of integrins with the b2-subunit in hepatocyte binding to HCP but not keratin substrates. To further investigate the postulate that keratin biomaterials support integrin-mediated cell adhesion, we evaluated downstream, integrin-mediated signaling following hepatocyte attachment to keratin and control substrates. To this end, keratin coatings as opposed to keratin films were used to allow for the isolation of intracellular protein. Quantification of hepatocyte adhesion to keratin coatings and control substrates is shown in Fig. 2, which confirms that similar trends in cell attachment are observed for both keratin coatings and films. Hepatocyte attachment to HCP was significantly inhibited by blockage of the b1-integrin subunit, whereas blocked b1-integrin function did not decrease attachment of hepatocytes to keratin. Expression levels of FAK and phosphorylated FAK (pFAK) following hepatocyte attachment to keratin coatings and HCP were assessed by Western blotting in order to evaluate the effects of downstream integrin-mediated signaling. Results are shown in Fig. 3. Data was normalized to corresponding GAPDH levels and protein expression levels of unseeded hepatocytes were subtracted as background. Therefore, the reported data is expressed in relation to the basal expression levels of FAK and phosphorylated-FAK found in hepatocytes prior to attachment. Integrin-mediated adhesion of hepatocytes to HCP resulted in an approximate 2.0-
Fig. 1. Involvement of b2-integrin subunit in hepatocyte attachment to HCP and keratin films. Attachment of normal cells and cells with blocked b2-integrin function was quantified after 5 h and data normalized to the positive control, hepatic culture plates (HCP). Inhibition of b2-integrin function decreased hepatocyte adhesion to HCP but not keratin films. Comparisons were made for each group between the relative attachment of unblocked and blocked cells (**p < 0.001).
fold upregulation in unphosphorylated FAK and a 3.2-fold increase in phosphorylated FAK. Moreover, the upregulation of FAK and its activation were substantially inhibited by blocking the function of the b1-integrin subunit on hepatocytes prior to seeding on HCP. Unphosphorylated FAK expression following attachment to keratin-coated plates showed a slight increase over unseeded hepatocytes. However, essentially none of the FAK was activated via autophosphorylation on keratin. Immunostaining of hepatocytes on HCP and keratin films revealed distinct differences in cellular structure and focal adhesion formation after 24 h of culture (Fig. 4). Most notably, hepatocytes on HCP had a flat morphology and were much larger in size than those cells on keratin films, which remained rounded and smaller. In addition, hepatocytes on HCP formed focal adhesion contacts as evidenced by the punctate staining for vinculin, which was absent in hepatocytes attached to keratin films. 3.2. Validation of ASGPR antibody Since the ASGPR antibody used for this study is not validated as a function-blocking antibody, a peptide was used to neutralize the
Fig. 2. Involvement of the b1-integrin subunit in hepatocyte attachment to HCP and keratin coatings. Attachment of normal cells and cells with blocked b1-integrin function was quantified after 5 h and data normalized to the positive control, hepatic culture plates (HCP). Inhibition of b1-integrin function decreased hepatocyte adhesion to HCP but not keratin coatings. Comparisons were made for each group between the relative attachment of unblocked and blocked cells (þp < 0.05; **p < 0.001).
7558
J.R. Richter et al. / Biomaterials 32 (2011) 7555e7561
3.3. Non-integrin-mediated hepatocyte attachment To test the involvement of the asialoglycoprotein receptor (ASGPR) in hepatocyte binding to keratin, an antibody against hepatic ASGPR was used to block receptor function. Inhibition of hepatic ASGPR receptor resulted in decreased adhesion to both HCP and keratin films (Fig. 6). 4. Discussion
Fig. 3. Hepatocyte expression of (A) FAK and (B) phosphorylated FAK after 24 h of culture on HCP and keratin (KTN)-coated plates. As a negative control, an anti-b1integrin mAb was used to block receptor function and inhibit integrin-mediated attachment and signaling on HCPs (b1-Blocked). Uninhibited hepatocyte attachment to HCPs served as the positive control (Unblocked) and was used to compare protein expression following attachment to keratin-coated plates. All quantitative data for protein expression levels are normalized to the corresponding GAPDH level and background levels of FAK and pFAK from unseeded hepatocytes are subtracted.
antibody and restrict its ability to bind to ASGPR on the hepatocyte surface. Blockage of hepatic ASGPR with the ASGPR antibody decreased hepatocyte attachment to HCP to 45% of unblocked levels (Fig. 5). Neutralization of the antibody with the blocking peptide restored attachment to 70% of unblocked levels. The blocking peptide itself did not interfere with cellular adhesion as compared to untreated hepatocytes. Thus, the ASGPR antibody was shown to have some function-blocking capabilities.
Biomimetic materials are capable of eliciting specific cellular responses by controlling biomolecular recognition and interactions. In many cases, these biomaterials mimic the natural role of ECM and provide cues for cellematrix interactions through specific ligands integrated within the biomaterial. Such capabilities enable cellular binding, which leads to a cascade of receptor-mediated, intracellular signaling that can ultimately be used to coordinate targeted physiological responses. The ability to understand the mechanisms of cellular binding to keratin is essential for the development of biomimetic keratin materials that are capable of mediating cellular activity via surface interactions. Human hair keratins contain LDV binding motifs and several researchers have speculated on the involvement of this ligand in cellular recognition of keratin biomaterial surfaces [7,8]. Presently, the only known integrin receptor for LDV is a4b1 [31]. Our finding that hepatocytes are capable of adhering to keratin substrates despite the fact that this cell type is not known to express the a4b1-integrin, suggests that cellular recognition of keratin biomaterials can be mediated by cellematrix interactions that do not involve b1-integrins. Hepatocyte adhesion to HCP was shown to be mediated by integrins of the b1-subunit (Fig. 2), which induced intracellular signaling necessary for FAK activation (Fig. 3) and focal adhesion formation (Fig. 4). In addition, hepatocytes on HCP surfaces exhibited a flattened and spread morphology, which is indicative of integrinmediated adhesion [32]. Adherence to keratin materials, however, was not diminished by blockage of b1-integrins nor did it lead to an upregulation of phosphorylated-FAK. Likewise, blockage of the b2integrin subunit did not inhibit hepatocyte attachment to keratin substrates as it did to HCP control surfaces (Fig. 1). Given that large numbers of hepatocytes attach to keratin surfaces, these data strongly suggest the role of another class of cell membrane receptors. The hepatic asialoglycoprotein surface receptor facilitates binding of hepatocytes to glycoproteins with glucose, galactose or N-acetylgalactosamine residues at their termini [33]. For liver tissue engineering, many researchers have taken advantage of this naturally occurring hepatocyte receptor and developed receptorrecognizable materials by the incorporation of chemical modifications with galactose moieties. Through specific interactions between the biomaterial ligand and hepatic receptor, these biomimetic glycopolymers have been shown to mimic the natural liver ECM by controlling hepatocyte attachment, proliferation, migration, differentiation and functionality [34e36]. In the current study, hepatocyte adhesion to HCP control surfaces was shown to be at least partially mediated by hepatic ASGPR as evidenced by the decrease in cellular attachment following inhibition of the glycoprotein receptor (Fig. 6). Together with the previously discussed results regarding the involvement of integrin-mediated binding to HCP, these results demonstrate that the HCP coating contains ligands that promote hepatocyte binding through both integrin(i.e. b1 and b2) and non-integrin- (i.e. ASGPR) mediated mechanisms. Inhibition of hepatic ASGPR was also shown to decrease hepatocyte attachment to keratin substrates, implying that this glycoprotein receptor may also participate in hepatocyte recognition and non-integrin-mediated binding to keratin biomaterials.
J.R. Richter et al. / Biomaterials 32 (2011) 7555e7561
7559
Fig. 4. Hepatocyte morphology and focal adhesion formation. Immunostaining was performed to visualize cytoskeletal actin (green), vinculin linker proteins (red) and cell nuclei (blue). Attachment to HCP resulted in cell spreading and focal adhesion formation (white arrows), whereas hepatocytes on keratin films remained rounded and did not develop focal adhesion contacts with the surfaces as evidenced by the lack of vinculin staining. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Hepatocytes on keratin films had a similar morphology to hepatocytes on glycopolymer materials [32], showing evidence of cellecell contacts and interactions, but remaining rounded and lacking focal adhesion contacts (Fig. 4). Similar to biomimetic glycopolymer systems, keratin biomaterials may be capable of supporting hepatocyte viability by suppressing intracellular signaling that would otherwise initiate anoikis.
As mentioned previously, ligands for hepatic ASGPR include galactose, glucose or N-acetylgalactosamine residues. Therefore, the involvement of this receptor in attachment to proteinaceous, keratin materials is somewhat perplexing. To our knowledge, the presence of carbohydrates or glycosylated proteins in extracted keratins has not been studied, and so it is unclear as to whether keratin biomaterials contain glycans that may serve as specific ligands for hepatic ASGPR. However, glycation of keratins within the native hair fiber does occur [37e40], and so it is possible that keratin biomaterials derived from human hair may contain appropriate ligands to
Fig. 5. Function blocking capability of the ASPGR antibody. Hepatocyte attachment was inhibited by pre-incubation with the ASGPR antibody. Neutralization of the antibody with a blocking peptide significantly prevented the inhibitory effect and restored attachment to 70% of untreated hepatocytes on HCP (**p < 0.001 compared to the ASGPR antibody). The blocking peptide itself did not effect hepatocyte attachment relative to the untreated control.
Fig. 6. Hepatocyte attachment to HCP and keratin films following inhibition of the asialoglycoprotein receptor (ASGPR). Attachment was quantified after 5 h and data normalized to the positive control, hepatic culture plates (HCP). Blockage of the hepatic ASGPR decreased hepatocyte attachment to both HCP and keratin films. Comparisons were made for each group between the relative attachment of unblocked and blocked cells (þp < 0.05; **p < 0.001).
7560
J.R. Richter et al. / Biomaterials 32 (2011) 7555e7561
facilitate hepatocyte binding through ASPGR. In addition to glycan recognition, however, the binding avidity of ASGPR is also dependent on ligand size, stoichiometry and spatial orientation [28]. Therefore, it could also be possible that polymerized keratin materials mimic the conformational structure and size of natural ASGPR ligands to facilitate cellematrix interactions even in the absence of a galactoseterminated glycan. Further research is needed to validate these postulates and to identify the ligand for hepatic ASGPR that appears to be present in keratin biomaterials. 5. Conclusion Understanding the innate mechanisms of cellular recognition of keratin biomaterials is important for exercising control over cellular activities and functions. In this work, we have verified that hepatocyte adhesion to keratin substrates is not mediated by integrins of the b1- or b2-subtype as shown through receptor function-blocking studies, activated FAK expression analysis, and focal adhesion immunostaining. The hepatic glycoprotein receptor, ASGPR, appeared to mediate hepatocyte adhesion to keratin materials, which exhibited a rounded morphology after 24 h of culture. These initial investigations into the mechanisms of keratinehepatocyte interactions have revealed similarities to hepatocyte attachment to biomimetic glycopolymer materials that also support attachment via ASGPR. Glycopolymer surfaces preserve the differentiated state of mature hepatocytes and help maintain their ability to perform liver-specific functions. Although further research is needed to identify the keratin ligands that are recognized by hepatic ASGPR and to clearly elucidate the effect of these interactions on long term viability and function of cultured hepatocytes, this work has important implications for keratin biomaterials researchers aiming to regulate physiological events through surface-mediated control of cellular behavior. Acknowledgments The authors gratefully acknowledge the funding support provided by KeraNetics, LLC and thank Mária Bahawdory for her invaluable assistant with antibody selections. We also thank Dr. Kathryn Clausen for providing her expertise and assistance with the Western blot experiments, Dr. Pedro Baptista for his help with hepatocyte culture, and Dr. Randall McClelland from SciKon Innovation, Inc. for technical and material assistance with hepatocyte culture. Mark Van Dyke holds stock and is an officer in the company, KeraNetics LLC, who has provided partial funding for this research. Wake forest Health Sciences has a potential financial interest in KeraNetics, LLC through licensing agreements. References [1] Zamir E, Geiger B. Molecular complexity and dynamics of cell-matrix adhesions. J Cell Sci 2001;114(Pt 20):3583e90. [2] Palecek SP, Loftus JC, Ginsberg MH, Lauffenburger DA, Horwitz AF. Integrinligand binding properties govern cell migration speed through cellsubstratum adhesiveness. Nature 1997;385(6616):537e40. [3] Shibue T, Weinberg RA. Integrin beta1-focal adhesion kinase signaling directs the proliferation of metastatic cancer cells disseminated in the lungs. Proc Natl Acad Sci U S A 2009;106(25):10290e5. [4] Mooney D, Hansen L, Vacanti J, Langer R, Farmer S, Ingber D. Switching from differentiation to growth in hepatocytes: control by extracellular matrix. J Cell Physiol 1992;151(3):497e505. [5] Boudreau N, Sympson CJ, Werb Z, Bissell MJ. Suppression of ICE and apoptosis in mammary epithelial cells by extracellular matrix. Science 1995;267(5199): 891e3. [6] Yamauchi K, Maniwa M, Mori T. Cultivation of fibroblast cells on keratincoated substrata. J Biomater Sci Polym Ed 1998;9(3):259e70.
[7] Verma V, Verma P, Ray P, Ray AR. Preparation of scaffolds from human hair proteins for tissue-engineering applications. Biomed Mater 2008;3(2): 025007. [8] Sando L, Kim M, Colgrave ML, Ramshaw JA, Werkmeister JA, Elvin CM. Photochemical crosslinking of soluble wool keratins produces a mechanically stable biomaterial that supports cell adhesion and proliferation. J Biomed Mater Res A 2010;95(3):901e11. [9] Tachibana A, Furuta Y, Takeshima H, Tanabe T, Yamauchi K. Fabrication of wool keratin sponge scaffolds for long-term cell cultivation. J Biotechnol 2002; 93(2):165e70. [10] Reichl S. Films based on human hair keratin as substrates for cell culture and tissue engineering. Biomaterials 2009;30(36):6854e66. [11] Sierpinski P, Garrett J, Ma J, Apel P, Klorig D, Smith T, et al. The use of keratin biomaterials derived from human hair for the promotion of rapid regeneration of peripheral nerves. Biomaterials 2008;29(1):118e28. [12] Sudhakaran PR. Hepatocyte-matrix interactions. Proc Indian Acad Sci (Chem Sci.) 1999;111(2):331e42. [13] Martinez-Hernandez A. The hepatic extracellular matrix. I. Electron immunohistochemical studies in normal rat liver. Lab Invest 1984;51(1):57e74. [14] Pinkse GG, Voorhoeve MP, Noteborn M, Terpstra OT, Bruijn JA, De Heer E. Hepatocyte survival depends on beta1-integrin-mediated attachment of hepatocytes to hepatic extracellular matrix. Liver Int 2004;24(3):218e26. [15] Scott MJ, Billiar TR. Beta2-integrin-induced p38 MAPK activation is a key mediator in the CD14/TLR4/MD2-dependent uptake of lipopolysaccharide by hepatocytes. J Biol Chem 2008;283(43):29433e46. [16] Bernfield M, Sanderson RD. Syndecan, a developmentally regulated cell surface proteoglycan that binds extracellular matrix and growth factors. Philos Trans R Soc Lond B Biol Sci 1990;327(1239):171e86. [17] Grewal PK, Uchiyama S, Ditto D, Varki N, Le DT, Nizet V, et al. The Ashwell receptor mitigates the lethal coagulopathy of sepsis. Nat Med 2008;14(6): 648e55. [18] Grewal PK. The Ashwell-Morell receptor. Methods Enzymol 2010;479: 223e41. [19] Smets FN, Chen Y, Wang LJ, Soriano HE. Loss of cell anchorage triggers apoptosis (anoikis) in primary mouse hepatocytes. Mol Genet Metab 2002; 75(4):344e52. [20] Geiger B, Spatz JP, Bershadsky AD. Environmental sensing through focal adhesions. Nat Rev Mol Cell Biol 2009;10(1):21e33. [21] Zaidel-Bar R, Itzkovitz S, Ma’ayan A, Iyengar R, Geiger B. Functional atlas of the integrin adhesome. Nat Cell Biol 2007;9(8):858e67. [22] Zachary I. Focal adhesion kinase. Int J Biochem Cell Biol 1997;29(7):929e34. [23] Hoshiba T, Nagahara H, Cho CS, Tagawa Y, Akaike T. Primary hepatocyte survival on non-integrin-recognizable matrices without the activation of Akt signaling. Biomaterials 2007;28(6):1093e104. [24] Yin C, Ying L, Zhang PC, Zhuo RX, Kang ET, Leong KW, et al. High density of immobilized galactose ligand enhances hepatocyte attachment and function. J Biomed Mater Res A 2003;67(4):1093e104. [25] Ying L, Yin C, Zhuo RX, Leong KW, Mao HQ, Kang ET, et al. Immobilization of galactose ligands on acrylic acid graft-copolymerized poly(ethylene terephthalate) film and its application to hepatocyte culture. Biomacromolecules 2003;4(1):157e65. [26] Du Y, Han R, Ng S, Ni J, Sun W, Wohland T, et al. Identification and characterization of a novel prespheroid 3-dimensional hepatocyte monolayer on galactosylated substratum. Tissue Eng 2007;13(7):1455e68. [27] Hodgkinson CP, Wright MC, Paine AJ. Fibronectin-mediated hepatocyte shape change reprograms cytochrome P450 2C11 gene expression via an integrinsignaled induction of ribonuclease activity. Mol Pharmacol 2000;58(5): 976e81. [28] Cho CS, Seo SJ, Park IK, Kim SH, Kim TH, Hoshiba T, et al. Galactose-carrying polymers as extracellular matrices for liver tissue engineering. Biomaterials 2006;27(4):576e85. [29] Goddard DR, Michaelis L. Derivatives of keratin. J Biol Chem 1935;112(1): 361e71. [30] McClelland R, Wauthier E, Uronis J, Reid L. Gradients in the liver’s extracellular matrix chemistry from periportal to pericentral zones: influence on human hepatic progenitors. Tissue Eng Part A 2008;14(1):59e70. [31] Komoriya A, Green LJ, Mervic M, Yamada SS, Yamada KM, Humphries MJ. The minimal essential sequence for a major cell type-specific adhesion site (CS1) within the alternatively spliced type III connecting segment domain of fibronectin is leucine-aspartic acid-valine. J Biol Chem 1991;266(23): 15075e9. [32] Kim SH, Kim JH, Akaike T. Regulation of cell adhesion signaling by synthetic glycopolymer matrix in primary cultured hepatocyte. FEBS Lett 2003;553(3): 433e9. [33] Weigel PH. Rat hepatocytes bind to synthetic galactoside surfaces via a patch of asialoglycoprotein receptors. J Cell Biol 1980;87(3 Pt 1):855e61. [34] Kobayashi A, Goto M, Kobayashi K, Akaike T. Receptor-mediated regulation of differentiation and proliferation of hepatocytes by synthetic polymer model of asialoglycoprotein. J Biomater Sci Polym Ed 1994;6(4):325e42. [35] Seo SJ, Kim IY, Choi YJ, Akaike T, Cho CS. Enhanced liver functions of hepatocytes cocultured with NIH 3T3 in the alginate/galactosylated chitosan scaffold. Biomaterials 2006;27(8):1487e95. [36] Yoon JJ, Nam YS, Kim JH, Park TG. Surface immobilization of galactose onto aliphatic biodegradable polymers for hepatocyte culture. Biotechnol Bioeng 2002;78(1):1e10.
J.R. Richter et al. / Biomaterials 32 (2011) 7555e7561 [37] Oimomi M, Igaki N, Masuda S, Hata F, Maeda Y, Matsumoto S, et al. Hair protein glycation as a long-term index of blood glucose in diabetics. Diabetes Res Clin Pract 1988;5(4):305e8. [38] Oimomi M, Masuda S, Igaki N, Nakamichi T, Hata F, Matsumoto S, et al. Glycation of hair protein in the assessment of long-term control of blood glucose. Jpn J Med 1988;27(3):277e80.
7561
[39] Kobayashi K, Igimi H. Glycation index of hair for non-invasive estimation of diabetic control. Biol Pharm Bull 1996;19(4):487e90. [40] Nissimov J, Elchalal U, Bakala H, Brownlee M, Berry E, Phillip M, et al. Method for chronological recording of antigen appearance in human head-hair shafts and its use for monitoring glycation products in diabetes. J Immunol Methods 2007;320(1e2):1e17.
Contents lists available at ScienceDirect
Biomaterials journal homepage: www.elsevier.com/locate/biomaterials
Mechanisms of hepatocyte attachment to keratin biomaterials Jillian R. Richter, Roche C. de Guzman, Mark E. Van Dyke* Wake Forest Institute for Regenerative Medicine, Wake Forest School of Medicine, Winston-Salem, NC 27157, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 May 2011 Accepted 25 June 2011 Available online 23 July 2011
Keratin biomaterials support cellular adhesion, proliferation and migration, which have led to their exploitation in a variety of biomedical applications. The mechanism of cell adhesion to keratin biomaterials, however, is poorly understood. Therefore, the goal of this work was to investigate the mechanisms by which human hair keratin-based biomaterials facilitate cellular adhesion. Hepatocytes were used as a model cell type due to the abundance of published data on cell adhesion mechanisms and their relatively copious attachment to keratin substrates. The roles of b1- and b2-integrins and the hepatic asialoglycoprotein receptor (ASGPR) in hepatocyte adhesion to keratin substrates were studied using attachment assays with and without function blocking antibodies. Blocking of the hepatic integrin subunits did not decrease hepatocyte attachment to keratin. Furthermore, adhesion to keratin did not result in the formation of focal complexes or focal adhesions, nor did it produce an upregulation of phosphorylated-focal adhesion kinase. However, inhibition of hepatic ASGPR decreased the ability of hepatocytes to attach to keratin substrates, which is indicative of the role of this glycoprotein receptor in hepatocyte binding to keratin biomaterials. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: Keratin Hepatocyte Cell adhesion Integrin Glycoprotein receptor Focal adhesion
1. Introduction Cellular interactions with the extracellular matrix (ECM) play a critical role in cell survival and function. These interactions are facilitated through specific recognition of ECM binding domains by cell surface receptors and lead to a cascade of signaling events that act to mediate cell adhesion [1], migration [2], proliferation [3], differentiation [4] and apoptosis [5]. Recently, the understanding of such biological mechanisms has become of great interest for biomaterials researchers aiming to control cellular behavior by artificially recapitulating natural cellematrix interactions. Biomaterials derived from human hair keratins have previously been shown to support cellular attachment [6], proliferation [7e10] and migration [11], suggesting that regenerated keratin substrates may mimic the physical and biochemical roles of native ECM. Keratins contain peptide binding motifs such as leucine-aspartic acid-valine (LDV) that have been suggested by other researchers to facilitate cellular adhesion [7,8]. However, the specific mechanisms responsible for cellular adhesion and subsequent bioactivity have not been thoroughly investigated. In this work, primary rat hepatocytes were used as a model cell type to study the mechanisms by which keratins promote cellular adhesion.
* Corresponding author. Tel.: þ1 336 713 7266; fax: þ1 336 713 7290. E-mail address: [email protected] (M.E. Van Dyke). 0142-9612/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2011.06.061
Hepatocytes contain a variety of integrin and non-integrin adhesion molecules that aid in their binding to ECM [12]. In normal hepatic lobules, hepatocytes are in direct contact with the perisinusoidal space which contains an abundant amount of fibronectin and small discrete bundles of type I and type IV collagens. Laminin is also present in the hepatic basement membranes and surrounds the central vein of the heptic lobule [13]. Hepatocytes readily interact with the arginine-glycineaspartic acid (RGD) binding domains of these matrix components via integrin receptors, most notably the avb1 and a1b1 integrins. In addition, rat hepatocytes express a2b1, a3b1, a5b1, b2-integrins and ICAMs although their involvement in attachment and spreading on ECM molecules has been shown to be less important [14,15]. Hepatocytes also contain non-integrin receptors such as heparan sulfate proteoglycans (i.e. syndecans) that recognize matrix fibronectin and laminin via their heparan binding domains and serve as important co-receptors for facilitating hepatocyte interactions with these matrix proteins [12,16]. Lectins are also present on the surface of hepatocytes and are responsible for mediating their adhesion to glycoproteins bearing glycan chains that lack sialic acid, termed asialoglycoproteins. Physiologically, the hepatic asialoglycoprotein receptor (ASGPR; also known as the AshwelleMorell receptor) facilitates the capture and removal of a wide range of exogenously administered and potentially deleterious glycoproteins with galactose or N-acetylgalactosamine residues at their termini. In addition, hepatic ASGPR has been found to mitigate the lethal coagulopathy of sepsis following systemic Streptococcus
7556
J.R. Richter et al. / Biomaterials 32 (2011) 7555e7561
pneumoniae infection by binding and eliminating platelets desialylated by the bacterium’s neuraminidase [17,18]. Hepatocyte interaction with their surrounding matrix is required for cell survival. Lost interactions with the ECM leads to increased Fas (CD95) production and intracellular signaling that ultimately results in detachment-induced cell death, or anoikis [19]. Hepatocyte adhesion to natural ECM-derived molecules is mediated predominantly by integrins of the b1-subunit [14]. Cells can attach to substrates in a variety of specific (i.e. receptor mediated adhesion) and non-specific (i.e. electrostatic attachment) ways. Integrin-mediated adhesion of cells to their ECM first results in the formation of focal complexes, which mature to become focal adhesions [20,21]. These macromolecular assemblies serve as anchorage sites for cells and allow for the transmission of regulatory signals that mediate cell behavior (outside-in signaling) and ECM adaptations (inside-out signaling). Focal adhesion kinase (FAK) is a non-receptor protein tyrosine kinase associated with focal adhesions via the intracellular domain of the integrin bsubunits. Upon integrin-mediated interactions with the ECM, FAK becomes autophosphorylated and activated. Activated FAK serves as a docking site for other intracellular proteins and signaling molecules that regulate cellular adhesion, growth, motility and survival [22]. Such signaling is essential for cell survival on native ECM, and its absence can lead to the activation of downstream caspases that induce irreversible cell damage and apoptosis [14]. Unlike adhesion to natural ECM, the viability and function of hepatocytes on non-integrin-recognizable surfaces is not dependent on integrin-mediated signaling. For example, hepatocyte binding to biomimetic glycopolymer surfaces through ASGPR does not lead to focal adhesion formation or FAK activation, and adhered cells exhibit a spherical as opposed to flattened morphology. Anoikis of hepatocytes on non-integrin-recognizable matrices, however, is still suppressed by a reduction in Fas production and signaling [23], which allows these materials to maintain cell survival and functionality independent of integrin signaling. The exact regulation mechanisms by which non-integrin-recognizable matrices regulate cell behavior remain unclear. However, galactose-carrying synthetic ECMs have been shown to preserve the differentiation status of mature hepatocytes in culture and maintain their physiological functions such as albumin, urea and cytochrome P450 production better than substrates that mediate adhesion via integrin-binding [24e26]. In fact, the flattened and spread morphology of integrin-bound hepatocytes is now associated with dedifferentiated cells that lack liverspecific gene expression such as CYP2C11 which encodes for cytochrome P450 [27]. Therefore, galactose-carrying synthetic ECMs are capable of maintaining a suitable milieu for hepatocyte viability and function and show great promise for applications involving liver tissue engineering [28]. The goal of this work was to investigate the mechanisms by which human hair keratin-based biomaterials facilitate hepatocyte attachment. It has previously been suggested that cells adhere to keratin biomaterial substrates through integrin-mediated signaling, and that this is due to the innate capability of the LDV motif found in certain keratins to act as a ligand for this class of receptors [7,8]. In this study we extended our own and other group’s initial examinations of cell adhesion by investigating the role of b1 and b2 integrins, downstream integrin-mediated signaling, and the involvement of ASGPR in hepatocyte attachment to keratin substrates. 2. Experimental methods 2.1. Extraction and purification of hair proteins Keratins were extracted from commercially available Chinese hair using a modified protocol from Goddard and Michaelis [29]. Proteins within the hair fibers
were first solubilized through the reduction of cystine bonds by means of a 15 h treatment with 0.5 M thioglycolic acid (TGA) titrated to pH 11.0 using sodium hydroxide. The reduction solution was retained and additional proteins were extracted from the reduced hair fibers using a 2 h treatment with 100 mM tris base solution, followed by another 2 h extraction with deionized (DI) water. All extractions took place at 37 C while vigorously shaking, and two complete extraction cycles (i.e. TGA, tris, and DI water) were completed over a 48 h period. Following the extractions, all solutions were combined and dialysis was used to obtain more purified keratin fractions. The a-keratin fraction contained high molecular weight complexes of type I and type II keratins, and the g-keratin fraction was a heterogeneous mixture of protein fragments and low molecular weight keratin associated proteins (KAPs), which do not contain any known integrinbinding motifs and were not further tested for cell adhesion in the present study. To obtain the a-fraction, the crude keratin extract was dialyzed against DI water using a 100 kDa nominal low molecular weight cutoff, tangential flow, spiral wound cartridge (Millipore, Billerica, MA) connected to a gear pump operating at a flow rate of about 1.5 L/min and a back pressure of 10 psi. This process retained only the higher molecular weight complexes of keratins and removed the lower molecular weight and fragmented proteins. The protein solution was dialyzed until five complete system washes were achieved while monitoring pH and electrical conductivity. Following dialysis, the keratin solution was shell frozen in liquid nitrogen and then lyophilized. The lyophilized protein was ground into a fine powder and stored under dry conditions at 80 C until use. 2.2. Preparation of keratin substrates for cell culture Keratin coatings (5 mg/cm2) were produced by a 2 h, 37 C incubation of tissue culture plates with aqueous keratin solutions followed by removal of excess solution and two washes with phosphate buffered saline (PBS). This resulted in an adsorbed coating of keratin. Keratin films were formed by adding 3% (w/v) keratin solutions to cultureware (5 mg/cm2) and evaporating the excess water by exposure to ambient air for an 8e12 h period at 37 C. This resulted in a thin film that presented a more three-dimensional substrate to seeded cells. Prior to cell culture work, all substrates were sterilized using a 1 Mrad dose of g-irradiation. Coatings (two dimensional) were used only for the Western blot analyses while the thin gel films (three dimensional) were used for all other experiments. 2.3. Antibodies Function blocking monoclonal antibodies (mAb) against CD29 (b1-integrin subunit) and CD18 (b2-integrin subunit) were purchased from Pharmingen (Palo Alto, CA) and AbD Serotec (Raleigh, NC). An antibody against the hepatic lectin ASPGR and blocking peptide were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). For Western blotting, antibodies for FAK (Cell Signaling Technology, Danvers, MA) and phosphorylated-FAK (Millipore, Billerica, CA) were used. Total GAPDH protein was detected with a mAb from Cell Signaling Technology. Immunocytochemical stainings were performed using mAbs for the focal adhesion protein vinculin and the F-actin binding peptide phalloidin, conjugated to fluorescein (Millipore). 2.4. Hepatocyte attachment assay Cryopreserved, primary rat hepatocytes, serum-free seeding medium and precoated 2D hepatic culture plates (HCP) were purchased from SciKon Innovation, Inc. (Chapel Hill, NC). A hepatocyte suspension of 1 106 viable cells/ml was incubated with 40 mg/ml of the either anti-CD29 or anti-CD18 or 20 mg/ml of antiASGPR for 10 min at 37 C. As a control for the ASGPR antibody, a blocking peptide was used to inhibit antibody specificity for its ligand. An excess amount of peptide was mixed with the antibody and incubated at room temperature for 30 min while gently shaking. Keratin film substrates were created as described above and were equilibrated with seeding medium for 1 h prior to seeding. HCP were used as a positive control since these surfaces contain important ECM-derived components to facilitate hepatocyte adhesion [30]. Hepatocytes were seeded at a density of 50,000 cells/cm2 and allowed to attach for 5 h. Attachment to the substrates was determined using the CyQuantÒ cell assay (Invitrogen Corporation, Palo Alto, CA). The reported results were obtained from one representative experiment (three independent experiments were performed), and data was expressed as percent attachment corresponding to the number of attached cells relative to the number of cells initially seeded and normalized to the percent attachment of unblocked attachment to the HCP positive control. 2.5. Western blotting The ability of keratins to upregulate the activation of FAK via phosphorylation was assessed after 24 h of culture on keratin and control coatings. Coatings were used for these experiments (as opposed to films) so that total cell lysate could be obtained. Hepatocytes grown on films become embedded in this three dimensional matrix, making the harvest of lysate more challenging. Protein was harvested from total cell lysates using RIPA extraction buffer, quantified by a DC Protein assay, and analyzed by Western blot. Proteins were resolved on a 4e12% SDS-PAGE gel and
J.R. Richter et al. / Biomaterials 32 (2011) 7555e7561
7557
transferred to a nitrocellulose membrane. Membranes were blocked for 2 h with 5% non-fat milk protein dissolved in 1% Tween-20 in phosphate buffered saline (PBST). Membranes were probed with antibodies to FAK (1:1000) and phosphorylated-FAK (1:1000) by overnight incubation at 4 C while shaking. Following four 5 min washes in PBST buffer, membranes were incubated for 45 min with secondary antibody (1:5000) conjugated with horseradish-peroxidase (goat anti-rabbit IgG, Sigma, St. Louis, MO). Membranes were washed four times with PBST buffer, and detection was performed using the ECL plus Western blotting detection system (Amersham Biosciences, Piscataway, NJ). Relative changes in protein expression between treatment groups were quantified by densitometry measurements using ImageJ (NIH, Bethesda, MA). All FAK and phosphorylated-FAK expression was normalized to its corresponding GAPDH internal control and the levels of expression for unseeded hepatocytes were subtracted as background. 2.6. Immunocytochemistry Focal adhesion formation and cytoskeletal rearrangement were visualized with vinculin and actin staining, respectively, on keratin films and HCP. After 24 h, culture medium was removed, and the hepatocytes were washed with 1X PBS and fixed with 4% paraformaldehyde for 20 min at room temperature. Following fixation, cells were washed twice with wash buffer (0.05% Tween-20 in 1X PBS) and permeabilized with 0.1% Triton X-100 in 1X PBS. After blocking with 1% bovine serum albumin, hepatocytes were incubated with the anti-vinculin primary antibody (1:100) for 1 h at room temperature. Cells were then washed three times with wash buffer and incubated with an Alexa FluorÒ 594-conjugated secondary antibody (1:500, antimouse IgG, Invitrogen Corporation) and phalloidin-fluorescein (1:100) for 1 h at room temperature. Cells were then washed three times with wash buffer and mounted with ProLongÒ Gold anti-fade reagent with DAPI (Invitrogen Corporation). 2.7. Statistical evaluations All data was reported as mean standard error of the mean (SEM), and statistical analyses were performed using GraphPad Software, Prism 4.0 (La Jolla, CA). A single-factor analysis of variance (ANOVA) with a Tukey’s multiple comparison post-hoc test was used to determine statistical differences between the various experimental and control groups. For the comparisons within a single group, unpaired t-tests were used.
3. Results 3.1. Integrin-mediated attachment and signaling The involvement of b1- and b2-integrins in hepatocyte attachment to keratin and control substrates was assessed using function blocking antibodies to inhibit these surface receptors. Blockage of the b2-integrins decreased hepatocyte attachment to HCP positive control, but did not affect attachment to keratin films (Fig. 1). This data suggests a role of integrins with the b2-subunit in hepatocyte binding to HCP but not keratin substrates. To further investigate the postulate that keratin biomaterials support integrin-mediated cell adhesion, we evaluated downstream, integrin-mediated signaling following hepatocyte attachment to keratin and control substrates. To this end, keratin coatings as opposed to keratin films were used to allow for the isolation of intracellular protein. Quantification of hepatocyte adhesion to keratin coatings and control substrates is shown in Fig. 2, which confirms that similar trends in cell attachment are observed for both keratin coatings and films. Hepatocyte attachment to HCP was significantly inhibited by blockage of the b1-integrin subunit, whereas blocked b1-integrin function did not decrease attachment of hepatocytes to keratin. Expression levels of FAK and phosphorylated FAK (pFAK) following hepatocyte attachment to keratin coatings and HCP were assessed by Western blotting in order to evaluate the effects of downstream integrin-mediated signaling. Results are shown in Fig. 3. Data was normalized to corresponding GAPDH levels and protein expression levels of unseeded hepatocytes were subtracted as background. Therefore, the reported data is expressed in relation to the basal expression levels of FAK and phosphorylated-FAK found in hepatocytes prior to attachment. Integrin-mediated adhesion of hepatocytes to HCP resulted in an approximate 2.0-
Fig. 1. Involvement of b2-integrin subunit in hepatocyte attachment to HCP and keratin films. Attachment of normal cells and cells with blocked b2-integrin function was quantified after 5 h and data normalized to the positive control, hepatic culture plates (HCP). Inhibition of b2-integrin function decreased hepatocyte adhesion to HCP but not keratin films. Comparisons were made for each group between the relative attachment of unblocked and blocked cells (**p < 0.001).
fold upregulation in unphosphorylated FAK and a 3.2-fold increase in phosphorylated FAK. Moreover, the upregulation of FAK and its activation were substantially inhibited by blocking the function of the b1-integrin subunit on hepatocytes prior to seeding on HCP. Unphosphorylated FAK expression following attachment to keratin-coated plates showed a slight increase over unseeded hepatocytes. However, essentially none of the FAK was activated via autophosphorylation on keratin. Immunostaining of hepatocytes on HCP and keratin films revealed distinct differences in cellular structure and focal adhesion formation after 24 h of culture (Fig. 4). Most notably, hepatocytes on HCP had a flat morphology and were much larger in size than those cells on keratin films, which remained rounded and smaller. In addition, hepatocytes on HCP formed focal adhesion contacts as evidenced by the punctate staining for vinculin, which was absent in hepatocytes attached to keratin films. 3.2. Validation of ASGPR antibody Since the ASGPR antibody used for this study is not validated as a function-blocking antibody, a peptide was used to neutralize the
Fig. 2. Involvement of the b1-integrin subunit in hepatocyte attachment to HCP and keratin coatings. Attachment of normal cells and cells with blocked b1-integrin function was quantified after 5 h and data normalized to the positive control, hepatic culture plates (HCP). Inhibition of b1-integrin function decreased hepatocyte adhesion to HCP but not keratin coatings. Comparisons were made for each group between the relative attachment of unblocked and blocked cells (þp < 0.05; **p < 0.001).
7558
J.R. Richter et al. / Biomaterials 32 (2011) 7555e7561
3.3. Non-integrin-mediated hepatocyte attachment To test the involvement of the asialoglycoprotein receptor (ASGPR) in hepatocyte binding to keratin, an antibody against hepatic ASGPR was used to block receptor function. Inhibition of hepatic ASGPR receptor resulted in decreased adhesion to both HCP and keratin films (Fig. 6). 4. Discussion
Fig. 3. Hepatocyte expression of (A) FAK and (B) phosphorylated FAK after 24 h of culture on HCP and keratin (KTN)-coated plates. As a negative control, an anti-b1integrin mAb was used to block receptor function and inhibit integrin-mediated attachment and signaling on HCPs (b1-Blocked). Uninhibited hepatocyte attachment to HCPs served as the positive control (Unblocked) and was used to compare protein expression following attachment to keratin-coated plates. All quantitative data for protein expression levels are normalized to the corresponding GAPDH level and background levels of FAK and pFAK from unseeded hepatocytes are subtracted.
antibody and restrict its ability to bind to ASGPR on the hepatocyte surface. Blockage of hepatic ASGPR with the ASGPR antibody decreased hepatocyte attachment to HCP to 45% of unblocked levels (Fig. 5). Neutralization of the antibody with the blocking peptide restored attachment to 70% of unblocked levels. The blocking peptide itself did not interfere with cellular adhesion as compared to untreated hepatocytes. Thus, the ASGPR antibody was shown to have some function-blocking capabilities.
Biomimetic materials are capable of eliciting specific cellular responses by controlling biomolecular recognition and interactions. In many cases, these biomaterials mimic the natural role of ECM and provide cues for cellematrix interactions through specific ligands integrated within the biomaterial. Such capabilities enable cellular binding, which leads to a cascade of receptor-mediated, intracellular signaling that can ultimately be used to coordinate targeted physiological responses. The ability to understand the mechanisms of cellular binding to keratin is essential for the development of biomimetic keratin materials that are capable of mediating cellular activity via surface interactions. Human hair keratins contain LDV binding motifs and several researchers have speculated on the involvement of this ligand in cellular recognition of keratin biomaterial surfaces [7,8]. Presently, the only known integrin receptor for LDV is a4b1 [31]. Our finding that hepatocytes are capable of adhering to keratin substrates despite the fact that this cell type is not known to express the a4b1-integrin, suggests that cellular recognition of keratin biomaterials can be mediated by cellematrix interactions that do not involve b1-integrins. Hepatocyte adhesion to HCP was shown to be mediated by integrins of the b1-subunit (Fig. 2), which induced intracellular signaling necessary for FAK activation (Fig. 3) and focal adhesion formation (Fig. 4). In addition, hepatocytes on HCP surfaces exhibited a flattened and spread morphology, which is indicative of integrinmediated adhesion [32]. Adherence to keratin materials, however, was not diminished by blockage of b1-integrins nor did it lead to an upregulation of phosphorylated-FAK. Likewise, blockage of the b2integrin subunit did not inhibit hepatocyte attachment to keratin substrates as it did to HCP control surfaces (Fig. 1). Given that large numbers of hepatocytes attach to keratin surfaces, these data strongly suggest the role of another class of cell membrane receptors. The hepatic asialoglycoprotein surface receptor facilitates binding of hepatocytes to glycoproteins with glucose, galactose or N-acetylgalactosamine residues at their termini [33]. For liver tissue engineering, many researchers have taken advantage of this naturally occurring hepatocyte receptor and developed receptorrecognizable materials by the incorporation of chemical modifications with galactose moieties. Through specific interactions between the biomaterial ligand and hepatic receptor, these biomimetic glycopolymers have been shown to mimic the natural liver ECM by controlling hepatocyte attachment, proliferation, migration, differentiation and functionality [34e36]. In the current study, hepatocyte adhesion to HCP control surfaces was shown to be at least partially mediated by hepatic ASGPR as evidenced by the decrease in cellular attachment following inhibition of the glycoprotein receptor (Fig. 6). Together with the previously discussed results regarding the involvement of integrin-mediated binding to HCP, these results demonstrate that the HCP coating contains ligands that promote hepatocyte binding through both integrin(i.e. b1 and b2) and non-integrin- (i.e. ASGPR) mediated mechanisms. Inhibition of hepatic ASGPR was also shown to decrease hepatocyte attachment to keratin substrates, implying that this glycoprotein receptor may also participate in hepatocyte recognition and non-integrin-mediated binding to keratin biomaterials.
J.R. Richter et al. / Biomaterials 32 (2011) 7555e7561
7559
Fig. 4. Hepatocyte morphology and focal adhesion formation. Immunostaining was performed to visualize cytoskeletal actin (green), vinculin linker proteins (red) and cell nuclei (blue). Attachment to HCP resulted in cell spreading and focal adhesion formation (white arrows), whereas hepatocytes on keratin films remained rounded and did not develop focal adhesion contacts with the surfaces as evidenced by the lack of vinculin staining. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Hepatocytes on keratin films had a similar morphology to hepatocytes on glycopolymer materials [32], showing evidence of cellecell contacts and interactions, but remaining rounded and lacking focal adhesion contacts (Fig. 4). Similar to biomimetic glycopolymer systems, keratin biomaterials may be capable of supporting hepatocyte viability by suppressing intracellular signaling that would otherwise initiate anoikis.
As mentioned previously, ligands for hepatic ASGPR include galactose, glucose or N-acetylgalactosamine residues. Therefore, the involvement of this receptor in attachment to proteinaceous, keratin materials is somewhat perplexing. To our knowledge, the presence of carbohydrates or glycosylated proteins in extracted keratins has not been studied, and so it is unclear as to whether keratin biomaterials contain glycans that may serve as specific ligands for hepatic ASGPR. However, glycation of keratins within the native hair fiber does occur [37e40], and so it is possible that keratin biomaterials derived from human hair may contain appropriate ligands to
Fig. 5. Function blocking capability of the ASPGR antibody. Hepatocyte attachment was inhibited by pre-incubation with the ASGPR antibody. Neutralization of the antibody with a blocking peptide significantly prevented the inhibitory effect and restored attachment to 70% of untreated hepatocytes on HCP (**p < 0.001 compared to the ASGPR antibody). The blocking peptide itself did not effect hepatocyte attachment relative to the untreated control.
Fig. 6. Hepatocyte attachment to HCP and keratin films following inhibition of the asialoglycoprotein receptor (ASGPR). Attachment was quantified after 5 h and data normalized to the positive control, hepatic culture plates (HCP). Blockage of the hepatic ASGPR decreased hepatocyte attachment to both HCP and keratin films. Comparisons were made for each group between the relative attachment of unblocked and blocked cells (þp < 0.05; **p < 0.001).
7560
J.R. Richter et al. / Biomaterials 32 (2011) 7555e7561
facilitate hepatocyte binding through ASPGR. In addition to glycan recognition, however, the binding avidity of ASGPR is also dependent on ligand size, stoichiometry and spatial orientation [28]. Therefore, it could also be possible that polymerized keratin materials mimic the conformational structure and size of natural ASGPR ligands to facilitate cellematrix interactions even in the absence of a galactoseterminated glycan. Further research is needed to validate these postulates and to identify the ligand for hepatic ASGPR that appears to be present in keratin biomaterials. 5. Conclusion Understanding the innate mechanisms of cellular recognition of keratin biomaterials is important for exercising control over cellular activities and functions. In this work, we have verified that hepatocyte adhesion to keratin substrates is not mediated by integrins of the b1- or b2-subtype as shown through receptor function-blocking studies, activated FAK expression analysis, and focal adhesion immunostaining. The hepatic glycoprotein receptor, ASGPR, appeared to mediate hepatocyte adhesion to keratin materials, which exhibited a rounded morphology after 24 h of culture. These initial investigations into the mechanisms of keratinehepatocyte interactions have revealed similarities to hepatocyte attachment to biomimetic glycopolymer materials that also support attachment via ASGPR. Glycopolymer surfaces preserve the differentiated state of mature hepatocytes and help maintain their ability to perform liver-specific functions. Although further research is needed to identify the keratin ligands that are recognized by hepatic ASGPR and to clearly elucidate the effect of these interactions on long term viability and function of cultured hepatocytes, this work has important implications for keratin biomaterials researchers aiming to regulate physiological events through surface-mediated control of cellular behavior. Acknowledgments The authors gratefully acknowledge the funding support provided by KeraNetics, LLC and thank Mária Bahawdory for her invaluable assistant with antibody selections. We also thank Dr. Kathryn Clausen for providing her expertise and assistance with the Western blot experiments, Dr. Pedro Baptista for his help with hepatocyte culture, and Dr. Randall McClelland from SciKon Innovation, Inc. for technical and material assistance with hepatocyte culture. Mark Van Dyke holds stock and is an officer in the company, KeraNetics LLC, who has provided partial funding for this research. Wake forest Health Sciences has a potential financial interest in KeraNetics, LLC through licensing agreements. References [1] Zamir E, Geiger B. Molecular complexity and dynamics of cell-matrix adhesions. J Cell Sci 2001;114(Pt 20):3583e90. [2] Palecek SP, Loftus JC, Ginsberg MH, Lauffenburger DA, Horwitz AF. Integrinligand binding properties govern cell migration speed through cellsubstratum adhesiveness. Nature 1997;385(6616):537e40. [3] Shibue T, Weinberg RA. Integrin beta1-focal adhesion kinase signaling directs the proliferation of metastatic cancer cells disseminated in the lungs. Proc Natl Acad Sci U S A 2009;106(25):10290e5. [4] Mooney D, Hansen L, Vacanti J, Langer R, Farmer S, Ingber D. Switching from differentiation to growth in hepatocytes: control by extracellular matrix. J Cell Physiol 1992;151(3):497e505. [5] Boudreau N, Sympson CJ, Werb Z, Bissell MJ. Suppression of ICE and apoptosis in mammary epithelial cells by extracellular matrix. Science 1995;267(5199): 891e3. [6] Yamauchi K, Maniwa M, Mori T. Cultivation of fibroblast cells on keratincoated substrata. J Biomater Sci Polym Ed 1998;9(3):259e70.
[7] Verma V, Verma P, Ray P, Ray AR. Preparation of scaffolds from human hair proteins for tissue-engineering applications. Biomed Mater 2008;3(2): 025007. [8] Sando L, Kim M, Colgrave ML, Ramshaw JA, Werkmeister JA, Elvin CM. Photochemical crosslinking of soluble wool keratins produces a mechanically stable biomaterial that supports cell adhesion and proliferation. J Biomed Mater Res A 2010;95(3):901e11. [9] Tachibana A, Furuta Y, Takeshima H, Tanabe T, Yamauchi K. Fabrication of wool keratin sponge scaffolds for long-term cell cultivation. J Biotechnol 2002; 93(2):165e70. [10] Reichl S. Films based on human hair keratin as substrates for cell culture and tissue engineering. Biomaterials 2009;30(36):6854e66. [11] Sierpinski P, Garrett J, Ma J, Apel P, Klorig D, Smith T, et al. The use of keratin biomaterials derived from human hair for the promotion of rapid regeneration of peripheral nerves. Biomaterials 2008;29(1):118e28. [12] Sudhakaran PR. Hepatocyte-matrix interactions. Proc Indian Acad Sci (Chem Sci.) 1999;111(2):331e42. [13] Martinez-Hernandez A. The hepatic extracellular matrix. I. Electron immunohistochemical studies in normal rat liver. Lab Invest 1984;51(1):57e74. [14] Pinkse GG, Voorhoeve MP, Noteborn M, Terpstra OT, Bruijn JA, De Heer E. Hepatocyte survival depends on beta1-integrin-mediated attachment of hepatocytes to hepatic extracellular matrix. Liver Int 2004;24(3):218e26. [15] Scott MJ, Billiar TR. Beta2-integrin-induced p38 MAPK activation is a key mediator in the CD14/TLR4/MD2-dependent uptake of lipopolysaccharide by hepatocytes. J Biol Chem 2008;283(43):29433e46. [16] Bernfield M, Sanderson RD. Syndecan, a developmentally regulated cell surface proteoglycan that binds extracellular matrix and growth factors. Philos Trans R Soc Lond B Biol Sci 1990;327(1239):171e86. [17] Grewal PK, Uchiyama S, Ditto D, Varki N, Le DT, Nizet V, et al. The Ashwell receptor mitigates the lethal coagulopathy of sepsis. Nat Med 2008;14(6): 648e55. [18] Grewal PK. The Ashwell-Morell receptor. Methods Enzymol 2010;479: 223e41. [19] Smets FN, Chen Y, Wang LJ, Soriano HE. Loss of cell anchorage triggers apoptosis (anoikis) in primary mouse hepatocytes. Mol Genet Metab 2002; 75(4):344e52. [20] Geiger B, Spatz JP, Bershadsky AD. Environmental sensing through focal adhesions. Nat Rev Mol Cell Biol 2009;10(1):21e33. [21] Zaidel-Bar R, Itzkovitz S, Ma’ayan A, Iyengar R, Geiger B. Functional atlas of the integrin adhesome. Nat Cell Biol 2007;9(8):858e67. [22] Zachary I. Focal adhesion kinase. Int J Biochem Cell Biol 1997;29(7):929e34. [23] Hoshiba T, Nagahara H, Cho CS, Tagawa Y, Akaike T. Primary hepatocyte survival on non-integrin-recognizable matrices without the activation of Akt signaling. Biomaterials 2007;28(6):1093e104. [24] Yin C, Ying L, Zhang PC, Zhuo RX, Kang ET, Leong KW, et al. High density of immobilized galactose ligand enhances hepatocyte attachment and function. J Biomed Mater Res A 2003;67(4):1093e104. [25] Ying L, Yin C, Zhuo RX, Leong KW, Mao HQ, Kang ET, et al. Immobilization of galactose ligands on acrylic acid graft-copolymerized poly(ethylene terephthalate) film and its application to hepatocyte culture. Biomacromolecules 2003;4(1):157e65. [26] Du Y, Han R, Ng S, Ni J, Sun W, Wohland T, et al. Identification and characterization of a novel prespheroid 3-dimensional hepatocyte monolayer on galactosylated substratum. Tissue Eng 2007;13(7):1455e68. [27] Hodgkinson CP, Wright MC, Paine AJ. Fibronectin-mediated hepatocyte shape change reprograms cytochrome P450 2C11 gene expression via an integrinsignaled induction of ribonuclease activity. Mol Pharmacol 2000;58(5): 976e81. [28] Cho CS, Seo SJ, Park IK, Kim SH, Kim TH, Hoshiba T, et al. Galactose-carrying polymers as extracellular matrices for liver tissue engineering. Biomaterials 2006;27(4):576e85. [29] Goddard DR, Michaelis L. Derivatives of keratin. J Biol Chem 1935;112(1): 361e71. [30] McClelland R, Wauthier E, Uronis J, Reid L. Gradients in the liver’s extracellular matrix chemistry from periportal to pericentral zones: influence on human hepatic progenitors. Tissue Eng Part A 2008;14(1):59e70. [31] Komoriya A, Green LJ, Mervic M, Yamada SS, Yamada KM, Humphries MJ. The minimal essential sequence for a major cell type-specific adhesion site (CS1) within the alternatively spliced type III connecting segment domain of fibronectin is leucine-aspartic acid-valine. J Biol Chem 1991;266(23): 15075e9. [32] Kim SH, Kim JH, Akaike T. Regulation of cell adhesion signaling by synthetic glycopolymer matrix in primary cultured hepatocyte. FEBS Lett 2003;553(3): 433e9. [33] Weigel PH. Rat hepatocytes bind to synthetic galactoside surfaces via a patch of asialoglycoprotein receptors. J Cell Biol 1980;87(3 Pt 1):855e61. [34] Kobayashi A, Goto M, Kobayashi K, Akaike T. Receptor-mediated regulation of differentiation and proliferation of hepatocytes by synthetic polymer model of asialoglycoprotein. J Biomater Sci Polym Ed 1994;6(4):325e42. [35] Seo SJ, Kim IY, Choi YJ, Akaike T, Cho CS. Enhanced liver functions of hepatocytes cocultured with NIH 3T3 in the alginate/galactosylated chitosan scaffold. Biomaterials 2006;27(8):1487e95. [36] Yoon JJ, Nam YS, Kim JH, Park TG. Surface immobilization of galactose onto aliphatic biodegradable polymers for hepatocyte culture. Biotechnol Bioeng 2002;78(1):1e10.
J.R. Richter et al. / Biomaterials 32 (2011) 7555e7561 [37] Oimomi M, Igaki N, Masuda S, Hata F, Maeda Y, Matsumoto S, et al. Hair protein glycation as a long-term index of blood glucose in diabetics. Diabetes Res Clin Pract 1988;5(4):305e8. [38] Oimomi M, Masuda S, Igaki N, Nakamichi T, Hata F, Matsumoto S, et al. Glycation of hair protein in the assessment of long-term control of blood glucose. Jpn J Med 1988;27(3):277e80.
7561
[39] Kobayashi K, Igimi H. Glycation index of hair for non-invasive estimation of diabetic control. Biol Pharm Bull 1996;19(4):487e90. [40] Nissimov J, Elchalal U, Bakala H, Brownlee M, Berry E, Phillip M, et al. Method for chronological recording of antigen appearance in human head-hair shafts and its use for monitoring glycation products in diabetes. J Immunol Methods 2007;320(1e2):1e17.