The creation of functional microtissues within microfluidic devices requires the stabilization of cell phenotypes by adapting traditional cell culture techniques to the limited spatial dimensions in microdevices. Modification of collagen allows the layer-by-layer deposition of ultrathin collagen assemblies that can stabilize primary cells, such as hepatocytes, as microfluidic tissue models.
Cite this ArticleCopy Citation | Download Citations | Reprints and Permissions
McCarty, W. J., Prodanov, L., Bale, S. S., Bhushan, A., Jindal, R., Yarmush, M. L., Usta, O. B. Layer-by-layer Collagen Deposition in Microfluidic Devices for Microtissue Stabilization. J. Vis. Exp. (103), e53078, doi:10.3791/53078 (2015).
Translate text to:
Although microfluidics provides exquisite control of the cellular microenvironment, culturing cells within microfluidic devices can be challenging. 3D culture of cells in collagen type I gels helps to stabilize cell morphology and function, which is necessary for creating microfluidic tissue models in microdevices. Translating traditional 3D culture techniques for tissue culture plates to microfluidic devices is often difficult because of the limited channel dimensions. In this method, we describe a technique for modifying native type I collagen to generate polycationic and polyanionic collagen solutions that can be used with layer-by-layer deposition to create ultrathin collagen assemblies on top of cells cultured in microfluidic devices. These thin collagen layers stabilize cell morphology and function, as shown using primary hepatocytes as an example cell, allowing for the long term culture of microtissues in microfluidic devices.
Although microfluidics allows for the exquisite control of the cellular microenvironment, culturing cells, especially primary cells, within microfluidic devices can be challenging. Many traditional cell culture techniques have been developed to sustain and stabilize cell function when cultured in tissue culture plates, but translating those techniques to microfluidic devices is often difficult.
One such technique is the culture of cells on or sandwiched between collagen gels as a model of the physiological 3D cell environment.1 Type I collagen is one of the most frequently used proteins for biomaterials applications because of its ubiquity in extracellular matrix, natural abundance, robust cell attachment sites, and biocompatibility.2 Many cells benefit from 3D culture with collagen, including cancer cells3,45, microvascular endothelial cells6, and hepatocytes7, among others. While the use of collagen gels is easy in open formats, such as tissue culture plates, the limited channel dimensions and enclosed nature of microfluidic devices makes the use of liquids that gel impractical without blocking the entire channel.
To overcome this problem, we combined the layer-by-layer deposition technique8 with chemical modifications of native collagen solutions to create ultrathin collagen assemblies on top of cells cultured in microfluidic devices. These layers can stabilize cell morphology and function similar to collagen gels and can be deposited on cells in microfluidic devices without blocking the channels with polymerized matrix. The goal of this method is to modify native collagen to create polycationic and polyanionic collagen solutions and to stabilize cells in microfluidic culture by depositing thin collagen matrix assemblies onto the cells. This technique has been used to stabilize the morphology and function of primary hepatocytes in microfluidic devices.9
Although layer-by-layer deposition has previously been reported with natural and synthetic polyelectrolytes10 to cover hepatocytes in plate culture11,12 and as a seeding layer for hepatocytes in microfluidic devices13,14, this method describes the deposition of a pure collagen layer on top of hepatocytes, mimicking the 3D collagen culture techniques. In this protocol, we use hepatocytes as example cells that can be maintained using 3D collagen layers. The many other types of cells that benefit from 3D culture in collagen may similarly benefit from culture after layer-by-layer deposition of an ultrathin collagen matrix assembly.
1. Preparation of the Native Soluble Collagen Solution
- Prepare or purchase 200 mg of acidified, soluble, type I collagen from rat tails at 1–3 mg/ml using standard isolation protocols, such as reported by Piez et al.15
- Scale the amount of starting material based on the desired end volume of modified collagen solutions. Approximately make 25–30 ml of methylated and 25–30 ml of succinylated collagen solutions, each at 3 mg/ml, from 200 mg of soluble native collagen.
2. Collagen Methylation
- Dilute 100 mg of the native, acidified (pH 2–3) collagen solution to a concentration of 0.5 mg/ml with ice cold sterile water and keep the solution on ice to prevent gelation.
- Adjust the pH of the collagen solution to 9–10 with a few drops of 1 N NaOH and stir at RT for 30 min. Observe the collagen precipitate, causing the solution to become cloudy.
- Spin down the precipitated collagen solution at 3,000 × g for 25 min. A clear, gel-like precipitate should be visible in the bottom of the tubes. Aspirate and properly dispose of the supernatant fluid.
- Resuspend the precipitated collagen in 200 ml of methanol with 0.1 N HCl and allow the methylation reaction to occur with stirring at RT for 4 days. The collagen will not dissolve, but should break apart into very small visible pieces that will make the solution cloudy.
- After the methylation, centrifuge the solution at 3,000 × g for 25 min to pellet the methylated collagen. Aspirate and dispose of the acidified methanol supernatant.
- Dissolve the methylated collagen in 25 ml of sterile PBS, giving a concentration of approximately 3 mg/ml, with repeated pipetting, and filter the solution through a 60 µm cell strainer. Adjust the pH of the solution to 7.3–7.4 using 20 µl increments of 1 N NaOH.
- Assess the concentration of the solution using a commercial rat collagen ELISA kit or hydroxyproline assay kit. Dilute the solution to 3 mg/ml with sterile PBS.
- Sterilize the methylated collagen solution by transferring it to a glass bottle with a screw cap, carefully layering 3 ml of chloroform at the bottom of the bottle, allow the bottle to set O/N at 4 °C, and then aseptically remove and store the top layer, which is the methylated collagen.
- Store the solution at 4 °C for use within 1 month.
3. Collagen Succinylation
- Dilute the other 100 mg of the native, acidified (pH 2–3) collagen solution to a concentration of 0.5 mg/ml with ice cold sterile water and keep the solution on ice to prevent gelation.
- Adjust the pH of the collagen solution to 9–10 with a few drops of 1 N NaOH and stir at RT for 30 min. The collagen should precipitate, causing the solution to become cloudy.
- Dissolve 40 mg of succinic anhydride (0.4 mg per mg of collagen) in 10 ml of acetone (1/20th the volume of the collagen solution). Slowly (in approximately 0.5 ml increments) add this mixture to the collagen solution with stirring while continuously monitoring the pH. Maintain the pH above 9.0 by adding 1 or 2 drops of 1 N NaOH as the pH approaches 9.0.
- Continue stirring at RT for 120 min after adding all of the succinic anhydride in acetone. Observe the solution to become clear as the succinylated collagen dissolves. Periodically check the pH to ensure it remains above 9.0.
- Adjust the pH of the solution to 4.0 with 20 µl increments of 1 N HCl. Observe the solution cloudy again, as the succinylated collagen precipitates.
- After the succinylation, centrifuge the solution at 3,000 × g for 25 min to pellet the succinylated collagen. Aspirate and dispose of the acidified supernatant with unreacted succinic anhydride.
- Dissolve the succinylated collagen in 25 ml of sterile PBS, giving a concentration of approximately 3 mg/ml, with repeated pipetting, and filter the solution through a 60 µm cell strainer. Adjust the pH of the solution to 7.3–7.4 using 20 µl increments of 1 N NaOH.
- Assess the concentration of the solution using a commercial collagen rat ELISA kit or hydroxyproline assay kit. Dilute the solution to 3 mg/ml with sterile PBS.
- Sterilize the succinylated collagen solution by transferring it to a glass bottle with a screw cap, carefully layering 3 ml of chloroform at the bottom of the bottle, allow the bottle to set O/N at 4 °C, and then aseptically remove and store the top layer, which is the succinylated collagen.
- Store the solution at 4 °C for use within 1 month.
4. Verification of the Collagen Modifications
- Prepare 1 ml samples from each of the native, methylated, and succinylated collagen solutions and dilute to a concentration of 0.1 mg/ml with sterile pure water for hydrogen ion titration. Further dilute the buffer at least 1,000-fold by dialysis against water through a 10 kDa cutoff membraneusing 3 solution changes for at least 4 hr each.
- Adjust the pH of the solutions to 7.3 with small amounts of NaOH and HCl. Using pH 7.3 as an arbitrary reference, perform hydrogen ion titration on each of the samples as described by Tanford16 to create titration curves for the native, methylated, and succinylated collagen solutions.
- Plot the change in pH per volume of acid added versus the number of bound H+ ions per molecule. The high pH “amino” range should show a shift in the succinylated collagen towards the neutral point (a loss of amine groups), and the low pH “carboxyl” range should show a leftward shift in the methylated collagen (a loss of carboxyl groups) and a rightward shift in the succinylated collagen (a gain in carboxyl groups), compared to the native collagen.
- Assess the efficacy of the succinylation reaction by determining the % of amino groups in the native collagen replaced by succinylation using the 2,4,6-trinitrobenzenesulfonic acid (TNBA) colorimetric method, following standard protocols17,18.
5. Fabrication of Microfluidic Devices and Cell Seeding
- Fabricate microfluidic devices using standard methods9. Use replica molding of PDMS from SU-8 masters on silicon defined using photolithography to create microfluidic cell culture chambers with 100 µm tall, 0.4–1.5 mm wide, and 1–10 mm long channels for cell growth.
- Use a plasma cleaner to oxidize the surfaces of the device and a glass slide, and then press together to bond. After sterilizing the device by exposure to UV light for at least 30 min, fill the chamber with 50 µg/ml fibronectin in sterile PBS and incubate at 37 °C for 45 min.
- Seed the device with cells, such as primary rat or human hepatocytes, which require a collagen gel for stabilization of phenotype or differentiation state. We use 20 µl of freshly isolated primary rat hepatocytes7,19 or commercially available cryopreserved primary human hepatocytes seeded at 14×106 cells/ml per device.
- Allow the cells to attach for 4–6 hr, and then wash out unattached cells and replace the plating media with growth media. Incubate O/N to ensure full cell spreading and create a confluent monolayer of cells.
6. Layer-by-layer Collagen Deposition
- In a laminar flow tissue culture hood, prepare sufficient volumes of methylated and succinylated collagen solutions for 10 applications of each solution per device, as well as a few mls of media. Keep the solutions on ice. We use 20 µl (10–15 times the total device volume) of collagen per layer per device.
- Beginning with the methylated (polycationic) solution, alternate flushing the devices with 20 µl of methylated and then succinylated collagen solutions, waiting 1 min between each application. Flush the device a total of 10 times per solution, which should take approximately 20 min. Work quickly to minimize the amount of time the cells are without media.
- Observe collagen slowly accumulate at the inlet/outlet depending on its size. If the resistance to fluid flow increases, flush the device once or twice with media, and then continue layering.
- After applying all of the layers, rinse the device twice with fresh media and return to the incubator. Depending on the cell type, the extracellular matrix-induced morphological changes, such as enhanced polarization in hepatocytes, should be visible within a few hours.
- Prepare representative devices for transmission electron microscopy using standard methods9 to verify the presence of a collagen matrix assembly on top of the cultured cells.
7. Stabilization of Cell Phenotype and Function
- Image the cell morphology, viability, and polarity using standard methods for the cell type. For hepatocytes, collect images over 14 days using phase microscopy, LIVE/DEAD staining for viability, and CMFDA dye for apical polarization to demonstrate the effects of the collagen deposition.
- For cell morphology, flush the device through the inlet by pipette with 20 µl of PBS to rinse, inject 20 µl of PBS, and image the device using phase contrast microscopy.
- For LIVE/DEAD staining, flush the device through the inlet by pipette with 20 µl of PBS to rinse, inject 20 µl of LIVE/DEAD staining solution with DAPI prepared following the manufacturer's instructions, incubate for 30 min at 37 °C, rinse the device again with 20 µl of PBS, and image the device using fluorescence microscopy.
- For bile canaliculi staining, flush the device through the inlet by pipette with 20 µl of PBS to rinse, inject 20 µl of 2 µM CMFDA staining solution with DAPI prepared following the manufacturer's instructions, incubate for 30 min at 37 °C, rinse the device again with 20 µl of PBS, and image the device using fluorescence microscopy.
- Collect the spent media and measure cellular metabolic products at appropriate time points over the culture duration to determine the cell function. For hepatocytes, measure the amount of albumin and urea in the spent media.
- Perform endpoint functional analyses on the cells themselves, such as assessing enzyme activity levels, antibody staining, or RNA analysis. For hepatocytes, induce and measure cytochrome P450 enzyme activities or phase II conjugation enzyme glutathione S-transferase.
Native collagen can be modified using methylation and succinylation to create polycationic and polyanionic collagen solutions for use in layer-by-layer deposition. Succinylation modifies the ε-amino groups of native collagen with succinyl groups, and methylation modifies the carboxyl groups of native collagen with a methyl group (Figure 1A). These modifications to the collagen protein amino acid side chains alter the pH titration curves for the solutions. Succinylation reduces the number of amino groups and increases the number of carboxyl groups, while methylation has no effect on the amino groups and reduces the number of carboxyl groups, compared to native collagen (Figure 1B). These modifications alter the charge characteristics of the collagen molecules. Succinylation removes positively charged groups and replaces them with negatively charged groups, and methylation removes negatively charged groups (Figure 1C).
Layer-by-layer deposition of the methylated and succinylated collagen solutions (COL LBL) creates an ultrathin collagen matrix assembly on top of charged surfaces, such as cells (Figure 2A). Primary hepatocytes seeded on fibronectin-coated glass within a microfluidic device (Figure 2B) can be coated with such a purely collagen matrix assembly (Figure 2C). Deposition of 10 bilayers on cells creates a collagen layer thickness of approximately 140 nm (Figure 2D).
The ultrathin collagen assembly layer deposited using the COL LBL technique stabilizes primary hepatocytes for over 14 days in microfluidic devices. Hepatocytes without a top matrix cover lose their differentiated phenotype, contract, and lift off the surface of microfluidic devices over time (Fig. 3A-C). In contrast, hepatocytes covered with an ultrathin collagen assembly maintain their differentiated morphology and polarization over 14 days (Figure 3D-F). The viability of the cells was maintained at greater than 90% (Figure 3G-I). In addition, the polarization of the hepatocytes treated with COL LBL increased over time, showing development of an apical biliary canalicular network (Figure 3J-L).
In addition to cell viability, morphology, and polarization, the COL LBL technique recovers and stabilizes the function of primary hepatocytes at levels similar to standard techniques for cells in plates. The secretion of albumin by hepatocytes is low immediately after plating, and remains low unless the cells are induced to polarize through a matrix covering. After COL LBL treatment, the albumin secretion by the hepatocytes increases over 8–10 days before stabilizing (Figure 4A). Urea production is high immediately after cell seeding and decreases over time in cells not covered with a matrix, urea production stabilizes after 8–10 days in hepatocytes stabilized with COL LBL (Figure 4B). Cytochrome P450 (CYP) enzyme expression in response to stimuli is one specialized function of hepatocytes. The induction of CYP1A1 in hepatocytes in response to 3-methylcholanthrene was recovered and stabilized up to 14 days by the COL LBL treatment (Figure 4C).
Figure 1. Rat tail collagen solutions were chemically modified to create net positively or net negatively charged polymer solutions. (A) Schematic representations of the succinylation and methylation reactions that modify carboxyl and ε-amino groups, respectively. (B) Hydrogen titration curve derivatives showing shifts in the relative numbers of H+ ions needed to alter solution pH after collagen modification. (C) The net charges (mean ± SD) of the modified collagen solutions calculated from the H+ shifts in (B) normalized to native collagen. MC: methylated collagen. SC: succinylated collagen. AA: amino acid residues. Re-print with permission from 9. Please click here to view a larger version of this figure.
Figure 2. Layer-by-layer deposition of modified collagen solutions created an ultrathin collagen matrix assembly on top of the hepatocytes. (A) Schematic representation of hepatocytes seeded on fibronectin on a microdevice and incubated with alternating methylated and succinylated collagen solutions. TEM cross-sections of representative (20 cells examined per group) control (B) and COL LBL-treated (C) hepatocytes after 2 days of culture. Arrows: LBL collagen layer. Scale bar: 2 µm. (D) COL LBL layer thickness (mean ± SD) per cell measured from TEM images (n=20 cells from 4 devices). Modified from 9. Please click here to view a larger version of this figure.
Figure 3. The COL LBL technique maintained hepatocyte morphology, viability, and polarization in microdevices over 14 days in culture. Hepatocytes with no top layer in microdevices (negative control) contract and lift off the surface over time (A–C). LBL deposition of modified collagens maintains hepatocyte morphology (D–F) and viability (G–I) in microdevices over 14 days. CMFDA is a generic cell marker that remains in the cytoplasm of non-polarized hepatocytes (J). Over time, the development of bile canaliculi can be seen by the excretion of the fluorophore into canalicular spaces (K) that develop around the cells over time (L). Image scale bar: 100 µm. Inset scale bar: 50 µm. Modified from 9. Please click here to view a larger version of this figure.
Figure 4. COL LBL stabilizes hepatocyte albumin and urea sections and CYP activity over 14 days in microfluidic devices. (A) Albumin secretion by hepatocytes in COL LBL started low and increased with time until reaching a plateau at day 10. (B) Urea secretion by COL LBL hepatocytes decreased with time until reaching a plateau at day 10. (C) The induction of CYP1A activity was low at day 2, but significantly increased at day 7, a level that was maintained through day 14. Mean ± SEM; COL LBL: hepatocytes cultured after LBL collagen deposition; NoTop: hepatocytes cultured with no top matrix layer; n=6–8 devices. Modified from9. Please click here to view a larger version of this figure.
Ultrathin pure collagen assemblies can be deposited on charged cells or material surfaces using layer-by-layer deposition of modified collagens. The results of this study demonstrate that methylation and succinylation of native collagen create polycationic and polyanionic collagen solutions (Figure 1) that can be used with the layer-by-layer technique to deposit ultrathin collagen matrix assemblies on cells (Figure 2) or other charged material surfaces. Such ultrathin matrix layers can stabilize the morphology, viability, and polarization of cells such as hepatocytes (Figure 3), after seeding in microfluidic devices, as well as their function (Figure 4).
For successful deposition of ultrathin collagen layers, the modification reactions creating the methylated and succinylated collagens are critical. Because the carboxyl group methylation reaction is not favorable, the collagen methylation reaction must be performed in acidified methanol to drive the reaction forward, based on Le Chatelier’s principle. Therefore, it is important to remove all excess water from the collagen after the pH precipitation before dissolving the precipitated collagen in the acidified methanol in order to increase the reaction efficiency. For the succinylation reaction, maintaining the pH of the solution above 9 while adding the succinic anhydride in acetone is critical. Continuous monitoring of the pH and slow addition of the succinic anhydride acetone solution are helpful for ensuring an efficient reaction. During the layer-by-layer deposition, it is critical to minimize the time that the cells go without media. The procedure for 10 bilayers takes approximately 20 min, which is about as long as most cells should be without media outside of the incubator. If more than 10 bilayers must be deposited, rest the cells in media in an incubator for 3–4 hr between collagen depositions to ensure the health of the cells. At the other extreme, too few layers will not be effective in maintaining cell morphology. In our testing, 3 layers did not maintain cell morphology, while 10 layers did.
Several modifications can be made to the procedures to troubleshoot problems. The time and temperature of the methylation reaction can be varied if problems arise. For example, decreasing the reaction temperature to 4 °C and increasing the time to 7 days may increase efficiency. Also, depending on the extent of polymerization during the pH extraction, the collagen will sometimes fully dissolve in the acidified methanol during the methylation reaction. If this occurs, add 10 M NaOH to bring the pH up to 9–10 after the reaction has completed in order to precipitate the collagen so that it can be pelleted during the next step. While applying the layer-by-layer deposition in microfluidic devices, small channels or ports can become clogged with collagen, indicated by an increase in the device fluidic resistance. Flushing the device with fresh media should remove any blockages, allowing deposition to continue.
The manual deposition of the layer-by-layer collagen as described here is a limitation in terms of the scaling and automation of microfluidic devices. However, this technique is compatible with standard microfluidic hardware, including valves and pumps that could be used to automate the technique and prepare arrays of devices at once. Another limitation of this technique for use in plate culture is that it requires a relatively large amount of native collagen. For microfluidic application, the volumes of methylated and succinylated collagens can be used to create ultrathin collagen layers in many devices. With standard tissue plate culture, on the other hand, the volumes required to coat wells become more prohibitive with increasing well size.
This layer-by-layer technique for the deposition of ultrathin collagen matrix layers could be further developed by using alternative or functional modifications, as well as by creating multiple layers of cells separated by the thin matrix assemblies. Although the relatively simple amino acid side chain modification reactions of methylation and succinylation were used here, alternative or additional reactions could be used instead in order to create functionalized collagens. The functionalized sidechains should still yield proteins with net overall positive and negative charges, but such functionalization may be useful for a variety of applications.
The authors have nothing to disclose.
This work was supported by grants from the National Institutes of Health, including a microphysiological systems consortium grant from the National Center for Advancing Translational Sciences (UH2TR000503), a Ruth L. Kirschstein National Research Service Award Postdoctoral Fellowship (F32DK098905 for WJM) and pathway to independence award (DK095984 for AB) from the National Institute of Diabetes and Digestive and Kidney Diseases.
|collagen type I, rat tail||Life Technologies||A1048301||option for concentrated rat tail collagen|
|collagen type I, rat tail||Sigma-Aldrich||C3867-1VL||option for concentrated rat tail collagen|
|collagen type I, rat tail||EMD Millipore||08-115||option for concentrated rat tail collagen|
|collagen type I, rat tail||R%D Systems||3440-100-01||option for concentrated rat tail collagen|
|succinic anhydride||Sigma-Aldrich||239690-50G||succinylation reagent|
|anhydrous methanol||Sigma-Aldrich||322415-100ML||methylation reagent|
|sodium hydroxide||Sigma-Aldrich||S5881-500G||pH precipitation reagent|
|hydrochloric acid||Sigma-Aldrich||320331-500ML||pH precipitation reagent|
|rat collagen type I ELISA||Chondrex||6013||option for detecting collagen content|
|hydroxyproline assay kit||Sigma-Aldrich||MAK008-1KT||option for detecting collagen content|
|hydroxyproline assay kit||Quickzyme Biosciences||QZBtotcol1||option for detecting collagen content|
- Pedersen, J. A., Swartz, M. A. Mechanobiology in the third dimension. Ann Biomed Eng. 33, (11), 1469-1490 (2005).
- Glowacki, J., Mizuno, S. Collagen scaffolds for tissue engineering. Biopolymers. 89, (5), 338-344 (2008).
- Vescio, R. A., et al. In vivo-like drug responses of human tumors growing in three-dimensional gel-supported primary culture. PNAS. 84, 5029-5033 (1987).
- Chandrasekaran, S., Guo, N. -h, Rodrigues, R. G., Kaiser, J., Roberts, D. D. Pro-adhesive and chemotactic activities of thrombospondin-1 for breast carcinoma cells are mediated by α3β1 integrin and regulated by insulin-like growth factor-1 and CD98. J Biol Chem. 274, (16), 11408-11416 (1999).
- Chen, S. S., et al. Multilineage differentiation of rhesus monkey embryonic stem cells in three‐dimensional culture systems. Stem Cells. 21, (3), 281-295 (2003).
- Whelan, M. C., Senger, D. R. Collagen I initiates endothelial cell morphogenesis by inducing actin polymerization through suppression of cyclic AMP and protein kinase A. J Biol Chem. 278, (1), 327-334 (2003).
- Dunn, J. C., Tompkins, R. G., Yarmush, M. L. Long-term in vitro function of adult hepatocytes in a collagen sandwich configuration. Biotechnol Prog. 7, (3), 237-245 (1991).
- Decher, G. Fuzzy Nanoassemblies: Toward Layered Polymeric Multicomposites. Science. 277, 1232-1237 (1997).
- McCarty, W. J., et al. A novel ultrathin collagen nanolayer assembly for 3-D microtissue engineering: Layer-by-layer collagen deposition for long-term stable microfluidic hepatocyte culture. TECHNOLOGY. 2, (01), 67-74 (2014).
- Swierczewska, M., et al. Cellular response to nanoscale elastin-like polypeptide polyelectrolyte multilayers. Acta Biomater. 4, (4), 827-837 (2008).
- Kim, Y., Larkin, A. L., Davis, R. M., Rajagopalan, P. The design of in vitro liver sinusoid mimics using chitosan-hyaluronic acid polyelectrolyte multilayers. Tissue Eng Pt A. 16, (9), 2731-2741 (2010).
- Larkin, A. L., Rodrigues, R. R., Murali, T., Rajagopalan, P. Designing a Multicellular Organotypic 3D Liver Model with a Detachable, Nanoscale Polymeric Space of Disse. Tissue Eng Pt C. 19, (11), 875-884 (2013).
- Kidambi, S., et al. Patterned Co‐Culture of Primary Hepatocytes and Fibroblasts Using Polyelectrolyte Multilayer Templates. Macromol Biosci. 7, (3), 344-353 (2007).
- Janorkar, A. V., Rajagopalan, P., Yarmush, M. L., Megeed, Z. The use of elastin-like polypeptide-polyelectrolyte complexes to control hepatocyte morphology and function in vitro. Biomaterials. 29, (6), 625-632 (2008).
- Piez, K. A., Eigner, E. A., Lewis, M. S. The Chromatographic Separation and Amino Acid Composition of the Subunits of Several Collagens*. Biochemistry. 2, (1), 58-66 (1963).
- Tanford, C. The interpretation of hydrogen ion titration curves of proteins. Adv Protein Chem. 17, 69-165 (1962).
- Cayot, P., Tainturier, G. The quantification of protein amino groups by the trinitrobenzenesulfonic acid method: a reexamination. Anal Biochem. 249, (2), 184-200 (1997).
- Kakade, M. L., Liener, I. E. Determination of available lysine in proteins. Anal Biochem. 27, (2), 273-280 (1969).
- Seglen, P. O. Preparation of isolated rat liver cells. Methods Cell Biol. 13, 29-83 (1976).