Evaluating kinetic and physiological features of rCHO-K1 cells cultured on microcarriers for production of a recombinant metalloprotease/disintegrin
M.C. da Silva
Financial support: Work supported by Fapesp (Proc. 01/02068-4) and CNPq scholarships (Proc.131241/2001-0).
Keywords: adhesion, animal cell, CHO-K1, growth, microcarrier, recombinant disintegrin.
present kinetic and physiological data regarding the culturing of
rCHO-K1 cells on various microcarriers, to evaluate the potential
of this culture strategy for mass production of these cells and expression
of a recombinant disintegrin. Cultures were performed in 500 mL spinner
flasks in DMEM culture medium with 10% v/v fetal calf serum, gently
A significant number of animal cell lines with industrial potential have an absolute requirement for adhesion to a solid substratum for growth and optimal synthesis of the product of interest. Such anchorage-dependent cells (ADC) grow in monolayers, and yields in biomass and product are thus limited by the size of the available adhesion area. To maintain high yields in large-scale production, the concept of culture on microcarriers introduced by van Wezel (1967) is being used with great success in research and industry. With this technique, the ADC attaches to and grows on the surface of small particles that are kept in suspension in the culture medium by gentle agitation. Fabricated from a variety of substrates, such particles are commercially available and supplied with diverse technological properties (surface charge, density, hydrophilicity, size, porosity, and adhesion-promoting coatings), for optimization of adhesion and cell growth, since many key cell processes and properties, such as the synthesis of DNA, RNA, and proteins, depend on correct cell adhesion and spreading (Varani et al. 1985; Nilsson et al. 1988; Varani et al. 1992).
The selection of the most suitable microcarrier for a new bioprocess involving the use of animal cells is an important stage, not only for technical reasons, since there is no microcarrier which is optimal for use in all applications, but also for economic reasons, since the cost of the microcarrier contributes significantly to the cost of the bioprocess. This rational approach is resulting in easier, more efficient and economical large-scale culture of animal cells for the manufacture of novel therapeutic proteins. Therefore, we decided to test microcarrier technology in the production of a biologically-active disintegrin from Agkistrodon contortrix laticinctus snake venom (Selistre-de-Araujo et al. 1997). Disintegrins were demonstrated to interfere with high affinity for integrins, cell-surface receptors that play crucial roles in physiological processes such as development and angiogenesis as well as in some diseases such as inflammation and cancer metastasis (Gould et al. 1990; Ramos and Selistre-de-Araujo, 2006). Therefore, disintegrins may be useful as integrin inhibitors in the treatment of cancer and inflammation. Snake venom disintegrins are synthesized in vivo as multimodular proteins that comprise a signal peptide, a pro-domain, a metalloprotease domain, a disintegrin domain, and a cysteine-rich domain (Bjarnason and Fox, 1994; Fox and Serrano, 2005; Ramos and Selistre-de-Araujo, 2006). Disintegrins can be released by proteolysis from their precursors, yielding soluble proteins with biological activity.
Recently, several disintegrins have been proposed as therapeutic drugs, in view of their antitumor, antiangiogenic and antithrombotic activities (Braud et al. 2000; Matsui et al. 2000; Zhou et al. 2000; Schmitmeier et al. 2003; Swenson et al. 2004; Markland et al. 2005; Yang et al. 2005). For such applications, a few cDNAs encoding disintegrins, originally from snake venom glands, have been cloned and expressed both in bacteria and in insect cells (Park et al. 1998; Fan et al. 1999; Jeon and Kim, 1999; Wang et al. 2003; Wang et al. 2004) by genetic engineering techniques. However, due to the structural complexity and high number of disulfide bonds of this type of protein, the yield and activity of the recombinant protein from most production processes are very low. Also, the expression of a metalloprotease in an active form has been shown to be very unstable (Selistre-de-Araujo et al. 2000). To overcome these difficulties, we decided to express the zymogen form of a snake venom metalloprotease with DCD-disintegrin and cysteine-rich domains from the snake Agkistrodon contortrix lacticinctus named ACLD (Selistre-de-Araujo et al. 1997). The main focus in the present study is on the utilization of a genetically modified CHO-K1 cell-line to express ACLD stably in a large-scale mammalian system. As no information is available in the literature on the development of a cell culture process with microcarrier technology for production of that sort of biomolecule, the objective was to evaluate kinetic and physiological features of a rCHO-K1 cell line when grown on four types of microcarriers with the purpose of identifying culture conditions that lead to high densities of active cells expressing the recombinant disintegrin. The cell performance was characterized by determination of adhesion, growth and metabolic parameters. The presence of the disintegrin was detected by immunoblotting.
mammalian cell-line CHO-K1 (Chinese hamster ovary) was genetically
modified by Selistre-de-Araujo et al. (1997), who cloned and
expressed the cDNA encoding a metalloprotease/disintegrin (ACLD) from
the venom gland of the snake Agkistrodon contortrix lacticinctus.
The cDNA was cloned into the plasmid vector pCDNA3 (Invitrogen),
which was in turn stably transfected into the CHO-K1 cells. The genetically
modified CHO-K1 cell line, here denominated CHOZMD (for zymogen
of metalloprotease/disintegrin), was cultured in Dulbecco’s
modified Eagle medium (DMEM, Gibco/BRL) supplemented with 110 IU.mL-1
penicillin (Sigma), 0.1 mg.mL-1
Four microcarriers were selected on account of the outstanding technical features claimed by the manufactures to enhance their performance in animal cell culture. The solid microcarriers Cytodex 1 (Pharmacia) and Pronectin F (Solohill) and the macroporous Cultispher S (Percell Biolytica) and Immobasil FS (Ashby Scientific) were tested under conditions as close as possible to those recommended by their manufacturers. The most noteworthy features of these microcarriers, their densities and the inoculum level used in the CHOZMD cell culture are presented in Table 1.
After attaining a sufficient quantity of cells in 150 cm2 T-flask cultures (Corning), through one passage in 25 and 75 cm2 T-flasks, 100 mL of culture medium, equilibrated with a predefined amount of microcarriers in a 500 mL glass spinner flask (Wheaton) were inoculated as specified in Table 1. During the first six hrs, in order to achieve a uniform and efficient cell attachment, the culture was performed with only 1/3 of the final volume and with intermittent stirring (5 min of stirring at 25-30 rpm every 30 min). After cell attachment, the volume of culture was made up to 300 mL and the stirring was kept constant at 60 rpm until the end of the experiment. Medium was replaced according to the requirements of each individual experiment; as soon as the pH of the medium decreased to approximately 7.1, 50% of its working volume was replaced by fresh medium.
For cell adhesion measurements, 1 mL samples were taken from the spinner flask every 2 hrs during the first 20 hrs of culture. 3 mL samples for monitoring of cell density and viability, as well as glucose, lactate, amino acid and ammonium concentrations were taken daily. The sample volume taken from the cultures was the same in all experiments to avoid differences in the remaining working volume.
The viability and density of cells in suspension were measured by the Trypan blue dye exclusion method (Doyle and Griffths, 1998). Adhered cell numbers were estimated by counting in haemocytometer the stained nuclei with crystal violet (GE Healthcare, 2005). The two ruled areas of the haemocytometer were counted two times. Glucose, amino acid and ammonium ion concentrations were measured by high performance liquid chromatography (HPLC). For glucose assays, a Waters chromatograph with a Lonpack KS 803 (Shodex) column was used. For amino acid and ammonium analysis a Shimadzu chromatograph with Amino Na and ISC-3050504 Na Shim-pack columns from Shimadzu were used. The lactate concentration was determined with an enzymatic kit from Sigma Diagnostics.
Cell adhesion was characterized by estimating the adhesion constants of a hyperbolic model which correlates the adhered cell density with time as follows:
where Na is the number of attached cells at time t (cell.mL-1), Ns is the suspended cell density at time t (cell.mL-1), Nso and NSf are the initial measured and fitted, cell densities in suspension (cell.mL-1), respectively, and Ka is the adhesion time constant (h). The parameters Naf (final number of attached cells (cell.mL-1)) and Ka were estimated by nonlinear curve fitting of experimental measurements of (NS0 - NS) at various t utilizing the program Origin 6.0, assuming no dependency between the parameters. Determinations of cells in suspension were preferred because its experimental error was much lower than the measurements of adhered cells.
To estimate the maximum specific growth-rate in the exponential growth phase, the following equation was utilized:
where µmax is maximum specific growth-rate, Na is the attached cell density at time t (cell.mL-1) and Nao is the initial attached cell density (cell.mL-1).
The disintegrin present in the medium was detected by two methods, SDS-PAGE (Laemmli, 1970) and dot blotting. For the gel electrophoresis, samples were taken just after the end of the exponential growth phase, from a culture in which a high final yield of cell was attained with the microcarrier Cytodex 1. At this stage of the culture, the DMEM medium with 10% v/v fetal bovine serum was replaced by similar medium with only 1% v/v serum to diminish the protein content. The cell culture was prolonged in this low serum medium for 24 hrs, samples being taken at 0, 3, 6, 9, 12 and 24 hrs. Samples of 30 mL were filtered (Millipore membrane, 0.45 μm) and concentrated by ultrafiltration in an Amicon system (MWCO 10 kDa, Millipore) followed by additional concentration in Centriprep (MWCO 30 kDa, Millipore) and Centricon (MWCO 30 kDa, Millipore), to give a final volume of 500 µL. Samples of 10 µL were then resolved by electrophoresis in a 15% polyacrylamide gel.
the dot blotting technique, samples of 30 mL were taken from a culture
of CHOZMD cells adapted to the serum-free medium CHO-SFM-II, growing
on the microcarrier Pronectin F, which was the one showing the highest
cell density in CHOZMD cultures. These samples were passed through
a 0.45 μm filter (Millipore), concentrated by ultrafiltration
to a final volume of 500 µL. Samples (200 µL) were then applied on
a nitrocellulose membrane (Immobilon-Millipore) in Bio Dot equipment
(BioRad). After a first washing
with TBST (
can be seen in Figure
the results for the adhesion of cells to the microcarriers (first
6 hrs of culture), presented in Figure 1, it was possible to
determine the adhesion parameters Naf and Ka
of a hyperbolic model displayed in Table
The results obtained by the nonlinear fitting algorithm presented in Table 2 show that, in general, the hyperbolic model describes better the behaviour of the cellular adhesion onto the microcarriers than the first-order model suggested by Ng et al. (1996). Using the correlation coefficient, R, as a measure of the quality of the adjustment, the value of that coefficient is, on average, slightly larger than the one for the first-order model. Another statistical parameter that reflects the fitting quality is the standard error (SE) of the estimated parameters. The SEs of the hyperbolic parameters Naf and Ka are smaller than those of the first order model Naf and ka. The mean of SEs (in %) of such parameters for the models is, respectively, 1.98 and 6.00 for the hyperbolic and 2.24 and 7.36 for the first order. An additional reason for the choice of the hyperbolic model is that it predicts values of the parameter Naf closer to the experimental values of inoculated cells (NS0) than those of the first order model (Figure 2).
A statistical analysis at a confidence level of 95% of the values of Naf in Table 2, reveal that only the microcarriers Cytodex 1 and Pronectin F have not significant difference. The results presented also indicate that the microcarrier with the highest expected number of adhered cells is Pronectin F, with Naf = 330,855 cells.mL-1, and in second place Cytodex 1 (Naf = 303,726 cells.mL-1), although it can take a long time to attain such asymptotic values. The higher rate of the Cytodex 1 is possibly due to the optimized positively-charged dextran matrix that enhances the binding between cells and particles. In third place is Cultispher S, with Naf = 280,350 cells.mL-1; this maximum value of adhered cells tends to be well below that initially inoculated in the suspension (NS0), possibly due to the low amount of microcarrier utilized in culture. Immobasil FS tended to stabilize with a low adhesion of cells (Naf = 250,763 cell.mL-1). This behaviour can be explained by the physical-chemical nature of the surface of the microcarrier, as will be reinforced later, in the analysis of cell growth.
Referring to Ka, a statistical analysis at a confidence level of 95% shows that only the values of Cytodex 1 and Immobasil FS are not significantly different. The values of Ka, the parameter that differentiates better the adhesion characteristics of the microcarriers, can be interpreted as a measure of the affinity for the cell, having a low value when the adhesion rate is high. In this sense, the microcarrier Cultispher S showed the highest affinity, followed by Immobasil FS, Cytodex 1 and Pronectin F.
Cell growth on the different types of adhesion substrate, presented in Figure 3a, can be compared through the maximum specific growth rate (µmax), estimated by means of equation 2, for the exponential growth phase of each culture. For Cytodex 1, Cultispher S and Immobasil FS, µmax was estimated in the period from 24 to 96 hrs. For Pronectin F, the period chosen was 48 to 120 hrs. The calculated values of µmax are presented in Table 3. After the anchored cell concentration fell at an advanced growth stage, probably due to key nutrient exhaustion, the experiments were stopped. Another parameter that can be used for comparison of cell growth on the microcarriers is the maximum cell density achieved in the cultures. Estimation of the error carried out with data obtained in our laboratories for adhered cell quantification by nuclei counting has shown standard deviation around 7% of the mean. The maximum values of cell density accomplished in the cultures with microcarriers were 2.76 x 106 cells.mL-1 for Pronectin F, 1.47 x 106 cells.mL-1 for Cytodex 1, 1.34 x 106 cells.mL-1 for Cultispher S and 3.84 x 105 cells.mL-1 for Immobasil FS. Each of these values was determined as the mean of a quadruplicate counting of nuclei and their statistical comparison at a confidence level of 95% gave as results: no significant difference between the cultures with Cytodex 1 and Cultispher S, the highest cell density being produced by Pronectin F and the lowest by Immobasil FS.
Analyzing the specific cell adhesion in terms of cells per particle, as plotted in Figure 3b, it may be noted that microcarriers coated with adhesion promoter substrates, namely collagen for Cultispher S and Pronectin™ for Pronectin F, showed a higher attached cell number than uncoated microcarriers such as Cytodex 1 and Immobasil FS. Immobasil FS, even though showing good adhesion and growth, did not attain a high cell density. The culture performed with the microcarrier Immobasil FS exhibited the lowest anchored cell density. The superior performance of Cultispher S may be attributed to its higher specific area. As Cultispher S has a very porous structure, cells grow on the inner as well as the outer surfaces, and the culture with this microcarrier produced approximately 780 cells/particle. The Pronectin F microcarrier produced 450 cells/particle, followed by Cytodex 1 (70 cells/particle) and Immobasil FS (15 cells/particle).
The statiscal analysis at 95% of confidence level of the calculated µmax values in Table 3 reveal no significant differences between the growth on Immobasil FS and Cytodex 1 and also between Cytodex 1 and Cultispher S. In addition, Pronectin F was the microcarrier that gave the highest specific growth rate. With Immobasil FS, even though it showed a low density of adhered cells, a higher µmax was achieved than with Cultispher S.
The profile of amino acids produced and consumed by the cultures with the four microcarriers followed patterns typical of animal cells in culture (Thomas, 1990), as can be observed in Figure 4a and Figure 4b. In all cultures the amino acids most consumed by the cells were arginine, isoleucine, leucine, lysine and serine and those most produced were alanine, glycine and glutamate. Appreciable differences in specific consumption of most of the amino acids occurred between the cultures with microcarriers coated with adhesion molecules and the uncoated microcarriers. The amount of amino acids consumed per 106 anchored cells was higher with the uncoated microcarriers. The same trend can be seen in Figure 4b, for amino acids produced by the cells: the cultures with uncoated microcarriers exhibited higher specific values of produced alanine and glycine.
remarkable feature of amino acid metabolism is the large quantity
of ammonia produced in cultures with the uncoated microcarriers, namely,
Cytodex 1 and Immobasil FS (21.5 and
consumption was similar in all cultures, the lowest concentration
being attained with the microcarrier Cytodex 1, Figure 5a. Lactate
production may also be considered similar in the various experiments,
Figure 5b, a higher level of production being observed with Cytodex
1 as consequence of the higher glucose consumption in this culture.
The lactate concentrations obtained in these experiments (up to
In order to overcome the problem caused by protein interference, a serum-free medium (CHO-SFM-II) was tested for CHOZMD cell culture. Combining this medium with the Pronectin F microcarrier, which showed high yields of cells, a culture was carried out by replacing gradually the 10% v/v serum DMEM by the serum-free medium. Even though a lower cell density was attained with this medium at 192 hrs, 1.8 x 106 cells/mL, as can be observed in Figure 6, against 2.8 x 106 cells/mL at 168 hrs with DMEM plus 10% v/v of serum (in Figure 3), the dot blot assay allowed a clearer demonstration of the presence of disintegrin in the medium CHO-SFM-II (Figure 7). The use of the serum-free medium is also interesting to facilitate the stage of purification of the disintegrin in a large-scale development of the bioprocess.
The diversified performance shown by the microcarriers can be ascribed to the variety and relevance of the main features built into each one. The best results for culturing the CHOZMD cells were obtained with the microcarrier Pronectin F. Even though the adhesion process on this microcarrier was slow, it allowed the highest specific growth rate (µmax) and the highest cell density achieved. Based on these results, since disintegrin production should be proportional to the cell number, Pronectin F is expected to be the most promising choice for larger-scale cultures.
With regard to cell attachment, the microcarrier Cultispher S showed the lowest adhesion time-constant Ka = 0.278 hrs (or the highest adhesion rate), followed, in increasing order, by the microcarriers Immobasil FS (0.846 hrs), Cytodex 1 (0.909 hrs) and Pronectin F (5.125 hrs ).
Comparing the growth, the highest value for specific maximum growth rate, µmax, was obtained with the Pronectin F microcarrier with 0.67d-1, followed, in decreasing order, by Immobasil FS with 0.52d-1, Cytodex 1 with 0.45d-1 and Cultispher S with 0.34d-1. The highest value of final cell density was obtained with the microcarrier Pronectin F (2.76 x 106 cells.mL-1), followed by Cytodex 1 (1.47 x 106 cells.mL-1), Cultispher S (1.34 x 106 cells.mL-1) and Immobasil FS (3.84 x 105 cells.mL-1).
The specific consumption and production of the amino acids (in grams x (106 adhered cells)-1) were higher in the CHOZMD cell cultures with the microcarriers coated with adhesive molecules, Pronectin F and Cultispher S, than in the cultures with uncoated microcarriers, Immobasil FS and Cytodex 1. The glucose consumption and lactate production did not differ significantly. In the case of ammonia production, the trends were clearly different when the cell was cultivated with the coated and uncoated microcarriers; it seems that the uncoated microcarriers having an abiotic adhesion surface, dissimilar to that found in vivo, stimulated the production of higher quantities of ammonia.
The results obtained in this work suggest that the use of the microcarrier Pronectin F for large-scale production of the recombinant disintegrin with the CHOZMD cell is very promising. A better and quantitative assay of the amount of the recombinant protein, such as an enzyme-linked immunoassay (ELISA), is under development and will allow the process to be followed more consistently.
BARTOLOMEO, Maria Paola and MAISANO, Federico. Validation of a reversed-phase HPLC method for quantitative amino acid analysis. Journal of Biomolecular Techniques, April 2006, vol. 17, no. 2, p.131-137.
BJARNASON, Jon Bragi and FOX, Jay William. Hemorrhagic metalloproteinases from snake venoms. Pharmacology and Therapeutics, 1994, vol. 62, no. 3, p. 325-372. [CrossRef]
BRAUD, Sandrine; BON, Cassian and WISNER, Anne. Snake venom proteins acting on hemostasis. Biochimie, September 2000, vol. 82, no. 9-10, p. 851-859. [CrossRef]
FASSNACHT, Dieter and PÖRTNER, Ralf. Experimental and theoretical considerations on oxygen supply for animal cell growth in fixed-bed reactors. Journal of Biotechnology, July 1999, vol. 72, no. 3, p. 169-184. [CrossRef]
FOX, Jay W. and SERRANO, Solange M.T. Structural considerations of the snake venom metalloproteinases, key members of the M12 reprolysin family of metalloproteinases. Toxicon, June 2005, vol. 45, no. 8, p. 969-985. [CrossRef]
HEALTHCARE. Microcarrier cell culture: Principles and methods.
General Electric Company,
GOULD, R.J.; POLOKOFF, M.A.; FRIEDMAN, P.A.; HUANG, T.F.; HOLT, J.C.; COOK, J.J. and NIEWIAROWSKI, S. Disintegrins: a family of integrin inhibitory proteins from viper venoms. Proceeding of the Society for Experimental Biology and Medicine, November 1990, vol. 195, no. 2, p.168-171.
JEON, O.H. and KIM, D.S. Cloning, expression, and characterization of a cDNA encoding snake venom metalloprotease. Biochemistry and Molecular Biology International, March 1999, vol. 47, no. 3, p. 417-425.
LAO, Mio-Sam and TOTH, Derek. Effects of ammonium and lactate on growth and metabolism of a recombinant Chinese hamster ovary cell culture. Biotechnology Progress, September-October 1997, vol. 13, no. 5, p. 688-691. [CrossRef]
MARKLAND, Francis S.; SWENSON, Stephen; COSTA, Fritz; MINEA, Radu; SHERWIN, Russell P.; YANG, Dongyun Y.; ERNST, William and FUJII, Gary. A snake venom disintegrin with potent antitumor and antiangiogenic activity. Toxin Reviews, 2005, vol. 24, no. 1, p. 113-142.
MATSUI, Taei; FUJIMARA, Yoshihiro and TITANI, Koiti. Snake venom proteases affecting hemostasis and thrombosis. Biochimica et Biophysica Acta, March 2000, vol. 1477, no. 1-2, p. 146-156. [CrossRef]
NAKARI-SETÄLÄ, Tiina; AZEREDO, Joana; HENRIQUES, Mariana; OLIVEIRA, Rosário; TEIXEIRA, José; LINDER, Markus and PENTTILÄ, Merja. Expression of a fungal hydrophobin in the Saccharomyces cerevisiae cell wall: effect on cell surface properties and immobilization. Applied and Environmental Microbiology, July 2002, vol. 68, no. 7, p. 3385-3391. [CrossRef]
NILSSON, Kjell; BIRNBAUM, Staffan and MOSBACH, Klaus. Microcarrier culture of recombinant Chinese hamster ovary cells for production of human immune interferon and human tissue-type plasminogen activator. Applied Microbiology and Biotechnology, January 1988, vol. 27, no. 4, p. 366-371. [CrossRef]
PARK, D.S.; KANG, I.; KIM, H.; CHUNG, K.; KIM, D.S. and YUN, Y. Cloning and characterization of novel disintegrins from Agkistrodon halys venom. Molecules and cells, October 1998, vol. 8, no. 5, p. 578-584.
RAMOS, O.H.P. and SELISTRE-DE-ARAUJO, H.S. Snake venom metalloproteases - structure and function of catalytic and disintegrin domains. Comparative Biochemistry and Physiology - Part C: Toxicology and Pharmacology, March-April 2006, vol.142, no. 3-4, p. 328-346. [CrossRef]
SCHMITMEIER, Stephanie; MARKLAND, Francis S.; RITTER, Matthew R.; SAWCER, David E. and CHEN, Thomas C. Functional effect of contortrostatin, a snake venom disintegrin, on human glioma cell invasion in vitro. Cell Communication and Adhesion, January-February 2003, vol. 10, no. 1, p. 1-16. [CrossRef]
SELISTRE-DE-ARAUJO, Heloisa S.; SOUZA, Dulce H.F. and OWNBY, Charlotte L. Analysis of a cDNA encoding a novel member of the snake venom metalloproteinase, disintegrin-like, cysteine-rich (MDC) protein family from Agkistrodom, contortrix lacticinctus. Biochimica et Biophysica Acta, October 1997, vol. 1342, no. 2, p. 109-115. [CrossRef]
SELISTRE-DE-ARAUJO, Heloisa S.; SOUZA, Eduardo L.; BELTRAMINI, Leila M.; OWNBY, Charlotte L. and SOUZA, Dulce H.F. Expression, refolding and activity of a recombinant non-hemorrhagic snake venom metalloprotease. Protein Expression and Purification, June 2000, vol. 19, no. 1, p. 41-47. [CrossRef]
SWENSON, Stephen; COSTA, Fritz; MINEA, Radu; SHERWIN, Russel P.; ERNST, William; FUJII, Gary; YANG, Dongyun and MARKLAND, Francis S. Jr. Intravenous liposomal delivery of the snake venom disintegrin contortrostatin limits breast cancer progression. Molecular Cancer Therapeutics, April 2004, vol. 3, no. 4, p. 499-511.
TAKAGI, Mutsumi; HIA, Hui Ching; JANG, Jin Ho and YOSHIDA, Toshiomi. Effects of high concentrations of energy sources and metabolites on suspension culture of Chinese hamster ovary cells producing tissue plasminogen activator. Journal of Bioscience and Bioengineering, 2001, vol. 91, no. 5, p. 512-521. [CrossRef]
VAN WEZEL, A.L. Growth of cell-strains and primary cells on micro-carriers in homogeneous culture. Nature, October 1967, vol. 216, no. 5110, p. 64-65. [CrossRef]
VARANI, James; DAME, Michael; REDISKE, John; BEALS, Ted F. and HILLEGAS, William. Substrate-dependent differences in growth and biological properties of fibroblasts and epithelial cells grown in microcarrier culture. Journal of Biological Standardization, January 1985, vol. 13, no. 1, p. 67-76. [CrossRef]
VARANI, James; SITRIN, Robert G. and HILLEGAS, William J. Expression of plasminogen activator and plasminogen activator inhibitor mRNA in human fibroblasts grown on different substrates. Cytotechnology, February 1992, vol. 9, no. 1-3, p. 157-162. [CrossRef]
WANG, Shao Hui; SHEN, Xiao Chuan; YANG, Guang Zheng and WU, Xiang Fu. cDNA cloning and characterization of Agkistin, a new metalloproteinase from Agkistrodon halys. Biochemical and Biophysical Research Communications, February 2003, vol. 301, no. 2, p. 298-303. [CrossRef]
WANG, Ji-Hong; WU, Yu; REN, Feng; LÜ, Li and ZHAO, Bao-Chang. Cloning and characterization of adinbitor, a novel disintegrin from the snake venom of Agkistrodon halys brevicaudus stejneger. Acta Biochimica et Biophysica Sinica, June 2004, vol. 36, no. 6, p. 425-429.
G. Differential carbohydrate metabolism in tumor and host. In: ARNOTT,
M.S.; VAN EYS, J. and WANG, Y.M. eds. Molecular interactions of
nutrition and cancer. Raven Press,
YANG, Rong-Sen; TANG, Chih-Hsin; CHUANG, Woei-Jer; HUANG, Tsang-Hai; PENG, Hui-Chin; HUANG, Tur-Fu and FU, Wen-Mei. Inhibition of tumor formation by snakevenom disintegrin. Toxicon, April 2005, vol. 45, no. 5, p. 661-669. [CrossRef]
ZHOU, Qing; SHERWIN, Russel P.; PARRISH, Catherine; RICHTERS, Valda; GROSHEN, Susan G.; TSAO-WEI, Denice and MARKLAND, Francis S. Contortrostatin, a dimeric disintegrin from Agkistrodon contortrix, inhibits breast cancer progression. Breast Cancer Research and Treatment, June 2000, vol. 61, no. 3, p. 249-260. [CrossRef]
Note: Electronic Journal of Biotechnology is not responsible if on-line references cited on manuscripts are not available any more after the date of publication.