The quality of collagenase containing tissue dissociation enzymes (TDEs) has a direct impact on the success of the cell isolation procedure (e.g., hepatocytes¹, islets²). Subsequent studies showed that both collagenase and a neutral protease were required to effectively dissociate cells from tissue. In the past, biochemical assays used to characterize the product served as a guide for purchase but the ultimate test was to assess the effectiveness of TDE product in a cell isolation procedure. VitaCyte’s characterization of purified class I (C1) and class II (C2) collagenase enzymes from Clostridium histolyticum by different biochemical assay procedures have provided useful comparative data for the most appropriate assays to characterize collagenase activity in any collagenase containing TDE products. The brief overview below provides an analysis of our current understanding of the pros and cons of different biochemical assays for collagenase.

Correlation of collagenase domain structure with functional assessment of collagenase: Historically, two broad categories of biochemical assays have been used for the characterize collagenase enzymes: assessment of enzyme functional activity by measuring the conversion of substrate to product or physical analysis of enzyme preparations. Collagenase activity was assessed by two broad classes of substrates: peptide substrates or macromolecular substrates that use native and denatured collagen. The most commonly used peptide substrates are the Pz and FALGPA peptides. The Pz peptide is used in the Wunsch assay. These assays are easy to perform, relatively consistent and are the primary assays many suppliers use to characterize collagenase products. The major disadvantages of these assays are they are biased to detect C2 as noted above. Further studies by Matsushita’s laboratory have shown that both degraded (no CBD) and intact C2 (1 CBD) are equivalent in degrading peptide substrates. These data are further supported by studies performed at VitaCyte that showed treatment of purified C1 or C2 with chymotrypsin resulted in > 85% loss of collagen degradation activity (CDA) when compared to untreated control. In contrast, only 7% of the Wunsch activity was lost when this same comparison was made.

Importance of understanding the biochemical properties and gene structure of Clostridium histolyticum collagenase: A critical assessment of collagenase assays requires an understanding of the biochemical characteristics of these enzymes. Research performed from 1950 to 1990 showed that there were two isoforms of C. histolyticum collagenase: type 1 or class 1 (C1) and type 2 or class 2 (C2)^3,4. These enzymes were initially identified by differences in substrate reactivity. C2 was more effective in degrading collagenase specific peptide substrates than C1. This difference is substantial: > 40 fold difference in the specific activity (units/mg protein) when the specific activity of purified C2 is compared to the specific activity of purified C1. These two enzymes degraded collagen synergistically with later reports indicating that each enzyme preferentially degrades different hyperactive sites on native collagen⁵.

At least seven different forms of C1 and C2 were identified by chromatographic purification methods⁴. It was not known if there were a number of C1 and C2 genes or if these were degraded forms of a “parent” C1 or C2 enzyme. This issue was resolved when Matsushita’s laboratory cloned the C1 and C2 genes and showed single copies of these genes were present in the genome, and subsequent expression led to synthesis of a single polypeptide chain⁶. They also showed that each enzyme contained four domains. The amino terminal ends of both C1 and C2 have a large catalytic domain that reflects 67 % of the amino acid residues in the protein. Following the catalytic domain and towards the carboxy end of the enzyme, C1 has one intervening domain followed by two collagen binding domains (CBDs). In contrast, C2 has two intervening domains followed by one CBD. The functions of the intervening domains are not known. The assignment of the domains as CBDs was determined by the ability of these portions of the molecule to bind native collagen independent of any other protein domains. Further studies by Matsushita’s laboratory using recombinant C1 and C2 confirmed the difference substrate reactivity of these enzymes. Only those enzymes with a catalytic domain and at least one CBD were effective in degrading native collagen. Intact or degraded C1 or C2 could cleave denatured collagen (i.e., gelatin) or peptide substrates.

The domain structures of the collagenase enzymes are shown below. At this time it is believed that the yellow sections reflect linking segments of the protein that are thought to be “random coil” and not part of any domain structure. It is these areas that are expected to be susceptible to degradation by other proteases. Other sites are likely to exist but are not identified at this time. This susceptibility to proteolytic cleavage explains the multiple forms of enzyme that were seen after purification of crude collagenase⁴.

click to enlarge

Correlation of collagenase domain structure with physical and functional assessment of the enzyme: Historically, two broad categories of biochemical assays have been used to characterize collagenase enzymes: assessment of enzyme functional activity by measuring the conversion of substrate to product or physical analysis of the enzyme preparations. Collagenase activity was assessed by two broad classes of substrates: peptide substrates or macromolecular substrates that use native or denatured collagen. The most commonly used peptide substrates are the Pz⁷ and FALGPA⁸ peptides. The Pz peptide is used in the Wunsch assay. These assays are easy to perform, relatively consistent and are the primary assays many suppliers use to characterize collagenase products. The major disadvantages of these assays are they are biased to detect C2 as noted above. Further studies by Matsushita’s laboratory have shown that both intact C2 (1 CBD) and degraded C2 (no CBD) have equivalent activity in degrading peptide substrates⁹. These data are further supported by studies performed at VitaCyte that showed treatment of purified C1 or C2 with chymotrypsin resulted in loss of > 84% C1 and > 97% of C2 collagen degradation activity (CDA) when compared to an untreated control. However, a similar comparison using the same materials showed only 7% loss of the C2 Wunsch activity¹⁰.

The macromolecular assays use denatured or native collagen as the substrate. Gelatinase assays use gelatin (i.e., denatured collagen) in a number of forms as the substrate (soluble or insoluble gelatin - azocoll)³. Assay of gelatinase activity provides acceptable reproducibility and linearity but their drawbacks are a lack of specificity¹¹. Many proteases can degrade gelatin as shown by the common use of this substrate in zymography methods¹². In addition, Matsushita’s studies have shown that intact or degraded collagenases (± CBD) can degrade gelatin. In contrast, assays using native collagen measure collagenase activity since proteases such as trypsin are ineffective in degrading collagen. The Mandl assay¹³ or its subsequent modifications¹⁴ measured CDA using insoluble collagen fibers as substrate. These assays are performed in two steps: step 1 is the degradation of collagen while step 2 assays the free α amino groups released in the supernatant after collagen degradation. These assays are time consuming (Mandl assay 5 hours incubation in step 1) and have poor reproducibility with limited linearity of response. VitaCyte has developed a simplified fluorescent CDA assay using fluorescein isothiocyanate conjugated to soluble calf skin collagen fibrils. The assay is performed using a microplate with the results read on a fluorescent microplate reader¹⁰. The primary advantages of this assay are shorter turnaround time (< 90 min), linearity, ability to run multiple samples, and specificity for collagenase. Purified bovine trypsin at 5.4 molar excess over substrate had a specific CDA of < 1% of purified C1 or C2¹⁰. A summary of these assays and their requirements for CBDs for activity are presented in the table below.

click to enlarge

Correlation of collagenase domain structure with physical analysis: An excellent assay used to assess the quality of collagenase is analytical anion exchange high pressure liquid chromatography (HPLC). This procedure was first used by Hefley to show that C1 and C2 could be separated by using an anion exchange resin¹⁵. The primary advantage of this analysis is that it is able to resolve different molecular forms of C1. Figure 1 is an HPLC chromatogram of representative analysis of a 60:40 C1:C2 purified enzyme mixture that contained C2 and three different forms of C1: intact C1, C1b, and C1c. Correlation of these peaks with CDA and SDS-PAGE showed the following. The earliest eluting peak is C2 with a retention time of ~13 min. Intact C1 enzyme with two collagen binding domains eluted next at ~ 20 min with the other two molecular forms, C1b and C1c, eluting at ~21 and ~ 23 min, respectively. Intact C2 and C1 had apparent molecular weights of about 116 kDa while C1b or C1c had apparent molecular weights of approximately 100 kDa. Correlation of fractions collected after HPLC separation with CDA and SDS-PAGE analysis showed that the specific CDA correlates with the number of CBDs present in the molecular form of the enzyme. Intact C1 contains two CBDs with proteolytic cleavage of the C terminal CBD resulting in the appearance of the 100 kDa form of the enzymes (C1b, C1c). Intact C1 has a specific CDA approximately 10 fold higher than enzymes forms that contain one CBD (intact C2, C1b, C1c)¹⁰.

Analytical HPLC Profile

click to enlarge

The main advantage of anion exchange HPLC analysis is characterization of the C1 collagenase. It is currently unable to provide the same resolution for the quality of the C2. In contrast, analysis of similar collagenase preparation analyzed above using reversed phase HPLC only revealed two peaks, C1 and C2. This difference reflects the resolution capabilities and conditions used to perform these analyses. The resin used in analytical anion exchange HPLC can separate proteins based on very small differences in charge. Intact C1, C1b, and C1c represent different molecular forms of the enzyme with intact C2, intact C1, C1b and C1c having different charges based on the charge of the amino acids found in each form of collagenase. In contrast, the hydrophobic resins used in the reversed phase HPLC procedure can only resolve hydrophobic differences in C1 and C2 and not in the degraded forms of either enzyme.

SDS-PAGE has also been used to characterize collagenase enzymes. However, unless performed under tightly controlled conditions, proteolytic artifacts are often created during sample preparation. Clostripain is a bacterial protease found in most C. histolyticum collagenase preparations. The contamination of this enzyme in purified collagenase preparations can be minimized but not eliminated. To obtain an accurate assessment of collagenase by SDS-PAGE, it is critical to neutralize the clostripain activity by supplementing the SDS-PAGE sample buffer with 15 mM EDTA and omitting the addition of reducing agent. These two steps will eliminate clostripain activity. Clostripain is a calcium dependent enzyme that expresses maximal activity under reducing conditions (i.e., sulfhydryl dependent enzyme). If these steps are not taken, the more thermostable clostripain will rapidly degrade the denatured C1 or C2, leading to artifactual bands.

Summary: The key criteria for successful analysis of collagenase containing tissue dissociation enzymes is the correlation of the biochemical assay results to the structure function capabilities of the enzyme. The review above indicates the limitations of Wunsch or FALGPA activity for assessing the quality of collagenase used in tissue dissociation applications. This is supported by data that showed total Wunsch activity used in human islet isolation was not predictive of the success of the isolation procedure ¹⁴. Assessment of gelatinase has similar limitations in addition to the lack of specificity for collagenase activity. The best combination of assays to assess the quality and activity of collagenase are analytical anion exchange HPLC and fluorescent microplate CDA assay. The analytical anion exchange HPLC provides and overall assessment of the quality of collagenase whereas the CDA assay provides an assessment of the functional activity required to degrade native collagen. These analyses are provided to customers in VitaCyte’s Certificate of Analysis. However, neither of these assays is commercially available but are offered as a service to customers. Please refer to the Products & Services page for additional details on these analyses.

Reference List

1. Berry M.N., Edwards A.M., and Barritt G.J. (1991) Isolated hepatocytes, preparation, properties, and applications. Elsevier, New York.
2. Barnett M.J., Zhai X., LeGatt D.F., Cheng S.B., Shapiro A.M.J., and Lakey J.R.T. (2005) Quantitative assessment of collagenase blends for human islet isolation. Transplantation 80, 723-728.
3. Mandl I., Keller S., and Manahan J. (1964) Multiplicity of Clostridium histolytcum collagenases. Biochemistry 3, 1737-1741.
4. Mookhtiar K.A. and Van Wart H.E. (1992) Clostridium histolyticum collagenases: a new look at some old enzymes. Matrix Suppl 1, 116-26.
5. French M.F., Bhown A., and Van Wart H.E. (1992) Identification of Clostridium histolyticum collagenase hyperreactive sites in type I, II, and III collagens: lack of correlation with local triple helical stability. Journal of Protein Chemistry 11, 83-97.
6. Matsushita O. and Okabe A. (2001) Clostridial hydrolytic enzymes degrading extracellular components. Toxicon 39, 1769-80.
7. Wünsch E. and Heidrich H.-G. (1963) Zur quantitativen bestimmung der kollagenase. Hoppe-Seyler’s Zeitschrift Physiologische Chemie 333, 149-151.
8. Van Wart H.E. and Steinbrink D.R. (1981) A continuous spectrophotometric assay for Clostridium histolyticum collagenase. Analytical Biochemistry 113, 356-65.
9. Matsushita O., Jung C.M., Minami J., Katayama S., Nishi N., and Okabe A. (1998) A study of the collagen-binding domain of a 116-kDa Clostridium histolyticum collagenase. Journal of Biological Chemistry 273, 3643-8.
10. McCarthy R.C., Spurlin B., Wright M.J., Breite A.G., Sturdevant L.K., Dwulet C.S., and Dwulet F.E. (2008) Development and characterization of a collagen degradation assay to assess purified collagenase used in islet isolation. Transplantation Proceedings 40, 339-42.
11. Poilane I., Karjalainen T., Barc M.C., Bourlioux P., and Collignon A. (1998) Protease activity of Clostridium difficile strains. Canadian Journal of Microbiology 44, 157-61.
12. Frederiks W.M. and Mook O.R. (2004) Metabolic mapping of proteinase activity with emphasis on in situ zymography of gelatinases: review and protocols. Journal of Histochemistry & Cytochemistry 52, 711-722.
13. Mandl I., MacLennan J.D., and Howes E.L. (1953) Isolation and characterization of proteinase and collagenase from Cl. histolyticum. Journal of Clinical Investigation 32, 1323-1329.
14. Peterkofsky B. (1982) Bacterial Collagenase. Methods in Enzymology 82, 453-471.
15. Hefley T.J. (1987) Utilization of FPLC-purified bacterial collagenase for the isolation of cells from bone. J Bone and Mineral Research 2, 505-16.

Tags: Assays, Collagenase
Categories: Applications, Products, Services