Improved method of producing collagenase

Abstract

To provide an improved method of producing collagenase.SOLUTION: A drug product contains isolated and purified collagenase and is essentially free of neutral protease.SELECTED DRAWING: None

Description

[Technical field]

[0001] Related Applications This application claims priority to U.S. Provisional Application No. 62 / 477,846, filed March 28, 2017, which is incorporated by reference in its entirety to the fullest extent permitted by law.

[0002] The present invention relates to the field of collagenase production and collagenase products, and in particular to improvements in the reproducibility, purity, and stability of collagenase I and collagenase II compositions, which are at least 95% area pure as measured by reverse phase high pressure liquid chromatography (RP-HPLC) and essentially free of neutral proteases. [Background technology]

[0003] A process for producing highly pure (at least 95% pure) collagenase from fermentation of C. histolyticum has been previously described in U.S. Patent No. 5,399,663. Collagenase I and II are enzymes that degrade collagen, the most abundant structural protein in mammals. These enzymes are used for the treatment of various collagen-mediated diseases, such as Dupuytren's contracture, Peyronie's disease, lipomas, and adhesive capsulitis. U.S. Patent No. 5,399,663 and U.S. Patent No. 5,399,663 disclose the use of collagenase preparations in the treatment of Dupuytren's disease. U.S. Patent No. 5,399,663 disclose the use of collagenase preparations in the treatment of Peyronie's disease. U.S. Patent No. 5,399,663 and U.S. Patent No. 5,399,663 disclose the use of collagenase in the treatment of lipomas. U.S. Patent No. 5,399,663 disclose the use of collagenase in the treatment of adhesive capsulitis. U.S. Patent No. 5,399,663 and U.S. Patent No. 5,399,663 disclose the use of collagenase to treat cellulite.

[0004] The main source of collagenase is from the fermentation of C. histolyticum. An injectable formulation containing C. histolyticum collagenase I and collagenase II is sold under the trade name XIAFLEX® and is approved by the U.S. Food and Drug Administration for the treatment of Dupuytren's contracture and Peyronie's disease. The injectable formulation is called XIAPEX® in Europe and other countries. The amino acid sequences for collagenase I and collagenase II are encoded by the colG and colH genes, respectively. The amino acid sequence for colG is described in GenBank Accession No. D87215 and Non-Patent Document 1, and the amino acid sequence for colH is described in GenBank Accession No. D29981 and Non-Patent Document 2. Collagenase AUXI has a single polypeptide chain of approximately 1000 amino acids with a molecular weight of about 113 kDa. Collagenase AUXII also has a single polypeptide chain of approximately 1000 amino acids with a molecular weight of about 112 kDa.

[0005] Collagenase I and collagenase II are metalloproteases that require tightly bound zinc and loosely bound calcium for their activity (Non-Patent Document 3). Both collagenase I and collagenase II have broad specificity for all types of collagen (Non-Patent Document 4). These collagenases digest collagen by hydrolyzing the triple helical region of collagen under physiological conditions (Non-Patent Document 4). Although each collagenase exhibits different specificity (i.e., each has a different preferred amino acid sequence for cleavage), together they have synergistic activity against collagen (Non-Patent Document 5; Non-Patent Document 6).

[0006] Collagenase for use in therapy may be obtained from a variety of sources, including mammalian, fungal, and bacterial sources. One common source of crude collagenase is from bacterial fermentation processes, specifically, from the fermentation of Clostridium histolyticum (C. histolyticum). Crude collagenase obtained from C. histolyticum can be purified using any of a number of chromatographic techniques. However, many other enzymes besides collagenase are secreted into the fermentation broth, including various toxins.

[0007] As previously described in US Patent No. 5,999,136, genome sequencing and analysis of the toxins secreted from C. histolyticum strain 004 was performed to investigate the function of these toxins during fermentation to produce collagenase. The toxins investigated were alpha toxin (lethal factor), beta toxin (type I and type II collagenases), gamma toxin (clostripain), delta toxin (neutral protease), and epsilon toxin (oxygen-labile hemolysin). Based on genome sequence analysis, only collagenase and clostripain were considered to be functional toxins of strain 004. The other toxins were considered to be non-functional based on essential amino acid sequence differences between the C. histolyticum proteins and each model protein.

[0008] Delta toxin, also known as neutral protease (NP), is a metalloprotease from C. histolyticum. Neutral protease is a member of the M4 family of metalloproteases, and according to the MEROPS peptidase database, neutral proteases share structural similarities with the M4 family, including consensus sequence motifs, substrate specificity, and cofactor requirements. Thermolysin is the most studied and understood enzyme of the M4 family. In connection with this study, thermolysin from Bacillus thermoproteolyticus was previously cloned and sequenced, and the information was deposited in GenBank under accession number CAA54291. Thermolysin is a zinc metalloprotease with a mature enzyme molecular weight of 34.6 kDa. Thermolysin and all members of the M4 peptide family consist of a signal peptide, a prosequence, and a mature sequence.

[0009] When thermolysin is produced by cells, it starts as an inactive enzyme (proenzyme) because a protein prosequence preceding the mature enzyme sequence inhibits thermolysin. The prosequence represents two-thirds of the size of the secreted proenzyme, while the mature enzyme represents one-third. The prosequence in thermolysin is autocatalytically cleaved, resulting in activation of the mature enzyme in its target environment. The thermolysin secretion strategy is shared with neutral proteases and was the basis for concluding that C. histolyticum neutral protease also starts as an inactive protease. Similarly, C. histolyticum neutral protease in its mature form has a molecular weight similar to thermolysin, as described in Herber (Patent Document 10) and Non-Patent Document 7.

[0010] The similarity of thermolysin in its synthesis, secretion, activation, and substrate specificity makes it a model for the neutral protease from C. histolyticum strain 004. These attributes between thermolysin and neutral protease are similar based on shared structural features, but sequence differences exist between the two enzymes. In relation to the present invention, previous genomic analysis of delta toxin and homology comparison to thermolysin in Herber (US Pat. No. 5,399,633) concluded that the neutral protease may have been secreted into the growth medium but was not active due to a different consensus sequence in the autocatalytic site.

[0011] A distinctive feature of thermolysin (and all neutral proteases) is the broad range of proteins it can act on. The presence of non-specific proteases in medicines, especially those that are injected into the body, is problematic because they, even when present in trace amounts, will undesirably degrade non-targeted enzymes, cells, and tissues in the body. [Prior art documents] [Patent documents]

[0012] [Patent Document 1] U.S. Patent No. 7,811,560 [Patent Document 2] U.S. Patent No. 6,086,872 [Patent Document 3] U.S. Patent No. 5,589,171 [Patent Document 4] U.S. Patent No. 6,022,539 [Patent Document 5] U.S. Patent No. 6,958,150 [Patent Document 6] U.S. Patent No. 7,842,673 [Patent Document 7] U.S. Patent Application Publication No. 2006 / 020448A1 [Patent Document 8] U.S. Patent Application Publication No. 2014 / 0335072 [Patent Document 9] U.S. Patent Application Publication No. 2016 / 0279046 [Patent Document 10] U.S. Patent Application Publication No. 2015 / 0010532 [Non-patent literature]

[0013] [Non-Patent Document 1] Matsushita et al. (1999), Journal of Bacteriology 181(3):923-933 [Non-Patent Document 2] Yoshihara et al. (1994), Journal of Bacteriology 176(21):6489-6496 [Non-Patent Document 3] Eddie L.Angleton and HEVan Wart,Biochemistry1988,27,7406-7412 [Non-Patent Document 4] Steinbrink, D; Bond, M and Van Wart, H; (1985), JBC, 260p2771-2776 [Non-Patent Document 5] Mandl, I., (1964), Biochemistry, 3: p. 1737-1741 [Non-Patent Document 6] Vos-Scheperkeuter, GH, (1997), Cell Transplantation, 6: p. 403-412 [Non-Patent Document 7] Maeda et al., Cloning a neutral protease of Clostridium histolyticum, determining its substrate specificity, and designing a specific substrate, Appl.Microbiol.Biotech.,99:10489-99(2015) Summary of the Invention [Problem to be solved by the invention]

[0014] Previous processes for making collagenase can sometimes result in degradation of the atypical collagenase. For efficient and effective commercial production of collagenase, reproducible production of toxin-free, highly pure collagenase is required. [Means for solving the problem]

[0015] The present disclosure provides an improved process for collagenase production that removes detectable amounts of neutral protease from collagenase I and II products. The present disclosure provides a collagenase product that is less susceptible to degradation and more pure than previous collagenase compositions for medical use. In one embodiment, an investigation was conducted to identify the cause of atypical collagenase I and II degradation from C. histolyticum during the manufacturing process. Neutral protease from C. histolyticum was found to be present in the fermentation broth of C. histolyticum in its active mature form. Furthermore, this neutral protease is highly active against collagenase I and collagenase II and may cause significant degradation of both collagenases. Its continued presence during the purification process may also degrade both collagenases.

[0016] Neutral proteases, previously thought to be inactive, unexpectedly become broadly more active in the presence of higher amounts of one or more metals, such as nickel and zinc. This metal-mediated activity produces high levels of collagenase product fragments during collagenase manufacturing processes in the presence of parts per million levels of such metals, causing purity defects at least in the ion exchange (IEX) chromatography step of the process.

[0017] The present invention also provides methods for isolating neutral protease from the manufacturing process, removing it from the process, and determining its presence throughout the manufacturing process. Additionally, the present invention provides improved reproducibility and stability of highly purified collagenase, provides collagenase I and II products that are essentially free of neutral proteases (i.e., below detectable limits), and provides safer and purer collagenase compositions that are essentially free of neutral proteases.

[0018] Manufacturing reproducibility is improved by controlling metals during the fermentation and purification process. Surprisingly, it has been found that the presence or absence of certain metals during purification of collagenase I and II from C. histolyticum affects the purity of the collagenase product. In one embodiment, zinc levels of about 80 ppm or more, or nickel levels of about 1 ppm or more, result in high levels of collagenase I and collagenase II product impurities as measured by SDS-PAGE (sodium dodecyl sulfate polyacrylamide gel electrophoresis), resulting in unacceptable levels of impurities for drug product.

[0019] In another embodiment, collagenase is not degraded into fragments when zinc levels are greater than about 2.75 ppm and less than about 80 ppm. Also, collagenase is not degraded into fragments when nickel levels are greater than about 0.4 ppm and less than about 1 ppm in at least one step of the purification process. Furthermore, controlling the drug purification process to maintain zinc levels of about 3 ppm or less, or nickel levels of about 0.5 ppm or less, improves the reliability and reproducibility of the production of collagenase products with high purity. Running a collagenase production process with trace amounts of metallic zinc and nickel (or other metals) that do not cause high levels of collagenase degradation provides significantly improved reliability and reproducibility compared to previous processes.

[0020] In certain embodiments, when nickel or zinc is present in the buffer for the hydrophobic interaction chromatography (HIC) step of the collagenase manufacturing process, neutral proteases aggressively degrade collagenase I and II into smaller fragments and result in collagenase products of low purity (less than 95% purity as measured by SDS-PAGE). In some embodiments, nickel (in amounts of about 0.4 ppm to about 1 ppm) and zinc (in amounts of about 2.8 ppm to 84 ppm) do not amplify neutral protease specific activity in the HIC and tangential flow filtration (TFF) unit operations and do not cause significantly increased product fragment impurities in the IEX unit operations. In one embodiment, control of the collagenase manufacturing process for nickel or zinc levels is important to obtain collagenase I and II products of high purity.

[0021] By eliminating or at least reducing neutral proteases from the collagenase manufacturing process so that the collagenase composition is essentially free of neutral proteases, the safety of the product is improved and the degradation of collagenase in the product by neutral proteases is prevented. The collagenase composition is improved by adding at least one neutral protease elimination step in the manufacturing process. The reduction or elimination of neutral proteases from the manufacture of collagenase products can be achieved by one or more of the following elimination steps: (a) excluding any fractions having less than about 90% collagenase I or II content and more than about 7% content of any single impurity as measured by SDS PAGE gel or HPLC; (b) excluding any fractions having detectable levels of neutral proteases as measured by SDS PAGE gel or zymography; or (c) excluding any fractions calculated to contain excessive levels of neutral proteases. Each can be used individually or in combination.

[0022] More specifically, the elimination step may involve rejecting fractions collected after passing collagenase through a separation column or filter (or other separation medium) and containing detectable levels of neutral protease when tested by SDS-PAGE or zymography assays prior to pooling of collagenase I or collagenase II fractions. In certain embodiments, fractions may be rejected even if they contain significant amounts of collagenase. The elimination may also entail rejecting or discarding any collagenase I or II fractions having less than about 90% collagenase I or II content and more than about 7% content of any single impurity as measured by SDS PAGE gel or HPLC. The elimination step may involve the use of an estimated amount of collagenase II produced in the process to control for which collagenase I fractions are pooled for collagenase I product. Any of these steps can be used in the collagenase manufacturing process to produce collagenase I and II products essentially free of neutral proteases.

[0023] One embodiment of neutral protease exclusion performs process control on the purity of ion exchange chromatography (IEX) fractions, where each fraction is analyzed for both purity and the most abundant impurity, and each fraction must meet this criterion in order to process the fraction forward (acceptable for use as a pool, drug product, and pharmaceutical formulation). In this embodiment, the positional nature of the neutral protease elution at the end of the collagenase I peak of the IEX step provides an effective mechanism for removal of neutral proteases by automatically rejecting fraction(s) that exhibit detectable limits of neutral protease by SDS-PAGE, SDS-PAGE with densitometry, or zymography (with or without densitometry) assays. In another embodiment, the peak fractions of collagenase I fractions collected from the IEX elution are processed forward based on fractions that meet the desired purity and impurity criteria, along with pooling only the collagenase I peak fractions until the approximate amount of pooled collagenase I matches the approximate amount of collagenase II produced by the process. This control strategy results in the elimination of further end fractions of the collagenase I peak, such that the pooled collagenase I fractions are essentially free of neutral proteases.

[0024] The present disclosure further provides a method for preparing a pharmaceutical formulation according to the present disclosure, and a method for treating a patient suffering from a collagen-mediated disease with the collagenase composition of the present invention. For example, a method of treating a collagen-mediated disease by administering an effective amount of collagenase, collagenase I, collagenase II, or a combination thereof is contemplated. In one embodiment, the present disclosure relates to a method of treating a collagen-mediated condition in a patient in need of treatment of the collagen-mediated condition (e.g., cellulite), comprising administering to the patient an effective amount of a collagenase composition, the collagenase composition comprising a collagenase I product and a collagenase II product combined in a mass ratio of about 0.6:1.4 collagenase I to collagenase II to about 1.4:0.6 collagenase I to collagenase II, the composition being essentially free of neutral proteases. Additionally, the present disclosure provides a more stable and safer pharmaceutical formulation.

[0025] Further embodiments of the present compositions and methods will be apparent from the following description, drawings, examples, and claims. As can be understood from the above and below description, any and all features described herein, and any and all combinations of two or more of such features, are included within the scope of the present disclosure, unless the features included in such combinations are mutually inconsistent. Furthermore, any feature or combination of features may be specifically excluded from any embodiment or aspect. Further aspects and embodiments are set forth in the following description and claims, particularly when considered in conjunction with the accompanying examples and drawings.

[0026] References to color drawings The application file contains at least one drawing executed in color. Copies of this application publication with color drawing(s) will be provided by the Office upon request and payment of the appropriate fee.

[0027] That is, the gist of the present invention relates to the following. Item 1 A drug product comprising isolated and purified collagenase, the drug product being essentially free of neutral proteases. Section 2 Item 1. The drug product of item 1, wherein the product contains less than about 100 ng of neutral protease per mg of the product. Section 3 Item 10. The drug product of item 1, wherein the product contains less than about 75 ng of neutral protease per mg of the product. Section 4 Item 1. The drug product of item 1, wherein the product contains less than about 50 ng of neutral protease per mg of the product. Section 5 Item 10. The drug product of item 1, wherein the product contains less than about 25 ng of neutral protease per mg of the product. Section 6 Item 2. The pharmaceutical product of item 1, wherein the collagenase comprises collagenase I and collagenase II from C. histolyticum. Section 7 Item 2. The drug product of item 1, wherein the collagenase is at least one of collagenase I and collagenase II. Section 8 Item 8. The pharmaceutical product of item 7, wherein the collagenase I and collagenase II are present in an approximately 1:1 ratio. Section 9 1. An isolated and purified collagenase I obtained or derived from C. histolyticum, the collagenase I being essentially free of neutral proteases. Section 10 Item 10. The collagenase I described in Item 9, wherein the collagenase I does not contain a detectable amount of neutral protease. Section 11 Item 10. The collagenase I of item 9, wherein the collagenase I contains less than about 1000 ng of neutral protease per mg of collagenase I. Section 12 Item 10. The collagenase I of item 9, wherein the collagenase I contains less than about 500 ng of neutral protease per mg of collagenase I. Section 13 A method for isolating and purifying collagenase I and collagenase II obtained or derived from C. histolyticum, the method comprising controlling metal levels during purification of collagenase I and collagenase II. Section 14 14. The method of claim 13, wherein the process is exposed to low levels of a metal selected from the group consisting of nickel, zinc, aluminum, arsenic, calcium, cadmium, chromium, copper, iron, magnesium, and lead. Item 15 15. The method of claim 14, wherein the nickel is present in an amount less than about 2.0 ppm. Section 16 15. The method of claim 14, wherein the nickel is present in an amount less than about 1.2 ppm. Section 17 15. The method of claim 14, wherein the nickel is present in an amount less than about 1.0 ppm. Section 18 15. The method of claim 14, wherein the nickel is present in an amount less than about 0.5 ppm. Section 19 15. The method of claim 14, wherein the nickel is present in an amount selected from the group consisting of less than about 0.9 ppm, less than about 0.8 ppm, less than about 0.7 ppm, less than about 0.6 ppm, less than about 0.5 ppm, less than about 0.4 ppm, less than about 0.3 ppm, less than about 0.2 ppm, less than about 0.1 ppm, less than about 0.09 ppm, less than about 0.08 ppm, less than about 0.07 ppm, less than about 0.06 ppm, less than about 0.05 ppm, less than about 0.04 ppm, less than about 0.03 ppm, less than about 0.02 ppm, or less than about 0.01 ppm. Section 20 15. The method of claim 14, wherein the zinc is present in an amount less than about 80 ppm. Section 21 15. The method of claim 14, wherein the zinc is present in an amount less than about 50 ppm. Section 22 15. The method of claim 14, wherein the zinc is present in an amount less than about 25 ppm. Section 23 15. The method of claim 14, wherein the zinc is present in an amount less than about 10 ppm. Section 24 15. The method of claim 14, wherein the zinc is present in an amount less than about 5 ppm. Section 25 15. The method of claim 14, wherein the zinc is present in an amount less than about 3 ppm. Section 26 15. The method of claim 14, wherein the zinc is present in an amount of less than about 1 ppm. Section 27 A method for making isolated and purified collagenase, comprising preparing said collagenase in a process that controls and limits metal content, and comprising a exclusion step for separating multiple highly pure fractions of collagenase from neutral proteases. Section 28 28. The method of claim 27, wherein the collagenase is collagenase I and the exclusion step isolates a collagenase I fraction. Section 29 28. The method of claim 27, wherein the collagenase is collagenase II and the elimination step isolates a fraction of collagenase II. Item 30 28. The method of claim 27, wherein the collagenase comprises collagenase I and collagenase II, and highly pure fractions of collagenase I and collagenase II are mixed, and the resulting mixture has less than 50 ng of neutral protease per mg of mixture. Section 31 28. The method of claim 27, wherein the process is limited to low levels of metals selected from the group consisting of nickel, zinc, aluminum, arsenic, calcium, cadmium, chromium, copper, iron, magnesium, and lead. Section 32 32. The method of claim 31, wherein the metal is nickel and the metal is present in an amount less than about 2.0 ppm. Item 33 32. The method of claim 31, wherein the metal is nickel and the metal is present in an amount less than about 1.2 ppm. Section 34 32. The method of claim 31, wherein the metal is nickel and the metal is present in an amount less than about 1.0 ppm. Section 35 32. The method of claim 31, wherein the metal is nickel and the metal is present in an amount less than about 0.5 ppm. Section 36 32. The method of claim 31, wherein the metal is nickel and the metal is present in an amount selected from the group consisting of less than about 0.9 ppm, less than about 0.8 ppm, less than about 0.7 ppm, less than about 0.6 ppm, less than about 0.5 ppm, less than about 0.4 ppm, less than about 0.3 ppm, less than about 0.2 ppm, less than about 0.1 ppm, less than about 0.09 ppm, less than about 0.08 ppm, less than about 0.07 ppm, less than about 0.06 ppm, less than about 0.05 ppm, less than about 0.04 ppm, less than about 0.03 ppm, less than about 0.02 ppm, or less than about 0.01 ppm. Section 37 32. The method of claim 31, wherein the metal is zinc and the metal is present in an amount less than about 80 ppm. Section 38 32. The method of claim 31, wherein the metal is zinc and the metal is present in an amount less than about 50 ppm. Section 39 32. The method of claim 31, wherein the metal is zinc and the metal is present in an amount less than about 25 ppm. Section 40 32. The method of claim 31, wherein the metal is zinc and the metal is present in an amount less than about 10 ppm. Section 41 32. The method of claim 31, wherein the metal is zinc and the metal is present in an amount less than about 5 ppm. Section 42 32. The method of claim 31, wherein the metal is zinc and the metal is present in an amount less than about 3 ppm. Section 43 32. The method of claim 31, wherein the metal is zinc and the metal is present in an amount of less than about 1 ppm. Section 44 32. The method of claim 31, wherein the elimination step comprises testing each fraction on an SDS-PAGE gel and pooling only those fractions that do not contain a detectable amount of neutral protease on the gel. Section 45 A method for producing isolated and purified collagenase I obtained or derived from C. histolyticum, the method comprising: (a) separating multiple highly pure fractions of collagenase I from neutral proteases; and (b) pooling the highly pure fractions of collagenase I into a collagenase I drug substance. Section 46 46. ​​The method of claim 45, wherein the separating step comprises testing each fraction for the presence of neutral protease using zymography, and the pooling step comprises pooling only fractions containing collagenase I that do not contain a detectable amount of neutral protease on a zymography gel. Section 47 46. ​​The method of claim 45, wherein said collagenase I drug substance is essentially free of neutral proteases. Section 48 46. ​​The method of claim 45, wherein the collagenase I drug substance contains less than 100 ng neutral protease per mg of drug substance. Section 49 46. ​​The method of claim 45, wherein the collagenase I drug substance contains less than 75 ng neutral protease per mg of drug substance. Item 50 46. ​​The method of claim 45, wherein the collagenase I drug substance contains less than 50 ng neutral protease per mg of drug substance. Section 51 46. ​​The method of claim 45, wherein the collagenase I drug substance contains less than 25 ng neutral protease per mg of drug substance. Section 52 46. ​​The method of claim 45, wherein the collagenase I drug substance contains less than 20 ng neutral protease per mg of drug substance. Section 53 46. ​​The method of claim 45, wherein the collagenase I drug substance contains less than 10 ng neutral protease per mg of drug substance. Section 54 1. A process for producing isolated and purified collagenase I obtained or derived from C. histolyticum, comprising: fermenting aC histolyticum; b. recovering a crude fermentation broth containing collagenase I and collagenase II; c. Purifying collagenase I and collagenase II, i. filtering the crude fermentation liquor through an anion exchange filter; ii. precipitating the filtrate from step (c)(i) with ammonium sulfate containing less than about 80 ppm zinc and less than about 1.2 ppm nickel; iii. resuspending the precipitated product from step (c)(ii); iv. subjecting the resuspension from step (c)(iii) to a hydrophobic interaction chromatography column; v. separating the collagenase I and collagenase II from the filtrate of step (c)(iv) using ion exchange chromatography; and vi. collecting the filtrate of step (c)(v) in fractions; d. estimating the total amount of collagenase II collected in each fraction from step (c)(vi); e. estimating the amount of collagenase I collected in each fraction from step (c)(vi); f. Pooling fractions containing collagenase I from the peak of collagenase I in the fraction toward the tail of collagenase I in the fraction until the amount of collagenase I pooled is approximately the same as the amount of collagenase II estimated in step (d). Section 55 55. The process of claim 54, wherein the ammonium sulfate contains less than about 80 ppm zinc and less than about 1.2 ppm nickel. Section 56 1. A process for producing isolated and purified collagenase I obtained or derived from C. histolyticum, comprising: fermenting aC histolyticum; b. recovering a crude fermentation broth containing collagenase I and collagenase II; c. Purifying collagenase I and collagenase II, i. filtering the crude fermentation liquor through an anion exchange filter; ii. precipitating the filtrate from step (i) with ammonium sulfate that is substantially free of transition metals; iii. resuspending the precipitated product from step (ii); iv. subjecting the resuspension from step (iii) to a hydrophobic interaction chromatography column; v. separating the collagenase I and collagenase II from the filtrate of step (iv) using ion exchange chromatography; and vi. collecting the filtrate of step (v) in fractions; d. separating multiple highly pure fractions of collagenase I from neutral protease; e. Pooling said highly pure fractions of collagenase I. Section 57 18. The process of claim 17, wherein the separating step (d) comprises testing each fraction on an SDS-PAGE gel, and the pooling step (e) comprises pooling only fractions containing collagenase I that do not contain detectable amounts of neutral protease on the gel. Section 58 1. A process for making a drug product consisting of isolated and purified collagenase I and collagenase II obtained or derived from C. histolyticum collagenase I and collagenase II, respectively, wherein said collagenase I and collagenase II have a mass ratio of about 1 to 1, and said drug product is at least 95% area pure as determined by reverse phase high performance liquid chromatography; fermenting aC histolyticum; b. recovering a crude fermentation broth containing collagenase I and collagenase II; c. purifying collagenase I and collagenase II from the crude recovery solution via filtration and column chromatography, comprising treating the collagenase I and collagenase II with an ammonium sulfate solution having a zinc level of less than about 80 ppm and a nickel level of less than about 1.2 ppm. Section 59 59. The method of claim 58, wherein the process is limited to low levels of metals selected from the group consisting of nickel, zinc, aluminum, arsenic, calcium, cadmium, chromium, copper, iron, magnesium, and lead. Section 60 60. The method of claim 59, wherein the nickel is present in an amount less than about 2.0 ppm. Section 61 60. The method of claim 59, wherein the nickel is present in an amount less than about 1.2 ppm. Section 62 60. The method of claim 59, wherein the nickel is present in an amount less than about 1.0 ppm. Section 63 60. The method of claim 59, wherein the nickel is present in an amount less than about 0.5 ppm. Section 64 60. The method of claim 59, wherein the nickel is present in an amount selected from the group consisting of less than about 0.9 ppm, less than about 0.8 ppm, less than about 0.7 ppm, less than about 0.6 ppm, less than about 0.5 ppm, less than about 0.4 ppm, less than about 0.3 ppm, less than about 0.2 ppm, less than about 0.1 ppm, less than about 0.09 ppm, less than about 0.08 ppm, less than about 0.07 ppm, less than about 0.06 ppm, less than about 0.05 ppm, less than about 0.04 ppm, less than about 0.03 ppm, less than about 0.02 ppm, or less than about 0.01 ppm. Section 65 60. The method of claim 59, wherein the zinc is present in an amount less than about 80 ppm. Section 66 60. The method of claim 59, wherein the zinc is present in an amount less than about 50 ppm. Section 67 60. The method of claim 59, wherein the zinc is present in an amount less than about 25 ppm. Section 68 60. The method of claim 59, wherein the zinc is present in an amount less than about 10 ppm. Section 69 60. The method of claim 59, wherein the zinc is present in an amount less than about 5 ppm. Section 70 60. The method of claim 59, wherein the zinc is present in an amount less than about 3 ppm. Section 71 60. The method of claim 59, wherein the zinc is present in an amount of less than about 1 ppm. Section 72 1. A method for producing a neutral protease obtained from C. histolyticum, comprising: fermenting aC histolyticum; b. recovering a crude fermentation liquor containing the neutral protease; c. purifying the neutral protease using an anion exchange filter and hydrophobic interaction chromatography to obtain said neutral protease. Section 73 73. The method of claim 72, further comprising the step of adding ammonium sulfate prior to step (c). Section 74 Item 74. The method according to Item 73, wherein the ammonium sulfate is about 0.8 M to about 1.2 M. Effect of the Invention

[0028] The present invention may provide an improved method for producing collagenase. [Brief description of the drawings]

[0029] The features of the foregoing embodiments will be more readily understood by reference to the following detailed description, taken in conjunction with the accompanying drawings, in which: [Figure 1] An example of a plot showing the linearity of response for the AUX-I / AUX-II bands. This is a plot of trace amounts of collagenase versus micrograms of a known reference standard over a minimal range of 1.125 micrograms to 1.875 micrograms, with an R value of greater than 98%. [Diagram 2] 1 is an example of an annotated image report generated by Bio-Rad Quantity One software. Lane 1 contains a molecular weight standard with bands of known molecular weight. Lane 2 contains the appropriate AUX intermediate reference standard. The remaining lanes contain samples from each IEX fraction. The main product band is the top band in each lane, and each of the additional bands further down represent impurities. [Diagram 3] The frequency of impurity occurrence per fraction and the mean and range of percent relative amount of impurity per fraction for the AUX-I 33 kD impurity are shown. [Figure 4] The frequency of impurity occurrence per fraction and the mean and range of percent relative amount of impurity per fraction for the AUX-I 45 kD impurity are shown. [Diagram 5] The frequency of impurity occurrence per fraction and the mean and range of percent relative amount of impurity per fraction for the AUX-I 55 kD impurity are shown. [Figure 6] The frequency of impurity occurrence per fraction and the mean and range of percent relative amount of impurity per fraction for the AUX-I 80 kD impurity are shown. [Figure 7] The frequency of impurity occurrence per fraction and the mean and range of the percent relative amount of the impurity per fraction for the AUX-I 90 kD impurity are shown. [Figure 8] The frequency of impurity occurrence per fraction and the mean and range of percent relative amount of impurity per fraction for the AUX-I 96 kD impurity are shown. [Figure 9] The frequency of impurity occurrence per fraction and the mean and range of the percent relative amount of the impurity per fraction for the AUX-II 25 kD impurity are shown. [Figure 10] The frequency of impurity occurrence per fraction and the mean and range of the percent relative amount of the impurity per fraction for the AUX-II 38 kD impurity are shown. [Figure 11] The frequency of impurity occurrence per fraction and the mean and range of percent relative amount of impurity per fraction for the AUX-II 50 kD impurity are shown. [Figure 12] The frequency of impurity occurrence per fraction and the mean and range of the percent relative amount of the impurity per fraction for the AUX-II 60 kD impurity are shown. [Figure 13] The frequency of impurity occurrence per fraction and the mean and range of the percent relative amount of the impurity per fraction for the AUX-II 80 kD impurity are shown. [Figure 14] The frequency of impurity occurrence per fraction and the mean and range of the percent relative amount of the impurity per fraction for the AUX-II 90 kD impurity are shown. [Figure 15] The frequency of impurity occurrence per fraction and the mean and range of percent relative amount of impurity per fraction for the AUX-II 92 kD impurity are shown. [Figure 16] The frequency of impurity occurrence per fraction and the mean and range of the percent relative amount of the impurity per fraction for the AUX-II 96 kD impurity are shown. [Figure 17] Typical grams of protein per IEX fraction are shown. [Figure 18] A typical estimate of grams of 90 kD impurity per IEX fraction is shown. [Figure 19] The estimated grams of 90 kD impurity per gram of protein per AUX peak is shown. [Figure 20] The control chart for the AUX-I impurity is shown. [Figure 21A-21B] The control chart for AUX-II impurities is shown. [Figure 22]1 is an SDS-PAGE Coomassie stained gel showing impurities in the AUX-II fraction after the IEX column during a control run from a zinc spiking study (DEV-25C). The AUX-II peak from the IEX column was collected in six fractions. The gel shows the typical purity of the AUX-II fraction and the typical pattern of product associated impurities. Lane 1: Molecular weight markers; Lane 2: Collagenase II; Lane 3: Fraction #1 - 94.9% pure (1.5 μg / lane); Lane 4: Fraction #2 - 98.1% pure (1.5 μg / lane); Lane 5: Fraction #3 - 98.4% pure (1.5 μg / lane); Lane 6: Fraction #4 - 97.3% pure (1.5 μg / lane); Lane 7: Fraction #5 - 96.3% pure (1.5 μg / lane); Lane 8: Fraction #6 - 89.9% pure (1.5 μg / lane); Lane 9: Blank; Lane 10: Collagenase I; Lane 11: Fraction #16 from the Collagenase I peak (1.5 μg / lane). [Figure 23] SDS-PAGE Coomassie stained gel showing impurities in the AUX-I fraction after the IEX column during a control run in a zinc spiking study (run DEV-25C). The AUX-I peak from the IEX column was collected in 10 fractions. The gel shows the typical purity of the AUX-I fractions and the typical pattern of impurities associated with the product. Lane 1: Molecular weight markers Lane 2: Collagenase I Lane 3: Fraction #7 - 87.7% purity Lane 4: Fraction #8 - 86.5% purity Lane 5: Fraction #9 - 92.0% purity Lane 6: Fraction #10 - 95.8% purity Lane 7: Fraction #11 - 97.8% purity Lane 8: Fraction #12 - 97.1% purity Lane 9: Fraction #13 - 97.0% purity Lane 10: Fraction #14 - 96.8% purity Lane 11: Fraction #15 - 96.5% purity [Figure 24]1 is an SDS-PAGE Coomassie stained gel showing impurities in the AUX-II fraction after the IEX column from a zinc spiking study (run DEV-25B) where zinc was spiked to 84 ppm. The AUX-II peak from the IEX column was collected in eight fractions. This gel shows the degradation pattern caused by the presence of zinc, with an increase in the number and intensity of impurities and a decrease in the purity of the AUX-II fractions. In this gel, none of the AUX-II fractions meet the purity criteria for pooling. Lane 1: Molecular weight markers; Lane 2: Collagenase II; Lane 3: Fraction #1 - 66.4% pure (1.5 μg / lane); Lane 4: Fraction #2 - 75.5% pure (1.5 μg / lane); Lane 5: Fraction #3 - 79.5% pure (1.5 μg / lane); Lane 6: Fraction #4 - 76.7% pure (1.5 μg / lane); Lane 7: Fraction #5 - 72.1% pure (1.5 μg / lane); Lane 8: Fraction #6 - 63.9% pure (1.5 μg / lane); Lane 9: Fraction #7 - 52.9% pure (1.5 μg / lane); Lane 10: Fraction #8 - 45.0% pure (1.5 μg / lane) [Diagram 25]1 is an SDS-PAGE Coomassie stained gel showing impurities in the AUX-I fraction after the IEX column from a zinc spiking study (run DEV-25B) where zinc was spiked to 84 ppm. The AUX-I peak from the IEX column was collected in 10 fractions. This gel shows the degradation pattern caused by the presence of zinc, with an increase in the number and intensity of impurities and a decrease in the purity of the AUX-I fraction. In this gel, none of the AUX-I fractions meet the purity criteria for pooling. Lane 1: Molecular weight markers; Lane 2: Collagenase I; Lane 3: Fraction #9 - 54.9% pure (1.5 μg / lane); Lane 4: Fraction #10 - 50.9% pure (1.5 μg / lane); Lane 5: Fraction #11 - 56.2% pure (1.5 μg / lane); Lane 6: Fraction #12 - 65.2% pure (1.5 μg / lane); Lane 7: Fraction #13 - 70.4% pure (1.5 μg / lane); Lane 8: Fraction #14 - 70.0% pure (1.5 μg / lane); Lane 9: Fraction #15 - 67.5% pure (1.5 μg / lane); Lane 10: Fraction #16 - 62.4% pure (1.5 μg / lane); Lane 11: Fraction #17 - 58.8% pure (1.5 μg / lane) Lane 12: Fraction #18 - 52.6% purity (1.5 μg / lane) [Figure 26]1 is an SDS-PAGE Coomassie stained gel showing impurities in the AUX-II fraction after the IEX column from a control run in the nickel spiking study (run DEV-25A). The AUX-II peak from the IEX column was collected in six fractions. The gel shows the typical purity of the AUX-I fraction and the typical pattern of product associated impurities. Lane 1: Molecular weight markers. Lane 2: Collagenase II. Lane 3: Fraction #1 - 95.9% pure (1.5 μg / lane). Lane 4: Fraction #2 - 97.1% pure (1.5 μg / lane). Lane 5: Fraction #3 - 99.1% pure (1.5 μg / lane). Lane 6: Fraction #4 - 98.2% pure (1.5 μg / lane). Lane 7: Fraction #5 - 96.9% pure (1.5 μg / lane). Lane 8: Fraction #6 - 92.3% pure (1.5 μg / lane). Lane 9: Blank. Lane 10: Collagenase I. Lane 11: Fraction #16 from the Collagenase I peak (1.5 μg / lane). [Figure 27] 1 is an SDS-PAGE Coomassie stained gel showing impurities in the AUX-I fraction after the IEX column from a control run in the nickel spiking study (run DEV-25A). The AUX-I peak from the IEX column was collected in 10 fractions. The gel shows the typical purity of the AUX-II fraction and the typical pattern of product associated impurities. Lane 1: Molecular weight markers; Lane 2: Collagenase I; Lane 3: Fraction #7 - 85.6% pure (1.5 μg / lane); Lane 4: Fraction #8 - 85.2% pure (1.5 μg / lane); Lane 5: Fraction #9 - 90.7% pure (1.5 μg / lane); Lane 6: Fraction #10 - 95.5% pure (1.5 μg / lane); Lane 7: Fraction #11 - 96.6% pure (1.5 μg / lane); Lane 8: Fraction #12 - 97.0% pure (1.5 μg / lane); Lane 9: Fraction #13 - 96.6% pure (1.5 μg / lane); Lane 10: Fraction #14 - 96.8% pure (1.5 μg / lane); Lane 11: Fraction #15 - 96.8% pure (1.5 μg / lane) [Figure 28]1 is an SDS-PAGE Coomassie stained gel showing impurities in the AUX-II fraction after the IEX column from a nickel spiking study (run DEV-25D) where nickel was spiked to 1.2 ppm. The AUX-II peak from the IEX column was collected in seven fractions. This gel shows the degradation pattern caused by the presence of nickel, with an increase in the number and intensity of impurities, and a decrease in the purity of the AUX-II fraction. In this gel, none of the AUX-II fractions meet the pooling criteria for purity. Lane 1: Molecular weight markers. Lane 2: Collagenase II. Lane 3: Fraction #1 - 74.8% pure (1.5 μg / lane). Lane 4: Fraction #2 - 81.8% pure (1.5 μg / lane). Lane 5: Fraction #3 - 84.6% pure (1.5 μg / lane). Lane 6: Fraction #4 - 85.6% pure (1.5 μg / lane). Lane 7: Fraction #5 - 83.0% pure (1.5 μg / lane). Lane 8: Fraction #6 - 72.3% pure (1.5 μg / lane). Lane 9: Fraction #7 - 64.9% pure (1.5 μg / lane). Lane 10: Blank. Lane 11: Collagenase I. Lane 12: Fraction #8 from the Collagenase I peak - 62.7% pure (1.5 μg / lane). [Figure 29]1 is an SDS-PAGE Coomassie stained gel showing impurities in the AUX-I fraction after the IEX column from a nickel spiking study (run DEV-25D) where nickel was spiked to 1.2 ppm. The AUX-I peak from the IEX column was collected in 10 fractions. This gel shows the degradation pattern caused by the presence of nickel, with an increase in the number and intensity of impurities, and a decrease in the purity of the AUX-I fraction. In this gel, none of the AUX-I fractions meet the pooling criteria for purity. Lane 1: Molecular weight markers; Lane 2: Collagenase I; Lane 3: Fraction #9 - 60.3% pure (1.5 μg / lane); Lane 4: Fraction #10 - 67.6% pure (1.5 μg / lane); Lane 5: Fraction #11 - 81.6% pure (1.5 μg / lane); Lane 6: Fraction #12 - 84.7% pure (1.5 μg / lane); Lane 7: Fraction #13 - 84.1% pure (1.5 μg / lane); Lane 8: Fraction #14 - 82.6% pure (1.5 μg / lane); Lane 9: Fraction #15 - 82.0% pure (1.5 μg / lane); Lane 10: Fraction #16 - 72.4% pure (1.5 μg / lane); Lane 11: Fraction #17 - 68.6% pure (1.5 μg / lane) [Diagram 30]Q-Sepharose High Performance (Q-HP) chromatogram and SDS-PAGE gel image with Coomassie staining. Load and merged AUX I and AUX II pooled fractions are shown. Q HP chromatogram represents purification of neutral protease from the waste stream of the HIC chromatography step using an IEX chromatography column. Gel image shows fractions collected over the IEX load and purification. Neutral protease elution at the end of the gradient is observed in fraction 3. N-termini of all impurities are determined by liquid chromatography-mass spectrometry (LC-MS) using Edman degradation and purified product fragments. Thermolysin was used as a model protease to determine whether the cut ends matched the thermolysin cleavage site in the protein analysis program Peptide Cutter at http: / / expasy.org. In all cases, the cut ends were hits for thermolysin cleavage. The purified neutral protease was sequenced by Edman degradation and LC-MS to identify the N-terminus of the current process neutral protease, which was also used for degradation studies to confirm that the isolated neutral protease of the process was functional (active). [Diagram 31] Deduced genomic sequence of C. histolyticum (CCH) neutral protease (δ-toxin, CHL_2576). The blue highlight identifies the N-terminus of the mature neutral protease from C. histolyticum strain 004, as identified by Edman degradation assay, and is highlighted in blue. The green highlight identifies the section of the mature neutral protease from C. histolyticum, as detected by Lys-C / trypsin LC-mass spectrometry. [Diagram 32] The protease subsite nomenclature of Schechter and Berger is described. [Diagram 33] Sequence alignment of rnprA and collagenase C. histolyticum (abbreviated as CCH) neutral protease (CLH_2576) showing complete homology between the two enzymes. [Diagram 34]12% Tris-Glycine Novex® casein zymography PAGE images of C. histolyticum (CCH) in-process product streams: Mustang-Q filtrate (MQF), HIC load, HIC eluate, and AUX-II pool from lot 100808. The gel used is a 12% Tris-Glycine Novex® zymogram casein gel. The manufacturing process that produced lot 100808 produced high levels of impurities and used outdated process control methods. In this image, neutral protease activity is observed in the MQF, HIC load, and HIC eluate, but not in the AUX-II pool. The low signal observed in the HIC eluate sample indicates the primary clearance of neutral proteases occurring in the HIC step. Lane 1: + control (thermolysin), 1 ng Lane 2: molecular weight reference Lane 3: blank Lane 4: MQF Dev-25A, 2 μg Lane 5: blank Lane 6: HIC load Dev-13, 3 μg Lane 7: blank Lane 8: blank Lane 9: HIC eluent Dev-25A, 4 μg Lane 10: blank Lane 11: Aux II pool 1000808, 10 μg Lane 12: blank [Diagram 35]C. histolyticum (CCH) in-process product streams: 12% Tris-Glycine Novex® casein zymography PAGE images of the first tangential flow filter (TFF-1) retentate, AUX-I pool, AUX-I intermediate, AUX-II intermediate, and drug product from a lot affected by HIC buffer with high zinc and nickel content. The gel used is a 12% Tris-Glycine Novex® zymogram casein gel. In this case, neutral protease was observed in the TFF-1 retentate, as well as low levels of AUX-I and drug product, but not in AUX-II. These data demonstrate neutral protease separation from product over the IEX step eluting in AUX-I. Lane 1: + control (thermolysin), 1 ng Lane 2: molecular weight reference Lane 3: blank Lane 4: TFF-1 concentrate 10000808, 10 μg Lane 5: blank Lane 6: AUX-I pool 10000808, 10 μg Lane 7: blank Lane 8: AUX-I intermediate 10000808, 10 μg Lane 9: blank Lane 10: AUX-II intermediate 10000808, 10 μg Lane 11: blank Lane 12: drug substance 10000808, 10 μg [Diagram 36]Figure 12% Tris-Glycine Novex® casein zymography PAGE image of collagenase C. histolyticum (CCH) IEX fraction retentate of lot 1000414 AUX-II fraction. The gel used is a 12% Tris-Glycine Novex® zymogram casein gel. These fractions are from a manufacturing lot produced before active neutral protease was found in the process. Neutral protease was not detected in any of the AUX-II fractions. Lane 1: + control (thermolysin), 1 ng Lane 2: molecular weight reference Lane 3: AUX-II fraction 1 1000414, 6.4 μg Lane 4: AUX-II fraction 2 1000414, 14.4 μg Lane 5: AUX-II fraction 3 1000414, 10 μg Lane 6: AUX-II fraction 4 1000414, 10 μg Lane 7: AUX-II fraction 5 1000414, 10 μg Lane 8: AUX-II fraction 6 1000414, 11.2 μg Lane 9: AUX-II fraction 7 1000414, 6.2 μg Lane 10: AUX-II fraction 8 1000414, 5.8 μg [Figure 37]Figure 12% Tris-Glycine Novex® casein zymography PAGE image of collagenase C. histolyticum (CCH) IEX fraction retentate of lot 1000414 AUX-I fraction. The gel used is a 12% Tris-Glycine Novex® zymogram casein gel. These fractions are from manufacturing lots produced before active neutral protease was found in the process. Neutral protease was not detected in any of the AUX-I fractions using this method. Lane 1: molecular weight reference Lane 2: AUX-I fraction 9 1000414, 8.8 μg Lane 3: AUX-I fraction 10 1000414, 10 μg Lane 4: AUX-I fraction 11 1000414, 10 μg Lane 5: AUX-I fraction 12 1000414, 10 μg Lane 6: AUX-I fraction 13 1000414, 10 μg Lane 7: AUX-I fraction 14 1000414, 10 μg Lane 8: AUX-I fraction 15 1000414, 10 μg Lane 9: AUX-I fraction 16 1000414, 14 μg Lane 10: AUX-I fraction 17 1000414, 10 μg [Figure 38]Figure 12 shows a 12% Tris-Glycine Novex® casein zymography PAGE image of collagenase C. histolyticum (CCH) IEX fraction retentate of lot 1000414 AUX-II fractions 1, 7 and AUX-I fractions 8, 17. The gel used is a 12% Tris-Glycine Novex® zymogram casein gel. The manufacturing process that produced lot 1000414 resulted in low levels of impurities and used older process control methods. On this gel, fractions were run at higher concentrations than on the original gels (Figures 16 and 17). Neutral protease was only detected in the last fraction of the AUX-I peak (fraction 17), demonstrating its positional nature in eluting at the end of the peak. Lane 1: + control (thermolysin), 1 ng Lane 2: molecular weight reference Lane 3: AUX-II fraction 1 1000414 Tris-Gly, 10 μg Lane 4: AUX-II fraction 1 1000414 LDS, 10 μg Lane 5: AUX-II fraction 7 1000414 Tris-Gly, 9.3 μg Lane 6: AUX-II fraction 7 1000414 LDS, 9.3 μg Lane 7: AUX-I fraction 8 1000414 Tris-Gly, 6.5 μg Lane 8: AUX-I fraction 8 1000414 LDS, 6.5 μg Lane 9: AUX-I fraction 17 1000414 Tris-Gly, 11 μg Lane 10: AUX-I fraction 17 1000414 LDS, 11 μg Tris-Gly and LDS are two buffer systems for SDS-PAGE. This gel was run to assess whether one system was superior to the other. No differences were observed. [Figure 39]Figure 12 is a 12% Tris-Glycine Novex® casein zymography PAGE image of collagenase C. histolyticum (CCH) IEX fraction retentate of lot 001138 AUX-I tail fractions 17, 18, and 19. The gel used is a 12% Tris-Glycine Novex® zymogram casein gel. This gel shows rejected fractions from the end of the AUX-I peak. Each of the three rejected fractions contained detectable neutral protease. Fractions 17 and 18 passed the in-process pooling criteria but were rejected based on the 1:1 product target pooling strategy. This demonstrates process control to remove residual neutral protease from the AUX-I and drug products. Lane 1: TFF-1 concentrate Dev-25A, 34 μg Lane 2: molecular weight reference Lane 3: blank Lane 4: AUX-I fraction 7 Dev-25A, 4.5 μg Lane 5: blank Lane 6: AUX-I fraction 17 Dev-25A, 5 μg Lane 7: blank Lane 8: AUX-I fraction 17 001138, 14.8 μg Lane 9: blank Lane 10: AUX-I fraction 18 001138, 10.3 μg Lane 11: blank Lane 12: AUX-I fraction 19 001138, 8.1 μg [Fig. 40A-40B]Figure 40A is a 12% Tris-Glycine Novex® casein zymography SDS-PAGE gel image of the collagenase C. histolyticum (CCH) manufacturing process, lot 0010987. The gel used is a 12% Tris-Glycine Novex® zymogram casein gel. This gel shows rejected fractions from the end of the AUX-I peak. Each of the three rejected fractions contained detectable neutral protease. Fractions 17 and 18 passed the in-process pooling criteria but were rejected based on the 1:1 product target pooling strategy. This demonstrates the process control to remove residual neutral protease from AUX-I and the drug product. Figure 40A shows the purification status for AUX-I throughout the process. Figure 40B shows the purification status for AUX-II throughout the process. Figure 40A Lane 1: Positive control (TL) Lane 2: Molecular weight reference Lane 3: None Lane 4: Mustang Q filtrate, 2.1 μg loaded Lane 5: None Lane 6: HIC eluate, 4.9 μg loaded Lane 7: None Lane 8: AUX-I pool, 10 μg loaded Lane 9: None Lane 10: AUX-I intermediate, 10 μg loaded Lane 11: None Lane 12: Assay blank Figure 40B Lane 1: Positive control (TL) Lane 2: Molecular weight reference Lane 3: None Lane 4: TFF-1 concentration, 10 μg loaded Lane 5: None Lane 6: AUX-II pool, 10 μg loaded Lane 7: None Lane 8: AUX-II intermediate, 10 μg loaded Lane 9: None Lane 10: Bulk drug substance (drug product), 10 μg loaded Lane 11: None Lane 12: Assay blank [Diagram 41] Collection of collagenase I fractions and collagenase II and their forward processing based on meeting purity criteria are shown. [Diagram 42] The percent purity of the AUX-I product relative to the amount of total protein in each AUX elution peak is provided. [Diagram 43] The percent purity of the AUX-II product relative to the amount of total protein in each AUX elution peak is provided. [Diagram 44]42 depicts a control diagram for the AUX-I 90 kDa impurity. The decrease in percent purity level of the AUX-I elution peak seen in FIG. 42 is primarily due to high levels of the AUX-I 90 kDa impurity. [Diagram 45] 43 depicts a control diagram of the AUX-II 96 kDa impurity. The decrease in percent purity level of the AUX-II elution peak seen in FIG. 43 is primarily due to high levels of the AUX-II 96 kDa impurity. [Fig. 46A-46D] The estimated percentages in the product peak for the various AUX peaks and bands are provided. The data provided by these figures were used in statistical analysis to evaluate the IEX fraction impurities after implementation of corrective and preventive actions related to deviation 3797. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0030] Various aspects and embodiments are now fully described herein. However, these aspects and embodiments may be embodied in many different forms and should not be construed as limiting. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the subject matter to those skilled in the art. All publications, patents, and patent applications cited herein, whether supra or infra, are hereby incorporated by reference in their entirety.

[0031] I. Definition Unless otherwise defined, all terms and phrases used herein include the meaning that those terms and phrases have acquired in the art unless expressly indicated to the contrary or apparent from the context in which the term or phrase is used. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, particular methods and materials are described herein.

[0032] Unless otherwise stated, the use of individual numerical values ​​is stated as an approximation, as if preceded by the word "about" or "approximately". Similarly, the various ranges of numerical values ​​specified in this application are stated as approximations, as if the minimum and maximum values ​​in the stated range were both preceded by the word "about" or "approximately", unless otherwise expressly indicated. In this manner, variations above and below the stated range can be used to achieve substantially the same results as the values ​​within the range. As used herein, the terms "about" and "approximately" when referring to numerical values ​​have their clear and ordinary meaning to those skilled in the art to which the subject matter of this disclosure most closely relates or to the range or element in question. The amount by which the numerical values ​​extend from the strict numerical boundaries depends on many factors. For example, some of the factors that may be considered include the importance of the element and / or the effect that a given amount of variation will have on the performance of the claimed subject matter, as well as other considerations known to those skilled in the art. As used herein, the use of different amounts of significant digits for different numerical values ​​does not mean to limit how the use of the word "about" or "approximately" acts to broaden a particular numerical value or range. Thus, as a general matter, "about" or "approximately" broadens a numerical value. The disclosure of a range is also intended as an expanded range given by the use of the term "about" or "approximately", in addition to a continuous range that includes all values ​​between the minimum and maximum values. As a result, the recitation of a range of numerical values ​​herein is merely intended to serve as a shorthand method of individually referring to each separate value within the range, and each separate value is incorporated herein as if it were individually recited herein.

[0033] As used herein, "collagenase" means to include one or more proteins that exhibit collagenase activity in a standard collagenase assay, e.g., collagenase I and / or collagenase II. A collagenase composition may include at least one of two microbial collagenases, sometimes called AUXI and AUXII. The terms "collagenase I," "ABCI," "AUXI," "AUX-I," "collagenase AUXI," and "collagenase ABCI" refer to the same enzyme having the same amino acid sequence and can be used interchangeably. Similarly, the terms "collagenase II," "ABCII," "AUXII," "AUX-II," "collagenase AUXII," and "collagenase ABCII" refer to the same enzyme and can also be used interchangeably. These collagenases are secreted by bacterial cells. In some embodiments, collagenases are isolated and purified from C. histolyticum culture supernatant by chromatographic methods. In other aspects, the collagenase is recombinantly isolated and purified and / or obtained from a variety of sources, including mammalian, fungal, and bacterial sources. Both collagenases are specialized proteases and share the same EC number (E.C3.4.24.3). In other aspects, the collagenase composition comprises collagenase I or collagenase II. In another embodiment, the collagenase composition comprises both collagenase I and collagenase II.

[0034] As used herein, "collagen-mediated condition" refers to any disease or illness that involves collagen.Examples of collagen-mediated conditions that can be treated by the compositions and methods described herein include, but are not limited to, Dupuytren's disease; Peyronie's disease; frozen shoulder (adhesive capsulitis); keloid; hypertrophic scar; depressive scar such as that resulting from inflammatory acne; post-operative adhesion; acne vulgaris; lipoma, and disfiguring conditions such as wrinkles, cellulite, and neoplastic fibrosis.

[0035] The phrase "collagenase I product" means collagenase I purified from fermentation, where the collagenase I elutes from the purification process and is collected in fractions, and any number of these fractions are pooled together.

[0036] The phrase "collagenase II product" means collagenase II purified from fermentation, the elution of collagenase II from the purification process being collected in fractions, and any number of these fractions being pooled together.

[0037] The phrase "derived from" Clostridium histolyticum means collagenase that originates from C. histolyticum. This includes both collagenase obtained from fermentation of C. histolyticum and from fermentation with other organisms that have recombinant collagenase.

[0038] The terms "drug substance," "drug product," or "collagenase composition" may be used interchangeably and are understood to refer to a collagenase I composition, a collagenase II composition, or a composition of both collagenase I and collagenase II, prior to any optional lyophilization.

[0039] As used herein, "essentially free of neutral proteases" means below the detectable limits of the particular detection assay being used (e.g., SDS PAGE, zymography, HPLC, etc.).

[0040] As used herein, "forward processed" refers to highly purified collagenase I or II fractions that are used for pooling to make the final collagenase I drug material or collagenase II drug material.

[0041] As used herein, "highly purified" or "high purity" or "highly pure" collagenase I or II, or a combination of collagenase I and II, means a collagenase that is at least 95 area% pure, or at least 96 area% pure, or at least 97 area% pure, or at least 98 area% pure, or at least 99 area% pure, or about 100 area% pure as measured by reverse phase HPLC (RP-HPLC). For further clarity, this includes all values ​​for purity from at least 95 area% to about 100 area% as measured by RP-HPLC.

[0042] As used herein, "most abundant impurity" refers to the collagenase fragment having the highest presence at a given step (e.g., after IEX) as measured by SDS-PAGE gels with or without densitometry, or as measured by another acceptable method. Collagenase fragments may be found throughout the purification process and may be of any size, but are most frequently found at molecular weights of about 33 kDa, 45 kDa, 55 kDa, 80 kDa, or 90 kDa for collagenase I fragments, and about 25 kDa, 38 kDa, 50 kDa, 60 kDa, 80 kDa, 90 kDa, 92 kDa, 92 kDa, or 96 kDa for collagenase II fragments.

[0043] As used herein, "neutral protease exclusion" or "neutral protease exclusion step" refers to the removal or rejection of the fraction collected during production after the collagenase filtrate has passed through a separation column or filter.

[0044] "Optional" or "optionally" means that the subsequently described element, component, or circumstance may or may not occur, such that the description includes cases where the element, component, or circumstance occurs as well as cases where it does not occur.

[0045] As used herein, the term "pharmaceutically acceptable carrier" or "excipient" means a non-toxic, inert solid, semi-solid, or liquid filler, diluent, encapsulating material, or formulation auxiliary. Some examples of materials which can serve as pharma- ceutically acceptable carriers are sugars such as lactose, glucose, and sucrose; starches such as corn starch and potato starch; cellulose and its derivatives such as sodium carboxymethylcellulose, ethylcellulose, and cellulose acetate; powdered tragacanth; malt; gelatin; talc; glycols such as propylene glycol; esters such as ethyl oleate and ethyl laurate; agar; buffers such as magnesium hydroxide and aluminum hydroxide; alginic acid; pyrogen-free distilled water; isotonic saline; Ringer's solution; ethyl alcohol, and phosphate buffer solutions, as well as other non-toxic compatible lubricants such as sodium lauryl sulfate and magnesium stearate, as well as coloring agents, releasing agents, coating agents, flavoring agents, preservatives, and antioxidants can also be present in the composition, according to the judgment of the formulator.

[0046] As used herein, a "purified fraction" means that the undegraded collagenase I has a purity of at least 90% as measured by SDS-PAGE gel densitometry, and the most abundant degraded collagenase I or II impurity is 7.5% or less. To qualify as a "purified fraction" for a collagenase II fraction from a collagenase manufacturing process step, the undegraded collagenase II has a purity of at least 90% as measured by SDS-PAGE gel densitometry, and the most abundant degraded collagenase I or II impurity is 7.5% or less. For example, a purified collagenase I fraction may have at least about 90% or about 91.2% undegraded AUX-I as measured by SDS-PAGE gel densitometry, and the most abundant degraded AUX-I or AUX-II impurity is about 5% or about 5.8% or less as measured by SDS-PAGE gel densitometry. As a second example, for an AUX-II fraction to qualify as a purified fraction, the fraction may have undegraded AUX-II of at least about 90% or about 91.2% purity as measured by SDS-PAGE gel densitometry, and the most abundant degraded AUX-I or AUX-II impurities are no more than about 5% or about 5.8% pure as measured by SDS-PAGE gel densitometry.

[0047] The terms "subject" or "patient" are used interchangeably herein and refer to a human or other mammal.

[0048] II. Introduction Collagenase can be produced by various fermentation methods well known to those skilled in the art. Crude collagenase obtained from C. histolyticum has been previously purified by various methods, including dye-ligand affinity chromatography, heparin affinity chromatography, ammonium sulfate precipitation, hydroxyapatite chromatography, size-exclusion chromatography, ion-exchange chromatography, and metal chelation chromatography. Crude and partially purified collagenase is commercially available from many sources, including Sigma Aldrich (SIAL-Millipore) and Advance Biofactures Corp., Lynbrook, NY. Fermentation and purification methods for crude collagenase obtained from C. histolyticum are also described in U.S. Patent No. 7,811,560.

[0049] As detailed below, the present invention provides improved methods for producing collagenase products, including (a) controlling metal content during purification, and (b) reducing or eliminating neutral proteases from the final collagenase product, e.g., collagenase I and collagenase II drug substance. The present disclosure is generally applicable to the production and use of collagenase, regardless of source. It is not intended to be limited to collagenase I and II derived from C. histolyticum.

[0050] III. Improved Production of Collagenase In one general aspect of the method of the invention for producing highly pure collagenase (e.g., collagenase I and II drug substance), the method includes the steps of (1) fermenting bacteria that secrete at least collagenase(s) and neutral protease into a liquid fermentation medium, (2) purifying the collagenase from the medium (optionally including treatment with a high salt composition), (3) separating collagenase I and collagenase II from each other, (4) reducing or eliminating neutral protease from one or both of the collagenase I and collagenase II portions so that they are essentially free of neutral protease, and (5) producing a highly pure collagenase drug substance after removal of the neutral protease. Optionally, pharma- ceutically acceptable excipients are added to such drug substance to form a pharmaceutical formulation as described elsewhere herein.

[0051] A. Fermentation to make collagenases such as collagenase I and collagenase II. The present disclosure encompasses the fermentation of any bacteria that secretes collagenase into the fermentation medium. Examples of such fermentations include those from Actinomadura madurae, Actinobacillus actinomycetemcomitans, Bacillus cereus, Clostridium histolyticum, Clostridium perfringens, Streptococcus mutans, Staphylococcus spp., Vibrio alginolyticus, and Vibrio vulnificus. This list is not exhaustive and other bacterial organisms that secrete collagenases known to those of skill in the art can also be used. See, for example, US Patent No. 7,811,560.Proteins secreted by C. histolyticum include type I and type II collagenases, and various toxins (eg, neutral protease, clostripain, aerolysin-like hemolysin, and oxygen-labile hemolysin).

[0052] The fermentation medium can be any medium into which collagenase is secreted. The fermentation medium may include a carbon source, a nitrogen source, salts, micronutrients, and water. The carbon source may include, but is not limited to, sugars or other carbohydrates. The nitrogen source may include, but is not limited to, a food, such as a soy food, or an extract, such as a yeast extract, or a pre-digested polypeptide, such as a peptone or tryptone. The peptone may be of animal origin, such as a bovine or porcine derived medium. The peptone may be of non-animal origin, such as a vegetable or other plant origin.

[0053] B. Purification of Collagenase I and Collagenase II Common purification techniques include, but are not limited to, filtration, chromatography, and precipitation. Certain aspects of the purification of collagenase I and II from C. histolyticum can be performed as is well known in the art. See, for example, U.S. Patent No. 7,811,560.

[0054] The collagenase purification process further includes at least one of the following steps: (1) controlling zinc and / or nickel concentration in at least the first anion exchange step, the HIC step, and the first buffer exchange step; (2) discarding the fraction after AUX-I from any of the anion exchange steps; (3) pooling the collagenase I-only peak fractions after separating collagenase I from collagenase II until the amount of collagenase I pooled from the peak fractions is approximately equal to the amount of collagenase II from the manufacturing process; and (4) setting purity criteria for the collagenase I, collagenase II, or both collagenase I and collagenase II pooling steps. The use of these steps produces a collagenase I product that is essentially free of neutral proteases. Similarly, the collagenase II product is essentially free of neutral proteases.

[0055] Filtration of collagenase and toxins from the medium can be accomplished by a number of methods, such as through the use of depth filters, centrifugation, microfiltration, tangential flow filtration, diatomaceous earth filter beds, and vacuum filtration, In one embodiment, filtration is performed using a depth filter followed by a 0.2 micron sterilizing grade filter.

[0056] In one embodiment, the purification process may include recovering collagenase from fermented C. histolyticum (e.g., porcine-derived medium such as protease peptone) through a depth filter (e.g., Millipore Millistak HC POD), followed by anion exchange filtration (e.g., Mustang Q filter), followed by filtration (e.g., 0.45μ filter), and conditioning of the filtrate by exposing the filtrate to a high salt composition (e.g., ammonium sulfate) at a concentration of about 0.8M to 1.2M, followed by a hydrophobic interaction chromatography column (e.g., Phenyl SFF low-sub), followed by buffer exchange (e.g., tangential flow filtration or dialysis), followed by ion exchange chromatography (e.g., Q Sepharose HP anion exchange column), followed by buffer exchange (e.g., tangential flow filtration or dialysis). Additionally, leupeptin may be added to the hydrophobic interaction chromatography step and maintained throughout the ion exchange step.

[0057] High salt compositions allow for separation of proteins based on solubility. Salts from the Hofmeister series may be used. In some embodiments, the composition is a solution selected from the group consisting of ammonium sulfate, sodium chloride, potassium chloride, sodium sulfate, or mixtures thereof. High salt compositions may be used in a variety of concentrations. For example, ammonium sulfate may be used in concentrations ranging from about 0.8 M to about 1.2 M, or from about 0.9 M to about 1.1 M, or about 1 M.

[0058] It is advantageous for the high salt composition to be free of metal contaminants or to have a low metal content. In some embodiments, the high salt composition meets one or more of the following specifications: [Table 1]

[0059] In another embodiment, the high salt composition may have zinc present in an amount less than about 85 ppm, or 80 ppm, or 75 ppm, or 70 ppm, or 65 ppm, or 60 ppm, or 55 ppm, or 50 ppm, or 45 ppm, or 40 ppm, or 35 ppm, or 30 ppm, or 25 ppm, or 20 ppm, or 15 ppm, or 10 ppm, or 9 ppm, or 8 ppm, or 7 ppm, or 6 ppm, or 5 ppm, or 4 ppm, or 3 ppm, or 2 ppm, or 1 ppm. Nickel may be present in the high salt composition in an amount less than about 1.5 ppm, or 1.4 ppm, or 1.3 ppm, or 1.2 ppm, or 1.1 ppm, or 1.0 ppm, or 0.9 ppm, or 0.8 ppm, or 0.7 ppm, or 0.6 ppm, or 0.5 ppm, or 0.4 ppm, or 0.3 ppm, or 0.2 ppm, or 0.1 ppm.

[0060] The metal content of the raw materials can be controlled by obtaining ultra-high purity raw materials from certified commercial sources. In one embodiment, the nickel concentration is less than about 1.2 ppm, or the zinc concentration is less than about 84.4 ppm, or the nickel concentration is less than about 1.2 ppm and the zinc concentration is less than about 84.4 ppm. In other embodiments, the nickel concentration, the zinc concentration, or the concentrations of both nickel and zinc are less than about 0.1 ppm.

[0061] In one embodiment, collagenase I and collagenase II from C. histolyticum are isolated and purified, and the purification process controls metal levels. In a further embodiment, collagenase I and collagenase II from C. histolyticum are isolated and purified, and the purification process controls levels of nickel, zinc, aluminum, arsenic, calcium, cadmium, chromium, copper, iron, magnesium, lead, or combinations thereof.

[0062] In a further embodiment, the present disclosure relates to a method for isolation and purification of collagenase I and collagenase II from C. histolyticum, wherein the purification process controls nickel and / or zinc levels.

[0063] In one embodiment of the invention, the nickel concentration during the manufacturing process for collagenase I and / or II and formation of the drug product is less than about 100 ppm, or less than about 75 ppm, or less than about 50 ppm, or less than about 40 ppm, or less than about 30 ppm, or less than about 20 ppm, or less than about 10 ppm, or less than about 5 ppm, or less than about 3.0 ppm, or less than about 2.9 ppm, or less than about 2.8 ppm, or less than about 2.7 ppm, or less than about 2.6 ppm, or less than about 2.5 ppm, or less than about 2.4 ppm, or less than about 2.3 ppm, or less than about 2.2 ppm, or less than about 2.1 ppm, or less than about 2.0 ppm, or less than about 1.9 ppm, or less than about 1.8 ppm, or less than about 1.7 ppm, or or less than about 1.6 ppm, or less than about 1.5 ppm, or less than about 1.4 ppm, or less than about 1.3 ppm, or less than about 1.2 ppm, or less than about 1.1 ppm, or less than about 1.0 ppm, or less than about 0.9 ppm, or less than about 0.8 ppm, or less than about 0.7 ppm, or less than about 0.6 ppm, or less than about 0.5 ppm, or less than about 0.4 ppm, or less than about 0.3 ppm, or less than about 0.2 ppm, or less than about 0.1 ppm, or less than about 0.09 ppm, or less than about 0.08 ppm, or less than about 0.07 ppm, or less than about 0.06 ppm, or less than about 0.05 ppm, or less than about 0.04 ppm, or less than about 0.03 ppm, or less than about 0.02 ppm, or less than about 0.01 ppm.

[0064] In another embodiment of the invention, the zinc concentration during the manufacturing process for collagenase I and / or II and formation of the drug product is less than about 1000 ppm, or less than about 950 ppm, or less than about 900 ppm, or less than about 850 ppm, or less than about 800 ppm, or less than about 750 ppm, or less than about 700 ppm, or less than about 650 ppm, or less than about 600 ppm, or less than about 550 ppm, or less than about Less than 500 ppm, or less than about 450 ppm, or less than about 400 ppm, or less than about 350 ppm, or less than about 300 ppm, or less than about 250 ppm, or less than about 200 ppm, or less than about 150 ppm, or less than about 125 ppm, or less than about 100 ppm, or less than about 95 ppm, or less than about 90 ppm, or less than about 85 ppm, or less than about 84 ppm, or less than about 80 ppm, or less than about 7 Less than 5 ppm, or less than about 70 ppm, or less than about 65 ppm, or less than about 60 ppm, or less than about 55 ppm, or less than about 50 ppm, or less than about 45 ppm, or less than about 40 ppm, or less than about 35 ppm, or less than about 30 ppm, or less than about 25 ppm, or less than about 20 ppm, or less than about 15 ppm, or less than about 10 ppm, or less than about 9 ppm, or less than about 8 ppm, or less than about 7 ppm pm, or less than about 6 ppm, or less than about 5 ppm, or less than about 4 ppm, or less than about 3 ppm, or less than about 2 ppm, or less than about 1 ppm, or less than about 0.9 ppm, or less than about 0.8 ppm, or less than about 0.7 ppm, or less than about 0.6 ppm, or less than about 0.5 ppm, or less than about 0.4 ppm, or less than about 0.3 ppm, or less than about 0.2 ppm, or less than about 0.1 ppm.

[0065] C. Separation of Collagenase I and Collagenase II Collagenase I and II can be separated by any technique known to those of skill in the art. For example, collagenase I and II can be separated from each other using ion exchange chromatography. One type of ion exchange chromatography can be an anion exchange column, such as a Q-Sepharose column.

[0066] D. Reduction or elimination of neutral proteases In one general embodiment, when multiple collagenases are present, the collagenases are separated from each other followed by a step of eliminating neutral proteases. The collagenases are separated using ion exchange chromatography during the elution step. The elution of collagenases generally occurs in multiple peaks, where collagenase I is in one peak and collagenase II is in the second peak. Each peak can be collected as a fraction over time for each elution peak.

[0067] Reduction or elimination of neutral proteases from collagenase I and collagenase II products may be accomplished by one or more of the following elimination steps: (a) eliminating any fractions having less than about 90% collagenase I or II content and about 5% or more content of any single impurity as measured by SDS PAGE gel or HPLC; (b) eliminating any fractions having detectable levels of neutral protease as measured by SDS PAGE gel or zymography; or (c) eliminating any fractions calculated to contain excessive levels of neutral protease. Each is described below and may be used individually or in combination.

[0068] The size of the fractions is not limited, but exemplary industrial size fractions can be less than about 200 mL, or 300 mL, or 400 mL, or 500 mL, or 600 mL, or 700 mL, or 800 mL, or 900 mL, or 1000 mL, or 1100 mL, or 1200 mL, or 1300 mL, or 1400 mL, or 1500 mL, or 1600 mL, or 1700 mL, or 1800 mL, or more.

[0069] The amount of collagenase per fraction is not limited, but exemplary industrial size fractions may contain a protein concentration of anywhere from about 0.1 g / L to about 5 g / L or more, such as about 0.2 g / L, or 0.5 g / L, or 0.8 g / L, or 1.1 g / L, or 1.4 g / L, or 1.7 g / L, or 2.0 g / L, or 2.3 g / L, or 2.6 g / L, or 2.9 g / L, or 3.2 g / L, or 3.5 g / L, or 3.8 g / L, or 4.1 g / L, or 4.4 g / L, or 4.7 g / L, or 5.0 g / L, or 6 g / L, or 7 g / L, or 8 g / L, or 9 g / L, or 10 g / L, or more.

[0070] The percent purity of collagenase may be determined for each fraction using any method known to one of skill in the art. One exemplary method is purity measurement using SDS-PAGE, or SDS-PAGE and densitometry. Densitometry software can be any software that converts SDS-PAGE bands on a gel into intensity quantities. One example of such software is Quantity One®. SDS-PAGE gels may have any number of lanes, with common lane numbers including 10 lanes or 15 lanes. The well size for each lane may vary, but common well sizes hold 15-50 microliters. Generally, any amount of protein that fits into the well can be loaded, but a typical amount sufficient for analysis of the product of interest is anywhere from about 0.5 to about 5 micrograms of protein.

[0071] Assays for the presence of neutral proteases in the fractions may also be performed. Any method known to one of skill in the art for determining the presence of neutral proteases in fractions may be used. Exemplary methods include SDS-PAGE, or SDS-PAGE with densitometry, as previously described. Other exemplary methods include casein zymography, or casein zymography with densitometry. Casein zymography is an SDS-PAGE gel that typically uses milk casein embedded in a gel on which neutral proteases can act, but collagenase cannot act. Thus, bands in the zymography gel that contain neutral proteases will appear as a different color than bands that contain collagenase or other impurities. In general, the limit of detection (LOD) for zymography assays related to the present disclosure is about 0.5 ng of protein using densitometry software and about 0.2-0.3 ng visually. However, smaller amounts may be detected depending on the exact method used.

[0072] Generally, when testing for impurities, a larger amount of protein is loaded into the well than that used to analyze the primary product(s) of interest. Impurities may be detected using the same amount of protein used for the primary product of interest, but may require loading at least about 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 12 or 15 or 20 times more than the amount loaded into the well to analyze or detect the primary product of interest. This is especially true as the purity of the product of interest increases. Generally, the amount of protein loaded into the well to detect neutral proteases related to the present disclosure is about 1 microgram to about 100 micrograms, for example, about 5 micrograms, 8 micrograms, or 10 micrograms, or 12 micrograms, or 15 micrograms, or 20 micrograms, or 30 micrograms, or 50 micrograms, or 75 micrograms, or more.

[0073] In certain embodiments, the present disclosure provides three different methods for reducing or eliminating neutral proteases.

[0074] 1. Exclusion Strategy 1 According to this strategy, fractions that do not meet the selected specifications for purity are filtered out and excluded from further forward processing. Each fraction is assayed using SDS-PAGE or SDS-PAGE with densitometry, and each band in a lane on the gel corresponds to the percent of the total protein loaded into the well above the lane. The bands may correspond to collagenase I or collagenase II, or to impurities in the form of fragments of collagenase or other proteins (e.g., neutral proteases). The acceptable level of collagenase I or collagenase II for the pooling step to generate collagenase I or collagenase II products, respectively, is at least 80 area % as measured by SDS-PAGE.

[0075] In another embodiment, an acceptable level of collagenase I for the pooling step to generate a collagenase I product is at least about 81%, or 82%, or 83%, or 84%, or 85%, or 86%, or 87%, or 88%, or 88.5%, or 89.0%, or 89.5%, or 90%, or 90.5%, or 91%, or 91.5%, or 92%, or 92.5%, or 93%, or 93.5%, or 94%, or 94.5%, or 95%, or 95.5%, or 96%, or 96.5%, or 97%, or 97.5%, or 98%, or 98.5%, or 99%, or 99.5%, or 100 area % as measured by SDS-PAGE.

[0076] In one embodiment, an acceptable level of collagenase II for the pooling step to generate a collagenase II product is at least about 81%, or 82%, or 83%, or 84%, or 85%, or 86%, or 87%, or 88%, or 88.5%, or 89.0%, or 89.5%, or 90%, or 90.5%, or 91%, or 91.5%, or 92%, or 92.5%, or 93%, or 93.5%, or 94%, or 94.5%, or 95%, or 95.5%, or 96%, or 96.5%, or 97%, or 97.5%, or 98%, or 98.5%, or 99%, or 99.5%, or 100 area % as measured by SDS-PAGE.

[0077] In one embodiment, the acceptable level of the most abundant impurity for a collagenase I drug substance (after it has been purified and separated from collagenase II, but before it is mixed with collagenase II) is less than about 20%, or 19%, or 18%, or 17%, or 16%, or 15%, or 14%, or 13%, or 12%, or 11%, or 10%, or 9%, or 8%, or 7%, or 6%, or 5%, or 4%, or 3%, or 2%, or 1%, or 0.75%, or 0.5%, or 0.25%, or 0.1% (w / w) of the collagenase I drug substance as measured by SDS-PAGE. The amount of the most abundant impurity may be measured by SDS-PAGE with densitometry.

[0078] In another embodiment, the acceptable level of the most abundant impurity for a Collagenase II Drug Material (after it has been purified and separated from Collagenase I, but before it is mixed with Collagenase I) is less than about 20%, or 19%, or 18%, or 17%, or 16%, or 15%, or 14%, or 13%, or 12%, or 11%, or 10%, or 9%, or 8%, or 7%, or 6%, or 5%, or 4%, or 3%, or 2%, or 1%, or 0.75%, or 0.5%, or 0.25%, or 0.1% (w / w) of the Drug Material as measured by SDS-PAGE. The amount of the most abundant impurity may be measured by SDS-PAGE followed by densitometry.

[0079] Additionally, acceptable levels of the most abundant impurity for mixed collagenase I and collagenase II drug substance are less than about 5%, or 4%, or 3%, or 2%, or 1%, or 0.75%, or 0.5%, or 0.25%, or 0.1% (w / w) of the drug substance as measured by SDS-PAGE. The amount of the most abundant impurity may be measured by SDS-PAGE followed by densitometry.

[0080] Moreover, in certain embodiments, once the selected fractions are pooled into the collagenase I or collagenase II product, the overall purity of such pooled fractions is at least about 95%, or 96%, or 97%, or 98%, or 99%, or 100% as measured by RP-HPLC, hi other embodiments, the overall purity of such pooled fractions is between about 80% and 100% as measured by RP-HPLC.

[0081] 2. Exclusion Strategy 2 According to this strategy, a fraction is rejected from further processing (forward processing) if it has detectable levels of neutral protease as measured by SDS-PAGE, SDS-PAGE with densitometry, casein zymography, casein zymography with densitometry, or a combination thereof. The concentration of protein in each fraction is determined by any method known to those skilled in the art. Each well of the gel can be loaded with any amount of protein. In some embodiments, the amount of protein loaded is about 0.1 micrograms to about 100 micrograms, for example, about 90, 80, 70, 60, 50, 40, 30, 25, 20, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1 micrograms. Any fraction containing detectable levels of neutral protease on the gel is rejected to be pooled.

[0082] 3. Exclusion Strategy 3 According to this strategy, collagenase I fractions are pooled based on the location of the fraction relative to the elution peak and the estimated amount of collagenase II pooled. In other words, the elution peak of collagenase II is used as a identifier to enable pooling of collagenase I fractions.

[0083] The collagenase I fractions pooled can be from any fraction(s) of the collagenase I elution peak. When multiple fractions are pooled, the fractions can be consecutive or non-consecutive. In some embodiments, starting from the collagenase I fraction elution peak maximum, fractions that move evenly on both sides of the peak toward the end are pooled. In another embodiment, starting from the first fraction that meets a particular purity criterion for collagenase I purity, or the most abundant impurity, or both, then fractions that move toward the collagenase I elution peak maximum are pooled. In some embodiments, starting from any fraction that does not contain detectable levels of neutral protease (as measured by any technique known to those skilled in the art), fractions that move toward the beginning of the collagenase I elution peak are collected.

[0084] The amount of collagenase I pooled may differ from the amount of estimated collagenase II (total collagenase II elution peak, pooled collagenase II amount, or otherwise). In some embodiments, the estimated amount of collagenase II may be about 2 times the estimated amount of collagenase I, or about 1.5 times the estimated amount of collagenase I, or about 3 times the estimated amount of collagenase I. Any particular ratio will work, such as about 1:1.2 collagenase I to collagenase II, or about 1:1.1, or 1:1.3, or 1:1.4, or more collagenase I to collagenase II, etc.

[0085] For example, a strategy may include (a) calculating the approximate amount of eluted collagenase II contained in fractions that pass the purity requirements described above, and (b) discarding tail fractions from the collagenase I elution peak that exceed the approximate amount of collagenase II that creates an approximately 1:1 mass ratio of collagenase I to the amount of collagenase II calculated in (a), with the pooling of collagenase I fractions starting with the first pass fractions at the tail of the collagenase I elution peak and moving toward the final collagenase I fraction. The collagenase composition produced by this method is essentially free of neutral proteases.

[0086] In another embodiment, the disclosure relates to a process for producing collagenase I and collagenase II from C. histolyticum, the process comprising the steps of: (a) separating collagenase I from collagenase II using chromatography; (b) collecting elution fractions from step (a) separately for collagenase I and collagenase II; and (c) collecting elution fractions having less than about 91.8% collagenase I or collagen II of the total amount of protein in the fractions as measured by SDS-PAGE with densitometric analysis or less than about 91.8% collagenase I or collagen II of the total amount of protein in the fractions as measured by SDS-PAGE with densitometric analysis. (d) calculating the approximate amount of eluted collagenase II contained in the fractions that pass the purity requirement of (c), and discarding tail fractions from the collagenase I elution peak that exceed the approximate amount of collagenase II that produces an approximately 1:1 mass ratio of collagenase I to the amount of collagenase II calculated in (d), where the discarding of tail collagenase I fractions starts about the tail of the collagenase I elution peak and moves toward the peak maximum. The collagenase composition produced by this method is essentially free of neutral proteases.

[0087] E. Finalization of Drug Products and Formulations Following Removal of Neutral Proteases The collagenase I drug material after the neutral protease exclusion step and before it is mixed with collagenase II may have less than about 50 ng of neutral protease per mg of collagenase I drug material as measured by casein zymography with or without densitometry. Alternatively, it may have less than about 40 ng, or 30 ng, or 20 ng, or 10 ng, or 5 ng, or 1 ng, or less, of neutral protease per mg of collagenase I drug material as measured by casein zymography with or without densitometry.

[0088] In another embodiment, the collagenase II drug material after the neutral protease exclusion step and before it is mixed with collagenase I may have less than about 50 ng neutral protease per mg of collagenase II drug material as measured by casein zymography with or without densitometry. Alternatively, it may have less than about 40 ng, or 30 ng, or 20 ng, or 10 ng, or 5 ng, or 1 ng, or less neutral protease per mg of collagenase II drug material as measured by casein zymography with or without densitometry.

[0089] Additionally, the final mixed collagenase I and collagenase II drug material may have less than about 50 ng of neutral protease per mg of collagenase I and II drug material as measured by casein zymography with or without densitometry, or it may have less than about 40 ng, or 35 ng or 30 ng, or 25 ng, or 20 ng, or 15 ng, or 10 ng, or 5 ng, or 4 ng, or 3 ng, or 2 ng, or 1 ng, or less, of neutral protease per mg of collagenase I and II drug material as measured by casein zymography with or without densitometry.

[0090] In another embodiment, the collagenase drug substance has less than about 1000 ng, or 750 ng, or 500 ng, or 250 ng, or 100 ng, or 75 ng, or 50 ng, or 40 ng, or 35 ng or 30 ng, or 25 ng, or 20 ng, or 15 ng, or 10 ng, or 5 ng, or 4 ng, or 3 ng, or 2 ng, or 1 ng or less neutral protease per mg of collagenase as measured by casein zymography with or without concentration measurement.

[0091] In one example, the amount of neutral protease present is determined by comparing the measured amount of neutral protease in a well with about 5 micrograms to about 20 micrograms of collagenase. For example, samples for the zymography test method are prepared with 0.5 ug of AUX-I per uL of sample and loaded at 20 uL on an SDS-PAGE gel. This results in a 10 ug AUX-I load on the gel. The zymography test method has a detection limit of 0.5 ng of neutral protease. Thus, the detection limit is 0.5 ng NP per 10 ug of AUX-I.

number

[0092] In another embodiment, the present disclosure relates to a collagenase composition derived from C. histolyticum, the collagenase composition comprising a collagenase I product and a collagenase II product, wherein the collagenase product is at least 95 area% pure as measured by reverse phase high pressure liquid chromatography (RP-HPLC), and the collagenase I product is essentially free of neutral proteases.

[0093] In some embodiments, the collagenase composition or pharmaceutical formulation has a mass ratio of collagenase I to collagenase II of approximately 1:1. However, other mass ratios of collagenase I to collagenase II are contemplated. The mass ratio can range anywhere from about 1:10 to about 10:1 collagenase I to collagenase II. Preferably, the mass ratio is about 1:3 to about 3:1 collagenase I to collagenase II. More preferably, the mass ratio is about 0.9:1 to about 1:1.4 collagenase I to collagenase II. For example, the mass ratio can be about 1:1, or 1:1.5, or 1.5:1, or 2.4:1, 1:2.1, or 1:1.3, or 1:1.4, or 1.4:1, or 1.3:1 collagenase I to collagenase II. In one aspect, the mass ratio of collagenase I to collagenase II is about 1: 1. In one embodiment, the collagenase concentrate has an extinction coefficient of 1.528.

[0094] The present invention further provides a collagenase composition comprising collagenase I and collagenase II derived from C. histolyticum, wherein the collagenase I and collagenase II have a mass ratio of collagenase I to collagenase II of about 0.9:1 to about 1:1.4 (e.g., about 1:1), and the collagenase composition has a purity of at least 90 area % as measured by RP-HPLC, or at least about 95 area % as measured by RP-HPLC, or at least 97 area % as measured by RP-HPLC, or at least 98 area % as measured by RP-HPLC, or at least 99 area % as measured by RP-HPLC.

[0095] Pharmaceutical formulations of the present invention comprise a therapeutically effective amount of a collagenase composition of the present invention formulated together with one or more pharma- ceutically acceptable carriers or excipients.

[0096] In another embodiment, the zinc and / or nickel concentrations are controlled during the manufacturing process of collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II, and the formation of the drug product results in a process reproducibility of about 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% for the collagenase product, or a collagenase I product or collagenase II product purity of at least about 95 area%, at least about 96 area%, at least about 97 area%, at least about 98 area%, at least about 99 area%, or 100 area% as measured by RP-HPLC (reverse-phase high pressure liquid chromatography).

[0097] In one embodiment, the present disclosure relates to a method for the isolation and purification of collagenase I and collagenase II from C. histolyticum, in which nickel and zinc levels are controlled during the collagenase purification process resulting in improved process reproducibility of about 95% ultra-pure collagenase I or collagenase II.

[0098] In one example, the purification process includes passing through a filter (e.g., Millipore Millistak HC POD), then adjusting the Mustang Q filtrate to 1M with ammonium sulfate concentration (e.g., at 60% saturation of ammonium sulfate) to promote binding of collagenase to a HIC column, then resuspending or dissolving the precipitated collagenase in ammonium sulfate, then anion ion exchange chromatography (e.g., using a Mustang Q column), then filtration (e.g., a 0.45μ filter), then buffer exchange (e.g., dialysis or tangential flow filtration), then running the filtrate over a hydrophobic interaction chromatography column (e.g., Q-Sepharose), then adding leupeptin, followed by recovering the collagenase from the fermented C. histolyticum (e.g., porcine derived medium (preferably protease peptone)) medium by buffer exchange.

[0099] In another example, the purification process includes passing through a depth filter (e.g., Millipore Millistak HC POD), followed by anion exchange chromatography (e.g., Mustang Q filtration), followed by addition of salt (preferably ammonium sulfate) to a concentration of about 0.8 M to about 1.2 M (e.g., about 1 M ammonium sulfate), followed by a hydrophobic interaction chromatography column (e.g., phenyl sepharose), followed by buffer exchange (e.g., using TFF to remove ammonium sulfate), followed by anion exchange chromatography (e.g., using a Q-sepharose column), followed by collecting fractions of the filtrate from anion exchange chromatography separately for AUX-I and AUX-II, then pooling the fractions of AUX-I together and pooling the fractions of AUX-II together to recover collagenase from fermented C. histolyticum (e.g., fermented in vegetable derived medium). Leupeptin may be added to the purification process in a hydrophobic interaction chromatography step, a first buffer exchange step, and a second anion exchange chromatography step.

[0100] An example of a purification process according to the present method includes filtering through a filter (e.g., Millipore Millistak HC POD), followed by filtration through at least one 0.2 micron filter, followed by anion exchange chromatography (e.g., Mustang Q filtration) where the effluent is adjusted to an ammonium sulfate concentration of about 0.8M to about 1.2M, followed by a hydrophobic interaction (HIC) column (e.g., phenyl sepharose), followed by buffer exchange (e.g., using TFF to remove ammonium sulfate), followed by anion exchange chromatography (e.g., Q-sepharose), followed by collecting fractions of the eluate from the anion exchange chromatography separately for AUX-I and AUX-II, then pooling the fractions of AUX-I together, pooling the fractions of AUX-II together, followed by buffer exchange into a pharmaceutical formulation buffer (e.g., into 10 mM Tris, 60 mM sucrose at pH 8 using tangential flow filtration) to recover the collagenase from the fermented C. histolyticum. Optionally, leupeptin may be added to the purification process in the hydrophobic interaction chromatography step, the first buffer exchange step, and the second ion exchange chromatography step.

[0101] In certain embodiments of the invention, the collagenase purification procedure comprises the steps of: a) filtering the crude recovery solution using ion exchange chromatography, preferably using an anion exchange capsule filter (e.g., MUSTANG Q); b) adding ammonium sulfate, preferably to a final concentration of about 1 M; c) filtering the crude recovery solution, preferably through a 0.45 μm filter; d) subjecting the filtrate to a hydrophobic interaction chromatography (HIC), preferably a Phenyl Sepharose 6 FF (low sub) column; e) adding leupeptin to the eluate, preferably to a final concentration of about 0.2 mM for the post-HIC elution product; f) removing ammonium sulfate and retaining leupeptin, preferably using buffer exchange by tangential flow filtration (TFF); g) filtering the mixture of step (f), preferably through a 0.45 μm filter; and h) filtering the crude recovery solution using ion exchange chromatography, preferably using a Q-Sepharose Fast (Q The method includes the steps of: separating collagenase I and collagenase II using a neutral protease (HP) column; i) performing a neutral protease exclusion step; Optionally, after step (i), further steps may be performed, including: j) performing a buffer exchange of collagenase I and II separately, preferably by preparing and formulating TFF concentrations separately for collagenase I and collagenase II, where the TFF is a tangential flow filter with a 10 and / or 30K MWCO (molecular weight cut-off) PES or RC-polyethersulfone or regenerated cellulose filter membrane (the TFF retains and concentrates selected proteins and provides a means to exchange proteins from one buffer solution to another); and k) filtering the buffer-exchanged collagenase I and II separately, preferably through a 0.2 μm filtration system.

[0102] One embodiment of the method includes a method for producing collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II, where the zinc and / or nickel concentrations are controlled in the process such that the nickel concentration during the manufacturing process is less than about 0.2 ppm and / or the zinc concentration during the manufacturing process is less than about 1 ppm, and these metal concentrations result in a process reproducibility of greater than about 90% and also achieve a purity of greater than about 95 area % as measured by RP-HPLC for collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II. Another embodiment of the method includes monitoring zinc and nickel levels during the manufacturing process of collagenase, collagenase I, collagenase II, or collagenase I and collagenase II, wherein the concentration of nickel during the manufacturing process is maintained at less than about 0.2 ppm and / or the concentration of zinc during the manufacturing process is maintained at less than about 1 ppm, such that these metal concentrations result in a process reproducibility of greater than about 95% of a collagenase composition that is greater than about 95% pure.

[0103] The present disclosure further includes methods for producing collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II, wherein the concentration of nickel during the manufacturing process is less than about 0.2 ppm, and / or the concentration of zinc during the manufacturing process is less than about 1 ppm, and the concentrations of zinc and / or nickel are monitored and / or controlled in the process such that these metal concentrations result in a process reproducibility of greater than about 90% achieving collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II that are greater than about 95% pure. Another embodiment of the method includes monitoring zinc and nickel levels in the manufacturing process of collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II, where the concentration of nickel is maintained at less than about 0.2 ppm during the manufacturing process and / or the concentration of zinc is maintained at less than about 1 ppm during the manufacturing process, such that these metal concentrations result in a process reproducibility of greater than about 95% and a purity of greater than about 95 area % as measured by RP-HPLC for collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II.

[0104] A further embodiment of the method includes a method for producing collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II, wherein the concentration of nickel during the manufacturing process is less than about 0.5 ppm, and / or the concentration of zinc during the manufacturing process is less than about 10 ppm, and the concentrations of zinc and / or nickel are monitored and / or controlled in the process such that these metal concentrations result in a process reproducibility of greater than about 90%, achieving a purity of greater than about 95 area % as measured by RP-HPLC for collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II. Yet another embodiment of the method includes a method for producing collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II, wherein the concentration of nickel during the manufacturing process is less than about 0.5 ppm, and / or the concentration of zinc during the manufacturing process is less than about 10 ppm, and the concentrations of zinc and / or nickel are monitored and / or controlled in the process such that these metal concentrations result in a process reproducibility of greater than about 95%, achieving a purity of greater than about 95 area % as measured by RP-HPLC for collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II.

[0105] A further embodiment of the method includes a method for producing collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II, wherein the zinc and / or nickel concentrations are controlled in the process such that the nickel concentration during the manufacturing process is less than about 0.5 ppm and / or the zinc concentration during the manufacturing process is less than about 10 ppm, and these metal concentrations result in a process reproducibility of greater than about 95%, achieving a collagenase composition that is greater than about 95 area % pure as measured by RP-HPLC.

[0106] Yet another embodiment of the method includes a method for manufacturing collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II, wherein the zinc and / or nickel concentrations are controlled in the process such that the nickel concentration during the manufacturing process is less than about 0.5 ppm and / or the zinc concentration during the manufacturing process is less than about 10 ppm, and these metal concentrations result in a process reproducibility of greater than about 95%, achieving a purity of greater than about 95 area % as measured by RP-HPLC for collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II.

[0107] Further disclosed are methods for producing collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II, wherein the zinc and / or nickel concentrations are controlled in the process such that the nickel concentration during the manufacturing process is less than about 0.5 ppm and / or the zinc concentration during the manufacturing process is less than about 10 ppm, and these metal concentrations result in a process reproducibility of greater than about 90%, achieving a purity of greater than about 97 area % as measured by RP-HPLC for collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II.

[0108] The present disclosure further contemplates a method for producing collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II, wherein the concentrations of zinc and / or nickel are controlled in the process such that the concentration of nickel during the manufacturing process is less than about 0.5 ppm and / or the concentration of zinc during the manufacturing process is less than about 10 ppm, and these metal concentrations result in a process reproducibility of greater than about 95%, achieving a purity of greater than about 97 area % as measured by RP-HPLC for collagenase, collagenase I, collagenase II, or both collagenase I and collagenase II.

[0109] In another embodiment, the disclosure relates to a process for producing collagenase I and collagenase II from C. histolyticum, the process comprising the steps of: (a) separating collagenase I from collagenase II using chromatography; (b) collecting elution fractions from step (a) separately for collagenase I and collagenase II; (c) assaying each fraction for neutral protease using SDS-PAGE, SDS-PAGE and densitometry, or zymography; and (d) collecting collagenase I from the C. histolyticum sample at a concentration of less than about 91.8% of the total amount of protein in the fraction as measured by SDS-PAGE with densitometry analysis. or collagenase II, or any single impurity greater than about 5.8% of the total amount of protein in the fraction as measured by SDS-PAGE with densitometric analysis, or exhibiting detectable levels of neutral protease when tested in (c), and (e) pooling only the fractions of collagenase I that pass for collagenase I product and pooling only the fractions of collagenase II that pass for collagenase II product, wherein the pooled collagenase I product or collagenase II product is at least 95 area % pure as measured by RP-HPLC. The collagenase composition produced by this method is essentially free of neutral proteases.

[0110] In another embodiment, the disclosure relates to a process for producing collagenase I and collagenase II from C. histolyticum, the process comprising the steps of: (a) separating collagenase I from collagenase II using chromatography; (b) collecting elution fractions from step (a) separately for collagenase I and collagenase II; and (c) collecting elution fractions from step (a) having collagenase I or collagenase II that is less than about 91.8% of the total amount of protein in the fraction as measured by SDS-PAGE with densitometric analysis. (d) calculating the approximate amount of eluted collagenase II contained in fractions that pass the purity requirement of (c), and discarding tail fractions from the collagenase I elution peak that exceed the approximate amount of collagenase I that produces an approximate 1:1 mass ratio of collagenase I to collagenase II calculated in (d), where the pool of collagenase I fractions starts about the first passing fraction at the tail of the collagenase I elution peak and moves toward the final collagenase I fraction. The collagenase composition produced by this method is essentially free of neutral proteases.

[0111] The present disclosure relates to a process for producing collagenase I and collagenase II from C. histolyticum, the process comprising the steps of: (a) separating collagenase I from collagenase II using chromatography; (b) collecting elution fractions from step (a) separately for collagenase I and collagenase II; and (c) collecting elution fractions from step (a) that are less than about 91.8% of the total amount of collagenase I or collagenase II protein in the fraction as determined by SDS-PAGE with densitometric analysis, less than about 91.8% of the total amount of collagenase I or collagen II protein in the fraction as determined by SDS-PAGE with densitometric analysis. discarding any fractions having any single impurity greater than about 5.8% of the total amount of protein, (d) calculating the approximate amount of eluted collagenase II contained in fractions that pass the purity requirement of (c), and discarding tail fractions from the collagenase I elution peak that exceed the approximate amount of collagenase I that produces an approximate 1:1 mass ratio of collagenase I to collagenase II amount calculated in (d), where discarding the tail collagenase I fractions starts about the tail of the collagenase I elution peak and moves toward the peak maximum. The collagenase composition produced by this method is essentially free of neutral proteases.

[0112] The present disclosure relates to a manufacturing process for collagenase I and collagenase II derived from C. histolyticum, wherein (a) neutral proteases are essentially eliminated from the collagenase I product during the manufacturing process, and (b) the concentrations of nickel, zinc, aluminum, arsenic, calcium, cadmium, chromium, copper, iron, magnesium, lead, or combinations thereof, are controlled during the manufacturing process.

[0113] In another embodiment, the present disclosure relates to a method for purifying collagenase I and collagenase II comprising: (a) filtering a fermentation broth from C. histolyticum through an anion exchange filter; (b) adding ammonium sulfate to the filtrate of step (a); (c) subjecting the filtrate of step (b) to a hydrophobic interaction chromatography column; (d) adding leupeptin to the eluate of step (c); (e) removing ammonium sulfate from the mixture of step (d); (f) filtering the mixture of step (e); and (g) separating collagenase I and II in the mixture of step (f) using ion exchange chromatography, wherein the concentrations of nickel, zinc, aluminum, arsenic, calcium, cadmium, chromium, copper, iron, magnesium, lead, or combinations thereof are controlled during the purification process.

[0114] In a further embodiment, the present disclosure relates to a method for purifying collagenase I and collagenase II, comprising: (a) filtering a fermentation broth from C. histolyticum through an anion exchange filter; (b) adding ammonium sulfate to the filtrate of step (a); (c) subjecting the filtrate of step (b) to a hydrophobic interaction chromatography column; (d) adding leupeptin to the eluate of step (c); (e) removing ammonium sulfate from the mixture of step (d); (f) filtering the mixture of step (e); and (g) separating collagenase I and II in the mixture of step (f) using ion exchange chromatography, wherein the nickel concentration or zinc concentration is controlled during the purification process.

[0115] The present disclosure also relates to a method for purifying collagenase I and collagenase II, comprising: (a) filtering a fermentation broth from C. histolyticum through an anion exchange filter; (b) adding ammonium sulfate to the filtrate of step (a); (c) subjecting the filtrate of step (b) to a hydrophobic interaction chromatography column; (d) adding leupeptin to the eluate of step (c); (e) removing ammonium sulfate from the mixture of step (d); (f) filtering the mixture of step (e); (g) separating collagenase I and II in the mixture of step (f) using ion exchange chromatography; and (h) removing neutral proteases from collagenase I or collagenase II from the eluate of step (g).

[0116] The present disclosure further relates to a method for purifying collagenase I and collagenase II, comprising: (a) filtering a fermentation broth from C. histolyticum through an anion exchange filter; (b) adding ammonium sulfate to the filtrate of step (a); (c) subjecting the filtrate of step (b) to a hydrophobic interaction chromatography column; (d) adding leupeptin to the eluate of step (c); (e) removing ammonium sulfate from the mixture of step (d); (f) filtering the mixture of step (e); and (g) separating collagenase I and II in the mixture of step (f) using ion exchange chromatography, wherein nickel and zinc levels are controlled during the collagenase purification process.

[0117] In another embodiment, the disclosure relates to a process for producing a drug product consisting of isolated and purified collagenase I and collagenase II from C. histolyticum, wherein collagenase I and collagenase II have a mass ratio of about 1 to 1, and the drug product is at least 95 area % pure as measured by reverse phase high performance liquid chromatography, the process comprising the steps of: (a) fermenting C. histolyticum; (b) recovering a crude fermentation broth comprising collagenase I and collagenase II; and (c) purifying collagenase I and collagenase II from the crude recovery broth via filtration and column chromatography, the steps including: (i) filtering the fermentation broth from C. histolyticum through an anion exchange filter; (ii) filtering the fermentation broth from C. histolyticum through an anion exchange filter; and (iii) filtering the fermentation broth from C. histolyticum through an anion exchange filter. (iii) subjecting the filtrate of step (ii) to a hydrophobic interaction chromatography column; (iv) adding leupeptin to the eluate of step (iii); (v) removing ammonium sulfate from the mixture of step (iv); (vi) filtering the mixture of step (v); and (vii) separating collagenase I and collagenase II in the mixture of step (vi) using ion exchange chromatography; and (d) combining the purified collagenase I and collagenase II from step (c) in an approximately 1:1 ratio, wherein the concentrations of nickel, zinc, aluminum, arsenic, calcium, cadmium, chromium, copper, iron, magnesium, lead, or combinations thereof are controlled during the manufacturing process.

[0118] The present disclosure further relates to a process for producing a drug product consisting of isolated and purified collagenase I and collagenase II from C. histolyticum, wherein collagenase I and collagenase II have a mass ratio of about 1 to 1, and the drug product is at least 95 area % pure as measured by reverse phase high performance liquid chromatography, the process comprising the steps of (a) fermenting C. histolyticum; (b) recovering a crude fermentation broth comprising collagenase I and collagenase II; and (c) purifying collagenase I and collagenase II from the crude recovery broth via filtration and column chromatography, the steps comprising: (i) filtering the fermentation broth from C. histolyticum through an anion exchange filter; (ii) filtering the fermentation broth from C. histolyticum through an anion exchange filter; and (iii) filtering the fermentation broth from C. histolyticum through an anion exchange filter. (iii) adding ammonium sulfate to the filtrate of step (i); (iv) adding leupeptin to the eluate of step (iii); (v) removing ammonium sulfate from the mixture of step (iv); (vi) filtering the mixture of step (v); and (vii) separating collagenase I and collagenase II in the mixture of step (vi) using ion exchange chromatography; and (d) combining purified collagenase I and collagenase II from step (c) in about a 1:1 ratio, wherein the nickel levels, zinc levels, or zinc and nickel levels are controlled during the manufacturing process.

[0119] The present disclosure relates to a process for producing a drug product consisting of isolated and purified collagenase I and collagenase II from C. histolyticum, wherein collagenase I and collagenase II have a mass ratio of about 1 to 1, and the drug product is at least 95 area % pure as measured by reversed-phase high performance liquid chromatography, the process comprising the steps of (a) fermenting C. histolyticum; (b) recovering a crude fermentation broth comprising collagenase I and collagenase II; and (c) filtering and column chromatography. A process for purifying collagenase I and collagenase II from a crude recovery solution via filtration, comprising the steps of: (i) filtering a fermentation broth from C. histolyticum through an anion exchange filter; (ii) adding ammonium sulfate to the filtrate of step (i); (iii) subjecting the filtrate of step (ii) to a hydrophobic interaction chromatography column; (iv) adding leupeptin to the eluate of step (iii); (v) removing ammonium sulfate from the mixture of step (iv); and (vi) eluting the mixture of step (v). (vii) separating collagenase I and collagenase II in the mixture of step (vi) using ion exchange chromatography, the separating step further comprising: (a) collecting elution fractions from (vii) separately for collagenase I and collagenase II; (b) assaying each fraction for neutral protease using SDS-PAGE or zymography; and (c) collecting less than about 91.8% of the total amount of protein in the fraction as measured by SDS-PAGE with densitometric analysis. and (d) pooling only the fractions of collagenase I that pass for the collagenase I product and pooling only the fractions of collagenase II that pass for the collagenase II product. The collagenase composition produced by this method is essentially free of neutral proteases. EXAMPLES

[0120] The following examples are included to demonstrate certain specific embodiments of the present disclosure. However, those skilled in the art will appreciate that in light of this disclosure, modifications can be made to the specific embodiments disclosed and still obtain similar or similar results without departing from the spirit and scope of the present invention. Accordingly, all matters shown are to be interpreted as illustrative and not limiting.

[0121] Example 1 - Categorization and trends of IEX fraction impurities when analyzed by optimized SDS-PAGE method IEX fractions were tested from 13 different lots of manufactured collagenase I and collagenase II obtained from C. histolyticum, as shown in Table 1. Each collagenase fraction collected from the IEX eluate in each of the lots was run on an SDS-PAGE gel and analyzed using gel densitometry using a Bio-Rad densitometer and accompanying densitometric analysis software. For each fraction of each lot, the percent purity of undegraded collagenase I or II, and the percent of each visible impurity (collagenase fragment or otherwise) was determined. [Table 2]

[0122] Using the molecular weight values ​​reported by the software and by comparing the vertical position of the impurities to the vertical position of known molecular weight bands located in the molecular weight marker lane, the impurities are stratified into four main categories (90 kD, 80 kD, 55 kD, and 40 kD). Impurities that do not fit into a category were classified as "other." The SDS-PAGE method utilizing gel densitometry is capable of resolving many more impurity bands than the current SDS-PAGE method alone.

[0123] A. Quantification of major and minor bands in collagenase C. histolyticum ion-exchange chromatography fractions by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and densitometry This method determines the purity and impurities of collagenase C. histolyticum ion exchange chromatography fractions using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using 4-12% NuPAGE gels (MES buffer) and colloidal Coomassie staining. The method uses densitometric analysis of the gels that allows for relative quantification of major bands (percent AUX-I, AUX-II, or both) and minor bands of product fragments or other impurities (percent minor bands).

[0124] Generally, the amount of collagenase loaded into the wells of a gel for SDS-PAGE is linear to the intensity of the bands as seen using densitometry software. For example, the linearity of response for collagenase bands is R of >0.98. 2 This can be demonstrated by plotting collagenase against μg of a known reference standard over a range of 1.123 to 1.875 μg of collagenase (75% to 125% of the target) with values. An example of a typical linear plot is shown in Figure 1.

[0125] A reference standard should be used for the sample being analyzed. For example, an AUX-I intermediate reference standard can be used to analyze the elution fraction of AUX-I from the IEX step. The amount of reference standard and sample needed to produce the final sample for the gel (C2 = 0.15 mg protein / mL) is calculated by the following formula: C1V1 = C2V2 C1 = initial concentration of the sample to be tested (mg / mL) V1 = volume of sample required at initial concentration (μL) C 2= 1mg / mL (final sample preparation concentration) V 2= Final sample preparation volume (μL)

[0126] B. Impurity Categorization i. Approaches to Impurity Classification Fraction impurities can be determined based on analysis of gels resulting from analyzing samples of fractions using SDS-PAGE with densitometry. Each gel is placed on a densitometer and the gel image is scanned into densitometer software (e.g., Bio-Rad Quantity One software) using a densitometer (e.g., Bio-Rad densitometer). The analyst verifies that the software has correctly identified the gel lanes and appropriate bands. The software produces an annotated image report, such as the representative image shown in Figure 2.

[0127] In Figure 2, lane 1 contains a molecular weight standard with bands of known molecular weight. Lane 2 contains the appropriate AUX intermediate reference standard. The remaining lanes contain samples from each IEX fraction. The main product band is the upper band in each lane, and each of the additional bands further down represent impurities.

[0128] Trend analysis of impurities observed in IEX fractions can provide important process performance information. Impurities should be consistently categorized based on molecular weight per lot for trend analysis to be successful. Although concentration measurement software may report an apparent molecular weight for each impurity, categories based on the reported apparent molecular weight will have some normal variation in the gel image, and the variation combined with nearby bands can lead to ambiguous and overlapping classification ranges. Instead, the recommended molecular weight categories for AUX-I and AUX-II impurities can be based on a visual comparison of the vertical (y-axis) position of the fraction impurity to the vertical position of the band in the known molecular weight lane.

[0129] A total of 128 AUX-I and 99 AUX-II fractions were analyzed. Impurities that appeared in 3 or more of the 13 batches were assigned to a category. A description of each category is provided below.

[0130] ii. Impurity classification results AUX-I The AUX-I fraction impurity categories include 33 kD, 45 kD, 55 kD, 80 kD, 90 kD, and 96 kD. A description of each impurity category is provided in Figures 3-8. Each figure includes the following: · First panel: Percent frequency with which a given impurity was observed in each fraction of the 13 lots included in the evaluation. · Second panel: Mean (dots) and range (error bars) of relative amounts (reported as percentages) for a given impurity per fraction number. · Third panel: Description of the visual vertical position of the bands relative to molecular weight standards. · Fourth panel: the range of apparent molecular weights reported by the software for a given band.

[0131] AUX-II The AUX-II fraction impurity categories include 25 kD, 38 kD, 50 kD, 60 kD, 80 kD, 90 kD, 92 kD, and 96 kD. A description of each impurity is provided in Figures 9-16. The figure panels are the same as those described for AUX-I.

[0132] Of the 877 impurity bands analyzed, three bands did not fit into the above categories and were classified as “other.” Details of these impurities are included in Table 2. [Table 3]

[0133] C. Impurity trends i. Approach to impurity trends After the impurities are categorized, they can be trended to monitor process performance. Fraction impurity trends can be complex because there are multiple impurities at different protein concentrations, multiple IEX fractions, and results are reported as relative amounts. However, trends can be simplified by reporting fraction impurities as a percentage of the impurity of interest relative to the total protein per AUX peak, as shown below.

[0134] The amount of protein in each IEX fraction was determined by the fraction volume and UV A280 The fraction protein concentration can be determined by multiplying the fraction protein concentration by the fraction. The resulting value is grams of protein per fraction. A typical profile of grams of protein per fraction is provided in FIG.

[0135] Utilizing SDS-PAGE and densitometry methods as described results in IEX fraction impurities as a percentage of total protein detected in each SDS-PAGE gel fraction lane. The same amount of protein (1.5 μg protein / lane) is loaded into each fraction lane. Therefore, grams of each impurity in each fraction can be estimated by converting the impurity percentage to a decimal (dividing by 100) and then multiplying by grams of protein per fraction. For example, Figure 18 shows a typical profile of a 90 kD impurity, expressed in grams, commonly observed in both AUX-I and AUX-II fractions.

[0136] Grams of protein for each AUX I, AUX II, or impurity can be calculated per fraction. Alternatively, grams of protein and impurities of interest per fraction can be summed for each AUX peak. A visual representation of this mathematical operation is provided in Figure 19.

[0137] The percentage of the impurity of interest relative to the amount of protein per AUX peak can then be calculated by dividing the grams of impurity of interest by the total AUX peak grams of protein, and then multiplying the resulting number by 100 to convert the value to a percentage. The equation for arriving at the final value expressed as a percentage of the impurity of interest relative to the total protein per AUX peak is summarized below:

number

[0138] The above equation can be used to trend the amount of each detected impurity to assess process performance. The above equation can also be easily modified to represent the relative amount of the impurity of interest in only the forward processed fraction.

[0139] Trending can be performed using a variety of methods. One method uses Shewhart control charts for individual values. Shewhart control charts for individual values ​​are statistical tools used to distinguish between results due to routine variation (within the control chart limits) and results due to exceptional variation (above the control limits). To estimate process variability, the moving range of two consecutive observations is used to calculate the control limits. The green horizontal line in the control chart represents the average result. The red horizontal lines represent the upper and lower control limits and are calculated as follows:

number

[0140] The limits produced using Shewhart control charts for individual values ​​are not intended to indicate product quality, therefore results outside the control limits do not automatically indicate a quality impact and need to be evaluated holistically in the context of all available data.

[0141] ii. Impurity trend results Individual Shewhart control charts were used to trend the impurities for the lots listed in Table 1. AUX-I impurities are trended in Figure 20 and AUX-II impurities are trended in Figures 21A and 21B. Lots are characterized by manufacturing site. A different x-axis scale was used for each control chart. Recommended impurity categories based on the evaluations presented in this report are included in Table 3. [Table 4]

[0142] Trending of IEX fraction impurities can provide important process performance information. Reporting impurities as a percentage of the impurity of interest relative to the total protein per AUX peak is recommended to simplify impurity trending.

[0143] Example 2 - Metals in the Refining Process A. Abnormal impurity levels in the manufacture of collagenase I and II products A substantial decrease in the purity of the collagenase I and II products occurred during the routine manufacturing process. Analysis was performed that included determining the identity of impurities seen after the ion exchange chromatography (IEX) step of the manufacturing process. The collagenase I product has a molecular weight (MW) of approximately 113 kDa and the collagenase II product has a MW of approximately 112 kDa. High levels of impurities inexplicably emerged during routine manufacturing of collagenase, apparently from the IEX step. These high levels of impurities appeared at 40 kDa, 50 kDa, 55 kDa, 80 kDa, 90 kDa, and 96 kDa on an optimized SDS-PAGE gel.

[0144] B. Potential Causes of Increased Impurity Products The cause of the high levels of impurities was then investigated. In addition to collagenase (AUX-I and AUX-II), C. histolyticum produces two proteases, clostripain and neutral protease. Both proteases require metal cofactors for activity. Calcium is a cofactor for clostripain and is required for activity. Calcium is also a cofactor for the neutral protease, but is required only for structural stability. If calcium is removed from the binding site of the neutral protease, the enzyme is autocatalytically degraded. Zinc is a cofactor for the neutral protease and is required for neutral protease activity.

[0145] Extensive product degradation in a manufacturing process stream was first detected after a hydrophobic interaction chromatography (HIC) unit operation during a manufacturing run. The HIC process buffer feedstocks investigated included ammonium sulfate (AS), Tris-base (tromethamine, or "Tris"), and approximately 37% concentrated hydrochloric acid (HCl). The investigation involved a determination of the metal content in the feedstock. Ion-coupled plasma mass spectrometry (ICP-MS) was selected to measure the metal content due to its sensitivity and the broad panel of metals it can detect and quantitate. ICP-MS has parts per billion (ppb) sensitivity to measure 34 metal ions, including calcium and zinc.

[0146] The trace metals tested by ICP-MS are presented below for each of the above raw materials with respect to calcium, nickel, and zinc levels. These particular metal levels were found at significantly higher concentrations in several lots of ammonium sulfate tested (Tables 4-6) compared to the Tris lots (Table 7) and HCl lots (Table 8). Tris-base and concentrated HCl were excluded from further testing based on the low levels of zinc, calcium, and nickel metals detected, as well as their relatively low concentrations used in the process buffers.

[0147] i. Ammonium sulfate test: [Table 5]

[0148] Calcium was ruled out as a cause of impurities because the calcium levels in the ammonium sulfate lots that produced low levels of impurities (approximately 5 area % or less as measured by SDS-PAGE) were higher than the ammonium sulfate lots that caused high levels of impurities (greater than 5 area % as measured by SDS-PAGE). These data ruled out calcium contamination as a potential cause of collagenase I and II degradation to impurities observed in the manufacturing runs. Tested ammonium sulfate lots that produced high levels of impurities during collagenase production had increased levels of zinc and nickel present. These data suggested a possible link between nickel contamination and the observed product degradation. The impact of zinc on the manufacturing process was not clear from these data because both lots contained significant high levels of nickel. The potential impact of zinc and nickel ion levels was explored in spiking studies discussed further below.

[0149] Historic lots of ammonium sulfate were also examined. [Table 6] [Table 7]

[0150] Calcium levels for the five historical ammonium sulfate lots (Table 5) ranged from 240 to 870 ppb. Furthermore, none of the elevated calcium levels produced the high levels of impurities seen in the manufacturing process anomalies. Zinc content ranged from about 130 to 2980 ppb. Nickel content of the historical lots ranged from about 140 to 180 ppb. Comparing these data with the nickel content of the two ammonium sulfate lots that produced high levels of product impurities, about 1150 ppb and about 1020 ppb, respectively, shows low tolerance to nickel contamination. The relative difference in nickel between the ammonium sulfate lots is 6 to 8 times, but this difference is much lower than other metals such as zinc, which varied by more than 10 times within the historical ammonium sulfate lots. Also, the low zinc content (250 ppb) of Avantor ammonium sulfate lot 99428 suggests that the high levels of impurities seen in the manufacturing process were highly nickel sensitive. In summary, the study demonstrated that the product of the refinery process was sensitive to both nickel and zinc levels in the process, but that the tolerance to zinc in the process was higher than the tolerance to nickel.

[0151] Calcium levels in the EMD-Millipore ammonium sulfate lots (Table 6) were lower than the deviations associated with the ammonium sulfate lots tested and were comparable to calcium levels in historical ammonium sulfate lots (Table 5). Zinc ion levels in the EMD-Millipore ammonium sulfate lots were generally lower than the Avantor lots tested (Table 3). Nickel ion levels were below the limit of quantitation (LOQ) of 100 ppb for the four developmental ammonium sulfate lots EMD-Millipore tested.

[0152] ii.Tris test: The two lots of Tris-base represent the raw material used in commercial production (Avantor lot 61712) prior to and through the manufacturing process that produced high levels of impurities, and the development material used for the laboratory scale exploratory study (Fisher lot 126928). As shown in Table 7, the Tris-base contained comparable levels of calcium and nickel between the batches. The notable difference between the two lots was the zinc content. The zinc content of the development material was below the LOQ, 100 ppb. Because the levels of nickel and zinc present during the manufacturing processes that produced the high and low levels of impurities were very similar, Tris was not considered to be a raw material that contributed to collagenase degradation. [Table 8]

[0153] iii. HCl test: Both HCl lots were raw materials used during manufacturing processes that produced high levels of impurities. Table 8 shows that calcium levels were comparable to those detected in the AS lots, while zinc and nickel contents were below the LOQ (100 ppb). Because the levels of nickel and zinc present for the manufacturing processes producing high and low levels of impurities were very similar and below the detectable limits, HCl was not considered a raw material contributing to collagenase degradation. [Table 9]

[0154] C. Zinc and Nickel Spiking Studies Laboratory-scale purification studies were performed using two different ammonium sulfate lots known to be essentially free of zinc or nickel (Fisher lot 1476214A, Table 4, and EMD lot AMO556316, Table 5) to determine whether the presence of zinc or nickel at the relative concentrations detected in the deviations associated with the ammonium sulfate lots was responsible for the product degradation observed in manufacturing processes that produced high levels of impurities. Initial exploratory studies at laboratory scale using ammonium sulfate Fisher lot 1476214A produced a representative number of pass-through AUX-I and AUX-II fractions, nearly all of which had an unresolved AUX-I / AUX-II purity of at least 91.2% as measured by SDS-PAGE gel densitometry, and the most abundant degraded AUX-I or AUX-II impurity was 5.8% or less as measured by SDS-PAGE gel densitometry. Testing of EMD ammonium sulfate lots showed zinc and nickel contents below the LOQ (<100 ppb).

[0155] Using these ammonium sulfate lots for the study, a control run consisting of qualified HIC and IEX small scale models was performed using previously frozen, thawed Mustang Q filtrate (MQF) from a manufacturing process that produced high levels of impurities as starting material. The MQF was processed through the HIC, TFF, and IEX steps. Both the HIC and IEX steps were loaded at the midpoint of the effective load range for the chromatography steps. All unit operations were performed under refrigerated conditions with the chromatography cabinet set at 10°C for the HIC step and 4°C for the IEX step. IEX fractions were analyzed by SDS-PAGE with concentration measurements. Fractions carried forward for pooling required undegraded collagenase I or II at 91.2% purity or greater and the most abundant degraded collagenase I or II impurities at 5.8% or less.

[0156] The zinc spiking study utilized ammonium sulfate from Fisher lot 1476214A. This study was preceded by a nickel spiking study utilizing ammonium sulfate from EMD lot AM0556316. Both ammonium sulfate lots were determined to have zinc and nickel levels below 100 ppb.

[0157] In the zinc spiking study, ammonium sulfate from Fisher lot 1476214A was spiked with ZnCl2 to raise the zinc concentration in the HIC buffer to approximately 84 ppm. The IEX fraction results from material processed with the zinc-spiked buffer were compared to IEX fraction results resulting from material processed without additional zinc added to the process buffer. For the nickel spiking study, ammonium sulfate from EMD lot AM0556316 was spiked to approximately 1.2 ppm with NiCl2. The studies are discussed separately below.

[0158] i. Zinc spiking studies: The control run, DEV-25C, produced typical results as determined by SDS-PAGE analysis of IEX fractions. DEV-25B, performed with zinc-spiked buffer, resulted in significant degradation of AUX-I and AUX-II.

[0159] IEX SDS-PAGE images for the control run, DEV-25C, are presented in Figures 22 and 23. Concentration data for these figures are tabulated in Tables 9 and 10 below. The AUX-II peak consisted of 6 fractions, of which fractions 1-5 met the pooling criteria. The last fraction did not meet the pooling criteria, which is consistent with typical process performance (see Figure 22 and Table 9). The AUX-I peak consisted of 10 fractions (fractions 7-16), of which fractions 9-15 met the pooling criteria. For an AUX-I fraction to meet the pooling criteria, the fraction had to be at least 91.2% pure in undegraded AUX-I and 5.8% or less pure in the most abundant degraded AUX-I or AUX-II impurity. Similarly, for an AUX-II fraction to meet the pooling criteria, the fraction had to be at least 91.2% pure in undegraded AUX-I and 5.8% or less pure in the most abundant degraded AUX-I or AUX-II impurity. Under these specific requirements, the first two AUX-I fractions (fractions 7 and 8) as well as the last AUX-I fraction (16) did not meet the pooling criteria (see Figure 23 and Table 10). A bell-shaped purity curve across the fractions is normal, as the fractions represent the peaks of AUX-I or AUX-II being purified. [Table 10] [Table 11]

[0160] In contrast, SDS-PAGE / densitometry analysis of IEX fractions produced from the 84 ppm zinc spiked run DEV-25B contained collagenase impurity fragments well above the maximum allowable limit of <5.8% (%, most abundant impurity). Neither the AUX-I nor AUX-II fractions met the pooling criteria (Figures 24 and 25, Tables 11 and 12). This study indicates that zinc contamination of the HIC buffer to levels found in ammonium sulfate from Avantor lot 70339 promotes collagenase degradation as observed in the manufacturing process and produced high levels of impurities. [Table 12] [Table 13]

[0161] ii. Nickel Spiking Studies: IEX SDS-PAGE images for the control run, DEV-25A, are presented in Figures 26 and 27. Densitometric data for these figures are tabulated in Tables 13 and 14 below. The AUX-II peak consisted of 6 fractions. All fractions met the pooling criteria (see Figure 26 and Table 13). The AUX-I peak consisted of 10 fractions, of which fractions 9-15 met the pooling criteria. The first three AUX-I fractions (fractions 7-9) and the last AUX-I fraction (fraction 16) did not meet the pooling criteria (see Figure 27 and Table 14). Again, a bell-shaped purity curve across the fractions is normal, as the fractions represent the peaks of AUX-I or AUX-II being purified. [Table 14] [Table 15]

[0162] In contrast, in run 4, which was spiked with nickel, the HIC buffer contained 1.2 ppm nickel chloride. The presence of nickel at 1.2 ppm in the HIC buffer produced product fragments in the AUX-I and AUX-II fractions comparable to those experienced in manufacturing processes that produce high levels of impurities. Neither the AUX-I nor AUX-II fractions met the pooling criteria (Figures 28 and 29, Tables 15 and 16). These results demonstrate that nickel contamination at 1.2 ppm in ammonium sulfate lots promotes collagenase degradation, consistent with the high levels of impurities seen from an aberrant manufacturing process. [Table 16] [Table 17]

[0163] D. Isolation and Identification of Neutral Proteases from HIC CIP Water Washes Prior to the present invention, impurities in the form of collagenase I or II fragments were believed to be the result of clostripain degradation during the collagenase purification process, a known host cell contamination. Clostripain is believed to cleave collagenase I and II into fragments, which were known to be present and active in C. histolyticum fermentation and during the collagenase purification process. Previous homology modeling and genome sequence analysis in Herber (US Patent Application Publication No. 2015 / 0010532) predicted that a neutral protease (another cleavage enzyme) was secreted during fermentation, but was non-functional due to significant mutations in the autocatalytic region between the C-terminal end of the prosequence region and the N-terminal end of the mature protein.

[0164] Increased levels of impurities in the routine manufacturing process necessitated further investigation. This investigation determined that the impurities were fragments of collagenase I (AUX-I) and collagenase II (AUX-II) produced by the neutral protease from C. histolyticum. The amino acid sequences of AUX-I and AUX-II were used in the Peptide Cutter program of the bioinformatics resource portal ExPASy.org to generate a theoretical protein cleavage site list using clostripain and thermolysin cleavage rules. Thermolysin was selected as a model protease for the neutral protease from C. histolyticum because it is in the haplotype of the M4 family. Examination of the cleavage sites produced by clostripain and thermolysin in the AUX-I and AUX-II sequences resulted in positive matches of terminal cleavage sites for the majority of the characterized routine product fragments (40 kDa, 50 kDa, 55 kDa, 80 kDa, 90 kDa, and 96 kDa) contained in the IEX fraction. The main product fragments and the characterization of their respective cleavage enzymes are summarized in Table 17 below: [Table 18]

[0165] A complete list of IEX fraction impurity categories from the IEX fractions is provided in Table 3 of Example 1. Impurities can occur at any one of the molecular weights (MW) recorded in Table 3. However, the most common impurities occur at 45 kDa, 55 kDa, 80 kDa, and 90 kDa for AUX-I and 38 kDa, 50 kDa, 60 kDa, 80 kDa, and 90 kDa for AUX-II. Occasionally, impurities at molecular weights other than those specified in Table 3 may occur (represented as "other" in Table 3).

[0166] To produce purified neutral protease from C. histolyticum, a method was developed to purify neutral protease from the clean-in-place (CIP) waste stream of a hydrophobic interaction chromatography (HIC) process. Two liter fractions of HIC CIP water wash effluent were manually collected across the peak. The fraction containing the peak maximum was concentrated 10-fold in volume and then buffer exchanged using 6 diavolumes of 10 mM Tris, 3 mM CaCl2, pH 8.0, 5°C. This procedure was carried out using a single 0.1 ml 2 The purification was performed using a Pellicon® 5 kDa nominal molecular weight cutoff tangential flow filter. The concentrated buffer exchanged wash water was loaded onto a Q Sepharose High Performance (Q HP) column equilibrated with 10 mM Tris, 3 mM CaCl2, pH 8.0. After loading the equilibration buffer, the column was washed before using a 25 column volume gradient from 0-50% buffer B (10 mM Tris, 3 mM CaCl2, 360 mM NaCl, pH 8.0). Two clusters were combined peaks eluted from the column and collected by automated fractionation. SDS-PAGE analysis with Coomassie staining showed that a 34 kDa protein eluted within the second peak cluster, without other protein species within the detectable limits (Figure 30, gel lane 3). The peak maximum sample was buffer exchanged into 20 mM potassium phosphate and stored frozen at -70°C.

[0167] This wastewater stream, the water wash step of the HIC cleaning in place (CIP) procedure, is produced by flushing the HIC column with water for injection (WFI) after completing the HIC unit operations prior to starting the CIP by flushing the column with 0.5 M NaOH. The water wash caused the elution of the neutral protease from the HIC column. Isolation of the neutral protease from the wastewater stream was successful, confirming the presence of the enzyme in the collagenase manufacturing process, albeit in the HIC process wastewater stream. The isolated neutral protease had an apparent molecular weight of 34 kDa.

[0168] The 80 kDa fragment was the only characterized fragment that did not fit any protease cleavage rules using the Peptide Cutter program due to a proline residue at the P2' position of the potential cleavage site. However, by mixing AUX-I with either process-isolated neutral protease or a commercial version of C. histolyticum neutral protease (from Serva), the 80 kDa fragment was again generated. Furthermore, evaluation of the cleavage site for the 80 kDa fragment using the MEROPS database revealed that thermolysin is capable of cleaving the Thr-693-Gly694 peptide bond to generate the 80 kDa fragment from AUX-I. Thus, all five AUX-I and AUX-II product fragments that typically appear during the production of collagenase products are proteolytic cleavage products of neutral proteases. Under production conditions, the 50 kDa and 96 kDa impurities may also be cleavage products from clostripain degradation.

[0169] The uncharacterized 96 kDa fragment present in the AUX-II fraction theoretically fit multiple cleavage sites, both N- and C-terminal, for both clostripain and neutral protease. The 96 kDa fragment was again generated by exposing AUX-II to neutral protease. This in-silico analysis and neutral protease digestion studies implicate neutral protease as the enzyme responsible for the formation of product fragments in C. histolyticum fermentation and / or purification.

[0170] As mentioned above, the neutral protease isolated from the collagenase production process had an apparent molecular weight of 34 kDa, indicating that the prosequence of the neutral protease was cleaved either by autoactivation or by another host enzyme, suggesting that the neutral protease is active in the C. histolyticum production process.

[0171] Next, an investigation was conducted to determine whether the prosequence would self-cleave itself to produce an active and mature neutral protease, although this was presumably not possible. Identification of the N-terminus of the neutral protease would provide the primary data to assess the autocatalytic competence of the proenzyme by comparing the consensus sequence requirements for thermolysin with the sequence adjacent to the N-terminus of the C. histolyticum neutral protease. The purified neutral protease was then subjected to N-terminal identification, which identified the Gln220 residue as the N-terminus of the enzyme. This N-terminus matches the N-terminus of the C. histolyticum neutral protease recently published by Maeda et al. in 2015.

[0172] Furthermore, neutral protease, and not clostripain, was found to be the proteolytic enzyme that produced all of the routine collagenase fragments observed during the collagenase manufacturing process. Use of isolated neutral protease from the HIC CIP water wash in AUX-I / AUX-II digestion studies produced product fragments consistent with the apparent molecular weights of the routine fragments as determined by SDS-PAGE analysis. These digestion studies used neutral protease isolated from the manufacturing process as described above.

[0173] Identification of the N-terminus of E. histolyticum neutral protease The N-terminal identification methods used were orthogonal: Edman degradation sequencing and N-terminal identification by LC-MS / MS after digestion with Lys-C / trypsin. Samples for both studies were prepared from the same purification run and represented different fractions of the neutral protease elution peak. Edman degradation analysis identified the following five residues in the N-terminal region of the neutral protease: Gln220-Ala224. LC-MS / MS digestion analysis identified a larger stretch of the N-terminal region of the neutral protease, which was detected as two peptides in the analysis: Gln220Ala-Arg227 and Gly228-Lys236. The identified amino acids are highlighted in Figure 31.

[0174] The identified N-terminus of the neutral protease was used to search for a hexapeptide consensus sequence region of the proenzyme. The consensus sequence corresponds to the three residues at the C-terminus of the prosequence of thermolysin combined with the N-terminus and the following two residues of the mature enzyme. For thermolysin, the consensus sequence is Val230LysSerIleTherGly236. Ile233 is the N-terminus of thermolysin. To describe the consensus sequence, the protease subsite nomenclature of Schechter & Berger is used (see Figure 32). The amino acid features required for autocatalysis of thermolysin are a non-polar residue at the P3 position (Gly, Ala, Ile, Leu, or Val), a polar residue or proline (Ser, His, Glu, P) at the P1 position, and a non-polar residue at the Ρ1' position. Inspection of the consensus sequence of thermolysin shows that it meets all three requirements.

[0175] Using the identified N-terminus (Gln220) of C. histolyticum neutral protease, the theoretical consensus sequence is Lys217SerCysGlnAlaThr222. This hexapeptide region does not follow any of the subsite characteristics for autocatalysis, except that Cys219 is a polar residue. However, it is not listed as a suitable residue in the rules. It is unclear from the literature whether the structural characteristics of cysteine ​​would exclude it as a suitable residue for the P1 site, and this has not been studied by researchers. Based on the rules of the thermolysin autocatalytic sequence, C. histolyticum neutral protease should not be an autocatalytic component.

[0176] The complete gene product sequence (secretory, pro-, and mature sequences) of C. histolyticum ATTC19401 neutral protease was recently published by Maeda et al. 2015. The gene product was cloned and expressed to allow for the study of neutral protease substrate specificity yielding active neutral protease. Recombinant neutral protease ( rA sequence alignment of the C. histolyticum neutral protease gene product sequence (nprA) and the C. histolyticum neutral protease gene product sequence resulted in 100% homology (see FIG. 33). Thus, it was determined that the neutral protease in the collagenase C. histolyticum (CCH) production process is autocatalytic and secreted as an inactive proenzyme precursor that autocleaves the prosequence into a functional mature enzyme in the growth medium, a typical feature of CCH fermentation.

[0177] F. Investigating the presence of neutral proteases during the manufacturing process An investigation was performed to determine whether the active neutral protease was specific to the abnormal manufacturing process run or a routine feature of the process. The collagenase fragment pattern during the manufacturing process was analyzed. The degradation patterns of the AUX-I and AUX-II IEX fractions during the abnormal manufacturing process run were consistent with the relative molecular weights of the typical fragments in a typical manufacturing process run, but at a much higher level. Evaluation of the cleavage sites of each characterized product fragment against clostripain and neutral protease cleavage rules suggested that the neutral protease produced a product fragment that was identified as an impurity. SDS-PAGE analysis of the entire HIC unit operation, including the CIP wastewater stream, was performed to determine whether the approximately 34 kDa protein band was present in both process streams. The water wash portion of the CIP process contained a band consistent with this apparent molecular weight by SDS-PAGE evaluation. The HIC water wash sample also showed degraded AUX-I and AUX-II in analytical spiking studies (similar to the pattern observed in the abnormal manufacturing process run).

[0178] Besides these steps in the collagenase manufacturing process, residual neutral protease was also found in the HIC eluate and TFF-1 retentate product streams. Examination of the IEX fractions, AUX-I and AUX-II IEX pools, AUX-I intermediates, AUX-II intermediates, and drug product revealed that the neutral protease was sequestered in the tail AUX-I fraction of the IEX unit operation where it co-elutes with the tail of the AUX-I peak. These fractions may be rejected from the product stream due to fractions not meeting the in-process purity limits of the SDS-PAGE / densitometric analysis.

[0179] Example 3 - Process Control Neutral protease activity was found in the Mustang Q filtrate, HIC eluate, and tangential flow filtration (TFF) concentrate process samples of three manufacturing process runs. IEX AUX-I fractions of a routine manufacturing run were tested in a neutral protease zymography activity assay, and protease showed activity in the tail fractions of these samples. These data suggest that residual neutral protease that co-elutes from the HIC column is sequestered in the tail AUX-I fractions that elute from the IEX column. The majority of the tail fractions do not meet the in-process purity limits of the SDS-PAGE densitometric fraction test, and therefore are not pooled to produce AUX-I intermediates. However, additional procedural controls were implemented to ensure that fractions containing small amounts of neutral protease were not processed forward. Additional procedures act to reject additional AUX-I tail fractions from forward processing. The distinct locational nature of neutral protease elution, as suggested by zymographic activity analysis, indicates that these procedures act to eliminate neutral proteases from the product stream prior to formulating the drug product. These process controls are discussed below.

[0180] Zymography is an electrophoresis-based enzyme activity assay that utilizes a substrate (casein) copolymerized within an SDS-PAGE gel and the ability of certain enzymes to regenerate upon removal of SDS from the gel. After removal of SDS, the regenerated enzyme is allowed to incubate for a period of time to degrade the copolymerized casein substrate. Activity is detected by Coomassie staining, where the degraded casein regions, corresponding to the apparent molecular weight of the enzyme, appear as white zones that do not absorb the stain. The remainder of the gel is stained blue by Coomassie dye, which complexes with the casein contained in the gel. Casein is a milk protein that is susceptible to proteolytic attack by C. histolyticum neutral protease, but not by C. histolyticum clostripain or collagenase. Thus, this PAGE-based activity assay provided the ability to screen manufacturing product streams for the presence of active neutral protease.

[0181] The assay demonstrated neutral protease signal in the Mustang Q filtrate, HIC eluate, TFF-1 retentate, and the final IEX AUX-I tail fraction. The neutral protease is largely removed in the HIC step, but remains strongly bound to the column and elutes during the water wash of the column cleaning process. Trace amounts of neutral protease are co-purified with collagenase from the HIC column and bind to the IEX column. This trace amount of neutral protease begins to desorb from the column at the end of product collection in the final AUX-I fraction. Neutral proteases can be removed by separating collagenase I from collagenase II and collecting each eluate in fractions across the eluate peak, followed by at least one of the following exclusion steps: (1) incorporating improved criteria limits for collagenase I, collagenase II, or both collagenase I and collagenase II fractions that can be pooled, (2) incorporating an improved fraction pooling strategy that rejects fractions that contain detectable levels of neutral proteases when tested by SDS-PAGE or zymography electrophoresis, and (3) incorporating an approximately 1:1 (equal yield) pooling strategy of AUX-I and AUX-II purified fractions. Zymography data and IEX process control are now considered.

[0182] A. Improvement of pooling - collecting only neutral protease-free fractions for casein zymography electrophoresis Samples from the lab-scale run of Mustang Q filtrate, HIC eluate, TFF-1 concentrate, AUX-I and AUX-II intermediates, and drug product were tested by zymography electrophoresis. Data are presented in Figures 34 and 35. These samples were from two different runs (Dev-13 and Dev25A). Zymography images were scanned with densitometry and annotated images used to help highlight the neutral protease bands as some sample types showed low neutral protease signal. To aid in band visualization, an inverted image of the gel was used to produce a typical Coomassie gel image (blue stained proteolytic bands). Neutral proteases were detected in early process samples such as MQF, HIC load, and HIC eluate (see Figure 34). The neutral protease band is readily observed in the MQF and HIC load samples at an apparent molecular weight of approximately 34 kDa. Note that although the total protein load for these three sample types is increased from 2 μg (MQF), 3 μg (HIC load), and 4 μg (HIC eluate), the HIC eluate neutral protease response is substantially lower. The neutral protease band in the HIC eluate sample was faint, but sufficient to call for highlighting using the "selected bands" feature of the concentration measurement software. This demonstrates significant clearance of neutral proteases over the HIC step.

[0183] A higher loading mass of 10 μg total protein for the TFF-1 concentrate, AUX-I and AUX-II pools, intermediates, and drug product was used to amplify potential neutral protease signals in these samples. The TFF-1 concentrate (Dev-25A) sample contained neutral protease at a sufficient concentration to produce a diffuse band at approximately 34 kDa. The AUX-I pool, AUX-I intermediate, and drug product contained detectable neutral protease, highlighted by the densitometry software and observable in lanes 6, 8, and 12 (see FIG. 35). The 34 kDa band was visible in the gel. However, the images obtained were not with sufficient fidelity to clearly see the band. The AUX-II intermediate did not contain detectable neutral protease, nor did its precursor AUX-II pool (FIG. 34). These data demonstrate that the neutral protease is separated from the product throughout the IEX step and co-elutes in the AUX-I peak as evidenced by the AUX-I pool and AUX-I intermediate. These data also demonstrate that any fractions containing detectable levels of neutral protease as measured by zymography should be rejected and not included in the pooling step.

[0184] IEX fractions of AUX-I and AUX-II eluate peaks from a typical manufacturing process were run in a casein zymography assay to determine whether neutral protease was detectable in the fractions and whether it was contained in a specific location. Figures 36 and 37 contained gel images from the assay. None of the fractions of AUX-II contained neutral protease at a concentration detectable by this method. Selected samples were run at higher concentrations to exclude the presence of neutral protease in the IEX fractions (see Figure 38).

[0185] A low neutral protease signal was detected in the last AUX-I fraction (17) of this typical manufacturing run. This positional nature of the neutral protease elution from the IEX column is illustrated in Figures 38 (lanes 9, 10) and 39 (lanes 8, 10, and 12). Again, this detection suggests that residual neutral protease (not eliminated in the HIC or TFF-1 steps) partially releases from the IEX column at the end of the AUX-I peak collection. The positional nature of the neutral protease contamination at the end of the AUX-I peak for that manufacturing run, and the fact that the older manufacturing process includes the last AUX-I fraction in the AUX-I pool, explain the presence of neutral protease detected in the AUX-I pool, AUX-I intermediates, and drug product of the manufacturing run seen in Figure 35. Thus, pooled fractions that do not contain any detectable levels of neutral protease as measured by casein zymography assay can yield a collagenase composition that is essentially free of neutral proteases.

[0186] The present invention may also include an auto-reject feature. Prior to the present invention, these AUX-I fractions were manually rejected during mixing of the passed AUX-I fractions to create the AUX-I pool. Auto-rejection of at least the last two AUX-I fractions prevents residual neutral proteases from entering the AUX-I intermediate since they usually co-elute in these fractions.

[0187] As described in the next paragraph, implementation of an improved IEX pooling strategy eliminates end contamination between pools of AUX-I and AUX-II fractions.

[0188] B. Refining the Pool - Collection of Fractions that Meet Minimum Purity Limits as Determined by SDS-PAGE Collagenase, AUX-I and AUX-II elute from the IEX column as discrete peaks during the AUX-II elution followed by gradient elution with AUX-I. The peaks are collected by manual separation in approximately 1 liter aliquots. These fractions are tested for purity using SDS-PAGE with densitometry. Successive AUX-I or AUX-II fractions are available for pooling to make collagenase I or collagenase II products. Historically, fractions of AUX-I or AUX-II that were at least 88.5% area pure as measured by SDS-PAGE with densitometry and had no single impurity greater than 10% area as measured by SDS-PAGE with densitometry were pooled to produce collagenase I or collagenase II products. Fractions that did not meet these limits were not processed forward and were therefore rejected from the product stream.

[0189] Improving the criteria for the flow-through not only improves the overall purity of the collagenase composition, but also helps to further remove neutral proteases from the AUX-I tail. By increasing the limit for the flow-through, neutral proteases can be removed if the AUX-I and AUX-II fractions have an AUX-I or AUX-II purity of at least 91.2 area % as measured by SDS-PAGE with densitometry, and the most abundant single impurity is 5.8 area % or less as measured by SDS-PAGE with densitometry.

[0190] As part of the investigation, a planned deviation was performed to pool the AUX-I and AUX-II fractions based on historical batch data. AUX-I fractions 12-16 and AUX-II fractions 2-5 were pooled and processed forward to create the drug substance (sometimes referred to as bulk drug substance or BDS). Release testing of the drug product resulted in a higher limit of out-of-specification (OOS) purity by SDS-PAGE and out-of-trend (OOT) purity by reversed-phase HPLC (RP-HPLC), confirming the lower purity of the IEX fractions. This planned deviation resulted in the data presented in Figure 35. Neutral protease detected in the AUX-I pool, AUX-I intermediates, and drug product from the manufacturing run was contributed by the tail AUX-I fraction (16) that was pooled due to the planned deviation. The limit of detection for the zymography electrophoresis assay is approximately 0.5 ng of neutral protease. The bands in FIG. 35 were visible on the actual casein zymography gel itself, but were slightly below the 0.5 ng limit - approximately 0.2-0.3 ng of neutral protease using a 10 microgram loading of the collagenase composition.

[0191] C. Pool refinement - Adapting AUX-II yield to minimize excess AUX-I yield The present invention may also include evaluating the theoretical gram amount of AUX-II produced by pooling the flow-through AUX-II fraction and pooling the flow-through AUX-I fraction to produce a pool of AUX-I to AUX-II at about a 1:1 mass ratio in the pooling step. For example, the amount of pooled AUX-I may be 1.3 times the grams of AUX-II. Any excess AUX-I or AUX-II may be discarded after manufacturing of the bulk drug substance (collagenase composition). This typically results in one or more tail AUX-I fractions that meet the initial in-process limit to be excluded from the AUX-I pool. This practice is a secondary process control to eliminate neutral proteases during the manufacturing process.

[0192] As an example, the peak fractions of collagenase I fractions collected from IEX elution are processed forward based on the fractions that meet the desired purity and impurity criteria, along with pooling only collagenase I peak fractions until the approximate amount of collagenase I pooled matches the approximate amount of collagenase II produced by the process. First, the AUX-II fractions are collected and the theoretical amount of total AUX-II from the process can be calculated after excluding the final tail fraction. Second, the theoretical amount of AUX-I in each fraction is determined. Then, starting from the first AUX-I fraction that contains at least 91.2 area % purity as measured by SDS-PAGE with densitometry and does not contain more than about 5.8% of a single impurity as measured by SDS-PAGE with densitometry, successive fractions for AUX-I are pooled until the amount of AUX-I pooled is about 1.3 times the total amount calculated for AUX-II. The collagenase composition produced by this method is essentially free of neutral proteases.

[0193] This control strategy results in the elimination of the further end fractions of the collagenase I peak, so that the pooled collagenase I fractions are essentially free of neutral proteases.

[0194] Example 4 - Clearance of neutral proteases from the process Amplification of inherent neutral protease activity due to nickel or zinc was demonstrated by the small-scale spiking study described above, resulting in higher impurity levels in the IEX fraction. Importantly, the neutral protease clearance study found detectable neutral protease activity in only one of the three HIC eluates included in the study (Lot 0011338), while all three lots contained detectable neutral protease activity in the TFF-1 concentrate sample. Although the trace levels present in typical HIC eluates and TFF-1 concentrates do not promote increased product fragments in the IEX fraction, the manufacturing process run still produced high levels of impurities, and the small-scale spiking study run contained increased fragments present in the HIC eluates, TFF-1 concentrates, and IEX fractions. These results suggest that nickel and zinc ions amplify neutral protease specific activity in the HIC and TFF-1 unit operations, causing increased product fragment impurities during the IEX unit operations.

[0195] A neutral protease clearance study was performed to determine whether it could be effectively removed from the product stream of the collagenase manufacturing process prior to making the drug product. This study was performed on three historical drug product lots by testing downstream process product stream retentate samples using two recently developed neutral protease activity-based test methods: (1) rejecting the AUX I elution fractions from the ion exchange step that contain detectable levels of neutral protease as measured by SDS-PAGE or zymography assay from the AUX-I pooling step, and (2) pooling the highest peak fractions of AUXI based on an approximate 1:1 mass ratio of AUXII pooled from the ion exchange step. The product streams tested in this study are listed below in Table 19. The test methods for detection of neutral protease activity in manufacturing product streams are a fluorescence-based microplate assay and an SDS-PAGE zymography method. Data from each of these test methods are discussed separately below. [Table 19]

[0196] A. Fluorescent Microplate Neutral Protease Activity Assay A semi-quantitative fluorescent activity assay was used to test the sample types listed in Table 19. The assay utilizes a commercial preparation of thermolysin (TL) for a standard curve as the only commercially available source of neutral protease that was not suitably pure for use. The limit of quantification (LOQ) for the assay is 0.0625 AU / mL (0.01866 AU / mg of TL). Only the Mustang Q filtrate contained quantifiable levels of neutral protease activity. All further downstream sample types were below the LOQ of the assay. Data for the three production lots tested are presented below in Table 20. [Table 20]

[0197] B. Zymography Assay Zymography assays for neutral protease activity resulted in positive neutral protease band detection in all Mustang Q filtrate samples, one lot of HIC eluate (lot 0011338), and all TFF-1 concentrate samples. None of the downstream sample types of TFF-1 concentrate show neutral protease activity. Data from C. histolyticum lots 0009749, 0010987, and 0011338 are presented below in Table 21. A zymography gel image of lot 0010987 is presented in Figure 40. The image is an inverted Coomassie stain representation, providing improved contrast to allow for low amplitude signal detection, e.g., neutral protease bands. [Table 21]

[0198] The neutral protease clearance study demonstrates that neutral protease is primarily cleared in the HIC chromatography step, as evidenced by the lack of detectable activity after the Mustang Q filtrate sample in the microplate assay, and the low level of activity present in one of the HIC eluate samples (Lot 0011338) as assessed by zymography assay. The zymography assay revealed low neutral protease activity in all three TFF-1 concentrate samples, confirming that the neutral protease is partially released from the HIC column during the elution step, while the majority of the enzyme remains bound to the column. The lack of neutral protease activity in the two HIC eluate samples, with the low activity detected in the respective TFF-1 concentrates, is due to the six-fold concentration that occurs before the start of diafiltration during this unit operation. Residual neutral protease present in the TFF-1 retentate is rejected from the product stream during the IEX unit operation where the neutral protease partially co-elutes with the AUX-I tail fraction (see Figure 39). IEX fraction purity process limits and procedural controls (automatic rejection of the last two AUX-I fractions, targeting equivalent masses of intermediates) exclude the tail AUX-I fraction from the AUX-I pool removing residual neutral protease from the manufacturing process prior to producing the AUX-I intermediates and drug product.

[0199] Example 5 - Effect of eliminating trace amounts of neutral protease As shown in Example 4, the smaller A at the tail end of the collagenase peak 280 Pooling the collagenase peak fractions also removed most, if not all, traces of neutral proteases, such that the collagenase I product made by pooling fractions of AUX I is essentially free of neutral proteases. Quantitative evaluation of the pooled elution fractions after the IEX purification step for each collagenase was performed to compare the effectiveness of the improved manufacturing process to the previous manufacturing process.

[0200] Trend analysis of the relative purity and impurity levels observed in the IEX fractions provides important process performance information. One such method is described in Example 1. In this example, collagenase I and collagenase II were tested from the following lots: [Table 22]

[0201] For each manufacturing run listed in Table 22, the AUX I and AUX II elution fractions were measured for product purity and impurity percentages as described in Example 1. The percentage of impurities in each fraction was measured both on a per fragment basis as well as on a total impurity basis.

[0202] The percent purity and percent of each impurity affected by the atypical ammonium sulfate lot were plotted against the amount of total protein in each AUX I or AUX II elution peak using the computer program Control Charts for Individuals. Control Charts for Individuals is a statistical tool used to distinguish between results due to routine variation (within the control chart limits) and results due to exceptional variation (above the control chart limits). To estimate process variability, the moving range of two consecutive observations is used to calculate the control limits. The green horizontal lines in the control charts represent the average results. The red horizontal lines represent the mean results.

number

[0203] The percent purity of the AUX-I and AUX-II products relative to the amount of total protein in each AUX elution peak is provided in Figure 42 and Figure 43. The decrease in percent purity levels of the AUX-I and AUX-II elution peaks seen in Figure 42 and Figure 43 is primarily due to high levels of the AUX-I 90 kDa and AUX-II 96 kDa impurities, as shown in Figure 44 and Figure 45, respectively.

[0204] The raw data for AUX I and AUX II product peak percentages for a typical manufacturing run are shown in Figures 46A and 46B, and the corresponding raw data for AUXI and AUX II major impurity percentages are shown in Figures 46B-46D. The major impurities are the 90 kDa and 96 kDa degradation products from collagenase I and collagenase II due to the presence of neutral proteases in the upstream manufacturing process.

[0205] As shown in Figures 46A-46D, the raw data for collagenase I purified using the previous processing method has an average purity of 94.5% + / - 1.0%. The raw data for collagenase I purified using the improved processing method has an improved average purity of 95.8% + / - 1.0%. The statistically significant improvement of 0.3% for the AUX I pool is further evidence that a collagenase I product essentially free of neutral proteases is produced using the improved manufacturing method. Not surprisingly, the purity of collagenase II (AUX-II) remains unchanged since neutral proteases do not elute from the ion exchange column with AUX II. Even with the improved process, the lingering presence of the 90 kDa and 96 kDa collagenase impurity peaks in the purified product is not unexpected since neutral proteases are still present in the earlier steps of purification even during the improved process.

[0206] The above-described embodiments of the method are intended to be merely illustrative, and various variations and modifications will be apparent to those skilled in the art. All such variations and modifications are intended to be within the scope of the present invention as defined in any of the appended claims.

Claims

[Claim 1] A drug product comprising isolated and purified collagenase, the drug product being essentially free of neutral proteases.

Images