Animal product free system and process for purifying a botulinum toxin

Abstract

Chromatographic processes and systems for purifying a botulinum toxin from an APF fermentation medium.

Claims

1. A process for purifying a botulinum toxin comprising the step of culturing Clostridium botulinum bacteria in an animal protein free (APF) culture medium and sequentially contacting a plurality of chromatography columns with an aqueous medium containing a botulinum toxin, thereby obtaining a purified botulinum toxin; wherein the plurality of chromatography columns comprise at least a first chromatography column and a second chromatography column; wherein the first chromatography column is an ion exchange chromatography column, and the second chromatography column is a hydrophobic interaction chromatography column; wherein the botulinum toxin is a botulinum toxin complex type A comprising a heavy chain and a light chain; and wherein the process is an animal protein free (APF) process.

2. The process of claim 1, wherein the plurality of chromatography columns further comprise a second ion exchange chromatography column.

3. The process of claim 1, wherein the plurality of chromatography columns further comprise a mixed mode chromatography column.

4. The process of claim 1, further comprising a filtration step comprising contacting the aqueous medium containing the botulinum toxin with a filter.

Description

DRAWINGS

(1) Aspects of the invention are explained or illustrated by the following drawings.

(2) FIG. 1 entitled N-Source (i.e. HySoy plus YE) % vs. Potency and pH” is a graph showing botulinum toxin activity as determined: (1) on the left side Y axis as mouse lethal dose 50 (MLD 50), and; (2) on the left side Y axis as SNAP 25 activity, of various APF media at the elapsed fermentation times shown at the top of the bars, for APF medium pH as shown on the right side Y axis the pH, for APF media with the wt % amount of hydrolyzed soy concentrate and yeast extract concentrate as shown by the X axis. All FIG. 1 media also contained 1% by wt glucose.

(3) FIG. 2 is a summary flow chart comparing a non-APF process for obtaining a botulinum toxin (the top half of FIG. 2) with an APF process, within the scope of the present invention, for obtaining a botulinum toxin (the bottom half of FIG. 2), through the cell bank creation, culture and fermentation steps. FIG. 2 omits the harvest and purification steps.

(4) FIG. 3 is a chromatograph obtained from hydrophobic interaction chromatography of an APF clarified culture (a 3.1.1 culture) on a Butyl Sepharose Fast Flow column. The X axis in FIG. 3 represents the volume in ml of liquid (effluent) which has passed through the column. The Y axis represents absorbance at 280 nm in mAU.

(5) FIG. 4 is a chromatograph obtained from ionic exchange chromatography of the eluent from the FIG. 3 Butyl column on an SP Sepharose high performance column. The axes in FIG. 4 are the same as they are for FIG. 3.

(6) FIG. 5A is an image of reduced SDS-PAGE of various fractions obtained from operation of the Butyl column of FIG. 3. The left hand side of FIG. 5A is marked vertically with descending molecular weights in thousands of Daltons (kDa). The numbers 1 to 8 along the bottom border of FIG. 5A represents the lanes in which fractions were loaded.

(7) FIG. 5B is an image of reduced SDS-PAGE of various fractions obtained from operation of the SP column of FIG. 4. The left and bottom sides of FIG. 5B are marked as they are in FIG. 5A.

(8) FIG. 6 is an image of reduced SDS-PAGE of various fractions obtained in post column steps (see FIG. 7), namely fractions from the UF/DF step, the sterile filtration step, and from the ammonium sulfate precipitation step. The left and bottom sides of FIG. 6 are marked as they are in FIG. 5A.

(9) FIG. 7 is a flow chart of a APF chromatographic botulinum toxin purification process within the scope of the present invention.

(10) FIG. 8 is a graph comparing the effect of a soy protein concentration on a botulinum toxin type A complex production in an APF fermentation process, where the fermentation medium contained 1 wt % glucose and 1 wt % of a yeast extract. In FIG. 8 the X axis represents the weight percent concentration in the fermentation medium of a particular hydrolyzed soy protein (HySoy), the left side Y axis represents potency of the final purified botulinum toxin complex and the right side Y axis represents the percent of cell lysis completed, as determined by the equation:

(11) $\mathrm{Cell}\mathrm{Lysis}\left(\%\right)=\frac{O{D}_{600\max }-O{D}_{600\mathrm{endpoint}}}{O{D}_{600\max }}\times 100$
where OD.sub.600 max corresponds to the optical density measured at 600 nm at the time of maximum growth, and OD.sub.600 endo int is at the time of fermentation harvest.

DESCRIPTION

(12) The present invention is based upon the discovery that a Clostridial toxin can be purified by use of an animal product fee (APF) system and process. The present invention encompasses a animal product free system and process for purifying a Clostridium botulinum neurotoxin. The Clostridium botulinum neurotoxin can be a botulinum toxin type A complex, such as a 300 kD, 500 kD or 900 kD (approximate molecular weights) complex or mixtures thereof. The Clostridium botulinum neurotoxin can be any one of the serotypes A, B, C, D, E, F or G or mixtures thereof. Additionally, the system and process can be practiced in conjunction with a recombinant, hybrid, chimeric or modified botulinum toxin (light chain, heavy chain, or both chains together).

(13) Significantly, the system and process disclosed herein is scalable, meaning that it can be used to purify the quantities of botulinum toxin obtained from an industrial or commercial process, as use for pharmaceutical production. Further, the system and process is also CGMP (certified good manufacturing practices) compliant, as required by the U.S. CFR (United States code of federal regulations), meaning that it can comply with regulatory requirements.

(14) Through experimentation, there were developed APF systems and processes to purify a Clostridial toxin, such as a Clostridium botulinum type A (Hall strain) neurotoxin complex. The Clostridial toxin is purified from the fermentation medium resulting from either a Schantz (non-APF) fermentation process or from an APF fermentation process. Schantz processes use animal derived products. Significantly, while an APF fermentation process can reduce or eliminate animal derived products (such as casein and meat broth) as nutrients from the media used to culture and ferment Clostridial bacteria, APF fermentation processes are typically followed by one or more purification steps which make use of animal derived products, such as the enzymes DNase and RNase. Purification of the fermentation medium is required to obtain bulk Clostridial toxin. Bulk Clostridial toxin (pure toxin or toxin complex) can be used for compounding a Clostridial toxin pharmaceutical composition.

(15) Preferably, the present invention is practiced in conjunction with an APF fermentation process. Practicing the present invention in conjunction with an APF fermentation process provides a combined APF fermentation process and an APF purification process. Additionally, systems and method of the present invention are optimized for operation upon an APF fermentation medium, as opposed to a casein or other animal protein based fermentation medium. Practicing the presently invention upon a non-APF fermentation can result in a lower yield and/or a lower potency of the purified botulinum toxin obtained.

(16) Thus, although both the Schantz and APF botulinum toxin purification processes use animal derived products such as benzamidine to stabilize the botulinum toxin and DNase and RNase to remove nucleic acids present with the botulinum toxin in the fermentation medium (see e.g. Examples 6 and 7), our invention permits a botulinum toxin can be purified without using such animal derived products.

(17) The present invention encompasses systems and processes for purifying a Clostridial toxin, such as a botulinum toxin complex. Typically a particular system within the scope of the present invention is operated in conjunction with a particular process within the scope of the present invention. A system within the scope of the present invention can comprise a plurality (preferably as a consecutive series) of chromatography steps. A process within the scope of the present invention can comprise passing a Clostridial toxin fermentation medium through the plurality of chromatography columns to thereby obtain a highly purified and highly potent Clostridial toxin. Such a purified Clostridial toxin is suitable for compounding a Clostridial toxin pharmaceutical composition. Important parameters of systems and processes within the scope of the present invention include the particular columns, buffers and operating (column running) conditions used.

(18) A first broad step in a particular embodiment of the invention can be to load a fermentation medium clarified culture onto a hydrophobic interaction column (such as a Butyl Sepharose Fast Flow [“FF”] column). This first column captures the Clostridial toxin (such as a botulinum toxin complex) and allows impurities to flow through the column. It was found that a hydrophobic interaction column provided an efficient capture of a botulinum toxin complex (a large protein with a particular tertiary and quaternary structure) from fermentation medium with retention of the biological activity of the botulinum toxin complex, while also separating (flow through) of many impurities present with the botulinum toxin in the fermentation medium. A suitable buffer is used to elute the captured (bound) Clostridial toxin from the hydrophobic interaction column.

(19) In a second broad step in a particular embodiment of the present invention, the eluent from the first column is loaded onto a second column to further purify the Clostridial toxin. It was found that preferably, if second column [provides a different mechanism for separation of Clostridial toxin from impurities, then a second column chromatography step can provide a further efficient purification step. Thus, preferably, the second chromatography step entails use of a different column, such as a SP Sepharose high performance [“HP”] column.

(20) In post chromatography (column) steps eluent from the second column can then be further processed to obtain highly purified bulk botulinum toxin complex. These additional processing steps can include buffer exchange by ultrafiltration and diafiltration, sterile filtration and preparation of an ammonium sulphate suspension of the purified botulinum toxin complex.

(21) Our invention encompasses a scalable and cGMP compliant system and process for purifying a botulinum toxin, which can result in obtaining a bulk botulinum toxin with the characteristics set forth in Table 1.

(22) TABLE-US-00002 TABLE 1 Purified Botulinum Neurotoxin Characteristics Appearance White to off-white suspension Concentration 2.0-3.6 mg/ml Nucleic Acids (A260/A278) Not more than 0.6 Specific Potency (MLD50 unit/mg) 2.4-5.9 × 10.sup.7 Immunological Identity Pass SDS-PAGE Conformed to standard SEC-HPLC 900 kDa toxin complex >95% total peak

(23) A commercially available botulinum toxin containing pharmaceutical composition is sold under the trademark BOTOX® (available from Allergan, Inc., of Irvine, Calif.). BOTOX® has the characteristics set forth in Table 1 above. BOTOX® consists of a purified botulinum toxin type A complex, human serum albumin, and sodium chloride packaged in sterile, vacuum-dried form. The botulinum toxin type A is made from a culture of the Hall strain of Clostridium botulinum grown in a medium containing N-Z amine casein and yeast extract (i.e. non-APF process). The botulinum toxin type A complex is purified from the culture solution by a series of precipitation (including acid precipitation) steps to a crystalline complex consisting of the active high molecular weight toxin protein and an associated hemagglutinin protein. The crystalline complex is re-dissolved in a solution containing saline and albumin and sterile filtered (0.2 microns) prior to vacuum-drying. BOTOX® can be reconstituted with sterile, non-preserved saline prior to intramuscular injection. Each vial of BOTOX® contains about 100 units (U) of Clostridium botulinum toxin type A complex, 0.5 milligrams of human serum albumin and 0.9 milligrams of sodium chloride in a sterile, vacuum-dried form without a preservative.

(24) To reconstitute vacuum-dried BOTOX® sterile normal saline without a preservative (0.9% Sodium Chloride injection) is used by drawing up the proper amount of diluent in the appropriate size syringe. Since BOTOX® is denatured by bubbling or similar violent agitation, the diluent is gently injected into the vial. It has been reported that BOTOX® has been administered thirty or more days after reconstitution with little loss of potency. During this time period, reconstituted BOTOX® is stored in a refrigerator (2° to 8° C.). Reconstituted BOTOX® is clear, colorless and free of particulate matter. The vacuum-dried product is stored in a freezer at or below −5° C.

(25) The present invention is based upon the discovery of media and processes which are free or substantially free of an animal product or an animal byproduct useful for culture and fermentation of an organism (such as a Clostridium botulinum bacterium) capable of producing biologically active botulinum toxin. The botulinum toxin obtained can be used for making botulinum toxin active ingredient pharmaceutical compositions. Thus, growth media are disclosed herein which have significantly reduced levels of meat or dairy by-products and preferred media embodiments are substantially free of such animal products.

(26) The present invention encompasses the surprising finding that animal-based products are not required in media for growth of Clostridium botulinum, and particularly that vegetable-based products can replace animal-based products typically employed in such media for the growth of Clostridium botulinum.

(27) Media that are in current use for growth and fermentation of bacteria usually comprise one or more animal derived ingredients, such as cooked meat. In accordance with the present invention, preferred media for growth of Clostridium botulinum contain animal derived ingredients which comprise no more than about five to about ten percent of the total weight of the media. More preferably, media within the scope of the invention comprise no more than about one to less than about five percent of the total weight of the media of animal-derived products. Most preferably, all media and cultures used for the growth of Clostridium botulinum for the production of botulinum toxin are completely free of animal derived products. These media include but are not limited to media for small and large scale fermentation of Clostridium botulinum, media for growth of cultures of Clostridium botulinum used to inoculate the seed (first) media and fermentation (second) media, as well as and media used for long-term storage of cultures of Clostridium botulinum (e.g. stock cultures).

(28) In certain preferred embodiments of the invention, the media for the growth of Clostridium botulinum and production of botulinum toxin can comprise soy based products to replace animal derived products. Alternately, instead of a soy based product there can be used debittered seed of Lupinus campestris. It is known the protein content of L. campestris seed is very similar to that of soybean. Preferably, these media include soybean or of L. campestris derived products that are hydrolyzed and that are soluble in water. However, insoluble soy or of L. campestris products can also be used in the present invention to replace animal products. Common animal derived products which can be substituted by soy or of L. campestris products include beef heart infusion (BHI), animal derived peptone products, such as Bacto-peptone, hydrolyzed caseins, and dairy by-products such as animal milk.

(29) Preferably media containing soy-based or of L. campestris based products for the growth of Clostridium botulinum are similar to commonly used growth media containing animal derived products except that substantially all animal-derived products are replaced with vegetable-derived products. For example, soy based fermentation media can comprise a soy based product, a source of carbon such as glucose, salts such as NaCl and KCl, phosphate-containing ingredients such as Na.sub.2HPO.sub.4, KH.sub.2PO.sub.4, divalent cations such as iron and magnesium, iron powder, and amino acids such as L-cysteine and L-tyrosine. Media used to grow cultures of Clostridium botulinum for inoculation (i.e. the seed or first medium) of the fermentation (second) media preferably contain at least a soy based product, a source of salt such as NaCl, and a carbon source such as glucose.

(30) The present invention provides a method for the growth of Clostridium botulinum that maximizes the production of a botulinum toxin using media that are substantially free of animal-derived products. Growth of Clostridium botulinum for production of botulinum toxin can take place by fermentation in media containing soy by-products that replace ingredients derived from animal by-products. The inoculant for the fermentation medium can be derived from a smaller scaled growth medium (a seed medium). Depending on the size and volume of the fermentation step, the number of successive growths in seed media to increase the biomass of the culture can vary. To grow a suitable amount of Clostridium botulinum for inoculating the fermentation medium, one step or multiple steps involving growth in a seed medium can be performed. For a method of growing Clostridium botulinum that is free of animal derived products, it is preferable that growth of Clostridium botulinum originates from a culture stored in non animal derived media. The stored culture, preferably lyophilized, is produced by growth in media containing proteins derived from soy and lacking animal by-products. Growth of Clostridium botulinum in a fermentation medium can take place by inoculation directly from a stored, lyophilized culture.

(31) In a preferred embodiment of the present invention, growth of Clostridium botulinum proceeds in two phases-seed growth and fermentation. Both of these phases are carried out in anaerobic environments. The seed growth phase is generally used to “scale-up” the quantity of the microorganism from a stored culture. The purpose of the seed growth phase) is to increase the quantity of the microorganism available for fermentation. In addition, the seed growth phase allows relatively dormant microbes in stored cultures to rejuvenate and grow into actively growing cultures. Furthermore, the volume and quantity of viable microorganisms used to inoculate the fermentation culture can be controlled more accurately from an actively growing culture than from a stored culture. Thus, growth of a seed culture for inoculation of the fermentation medium is preferred. In addition, any number of consecutive steps involving growth in seed media to scale-up the quantity of Clostridium botulinum for inoculation of the fermentation medium can be used. It is noted that growth of Clostridium botulinum in the fermentation phase can proceed directly from the stored culture by direct inoculation.

(32) In the fermentation phase, a portion of a seed medium or all of a seed medium containing Clostridium botulinum from the seed growth is used to inoculate a fermentation medium. Preferably, approximately 2-4% of a seed medium having Clostridium botulinum from the seed growth phase is used to inoculate the fermentation medium. Fermentation is used to produce the maximum amount of microbe in a large-scale anaerobic environment (Ljungdahl et al., Manual of industrial microbiology and biotechnology (1986), edited by Demain et al, American Society for Microbiology, Washington, D.C. page. 84).

(33) A botulinum toxin can be isolated and purified using methods of protein purification well known to those of ordinary skill in the protein purification art. See e.g. Coligan et al. Current Protocols in Protein Science, Wiley & Sons; Ozutsumi et al. Appl. Environ. Microbiol. 49; 939-943:1985.

(34) For production of botulinum toxin, cultures of Clostridium botulinum can be grown in a seed medium for inoculation of the fermentation medium. The number of successive steps involving growth in a seed medium can vary depending on the scale of the production of botulinum toxin in the fermentation phase. However, as previously discussed, growth in the fermentation phase may proceed directly from inoculation from a stored culture. Animal-based seed media generally are comprised of BHI, bacto-peptone, NaCl, and glucose for growth of Clostridium botulinum. As previously discussed, alternative seed media may be prepared in accordance with the present invention in which animal-based components are substituted with non-animal-based components. For example but without limitation, soy-based products can substitute for BHI and bacto-peptone in the seed medium for growth of Clostridium botulinum and production of botulinum toxin. Preferably, the soy-based product is soluble in water and comprises hydrolyzed soy, although cultures of Clostridium botulinum can grow in media containing insoluble soy. However, levels of growth and subsequent toxin production are greater in media derived from soluble soy products.

(35) Any source of soy-based products may be used in accordance with the present invention. Preferably, the soy is hydrolyzed soy and the hydrolyzation has been carried out using non-animal enzymes. Sources of hydrolyzed soy are available from a variety of commercial vendors. These include but are not limited to Hy-Soy (Quest International), Soy peptone (Gibco) Bac-soytone (Difco), AMISOY (Quest), NZ soy (Quest), NZ soy BL4, NZ soy BL7, SE50M (DMV International Nutritionals, Fraser, N.Y.), and SE50MK (DMV). Most preferably, the source of hydrolyzed soy is Hy-Soy or SE50MK. Other potential sources of hydrolyzed soy are known.

(36) Concentrations of Hy-Soy in the seed medium in accordance with the present invention range between 25-200 g/L. Preferably, the concentration of Hy-Soy in the seed medium ranges between 50-150 g/L. Most preferably the concentration of Hy-Soy in the seed medium is approximately 100 g/L. In addition, the concentration of NaCl ranges between 0.1-2.0 g/L. Preferably the concentration of NaCl ranges between 0.2-1.0 g/L. Most preferably, the concentration of NaCl in the seed medium is approximately 0.5 g/L. The concentration of glucose ranges between 0.1 g/L and 5.0 g/L. Preferably, the concentration of glucose ranges between 0.5-2.0 g/L. Most preferably, the concentration of glucose in the seed medium is approximately 1.0 g/L. It is also preferred but not necessary for the present invention that the glucose is sterilized by autoclaving together with the other components of the seed medium. The pH level of the seed medium prior to growth can be 7.5-8.5. For example, the pH of the seed medium prior to growth of Clostridium botulinum can be approximately 8.1.

(37) Growth of Clostridium botulinum in the seed medium can proceed in one or more stages. Preferably, growth in the seed medium proceeds in two stages. In stage one, a culture of Clostridium botulinum is suspended in a quantity of seed medium and incubated at 34±1° C. for 24-48 hours in an anaerobic environment. Preferably, growth in stage one proceeds for approximately 48 hours. In stage two, a portion or all of the stage one medium containing Clostridium botulinum is used to inoculate a stage two seed medium for further growth. After inoculation, the stage two medium is incubated at 34±1° C. for approximately 1-4 days also in an anaerobic environment. Preferably, growth in the stage two seed medium proceeds for approximately 3 days. It is also preferable that growth in seed media in any stage does not result in cell lysis before inoculation of fermentation media with the final growth in seed medium.

(38) Standard fermentation media containing animal by-products for the growth of Clostridium botulinum can be based on a recipe of Mueller and Miller (MM; J. Bacteriol. 67:271, 1954). The ingredients in MM media containing animal by-products include BHI and NZ-CaseTT. NZ-CaseTT is a commercially available source of peptides and amino acids which are derived from the enzymatic digestion of caseins, a group of proteins found in animal milk. The present invention demonstrates that non-animal based products may be substituted for BHI and NZ-CaseTT in fermentation media. For example but without limitation, soy-based products can replace the animal-based components of MM media used for fermentation of Clostridium botulinum. Preferably, the soy-based products are water-soluble and derived from hydrolyzed soy, although as previously discussed, insoluble soy products can also be used to practice the present invention.

(39) Any source of soy-based products may be used in accordance with the present invention. Preferably, the hydrolyzed soy is obtained from Quest International (Sheffield) under the tradename, Hy-Soy or from DMV International Nutritionals (Fraser, N.Y.) under the tradename, SE50MK. Soluble soy products can be also obtained from a variety of sources including but not limited to Soy peptone (Gibco) Bac-soytone (Difco), AMISOY (Quest), NZ soy (Quest), NZ soy BL4, NZ soy BL7, and SE50MK (DMV International Nutritionals, Fraser, N.Y.).

(40) In another preferred embodiment of the present invention, the medium used for fermentation of Clostridium botulinum is free of animal by-products and comprises hydrolyzed soy, glucose, NaCl, Na.sub.2HPO.sub.4, MgSO.sub.47 H.sub.2O, KH.sub.2PO.sub.4, L-cysteine,L-tyrosine, and powdered iron. As disclosed for the seed medium, hydrolyzed soy can replace animal by-products in fermentation medium. These animal by-products include BHI and NZ-CaseTT (enzymatically digested casein).

(41) The concentration of Hy-Soy in the fermentation medium for production of botulinum toxin preferably ranges between approximately 10-100 g/L. Preferably, the concentration of Hy-Soy ranges between approximately 20-60 g/L. Most preferably, the concentration of Hy-Soy in the fermentation medium is approximately 35 g/L. For maximal production of botulinum toxin, particularly preferred concentrations of components in the fermentation medium are approximately 7.5 g/L, glucose; 5.0 g/L NaCl; 0.5 g/L Na.sub.2HPO.sub.4; 175 mg/L KH.sub.2PO.sub.4; 50 mg/L MgSO.sub.47H.sub.2O; 125 mg/L L-cysteine; and 125 mg/L L-tyrosine. The amount of powdered iron used can range from 50 mg/L to 2000 mg/L. Preferably, the amount of powdered iron ranges between approximately 100 mg/L and 1000 mg/L. Most preferably, the amount of powdered iron used in fermentation media ranges between approximately 200 mg/L and 600 mg/L.

(42) For optimal levels of toxin production, the initial pH (before autoclaving) of the soy-based fermentation media ranges preferably between approximately 5.0 to 7.1. We found that pH control improves botulinum toxin recovery. Preferably the initial pH of the fermentation medium is about pH 7. As explained in Example 7, we have found that a high yield of stable botulinum toxin can be obtained if the pH is thereafter reduced to and maintained between pH 5-5.5. As described for the seed medium, the components of the fermentation medium, including glucose and iron, are preferably autoclaved together for sterilization.

(43) Preferably, a portion of the second stage seed medium used for growth of Clostridium botulinum is used to inoculate the fermentation medium. Fermentation occurs in an anaerobic chamber at approximately 34.±1° C. for approximately 7 to 9 days. Bacterial growth can be monitored by measuring the optical density (O.D.) of the medium. Fermentation preferably is stopped after cell lysis has proceeded for at least 48 hours as determined by growth measurement (optical density). As cells lyse, the O.D. of the medium decreases.

(44) In a preferred embodiment of the present invention, cultures of Clostridium botulinum used for long-term storage of Clostridium botulinum and inoculation of the seed medium are grown and lyophilized in soy-milk prior to storage at 4° C. Cultures of Clostridium botulinum in animal milk lyophilized for storage can also be used for the production of botulinum Toxin. However, to maintain media that are substantially free of animal by-products throughout the production of botulinum toxin, it is preferred that the initial culture of Clostridium botulinum be preserved in soy milk and not animal milk.

EXAMPLES

(45) The following examples set forth specific methods encompassed by the present invention and are not intended to limit the scope of the invention. Unless explained otherwise in these Examples “toxin” or “botulinum toxin” means a botulinum toxin type A complex with a molecular weight of about 900 kDa. Our invention is not limited to systems and method for purifying a botulinum toxin type A complex with a molecular weight of about 900 kDa, having ready applicability to the purification of 150 kDa, 300 kDa, 500 kDa and well as other molecular weight toxins, complexes and botulinum toxin serotypes.

Example 1

Preparation of an Animal Product Free Seed Medium for Clostridium Botulinum

(46) A control seed medium can be prepared using the following ingredients for each one 1 liter of medium: NaCl (5 g), Bacto-peptone (10 g), glucose (10 g), BHI (to 1 liter), pH 8.1 (adjusted with 5 N NaOH).

(47) A test (animal product free) seed medium can be prepared using the following ingredients for each one 1 liter of medium: NaCl (5 g), Soy-peptone (10 g), glucose (10 g), Hy-Soy (35 g/liter, to make up 1 liter of media fluid), pH 8.1 (adjusted with 5 N NaOH).

Example 2

Culturing Clostridium Botulinum in an Animal Product Free Seed Medium

(48) A lyophilized culture of the Clostridium botulinum can be suspended in 1 ml of each of the control and test seed medium of Example 1, divided (each seed media) into two tubes of which each can contain 10 ml of the respective seed media, and then incubated at 34° C. for about 24-48 hours. One ml of culture can be then used to inoculate a 125 ml DeLong Bellco Culture Flask containing 40 ml of (the respective) seed media. The inoculated culture can be incubated at 33° C.±1° C. for 24 hours in a Coy Anaerobic Chamber (Coy Laboratory Products Inc., Grass Lake, Mich.).

Example 3

Preparation of an Animal Product Free Fermentation Media for Clostridium Botulinum

(49) A basal fermentation medium can be prepared using the following ingredients for each two liters of medium: glucose (15 g), NaCl (10 g), NaH.sub.2PO.sub.4 (1 g), KH.sub.2PO.sub.4 (0.350 g), MgSO.sub.47H.sub.2O (0.1 g), cysteine-HC (0.250 g), tyrosine-HCl (0.250 g), powdered iron (1 g), ZnCl.sub.2 (0.250 g), and MnCl.sub.2 (0.4 g).

(50) A control fermentation medium can be prepared using the following ingredients for each two liters of medium prepared: BHI (500 ml; this corresponds to about 45.5 grams of dry weight beef heart infusion), NZ-CaseTT (30 g), and basal medium (to 2 liters), pH 6.8.

(51) The basal fermentation medium can be prepared first and adjusted to pH 6.8. The beef heart infusion (BHI) BHI can then be prepared and it's pH adjusted to 0.8 with 5 N NaOH. The BHI can then be added to the basal medium. Next the NZ-CaseTT can be prepared. The NZ-CaseTT is then added the basal medium to which the beef heart infusion has already been added, and dissolved by addition of HCl. The pH can then be adjusted to 6.8 with 5 N NaOH. This medium can then be separated into 8 ml portions into each of sixteen 100 mm test tubes, following by autoclaving for 25 minutes at 120° C.

(52) A test fermentation medium (animal product free) can be prepared by substituting a test nitrogen source for the BHI present in the control fermentation medium. Suitable test fermentation medium nitrogen sources include: Hy-Soy (Quest), AMI-Soy (Quest), NZ-Soy (Quest), NZ-Soy BL4 (Quest), NZ-Soy BL7 (Quest), Sheftone D (Sheffield), SE50M (DMV), SE50 (DMV), SE %) MK (DMV), Soy Peptone (Gibco), Bacto-Soytone (Difco), Nutrisoy 2207 (ADM), Bakes Nutrisoy (ADM) Nutrisoy flour, Soybean meal, Bacto-Yeast Extract (Difco) Yeast Extract (Gibco), Hy-Yest 412 (Quest), Hy-Yest 441 (Quest), Hy-Yest 444 (Quest), Hy-Yest (455 (Quest) Bacto-Malt Extract (Difco), Corn Steep, and Proflo (Traders).

(53) The test fermentation medium can be prepared as set forth above for a control fermentation medium except that BHI is excluded and the relevant nitrogen source can be first adjusted to pH 6.8 with 3 N HCl or with 5 N NaOH. The media can be allocated to in 8 ml portions to sixteen 100 mm test tubes, followed by autoclaving for 20-30 minutes at 120° C.

Example 4

Growth of Clostridium Botulinum in an Animal Product Free Fermentation Medium

(54) A 40 μl portion of the test seed medium culture (animal product free) can be used to inoculate each 8 ml control or test fermentation medium aliquot in an 8 ml 16×100 mm test tube. The cultures can then be incubated at 33±1° C. for 24 hours. Tubes can then be incubated in an anaerobic chamber to allow for growth of the bacterium. Each medium assay can be performed in triplicate (i.e. can involve three independent inoculations of the same medium), and can also include a non-inoculated control, which can be used as the blank for the spectrophotometer). Growth (as determined by optical density, OD) can be measured every 24 hours with a Turner Spectrophotometer (Model 330) at 660 nm. Cultivation should be stopped after cell lysis has lasted for about 48 hours and botulinum toxin production can then be measured.

(55) Additional experiments can be carried out with a Hy-Soy fermentation medium containing the following ingredients for each 500 ml of the medium: Hy-Soy (17.5 g), glucose (3.75 g); NaCl (2.5 g); Na.sub.2HPO.sub.4 (0.25 g), MgSO.sub.47H.sub.2O (0.025 g), KH.sub.2PO.sub.4 (0.0875 g), L-cysteine (0.0625 g), L-tyrosine (0.0625 g), powdered iron (0.25 g), pH 6.8.

Example 5

Determination of Botulinum Toxin Production by Clostridium Botulinum Grown in an Animal Product Free Fermentation Medium

(56) The cultured cells of Example 4 can be centrifuged, and the pH of the supernatant then determined. The levels of botulinum toxin in a given sample can be measured by adding a standard antitoxin and measuring the elapsed time before flocculation. Both Kf (the time required for flocculation to occur, in minutes) and Lf (the limit of flocculation; equivalent to 1 international unit of standard antitoxin, as established by flocculation) can be determined. 4 ml of fermentation broth can be taken from each fermentation tube for a given culture, and can be combined together so that 12 ml total can be mixed in a 15 ml centrifuge tube. The tubes can be centrifuged at 5000 rpm (3400 g) for 30 min at 4° C. 1 ml aliquots of supernatant can be added to tubes containing 0.1-0.6 ml of standard botulinum toxin antiserum, and the tubes can be carefully shaken to mix their contents. The tubes can then be placed in a water bath at 45±1° C. and the initial time can be recorded. The tubes can be checked frequently, and the time at which flocculation began can be recorded as Kf. The concentration of toxin in the tube in which flocculation can be first initiated can be designated LfFF. The concentration of toxin in the tube in which flocculation can be initiated second can be designated LfF.

(57) Parallel fermentation, growth and toxin production assays can be carried out for both of: (a) the control seed medium (used to inoculate the control fermentation medium) and the control fermentation medium, and; (2) the (animal product free) test seed medium (used to inoculate the test fermentation medium) and the (animal product free) test fermentation medium. Significantly, it can be determined that the fermentation of Clostridium botulinum in media free of animal products and inoculated from cultures also free of animal products (with soy-base products replacing the animal products) can result in an Lf.sub.toxin of approximately 50 or more. Minimally, Lf.sub.toxin equals approximate 10. Preferably the Lf.sub.toxin at least 20. Most preferably the Lf.sub.toxin greater than 50.

(58) Additionally, it can be determined that various soy products support Clostridium botulinum growth in fermentation media lacking BHI. Thus soluble soy preparations can replace BHI for growth of Clostridium botulinum. The best concentration can be 12.5 or 25 g/L. Hy-Soy (Sheffield) can give the highest growth. Insoluble soy preparations can be less effective.

(59) Furthermore, results can be obtained to show that Quest Hy-Soy, DMV SE50MK, and Quest NZ-Soy can be effective soy products in terms of their ability to replace BHI for Clostridium botulinum growth. The results can reveal that the soy products (such as Quest Hy-Soy, DMV SE50MK, and Quest NZ-Soy) that may be optimal for growth can also be effective at replacing BHI for toxin production. The best soy product for toxin production can be Quest Hy-Soy at 22.75 g/I. Higher concentrations of this product may produce better growth but not improve toxin production. Similar results can, it is proposed, be obtained with SE50MK, for which a higher concentration may generate increased growth, but not increase toxin production. NZ-Soy, on the other hand, may give higher growth and higher toxin production at its higher concentration.

(60) Finally, it can be determined that soy products can effectively replace BHI as well as the NZ-CaseTT. Removal of NZ-CaseTT from soy-based media can reduce growth of about 2-4 fold. The best soy product for growth both in the presence and the absence of NZ-CaseTT can be SE50MK. HY-Soy can replace both BHI and NZ-CaseTT for toxin production. However, a longer fermentation cycle of 1 or 2 days may be necessary. HY-Soy could replace both BHI and NZ-CaseTT in media for toxin production. However, it can be determined that yeast extracts can be inhibitory to toxin production.

(61) It can be determined that HY-Soy at 22.75 g/l may completely replace both BHI and HY-CaseTT for toxin production. Unlike the effect on growth where 56.88 g/l HY-Soy can be best, 34.13 g/l HY-Soy can be best for the toxin production phase.

(62) Thus, it has surprisingly been determined if Hy-Soy or [Hy-Soy+Hy-Yest] can replace BHI and Bacto-peptone in media for seed growth of Clostridium botulinum. In addition, experiments can be designed to determine the optimum concentrations of components in seed media to produce the maximum levels of botulinum toxin production by the Clostridium botulinum. Toxin production by Clostridium botulinum grown in seed medium and fermentation medium that is free of BHI and NZ-CaseTT can reach or exceed levels attained in media containing BHI and NZ-CaseTT.

(63) It can be determined that the optimum concentrations of Hy-Soy or [Hy-Soy+Hy-Yest] for growth in the seed medium. Experiments can confirm that Hy-Soy can replace BHI and Bacto-peptone as the nitrogen source in seed medium for growth of Clostridium botulinum and for production of botulinum toxin in the subsequent fermentation phase. Also, Hy-Soy as nitrogen source in the seed medium, as compared to Hy-Soy plus Hy-Yest, can produce higher levels of botulinum toxin in the subsequent fermentation step. The concentrations of Hy-Soy in seed medium that produce the best levels of toxin range from approximately 62.5 g/L to 100 g/L.

(64) Additional experiments can be designed to determine the optimum concentrations of Hy-Soy in the seed medium for the maximum production of botulinum toxin by Clostridium botulinum by fermentation. Thus, 30 g, 50 g, 75 g and 100 g of Hy-Soy in the seed medium can all resulted in production of botulinum toxin by fermentation of Clostridium botulinum and this is comparable or exceeds levels of botulinum toxin made in seed medium containing BHI and Bacto-peptone as a nitrogen source.

(65) It can be found that a concentration of 100 g/L Hy-Soy in the seed medium resulted in the highest levels of toxin production in the subsequent fermentation step. In addition, the data indicate that seed step-1 of Hy-Soy seed medium produced greater growth after 48 hours than after 24 hours.

Example 6

Non-APF Process for Obtaining a Botulinum Toxin

(66) A Clostridial toxin was obtained by fermentation of a Clostridium botulinum bacterium. Thus, a modified Schantz (non-APF) process was carried out to obtain highly potent and highly purified Clostridium botulinum toxin (i.e. bulk toxin) as follows. A modified Schantz (non-APF) process can provide a high yield of botulinum toxin. Both Schantz and modified Schantz processes use casein in all the fermentation media.

(67) Stock Culture Preparation

(68) Various Clostridial bacteria are available from the American Type Culture Collection (ATCC), Manassas, Va. Alternately, a Clostridium botulinum cell bank vial can be prepared by isolating Clostridium botulinum from various sources, including soil or by deep sampling (at anaerobic or at quasi-anaerobic locations) of putrefying animal carcasses. Commonly, Clostridium botulinum can be obtained from a sample of a physiological fluid (i.e. a wound swap from a patient with wound botulism) of a patient diagnosed with botulism. The top half of FIG. 1 summarizes the non-APF process used for preparation of a cell bank vial, and for the culture and fermentation of a botulinum toxin.

(69) The Clostridium botulinum obtained from a natural or patient source is cultured on blood agar plates, followed by inoculation of high growth colonies into a cell bank vial medium. The cell bank vial medium used for Clostridium botulinum was a cooked meat medium which contains chopped fresh beef. Actively growing cultures were mixed with glycerol to prepare a cell bank vial (i.e. a stock culture) of the Clostridium botulinum bacterium which was frozen for later use.

(70) Seed Cultivations

(71) A Clostridium botulinum cell bank vial was thawed at room temperature, followed by four cultivation steps. (1) To select colonies with a suitable morphology, aliquots from the thawed cell bank vial were cultivated by streaking the bacterium on pre-reduced Columbia blood agar plates and anaerobically incubating for 30-48 hours at 34° C.±1°. (2) Selected colonies were then inoculated into test tubes containing a casein growth medium for 6-12 hours at 34° C. The contents of the tube with the most rapid growth and highest density (growth selection step) were then further cultivated through two step-up anaerobic incubations: (3) a first 12-30 hour incubation at 34° C. in a one liter seed cultivation bottle, followed by (4) a second cultivation in a 25 liter seed fermenter containing a casein growth medium for 6-16 hours at 35° C. These two step-up cultivations were carried out in a nutritive media containing 2% casein hydrolysate (a casein [milk protein] digest), 1% yeast extract and 1% glucose (dextrose) in water at pH 7.3.

(72) Fermentation

(73) The step-up cultivations were followed by a further incubation for 60-96 hours at 35° C. in a commercial scale (i.e. 115 liter) fermenter in a casein containing medium under a controlled anaerobic atmosphere. Growth of the bacterium is usually complete after 24 to 36 hours, and during the 60-96 hour fermentation most of the cells undergo lysis and release botulinum toxin. Control of the fermentation medium pH is not required in a Schantz or modified Schantz process. It is believed that toxin is liberated by cell lysis and activated by proteases present in the culture broth. Optionally, a filtration of this culture medium using a single layer depth filter to remove gross impurities (i.e. whole and ruptured cells) can be prepared to obtain a clear solution referred to a clarified culture.

(74) Harvest

(75) Harvest of toxin can be accomplished by lowering the pH to 3.5 with sulfuric acid to precipitate the raw toxin at 20° C. The raw toxin was then concentrated by ultramicrofiltration followed by diafiltration.

(76) Purification

(77) The harvested crude toxin was then transferred to a digestion vessel and stabilized by addition of the protease inhibitor benzamidine hydrochloride. DNase and RNase were added to digest nucleic acids. The toxin containing material was subjected to UF/DF and three precipitation steps (cold ethanol, hydrochloric acid and ammonia sulfate precipitations). The purified botulinum neurotoxin complex (bulk toxin) was stored as a suspension in a sodium phosphate/ammonium sulphate buffer at 2-8 degrees C.

(78) The resulting bulk toxin was a high quality crystalline 900 kD botulinum toxin type A complex made from the Hall A strain of Clostridium botulinum with a specific potency of ≧3×10.sup.7 U/mg, an A.sub.260/A.sub.278 of less than 0.60 and a distinct pattern of banding on gel electrophoresis, and suitable for use for the compounding of a botulinum toxin pharmaceutical composition.

(79) Compounding can encompass a many fold dilution of the bulk toxin, mixing with one or more excipients (such as albumin and sodium chloride) to thereby form a toxin composition, and preparation of a storage and shipment stable form of the toxin composition, as by lyophilizing, freeze drying or vacuum drying the composition.

(80) The purified botulinum toxin complex obtained from a Schantz or modified Schantz process can be eluted from an ion exchange column in a pH 7-8 buffer to disassociate the non toxin complex proteins from the botulinum toxin molecule, thereby providing (depending upon the type of Clostridium botulinum bacterium fermented) pure botulinum toxin type A with an approximately 150 kD molecular weight, and a specific potency of 1-2×10.sup.8 LD.sub.50 U/mg or greater; or purified botulinum toxin type B with an approximately 156 kD molecular weight and a specific potency of 1-2×10.sup.8 LD.sub.50 U/mg or greater, or purified botulinum toxin type F with an approximately 155 kD molecular weight and a specific potency of 1-2×10.sup.7 LD.sub.50 U/mg or greater.

(81) As set forth supra, in one aspect our invention eliminates the harvest purification steps set forth in this Example 6 carried out upon clarified culture, including elimination of use of the animal derived products, such as RNase and DNase.

Example 7

APF Media and Process for Obtaining a Botulinum Toxin

(82) This Example 7 sets forth an APF process carried out to obtain highly potent and highly purified Clostridium botulinum toxin type A (i.e. bulk toxin). The process can be used with other botulinum toxin serotypes.

(83) Stock Culture Preparation

(84) As set forth in Example 6, Clostridial botulinum can be obtained from the ATCC, from various sources in nature or from a botulism patient. The bottom half of FIG. 1 summarizes the APF process used for preparation of a cell bank vial, and for the culture and fermentation of a botulinum toxin. APF cell bank vials were prepared by culturing Clostridium botulinum on plant agar plates. The plant agar plates were made by mixing the soy derivative HySoy (Quest) with a yeast extract and glucose in a 3:1:1 (weight percent) ratio with agar and allowing setting. Other commercially available APF agar plates or dehydrated powder for making the plates were also found to be suitable. Selected high growth colonies were then inoculated into an APF cell bank vial medium. The APF cell bank vial medium used comprised hydrolyzed soy protein, yeast extract (no animal product was used in either the cultivation of the yeast or in the process for preparation of the yeast extract made therefrom) and glucose in the same 3:1:1 ratio. Other nutrient ratios (i.e. 6:1:1, 6:0:1 and 6:3:1 were also found to be suitable). The hydrolyzed soy (HySoy) and yeast extract (HyYest) concentrates used were obtained from Quest International. The Clostridium botulinum culture in the APF medium was combined with glycerol, aliquoted to cryovials and frozen for later use. The APF media developed can be used to store the Clostridial botulinum bacteria for a period of one year or longer without loss of viability. These frozen culture and glycerol mixtures in cryovials are the APF cell bank vials.

(85) Seed Cultivations

(86) An APF cell bank vial was thawed at room temperature, followed by a single cultivation step: a one liter seed culture bottle was then inoculated directly (i.e. without an intervening blood agar culture or tube growth steps) with the APF cell bank vial contents using the same APF medium (the APF cell bank vial [storage] medium can be different from the APF fermentation [growth] medium) and maintained at 35° C. for 15 to 24 hours, with an initial medium pH of 7.0 in an anaerobic (nitrogen) atmosphere.

(87) Fermentation

(88) Next the seed bottle culture was transferred to a commercial scale 10 liter production fermenter containing the APF medium (hydrolyzed soy protein, yeast extract and 1% glucose) maintained at 35° C. for 52-72 hours, with an initial medium pH of 7.0, in an anaerobic (nitrogen) atmosphere. Approximately 15 hours after commencement of the fermentation (the culture pH has naturally decreased to below 6.0), a pH control program at range of pH 5.0-5.5 was initiated by adding HCl to the culture. It was found that it was necessary to control the pH of the APF fermentation medium within the narrow range in order to obtain an acceptable yield of active botulinum toxin. Thus, it was found that this pH control to between pH 5.0-5.5 substantially prevented degradation and loss of potency of the botulinum toxin. It is believed that during the fermentation most of the cells undergo lysis and release botulinum toxin and that toxin liberated by cell lysis is activated by proteases present in the culture broth. Filtration of this culture medium using a single layer depth filter removes gross impurities (i.e. whole and ruptured cells) and results in a clear solution referred to a clarified culture.

(89) Harvest

(90) Harvest of botulinum toxin can then proceed as in Example 6 (i.e. sulfuric acid precipitation, followed by concentrated by microfiltration followed by diafiltration).

(91) Purification

(92) Purification of the toxin can then proceed as set forth in Example 6: i.e. addition of benzamidine hydrochloride, and DNase and RNase, sulfuric acid precipitation, cold ethanol precipitation, phosphate buffer extraction, hydrochloric acid precipitation, phosphate buffer extraction and bulk toxin storage.

(93) As an alternative to the Example 6 harvest and purification process, a column chromatography process of the present invention can be carried out.

(94) The resulting bulk toxin is a high quality crystalline 900 kD botulinum toxin type A complex made from the Hall A strain of Clostridium botulinum with a specific potency of ≧3×10.sup.7 U/mg, an A.sub.260/A.sub.278 of less than 0.60 and a distinct pattern of banding on gel electrophoresis, and suitable for use for the compounding of a botulinum toxin pharmaceutical composition. Thus, this APF process for a botulinum toxin can generate high quality toxin.

(95) The purified botulinum toxin complex obtained from an APF process can be passed through and eluted from an ion exchange column in a pH 7-8 buffer to disassociate the non toxin complex proteins from the botulinum toxin molecule, thereby providing (depending upon the serotype of Clostridium botulinum bacterium fermented) botulinum toxin with an approximately 150 kD molecular weight, and a specific potency of 1-2×10.sup.8 LD.sub.50 U/mg or greater; or purified botulinum toxin type B with an approximately 156 kD molecular weight and a specific potency of 1-2×10.sup.8 LD.sub.50 U/mg or greater, or purified botulinum toxin type F with an approximately 155 kD molecular weight and a specific potency of 1-2×10.sup.7 LD.sub.50 U/mg or greater. For example, by use of our APF medium we were able to obtain a botulinum toxin type A complex with a specific potency of 1.02×10.sup.8 LD.sub.50 U/mg of the botulinum toxin.

(96) In this Example 7 APF media with either 1% by wt or 2% by wt glucose were used (note that 1% glucose means 1 g of glucose per 100 ml of the culture medium and 2% glucose means 2 g of glucose were present for each 100 ml of the culture medium) and it was determined that maximal bacterium growth (as determined by peak optical density [optical density was measured at 600 nm] of the culture) occurred after about 20 hours of fermentation in the 1% glucose APF medium vs after about 40 hours of fermentation in the 2% glucose APF medium, but that the peak optical densities did not differ significantly as the glucose content of the media was so varied. It was believed that cell autolysis and toxin release resulted in a maximal amount of active botulinum toxin in the 1% glucose APF media (as determined by a SNAP-25 assay for active toxin) after about 55 hours of fermentation, but that with the 2% glucose APF media the amount of active botulinum toxin present in the medium at a later time (as determined by a SNAP-25 assay for active toxin) and was still increasing after 65 hours of fermentation. Thus, a more rapid release of botulinum toxin occurred with use of the lower (1%) glucose APF medium amount present, indicating that a more efficient toxin production process (i.e. more amount of toxin obtained per unit of time) can be carried out with use of the lower (1%) glucose APF medium.

(97) As shown by FIG. 1, it was also determined that optimal parameters for production of botulinum toxin in an APF medium were the combination of the following parameters: (1) about 6% by weight of a hydrolyzed soy concentration (“HySoy Conc.” in FIG. 1) in the APF fermentation medium. 6% soy means 6 g of the soy protein per 100 ml of the culture medium; (2) 0% to 3% yeast extract concentrate (“YE Conc.” In FIG. 1) in the APF fermentation medium; (3) 50-72 hours of fermentation at a temperature of 33-35° C. under anaerobic (nitrogen atmosphere) conditions; (4) pH of the fermentation medium maintained between about pH 5.0 to 5.5 throughout the fermentation period after the initial cell growth, and (5) 1 wt % glucose in the APF fermentation medium.

(98) Thus, as shown by FIG. 1 as more protein is present in the APF medium (as the total amount of HySoy and YE) the pH of the medium tends to increase with resulting lower toxin stability and that when the pH was lowered with the same total protein nutrient content in the medium, toxin production yield increased dramatically. In the non-APF process the total protein content is lower so that pH does not tend to rise and therefore there is no elevated pH to have a deleterious effect on toxin production. FIG. 1 shows that there was consistently more activity (as determined by the MLD50 and SNAP-25 assays) when the pH of the medium was controlled to within a narrow range of about 5.3 to 5.5. FIG. 1 also shows that the highest toxin yield (as determined by the SNAP 25 assay) was obtained with a medium which comprised 6% hydrolyzed soy and 1% yeast extract. FIG. 8 shows that when the yeast and glucose nutrients were both at 1%, that cell lysis between 68-100% and potency as high as about 38×10.sup.5 units/mL of toxin was obtained, as soy protein was varied from 1 to 6% weight.

(99) The SNAP-25 assay used was an ELISA based method to measure SNAP-25 proteolytic activity of the botulinum toxin. SNAP-25 is an abbreviation for synaptosome associated protein of 25 kDa molecular weight. SNAP-25 is a 206 amino acid plasma membrane protein involved in neuronal exocytosis. The assay is based on the method disclosed in Ekong T., et al., Recombinant SNAP-25 is an effective substrate for Clostridium botulinum type A toxin endopeptidase activity in vitro, Microbiology (1997), vol 143, pages 3337-3347. The assay uses a truncated SNAP-25 protein (the 206 amino acid residue peptide) bound to polystyrene 96 well microtiter plates and a monoclonal antibody that recognizes the cleaved product (a 197 amino acid residue peptide) which is made by enzymatic hydrolysis between amino acids 197 and 198 of the SNAP-25 by reduced botulinum toxin type A. The monoclonal antibody bound to the cleaved product is then detected with a secondary antibody (goat anti-mouse IgG conjugated to horseradish peroxidase [HRP)], which produces a color change in the presence of a chromogenic substrate (TMB).

(100) The MLD50 (mouse 50% lethal dose) assay is a method for measuring the potency of a botulinum toxin by intraperitoneal injection of the botulinum toxin into female mice (about four weeks old) weighing 17-22 grams each at the start of the assay. Each mouse is held in a supine position with its head tilted down and is injected intraperitoneally into the lower right abdomen at an angle of about 30 degrees using a 25 to 27 gauge ⅜″ to ⅝″ needle with one of several serial dilutions of the botulinum toxin in saline. The death rates over the ensuing 72 hours for each dilution are recorded. The dilutions are prepared so that the most concentrated dilution produces a death rate of at least 80% of the mice injected, and the least concentration dilution produces a death rate no greater than 20% of the mice injected. There must be a minimum of four dilutions that fall within the monotone decreasing range of the death rates. The monotone decreasing range commences with a death rate of no less than 80%. Within the four or more monotone decreasing rates, the two largest and the two smallest rates must be decreasing (i.e. not equivalent). The dilution at which 50% of the mice die within the three day post injection observation period is defined as a dilution which comprises one unit (1 U) of the botulinum toxin.

(101) Significantly, the APF process of this Example 7 differs from the Example 6 non-APF process, by at least: (1) replacing the cell bank vial cooked meat medium with an APF medium; (2) eliminating the blood agar colony selection step; (3) eliminating the subsequent casein medium based tube growth step, and; (4) replacing the non-APF fermentation media with APF media throughout.

(102) FIG. 2 presents a summary of the differences between an industrial scale (non-APF) Schantz process (Example 6 and the industrial scale APF process of Example 7, through the cell bank creation, culture and fermentation steps. FIG. 2 omits the harvest and purification steps.

(103) The APF media can be used to select for Clostridium botulinum bacteria. Thus, concurrent practice of the Examples 6 and 7 initial culture steps permits isolation and growth of a Clostridium botulinum culture with characteristics conducive to growth and production of botulinum toxins in or on an APF medium. The transfer of Clostridium. botulinum bacteria from a non-APF medium to an APF medium enriches for and selects for bacteria that can either adapt to the new environment or through selective die off of bacteria that cannot grow and produce in the new environment.

Example 8

Chromatographic Systems and Methods for Purifying a Botulinum Toxin

(104) The chemicals used in the experiments set forth in Examples 8 and following included: 10N NaOH (Mallinckrodt, VWR Cat # MKH38505) Acetic Acid, USP/FCC Grade, 99.5-100.5% (J.T.Baker, Cat # JT9522-2) Ammonium Sulfate, Ultrapure, 99% (ICN, Cat # IC808211) Citric Acid, USP/FCC Grade, 99.5-100.5% (J.T.Baker, Cat # JT0119-1) Ethanol, anhydrous, denatured (JT Baker, Cat #9299-1) Hydrochloric acid, NF/FCC Grade, 36.5-38%-Mallinckrodt-MK2612-14 Phosphoric acid, NF/FCC, 85%-88% (Mallinckrodt, Cat # MK278814) Sodium acetate trihydrate, 99%-101%, USP/FCC (Mallinckrodt, Cat # MK735602) Sodium chloride, USP/FCC Grade, 99.0-101.0 (Mallinckrodt, Cat # MK753204) Sodium citrate, USP/FCC Grade, 99.0-100.5% (J.T.Baker, Cat # JT3650-1) Sodium hydroxide, NF/FCC Grade, 95.0-100.5%-Mallinckrodt-MK768004 Sodium phosphate, dibasic Heptahydrate, USP (Mallinckrodt, Cat # MK789604) Sodium phosphate, monobasic monohydrate, USP/FCC (Mallinckrodt, Cat # MK786812)

(105) The chromatography resins use in the experiments below included: Bakerbond ABx Prepscale (JT Baker, Cat #7269-02) Butyl Sepharose FF (Amersham Biosciences, Cat #17-0980-02) Ceramic Hydroxyapatite, Type I (Bio-Rad, Cat #158-4000) Ceramic Hydroxyapatite, Type II (Bio-Rad, Cat #157-4200) HiTrap HIC Selection Kit (Amersham Biosciences, Cat #17-1349-01) HiTrap IEX Selection Kit (Amersham Biosciences, Cat #17-6002-33) MEP Hypercel (Ciphergen, sample) SP Sepharose HP (Amersham Biosciences, Cat #17-1087-03)

(106) The equipment and accessories used is the experiments below included: AKTA Purifier and AKTA FPLC Chromatography System (Amersham Biosciences) Bottle-top 0.22 μm vacuum sterile filter (Nalgene) Labscale TFF system and Pellicon XL50 with Biomax 100 membrane (Millipore) (this is the ultrafiltration equipment). Masterflex L/S pump Model #77201-62 (Cole-Parmer) Pellicon 2 Mini Holder (Millipore) XK and HR columns (Amersham Biosciences)

(107) The buffers used in our experiments are listed in Table 2.

(108) TABLE-US-00003 TABLE 2 Buffers used in the APF purification process Purification Steps Buffers used Butyl Sepharose 1. 50 mM NaPi, 4M NaCl, pH 6.0 FF Chromatography 2. 50 mM NaPi, 2M NaCl, pH 6.0 3. 50 mM NaPi, 1M NaCl, pH 6.0 4. 50 mM NaPi, pH 6.0 SP Sepharose HP 5. 20 mM Na citrate, pH 4.0 Chromatography 6. 20 mM Na citrate, 300 mM NaCl, pH 4.0 7. 20 mM Na citrate, 400 mM NaCl, pH 4.0 8. 20 mM Na citrate, 1M NaCl, pH 4.0 Post Purification Steps Solutions used Post-column processes  9. 50 mM NaAc, pH 4.0 10. 3.5M ammonium sulfate Miscellaneous 11. 0.1N NaOH 12. 1N NaOH

(109) In Table 2: buffers 1 and 2 were used to wash impurities off the column; buffers 3 and 4 was used to elute bound toxin from the column; buffer 5 was used to dilute the eluent from the Butyl column; buffer 6 was used to wash impurities off the column; buffers 7 and 8 were used to elute bound toxin from the column; buffer 8 was the UF/DF dialysis buffer; solution 9 was used to precipitate toxin, and solutions 10 and 11 were used to inactivate (clean) any toxin remaining in the columns after use.

Example 9

Selection of Preferred Chromatography Columns for Use in an APF Column

Chromatographic Botulinum Toxin Purification (Capture Step) Process

(110) This experiment established preferred chromatography columns and techniques for initial purification of a botulinum toxin type A complex from attendant impurities in a fermentation medium.

(111) Feed Materials

(112) Both a filtered cell culture (clarified culture) obtained from an APF process fermentation and an extract thereof prepared by hydrochloric acid precipitation were assessed as chromatography column feed materials. It was found that direct loading of the clarified culture onto a column prevented toxin precipitation and that a clarified culture feed material was much easier to handle and validate. On the other hand, use as the feed material of a clarified culture extract prepared by acid precipitation removed additional impurities and provided virus inactivation. With regard to the characteristics of process robustness, a clarified culture was determined to be the preferred feed material, as opposed to use of a hydrochloric acid precipitation preparation as the bulk botulinum toxin complex chromatography resin feed material. Hence, clarified culture was the preferred feed material.

(113) Our studies showed that as the pH was lowered proteins (i.e. the botulinum toxin complex) started to precipitate at about pH 5, that small amounts of toxin was extracted (as most had precipitated out) at about pH 4.0, and that essentially all of the toxin had precipitation out of the solution at between pH 3.5 to 3.8. On the other hand, we found (based for example on SDS-PAGE and Western blotting) that most impurities were co-extracted with the botulinum toxin at a pH of 6.8. Hence, a preferred feed liquid pH for carrying out our purification process invention was between about pH 5-6.8, with a more preferred pH being about pH 5.5 for extraction, that is separation of the botulinum toxin from attendant impurities.

(114) Capture Step

(115) For the capture step botulinum toxin type A (Hall strain) cell culture filtrates were incubated with a number of chromatography resins (see below) under the manufacturer specified conditions for use of each particular column.

(116) After washing the columns, the column bound proteins were eluted with the specified elution buffers. All eluted fractions were collected and analyzed by SDS-PAGE. The results obtained (Table 3) were confirmed by chromatography using 1 ml HiTrap or HR5/5 columns.

(117) TABLE-US-00004 TABLE 3 Summary of Capture Step Results Separation Toxin in Toxin in Separation Technique Resin Flowthru Eluate Observed Hydrophobic Phenyl FF (HS) − + + Interaction Octyl FF − + + Butyl FF − + + Ion Exchange Q FF + − + SP FF + − − Mixed Mode HA Type I + − − HA Type II + − − Abx + − − Hydrophobic MEP − + − Charge-Induction Immobilized Chelating FF + − − Metal-ion Affinity

(118) This experiment clearly showed that the desired separation of the botulinum toxin from other substances present was best achieved by use of hydrophobic type column chromatography. Thus, we found that the botulinum toxin bound to hydrophobic columns, but that it did not bind to an ion exchange column, such as the Q Sepharose FF column.

(119) Among the hydrophobic columns evaluated, the weakly hydrophobic Butyl Sepharose FF gave the best resolution. Therefore, either Butyl Sepharose FF in binding mode or Q Sepharose FF in flowthru mode provided a preferred botulinum toxin capturing step.

(120) Thus, we determined that an efficient capturing step can be carried out using a hydrophobic column, such as the Butyl Sepharose column chromatography. Presumably, the toxin binds to the Butyl column via a hydrophobic interaction. Prior to this experiment it was unknown that a botulinum toxin complex could be purified toxin directly from clarified culture using a hydrophobic chromatography column. We found that the Butyl Sepharose Fast Flow column has high binding capacity, allows fast flow rate with low back pressure and is therefore suitable for the capturing step that requires fast removal of impurities.

Example 10

Four Column APF Chromatographic System and Process for Purifying a Botulinum Toxin Complex

(121) Intermediate and Polishing Purification Steps

(122) Additional (intermediate and polishing) toxin purification steps were carried out using the toxin-containing fractions obtained from the preferred Q and Butyl columns of Example 9.

(123) Three types of chromatography columns were found effective for such further purification of the botulinum toxin complex. A Hydroxyapatite (HA) type I column was a preferred column we used because it showed separation, but some toxin was found in the flowthru. Gel filtration with a Superdex 200 column was a more preferred column to use because it permitted purification of the 900 kDa botulinum toxin complex from the impurities, but a minor impurity band was still present on SDS-PAGE.

(124) A most preferred column was a SP Sepharose HP column which we found to separate the botulinum toxin from impurities with very good resolution. The botulinum toxin was pure after SP Sepharose HP chromatography, based on analysis by SDS-PAGE.

(125) TABLE-US-00005 TABLE 4 Summary of Column Chromatography Purification Steps Separation technique Resin Summary Mixed mode Hydroxyapatite Toxin in flowthru mode, separated type I some impurities. Gel filtration Superdex 200 Partially purified toxin, difficult to scale-up, low productivity. Ion exchange SP Sepharose HP High resolution separation, pure toxin obtained.

(126) Based on the results of Examples 8 and 9, and as shown by Table 4, the following four column chromatography purification process was developed: 1. use of a Q Sepharose FF column for initial purification of a clarified culture. In this step impurities bound to the column and the toxin flowed through the column; 2. the eluent from the Q Sepharose FF column step 1 was then passed through a Butyl Sepharose FF column. The toxin bound to the column and was eluted off with a suitable buffer; 3. the eluent from the Butyl Sepharose FF was then passed through a Hydroxyapatite type I column. Impurities bound to the column and the toxin flowed through the column; 4. the eluent from the Hydroxyapatite type I was then passed through an SP Sepharose column. The toxin bound to the column and was eluted off with a suitable buffer.

(127) This four column toxin purification process can be summarized as: APF clarified culture=>Q(flowthru)=>Butyl(binding)=>HA (flowthru)=> SP (binding)=>purified toxin complex

(128) This four column bulk botulinum toxin complex process allowed direct loading of filtered culture supernatant onto the Q column (step 1). The flowthru was supplemented with ammonium sulfate to 0.8M before the second step of loading onto the Butyl column. For the third step, the butyl eluate was loaded onto the HA column directly, while the flowthru of the HA was diluted 4 times with deionized water and the pH was adjusted to 4.0 before loading onto the SP column for the fourth column step. This four column process required minimal sample handling at each step, and ensured that the toxin was exposed to mild buffering conditions throughout the four steps of this purification process.

(129) A scale up of the four column purification process set forth above was used carried out upon 680 ml of filtered culture supernatant obtained from an APF botulinum toxin type A fermentation process. The results (see Table 5) show that this four column process resulted in highly in a high yield of highly purified botulinum toxin type A complex.

(130) TABLE-US-00006 TABLE 5 Results of a Scale Up Purification using the Four Column Purification process. Toxin yield ~30 mg per L culture based on UV and Hc-ELISA. Toxin purity >98%, monodisperse, 900 kDa complex based on SEC- HPLC and LS. Pure on SDS-PAGE, western blotting conforms to standard. Toxin potency 3-5 × 10.sup.7 MLD.sub.50 units per mg based on mouse toxicity assays.

Example 11

Additional Multi-Column APF Chromatography Processes for Purifying a Botulinum Toxin Complex

(131) Using the same procedures set forth in Examples 9 and 10 additional column combinations were evaluated. It was determined that each of the following four additional column combinations provided APF methods for obtaining highly purified botulinum toxin complex, as determined by SDS-PAGE. 1. Q (flowthru)=>Butyl=>SP 2. Butyl=>Q or HA (flowthru)=>SP 3. Butyl=>SP=>Q or HA (flowthru) 4. Butyl=>SP

(132) The purified toxins were further analyzed by SEC-HPLC with light scattering, capillary electrophoresis, residual DNA assay, Hc-ELISA, and MLD50. No significant differences were found among the toxins from the four different processes set forth above. The results are summarized in Table 6.

(133) TABLE-US-00007 TABLE 6 Quality summary of toxin samples purified by different APF processes 1.4. above. SEC-HPLC/LS Purity >99%, purer than BCC2030, but less homogeneous than BCC2030. Capillary Identical to one another, similar to 19P and 20P eletrophoresis Research Grade APF Toxin, but slightly different from BCC2030. Picogreen DNA 2-6 ng/ml, significantly lower than BCC2030. assay Mouse toxicity Toxin potency 3.1-4.8 × 10.sup.7 MLD.sub.50 units/mg toxin assay, Hc-ELISA (by UV), or 3.8-12 × 10.sup.7 MLD.sub.50 units/mg toxin (by Hc-ELISA). Silver staining Identical to one another. SDS-PAGE

Example 12

Two Column APF Chromatography Process for Purifying a Botulinum Toxin Complex

(134) Based on the results obtained in Example 11 a two column (Butyl=>SP) column chromatography process was selected for further development.

(135) Optimization of the First Step: Butyl Sepharose FF Toxin Capture

(136) Feed: Feed is to the clarified culture loaded on the column. Since ammonium sulfate can affect the buffer pH, the use of NaCl to replace ammonium sulfate in Butyl column was evaluated. We found that addition of NaCl to the feed sufficient to 2M NaCl allowed the botulinum toxin complex to bind to the butyl column. Subsequently, we determined that feed at a 4M NaCl increased the binding of botulinum toxin complex to the Butyl column, such that the yield of toxin from the Butyl column was increased by 30% to 50%, as determined by Hc-ELISA, as compared to use of feed at 2M NaCl.

(137) The addition of NaCl to the clarified culture (the feed) caused a small pH shift. However, the acceptable feed pH was established between pH 5 and pH 6 and the final pH of the feed after NaCl addition was within pH 5 and pH 6. Hence the preferred feed to use in this first step of a two column purification process has a 4M NaCl concentration and is at pH 5-6. Solid NaCl was added to the clarified culture directly to obtain the 4M NaCl concentration and this feed was then added to the Butyl column. The bound toxin was eluted from the column using a 1M NaCl elution buffer.

(138) It was surprising that most of impurity proteins could be washed away from the column and most of toxin bound to the column could be eluted with a 1M NaCl buffer because column purification processes typically consist of 3 or more columns, except for an affinity column process. We determined that this butyl column is unique as it has the ability to remove many of the impurity proteins. Thus, after use of this column the botulinum toxin complex purity was approximately 50%.

(139) A wash step was then carried out to remove impurities from a column. The impurities in the column came from the clarified culture feed (containing 4M NaCl) used. The optimized washing steps were: 1) Wash #1: 5CV of 50 mM NaPi, 4M NaCl, pH 6.0, and 2) Wash #2: 12CV of 50 mM NaPi, 2M NaCl, pH 6.0. When 12CV and 5CV were compared, it was found that 5CV is not sufficient in removing the impurities. While the wash is to remove impurities after loading the clarified culture in this case.

(140) Elution (to remove toxin bound to a column). Toxin elution with 1.2M, 1.0M and 0.8M NaCl were evaluated. It was chosen to elute toxin with 1M NaCl in 50 mM NaPi, pH 6.0, based on toxin recovery and impurity removal.

(141) Low salt wash: After elution, the column was further washed with 50 mM NaPi, pH 6.0 to remove residual impurities bound to the column for the characterization of purification process.

(142) Cleaning: the column was cleaned with 3CV of 0.1 N NaOH to inactivate any residual toxin before the disposal of used resin.

(143) Running flow rate: The typical flow rate was 100 cm/h. The loading flow rate was between 90 cm/h and 120 cm/h depending on the back pressure.

(144) Loading capacity: Typical loading capacity was 12.7 ml culture per ml bed, or at production scale, 10 L culture for 785 ml resin bed (BPG 100 column at 10 cm bed height).

(145) Bed height: All columns were packed with standard 10 cm bed height.

(146) Optimization of the Second Step: SP Sepharose HP Purification Feed conditioning: The Butyl eluate was diluted 5 times with 20 mM Na citrate buffer, pH 4.0, and the feed pH was adjusted to 4.0. The five times dilution step was carried out to condition the hydrophobic interaction chromatography eluent for use in ion exchange chromatography. We found that the optimal feed pH for best toxin recovery was within the range of pH 4.0±0.2.

(147) Wash step: After loading, the column was washed with 1) 5CV of 20 mM Na citrate, pH 4.0, followed by 2) 3-5CV of 20 mM Na citrate, 300 mM NaCl, pH 4.0 to remove impurities before the elution of bound toxin.

(148) Elution step: The toxin was eluted with 20 mM Na citrate, 400 mM NaCl, pH 4.0.

(149) High salt washing step: After elution, the column was further washed with 20 mM Na citrate, 1M NaCl, pH 4.0 to remove strongly bound impurities.

(150) Column cleaning: The column was cleaned with ˜3CV of 0.1 N NaOH to inactivate residual toxin before the disposal of used resin.

(151) Flow rate: The typical flow rate was 100 cm/h.

(152) Load: The entire Butyl eluate was loaded onto the SP column.

(153) Bed height: All columns were packed with standard 10 cm bed height.

(154) Detailed operating procedures carried out with regard to this two column botulinum toxin complex purification process set forth in this Example 12 are set forth below. 1. Butyl Hydrophobic Interaction Column Materials and Reagents Used Chromatography System: AKTA purifier 100, Amersham Biosciences Resin Type: Butyl Sepharose FF, Amersham Pharmacia Detection: UV (280 nm) Equilibration Buffer/Wash Buffer #1: 50 mM NaPi, 4 M NaCl, pH 6.0 Wash Buffer #2: 50 mM NaPi, 2 M NaCl, pH 6.0 Elution Buffer: 50 mM NaPi, 1 M NaCl, pH 6.0 Low Salt Wash Buffer: 50 mM NaPi, pH 6.0 Cleaning Solution: 0.1 N NaOH Titration Buffer: 500 mM NaPi, pH 7.2
Procedure
Column Packing and Conditioning Equilibrate the column with at least 5-10 CV of Equilibration Buffer or until outlet pH is equivalent to inlet pH.
Sample Preparation
Measure the pH of the Starting Material.

(155) Add solid NaCl to the clarified culture to the final NaCl concentration to 4

(156) M. Addition of 4M NaCl is an example of how to condition the clarified culture for use of the clarified culture as a feed liquid in hydrophobic interaction chromatography. Adjust the pH to 5.0 to 6.0 if needed with Titration Buffer.

(157) Column Loading

(158) Load the clarified culture (containing 4M NaCl) and collect the flow through fraction for analysis.

(159) Column Wash #1 (4 M NaCl Wash)

(160) Wash the column proteins with 5CV of Equilibration Buffer to remove impurity. Collect the wash fraction for analysis and record the volume.

(161) 3.5. Column Wash #2 (2 M NaCl Wash)

(162) Wash the column with 15CV of Wash Buffer #2 to remove additional impurity proteins. Collect the wash fraction for analysis and record the volume.

(163) Elution (1 M NaCl Toxin Peak Elution)

(164) Elute the bound toxin with 5 CV of Elution Buffer. Monitor the 280 nm absorbance of eluate, begin the collection of eluate when the 280 nm absorbance starts to increase and stop the collection of the eluate peak when the 280 nm absorbance reaches the baseline. Record the volume of toxin elution fraction.

(165) Low Salt Wash (0 M NaCl Impurity Peak Elution)

(166) Wash the column with 4CV of Low Salt Wash Buffer to remove residual impurity proteins. Collect the fraction for analysis and record the volume.

(167) Column Cleaning (0.1 N NaOH)

(168) Clean the column with 3 CV of Cleaning Buffer to inactivate the residual toxin before the disposal of used resin.

(169) 2. SP Cation Exchange (Post Butyl) Column

(170) Materials and Reagents Used Chromatography System: AKTA purifier 100, Amersham Biosciences Resin Type: SP Sepharose HP, Amersham Pharmacia Detection: UV (280 nm) Dilution, Equilibration and Wash Buffer #1: 20 mM NaCitrate, pH 4.0 Wash Buffer #2: 20 mM NaCitrate, 300 mM NaCl, pH 4.0 Elution Buffer: 20 mM NaCitrate, 400 mM NaCl, pH 4.0 High Salt Buffer: 20 mM NaCitrate, 1 M NaCl, pH 4.0 Cleaning Solution: 0.1 N NaOH
Procedure
Column Packing and Conditioning

(171) Equilibrate the column with 5-10 CV of Equilibration Buffer or until outlet pH is equivalent to inlet pH.

(172) Sample Preparation

(173) Dilute one volume of 1M NaCl Butyl eluate with 4 volume of Dilution Buffer. Measure the conductivity and pH of the load. Adjust the pH to 4.0 if needed.

(174) Column Loading

(175) Apply the above diluted Butyl eluate to SP column and collect the flow through fraction.

(176) Column Wash #1 (Equilibration Buffer Wash)

(177) Wash the SP column with 5CV of Equilibration Buffer. Continue to collect the eluate as flow through fraction.

(178) Column Wash #2 (300 mM NaCl Wash)

(179) Wash the SP column with 4CV of Wash Buffer #2 to remove impurity proteins. Record the volume of the wash #2 fraction.

(180) Elution (400 mM NaCl Elution)

(181) Elute the bound toxin with 3CV of Elution Buffer. Monitor the 280 nm absorbance of eluate, begin the collection of eluate when the 280 nm absorbance starts to increase and stop the collection of the eluate peak when the 280 nm absorbance reaches the baseline. Record the volume of toxin elution fraction.

(182) High Salt Elution (1 M NaCl)

(183) Elute the strongly bound impurity proteins with 3CV of High Salt Buffer. Collect the fraction for analysis and record the volume.

(184) Column Cleaning

(185) Clean the SP column with 3 CV of Cleaning Solution to inactivate the residual toxin before the disposal of used resin.

Example 13

Robustness of the Two Column APF Chromatography Process for Purifying a Botulinum Toxin Complex

(186) The robustness of the two column method of Example 12 was studied in a series of experiments, as set forth below.

(187) Culture pH

(188) The effect of culture pH on toxin purification was evaluated. A study using cultures grown at pH 5.5 and pH 6.5 as the starting material for the purification was performed, and it was found that the recovery from the pH 6.5 culture was slightly lower than that from the pH 5.5 culture, based on Hc-ELISA results.

(189) Storage Time

(190) Toxin was purified from a culture grown at pH 5.5 on the day of harvesting and after 4-day storage of the culture at 2-8° C. No difference was found, based on toxin recovery, Butyl and SP chromatograms, SDS-PAGE, and Hc-ELISA results.

(191) Column Binding Capacity

(192) The proposed load on the Butyl column was 12.7 ml culture per ml resin, or 10 L culture for BPG100 column (with 10 cm bed height). Butyl and SP columns were tested by loading 4× more culture. SDS-PAGE and Hc-ELISA results indicated little toxin in the flowthru fractions for both Butyl and SP columns. The capacity of Butyl and SP column is at least four times greater than that of the current load. The toxin in SP eluate was pure on SDS-PAGE. The recovery of Butyl column is 48% and the recovery of SP column was 74%, based on Hc-ELISA. The overall yield is 16 mg toxin per L culture, based on UV result.

(193) Process Hold Time

(194) After harvesting, the culture was processed through Butyl column on the same day or after overnight storage. Butyl eluate was normally stored overnight before loading onto the SP column. A preliminary study showed that the Butyl eluate was stable for up to 4 days, which gave identical chromatogram and SDS-PAGE patterns. The stability of SP eluate was evaluated by capillary electrophoresis (CE) and SEC-HPLC. The results showed no difference among samples stored for up to 2 days. The recovery of toxin after filtration was also evaluated for these samples. Toxin recovery was slightly decreased on day 2 compared to day 0, but it was not clear whether such decrease was due to storage or experimental variation.

(195) Cell Density of Culture Two times concentrated culture and 2× diluted culture were evaluated by Butyl column chromatography to study the effect of culture cell density on toxin purification. The chromatograms from both runs looked identical. The impurity and toxin profile from both runs were identical on SDS-PAGE. The Hc-ELISA results (Table 7) showed that the mass balance from both runs were >90%, while the recovery of 2× concentrated culture was significantly lower than that of 2× diluted culture. Twenty-nine percent toxin was lost before toxin elution for 2× concentrated culture, compared with 4% loss for 2× diluted culture.

(196) TABLE-US-00008 TABLE 7 APF toxin mass balance analyzed by Hc-ELISA. Mass 2M 1M 0M Run balance FT Wash Elution Elution 2x concent. 91% 11% 18% 53%  9% 2x diluted 97%  0%  4% 74% 19%

(197) Bioburden Studies

(198) Bioburden was monitored at different steps of the process. Samples of Butyl load, Butyl eluate after 3 day storage, SP load, SP eluate, and SP eluate after overnight storage were evaluated. Some contaminants were noted (˜<1 CFU/ml to 35 CFU/ml). The sample with the highest number of contaminants was the Butyl eluate. Contaminants may be due to the uncontrolled environment in which purification process was performed.

(199) Effect of 4M NaCl

(200) In order to evaluate the effect of 4M NaCl on the toxin in culture, the culture containing 4M NaCl was kept at 4° C. overnight and then Butyl and SP columns were performed. The chromatographic result, SDS-PAGE and Hc ELISA showed there was no effect of 4M NaCl on the toxin in the culture after overnight storage.

(201) Culture Media (3:1:1 Vs 5:1:1)

(202) Two point five liters of 5:1:1 and 3:1:1 cultures were processed. The toxin recovery for each of the purifications analyzed by Hc-ELISA is summarized in Table 8. The toxin purified from both cultures was pure on SDS-PAGE, which indicates that the process developed with 3:1:1 culture can be used to purify toxin from 5:1:1 culture.

(203) TABLE-US-00009 TABLE 8 Toxin recovery based on Hc-ELISA Step 3:1:1 culture 5:1:1 culture Butyl 46% 43% SP 63% 44%

(204) Working pH for SP Sepharose HP Chromatography

(205) SP Sepharose HP chromatography was carried out at different pH values: 3.5, 4.2, and 4.5. It was found that pH 3.5 caused toxin precipitation in the column, no toxin was eluted with 400 mM NaCl and very little toxin came out with 1M NaCl. At pH 4.5, toxin did not bind to the SP column. Preliminary results obtained at pH 4.2 showed that the toxin did not bind as strongly as at pH 4.0 and was eluted as a broad peak after the wash peak at 300 mM NaCl. The results indicate that the pH at this step was critical and that the optimal pH range was narrow.

Example 14

Evaluation of Two Column APF Chromatography Process for Purifying a Botulinum Toxin Complex

(206) Various eluents from each of the two columns of the purification process of Example 12 were evaluated as set forth below.

(207) A. Butyl Sepharose FF Chromatography

(208) Filtered 3:1:1 culture was used as the feed for this experiment. Before loading the feed (clarified culture obtained from a Schantz fermentation of a Clostridium botulinum type A [Hall strain]) onto the Butyl Sepharose FF column (XK50/10, column diameter 5 cm, bed height 10 cm, column volume: 196 ml), 584.4 g of NaCl was added to 2500 ml of culture with stirring for ˜30 min. Atypically, the feed pH was adjusted to 5.81 and the running flow rate was set at 92 cm/h (normal flow rate is 100 cm/h). The loading volume was 2800 ml.

(209) After loading, the column was washed with 5CV or 1000 ml of 50 mM NaPi, 4M NaCl, pH 6.0, followed by 15CV or 3000 ml of 50 mM NaPi, 2M NaCl, pH 6.0. The bound botulinum toxin type A complex was then eluted from the column with 5CV or 1000 ml of 50 mM NaPi, 1M NaCl, pH 6.0. After the elution of the botulinum toxin complex, the strongly bound impurities were washed off the column with 4CV or 800 ml of 50 mM NaPi, pH 6.0. The column was next washed with 2CV (400 ml) of 0.1 N NaOH to inactivate residual toxin before the disposal of used resin. The chromatogram of the toxin eluent is shown in FIG. 3.

(210) FIG. 3 shows that the Butyl column used can provide good separation of botulinum toxin complex from impurities present with it in the clarified culture feed liquid. As measured by UV280 nm, FIG. 3 shows the flow through peak and the peaks of 2M NaCl, 1M NaCl, 0M NaCl and 0.1 N NaOH. Based on the peak size, it was determined that most impurities were removed in the flow through fraction. A significant amount of impurities were also removed in 2M NaCl fraction before the elution of toxin in the 1M NaCl fraction.

(211) FIG. 3 is a chromatograph obtained from passage of an APF clarified culture (a 3.1.1 culture) through a Butyl hydrophobic interaction column. The X axis represents the volume in ml of liquid (effluent) which has passed through the column. The Y axis represents the UV absorbance at 280 nm in mAU. In addition, the conductivity (separate graph line) was monitored during chromatography.

(212) As shown by FIG. 3, many protein impurities passed through the column in about the first approximately 3000 mls. The 4M NaCl and 2M NaCl washes buffer cause subsequent, though smaller peaks, showing removal of additional impurities. Use of the 1M NaCl (at about the 7000 ml volume) caused elution of bound toxin complex from the column and this was the fraction loaded onto the second column.

(213) B. SP Sepharose HP Chromatography

(214) The axes in FIG. 4 are the same as they are for FIG. 3. The steps carried out to obtain the FIG. 4 chromatograph were as follows: (1) one hundred ml of the Butyl eluate obtained from Example 12 (the Butyl column eluent resulting from FIG. 3) was diluted with 400 ml of 20 mM Na citrate buffer at pH 4.0 (a five times dilution therefore). The pH of this diluted Butyl eluent was 4.1. (2) four hundred and sixty-six ml of this feed was then loaded onto the SP Sepharose HP column (XK26/10, column diameter 2.6 cm, bed height 10 cm, column volume:53 ml). (3) after loading the column was washed (at about the volume 450 ml point on the x axis of FIG. 4) with 5CV or 250 ml of 20 mM Na citrate, pH 4.0. (4) the column was then washed with 4CV or 200 ml of 20 mM Na citrate, 300 mM NaCl, pH 4.0 (at about the volume 725 ml point on the x axis of FIG. 4). (5) the column bound botulinum toxin complex toxin was then eluted with 3CV or 150 ml of 20 mM Na citrate, 400 mM NaCl, pH 4.0 (at about the volume 925 ml point on the x axis of FIG. 4). (6) after elution of the column bound toxin complex, the column was further washed with 3CV or 150 ml of 20 mM Na citrate, 1M NaCl, pH 4.0 to elute strongly bound impurities (at about the volume 1050 ml point on the x axis of FIG. 4). (7) the column was then cleaned with 3CV or 150 ml of 0.1N NaOH (just after the volume 1200 ml point on the x axis of FIG. 4).

(215) The FIG. 4 chromatogram shows elution of a botulinum toxin type A complex (about 900 kDa molecular weight) just before the 1000 ml volume point on the x axis of FIG. 4.

(216) FIG. 4 shows that high purified botulinum toxin complex can be obtained by use of the SP sepharose column subsequent to the Butyl column. FIG. 3 shows that there was a broad flow through peak, a small 300 mM NaCl wash peak, 400 mM toxin elution peak and 1M NaCl cleaning peak. As analyzed by SDS-PAGE in FIG. 5B, there was no visible protein band in flow through fraction, some impurity protein bands in 300 mM NaCl wash fraction and 1M NaCl cleaning fraction. Toxin was eluted in 400 mM NaCl elution fraction.

(217) C. Analytical Results:

(218) SDS-PAGE: The elution fractions from the Butyl and SP column chromatography columns were analyzed by SDS-PAGE and the typical result is shown in FIG. 5A (Butyl column) and FIG. 4B (SP column).

(219) FIGS. 5 and 6 are gel electrophoresis records obtained by use of reduced SDS-PAGE. The left had side of the FIGS. 5 and 6 gel electrophoresis records is marked vertically with ascending molecular weights in thousands of Daltons (kDa). The numbers 1 to 6, 1 to 7 or 1 to 8 is FIGS. 5 and 6 represent the fractions loaded onto the gels.

(220) In FIG. 5A: item 1 (gel lane 1) “Mark 12” is the Novex molecular weight marker of standard molecular masses; lane 2 is the clarified culture feed liquid; lane 3 is an aliquot from the wash resulting from use of the flow through (“FT”) and 4M wash in the Butyl column; lane 4 is an aliquot from use of the 2M wash; lane 5 is an aliquot from the tail fraction of the 2M wash; lane 6 is an aliquot from the fraction of 1M elution; lane 7 is an aliquot from the tail fraction of 1M elution, and; lane 8 is an aliquot from the 0M wash.

(221) FIG. 5A shows that the Butyl column removed many impurities (see columns 3-5 in FIG. 5A) and provided initially purified botulinum toxin (see columns 6-8 in FIG. 5A).

(222) In FIG. 5B: item 1 (gel column 1) “Mark 12” is the same molecular weight marker used in FIG. 5A; column 2 is the diluted Butyl column eluent; column 3 is an aliquot of the column flow through; column 4 is an aliquot from the 300 mM wash; column 5 is an aliquot from eluant from the column; column 6 is an aliquot from the 1M wash. FIG. 5B shows that use of an SP column subsequent to use of a Butyl column provided highly purified botulinum toxin (see column 5 in FIG. 5B).

(223) Hc-ELISA

(224) Toxin concentration was analyzed by Hc-ELISA, an ELISA assay to determine the toxin concentration based on the concentration of toxin heavy chain, and toxin mass balance during the purification was estimated. Table 9 shows the toxin concentration and step recovery during Butyl and SP column steps from a typical purification run. The overall recovery after Butyl and SP was 28.6%.

(225) SEC-HPLC

(226) The results from SEC-HPLC showed that the step recovery for SP chromatography was 42.9%, compared to 62.5% from Hc-ELISA. This shows that the recovery of botulinum toxin after the SP column step was approximately 50%.

(227) Normalized Yield

(228) The toxin yield was normalized as 22.3 mg (by SEC-HPLC) or 23.4 mg (by Hc-ELISA) per L culture after Butyl chromatography, and 9.6 mg (by SEC-HPLC) or 8.9 mg (by Hc-ELISA) per L culture after SP chromatography from one run. Thus, using our two column system and process set forth herein, between about 50 mg to about 90 mg of botulinum toxin complex can be purified from each 10 L of fermentation medium clarified culture (as obtained for example from the Example 6 or Example 72 fermentation processes).

(229) TABLE-US-00010 TABLE 9 Toxin concentration and mass balance in typical Butyl and SP chromatography steps. Volume Conc Toxin Amt (ml) (μg/ml) (mg) % Recovery Butyl samples Butyl Load 2800 45.5 127.4 100 Flowthru and Wash 2634 N/A N/A N/A 2M NaCl Wash Peak 336 32.5 10.9 8.6 1M NaCl Elution Peak 404 144.5 58.4 45.8 1M NaCl Post Elution 443 N/A N/A N/A 0M NaCl Wash 369 19 7.0 5.5 SP Samples SP Load 466 19 8.9 100.0 Flowthru 708 N/A N/A N/A 300 mM Wash Peak 54 N/A N/A N/A Elution Peak 35 158 5.5 62.5 1M Wash Peak 13 N/A N/A N/A Cleaning Peak 44 N/A N/A N/A

Example 15

Process for Post Column Chromatography Toxin Complex Stabilization and Storage

(230) 1. Development Rationale

(231) After column chromatography, it is preferred to transfer the purified botulinum toxin complex into a stable buffer at a desired concentration by a UF/DF step, followed by sterile filtration to thereby obtain a toxin suitable for use in a compounding of a botulinum toxin pharmaceutical composition. The purified botulinum was stored either in a soluble form in acetate buffer or as an ammonium sulfate suspension.

(232) 2. UF/DF Step

(233) A polyethersulfone Biomax-10 membrane (NMWCO: 10 kDa, Millipore) was used in the UF/DF step. 50 mM NaAc, pH 4.0 was chosen as the diafiltration buffer. The SP eluate was ultrafiltered to −1 mg/ml, then diafiltered with 8 diafiltration volumes (DV) of 50 mM NaAc, pH 4.0.

(234) Ultrafiltration (UF) is a process for separating extremely small particles and dissolved molecules from fluids. The primary basis for the separation is molecular size although secondary factors such as molecule shape and charge can play a role. Materials ranging in size from 1,000 to 1,000,000 molecular weight are retained by ultrafilter membranes, while salts and water pass through. Colloidal and particulate matter can also be retained.

(235) Diafiltration (DF) is the fractionation process that washes smaller molecules through a membrane and leaves larger molecules in the retentate without ultimately changing concentration. DF can be used to remove salts or exchange buffers. DF can also remove ethanol or other small solvents or additives. There are several ways to perform diafiltration. In continuous diafiltration, the diafiltration solution (water or buffer) is added to the sample feed reservoir at the same rate as filtrate is generated. In this way the volume in the sample reservoir remains constant, but the small molecules (e.g. salts) that can freely permeate through the membrane are washed away. Using salt removal as an example, each additional diafiltration volume (DV) reduces the salt concentration further. (A diafiltration volume is the volume of sample before the diafiltration solution is added.) Using 5 diafiltration volumes will reduce the ionic strength by ˜99% with continuous diafiltration. In discontinuous diafiltration, the solution is first diluted and then concentrated back to the starting volume. This process is then repeated until the required concentration of small molecules (e.g. salts) remaining in the reservoir is reached. Each additional diafiltration volume (DV) reduces the salt concentration further. A diafiltration volume is the volume of sample before the diluting solution is added. Using 5 diafiltration volumes will reduce the ionic strength by ˜96% with discontinuous diafiltration. Continuous diafiltration requires less filtrate volume to achieve the same degree of salt reduction as discontinuous diafiltration.

(236) 3. 0.22 μm Filtration Step

(237) The low-protein-binding 0.22 μm cellulose acetate (CA) vacuum bottle-top filter was selected for the filtration step.

(238) 4. Ammonium Sulfate Precipitation Step

(239) Ammonium sulfate precipitation was then carried out: 3.5M ammonium sulfate was added to the 0.22 μm filtered toxin solution with gentle stirring until the first appearance of opalescence. The purified bulk toxin was then stored at 2-8° C.

(240) 5. Results from a Typical Post-Column Process

(241) SP eluate was concentrated from 70.5 ml to 18 ml using Pellicon Biomax-10 (50 cm.sup.2 surface area, Millipore) on a Labscale TFF system (Millipore) and diafiltered with 8DV of 50 mM NaAc, pH 4.0. The retentate (post-UF/DF fraction) was collected and was filtered with Corning 0.22 μm CA filter (Corning 431154). The UF/DF system was rinsed with acetate buffer. The rinse fraction was collected. Ten ml of the post 0.22 μm filtrate was stored at 2-8° C. for stability studies. Eight ml of the post 0.22 μm filtrate was subjected to ammonium sulfate precipitation. A total of 2.8 ml of 3.5 M ammonium sulfate was added into the filtrate until it became opalescent.

(242) Toxin recovery was estimated based on UV measurement, which is shown in Table 10. SDS-PAGE results are shown in FIG. 4.

(243) In FIG. 6 the lanes shown represent: Lane 1 is M12, molecular weight standards Lane 2 is SP column eluate Lane 3 is UF/DF retentate: UF/DF retentate after UF/DF of SP eluate, diluted to the same amount of loaded protein as Lane 2, for comparison Lane 4 is UF/DF rinse solution from rinsing UF/DF membrane after completion of UF/DF membrane Lane 5 is post 0.2 μm filtration; after UF/DF process and after the sample was filtered with the 0.22 μm filter Lane 6 is post column ammonium sulphate suspension; after 0.22 μm filter filtration, the sample was precipitated with ammonium sulphate because the botulinum toxin complex is stable in ammonium sulphate Lane 7 is UF/DF retentate (same as lane 3), but undiluted, to show the details.

(244) FIG. 6 tells us that the post column purification process steps of UF/DF, 0.22 μm filtration, and ammonium sulphate precipitation do not affect the purity of the botulinum toxin complex, as determined by SDS-PAGE analysis. Significantly, the MLD.sub.50 results showed that the potency of the purified bulk botulinum toxin complex was 2.9-3.7×10.sup.7 MLD.sub.50 units/mg.

(245) TABLE-US-00011 TABLE 10 Toxin recovery based on UV measurement Toxin conc. by UV Vol. Total toxin Recovery Fraction (mg/ml) (ml) (mg) (%) SP eluate 0.389 70.5 27.4 100 (defined) Post UF/DF 1.260 18.0 22.7 82.8 UF/DF rinse 0.220 14.0 3.1 11.3 Post filtration 1.270 18.0 22.8 83.2 Post AS ppt* N/A (~0.94) ~10.8 N/A N/A *from 8 ml post filtration fraction.

(246) FIG. 7 is a flowchart of a preferred animal protein free, two column chromatographic method for purifying a botulinum toxin type A complex. This is a robust, scalable and cGMP compliant process for obtaining purified Clostridium botulinum toxin 900 kDa complex. In FIG. 7 it can be noted that the Butyl eluate is conditioned for ion exchange chromatography by a five times dilution with a pH 4 sodium citrate buffer.

(247) The FIG. 7 process can also be used to obtain pure (i.e. 150 kDa botulinum toxin free of the non-toxin complex proteins) by loading the SP column eluent onto an ion exchange column in a pH 8 buffer to disassociate the non toxin complex proteins from the 150 kDa botulinum toxin molecule, thereby providing (in the flow through from the column) a botulinum toxin type A (neurotoxic component) with an approximately 150 kD molecular weight, and a specific potency of 1-2×10.sup.8 LD.sub.50 U/mg or greater. This process can also be used to obtain other non toxin components of botulinum toxin complex (i.e. non toxin hemagglutinin proteins and/or non toxin non hemagglutinin proteins) by dissociating the complex into its components and next purifying the dissociated components.

(248) The purified toxin complex obtained by our process meets or exceeds the specifications set forth in Table 1. Additionally, the typical yield was approximately 100 mg of 900 kDa toxin complex from a 10 L cell culture, which is higher than the yield obtained from a Schantz (non-APF) process.

(249) Advantages of our invention include: 1. No component or substance derived from animal source is used in the process. Specifically, use of DNase and RNase are eliminated. 2. More than about 50 mg per purified botulinum toxin type A complex with the characteristics set forth in Table 1 can be obtained per 10 liters of fermentation medium. 3. The purified toxin is obtained from a process which is robust, scalable, validatable, and cGMP compliant. Robust means the process is reproducibility even upon an about ±10% change in one or more of the process parameters. Validatable means the process consistently yield purified toxin with the table 1 characteristics. cGMP means that the process can be easily converted to a manufacturing process that complies with FDA required current Good Manufacturing Practices. 4. The potency of the final purified botulinum toxin complex meets or exceeds the potency (as determined by the MLD50 assay) of purified botulinum toxin complex obtained from a Schantz or modified Schantz process. 5. Elimination of any precipitation steps to purify a botulinum toxin complex.

(250) Various publications, patents and/or references have been cited herein, the contents of which, in their entireties, are incorporated herein by reference.

(251) Although the present invention has been described in detail with regard to certain preferred methods, other embodiments, versions, and modifications within the scope of the present invention are possible. For example, a wide variety of animal product free systems and processes (including chromatographic botulinum toxin purification processes) are within the scope of the present invention.

(252) Accordingly, the spirit and scope of the following claims should not be limited to the descriptions of the preferred embodiments set forth above.