TITLE OF THE INVENTION
METHOD FOR THE PRODUCTION OF AN ANTIBIOTIC AGENT
BACKGROUND OF THE INVENTION
The fungus Glarea lozoyensis produces a family of structurally-related, pharmaceutically-important compounds known as the pneumocandins. These acylated hexapeptides are interesting in that all the constituent amino acids contain one or more hydroxyl groups. Compound I of the formula
currently the major pneumocandin produced in this fermentation, possesses the amino acids threonine, 4-hydroxyproline, 3,4-dihydroxytyrosine, 3-hydroxyglutamine, 3- hydroxyproline, and 4,5-dihydroxyornithine in addition to a dimethylmyristate side chain. Moreover, differences in the hydroxylation and amino acid substitution patterns result in over ten biosynthetically derived pneumocandins. The rate of synthesis of compound I and the levels of the structural analogues are affected by certain amino acid supplementations, certain trace elements, and osmolarity. The present invention relates to an improved process for the production of compound I, as well as certain other findings relating to the use of certain amino acids and trace elements. Compound I is disclosed in U.S. Patent No. 5,202,309 which issued
April 13, 1993. Compound I is produced by cultivating the fungus Glarea lozoyensis
(formerly identified as Zalerion arboricola) under aerobic conditions. A process for the production of Compound I is disclosed in U.S. Patent 5,194,377 which issued March 16, 1993. Compound I is produced by cultivating Glarea lozoyensis, ATCC No. 20868, deposited under the Budapest Treaty in the Culture Collection of the American Type Culture Collection at 12301 Parklawn Drive, Rockville, Md. 20852.
SUMMARY OF THE INVENTION
The present invention relates to an improved process for the production of Compound I of the formula
There are also disclosed novel compounds which are structural analogues of Compound I produced during the fermentation of Compound I. These include compounds of the formula
wherein substituents Ri to Rfi are as defined below:
COMPOUND R, R2 R3 R4 R5 Rδ
II OH H CH3 OH H OH
III OH =O CH3 OH H OH rv OH OH H OH H OH
V OH OH CH3 H H OH
VI OH OH CH3 OH H H
VII H H CH3 OH OH OH
Other structural analogues are known in the art and include compounds of the formula
wherein substituents Ri to Rfi are as defined below:
COMPOUND Ri R2 R3 R4 R5 R6 vm H OH CH3 OH H OH
IX H H CH3 OH H OH
X OH OH CH3 OH OH H
XI OH OH CH3 OH OH OH
XII OH OH CH3 OH CH3 OH
Each of the compounds (II-XII) exhibits antifungal activity.
The invention also relates to the use of certain amino acids, trace elements and sugar content to enhance the production of Compound I and impact the production of certain analogue impurities.
In the production of Compound I, it has been found that certain trace elements such as divalent cations preferably zinc and cobalt, which are known to be inhibitors of α-ketoglutarate-linked dioxygenases, reduced the titer of Compound I and increased the levels of structural analogues which possess altered proline, ornithine and tyrosine hydroxylation patterns. Nickel, which is also an inhibitor of
α-ketoglutarate-linked dioxygenases, has no effect on the Compound I titer but altered the hydroxylation pattern of the tyrosine residue.
Thus, certain amino acids and trace elements impact the fermentation of Compound I. In particular, supplementation of the fermentation media with the amino acid threonine controls the level of the serine analogue, Compound IV.
Culture
The fungus Glarea lozoyensis (ATCC 74030) is used to produce Compound I and the structurally related analogues. This improved production strain was derived ultimately from the wild-type organism, ATCC 20868, (isolated from a sample of fresh water) by sequential steps of N-methyl-N'-nitro-iV-nitrosoguanidme mutagenesis. The culture was maintained as aliquots of a mycelial suspension in 5% (v/v) glycerol stored at -70°C.
Seed and Production Media
The compositions of the seed and production media can be composed of a variety of carbon sources, nitrogen sources, inorganic salts, and trace nutrients in a variety of proportions. Where applicable these nutrients can be organic or inorganic, simple or complex. Each nutrient is present at a concentration appropriate and in proportion to the other nutrients in the medium. Typical useful seed and production media are listed below in Table 1 and Table 2.
Table 1 : LYCP-5 seed medium
Adjust pH to 6.0, sterilize 25 minutes at 121 °C
Table 2: FGY production medium
Adjust pH to 5.5, sterilize 30 minutes at 121°C
The following examples illustrate the invention but are not to be construed as limiting the invention disclosed herein.
EXAMPLE 1
Shake-Flask Scale Fermentations
Control Process
A 250 ml Erlenmeyer flask containing 50 ml of LYCP-5 medium
was inoculated aseptically with 1 ml of a thawed culture stock. This first stage seed culture was incubated at 25°C with 220 rpm agitation for 3-5 days. A 1 ml aliquot of the first stage seed was transferred to a second 250 ml Erlenmeyer flask containing 50 ml of LYCP-5 medium. This second stage seed culture was incubated as above for 3 days.
For each variable tested (i.e., treatment group), several 250 ml Erlenmeyer flasks each containing 25 ml FGY medium or a variation thereof (described below) were inoculated at 5% (v/v) with second stage seed. The flasks were incubated at 25°C with 220 rpm agitation for 14 days. The pH for each treatment group was adjusted as required by removing one flask from the group, adding acid or base to return the pH to 5.0-5.5, and then adding this same volume of sterile titrant to the remaining flasks in the group. Where required, a volume of a sterile fructose solution was added during the fermentation to maintain the residual concentration within a specific range. Analysis of the pneumocandins produced was carried out by extracting the whole broth with organic solvent followed by chromatographic analysis using standard reverse phase and normal phase procedures. The titer of Compound I is expressed as arbitrary "units". The levels of the structural analogues is expressed as a ratio percent of the amount of Compound I produced.
Amino Acid Supplementation
On or about day 6 (i.e., mid-cycle) of the fermentation, sterile solutions of L-proline, trαrø-3-hydroxy-L-proline, trøws-4-hydroxy-L-proline, threonine, serine, arginine, ornithine or glutamine were added to the fermentation to give appropriate final concentrations. Pneumocandin extraction and analysis was carried out after 14 days of fermentation.
Increasing the proline concentration in the base medium (0-15 gm/1) resulted in a dose-dependent reduction in the levels of Compounds X and XI while the level of Compound VI increased as a function of proline concentration (Table 3).
A 15 gm/1 addition of proline to each of these treatments on or around day 6 resulted in comparable titers for each treatment but was unable to off-set the effects of the initial level of proline in the medium.
Table 3: Effect of varying the initial proline concentration in the base medium
The mid-cycle addition of hydroxyprolines impacted the fermentation as well (Table 4). A 5 gm/L addition of trα«5-3-hydroxy-L-proline resulted in a 50% improvement in titer with the levels of Compounds X, VI and XI reduced dramatically. Conversely, a 5 gm/L addition of tra«s-4-hydroxy-L-proline resulted in a doubling of the level of Compound X with minimum impact on the other analogues or the titer of Compound I.
Table 4: Effect of 15 gm/L trans-3 and tra«s-4-hydroxyproline added on or about day 6
Amino acids such as glutamine, arginine, and ornithine which can be metabolized to Δ -pyrroline-5-carboxylate (P5C) also appear to have an impact on the analogues which are defined by the specific amino acid incorporated at the position "occupied" by 3-hydroxyproline in Compound I (Table 5).
Table 5: Effect of proline "related" amino acids added (5 gm/L) on or about day 6.
Supplementation of the medium with 5 gm/L threonine or serine resulted in a complete elimination or large increase in the level of Compound IV respectively (Table 6). In both cases, the titer of Compound I was reduced by 30%. Additional work has shown that 1 gm/L threonine is sufficient to maintain Compound IV at acceptable levels while having no impact on the titer of Compound I.
Table 6: Effect of adding 5 gm/L serine or threonine on or about day 6
Effect of Trace Elements
Several trace elements were examined for their impact on the titer of Compound I and the spectrum of structural analogues produced. When added at concentrations equal to the ferrous salt, zinc, cobalt, and nickel salts had the most pronounced effects (Table 7). Zinc reduced the titer of Compound I by 50% and doubled the level of Compound VI. Cobalt affected a 25% reduction in the titer of Compound I while increasing the levels of Compounds VI, VIII, IX and V. The addition of nickel had no impact on the titer of Compound I but increased the level of Compound V.
Table 7: Effects of trace elements
Osmolarity
In this fermentation, osmolarity can be controlled by maintaining the residual fructose concentration at high (>75 gm/L) or low (<30 gm/L). The initial fructose concentration in the control process is 125 gm/L and is kept high by making two 50 gm/L additions during the 14 cycle. Alternatively, the initial fructose concentration can be lowered to 40 gm/L and several 25 gm/L additions made during the course of the fermentation to maintain a low residual sugar level. When the "low" fructose process is run, there is a increase in the titer of Compound I along with an increase in the level of Compound X (Table 8). This increase in the level of Compound X can be offset by adding an inorganic salt such as sodium chloride or sodium sulfate. The addition of inorganic reduces the effects of running at a reduced concentration of fructose. These results suggest that osmolarity plays a role in pneumocandin synthesis.
Table 8: Effect of osmolarity
In summary, hydroxylation patterns of amino acids of Compound I are sensitive to zinc, cobalt and nickel. Additionally, amino acid additions to the production medium have a direct effect on the pneumocandins produced by the fermentation. Supplementation ofthe production medium with proline, trans-3- hydroxyproline and trα«s-4-hydroxyproline effects the incorporation of trans-3 or trαn.s-4-hydroxyproline residues in Compound I. The addition of threonine to the fermentation controls the level ofthe serine analogue, Compound IV.
Thus, the impact of amino acids and trace elements on the fermentation provides insights into factors affecting the biosynthesis of Compound I and has provided for an improved fermentation process by decreasing the levels of structural analogue and increasing the titer of Compound I.