Changes in GE2270 antibiotic production in Planobispora rosea through modulation of methylation metabolism

Luciano Gastaldo and Flavia Marinelli

Biosearch Italia S.p.A., Via R. Lepetit, 34, 21040 Gerenzano (VA), Italy

Correspondence
Flavia Marinelli
fmarinelli{at}biosearch.it


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Thiazolylpeptide GE2270 is a potent antibiotic inhibiting protein synthesis in Gram-positive bacteria. It is produced as a complex of 10 related metabolites, differing mainly in the degree of methylation, by fermentation of the rare actinomycete Planobispora rosea ATCC 53773. Addition of vitamin B12 to the fermentation medium doubled total complex production and markedly changed the relative production of the various GE2270 metabolites, enhancing the biosynthesis of the more methylated component A. Among methylation inhibitors, the addition of sinefungin increased the amount of factor D2, which differs from component A in the lack of a methyl group. Since sinefungin is an S-adenosyl-L-methionine methyltransferase-specific inhibitor, these results indicate that the methylation step converting D2 into A involves an S-adenosyl-L-methionine methyltransferase. Simultaneous supplementation of vitamin B12 and sinefungin led to a twofold increase in D2 concentration, showing that vitamin B12, in addition to having an effect on the late methylation step, exerts a stimulating action on antibiotic backbone synthesis. This is possibly due to its role in an unusual pathway of serine synthesis peculiar to P. rosea metabolism. Finally, fermentation medium modifications were shown to be useful for the production of industrially valuable levels of components A or D2 in the GE2270 complex as starting points for the production of new interesting semi-synthetic antibiotics.


Abbreviations: CV, coefficient of variation


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Antibacterial and antifungal agents belonging to totally new chemical classes and endowed with innovative mechanisms of action are urgently needed for treatment of infectious diseases caused by bacteria and fungi multiresistant to currently available therapies (National Institutes of Health, 2001).

GE2270 is a novel potent antibiotic inhibiting protein synthesis in Gram-positive bacteria (Goldstein et al., 1993; Selva et al., 1991). It is the natural precursor of clinical candidate BI-K0376 specifically active against Propionibacterium acnes, including isolates resistant to currently available therapies (Jabes et al., 2002). GE2270 was discovered from a targeted screen for elongation factor Tu binding antibiotics applied to 34 000 actinomycetes (Selva et al., 1997). It is characterized by a highly modified peptide backbone consisting of six thiazole rings, one pyridine ring and one oxazoline ring (Fig. 1) (Colombo et al., 1995; Kettenring et al., 1991; Tavecchia et al., 1994). It is thus structurally related to other thiazolylpeptide antibiotics such as thiostrepton and nosiheptide, which also consist of highly modified peptides with several thiazole rings and which inhibit bacterial protein synthesis. GE2270, and the closely related molecules amythiamicin (Shimanaka et al., 1995) and GE37468 (Stella et al., 1995), act as selective inhibitors of bacterial elongation factor Tu (Anborgh & Parmeggiani, 1991; Heffron & Jurnak, 2000; Hogg et al., 2002; Landini et al., 1996), whereas thiostrepton and related molecules inhibit bacterial protein synthesis by binding to the L11 protein in the 23S ribosomal complex (Cameron et al., 2002; Porse et al., 1998).



View larger version (20K):
[in this window]
[in a new window]
 
Fig. 1. Chemical structure of the GE2270 skeleton and differences among the components of the antibiotic complex as produced in P. rosea fermentations.

 
GE2270 is produced by fermentation of Planobispora rosea ATCC 53773 (Selva et al., 1991), a strain belonging to one of the most rare genera of actinomycetes, as a complex of 10 structurally related compounds (Selva et al., 1995). Under most of the fermentation conditions tested, the strain produced preponderantly GE2270 component A (50–60 % of the total complex) and small quantities of other metabolites, differing mainly in the degree of methylation (Selva et al., 1991). GE2270 A (Fig. 1) is the component with the highest number of methyl groups (Kettenring et al., 1991; Tavecchia et al., 1994). Improvements in the analysis and structural elucidation of the minor components of the GE2270 complex (Colombo et al., 1995; Selva et al., 1995) has allowed the identification of de-methylated metabolites. The building blocks of the thiazolyl peptide skeleton and the sequence are conserved in all components, which differ in the substituents at position 5 in thiazole rings D and E, and on NH2 of asparagine (Fig. 1). Under standard fermentation conditions there is a significant production of components E (ca 15 %), D1 (ca 10 %) and D2 (ca 15 %), lower amounts of B1 and C1 and traces of C2 and B2. Component T differs from all the others by the oxidation of the oxazoline ring and it is produced in almost undetectable trace amounts (Selva et al., 1995).

In the biosynthesis of GE2270, as for other thiazolylpeptide antibiotics, the thiazole rings derive from one molecule of cysteine and the carboxy group of the adjacent amino acid, leading to the formation of unsubstituted rings; the pyridine ring is derived from the head to head condensation of two serine residues; and the oxazoline ring results from cyclization of serine (De Pietro et al., 2001; Mocek et al., 1993a, b). Labelling studies on GE2270 component A biosynthesis showed that the methyl groups at ring E (R2), the N-methyl group of asparagine (R3) and the methylene and O-methyl groups at ring D (R1) (Fig. 1) are derived from a cellular methyl donor (De Pietro et al., 2001). It was discovered that in P. rosea [2-13C]glycine and [3-13C]serine are effective sources of methylating equivalents; the results of isotopic enrichment are consistent with the existence of a pathway whereby glycine is converted to serine by methylation, and C-2 of glycine can generate methyl groups (De Pietro et al., 2001).

Vitamin B12 (cobalamin) is a coenzyme involved in carbon skeletal rearrangements and it acts as a cofactor for several enzymes by catalysing methyl transfer reactions, including the methionine synthase (Matthews, 2001). This enzyme transfers a methyl group from N5-methyl tetrahydrofolate to homocysteine as the final step in methionine synthesis. The methyl group of methionine may be derived from serine (or glycine) via N5,N10-methylenetetrahydrofolate, i.e. the folate branch of the methionine pathway (Blanco et al., 1998). Methionine, via its activated form, S-adenosyl-L-methionine, is considered as the general source of methyl groups in secondary metabolism (Roth et al., 1996). O-, N- and C-methyltransferases that use S-adenosyl-L-methionine as methyl donor are ubiquitous in the biosynthesis of secondary metabolites such as staurosporine (Yang et al., 1999), citreamicin (Pearce et al., 1991), avermectins (Schulman et al., 1986), erythromycin (Haydock et al., 1991), avilamycin (Weitnauer et al., 2002), daunomycin (Connors & Strohl, 1993), FK-506 (Shafiee et al., 1994), tylosin (Kreuzman et al., 1988; Seno & Baltz, 1981) and A10255 (Favret et al., 1992), the latter being a thiopeptide antibiotic belonging to the same class as GE2270. Much less is known about different methyltransferases involved in the biosynthesis of other antibiotics such as fortimicin, bialaphos and fosfomycin that seem to utilize methylcobalamin as a direct methyl donor (Kamigiri et al., 1992; Kuzuyama et al., 1992, 1995) and not as a prosthetic group. Different components of the above described methylation pathways – i.e. vitamin B12, commercially available as cyanocobalamin, methylcobalamin, cobalt ion, methionine and folic acid – were assayed in P. rosea fermentations to investigate their effect on GE2270 antibiotic complex production.

Among inhibitors of methylation reactions, sinefungin, a natural nucleoside structurally analogous to S-adenosyl-L-methionine, acts as a specific inhibitor of S-adenosyl-L-methionine-dependent methyltransferases (Barbes et al., 1990; Borchardt et al., 1979). Sinefungin was found to inhibit methyltransferases involved in the biosynthesis of secondary metabolites, causing the accumulation of de-methylated antibiotics, as seen for avermectins (Schulman et al., 1985, 1987), citreamicin (Pearce et al., 1991), tylosin (Kreuzman et al., 1988), pradimicins (Kakinuma et al., 1993; Saitoh et al., 1995) and staurosporine (Yang et al., 1999). S-Adenosyl-L-homocysteine, which is the common nucleoside reaction product, is often also employed to inhibit S-adenosyl-L-methionine-dependent methyltransferases in antibiotic biosynthesis (Bauer et al., 1988; Connors & Strohl, 1993; Kreuzman et al., 1988; Shafiee et al., 1994). Other methylation inhibitors are aminopterin (4-aminofolate), amethopterin (10-methylaminofolate), D-methionine and DL-ethionine. Aminopterin and amethopterin are structural analogues of folic acid and affect the biosynthesis of tetrahydrofolate derivatives, which act as coenzymes of one-carbon group transferases (Brown & Williamson, 1982). Sulfamethizole, as for the other sulfonamides, competes with p-aminobenzoate, which is the biosynthetic precursor of folic acid (Brown & Williamson, 1982). Aminopterin has been used to direct biosynthesis toward demethylated metabolites in the monensin producer Streptomyces cinnamoniensis (Pospisil & Zima, 1987) and citreamicin in Micromonospora citrea (Pearce et al., 1991). D-methionine and ethionine act as antimetabolites of L-methionine. The addition of DL-ethionine, D-methionine and aminopterin to the tetracycline producers yields a number of demethylated tetracycline precursors (Miller et al., 1964).

In this paper, we investigate the effect of coenzymes and inhibitors involved in the methylation pathway on the production of GE2270 antibiotic with the aim of better understanding its biosynthetic pathway and of modulating its complex composition towards the accumulation of desired metabolites.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Micro-organism.
Planobispora rosea GE2270/9-1 used in this study was isolated as a high-producing natural variant from the parental ATCC 53773 strain, which was described in a previous paper (Selva et al., 1991).

Flask fermentation.
Frozen cultures in D/Seed medium were maintained at -80 °C as a working cell bank and 5 ml was used to inoculate 500 ml Erlenmeyer flasks containing 100 ml medium C. The inoculated flasks were incubated for 48 h at 28 °C with shaking at 200 r.p.m. Then 5 ml aliquots were transferred to other shake flasks containing the same medium with added methylation inhibitors and/or vitamins. Fermentations were carried out for 7 days at 28 °C with shaking at 200 r.p.m. Biomass was measured as packed mycelium volume after centrifugation at 1200 g for 10 min in a 10 ml graduated conical glass tube. Production of the antibiotic and its complex composition were measured by HPLC as described below. Data represent the mean values of three replicas of each fermentation experiment and each experiment was repeated at least twice. Standard deviation and coefficient of variation (CV) were estimated.

Bioreactor fermentations.
Two millilitres of the working cell bank was inoculated into six 500 ml Erlenmeyer flasks containing 100 ml D/Seed medium. After 72 h at 28 °C with shaking at 200 r.p.m., the cultures were transferred to Chemap Braun CF 3000 fermenters, containing 14 l medium C. Fermentations were carried out at 28 °C, stirred at 900 r.p.m. with an aeration rate of 0·5 l air (l culture)-1 min-1. Biomass was measured as packed mycelium volume after centrifugation at 1200 g for 10 min in a 10 ml graduated conical glass tube. Production of the antibiotic and its complex composition were measured by HPLC as described below.

Media composition.
Vegetative medium D/Seed had the following composition (g l-1): soluble starch (Cerestar), 20; polypeptone (BBL), 5; yeast extract (Costantino), 3; meat extract (Costantino), 2; soybean meal (Cargill), 2; calcium carbonate (Carlo Erba), 1. Medium was prepared in distilled water and the pH was adjusted to 7·0 before sterilization at 121 °C for 20 min.

Fermentation medium C had the following composition (g l-1): glucose (Baker), 10; soluble starch (Cerestar), 35; hydrolysed casein (Costantino), 5; yeast extract (Costantino), 8; meat extract (Costantino), 3·5; soybean meal (Cargill), 3·5; calcium carbonate (Carlo Erba), 2. Medium was prepared in distilled water and the pH was adjusted to 7·2 before sterilization at 121 °C for 30 min. All media were supplemented with 0·03 % (v/v) Hodag AFM-5 (Prodotti Gianni) as antifoaming agent.

HPLC analysis of the GE2270 complex.
One volume of fermentation culture was mixed with 2 vols acetonitrile and shaken at room temperature for 20 min. After centrifugation at 15 000 g for 10 min, supernatant was collected and injected in an analytical HPLC apparatus (1082 Hewlett Packard) equipped with a UV detector at 254 nm. For the analysis of GE2270 component A (HPLC method 1), the Brownlee RP-18 column (4·6x220 mm, 5 µm) was eluted with a mobile phase composed of 20 mM NaH2PO4/ acetonitrile (9 : 1) as eluent A and 20 mM NaH2PO4/acetonitrile (3 : 7) as eluent B. A linear gradient from 45 to 75 % (v/v) eluent B in 20 min was run with a flow of 1·5 ml min-1.

To better resolve the various GE2270 metabolites (HPLC method 2), the Bakerbond C-8 column (4·6x220 mm, 5 µm) was eluted with a mobile phase composed of 2·5 g ammonium formate l-1/tetrahydrofuran/acetonitrile (80 : 10 : 10) as eluent A and 2·5 g ammonium formate l-1/tetrahydrofuran/acetonitrile (20 : 40 : 40) as eluent B. The separation was accomplished by a linear gradient from 20 to 35 % (v/v) eluent B in 20 min with a flow of 1·5 ml min-1.

In both methods, a solution of 1 mg pure GE2270 A ml-1 was used as external standard (STD). Relative production of each GE2270 metabolite (GE2270 m) was calculated as follows:

where C (GE2270 m) is the concentration of each GE2270 metabolite in the sample, 3 is the diluting factor from sample preparation, C (STD) is the concentration of GE2270 A in the standard solution, A (GE2270 m) is the area of each GE2270 metabolite peak in the sample and A (STD) is the area of GE2270 A in the standard solution.

GE2270 total complex production was calculated as the sum of the concentrations of all the GE2270 metabolites, identified as previously reported (Colombo et al., 1995; Selva et al., 1995).

All the HPLC chromatograms reported in this paper were recorded using the same value of attenuation.

Methylation inhibitors and cofactors.
Sinefungin, S-adenosyl-L-homocysteine, amethopterin (metotrexate), aminopterin, sulfamethizole, DL-ethionine, D-methionine, vitamin B12 (cyanocobalamin), methylcobalamin and folic acid were obtained from Sigma. Sinefungin, vitamin B12 and methylcobalamin were dissolved in water; amethopterin, aminopterin and folic acid in NaOH aqueous solution; DL-ethionine and D-methionine in acidified water; s-adenosyl-L-homocysteine in dimethylformamide and then in water. All the solutions were sterilized by filtration through 0·22 µm Millipore filters before addition to the cultures.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Effect of coenzymes involved in methylation reactions
Fig. 2 shows that the addition of increasing concentrations of vitamin B12 (cyanocobalamin) to P. rosea fermentations exerted two significant effects on GE2270 productivity. Total complex productivity was doubled by the addition of vitamin B12 in the range 0·001–0·01 µg ml-1. A further effect was the increase in GE2270 component A synthesis, whose relative abundance in the complex changed from about 60 % in the standard fermentation conditions to more than 90 %, following the addition of vitamin B12. HPLC profiles in Fig. 3(a) and (b) show the complex composition without and with 0·01 µg vitamin B12 ml-1. In normal fermentation conditions there was a significant production of less methylated factors, which were strongly reduced by the addition of vitamin B12. As shown in Table 1, equivalent molar additions of methylcobalamin to P. rosea fermentations gave an effect similar to cyanocobalamin in both antibiotic potency and component A selective biosynthesis. Since cobalt is a central component of cyanocobalamin and methylcobalamin, the effect of its supplementation on methylated metabolite biosynthesis in the GE2270 complex was tested. As shown in Table 2, supplementing the P. rosea cultures with cobalt chloride gave only a moderate enhancement of GE2270 A production and a relatively poor increase in total antibiotic production. Thus cobalt cannot replace the vitamin B12 effect, suggesting that under our conditions the biosynthesis of vitamin B12 is not significantly limited by cobalt concentration.



View larger version (19K):
[in this window]
[in a new window]
 
Fig. 2. Improvement of relative GE2270 complex productivity (diamonds) and percentage component A (squares) after the addition of vitamin B12. In standard fermentations without vitamin B12, total GE2270 productivity (150±10 µg ml-1) was set equal to 100 %, of which 60 % is represented by component A.

 


View larger version (11K):
[in this window]
[in a new window]
 
Fig. 3. HPLC profiles of extracts from cultures of P. rosea without (a) and with (b) the addition of 0·01 µg vitamin B12 ml-1. HPLC method 1 is described in Methods.

 

View this table:
[in this window]
[in a new window]
 
Table 1. Effect of vitamin B12 (cyanocobalamin) and methylcobalamin on the biosynthesis of the GE2270 complex and on the enrichment of component A

 

View this table:
[in this window]
[in a new window]
 
Table 2. Effect of cobalt chloride and folic acid on the biosynthesis of the GE2270 complex and on the enrichment of component A

 
Since GE2270 A is the metabolite with the highest number of methyl groups in the GE2270 complex, other components involved in the methylation pathway were added to P. rosea cultures. The effect of methionine, which is the general source of O-methyl and N-methyl groups in secondary metabolism, was investigated. Addition of methionine up to 250 µg ml-1 to P. rosea cultures had no detectable effect on the synthesis of component A and/or on complex composition. Addition of higher amounts caused a reduction in microbial growth: 1 mg methionine ml-1 caused an inhibition of 25 % in biomass production followed by a 50 % decrease in GE2270 productivity (data not shown). Even if it cannot be ruled out that P. rosea possesses a poorly efficient system to use exogenously supplied methionine, it is likely that exogenously supplied methionine exerts a feedback regulation on the generation of methylating equivalents. Addition of folic acid, which is involved in an earlier step of the methylation pathway, showed no detectable effects (Table 2), indicating that the amount of this co-factor is not limiting the flow of one-carbon units.

Effect of methylation inhibitors
Table 3 shows the effect of different methylation inhibitors added to P. rosea fermentations. Sinefungin, a structural analogue of S-adenosyl-L-methionine, markedly changed GE2270 complex composition, as shown by HPLC profiles of fermentation broths in Fig. 4(a) and (b). Concomitant improvements in the analysis (second HPLC method described in Methods) allowed better identification of the demethylated components. As shown in Fig. 4(a), in standard fermentation conditions there is a significant production of components E, D1 and D2, a lower amount of B1 and C1 and traces of C2 and B2 (see also Fig. 1 and Table 4). Component T, which differs from all the others by the oxidation of the oxazoline ring, is the only one eluting after component A and it is produced in almost undetectable trace amounts. As shown in Fig. 4(b), in the presence of sinefungin, component A was strongly reduced and component D2 became the main component in the antibiotic complex. Total complex productivity was only slightly reduced, but the ratio between D2, which differs from A only in the free hydroxyl group on the D ring (Fig. 1), to A rose from 0·24 to 5·51 (Table 3).


View this table:
[in this window]
[in a new window]
 
Table 3. Effect of methylation inhibitors on the biosynthesis of the GE2270 complex and on the enrichment of component D2

 


View larger version (14K):
[in this window]
[in a new window]
 
Fig. 4. HPLC profiles of extracts from cultures of P. rosea without (a) and with (b) the addition of 60 µg sinefungin ml-1. HPLC method 2 is described in Methods.

 

View this table:
[in this window]
[in a new window]
 
Table 4. Effect of sinefungin on the biosynthesis of different metabolites in the GE2270 complex and on the enrichment of component D2

 
The other methylation inhibitors tested, which affect tetrahydrofolate synthesis or act as L-methionine antimetabolites, did not cause dramatic changes in GE2270 production. Aminopterin and amethopterin both gave a slight increase in the D2/A ratio to about 0·5. Sulfamethizole reduced the productivity of GE2270 antibiotic without major modifications in D2/A ratio. D-methionine and DL-ethionine did not show any significant effect on GE2270 synthesis.

S-Adenosyl-L-homocysteine, which is the nucleoside product of S-adenosyl-L-methionine-dependent methylations, did not produce significant changes in GE2270 complex composition and total productivity (data not shown), when assayed at the same concentrations as sinefungin (see below and Table 4). Even if it is likely that higher concentrations of S-adenosyl-L-homocysteine inhibit the methyltransferase converting D2 into A component, these results point out a significantly different sensitivity of the enzyme to sinefungin and to S-adenosyl-L-homocysteine.

Table 4 shows a more detailed investigation of the effect of increasing amounts of sinefungin on the different measurable components of the GE2270 complex. A dose-dependence was found between the increased concentration of sinefungin and the selective production of component D2.

When sinefungin was added at 60 µg ml-1 (0·158 mM), the D2/A ratio was 24-fold enhanced in comparison to the fermentation in standard conditions: D2 composed more than 70 % of the complex, whereas A composed merely 10 %. Addition of higher concentrations of sinefungin than those reported in Table 4 interfered with growth and caused a reduction of GE2270 productivity (data not shown).

Other detectable factors in the GE2270 complex showed some fluctuations when sinefungin was supplemented, but, as shown in Table 4, none of them was significantly reduced, except for the disappearance of component B1. Component B1 bears the same methylated group at the D ring as does component A, but it is not N-methylated at the asparagine moiety (Fig. 1). A concomitant accumulation of the demethylated component E was not detected, suggesting that this methylation is not a bottleneck in the biosynthetic flux towards the synthesis of GE2270 component D2 and then A. However, we cannot exclude a chromatographic limit of the detection method, which hampered a more precise quantitative analysis and resolution of minor factors in the GE2270 complex.

These results, together with those of vitamin B12, indicate that methylation of the hydroxy group at ring D is a late methylation reaction in GE2270 biosynthesis and it is catalysed by an S-adenosyl-L-methionine methyltransferase.

Combined effect of vitamin B12 and sinefungin on D2 biosynthesis
As described above, addition of vitamin B12 to P. rosea favoured the production of component A, whereas sinefungin supplementation stopped conversion of component D2 to A. When vitamin B12 and sinefungin were simultaneously added to the fermentation broths, maximum volumetric productivity of D2 was almost doubled, as reported in Fig. 5. These data show that vitamin B12 has two quite different effects on GE2270 biosynthesis: one is on the late methylation reaction converting D2 into A, and the other is a stimulating action on antibiotic backbone synthesis.



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 5. Effect of vitamin B12 (B12) and sinefungin (Sin) on the biosynthesis of components A (white bars) and D2 (black bars) in the GE2270 antibiotic complex, in comparison with standard fermentation conditions (Ctrl).

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Studies on GE2270 component A biosynthesis have shown that in the molecule four carbons derived from one-carbon units are present. These are the methyl groups at ring E (R2), the N-methyl group of asparagine (R3) and the methylene and O-methyl groups at ring D (R1) (De Pietro et al., 2001). The specific action of catalytic amounts of vitamin B12 (0·01 µg ml-1) in directing the biosynthesis towards the methylation of the hydroxy group at ring D suggests that this reaction is catalysed by an S-adenosyl-L-methionine methyltransferase. A similar effect of vitamin B12 was previously reported for the methylated antibiotics fortimicin A (Yamamoto et al., 1977), coumermycin A1 (Claridge et al., 1966) and thiazolylpeptide A10255 B and E (Favret & Boeck, 1992). Incorporation studies with L-[methyl-13C]methionine demonstrated that methionine is the methyl donor for additional methyl groups in coumermycin A1 (Claridge et al., 1966), fortimicin A (Okumura et al., 1981; Yamamoto et al., 1977) and A10255 metabolites (Favret & Boeck, 1992; Favret et al., 1992). Nevertheless, when methionine was added to these fermentations, it did not significantly change the antibiotic production and complex composition as it occurs in GE2270 biosynthesis, suggesting that the intracellular pool of methionine is not altered by its exogenous supply. The evidence that cyanocobalamin and methylcobalamin exert the same stimulatory effect on GE2270 biosynthesis confirms that both of these forms of vitamin B12 act as prosthetic groups, and consequently excludes an involvement of methylcobalamin in directly supplying methyl groups, as in the case of fortimicin biosynthesis (Kuzuyama et al., 1995; Okumura et al., 1981). Results of our experiments with folic acid addition confirm that availability of one-carbon units is not a limiting factor for methylations.

In addition to the selective production of component A, the twofold increase in GE2270 total antibiotic yield achieved by the addition of vitamin B12 to the fermentation of P. rosea represents an innovative and industrially relevant aspect of this study (Rizzo & Gastaldo, 1999). In previous studies on production of methylated metabolites in fortimicin (Yamamoto et al., 1977), coumermycin (Claridge et al., 1966) and A10255 (Favret & Boeck, 1992) antibiotic complexes, no similar improvement of total productivity was stimulated by vitamin B12 or correlated molecules. However, vitamin B12 was previously identified as the compound needed to restore thienamycin production (Chen et al., 1993) in biosynthetic studies by blocked mutants of Streptomyces cattleya.

Confirmation of the type and the role of the methyltransferase involved in component A formation resulted from the use of methylation inhibitors. Sinefungin, a natural nucleoside structurally analogous to S-adenosyl-L-methionine, strongly enhanced the accumulation of component D2, which differs from component A only in the lack of a methyl group at ring D. Thus methylation of D2 into A is a late step in the GE2270 biosynthetic pathway and the responsible O-methyltransferase, being specifically inhibited by sinefungin, is an S-adenosyl-L-methionine methyltransferase. Mass balance of this reaction is in agreement with the proposed pathway because the amount of D2 metabolite that accumulated corresponded to the amount of A component lost. Sinefungin did not provoke other marked changes in complex composition, suggesting that different methyltransferases are involved in the addition of the other one-carbon units. It is interesting to note that component B1, which bears the same methoxy group at ring D as does factor A, but which is not N-methylated at the asparagine moiety, disappeared in the presence of sinefungin. A concomitant accumulation of component E was not detected, suggesting that this methylation is not a bottleneck in the biosynthetic flux towards the synthesis of GE2270 component D2 and then A.

All of the other methylation inhibitors screened failed to enhance production or give rise to new metabolites, nor was the biosynthesis of GE2270 blocked. A complete understanding of the effect of these inhibitors would require testing over a greater range of concentrations and generally a deeper investigation. For those inhibitors, such as structural analogues of folate and of p-aminobenzoate, or methionine antimetabolites, which showed little or no effect, our work cannot rule out other explanations for their partial or total inefficacy, such as the impermeability of P. rosea cells to some of them or toxicity for others. The different activity of methylation inhibitors on secondary metabolism has been reported (Miller et al., 1964; Pearce et al., 1991; Pospisil & Zima, 1987). In the case of staurosporine biosynthesis, among the different methyltransferase inhibitors assayed, sinefungin was the only one able to produce a demethyl derivative and it was shown to specifically inhibit the O-methyltransferase, but not the N-methyltransferase, even if both the O- and N-methyl groups of staurosporine derived from methionine (Yang et al., 1999). Sinefungin was also the most active and the least toxic methyltransferase inhibitor in citreamicin-producing fermentations, but it stopped only one of the two O-methyltransferases involved in the modification of two adjacent hydroxy groups (Pearce et al., 1991). Sinefungin was added to parental or mutant strains of Actinomadura spp. and Streptomyces avermitilis to understand biosynthesis and to produce new analogues of pradimicin and avermectin, respectively (Kakinuma et al., 1993; Saitoh et al., 1995; Schulman et al., 1985). Sinefungin differs from S-adenosyl-L-methionine and s-adenosyl-L-homocysteine in the side chain: an ornithine residue is linked to the 5' position of adenosine through a carbon–carbon bond (Barbes et al., 1990). Different susceptibilities to sinefungin and to S-adenosyl-L-homocysteine were previously reported for methyltransferases involved in primary and secondary metabolism (Barbes et al., 1990; Borchardt et al., 1979). In staurosporine production, 3'-O-methyltransferase was selectively inhibited by sinefungin and not by S-adenosyl-L-methionine (Yang et al., 1999). The two rate-limiting O-methyltransferases which catalyse the penultimate and the terminal steps in tylosin biosynthesis (Bauer et al., 1988; Kreuzman et al., 1988) are sensitive differentially to inhibition by adenine-containing compounds and particularly to sinefungin. Sinefungin is a potent inhibitor for macrocin O-methyltransferase, but a poor inhibitor for demethylmacrocin methyltransferase. The inhibition efficiency appears to be governed by the side chain of the nucleosides (Kreuzman et al., 1988).

One surprising result was the effect of the simultaneous addition of vitamin B12 and sinefungin to P. rosea cultures. Once the last methylation reaction converting D2 into A was selectively inhibited by sinefungin, it clearly appeared that vitamin B12 exerted a major stimulatory effect on earlier steps of GE2270 biosynthesis. Further studies are needed to understand how vitamin B12 acts in the synthesis of GE2270. Recent investigations (De Pietro et al., 2001) by adding labelled amino acid precursors have shown that peptide skeleton formation in GE2270 is very similar to that of other thiazolylpeptides (Mocek et al., 1993a, b). The assembly of the amino acids is achieved by the well known multienzyme complex peptidyl synthetase. Neither the assembly, nor the biosynthesis of the amino acids constituting the chain appears to involve reactions known to be dependent on cobalamins. However, the pattern of 13C-labelled precursor incorporation revealed the existence in P. rosea of an uncommon pathway by which serine, the precursor of the several cysteine units present in GE2270, originates from glycine by the addition of a one-carbon unit. This differs from the common pattern by which glycine originates from serine with simultaneous generation of methylating equivalents. Furthermore, the results reported by De Pietro and co-workers give clear evidence that all the one-carbon units participating in GE2270 formation derive from carbon 2 of glycine. This requires the cleavage of glycine possibly with the formation of CO2, NH+4 and 13C-methylenetetrahydrofolate, as proposed in the biosynthesis of A10255 by Streptomyces gardneri (Favret et al., 1992). The observation that the methylene group of glycine is an effective source of methylating equivalents has also been made during the biosynthesis of the thiopeptide sulfomycin by Streptomyces arginensis (Fate et al., 1996). It is tempting to suggest that the enhancing effect of vitamin B12 on GE2270 antibiotic production can be explained by assuming that cobalamins act as coenzymes in one of these poorly studied reactions.

From a more applied point of view, the work reported in this paper allowed the production of industrially valuable levels of two different GE2270 metabolites as starting points for the synthesis of interesting semi-synthetic derivatives (Lociuro et al., 1997). In particular, the GE2270 component D2 was used as a starting scaffold bearing a free hydroxy group, which represents a new ‘handle’ for chemical modifications.


   ACKNOWLEDGEMENTS
 
The authors thank Giancarlo Lancini for helpful discussions and support over the years.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Anborgh, P. H. & Parmeggiani, A. (1991). New antibiotic that acts specifically on the GTP-bound form of elongation factor Tu. EMBO J 10, 779–784.[Abstract]

Barbes, C., Sanchez, J., Yebra, M. J., Robert-Gero, M. & Hardisson, C. (1990). Effects of sinefungin and S-adenosylhomocysteine on DNA and protein methyltransferases from Streptomyces and other bacteria. FEMS Microbiol Lett 57, 239–243.[Medline]

Bauer, N. J., Kreuzman, A. J., Dotzlaf, J. E. & Yeh, W. K. (1988). Purification, characterization, and kinetic mechanism of S-adenosyl-L-methionine : macrocin O-methyltransferase from Streptomyces fradiae. J Biol Chem 263, 15619–15625.[Abstract/Free Full Text]

Blanco, J., Coque, J. J. & Martin, J. F. (1998). The folate branch of the methionine biosynthesis pathway in Streptomyces lividans: disruption of the 5,10-methylenetetrahydrofolate reductase gene leads to methionine auxotrophy. J Bacteriol 180, 1586–1591.[Abstract/Free Full Text]

Borchardt, R. T., Eiden, L. E., Wu, B. & Rutledge, C. O. (1979). Sinefungin, a potent inhibitor of S-adenosylmethionine : protein O-methyltransferase. Biochem Biophys Res Commun 89, 919–924.[Medline]

Brown, G. M. & Williamson, J. M. (1982). Biosynthesis of riboflavin, folic acid, thiamine, and pantothenic acid. Adv Enzymol Relat Areas Mol Biol 53, 345–381.[Medline]

Cameron, D. M., Thompson, J., March, P. E. & Dahlberg, A. E. (2002). Initiation factor IF2, thiostrepton and micrococcin prevent the binding of elongation factor G to the Escherichia coli ribosome. J Mol Biol 319, 27–35.[CrossRef][Medline]

Chen, T. S., Arison, B. H., Ruby, C. L., Dombrowski, A. W. & Inamine, E. S. (1993). A cofactor for thienamycin biosynthesis produced by a blocked mutant of Streptomyces cattleya. J Ind Microbiol 12, 66–67.[Medline]

Claridge, C. A., Rossomano, V. Z., Buono, N. S., Gourevitch, A. & Lein, J. (1966). Influence of cobalt on fermentative methylation. Appl Microbiol 14, 280–283.[Medline]

Colombo, L., Stella, S. & Selva, E. (1995). Contribution of mass spectrometry to the structural confirmation of components of the antibiotic GE2270 complex. Rapid Commun Mass Spectrom 9, 717–722.[Medline]

Connors, N. C. & Strohl, W. R. (1993). Partial purification and properties of carminomycin 4-O-methyltransferase from Streptomyces sp. strain C5. J Gen Microbiol 139, 1353–1362.[Medline]

De Pietro, M. T., Marazzi, A., Sosio, M., Donadio, S. & Lancini, G. (2001). Biosynthesis of the thiazolylpeptide antibiotic GE2270. J Antibiot (Tokyo) 54, 1066–1071.[Medline]

Fate, G. D., Benner, C. P., Grode, S. H. & Gilbertson, T. J. (1996). The biosynthesis of sulfomycin elucidated by isotopic labeling studies. J Am Chem Soc 118, 11363–11368.[CrossRef]

Favret, M. E. & Boeck, L. D. (1992). Effect of cobalt and cyanocobalamin on biosynthesis of A10255, a thiopeptide antibiotic complex. J Antibiot (Tokyo) 45, 1809–1811.[Medline]

Favret, M. E., Paschal, J. W., Elzey, T. K. & Boeck, L. D. (1992). Biosynthesis of thiopeptide antibiotic A10255: incorporation of isotopically-labeled precursors. J Antibiot (Tokyo) 45, 1499–1511.[Medline]

Goldstein, B. P., Berti, M., Ripamonti, F., Resconi, A., Scotti, R. & Denaro, M. (1993). In vitro antimicrobial activity of a new antibiotic, MDL 62,879 (GE2270 A). Antimicrob Agents Chemother 37, 741–745.[Abstract]

Haydock, S. F., Dowson, J. A., Dhillon, N., Roberts, G. A., Cortes, J. & Leadlay, P. F. (1991). Cloning and sequence analysis of genes involved in erythromycin biosynthesis in Saccharopolyspora erythraea: sequence similarities between EryG and a family of S-adenosylmethionine-dependent methyltransferases. Mol Gen Genet 230, 120–128.[Medline]

Heffron, S. E. & Jurnak, F. (2000). Structure of an EF-Tu complex with a thiazolyl peptide antibiotic determined at 2·35 Å resolution: atomic basis for GE2270A inhibition of EF-Tu. Biochemistry 39, 37–45.[CrossRef][Medline]

Hogg, T., Mesters, J. R. & Hilgenfeld, R. (2002). Inhibitory mechanisms of antibiotics targeting elongation factor Tu. Curr Protein Pept Sci 3, 121–131.[Medline]

Jabes, D., Cavaleri, M., Romanò, G., Mosconi, G., Bojar, R. & Cunliffe, W. (2002). Preclinical profile of BI-K0376: a new promising topical treatment for acne. In 20th World Congress of Dermatology, Paris, July 2002.

Kakinuma, S., Suzuki, K., Hatori, M., Saitoh, K., Hasegawa, T., Furumai, T. & Oki, T. (1993). Biosynthesis of the pradimicin family of antibiotics. III. Biosynthetic pathway of both pradimicins and benanomicins. J Antibiot (Tokyo) 46, 430–440.[Medline]

Kamigiri, K., Hidaka, T., Imai, S., Murakami, T. & Seto, H. (1992). Studies on the biosynthesis of bialaphos (SF-1293) 12. C–P bond formation mechanism of bialaphos: discovery of a P-methylation enzyme. J Antibiot (Tokyo) 45, 781–787.[Medline]

Kettenring, J., Colombo, L., Ferrari, P., Tavecchia, P., Nebuloni, M., Vekey, K., Gallo, G. G. & Selva, E. (1991). Antibiotic GE2270 A: a novel inhibitor of bacterial protein synthesis. II. Structure elucidation. J Antibiot (Tokyo) 44, 702–715.[Medline]

Kreuzman, A. J., Turner, J. R. & Yeh, W. K. (1988). Two distinctive O-methyltransferases catalyzing penultimate and terminal reactions of macrolide antibiotic (tylosin) biosynthesis. Substrate specificity, enzyme inhibition, and kinetic mechanism. J Biol Chem 263, 15626–15633.[Abstract/Free Full Text]

Kuzuyama, T., Hidaka, T., Kamigiri, K., Imai, S. & Seto, H. (1992). Studies on the biosynthesis of fosfomycin. 4. The biosynthetic origin of the methyl group of fosfomycin. J Antibiot (Tokyo) 45, 1812–1814.[Medline]

Kuzuyama, T., Seki, T., Dairi, T., Hidaka, T. & Seto, H. (1995). Nucleotide sequence of fortimicin KL1 methyltransferase gene isolated from Micromonospora olivasterospora, and comparison of its deduced amino acid sequence with those of methyltransferases involved in the biosynthesis of bialaphos and fosfomycin. J Antibiot (Tokyo) 48, 1191–1193.[Medline]

Landini, P., Soffientini, A., Monti, F., Lociuro, S., Marzorati, E. & Islam, K. (1996). Antibiotics MDL 62,879 and kirromycin bind to distinct and independent sites of elongation factor Tu (EF-Tu). Biochemistry 35, 15288–15294.[CrossRef][Medline]

Lociuro, S., Tavecchia, P., Marzorati, E., Landini, P., Goldstein, B. P., Denaro, M. & Ciabatti, R. (1997). Antimicrobial activities of chemically modified thiazolyl peptide antibiotic MDL 62,879 (GE2270A). J Antibiot (Tokyo) 50, 344–349.[Medline]

Matthews, R. G. (2001). Cobalamin-dependent methyltransferases. Acc Chem Res 34, 681–689.[CrossRef][Medline]

Miller, P. A., Saturnelli, A., Martin, J. H., Itscher, L. A. & Bohonos, N. (1964). A new family of tetracycline precursors. N-Demethylanhydrotetracyclines. Biochem Biophys Res Commun 16, 285–291.[Medline]

Mocek, U., Knaggs, A. R., Tsuchiya, R., Nguyen, T., Beala, J. M. & Floss, H. (1993a). Biosynthesis of the modified peptide antibiotics nosiheptide in Streptomyces actuosus. J Am Chem Soc 115, 7557–7568.

Mocek, U., Zeng, Z., O'Hagan, D., Zhou, P., Fan, L. D. G., Beala, J. M. & Floss, H. (1993b). Biosynthesis of the modified peptide antibiotic thiosptrepton in Streptomyces azureus and Streptomyces laurentii. J Am Chem Soc 115, 7992–8001.

National Institutes of Health (2001). NIAID Global Health Research Plan for HIV/AIDS, Malaria and Tuberculosis.

Okumura, S., Deguchi, T. & Marumo, H. (1981). Biosynthetic incorporation of methyl groups into fortimicins. J Antibiot (Tokyo) 34, 1360–1362.[Medline]

Pearce, C. J., Carter, G. T., Nietsche, J. A., Borders, D. B., Greenstein, M. & Maiese, W. M. (1991). The effect of methylation inhibitors on citreamicin biosynthesis in Micromonospora citrea. J Antibiot (Tokyo) 44, 1247–1251.[Medline]

Porse, B. T., Leviev, I., Mankin, A. S. & Garrett, R. A. (1998). The antibiotic thiostrepton inhibits a functional transition within protein L11 at the ribosomal GTPase centre. J Mol Biol 276, 391–404.[CrossRef][Medline]

Pospisil, S. & Zima, J. (1987). Biosynthesis of monensins and 3-O-demethylmonensins in Streptomyces cinnamonensis in the presence of methylation inhibitors. FEMS Microbiol Lett 44, 283–287.[CrossRef]

Rizzo, A. M. & Gastaldo, L. (1999). Process for the selective increase of production of antibiotic GE 2270 factor A by adding vitamin B12 to nutrient medium. US Patent no. 5 882 900.

Roth, J. R., Lawrence, J. G. & Bobik, T. A. (1996). Cobalamin (coenzyme B12): synthesis and biological significance. Annu Rev Microbiol 50, 137–181.[CrossRef][Medline]

Saitoh, K., Furumai, T., Oki, T., Nishida, F., Harada, K. & Suzuki, M. (1995). Pradimicin S, a new pradimicin analog. III. Application of the frit-FAB LC/MS technique to the elucidation of the pradimicin S biosynthetic pathway. J Antibiot (Tokyo) 48, 162–168.[Medline]

Schulman, M. D., Valentino, D., Hensens, O. D., Zink, D., Nallin, M., Kaplan, L. & Ostlind, D. A. (1985). Demethylavermectins. Biosynthesis, isolation and characterization. J Antibiot (Tokyo) 38, 1494–1498.[Medline]

Schulman, M. D., Valentino, D., Nallin, M. & Kaplan, L. (1986). Avermectin B2 O-methyltransferase activity in ‘Streptomyces avermitilis’ mutants that produce increased amounts of the avermectins. Antimicrob Agents Chemother 29, 620–624.[Medline]

Schulman, M. D., Valentino, D., Streicher, S. & Ruby, C. (1987).Streptomyces avermitilis’ mutants defective in methylation of avermectins. Antimicrob Agents Chemother 31, 744–747.[Medline]

Selva, E., Beretta, G., Montanini, N. & 8 other authors (1991). Antibiotic GE2270 A: a novel inhibitor of bacterial protein synthesis. I. Isolation and characterization. J Antibiot (Tokyo) 44, 693–701.[Medline]

Selva, E., Ferrari, P., Kurz, M. & 7 other authors (1995). Components of the GE2270 complex produced by Planobispora rosea ATCC 53773. J Antibiot (Tokyo) 48, 1039–1042.[Medline]

Selva, E., Montanini, N., Stella, S., Soffientini, A., Gastaldo, L. & Denaro, M. (1997). Targeted screening for elongation factor Tu binding antibiotics. J Antibiot (Tokyo) 50, 22–26.[Medline]

Seno, E. T. & Baltz, R. H. (1981). Properties of S-adenosyl-L-methionine : macrocin O-methyltransferase in extracts of Streptomyces fradiae strains which produce normal or elevated levels of tylosin and in mutants blocked in specific O-methylations. Antimicrob Agents Chemother 20, 370–377.[Medline]

Shafiee, A., Motamedi, H. & Chen, T. (1994). Enzymology of FK-506 biosynthesis. Purification and characterization of 31-O-desmethylFK-506 O : methyltransferase from Streptomyces sp. MA6858. Eur J Biochem 225, 755–764.[Abstract]

Shimanaka, K., Iinuma, H., Hamada, M., Ikeno, S., Tsuchiya, K. S., Arita, M. & Hori, M. (1995). Novel antibiotics, amythiamicins. IV. A mutation in the elongation factor Tu gene in a resistant mutant of B. subtilis. J Antibiot (Tokyo) 48, 182–184.[Medline]

Stella, S., Montanini, N., Le Monnier, F. & 8 other authors (1995). Antibiotic GE37468 A: a new inhibitor of bacterial protein synthesis. I. Isolation and characterization. J Antibiot (Tokyo) 48, 780–786.[Medline]

Tavecchia, P., Gentili, P., Kurz, M., Sottani, C., Bonfichi, R., Lociuro, S. & Selva, E. (1994). Revised structure of the antibiotic GE 2270A. J Antibiot (Tokyo) 47, 1564–1567.[Medline]

Weitnauer, G., Gaisser, S., Kellenberger, L., Leadlay, P. F. & Bechthold, A. (2002). Analysis of a C-methyltransferase gene (aviG1) involved in avilamycin biosynthesis in Streptomyces viridochromogenes Tu57 and complementation of a Saccharopolyspora erythraea eryBIII mutant by aviG1. Microbiology 148, 373–379.[Abstract/Free Full Text]

Yamamoto, M., Okachi, R., Kawamoto, I. & Nara, T. (1977). Fortimicin A production by Micromonospora olivoasterospora in a chemically defined medium. J Antibiot (Tokyo) 30, 1064–1072.[Medline]

Yang, S. W., Lin, L. J., Cordell, G. A., Wang, P. & Corley, D. G. (1999). O- and N-methylation in the biosynthesis of staurosporine. J Nat Prod 62, 1551–1553.[CrossRef][Medline]

Received 28 November 2002; revised 20 February 2003; accepted 3 March 2003.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Gastaldo, L.
Articles by Marinelli, F.
Articles citing this Article
PubMed
PubMed Citation
Articles by Gastaldo, L.
Articles by Marinelli, F.
Agricola
Articles by Gastaldo, L.
Articles by Marinelli, F.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2003 Society for General Microbiology.