INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

-Lactam biosynthesis in Acremonium chrysogenum (formerly known as Cephalosporium acremonium) begins with the nonribosomal condensation of the three precursor amino acids L--aminoadipic acid, L-cysteine, and L-valine (1, 21, 23). There are two ways to synthesize cysteine in filamentous fungi (35, 26). One pathway, the so-called autotrophic pathway, converts inorganic sulfur to cysteine via serine O-acetyltransferase and O-acetylserine sulfhydrilase (Fig. 1). In the second route, cysteine can be obtained via the reverse transsulfuration pathway, in which the sulfur atom of methionine is transferred to cysteine through S-adenosylmethionine, S-adenosylhomocysteine, homocysteine, and cystathionine as intermediates.

View larger version (18K): [in this window] [in a new window]   FIG. 1.   Cysteine biosynthesis in A. chrysogenum from sulfate (autotrophic pathway) and from methionine (reverse transsulfuration pathway). The mecB gene encoding cystathionine--lyase is shaded. This gene has been disrupted in transformants T6 and T24 (see the text). In the text, genes designated met belong to the methionine biosynthesis (direct transsulfuration) pathway and those named mec are involved in the reverse transsulfuration pathway. SAH, S-adenosylhomocysteine; SAM, S-adenosylmethionine.

The predominant route of the cysteine supply for -lactam biosynthesis depends on the producer microorganism (6). Cysteine for penicillin biosynthesis in Penicillium chrysogenum is obtained mainly from sulfate reduction (31). In this microorganism, the transsulfuration pathway has been considered to have a minor role in penicillin biosynthesis. On the other hand, the sulfur atom for -lactam biosynthesis by A. chrysogenum is believed to derive preferentially from methionine via the reverse transsulfuration pathway (3, 24). However, there is no genetic evidence for, or against, this hypothesis.

DL-Methionine is known to stimulate cephalosporin biosynthesis in A. chrysogenum (7, 16). For many years it was unclear whether the stimulatory effect of DL-methionine was due to a precursor effect (one providing cysteine) or to an inducing effect (8, 20, 23). Sawada and coworkers established that methionine increases isopenicillin N synthase, deacetoxycephalosporin C synthase (30), and -aminoadipyl-cysteinyl-valine synthetase activities (38) in Acremonium cultures. Later, Velasco et al. (35) showed that methionine induces the transcription of three of the four known cephalosporin biosynthetic genes.

Therefore, methionine has presumably a double effect on cephalosporin biosynthesis: (i) it could be the main supplier of cysteine via the reverse transsulfuration pathway and (ii) it has an induction effect on cephalosporin biosynthetic genes.

The mecB gene of A. chrysogenum has recently been cloned and characterized in our laboratory (19). This gene encodes a cystathionine--lyase activity which splits cystathionine into cysteine and -ketobutyrate. Several years ago it was proposed that cystathionine--lyase is required for cephalosporin biosynthesis (34). In order to study the influence of the transsulfuration pathway on cephalosporin biosynthesis, mecB was inactivated by a novel disruption technique in this fungus and the effect of this inactivation on cephalosporin production was studied. Results show unequivocally that the splitting of cystathionine is essential for high-level production of cephalosporin but that reverse transsulfuration is not the only pathway for cysteine formation for cephalosporin since the mecB-disrupted mutants were still able to produce cephalosporin.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Microbial strains. A. chrysogenum C10 (ATCC 48278), a strain producing high levels of cephalosporin C, released by Panlabs (6, 28), was used for mecB inactivation experiments. Aspergillus nidulans C47 (mecB cysB) and M63 (mecB) were obtained from A. Paszewski and used as controls to test for methionine auxotrophy following gene disruption. Escherichia coli ESS2231, a -lactam-supersensitive strain, was used for routine bioassays. E. coli DH5 competent cells were used in routine DNA manipulations.

Media and growth conditions for cephalosporin production. A. chrysogenum was grown in seed medium (32) for 48 h; 10 ml of seed culture was used to inoculate Shen's defined production (MDFA) medium (32) with or without supplementation with DL-methionine (3 g/liter). The cultures were incubated in triple-baffled flasks (500 ml; Bellco) containing 100 ml of medium at 25°C in a rotary shaker. Samples were taken every 24 h, and levels of cephalosporin antibiotics were determined by bioassays against E. coli ESS2231.

A. chrysogenum transformation and isolation of genomic DNA. Transformation of A. chrysogenum protoplasts was carried out as described previously (12). Transformants were selected in tryptic soy agar (Difco) with sucrose (10.3%), supplemented with hygromycin at 30 µg/ml. Genomic DNA of A. chrysogenum was isolated as described previously (12).

Site-directed mutagenesis. In vitro mutagenesis was performed with a Quickchange mutagenesis kit (Stratagene, La Jolla, Calif.) by following the manufacturer's instructions. Oligonucleotides used in the formation of an EcoRI site in the mecB gene were Ia (5'-CCTTATGTGCAGAATTCGCTCGACCTC) and Ib (5'-GAGGTCGAGCGAATTCTGCACATAAGG).

Southern blotting and hybridization. Three-microgram samples of genomic DNAs from A. chrysogenum C10 and its transformants were digested with EcoRI and separated in 0.8% agarose gel. The gel was blotted onto Hybond-N membrane (Amersham Pharmacia Biotech) as described by Sambrook et al. (29). Digoxigenin-alkaline phosphatase labeling, hybridization, and detection were done with a Genius kit (Boehringer Mannheim) according to the manufacturer's protocol. Hybridizations were performed at 68°C, and the blots were washed twice with 2× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate)-0.1% sodium dodecyl sulfate for 5 min at room temperature and twice with 0.1× SSC-0.1% sodium dodecyl sulfate for 15 min at 68°C.

Cell extracts and cystathionine--lyase activity assay. Frozen mycelia (500 mg) collected every 24 h were ground in a mortar with liquid nitrogen and resuspended in 100 mM sodium phosphate buffer (pH 7.3) containing 1 mM EDTA and 0.1 mM pyridoxal 5'-phosphate. Cell debris was removed by centrifugation at 15,000 × g for 20 min at 4°C. Cystathionine--lyase activity was assayed by determining the conversion of L-cystathionine into L-cysteine and -ketobutyrate. The reaction mixture consisted of 4 mM L-cystathionine, 5.5 × 10² mM pyridoxal 5'-phosphate, 7 mM EDTA, 2 mM dithiothreitol, and 0.1 mg of protein in a final volume of 0.5 ml. The reaction mixtures were incubated for 30 min at 30°C, and the reactions were stopped by the addition of 1 ml of Gaitonde's reagent and boiled for 5 min. The precipitated proteins were removed by centrifugation, and the amount of cysteine in 1 ml of supernatant was determined by using the acid ninhydrin assay. This assay is highly specific and gives essentially no reaction with cystathionine or methionine (10). Total protein in cell extracts was determined by the Bradford method with ovalbumin as a standard.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Gene disruption in A. chrysogenum by the double-marker technique. Transforming DNA is integrated into the genome of A. chrysogenum at a very low efficiency, mainly via heterologous recombination (14, 36). The frequency of gene disruption in A. chrysogenum is one event every 200 transformants screened (36). Since target inactivation of the mecB gene, which encodes cystathionine--lyase, would not generate any easily detectable phenotype (i.e., auxotrophy, resistance, color), we tried a technique to enrich the proportion of gene disruption events when inactivation is targeted to nonselectable genes (18). This technique, which has not been applied before to filamentous fungi, is based on the use of two selectable markers (Fig. 2). Marker A has two functions: (i) to inactivate the desired gene and (ii) to serve as the transformation marker. When double recombination takes place, the second marker (marker B) is lost. Thus, transformants are selected as marker A resistant and then screened for marker B sensitivity.

View larger version (20K): [in this window] [in a new window]   FIG. 2.   Scheme of the double-marker enrichment technique for targeted gene disruption. Marker A is the transformation marker. Marker B is the enrichment marker (see the text). Transformants having gene A disrupted were selected as marker A resistant and marker B sensitive.

View larger version (7K): [in this window] [in a new window]   FIG. 3.   Physical map of the pCGL::hph plasmid constructed for replacement of the mecB gene. The plasmid contains as marker A the hygromycin resistance cassette (hph) and as marker B the phleomycin resistance cassette (ble) (see the text for details of the resistance cassettes). B, BamHI; EV, EcoRV; S, SstI; E, EcoRI; H, HindIII.

Southern blot analysis of transformants sensitive to phleomycin and resistant to hygromycin. About 600 transformants resistant to hygromycin were selected by transformation of A. chrysogenum C10 with pCGL::hph and screened for resistance to phleomycin. Ninety-nine out of 600 transformants were sensitive to phleomycin and resistant to hygromycin. There are two possibilities to explain the phenotype of these phleomycin-sensitive, hygromycin-resistant transformants: (i) phleomycin resistance is lost because double recombination took place or (ii) the transforming DNA is integrated into the genome through the phleomycin resistance cassette, thus losing its function.

View larger version (40K): [in this window] [in a new window]   FIG. 4.   Replacement of mecB by an inactive copy using the double-marker technique and molecular analysis of the transformants. (A) Disruption of mecB with the pCGL::hph plasmid. The 11-kb EcoRI fragment in the genome is changed into two fragments of 9 and 7 kb B, BamHI; E, EcoRI. (B) Southern blot hybridization of EcoRI-digested genomic DNA (two Southern blots are shown) using as the probe the 7.7-kb BamHI fragment containing mecB. Lane 1, A. chrysogenum C10; lanes 2 to 24 transformants T2 to T24. The sizes of the hybridization bands are indicated on the right.

Transformants T6 and T24 with the disrupted mecB gene lack cystathionine--lyase activity. To study if transformants T6 and T24 had lost cystathionine--lyase activity (encoded by mecB), fermentation in defined production medium was performed (see Materials and Methods) using strains A. chrysogenum C10, the transformants T6 and T24 (with a disrupted mecB region), and transformant T99 (unmodified mecB region). Results showed (Fig. 5) a typical profile of a primary metabolism enzyme for A. chrysogenum C10; a very similar level of cystathionine--lyase activity was obtained with transformant T99. However, transformants T6 and T24 showed no detectable cystathionine--lyase activity. These results indicated that those transformants that suffered the canonical recombination process lost the mecB gene in the gene replacement process. These results confirmed that there is a single copy of the mecB gene in A. chrysogenum, as was proposed before on the basis of Southern hybridization evidence (19).

View larger version (45K): [in this window] [in a new window]   FIG. 5.   Specific cystathionine--lyase activity in A. chrysogenum C10 (), mutant T99 (), and mutants T6 and T24 (); the last two mutants showed no activity. (Inset) Growth of the T6 and T24 mutants in minimal Czapek medium. The parental strain, A. chrysogenum C10, is also a prototroph. Aspergillus nidulans C47 (mecB cysB), used as control, is unable to grow in minimal medium since it lacks both cysteine biosynthesis pathways, whereas Aspergillus nidulans M63 (mecB) behaves like the A. chrysogenum transformants T6 and T24.

MecB-inactivated mutants of A. chrysogenum are not auxotrophs. Transformants T6 and T24, deficient in cystathionine--lyase activity, lack the reverse transsulfuration pathway, but they are not auxotrophs, i.e., they are able to synthesize cysteine from sulfate and convert cysteine to methionine to meet the methionine requirement of the cell (Fig. 5, inset). The control Aspergillus nidulans strain C47 (mecB cysB), a double auxotroph defective in the autotrophic as well as in the reverse transsulfuration pathway, was unable to grow in minimal Czapek medium. These results indicate that cystathionine--lyase (catalyzing the splitting of cystathionine in the reverse transsulfuration pathway) is different from cystathionine--synthase, which catalyzes the formation of cystathionine from cysteine and O-acetylserine in direct transsulfuration (conversion of cysteine to methionine); mutants lacking cystathionine--synthase are known to be methionine auxotrophs, and this is not the case with transformants T6 and T24.

Inactivation of the mecB gene decreases cephalosporin production in methionine-supplemented cultures. The effect of inactivation of the mecB gene on cephalosporin biosynthesis was studied in cultures of the transformants T6 and T24 in triplicate flasks, and the same fermentation was repeated three times. Fermentations to evaluate the effect of inactivation of mecB were performed in MDFA medium with (3 g/liter) and without DL-methionine. Results showed that in medium without methionine (Fig. 6A to C), the growth behavior of the inactivated strains (T6 and T24) and their specific levels of cephalosporin production were similar to those of the control strain, A. chrysogenum C10. The lack of cystathionine--lyase activity (Fig. 6C) did not affect growth in MDFA medium.

View larger version (27K): [in this window] [in a new window]   FIG. 6.   Growth kinetics (A and D), cephalosporin production (B and E), and cystathionine--lyase activity (C and F) in A. chrysogenum C10 () and mutant T6 or T24 () in MDFA medium without methionine (left) and in MDFA medium supplemented with 3 g of DL-methionine per liter (right). Note the different scales in panels B and E. The increase in cephalosporin C (CPC) production (E) is due to the induction of cephalosporin biosynthesis by methionine. Results are the averages of three determinations from triplicate flasks. Vertical bars indicate the standard deviations from the averages using nine determinations.

DL-Methionine still induces cephalosporin gene expression in mecB-disrupted mutants. It is well known that exogenous DL-methionine induces the transcription of cephalosporin C biosynthetic genes (35). It was unclear, however, whether the methionine induction effect could be mediated by -ketobutyrate (a product of cystathionine splitting by cystathionine--lyase) or by the cystathionine--lyase protein itself acting as a regulatory protein. To determine if the methionine induction effect was still present in mecB-disrupted mutants lacking cystathionine--lyase activity, Northern blot analysis was performed using RNAs extracted from mycelia of A. chrysogenum C10 and mutant T6, which were grown for 48 h in MDFA without and with methionine (3 g/liter). Results showed (Fig. 7) that the induction effect on expression of the A. chrysogenum pcbC gene was still present in the strain lacking cystathionine--lyase activity. These results indicate that cystathionine--lyase is not involved in cephalosporin gene expression. The involvement of -ketobutyrate in cephalosporin induction cannot be ruled out since -ketobutyrate may still be formed in the cell by other catabolic pathways.

View larger version (22K): [in this window] [in a new window]   FIG. 7.   Methionine stimulation of transcription of the pcbC gene in the parental strain A. chrysogenum C10 and the A. chrysogenum T6 mutant, disrupted in mecB. Northern hybridizations were performed with a 0.8-kb NcoI/KpnI fragment containing the actin gene of Aspergillus nidulans and a 1.8-kb SalI/BamHI fragment that contains the pcbC gene of A. chrysogenum.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Gene disruption is a technique difficult to apply to filamentous fungi (9). However, there are significant differences among filamentous fungi. Some of them are able to efficiently integrate homologous DNA into the genome, as is the case with Aspergillus nidulans (33, 37), Ascobolus immersus (11), and Cochiobolus heterostrophus (15). On the other hand, with other filamentous fungi such as A. chrysogenum (14, 36), Podospora anserina (2), and Penicillium chrysogenum (5), the frequency of homologous recombination is very low (usually lower than 1 in 100 transformants).

Only two cases of targeted gene disruption have been described for A. chrysogenum. Hoskins et al. (14) inactivated the pcbAB gene by the single-crossover integration approach. However, this technique generates direct repeats in the genome and it has been shown to be highly unstable in the genomes of filamentous fungi (4). Waltz and Kück (36) studied the frequency of gene disruption by double crossover in A. chrysogenum and established that only in 0.5% of the transformants was the targeted gene inactive.

In this work, we have adapted the gene disruption enrichment method developed for mouse embryo-derived stem cells (18) based on the use of two selection markers. This technique resulted in obtaining 5% gene disruption (one event in every 20 transformants); thus, the expected gene disruption efficiency (0.5%) was increased 10-fold. The use of this enrichment method in our laboratory has resulted in very high efficiencies (30% of gene disruption events) for the inactivation of other A. chrysogenum genes (G. Liu, J. Casqueiro, and J. F. Martín, unpublished results).

In Saccharomyces cerevisiae and filamentous fungi, cysteine is synthesized by two pathways (25, 26) (Fig. 1). The L-cysteine molecule required for the biosynthesis of the ACV tripeptide in cephalosporin-producing A. chrysogenum strains was proposed to be synthesized mainly from methionine via the reverse transsulfuration pathway through homocysteine and cystathionine (Fig. 1) (23). Treichler et al. (34) at CIBA-Geigy suggested that cystathionine--lyase, the enzyme that splits cystathionine into cysteine and -ketobutyrate, is a key enzyme for cephalosporin biosynthesis. An A. chrysogenum mutant isolated by random mutagenesis and defective in cystathionine--lyase activity showed a drastic reduction in cephalosporin biosynthesis (34). However, this mutant might have contained a second mutation affecting cephalosporin production and, unfortunately, the CIBA-Geigy mutant is not available for further studies (J. Heim, personal communication). In this article, we have inactivated the mecB gene, which encodes cystathionine--lyase activity in A. chrysogenum (19). The inactivation of the mecB gene does not affect cephalosporin biosynthesis when the sulfur source is sulfate, i.e., cysteine for cephalosporin biosynthesis may be synthesized from either sulfate or methionine. Addition of methionine stimulates cephalosporin biosynthesis in the parental strain A. chrysogenum C10. But when mecB is inactivated (mutants T6 and T24), there is a clearly lower production of cephalosporin, indicating that the supply of cysteine from methionine via the reverse transsulfuration pathway is required for high-level production of cephalosporin. A correlation between the positive effect of methionine and cystathionine--lyase activity in A. chrysogenum C10 was observed (Fig. 6). The cystathionine--lyase activity is higher in the presence of methionine (as previously described by Zanca [38]), in agreement with a direct supply of cysteine from the reverse transsulfuration pathway. The growth of mutants T6 and T24 is retarded in methionine-supplemented medium, probably because in the presence of methionine the utilization of sulfate is depressed (17). Sulfate utilization is needed for cysteine biosynthesis in these mecB mutants because they cannot synthesize cysteine through the reverse transsulfuration pathway. Complementation tests of the mecB mutants would be desirable to confirm this hypothesis.

DL-Methionine greatly stimulates cephalosporin biosynthesis (16, 20) by inducing expression of the cephalosporin biosynthesis genes pcbAB, pcbC, and cefEF (30, 35). Disruption of the mecB gene did not affect the induction of pcbC by methionine, i.e., induction of the cephalosporin genes is not mediated by a putative regulatory mechanism exerted by the cystathionine--lyase protein. The induction mechanism may be triggered by methionine itself or by a catabolite derived from methionine, e.g., -ketobutyrate. -Ketobutyrate (a product of cystathionine splitting) cannot be formed in mecB mutants. However, the involvement of -ketobutyrate as an inducer in A. chrysogenum cannot be ruled out since it may still be formed by methionine-deaminases similar to those that occur in Aspergillus nidulans (R. E. Cardoza and J. F. Martín, unpublished results), although their presence in A. chrysogenum has not yet been demonstrated.

In conclusion, targeted inactivation of the mecB gene shows that methionine has two positive effects on cephalosporin biosynthesis; one of them, the supply of cysteine from the reverse transsulfuration pathway, required for high-level cephalosporin biosynthesis, is catalyzed by cystathionine--lyase.