Genetic Engineering of Acremonium chrysogenum, the Cephalosporin C Producer

1. Introduction

Acremonium chrysogenum, belongs to Filamentous fungi, is an important industrial microorganism. One of its metabolites, cephalosporin C (CPC), during fermentation is the major resource for production of 7-amino cephalosporanic acid (7-ACA), an important intermediate for the manufacture of many first-line anti-infectious cephalosporin-antibiotics, in industry.

Cephalosporins belong to the family of beta-lactam antibiotics. Comparing the first-discovered penicillin, cephalosporins have obvious advantages since they are more stable to penicillinase and are more effective to many penicillin-resistant strains. The incidence of adverse effects for cephalosporins is also lower than that for penicillins and other anti-infectious agents. Thus, cephalosporins are among the most-widely used anti-infectious drugs clinically. In China, the research on cephalosporins started from the 1960s, and cefoxitin was first developed in 1970. In the past 30 years, cephalosporin-antibiotics are one the most developed medicines on the domestic market. They accounts for more than 40% of the anti-infectious drug market share.

As the major resource for manufacturing 7-ACA, the production and cost of CPC is of the utmost importance in the cephalosporin-antibiotics market. The Ministry of Science and Technology of China has listed the fermentation of CPC as the major scientific and technical project in the past 30 years due to the continuous demand of strain improvement for the CPC-producing Acremonium chrysogenum.

Because of the limitation of traditional techniques on strain improvement for A. chrysogenum, along with the ubiquitous applications of molecular biology, genetic engineering has become a powerful tool to manipulate the antibiotic producing strain and to obtain a high-yielding mutant strain. This paper will summarize the most recent developments on genetic manipulation of A. chrysogenum.

2. Biosynthesis of CPC

The industrialization of CPC fermentation has been established tens of years ago with the breakthrough in key technologies including fermentation yield, fermentation regulation and preparation and purification. Nevertheless, there has been a lot of publications, recently on the improvement of CPC-producing strain by traditional methods, such as UV [1] or NTG [2]mutagenesis, and optimization of fermentation process [3], as well. However, most of the latest strain breeding techniques are at the molecular level, and the most important approach has been the research on the biosynthesis of the target metabolite.

The biosynthesis of CPC during the fermentation of A. chrysogenumhas been well investigated. There are two gene clusters on the chromosome that are involved in the biosynthesis of CPC. The “early” cluster consists of pcbAB-pcbCand cefD1-cefD2. The pcbAB-pcbCencode two enzymes responsible for the first two steps in CPC biosynthesis [4]. While the cefD1-cefD2encode proteins that epimerize isopenicillin N (IPN) to penicillin N [5]. The “late” cluster consists of cefEFand cefGgenes, which encode enzymes responsible for the last two steps [6].

The biosynthesis pathway of CPC is illustrated in figure 1. The ACV synthase, encoded by the pcbABgene, condenses 3 precursors L-α-aminoadipic acid, L-cysteine, L-valine to the ACV tripeptide. The ACV is then cyclized into IPN by IPN synthase encoded by pcbCgene. The step from IPN to penicillin N is catalyzed by a two-component epimerization system encoded by cefD1-cefD2. The cefEFencodes a unique bi-functional enzyme, deacetyloxy-cephalosporin C (DAOC) synthase-hydroxylase which successively transforms penicillin N into DAOC and deacetyl-cephalosporin C (DAC). The last step in CPC biosynthesis is catalyzed by a DAC-acetyltransferase (DAC-AT) which is encoded by cefG. The crystal structure of DAC-AT has been published [7]. It has been shown that DAC-AT belongs to α/β hydrolase family according to the formation of DAC-enzyme complex [7]. Among these, pcbAB, cefEFand cefGwere considered as the rate-limiting steps in CPC biosynthesis [8].

Figure 1.

The biosynthesis pathway of CPC

In recent years, some other regulatory proteins, which have been found to be important in CPC biosynthesis, as well as their coding genes have been discovered. For example, AcveA, a homologue of veAfrom Aspergillus, regulates the transcription of all 6 major CPC biosynthesis genes including pcbAB, pcbC, cefD1, cefD2, cefEFand cefG. Disruption of AcveAleads to a dramatic reduction of CPC yield.

A cefPgene located in the early cluster of CPC biosynthesis cluster has just been characterized. This gene encodes a transmembrane protein anchored in a peroxisome. It regulates the epimerization of IPN to penicillin N catalyzed by CefD1-CefD2 two-component enzyme complex in peroxisome. The cefPdisruptant accumulated IPN and lost CPC production [10]. To compensate for the disruption of cefP, both cefPand cefRneed to be introduced simultaneously. The CefR is the repressor of CefT, and stimulates the transcription of cefEF. A mutant A. chrysogenumwithout cefRshowed delayed transcription of cefEFand accumulation of penicillin N resulted in reduction of CPC yield [11].

A cefMgene was also found downstream of cefD1. Disruption of cefMaccumulates penicillin N with no CPC production at all [12]. It is suggested that CefM may be involved in the translocation of penicillin N from the peroxisome to the cytoplasm. Without cefM, cells are unable to transport penicillin N which gets epimerized in peroxisome into cytoplasm, from where CPC is synthesized.

3. Techniques for molecular breeding

Acremonium chrysogenumbelongs to the family of Filamentous fungi. The techniques for genetic breeding are somehow difficult to manipulate due to its complicated structure of the cell wall and the special life cycle. Our laboratory has started the molecular breeding of A. chrysogenumat a relatively early stage based on some published results from host, transformation, homologuous recombination and selectable marker of A. chrysogenum[13, 14].

To introduce exogenous DNA into A. chrysogenum, a traditional PEG-mediated protoplast transformation method is commonly used [15]. Since we are focusing on high-yield, or industrial strains, which usually have a stronger restriction-modification system than type strain, the traditional transformation method is not efficient enough for foreign gene introduction.

Agrobacterium tumefaciensmediated transformation has been widely used in plant genetic engineering, and in some of the Filamentous fungi including Penicillium chrysogenumand Aspergillus nidulansas well [17]. We have developed an adapted A. tumefaciensmediated transformation protocol for A. chrysogenum, which has a higher transformation efficiency than the PEG- mediated method [17], and more importantantly, this protocol can also be applied in A. chrysogenumhigh-yield strain. This is the first report of A. tumefaciensmediated A. chrysogenumtransformation in the world.

Considering the significant improvement after introduction of vgb, VHb protein coding gene, we use error-prone PCR together with DNA shuffling to artificially evolve the vgbgene in vitro. After primary and secondary screening, a higher active mutant protein was obtained. E. colibearing this mutant VHb produce 50% more biomass than its counterpart bearing the original VHb under limited oxygen environment [18].

A lot of basic research was done to facilitate the genomic DNA extraction [19] and endogenous promoter capture [20] from the chromosome of A. chrysogenum. A notable progress is the cloning of pcbAB-pcbCbi-directional promoter from the chromosome of A. chrysogenum[21]. This allows for the convenient manipulation of A. chrysogenumby introduction of multiple genes.

The last step in CPC biosynthesis, DAC transformed into CPC catalyzed by DAC acetyltransferase, was further investigated, as many reports have demonstrated that this is the rate-limiting step while DAC acetyltransferase coding gene, cefGhas a low transcription rate in vivo. Our study showed that recombinant expressed DAC acetyltransferase can transform DAC into CPC in vitroin the presence of acetyl CoA [22]. The enzymological and kinetic study of the recombinant DAC acetyltransferase help us better understand the catalytic mechanism of the enzyme and make it possible to improve its enzymatic activity in vivo[23].

4. Molecular breeding of Acremonium chrysogenum

Among the three rate-limiting enzymes, PcbAB is relatively difficult to manipulate due to its larger coding gene. Thus, researchers focus on cefEFand cefGfor molecular breeding of A. chrysogenum. Besides, extra copy numbers of cefTcould increase the yield of CPC in the mutant A. chrysogenum[24]. And, overexpression of cefPand cefRin A. chrysogenumcan decrease the accumulation of penicillin N and promote the yield of CPC by about 50% [11].

The fermentation process of A. chrysogenumis an extreme oxygen-consumption procedure. All the rate-limiting enzymes are oxygen-requiring enzymes. The VitreoscillaHemoglubin (VHb) is very attractive since it is capable of oxygen transmission in oxygen-limiting environments. A recombinant strain bearing VHb can significantly improve the usage of oxygen during the fermentation process and increase the product yield, which has been proven in Aspergillus[25]. Introduction of vgb, the coding gene for VHb, into A. chrysogenumcan also maintain a higher specific growth rate and specific production rate resulting in a 4-5 fold higher yield of the mutant strain [26]. Actually, there are many industrial A. chrysogenumstrains that express a recombinant vgb.

The earliest report on genetic modification for A. chrysogenumwas published in 1989, when researchers from Eli Lilly Co. introduced an extra copy of cefEF-cefGfragment into A. chrysogenumwhich resulted in a 15%-40% higher producing mutant strain [27]. This was the first evidence that molecular breeding could be a powerful tool in strain improvement of A. chrysogenum.

Although controlled by the same bi-directional promoter, the transcription levels of cefEFand cefGshowed a huge difference as shown by RT-PCR. The transcription of cefGis much lower than that of cefEF. This leads to the accumulation of DAC in the metabolites since they can not be efficiently transformed into CPC. As a matter of fact, CPC/DAC ratio is a quality control parameter in the industrial production of CPC fermentation. Thus, the introduction of extra copy numbers of cefGproduced an engineering strain whose CPC yield is 3 folds higher than the parental strain [28].

There is another report on the introduction of cefTinto A. chrysogenum,where the resulting mutant doubled the CPC yield [29]. This could be attributed to the enhancement of CefT, the efflux pump protein, so that the feedback inhibition in vivotriggered by the fermentation product was attenuated, resulting in a higher product yield.

Using molecular breeding technology, some CPC derivatives can be directly produced by engineering A. chrysogenumfermentation. For example, by disruption of cefEFand introduction of cefEoriginating from Streptomyces clavuligerus, a novel DAOC producing strain was obtained, which if followed by two enzymatic transformations, the industrially important 7-ADCA can be produced [30]. By introduction of the coding genes simultaneously into A. chrysogenumfor the two enzymes used the industrial production of 7-ACA by immobilized enzymatic transformation, the engineering strain can produce 7-ACA by fermentation [31].

Besides the introduction of exogenous genes, disruption and/or silencing of the endogenous genes is also a common strategy for genetic breeding of a certain strain. The recently developed RNA interference (RNAi) technique can be used as an alternative to silence the transcription of target genes instead of homologous recombination. RNAi in A. chrysogenumwas first published in 2007 [32]. The latest report was silencing of pcbCgene in Penicillium chrysogenumand cefEFgene in A. chrysogenumby RNAi [33]. These reports demonstrated the feasibility of RNAi technique in Filamentous fungi.

There is another interesting research for the molecular breeding of A. chrysogenumin a different idea. As we mentioned before, CefD1-CefD2 is a two-component enzyme complex that transforms IPN into penicillin N by an epimerization system located in the peroxisome. cefD1-cefD2block mutant lacking this epimerization system accumulated a large amount of IPN to more than 650μg/mL, almost the total relative CPC yield. With this mutant, the unstable IPN, which has never been purified before, could be now be purified by several steps using chromatography [34]. Characterization of its half-life and stability under a variety conditions can greatly help in the investigation of IPN.

It is worth noting that all of the above genetic breeding reports were on the background of an A. chrysogenumtype strain C10, whose CPC yield is only 1 mg/mL, far less than the industrial production level. Although some good achievements were obtained in improvement of A. chrysogenumfermentation and modification of metabolic products, those achievements are still far away from application in industry.

5. Industrialization research on molecular breeding of A. chrysogenum

Our research is focused on the molecular breeding of A. chrysogenumhigh-yield and/or industrial strains. We introduced different combinations of cefG/cefEF/cefT/vgbgenes into CPC high-producing strain and found that an extra copy of cefGhas a significant positive effect on CPC fermentation level. Since random integration occurring in A. chrysogenum, different transformants with cefGintroduction showed different elevated levels, with some at 100%. An extra copy of vgbgene also displayed a significant improvement up to 30% more of the CPC yield. Meanwhile, introduction of cefEFand cefThas no obvious effect on CPC production in the high-yield strain [35]. This revealed the apparent discrepancy between the genetic background of the type strain and the high-yield strain, and also suggested that endogenous cefEFand cefTmay already achieve high bioactivity after several rounds of mutagenesis breeding that a high-yield strain usually undertaken.

We then applied this achievement to a CPC industrial strain. Although we didn’t obtain a mutant that doubled the CPC yield, we did obtain an engineering strain whose CPC yield was increased by 20%, which has a promising industrialized potential.

We also tried the RNAi technique in the high-yield strain. A plasmid vector containing cefGdouble strain RNA transcription unit was constructed and transformed into high-yield A. chrysogenum. The cefGtranscription level in the transformants was measured by quantitative RT-PCR. Two mutant strains were found to have a decreasing cefGtranscription level of up to 80%. Their CPC yield was also found to decrease by 34.6% and 28.8%, respectively [36]. This result demonstrated the feasibility of RNAi application in high-yield A. chrysogenumand possible, industrial strain. Moreover, this is important for metabolic pathway reconstitution and novel CPC derivatives fermentation in A. chrysogenum.

The fermentation product of A. chrysogenum, CPC, is the major resource for industrial manufacturing of 7-ACA, the important intermediate of a large variety of cephalosporins antibiotics. A common producing route of 7-ACA is the chemical semi-biosynthesis. To date, the more environmental-friendly biotransformation has been widely used in industry. Although two step transformation dominates in the market [37], research on one step transformation from CPC to 7-ACA is still hot. However, the substrate specificity of CPC acylase still remains unsolved [38].

Whether two-step or one-step, fermentation of CPC is the prerequisite followed by enzymatic biotransformation in vitro. We are thinking of introducing CPC acylase gene into A. chrysogenumto construct the engineering strain that can produce 7-ACA directly by fermentation, a breakthrough in the production of 7-ACA.

A CPC acylase gene was designed according to the codon bias of A. chrysogenumand introduced into an industrial strain. Our result showed that this CPC acylase was expressed in A. chrysogenumwith bioactivity. The recombinant acylase can transform the original product CPC into 7-ACA in vivo, makes the engineering strain capable of direct fermentation of 7-ACA. Based on enzymological profiles of CPC acylase in vitro, we performed a preliminary optimization of medium composition and culture condition and the CPC yield was increased significantly with as least 30% of the CPC fermented being transformed into 7-ACA [40]. We believe this in vivoconversion can be more effective if a more powerful transcription cassette and more copy number can be introduced, with the incorporation of traditional breeding technology, and finally, bring this technique to industry.

6. Perspectives

As a novel tool for strain improvement, genome shuffling is of widespread concern in the field of industrial microbiology since it was first reported [41]. This has been applied in Bacteria and Streptomyces, and the yield of a lot of metabolites got a substantial increase by genome shuffling. However, genome shuffling in Filamentous fungi is rare, maybe due to the undeveloped genetic manipulation system. In 2009, the cellulase production in Penicillin decumbenswas reported to be increased by 40% with the help of genome shuffling [42]. But this achievement resulted largely in primary metabolites. As we all know, the regulation of secondary metabolites, as well as the genetic manipulation of A. chrysogenumis much more complicated. Since the exogenous genes were randomly integrated in the chromosome of A. chrysogenum, we suggest that genome shuffling can effectively improve the fermentation of the strains based on the established genetic techniques in our laboratory.

The biosynthesis of CPC in A. chrysogenumhas been investigated thoroughly. However, the mechanism of its regulation as well as the biosynthesis of precursors in primary metabolism is still unclear [43]. The full sequence of A. chrysogenumis yet to be completed, although there are more than 10 species belonging to the Filamentous fungi that have already been sequenced [44]. To better understand the genetic basis of A. chrysogenum, we realize that comparative proteomics could be used to study the molecular breeding without the genomic sequence of A. chrysogenum. By identifying those different expressed proteins during CPC fermentation, fermentation may be proposed based on the popular theory of metabolic engineering and system biology [45].

Besides its use in studying the mechanism of disease development, the application of comparative proteomics in antibiotic-producing microorganisms also showed promise. For instance, 345 different proteins were identified as critical during the conversion from primary to secondary metabolism in Streptomyces coelicolor[46]. Another example is research on Penicillin chrysogenumwhere 950 proteins involved in precursor biosynthesis, stress response and pentose phosphate pathway were found to be related to the fermentation yield in 3 penicillin-producing strains [47].

Thus, we believe that the molecular breeding of A. chrysogenumshould consist of genome shuffling, optimization of secondary metabolism, improvement of precursor biosynthesis and energy metabolism as well. Although there are still big effects need to be put in the basic and practical research of A. chrysogenum, the molecular bred engineering strains will play an important role in the industrial production of CPC and its derivatives.