RU2307169C1 - Method for investigation and prognosis of methyonin biosynthesis rout in related corynebacteria genomes - Google Patents

Abstract

FIELD: bioinformatics, in particular pharmaceutical industry and biothectnology.

SUBSTANCE: of methyonon biosynthesis rout in related corynebacteria genomes is forecasted on the base of screening of possible binding sites of McbR regulone among predicted and suggested regulatory regions before genes of enzymes, participated in methyonon biosynthesis and orthologs of methyonon biosynthesis rout in corynebacteria genomes, as well as determination of operone composition. Obtained data are represented in form of table of gene compatibility to possible binding sites of McbR regulone and related corynebacteria genomes and diagram of McbR regulone operone structure. Participates and courses pf methyonon biosynthesis rout are forecasted by position comparison of content of said table and diagram.

EFFECT: improved method for prognosis of methyonin biosynthesis rout in related corynebacteria; large-scale methyonin production of decreased cost.

4 cl, 7 dwg

Description

The invention relates to computer systems using genetic models in the development of new compounds for pharmaceuticals, biotechnology and other industries, for which the use of new protein compounds based on amino acids is essential.

Methionine is one of the essential amino acids and, in addition, the first amino acid in most protein molecules, so much attention is paid to the possibility of its synthesis on an industrial scale, which, in turn, requires a detailed study of the methionine biosynthesis pathway.

It is known that some mutants of Escherichia coli that are resistant to methionine analogues can produce methionine.

It was found that mutants of Escherichia coli, resistant to ethionine and having lost the ability to inhibit the final product, are able to produce methionine (Abelberg E.A., J. Bacteriol., 76, 326-328 (1958)).

The production of methionine by Escherichia coli mutants resistant to norleucine has also been described (Rowbury R. J., Nature, 206, 962 (1965)).

But only trace amounts of methionine were obtained.

A microorganism is described, obtained by introducing into a recipient strain of Escherichia a hybrid plasmid containing a DNA fragment isolated from the donor strain of Escherichia resistant to ethionine (UK patent GB2075055). A variation type of the metJ gene is also described with a specific sequence encoding the methionine biosynthesis-related repressor protein from Escherichia coli, in which repression activity is reduced and which is useful in methionine production (JP 157267 A2). But the yields in the production of methionine by these mutants (about 0.3-0.5 g / l) cannot satisfy the growing demand for this amino acid.

A known method of producing methionine, in which it is obtained by growing the bacteria Escherichia coli, with the ability to produce at least 0.5 g / l methionine in the process of growing on minimal medium and resistant to the methionine analogue. These properties are possessed by the strain Escherichia coli 218 (VKPM B-8125). The accumulated amino acid is isolated from the culture fluid. This invention allows to obtain methionine with a relatively high degree of efficiency (RU 2209248, 2001). However, this method cannot satisfy the needs of industry in methionine.

All of the above methods for producing methionine, in addition to low productivity, are time consuming and expensive.

In this regard, the search for genetic models and the prediction of new ways of producing methionine using accumulated experience, information technology and comparative genomics is an urgent need.

The species Corynebacterium glutamicum is represented by free-living soil bacteria that are used in the industrial production of a large number of amino acids, mainly lysine, arginine, threonine, isoleucine, valine, serine, tryptophan, phenylalanine and histidine [1], but not methionine, despite the common pathways biosynthesis of lysine and threonine with methionine biosynthesis. In recent years, the biosynthesis of methionine in C. glutamicum has been studied in detail in order to use this bacterium as an industrial producer of methionine.

The structural genes of the genetic model of the methionine synthesis pathway can be divided into two parts - the genes of the pathway for incorporating sulfur into the carbon skeleton and the biosynthesis pathway of methionine directly (see Fig. 1 - biosynthesis scheme). The main pathway for sulfur assimilation is the pathway for reducing sulfate to sulfide. Sulfur, which is part of the sulfide, then enters the biosynthesis pathway of methionine either as part of cysteine or as part of cystathionine, depending on the microorganism [2]. The sulfate reduction enzyme genes form a single operon in the C. glutamicum genome (Fig. 4, position A, C. glutamicum). In addition to sulfate, other compounds, such as sulfate esters and sulfonates, whose utilization requires the presence of appropriate enzymes, can also serve as a source of sulfur for soil bacteria. They are also organized into single operons.

For genes of the methionine synthesis model itself, unification into large operons is not characteristic. The methionine biosynthesis proceeds from aspartate, followed by the formation of homoserine, which, in turn, is converted either to acetyl homoserine in Neurospora crassa, Saccharomyces cerevisiae and C. glutamicum, or to succinyl homoserine in Escherichia coli and Pseudomonas aeruginosa. Then there is the formation of homocysteine, which can occur in two ways: along the direct sulfhydrylation pathway (C. glutamicum, Pseudomonas aeruginosa and Saccharomyces cerevisiae) or along the trans sulfurylation pathway with the formation of an intermediate product - cystathionine, this pathway is characteristic of Escherichia coli, Neurospora crassa .glutamicum [2; 3]. Thus, unlike most microorganisms that use one of the pathways for the synthesis of homocysteine, C. glutamicum is characterized by the presence of both fully functional pathways of the biosynthesis model [3] (see Fig. 1).

Genes for methionine biosynthesis in the C. glutamicum genome are regulated by at least two regulators. One of them is an indirect regulator encoded by the cg3031 gene. The corresponding cg3031p protein inhibits the methionine biosynthesis, but is not capable of binding to DNA [4].

Known for Corynebacterium glutamicum [5; 6] a transcriptional regulator of genes for the methionine biosynthesis pathway, mcbR (methionine and cysteine biosynthesis represser gene) and the putative operon structure of the mcbR regulon.

This is a dimeric transcriptional repressor belonging to the TetR family that binds to the palindromic sequence TAGAC-N6-GTCTA [6] in the regulatory regions of the metabolic pathway genes. Most of the binding sites of this regulator in C. glutamicum were predicted by aligning the regulatory regions of the target genes and the prediction was experimentally verified and confirmed for five regulated genes, such as hom, cysI, cysK, metK and mcbR [6]. In general, in [6], the operon organization of genes of the methionine biosynthesis pathway was predicted (see Fig. 4, C. glutamicim) and potential binding sites of the transcriptional repressor were presented in front of most operons in the C. glutamicum genome (see Fig. 4, C. glutamicum is the operon structure from [6] and the corresponding sites).

Despite the fact that the biosynthesis of methionine has been studied in detail in most taxonomic groups of bacteria, very little is known about this metabolic pathway for actinomycetes.

The objective of the invention is to expand the functionality of the forecasting method, increase its efficiency, and also, in the future, reduce the cost of industrial production of methionine.

This problem is solved in the method of research and prediction of the pathway of methionine biosynthesis in related genomes of corynebacteria, including the stages in which:

a) compile the initial training sample based on previously predicted and experimentally confirmed regulatory signals before the enzymes involved in methionine biosynthesis, and record it on an information carrier;

b) enter into the computer the source training sample recorded on the information carrier and process it with the SignalX program written in the computer memory in advance to construct the matrix of positional nucleotide weights;

c) the initial training set is processed using algorithms previously stored in the computer’s memory from the Genome Explorer software package, respectively, to search for potential McbR regulon binding sites and to analyze protein orthologs, while the search for potential binding sites is limited to the range of 400 ... + 100 nucleotides relative to the start gene translation, and the gene itself is considered to belong to the operon in the case of the same reading direction and the distance of not more than 100 pairs of nucleotides from other genes of this operon and record the result you are processing in the computer memory;

d) they search for orthologs of proteins of the methionine biosynthesis pathway in all related genomes, and the results of the search and processing for c) are divided into two output text files, the first of which is displayed on an information medium in the form of a table of correspondence of the genes of the methionine biosynthesis pathway with potential McbR regulatory binding sites and related genomes of corynebacteria, and the second in the form of a diagram of the structure of operons of the McbR regulon;

e) based on the positional comparison of the contents of both output text files, taking into account the search results for regulatory binding sites in more phylogenetically distant genomes for the first output text file and taking into account the characteristic evolutionary events of the formation of the structure of operons for the second output text file, the direction of the methionine biosynthesis path is predicted.

In addition, non-orthologic substitutions, operon rearrangements, paralogs and duplications, loss of orthologs, or complete conservatism in the structure of the operon are recorded as evolutionary events.

In addition, when fixing the presence of paralogs and duplications, evolutionary trees are used, the construction of which is carried out by the maximum likelihood method using the program from the PHYLIP 3.63 software package.

In addition, when compiling the initial training sample, complete sequences of actinobacteria genomes are formed on the basis of the Gen Bank database, homologs are searched in the database using the BLAST program, and multiple alignments of amino acids and nucleotide sequences are constructed using the algorithms of the ClustalX program.

The proposed method was applied to the mcbR regulon in all genomes where this regulator is present — these are seven genomes of the Actinomycetales order.

In figure 1. a genetic model of methionine biosynthesis scheme is presented; (alternative) direct sulfhydrylation and trans-sulfurylation pathways used by various microorganisms (abbreviations of genome names: Cgl - C.glutamicum, Cdp - C. diphtheriae, Cef - C.effiiciens, Eco - Escherichia coli, Pae - Pseudomonas aeruginosa);

figure 2 shows the evolutionary tree McbR, highlighted the branch on which McbR C.glutamicum and its orthologs are located;

figure 3 shows the output of the first text file in the form of a table of potential McbR binding sites in front of the genes of the C. glutamicum methionine biosynthesis pathway and their orthologs in related genomes, the names of the genes and orthologs correspond to the names of the genes in the C. glutamicum genome except for the Dip0611 C. diputriae genes , which lacks an ortholog in the C. glutamicum genome and an ortholog of the C.efficiens CE2636 gene in C. diphtheriae;

figure 4. the output of the second text file is presented in the form of the structure of five operons of the McbR regulon that underwent the largest evolutionary rearrangements, trivial gene names are given for C. glutamicum, for orthologs with different names in GenBank, gene names are shown in brackets above the corresponding gene, orthologous genes have the same contrast (genome abbreviations: Cgl - C.glutamicum, Cdp - C. diphtheriae, Cef - C.effiiciens, Cjk - C.jeikeium, Nfa - Nocardia faricnica, Lxl - Leifsonia xyli and Scc - Streptomyces coelicolor);

figure 5 shows the evolutionary tree Fpr, highlighted the branch on which Fpr C.jeikeium and its orthologs are located;

figure 6 shows the evolutionary tree CysN, the branches of the orthologs of the proteins CysN and CysN2, the protein Nfa33920 and its closest homologues from the group of alpha proteobacteria are highlighted;

7 shows the evolutionary tree of CysD. Branches of orthologs of the CysD and CysD2 proteins, the Nfa33930 protein, and its closest homologues from the gamma-proteobacteria group were identified.

The method is as follows.

Definition of members of regulon

To search for potential mcbR binding sites, the method of matrices of positional nucleotide weights was used [Mironov A.A., Koonin E.V., Roytberg M.A., Gelfand M.S. 1999. Computer analysis of transcription regulatory patterns in completely sequenced bacterial genomes. Nucleic Acid Res. 27.2981-2989]. The procedure for constructing such a matrix for searching mcbR sites is described in detail below.

To determine whether a gene belongs to the regulon, a conformance check method was used. According to this method, a gene is considered a member of the regulon if a potential site is found in its regulatory region, which is stored before the orthologous genes in related genomes (ibid.).

Search for potential sites was carried out in the region of -400 ... + 100 pairs of nucleotides relative to the start of gene translation. In the case when the operon structure of this fragment of the sequence is unknown, the genes were considered to belong to the same operon if they had the same reading direction and the distance between them did not exceed 100 nucleotide pairs

Software

To search for potential mcbR binding sites, as well as the analysis of orthologs, the GenomeExplorer software package was used [A. Mironov, N. Vinokurova, M. S. Gelfand 2000. Bacterial Genome Analysis Software. Like Biol. 34, 253-262]. The SignalX program [ibid] was used to construct positional weight matrices. The homologs in the databases were searched using the BLAST program [Altschul S.F., Madden T.L., Schaffer A.A., Zhang J., Zhang Z M.W., Lipman D.J. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acid Res. 25, 3389-3402]. The ClustalX program [Thompson J.D., Gibson T.J., Plewniak F., Jeanmougin F., Higgins D.G. was used to construct multiple alignments of amino acid and nucleotide sequences. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acid Res. 25, 4876-4882]. Evolutionary trees were constructed using the maximum likelihood method (Felsenstein J. 1996. Inferring phylogenies from protein sequences by parsimony, distance and likelihood methods. Methods Enzymol. 266, 418-427] using the proml program from the PHYLIP 3.63 software package (http://evolution.genetics.washington.edu/).

A total of 7 actinobacterial genomes were examined. Complete genome sequences:

Corynebacterium glutamicum ATCC 13032 [Kalinowski J., Bathe B., Bartels D., Bischoff N., Bott M., Burkovski A., Dusch N., Eggeling L., Eikmanns BJ, Gaigalat L., Goesmann A., Hartmann M ., Huthmacher K., Kramer R., Linke B., McHardy AC, Meyer F., Mockel B., Pfefferle W., Puhler A., Rey DA, Ruckert C., Rupp O., Sahm H., Wendisch VF , Wiegrabe I., Tauch A. 2003. The complete Corynebacterium glutamicum ATCC 13032 genome sequence and its impact on the production of L-aspartate-derived amino acids and vitamins. J. Biotechnol. 104, 5-25],

Corynebacterium diphtheriae NCTC 13129 [Cerdeno-Tarraga AM, Efstratiou A., Dover LG, Holden MT, Pallen M., Bentley SD, Besra GS, Churcher C., James KD, De Zoysa A., Chillingworth T., Cronin A., Dowd L., Feltwell T., Hamlin N., Holroyd S., Jagels K., Moule S., Quail MA, Rabbinowitsch E., Rutherford KM, Thomson NR, Unwin L., Whitehead S., Barrell BG, Parkhill J 2003. The complete genome sequence and analysis of Corynebacterium diphtheriae NCTC13129. Nucleic Acids Res. 31, 6516-6523],

Corynebacterium efficiens YS-314 [Nishio Y., Nakamura Y., Kawarabayasi Y., Usuda Y., Kimura E., Sugimoto S., Matsui K., Yamagishi A., Kikuchi H., Ikeo K., Gojobori T. 2003 Comparative complete genome sequence analysis of the amino acid replacements responsible for the thermostability of Corynebacterium efficiens. Genome Res. 13, 1572-1579],

Corynebacterium jeikeium K411 [Tauch A., Kaiser O., Hain T., Goesmann A., Weisshaar B., Albersmeier A., Bekel T., Bischoff N., Brune I., Chakraborty T., Kalinowski J., Meyer F ., Rupp O., Schneiker S., Viehoever P., Puhler A. 2005. Complete genome sequence and analysis of the multiresistant nosocomial pathogen Corynebacterium jeikeium K411, a lipid-requiring bacterium of the human skin flora. J. Bacteriol. 187, 4671-4682],

Nocardia farcinica IFM 10152 [Ishikawa J., Yamashita A., Mikami Y., Hoshino Y., Kurita H., Hotta K., Shiba T., Hattori M. 2004. The complete genomic sequence of Nocardia farcinica IFM 10152. Proc. Natl. Acad. Sci. USA 101, 14925-14930],

Leifsonia xyli subsp. xyli str. CTCB07 [Monteiro-Vitorello CB, Camargo LE, Van Sluys MA, Kitajima JP, Truffi D., do Amaral AM, Harakava R., de Oliveira JC, Wood D., de Oliveira MC, Miyaki C., Takita MA, da Silva AC, Furlan LR, Carraro DM, Camarotte G., Almeida NF Jr., Carrer H., Coutinho LL, El-Dorry HA, Ferro MI, Gagliardi PR, Giglioti E., Goldman MH, Goldman GH, Kimura ET, Ferro ES, Kuramae EE, Lemos EG, Lemos MV, Mauro SM, Machado MA, Marino CL, Menck CF, Nunes LR, Oliveira RC, Pereira GG, Siqueira W., de Souza AA, Tsai SM, Zanca AS, Simpson AJ, Brumbley SM, Setubal JC 2004. The genome sequence of the gram-positive sugarcane pathogen Leifsonia xyli subsp. xyli. Mol. Plant. Microbe. Interact 17, 827-836] and

Streptomyces coelicolor A3 (2) [Redenbach M., Kieser H.M., Denapaite D., Eichner A., Cullum J., Kinashi H., Hopwood D.A. 1996. A set of ordered cosmids and a detailed genetic and physical map for the 8 Mb Streptomyces coelicolor A3 (2) chromosome. Mol. Microbiol. 21, 77-96, as well as Bentley SD, Chater KF, Cerdeno-Tarraga AM, Challis GL, Thomson NR, James KD, Harris DE, Quail MA, Kieser H., Harper D., Bateman A., Brown S., Chandra G., Chen CW, Collins M., Cronin A., Fraser A., Goble A., Hidalgo J., Homsby T., Howarth S., Huang CH, Kieser T., Larke L., Murphy L., Oliver K., O'Neil S., Rabbinowitsch E., Rajandream MA, Rutherford K., Rutter S., Seeger K., Saunders D., Sharp S., Squares R., Squares S., Taylor K., Warren T., Wietzorrek A., Woodward J., Barrell BG, Parkhill J., Hopwood DA 2002. Complete genome sequence of the model actinomycete Streptomyces coelicolor A3 (2). Nature. 417, 141-147]

were taken from the GenBank database [Benson D.A., Boguski M.S., Lipman D.J., Ostell J., Ouellette B.F., Rapp B.A., Wheeler D.L. 1999. GenBank. Nucleic Acids Res. 27, 12-17].

The evolutionary tree of the mcbR regulator (Fig. 2) indicates that this regulator is preserved in the genomes of the genus Corynebacterium (C. glutamicum, C. diphtheriae, C.efficiens and C.jeikeium), as well as in other genomes of the Actinomycetales order, such as Nocardia farcinica, Leifsonia xyli and Streptomyces coelicolor. Therefore, the steps of the method were applied only to these genomes.

Positional comparison

Genes of the metabolic pathway and their organization into operons have already been shown for C. glutamicum [6]. When implementing the proposed method, a search was made for protein orthologs of the pathway of methionine biosynthesis in related genomes and the prediction of their organization into operons (see Fig. 4). An analysis of the predicted operon structures made it possible to identify several types of evolutionary events:

1. Non-orthologic substitutions. C.jeikeium, Nocardia farcinica, Leifsonia xyli and Streptomyces coelicolor use threonine synthase (thrC), which is typical for most actinomycetes. At the same time, threonine synthase is part of the operon with the genes homoserine dehydrogenase (hom or thrA) and homoserine kinase thrB. The sequence of genes in the operon can vary, so in C.jeikeium and Nocardia farcinica the operon looks like hom-thrB-thrC, and in Leifsonia xyli and Streptomyces coelicolor-like hom-thrC-thrB (Fig. 4, position D).

At the same time, in the genomes of C. glutamicum, C. diphtheriae and C.efficiens, the function of threonine synthase is performed by a protein that is not orthologous to thrC actinomycete. Its mapping and functionality has been tested experimentally, although the origin of this protein is not obvious. Characteristically, this threonine synthase is no longer part of the operon and thus two independent operons are formed: hom-thrB and unregulated monocistronic operon thrC, located in different parts of the genome (Fig. 4, position D).

For soil bacteria, sulfate esters and sulfonates are one of the main sources of sulfur. Sulphate esters are utilized through sulfatases, which are present in almost every microorganism. Nevertheless, the C. glutamicum genome for assimilation of sulfate esters contains a separate seuABC operon encoding three potential desulfurization enzymes [6], which also includes ssuD2 (see below). The disposal of sulfonates requires the presence of specific enzymes. In all the cases described, the sulfonate utilization genes form one operon, which includes the ABC transporter and NAD (P) H-dependent monooxygenase. For C. glutamicum, this is the ssuABC-D1 operon (ssu - sulfonate-sulfur utilization) operon described in [Koch DJ, Ruckert C., Rey DA, Mix A., Puhler A., Kalinowski J. 2005. Role of the ssu and seu genes of Corynebacterium glulamicum ATCC 13032 in utilization of sulfonates and sulfonate esters as sulfur sources. Appl. Environ. Microbiol. 71, 6104-6114]. Interestingly, in both genomes studied, both operons are almost completely lost (see below), however, a new potential member appears in the genome of C.efficiens in the cysIXHDNYZ sulfate utilization operon (Fig. 4, position A), the nearest orthologs of which are members of the protein luciferase families from proteobacterial genomes. A search for conservative protein domains in GenBank allows us to more accurately predict the function of the CE2637 protein as an alkanesulfonate monooxygenase, which appears to have appeared in the C.efficiens genome due to non-orthologous substitution. The sulfate utilization operon does not have an ABC transporter; however, the substrate for alkanesulfonate monooxygenase is probably supplied by one of two ABC transporters (cg0735-cg0737 and cg2675-cg2678 in the C. glutamicum genome) with unknown specificity. Both transporters and regulatory signals in front of the operons encoding them (see below) are stored in the C.efficiens genome (Fig. 4, position A).

2. Operon rearrangements. The situation with gene permutation in the thrABC operon has already been mentioned. Similar cases are characteristic of other operons. So, the genes of homoserine acetyltransferase (metX) and acetyl homoserine lyase (metY) in different genomes are located differently relative to each other (Fig. 4, position D). In the C. glutamicum genome, they are nearby, are transcribed in one direction, the distance between the genes is relatively small - 144 pairs of nucleotides and, nevertheless, the genes are separated by a potential terminator [6]. In the genomes of C. diphtheriae and Nocardia farcinica, these genes are also transcribed in the same direction, but they are already widely spaced across the genome. The genomes of C.efficiens and C.jeikeium are characterized by the formation of a single metXY operon. In addition, a divergent arrangement of these genes is possible when metY (here cysD) is located convergent with the potential dehydrogenase gene, and the metX gene (here metA) is diverged to them. This situation is observed in the genome of Leifsoma xyli (Fig. 4, position D).

Intraperon gene permutations are also characteristic of the cg0735-cg0737 operon encoding the ABC carrier of an unknown substrate (Fig. 4, position B). In the genomes of the genus Corynebacterium (except C. diphtheriae), as in the genomes of Nocardia farcinica and Leifsonia xyli, the structure of the operon is preserved, while in the genome of Streptomyces coelicolor it is changed to cg0737-cg0735-cg0736. In the genome of C. diphtheriae, the operon is elongated due to the appearance of the paralog of protein cg0737 before the operon. It is possible that the lengthening of the operon is also observed using the mcbR regulator as an example (see below).

In the sulfate recycling operon already mentioned above with the Streptomyces coelicolor genome, a new member appears in the center of the cysIXHCDN operon, cysC, potentially encoding adenylyl sulfate kinase (Fig. 4, position A). The origin of this protein is not obvious, since its orthologs are present only in the genomes of the genus Streptomyces. In the homologous operon cysIXHDNYZB C.efficiens, in addition to the alkanesulfonate monooxygenase (see above), another new potential termulon member appears (Fig. 4, position A) encoding methylthioadenosine nucleosidase (see below). Orthologs of this protein among the genomes under consideration are also in C. glutamicum and C. diphtheriae, where they form monocistronic operons.

The operon cg2679 C.glutamicum divergent to the operon cg2674-cg2678 [6] may include the cg2680 gene (encoding aminotransferase), since the intergenic distance is only 52 pairs of nucleotides and a potential Rho-independent terminator between these genes was not predicted [ 6]. The divergon cg2574-cg2680 is conservative only in the C.efficiens genome and is completely lost in all other genomes. Only for the cg2680 gene is there an ortholog in the C.jeikeium genome, where it is part of another regulated operon (Fig. 4, position B), which also includes aecD (cystathionine-begaliasis) and ywjA (ATPase ABC transporter). In the genomes of C. glutamicum and C.efficiens, the orthologs of the ywjA gene form monocistronic unregulated operons, and the orthologs of the aecD gene form part of the unregulated operon, which also includes the brnQ gene (branched chain amino acid carrier) and the alkane-monooxygenase encoded by (Fig. 25) .4, position B). In the C. diphtheriae genome, the ortholog of the ywjA gene is part of the unregulated operon aecD-brnQ-DIP1738 [DIP 1738 encodes alkanal monooxygenase). In the C.jeikeium genome, orthologous genes of the alkanal monooxygenase and the branched chain amino acid transporter form unregulated monocistronic operons (Fig. 4, position B).

3. Paralogs and duplications. The situation with the appearance of the cg0737 protein paralog in C. diphtheriae in an operon orthologous to cgl0735-cg0737 C.glutamicum has already been described above (Fig. 4, position B). In addition, the ssuD protein of Nocardia farcinica encoding the alkanesulfonate monooxygenase duplicated in the C.glufamicum genome to form ssuD1 and ssuD2 paralogs, probably performing the same function. The ssuDl gene is part of the ssuABC-D1 operon, which is responsible for the disposal of sulfonates. The ssuD2 gene is included in the seuABC-ssuD1 sulfate ester utilization operon. Both operons are completely lost in all genomes studied, except Nocardia farcinica, where, in addition to the ssuD protein, the ortholog of the seuA gene encoding the desulfurization enzyme is also preserved. Both genes form an operon, which also probably contains fadE14, which encodes a potential acylCoA dehydrogenase.

The cysJ / fpr2 gene (sulfate reductase) is located divergent to the sulfate utilization operon (cysIXHDNYZ), and it is conservative in all genomes studied, despite the fact that the main operon can be completely lost (Fig. 4, position A). It should be noted that on the evolutionary tree (Fig. 5), all orthologs are located on the same branch except for the fpr protein Leifsonia xyli, which is located closer to Arthobacter and Brevibacterium. In each of the studied genomes, the fpr protein is represented as a single copy, while in the genomes of C.efficiens and C.glutamicum, in the form of two paralogs, fpr1 and frp2 (Fig. 5). In each of the two genomes, one of the paralogs is part of the sulfate utilization divergon, and the second forms a monocistronic unregulated operon.

For the same sulfate utilization operon (cysIHDNY) in different parts of the Nocardia farcinica genome, there are two additional pairs of genes orthologous to the cysDN genes encoding ATP sulfurylase, cysD2-N2 and nfa33930-nfa33920. On the evolutionary trees of the CysN proteins (Fig. 6) and CysD (Fig. 7), it is seen that the closest homologs of the CysD2 and CysN2 proteins in Nocardia farcinica are bacteria proteins of the genus Mycobacterium and, most likely, a pair of cysD2-N2 genes appeared in the Nocardia farcinica genome due to horizontal transfer from other bacteria of the Actinomycetales order (most likely, from bacteria of the genus Mycobacterium). Similarly, for the proteins encoded by the nfa33930-nfa33920 genes, the closest homologs are the proteins of the proteobacteria (Figs. 6 and 7) and, most likely, the pair of nfa33930-nfa33920 genes in the Nocardia farcinica genome also appeared as a result of horizontal transfer, but in this case from the genomes proteobacteria. In both cases (cysD2-N2 and nfa33930-nfa33920), genes presumably form operons (for regulation, see below).

4. The loss of orthologs. A complete absence of orthologs is observed in the case of the following C. glutamicum genes: ssuR (regulator of the utilization of sulfonate and sulfate ester genes), cg3372 (protein with unknown function), ssul (flavin reductase), seuBC and ssuABC (see above). Many orthologs are lost only in phylogenetically more distant genomes. Thus, in the genomes of Nocardia farcinica, Leifsonia xyli, and Streptomyces coelicolor, there are no orthologs of the cysK (acetylserine-lyase), cysE (serine-acetyltransferase), aecD (see above) and brnQ (see above) and brnQ orthologs of the gene ywj orthologs .jeikeium. The distant genomes also lack an ortholog of the cg1739 gene (see below). Some orthologs were lost not only in the more distant genomes, but also in the genomes of C. diphtheriae and C. jeikeium - these are almost all genes of the operon orthologous to the cg2674-2680 C. glutamicum operon (see above), and the orthologs of the cg3132 gene with unknown function. In the genomes of all distant organisms and in C.jeikeium there are no orthologs of the potential transcriptional regulator cg0156.

Among more specific cases, one can single out the loss of orthologs of the seuA and ssuD genes (see above) in all studied genomes except Nocardia farcinica. No metX and metY gene orthologs were found in the Strepromyces coelicolor genome, metB gene (cystathionine-gamma-synthase) in the C. diphtheriae genome, metH gene (homocysteine methyltransferase) in the C.jeikeium and Leifsonia xyli genomes, etfAB genes (etfAB the genome of Leifsonia xyli.

Orthologs of the metE gene (homocysteine methyltransferase) and cysIXHDNYZ sulfate utilization operon (Fig. 4, position A) are completely lost in the genomes of C. diphtheriae and Leifsonia xyli. The cysIXHDNYZ C. glutamicum operon contains the short cysX protein (potentially involved in electron transfer), whose orthologs are preserved in the genomes of C.efficiens, C.jeikeium and Streptomyces coelicolor (Fig. 4, position A). Among the genomes that preserved the sulfate utilization operon, the genome of Streptomyces coelicolor lacks the orthologs of the cysYZ genes (cys Y takes part in the cofactor biosynthesis, cysZ is the potential sulfate permease), and the orthologs of the cysX and cysZ proteins are absent in the Nocardia farcinica genome.

5. Complete conservatism in the structure of the operon in all studied genomes is observed only in the case of the metK gene (adenosylmethionine synthase) and, possibly, the mcbR gene (see below).

In [6], regulatory motifs were predicted for almost all members of the McbR regulon in C. glutamicum, and some of them were confirmed experimentally. A consensus palindromic binding site of a regulator of the species TAGAC-N6-GTCTA was also presented. In the implementation of the proposed method, a consensus search confirmed the prediction of regulatory signals before the enzymes involved in methionine biosynthesis, and based on the predicted and experimentally confirmed signals, an initial training sample was compiled and a matrix was constructed to search for McbR binding sites.

In the genomes of the genus Corynebacterium, sites with a weight above the threshold value of 4.1 were searched. All potential McbR binding sites found are shown in FIG.

In addition to the signals already predicted [6], a new potential McbR binding site was discovered in the C. glutamicum genome - in front of the cg1739 gene, which encodes a protein with a conserved glutamine amidotransferase domain. Orthologs of this protein are present in all genomes of the genus Corynebacterium and are absent in more phylogenetically distant organisms. Before all orthologs of this protein, a binding site with a weight higher than a predetermined threshold value was predicted (Fig. 3), thus, according to the method of checking the correspondence, the cg1739 gene is a new potential member of the McbR regulon. An indirect experimental confirmation of this prediction is the activation of the expression of this gene in the deletion mcbR mutant C.glutamicum [6].

Before the main part of the regulated orthologs of the C. glutamicum genes in other genomes of the genus Corynebacterium, potential binding sites of the McbR regulator are also predicted (Fig. 3). So, the signal is saved before all the orthologs of the metK, metH, metE, metB, cysK and cg3132 genes found in these genomes and before the cg2674-cg2680 divergon genes.

However, sometimes even in close genomes of the genus Corynebacterium, regulatory signals are absent before orthologous genes. So, the hom-thrB operon regulated in C. glutamicum contains potential signals for homologous operons only in the genomes of C. efficiens and C. jeikeium, and not in C. diphtheriae (Fig. 4, position D). For the cg0156 gene, the situation is the opposite: the ortholog of this gene in C. diphtheriae maintains the binding site of the regulator, while the ortholog in C.efficiens does not (Fig. 3). In the case of the etfAB operon, regulation is preserved only before the orthologous operon in the C.efficiens genome.

More specific cases include the regulation of homoserine acetyltransferase (metX) and acetyl homoserine lyase (metY) genes. The metX and metY genes in the C. glutamicum genome are located in tandem, but separated by a potential Rho-independent terminator [6] and there is a McbR binding site in front of each of the genes. In the genome of C. diphtheriae, these genes form monocistronic operons spaced across the genome, a regulatory signal is stored in front of each of the genes (Fig. 3). In the genomes of C.efficiens and C.jeikeium there is already a single regulated bicistronic operon metYX, the McbR binding site is located in front of the first gene of the operon metY (Fig. 4, position D).

In the C.diphtheriae genome, before the operon DIP0608-DIP0610, homologous to the operon cg0735-cg0737 C.glitamicum, the paralogue of the DIP0610-DIP0611 gene appears, which forms an independent monocistronic operon. Regulator binding sites are located in front of all orthologs of the cg0737 gene in the genomes of the genus Corynebacterium, as well as in front of the DJP0611 gene (Fig. 4, position B, Fig. 3).

For the transcriptional regulator MbR, autoregulation was predicted and experimentally confirmed. Among regulator orthologs in the genomes of the genus Corynebacterium, autoregulation is preserved in the genome of C. diphtheriae and is lost in the genome of C.jeikeium. For the C.efficiens genome, a tandem is predicted in front of the mcbR gene, overlapping with it by 28 pairs of nucleotides, a short open reading frame encoding a hypothetical protein of 57 amino acid residues. A potential mcbR binding site is located in the regulatory region of this hypothetical gene. However, for this hypothetical protein, there is not a single ortholog among all the proteins of the GenBank database and, possibly, it is the result of re-prediction during mapping of the C.efficiens genome. If this is true, the mcbR operon is completely conserved in all studied genomes, and the regulatory binding site in the C.efficiens genome is located at a distance of 143 nucleotides from the translation start point of the mcbR gene.

The operon rearrangements of orthologs were described above, in which the orthologs of the aecD, brnQ genes and C. glutamicum monooxygenase and the orthologs of the C. jeikeium ywjA gene are involved (Fig. 4, position B). In the genomes of C. glutamicum, C. diphtheriae and C.efficiens, all of these genes (and their orthologs) are not regulated by McbR. In the C.jeikeium genome, genes encoding monooxygenase and brnQ protein form two unregulated monocistronic operons, while the aecD, ywjA genes and the ortholog of the cg2680 C.glutamicum gene are in a single, potentially regulated operon - the predicted McE binding site is in front of the aecD gene (Mcb .4, position B, Fig. 3).

Divergons homologous to the cysJIXHDNZY C.glutamicum sulfate utilization divergon retain McbR regulation in the C.efficiens and C.jeikeium genomes in which this divergon has survived (Fig. 4, position A). The structure of the operon is conservative in all three genomes, but the intergenic distances are different. In the C. glutamicum and C.efficiens genomes, the distance between the cysXvi cysH genes is 3 nucleotide pairs, while in the C.jeikeium genome there are 247 nucleotide pairs. The fact that the cysHDNYZ genes form an independent operon is confirmed by the presence of a potential McbR binding site in front of the cysH gene (Fig. 4, position A, Fig. 3). Thus, the sulfate utilization genes in C.jeikeium form the cysJIX divergon and the cysHDNYZ operon.

As already mentioned, in the C.efficiens genome, the cysIXHDNZY operon includes two additional members encoding an alkanesulfonate monooxygenase (CE2637) and methylthioadenosine nucleosidase (CE2636). The CE2637 gene has no orthologs in any of the genomes under consideration (see above). Orthologs of the CE2636 gene are present in the genomes of C. glutamicum and C. diphtheriae in the form of monocistronic operons, with the McbR binding site located in front of the ortholog in the C. diphtheriae genome (Fig. 4, position A, Fig. 3). The regulatory region in front of the orthologous gene in the C. glutamicum genome also contains a potential binding site, but the weight of this signal is below a predetermined threshold value and is 3.79. Thus, the fact that the CE2636 gene belongs to the McbR regulon is highly hypothetical, although it deserves experimental verification.

A positional weight matrix based on McbR binding sites in the C. glutamicum genome made it possible to search for regulatory sites in the genomes of the genus Corynebacterium. However, when searching for signals in more phylogenetically distant genomes, this matrix found only one strong binding site for the regulator (site weight 5.02) in front of the genes of the metabolic pathway under study - the mcbR autoregulation site in the Nocardia farcinica genome (Fig. 3). All other genes of the methionine biosynthesis pathway in these genomes have signals with too low a weight (less than 3.0). A possible explanation for this observation may be that the regulator could change the DNA-binding domain and, accordingly, the signal it recognizes. To search for a signal specific for more distant genomes, the following assumptions were checked in each genome:

- the possibility of a slight change in the signal. An attempt was made to construct a new matrix based on a sample of those weak sites that the original matrix recognized;

- the possibility of a significant change in the signal. Given the McbR affiliation with the TetR family of transcriptional regulators, it could be assumed that the signal structure remained palindromic. A search was made for a more or less strict palindromic signal of different parity and length in the 5'-regions of the main and, separately, "secondary" genes of the metabolic pathway.

None of the approaches to creating a new matrix yielded results. A possible explanation for this is the extremely high GC composition of all three distant genomes: the average level of GC pairs in the genome is on average 70% (for C. glutamicum - 54%), which greatly complicates the search for a specific signal.

As a result, we analyzed the structure and potential regulation of the McbR regulon in all seven actinobacteria genomes in which close orthologs of this regulator are preserved.

Despite the impossibility of studying regulation in phylogenetically more distant genomes due to the significantly increased GC composition of these genomes, the study of potential signals in the four genomes of the genus Corynebacterium suggests that the majority of McbR regulon genes maintain regulation on the part of this repressor. In most cases, the potential signal is stored before the studied orthologous gene, however, in some situations, due to the different relative locations of the genes, the potential binding site may be located in the regulatory region of another gene. For example, the metX and metY genes can form independently regulated monocistronic loci or a single operon with a signal in front of the metY gene (Fig. 4, position D), and the divergon of cysJIHDNZY sulphate utilization C.glutamicum in the C.jeikeium genome breaks into independently regulated cysJIX divergon and operon cysHDNYZ (Fig. 4, position A).

A positional comparison and study of the regulation of orthologous regulons made it possible to predict the presence of a new potential member of the McbR regulon in the genomes of the genus Corynebacterium, the cgl1739 gene C.glutamicum. A search for conserved domains in GenBank reveals the similarity of the encoded protein with the guaA glutamine amidotransferase domain of Bacillus subtilis [1312531] and the distant protein homology with the homoserine succinyltransferase domain. Homoserin-succinyltransferase performs a function similar to that of homoserin-acetyltransferase in the genomes of some bacteria, such as Escherichia coli and Pseudomonas aeruginosa [Lee HS, Hwang BJ 2003. Methionine biosynthesis and its regulation in Corynebacterium glutamictrum sulfrum parallel. Appl. Microbiol. Biotechnol. 62, 459-467]. The simultaneous presence of these two enzymes has not yet been shown and, moreover, it has been experimentally shown that the only homoserine utilization enzyme in the C. glutamicum genome is homoserine acetyltransferase [Park SD, Lee JY, Kim Y., Kim JH, Lee HS 1998. Isolation and analysis of metA, a methionine biosynthetic gene encoding homoserine acetyltransferase in Corynebacterium glutamicum. Mol. Cells 8, 286-294.]. The possible role of a protein homologous to B.subtilis guaA protein in methionine biosynthesis is also not obvious. The function of this enzyme is to hydrolyze an ammonium ion (NH4 ⁺ ) from a glutamine molecule to form a glutamate molecule and then attach the ammonium ion to the acceptor molecule. An explanation of the role of this enzymatic activity in methionine / cysteine biosynthesis could be the possible regulation of biosynthesis at the stage of formation of aspartate from glutamate and oxaloacetate. But such an interpretation seems unlikely due to the fact that glutamate is involved in most of the biosynthesis processes and the regulation of its level in the cell by a specific regulator of methionine biosynthesis is not advisable.

In addition, the results indicate that C.efficiens CE2636p protein and its orthologs in C.glutamicum and C. diphtheriae can also be included in the McbR regulon. This protein has a significant degree of similarity with members of the nucleoside phosphorylase family, which includes 5'-methylthioadenosine nucleosidases (MTA nucleosidases). In the biosynthesis of polyamines, one of the by-products is MTA. The recovery of methionine from MTA consists of several reactions, the first and the limiting of which is the cleavage of MTA into adenine and 5'-methylthioribose catalyzed by MTA nucleosidases. Thus, an increase in the methionine pool due to MTA is also potentially regulated by McbR, at least in the genomes of C. diphtheriae and C. efficiencyiens.

Despite the general high stability of the genomes of corynebacteria [Nakamura Y., Nishio Y ,, Ikeo K ,, Gojobori T. 2003. The genome stability in Corynebacterium species due to lack of the recombinational repair system. Gene. 317, 149-155], in many cases, the loss of orthologous genes was recorded in both distant and closely related genomes. Most of these situations are associated with the characteristics of the parasitic lifestyle characteristic of four of the seven microorganisms described - C. diphtheriae, C.jeikeium, Nocardia farcinica and Leifsonia xyli. Apparently, these microorganisms are able to obtain sulfur-containing organic substances from the host organisms, due to which there is no acute need for enzymes in the methionine biosynthesis pathway, and, as a result, the methionine biosynthesis pathway in these microorganisms is reduced. So, for example, in the genome of C. diphtheriae, the sulfate utilization operon and both alkanesulfonate utilization operons are completely lost, which is only partially offset by the appearance of a new alkanesulfonate monooxygenase as a result of non-orthologous substitution. In addition, the metB gene (cystathionine gamma synthase), which encodes the first enzyme of the trans-sulfurylation pathway, has been lost, and the direct sulfhydrylation pathway involving MetY protein (acetyl homoserine lyase) remains the only possible way to synthesize homocysteine. The metE gene encoding highly efficient methionine synthase has been lost. However, the methionine biosynthesis pathway remains functional due to the presence of MetH methionine synthase, the effectiveness of which is lower.

A positional comparison of amino acid biosynthesis genes in the genomes of C. glutamicum, C.efficiens, and C. diphtheriae revealed the absence of a large number of orthologs in the C. diphtheriae genome, which led to a noticeably smaller genome of this bacterium. Among the microorganisms studied, leading a parasitic lifestyle is also characterized by a markedly reduced genome size (with the exception of the Nocardia farcinica genome), which may indicate a similar mass loss of orthologs and, possibly, indirectly indicate the use of the necessary compounds of the host organism.

In some cases, the absence of orthologs is compensated for by the presence of homologous but not orthologous proteins, as has been described for the functions of threonine synthase and alkanesulfonate monooxygenase. Thus, the genomes of C. glutamicum, C.efficiens, and C. diphtheriae carry a threonine synthase, which is not characteristic of most actinomycetes, obtained, apparently, by the common ancestor of these three bacteria due to horizontal transfer. A more particular analogous case of horizontal gene transfer is the appearance in the genome of C.efficiens of an alkanesulfonate monooxygenase characteristic of proteobacteria.

The application of the proposed method to the mcbR regulon demonstrated numerous and diverse operon rearrangements, despite the previously shown genomic stability, and, in addition, allowed to predict and present several new potential members of this regulon based on positional comparison of the genomes and study of the binding sites of the mcbR regulator. Enzymatic activities were previously attributed to most of the new potential members of the regulon, which made it possible to predict the roles of encoded proteins in the methionine biosynthesis pathway in Actinomycetales genomes.

Thus, the present invention allows to solve the problem of expanding the functionality of the forecasting method, increasing its efficiency, as well as, in the future, reducing the cost of industrial production of methionine.

Information sources

1. Nishio Y., Nakamura Y., Usuda Y., Sugimoto S., Matsui K., Kawarabayasi Y., Kikuchi H., Gojobori T., Ikeo K. 2004. Evolutionary process of amino acid biosynthesis in Corynebacterium at the whole genome level. Mol. Biol. Evol. 21, 1683-1691.

2. Lee H.S., Hwang B.J. 2003. Methionine biosynthesis and its regulation in Corynebacterium glutamicum: parallel pathways of transsulfuration and direct sulfhydrylation. Appl. Microbiol. Biotechnol. 62, 459-467.

3. Hwang B.J., Yeom H.J., Kim Y., Lee H.S. 2002. Corynebacterium glutamicum utilizes both transsulfuration and direct sulfhydrylation pathways for methionine biosynthesis. J. Bacteriol. 184, 1277-1286.

4. Mampel J., Schroder H., Haefner S., Sauer U. 2005. Single-gene knockout of a novel regulatory element confers ethionine resistance and elevates methionine production in Corynebacterium glutamicum. Appl. Microbiol. Biotechnol. 68, 228-236.

5. Rey D.A., Puhler A., Kalinowski J. 2003. The putative transcriptional represser McbR, member of the TetR-family, is involved in the regulation of the metabolic network directing the synthesis of sulfur containing amino acids in Corynebacterium glutamicum. J. Biotechnol. 103, 51-65.

6. Rey DA, Nentwich SS, Koch DJ, Ruckert C., Puhler A., Tauch A., Kalinowski J. 2005. The McbR repressor modulated by the effector substance S-adenosylhomocysteine controls directly the transcription of a regulon involved in sulphur metabolism of Corynebacterium glutamicum ATCC 13032. Mol. Microbiol. 56, 871-887.

Claims (4)

1. A method for predicting the pathway of methionine biosynthesis in related genomes of corynebacteria based on the search for potential binding sites of the McbR regulator, including a) compiling an initial training sample, for which complete sequences of corynebacterial genomes are formed on the basis of the Gen Bank database and previously predicted and experimentally confirmed regulatory regions in front of the enzyme genes involved in methionine biosynthesis, and recorded on an information carrier; b) introducing into the computer the initial training sample recorded on the information carrier and processing it in advance with the SignalX program written in the computer memory to construct the matrix of positional nucleotide weights; c) searching for potential binding sites of the McbR regulator and predicting the composition of operons, if the operon structure is unknown, by processing the initial training set with algorithms from the Genome Explorer software package pre-stored in the computer’s memory, and the search for potential binding sites is limited to the range of -400 ... + 100 nucleotide pairs relative to the start of gene translation, and genes are considered to belong to the operon in the case of the same reading direction and the distance between them is not more than 100 nucleotide pairs; d) recording the results obtained at the stage by sub-item (c) in the computer memory; e) the search for orthologs of the proteins of the methionine biosynthesis pathway in all related genomes using the GenomeExplorer software package, presenting the results of this search and the steps in p (c) in two output text files, the first of which is displayed on an information medium in the form of a table of path gene matching methionine biosynthesis to potential binding sites of the McbR regulon and related genomes of corynebacteria, and the second - in the form of a diagram of the structure of McbR regulon operons; e) predicting the participants and the direction of the methionine biosynthesis pathway in related genomes of corynebacteria based on positional comparison of the contents of both output text files, taking into account the search results for regulatory binding sites in more phylogenetically distant genomes, for the first output text file, and taking into account the characteristic evolutionary events of the formation of the structure of operons , for the second output text file. 2. The method according to claim 1, characterized in that non-orthologic substitutions, operon rearrangements, paralogs, loss of orthologs, or complete conservatism in the structure of the operon are recorded as evolutionary events. 3. The method according to claim 2, characterized in that when fixing the presence of paralogs, evolutionary trees are used, the construction of which is carried out by the maximum likelihood method using the program from the PHYLIP 3.63 software package. 4. The method according to claim 1, characterized in that when searching for protein orthologs of the methionine biosynthesis pathway in the related genomes of corynebacteria, the search for homologous sequences in the database is performed using the BLAST program, and the construction of multiple alignments of amino acid and nucleotide sequences is carried out using the algorithms of the ClustalX program.

Images