Exploring sequenced metagenomes for novel BVMOs

Yeast Infection No More

Natural Solution for Candida Albicans

Get Instant Access

Nowadays, the genomes of a wide variety of organisms have been sequenced and are publicly available, offering a new and efficient way of retrieving BVMO genes. Currently, as genome sequences of over 600 microbes are available (see http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi), it is attractive to look directly into this pool of genes in silico without growing and isolating any microorganism. In addition to the database of partially and fully sequenced genomes, it is also informative to survey the database of sequenced environmental genomes. Especially, the 'Sargasso sea' metagenome database is rich in new sequences.38 Metagenomic databases contain a large number of unexplored genes (>1 million sequences in the Sargasso database!). However, in an in silico (meta)genome mining approach it is fundamental to have bioinformatic tools to identify, with some certainty, genes that encode BVMOs. By simply searching for sequences that show homology with known BVMOs, novel putative BVMO genes may be found. However, many sequence-related genes may represent flavoprotein monooxygenases that do not catalyze Baeyer-Villiger reactions. A more reliable identification of BVMO genes has become feasible since a type I BVMO-specific motif (FxGxxxHxxxWD/P) was identified by comparing sequences of characterized BVMOs.39 This allows an effective survey of all (meta)genome databases concerning the occurrence of type I BVMOs (Table 2: note that this table only includes completely sequenced genomes). The identification of new BVMO genes will obviously facilitate production of novel biocatalysts. Except for obtaining the new gene sequence of a specific BVMO, a thorough genome analysis may also provide more valuable information. As genes belonging to specific degradation pathways are often clustered on microbial genomes, analyzing the sequence regions flanking a BVMO gene may give hints concerning the physiological role of the enzyme and accordingly provide clues for the corresponding substrate specificity. In fact, we have found that many BVMO genes are flanked by an esterase/hydrolase gene. The corresponding esterase or lactonase activity would hydrolyze the ester or lactone formed by BVMO activity. Such a co-localization of BVMO with an esterase/lactonase has also been observed in sequenced genome fragments containing a BVMO gene, e.g., in the case of the CHMO, cyclopentanone monooxygenase, PAMO, HAPMO and cyclopentadecanone genes.17'22'23'40-42 This suggests that BVMOs often play a role in a specific catabolic pathway and is in agreement with the fact that most described BVMOs are part of a degradation pathway.

A genomic trawl using the above-mentioned type I BVMO motif as filter currently yields 174 putative BVMO genes when searching for all finished micro-bial genomes (Table 2). This indicates that type I BVMOs are not very rare but are frequently utilized by microbes. All identified BVMO genes originate from bacteria or fungi while none could be found in archaebacteria, plants, animals or the human genome. On average, roughly one out of two (174/348) micro-bial genomes contain a BVMO gene. This suggests that at present ~400 novel

Table 2

Genomic occurrence of BVMOs

Table 2

Genomic occurrence of BVMOs

Source

Number of

Number

Examples of genomes

screened

of putative

containing multiple BVMOsb

genomes

BVMOsa

Bacteria

293

138

Mycobacterium aviumc (12) Novosphingobium aromaticivorans (8) Rhodococcus RHA1 (20)

Archaea

27

0

-

Fungid

28

FGSC-4A (16) Gibberella zeae PH-1 (10) Magnaporthe grisea 70-15 (7)

Other

22

0

-

eukaryotese

Sargasso sea

-

32

-

Total

370

206

-

a Genomes were surveyed for the presence of putative type I BVMOs by: (1) searching for sequence homologs of phenylacetone monooxygenase and (2) filtering for sequences that contain the type I BVMO-sequence motif.39 b The number of BVMO genes for each genome is indicated in brackets. c Mycobacterium avium subsp. paratuberculosis K-10.

d Sixteen yeast and 12 fungal genomes have been screened: 33 BVMOs discovered in fungi and three in the yeast Candida albicans SC5314. e Other eukaryotes included Apis mellifera, Bombyx mori, Caenorhabditis elegans, Drosophila melanogaster, Homo sapiens, Mus musculus, Tetraodon nigroviridis.

a Genomes were surveyed for the presence of putative type I BVMOs by: (1) searching for sequence homologs of phenylacetone monooxygenase and (2) filtering for sequences that contain the type I BVMO-sequence motif.39 b The number of BVMO genes for each genome is indicated in brackets. c Mycobacterium avium subsp. paratuberculosis K-10.

d Sixteen yeast and 12 fungal genomes have been screened: 33 BVMOs discovered in fungi and three in the yeast Candida albicans SC5314. e Other eukaryotes included Apis mellifera, Bombyx mori, Caenorhabditis elegans, Drosophila melanogaster, Homo sapiens, Mus musculus, Tetraodon nigroviridis.

type I BVMO genes are present in the genome sequence database (including the unfinished genomes). Strikingly, BVMO genes are unevenly distributed among microbial genomes with only a few microorganisms containing a large number of BVMOs while the majority of genomes are devoid of putative type I BVMOs (Fig. 4). In fact, the number of genomes containing only one BVMO is more or less equal to the number of genomes containing four or more BVMO genes. It is also worth noting that a relatively large number of bacterial BVMO genes (80) were found in actinomycetes. This may suggest a role of BVMOs in the synthesis of secondary metabolites. Also, the Sargasso metagenome database contains a significant number of BVMO genes indicating that in the sea environment also many microbes employ BVMOs for specific but yet unknown metabolic routes (Table 2). These genes cannot simply be obtained using PCR techniques as the microbes from which the genes originate have not been isolated. However, by gene synthesis it is in principle feasible to explore these newly identified putative biocatalysts.

I Bacteria ] Fungi

Number of BVMOs/genome

Figure 4: Distribution of putative type I BVMO genes among bacterial and fungal genomes (see Table 2).

In contrast to the type I BVMOs discussed above, type II BVMOs have been explored to a limited extent. In fact, an in silico search for type II BVMOs in the genome sequence database is hampered by the fact that only one type II BVMO sequence (limonene monooxygenase, gi47116765) has been deposited in the database. A BLAST search with the limonene monooxygenase sequence at NCBI (http://www.ncbi.nlm.nih.gov) reveals that only 12 bacterial sequences show a relatively high sequence homology (>40% sequence identity). Except for these sequence homologs, a large number of other sequences also show limited sequence homology and appear to belong to the luciferase class of flavin-dependent monooxygenases. This hints to an evolutionary relationship between type II BVMOs and luciferases. In the Sargasso sea database, only five sequences can be found that display high sequence identity (>40%) with limonene monooxygenase. These findings suggest that type II BVMOs are less widespread when compared with type I BVMOs explaining to some extent why these BVMOs have been reported in the literature less frequently.

Recently, several BVMOs have been reported in the literature that had been found by genome mining. The first example concerned the discovery of a thermostable BVMO. The well-studied BVMO, CHMO, is not a very robust biocata-lyst. Often, conversions using this biocatalyst suffer from enzyme inactivation. To circumvent this problem, it was of interest to obtain a more (thermo)stable BVMO. As no BVMO genes have been identified in genomes of archaebacteria, genomes of (semi)thermophilic bacteria were surveyed. Using the above-mentioned BVMO sequence motif, it was found that the genome of Thermobifida fusca contains two type I BVMO genes. This actinomycete typically grows at 55-60°C and therefore should yield thermostable biocatalysts. The two genes have been cloned and the corresponding enzymes have been overexpressed in Escherichia coli. Only one of the two expressed BVMOs could be purified and characterized and was shown to be primarily active on a range of aromatic ketones and sulfides. The highest catalytic efficiency has been obtained with phenylacetone and hence its name phenylacetone monooxygenase. The genes flanking the PAMO gene also suggest a role in the degradation of aromatic compounds like phenylacetone. Except for phenylacetone, a number of other ketones and sulfides are accepted by the enzyme.17 36 With several aromatic prochiral ketones and sulfides, excellent enantioselectivity was observed. However, the enzyme is only marginally active with cyclic aliphatic ketones and therefore has only a limited overlap in substrate specificity when compared with CHMO (Fig. 3). The enzyme indeed proves to be of superior stability when compared with other known BVMOs as it is stable for days when stored at moderate temperatures (<40°C). The enzyme is also active in the presence of organic solvents and, interestingly, it was found that organic solvents can tune the enantioselectivity.43

More recently the group of Grogan reported on another genome mining project targeting BVMOs from Mycobacterium tuberculosis.44 The genome of M. tuberculosis harbors six putative type I BVMO genes. Also, other mycobacterial genomes are relatively rich in BVMO genes with M. marinum setting the record at present: 15 BVMO genes. All six type I BVMOs genes that are part of the genome of M. tuberculosis were cloned and four of them could be overexpressed in E. coli. Unfortunately, by using three test substrates, Baeyer-Villiger activity of only three mycobacterial BVMOs could be confirmed. One of them corresponds to ethionamide monooxygenase that has been shown to be involved in the activation of commonly used antitubercular drugs.45 46 In a previous study, Baeyer-Villiger activity of this monooxygenase had already been verified with a large number of ketones.47 However, another overexpressed BVMO from M. tuberculosis exhibited exquisite enantioselectivity with racemic bicyclo[3.2.0]hept-2-en-6-one.44 This bicyclic aliphatic ketone is often used to probe the biocatalytic potential of BVMOs. This chiral ketone is of high value for the synthesis of fine chemicals when available in its enantiomerically pure form. Also, the chiral lactones formed from this ketone by Baeyer-Villiger oxidation are highly interesting for synthetic purposes. For this reason, CHMO has been exploited in several biocatalytic studies as mentioned above.3032 Using whole cells containing the novel mycobacterial biocatalyst, it was possible to perform a resolution of racemic bicyclo[3.2.0]hept-2-en-6-one yielding enantiomerically pure (1R,5S)-(+)-bicyclo[3.2.0]hept-2-en-6-one. This was not feasible with CHMO or any other available BVMO and demonstrates again the value of a genome mining approach. By exploring the catalytic properties of novel BVMOs identified only by sequence, new substrate specificities and regio- and/or enantioselectivities may be uncovered.

Was this article helpful?

0 0
Natural Remedy For Yeast Infections

Natural Remedy For Yeast Infections

If you have ever had to put up with the misery of having a yeast infection, you will undoubtedly know just how much of a ‘bummer’ it is.

Get My Free Ebook


Post a comment