Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Aug 25.
Published in final edited form as: Chem Biol. 2008 Aug 25;15(8):790–798. doi: 10.1016/j.chembiol.2008.07.012

ROSETTA STONE FOR DECIPHERING DEAZAPURINE BIOSYNTHESIS: PATHWAY FOR PYRROLOPYRIMIDINE NUCLEOSIDES TOYOCAMYCIN AND SANGIVAMYCIN

Reid McCarty 1, Vahe Bandarian 1,*
PMCID: PMC2603307  NIHMSID: NIHMS67376  PMID: 18721750

Summary

Pyrrolopyrimidine nucleosides analogs, collectively referred to as deazapurines, are an important class of structurally diverse compounds found in a wide variety of biological niches. In this report, a cluster of genes from Streptomyces rimosus involved in production of the deazapurine antibiotics sangivamycin and toyocamycin was identified using forward genetics methods. The cluster includes toyocamycin nitrile hydratase, an enzyme that catalyzes the conversion of toyocamycin to sangivamycin. In addition to this rare nitrile hydratase, the cluster encodes a GTP cyclohydrolase I, linking the biosynthesis of deazapurines to folate biosynthesis, and a set of purine salvage genes, which presumably convert the guanine moiety from GTP to the adenine-like deazapurine base found in toyocamycin and sangivamycin. The gene cluster presented here could potentially serve as a “Rosetta stone” to inform on deazapurine biosynthesis in other bacterial species.

Pyrrolopyrimidine functional groups have been found in many secondary metabolites produced by Streptomyces bacteria. The molecules in which they are found range from cofactors involved in the biosynthesis of tetracycline antibiotics (McCormick and Morton, 1982; Miller et al., 1960) and DNA repair (Kuo et al., 1989) to compounds with herbicidal, antibacterial, antifungal and antineoplastic activities (Isaac et al., 1991; Nishioka et al., 1991; Suhadolnik, 1970a). The breadth of structural diversity in deazapurine-containing family of compounds is remarkable (Figure 1a). The strong resemblance between pyrrolopyrimidine nucleosides and purines suggest that these molecules could disrupt nucleic acid metabolism and kinase-related signaling cascades and be of potential use as therapeutic agents. Indeed, the ability of deazapurines to enter nucleic acid pools of cells is well documented (Ritch and Glazer, 1982; Suhadolnik et al., 1967; Suhadolnik et al., 1968; Tavitian et al., 1969). Despite the ubiquity and potential usefulness of these molecules both as tools and therapies, their biosynthetic pathways have remained elusive.

Figure 1.

Figure 1

Open in a new tab

Deazapurines are a diverse class of purine-based secondary metabolites. Deazapurine secondary metabolites produced by Streptomyces (a). (b) Deazapurine secondary metabolites are derived from purines in a pathway that retains C-2, as well as C-1′, C-2′ and C-3′ of the starting purine, but loses the C-8. Toyocamycin nitrile hydratase catalyzes the hydration of toyocamycin to produce sangivamycin. S. rimosus produces both sangivamycin and toyocamycin.

The outlines of the biosynthetic steps leading to these molecules were established by elegant radiotracer experiments (summarized in Figure 1b) in which deazapurine base was shown to be derived from a purine precursor (Smulson and Suhadolnik, 1967; Suhadolnik and Uematsu, 1970; Uematsu and Suhadolnik, 1970), by way of a process mirroring transformations involved in the formation of folate (Burg and Brown, 1966; Reynolds and Brown, 1964) and riboflavin (Bacher and Mailänder, 1973). When [2-14C]- or [8-14C]-labeled adenine and guanine were fed to S. rimosus cells in culture, the isolated toyocamycin was shown to have retained label at carbon-2 but lost the label at carbon-8. Intriguingly, the carbons 5 and 6 of the pyrrole moiety, and the cyano carbon appeared to derive from the C1′, C2′, and C3′ of the proffered ribose, respectively.

While exact nature of the enzymatic transformations that underlie the biosynthesis of deazapurines remains to be established, two studies have provided substantial hints (Elstner and Suhadolnik, 1971; Elstner and Suhadolnik, 1975; Suhadolnik, 1970a; Uematsu and Suhadolnik, 1974; Uematsu and Suhadolnik, 1975). First, Suhadolnik and coworkers isolated a GTP cyclohydrolase protein from cells producing sangivamycin, whose activity paralleled the appearance of sangivamycin in the growth medium (Elstner and Suhadolnik, 1971; Elstner and Suhadolnik, 1975); the protein was shown to produce a neopterin-like molecule, much like the product of GTP cyclohydrolase I (GCH I) of E. coli, which had been characterized by Brown and colleagues (Burg and Brown, 1968). The pyrimidine ring of sangivamycin or toyocamycin, however, resembles adenine and not guanine, suggesting that additional transformations must be involved, if the protein described in these early studies indeed catalyzes the first step in the biosynthesis of these deazapurines. The second intriguing observation was description of a toyocamycin nitrile hydratase (TNHase) activity involving addition of a water molecule across the nitrile moiety of toyocamycin to produce sangivamycin (Uematsu and Suhadolnik, 1974; Uematsu and Suhadolnik, 1975) (see Figure 1b).

The hypermodified tRNA base, queuosine (Kasai et al., 1975), is structurally homologous to the deazapurine base of pyrrolopyrimidine nucleosides that are produced by Streptomyces. Recently, four genes encoding proteins required for the biosynthesis of queuosine have been identified in Bacillus subtillis (Bai et al., 2000; Van Lanen et al., 2005). The importance of the four B. subtillis genes in the early steps was verified by knockout experiments. While bioinformatics studies have revealed the potential biochemical functions for these proteins, in vitro characterization has only been carried out for QueF, which has been shown to catalyze the NADPH dependent conversion of 7-cyano-7-deazaguanine (see preQ0 in Figure 3b) to 7-aminomethyl-7-deazaguanine (preQ1) (Van Lanen et al., 2005). However, the reaction(s) catalyzed for the remaining three have yet to be established. Recent X-ray crystallographic studies of QueC (Cicmil and Huang, 2008), QueD (Cicmil and Shi, 2008), and QueF (Swairjo et al., 2005) may provide additional insights into their biochemical function.

Figure 3.

Figure 3

Open in a new tab

Toyocamycin and sangivamycin are produced by a cluster of genes in S. rimosus. Organization of gene cluster (a) and putative biosynthetic pathway for production of sangivamycin and toyocamycin (b). Radiotracer experiments (Suhadolnik and Uematsu, 1970) have shown that, during the conversion of purines to toyocamycin, carbon 2 (orange) is retained and carbon 8 (green) is lost and that the pyrrole carbons in toyocamycin (red) are derived from the ribose on the nucleoside precursor (Uematsu and Suhadolnik, 1970.) Predicted functions of gene products are based on hits in PSI-BLAST analysis. The putative functions of the proteins are as follows: toyA, LuxR transcriptional regulator; toyB, 6-pyruvoyltetrahydropterin synthase; toyC, radical SAM family protein; toyE, GMP reductase; toyF, adenylosuccinate lyase; toyG, adenylosuccinate synthetase; toyH, phosphoribosylpyrophosphate transferase; toyI, phosphatase; toyK, QueC related protein. Functions of ToyD (GTP cyclohydrolase I) and ToyJKL (toyocamycin nitrile hydratase) were demonstrated biochemically in this study.

Streptomyces tend to organize genes that are responsible for the biosynthesis of secondary metabolites in distinct clusters (Hopwood, 1997); therefore, identification of a cluster of genes involved biosynthesis of any deazapurine secondary metabolite produced by Streptomyces could potentially serve as a “Rosetta stone” to inform on the biosynthetic pathway for deazapurine-containing compounds in nature. The genome sequence of S. rimosus is not available; therefore, identification of the genes responsible for biosynthesis of these secondary metabolites required a different approach. In this work we utilized forward genetics methods to identify the gene for TNHase in S. rimosus. Additional biochemical and bioinformatics analysis leave little doubt that this cluster is indeed the elusive deazapurine secondary metabolite biosynthesis cluster and it appears that this gene cluster can serve as a Rosetta stone for deciphering the language of deazapurine biosynthesis in bacteria.

Results

We undertook a forward genetic approach to isolate the cluster of genes that are responsible for the biosynthesis of deazapurines in Streptomyces, which are known to produce a large variety of deazapurine-containing secondary metabolites in the stationary phase of growth (Suhadolnik, 1970a). S. rimosus was chosen as the model organism in this study since it is known to produce two compounds, sangivamycin and toyocamycin (Elstner and Suhadolnik, 1971; Suhadolnik, 1970b; Suhadolnik and Uematsu, 1970; Uematsu and Suhadolnik, 1974; Uematsu and Suhadolnik, 1970). Furthermore, Suhadolnik and colleagues had identified a TNHase in this organism(Uematsu and Suhadolnik, 1974; Uematsu and Suhadolnik, 1975); as Streptomyces tend to cluster the genes for the biosynthesis of secondary metabolites (Hopwood, 1997), we reasoned that TNHase would be a good marker for the location of the biosynthetic cluster. The S. rimosus strain used in these studies was obtained from American Type Culture Collection (ATCC strain 14673).

Isolation of the TNHase bearing cluster from S. rimosus (ATCC 14673)

TNHase was partially purified from a 3-day old culture growth by a combination of ammonium sulfate precipitation and seven chromatographic steps, which included anion exchange, hydrophobic interactions, hydroxyapatite and size exclusion chromatographic steps. At each step of the purification the TNHase activity present in the fractions was monitored by an HPLC assay, which followed the conversion of toyocamycin to sangivamycin. SDS-PAGE of the resulting partially purified protein eluting from the final column (Fig. 2a) revealed a complex mixture; however, the intensities of three bands appeared to correlate with the magnitude of the toyocamycin→sangivamycin activity that was observed in each fraction (Supplementary Figure S1). Therefore, we reasoned that these proteins were likely to comprise the TNHase protein that we had sought to purify.

Figure 2.

Figure 2

Open in a new tab

Identification of toyocamycin nitrile hydratase of S. rimosus. SDS PAGE analysis of proteins contained in consecutive fractions eluting from a HiPrep 16/60 Sephacryl S-200 analytical size exclusion column during purification of toyocamycin nitrile hydratase activity from S. rimosus (a). The intensities of protein bands 1, 2 and 3 correlate with TNHase activity over the range of fractions shown (Supplementary Fig. S1). N-terminal sequences of bands 1, 2, and 3 (b) were obtained by Edman degradation. TNHase is a three-subunit nitrile hydratase protein (c) with sequence similarity to nitrile and thiocyanate hydratase proteins. Each of these proteins has an metal ion in the active site (cobalt or iron), which interacts with two post-translationally modified cysteine residues (shown in red).

The N-terminal amino acid sequences (≥30 amino acids) of each of the three bands were obtained by Edman degradation (Figure 2b). The 40 amino acid read obtained from the N-terminus of band 3 permitted degenerate primers corresponding to N- and C-terminal portions of the sequence to be designed and used to amplify by PCR a 78 bp region from the S. rimosus genomic DNA. Cloning and sequencing of the PCR product revealed the genomic sequence encoding the amino acids 9–34 of band 3. A synthetic DNA oligomer containing this 78 bp sequence was used to probe a S. rimosus cosmid library for the corresponding gene. Cosmids that hybridized to the probe were isolated, digested into smaller fragments by a variety of restriction endonucleases, and subcloned. Sequencing of these overlapping fragments revealed a 14.5 kbp stretch of DNA that contains the toyocamycin and sangivamycin biosynthesis cluster of S. rimosus, including three subunits that encode toyocamycin nitrile hydratase observed in the purification (Fig. 2c).

The open reading frames that encode the deazapurine biosynthesis cluster in this organism were identified and annotated manually (Fig. 3a). Potential ORFs were identified by the presence of potential start codons, ATG, GTG, and TTG, preceded by reasonable ribosome binding sites (containing a series of adenosine and guanosine bases), and in-frame stop codons. The putative functions of each orf were gleaned from analyzing the results of PSI-BLAST (Altschul et al., 1997) searches of the sequences against bacterial protein databases.

The pyrrolopyrimidine nucleoside antibiotic biosynthesis gene cluster of S. rimosus

Sequence analysis revealed at least 13 putative ORFs clustered together, including the TNHase gene (Fig. 3a). The ORFs that comprise the gene cluster will designated toyABCDEFGHIJKLM; the ORFs encoding protein bands 1, 2, and 3 of the TNHase protein isolated from S. rimosus will be designated toyJ, L, and K, respectively. The orfs in this cluster are listed in Table 1.

Table 1.

ORFs encoded by the sangivamycin/toyocamycin biosynthesis cluster of S. rimosus.

ORF Ribosome binding site Start (Stop) % (G+C) MW (residues) Predicted/Known Function
toyA AAAGA ATG (TGA) 74.5 101436 (961) LuxR transcriptional regulator
toyB GGGGAA TTG (TGA) 65.4 15114 (131) Pyruvoyl-tetrahydropterin synthase
toyC GGGAAGGGG GTG (TGA) 71.1 26743 (243) Radical SAM
toyD AGGAGAG ATG (TGA) 71.2 22140 (200) GCH I (verified, this study)
toyE AGAGAAAGAA GTG (TGA) 74.0 39492 (384) GMP reductase
toyF GGAGG GTG (TGA) 72.6 49005 (452) Adenylosuccinate lyase
toyG AAGGAGGA ATG (TGA) 73.7 36855 (351) Adenylosuccinate synthetase
toyH AAGGAG ATG (TGA) 73.3 18859 (176) Phosphoribosyl-pyrophosphate transferase
toyI GGAGG GTG (TGA) 75.3 25466 (241) Haloacid dehalogenase superfamily
toyJ GGAGGAGAAA ATG (TGA) 71.8 21215 (195) Nitrile hydratase, α subunit (verified, this study)
toyK AGAAGGGGA (GTG) (TGA) 71.2 10105 (91) Nitrile hydratase, γ subunit (verified, this study)
toyL GGGAAA ATG (TGA) 66.3 11543 (103) Nitrile hydratase, β subunit (verified, this study)
toyM AGGAGG TTG (TGA) 69.2 25583 (238) ExsB family
Open in a new tab

In addition to the TNHase genes, this group also includes a GCH I homolog, which has long been suspected to be one of the key enzymes required for the biosynthesis. Indeed, this result is completely in accord with the work of Suhadolnik and coworkers showing appearance of GCH I in cells during production of sangivamycin (Elstner and Suhadolnik, 1971; Elstner and Suhadolnik, 1975). Interestingly, the cluster also includes genes encoding a putative 6-pyruvoyltetrahydropterin synthase (PTPS), a member of the radical SAM protein family (Sofia et al., 2001), and a member of the ExsB protein family. Similar proteins have previously been shown to be involved in queuosine biosynthesis (Gaur and Varshney, 2005; Reader et al., 2004). Most importantly, however, our data show how the organism converts a guanine-based purine to an adenine-based deazapurine. This task appears to be accomplished by a group of purine salvage proteins (phosphoribosylpyrophosphate transferase, GMP reductase, and adenylosuccinate synthetase and lyase homologs).

Activity of heterologously expressed S. rimosus TNHase

To confirm that the toyJKL genes indeed encode the TNHase protein, which was the basis for the isolation of the gene cluster, these genes were cloned into vectors with compatible origins of replication and transformed into E. coli BL21(DE3) cells. Cell lysates of E. coli containing exogenously expressed TNHase subunits ToyJ, K, and L were incubated with toyocamycin, and analyzed by HPLC. Control incubations of cell lysate from E. coli transformed with the corresponding expression vectors alone were performed as well. The HPLC analysis revealed conversion of toyocamycin to sangivamycin by cell lysate from cells containing expression constructs for ToyJ, K, and L (Fig. 4). By contrast, TNHase activity was not observed in lysates of vector only controls. Moreover, in preliminary studies, purified TNHase (ToyJKL) exhibits the toyocamycin→sangivamycin activity (McCarty & Bandarian, unpublished observation).

Figure 4.

Figure 4

Open in a new tab

ToyJKL encode toyocamycin nitrile hydratase. Trace (a) is a standard mixture of sangivamycin and toyocamycin. The activity was assayed in crude extracts from E. coli strains that contained either the toyJKL genes (c) or the corresponding empty vectors (pACYCDuet-1 and pET29a) (b).

The three-subunit TNHase protein isolated in the course of these studies is unique in a number of respects. First, it catalyzes hydration of a cyanide moiety appended to a complex organic molecule, substantially expanding the repertoire of substrates for hydration, and may represent a mild alternative to chemical hydration of cyanide-containing compounds to amides. Second, TNHase contains three subunits while other nitrile hydratase (NHase) enzymes studied to date contain only two (Kobayashi et al., 1992) (see Fig. 2c). BLAST sequence analysis reveals that ToyJ is similar to the α subunits found in other NHase enzymes. However, ToyL and ToyK show identity with the N- and C- terminal halves, respectively, of the β subunit of NHase enzymes.

Curiously, thiocyanate hydrolase (SCNase) from Thiobacillus thioparus THI115, which catalyzes the conversion of thiocyante to carbonyl sulfide and ammonia, contains three subunits (Katayama et al., 1992), which are homologous to those that comprise S. rimosus TNHase. The α and β subunits of T. thioparus SCNase share similarities with the N- and C-terminal halves of the β subunit of typical NHase enzymes (Katayama et al., 1998); the γ subunit of thiocyanate hydrolase is homologous to the a subunit of NHase and TNHase (see Fig. 2c). Therefore, TNHase appears to be a new member of the thiocyanate hydrolase class of proteins with novel substrate spectrum.

The rarity of NHase enzymes is a consequence not only of the fact that they catalyze hydration of an uncommon biological moiety, cyanide, but they do so using either a mononuclear non-heme iron or non-corrinoid cobalt (Banerjee et al., 2002). The metal ion is bound to the active site in an unusual mononuclear six-coordinate metal site and sequence analysis reveals a signature amino acid sequence Val-Cys1-Ser/Thr-Leu-Cys2-Ser-Cys3 in the α subunit; in all cases examined to date, two of the three Cys residues (Cys2 and Cys3) are modified to cysteine sulfinic acid (Cys2-SO2H) and cysteine sulfenic acid (Cys3-SOH), respectively(Arakawa et al., 2007; Nagashima et al., 1998; Song et al., 2007). Iron or cobalt preference of the proteins is thought to be dictated as follows: proteins that have a Ser residue between Cys1 and Cys2 bind iron, where as the presence of Thr directs binding of cobalt(Banerjee et al., 2002; Payne et al., 1997). ToyJ encodes the amino acid sequence V-C-T-L-C-S-C; therefore, we presume that it is a cobalt-type nitrile hydratase. Additional studies on this fascinating protein are currently in progress.

Activity of heterologously expressed GCH I homolog from S. rimosus

To determine if toyD encodes a GCH I homolog that catalyzes the canonical reaction, the gene was cloned, ToyD was overexpressed, purified and assayed for it GCH I activity. GCH I is a zinc-containing protein (Auerbach et al., 2000; Kaiser et al., 2002) and catalyzes the conversion of GTP to 7,8-dihydroneopterin triphosphate (H2NTP). Recombinant purified S. rimosus ToyD was shown to bind 0.76 mol of zinc per mole of monomer.

The conversion of GTP to 7,8-dihydroneopterin triphosphate (H2NTP) is readily monitored by UV-visible spectrophotometry. Changes in the spectral properties of GTP upon incubation with ToyD are shown in Fig. 5a. Incubation with ToyD leads to a decrease in absorbance at 253 nm due to GTP, with a concomitant increase at 330 nm, which is consistent with conversion of GTP to H2NTP (Pfleiderer, 1985). The turnover number for this reaction was determined to be ~0.4 min−1 (at pH 8.0). While low, this value is consistent with the relatively low (0.2–2.5 min−1) turnover numbers that have been observed with GTP cyclohydrolase I homologs studied to date (Bracher et al., 1999; El Yacoubi et al., 2006; Rebelo et al., 2003). The formation of H2NTP was confirmed by LC/MS analysis of the reaction (Fig. 5b). In these experiments, we compared the mass and retention time for the product produced by ToyD with that produced with E. coli GCH I (FolE, B2153). Under the conditions of the assays and analysis, H2NTP is oxidized to neopterin triphosphate, as has been observed in the past (El Yacoubi et al., 2006). Extracted ion chromatograms of the GCH I reactions (in the negative ion mode) reveal a peak with product with m/z of 492 amu (Fig 5b), which elutes at an identical position as the product of ToyD. Furthermore, when equal quantities of the E. coli GCH I and S. rimosus ToyD reaction mixtures are combined and injected, a single peak is observed. Finally, when enzyme is excluded, the neopterin triphosphate is not observed. Collectively, these results confirm that ToyD is indeed a GCH I homolog. Additional control experiments demonstrated that, as with GCH I, ToyD does not catalyze cyclohydrolase chemistry with GDP, GMP, ATP, ADP or AMP (data not shown). To confirm that the GCH I activity observed in purified recombinant ToyD, exogenously expressed in E. coli, is not due to contaminating E. coli FolE, samples of the two purified proteins used in this study (ToyD and FolE) were compared by SDS-PAGE (Supplementary Figure S2). Since FolE has a calculated molecular weight ~2.7 kDa greater than ToyD, the two are resolved by SDS-PAGE to reveal that the ToyD used in these experiments is not contaminated by E. coli FolE (Figure S2). The SDS-PAGE experiment confirms that the stock is composed of ToyD alone.

Figure 5.

Figure 5

Open in a new tab

ToyD is a GTP cyclohydrolase I. UV spectrophotometric assays showing conversion of GTP to H2NTP by S. rimosus ToyD (a) were carried out as described in Experimental Procedures. The identity of the product was confirmed by LC-MS analysis of the reaction and comparison to that observed with E. coli GCH I (FolE) (b). Extracted negative ion chromatograms of the GCH I product (at 492 amu for neopterin triphosphate) show peaks with identical retention times for the product of E. coli FolE and S. rimosus toyD. Identity of the product was confirmed by co-injection of a sample containing equal volumes of E. coli FolE and S. rimosus ToyD reaction mixtures. The product is not observed when the enzyme is omitted. 7,8-Dihydroneopterin triphosphate, which is the initial product of both reactions, is oxidized to neopterin triphosphate in the course of the analysis.

Whether ToyD is the same protein that Suhadolnik and coworkers purified from the organism in 1975 (Elstner and Suhadolnik, 1975) cannot be known for certain, as S. rimosus may encode multiple open reading frames that catalyze similar or identical reactions, as we have shown in the past for GTP cyclohydrolase II in S. coelicolor (Spoonamore and Bandarian, 2008; Spoonamore et al., 2006). Therefore, at present, we cannot rule out the presence of additional homologs of GCH I in S. rimosus.

Discussion

We have identified the gene encoding the toyocamycin nitrile hydratase enzyme in S. rimosus and have sequenced a series of adjacent putative ORFs comprising what appears to be a cluster of genes for the biosynthesis of the pyrrolopyrimidine nucleosides sangivamycin and toyocamycin. A pathway for conversion of GTP to sangivamycin or toyocamycin is depicted in Figure 2b. The biosynthesis occurs in three phases: in the first phase, the starting purine nucleotide is tailored to the deazapurine base, the second phase involves conversion of the guanine-like base to an adenine-like base, and modifications in the 7-position occur in the third phase.

Our working model is that first phase in the biosynthesis of toyocamycin is achieved through 4 steps catalyzed by ToyD, ToyB, ToyC, and ToyM. According to this scheme, ToyD catalyzes the conversion of GTP to H2NTP; in this paper we show that indeed, the protein has the expected enzymatic activity. While the involvement of a GCH I protein in the biosynthetic pathways to deazapurines has been hypothesized previously (Elstner and Suhadolnik, 1975; Reader et al., 2004), this report is the first demonstration that a GCH I protein with canonical activity is co-localized with the genes for the biosynthesis of a purine analog. The additional steps involved in the transformation of H2NTP to the deazapurine, preQ0, occur through the action of the ToyB, ToyC and ToyM trio. The nature of the chemistry and order in which these enzymes act remain to be established. However, homologs of ToyC and ToyM have been postulated to be involved in the biosynthesis of preQ0, which is a precursor the hypermodified base queuosine (Reader et al., 2004). The chemical transformations catalyzed by each remain to be established; however, the presence of the same genes in the toyocamycin biosynthesis pathway provides a universal paradigm for the biosynthesis of deazapurine-containing compounds in nature. The origin of the cyano nitrogen is not known. The most parsimonious path, however, is one where it is derived from N-7 of the starting purine, we cannot exclude the possibility that it is eliminated and the cyano nitrogen is derived from another source.

We note in passing that the gene encoding the radical SAM homolog in the toyocamycin biosynthesis pathway encodes a TGA stop codon 30 bp from the GTG, which we have assigned as the start codon for the gene. Read-through of TGA stop codons, generally by insertion of a selenocysteine, are rare, but precedented. To our knowledge, Streptomyces species have not been shown to produce selenoproteins; moreover, the completed genome sequences of S. avermitilis and S. coelicolor lack genes encoding selenocysteyl tRNA, or other homologs of the bacterial selenocysteine incorporation systems. At this point, one cannot rule out is that the selenocysteine incorporation pathways have diverged. Although PSI-BLAST searches reveal very few examples of homologs that have similarity over the first ~ 11 amino acids of ToyC, significant similarity, including in the CXXXCXXC SAM radical motif (Sofia et al., 2001), is found between the in-frame TGA and the next potential start codon.

In the second phase of the biosynthetic pathway, the preQ0 base is converted to the toyocamycin 5′-monophosphate by the successive action of ToyH, ToyE, ToyG, and ToyF as shown in Fig. 3b. The functional assignments are based on sequence similarities. Intriguingly, these enzymes may have been “borrowed” from those involved in purine salvage; it would be interesting to know if intragenomic homologs of these proteins, which carry-out housekeeping functions, are present elsewhere in the chromosome. In this phase of the pathway, the only orf whose functional assignment is questionable is ToyI. The toyI gene encodes a protein which is homologous to the HAD superfamily of enzymes, which includes a large family of phosphatases (Burroughs et al., 2006); we posit that ToyI could be involved in dephosphorylation of toyocamycin monophosphate, setting the stage for the third and final phase of the biosynthetic pathway.

The third phase of the pathway involves addition of water across the nitrile moiety of toyocamycin to yield sangivamycin. ToyJKL belongs to a rare metal-dependent nitrile hydratase family of proteins. Heterologous expression of the S. rimosus TNHase enzyme in E. coli and the observation of TNHase activity in the lysate from these cells establishes a solid connection between the S. rimosus protein bands for which N-terminal sequences were obtained and TNHase activity. Assays with the purified ToyJKL confirm this observation (McCarty & Bandarian, unpublished observations); studies on the mechanisms of the reaction(s) catalyzed by this protein are currently underway.

The pathway shown in Fig. 3b is not the only possible one that could have been proposed based on the available genes. For instance, one may imagine that the cyano group of preQ0 could be hydrated first, prior to being appended to a ribose moiety. Access to the entire set of recombinant proteins encoded by the Toy cluster permits these questions to be addressed.

Significance

Deazapurine-containing compounds are widespread in nature. They are utilized in biological niches as diverse as the hypermodified tRNA base queuosine, which is found in nearly all organisms, to secondary metabolites that are produced by various strains of Streptomyces. Deazapurine containing secondary metabolites, such as sangivamycin, and their chemically modified derivatives are of importance clinically as they have long been known to partition into nucleic acid pools and have demonstrated antineoplastic, antiviral, and antimicrobial activity. Current research efforts aimed at examining the therapeutic potential of natural and derivatized deazapurines, as well as development of novel deazapurine synthetic routes, are active subjects of investigation. While early radiotracer work has suggested that the biosynthetic routes toward diverse deazapurines proceed by way of similar steps, the absence of a complete set of deazapurine biosynthetic genes has made the exact chemical transformations difficult to decipher. The cluster presented here represents the first apparently complete set of genes required for the biosynthesis of a deazapurine secondary metabolite and is of significance for a number of reasons. First, the data clearly show a biosynthetic link between production of queuosine and formation of deazapurine-based secondary metabolites produced by Streptomyces. Second, access to the proteins involved in the biosynthesis of deazapurines could lead to development of semi-synthetic paths toward known or novel deazapurines. Third, the gene cluster presented in this work represents a Rosetta stone for identification of deazapurines and deazapurine biosynthetic clusters that have yet to be discovered.

Experimental Procedures

Isolation of the sangivamycin/toyocamycin biosynthesis cluster

Partial purification of TNHase, N-terminal sequencing, preparation and screening of the cosmid library, and annotation of the cluster are described in Supplementary Information. The sequence has been deposited at NCBI and has an accession number of 1074787 (BankIT number from GenBank).

Cloning, Expression and purification of recombinant S. rimosus GCH I (ToyD)

The toyD ORF was amplified from S. rimosus genomic DNA by PCR, subcloned into pGEM-T easy vector and finally into the HindIII and NdeI digested pET29a for expression of native recombinant protein. Constructs were introduced by electroporation into E. coli BL21(DE3) for protein expression. One colony from transformants containing the expression plasmid, grown at 37 °C on LB media containing 34 μg/mL kanamycin, was used to inoculate 0.05 L LB starter culture containing 34 μg/ml kanamycin. The overnight starter culture was used to inoculate four 1 L of LB media containing 34 μg/mL kanamycin. Cells were grown at 18 °C to an OD600nm ~1, at which point Zn2SO4 was added (0.1 mM final) and protein expression was induced by adding IPTG (0.1 mM). Cells were harvested after 16 h by centrifugation (20,500 × g). All subsequent steps were carried out at 4 °C. Cells were suspended in 20 mM Tris·SO4 (pH 8.0) containing 5 mM dithiothreitol (Buffer A) and 1 mM PMSF and lysed using a Branson 450D sonifier at 60% power. The cell lysate was centrifuged at 26,500 × g for 30 min. The soluble extract was loaded on a Q-Sepharose column (2.6 × 12.5 cm), which had been pre-equilibrated in Buffer A. The column was developed with a linear gradient in Buffer A first to 0.15 M NaCl over 0.1 L, then washed with Buffer A containing 0.15 M NaCl. Fractions containing ToyD (as judged by SDS-PAGE) were pooled, taken to 1 M ammonium sulfate by addition of an equal volume of a solution of Buffer A containing 2 M ammonium sulfate, and loaded on a Butyl Sepharose column (2.6 × 12.5 cm) in Buffer A containing 1 M ammonium sulfate. A 0.7 L linear gradient in Buffer A from 1 M ammonium sulfate to Buffer A containing 30% ethylene glycol was applied. Fractions containing ToyD (as judged by SDS-PAGE) were pooled, dialyzed against 2 L of Buffer A with one buffer change. The protein was concentrated in an Amicon pressure cell (YM-10 membrane). Protein was quantified with a BCA assay and metal content was determined by ICP-OES by Garratt-Callahan Company.

Activity assays for S. rimosus toyD and E. coli FolE

UV spectrophotometric assays used to determine kcat for ToyD GCH I activity were carried out in the presence of 0.1 M Tris·HCl (pH 8.0), 0.1 M KCl, and 0.2 mM GTP. The UV spectra presented in Figure 5 were obtained from a reaction mixture containing 20 mM Tris·HCl (pH 8.0), 1 mM dithiothreitol and 0.1 mM GTP, spectra shown were obtained at 5 min intervals. UV-visible spectra were recorded on an Agilent 8453 spectrophotometer after addition of 10 μM protein. The rate of conversion of GTP to H2NTP was monitored by observing changes in absorbance at 330 nm (ε = 6,300 M−1 cm−1). GCH I assays for analysis by LC-MS were carried out in the presence of 0.02 M Tris·SO4 (pH 8.0), 2 mM dithiothreitol, 0.1 M KCl, and 100 μM GTP. Reactions were allowed to proceed for 120 min at ambient temperature and quenched by centrifugation for 5 min at 14,000 × g through Nanosep (10K MWCO) Omega centrifugal devices (Pall life sciences) to remove enzyme prior to being analyzed by reverse-phase HPLC. An aliquot (20 μl) of each reaction mix was injected onto a 100 × 4.6mm 5μm Hypercarb column (Thermo Scientific) which had been pre-equilibrated in a buffer containing 95% 50 mM ammonium acetate, 0.1% diethylamine and 5% acetonitrile at a flow rate of 0.5 ml/min. The column was developed with a linear gradient from 5% to 18% ACN between 2 to 14 minutes followed by and increase from 18% to 95% ACN between 17 to 27 minutes. UV-visible spectra were obtained from 220 to 500 nm, using a ThermoFinnigan Surveyor photodiode array detector. Mass spectra were obtained in negative ion mode, scanning the m/z range of 300 to 550 amu, using ES ionization-equipped LCQ ThermoFinnigan Deca XP mass spectrometer. The instrument was set at a 42-V ionization energy and a 300 °C ion source temperature.

Cloning and assays of S. rimosus TNHase

Genes for the three subunits of TNHase (toyJ, toyK and toyL) were amplified from S. rimosus genomic DNA by PCR. DNA encoding toyJ and toyK were cloned between NdeI/XhoI and NcoI/HindIII sites of pACYC DUET-1, respectively, for simultaneous expression of both proteins. DNA encoding toyL was cloned between NdeI and HindIII sites in pET29 vector. For the assays, pACYCDuet-1 containing toyJ and toyK and pET29 containing toyL were electroporated into E. coli BL21(DE3). Control strains containing the empty pACYCDuet-1 and/or pET29 vectors were also prepared. Transformants were plated on LB agarose plates. In this and all subsequent fermentations, 34 μg/mL each of kanamycin and chloramphenicol were included. A single colony was used to inoculate a 1 mL overnight culture containing the appropriate antibiotic. The overnight culture was added to a flask containing 0.05 L of LB with the appropriate antibiotic(s) and grown to OD600nm~0.5; protein expression was induced by addition of IPTG to a final concentration of 0.1 mM. Since we suspected that TNHase is a cobalt-dependent protein (see text), the cultures were also supplemented with 0.1 mM CoCl2. After 4 h, 1 mL of the culture was removed, cells were pelleted and suspended in 50 mM KPi (pH 7.4). Cells were lysed by sonication and clear lysates were obtained by centrifugation. Total protein concentration in the cleared lysate was determined by a BCA protein assay. A volume of cell lysate containing 5 μg total protein was combined with a 0.1 mL solution containing 0.05 M KPi (pH 7.4) and 0.105 mM toyocamycin. The samples were incubated 15 min at room temperature, enzyme was removed using a Microcon YM-10 centrifugal filtration device. An aliquot of the flow-through (10 μL) was analyzed by HPLC over a Zorbax Eclipse C-18 column (4.6 × 250 mm). Baseline separation of sangivamycin and toyocamycin was achieved by isocratic elution with a mixture of 85% 0.1 M triethylammonium acetate (pH 6.8) and 15% methanol. Under these conditions, sangivamycin and toyocamycin elute at 15.2 and 19.7 min, respectively.

Supplementary Material

01

Acknowledgments

The authors wish to thank Dr. Robert Lyons of the University of Michigan DNA Sequencing Core for his help and insight on difficult to sequence DNA regions. The authors also acknowledge Alberto Rascón for a gift of the E. coli GCH I protein. RM wishes to acknowledge Science Foundation Arizona for a Graduate Fellowship. Support from the National Institutes of Health (NIH) grant GM 72623 to V.B. is gratefully acknowledged. In addition, the research of V.B. is supported (in part) by a Career Award in Biomedical Sciences from the Burroughs Wellcome Fund.

Footnotes

Data deposition: Cluster of genes described in this paper have been deposited at NCBI (Accession number bankit1074787 and EU573979)

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 1997;25:3389–3402. doi: 10.1093/nar/25.17.3389. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arakawa T, Kawano Y, Kataoka S, Katayama Y, Kamiya N, Yohda M, Okada M. Structure of thiocyanate hydrolase: A new nitrile hydratase family protein with a novel five-coordinate cobalt(III) center. J Mol Biol. 2007;366:1497–1509. doi: 10.1016/j.jmb.2006.12.011. [DOI] [PubMed] [Google Scholar]
  3. Auerbach G, Herrmann A, Bracher A, Bader G, Gutlich M, Fischer M, Neukamm M, Garrido-Franco M, Richardson J, Nar H, et al. Zinc plays a key role in human and bacterial GTP cyclohydrolase I. Proc Natl Acad Sci U S A. 2000;97:13567–13572. doi: 10.1073/pnas.240463497. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bacher A, Mailänder B. Biosynthesis of riboflavin: The structure of the purine precursor. J Biol Chem. 1973;248:6227–6231. [PubMed] [Google Scholar]
  5. Bai Y, Fox DT, Lacy JA, Van Lanen SG, Iwata-Reuyl D. Hypermodification of tRNA in Thermophilic Archaea. J Biol Chem. 2000;275:28731–28738. doi: 10.1074/jbc.M002174200. [DOI] [PubMed] [Google Scholar]
  6. Banerjee A, Sharma R, Banerjee US. The nitrile-degrading enzymes: Current status and future prospects. Appl Microbiol Biotechnol. 2002;60:33–44. doi: 10.1007/s00253-002-1062-0. [DOI] [PubMed] [Google Scholar]
  7. Bracher A, Fischer M, Eisenreich W, Ritz H, Schramek N, Boyle P, Gentili P, Huber R, Nar H, Auerbach G, Bacher A. Histidine 179 mutants of GTP cyclohydrolase I catalyze the formation of 2-amino-5-formylamino-6-ribofuranosylamino-4(3H)-pyrimidinone triphosphate. J Biol Chem. 1999;274:16727–16735. doi: 10.1074/jbc.274.24.16727. [DOI] [PubMed] [Google Scholar]
  8. Burg AW, Brown GM. The biosynthesis of folic acid. VI. Enzymatic conversion of carbon atom 8 of guanosine triphosphate to formic acid. Biochim Biophys Acta. 1966;117:275–278. doi: 10.1016/0304-4165(66)90180-2. [DOI] [PubMed] [Google Scholar]
  9. Burg AW, Brown GM. The biosynthesis of folic acid. 8. Purification and properties of the enzyme that catalyzes the production of formate from carbon atom 8 of guanosine triphosphate. J Biol Chem. 1968;243:2349–2358. [PubMed] [Google Scholar]
  10. Burroughs A, Dunaway-Mariano D, Aravind L. Evolutionary genomics of the HAD superfamily: Understanding the structural adaptations and catalytic diversity in a superfamily of phosphoesterases and allied enzymes. J Mol Biol. 2006;361:1003–1034. doi: 10.1016/j.jmb.2006.06.049. [DOI] [PubMed] [Google Scholar]
  11. El Yacoubi B, Bonnett S, Anderson JN, Swairjo MA, Reuyl-Iwata D, de Crecy-Lagard V. Discovery of a new prokaryotic type I GTP cyclohydrolase family. J Biol Chem. 2006;281:37586–37693. doi: 10.1074/jbc.M607114200. [DOI] [PubMed] [Google Scholar]
  12. Elstner EF, Suhadolnik RJ. The biosynthesis of the nucleoside antibiotics. IX. Purification and properties of guanosine triphosphate 8-formylhydrolase that catalyzes production of formic acid from the ureido carbon of guanosine triphosphate. J Biol Chem. 1971;246:6973–6981. [PubMed] [Google Scholar]
  13. Elstner EF, Suhadolnik RJ. Guanosine triphosphate-8-formylhydrolase. Methods Enzymol. 1975;43:515–520. doi: 10.1016/0076-6879(75)43113-5. [DOI] [PubMed] [Google Scholar]
  14. Gaur R, Varshney U. Genetic analysis identifies a function fo rthe queC (ybaX) gene product at an initial step in the queuosine biosynthetic pathway in Escherichia coli. J Bacteriol. 2005;187:6893–6901. doi: 10.1128/JB.187.20.6893-6901.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hopwood DA. Genetic Contributions to the Understanding Polyketide Synthases. Chem Rev. 1997;97:2465–2497. doi: 10.1021/cr960034i. [DOI] [PubMed] [Google Scholar]
  16. Isaac BG, Ayer SW, Letendre LJ, Stonard RJ. Herbicidal nucleosides from microbial sources. J Antibiotics. 1991;44:729–732. doi: 10.7164/antibiotics.44.729. [DOI] [PubMed] [Google Scholar]
  17. Kaiser J, Schramek N, Puttmer S, Schuster M, Eberhardt S, Rebelo J, Auerbach G, Bader G, Bracher A, Nar H, et al. Essential zinc ions at the catalytic sites of GTP cyclohydrolases. In: Chapman S, Perham R, Scrutton N, editors. Flavins and flavoproteins. Berlin: Rudolf Weber; 2002. pp. 241–246. [Google Scholar]
  18. Kasai H, Ohashi Z, Harada F, Nishimura S, Oppenheimer NJ, Cain PF, Leiehr JG, Minden DLv, McCloskey JA. Structure of the modified nucleoside Q isolated from Escherichia coli transfer ribonucleic acid. 7-(4,5-cis-dihydroxy-1-cyclopenten-3-ylaminomethyl)-7-deazaguanosine. Biochemistry. 1975;14:4198–4208. doi: 10.1021/bi00690a008. [DOI] [PubMed] [Google Scholar]
  19. Katayama Y, Matsushita Y, Kaneko M, Kondo M, Mizuno T, Nyunoya H. Cloning of genes coding fo rthe three subunits of thiocyanate hydrolase of Thiobacillus thioparus THI 115 and their evolutionary relationship to nitrile hydratase. J Bacteriol. 1998;180:2583–2589. doi: 10.1128/jb.180.10.2583-2589.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Katayama Y, Narahara Y, Inoue Y, Amano F, Kanagawa T, Kuraishi H. A thiocyanate hydrolase of Thiobacillus thioparus. A novel enzyme catalyzing the formation of carbonyl sulfide from thiocyanate. J Biol Chem. 1992;267:9170–9175. [PubMed] [Google Scholar]
  21. Kobayashi M, Nagasawa T, Yamada H. Enzymatic synthesis of acrylamide: A success story not yet over. Trends Biotechnol. 1992;10:402–408. doi: 10.1016/0167-7799(92)90283-2. [DOI] [PubMed] [Google Scholar]
  22. Kuo MT, Yurek DA, Coats JH, Li GP. Isolation and identification of 7,8-didemethyl-8-hydroxy-5-deazaflavin, an unusual cosynthetic factor in streptomycetes, from Streptomyces lincolnensis. J Antibiotics. 1989;42:475–478. doi: 10.7164/antibiotics.42.475. [DOI] [PubMed] [Google Scholar]
  23. McCormick JRD, Morton GO. Identity of cosynthetic factor 1 of Streptomyces aureofaciens and fragment F0 from coenzyme F420 of Methanobacterium species. J Am Chem Soc. 1982;104:4014–4015. [Google Scholar]
  24. Miller PA, Sjolander NO, Nalesnyk S, Arnold N, Johnson S, Doerschuk AP, McCormick JRD. Cosynthetic factor I, a factor involved in hydrogen-transfer in Streptomyces aureofaciens. J Am Chem Soc. 1960;82:5002–5003. [Google Scholar]
  25. Nagashima S, Nakasako M, Dohmae N, Tsujimura M, Takio K, Okada M, Yohda M, Kamiya N, Endo I. Novel non-heme iron center of nitrile hydratase with a claw setting of oxygen atoms. Nat Struct Biol. 1998;5:347–351. doi: 10.1038/nsb0598-347. [DOI] [PubMed] [Google Scholar]
  26. Nishioka H, Sawa T, Nakamura H, Iinuma H, Ikeda D, Sawa R, Naganawa H, Hiyashi C, Hamada M, Takeuchi T, et al. Isolation and structure determination of novel phosphatidylinositol kinase inhibitors, echiguanines A and B, from Streptomyces sp. J Nat Prod. 1991;54:1321–1325. doi: 10.1021/np50077a014. [DOI] [PubMed] [Google Scholar]
  27. Payne MS, Wu S, Fallon RD, Tudor G, Stieglitz B, Turner IMJ, Nelson MJ. A stereoselective cobalt-containing nitrile hydratase. Biochemistry. 1997;36:5447–5454. doi: 10.1021/bi962794t. [DOI] [PubMed] [Google Scholar]
  28. Pfleiderer W. Chemistry of naturally occurring pterins. In: Blakley RL, Benkovic SJ, editors. Folates and pterins. New York: John Wiley & Sons; 1985. pp. 44–114. [Google Scholar]
  29. Reader JS, Metzgar D, Schimmel P, de Crecy-Lagard V. Identification of four genes necessary for biosynthesis of the modified nucleoside queuosine. J Biol Chem. 2004;279:6280–6285. doi: 10.1074/jbc.M310858200. [DOI] [PubMed] [Google Scholar]
  30. Rebelo J, Auerbach G, Bader G, Bracher A, Nar H, Hosl C, Schramek N, Kaiser J, Bacher A, Huber R, Fischer M. Biosynthesis of pteridines. Reaction mechanism of GTP cyclohydrolase I. J Mol Biol. 2003;326:503–516. doi: 10.1016/s0022-2836(02)01303-7. [DOI] [PubMed] [Google Scholar]
  31. Reynolds JJ, Brown GM. The biosynthesis of folic acid: IV. Enzymatic synthesis of dihydrofolic acid from guanine and ribose compounds. J Biol Chem. 1964;239:317–325. [PubMed] [Google Scholar]
  32. Ritch PS, Glazer RI. Preferential incorporation of sangivamycin into ribonucleic acid in Sarcoma 180 cells in vitro. Biochem Pharmacol. 1982;31:259–261. doi: 10.1016/0006-2952(82)90222-2. [DOI] [PubMed] [Google Scholar]
  33. Smulson ME, Suhadolnik RJ. The biosynthesis of the 7-deazaadenine ribonucleoside, tubercidin, by Streptomyces tubercidicus. J Biol Chem. 1967;242:2872–2876. [PubMed] [Google Scholar]
  34. Sofia HJ, Chen G, Hetzler BG, Reyes-Spindola JF, Miller NE. Radical SAM, a novel protein superfamily linking unresolved steps in familiar biosynthetic pathways with radical mechanisms: functional characterization using new analysis and information visualization methods. Nucleic Acids Res. 2001;29:1097–1106. doi: 10.1093/nar/29.5.1097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Song L, Wang M, Shi J, Xue Z, Wang MX, Qian S. High resolution X-ray molecular structure of the nitrile hydratase from Rhodococcus erythropolis AJ270 reveals posttranslational oxidation of thw cysteines into sulfinic acids and a novel biocatalytic nitrile hydration mechanism. Biochem Biophys Res Commun. 2007;362:319–324. doi: 10.1016/j.bbrc.2007.07.184. [DOI] [PubMed] [Google Scholar]
  36. Spoonamore JE, Bandarian V. Understanding functional divergence in proteins by studying intragenomic homologues. Biochemistry. 2008;47:2593–2600. doi: 10.1021/bi702263z. [DOI] [PubMed] [Google Scholar]
  37. Spoonamore JE, Dahlgran AL, Jacobsen NE, Bandarian V. Evolution of New Function in the GTP Cyclohydrolase II Proteins of Streptomyces coelicolor. Biochemistry. 2006;45:12144–12155. doi: 10.1021/bi061005x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Suhadolnik RJ. Nucleoside Antibiotics. New York: Wiley-Interscience; 1970a. pp. 298–353. [Google Scholar]
  39. Suhadolnik RJ. Nucleoside antibiotics. New York: Wiley-Interscinece; 1970b. Pyrrolopyrimidine nucleosides; pp. 298–353. [Google Scholar]
  40. Suhadolnik RJ, Uematsu T. Biosynthesis of the pyrrolopyrimidine nucleoside antibiotic, toyocamycin. VII. Origin of the pyrrole carbons and the cyano carbon. J Biol Chem. 1970;245:4365–4371. [PubMed] [Google Scholar]
  41. Suhadolnik RJ, Uematsu T, Uematsu H. Toyocamycin: Phosphorylation and incorporation into RNA and DNA and the biochemical properties of the triphosphate. Biochim Biophys Acta. 1967;149:41–49. doi: 10.1016/0005-2787(67)90689-2. [DOI] [PubMed] [Google Scholar]
  42. Suhadolnik RJ, Uematsu T, Uematsu H, Wilson RG. The incorporation of sangivamycin 5′-triphosphate into polyribonucleotide by ribonucleic acid polymerase from Micrococcus lysodeikticus. J Biol Chem. 1968;243:2761–2766. [PubMed] [Google Scholar]
  43. Tavitian A, Uretsky SC, Acs G. The effect of toyocamycin on cellular RNA synthesis. Biochim Biophys Acta. 1969;179:50–57. doi: 10.1016/0005-2787(69)90121-x. [DOI] [PubMed] [Google Scholar]
  44. Uematsu H, Suhadolnik RJ. In vivo and enzymatic conversion of toyocamycin to sangivamycin by Streptomyces rimosus. Arch Biochem Biophys. 1974;162:614–619. doi: 10.1016/0003-9861(74)90223-9. [DOI] [PubMed] [Google Scholar]
  45. Uematsu T, Suhadolnik RJ. Nucleoside antibiotics. VI. Biosynthesis of the pyrrolopyrimidine nucleoside antibiotic toyocamycin by Streptomyces rimosus. Biochemistry. 1970;9:1260–1266. doi: 10.1021/bi00807a030. [DOI] [PubMed] [Google Scholar]
  46. Uematsu T, Suhadolnik RJ. Toyocamycin nitrile hydrolase. Methods Enzymol. 1975;43:759–762. doi: 10.1016/0076-6879(75)43143-3. [DOI] [PubMed] [Google Scholar]
  47. Van Lanen SG, Reader JS, Swairjo MA, de Crecy-Lagard V. From cyclohydrolase to oxidoreductase: Discovery of nitrile reductase activity in a common fold. Proc Natl Acad Sci U S A. 2005;102:4264–4269. doi: 10.1073/pnas.0408056102. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS67376-supplement-01.pdf (139.3KB, pdf)

RESOURCES