Genome-Wide Analysis Reveals the Secondary Metabolome in Streptomyces kanasensis ZX01

Streptomyces kanasensis ZX01 produces some antibiotics and a glycoprotein with antiviral activity. To further evaluate its biosynthetic potential, here we sequenced the 7,026,279 bp draft genome of S. kanasensis ZX01 and analyzed all identifiable secondary gene clusters for controlling natural products. More than 60 putative clusters were found in S. kanasensis ZX01, the majority of these biosynthetic loci are novel. In addition, the regulators for secondary metabolism in S. kanasensis ZX01 were abundant. The global regulator nsdA not only controls biosynthesis of some antibiotics, but also enhances production of glycoprotein GP-1 with antiviral activity. This study importantly reveals the powerful interplay between genomic analysis and studies of traditional natural product purification/production increasing.


Introduction
Actinomyces are an especially abundant source of secondary metabolites, contributing more than half of microbial antibiotics. Moreover, a large number of these metabolites are derived from one genus, Streptomyces. Since the publication of the genome of Streptomyces coelicolor, researchers were encouraged to mine actinomyces as a source of novel secondary metabolites [1]. Meanwhile, approaches based on genomic or bioinformatics for discovering natural products have been well developed [2]. Genomic information is very helpful to microbial natural products studies, because secondary metabolites of Streptomyces are often biosynthesized by complex gene clusters [3]. In addition, it is the genomic information that provides an opportunity for regulating the antibiotic biosynthesis at the molecular level [4]. Studies of metabolic regulation may provide a vital way for increasing production of antibiotics.
Here we reported all predicted biosynthetic gene clusters for secondary metabolites and abundant regulatory genes from the genome sequence of S. kanasensis ZX01. In addition, we evaluated the effect of nsdA, a globle regulator, on the production of glycoprotein GP-1 and on the biomass of S. kanasensis ZX01.

Strains, Plasmids and Growth Conditions
The wild-type strain of S. kanasensis ZX01 (CGMCC 4893) was used in this study. Gause-I agar was used to observe morphology, prepare spore suspensions and plate out conjugations [5]. R2YE medium was used for protoplast transformation [7]. GCBY (glucose 10 g/L, casein hydrolysate 2 g/L, beef extract 1 g/L, Yeast extract 1 g/L, pH 7.2) was used as the liquid medium. All Streptomyces cultivations were carried out at 28 • C. Escherichia coli ET12567 (A donation from Dr. Keqian Yang), containing a non-transmissible plasmid pUZ8002, was used to propagate unmethylated DNA into S. kanasensis ZX01 by conjugation. pDN221 (Invitrogen, Carlsbad, California, USA) was used as the template for the amplification of the kanamycin resistance gene. E. coli-Streptomyces shuttle plasmid, pKC1139 (A donation from Dr. Keqian Yang), was used to construct the nsdA gene-replacement vector [8].

Genome Sequencing, Annotation, and Analysis
The nucleotide sequencing was performed by Chinese National Human Genome Center (CHGC) using a Roche 454 Genome Sequencer FLX (Life Sciences, Branford, CT, USA) and was assembled using Newbler 2.3 (Life Sciences, Branford, CT, USA). Putative protein coding sequences were identified by Glimmer 3.02 [9] and GeneMark [10]. Functional annotation was based on BLASTP results with NR and KEGG databases. The tRNA and rRNA genes were predicted by tRNAscan-SE [11] and RNAmmer [12], respectively. The whole-genome shotgun project was deposited at DDBJ/EMBL/GenBank database under the accession no. LNSV00000000. The draft genome was submitted to antiSMASH for analysis of secondary metabolite gene cluster [13].

Purification and Determination of Glycoprotein GP-1
The supernatant liquid from the culture of S. kanasensis ZX01 or mutant was filtrated using a 10 kDa ultrafiltration membrane. The filtrate was lyophilized and purified by DEAE-52 column (10 × 2 cm). The 0.1 M NaCl eluent was collected, dialyzed and freeze-dried after water elution. The freeze-dried sample was dissolved in water again and then purified by HiTrap™ ConA 4B column (GE, Branford, CT, USA). The eluent was collected, dialyzed and freeze-dried. Determination of glycoprotein GP-1 was performed by HPLC (Ailgent 1260, Palo Alto, CA, USA) with TSK-GEL G2000SWXL column (7.8 × 300 mm, 5 µm). Flowing phase: ultrapure water; flow rate: 0.5 mL/min; temperature of column: 30 • C; wavelength: 280 nm.

Statistical Analyses
Statistical comparisons were analyzed by one-way analysis of variance (ANOVA). Differences were considered significant when p < 0.05.

Sequencing and Gene Annotation of the S. kanasensis ZX01
The size of S. kanasensis ZX01 draft genome is 7,026,279 bp, comprising 225 contigs (>500 bp). The genome contains at least 6245 predicted protein coding sequences (CDSs). A total of 4176 (66.87%) CDSs were assigned to known or putative functions, and 2069 (33.13%) CDSs were annotated as hypothetical protein genes. The average of GC content in this genome is 73.88%. The genome also contains seven rRNA operons and 65 tRNA genes.
The butyrolactone synthetase gene divided cluster 18 into type-I PKS domains and NRPs domains ( Figure 1). γ-Butyrolactone (GBL) autoregulators are regarded as microbial hormones that trigger secondary metabolism and morphogenesis in Streptomyces [17][18][19]. We found chemical GBL could induce expression of some PKS or NRPs gene. The expression of Ska1968, 3482, 3506 and 4292 increased dramatically under induction condition of chemical GBL (Figure 2). Ska4292 in cluster 17 did not express under general condition (wild strain was inoculated in GCBY liquid medium, then cultivated on shaker at 120 rev/min, 28 • C for 72 h); however, it strongly expressed in the presence of high concentration of chemical GBL. Perhaps, there is a relation between chemical GBL and biological GBL autoinducers. Further experiments are required to confirm this relation.

Regulators for Secondary Metabolism in S. kanasensis ZX01 Genome
Secondary metabolism has a complex regulatory network in Streptomyces. Pathway-specific regulatory genes, where each controls one antibiotic biosynthetic pathway, are at the bottom of the regulatory network, such as actII-orf4, redD, cdaR and mmyR [20]. Global regulators perform the highest-level regulation and affect the morphological and physiological differentiation. There are at least 466 putative regulators in S. kanasensis ZX01 genome, belonging to LuxR, MerR, MarR, AraC, GntR, ArsR family, et al. Some of them, as positive or negative regulators, probably have effects on the secondary metabolism. For examples, nsdA negatively regulates the sporulation and antibiotic production in Streptomyces. The nsdA disruption caused the overproduction of spores and three antibiotics (actinorhodin, calcium-independent antibiotic, epoxycyclopentanone methylenomycin) in S. coelicolor [21]. The Streptomyces bingchengensis ∆nsdA mutant produced more pigments and spores; meanwhile, milbemycin A4 and nanchangmycin were over-produced in the ∆nsdA mutant [22]. could induce expression of some PKS or NRPs gene. The expression of Ska1968, 3482, 3506 and 4292 increased dramatically under induction condition of chemical GBL (Figure 2). Ska4292 in cluster 17 did not express under general condition (wild strain was inoculated in GCBY liquid medium, then cultivated on shaker at 120 rev/min, 28 °C for 72 h); however, it strongly expressed in the presence of high concentration of chemical GBL. Perhaps, there is a relation between chemical GBL and biological GBL autoinducers. Further experiments are required to confirm this relation.

Regulators for Secondary Metabolism in S. kanasensis ZX01 Genome
Secondary metabolism has a complex regulatory network in Streptomyces. Pathway-specific regulatory genes, where each controls one antibiotic biosynthetic pathway, are at the bottom of the regulatory network, such as actII-orf4, redD, cdaR and mmyR [20]. Global regulators perform the highest-level regulation and affect the morphological and physiological differentiation. There are at least 466 putative regulators in S. kanasensis ZX01 genome, belonging to LuxR, MerR, MarR, AraC, GntR, ArsR family, et al. Some of them, as positive or negative regulators, probably have effects on the secondary metabolism. For examples, nsdA negatively regulates the sporulation and antibiotic production in Streptomyces. The nsdA disruption caused the overproduction of spores and three antibiotics (actinorhodin, calcium-independent antibiotic, epoxycyclopentanone methylenomycin) in S. coelicolor [21]. The Streptomyces bingchengensis ΔnsdA mutant produced more pigments and spores; meanwhile, milbemycin A4 and nanchangmycin were over-produced in the ΔnsdA mutant [22].
In the genome of S. kanasensis ZX01 nsdA is also present, and affects secondary metabolism and antibiotics. Glycoprotein GP-1 is a secondary metabolite with antiviral activity, producing by S. kanasensis ZX01. The ΔnsdA mutant of S. kanasensis ZX01 produced more spores and aerial hyphae than the wild strain on plate medium or liquid medium, and the production of GP-1 dramatically increased as well (Figure 3). Therefore, the global regulator, nsdA, affects colonial morphology and In the genome of S. kanasensis ZX01 nsdA is also present, and affects secondary metabolism and antibiotics. Glycoprotein GP-1 is a secondary metabolite with antiviral activity, producing by S. kanasensis ZX01. The ∆nsdA mutant of S. kanasensis ZX01 produced more spores and aerial hyphae than the wild strain on plate medium or liquid medium, and the production of GP-1 dramatically increased as well (Figure 3). Therefore, the global regulator, nsdA, affects colonial morphology and secondary metabolism of antibiotic or non-antibiotic.
In the genome of S. kanasensis ZX01 nsdA is also present, and affects secondary metabolism and antibiotics. Glycoprotein GP-1 is a secondary metabolite with antiviral activity, producing by S. kanasensis ZX01. The ΔnsdA mutant of S. kanasensis ZX01 produced more spores and aerial hyphae than the wild strain on plate medium or liquid medium, and the production of GP-1 dramatically increased as well (Figure 3). Therefore, the global regulator, nsdA, affects colonial morphology and secondary metabolism of antibiotic or non-antibiotic. At least 46 two-component systems and 7 multi-component systems exist in S. kanasensis ZX01 genome. Several two-component systems directly or indirectly affect antibiotics production. The two-component system PhoR-PhoP is the major signal transduction system for phosphate control in Streptomyces. In addition, it affects antibiotics production by influencing the transcription of biosynthesis genes, such as, S. coelicolor (actinorhodin ACT and prodigiosins RED) [23], Streptomyces griseus (candicidin) [24], Streptomyces rimosus (oxytetracycline) [25], Streptomyces natalensis (pimaricin) [26]. Similar two-component systems, which affect antibiotic production, were predicted in genome of S. kanasensis ZX01, including PhoR-P (Ska6032-6034), AfsQ1-Q2 (Ska1698-1697), RapA1-A2 (Ska2296-2297) and CutS-R (Ska4345-4346).
Guanosine tetraphosphate (ppGpp) and pentaphosphate (pppGpp) activate the expression of antibiotic biosynthetic genes, when the supply of amino acids becomes rate limiting for protein synthesis. The gene of (p)ppGpp synthase enzyme relA disruption mutant did not produce ACT in S. coelicolor [27], but, (p)ppGpp overaccumulation increased ACT and RED production [28]. Ska1310, 3074, 3731 and 3887 probably regulate secondary metabolism of S. kanasensis ZX01, because they are all related to (p)ppGpp synthesis. Furthermore, GBL is a signaling molecule as well [29], GBL synthase (ScbA) and receptor (ArpA) affect antibiotic biosynthesis [4]. Ska3485, homologous gene of ScbA, has huge potential for affecting production of secondary metabolites in S. kanasensis ZX01.
Many putative regulators for secondary metabolism were found in S. kanasensis ZX01 genome. However, most of them are unknown in the function. Hence, mining novel regulatory genes is an important work, especially global regulators, since they serve multiple purposes. Moreover, products of some gene clusters might accurately be labelled 'stress metabolites', predicted to combat stresses of a physical (desiccation, low temperature), chemical (low iron) or biological (competition) nature. Therefore, searching for factors that cause the gene clusters to be silent is a focus as well.
Genome sequencing of S. kanasensis ZX01 revealed plenty of novel biosynthetic gene clusters, the majority of them were unexpected and excited. With this available information, there is a need for more genome-guided isolation studies of natural products. Meanwhile, our regulator analysis will facilitate the process of natural product discovery and structure elucidation. Because, studies of regulator may suggest ways of increasing production levels, both at the early stages of characterizing new products and at the level of large-scale industrial production. They may also provide routes to the activation of "silent" gene cluster that are revealed by genome sequencing.