Deletion of the accramycin pathway-specific regulatory gene accJ activates the production of unrelated polyketide metabolites

Abstract The manipulation of regulatory genes has been employed to awaken cryptic metabolites in Streptomyces. Of particular interest in recent years is the effect of disruption of a pathway-specific gene to other biosynthetic pathways. Herein, we report the inactivation of the accramycin pathway-specific regulatory gene, accJ in Streptomyces sp. MA37 resulted in the production of unrelated polyketide metabolites. Through detailed mass spectrometric and spectroscopic analyses, and comparison with literature data, their structures were deduced as 3-methoxy-2-methyl-4H-pyran-4-one (1), zanthopyranone (2), propioveratrone (3), and TW94a (4). To the best of our knowledge, this is the first report of the isolation of 1–3 from bacteria. Compounds 1, 2, and 4 showed weak to moderate activity against Staphylococcus aureus, Enterococcus faecalis, and Enterococcus faecium. Propioveratrone (3) displayed better inhibitory activity (MIC = 6.3 μg/mL) than ampicillin against multi-drug resistant E. faecium K60–39 clinical isolate (MIC = 25 μg/mL), suggesting a promising structural template for the drug development targeting Enterococcus isolates. Graphical Abstract


Introduction
Streptomyces is the largest antibiotic-producing genus discovered to date, producing over two-thirds of the naturally derived antibiotics in current clinical use (Newman and Cragg 2020). This includes polyketides, terpenoids, and alkaloids that are generated by means of complex metabolic pathways (de Lima Proc opio et al. 2012).
Recent advances in genome mining have revealed that the genus Streptomyces possess a far greater chemical diversity of secondary biosynthetic gene clusters (BGCs) than the secondary metabolites that have been discovered to date. Most of these BGCs are silent or cryptic under typical laboratory culture conditions and their expression is controlled by regulatory biosynthetic genes . Although it is the most studied genus for antibiotic production, the regulation of antibiotic biosynthesis in Streptomyces remains poorly understood (Millan-Oropeza et al. 2020;Wei et al. 2018). Consequently, considerable research efforts worldwide have been directed toward a better understanding of the molecular processes underpinning such regulation for strain improvement and antibiotic discovery (Wei et al. 2018).
In recent years, there has been a growing interest in the effect of disruption of one biosynthetic gene on other metabolic pathways (Wei et al. 2018;Xia et al. 2020). Few examples of gene inactivation causing enhanced titre of unrelated biosynthetic gene cluster-encoded metabolites have been reported previously. For instance, disruption of a methyltransferase gene in the actinomycin G biosynthetic gene cluster in Streptomyces iakyrus enhanced the production of phenazinomycin (Qin et al. 2014). In another study, the inactivation of genes responsible for the biosynthesis of clavulanic acid in Streptomyces clavuligerus resulted in the overproduction of holomycin antibiotic (Qin et al. 2013).
We previously reported that deletion of accJ gene, which encoded multiple antibiotic resistance regulator (MarR) resulted in the overproduction of accramycin metabolites in Streptomyces sp. MA37, indicating that accJ is a repressor gene . Likewise, ForJ, a homolog of AccJ was also shown to repress the production of formicamycins in Streptomyces formicae KY5 (Devine et al. 2021). Herein, we report that knocking out of the accramycin pathway-specific regulatory gene, accJ resulted in previously unnoticed metabolites in MA37, leading to the hypothesis of redirecting the carbon flux to other polyketide pathways. Further chemical investigation of the DaccJ extract led to the isolation and characterization of four polyketides that were unrelated to the accramycin biosynthetic pathway, namely, 3-methoxy-2-methyl-4H-pyran-4-one (1), zanthopyranone (2), propioveratrone (3), and TW94a (4). Compounds 1-3 were identified for the first time from bacteria. Their structures were elucidated by analysis of the nuclear magnetic resonance and mass spectrometric data, and by comparison with literature data.

Results and discussion
The DaccJ mutant strain was cultured in ISP2 broth at 28 C at 180 rpm for 7 days. Subsequently, diaionV R HP-20 was added to the cultures, followed by overnight incubation at the same temperature and shaking conditions. The cultures were then filtered, the resin was extracted with methanol, and the collected filtrate was partitioned with ethyl acetate. The methanolic extracts produced 13 accramycin analogues with 330fold higher yields than the WT strain . Mass spectrometric and HPLC-UV analyses of the ethyl acetate extract revealed several isolable metabolites that were not observed in the methanol extract, particularly non-accramycin compounds (Maglangit et al. 2019;. Preliminary bioactivity screening also revealed that the ethyl acetate extract showed activity against Gram-positive pathogens. Hence, we subjected this fraction to repeated rounds of chromatographic fractionation and isolation leading to purified compounds 1-4. An exhaustive search in the literature and chemical databases (e.g. AntiBase) using the masses of compounds 1-3 as queries showed that they were known polyketides previously isolated from fungi or plants (Figure 1). To the best of our knowledge, this is the first report of the isolation of 1-3 from bacteria. Compound 1 was identified to be 3-methoxy-2-methyl-4H-pyran-4-one, which was previously isolated from the flavour concentrate of Shoyu (soy sauce) (Nunomura et al. 1980), and from the fungal endophyte, Penicillium citrinum isolated from the flowers of Ocimum tenuiflorum (Lamiaceae) (Lai et al. 2013). Compound 2 was zanthopyranone, which was previously isolated from the stems of Formosan Zanthoxylum simulans (Yang et al. 2002). Propioveratrone 3 was previously obtained from the plant roots of Baliospermum montanum (Euphorbiaceae) (Pipatrattanaseree et al. 2019) and rhizome of Acorus gramineus (Araceae) (Park et al. 2011). Compound 4 was a known fungal and bacterial metabolite named TW94a. Although 4 has been previously reported from the fungal cultures of Scolecotrichum graminis Fuckel (Tabuchi et al. 1994), Monascus purpureus BCRC 38110 (Wu et al. 2019), and from various Streptomyces species such as S. puniceus AS13 (Hussain et al. 2018) and S. coelicolor YU105 (Yu et al. 1998), it is also reported from this strain together with its antibacterial activities. This is the first report of the bioactivity of 4 against the tested pathogens. Mass spectrometric (MS) analysis of 1-4 also revealed that these compounds did not exist in the WT strain of MA37.
It is noteworthy that deletion of the accJ repressor gene in Streptomyces sp. MA37 did not only activate the accramycin biosynthetic pathway but also other unrelated pathways for 1-4 production. The corresponding biosynthetic gene cluster involved in the biosynthesis of 1-4 will be part of a future investigation in our laboratory. Understanding the regulation of antibiotic biosynthesis is of crucial importance to exploiting the great metabolic diversity of Streptomyces bacteria.

Conclusions
Disruption of a pathway-specific regulatory gene perturbs other unrelated metabolic pathways. The deletion of the accJ gene in the accramycin BGC in Streptomyces sp. MA37 activated other pathways to produce polyketide metabolites, 3-methoxy-2methyl-4H-pyran-4-one (1), zanthopyranone (2), propioveratrone (3), and TW94a (4). This is the first report of 1-3 as bacterial natural products. Compounds 1, 2, and 4 exhibited weak to moderate activity against S. aureus (ATCC 25923), E. faecalis (ATCC 29212), and multi-drug resistant E. faecium clinical isolates (K59-68 and K60-39). Remarkably, compound 4 showed a MIC of 6.3 lg/mL against E. faecium K60-39, better than the antibiotic ampicillin (MIC ¼ 25 lg/mL). The results also suggested that 4 is a promising structural template for the drug development targeting Enterococcus isolates. The discovery of compounds that display better bioactivity than current antibiotics for multidrug-resistant clinical strains is of great interest in the quest for novel drugs and drug leads. The rise of antibiotic resistance reinforces the need to re-visit and re-assess the natural products that have been discovered over the past years. Understanding the complex regulatory mechanisms involved in the biosynthesis of antibiotics in Streptomyces will decipher cryptic gene clusters and fine-tune antibiotic production.