Identification of RimR2 as a positive pathway-specific regulator of rimocidin biosynthesis in Streptomyces rimosus M527

Abstract

Background: Streoptomyces rimosus M527 is a producer of the polyene macrolide rimocidin which shows activity against various plant pathogenic fungi. Notably, the regulatory mechanisms underlying rimocidin biosynthesis are yet to be elucidated.

Results: In this study, using domain structure and amino acid alignment and phylogenetic tree construction, rimR2, which located in the rimocidin biosynthetic gene cluster, was first found and identified as a larger ATP-binding regulators of the LuxR family (LAL) subfamily regulator. The rimR2 deletion and complementation assays were conducted to explore its role. Mutant M527-ΔrimR2 lost its ability to produce rimocidin. Complementation of M527-ΔrimR2 restored rimocidin production. The five recombinant strains, M527-ER, M527-KR, M527-21R, M527-57R, and M527-NR, were constructed by overexpressing rimR2 gene using the promoters permE^*, kasOp^*, SPL21, SPL57, and its native promoter, respectively, to improve rimocidin production. M527-KR, M527-NR, and M527-ER exhibited 81.8%, 68.1%, and 54.5% more rimocidin production, respectively, than the wild-type (WT) strain, while recombinant strains M527-21R and M527-57R exhibited no obvious differences in rimocidin production compared with the WT strain. RT-PCR assays revealed that the transcriptional levels of the rim genes were consistent with the changes in rimocidin production in the recombinant strains. Using electrophoretic mobility shift assays, we confirmed that RimR2 can bind to the promoter regions of rimA and rimC.

Conclusion: A LAL regulator RimR2 was identified as a positive specific-pathway regulator of rimocidin biosynthesis in M527. RimR2 regulates the rimocidin biosynthesis by influencing the transcriptional levels of rim genes and binding to the promoter regions of rimA and rimC.

Keywords: LAL regulator; RimR2; Rimocidin; Streptomyces rimosus.

Fig. 1

Gene organization of rim gene cluster in the genome of Streptomyces rimosus M527 and rimocidin biosynthetic pathway. Module 0, Module 1, Module 2, Module 3, Module 4, type I polyketide synthase; rimK, acetyltransferase; rimJ, crotony-CoA reductase; rimH, ferredoxin; rimG, cytochrome P450 monooxygenase; rimF, aminotransferase; rimE, glycosyl transferase; rimD, cholesterol oxidase; rimC, tyrosine phosphatase; rimRl, PAS-LuxR family transcriptional regulator; rimR2, rimR3, rimR4, LAL family transcriptional regulator. The arrows in gene cluster represent putative promoters. Proposed model for rimocidin and CE-108 biosynthesis in S. diastaticus var. 108 [38]

Fig. 2

Domain structure and amino acid alignment of RimR2 and related LAL family regulators. Sequence comparisons of the N-terminal Walker A and Walker B domains and C-terminal HTH between RimR2 and well-studied LAL family regulators. AmphRI, a regulator of Amphotericin biosynthesis from Streptomyces nodosus; FscRII, FscRIII, FscRIV, regulators of Candicidin biosynthesis from Streptomyces sp. FR-008; NysRI, NysRIII, regulators of nystatin biosynthesis from Streptomyces noursei ATCC 11455; TtmRI, TtmRII, TtmRIII, regulators of tetramycin biosynthesis in Streptomyces ahygroscopicus. The NCBI database accession numbers of the sequences used in this analysis are as follows: AAV37059.1(AmphRI), AAQ82552.1(FscRII), AAQ82553.1(FscRIII), AAQ82554.1 (FscRIV), AAF71778.1(NysRI), AAF71780.1 (NysRIII), AFW98290.1(TtmRI), AFW98288.1(TtmRII), and AFW98289.1 (TtmRIII)

Fig. 3

Phylogenetic analysis based on RimR2 of S. rimosus M527 and some polyene macrolide biosynthesis regulators from other Streptomyces species. Phylogenetic analysis was performed with MEGA 7.0, using the neighbor-joining method in the Jukes-Cantor model. Bootstrap values (> 50%) based on 1000 replicates were shown at the branch nodes. Bar, 0.20 substitutions per nucleotide positions

Fig. 4

Comparison of the transcriptional levels of rim genes involved in rimocidin biosynthetic gene cluster by using qRT-PCR in WT strain S. rimosus M527, mutant S. rimosus M527-ΔrimR2, complemented strain S. rimosus M527-ΔrimR2/pSET152::rimR2.^**indicates highly statistically significant results (P-value < 0.01)

Fig. 5

In vitro electrophoretic mobility-shift assay (EMSA) assay of RimR2 binding to the promoter regions of the rimocidin biosynthetic genes rimA(a), rimC (b), rimD (c), rimF (d), rimG (e), rimH (f), rimR1 (g), and its own gene rimR2 (h). The 5′-biotin labeled DNA probe containing tested promoter regions were incubated with His₆-tagged RimR2 protein. A 100-fold excess of unlabeled specific competitor was added to the competition assay, respectively. RimR2 protein binding putative promoter region of rimA gene (a), rimC gene (b), rimD gene (c), rimF gene (d), rimG gene (d), rimH gene (d), rimR1 gene (g), rimR2 gene (h). The symbols “ + ” or “ − ”in the top row indicate the presence or absence of probes and competitors. Lane 1: biotin-labeled DNA probe; lane 2: biotin-labeled DNA probe plus RimR2 protein; lane 3: a 100-fold excess of unlabeled specific competitor plus RimR2 protein. All binding experiments were performed using 0.04 pmol/μl of biotin-labeled DNA probe and 10 μg of RimR2 protein

Fig. 6

Detection and comparison of rimocidin production (a) and cell dry weight (b) of WT strain S. rimosus M527(●), recombinant strains M527-ER(■), M527-NR(▲), M527-KR(▼), M527-21R(◆) and M527-57R(○) in shake-flask culture experiment. All shake-flask fermentations were carried out in 250 ml flasks with a working volume of 40 ml at 200 rpm and 28 °C. The medium was inoculated at 5% (v/v). The error bars were calculated from three different batches of fermentation

Fig. 7

Comparison of the transcription levels of rim genes involved in rimocidin production in different strains obtained by quantitative reverse transcription-PCR (qRT-PCR). M527: S. rimosus M527; M527-KR: S. rimosus M527-KR; M527-NR: S. rimosus M527-NR; M527-ER: S. rimosus M527-ER; M527-21R: S. rimosus M527-21R; M527-57R: S. rimosus M527-57R. The cells were harvested from the fermentation broth after 36 and 72 h. Error bars were calculated by measuring the standard deviations of the data from three replicates of each sample. (^**) indicates highly statistically significant results (P-value < 0.01)