Free access
Biotechnology
Research Article
15 February 2022

Rid7C, a Member of the YjgF/YER057c/UK114 (Rid) Protein Family, Is a Novel Endoribonuclease That Regulates the Expression of a Specialist RNA Polymerase Involved in Differentiation in Nonomuraea gerenzanensis

ABSTRACT

The YjgF/YER057c/UK114 (Rid) is a protein family breadth conserved in all domains of life and includes the widely distributed archetypal RidA (YjgF) subfamily and seven other subfamilies (Rid1 to Rid7). Among these subfamilies, RidA is the only family to have been biochemically well characterized and is involved in the deamination of the reactive enamine/imine intermediates. In this study, we have characterized a protein of the Rid7 subfamily, named Rid7C, in Nonomuraea gerenzanensis, an actinomycete that is characterized by the presence of two types of RNA polymerases. This is due to the coexistence in its genome of two RNA polymerase (RNAP) β chain-encoding genes, rpoB(S) (the wild-type rpoB gene) and rpoB(R) (a specialist, mutant-type rpoB gene) that controls A40926 antibiotic production and a wide range of metabolic adaptive behaviors. Here, we found that expression of rpoB(R) is regulated posttranscriptionally by RNA processing in the 5′ untranslated region (UTR) of rpoB(R) mRNA and that the endoribonuclease activity of Rid7C is responsible for mRNA processing, thereby overseeing several tracts of morphological and biochemical differentiation. We also provide evidence that Rid7C may be associated with RNase P M1 RNA, although M1 RNA is not required for rpoB(R) mRNA processing in vitro, and that Rid7C endoribonuclease activity is inhibited by A40926, suggesting the existence of a negative feedback loop in A40926 production and a role of the endogenous synthesis of A40926 in the modulation of biochemical differentiation in this microorganism.
IMPORTANCE The YjgF/YER057c/UK114 family includes many proteins with diverse functions involved in detoxification, RNA maturation, and control of mRNA translation. We found that Rid7C is an endoribonuclease that is involved in processing of rpoB(R) mRNA, coding for a specialized RNA polymerase beta subunit that oversees morphological differentiation and A40926 antibiotic production in Nonomuraea gerenzanensis. Rid7C-mediated processing promotes rpoB(R) mRNA translation and antibiotic production, while Rid7C endoribonuclease activity is inhibited by A40926, suggesting a role of the endogenous synthesis of A40926 in modulation of biochemical differentiation in this microorganism. Finally, we show that recombinant Rid7C copurified with M1 RNA (the RNA subunit of RNase P) from Escherichia coli extract, suggesting a functional interaction between Rid7C and M1 RNA activities.

INTRODUCTION

YjgF/YER057c/UK114 (also known as Rid family) is a large family of small proteins found across all domains of life and comprises members with diverse activities; most of them remain enigmatic. Eight subfamilies, RidA and Rid1 to Rid7, have been described (1). The RidA subfamily includes members from Eukarya and Archaea and is widely distributed among Eubacteria species, with the best-characterized member being RidA from Salmonella enterica.
In Eukarya, a series of poorly defined enzymatic activities have been reported for RidA, including endoribonuclease activity, inhibitor of protein synthesis, activator of the μ-calpain protease, and mRNA decay (24). However, most biochemical studies agree on the attribution of the reactive enamine/imine intermediate deamination activity to all the members of the RidA subfamily.
S. enterica YjgF protein was demonstrated to deaminate the 2-amino acrylate (2AA) generated by pyridoxal 5′-phosphate (PLP)-dependent serine/threonine dehydratases, and hence, YjgF was renamed RidA (reactive intermediate imine deaminase A) (5). As for S. enterica YjgF, the function of most of the RidA proteins was associated with the ability to eliminate toxic enamine/imine intermediates from reactions catalyzed by PLP-dependent enzymes involved in several metabolic pathways such as the isoleucine and purine biosynthesis as well as the threonine anaerobic degradation. The enamine/imines react with the PLP cofactor and inactivate PLP-dependent enzymes (6, 7) and form Michael adducts with cysteines (8), thereby causing significant cellular damage if they accumulate within the cells, and RidA would be responsible for clearance of these reactive species (9). YjgF/YER057c/UK114 proteins have been characterized also in lower Eukarya, including Mmf1p (YIL051c) and Hmf1p YER057c from Saccharomyces cerevisiae (10). In particular, the accumulation of 2AA in the mitochondria of S. cerevisiae results in the irreversible loss of mitochondrial DNA, and Mmf1p (YIL051c) prevents stress derived from the metabolism of mitochondrial amino acid thanks to the deamination of 2AA (1). RidA was also characterized in Arabidopsis thaliana, whereby deaminating the toxic enamine/imine intermediates prevents the inactivation of many functionally important PLP-containing enzymes in plants such as branched-chain aminotransferase BCAT (IlvE) (11). Importantly, all members of the RidA subfamily have the key arginine residue (105 in S. enterica) essential for its deaminase activity.
The YjgF/YER057c/UK114 proteins are characterized by a well-preserved structural organization. They exist as homotrimers and are characterized by intersubunit clefts that enable these proteins to bind a variety of small molecules. For instance, Escherichia coli TdcF was complexed with a variety of ligands with a carboxylate group such as 2-ketobutyrate, propionate, serine, acetate, and benzoate, although the interaction with 2-ketobutyrate appeared to be the most well defined (12). Binding to these ligands could also enable the proteins to allosterically modulate their activity in response to environmental cues. It may be also relevant to note the remarkable stability of RidA structure in S. enterica (5) and in goat (13), although its functional significance is still unclear.
In this study, we have characterized a YjgF/YER057c/UK114 family protein, Rid7C, with associated endoribonuclease activity in Nonomuraea gerenzanensis. N. gerenzanensis is an industrially important actinomycete (14) that is known for its ability to produce the teicoplanin-like glycopeptide A40926 with anti-Neisseria activity (15, 16). A40926 is the precursor of dalbavancin, a novel second-generation lipoglycopeptide antimicrobial with excellent activity against resistant Gram-positive pathogens, including methicillin-resistant Staphylococcus aureus (17).
A peculiar feature of this microorganism is the presence of two types of RNA polymerases. This is due to coexistence in its genome of two RNA polymerase (RNAP) β chain-encoding genes, rpoB(S) (the wild-type rpoB gene) and rpoB(R) (a mutant-type rpoB gene) (18). With respect to the rpoB(S) gene product, the product of rpoB(R) is characterized by a six-amino-acid deletion in a hypervariable region of the β lobe domain and five amino acid substitutions located in the RNA polymerase fork domain. Two out of the five amino acid substitutions in this domain, i.e., a histidine-to-asparagine substitution (H526N in E. coli numbering) in rif cluster I and serine-to-tyrosine substitution (S531Y in E. coli) in rif cluster II were associated with resistance to rifamycins and sorangicin, respectively (18).
The presence of both wild-type and mutant rpoB genes in the same genome may represent an elaborate strategy enabling certain actinomycetes to cohabit with microorganisms that produce antibiotics targeting the bacterial RNAP, minimizing, at the same time, the fitness cost often associated with antibiotic resistance. However, the more intriguing possibility is that rpoB duplication contributes to the developmental strategy of these bacteria. Indeed, it was shown that rpoB(R) controls A40926 antibiotic production and a wide range of metabolic adaptive behaviors in response to environmental pH (19). This hypothesis is supported by the evidence that Nonomuraea terrinata strains with duplicated rpoB genes exhibited in vitro much greater capability than single rpoB strains for growth, sporulation, and antibiotic production under certain stressful conditions (20). Furthermore, rpoB(R) markedly activated antibiotic biosynthesis in the wild-type Streptomyces lividans strain 1326, and also in strain KO-421, a “relaxed” mutant unable to produce the regulatory nucleotide guanosine tetraphosphate (“magic spot”), and the RpoB(R)-specific histidine-to-asparagine substitution was essential for the activation of secondary metabolism by mimicking a “stringent” phenotype (20).
Here, we provide evidence that Rid7C-associated endoribonuclease activity is involved in regulating the expression of rpoB(R) at the posttranscriptional level, thereby overseeing several tracts of morphological and biochemical differentiation in N. gerenzanensis.

RESULTS

Identification of genes coding for YjgF/YER057c/UK114 family proteins in N. gerenzanensis.

YjgF/YER057c/UK114 family members can be classified by phylogenetic analysis into eight subfamilies, including the apparently archetypal RidA subfamily and seven other subfamilies (Rid1 to Rid7) and eukaryotic subfamilies AANH 1 and 2. RidA subfamily members are widespread among prokaryotes (Bacteria and Archaea) and Eukarya, while members of subfamilies from Rid1 to Rid7 have only been found in prokaryotes (21, 22). RidA, Rid1, Rid2, and Rid3 contain a conserved arginine residue (arginine 105 according to S. enterica numbering) described as essential for their imine deaminase activity. Subfamily members from Rid4 to Rid7 lack this conserved residue, suggesting a new role for these proteins (21).
A genome search revealed in N. gerenzanensis ATCC 39727 10 distinct YjgF/YER057c/UK114 family proteins. Two methods have been used to assign a subfamily to N. gerenzanensis Rid proteins, the analysis of (i) the molecular phylogeny by using sequences downloaded from the Conserved Domain Database (CDD; https://www.ncbi.nlm.nih.gov/Structure/cdd/cl10015) (Fig. 1), and (ii) the conservation of residues Y17, S30, G31, R105, and E120 (5, 23). Specifically, we used the cladogram obtained to identify the clades of each subfamily. These methods allowed us to tentatively assign proteins to subfamilies as follows: Rid7, SBO96935.1 (Rid7A), SBO95965.1 (Rid7B), SBP00267.1 (Rid7C), and SBO92286.1; Rid1, SBO91465.1; Rid 3, SBO98760.1; Rid6, SBO94674.1; and RidA, SBO92579.1, SBO90862.1, and SBO96592.1 (Fig. 1).
FIG 1
FIG 1 Cladogram of N. gerenzanensis YjgF/YER057c/UK114 family members. (A) Cladogram illustrates the taxonomic position of N. gerenzanensis YjgF/YER057c/UK114 family members (in red) with respect to reference members (in black or blue). The Rid proteins with assigned function were reported in blue, while the Rid proteins with a crystal structure were reported in pink. Dashed lines evidence proteins with uncertain assignation into a family. (B) Alignment shows the conservation of R105 (in red), according to nomenclature of RidA from S. Typhimurium LT2.
Y17 and E120 are present in Rid7A (SBO96935.1), Rid7B (SBO95965.1), Rid7C (SBP00267.1), and in most of the Rid7 proteins included in the cladogram (Table 1; see Fig. S1 in the supplemental material). Moreover, G31 is conserved in all Rid7 proteins, while S30 is absent in Rid7 proteins of N. gerenzanensis that show an alanine at position 30 (Table 1 and Fig. S1). R105, which was associated with deaminase activity, is not present in the N. gerenzanensis Rid7 proteins, and, in particular, Rid7B (SBO95965.1) contains the R105A missense that has been described as interfering with deaminase activity (Table 1 and Fig. S1) (5).
TABLE 1
TABLE 1 Conservation of the residues involved in deaminase activity (Y17, R105, and E120) or conserved in Rid family (S30 and G31)
Protein IDSubfamilyConservationb of:
Y17S30G31R105E120
SBO91465.1Rid1+a++++
SBO96592.1Rid2− (Y17F)a++++
SBO90862.1Rid2− (Y17F)a++++
SBO92579.1Rid2+++++
SBO98760.1Rid3++
SBO94674.1Rid6++++
SBO92286.1Rid7+++
SBO96935.1Rid7A+++++
SBO95965.1Rid7B++− (R105A)a+
SBP00267.1Rid7C+++
a
Aminoacidic substitution that cause a reduction in deaminase activity.
b
+, conserved; −, not conserved.
In bacteria, the position of a gene in the genome relative to neighboring genes may give some indication of their function. The genome location of N. gerenzanensis ATCC 39727 genes coding for YjgF/YER057c/UK114 family proteins is depicted (Fig. 2; Table S1). The genetic maps allowed us to infer a putative function of SBO90862.1 and SBO96592.1, whose genes were located in duplicated nba gene clusters coding for enzymes involved in 2-nitrobenzoate degradation. A similar gene cluster was previously described in Pseudomonas fluorescens. In this cluster, nbaF was reported to encode a YjgF/YER057c/UK114 family protein with 2-aminomuconate deaminase activity (24).
FIG 2
FIG 2 Maps of genetic loci containing N. gerenzanensis YjgF/YER057c/UK114 family member-encoding genes. The YjgF/YER057c/UK114 family member-encoding genes are marked in red. rid7A, rid7B, and rid7C were analyzed in this study.
The SBO92579.1 gene was located close to dsdC, coding for PLP-dependent d-serine deaminase, while the SBO94674.1 gene was placed in his cluster coding for genes involved in the histidine biosynthesis. Among these genes, hisC codes for PLP-dependent histidinol-phosphate aminotransferase. It is likely that the SBO92579.1 and SBO94674.1 may have a detoxifying activity by eliminating toxic intermediates in PLP-dependent reactions.
Little can be deduced about the possible functions of the remaining YjgF/YER057c/UK114 proteins. However, we can see that the SBO98760.1 gene is located downstream of yobN, coding for monoamine oxidase, and that the genes coding for SBO96935.1 and SBO95965.1 are located close and are transcribed in opposite orientations with respect to two genes coding for two DeoR and HxlR family transcriptional regulators, respectively, of unknown function in N. gerenzanensis.
We performed a genomic neighborhood and co-occurrence analysis (Enzyme Function Initiative-Enzyme Similarity Tool [EFI-EST]/GNT) (25, 26) to analyze the conservation of the Rid genomic regions shown in Fig. 2 (Fig. S2 to S5 and Tables S2 to S4). The co-occurrence analysis was performed using the EFI program (EFI-EST/GNT) (25, 26) and considering all the genomes identified by the tool and annotated by UniProt. The genomic regions proximal and distal to SBO90862.1- and SBO96592.1-encoding genes (associated with the nba cluster) were extremely conserved (Fig. S2). The genomic regions proximal and distal to SBO94674.1- and SBO95965.1-encoding genes were also highly conserved (Fig. S3 and S4). In contrast, when the genomic context analysis was carried out with the genomic regions encompassing SBO96935.1 (Rid7A)-, SBO95965.1 (Rid7B)-, SBO98760.1-, and SBO92579.1-encoding genes, a low degree of conservation was found (Fig. S5). Finally, no conservation was found with the genomic regions encompassing SBP00267.1 (Rid7C)- and SBO92286.1-encoding genes (Fig. S5). Based on the genome context, Rid1 (SBO91465.1) and Rid2 (SBO90862.1, SBO96592.1, and SBO92579.1) could exert a metabolic function. In contrast, no function can be clearly attributed to the other Rid proteins.

Expression of rid7C in S. lividans harboring N. gerenzanensis rpoB(R) boosts antibiotic production.

To gain some functional insight into the role of YjgF/YER057c/UK114 family proteins from N. gerenzanensis on secondary metabolism, we decided to express some of them in a heterologous host, the model actinomycete S. lividans strain 1326. We focused our attention on Rid7 subfamily homologs SBO96935.1 (Rid7A), SBO95965.1 (Rid7B), and SBP00267.1 (Rid7C), whose function was completely unknown.
We chose S. lividans 1326 as host because, on the basis of genomic data, this strain would encode only two YjgF/YER057c/UK114 family proteins, EOY52403.1 and EOY47019.1, that are very distantly related to the N. gerenzanensis Rid7B and Rid7C, with identities at the level of amino acid sequences of 47% and 44%, respectively, thereby suggesting different functions with respect to N. gerenzanensis proteins. Moreover, importantly, we noticed a different pattern of transcript initiation sites of rpoB(R) mRNA in S. lividans transconjugant compared to N. gerenzanensis (20) (see below) and imputed this difference to a specific mRNA processing event that occurred in N. gerenzanensis and not in S. lividans. We therefore hypothesized a possible involvement of YjgF/YER057c/UK114 proteins (with endoribonuclease activity) that could not be associated with known metabolic/detoxifying functions also on the basis of the localization of the corresponding genes in the genetic map.
N. gerenzanensis Rid7A-, Rid7B-, or Rid7C-encoding genes, including their respective promoter region, were subjected to PCR amplification, and the amplicons were inserted into the HindIII-cloning site of the E. coli-Streptomyces shuttle plasmid vector pTYM-18. The same amplicons were also inserted into pTYM-rpoB(R), a pTYM-18 derivative plasmid harboring the rpoB(R) gene from N. gerenzanensis. The resulting plasmids pTYM-Rid7A, pTYM-Rid7B, and pTYM-Rid7C (pTYM-18 derivatives), and pTYM-rpoB(R)-Rid7A, pTYM-rpoB(R)-Rid7B, and pTYM-rpoB(R)-Rid7C were transferred to S. lividans strain 1326 by conjugation. The corresponding transconjugants are here indicated as S. lividans pTYM-18, S. lividans rid7A, S. lividans rid7B, S. lividans rid7C, S. lividans rpoB(R), S. lividans rpoB(R) rid7A, S. lividans rpoB(R) rid7B, and S. lividans rpoB(R) rid7C.
Growth and antibiotic production by S. lividans 1326 and its transconjugants were analyzed in R3 and R4 broths. R4 broth was characterized by a lower content of yeast extract than that of R3 broth and no added phosphate, potentially resulting in phosphate-limited growth. Under these conditions, production of undecylprodigiosine (Red), a mycelium-associated, red-pigmented antibiotic, and actinorhodin (Act), a diffusible, blue-pigmented antibiotic, is not subject to stringent control in Streptomyces coelicolor A3(2) (27, 28). All transconjugants exhibited similar growth curves in both R3 and R4 broths (Fig. 3 and Fig. S6), demonstrating that Rid7A-, Rid7B-, and Rid7C-encoding genes did not affect growth kinetics. Also, Red and Act productions did not change significantly in S. lividans pTYM-18, S. lividans rid7A, S. lividans rid7B, and S. lividans rid7C compared to the wild-type strain S. lividans 1326 (Fig. S6).
FIG 3
FIG 3 Growth and antibiotic production in S. lividans transconjugants. (A and B) Growth curves of S. lividans transconjugants rpoB(R), rpoB(R) rid7A, rpoB(R) rid7B, and rpoB(R) rid7C cultivated in either R3 (A) or R4 (B) broth. Biomass is expressed as dry weight (g/L). (C and D) Undecylprodigiosine (Red) production at 24, 48, 72, and 96 h by S. lividans transconjugants cultivated in either R3 (C) or R4 (D) broth. Red production was evaluated spectrophotometrically measuring the absorbance at 633 nm. (E and F) Actinorhodin (Act) production at 24, 48, 72, and 96 h by S. lividans transconjugants cultivated in either R3 (E) or R4 (F) broth. Act production was evaluated spectrophotometrically measuring the absorbance at 533 nm. The concentration of antibiotics (Red and Act) was measured in triplicate and reported as milligram of antibiotic per gram of biomass ± SD.
Consistent with previous findings (20), Red and Act yields were dramatically increased in S. lividans rpoB(R) with respect to the wild type or S. lividans pTYM-18 control strain (Fig. 3 and Fig. S6), and here, we found that antibiotic yields were further and significantly (P < 0.05) increased by a factor of 1.4 to 2.2 (depending on the antibiotic and the time point) in S. lividans rpoB(R) rid7C (Fig. 3). In contrast, Red and Act productions did not change significantly in S. lividans rid7A compared to S. lividans rpoB(R), while they were slightly increased in S. lividans rid7B only at some time points (Fig. 3). This finding demonstrated an ability of rid7C to further boost antibiotic production in the presence of rpoB(R).

Rid7C affects processing of rpoB(R) 5′ untranslated region mRNA in S. lividans.

The effects of rid7C on Red and Act production in S. lividans prompted us to explore the possibility that Rid7C could be involved in regulation of rpoB(R) expression. It was previously shown that in N. gerenzanensis, two major developmentally regulated rpoB(R) transcription start sites, TSS1 and TSS2 (Fig. 4A), are located 43 and 126 nucleotides (nt) upstream of the putative translational start codon, respectively. In particular, it was reported that the levels of transcripts starting at TSS2 increased markedly during late growth, thereby reinforcing rpoB(R) expression during the stationary phase (20). It was also shown that, in contrast to what happens in N. gerenzanensis, the rpoB(R) transcript in the S. lividans transconjugant did not start at the TSS2 but only at the TSS1.
FIG 4
FIG 4 Structural features and primer extension analysis of rpoB(R) 5′-UTR mRNA. (A) Minimum free energy structure of rpoB(R) 5′-UTR mRNA obtained by using RNAfold tool from ViennaRNA package. The color scale shows base pair probability as defined by the package. Transcription start sites 1 (TSS1) and 2 (TSS2), UUG translation start site, and Shine-Dalgarno sequence are indicated. (B) Primer extension analysis. (B, Left) Total RNAs from N. gerenzanensis grown in YS broth for 72, 120, or 168 h at either pH 6.0 or 9.5, and from S. lividans rpoB(R) grown in R3 broth for 72 h were analyzed by primer extension using the Pex2 primer spanning nucleotides 11 to 30 from the rpoB(R) translational start codon (TTG). (B, Right) N. gerenzanensis was cultivated in YS broth at pH 6.0 for 72 h, and then RNA synthesis was blocked by addition of actinomycin D. Total RNAs were extracted and analyzed by primer extension using the Pex2 primer. (C) Total RNAs from N. gerenzanensis grown in YS broth for 120 h at pH 6.0 and from S. lividans rpoB(R) rid7C grown in R3 broth for 72 h were analyzed by primer extension using the Pex2 primer. Bands corresponding to TSS1 and TSS2 are shown.
The physiological significance of the two transcription start sites could be to enhance the production of different classes of secondary metabolites (tabtoxin-like β-lactams, angucycline, lantipeptide, and A40926 biosynthetic precursors) and sulfur and nitrogen uptake- and metabolism-related proteins whose expression is positively controlled by rpoB(R) to better face harsh conditions during the stationary phase and/or growth in very alkaline soils (19).
We here confirmed these findings by S1 mapping experiments showing that the rpoB(R) transcripts starting at TSS2 increased during late growth and that rpoB(R) transcription levels were high under alkaline (pH 9.5) than under moderately acidic (pH 6.0) conditions (Fig. 4B; Fig. S7A and B), consistent with a role of rpoB(R) in the adaptation to alkaline conditions. Moreover, we demonstrated that the amount of transcript starting at TSS2 accumulated after transcription inhibition by actinomycin D treatment, while, under the same treatment, the amount of transcript starting at TSS1 almost disappeared (Fig. 4B). This result strongly suggested that TSS2 transcript originates from an mRNA processing event.
The processing event could affect rpoB(R) expression. Indeed, an in silico analysis indicated that the rpoB(R) 5′ untranslated region (5′-UTR) mRNA may fold into a secondary structure that could sequester the Shine-Dalgarno sequence, possibly preventing full rpoB(R) expression (Fig. 4A). The mRNA cleavage at TSS2 would facilitate the exposure of the Shine-Dalgarno sequence.
Intriguingly, the processing event did not occur in S. lividans rpoB(R) (Fig. 4B), possibly limiting rpoB(R) expression, and it may be interesting to note that the processing event could be restored by heterologous rid7C expression in S. lividans (Fig. 4C), suggesting an endoribonuclease activity of Rid7C.

N. gerenzanensis Rid7C performs site-specific processing of rpoB(R) 5′-UTR mRNA in vitro.

To prove a possible endoribonuclease activity associated with the Rid7 proteins of N. gerenzanensis, rid7A, rid7B, and rid7C genes were cloned into pET21b plasmid, and histidine-tagged Rid7A, Rid7B, and Rid7C proteins were overexpressed in E. coli BL21(DE3) and purified by affinity chromatography (Fig. S8). In parallel, a DNA fragment corresponding to the rpoB(R) 5′-UTR mRNA was cloned into pT7/7 vector plasmid (see Materials and Methods) in order to obtain a 276-nt-long suitable RNA substrate containing the secondary structure of the rpoB(R) 5′ UTR for endoribonuclease activity assays (Fig. 5A).
FIG 5
FIG 5 RNase activity assay. (A) Minimum free energy structure of the RNA substrate containing the rpoB(R) 5′ UTR, which was produced by in vitro transcription with pT7-rpoB(R)-5′-UTR plasmid. The RNA structure was obtained by using RNAfold tool from ViennaRNA package. EcoRI and HindIII cloning sites for pT7-rpoB(R)-5′-UTR construction starting from pT7/7 plasmid vector are shown. pT7-rpoB(R)-5′ UTR was linearized by HindIII restriction before in vitro transcription. (B) The RNA substrate containing the rpoB(R) 5′ UTR was incubated with purified recombinant Rid7A, Rid7B, Rid7C, and RNA free-Rid7C* in 50 mM Tris-HCl at 30°C for 90 min, with or without MgCl2 at pH 7.4 and with MgCl2 at pH 7.8. Reaction products were analyzed by electrophoresis in 2% agarose gel. In the first lane, the RNA substrate was incubated in the absence of proteins as a control. The blue arrows indicate the RNA bound to Rid7C; the riboprobe and its degradation products are indicated by red and black arrows, respectively. (C) Products of RNase activity assay at pH 7.8 described for panel B were also electrophoresed on polyacrylamide denaturing gels.
The purified proteins were incubated in the presence of the RNA substrate either in the absence (Fig. 5B, left; Fig. S7C) or in the presence of 1 mM MgCl2 (Fig. 5B, middle and right; Fig. S7D and E) under different pH conditions. The reaction products were analyzed by electrophoresis on nondenaturing agarose gel. In all the samples incubated with Rid7C, the presence of an additional band migrating slower than that of the RNA substrate could be noted. This band was not visible in the samples incubated with either Rid7A or Rid7B. Considering the high stability shown by RidA proteins (5, 13), a further purification step of the Rid7C protein performed using denaturing conditions such as 0.1 M NaOH or 8 M urea was able to remove the additional band from the protein (here referred to as Rid7C*) (Fig. 5B).
Incubation in the absence of MgCl2 had no effect on the RNA substrate. When the samples were incubated with MgCl2 at pH 7.4, no changes were observed with respect to the control sample consisting of the untreated substrate RNA (Fig. 5B, middle). Under these conditions, a band migrating slightly faster than that of the RNA substrate was visible, possibly determined by conformational changes of the RNA substrate induced by MgCl2. In contrast, when the samples were incubated with MgCl2 at 30°C under more alkaline conditions (pH 7.8), degradation products could be detected in the samples incubated with either Rid7A or Rid7B, and mostly in the sample incubated with either Rid7C or Rid7C* (Fig. 5B, right). Under denaturing conditions, time course experiments showed a gradual and faster reduction of riboprobe amount in samples treated with either Rid7C or Rid7C* with respect to those treated with either Rid7A or Rid7B (Fig. 5C). A concomitant increment of sharper and more persistent residual fragments was observed in the Rid7C*-treated samples and, to a lesser extent, in those treated with Rid7C.
The 5′ ends of the degradation products were then identified and mapped by 5′-rapid amplification of cDNA ends (5′-RACE) experiments (Fig. S9). In these experiments, the RNA substrate was incubated in the presence of Rid7C for 90 min at 30°C and pH 7.8, and the resulting products were analyzed as detailed in Materials and Methods. DNA sequencing of 5′ RACE products led to identify the 5′ ends of two degradation products, S2 (Fig. S9A, electropherogram on the top), and S3 (Fig. S9A, electropherogram on the bottom). S2 and S3 cleavage sites were mapped in the secondary structure of the RNA substrate (Fig. 5A). The data demonstrated that S2 corresponded to the same TSS2 detected in vivo (Fig. 4A), while the additional S3 cleavage sites were located in a distinct loop 30 nt upstream of S2 (Fig. 5A). Importantly, the same S2 and S3 cleavage sites were also mapped in degradation products of the RNA substrate obtained with Rid7C* (Fig. S9B).

Molecular docking simulations reveal Rid7A, Rid7B, and Rid7C complexed with rpoB(R) leader mRNA.

Structural analysis of the models obtained with I-TASSER showed that Rid7A, Rid7B, and Rid7C were characterized by a similar structure formed by two α-helices and a β-sheet (Fig. 6A to C). Homomer prediction suggested that these proteins may form a trimeric complex in which each subunit is arranged by turning the sheet inside, while the α-helices are facing outward (Fig. 6D to F). Various three-dimensional crystal structures were used as the templates to build I-TASSER models, including 1ONI (human p14.5, a translational inhibitor of the Yjgf/UK114 protein family), while 4O4I (tubulin-laulimalide-epothilone A complex) three-dimensional crystal structure was used to predict the Rid7C ligand binding site. The quality scores of the obtained models (both monomeric and trimeric structures) are reported in Table S5.
FIG 6
FIG 6 In silico models of Rid7 proteins and docking with rpoB(R) riboprobe model. (A to C) Rid7A (A), Rid7B (B), and Rid7C (C) models obtained by I-TASSER. Surface views of trimeric structure predicted by Homomer (left, front view; middle, back view; right, lateral view). (D to F) Trimeric structure predicted by Homomer of Rid7A (D), Rid7B (E), and Rid7C (F). (G to I) In silico interaction between Rid7A (G), Rid7B (H), and Rid7C (I) with riboprobe. (J) Focus on the docking complex showing the H-bond between the cut site and Rid7C. TSS1 and TSS2 cut sites were colored using pink.
Rid7A, Rid7B, and Rid7C I-TASSER models were used for protein-RNA docking simulations by using HDOCK. The three-dimensional structure of rpoB(R) leader mRNA was predicted by RNAComposer. The rpoB(R) start codon is marked yellow, while the S2 and S3 cleavage sites are marked pink. Rid7C-rpoB(R) leader mRNA docking complex is illustrated in Fig. 6G. Rid7C appears to fit into a V-shaped structure formed by the rpoB(R) leader mRNA establishing contacts with the two cleavage sites (Fig. 6G). The same arrangement was found in Rid7A- and Rid7B-rpoB(R) leader mRNA-docking complexes (Fig. 6H and I). Contacts (red) and the H-bonds (cyan) were visualized by Find Cash/Contact and Find H-bond tools of USCF Chimera package (Fig. 6J). The estimated GES (global energy value; kcal/mol) for each complex was calculated using HDOCK (29) and FireDock (30) (Fig. S10). The GES values suggest that all examined complexes have favorable binding energies. However, only the complex obtained using the Rid7C model showed a hydrogen bond positioned exactly at one of the cleavage sites.

Recombinant N. gerenzanensis Rid7C copurifies with E. coli RNase P RNA component.

The ethidium bromide-stained band migrating slower than that of the RNA substrate in samples treated with Rid7C (Fig. 5B, blue arrow) caught our attention because there is very recent evidence that human RNase UK114 (also known as HRSP12, RIDA, or 14.5-kDa translational inhibitor protein) directly interacts with RNase P/MRP and YTHDF2 (N6-methyladenosine reader protein) and that a YTHDF2-HRSP12-RNase P/MRP axis contributes to m6A-mediated RNA decay (31, 32). Eukaryotic RNase P and its close relative RNase MRP are essential endoribonucleases that are formed by ribonucleoprotein complexes containing common protein components and specific protein and RNA components (33). Eukaryotic RNase P and RNase MRP function as ribozymes similar to bacterial RNase P that comprises an RNA catalytic component (M1 RNA encoded by rnpB) and a noncatalytic protein component (C5 protein encoded by rnpA) (34).
These premises led us to investigate whether the slower migrating band (Fig. 5B, blue arrow) could indicate the presence of the RNA component of the E. coli RNase P. A test was carried out by reverse transcriptase PCR (RT-PCR) using 200 ng of purified recombinant Rid7A, Rid7B, and Rid7C protein samples and rnpB-specific primers (Fig. 7A). Control RT-PCRs with rid7A-, rid7B-, and rid7C-specific primers targeting these overexpressed genes were performed in parallel to rule out the possibility of nonspecific RNA contamination of the recombinant protein samples. The results demonstrated that an rnpB-specific amplicon of expected size (377 bp) could be clearly detected in the Rid7C sample and barely detected also in Rid7A and Rid7B samples. No M1 amplimer was detected when purified Rid7C* was used as the sample in the RT-PCR. DNA sequencing confirmed that the amplicon corresponded to the M1 RNA component of E. coli RNase P. Sequence similarity between E. coli M1 RNA and N. gerenzanensis M1 RNA was determined with Clustal Omega. According to this analysis, more than 65% of nucleotides were identical, while the overall similarity was 80% (Fig. S11).
FIG 7
FIG 7 Copurification of recombinant N. gerenzanensis Rid7C with E. coli RNase P RNA component. (A) Purified recombinant Rid7A, Rid7B, Rid7C, and Rif7C* protein samples were subject to RT-PCR analysis rnpB-specific primers. Arrowhead on the left indicates a specific RT-PCR product whose size is consistent with the RT-PCR-amplified M1 RNA component of E. coli RNase P. (B) RT-qPCR was used to compare the quantity of the RNA M1 in Rid7C protein pretreated or not with A40926. The antibiotic A40926 was added before the purification of His-tagged Rid7C protein.

A40926 binding to Rid7C negatively modulates its RNase activity in vitro.

The dual role of rpoB(R) in the induction of A40926 antibiotic production and rifampin resistance in N. gerenzanensis (18, 19) led us to investigate the possible effects of A40926 and rifamycin B (the natural rifampin precursor) on Rid7C RNase activity. Preincubation of Rid7C with 1:50 molar excess of A40926 (at the maximum concentration, 3 mM) resulted in a marked inhibition of RNase activity on rpoB(R) 5′-UTR RNA substrate compared to control (Fig. 8A). In contrast, neither rifamycin B nor spiramycin (a macrolide antibiotic that was used as a negative control) was able to affect the Rid7C RNase activity. Moreover, the addition of A40926 to the Rid7C protein without the bound RNA (Rid7C*) showed a similar effect (Fig. S12).
FIG 8
FIG 8 Binding of A40926 to Rid7C and effect on endoribonuclease activity. (A) The rpoB(R) 5′-UTR riboprobe was incubated with purified recombinant Rid7C in 50 mM Tris-HCl at pH 7.8 in the absence or in the presence of 0.5 mM of the indicated antibiotic. Reaction products were analyzed by electrophoresis in 6% denaturing polyacrylamides gel and stained with ethidium bromide. (B) Rid7C or bovine serum albumin (BSA) was incubated with the indicated antibiotics in the range of 0.5 to 3 mM. After incubation, the reaction was analyzed on 2% native agarose gel and stained with ethidium bromide and with Coomassie brilliant blue.
We then used native agarose gels stained with either ethidium bromide or Coomassie brilliant blue to analyze the interaction of A40926 with Rid7C. It may be noted that ethidium bromide and Coomassie brilliant blue detected the same signals consistent with the presence of RNA in Rid7C (Fig. 8B). Moreover, preincubation of Rid7C with A40926 resulted in an increasingly enhanced mobility of the Rid7C complex as a function of A40926 concentration (ranging from 0.5 to 3 mM) (Fig. 8B; Fig. S12C). In contrast, Rid7C mobility was not affected by preincubation with either rifamycin B or spiramycin. This finding was suggestive of an interaction between A40926 and Rid7C, which may modify either the overall charge or the oligomerization status and, hence, the enzymatic activity of Rid7C. Inhibition of the enzymatic activity of Rid7C by A40926 was expected to inhibit rpoB(R) mRNA translation and, hence, to decrease rifamycin B resistance levels. The evidence that subinhibitory A40926 concentrations considerably enhanced rifamycin B sensitivity of N. gerenzanensis growing in yeast-starch (YS) broth supports this hypothesis (Fig. S13).
The possible interaction between A40926 and Rid7C was modeled in silico. A protein-ligand AutoDock Vina docking simulation was performed using monomeric or trimeric model of Rid7C as receptor. Six different docking sites have been chosen on Rid7C surface, i.e., the α-helices on the monomer and outside the trimer barrel (sites 1 and 3, respectively), the monomer β-sheet (site 2), the two opposite barrel bases (sites 4 and 5), and, finally, the cleft between two monomers (site 6). The rifamycin B was also evaluated as ligand in docking simulation, owing to its role in modulating A40926 production in N. gerenzanensis. According to the results (Fig. S14B; Table S6), site 4 (trimer) showed the best affinity to A40926, while site 2 (monomer, β-sheet) showed the best affinity to rifamycin B. Site 4 was used to perform another virtual screening using 5 different antibiotics, including rifampin, a rifamycin B-derived antibiotic, and spiramycin and its related laulimalide, the latter identified as a potential Rid7C ligand through the I-TASSER tool (Fig. S14C; Table S7). The results confirmed that A40916 presented the best affinity energy. Finally, color Coulomb potential (Chimera USCF) suggested that A40926 may bind the Rid7C on a hydrophobic pocket through its lipid tail (Fig. S14D).
Reverse transcriptase quantitative PCR (RT-qPCR) was then performed to evaluate the possible effect of the A40926 on the binding between M1 RNA and Rid7C. We added 5 mM A40926 to the bacterial lysate containing the overexpressed Rid7C, and then the recombinant protein was purified. RT-qPCR was carried out on this sample using the primers reported in Table 2, while Rid7C purified without antibiotic pretreatment was used as a control. The results (Fig. 7B) showed that the addition of A40926 to bacterial lysate reduced M1 RNA binding by about 70%. Altogether, these findings are consistent with Rid7C-mediated modulation of rpoB(R) expression by its endogenous regulator A40926.
TABLE 2
TABLE 2 Oligonucleotides used in this study
NameSequenceaApplication
Rid7A.1-F5′-cGCTGAAGCTTCGCAGCGAGGAGACCGACG-3′rid7A cloning in pTYM-18 or pTYM-rpoB(R)
Rid7A.1-R5′-gCGACGGCGGTAAGCTTCCAGCTCGACCATC-3′rid7A cloning in pTYM-18 or pTYM-rpoB(R)
Rid7B.1-F5′-gTCGATAAGCTTTGCGGCTGGACCCACGAGTCTTC-3′rid7B cloning in pTYM-18 or pTYM-rpoB(R)
Rid7B.1-R5′-cAGTCGAAGCTTGCGGTGACCTCGGCCTCG-3′rid7B cloning in pTYM-18 or pTYM-rpoB(R)
Rid7C.1-F5′-gATGGACAAGCTTGGCGACTCTCCCTTGAGCTGAC-3′rid7C cloning in pTYM-18 or pTYM-rpoB(R)
Rid7C.1-R5′-cGAGCAAGCTTTCACCGCCACCGCCTCCAC-3′rid7C cloning in pTYM-18 or pTYM-rpoB(R)
Rid7A.2-F5′-aAGCTTCATATGACGATGCAGCGAACGG-3′Expression of His-tagged rid7A in E. coli
Rid7A.2-R5′-gAATTCAAGCTTCGCGACGGCGGTCCCCT-3′Expression of His-tagged rid7A in E. coli
Rid7B.2-F5′-aAGCTTCATATGGCTGTCACGCTCATCA-3′Expression of His-tagged rid7B in E. coli
Rid7B.2-R5′-gAATTCAAGCTTGTCGATGACCGCGGTGACCT-3′Expression of His-tagged rid7B in E. coli
Rid7C.2-F5′-aAGCTTCATATGACATCAGGAGTTCGAC-3′Expression of His-tagged rid7C in E. coli
Rid7C.2-R5′-gAATTCAAGCTTCGAGCCCGCGATCACCGC-3′Expression of His-tagged rid7C in E. coli
EtvrC5′-gTTGATGAATTCATGGTTTGACGAGGGGGCGG-3′Cloning into pT7/7 for in vitro transcription
HtvrC5′-gAAGGAAAGCTTGCGGGGACCGGCGG-3′Cloning into pT7/7 for in vitro transcription
Pex25′- TACGGGAGAGGCGTTGCGCG -3′Primer extension
M1_1_FGAAGCTGACCAGACAGTCGCRT-PCR and RT-qPCR used to detect RNA M1
M1_1_RAGGTGAAACTGACCGATAAGCRT-PCR and RT-qPCR used to detect RNA M1
M1_2_RCCTTACCTGATCCCGCTTGCRT-PCR and RT-qPCR used to detect RNA M1
a
Restriction sites underlined as follows: CATATG, NdeI; AAGCTT, HindIII; GAATTC, EcoRI.

DISCUSSION

Based on phylogeny, the YjgF/YER057c/UK114 family proteins can be distinguished into a widely distributed “archetypal” RidA (previously known as YigF in bacteria) subfamily, which is represented in all domains of life, and seven other subfamilies (Rid1 to Rid7) that are mostly confined to bacteria often co-occurring in the same microorganism with RidA (22).
Generally, there is more than one Rid protein in a bacterial genome. Some Gram-positive bacteria produce a few numbers of Rid proteins, such as Firmicutes (Clostridium difficile, 3 proteins; Bacillus subtilis, 1 protein), while other Gram-positive bacteria may produce numerous Rid proteins, such as Actinobacteria (Streptomyces coelicolor, 12 proteins) (21). Gram-negative bacteria may also produce a variable number of Rid proteins, ranging from 2 in Neisseria meningitidis to 17 in Burkholderia cepacia (21). In addition, members of RidA and Rid1 to Rid3 subfamilies are generally more numerous (S. coelicolor, 10 proteins; B. cepacia, 13 proteins) than members of Rid4 to Rid7 subfamilies (S. coelicolor, 2 proteins; B. cepacia, 4 proteins) (21).
Although RidA subfamily members share amino acid sequence and structure similarity, they have been involved in different functions. In Bacteria, YjgF was shown to have an enamine/imine deaminase activity that is responsible for accelerating the release of ammonia from reactive enamine/imine intermediates of the PLP-dependent reaction catalyzed by threonine dehydratase (5). As a consequence, YjgF proteins were renamed reactive intermediate/imine deaminase A (RidA) (9). In budding yeast, the RidA subfamily protein YER057C (also known as Mmf1p) is involved in maintenance of the mitochondrial genome by preventing the accumulation of 2AA (10). Human RidA has also been shown to have enamine deaminase activity (35). In mammals, RNase UK114 (also known as HRSP12, RIDA, or 14.5-kDa translational inhibitor protein) is an endoribonuclease responsible for the inhibition of the translation by cleaving mRNA (2, 3). Such disparate functions are difficult to trace back to a single mechanistic model of functioning of these proteins. The possibility that these proteins are bifunctional or multifunctional (i.e., moonlighting proteins) and that the different functions can be modulated by allosteric ligands and/or posttranslational modifications cannot be excluded.
In the current study, we demonstrate that in the actinomycete N. gerenzanensis, the Rid subfamily proteins Rid7A, Rid7B, and Rid7C (Fig. 1 and 2) are directly involved in mRNA processing. Our data therefore bring the functioning of these proteins closer to that of RNase UK114, an endoribonuclease with unique properties lacking both significant sequence homology and signature sequences with other ribonucleases (2, 3, 36, 37). This protein was shown to form a trimeric structure (36, 37) and cleave RNA substrates in single-stranded regions (2). More recently, UK114/HRSP12/RIDA was demonstrated to play an essential role in the formation of a functionally active glucocorticoid receptor-mediated (GMD) mRNA decay complex (32, 38).
Consistent with this finding, we mapped Rid7C-dependent cleavage sites in single-stranded regions of the rpoB(R) 5′-UTR mRNA both in vivo (Fig. 4) and in vitro with recombinant Rid7C (see Fig. S8 in the supplemental material). The rpoB(R) 5′-UTR mRNA may fold into a secondary structure, and the mRNA processing event would facilitate the exposure of Shine-Dalgarno sequence (Fig. 4A), thereby reinforcing rpoB(R) expression during the late growth phase (Fig. 4B). This hypothesis was supported by the evidence that rid7C (and partially rid7B) expression further increased Red and Act production in S. lividans transconjugants harboring rpoB(R) (Fig. 3). The effects of rid7C on rpoB(R) expression could link Rid7C activity in actinomycetes to morphological and biochemical differentiation. Indeed, in N. gerenzanensis, rpoB(R) controls the production of different classes of secondary metabolites, including A40926, and a wide range of metabolic adaptive behaviors by overseeing a global metabolic switch during the biochemical differentiation (19). The evidence that A40926 inhibits the activity of Rid7C RNase that is, in turn, required for rpoB(R) expression may suggest the existence of a negative feedback loop on A40926 production and a role of the endogenous synthesis of A40926 in modulation of biochemical differentiation in this microorganism. To our best knowledge, there is no evidence of other YjgF/YER057c/UK114 proteins known to regulate natural product biosynthesis. Our finding also leaves open the possibility that Rid family proteins could be new targets of glycopeptide antibiotics. These hypotheses must be verified in suitable experimental settings.
Thus, it would also be interesting to investigate the regulation of rid7C expression as well as Rid7C activity during the growth of this microorganism. In addition to transcriptional and posttranscriptional levels, posttranslational level may be involved, as there is evidence that, in mammals, UK114/HRSP12/RIDA is subject to lysine succinylation (35, 39), a posttranslational modification that is also frequently occurring in Bacteria (40).
Here, we provide evidence that Rid7C activity is negatively modulated by A40926 (Fig. 8A), consistent with Rid7C-mediated modulation of rpoB(R) expression by an endogenously produced antibiotic. In silico data predict that A40926 may bind the Rid7C on a hydrophobic pocket through its lipid tail (Fig. S14). Moreover, binding experiments with native agarose gels demonstrate that preincubation of Rid7C with A40926 resulted in an increasingly enhanced mobility of the Rid7C complex as a function of A40926 concentration, suggesting that the interaction between A40926 and Rid7C may modify either the overall charge or the oligomerization status and, hence, the enzymatic activity of Rid7C (Fig. 8B). Of note, we show that recombinant Rid7C (and, eventually, Rid7A and Rid7B also) from E. coli extracts copurifies with the M1 RNA, the catalytic RNA subunit of E. coli RNase P (Fig. 7A), and we found that the addition of A40926 to bacterial lysate reduced M1 RNA binding by about 70% (Fig. 7A). These results open the possibility that A40926 interferes with the association of M1 RNA to Rid7C.
Although the ability of the Rid7 subfamily proteins to form complexes with N. gerenzanensis M1 RNA remains to be verified, our finding is consistent with very recent data on RNase P and RNase MRP in Eukarya (31, 32). A central point of this story is the role that the RNase P (or the simple catalytic RNA component) would play in the catalysis of the cleavage reactions. Characterization of the cleavage products could help define this point. Indeed, it was reported that UK114/HRSP12/RIDA cleaves RNA substrates in vitro by a mechanism involving possibly 2′,3′-cyclic phosphate intermediates and producing 3′-phosphate ribonucleotide ends (2). In contrast, RNase P cleaves RNA various substrates in vitro, generating 5′-phosphates and 3′-hydroxyls as cleavage products, and its activity is dependent on the presence of Mg2+ ions (41). We also observed strong Mg2+ dependency on processing of our substrate. However, we demonstrated the M1 RNA is not required for rpoB(R) 5′-UTR mRNA processing in vitro (Fig. 5B, right; Fig. 5C; Fig. S7B). Moreover, it may be interesting to note that in protein-RNA docking simulation Rid7C established contacts precisely in proximity of the S2 and S3 cleavage sites in the rpoB(R) 5′-UTR mRNA, which appears to wrap around the proteins with a V-shaped structure (Fig. 6). This finding seems to confirm that Rid7 subfamily proteins are directly involved in the catalytic cleavage of the rpoB(R) 5′-UTR mRNA. A dedicated and in-depth study will provide further information on the Rid7 amino acid residues involved in the interaction with RNA substrate and RNase activity.
Thus, the role of Rid7C, in addition to its intrinsic endoribonuclease activity, could be to convey the RNase P (or simply its catalytic RNA component) to specific RNA targets. Indeed, despite an original involvement of RNase P and RNase MRP only in maturation of tRNA and mitochondrial RNA processing of replication primers, respectively, it was later demonstrated that these enzymes carry out the cleavage of a wide range of RNA substrates, such as rRNAs, long noncoding RNAs, and mRNAs (33, 4244). Genetic evidence that, in bacteria, RNase P is involved in processing polycistronic mRNA dates back to 1994 (45). Therefore, a new biological role of Rid proteins in actinomycetes could be to convey RNase P on target RNA substrates. Due to the ability of Rid family proteins to bind metabolites, it can be also hypothesized that the adapter function of Rid on RNA substrates may be modulated by endogenous or exogenous metabolites. Leaving aside the interesting association with the RNA component of the RNase P whose functional implications deserve further investigation, this study demonstrated an unprecedented role of some Rid proteins in posttranscriptional regulation.

MATERIALS AND METHODS

Bacterial strains, media, and culture conditions.

Nonomuraea gerenzanensis ATCC 39727 was obtained from the American Type Culture Collection (ATCC). This strain was deposited under the genus name Actinomadura and later reclassified as Nonomuraea (46) and, recently, as type strain of novel species N. gerenzanensis (14). Yeast-starch (YS) medium was used for growth, and culture conditions were as reported previously (16). The YS medium was used as basal medium to analyze the effects of pH (which was adjusted to the desired values using HCl or NaOH) (19).
To determine rifamycin B sensitivity of N. gerenzanensis either in the absence or in the presence of A40926, bacteria were seeded onto oatmeal-yeast agar (OMYA) (18) and incubated at 28°C for 240 h. Then, the mycelium was used to inoculate 30 mL of oatmeal-yeast (OMY) (18) broth in a 300-mL baffled Erlenmeyer flask, and the culture was grown at 28°C for 72 h in a rotary shaker at 180 rpm. This preculture was used as inoculum (1:100 dilution) for determining rifamycin B sensitivity either in the absence or in the presence of A40926. To this purpose, rifamycin B or A40926 were dissolved at 10 mg/mL using dimethyl sulfoxide (DMSO) as solvent. Bacteria were then grown in 2 mL YS broth in 13-mL plastic tubes containing different rifamycin B and/or A40926 concentrations. After 24 h of growth at 28°C under shaking (180 rpm), bacterial biomass was determined using either optical density at 600 nm (OD600) or wet weight.
S. lividans strains 1326 (wild type), KO-421 (relC), and KO-422 (relC rif1) used in this study were described previously (20, 47, 48). These strains were grown on R3 or R4 medium (49) at 30°C. R4 medium was the same as R3 medium but with a reduced amount of yeast extract (0.1% instead of 0.5%) and no added KH2PO4. Culture conditions for S. lividans were as reported previously for S. coelicolor A3(2) (50).
Undecylprodigiosine (Red) and actinorhodin (Act) production was determined as described previously (51).

Genetic maps of N. gerenzanensis ridA loci and phylogenetic analysis of RidA proteins.

Starting from the annotated genome of N. gerenzanensis from the NCBI database (GenBank accession no. LT559118.1; Nonomuraea sp. ATCC 39727 isolate nono1 genome assembly, chromosome I), 10 genetic loci encoding Rid family proteins (YjgF/YER057c/UK114) were identified. The genetic maps were implemented using IBS server (Illustrator for Biological Sequences, version 1.0) (52) and coordinates of the genes reported in the genome. For each Rid gene, three genes upstream and three genes downstream were reported in the genetic map. Only for the SBO94674.1-encoding gene, seven genes mapping upstream were included because these genes were clearly related to SBO94674.1 function.
Phylogenetic trees of Rid family proteins were built using Clustal Omega (ClustalW algorithm, neighbor joining) (53), FigTree was used to manage the tree (http://tree.bio.ed.ac.uk/software/figtree/). The protein sequences used in this step were identified in the cd00448 family (YjgF_YER057c_UK114_family) using the Conserved Domain Database (CDD) (54). In this database, 9 subfamilies were reported, YjgF_YER057c_UK114_like_1 (cd02199), YjgF_YER057c_UK114_like_2 (cd06150), YjgF_YER057c_UK114_like_3 (cd06151), YjgF_YER057c_UK114_like_4 (cd06152), YjgF_YER057c_UK114_like_5 (cd06153), YjgF_YER057c_UK114_like_6 (cd06154), YjgH_like (cd02198), eu_AANH_C_1 (cd06155), and eu_AANH_C_2 (cd06156). We renamed these families Rid1 (cd02199), Rid2 (cd06150), Rid3 (cd06151), Rid4 (cd06152), Rid5 (cd06153), Rid6 (cd06154), Rid7 (cd02198), and Rid8 (AANH_C1 and AANH_C2). For each family, 5 protein sequences were downloaded from NCBI (representative proteins section). Moreover, we added 4 well-known members of the archetypal RidA subfamily, Q7CP78 RidA from S. enterica serovar Typhimurium LT2, P52758 and P52759.3 Rat-L from Rattus norvegicus, Q94JQ4 chloroplastic RidA from Arabidopsis thaliana, and P0AF93 RidA from Homo sapiens.

Gene neighborhood and co-occurrence.

The sequence of each Rid protein was submitted to the EFI-EST tool (25, 26) (option A; alignment score threshold, 10) to generate a sequence similarity network (SSN) by formed by the closest homologs in the UniProt database.
The representative node (RepNode) network was chosen (identity threshold, 100%) and submitted to EFI-GNT (neighborhood size, 10; minimal co-occurrence percentage lower limit, 20) (26). The values of co-occurrence were extracted from PFAM (55) Family Hub-Nodes Genome Neighborhood Network (GNN). The gene neighborhood analysis was performed using Genome Neighborhood Diagrams (GNDs) (26). EFI-EST/GNT automatically selected about 900 genomes for each protein, retrieving the Pfam annotation from UniProt.

Construction of recombinant S. lividans strains.

Plasmid vector pTYM-18 and derivative plasmid pTYM-rpoB(R), carrying the N. gerenzanensis rpoB(R) gene, were previously described (20, 56). To construct pTYM-Rid7A, pTYM-Rid7B, and pTYM-Rid7C plasmids, N. gerenzanensis Rid7A-, Rid7B-, or Rid7C-encoding genes, including their promoter regions, were amplified by PCR by using the primer pairs reported in Table 2, and the amplicons were inserted into the HindIII cloning site of the E. coli-Streptomyces shuttle plasmid vector pTYM-18, respectively. The oligonucleotides that were used as primer in PCR amplification are reported in Table 2. Plasmids pTYM-rpoB(R)-Rid7A, pTYM-rpoB(R)-Rid7B, and pTYM-rpoB(R)-Rid7C were similarly obtained by inserting the same amplicons into the HindIII cloning site of pTYM-rpoB(R), respectively. Plasmids were introduced into the S. lividans 1326 by conjugation with Escherichia coli GM2929/pUB307::Tn7 as described previously (47). To allow plasmid selection, conjugation medium was supplemented with kanamycin (25 μg/mL). Transconjugants were confirmed by PCR amplification and DNA sequencing.

Purification of histidine-tagged Rid7A, Rid7B, and Rid7C proteins from E. coli.

To overexpress rid7A, rid7B, and rid7C genes in E. coli Rid7A-, Rid7B-, or Rid7C-encoding genes were amplified from N. gerenzanensis genome using the primer pairs reported in Table 2. The amplified DNA fragments were digested with NdeI and HindIII and then ligated to the pET21b plasmid, obtaining the pET-Rid7A, pET-Rid7B, and pET-Rid7C constructs. Purified recombinant N. gerenzanensis Rid7A, Rid7B, and Rid7C were obtained by introducing pET-Rid7A, pET-Rid7B, and pET-Rid7C into E. coli BL21(DE3). An isolated bacterial colony carrying the recombinant plasmid was selected and grown in 50 mL of LB medium supplemented with ampicillin (100 μg/mL) to an OD600 of 0.6 at 37°C with shaking (250 rpm). Protein production was induced by addition of 1 mM IPTG (isopropyl-β-d-thiogalactopyranoside) and incubation of bacteria for 5 h at 37°C with shaking (250 rpm). His6-tagged recombinant proteins were purified by using the Ni-immobilized metal ion affinity chromatography (IMAC) resin (Bio-Rad Laboratories). The scheme used for the protein purification has been reported in Fig. S6 in the supplemental material. Buffer A includes 50 mM Tris-HCl [pH 8.0], 10 mM KCl, 5 mg/mL lysozyme, 1 mM dithiothreitol [DTT], 1.5 mM MgCl2, 0.1% Triton X-100, 2% sodium lauroyl sarcosinate, and 3 μg/mL DNase I.

RNA extraction and primer extension experiments.

Total bacterial RNA was extracted from N. gerenzanensis ATCC 39727 grown in 50 mL of YS broth at 28°C for 72 to 168 h and from S. lividans rpoB(R) transconjugants grown in 50 mL of R3 broth at 30°C for 72 h. For RNA decay experiments, N. gerenzanensis was cultivated in YS broth at pH 6.0 for 72 h, and then RNA synthesis was blocked by addition of 10 μg/mL actinomycin D as described (18). Bacteria were harvested after 0, 5, 10, 15, and 30 min of actinomycin D treatment by rapidly chilling and centrifugation. After centrifugation, mycelia were resuspended in 10 mL of TE buffer (100 mM Tris-HCl [pH 7.5], 10 mM EDTA) containing 5 mg/mL of lysozyme, and the suspensions were incubated for 10 min at room temperature. Samples were sonicated three times for 30 s each time at 0°C with a Sonifier model 250/240 sonicator (Braun Ultrasonic), and total RNA was extracted by the sodium dodecyl sulfate-hot phenol procedure (47). RNase-free DNase I was used to eliminate traces of DNA in the samples in accordance with the instructions of the manufacturer (Promega, Madison, WI). The quantity and quality of RNA preparations were assessed by spectrophotometry and agarose gel electrophoresis. Primer extension experiments were performed using the 5′-end-labeled primer Pex2 (Table 2) and SuperScript reverse transcriptase (Gibco-BRL, Grand Island, NY) according to standard procedures (57).

In vitro transcription, RNase activity assay, and 5′-RACE experiments.

To prepare the RNA substrate (riboprobe) for RNA degradation experiments, a 173-bp DNA fragment corresponding to the leader (129 bp) and the first coding (44 bp) regions of rpoB(R) was amplified by PCR using N. gerenzanensis genome as a template and the primer pair EtvrC/HtvrC (Table 2). The amplicon was digested with EcoRI and HindIII and cloned into pT7-7 plasmid, thereby obtaining pT7-rpoB(R)-5′-UTR plasmid.
After linearization with HindIII, in vitro transcription was carried out in 60 μL of reaction mixture containing 3 μg of linearized plasmid, 30 U T7 RNA polymerase, and 200 nM nucleoside triphosphates (A, G, C, and U). The reaction mixture was incubated for 60 min at 37°C and stopped by adding 3 U RNase-free DNase I for 15 min at 37°C. For in vitro RNase activity, 50 ng of purified recombinant Rid7C was incubated with 300 ng of riboprobe for 90 min in 20 μL of a reaction buffer containing in 50 mM Tris-HCl and 1 mM MgCl2. The product of nuclease activity was analyzed by electrophoresis in 2% agarose gel or in 6% denaturing (8 M urea) polyacrylamide (acrylamide/bis ratio, 19:1) gels. Four synthetic oligonucleotides (20, 30, 40, and 70 nt) and the 173-nt rpoB(R) riboprobe were used as molecular marker to calibrate the riboprobe digestion products.
We performed 5′ rapid amplification of cDNA ends (5′-1RACE) with RNA fragments obtained from Rid7C RNase activity, according to manual instruction. Specific primers for the reverse transcription and PCR amplification steps were HtvrC (Table 2), the anchor primer 5′-AGATATCGAATTCCTCGAGTTTTTTTTTTTTTTTTT-3′, and the adaptor primer 5′-AGATATCGAATTCCTCGAG-3′. The resulting amplimers were fractionated by electrophoresis on 2% agarose gel, eluted, and digested with EcoRI and HindIII. After ligation of fragments in pBluescript II, a final step of PCR was carried out directly on ligation products with the adaptor and M13 Rev primers. The sequences of the amplicons were obtained by the dye terminator method (Eurofins Genomics, Milan, Italy).

Rid7C-bound M1 RNA quantification by RT-PCR and RT-qPCR.

Reverse transcription was carried out in 20 μl with M1_R reverse primer (Table 2) and 200 ng of purified recombinant Rid7A, Rid7B, and Rid7C purified from bacterial lysate preincubated with 5 mM A40926. We used 2 μL of cDNA in the amplification reaction, using M1_F and M1_R primers. Quantitative gene expression analysis was carried out on CFX Connect real-time PCR detection system (Bio-Rad), using SYBR Select master mix for CFX (Life Technologies) and M1_F and M_2R primers (Table 1). For the normalization, an aliquot of purified proteins (200 ng) was analyzed by Western blotting. After electrophoresis, proteins were transferred electrophoretically onto nitrocellulose filter (Pall, East Hills, NY) (58). Blots were incubated with a monoclonal antibody against the His epitope (catalog no. sc-8036; Santa Cruz Biotechnology). The immune complexes were detected using peroxidase-conjugated secondary antibodies by chemiluminescence (Pierce ECL Western blotting substrate). Densitometric analysis was carried out on the Western blots using the NIH Image 1.62 software (National Institutes of Health, Bethesda, MD).
The sequence of M1 RNA from E. coli strain K-12 substrate MG165 (U00096.3) was downloaded from Rfam (https://rfam.xfam.org/) and submitted as a query sequence to BLAST to perform research, including only the taxon N. gerenzanensis (taxonomy ID 93944). Moreover, Clustal Omega was used to calculate the similarity between M1 RNA from E. coli and M1 RNA from N. gerenzanensis.

Three-dimensional modeling and docking simulations.

I-TASSER software (59) was used to construct models of N. gerenzanensis Rid7A (SBO96935.1), Rid7B (SBO95965.1), and Rid7C (SBP00267.1). The PDB files obtained by I-TASSER were used to predict homo-oligomers by Homomer (60). Secondary structures in rpoB(R) 5′-UTR mRNA were predicted by using RNAfold tool from ViennaRNA package (61, 62) at the RNAfold WebServer (http://www.viennarna.at/cgi-bin/RNAWebSuite/RNAfold.cgi). RNAComposer (63, 64) was used to model the tertiary structure of rpoB(R) mRNA leader using Centroid Fold as a secondary structure prediction method. RNA-protein docking simulations were performed by HDOCK (65), using Rid7A, Rid7B, and Rid7C PDB models as receptors and the rpoB(R) mRNA leader PDB model as a ligand. Results were visualized by UCSF Chimera (66).
Docking simulations of complexes involving Rid7C and different antibiotics were carried out by using AutoDock Vina (67). Either monomeric or trimeric Rid7C was used as receptor, while 5 antibiotic structures downloaded from PubChem, A40926 (CID 133082069), rifamycin B (CID 5459948), rifampin (CID 135398735), spiramycin (CID 5284619), and laulimalide (CID 6918457), were used as ligands.

Analysis of A40926, rifamycin B, and spiramycin binding on Rid7C.

Analysis of the interaction between A40926, or rifamycin B, or spiramycin and Rid7C was performed in 15 μL of a binding reaction mixture containing 50 mM Tris-HCl (pH 7.8), 1 mM MgCl2, 200 ng Rid7C, and increasing concentrations (0.5, 1, 2, 3 mM) of the antibiotic for 10 min at 37°C. The reaction was analyzed by electrophoresis in 2% agarose gel with Tris-acetate-EDTA (TAE) 1× buffer. After the electrophoresis, gel was stained with ethidium bromide for 15 min and subsequently with Coomassie blue staining solution (Coomassie blue R-250 0.005% [wt/wt], 40% methanol, and 10% acetic acid) overnight. The effect of the antibiotic-Rid7C interaction on RNase activity was also evaluated. RNase activity of antibiotic-Rid7C complex was also analyzed as previously described, including 0.5 mM A40926, rifamycin B, or spiramycin in the reaction buffer. The product of nuclease activity was analyzed by electrophoresis in denaturing 6% (19:1) polyacrylamide gel with 8 M urea.

Statistical analysis.

Two-tailed paired Student’s t tests were used to calculate the P values. Differences with P values of <0.05 were considered statistically significant.

ACKNOWLEDGMENTS

This work was supported partially by grants from the Italian MIUR to P.A. (PRIN 2017, grant 2017SFBFER) and from Consorzio Interuniversitario Biotecnologie (CIB; grant N. 86/19) to P.A.
P.A. conceived the experimental design and conceptualized the study; F.D., L.G., D.P., S.M.T., and A.T. performed the in vivo and in vitro experiments, and M.C. performed genomic, phylogenetic, and in silico analyses. F.D., M.C., and P.A., wrote the manuscript draft; P.A. and L.S. reviewed the manuscript draft and interpreted the experimental results. All authors critically reviewed and finally approved the manuscript.
We declare no competing financial interest.

Supplemental Material

File (jb.00462-21-s0001.pdf)
File (jb.00462-21-s0002.xlsx)
ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.

REFERENCES

1.
Ernst DC, Downs DM. 2018. Mmf1p couples amino acid metabolism to mitochondrial DNA maintenance in Saccharomyces cerevisiae. mBio 9:e00084-18.
2.
Morishita R, Kawagoshi A, Sawasaki T, Madin K, Ogasawara T, Oka T, Endo Y. 1999. Ribonuclease activity of rat liver perchloric acid-soluble protein, a potent inhibitor of protein synthesis. J Biol Chem 274:20688–20692.
3.
Oka T, Tsuji H, Noda C, Sakai K, Hong YM, Suzuki I, Muñoz S, Natori Y. 1995. Isolation and characterization of a novel perchloric acid-soluble protein inhibiting cell-free protein synthesis. J Biol Chem 270:30060–30067.
4.
Schmiedeknecht G, Kerkhoff C, Orsó E, Stöhr J, Aslanidis C, Nagy GM, Knuechel R, Schmitz G. 1996. Isolation and characterization of a 14.5-kDa trichloroacetic-acid-soluble translational inhibitor protein from human monocytes that is upregulated upon cellular differentiation. Eur J Biochem 242:339–351.
5.
Lambrecht JA, Flynn JM, Downs DM. 2012. Conserved YjgF protein family deaminates reactive enamine/imine intermediates of pyridoxal 5'-phosphate (PLP)-dependent enzyme reactions. J Biol Chem 287:3454–3461.
6.
Likos JJ, Ueno H, Feldhaus RW, Metzler DE. 1982. A novel reaction of the coenzyme of glutamate decarboxylase with L-serine O-sulfate. Biochemistry 21:4377–4386.
7.
Ueno H, Likos JJ, Metzler DE. 1982. Chemistry of the inactivation of cytosolic aspartate aminotransferase by serine O-sulfate. Biochemistry 21:4387–4393.
8.
Cooper AJ, Krasnikov BF, Niatsetskaya ZV, Pinto JT, Callery PS, Villar MT, Artigues A, Bruschi SA. 2011. Cysteine S-conjugate β-lyases: important roles in the metabolism of naturally occurring sulfur and selenium-containing compounds, xenobiotics and anticancer agents. Amino Acids 41:7–27.
9.
Lambrecht JA, Schmitz GE, Downs DM. 2013. RidA proteins prevent metabolic damage inflicted by PLP-dependent dehydratases in all domains of life. mBio 4:e00033-13.
10.
Kim JM, Yoshikawa H, Shirahige K. 2001. A member of the YER057c/yjgf/Uk114 family links isoleucine biosynthesis and intact mitochondria maintenance in Saccharomyces cerevisiae. Genes Cells 6:507–517.
11.
Niehaus TD, Nguyen TN, Gidda SK, ElBadawi-Sidhu M, Lambrecht JA, McCarty DR, Downs DM, Cooper AJ, Fiehn O, Mullen RT, Hanson AD. 2014. Arabidopsis and maize RidA proteins preempt reactive enamine/imine damage to branched-chain amino acid biosynthesis in plastids. Plant Cell 26:3010–3022.
12.
Burman JD, Stevenson CE, Sawers RG, Lawson DM. 2007. The crystal structure of Escherichia coli TdcF, a member of the highly conserved YjgF/YER057c/UK114 family. BMC Struct Biol 7:30.
13.
Degani G, Barbiroli A, Regazzoni L, Popolo L, Vanoni MA. 2018. Imine deaminase activity and conformational stability of UK114, the mammalian member of the Rid protein family active in amino acid metabolism. Int J Mol Sci 19:945.
14.
Dalmastri C, Gastaldo L, Marcone GL, Binda E, Congiu T, Marinelli F. 2016. Classification of Nonomuraea sp. ATCC 39727, an actinomycete that produces the glycopeptide antibiotic A40926, as Nonomuraea gerenzanensis sp. nov. Int J Syst Evol Microbiol 66:912–921.
15.
Goldstein BP, Selva E, Gastaldo L, Berti M, Pallanza R, Ripamonti F, Ferrari P, Denaro M, Arioli V, Cassani G. 1987. A40926, a new glycopeptide antibiotic with anti-Neisseria activity. Antimicrob Agents Chemother 31:1961–1966.
16.
Technikova-Dobrova Z, Damiano F, Tredici SM, Vigliotta G, di Summa R, Palese L, Abbrescia A, Labonia N, Gnoni GV, Alifano P. 2004. Design of mineral medium for growth of Actinomadura sp. ATCC 39727, producer of the glycopeptide A40926: effects of calcium ions and nitrogen sources. Appl Microbiol Biotechnol 65:671–677.
17.
Anderson VR, Keating GM. 2008. Dalbavancin. Drugs 68:639–651.
18.
Vigliotta G, Tredici SM, Damiano F, Montinaro MR, Pulimeno R, di Summa R, Massardo DR, Gnoni GV, Alifano P. 2004. Natural merodiploidy involving duplicated rpoB alleles affects secondary metabolism in a producer actinomycete. Mol Microbiol 55:396–412.
19.
D'Argenio V, Petrillo M, Pasanisi D, Pagliarulo C, Colicchio R, Talà A, de Biase MS, Zanfardino M, Scolamiero E, Pagliuca C, Gaballo A, Cicatiello AG, Cantiello P, Postiglione I, Naso B, Boccia A, Durante M, Cozzuto L, Salvatore P, Paolella G, Salvatore F, Alifano P. 2016. The complete 12 Mb genome and transcriptome of Nonomuraea gerenzanensis with new insights into its duplicated “magic” RNA polymerase. Sci Rep 6:18.
20.
Talà A, Wang G, Zemanova M, Okamoto S, Ochi K, Alifano P. 2009. Activation of dormant bacterial genes by Nonomuraea sp. strain ATCC 39727 mutant-type RNA polymerase. J Bacteriol 191:805–814.
21.
Irons JL, Hodge-Hanson K, Downs DM. 2020. RidA proteins protect against metabolic damage by reactive intermediates. Microbiol Mol Biol Rev 84:e00024-20.
22.
Niehaus TD, Gerdes S, Hodge-Hanson K, Zhukov A, Cooper AJ, ElBadawi-Sidhu M, Fiehn O, Downs DM, Hanson AD. 2015. Genomic and experimental evidence for multiple metabolic functions in the RidA/YjgF/YER057c/UK114 (Rid) protein family. BMC Genomics 16:382.
23.
Borchert AJ, Ernst DC, Downs DM. 2019. Reactive enamines and imines in vivo: lessons from the RidA paradigm. Trends Biochem Sci 44:849–860.
24.
Muraki T, Taki M, Hasegawa Y, Iwaki H, Lau PC. 2003. Prokaryotic homologs of the eukaryotic 3-hydroxyanthranilate 3,4-dioxygenase and 2-amino-3-carboxymuconate-6-semialdehyde decarboxylase in the 2-nitrobenzoate degradation pathway of Pseudomonas fluorescens strain KU-7. Appl Environ Microbiol 69:1564–1572.
25.
Gerlt JA, Bouvier JT, Davidson DB, Imker HJ, Sadkhin B, Slater DR, Whalen KL. 2015. Enzyme Function Initiative-Enzyme Similarity Tool (EFI-EST): a web tool for generating protein sequence similarity networks. Biochim Biophys Acta 1854:1019–1037.
26.
Zallot R, Oberg N, Gerlt JA. 2021. Discovery of new enzymatic functions and metabolic pathways using genomic enzymology web tools. Curr Opin Biotechnol 69:77–90.
27.
Chakraburtty R, Bibb M. 1997. The ppGpp synthetase gene (relA) of Streptomyces coelicolor A3(2) plays a conditional role in antibiotic production and morphological differentiation. J Bacteriol 179:5854–5861.
28.
Sun J, Hesketh A, Bibb M. 2001. Functional analysis of relA and rshA, two relA/spoT homologues of Streptomyces coelicolor A3(2). J Bacteriol 183:3488–3498.
29.
Yan Y, Zhang D, Zhou P, Li B, Huang SY. 2017. HDOCK: a web server for protein-protein and protein-DNA/RNA docking based on a hybrid strategy. Nucleic Acids Res 45:W365–W373.
30.
Mashiach E, Schneidman-Duhovny D, Andrusier N, Nussinov R, Wolfson HJ. 2008. FireDock: a web server for fast interaction refinement in molecular docking. Nucleic Acids Res 36:W229–W232.
31.
Lee Y, Choe J, Park OH, Kim YK. 2020. Molecular mechanisms driving mRNA degradation by m6A modification. Trends Genet 36:177–188.
32.
Park OH, Ha H, Lee Y, Boo SH, Kwon DH, Song HK, Kim YK. 2019. Endoribonucleolytic cleavage of m6A-containing RNAs by RNase P/MRP complex. Mol Cell 74:494–507.
33.
Jarrous N. 2017. Roles of RNase P and its subunits. Trends Genet 33:594–603.
34.
Altman S. 1989. Ribonuclease P: an enzyme with a catalytic RNA subunit. Adv Enzymol Relat Areas Mol Biol 62:1–36.
35.
Siculella L, Giannotti L, Di Chiara Stanca B, Calcagnile M, Rochira A, Stanca E, Alifano P, Damiano F. 2021. Evidence for a negative correlation between human reactive enamine-imine intermediate deaminase A (RIDA) activity and cell proliferation rate: role of lysine succinylation of RIDA. Int J Mol Sci 22:3804.
36.
Mistiniene E, Luksa V, Sereikaite J, Naktinis V. 2003. Oligomeric assembly and ligand binding of the members of protein family YER057c/YIL051c/YJGF. Bioconjug Chem 14:1243–1252.
37.
Mistiniene E, Pozdniakovaite N, Popendikyte V, Naktinis V. 2005. Structure-based ligand binding sites of protein p14.5, a member of protein family YER057c/YIL051c/YjgF. Int J Biol Macromol 37:61–68.
38.
Park OH, Park J, Yu M, An HT, Ko J, Kim YK. 2016. Identification and molecular characterization of cellular factors required for glucocorticoid receptor-mediated mRNA decay. Genes Dev 30:2093–2105.
39.
Rardin MJ, He W, Nishida Y, Newman JC, Carrico C, Danielson SR, Guo A, Gut P, Sahu AK, Li B, Uppala R, Fitch M, Riiff T, Zhu L, Zhou J, Mulhern D, Stevens RD, Ilkayeva OR, Newgard CB, Jacobson MP, Hellerstein M, Goetzman ES, Gibson BW, Verdin E. 2013. SIRT5 regulates the mitochondrial lysine succinylome and metabolic networks. Cell Metab 18:920–933.
40.
Weinert BT, Schölz C, Wagner SA, Iesmantavicius V, Su D, Daniel JA, Choudhary C. 2013. Lysine succinylation is a frequently occurring modification in prokaryotes and eukaryotes and extensively overlaps with acetylation. Cell Rep 4:842–851.
41.
Kirsebom LA, Trobro S. 2009. RNase P RNA-mediated cleavage. IUBMB Life 61:189–200.
42.
Coughlin DJ, Pleiss JA, Walker SC, Whitworth GB, Engelke DR. 2008. Genome-wide search for yeast RNase P substrates reveals role in maturation of intron-encoded box C/D small nucleolar RNAs. Proc Natl Acad Sci USA 105:12218–12223.
43.
Maida Y, Yasukawa M, Furuuchi M, Lassmann T, Possemato R, Okamoto N, Kasim V, Hayashizaki Y, Hahn WC, Masutomi K. 2009. An RNA-dependent RNA polymerase formed by TERT and the RMRP RNA. Nature 461:230–235.
44.
Wilusz JE, Freier SM, Spector DL. 2008. 3' End processing of a long nuclear-retained noncoding RNA yields a tRNA-like cytoplasmic RNA. Cell 135:919–932.
45.
Alifano P, Rivellini F, Piscitelli C, Arraiano CM, Bruni CB, Carlomagno MS. 1994. Ribonuclease E provides substrates for ribonuclease P-dependent processing of a polycistronic mRNA. Genes Dev 8:3021–3031.
46.
Monciardini P, Sosio M. 2004. Reclassification as a Nonomuraea sp. of the strain ATCC 39727, producing the glycopeptide antibiotic A40926. J Antibiot 57:68–70.
47.
Kieser T, Bibb MJ, Buttner MJ, Chater KF, Hopwood DA. 2000. Practical Streptomyces genetics. The John Innes Foundation, Norwich, UK.
48.
Lai C, Xu J, Tozawa Y, Okamoto-Hosoya Y, Yao X, Ochi K. 2002. Genetic and physiological characterization of rpoB mutations that activate antibiotic production in Streptomyces lividans. Microbiology (Reading) 148:3365–3373.
49.
Shima J, Hesketh A, Okamoto S, Kawamoto S, Ochi K. 1996. Induction of actinorhodin production by rpsL (encoding ribosomal protein S12) mutations that confer streptomycin resistance in Streptomyces lividans and Streptomyces coelicolor A3(2). J Bacteriol 178:7276–7284.
50.
Hu H, Ochi K. 2001. Novel approach for improving the productivity of antibiotic-producing strains by inducing combined resistant mutations. Appl Environ Microbiol 67:1885–1892.
51.
Hu H, Zhang Q, Ochi K. 2002. Activation of antibiotic biosynthesis by specified mutations in the rpoB gene (encoding the RNA polymerase beta subunit) of Streptomyces lividans. J Bacteriol 184:3984–3991.
52.
Liu W, Xie Y, Ma J, Luo X, Nie P, Zuo Z, Lahrmann U, Zhao Q, Zheng Y, Zhao Y, Xue Y, Ren J. 2015. IBS: an illustrator for the presentation and visualization of biological sequences. Bioinformatics 31:3359–3361.
53.
Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Söding J, Thompson JD, Higgins DG. 2011. Fast, scalable generation of high‐quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol 7:539.
54.
Marchler-Bauer A, Derbyshire MK, Gonzales NR, Lu S, Chitsaz F, Geer LY, Geer RC, He J, Gwadz M, Hurwitz DI, Lanczycki CJ, Lu F, Marchler GH, Song JS, Thanki N, Wang Z, Yamashita RA, Zhang D, Zheng C, Bryant SH. 2015. CDD: NCBI's conserved domain database. Nucleic Acids Res 43:D222–D226.
55.
El-Gebali S, Mistry J, Bateman A, Eddy SR, Luciani A, Potter SC, Qureshi M, Richardson LJ, Salazar GA, Smart A, Sonnhammer ELL, Hirsh L, Paladin L, Piovesan D, Tosatto SCE, Finn RD. 2019. The Pfam protein families database in 2019. Nucleic Acids Res 47:D427–D432.
56.
Onaka H, Taniguchi S, Ikeda H, Igarashi Y, Furumai T. 2003. pTOYAMAcos, pTYM18, and pTYM19, actinomycete-Escherichia coli integrating vectors for heterologous gene expression. J Antibiot 56:950–956.
57.
Sambrook J, Maniatis T, Fritsch EF. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
58.
Damiano F, Tocci R, Gnoni GV, Siculella L. 2015. Expression of citrate carrier gene is activated by ER stress effectors XBP1 and ATF6α, binding to an UPRE in its promoter. Biochim Biophys Acta 1849:23–31.
59.
Yang J, Yan R, Roy A, Xu D, Poisson J, Zhang Y. 2015. The I-TASSER Suite: protein structure and function prediction. Nat Methods 12:7–8.
60.
Baek M, Park T, Heo L, Park C, Seok C. 2017. GalaxyHomomer: a web server for protein homo-oligomer structure prediction from a monomer sequence or structure. Nucleic Acids Res 45:W320–W324.
61.
Gruber AR, Lorenz R, Bernhart SH, Neuböck R, Hofacker IL. 2008. The ViennaRNA Websuite. Nucleic Acids Res 36:W70–W74.
62.
Lorenz R, Bernhart SH, Höner Zu Siederdissen C, Tafer H, Flamm C, Stadler PF, Hofacker IL. 2011. ViennaRNA package 2.0. Algorithms Mol Biol 6:26.
63.
Antczak M, Popenda M, Zok T, Sarzynska J, Ratajczak T, Tomczyk K, Adamiak RW, Szachniuk M. 2016. New functionality of RNAComposer: an application to shape the axis of miR160 precursor structure. Acta Biochim Pol 63:737–744.
64.
Popenda M, Szachniuk M, Antczak M, Purzycka KJ, Lukasiak P, Bartol N, Blazewicz J, Adamiak RW. 2012. Automated 3D structure composition for large RNAs. Nucleic Acids Res 40:e112.
65.
Yan Y, Tao H, He J, Huang SY. 2020. The HDOCK server for integrated protein-protein docking. Nat Protoc 15:1829–1852.
66.
Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE. 2004. UCSF Chimera—a visualization system for exploratory research and analysis. J Comput Chem 25:1605–1612.
67.
Trott O, Olson AJ. 2010. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem 31:455–461.

Information & Contributors

Information

Published In

cover image Journal of Bacteriology
Journal of Bacteriology
Volume 204Number 215 February 2022
eLocator: e00462-21
Editor: Anke Becker, Philipps University Marburg
PubMed: 34694905

History

Received: 10 September 2021
Accepted: 21 October 2021
Accepted manuscript posted online: 25 October 2021
Published online: 15 February 2022

Permissions

Request permissions for this article.

Keywords

  1. YjgF/YER057c/UK114 family proteins
  2. Rid endoribonuclease
  3. RNase P
  4. Actinomycetes
  5. secondary metabolism

Contributors

Authors

Fabrizio Damiano
Department of Biological and Environmental Sciences and Technologies (DiSTeBA), University of Salento, Lecce, Italy
Matteo Calcagnile
Department of Biological and Environmental Sciences and Technologies (DiSTeBA), University of Salento, Lecce, Italy
Daniela Pasanisi
Department of Biological and Environmental Sciences and Technologies (DiSTeBA), University of Salento, Lecce, Italy
Adelfia Talà
Department of Biological and Environmental Sciences and Technologies (DiSTeBA), University of Salento, Lecce, Italy
Salvatore Maurizio Tredici
Department of Biological and Environmental Sciences and Technologies (DiSTeBA), University of Salento, Lecce, Italy
Laura Giannotti
Department of Biological and Environmental Sciences and Technologies (DiSTeBA), University of Salento, Lecce, Italy
Luisa Siculella [email protected]
Department of Biological and Environmental Sciences and Technologies (DiSTeBA), University of Salento, Lecce, Italy
Pietro Alifano [email protected]
Department of Biological and Environmental Sciences and Technologies (DiSTeBA), University of Salento, Lecce, Italy

Editor

Anke Becker
Editor
Philipps University Marburg

Notes

Fabrizio Damiano and Matteo Calcagnile contributed equally to this work. Author order was determined in order of decreasing seniority.
The authors declare no conflict of interest.

Metrics & Citations

Metrics

Note:

  • For recently published articles, the TOTAL download count will appear as zero until a new month starts.
  • There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.
  • Citation counts come from the Crossref Cited by service.

Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.

View Options

Figures and Media

Figures

Media

Tables

Share

Share

Share the article link

Share with email

Email a colleague

Share on social media

American Society for Microbiology ("ASM") is committed to maintaining your confidence and trust with respect to the information we collect from you on websites owned and operated by ASM ("ASM Web Sites") and other sources. This Privacy Policy sets forth the information we collect about you, how we use this information and the choices you have about how we use such information.
FIND OUT MORE about the privacy policy