β-Lactam Resistance in Azospirillum baldaniorum Sp245 Is Mediated by Lytic Transglycosylase and β-Lactamase and Regulated by a Cascade of RpoE7→RpoH3 Sigma Factors

ABSTRACT

Bacterial resistance to β-lactam antibiotics is often mediated by β-lactamases and lytic transglycosylases. Azospirillum baldaniorum Sp245 is a plant-growth-promoting rhizobacterium that shows high levels of resistance to ampicillin. Investigating the molecular basis of ampicillin resistance and its regulation in A. baldaniorum Sp245, we found that a gene encoding lytic transglycosylase (Ltg1) is organized divergently from a gene encoding an extracytoplasmic function (ECF) σ factor (RpoE7) in its genome. Inactivation of rpoE7 in A. baldaniorum Sp245 led to increased ability to form cell-cell aggregates and produce exopolysaccharides and biofilm, suggesting that rpoE7 might contribute to antibiotic resistance. Inactivation of ltg1 in A. baldaniorum Sp245, however, adversely affected its growth, indicating a requirement of Ltg1 for optimal growth. The expression of rpoE7, as well that of as ltg1, was positively regulated by RpoE7, and overexpression of RpoE7 conferred ampicillin sensitivity to both the rpoE7::km mutant and its parent. In addition, RpoE7 negatively regulated the expression of a gene encoding a β-lactamase (bla1). Out of the 5 paralogs of RpoH encoded in the genome of A. baldaniorum Sp245, RpoH3 played major roles in conferring ampicillin sensitivity and in the downregulation of bla1. The expression of rpoH3 was positively regulated by RpoE7. Collectively, these observations reveal a novel regulatory cascade of RpoE7-RpoH3 σ factors that negatively regulates ampicillin resistance in A. baldaniorum Sp245 by controlling the expression of a β-lactamase and a lytic transglycosylase. In the absence of a cognate anti-sigma factor, addressing how the activity of RpoE7 is regulated by β-lactams will unravel new mechanisms of regulation of β-lactam resistance in bacteria.

IMPORTANCE Antimicrobial resistance is a global health problem that requires a better understanding of the mechanisms that bacteria use to resist antibiotics. Bacteria inhabiting the plant rhizosphere are a potential source of antibiotic resistance, but their mechanisms controlling antibiotic resistance are poorly understood. A. baldaniorum Sp245 is a rhizobacterium that is known for its characteristic resistance to ampicillin. Here, we show that an AmpC-type β-lactamase and a lytic transglycosylase mediate resistance to ampicillin in A. baldaniorum Sp245. While the gene encoding lytic transglycosylase is positively regulated by an ECF σ-factor (RpoE7), a cascade of RpoE7 and RpoH3 σ factors negatively regulates the expression of β-lactamase. This is the first evidence showing involvement of a regulatory cascade of σ factors in the regulation of ampicillin resistance in a rhizobacterium.

INTRODUCTION

The rhizosphere is a zone of intense microbial competition for the carbon sources secreted from the roots as root exudates (1). The resistance of microbes to antibiotics is an ancient and widespread naturally occurring phenomenon which precedes modern use of clinical antibiotics (2, 3). In order to compete with each other, bacteria use several strategies, which include production and hydrolysis of antibiotics (4). Hence, rhizospheres are potential hot spots for the evolution of genes and enzymes that can inactivate antibiotics (5, 6). The ability to produce enzymes such as β-lactamase and other factors that can inactivate antibiotics confers rhizocompetence to rhizosphere bacteria (7). Bacteria of the genus Azospirillum, which are among the most rhizocompetent plant-growth-promoting bacteria, can resist high levels of trimethoprim (>1,000 μg/mL) and β-lactam antibiotics (>500 μg/mL) (8). They colonize a wide variety of crops and grasses and promote their growth by multiple mechanisms, including production of phytohormones and siderophores (9). Due to their plant-growth-promoting abilities, they are also used as bioinoculants for improving yields of cereals and several horticultural crops. The ability of Azospirillum brasilense and Azospirillum lipoferum to resist high levels of β-lactam antibiotics is a characteristic feature that distinguishes them from many other rhizobacteria (8). It is likely to be advantageous for their survival and multiplication in the rhizosphere.

β-Lactam antibiotics inhibit the synthesis of the bacterial cell wall, which is made up of peptidoglycan (PG) and comprises repeating disaccharide N-acetylglucosamine and N-acetyl muramic acid (NAG-NAM) glycan chains that are cross-linked with peptides to form a matrix-like structure (10–12). The maintenance and turnover of PG is a dynamic process that requires the concerted action of several enzymes, which include amidases, endopeptidases, carboxypeptidases, transpeptidases, and lytic transglycosylases (LTs) (13). LTs play an important role in cell wall synthesis by causing lysis of peptidoglycan without hydrolysis, leading to the production of 1,6-anhydromuropeptides (anhMPs), which are transported to the cytosol with the help of an inner membrane permease (AmpG) for recycling (10, 14). The transpeptidases link NAG and NAM, whereas endopeptidase and carboxypeptidase control the extent of the cross-linking of PG (10). In the presence of β-lactam antibiotics, cell wall transpeptidases are inhibited, leading to the accumulation of non-cross-linked strands of PG, and the cell is not able to process all of the anhMPs for recycling to the periplasmic space for their integration in the PG (15, 16). This inability of the cell to process the enhanced levels of anhMPs leads to anhMP accumulation in the cytoplasm. The anhMPs act as a ligand for AmpR and enhance the affinity of AmpR affinity to its binding sites, resulting in the induction of expression of AmpC-type β-lactamase, which is then transported to the periplasmic space where it inactivates β-lactam antibiotics (17). Thus, by producing anhMPs, lytic transglycosylases regulate the expression of β-lactamase conferring resistance against β-lactams.

Bacteria sense the fluctuations in their external environment and respond by inducing expression of genes required for adapting to the altered environmental condition. The expression of new sets of genes is initiated at promoter sequences recognized explicitly by RNA polymerase with the help of specific σ factors. While a primary housekeeping σ factor initiates gene expression in the exponentially growing cells, alternative σ factors are activated under specific conditions to control the expression of specific set of genes by recognizing alternative promoter sequences (18, 19). Based on their sequence, domain architecture, and function, σ factors of the σ⁷⁰ family are divided into four groups (19, 20). The primary housekeeping σ factor belongs to group 1 and contains four highly conserved domains (designated σ¹ through σ⁴) along with a nonconserved region (20). Group 2 σ factors are closely related to group 1 but are not essential for growth. However, group 3 σ factors lack a σ¹ domain, and control cellular processes such as sporulation, flagellum biosynthesis, or heat shock response.

Since group 4 σ factors constitute the largest and the most diverse group of σ factors that regulate cellular response to extracellular stimuli, they are known as extracytoplasmic function (ECF) σ factors (18–20). In contrast to the other σ⁷⁰ family members, the ECF σ factors contain only two of the four conserved domains, σ² and σ⁴, which are enough for promoter recognition and interaction with core enzyme. The ECF σ factors are usually cotranscribed with a gene encoding their cognate anti-σ factors, which regulate σ factor activity (18–20). They are also characterized by autoregulation of their promoters (21). After their expression, anti-σ factors sequester their cognate ECF σ factors to occlude their binding to the core enzyme and their cognate promoters. Specific intracellular or extracellular stimuli inactivate anti-σ factors either by a change in their conformation or by proteolytic degradation (22, 23). This sets the ECF σ factor free to associate with the core enzyme to initiate transcription from its target promoters. Some ECF σ factors, however, are not associated with anti-σ factors, and hence they are regulated by mechanisms other than anti-σ factors (24).

Azospirillum baldaniorum Sp245 (formerly A. brasilense Sp245 [25, 26]) is an important strain of Azospirillum which was used frequently in studies of its physiology and genetics (25). The genome of A. baldaniorum Sp245 encodes 23 σ factors, out of which 5 belong to group III (RpoH class) and 10 belong to group IV (RpoE class or ECF). Out of the 10 ECF σ factors, only 5 are accompanied with an anti-σ factor. So far, we have studied the role and regulation of 5 RpoH σ factors and 3 RpoE σ factors (27–34). We have shown that RpoE and RpoH σ factors operate as a network of multiple regulatory cascades in which RpoE1 regulates the expression of RpoH2 and RpoH5, and RpoE2 regulates the expression of RpoH1 and RpoH4 (33). While RpoH1 and RpoH2 are involved in responding to the heat and photo-oxidative stress, respectively (29, 32, 33), RpoH3 and RpoH5 are involved in coping with oxidative stress (33). In addition, we have recently shown that RpoE10 belongs to ECF41 group of ECF σ factors, and regulates biogenesis of lateral flagella and motility in A. baldaniorum Sp245 (34). Out of the 5 RpoE σ factors that are not in an operon-like organization with an anti-σ factor in A. baldaniorum Sp245, we found that RpoE7 was organized divergently to a gene encoding a lytic transglycosylase, which plays an important role in inducing resistance to β-lactam antibiotics in other bacteria (35). Hence, we investigated the role of RpoE7 in regulating ampicillin resistance in A. baldaniorum Sp245. Here, we show that RpoE7 regulates the expression of the divergently organized lytic transglycosylase. We also show that a cascade of RpoE7-RpoH3 σ factors controls the expression of an AmpC-type β-lactamase to control ampicillin resistance in A. baldaniorum Sp245 (26).

RESULTS

Organization of the gene encoding RpoE7 in A. baldaniorum Sp245.

Examination of the genomic organization of the genes encoding 10 ECF σ factors (RpoE) in the genome of A. baldaniorum Sp245 showed that rpoE7 (AZOBR_150067) is organized divergently to a gene encoding a lytic transglycosylase (Ltg) (AZOBR_150069) (Fig. 1A). The search for synteny of the rpoE7 gene organization revealed a similar kind of organization in A. brasilense Sp7, Azospirillum thiophilum, Azospirillum lipoferum, and Azospirillum humicireducens (see Fig. S1 in the supplemental material). In addition, Rhodospirillum centenum, which is a close relative of A. baldaniorum Sp245, also possesses a similar gene organization but also shows the presence of a gene encoding an anti-σ factor downstream of the rpoE7 ortholog.

FIG 1.

(A) Genomic organization of the gene encoding rpoE7 in A. baldaniorum Sp245, showing a divergently oriented lytic transglycosylase (Ltg) and a PepSY_2 domain protein-encoding gene located downstream of rpoE7. (B) Culture tubes of A. baldaniorum Sp245 and the rpoE7::km mutant during the exponential phase at 22 h show difference in their ability to flocculate. (C) Scanning electron microscopy (SEM) images of A. baldaniorum Sp245 and the rpoE7::km mutant (magnification, ×5,000. (D) Comparison of biofilm-forming abilities of A. baldaniorum Sp245 and the rpoE7::km mutant. (E) Comparison of exopolysaccharide production by A. baldaniorum Sp245 and the rpoE7::km mutant.

Since rpoE7 and ltg1 lie close to each other and lytic transglycosylases are known to be involved in mediating antibiotic resistance (17), we wanted to determine whether there is any interconnection between rpoE7 and lgt1 and whether this plays a role in antibiotic resistance. For this, we first made rpoE deletion and ltg1 deletion strains and examined their growth phenotypes.

Inactivation of the gene encoding rpoE7 increases cell aggregation and biofilm formation in A. baldaniorum Sp245.

To study the role of RpoE7 in the physiology of A. baldaniorum Sp245, the rpoE7 gene was inactivated by inserting a kanamycin (Km) resistance gene. A comparison of the growth of the rpoE7::km mutant with that of its parent in liquid culture showed that the mutant displayed a tendency to form flocs or aggregates (Fig. 1B) during the exponential phase (optical density at 600 nm [OD₆₀₀] = 0.6 to 0.7). This observation was unusual as A. baldaniorum Sp245 shows aggregate or floc formation during the stationary phase, not during the exponential phase. Hence, we compared the growth of both the parent and the mutant by estimating the protein content of the cultures at different time points, which showed that the mutant grew slower than its parent (Fig. S2). Observation of the cultures of A. baldaniorum Sp245 and the rpoE7::km mutant under scanning electron microscope (SEM) also showed that the cells of the wild type were segregated from each other, whereas the cells of rpoE7::km mutant adhered to each other in the form of clumps or aggregates (Fig. 1C). A comparison of the biofilm-forming ability of rpoE7::km mutant and its parent showed that the cells of the rpoE7::km mutant possess an increased ability to form biofilm in the microtiter plate assay. (Fig. 1D). To examine if there is any difference in the ability of the parent and the mutant to produce exopolysaccharides (EPS), we estimated EPS content and found that rpoE7::km mutant produced increased EPS relative to that produced by its parent (Fig. 1E).

Lytic transglycosylase (Ltg1) is required for the optimal growth of A. baldaniorum Sp245.

The gene encoding lytic transglycosylase (AZOBR_150069) encodes a protein of 331 amino acids (aa) including a signal peptide of 20 aa residues and two membrane-spanning domains (Fig. S3), suggesting it to be a membrane-bound protein that might be secreted out of the inner membrane into the periplasmic space. To examine its physiological role, we inactivated ltg1 by inserting a kanamycin resistance gene, and compared the growth of ltg1::km mutant with its parent. The mutant grew considerably slower than the parent (Fig. 2A). Complementation of ltg1::km mutant by expressing a plasmid located copy of ltg1 gene (pAK007) showed considerable recovery of growth. Thus, ltg1 is essential for the optimal growth of A. baldaniorum Sp245.

FIG 2.

(A) Growth curves of A. baldaniorum Sp245, the ltg1::km mutant, and the ltg1::km complemented strain in minimal malate medium. Each point of the curve shows the mean of three replicates obtained from three different experiments, and error bars indicate standard deviation (SD) at each point. (B) Effect of overproduction of Ltg1 of A. baldaniorum Sp245 on the growth of Escherichia coli DH5α in LB broth at 37°C with and without isopropyl-β-d-thiogalactopyranoside (IPTG) induction of the plasmid pAK007 and pAK008 expressing ltg1 from pMMB206 and pAK032, respectively. I, induced; UI, uninduced. (C) Relative expression of the 6 lytic transglycosylase paralogs of A. baldaniorum Sp245 by quantitative reverse transcription-PCR (RT-PCR) by using threshold cycle values obtained from RNA samples from A. baldaniorum Sp245 and its rpoE7::km mutant. Each bar shows the mean and standard deviation of values obtained from the three replicates.

Since lytic transglycosylases are known to cause anhydrolytic cleavage of peptidoglycan (14), its overexpression is likely to have an adverse effect on the integrity of cell envelope and survival. Hence, we examined the effect of expression of Ltg1 on Escherichia coli DH5α. The growth of E. coli DH5α expressing ltg1 from a low-copy-number vector, pMMB206 (2 to 5 copies per cell) (36) and from a medium-copy-number vector, pAK032 (20 to 40 copies per cell) (37) was compared. The expression of Ltg1 from a low-copy-number vector did not show any noticeable effect on the growth of E. coli, whereas its expression from a medium-copy-number vector led to an almost complete arrest of growth, which seems to be due to the lysis of E. coli cells (Fig. 2B). These observations suggest that the overexpression of Ltg1 of A. baldaniorum Sp245 in E. coli has an adverse effect on the integrity of the cell envelope, leading to its killing.

Expression of ltg1 and ltg5 is positively regulated by RpoE7.

Having shown that RpoE7 regulates ltg1 expression, we next asked if the A. baldaniorum Sp245 genome encodes more lytic transglycosylases and whether some of them are also regulated by this σ factor. A search for the paralogs of lytic transglycosylases (Ltg) in the genome of A. baldaniorum Sp245 revealed the presence of 5 additional paralogs. Comparison of the phylogenetic relatedness of all the 6 Ltg paralogs of A. baldaniorum Sp245 among themselves and with those from E. coli showed that Ltg1 (AZOBR_150069) and Ltg4 (AZOBR_p130174) are phylogenetically closely related (Fig. S5) and showed closer similarity to Slt70 and MltD, the two membrane-anchored lytic transglycosylases of E. coli (38). A comparison of the relative expression of the 6 ltg paralogs in A. baldaniorum Sp245 and its rpoE7::km mutant showed (Fig. 2C) that the expression of ltg1 and ltg5 was strongly downregulated in the rpoE7::km mutant indicating that the ltg1 and ltg5 (AZOBR_200055) genes are positively regulated by RpoE7. The differences in the levels of transcripts of ltg2 (AZOBR_p470003), ltg3 (AZOBR_110113), ltg4 (AZOBR_p130174), and ltg6 (AZOBR_70170) in A. baldaniorum Sp245 and the rpoE7::km mutant were not significant.

Overexpression of RpoE7 and Ltg1 confers ampicillin sensitivity to A. baldaniorum Sp245 and rpoE7::km mutant.

The above-described results and the genomic organization of rpoE7 and ltg1 strongly suggested that an interplay between rpoE7 and ltg1 might control antibiotic resistance in A. baldaniorum Sp245. Hence, we compared the sensitivity of rpoE7::km mutant with its parent to the different types of antibiotics, including ampicillin. It was only in the case of ampicillin that the rpoE7::km mutant showed resistance compared to its parent. Complementation of the rpoE7::km mutant with the rpoE7 gene via a broad-host-range vector pMMB206 (designated pAK006) restored the ampicillin sensitivity of the rpoE7::km mutant to the parent level (Fig. 3A). Moreover, overexpression of rpoE7 in A. baldaniorum Sp245 also conferred sensitivity to ampicillin (Fig. S4). These data demonstrate that a functional RpoE7 confers ampicillin sensitivity to A. baldaniorum Sp245.

FIG 3.

Effect of expression of rpoE7 and ltg1 on ampicillin sensitivity of the rpoE7::km mutant of A. baldaniorum Sp245 on an Minimal Malate Agar (MMA) plate supplemented with 450 μg ampicillin and triphenyl tetrazolium chloride (TTC; 50 mg/mL) at 10-fold dilutions (10⁻¹, 10⁻², and 10⁻³⁾ of the culture.

Since the ltg1::km mutant grew considerably slower than its parent, we investigated the effect of expression of ltg1 on ampicillin sensitivity of A. baldaniorum Sp245 and the rpoE7::km mutant. Expression of ltg1 conferred ampicillin sensitivity to both A. baldaniorum Sp245 and the rpoE7::km mutant (Fig. 3B), suggesting a possible link between ltg1 and rpoE7 in controlling ampicillin resistance in A. baldaniorum Sp245.

RpoE7 regulates its own promoter and ltg1 promoter.

To study the regulation of expression of rpoE7, we determined its transcription start site (TSS) using 5′ rapid amplification of cDNA ends (RACE). The TSS of rpoE7 was a “G” that was located 144 nucleotides upstream of the start codon, ATG (Fig. 4A and B). Based on the TSS, the rpoE7 promoter consists of ATTACG as a −10 motif and CGGTCA as a −35 motif with a distance of 17 nucleotides in between the two motifs (Fig. 4B). To study the regulation of expression of rpoE7, we constructed an rpoE7::lacZ fusion (pAK009) and mobilized it into A. baldaniorum Sp245 and the rpoE7::km mutant to examine whether, like promoters of other ECF σ factors, the rpoE7 promoter is autoregulated. A comparison of the β-galactosidase activity of the rpoE7::lacZ fusion showed that the β-galactosidase activity in rpoE7::km mutant was negligible, whereas it was ≈300 Miller units in A. baldaniorum Sp245 (Fig. 4C). Thus, RpoE7 is required for the expression of its own promoter, and hence its expression is autoregulated. To examine whether the expression of rpoE7 is induced by ampicillin and other stress agents, we studied the effect of ampicillin (100 μg/mL), polymyxin B (100 μg/mL), ethanol (1%), hydrogen peroxide (1 mM), and NaCl (150 mM) on the activation of the rpoE7::lacZ fusion in A. baldaniorum Sp245. Figure 4D shows that rpoE7 is upregulated by ampicillin and NaCl, but not by polymyxin B, ethanol, or hydrogen peroxide.

FIG 4.

(A) Chromatogram showing transcription start site of rpoE7 determined by 5′ rapid amplification of cDNA ends (RACE) with a “G” (encircled in red) as a transcriptional start point. (B) Nucleotide sequence upstream of rpoE7 showing −10 and −35 hexamers of the promoter are underlined. Upstream region of ltg1 gene showing predicted possible −35 and −10 elements (underlined). (C) Comparison of the β-galactosidase activities of rpoE7::lacZ and ltg1::lacZ fusions in A. baldaniorum Sp245 and its rpoE7::km mutant. 1, Sp245 (rpoE7:lacZ); 2, rpoE7::km(rpoE7::lacZ); 3, Sp245 (ltg1::lacZ); 4, rpoE7::km(ltg1::lacZ). Error bars show standard deviations (SD) for triplicates of three independent experiments. (D) Comparison of the β-galactosidase activities of the rpoE7::lacZ fusion in A. baldaniorum Sp245 under different abiotic stresses. Error bars show standard deviations (SD) for triplicates of three independent experiments.

Since ltg1 is divergently organized to rpoE7, we tested the possibility of its regulation by RpoE7. For this, we constructed an ltg1::lacZ fusion (pAK0010) and conjugatively mobilized it into A. baldaniorum Sp245 and the rpoE7::km mutant. The β-galactosidase activity from the ltg1::lacZ fusion in the rpoE7::km mutant was barely detectable, whereas in A. brasilense Sp245 it was ≈170 Miller units (Fig. 4C). This finding indicates that RpoE7 is required for the expression of ltg1.

RpoE7 regulates the expression of one of the five paralogs of β-lactamase, Bla1.

Having shown that RpoE7 regulates ampicillin resistance through Ltg1, we next explored the role of β-lactamases in the ampicillin resistance A. baldaniorum Sp245 and if RpoE7 plays any role in regulating the expression of β-lactamases. We found that A. baldaniorum Sp245 genome encodes 5 β-lactamase paralogs. According to the Ambler classification of β-lactamases, AZOBR_p220102 (Bla1) showed similarities to class A β-lactamase AZOBR_180252 [Bla2] to class C β-lactamases (serine-type d-Ala-d-Ala carboxypeptidases) and AZOBR_130007 (Bla3), AZOBR_140057 (Bla4), and AZOBR_10425 (Bla5) to class B β-lactamases (metal-dependent hydrolases) (39). To determine whether expression of any of the 5 paralogs of β-lactamase is dependent on RpoE7, we examined the relative expression of the 5 bla paralogs in A. baldaniorum Sp245 and its rpoE7::km mutant. Figure 5A shows that the expression of bla2 (AZOBR_180252), bla3 (AZOBR_130007), and bla4 (AZOBR_140057) was slightly inhibited in the rpoE7::km mutant, while relative expression of bla1 (AZOBR_p220102) and bla5 (AZOBR_10425) was significantly upregulated. The expression of bla1 in the rpoE7::km mutant was 3.5-fold higher than that in A. baldaniorum Sp245. This indicated that rpoE7 negatively regulates the expression of bla1. To validate this result further, a bla1::lacZ fusion (designated pAK011) was constructed and mobilized in A. baldaniorum Sp245 and the rpoE7::km mutant. Comparison of the β-galactosidase activity in the two strains indicated that bla1 expression in the rpoE7::km mutant was more than 3-fold higher than that observed in A. baldaniorum Sp245 (Fig. 5B). The expression of rpoE7 via pAK006 in rpoE7::km (bla1::lacZ) restored the parental level of expression of bla1 in the rpoE7::km mutant. This result confirms that the expression of bla1 is negatively regulated by RpoE7.

FIG 5.

(A) Relative expression of the 5 β-lactamase paralogs of A. baldaniorum Sp245 by quantitative RT-PCR using threshold cycle values obtained from RNA samples of A. baldaniorum Sp245 and its rpoE7::km mutant. Each bar shows the mean and standard deviation of values obtained from the three replicates. Error bars show standard deviations (SD) for triplicates of three independent experiments. (B) Comparison of the β-galactosidase activities from bla1::lacZ in A. baldaniorum Sp245 (no. 1), the rpoE7::km mutant (no. 2), and the rpoE7::km complemented strain (no. 3).

Bla1 confers ampicillin resistance to E. coli and A. baldaniorum Sp245.

None of the 5 genes encoding β-lactamase paralogs have an AmpR encoding gene in their vicinity except AZOBR_p220102, which is organized divergently from ampR. A closer examination of the intergenic region between bla1 and ampR revealed two potential sites for the binding of a LysR-type regulator (Fig. S6). A typical T-N₁₁-A motif (having 11 non-conserved nucleotides between the conserved T and A conserved nucleotides) was found 43 nucleotides upstream of the start codon of Bla1 and 33 nucleotides upstream of the start codon of AmpR. An alignment of the deduced amino acid sequence of AZOBR_p220102 with that of other orthologs showed four conserved motifs, SXXK, SDN, EXXN, and KTG, that are characteristic of the class A β-lactamases (Fig. S7). A BLAST search of the Beta-Lactamase DataBase (BLDB) showed maximum similarity of AZOBR_p220102 amino acid sequences with the β-lactamases (XCC-16, XCC-21, XCC-22, XCC-19, XCC-20) of different ampicillin-resistant strains of Xanthomonas campestris.

To verify if AZOBR_p220102 is a functional β-lactamase, the gene encoding Bla1 was cloned in pAK002 (resulting in the recombinant pAK023) and transformed in E. coli DH5α. Comparison of the growth of E. coli DH5α (pAK002) and DH5α (pAK023) on Luria-Bertani (LB) agar plates showed that DH5α (pAK002) was sensitive to ampicillin, whereas DH5α (pAK023) showed very good growth on LB agar plate containing 400 μg ampicillin/mL (Fig. 6A). This clearly indicates that the cloned bla1 conferred ampicillin resistance to E. coli DH5α. Screening of the sensitivity of E. coli DH5α (pAK023) to other β-lactam antibiotics using a disc diffusion assay revealed that Bla1 conferred complete resistance to cephalexin, cefadroxil, cefazolin, cefoperazone, piperacillin, and ticarcillin (Table 1). The resistance of E. coli DH5α (pAK023) declined in the order to the following drugs: cefaclor, carbenicillin, ampicillin, cefotaxime, cefuroxime, cefixime, ceftazidime, ceftriaxone, imipenem, and aztreonam.

FIG 6.

(A) LB agar plate (plate a) depicting ampicillin resistance in E coli strain DH5α overexpressing bla1 (pAK002::bla1↑) of A. baldaniorum Sp245. E coli DH5α lacking bla1 (pAK002) (plate b) is sensitive to ampicillin (450 μg/mL). (B) Plate c shows normal growth of Sp7 and bla1::km mutant without antibiotic. MMA plates showing resistance of A. baldaniorum Sp245 (plate d) and sensitivity of the bla1::km mutant (plate d) to ampicillin. Plate e shows resistance of the bla1::km mutant to kanamycin.

TABLE 1.

Effect of different β-lactam antibiotics on E. coli DH5α overexpressing bla1 of A. baldaniorum Sp245

Sample no.	Antibiotic (dose in μg)	Zone of inhibition (cm) for:
		DH5α	DH5α (bla1↑)
1	Cefalexin (30)	1.90 ± 0.14	0.00 ± 0.00
2	Cefadroxil (30)	1.95 ± 0.35	0.00 ± 0.00
3	Cefazolin (30)	2.00 ± 0.00	0.00 ± 0.00
4	Cefoperazone (75)	2.90 ± 0.14	0.00 ± 0.00
5	Piperacillin (100)	4.65 ± 0.21	0.00 ± 0.00
6	Ticarcillin (75)	4.90 ± 0.14	0.00 ± 0.00
7	Cefaclor (30)	3.10 ± 0.14	0.5 ± 0.71
8	Carbenicillin (100)	4.50 ± 0.14	1.15 ± 0.07
9	Ampicillin (100)	5.50 ± 0.00	1.20 ± 0.00
10	Cefotaxime (30)	5.45 ± 0.07	1.40 ± 0.14
11	Cefuroxime (30)	6.05 ± 0.35	1.70 ± 0.14
12	Cefixime (5)	3.45 ± 0.21	1.85 ± 0.07
13	Ceftazidime (30)	4.15 ± 0.21	2.05 ± 0.21
14	Ceftriaxone (30)	6.35 ± 0.21	2.15 ± 0.07
15	Imipenem (10)	5.70 ± 0.00	2.70 ± 0.00
16	Aztreonam (30)	5.65 ± 0.07	3.00 ± 0.14

The role of bla1 in A. baldaniorum Sp245 was examined by inactivating bla1 by inserting a kanamycin resistance gene cassette in its bla1 gene and replacing the wild copy of the genome with the mutant copy via homologous recombination. Comparison of the ampicillin sensitivity of the bla1::km mutant with that of its parent showed that the bla1::km mutant was sensitive to ampicillin, whereas the parent grew well even at 450 μg ampicillin/mL. (Fig. 6B). This result shows that Bla1 is exclusively responsible for the high level of resistance to ampicillin in A. baldaniorum Sp245, and other β-lactamases seem to have no notable role in conferring high-level resistance of A. baldaniorum Sp245 to ampicillin.

Effect of overexpression of rpoH paralogs on ampicillin resistance in A. baldaniorum Sp245.

Since the regulation by ECF σ factors is often mediated via RpoH σ factors (33, 40–42), we examined whether any of the 5 RpoH paralogs encoded in the genome of A. baldaniorum Sp245 are involved in mediating the regulation of ampicillin resistance by RpoE7. Figure 7A shows that the expression of rpoH2, rpoH3, and rpoH5 in A. baldaniorum Sp245 rendered the bacteria sensitive to ampicillin to various degrees. However, the effect of rpoH3 expression was the most pronounced in making A. baldaniorum Sp245 sensitive to ampicillin. This suggests that the expression of β-lactamase in A. baldaniorum Sp245 may be negatively regulated by RpoH3. We also investigated the effect of expression of the 5 rpoH paralogs on the expression of the bla1::lacZ fusion. Figure 7B shows that the expression of rpoH3 had the most pronounced adverse effect on the β-galactosidase activity from bla1::lacZ in A. baldaniorum Sp245. This result shows that the increased sensitivity of A. baldaniorum Sp245 to ampicillin due to the expression of rpoH3 was due to the reduced production of β-lactamase.

FIG 7.

(A) Ampicillin sensitivity of A. baldaniorum Sp245 carrying broad-host-range vector pAK032 and A. baldaniorum Sp245 derivatives harboring different rpoH paralogs cloned in pAK032 in MM medium amended with ampicillin (450 μg/mL) and TTC (50 mg/mL) at 10-fold dilutions (10⁻¹, 10⁻², and 10⁻³) of the culture. (B) Comparison of the β-galactosidase activities from bla1::lacZ in A. baldaniorum Sp245; 2, Sp245 (RpoH1↑); 3, Sp245 (RpoH2↑); 4, Sp245 (RpoH3↑); 5, Sp245 (RpoH4↑); 6, Sp245 (RpoH5↑). Error bars show standard deviations (SD) for triplicates of the three independent experiments. (C) Effect of the overexpression of rpoE7 on the expression of the rpoH3::lacZ fusion and of rpoH3 overexpression on the expression of the rpoE7::lacZ fusion in A. baldaniorum Sp245. Error bars show standard deviations (SD) for triplicates of the three independent experiments.

Expression of rpoH3 is regulated by RpoE7 in A. baldaniorum Sp245.

So far, we found that A. baldaniorum Sp245 becomes sensitive to ampicillin due to the expression of rpoE7 or rpoH3. We have also shown that rpoH3 expression inhibits the expression of the bla1::lacZ fusion. Based on these observations, it can be assumed that the expression of bla1 might be controlled by a regulatory cascade consisting of RpoE7 and RpoH3. We used a two-plasmid system (32, 43) consisting of an rpoH3::lacZ fusion (pAK022) and an rpoE7-overexpressing derivative (pAK006) on a compatible vector to examine if RpoE7 is able to regulate rpoH3 promoter in A. baldaniorum Sp245. Higher β-galactosidase activity from the rpoH3::lacZ fusion in the rpoE7-overexpressing derivative (563 Miller units) than that in the parent (216 Miller units) indicated that rpoE7 is required for an enhanced expression of rpoH3 in A. baldaniorum Sp245 (Fig. 7C). When a reverse experiment was performed by examining the effect of overexpression of rpoH3 on the expression of the rpoE7::lacZ fusion (pAK009), no difference was found in the β-galactosidase activity of the two strains (Fig. 7C). This suggested that the expression of rpoE7 was not dependent on RpoH3. These observations confirmed that ampC expression in A. baldaniorum Sp245 is controlled by a regulatory cascade in which RpoE7 regulates the activity of RpoH3.

DISCUSSION

The survival and multiplication of bacteria in the rhizosphere requires an ability to tolerate multiple stresses, including the ability to maintain the integrity of their cell wall against β-lactam antibiotics. Gram-negative bacteria often harbor multiple paralogs of lytic transglycosylases (LT) and β-lactamases to cope with the assault by β-lactams. Out of the eight LTs encoded in the genome of E. coli, seven are membrane-anchored LTs (mLTs) and one is a cytosolic LT (sLT) (44). Similarly, in A. baldaniorum Sp245, out of the 6 paralogs of LT encoded by its genome, all except one are membrane bound. The A. baldaniorum Sp245 genome also encodes 5 paralogs of β-lactamases, out of which 2 belong to the serine-type d-Ala-d-Ala carboxypeptidases, 2 to metallohydrolases, and 1 to class A β-lactamases. It is only the gene encoding Bla1 that is accompanied by a divergently organized gene encoding an AmpR regulator.

RpoE7 is one of the atypical ECF σ factors of A. baldaniorum Sp245 which lacks a gene encoding an anti-σ factor in its vicinity (34). The rpoE7::km mutant of A. baldaniorum Sp245 showed enhanced resistance to ampicillin, indicating that RpoE7 controls ampicillin resistance in this bacterium. We also showed that RpoE7 regulates the expression of ltg1, which confers ampicillin sensitivity in A. baldaniorum Sp245. This type of role of Ltg1 of A. baldaniorum Sp245 is similar to that of SltB1 and MltB lytic transglycosylases from Pseudomonas aeruginosa, the loss of which results in an increased resistance to β-lactams (35). A drastic decline in the growth of E. coli due to the overexpression of ltg1 indicated that hyperactivity of lytic transglycosylases adversely affects integrity of the bacterial cell envelope. However, reduced growth of the ltg1::km mutant of A. brasilense Sp245 suggested that Ltg1 is also required for the optimal growth of A. baldaniorum Sp245. Survival of the ltg1::km mutant shows that Ltg1 is not absolutely essential for the growth of A. baldaniorum Sp245, most likely due to some compensation by other Ltg paralogs encoded in the genome (45). Lytic transglycosylases regulate ampicillin resistance in bacteria by producing 1,6-anhydroMurNAc peptides (13, 46) that bind to the AmpR regulator to induce the expression of β-lactamase, which hydrolyzes β-lactam antibiotics and inactivates them (47). In the absence of β-lactams, UDP-MurNAc-pentapeptide binds to AmpR and causes repression of ampC transcription. The β-lactams disrupt PG metabolism and produce 1,6-anhydroMurNAc peptides that are transported to the cytosol by AmpG and are thought to competitively displace UDP-MurNAc-pentapeptides from AmpR to convert it into an inactive repressor of ampC transcription (47). The expressed AmpC is then exported to the periplasm, where it hydrolyzes β-lactam antibiotic to reestablish normal cell wall homeostasis.

Since one of the characteristic features of A. baldaniorum Sp245 is its high level of resistance to ampicillin, we assumed that this feature might be due to the presence of multiple paralogs of the gene encoding β-lactamases (48). The divergent organization of the genes encoding AmpC and AmpR, as found in the case of Bla1 in A. baldaniorum Sp245, is found in many other β-lactam-resistant bacteria (49). By demonstrating that the expression of Bla1 from A. baldaniorum Sp245 in E. coli can confer a high level of resistance against multiple β-lactams, including ampicillin, we have shown that Bla1 is a functional β-lactamase, which alone can confer high level of resistance to β-lactam antibiotics. In addition, complete loss of ampicillin resistance in the bla1::km mutant of A. baldaniorum Sp245 indicates that β-lactamase homologs, other than bla1, do not play any significant role in conferring ampicillin resistance in A. baldaniorum Sp245. This observation is not surprising, as enzymes harboring the β-lactamase fold have been shown to be present in numerous enzyme families responsible for diverse biological processes (50, 51). Since the deduced amino acid sequence of Bla1 of A. baldaniorum Sp245 showed maximum similarity to that of the constitutively expressed β-lactamases (XCC-16, XCC-19, XCC-20, XCC-21, and XCC-22) of ampicillin-resistant strains of X. campestris pathovar campestris, it is quite likely that Bla1 is also expressed constitutively in A. baldaniorum Sp245 and, like X. campestris β-lactamases, is upregulated by ampicillin (52, 53). This study suggests that the unusually high level of resistance to ampicillin found in A. baldaniorum Sp245 might be due to the Bla1.

Besides their role in many other extracytoplasmic functions, ECF σ factors are also known to regulate β-lactam resistance in Gram-negative and Gram-positive bacteria (40, 54–56). In P. aeruginosa, β-lactam antibiotics induce the expression of the AlgU σ factor, leading to the production of alginate as well as that of β-lactamase (54). AlgU regulates the expression of ampC β-lactamase by regulating the expression of ampR. In Bacillus cereus and Bacillus thuringiensis, ampicillin resistance is regulated by the σ^P and RsiP pair of σ and anti-σ factors (55, 56). Upon exposure to β-lactams, a proteolytic cascade leads to the destruction of RsiP, resulting in the release of σ^P, which induces the expression of β-lactamase and results in resistance to β-lactams. Activation of σ^P is specific to a subset of β-lactams, and σ^P is required for resistance to these β-lactams (56). However, it is not required for resistance to the other cell wall antibiotics, including vancomycin, nisin, and bacitracin, suggesting specificity in resistance to β-lactams and not a general cell envelope stress response. Unlike AlgU of P. aeruginosa and σ^P of B. thuringiensis, RpoE7 of A. baldaniorum Sp245 is not accompanied by a cognate anti-σ factor in its genomic vicinity, nor does it confer ampicillin resistance. Instead, a functional RpoE7 confers ampicillin sensitivity.

Usually, σ factors positively regulate the activity of their target genes. However, we found that the expression of β-lactamase and ampicillin resistance in A. baldaniorum Sp245 is negatively regulated by RpoE7 as well as RpoH3 (Fig. 8). The negative regulation of β-lactam resistance by an ECF σ factor has also been shown in Salmonella enterica serovar Typhi (57), in which the loss of ampicillin resistance is thought to be due to the downregulation of porins (OmpF and OmpC) and upregulation of efflux systems (57). Porins are known to facilitate penetration by antibiotics, including β-lactams, through the outer membrane (48), and OmpF is believed to be the principal pathway for the transport of β-lactam antibiotics (58). Regulatory cascades of RpoE-RpoH σ factors regulate gene expression in E. coli, Rhodobacter sphaeroides, and A. brasilense (32, 33, 41). Although we do not yet know why ampicillin resistance in A. baldaniorum Sp245 is negatively regulated by a RpoE7-RpoH3 cascade, we report for the first time an involvement of a regulatory cascade of alternative σ factors in controlling ampicillin resistance, which may be one of the important factors responsible for the rhizocompetence of A. baldaniorum Sp245, which colonizes the rhizosphere of a wide variety of plants.

FIG 8.

Proposed scheme of alternative sigma factors regulating the expression of the bla1 gene conferring ampicillin resistance in A. baldaniorum Sp245. RpoE7, along with the core enzyme, activates the promoters of ltg1 and rpoE7. Then, RpoE7 activates the expression of RpoH3, which, in turn negatively regulates the expression of AmpC type β-lactamase. Thus, a cascade of RpoE7-RpoH3 negatively regulates bla1 expression and ampicillin resistance in A. baldaniorum Sp245.

MATERIALS AND METHODS

Bacterial strains, growth conditions, chemicals, and plasmids.

A. baldaniorum Sp245 was grown at 30°C in minimal malate (MM) medium (59). E. coli strains (DH5α and S17.1) were grown in Luria-Bertani (LB) medium at 37°C. Bacterial strains and plasmids used in this study are described in Table 2. All of the media used for bacterial growth were purchased from Sigma-Aldrich (America) or HiMedia (India), and enzymes used for cloning and other DNA modifications were from New England Biolabs. Sequences of the primers used for PCR amplification of different genes are listed in Table 3.

TABLE 2.

Bacterial strains and plasmids used

Strain or plasmid	Relevant propertiesa	Reference or source
Bacterial strains
E. coli DH5α	ΔlacU169 hsdR17 recA1 endA1 gyrA96 thiL relA1	Gibco-BRL
E. coli S17-1	Smr; recA thiprohsdRRP4-2 (Tc::Mu; Km::Tn7)	61
A. baldaniorum Sp245	Wild-type strain	26
rpoE7::km	A. baldaniorum Sp245 derivative with insertion of a kanamycin resistance gene cassette in the rpoE7 gene	This work
ltg1::km	A. baldaniorum Sp245 derivative with insertion of a kanamycin resistance gene cassette in the ltg1 gene	This work
bla1::km	A. baldaniorum Sp245 derivative with insertion of a kanamycin resistance gene cassette in the bla1 gene	This work
Plasmids
pSUP202	ColE1 replicon, mobilizable, suicide vector for A. brasilense; Apr, Cmr, Tcr	61
pUC4K	Plasmid containing kanamycin resistance gene cassette	GE Healthcare
pCZ750	Tcr, Apr; broad-host-range vector for construction of promoter lacZ fusion	68
pMMB206	Cmr; broad-host-range, low-copy-number expression vector	36
pAK032	Tcr; broad-host-range, medium-copy-number expression vector	37
pAK001	rpoE7::km in pSUP202	This work
pAK002	pBBR1 MCS-3 derivative containing constitutive kanamycin resistance gene (apt) promoter	This work
pAK003	rpoE7 disruption plasmid harboring kanamycin resistance gene (apt) used for generating rpoE7::km mutant	This work
pAK004	ltg1 disruption plasmid harboring kanamycin resistance gene (apt) used for generating ltg1::km mutant	This work
pAK005	bla1 disruption plasmid harboring kanamycin resistance gene (apt) used for generating bla1::km mutant	This work
pAK006	rpoE7 gene of A. baldaniorum Sp245 cloned in vector pMMB206	This work
pAK007	ltg1 gene of A. baldaniorum Sp245 cloned in vector pMMB206	This work
pAK008	ltg1 gene of A. baldaniorum Sp245 cloned in vector pAK032	This work
pAK009	rpoE7 promoter of A. baldaniorum Sp245 cloned upstream of lacZ in pCZ750	This work
pAK010	ltg1 promoter of A. baldaniorum Sp245 cloned upstream of lacZ in pCZ750	This work
pAK011	bla1 promoter of A. baldaniorum Sp245 cloned upstream of lacZ in pCZ750	This work
pAK012	rpoH1 gene of A. baldaniorum Sp245 cloned in vector pAK032	This work
pAK013	rpoH2 gene of A. baldaniorum Sp245 cloned in vector pAK032	This work
pAK014	rpoH3 gene of A. baldaniorum Sp245 cloned in vector pAK032	This work
pAK015	rpoH4 gene of A. baldaniorum Sp245 cloned in vector pAK032	This work
pAK016	rpoH5 gene of A. baldaniorum Sp245 cloned in vector pAK032	This work
pAK017	rpoH1 gene of A. baldaniorum Sp245 cloned in vector pMMB206	This work
pAK018	rpoH2 gene of A. baldaniorum Sp245 cloned in vector pMMB206	This work
pAK019	rpoH3 gene of A. baldaniorum Sp245 cloned in vector pMMB206	This work
pAK020	rpoH4 gene of A. baldaniorum Sp245 cloned in vector pMMB206	This work
pAK021	rpoH5 gene of A. baldaniorum Sp245 cloned in vector pMMB206	This work
pAK022	rpoH3 promoter of A. baldaniorum Sp245 cloned upstream of lacZ in pCZ750	This work
pAK023	bla1 gene of A. baldaniorum Sp245 cloned in vector pAK002	This work

^aSm^r, spectinomycin resistance; Ap^r, ampicillin resistance; Cm^r, chloramphenicol resistance; Tc^r, tetracyclin resistance; Km^r, kanamycin resitance.

TABLE 3.

List of primers used

Primer	Sequence (5′ to 3′)a
RpoE7:A:FP	CGGAATTCCATCAGTACGTAATGGCG
RpoE7:A:RP	GAAGATCTCTCCTGCTTGAGGCGCTG
RpoE7:B:FP	GAAGATCTCATGGTCACCAAGATGGC
RpoE7:B:RP	AACTGCAGGTGAAGAAATCCTTGGAC
Ltg-1:A:FP	AACTGCAGCAGAGATGCAGGCGCGAC
Ltg-1:A:RP	GAAGATCTCAGAAAGGAGGCGATCAGC
Ltg-1:B:FP	GAAGATCTCGCTTCGAAACGCTGACG
Ltg-1:B:RP	CGGAATTCCTTACCGGCATCAACGATC
RpoE7_pMMB:FP	CGGAATTCGATGGCGAACCACGTCGCG
RpoE7_pMMB:RP	AACTGCAGTCAATCGTCATCGGTCGG
Ltg-1_pMMB:FP	CGGAATTCCGTGGCCCTGGCCGCCGG
Ltg-1_pMMB:RP	AACTGCAGTCAGCGTTTCGAAGCGCCG
Ltg-1_032:FP	CCGCTCGAGGTGGCCCTGGCCGCCGGC
Ltg-1_032:RP	TCCCCCCGGGTCAGCGTTTCGAAGCGCCG
RpoE7_GSP1	CGACGGATGCCGTTGATGTG
RpoE7_GSP2	CTCCAGACATTCCTGCAC
RpoE7_GSP3	ATCTCCGCCTCGATCTGCGCGATG
Oligo(dT) anchor primer	GACCACGCGTATCGATGTCGACTTTTTTTTTTTTTTTTV
PCR anchor primer	GACCACGCGTATCGATGTCGAC
RpoE7_pCZ:FP	GCTCTAGAGCGGATGGATGAATGAGCG
RpoE7_pCZ:RP	CCCAAGCTTCGGTGGAACGCTGCCTCC
Ltg-1_pCZ:FP	GCTCTAGACGTAATGGCGCACCACAGG
Ltg-1_pCZ:RP	CCCAAGCTTGCGCTTTTCACCGTCTTGAC
AmpC_pCZ:FP	GCTCTAGAGAAGAGCCATTAGTTTTTCTTG
AmpC_pCZ:RP	CCCAAGCTTCGTGTCCTCGTCTCATGCC
RpoH1_032:FP	GGAATTCCATATGGCGACGATATCCAGCG
RpoH1_032:RP	AACTGCAGTCAACCGGCCAGCTTCTGC
RpoH2_032:FP	GGAATTCCATATGGCCTACATCGACGATCCCG
RpoH2_032:RP	AACTGCAGCTACGCGTCCGCCAGCAGATC
RpoH3_032:FP	GGAATTCCATATGACGGAACTGGCAGTTTCTG
RpoH3_032:RP	AACTGCAGGTCACACCGGCAGCAGCGC
RpoH4_032:FP	GGAATTCCATATGAGCACCGCATTGGCACTG
RpoH4_032:RP	AACTGCAGCCTCCCTCTTACGCCGCTGC
RpoH5_032:FP	GGAATTCCATATGCTCAAGAAACTCCTGACTGGTG
RpoH5_032:RP	CCGCTCGAGTCAGGCTGCCATGTGG
RpoH1_pMMB:FP	CGGAATTCATGGCGACGATATCCAGCG
RpoH1_pMMB:RP	AACTGCAGTCAACCGGCCAGCTTCTGC
RpoH2_pMMB:FP	CGGAATTCATGGCCTACATCGACGATCC
RpoH2_pMMB:RP	AACTGCAGCTACGCGTCCGCCAGCAG
RpoH3_pMMB:FP	CGGAATTCTTGACGGAACTGGCAGTTTC
RpoH3_pMMB:RP	AACTGCAGTCACACCGGCAGCAGCGC
RpoH4_pMMB:FP	CGGAATTCATGAGCACCGCATTGACGC
RpoH4_pMMB:RP	AACTGCAGTTATGCCGCTGCGGCGAG
RpoH5_pMMB:FP	CGGAATTCGTGCTCAAGAAACTCCTGAC
RpoH5_pMMB:RP	AACTGCAGTCAGGCTGCCATATGGGCC
RpoH3_pCZ:FP	GCTCTAGACCTCCATTGGCGGTCACTG
RpoH3_pCZ:RP	CCCAAGCTTCTCCTGATTCCGGCATCCTC
AmpC_002:FP	CCGCTCGAGGATGATCGGACGGCGGGCT
AmpC_002:RP	TCCCCCCGGGTCAGCCCTTGAGCGCGTC

^aUnderlined portions of sequence represent restriction enzyme sites used for cloning of insert.

Genetic organization and phylogenetic analysis.

A genetic map of the genes flanking the gene encoding rpoE7 was prepared using Vector NTI software (Invitrogen). The sequences of RpoE7 paralogs found in A. baldaniorum Sp245 and orthologs from other bacteria were retrieved from the NCBI database (http://www.ncbi.nlm.nih.gov/) and the EBI server (http://www.ebi.ac.uk) and aligned by using CLC Sequence Viewer software and ClustalW (60). A phylogenetic tree was constructed using MEGA 5.05 software using 1,000 bootstraps and the Pearson model.

Insertional inactivation of genes encoding RpoE7 (rpoE7), lytic transglycosylase (ltg1), and β-lactamase (bla1).

The genes encoding RpoE7 and Ltg1 of A. baldaniorum Sp245 were insertionally inactivated and placed in the genome via allele replacement using a suicide plasmid vector (pSUP202) as described previously (28, 61). For the construction of an rpoE7-null mutant, two amplicons of approximately 1-kb size each, amplicon A and amplicon B, were PCR amplified. Amplicon A included half of the 5′ region of the rpoE7 open reading frame (ORF), along with its flanking region containing EcoRI and BglII restriction sites on its termini, and amplicon B included the remaining half of the rpoE7 ORF; i.e., the 3′ region of the rpoE7 gene with its flanking region containing BglII and PstI restriction sites at its termini. The two amplicons were first cloned separately in pGEM-T easy vector through TA cloning, and excised via digestion with restriction enzymes EcoRI/BglII and BglII/PstI, respectively. Both digested fragments A and B were ligated in pSUP202 using 3-fragment ligation, and the resulting recombinant (pSUP202AB) was selected on an LB plate amended with tetracycline. Furthermore, a kanamycin resistance (Km) cassette derived from pUC4K after digestion with BamHI was ligated in between amplicon A and amplicon B at BglII-linearized pSUP202AB (pAK003). The resulting plasmid was transformed in E. coli S17-1, and recombinants were selected on an LB agar plate supplemented with kanamycin and tetracycline. pAK003 was conjugatively mobilized in A. baldaniorum Sp245, and exconjugants were selected on kanamycin plates. Insertion of a Km^r gene cassette into the genomic copy of rpoE7 was confirmed by PCR amplification using a gene-specific (rpoE7::F and rpoE7::R) set of primers. The same strategy was also applied for the construction of ltg1::km (pAK004) and bla1::km (pAK005) mutants.

Cloning of genes encoding RpoE7, Ltg1, all RpoH paralogs, and Bla1 in different broad-host-range expression vectors (pMMB206, pAK032, and pAK002).

The gene encoding the complete ORF of RpoE7 was cloned in a low-copy-number expression vector, pMMB206, using the RpoE7::pMMB::FP and RpoE7::pMMB::RP set of primers having EcoRI and PstI restriction overhangs to produce the recombinant plasmid pAK006. The ORF of Ltg1 was amplified and cloned in pMMB206 and pAK032 using the primer pairs Ltg1::pMMB::FP/Ltg1::pMMB::RP and Ltg1_032::FP/Ltg1_032::RP into EcoRI/PstI and NdeI/PstI restriction sites, respectively, forming plasmids pAK007 and pAK008. The gene encoding bla1 was amplified using the primer pair Bla1_002::FP and Bla_002::RP and cloned into the restriction sites XhoI and XmaI of expression vector pAK002, forming plasmid pAK023. The complete ORFs of all 5 RpoH paralogs were also cloned in pAK032 (pAK012, pAK013, pAK014, pAK015, and pAK016) and pMMB206 (pAK017, pAK018, pAK019, pAK020 and pAK021) following the same method (Table 2). Once the cloning was confirmed by colony PCR and restriction digestion, each of the recombinant plasmids were transformed individually in E. coli S17.1, which was then used as donor for the conjugative mobilization of the recombinant plasmids into A. baldaniorum Sp245. The resulting exconjugants were selected on minimal malate agar plates amended with respective antibiotics.

Construction of rpoE7::lacZ, ltg1::lacZ, bla1::lacZ, and rpoH3::lacZ.

Approximately 500-bp regions upstream of the start codons of rpoE7, ltg1, bla1, and rpoH3 genes were PCR amplified and digested with XbaI and HindIII. The digested amplicons were ligated in the similarly digested vector pCZ750 to create promoter lacZ transcriptional fusions in pAK009, pAK010, pAK011, and pAK022, respectively. The correctness of the constructs was first confirmed by sequencing, and then plasmids were mobilized into A. baldaniorum Sp245 and the rpoE7::km mutant via biparental conjugation using E. coli S17-1 as a donor. The exconjugants were further selected on agarized minimal malate medium plates amended with tetracycline.

β-Galactosidase assay.

Cultures of A. baldaniorum Sp245 and rpoE7::km mutants harboring rpoE7::lacZ, ltg1::lacZ, bla1::lacZ, and rpoH3::lacZ fusions were inoculated in three flasks (containing 20 mL MM medium) each and allowed to grow up to an OD₆₀₀ of ∼1. Cells were pelleted at 6,000 rpm and resuspended in lysis buffer (50 mM phosphate buffer [pH 7.0], 0.1% SDS, 0.27% β-mercaptoethanol, and 100 μL chloroform). The β-galactosidase activity in the supernatant of the cell lysate was measured as described previously (62) and expressed as Miller units using the following formula: 1,000 × (OD₄₂₀ × 1.75 − OD₅₅₀)/reaction duration (min) × volume of culture assayed.

Aggregation, EPS estimation, and biofilm formation.

For comparing cell-cell aggregation between A. baldaniorum Sp245 and the rpoE7::km mutant, bacteria were grown overnight in MM medium and reinoculated in the same medium (10 mL); after 22 h of growth, tubes were kept in a standing position without shaking for 10 min before taking the photograph. EPS was estimated as described previously (63). For comparing biofilm formation, A. baldaniorum Sp245 and the rpoE7::km mutant were inoculated in 20 mL of LB medium in a 100-mL flask, incubated at 30°C with shaking (100 rpm) for 16 h to reach an optical density at 600 nm (OD₆₀₀) of 1.1 to 1.4. The cell suspension was diluted 1:100 in minimal malate broth without NH₄Cl medium, and 2 mL was transferred to each well of a 12-well microtiter plate and incubated under static conditions for 7 days at 30°C. Biofilm production was determined using crystal violet (CV), which was added to each well (0.1% [vol/vol] of 1% CV [wt/vol]). The CV was extracted with 2 mL of acetic acid 33% (vol/vol) and its concentration determined using a spectrophotometer at OD₅₉₀ (64).

Antibiotic sensitivity test.

The overnight-grown cultures of A. baldaniorum Sp245 and the rpoE7::km mutant were reinoculated in minimal malate medium and allowed to grow up to an OD₆₀₀ of 0.8, from which cultures of equal OD were taken for serial dilution. Culture aliquots (2 μL) were spotted from 10⁻¹ to 10⁻³ dilutions on minimal malate agar plates supplemented with ampicillin (450 μg/mL). We also amended the plates with low concentrations of triphenyl tetrazolium chloride (TTC) (50 mg/mL) to improve the visibility of A. baldaniorum Sp245 colonies. Plates were incubated for 48 h at 30°C before taking the picture. For studying the effect of expression of bla1 on the ampicillin resistance of E. coli, E. coli DH5α harboring pAK023 and empty plasmid pAK002 was streaked on LB agar plates amended with tetracycline (10 μg/mL) and tetracycline with ampicillin (450 μg/mL). To examine the effect of bla1 overexpression on resistance of E. coli to various β-lactam antibiotics, a disc diffusion assay for drug susceptibility was performed by the Kirby-Bauer method as described earlier (65) and the zone of inhibition measured. Similarly, A. baldaniorum Sp245 and the bla1::km mutant were also streaked on minimal malate medium plates without antibiotic and with ampicillin and kanamycin, respectively.

Identification of transcription start site by of 5′ RACE.

The 5′ end of the mRNA of rpoE7 was identified by 5′ RACE as described previously (66). Briefly, 2 μg RNA was reverse transcribed to cDNA using gene-specific reverse primer 1 (GSP1), followed by cDNA purification and poly(dA) tailing. Poly(dA)-tailed cDNA was then PCR amplified using GSP2 and oligo(dT) anchor primers. The amplicon so obtained was further used as a template in the next round of nested PCR using anchor and GSP3 primers. The final PCR product was then cloned in pGEM-T easy vector (Promega), and nucleotide sequences were determined by the chain termination method.

RT-PCR.

Reverse transcription PCR (RT-PCR) was performed as described previously (67). Briefly, cDNA was synthesized using 3 μg RNA according to the protocol of the Fermentas kit. No-RT PCR controls were carried out to check the DNA contamination for each RNA sample using housekeeping gene (rpoD). Relative expression of genes was done by real-time PCR using SYBR green I (Roche) in a LightCycler 480 II instrument. The real-time PCR was carried out according to the manufacturer’s instruction (Roche). The protocol used was as follows. The real-time PCR mixture contained 5 μL of 2× LightCycler 480 SYBR green I, 0.5 μM each primer, and 1 μL (2 to 5 ng) of cDNA. The cycling conditions comprised an initial incubation step at 95°C for 5 min, followed by 45 cycles of amplification for 10 s at 95°C, 10 s at 62°C (single acquisition), and 12 s at 72°C. The final cooling step was performed at 40°C for 30 s.

Scanning electron microscopy.

The morphological features of the wild type and the rpoE7::km mutant were examined by scanning electron microscopy (SEM). For this purpose, culture of both strains, i.e., the wild type and the rpoE7::km mutant, were taken (OD₆₀₀ 0.8), and fixed in 0.1 M sodium phosphate buffer containing 2.5% glutaraldehyde and 2% paraformaldehyde by vacuum infiltration for 2 h at room temperature and subsequent treatment for 6 to 12 h at 4°C. The fixative was removed and the samples washed twice with the same buffer. Dehydration was carried out in an ascending gradient series of acetone at 4°C, and the samples were dried in a critical point dryer in a CO₂ atmosphere. The dried samples were fixed to stubs with carbon cement, sputter-coated with gold (10 nm, Balzer Union SCD 020 sputter coating unit) and examined at different magnification (up to 8,000×) by a Leo 435 (variable-pressure) scanning electron microscope fitted with a Zeiss-Leica lens, equipped with digital imaging and a 35-mm photography system, and operating in high vacuum between 15 and 30 kV. (68)

Data availabilty.

All data are given in this article and the supplemental material.