Correspondence: Jitendra Singh, School of Biotechnology, Gautam Buddha University, Yamuna Express Way, Greater Noida 201308, Uttar Pradesh, India. Tel.: 0120 2344286; fax: 0120 2344202; e-mail: email@example.com
Bacteriocins are the toxic proteins produced by bacteria under stress condition to inhibit the growth of closely related bacterial strain(s). In our earlier study, purified recombinant xenocin–immunity protein complex from Xenorhabdus nematophila showed detrimental effect on six different insect gut residing bacteria. In this study, endogenous toxicity assay with xcinA and its catalytic domain under tightly regulated ara promoter was performed. Multiple sequence alignment and homology modelling revealed six conserved amino acid residues in the catalytic domain of xenocin. Site-directed mutagenesis was performed in all the conserved residues, followed growth profile analysis of all the mutants by endogenous toxicity assay. Among the six different conserved sites in catalytic domain of xenocin, we have identified one position where mutation resulted in no measurable reduction in the endogenous toxicity (K564), three positions with measurable reduction in the endogenous toxicity (E542, H551 and R570) and two positions where mutation caused a significant reduction in the toxicity (D535 and H538). Endogenous toxicity assay is validated by in vitro RNA degradation assay. Structural integrity of purified recombinant proteins was confirmed through circular dichroism and fluorescence spectroscopy. Our results indicate that D535 and H538 act as the acid–base pair for RNA hydrolysis.
Bacteriocins are ribosomally encoded, structurally, functionally and ecologically diverse toxins produced by bacteria to inhibit the growth of closely related bacterial strain(s) (Riley & Wertz, 2002; Gordon et al., 2007). They are produced by almost all the major lineages of Eubacteria and Archaebacteria (Riley & Wertz, 2002). Bacteriocins are generally secreted in the extracellular medium by the producer where they target specific receptors on the surface of target cells. Inhibition of target cells occurs by different mechanisms such as enzymatic nuclease (DNase or RNase) as well as pore formation in the cytoplasmic membrane (Cascales et al., 2007). Their structural gene encodes three distinct domains: (1) a domain involved in the recognition of specific receptor, (2) a domain involved in the translocation and (3) a domain responsible for their toxic activity. The average molecular mass of a typical ribosomally encoded bacteriocin lies within the range of ~ 25 to 80 kDa (Cursino et al., 2002).
Xenorhabdus nematophila is a motile, gram-negative entomopathogenic bacterium belonging to the family Enterobacteriaceae and is known to form symbiotic association in the gut of a soil nematode of the family Steinernematidae (Boemare & Akhrust, 1988; Herbert & Goodrich-Blair, 2007). Under standard laboratory conditions, X. nematophila secretes several extracellular products, which include lipase(s), phospholipase (s), protease(s) and several broad spectrum antibiotics (Akhurst, 1982; Nealson et al., 1990). These degradative enzymes are believed to be secreted in the insect haemolymph during the stationary phase of bacterial growth and are responsible for the breakdown of macromolecules of the insect cadaver to provide nutrient to the developing nematode, while the antibiotics play a major role in the suppression of contamination of the cadaver by other soil microorganisms. In our earlier study, we have isolated and characterized xenocin operon encoded by the genome of X. nematophila. Results showed that the transcription of xenocin was upregulated by iron-depleted condition, high temperature and in the presence of mitomycin C. Recombinant xenocin–immunity protein complex showed broad range of antibacterial effect, not only limited to the laboratory strains, but also to six other bacteria isolated from the gut of Helicoverpa armigera (Singh & Banerjee, 2008). These results compel us to study the structure of such an important antibacterial protein in detail. Therefore, in our recent studies, three-dimensional structure of xenocin has been deciphered by automated homology modelling (Singh, 2012). It is a multi-domain protein consisting of 576 amino acid residues. First 327 amino acid residues from the N′ terminal region form translocation domain (T), 328–476 amino acid residues form middle receptor domain (R) and amino acid residues from 477 to 576 form catalytic domain (C) at C′ terminal (Singh, 2012). In this study, xcinA as well as its catalytic domain was cloned under tightly regulated ara promoter, and endogenous toxicity assay was performed in the presence of arabinose. Six conserved amino acid residues in the catalytic domain were identified by multiple sequence alignment and their location on the surface was viewed by homology model structure. Site-directed mutagenesis was performed in all the conserved amino acids. Growth profile of wild-type catalytic domain and its mutant variant was analysed by performing endogenous toxicity assay. Homogeneity of the purified recombinant wild-type catalytic domain and its mutants was confirmed by Western blot. Structural integrity of the purified recombinant proteins was analysed by intrinsic tryptophan fluorescence and circular dichroism.
Material and methods
Bacterial strain, media and culture conditions
Escherichia coli strain DH5α (Bethesda Research Laboratories) was used as the host for cloning. The E. coli strains, TOP10 and BL 21(DE3) pLysS, were used in the expression studies. E. coli XL-Blue cells were used for the site-directed mutagenesis studies. The plasmid vector pGEM-T Easy from Promega (Madison, WI) was used for PCR cloning. LB medium was used for growing bacterial strains. Ampicillin, kanamycin and chloramphenicol were used at 100, 35 and 25 μg mL−1, respectively.
Cloning of xcinA and its catalytic domain under ara promoter
Primer PColF with PstI site at 5′ and primer 2 with HindIII site at 3′ were used to amplify xcinA alone, from the 4.3-kb genomic DNA fragment. The amplified 1.7-kb product was ligated in pGEM-T Easy vector producing pJS2 plasmid. Plasmid was digested with PstI and HindIII, and the released DNA fragment of 1.7 kb was ligated to pBAD vector resulting in plasmid pJSR2.
For catalytic domain, forward primer PDomF with PstI and backward primer 2 with HindIII site were used for PCR amplification. Amplified 318-bp product was cloned in pGEM-T Easy vector producing pJS3. 318 bp was excised from pJS3 by digestion with PstI and HindIII and ligated to pBAD vector, resulting pJSR3 construct. pJSR2, pJSR3 and pBAD without insert were finally electroplated in the E. coli TOP10 cells and gave rise to JSR2, JSR3 and JSR4 strains, respectively. All these strains were studied by endogenous toxicity assays.
Separation of protein complexes
For the isolation of individual domain proteins, the Ni-NTA purified catalytic–immunity domain protein complex was dialysed against 20 mM glycine–HCl buffer, pH 3.0, overnight and purified by a Sepharose-SP column (HiTrap SP; Amersham Biosciences) as described earlier (Singh & Banerjee, 2008). The catalytic domain was eluted first with NaCl gradient (0–2 M, pH 3) followed by the immunity domain with 20 mM sodium phosphate buffer, pH 8.0. The individual domains were dialysed against 50 mM sodium phosphate buffer, pH 8.0, for further studies.
Antiserum and Western blotting
The 64-kDa xenocin was purified as described earlier (Singh & Banerjee, 2008), and SDS-PAGE was performed by following the procedure described by Laemmli (1970). Antiserum against purified recombinant xenocin was raised in rabbit with standard protocol. Western blot was performed with 500 ng each purified sample with standard molecular protocol using anti-xenocin serum (1 : 2000 dilution).
Site-directed mutagenesis in the catalytic domain of xenocin was performed by Quick Change Site-Directed kit (stratagene) as per recommended protocol by manufactures. Ligation and transformation of competent E. coli XL1-Blue cells were performed using standard molecular biology procedures. In brief, Asp-535, His-538, Glu-542, His-551, Lys-564 and Arg-570 were altered to alanine with both strands harbouring a mutation in the middle were synthesized and used in PCR. Construct pJSR3 (for endogenous toxic studies) and pJC4 (for protein purification) were used as template to amplify a double-stranded nicked circle using different primers as listed in Table 1 resulting in pD535A, pH538A, pE542A, pH551A, pK564A, pR570A and pJC4(D535A), pJC4(H538A), pJC4(E542A), pJC4(H551A), pJC4(K564A), pJC4(R570A), respectively. All the constructs of pBAD were transformed in E. coli TOP10 resulting in D535A, H538A, E542A, H551A, K564A and R570A strains, and constructs in pET28 were transformed in BL 21 (DE3) pLysS resulting in JC4(D535A), JC4(H538A), JC4(E542A), JC4(H551A), JC4(K564A) and JC4(R570A) strains. All the strains and plasmid used in this study are listed in Table 2.
pGEM-T Easy containing xcinA between PstI and HindIII
pBAD His(c) containing xcinA
pGEM-T Easy containing catalytic domain between PstI and HindIII
pBAD His(c) containing 318 bp catalytic domain
pBAD His(c) alone without insert
pJSR3 with D535A xcinA
pJSR3 with H538A xcinA
pJSR3 with E542A xcinA
pJSR3 with H551A xcinA
pJSR3 with K564A xcinA
pJSR3 with R570A xcinA
pJSR4 with D535A xcinA
pJSR4 with H538A xcinA
pJSR4 with E542A xcinA
pJSR4 with H551A xcinA
pJSR4 with K564A xcinA
pJSR4 with R570A xcinA
Endogenous toxicity assays
Endogenous toxicity assays were performed in E. coli TOP10, as all the constructs of pBAD were transformed in E. coli TOP10 resulting in D535A, H538A, E542A, H551A, K564A and R570A strains. For endogenous toxicity assay, overnight cultures were diluted 100-fold in fresh medium and grown till log phase [optical density at 600 nm (OD600) = 0.4–0.5] and then diluted again to OD600 = 0.01 in fresh medium with 0.2% l-(+)-arabinose (Sigma, St. Louis, MO). Optical density was monitored at 600 nm using a spectrophotometer. All cultures were grown at 37 °C in LB medium containing 100 mg of ampicillin mL−1, with continuous shaking of ≥ 225 r.p.m. All the experiments were performed in triplicates, and mean values of three results were used to show the growth in percentage (%) at different interval of time.
In vitro RNase activity
In vitro RNase degradation assay was performed as per protocol described earlier (Singh & Banerjee, 2008) with purified recombinant wild-type catalytic domain and its all mutant variants. Briefly, RNase activity was measured using total bacterial RNA from E. coli strain BL 21(DE3)/pLysS as the substrate. The reaction mixture (20 μL) contained 1.2 μg of RNA in 50 mM Tris–HCl buffer (pH 7.5), 50 mM NaCl, 5 mM EDTA and the protein sample to be tested. After 1.5 h of incubation at 37 °C, 2.5 μL of the loading buffer (40% sucrose, 0.125 M EDTA, 0.5% sodium dodecyl sulfate; pH 8) was added, and the mixture was heated at 95 °C for 2 min and resolved on a 1% agarose gel containing ethidium bromide.
Intrinsic tryptophan fluorescence
Intrinsic tryptophan fluorescence spectra of wild-type catalytic domain and its mutants were measured by Varian spectrofluorometer. Spectra were recorded in 20 mM sodium phosphate buffer at protein concentration of 1 μM using excitation wavelength of 295 nm with excitation and emission slit width set at 5 nm.
The far UV CD spectrum was recorded between 190 and 250 nm (500 μL sample volume) on a Jasco J-810 spectropolarimeter equipped with a Jasco Peltier temperature controller at 20 °C using 1-mm optical path length quartz cells, the step size was 0.5 nm with 1-nm bandwidth at a scan speed of 50 nm min−1. Averages of five scans were obtained for blank and protein spectra, and data were corrected for buffer contribution. Measurement was taken at protein concentration between 1 and 2 μM under nitrogen flow. The results are expressed as mean residue ellipticity in units of degree cm−2 dmol−1.
Results and discussion
Endogenous toxicity assay
Xenocin is a multi-domain toxic protein consisting of translocation domain, receptor domain and catalytic domain. Toxicity of xenocin lies in its catalytic domain. To study the detrimental effect of xenocin alone, it was cloned under tightly regulated ara promoter. Xenorhabdus nematophila was not able to uptake arabinose, which is inducer for ara promoter. Therefore, all the endogenous toxicity assays were performed in the E. coli TOP10, the recommended host for the expression vector containing ara promoter like pBAD. In the endogenous toxic assay, growth profile of arabinose-induced JSR4 strain containing vector alone was considered as 100% and compared with induced JSR2 strain containing xenocin alone. Results showed that there was no change in growth profile of JSR2 strain after first hour of induction; however, growth was inhibited by 50% after second hour and was further declined in consecutive hours as shown in Fig. 1. In case of catalytic domain, growth declined immediately after induction and it was inhibited by almost 70% in first hours of induction, 80% in second hour and was further declined in the consecutive hours as shown in Fig. 1.
Site-directed mutagenesis and endogenous toxicity assay
In our previous work, we have shown that catalytic domain of xenocin has RNase activity (Singh & Banerjee, 2008). On the basis of multiple sequence alignment (Supporting Information, Fig. S1) and homology model, six conserved amino acids residues were predicted to form active site in catalytic domain including D535, H538, E542, H551, K564 and R570 as shown in Fig. 2a. Catalytic mechanism of RNA hydrolysis has been thoroughly studied by protein engineering and crystallography (Gilliland, 1997). RNase A has two active histidine residues that cooperate during the catalytic cycle (Raines, 1998; Scheraga et al., 2001). Other ribonuclease, such as barnase and colicin E3, precede probably through the similar mechanism, but in these cases, histidine and glutamic acid act as catalytic residues (Walker et al., 2004) Figs. S2, S3, S4 and S5. Killing of the target cells by multi-domain E colicins occur in three different stages. First step to bind with receptor, followed by its translocation into the periplasmic space and finally endogenous toxicity in the cytoplasm of target cells by its catalytic domain (Carr et al., 2000).
Primary sequence of catalytic domain from xenocin revealed the presence of four histidine residues. Interestingly, three of them were found conserved in multiple sequence alignment (Fig. S1). Three-dimensional structure of catalytic domain highlighted that these two (H538 and H551) conserved histidine residues are present on the surface as shown in Fig. 2b. Therefore, they may be responsible for the hydrolysis of RNA by a mechanism similar to RNase A. However, due to localization of aspartic acid (D535) on the surface of catalytic domain as shown in Fig. 2b, its role in RNA hydrolysis by mechanism similar to barnase and colicin E3 cannot be ruled out. Therefore, to determine individual role of conserved amino acid residues in the putative active site of catalytic domain of xenocin, site-directed mutagenesis was performed. All the conserved amino acid residues were mutated to alanine, and endogenous toxicity assay was performed with each mutant strain.
Growth profile of JSR4 strain–containing vector alone was considered as 100% and compared with growth profile of D535A, H538A, E542A, H551A, K564A and R570A strains. From the predicted structure of catalytic domain of xenocin as shown in Fig. 2b, it was observed that H538 was the most surface-exposed histidine residue among the four other present in the catalytic domain. Endogenous assay showed that mutation at H538 position results in the reduction of toxicity by more than 90% after 8 h postinduction as shown in Fig. 2c, which confirmed the role D535 as an important residue of the putative active site. As second conserved histidine residues H551 was nearer to H538 and exposed on the surface, it may behave as the second histidine residue required for the hydrolysis of RNA by a mechanism similar to RNase A ribonuclease. Therefore, H551 was mutated to alanine, and endogenous assay was performed. Results showed that there was only 50% reduction in endogenous toxicity in H551A strain after 8 h of induction as shown in Fig. 2c. One reason for such minimum reduction in endogenous toxicity in H551A strain is that it could be due to partial exposure of H551 to the surface as compared to H538 as revealed by the surface view model of catalytic domain as shown in Fig. 2b. This result indicates that RNA hydrolysis mechanism of catalytic domain of xenocin is different from RNase A ribonuclease.
D535 and E542 are two acidic amino acid residues that are conserved, exposed to surface as well as close to the H538 as shown in Fig. 2a and b. These two residues may be responsible for the hydrolysis of RNA by mechanism similar to barnase and colicin E3. Therefore, these two residues were mutated to alanine and analysed by endogenous assay. Endogenous toxicity assay result showed that toxicity was reduced by 70% after 8 h postinduction in E542A strain as shown in Fig. 2c. Structural studies showed that E542 was also a part of cleft formed by D535 and H538, which is exposed to the surface as shown in Fig. 2b. However, studies with D535 strain showed significant reduction (88%) in the endogenous toxicity after 8 h postinduction as shown in Fig. 2c; moreover, D535 was the closest amino acid residue with respect to H538 as shown in Fig. 2a. This result further confirmed the role of D535 as a second critical amino acid residue in the formation of putative active in catalytic domain of xenocin.
Studies with R570A strain resulted in 60% reduction in toxicity after 8 h postinduction as shown in Fig. 2c, which indicate the importance of this residue in the activity of catalytic domain. Although in primary sequence, R570 is located far from H535, H538 and E542, due to the protein conformation, it became a part of the cleft formed by these amino acids as shown in Fig. 2b. Moreover, it might be possible that positive charge on the R570 assists in the binding of RNA at putative active site by neutralizing the negative charges present on the backbone of RNA due to phosphate group. Interestingly, there was no reduction in toxicity in K564A strain whose growth profile was similar to wild type as shown in Fig. 2c. In three-dimensional structure of catalytic domain as shown in Fig. 2a, K564 lies very far from other conserved residue hence it is not part of putative active site but may assist in binding of RNA to the active due to its positive charge. Hence, we concluded that D535 and H538 act as acid–base pair to hydrolyse RNA, and D535, H538, E542 and R570 formed the active site in catalytic domain of xenocin.
Purification of catalytic domain variants and Western blot
To confirm that the loss of endogenous toxicity in catalytic domain variant strains was not due to the conformational change of the protein induced by site-directed mutagenesis, site-directed mutations were performed in pJC4 construct containing catalytic-immunity domain complex at all the six conserved sites. Wild-type catalytic-immunity domain complex and all the mutant complexes were purified with Ni-NTA chromatography under native conditions. Further, domains were separated and purified by ion exchange chromatography as discussed in 'Material and methods'. The homogeneity of purified catalytic domain variants was further confirmed by Western blot analysis using anti-rabbit serum generated against full-length xenocin protein as shown in Fig. 3a. Expression and purification of the immunity domain with the mutated catalytic domains indicate that mutation did not affect the formation of stable protein complexes. From this observation, we may hypothesize that catalytic domain consists of two functional regions. N′ terminal region of catalytic domain is responsible for the binding of immunity protein, whereas C′ terminal consists of active site.
In vitro RNase degradation assay with catalytic domain variants
To validate the endogenous toxicity assay, in vitro RNase degradation assay was performed with recombinant catalytic wild-type domain and its mutant variants. Result showed that total RNA isolated from E. coli BL 21(DE3)/pLysS cell was intact and not degraded when incubated with purified recombinant domain D535A and H538A mutant protein as shown in Fig. 3b lane 2 and 3, respectively. Moreover, these results were comparable to negative control experiment, which was performed without protein as shown in Fig. 3b lane 1. Therefore, we inferred that the D535 and H538 are the key amino acid residues of the active site of the catalytic domain of xenocin. As the growth profile of K564A strain was similar to wild type in endogenous toxicity assay, in vitro RNA degradation by K564A mutant protein was almost equivalent to a positive control wild-type catalytic domain as shown in Fig. 3b lane 6 and lane 7, respectively. There was partial degradation of RNA by E542A mutant protein as shown in Fig. 3b lane 8, which corroborate with its endogenous toxicity assay which showed 70% reduction in the toxicity. Similarly, H551A and R570A showed 50% and 60% reduction in endogenous toxicity, which corroborates with their in vitro RNA degradation assay as shown in Fig. 3b lane 4 and 5, respectively. Therefore, with in vitro RNA degradation assay, we have validated our endogenous toxicity assay performed with wild-type catalytic domain and its mutant variants.
Intrinsic tryptophan fluorescence spectra were obtained reflecting changes in the secondary and tertiary structure of the protein. The λmax of tryptophan in the solution is 345 nm, indicating the degree of solvent exposure. Wild-type catalytic domain showed a fluorescence emission spectra characteristic of a folded protein with tryptophan side chain buried in a protein core displaying a λmax of 326 nm as shown in Fig. 4a. All the mutants had the same λmax (326 nm) as compared to wild-type catalytic domain as shown in Fig. 4a. This result indicated that mutation in the catalytic domain at different positions did not change the secondary conformation. Hence, we confirmed that reduction in toxicity in the endogenous toxicity assays of different mutants is due to the absence of particular residues in the active site and not due to the conformational changes. These results were further confirmed by circular dichroism studies with purified recombinant wild-type and mutant variants. Far UV spectra of wild type catalytic domain displayed maxima at 227 nm and minima at 202 nm respectively as shown in Fig. 4b. All the mutants also displayed maxima at 227 nm and minima at 202 nm in far UV CD spectra. Thus, consistency between fluorescence data and CD measurement indicates that the structures of the mutant proteins are similar to the wild-type catalytic domain.