Functional and transcriptional analysis of chromosomal encoded hipBAXn2 type II toxin-antitoxin (TA) module from Xenorhabdus nematophila.

Xenorhabdus nematophila is an entomopathogenic bacterium that synthesizes numerous toxins and kills its larval insect host. Apart from such toxins, its genome also has a plethora of toxin-antitoxin (TA) systems. The role of TA systems in bacterial physiology is debatable; however, they are associated with maintaining bacterial genomic stability and their survival under adverse environmental conditions. Here, we explored the functionality and transcriptional regulation of the type II hipBAXn2 TA system. This TA system was identified in the genome of X. nematophila ATCC 19061, which consists of the hipAXn2 toxin gene encoding 278 amino acid residues and hipBXn2 encoding antitoxin of 135 amino acid residues. We showed that overexpression of HipAXn2 toxin reduced the growth of Escherichia coli cells in a bacteriostatic manner, and amino-acids G8, H164, N167, and S169 were key residues for this growth reduction. Promoter activity and expression profiling of the hipBAXn2 TA system was showed that transcription was induced in both E. coli as well as X. nematophila upon exposure to different stress conditions. Further, we have exhibited the binding features of HipAXn2 toxin and HipBXn2 antitoxin to their promoter. This study provides evidence for the presence of a functional and well-regulated hipBAXn2 TA system in X. nematophila.


Introduction
Bacteria have a typical genetic system that is responsible for the extensive potential for survival in various adverse environments like temperature variance, nutritional starvation, antibiotics, and other unfavorable conditions (Boutte and Crosson 2013). Under such conditions, Xenorhabdus nematophila (X. nematophila) is a highly successful pathogen with an ability to kill insects (Stilwell et al. 2018).
X. nematophila is an entomopathogenic bacterium and symbiotic to Steinernema carpocapsae nematode (Park and Kim 2000). This bacterial association of nematode is very lethal to many insects and considered as the cause of death in insect larvae (Mahmood et al. 2020). At the juvenile stage, free-living, nematode S. carpocapsae invade through the digestive tract in the insect. Nematodes penetrate the insect larvae body chambers and liberate X. nematophila into the hemolymph (Martens et al. 2003). Bacteria doubles quickly and thus kill larvae by secreting toxins. By whole-genome sequencing, it was observed that X. nematophila is a reservoir of numerous biological molecules and have great biosynthetic potential (Bentley et al. 2002;Chaston et al. 2011). A range of regulatory proteins and alarming molecules like TA systems are abundantly discovered in numerous bacteria and archaea (Song and Wood 2020). These are primarily consisting of a protein toxin and an RNA or protein antitoxin component. The toxin may suspend some essential cellular processes in a similar pattern like an antibiotic while antitoxin covers up the toxin's activity. In bacterial cell physiology, TA systems are involved in bio lm formation, phage inhibition, genetic element maintenance, persister cell formation, and growth diminution during stress (Page and Peti 2016; Song and Wood 2020). TA systems have been classi ed in eight different classes and among them, the type II TA system is the most investigated (Song and Wood 2020).
In our previous study, we have analyzed TAome for type II TA systems in X. nematophila (Yadav and Rathore 2018a) and found there were three different hipBA TA homolog loci present in this TAome. One homolog named as hipBA Xn was already described as a bona de type II TA system and hipBA Xn3 was a pseudo-type TA system (Mohit Yadav & Jitendra Singh Rathore 2020) while other hipBA Xn2 TA system was still not studied. Thus, to determine that these hipBA homolog operons encode active TA systems, it is needed to study them. In this study, we are exploring the activity of type II hipBA Xn2 TA operon on the chromosome of X. nematophila. We have performed the functional and transcriptional attributes of a typical type II TA system to study the activity of the hipBA Xn2 TA system.

Materials And Methods
Bacterial Strains, primers, and culture conditions Bacterial strains and plasmids are listed in Table S1 and primers are in Table S2. We used X. nematophila strain 19061 and cultured it at 28ºC with 220 rpm shaking conditions. Other strains were E. coli DH5 (Bethesda Research Laboratories) used as a cloning host, E. coli TOP10 cells (Invitrogen) used for toxicity assay, and E. coli BL21 (DE3) (Novagen) used in protein expression analysis. These strains were cultivated in Luria-Bertani (LB) medium with 220 rpm shaking conditions at 37°C. With the requirement, culture media was supplemented with 100 g mL −1 ampicillin and 50 g mL −1 kanamycin. Primers were synthesized by Integrated DNA Technologies (IDT) and other chemicals were used from HiMedia laboratories. Enzymes were purchased from New England Biolabs (NEB).
Bioinformatics of Putative hipBA Xn2   Promoter analysis of the upstream region of the hipBA Xn2 TA system was performed by BPROM (Solovyev and Salamov 2010).
Cloning of hipBA Xn2 TA genes The primer sequences and recombinant constructs used in this work were listed in Table S2 and Table S1, respectively. Cloning strategy involved the primer 1 and primer 2 with PstI and HindIII restriction enzymes sites respectively to amplify 813 bp of hipA Xn2 toxin gene from the genome of X. nematophila by polymerase chain reaction (PCR). ara promoter characterized vector pBAD/His C and ampli ed PCR products were digested with PstI and HindIII. These digested products were further ligated to produce a recombinant construct pJSM1. In the cloning of the hipB Xn2 antitoxin gene, primer 3 and primer 4 containing BamHI and HindIII sites respectively were used to amplify 408 bp of antitoxin gene. Expression vector pET28 (a) and ampli ed PCR products were digested with BamHI and HindIII and ligated to produce a recombinant construct pJSM2. Likewise, a full hipBA Xn2 TA operon of 1221 bp size, comprising the hipB Xn2 antitoxin gene and hipA Xn2 toxin gene was PCR ampli ed with primer 5 and primer 6. Plasmid vector pBAD/His C and PCR ampli ed products were digested with PstI and HindIII. These digested products were further ligated to produce a recombinant construct pJSM3. Further, all the above recombinant plasmids were transformed in E. coli DH5 cells with standard protocol.

Protein expression and puri cation
Expression and puri cation of hipBA Xn2 TA proteins were analyzed in E. coli cells. For this, recombinant constructs pJSM1 and pJSM3 were transformed into E. coli TOP10 cells and pJSM2 was transformed into E. coli BL21 (DE3) cells. The primary culture was prepared by inoculating transformed cells in LB broth supplemented with 100µg/mL Ampicillin (JSM1 and JSM3) or 50µg/mL Kanamycin (JSM2) and incubated at 37°C for overnight. 1% v/v of primary culture was used in 50ml LB broth for preparing secondary culture incubated at 37°C with a 220 rpm shaking condition. In the case of JSM1 and JSM3, culture was induced with 0.2% of L-arabinose while for JSM2; it was induced with 1mM of isopropyl-β-Dthiogalactopyranoside (IPTG), at the OD 600 value of 0.5 for 6 hours. 50mL of induced culture was harvested and pelleted to dissolve in 10 mL of lysis buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, pH 8.0). To prepare a total cell lysate, cells were sonicated on ice for 20 cycles (10 s on/off) and centrifuged at 12,000 x g for 30 minutes at 4˚C. Ni-NTA a nity chromatography was used to separate recombinant HipA Xn toxin and HipB Xn antitoxin proteins. Supernatant having soluble proteins was loaded on the Ni-NTA super ow column (Qiagen). After passing proteins, column was washed with washing buffer (50 mM NaH 2 PO 4 , 300mM NaCl, 20 mM Imidazole, pH 8.0) and hipBA Xn2 proteins were eluted by 15 mL of elution buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 250 mM Imidazole, pH 8.0). These eluted proteins were dialyzed against dialysis buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 2 mM β-mercaptoethanol, 20% glycerol, pH 8.0) at 4˚C overnight. Thermo Scienti c™ NanoDrop 2000 was utilized for measuring the purity and concentration of these puri ed proteins. Further, these puri ed proteins were analyzed on 15% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and con rmed by Western blotting probed with mouse anti-His monoclonal primary antibody (Bio-Rad, USA) as per the protocol described elsewhere (Mohit Yadav & Jitendra Singh Rathore 2020).

Toxicity assessment
The toxicity assessment was performed in liquid and solid LB medium. Recombinant constructs pJSM1 and pJSM3 have transformed into E. coli TOP 10 cells resulting in JSM4 and JSM5 strains. These strains were inoculated in liquid LB medium supplemented with 100μg/mL of ampicillin and incubated at 37°C overnight with a 220 rpm shaking condition. For assay toxicity in the liquid medium, overnight grown culture of JSM4 and JSM5 was inoculated with 1:100 in LB medium and incubated at 37°C. Cultures were induced with 0.2% of L-arabinose at the OD 600 of 0.1. Samples were harvested at each hour postinduction and OD 600 was measured with a spectrophotometer (Perkin Elmer, Waltham, MA). For assay toxicity on solid medium, these harvested samples were serially diluted and spotted on LB agar plates containing 100μg/mL of ampicillin. After overnight incubation at 37°C, colony-forming units (CFU) were counted. Mean value of three independent experiments were used to show the growth parameters at different interval of time.

Site-directed mutagenesis
Site-directed mutagenesis at the active site residues of hipA Xn2 toxin was performed by following the protocol as described elsewhere (Singh 2013; Yadav and Rathore 2020). Primers used are listed in Table   S2. Active site residues Gly-8, Ser-129, His-164, Asn-167, Ser-169, Asp-185, and Thr-220 were substituted with alanine. In brief, the pJSM1 construct was used as a template and mutated in pG8A, pS129A, pH164A, pN167A, pS169A, pD185A, and pT220A recombinant plasmid constructs. These recombinant constructs were transformed in E. coli TOP10 cells using a standard protocol and resulted in mutants G8A, S129A, H164A, N167A, S169A, D185A, and T220A as described in table 1. Further, we also performed the toxicity assessment of mutants and wild type hipA Xn2 toxin by following the method as described before in materials and methods.

Reporter assay
For the reporter β-galactosidase assay, 529 bp upstream region of the hipBA Xn2 TA system was used as the promoter. The method for determining β-galactosidase assay was used as described elsewhere (Yadav and Rathore 2018a, 2020). In brief, constructs/strains used in this study are listed in Table S1 and primer sequences details are given in Table S2. Promoter activity in different stress conditions (elevated temperature, antibiotic, and nutrient starvation) was measured and expressed in Miller units (MU) (Miller 1972). The hipBA Xn2 TA promoter was cloned in the pGEM-T Easy vector and resulting in the JSM6 strain containing the promoter-lacZ fusion. For primary culture, the fusion construct was incubated overnight at 30°C with 220 rpm shaking in LB medium supplemented with 100µg/mL ampicillin. 1:100 of primary culture was inoculated in secondary culture with LB media and incubated at 30°C. All samples were exposed to different stress conditions at OD 600 of 0.1 and promoter activity was measured.

Reverse Transcriptase Polymerase Chain Reaction (RT-PCR)
Total RNA and RT-PCR analysis from X. nematophila were performed by following the method as described previously (Yadav and Rathore 2018b; Mohit Yadav & Jitendra Singh Rathore 2020). In brief, 1% of the overnight X. nematophila culture was inoculated in nutrient broth and subjected to different stress conditions described as elevated temperature, antibiotic, and nutrient starvation. At different time intervals, cells were harvested and total RNA was isolated for RT-PCR analysis. Primer details for RT-PCR are given in Table S2. A sequence of 16s RNA from X. nematophila was used as a control gene. One-step RT-PCR kit (Qiagen) was utilized for this analysis. PCR reaction included 80 ng RNA (DNase treated), 10 mM of deoxynucleoside triphosphate (dNTP) mixture, 5 mM of primers, 5.0 µL of RT-PCR buffer and l µL of enzyme mixture in 50 µL reactions. PCR was analyzed on 1 % agarose gel by EtBr staining.

Gel Shift Assay
Gel shift assay was performed as described elsewhere (Mohit Yadav & Jitendra Singh Rathore 2020).
Puri ed recombinant HipA Xn2 toxin and HipB Xn2 antitoxin proteins dialyzed against 1X assay buffer (1.8M NaCl pH 7.5, 1M Tris pH 7.2, 1% SDS) and concentrated by Amicon Ultra-0.5 device (Merck). The promoter region of the hipBA Xn2 TA system (529 bp) was ampli ed and puri ed as described before in materials and methods. A concentration gradient of puri ed HipA Xn2 toxin and HipB Xn2 antitoxin proteins was incubated with 250 ng of puri ed promoter DNA containing 1X assay buffer for 30 minutes at room temperature. These prepared samples were analyzed on 8% Native PAGE gel at 4˚C. Gels were visualized by Ethidium bromide (EtBr) dye.

Results
Genetic organization of a putative hipBA Xn2 TA system A homolog of hipBA TA system in the genome of X. nematophila ATCC 19061 (NCBI Refseq NC_014228) was identi ed at position 3774379-3775635 bp on the negative strand under XNC1_operon 0746 locus tag and named as hipBA Xn2 TA system. This TA system was comprised of an 813 bp hipA Xn2 toxin gene which encodes for a 270 amino-acid residue protein and a 408 bp hipB Xn2 antitoxin gene which encodes for a protein having 135 amino acids as shown in Fig. 1. The chromosomal location of the hipA Xn2 toxic gene was 3774379-3775191 bp with locus tag XNC1_3911 and the hipB Xn2 antitoxin gene was positioned at 3775228-3775635 bp with locus tag XNC1_3912. The intragenic space between these two genes was 37 bp. Upstream region of the hipB Xn2 antitoxin gene was also explored for the predicted 529 bp promoter entity of the hipBA Xn2 TA system with -10 box 5'TGCTATTAT3' having a score of 74 and -35 box 5'TTACAA3' having a score of 32 as depicted in Fig. 1. Binding sites for transcription factors ihf and rpoS17 were also identi ed in this promoter region with sequences 5'AATAAAAT3' and 5'CTATTATA3' respectively as illustrated in Fig. 1.

Phylogenetic analysis of HipA Xn2 toxin and HipB Xn2 antitoxin
To relate the homolog of the hipBA TA system, a detailed phylogenetic analysis was performed at the protein level for the hipBA Xn2 TA system. Based on BLASTP algorithm results, for HipA Xn2 toxin phylogenetic analysis, different bacteria namely, Photorhabdus luminescens, Xenorhabdus khoisanae, Lelliottia amnigena, Morganella morganii, Morganella psychrotolerans, Cedecea neteri, Klebsiella pneumoniae, Serratia marcescensEnterobacter sp. 638 and E. coli were selected as the source of different serine/threonine protein kinases. These kinases had a higher degree of similarity with HipA Xn2 toxin as illustrated in Fig. 2a. Closely related bacterium Photorhabdus luminescens' kinase forms a separate cluster with HipA Xn2 toxin and it was found that HipA Xn2 toxin was more closely related to kinase from E. coli with a clade credibility value of 100. However, distantly related kinases from Morganella psychrotolerans and Xenorhabdus khoisanae had their separate cluster with a clade credibility value of 92 as depicted in Fig. 2a. Similarly, for HipB Xn2 antitoxin phylogenetic analysis, according to BLASTP results, a series of transcriptional regulatory proteins from different bacteria namely Xenorhabdus szentirmaii, Photorhabdus luminescens, Xenorhabdus khoisanae, Xenorhabdus eapokensis, Halomonas saccharevitans, Edwardsiella hoshinae,Leclercia adecarboxylata, Salinicola sp. MIT1003 and E. coli were screened. These regulatory proteins had a higher degree of similarity to HipB Xn2 antitoxin as shown in To characterize the hipBA Xn2 TA system, both identi ed TA proteins were structurally modeled as depicted in Fig.2. Multiple templates for these models were analyzed by LOMETS from the PDB library. Z-score was calculated to measure the signi cance of the models. For HipA Xn2 toxin, 4pu3A was the most signi cant template with a normalized Z-score of 1.18 as illustrated in Fig. 2a and for HipB Xn2 antitoxin, it was 1b0nA with a normalized Z-score 1.52 as shown in Fig. 2b. Further, these models were characterized based on C-score, TM-score, and RMSD. For the HipA Xn2 toxin model, C-score was 0.85 and other parameters including TM-score and RMSD were 0.61±0.14 and 7.9±4.4Å respectively. While for HipB Xn2 antitoxin model C-score was 3.17 and other parameters including TM-score and RMSD were 0.36±0.12 and 11.8±4.5Å respectively. Moreover, Protein Structure Validation Software Suit (PSVS version 1.5) was also used to validate these models. Ramachandran plot analysis and statistics showed that 75.70 % of residues in the HipA Xn2 model were in the favored regions and 93.80 % in the allowed regions as Expression and puri cation analysis of hipBA Xn2 TA system proteins Different recombinant constructs were developed for the expression and puri cation study of HipA Xn2 toxin and HipB Xn2 antitoxin from X. nematophila, and details are described in Table S1. E. coli cells were used as host cells for this purpose due to its lavishly controlled expression features ( Thomet et al. 2019). hipA Xn2 toxin gene with a size of 813 bp that encoding a 31.04 kDa protein was cloned in pBAD/His C vector for its tight regulation attribute. The expression pro le of the hipA Xn2 toxin gene (recombinant strain JSM4) was analyzed on 15% SDS-PAGE as depicted in Fig.   3a. As the protein band of HipA Xn2 toxin was not clearly distinguished in the total lysate (lane TL) therefore, under native conditions, HipA Xn2 toxin protein was puri ed with Ni-NTA a nity chromatography and a clear single band was detected in the fractions E1 to E6 at ~31 kDa that rati ed the size of the recombinant HipA Xn2 toxin protein with 6XHis-tag as shown in Fig. 3a.
Further, the alone hipB Xn2 antitoxin gene (408 bp) encoding the 14.90 kDa protein was cloned under the strong T7 promoter in the pET28a vector with N-terminal 6XHis-tag. To study the expression level of the hipB Xn2 antitoxin gene, recombinant strain JSM7 was used and the expression pro le was observed as illustrated in Fig. 3b. A band labeled with hipB Xn2 antitoxin was not distinct in the total lysate (lane TL), therefore after puri cation with Ni-NTA a nity chromatography under native conditions, a single band was observed in the fractions E1 to E6 at ~15 kDa that corroborate with the size of the recombinant HipB Xn2 antitoxin having 6XHis-tag as shown in Fig. 3b.
The same approach was also used to clone the hipBA Xn2 operon comprised of hipA Xn2 toxin and hipB Xn2 antitoxin genes in cis-form and expression pro le with recombinant strain JSM5 of both genes was illustrated in Fig. 3c. As the protein bands of HipA Xn2 toxin and HipB Xn2 antitoxin were not differentiable in the total lysate (lane TL) and thus, after puri cation, under native conditions by using Ni-NTA a nity chromatography, distinguishable bands of HipA Xn2 toxin and HipB Xn2 antitoxin proteins were observed in the fractions from E1 to E6 at ~31 kDa and ~15 kDa which vari ed the size of the recombinant HipA Xn2 toxin and HipB Xn2 antitoxin proteins with 6XHis-tag as shown in Fig. 3c. These puri ed recombinant HipA Xn2 toxin and HipB Xn2 antitoxin proteins were further detected by Western blot analysis. A singleband was observed at a size of ~31 kDa and ~15 kDa which rati ed the size of the recombinant HipA Xn2 toxin protein and HipB Xn2 antitoxin proteins respectively, as shown in Fig. 3d.
Functional assessment of hipBA Xn2 TA system Recombinant strains JSM4 and JSM5 containing hipA Xn2 toxin and hipA Xn2 -hipB Xn2 TA complex genes respectively were used to analyze the functionality of hipBA Xn2 TA system. We observed the growth pro les of these strains after the overexpression of hipA Xn2 toxin and hipA Xn2 -hipB Xn2 TA genes in liquid and on solid medium. In liquid media, post-induction, at each hour OD 600 was measured and E. coli Top10 cells with an empty pBAD/His C vector was used as a control as depicted in Fig. 4a. After 3h of induction, it was observed that cells overexpressing HipA Xn2 toxin inhibit the growth by two-fold as compared to control cells and as induction time increased, a sudden increment in the growth of control cells was also noticed and thus, after 4h to 7h of induction, the growth of cells containing hipA Xn2 toxin gene was retarded by more than 2.5-fold as compared to control as illustrated in Fig 4a. While a different growth pro le pattern was observed in cells overexpressing hipBA Xn2 TA complex genes as shown in Fig. 4a, due to the effect of HipB Xn2 antitoxin on HipA Xn2 toxin. At an initial 2h of postinduction, the toxicity of HipA Xn2 toxin was not signi cantly neutralized by HipB Xn2 antitoxin, but a clear effect of HipB Xn2 antitoxin overexpression was observed from 3rd hour of induction as depicted in Fig.   4a. After 3h to 7h of post-induction, it was found that the cellular growth of hipBA Xn2 TA complex genes harboring cells was resumed by more than two-fold as compared to only the hipBA Xn2 toxin genecontaining cells. Additionally, these observations were supported by growth assay on solid media as depicted in Fig. 4b. Colony-forming units (CFU) of recombinant strains JSM4, JSM5, and control cells were calculated. At the beginning of 3h post-induction, CFU counts of these strains were almost similar although in later hour i.e. after 4h to 7h, due to overexpression of HipA Xn2 toxin in JSM4, the number of viable cells was declined as compared to JSM5 and control cells as depicted in Fig. 4b. All experiments were conducted three times to verify the observations. Determination of active site residues responsible for the toxicity of HipA Xn2 toxin To access the function of active site residues in HipA Xn2 toxin, we have performed site-directed mutagenesis (SDM) and followed a toxicity assay analysis. Active site residues were screened by bioinformatics analysis as described elsewhere (Mohit Yadav & Jitendra Singh Rathore 2020) and by sequence alignment with other bacterial pathogens as depicted in Fig. 5. The selected active site residues were Gly-8, Ser-129, His-164, Asn-167, Ser-169, Asp-185, and Thr-220, as shown in Fig. 6a and these were substituted with alanine to avoid any structural complexity. In toxicity assay analysis, we have included mutant strains G8A, S129A, H164A, N167A, S169A, D185A, T220A, a control (E. coli cells with empty pBAD/His C vector) and a wild type (WT) strain i.e.E. coli cells harboring pBAD/His C hipA Xn toxin (for details see Table S1). Initial 3h of post-induction, all mutants and WT were shown almost a similar growth pro le pattern however from the fourth hour onwards mutants G8A, H164A, N167A, and S169A were showing elevated growth level compare to WT strain and approaches to the control as shown in Fig.   6b. Mutants S129A, D185A, and T220A were not signi cantly different from the growth pattern of WT strain throughout the growth study. In 4 th hour, percent growth inhibition for strain G8A, H164A, N167A, S169A, and WT was 46.34%, 41.46%, 56.09%, 46.34%, and 61.5% respectively as compared to control and in 5 th hour, percent growth inhibition for these strains was reached to 53.81%, 50.26%, 57.37%, 53.81%, and 65% respectively as compared to control. While in 6 th hour, percent growth inhibition for strain G8A, H164A, N167A, S169A, and WT were estimated to 50%, 43.70%, 47%, 47%, and 63% respectively as compared to control and in 7 th hour, this was approaching to 41.17%, 31.88%, 38.08%, 31.88%, and 50%.
Further, we examined these results by evaluating viable cells on solid LB media as illustrated in Fig. 6c.
We determined the colony-forming units (CFU) of E. coli cells overexpressing HipA Xn2 toxin and its mutants. After induction, at each hour, samples were harvested and diluted to spot on the LB medium. With overnight incubation at 37ºC, CFUs were counted and plotted against time post-induction. There was no signi cant difference observed in CFU counts of mutants S129A, D185A, and T220A as compared to WT as shown in Fig. 6c. While, from an initial 2h of post-induction, mutants G8A, H164A, N167A, and S169A were exhibiting increased CFU counts as compared to WT as depicted in Fig. 6c. In the 6-7h of post-induction, the CFU counts for these mutants were approaching the control and it was increased with more than 1.5-fold as compared to WT. Thus, the absorbance and viable cell count results show that four active site residues namely, Gly-8, His-164, Asn-167, and Ser-169 are crucial for the toxicity of hipA Xn2 toxin.
Transcriptional regulation of hipBA Xn2 TA system under stress conditions To analyze the transcription regulation of the hipBA Xn2 TA module under stress conditions like elevated temperature, antibiotics, and nutrient starvation, we have cloned 529 bp promoter sequences in reporter plasmid pJSM6. Thus, the β-galactosidase activity of the hipBA Xn2 TA promoter, in E. coli strain JSM6, was measured in the form of Miller unites (MUs) (Miller 1972) as illustrated in Fig. 7. This MUs examination was further con rmed by Reverse Transcriptase Polymerase Chain Reaction (RT-PCR) analysis in X. nematophila as depicted in Fig. 7. In this analysis, loading control was 16s RNA gene from X. nematophila, while primer details and gene sizes are depicted in Fig. 7a.
Under elevated temperature conditions, after two hours of incubation at 42˚C, the promoter activity was increased by three-fold and at 37˚C it was higher as compared to 30˚C as shown in Fig. 7b. In the RT-PCR analysis, this up-regulation was validated by observed higher intensity bands of hipA Xn2 toxin, hipB Xn2 antitoxin, and hipBA Xn2 operon genes under these temperature conditions as shown in Fig. 7e. For antibiotics stress conditions, two antibiotics cipro oxacin and o oxacin were used. The concentration of antibiotics dose were 1 and 3 μg/mL as described in our previous studies (Yadav and Rathore 2018; Mohit Yadav & Jitendra Singh Rathore 2020). Higher promoter activity was observed after the 20 minutes of o oxacin administration while it was activated by almost two-fold in the case of cipro oxacin treatment as compared to control cells (without antibiotics) depicted in Fig. 7c. Under the same conditions, RT-PCR analysis of hipA Xn2 toxin, hipB Xn2 antitoxin, and hipBA Xn2 operon genes was also performed, and up-regulation was further con rmed by intense bands as shown in Fig. 7f.
Nutrient starvation conditions were generated as described elsewhere (Yadav and Rathore 2018a; Mohit Yadav & Jitendra Singh Rathore 2020). Under these conditions, it was estimated that promoter activity was decreased by two-fold as compared to activity in M9 medium of stressed condition as illustrated in Fig. 7d. This down-regulation was further validated with low-intensity RT-PCR bands of hipA Xn2 toxin, hipB Xn2 antitoxin, and hipBA Xn2 operon genes as shown in Fig. 7g.
Interaction analysis of recombinant hipBA Xn2 TA proteins with its promoter The regulation of the hipBA Xn2 TA system was further investigated by analyzing the interaction of recombinant HipA Xn2 toxin and HipB Xn2 antitoxin proteins with its promoter region. A gel shift assay was performed with 200 ng of puri ed 529 bp hipBA Xn2 promoter and a concentration gradient (0.25 to 1.5 μM) of puri ed recombinant HipA Xn2 toxin and HipB Xn2 antitoxin proteins as illustrated in Fig. 8. There was no interaction observed, when the hipBA Xn2 TA promoter was subjected to recombinant HipA Xn2 toxin protein as depicted in Fig. 8a. However, in the case of recombinant HipB Xn2 antitoxin protein, a clear shift of the hipBA Xn2 TA promoter band was observed in the range of 0.25 to 1.0 μM of antitoxin protein as shown in Fig. 8b; and when we added both recombinant HipA Xn2 toxin and HipB Xn2 antitoxin proteins of 0.25 to 1.5 μM concentration range to the promoter, the band intensity of shift was further increased as shown in Fig. 8c. For control, we used non-interacting Bovine Serum Albumin (BSA) protein in the same concentration range to interact with the hipBA Xn2 TA promoter and no shift was found as illustrated in Fig. 8d.

Discussion
Under stressful environmental conditions in particular oxidative stress, nutritional depletion, temperature variation, etc., TA systems have differentially induction phenomena (Huang et al.; Ramage et al. 2009;Singh et al. 2010). The genome of X. nematophila has a copious amount of TA systems (Yadav and Rathore 2018) that make the basis to explore their role in its physiology. The X. nematophila ATCC 19061 chromosome contains three hipBA TA homolog operons (Yadav and Rathore 2018) (hipBA Xn , hipBA Xn2 , and hipBA Xn3 ) and possibly each will have a different effect on bacterial growth, molecular structure, expression pro le, gene regulation pattern and cellular viability. Among these three operons, previously, hipBA Xn TA operon was identi ed as a bona de type II TA system and the hipBA Xn3 TA system was a pseudo-type TA system (Yadav and Rathore 2018; Mohit Yadav & Jitendra Singh Rathore 2020) and the role of hipBA Xn2 TA system was unknown.
Thus, the present study was conducted to investigate the activity of the second hipBA TA homolog operon named as hipBA Xn2 TA system from X. nematophila. Here, we identi ed the chromosomal location of the hipBA Xn2 TA system (Fig. 1) and by phylogenetic analysis, we con rmed that the hipA Xn2 gene encoded a serine/threonine-protein kinase (HipA Xn2 toxin) while hipB Xn2 gene encoded a transcriptional regulatory protein (HipB Xn2 antitoxin) as depicted in Fig. 2 Rathore 2020) (Fig. 3). As expected, heterologous overexpression of HipA Xn2 toxin in E. coli cells majorly retard the growth of these cells, and the toxicity of this toxin was neutralized when it was co-expressed with its cognate HipB Xn2 antitoxin partner (Fig. 4), these results are consistent with our previous ndings of rst validated hipBA Xn TA system (Yadav and Rathore 2020). To further elucidate the function of HipA Xn2 toxin, we constructed various mutant strains devoid of activity associated with this toxin. In HipA toxin from E. coli, residues Gly-22, Asp-88, Ser150, Asp-291, Asp-309, and Asp-332 were very crucial for its activity (Leberman et al. 1980;Korch et al. 2003;Correia et al. 2006;Kaspy et al. 2013;Schumacher et al. 2015) while residues Ser149, Asp-306, and Asp-329 had great importance in the activity of HipA Xn toxin (Yadav and Rathore 2020). Thus, it was much needed to explore the role of such residues in HipA Xn2 toxin and we found that four residues Gly-8, His-164, Asn-167, and Ser-169 are very essential for its toxicity as shown in Fig. 6.
TA systems are known as stress-adaptive entities and our observation that the hipBA Xn2 TA system is activated under different adverse conditions (Fig. 7) supports the perception that this TA locus contributes to X. nematophila physiology. For instance, the increases in hipBA Xn2 TA transcripts in X. nematophila upon exposure to elevated temperature, nutrient starvation, and antibiotics (Fig.7) indicate that this TA might play a pivotal role in bacterial adaptation to such conditions. The regulated activation of TA systems in similar stress associated conditions has also been observed (Gupta et al. 2017; and Rathore 2020). RT-PCR analysis exhibited the distinctive induction pro le of the hipBA Xn2 TA system, and under most of the stress environment tested, the levels of hipA Xn2 toxin transcripts were higher than the levels of their cognate hipB Xn2 antitoxin transcripts as illustrated in Fig. 7. Such divergence in the transcript extents may be due to the distinctive stability of the transcripts or the expression of hipA Xn2 toxin is driven from multiple promoters. Likewise, post-transcriptional regulation of TA modules has also been reported (Singh et al. 2010; Yadav and Rathore 2020).
Transcriptional control of the hipBA Xn2 TA system was further supported by the gel shift assay as shown in Fig. 8. The mobility of the hipBA Xn2 TA promoter was hindered by HipB Xn2 antitoxin and HipBA Xn2 toxin-antitoxin while alone HipA Xn2 toxin does not affect its mobility as illustrated in Fig. Therefore, HipB Xn antitoxin may be a repressor protein in the transcriptional regulation of the hipBA Xn2 TA system whereas the association of HipA Xn2 toxin protein in it works as a corepressor. Similarly, Transcriptional control studies have also been done before in different TA systems that also support our assumptions (Overgaard et al. 2008; Kędzierska and Hayes 2016; Yadav and Rathore 2020). According to these observations, a proposed transcription regulation model of all three hipBA TA homologs from X. nematophila is illustrated in Fig. 9.
As hipBA Xn3 is a pseudo-type TA system, thus, the distinct response of hipBA Xn and hipBA Xn2 TA systems most likely suggests the components of TA systems have different substrates or speci city for their activity. Other TA systems like parDE, mazEF, and hipBA in E. coli have been studied for different substrates or speci city (Huang et al.; Zhang et al. 2003;Monti et al. 2007). Therefore, by these results, it may be conceptualized that each TA system has separate and de ned interactome to produce a differential effect of TAs in bacterial physiology. Moreover, the multiple numbers of TA loci in X. nematophila may hence re ect the necessity for extra control over the general expression of typical protein subsets as compare to having less or no TAs such as E. coli, which has only 5 TA systems (Pandey and Gerdes).
Future experiments include identifying cellular targets for hipBA Xn2 TA and elucidating the roles of these TA systems in X. nematophila persistence. Conclusively, we have been revealed some crucial points regarding hipBA Xn TA system such as it is an organized operon with two genes, both genes are protein in nature in which one encodes a toxin and other encodes an antitoxin, toxin protein inhibits the bacterial growth while antitoxin protein neutralized it, the formation of a TA complex, and activation of this TA system in stress conditions. All these attributes a typical type II TA system (  Chromosomal organization of hipBAXn2 TA system in the bacterium X. nematophila ATCC 19061, here, hipBXn2 antitoxin gene (cyan color) is located upstream to hipAXn2 toxin gene (blue color). The promoter sequence is also identi ed with -10 region underline in red color and -35 region underline with cyan color while rpoS17 and ihf transcription factor binding sites are indicated in red and cyan color boxes respectively     Site-Directed Mutagenesis analysis in the active site of HipAXn2 toxin. A The active site residues for the mutation in HipAXn2 toxin, cartoon structure on the transparent surface of this protein is shown and mutated residues are labeled. b Growth pattern of hipAXn2 toxin and its mutants, here, black triangle is Control and wild type (WT) is red triangle while mutants Gly-8, Ser-129, His-164, Asn-167, Ser-169, Asp-185, and Thr-220 are depicted with an orange circle, violet square, brown triangle, green square, blue square, grey circle, and yellow square respectively, c Graph between the number of viable E. coli cells (Log10CFU/mL) overexpressing HipAXn2 toxin and its mutants against time post-induction, here, control strain is shown with the black triangle (Control), wild type (WT) is shown with a red triangle and while mutants G8A, S129A, H164A, N167A, S169A, D185A, and T220A are shown with an orange circle, violet square, brown triangle, green square, blue square, grey circle, and yellow square respectively Transcriptional regulatory analysis of the hipBAXn2 TA system under different stress conditions. a The diagrammatic representation of TA genes and primers used in the RT-PCR expression analysis; genes sizes are in bp and arrow numbers represent the number of primers as described in Table S2  A schematic model proposed for the transcriptional regulation of hipBAXn, hipBAXn2, and hipBAXn3 TA systems in X. nematophila. In normal growth conditions, these TA complexes work as a repressor and negatively regulate the transcription. In stress conditions, these antitoxins are selectively degraded by cellular proteases and the cognate toxins are free to inhibit translation by affecting transcriptional machinery. Thus, these TAs are involved in bacterial physiology by reprogramming cells to reduce cellular growth with up and down regulating necessary genes to facilitate cell survival in the different stresses.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.