A novel chaperone-effector-immunity system identified in uropathogenic Escherichia coli UMN026

Bacterial strains and growth conditions

A total of 99 previously isolated UPEC clinical isolates were used (Martinez-Santos et al., 2021). The reference strains and clinical isolates used for RT-PCR and antibacterial activity assays are described in Table 1. Bacteria were grown at 37 °C under static or shaking conditions in LB broth (5 g/L NaCl, 5 g/L yeast extract, 10 g/L casein peptone) (Dibico, Mexico) or filtered urine. Mid-stream urine was collected from healthy volunteers by the mid-stream-clean-catch technique (MSCC) (Maher, Brown & Gatewood, 2017), centrifuged at 4,000 rpm at 4 °C for 10 min, and filtered with 0.22 µm sterile Millex-GP syringe filters (Merck Millipore, Germany). When necessary, the medium was supplemented with kanamycin (Km, 40 μg/mL).

Oligonucleotides

The oligonucleotides used in this work were synthesized by the Oligonucleotide Synthesis Facility at Instituto de Biotecnologia (UNAM, Cuernavaca, Mexico) and are listed in Table 2.

DNA extraction

Total DNA was obtained from the clinical isolates by thermal shock as described previously (Hernandez-Vergara et al., 2016). Briefly, 5–6 colonies were resuspended in 100 μL of sterile water, frozen at −70 °C for 30 min, and boiled at 96 °C for 10 min. The tubes were centrifuged at 13,200 rpm for 10 min, the supernatant was recovered, and the DNA was quantified on a NanoDrop 2000 (Thermo Scientific, Waltham, MA, USA).

Amplification of ECUMN genes

The frequency of the ECUMN genes _0231, _0232, and _0233, as well as the vgrG gene, was determined by PCR using specific primers (Table 2). The reactions were performed in a final volume of 12.5 μL, containing: 1X buffer, 25 mM MgCl₂, 1 mM dNTPs, 10 μM of each oligonucleotide, 150 ng of DNA, and 1.25 U of Taq DNA polymerase (Thermo Scientific, Waltham, MA, USA). The following conditions were used: 1 cycle at 95 °C for 5 min; 30 cycles at 95 °C for 1 min, 55 °C for 1 min, 72 °C for 15 s; and 1 cycle at 72 °C for 10 min for vgrG, ECUMN_0231, and ECUMN_0233. For ECUMN_0232, the following conditions were used: 1 cycle at 95 °C for 5 min; 30 cycles at 95 °C for 1 min, 58 °C for 1 min, 72 °C for 15 s; and 1 cycle at 72 °C for 10 min. The PCR products were analyzed on 2% agarose gels stained with ethidium bromide.

RNA purification and retro-transcription

Three mL of LB broth were inoculated with UPEC strain UMN026 and incubated overnight at 37 °C. Then, 25 mL of LB broth or urine were inoculated with 250 μL of the preinoculum and incubated at 37 °C with or without shaking for 4 (LB) or 15 h (urine). Bacterial cells were collected by centrifugation at 4,000 rpm for 15 min at 4 °C, and the pellets were stored overnight at −80 °C. Total RNA was purified with TRIzol. RNA integrity was verified in 1% agarose gels with 6% bleach (Aranda, LaJoie & Jorcyk, 2012). Purified RNA was treated with DNAse I (Thermo Scientific, Waltham, MA, USA) at 37 °C for 30 min. The reaction was performed in a final volume of 15 μL, containing 10 μL of RNA, 1.5 μL of 10X buffer, and 1.5 μL of DNAse I (1 U/mL). The enzyme was inactivated by adding 1 μL of 50 mM EDTA and incubating at 65 °C for 10 min. cDNA was synthetized using 1 μg of total RNA treated with DNAse I and the Verso cDNA synthesis kit (Thermo Scientific, Waltham, MA, USA), according to the manufacturer’s instructions.

Expression of the ECUMN genes

The expression of the ECUMN genes was determined by PCR using specific primers (Table 2). The reactions were performed in a final volume of 12.5 μL, containing: 1X buffer, 25 mM MgCl₂, 1 mM dNTPs, 10 μM of each oligonucleotide, 150 ng/μL cDNA, and 1.25 U Taq DNA polymerase (Thermo Scientific, Waltham, MA, USA). The following conditions were used: 1 cycle at 95 °C for 5 min; 30 cycles at 95 °C for 1 min, 55 °C for 1 min, and 72 °C for 15 s; and 1 cycle at 72 °C for 10 min for vgrG, ECUMN_0231, and ECUMN_0233. For the expression of ECUMN_0232, the following conditions were used: 1 cycle at 95 °C for 5 min; 30 cycles at 95 °C for 1 min, 58 °C for 1 min, and 72 °C for 15 s; and 1 cycle at 72 °C for 10 min. The PCR products were analyzed in 2% agarose gels stained with ethidium bromide.

Antibacterial activity (bacterial competition assays)

Prey strains (4012, 4014, S. aureus, E. faecalis, and K. pneumoniae), as well as the attacking strains (UMN026 and 4018), were grown overnight at 37 °C in 3 mL of supplemented or unsupplemented LB broth or urine, respectively, with Km. One mL of each overnight culture was centrifuged at 13,000 rpm for 3 min, after which the pellets were washed with 1 mL of LB broth and then resuspended in LB broth, after which the OD₆₀₀ was adjusted to 1. Cultures of attacking and prey strains were mixed at a 10:1 ratio in a total volume of 200 μL, spotted on LB plates, and incubated at 37 °C for 24 h. Bacterial growth was recovered in 2 mL of LB broth, the OD₆₀₀ was normalized to 0.5, and 10 μL of serial dilutions (10⁻¹ to 10⁻⁶) were spotted on LB plates supplemented with Km and incubated at 37 °C for 24 h. The colonies were photographed and recovered in 1 mL of LB broth, after which the OD₆₀₀ was determined (Basler, Ho & Mekalanos, 2013; Fitzsimons et al., 2018; Ma et al., 2020). Antibacterial activity assays were performed in triplicate.

Statistical analysis

Statistical analysis was performed using the STATA software v. 13.1. The relationships between quantitative variables were determined using the chi² test, and p < 0.05 was considered to indicate statistical significance.

Bioinformatic analysis

Identification of the possible functions of the three proteins ECUMN_0231 (GenBank: CAR11446.1), ECUMN_0232 (GenBank: CAR11447.1), and ECUMN_0233 (GenBank: CAR11448.1) was performed by available bioinformatic programs, using the online tools NCBI BLAST (https://blast.ncbi.nlm.nih.gov/) (Altschul et al., 1990), NCBI Conserved Domain Search (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) (Lu et al., 2020; Marchler-Bauer et al., 2017; Wang et al., 2023), KEGG (https://www.genome.jp/kegg/) (Kanehisa & Goto, 2000), UNIPROT (https://www.uniprot.org) (UniProt, 2023), InterPro (https://www.ebi.ac.uk/interpro/) (Paysan-Lafosse et al., 2023), SignalP-6.0 (https://services.healthtech.dtu.dk/services/SignalP-6.0/) (Teufel et al., 2022), and Bastion6 (https://bastion6.erc.monash.edu/) (Wang et al., 2018). Protein sequences were used to identify functional domains and features. With the information collected, representative images of the conserved regions of each protein were constructed.

Prediction of protein-protein interactions (docking)

Modeling of protein interactions was performed using the software UCSF ChimeraX, which has an integrated protein structure prediction tool based on the AlphaFold2-multimer code that uses ColabFold.V1.5.5 (free access version) (Meng et al., 2023). The amino acid sequences of the ECUMN_0231, ECUMN_0232, and ECUMN_0233 proteins (accession numbers B7N896_ECOLU, B7N897_ECOLU and B7N898_ECOLU, respectively) were downloaded from UniProt. Possible protein-protein interactions (docking) were probed between proteins ECUMN_0231_vs_ECUMN_0232 and ECUMN_0232_vs_ECUMN_0233 entering the downloaded sequences separated by commas. ColabFold.V1.5.5 finished the run and automatically opened the best model obtained after several cycles. The “contacts” tool was used on UCSF ChimeraX to calculate the contacts with default options between proteins, showing the distances calculated by the software. Finally, the names and numbers of the residues involved in interactions were added.

The ECUMN and vgrG genes are conserved in isolates from symptomatic and asymptomatic UTIs

As stated in the introduction, the ECUMN genes _0231, _0232, and _0233 were proposed as biomarkers of UTIs since they are conserved in strains isolated from patients with symptomatic UTIs (Nielsen et al., 2017). To corroborate whether these genes are conserved only in isolates from symptomatic patients, we determined the presence of these genes in clinical isolates obtained from patients with cystitis, pyelonephritis, and AB (Table 3). We found that of the 99 clinical isolates analyzed, 33 (33.3%) had at least one ECUMN gene, independent of the symptomatology of the patients, indicating that these genes are conserved not only in isolates from symptomatic UTI patients, but also in isolates from AB patients. Since the ECUMN genes are encoded in the T6SS cluster, we also determined the frequency of the vgrG gene. This gene was more frequent (48.5%) than the ECUMN genes, and it was also present in isolates from patients with and without symptoms.

We also analyzed the individual conservation of these genes. As shown in Fig. 1, most of the isolates have only ECUMN_0231 (36.36%, 12/33) or ECUMN_0233 (30.3%, 10/33). Two isolates (6.06%) had only ECUMN_0232, or the combinations ECUMN_0232/_0233 and ECUMN_0231/_0233. Only five isolates (15.15%) had these three genes. Interestingly, no isolate contained the combination ECUMN_0231/_0232. Notably, of the five isolates with the three genes, three were from patients with pyelonephritis, and two were from patients with AB.

The ECUMN and vgrG genes are present in pathogenic and nonpathogenic phylogroups

It was previously reported that the ECUMN genes are conserved only in UPEC strains belonging to phylogroups B2 and D (Nielsen et al., 2017). To test this hypothesis, we also analyzed the frequency of the ECUMN genes according to phylogenetic group (Table 4). Our results showed that the ECUMN genes were more frequent in isolates belonging to phylogroup B2, followed by isolates of phylogroup D (both of which are considered pathogenic), although they were also found in isolates from commensal phylogroups, mainly phylogroup F. On the other hand, similar results were obtained for the vgrG gene, although in this case, the presence of the gene is not related to the phylogroup.

The ECUMN and vgrG genes are expressed in LB and urine

Since the ECUMN genes are annotated as putative genes and there seems to be no relationship with symptoms, we analyzed their expression under laboratory (LB broth) and UTI-like conditions (urine) in the prototype strain UMN026. As shown in Fig. 2, all the genes were expressed in both LB broth and urine under shaking or static conditions. The same expression pattern was observed for the vgrG gene, indicating that the ECUMN genes are expressed under the same conditions as the T6SS.

To determine whether the clinical isolates also expressed these genes, we analyzed their expression in the clinical isolates that possessed all three genes. As shown in Figs. 3A and 3B, ECUMN_0231 was expressed in all the clinical isolates in LB, while in urine the clinical isolates 4087, 4012, 4039, and 4079, but not isolate 4073, expressed this gene (Figs. 3C and 3D).

On the other hand, ECUMN_0232 was expressed in all the isolates in LB and urine (Fig. 4). A similar result was obtained with ECUMN_0233 (Fig. 5).

Finally, the analysis of vgrG expression showed that this gene is expressed in all isolates when grown in LB, while it is only expressed in UPEC UMN026 and the isolate 4087 when grown in urine (Fig. 6).

UPEC UMN026 has inter- and intraspecific antibacterial activity

Our results showed that there was no relationship between the presence of the ECUMN genes and the symptomatology. However, T6SSs have also been shown to be involved in bacterial competence. To determine whether UPEC UMN026 is able to kill other bacteria, we performed bacterial competition assays against two UPEC clinical isolates (4012 and 4014), as well as against S. aureus, K. pneumoniae, and E. faecalis (Table 1). Our results showed that UPEC UMN026 did not have bacteriolytic activity against isolate 4012 (data not shown), but it has antibacterial activity against the clinical isolate 4014 when cultured both in LB and urine (Fig. 7A, Fig. S1A). When the assays were performed against other species, our results showed that UPEC UMN026 had no activity against K. pneumoniae or S. aureus (data not shown), but it had activity against E. faecalis (Fig. 7B, Fig. S1B). In this case, the effect was clearer in urine than in LB. These results indicate that UPEC UMN026 has intra- and interspecific antibacterial activity.

ECUMN genes form a chaperone-effector-immunity system

To understand the function of the ECUMN genes, we performed a bioinformatic analysis. Blastp, UniProt, and KEGG analyses revealed that ECUMN_0231 has a DUF4123 domain (pfam13503) (residues 9–122) (Fig. 8B), which has approximately 120 residues and four conserved motifs. Genes encoding proteins containing the DUF4123 motif are commonly found downstream of vgrG or putative effector-encoding genes (Moriel et al., 2021; Unterweger et al., 2015). This domain has been shown to be present in T6SS effector chaperones (TECs) in several bacterial species. These chaperones have been shown to be necessary only to load the cognate effector onto the T6SS to be secreted (Liang et al., 2015; Unterweger et al., 2015) or to stabilize its cognate effector, maintain its levels and aid in its secretion (Bondage et al., 2016). The NCBI Conserved Domains tool classifies ECUMN_0231 as a DUF4123 domain-containing protein similar to the Vibrio cholerae accessory protein VasW that plays an accessory role in VasX-mediated bacterial killing. VasW is encoded immediately upstream of vasX, which encodes an effector protein, and is proposed to mediate the interaction between VasX and the T6SS tip protein PAAR, so VasX can be secreted (Miyata et al., 2013). Another DUF4123 domain-containing protein is Proteus mirabilis IdsC, which stabilizes the effector protein IdsD in subcellular clusters, maintaining its levels and aiding in its secretion through the T6SS (Zepeda-Rivera, Saak & Gibbs, 2018). Thus, this result suggested that ECUMN_0231 encodes a putative TEC (Fig. 8A). On the other hand, our in silico analysis using BLAST and Conserved Domains Search showed that ECUMN_0232 has no conserved domains; however the Bastion6 tool predicts that ECUMN_0232 is a T6S effector. Accordingly, its localization (Fig. 8A) suggests that it could be an effector protein since genes that encode T6S effectors are often located adjacent to or near vgrG (Koskiniemi et al., 2013; Ma et al., 2014; Russell et al., 2013), and downstream of genes that encode DUF4123 domain-containing proteins (Liang et al., 2015). This finding also suggested that ECUMN_0231 could be a chaperone of ECUMN_0232 since the DUF4123 domain containing proteins seem to be specific for the effector encoded immediately downstream (Bondage et al., 2016; Liang et al., 2015). Finally, UniProt and InterPro analyses revealed that ECUMN_0233 has two features, the first being a signal peptide on the first 18 residues of its N-terminal domain (MKAKHLISVILLSGVVMG; n-region residues 1–5, h-region residues 6–13, and c-region residues 14–18) (Fig. 8B). The second feature is a chain or a noncytoplasmic domain, which is described as the extent of a polypeptide chain in the mature protein following processing or proteolytic cleavage. Further analysis using SignalP-6.0 showed that ECUMN_0233 has a lipoprotein signal peptide (Sec/SPII) with a likelihood of 0.7699, and a cleavage site between positions 21 and 22 with a probability of 0.662976. Lipoprotein signal peptides are transported by the Sec translocon and cleaved by signal peptidase II (Lsp) (Teufel et al., 2022). This result suggested that this protein could be translocated to the periplasm through the Sec system. This finding suggested that ECUMN_0233 could be an immunity protein located in the periplasm or in the outer membrane of bacteria. Taken together these findings indicate that ECUMN_0231, ECUMN_0232, and ECUMN_0233 encode proteins that form a chaperone/effector/immunity system.

To support this hypothesis, we performed a prediction of protein-protein interactions. The docking between ECUMN_0231 and ECUMN_ 0232 predicted 64 true contacts. The predicted true contact distances ranged between 2.679 Å, the nearest interaction, and 4.132 Å, the farthest. The contact points in ECUMN_0231 were distributed along the protein, whereas the contact residues in ECUMN_0232 were located only at the N-terminal end (Fig. 9A). The predicted ECUMN_0231 residues involved in the interaction are L101, D227, D226, A225, P262, and Y261, while the residues in ECUMN_0232 are C12, S6, T7, A8, S39, and K40 (Fig. 9B). On the other hand, the docking between ECUMN_0232 and ECUMN_0233 showed 444 true contacts. The distances between the contacts ranged from 2.607 Å to 4.160Å. In contrast to the interaction with ECUMN_0231, the residues in ECUMN_0232 that interact with ECUMN_0233 were located on the C-terminal end, whereas for ECUMN_0233, the residues involved were distributed along the protein (Fig. 10).