The pcnB gene impairs Shigella growth at 37°C, but promotes intestinal epithelial cell colonization

Virulence gene expression implies energetic and metabolic expenditures, including use of extra resources and the translation of chaperones and other regulators [1]. Therefore, loss of virulence gene expression has a beneficial effect on bacterial growth (hereafter also denoted “fitness”), as exemplified by the enhanced replication speed of Shigella clones showing impaired Congo Red binding (virulence off) [39]. However, mapping the global cost of virulence gene expression is challenging and it has only been achieved in a few cases [40,41].

We recently mapped the genome-wide cost of virulence gene expression in Shigella [33]. Among chromosomally located genes, we found that the ΔpcnB mutant grows significantly better at 37°C (virulence on) than at 30°C (virulence off), similarly to the ΔvirF and ΔvirB virulence mutants (Fig 1A). In line with this, a clean reconstructed ΔpcnB mutant grown at 37°C showed visibly larger colonies and a reduced Congo Red binding ability compared to the wt strain, indicating faster growth and reduced virulence (S1A Fig). To validate these findings we employed our recently developed barcoded competition assay, encompassing defined consortia of genetically barcoded wt and mutant Shigella strains [33,42,43]. Each strain harbours a unique nucleotide barcode that can be quantified by qPCR in genomic DNA extracted from the “input consortium” and the “output consortium” following growth under specific condition(s), or following an infection assay. We here employed four replicates of a mixed 1:1:1:1:1:1 consortium, containing six strains: two wt (tagA, tagB), two ΔmxiD (tagC, tagD) and two ΔpcnB (tagE, tagF) strains. The ΔmxiD mutants, lacking one of the structural components of the T3SS and therefore non-invasive, represent internal controls. All strains were equally represented in the input consortium (Fig 1B). We then grew these consortia overnight (ON), either at 30° or 37°C. The ΔpcnB mutant showed a minor growth defect at 30°C compared to the wt and the ΔmxiD strains, but a notable and consistent ≥2-fold increase in the relative strain abundance at 37°C (Fig 1C).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. pcnB contribution to Shigella fitness and colonization capacity of human intestinal epithelium.

(A) Replot from Di Martino et al, 2025 [33] - MA plot of relative mutant fitness expressed as Log₂FC between Growth at 37°C vs Growth at 30°C. Each dot represents a gene. Mutants with a Log₂FC with significant FDR (FDR ≤ 0.01) are shown in dark grey; non-significant mutants in light grey. pcnB mutant is highlighted in red among the detrimental genes. Log₂FC ≥ 1; FDR ≤ 0.01. (B-C) Barcoded competition assay in which a mixed barcoded consortium comprising two Shigella wt (tagA, tagB), two ΔmxiD (tagC, tagD) and two ΔpcnB (tagE, tagF) strains was grown ON at 30°C or 37°C. (B) The consortium was mixed from individual ON cultures grown at 30 °C and strain tag abundance in extracted gDNA was analysed by qPCR. The relative abundance of each strain is plotted as percentage of the total input population. (C) Quantification of relative tag abundances in the Shigella barcoded consortium cultures grown at 30° or 37°C, as analyzed by qPCR. Relative abundances were normalized to the input inoculum. Shown are data for four independently generated consortia. Statistical significance determined by paired t-test between the normalized output and the normalized input abundances (see Methods). ***p < 0.001. (D) Replot from Di Martino et al, 2025 [33] - MA plot showing Shigella gene mutant relative abundance upon infection of enteroids, as quantified by TraDIS. Shown is the Log₂FC in the output library on the y-axis and the Log₂CPM (average log₂ counts/million) on the x-axis. Each dot represents a gene. Mutants with a Log₂FC with significant FDR (FDR ≤ 0.2) are shown in dark grey; non-significant mutants in light grey. pcnB mutant is highlighted in red. (E-F) Enteroids were infected with a mixed barcoded consortium comprising two wt (tagA, tagB), two ΔmxiD (tagC, tagD) and two ΔpcnB mutant (tagE, tagF) strains at MOI 40 for 6h. (E) The graphs depict the composition of barcoded Shigella consortia used as input inocula. The relative abundance of each tag in the input consortia is plotted as percentage of the total population, as analysed by qPCR. (F) Quantification of relative tag abundance in the intracellular population, normalized against the corresponding input, as analysed by qPCR. Shown are data for four independently generated consortia for each infection. Statistical significance determined by paired t-test for each specific mutant between the normalized output and the normalized input abundances (see Methods). ***p < 0.001. (G) Bar graph showing the Colonization Index for ΔpcnB enteroid infections (derived from data in Fig 1E–1F and calculated as 1-(wt/mut)). wt is set as 0. (H) Intracellular Shigella population sizes in enteroids infected with Shigella wt, ΔmxiD or ΔpcnB strains for 3h at MOI 40. Results are shown as mean ± SD; CFU counts come from 6 biological replicates pooled from 2 independent experiments and were normalized to wt. Statistical significance determined by Mann-Whitney U-test; **p < 0.01.

https://doi.org/10.1371/journal.ppat.1013727.g001

In recent work, we also established a genome wide-map of Shigella genes required for the colonization of human intestinal epithelium [33]. Interestingly, in that dataset we observed a trend for the ΔpcnB mutant to be less capable of colonizing human epithelial enteroids (Fig 1D). Again, we employed barcoded assays to quantify this effect, mixing two wt (tagA, tagB), two ΔmxiD (tagC, tagD) and two ΔpcnB (tagE, tagF) strains and infecting enteroids at MOI 40. At 6h post infection we recovered the Shigella intracellular population and enriched it alongside with a dilution of the input inoculum for 14 h ON at 30°C. As expected, we again observed a slight growth defect for the ΔpcnB mutant after enrichment at 30°C in the input inoculum (Fig 1E). However, also after normalizing for this subtle effect, the ΔpcnB mutant was still ~3-fold less abundant in the intracellular population with respect to the input (Fig 1F-G), confirming a lower capacity to colonize human enteroids. To further validate these results in the absence of secondary growth enrichment steps, we infected enteroids with the wt, the ΔpcnB and the ΔmxiD (non-invasive control) strains at MOI 40. At 3h p.i. enteroids were lysed and bacteria in the intracellular populations were quantified by plating on selective plates. This yet again revealed ~2–3 fold less CFUs for the ΔpcnB mutant compared to the wt strain (Fig 1H). From these data we conclude that the pcnB gene reduces Shigella fitness at 37°C, while at the same time sustains the bacterium´s ability to colonize intestinal epithelial cells. This suggests a pcnB-potentiating effect on Shigella virulence gene expression.

The pcnB gene enhances T3SS expression

To shed light on the pcnB effect on Shigella virulence, we analysed expression of select virulence genes in the wt strain as well as in the ΔpcnB and ΔvirF mutants by RT-qPCR (Fig 2A). As expected, all virulence genes analysed showed minimal expression levels in the ΔvirF mutant, as well as in the wt strain grown at 30°C. In comparison, virF, virB and icsA expression was modestly lowered in the ΔpcnB mutant with respect to the wt strain grown at 37°C. However, downstream genes belonging to the T3SS-encoding mxi-spa operons, i.e. mxiD and mxiE, were ~2-fold less expressed in the ΔpcnB mutant than in the wt strain. This observation is compatible with the amplification cascade of Shigella virulence regulation, where even small changes in expression of the upstream transcriptional regulators result in significant reduction of downstream virulence gene expression [32].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Global effect of pcnB on pINV-encoded proteins.

(A) Expression levels of virF, virB, icsA, mxiD and mxiE analysed in Shigella wt strain grown at 30°C or 37°C at OD600 0.7, and in the ΔpcnB and ΔvirF mutants grown at 37°C at OD600 0.7. Data were generated from four biological replicates and are presented as mean ± SD. A minimum of 2 technical replicates were run in each experiment. Statistical significance determined by One-way Anova; ns - non-significant; **p < 0.01, ***p < 0.001. (B) Volcano plot showing differentially expressed proteins in the Shigella ΔpcnB mutant vs the wt strain. Each dot represents a protein. Differentially expressed proteins located on the pINV are shown in pink, differentially expressed proteins located on the chromosome in black, and non-significant differentially expressed proteins in grey. Log₂FC ≥ 0.5; adj_pvalue ≤0.01. (C) Volcano plot showing differentially expressed proteins in the Shigella ΔvirF mutant vs the wt strain. Each dot represents a protein. Differentially expressed proteins located on the pINV are shown in pink, differentially expressed proteins located on the chromosome in black, and non-significant differentially expressed proteins in grey. Log₂FC ≥ 0.5; adj_pvalue ≤0.01. (D) Bar graph showing Log₂FC of all differentially expressed proteins located on the pINV for the Shigella ΔpcnB vs wt and ΔvirF vs wt comparisons.

https://doi.org/10.1371/journal.ppat.1013727.g002

To widen our understanding of the impact of pcnB loss on the virulence cascade, we next performed proteomic profiling. For the 2,152 identified proteins in the ΔpcnB mutant, 85 proteins were significantly downregulated, and 17 significantly upregulated in comparison to wt (Fig 2B; S1 Table; |log₂FC| ≥ 0.5; adj_p_value≤0.01). Among the downregulated proteins, 57 were proteins encoded on pINV, including the VirF and VirB transcriptional regulators, the adaptor protein for host-cell actin nucleation IcsA and nearly all T3SS structural components and secreted effectors encoded by the ipa-mxi-spa operons. This extends a recent observation showing that a Shigella ΔipaH2.5ΔpcnB mutant secretes lower amount of the IpaB, IpaC and IpaD proteins [44]. In the ΔvirF mutant, 74 proteins were significantly downregulated, and 10 significantly upregulated in comparison to wt (Fig 2C; S1 Table; |log₂FC| ≥ 0.5; adj_p_value≤0.01). Among the downregulated proteins, 48 were virulence proteins encoded on pINV, in agreement with previous reports. Overall, pcnB confers a similar pattern of attenuation of the virulence-relevant proteome as the master regulator VirF, although at a lower response magnitude (Fig 2B-D). Among the differences, it is further interesting to note that in the ΔpcnB mutant proteins with plasmid housekeeping functions were also less abundant, including the partitioning system StbA/B and the plasmid replication repressor CopB (also known as RepA2) (Fig 2D; S1 Table). This may hint to a decreased plasmid copy number, since copB is constitutively expressed in similar systems (e.g., R1 plasmid; [26]). These data suggest that the lower epithelial colonization capacity of the ΔpcnB mutant is driven by global attenuation of virulence protein expression. It appeared highly plausible that this is steered by reduced pINV replication in the absence of pcnB.

The pcnB gene governs pINV abundance at both 30° and 37°C

In E. coli, lack of pcnB results in lower copy number of R1 and ColE1 plasmids, both characterized by sRNA-dependent replication [36,37]. To assess if pcnB also affects Shigella pINV abundance, we determined the pINV relative copies per chromosome by droplet digital PCR (ddPCR) [45]. First, we optimized the ddPCR assay for Shigella pINV detection, validating linearity for two primer pairs located on the pINV (in the virF or the ipaA gene, respectively) and one primer pair located close to the ter region on the chromosome (in the manA gene) to avoid replication biases (S2A-E Fig). Second, we extracted gDNA from wt and ΔpcnB Shigella cultures grown either at 30°C (virulence off) or at 37°C (virulence on) in exponential phase, employing the crude bead beating extraction method. In this setting, pINV relative copies were not affected by temperature in the wt strain, with essentially identical results when using virF or ipaA primers to detect pINV (Figs 3A; S2F). No difference in pINV relative copies was found also in the hypersecreting mutant ΔipaD [46] grown at different temperatures compared to the wt strain (S2G-H Fig). pINV copy number, moreover, showed some degree of variation in absolute values depending on the DNA extraction method (S2I-K Fig). Most importantly, in all cases we observed a consistent ~2.5-fold reduction in pINV abundance in the ΔpcnB mutant compared to the wt strain, which was evident at both 30°C and 37°C (Fig 3A-B). In further agreement with this, we observed no changes in pcnB gene expression between the Shigella wt strain grown at 30°C or 37°C (Fig 3C – left panel). No changes in pcnB expression were also elicited by VirF expression (Fig 3C – right panel). Finally, complementation of pcnB in the ΔpcnB mutant strain restored pcnB expression and also completely restored pINV copy number to wt levels (Fig 3D-E). Altogether, these data suggest that the pcnB gene sustains pINV replication in Shigella under both virulence non-inducing and inducing conditions.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. pcnB contribution in sustaining Shigella pINV abundance.

(A) pINV relative copies per chromosome was determined by ddPCR using virF primers for pINV and manA primers for chromosome detection. gDNA was extracted by bead beating from Shigella wt or ΔpcnB mutant grown at 30°C or 37°C at OD600 0.7. Data comes from 8 biological replicates from two independent experiments. Statistical significance determined by Two-way Anova; ns - non-significant; ***p < 0.001. (B) pINV ratio relative to wt using gDNA from the ΔpcnB mutant extracted by bead beating, Qiagen DNeasy Blood & Tissue Kit or Thermo Scientific GeneJET Genomic DNA Purification Kit. pINV ratio was calculated from data from Fig 3A and S2J-K Fig. wt is set as 1. (C) Bar graphs showing RNA-seq Log₂FC of pcnB and select differentially expressed genes for the Shigella wt - 37° vs 30°C (right panel) and wt vs ΔvirF – 37°C (left panel) comparisons. RNA-seq data were replotted from Di Martino et al, 2025 [33]. Significance determined by two-sided F-test with Benjamini-Hochberg correction for multiple hypothesis testing. Log₂FC ≥ 1; FDR ≤ 0.01. (D) pcnB mRNA expression levels (2^−ΔΔCt) in Shigella wt or ΔpcnB strains carrying the p-Empty or p-pcnB plasmid, grown at 37°C at OD600 0.7. Data come from 7 biological replicates and were normalized to the pcnB expression in the Shigella wt strain. Statistical significance determined by Kruskal-Wallis test. ns - non-significant; *p < 0.05; ***p < 0.001. (E) pINV relative copies per chromosome was determined by ddPCR using virF primers for pINV and manA primers for chromosome detection. gDNA was extracted by bead beating from Shigella wt or ΔpcnB strains carrying the p-Empty or p-pcnB plasmid, grown at 37°C at OD600 0.7. Data comes from 4 biological replicates. Statistical significance determined by One-way Anova; ns - non-significant; ***p < 0.001.

https://doi.org/10.1371/journal.ppat.1013727.g003

pcnB promotes decay of the pINV antisense RNA CopA

The pcnB gene encodes for the polyadenylase poly-A polymerase I (PAP-I) [34]. Mutations in pcnB were previously associated with the accumulation of RNase E-dependent cleavage products of antisense RNAs RNA-I and CopA, responsible for plasmid replication control of ColE1 and R1 plasmids, respectively [36,37]. These RNase E-dependent cleavage products were rapidly degraded in a wt strain (pcnB⁺), following polyadenylation at the 3’-end by PcnB/PAP-I activity [36]. Shigella CopA sRNA has 93% sequence identity with R1 CopA, suggesting a conserved RNA secondary structure (Figs 4A, S3A). Importantly, sequences important for RNase E cleavage (linker) and initial interaction with the target mRNA CopT (SL-II loop bulge) are fully conserved [47] (S3A Fig). To formally test the link between pcnB and pINV CopA processing and stability, we performed northern blot analysis, using a CopA probe located at the sRNA 3’end (Figs 4A, S3A). No CopA was detected in RNA extracted from an E. coli DH10b strain lacking the pINV. When analysing RNA extracted from the Shigella wt strain both at 30°C and 37°C, two major CopA species were observed: full-length CopA (CopA-FL) of ~90 nt, and CopA-SL-E, a fragment of ~65 nt previously shown to be generated by RNase E cleavage between the CopA stem loop-I and stem loop-II (Fig 4A-B). Therefore, CopA-SL-E is a processed form of CopA, which only harbours stem loop-II. Both CopA-FL and CopA-SL-E were previously found capable of inhibiting R1 plasmid replication [47,48]. However, CopA-SL-E half-life increased dramatically in the absence of pcnB, resulting in lower R1 copy number [36]. In our present experiments, pcnB deletion in Shigella caused a marked increase in the CopA-SL-E abundance relative to the wt, evident at both 30 and 37°C (~2.5-fold; Fig 4B-C). This indicates that the pcnB encoded enzyme PAP-I favours the degradation of the Shigella pINV CopA-SL-E, similarly to R1 plasmid CopA-SL-E [36] (Fig 4B-C). Smaller CopA fragments (indicated as α and β) were also observed in the ΔpcnB mutant (Fig 4B). Likely, these forms are extremely unstable in the wt strain and therefore barely detectable. Complementation of pcnB was not able to restore CopA-FL relative abundance (Fig 4B-D). However, complementation fully restored the CopA-SL-E form to wt levels (Fig 4B-D), suggesting that the half-life of the CopA-SL-E transcript was restored. Notably, the total amount of CopA transcripts (FL + SL-E), both functional in replication control, matched wt levels upon complementation (Fig 4B-D), likely explaining the pINV copy number restoration (Fig 2E). Altogether, these data suggest that the pcnB-encoded poly-A polymerase activity targets CopA, ensuring prompt processing and degradation of the sRNA, which in turn results in suitable pINV replication levels. This warrants a beneficial trade-off between fitness and virulence, resulting in well-balanced virulence gene expression.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Effect of pcnB on Shigella pINV CopA stability.

(A) Schematic of CopA RNA secondary structure, as resolved in [86]. Partially made with BioRender. (B) Representative northern blot of CopA species in Shigella wt or ΔpcnB mutant grown at 30°C or 37°C at OD600 0.7 and in Shigella wt or ΔpcnB strains carrying the p-Empty or p-pcnB plasmid, grown at 37°C at OD600 0.7. RNA from DH10b: negative control. Used as replicate 1 for quantification. (C) Relative abundance of CopA-FL + CopA-SL-E in Shigella wt or ΔpcnB mutant grown at 30°C or 37°C at OD600 0.7. CopA species abundance was determined by quantification of northern blots. Shown is the mean and SD of 3 independent experiments (See S3B Fig for replicate 2 and 3). Statistical significance determined by Two-way Anova; **p < 0.01, ***p < 0.001. Values in light grey refer to CopA-FL, in dark grey to CopA-SL-E. (D) Relative abundance of CopA-FL + CopA-SL-E in the Shigella wt or ΔpcnB strains carrying the p-Empty or p-pcnB plasmid was determined by quantification of northern blots. Shown is the mean and standard deviation of 3 independent experiments (See S3B Fig for replicate 2 and 3). Statistical significance for total CopA-FL + CopA-SL-E levels determined by Two-way Anova; ns - non-significant; *p < 0.05.

https://doi.org/10.1371/journal.ppat.1013727.g004

pcnB does not affect pINV stability, but imposes virulence counter selection pressure

Our results show a clear role for the pcnB gene in sustaining pINV replication (Figs 3, 4). However, the reduced pINV copy number observed in the ΔpcnB mutant could potentially also include higher rates of plasmid loss [44]. It has been previously shown that culturing Shigella wt bacteria for 50 generations at 37°C results in ~60% avirulent CR- colonies. However, nearly all of them still retained pINV [49]. Instead, the CR- phenotype depended on the high plasticity of the pINV genetic architecture and was linked to the specific excision of the upstream virulence regulators virF, virB or the entire PAI loci [49]. Plasmid-free bacteria only arose in the presence of a defective MvpAT Toxin-Antitoxin (TA) system, the main TA system responsible for pINV stability at 37°C [7,49].

To assess the impact of pcnB on pINV stability, we introduced the pcnB deletion into an ipaH9.8-tagged strain. In this strain, the ipaH9.8 gene, encoding for a T3SS effector located on pINV, has been disrupted by a kanamycin resistance cassette. The Shigella ipaH9.8 mutant retains a CR⁺ phenotype (S4A Fig) and, as previously shown, is not significantly impaired for invasion or early intraepithelial expansion in the enteroid infection model [31]. It can therefore be used here as a reporter strain, behaving similarly to a wt Shigella strain during infection while providing a selectable marker on pINV. We refer to this control strain as pINV::Km. Deletion of pcnB in the pINV::Km strain background resulted again in a CR- phenotype (S4A Fig, compare with S1A Fig). We cultured both the control pINV::Km strain and the pINV::Km ∆pcnB mutant overnight at 30°C, again noting a slight growth defect for the ∆pcnB mutant compared to the control strain (S4B Fig, compare with Fig 1B). We then passaged the cultures daily at 37°C, a condition under which virulence gene expression is active and selection pressure for loss of virulence therefore present (Fig 5A). During early passages, the pINV::Km ∆pcnB mutant exhibited a growth advantage (S4B Fig; compare with Fig 1B). Interestingly, the pINV::Km ∆pcnB growth advantage faded over time (S4B Fig), compatible with a loss of virulence gene expression and consequently higher fitness in the control strain. We plated the inoculum as well as the passaged cultures after ~25 (p2) and ~50 (p4) generations on LB agar plates. Subsequently, 100 individual colonies per replicate were patched onto both LB and kanamycin-containing agar plates. 100% kanamycin resistant colonies were detected after 50 generations at 37°C for both the control and ∆pcnB stains, indicating pINV retention in either strain (Fig 5B). However, we observed the gradual emergence of bigger colonies in the pINV::Km control strain (Fig 5C and 5D; ~50% of the population after 50 generations), which likely reflected increased fitness due to loss of virulence gene expression (S4B Fig). This observation correlated with the appearance of ~50% CR- colonies after 50 generations at 37°C (Fig 5E), in full agreement with previous studies [49].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. The contribution of pcnB to pINV stability and virulence counterselection.

(A) Schematic representation of the pINV stability assay. Partially made with Biorender. (B) Stability assay of pINV plasmid in Shigella pINV::Km and pINV::Km ΔpcnB strains in the inoculum and after growth for ~25 and ~50 generations at 37˚C. The percentage of cells retaining the plasmid calculated from four biological replicates. (C) Representative images of Shigella pINV::Km and pINV::Km ΔpcnB colonies grown at 37°C for ~50 generations. Scale bar: 10mm.(D) Percentage of big colonies in Shigella pINV::Km and pINV::Km ΔpcnB strains in the inoculum and after growth for ~25 and 50 generations at 37˚C. Data come from four biological replicates. (E) Percentage of CR- colonies in Shigella pINV::Km and pINV::Km ΔpcnB strains after growth for ~50 generations at 37˚C. Data come from four biological replicates. (F) Left panel. Percentage of colonies retaining pINV as scored by PCR. Right panel. Percentage of colonies lacking specified virulence loci genes. Others refers to colonies that contain virB, virF and the pINV. Twenty small colonies and 20 big colonies obtained from pINV::Km in four biological repeats and forty colonies obtained from pINV::Km ΔpcnB were analysed by parallel PCRs.

https://doi.org/10.1371/journal.ppat.1013727.g005

Next, we performed PCR analysis on a subset of the patched colonies after 50 generations at 37°C for the presence of pINV and the virF and virB loci. 100% of the tested colonies, derived from both control and ∆pcnB strains, retained pINV (Fig 5F, left panel). Notably, 100% of the big colonies from the control strain had lost either virF, virB or other loci resulting in a CR- phenotype, while small colonies retained these loci as well as the CR binding ability (Fig 5F, right panel). Finally, the vast majority of pINV::Km ∆pcnB colonies retained the virulence loci tested, while showing a CR- phenotype (Fig 5F, right panel).

Altogether, these results suggest that, in the presence of active TA systems, pcnB does not impact pINV stability or maintenance. However, the higher rate of pINV replication sustained by pcnB activity increases virulence counterselection pressure under virulence inducing conditions. Upon prolonged exposure to such conditions, this ultimately results in the emergence of avirulent colonies, still harbouring pINV but lacking key virulence loci.

The pcnB gene fuels Shigella intraepithelial expansion

As noted earlier, we observed a ~ 3-fold early colonization defect for the Shigella ΔpcnB mutant compared to wt (3-6h p.i.) in 3D human enteroid infections (Fig 1F–1H). Upon invasion, Shigella quickly escapes the endocytic vacuole and gains access to the cytosol [4]. Subsequently, Shigella spreads between adjacent cells via IcsA-dependent host cell actin nucleation [11], while evading innate immune signaling [31].

To follow the fate of pcnB-proficient and -deficient intraepithelial Shigella populations after intestinal epithelial cell invasion, we performed time-lapse imaging of individual invasion foci in infected enteroid-derived 2D monolayers. The approach relied on Shigella wt or ΔpcnB strains harbouring the intracellular reporter uhpT-GFP, as previously described [31]. The membrane-impermeable DRAQ7 dye was used as a marker for intestinal epithelial cell death and permeabilization. Strikingly, while Shigella wt spread efficiently within the enteroid epithelium, the ΔpcnB mutant completely failed to expand over time (6-20h p.i.) (Fig 6A-B; S1 Movie). To exclude a bias in fluorescent reporter intensity between the two strains, we quantified GFP mean fluorescence intensity (MFI) for each invasion focus. We observed no difference in GFP intensity between the two strains initially (4h p.i.), indicating a similar reporter intensity. Over time, GFP MFI increased for the Shigella wt invasion foci, but remained stable in the Shigella ΔpcnB invasion foci. This likely depends on increased number and density of intraepithelial bacteria for the Shigella wt invasion foci (Fig 6C). Since the live imaging setup was based on a GFP-reporter plasmid, which could potentially be affected by the lack of pcnB, we repeated the infections of enteroid-derived 2D monolayers with Shigella wt or ΔpcnB strains and performed immunostaining using a polyclonal anti-Shigella antibody. At 12 hours p.i, we again observed minimal spread of the ΔpcnB mutant within the intestinal epithelium, confirming and reinforcing the conclusion that the pcnB gene promotes Shigella intraepithelial expansion (Fig 6D-E). Failure at intraepithelial expansion for the ΔpcnB mutant could encompass multiple factors, including: i) slower/defective escape from the endocytic vacuole; ii) lower intracellular replication; iii) impaired actin-based spread, and; iv) poorer immune evasion. This aligns with the reduced expression of a broad range of virulence proteins required for the different stages of Shigella intracellular life in the ΔpcnB mutant (e.g., IpaB, IpaC and IpgD involved in vacuolar escape, IcsA involved in actin-based spread; OspC3, IpaH7.8, OspG involved in immune evasion; Fig 2B–2D; S1 Table; reviewed in [3]). Altogether, we conclude that a suitable level of pINV, sustained by pcnB activity, is required to raise Shigella’s chances of invading the intestinal epithelium and subsequently to boost intraepithelial bacterial population expansion.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. The contribution of pcnB to Shigella intraepithelial expansion.

(A) Representative time-lapse series of invasion foci and DRAQ7-positive cells in enteroid-derived IEC monolayers infected with Shigella wt or ΔpcnB strains harbouring the cytosolic reporter puhpT-GFP. Scale bars: 50 µm. (B) Quantification of the GFP-positive area for Shigella wt or ΔpcnB infection foci. (C) Quantification of the GFP mean fluorescence intensity over time in Shigella wt or ΔpcnB invasion foci. Plotted are mean and standard deviation for at least 6 individually infected wells per strain with at least 3 FOV (Fields of View)/well; statistical significance determined by Two-way Anova. (D) Representative fluorescent microscopy images of invasion foci in enteroid-derived IEC monolayers infected with Shigella wt or ΔpcnB strains. scale bars: 50 µm. Roman numbers represent different examples of infection foci. (E) Quantification of the GFP-positive area for Shigella wt or ΔpcnB infection foci. Plotted are mean and standard deviation of at least 50 FOVs (Fields of Views)/strain derived from 4 individually infected wells per strain; statistical significance determined by Mann-Whitney U-test; ***p < 0.001.

https://doi.org/10.1371/journal.ppat.1013727.g006

Ethics statement

Pseudonymized human adult stem cell-derived jejunal enteroids were established in earlier studies [67–69]. Intestinal tissue resected during bariatric surgery was used as starting material. Formal written consent was obtained from each donor. All procedures were approved under ethical permit number 2010–157 with addenda 2010-157-1 and 2020-05754, and permit nr 2023-01524-01, by the local governing body (Etikprövningsmyndigheten, Sweden).

Bacterial strains, plasmids and general microbiological procedures

Bacterial strains are listed in S2 Table. M90T, a Shigella flexneri serotype 5 strain [70] was used as wt and for mutant construction. The ΔvirF and ΔmxiD Shigella mutants were previously generated [71,72]. For the ΔpcnB and ΔipaD Shigella mutants, deletions were introduced by Lambda Red recombination in the Shigella wt strain. wt and ΔmxiD Shigella barcoded strains were previously constructed and harbour TagA-B and TagC-D integrated between genes ybhC and ybhB, respectively [33]. Each genetic tag is unique and consists of 40 nucleotides [73]. For the ΔpcnB Shigella barcoded mutants, the pcnB deletion was introduced by Lambda Red recombination in the Shigella wt barcoded strains harbouring TagE-F [33]. The pINV::Km Shigella strain (ΔipaH9.8 mutant) was previously constructed [31]. For the pINV::Km ΔpcnB Shigella mutant, the pcnB deletion, bringing a CmI cassette, was introduced by Lambda Red recombination in the Shigella pINV::Km strain. All newly constructed mutants were verified by PCR. puhpT-GFP [31,74] carries a cytosolic fluorescent reporter (S2 Table). p-Empty carries a p15A ori and AmpR [33]. An in vitro-synthetized DNA fragment containing pcnB including its native promoter was cloned in the BamHI site of the p-Empty vector, obtaining p-pcnB. p-pcnB was designed in silico and purchased from GenScript. Bacteria were routinely grown in Luria Bertani broth (LB) at the indicated temperatures. When necessary, antibiotics were supplemented at the following concentrations: chloramphenicol, 12.5μg/ml; kanamycin, 50μg/ml; ampicillin, 50μg/ml. Plasmid DNA extraction, DNA transformation, electrophoresis, purification of DNA fragments and sequencing were performed as described previously [75]. PCR reactions were performed using DreamTaq DNA polymerase (Thermo Fisher Scientific, #EP0702) or Phusion DNA polymerase (Thermo Fisher Scientific, #F-530L). All oligonucleotide primers and plasmids used in this study are listed in S2 Table.

Enteroid maintenance culture

Human enteroids (pseudonym ID: 18–9jej) were previously established and thoroughly characterized elsewhere [67–69]. Human enteroids were passaged weekly by mechanical dissociation and incubation with gentle cell dissociation reagent (StemCell; #07174). Enteroid fragments were washed with DMEM-F12 (Dulbecco’s modified Eagle’s medium; Gibco, #11039021) containing 0.25% BSA (bovine serum albumin; Gibco; #15260-037) and re-embedded into Matrigel domes (Corning; #356230), using a 1:4 – 1:10 splitting ratio. OGM growth medium (StemCell; #06010) was exchanged every 2–4 days. Cultures were kept at 37°C in 5% CO₂.

Enteroid bulk infections

Medium sized enteroids were used for bulk infections as previously described [33]. Briefly, enteroids were extracted from Matrigel domes and incubated in Cell Recovery Solution (Corning, #354253) for at least 1h on ice on a rotating table. Subsequently, enteroids were collected by sedimentation, washed with DMEM-F12/0.25% BSA, re-suspended in OGM growth medium containing 8% cold Matrigel to maintain basolateral polarity (basal-out), and aliquoted in ultra-low attachment 24-well tissue culture plates (Corning Costar; #CLS3473-24EA). Basal-out suspension cultures were incubated at 37°C with 5% CO₂ for 1 day prior to infection. 1h before the infection, enteroids were washed three times with DMEM-F12/0.25% BSA, using 40µm mini cell strainers (Funakoshi, #HT-AMS-14002), re-suspended in OGM and aliquoted in ultra-low attachment 24-well tissue culture plates (Corning Costar; #CLS3473-24EA). The indicated Shigella strains were grown ON in LB containing appropriate antibiotics at 30°C, diluted 1:50, subcultured for 2h at 37°C without antibiotics and diluted in OGM media to obtain the desired MOI. Tagged strains were mixed in equal ratios. Bacteria were added to each well, spun down at 300g for 10min and incubated at 37°C with 5% CO₂. At 1h p.i., enteroids were washed 4 times with DMEM-F12/0.25% BSA using 25µm mini strainers (Funakoshi, #HT-AMS-12502) and incubated with media containing 200µg/ml gentamicin for 2h. For infections longer than 3h, media was replaced with 20µg/ml gentamicin media up to 6h p.i.. Infected enteroids were washed 6 times with DMEM-F12/0.25% BSA, recovered from the strainers and lysed in 0.1% Na-deoxycholate by homogenization with a Tissue Lyser (Qiagen). For barcoded infections, recovered intracellular bacteria were enriched in LB at 30°C for 14h. A dilution of the input inoculum was enriched similarly. For CFU counting, intracellular bacterial populations were serially diluted and plated on LB plates containing appropriate antibiotics.

Barcoded assays – tag quantification

For barcoded competition assays and barcoded infections, the indicated tagged strains were mixed in equal ratios, following ON growth at 30°C. Genomic DNA was extracted using the the DNeasy Blood & Tissue Kit (Qiagen; #69504) from the “input consortium” and the “output consortium” (following growth under the indicated conditions, or following an infection assay). qPCR was performed to quantify each tag using Maxima SYBR green/ROX qPCR master mix (2X) (Thermo Fisher Scientific, #K0222) on a CFX384 Touch Real-Time PCR Detection System (Biorad), using 9ng of gDNA and tag-specific primers (S2 Table). The relative abundance of each strain was normalized to the abundance in the inoculum. For barcoded infection assays, a Colonization Index was calculated as 1-(avg relative abundance^WT/ relative abundance^MUT) as previously described [33]. Statistical significance was determined with a paired t-test between the relative input and output abundances of each strain normalized by the respective mean WT abundance across all barcoded consortium infections or competition assays.

qRT-qPCR

Total RNA extraction was performed with acid phenol and cDNA synthesis was performed as previously described [76]. qRT-PCR was performed using Maxima SYBR green/ROX qPCR master mix (2X) (Thermo Fisher Scientific, #K0222) on a CFX384 Touch Real-Time PCR Detection System (Bio-Rad). The levels of virF, virB, icsA, mxiD, mxiE and pcnB transcripts were analysed using the 2^−ΔΔCT (cycle threshold [CT]) method and results are reported as relative expression with respect to the reference. The gene nusA was used for normalization. The oligonucleotide primers used are listed in S2 Table.

Mass spectrometry-based proteomics

Shigella wt, ΔpcnB and ΔvirF strains were grown at 37°C until OD600 reached ~0.7. 2ml of bacterial cultures were pelleted and washed once in PBS. Four biological replicates were collected for each strain. Bacterial cells were lysed in 50µl 2% SDS and boiling at 98°C for 10min. 5μg of each sample were denatured in a final concentration of 2% SDS and 20mM TCEP. Samples were digested as previously described [77], with a modified sp3 protocol [78]. Briefly, samples were added to a bead suspension (10μg of beads; Sera-Mag Speed Beads, 4515-2105-050250, 6515-2105-050250, in 10μl 15% formic acid and 30μl ethanol) and incubated shaking for 15min at room temperature. Beads were then washed four times with 70% ethanol. Proteins were digested overnight by adding 40μl of 5mM chloroacetamide, 1.25mM TCEP, and 200ng trypsin in 100mM HEPES pH 8.5. Peptides were eluted from the beads and dried under vacuum. Peptides were then labelled with TMTpro (Thermo Fisher Scientific), pooled and desalted with solid-phase extraction using a Waters OASIS HLB μElution Plate (30μm). Samples were fractionated onto 48 fractions on a reversed-phase C18 system running under high pH conditions, with every sixth fraction being pooled together. Samples were analysed by LC-MS/MS, using a data-dependent acquisition strategy on a Thermo Fisher Scientific Vanquish Neo LC coupled with a Thermo Fisher Scientific Orbitrap Exploris 480. Raw files were processed with MSFragger [79] against the NCBI Shigella flexneri 5a strain M90T genome (CP037923 and CP037924) using standard settings for TMT. Data was normalized using vsn [80] and statistical significance was determined using limma [81]. Peptide intensities stemming from the deleted genes due to peptide co-isolation were set to zero [82].

gDNA extraction and pINV copy number determination by droplet digital PCR (ddPCR)

Shigella flexneri M90T wt, M90T ΔpcnB and M90T ΔipaD strains were grown in LB broth at the indicated temperature to an OD 600 nm of ~0.7. Genomic DNA extraction was performed by bead beating and phenol/chloroform purification. Briefly, 1.5ml of bacterial cultures were spun down and resuspend in 250µl of a 1:1 mix of Fast lysis buffer for bacterial lysis (QIAGEN) and nuclease-free water. The resulting samples were added to 2ml tubes containing 0.75g of acid-washed glass beads (diameter: 212–300μm; Sigma Aldrich, #G1277) together with 250μl of Phenol:Chloroform:Isoamyl Alcohol (25:24:1) (Sigma Aldrich; #77617–100ML). After 2 min ice incubation, the samples were bead beated twice with the following settings: 6.5 m/s, 24x2, 20 seconds using a FastPrep-24 Classic homogenizer (MP Biomedicals, #116004500). Subsequently the lysates were spun down for 3 min at 8,000 rpm at 4°C. The aqueous phase was transferred to a clean tube, and 1 volume of Phenol:Chloroform:Isoamyl Alcohol was added to each tube. The samples were mixed briefly by vortexing and centrifuged 3 min at 13,000 rpm at 4°C. The aqueous phase containing gDNA was transferred to a clean tube and used in subsequent ddPCR quantification. When indicated gDNA extraction was performed using the DNeasy Blood & Tissue Kit (Qiagen; #69504) or the GeneJET Genomic DNA Purification Kit (Thermo Scientific, #K0721). gDNA preparations were quantified by Qubit with the dsDNA HS Assay Kit (Invitrogen, #Q32851). ddPCR to determine pINV copy number was performed with the QX200 Droplet Digital PCR System (Bio-Rad) in 25µl reactions containing 2x EvaGreen Supermix (Biorad, #1864034), 5U of HindIII (Thermo Scientific; #ER0502), gDNA template (final concentration: 0.78pg/µl unless otherwise indicated) and primers (200nM each). For optimization, 6.25pg/µl of gDNA extracted by bead beating from Shigella wt were diluted in two-fold serial dilutions (up to 0.097656pg/ µl). To minimize pipetting errors, a pre-mix containing all components except the primers was prepared for each sample, aliquoted and primers were added last to each well. Two primer pairs located on pINV and one primer pair located on the chromosome were used (primer sequence in S2 Table; primer location in S2A Fig). Droplet generation was performed using the QX200 automated droplet generator. Subsequent PCR amplification was performed with a CFX deep-well PCR (Bio-Rad) with the following settings: initial activation at 95°C for 5 min, 39 cycles of denaturation at 94°C for 30 sec, annealing and elongation at 56°C for 1 min and one cycle of stabilization at 4°C for 10min and 90°C for 5 min. Following PCR amplification, droplets were detected with a QX200 Dropler Reader (Bio-Rad) and pINV copy number for each sample was calculated as following: nr of positive droplets with the pINV primers/ nr of positive droplets with the chromosome primers. Total droplet count of at least 15,000 was considered reliable.

Northern blot analysis

Shigella flexneri M90T wt or ΔpcnB strains were grown in LB broth at 30°C or 37°C to an OD 600 nm of ~0.7. Shigella wt or ΔpcnB strains carrying the p-Empty or p-pcnB plasmid were grown at 37°C at OD600 0.7. E.coli. Dh10B strain, used as a negative control, was grown at 37°C to an OD 600 nm of ~0.7. Total RNA was extracted with acid phenol as previously described [33]. Northern blotting was performed as previously described [83], with the following modifications. Five micrograms of RNA were separated on denaturing gels (6% polyacrylamide, 7M Urea, 1x TBE) and the RNA was transferred to Amersham Hybond-N+ nylon membranes (Cytiva, #RPN203B). After crosslinking at 1200 mJ/cm², the membranes were equilibrated with Church buffer [84], rolling at 42°C for 45 min. Radioactively 5’ end labelled and denatured DNA probes were added and the membranes were incubated rolling at 42°C over night. The membranes were washed twice with wash buffer (2x SSC/ 0.1% SDS), and the radioactive signals were exposed to phosphor screens and detected using a Typhoon 5 phosphoimager (Cytiva). The sequences of the DNA probes used to detect copA mRNA (copA_Shig) and 5S rRNA (5S) are found in S2 Table. The DNA probes were labeled using T4 polynucleotide kinase (Thermo Fisher Scientific, #EK0031) and [γ-32P]-ATP (Revvity). Unincorporated label was removed using Amersham MicroSpin G-50 columns (Cytiva, #27533001).

pINV stability assay

Shigella flexneri pINV::Km or pINV::Km ΔpcnB strains were grown in LB broth at 30°C ON. Following this initial ON growth, cultures were diluted daily 1:5000 in 10 ml of LB medium and grown at 37°C with shaking at 180 rpm. To assess plasmid stability, samples were taken after approximately 25 and 50 generations, and cells were plated on non-selective LB-agar plates. After overnight incubation at 37°C, one hundred clones were patched on non-selective or Kanamicin- supplemented LB-agar plates to determine the presence of pINV. Plasmid stability was calculated as the percentage of colonies resistant to Km relative to the total number of colonies tested. To monitor virulence loss, samples taken at 50 generations were also plated on CR plates and incubated at 37°C. To monitor virulence loci loss, 10% of the patched colonies after 50 generations were tested by PCR for the presence of pINV (ipaH9.8-Km locus) and the virF and virB loci. The oligonucleotide primers used are listed in S2 Table.

Infection and live-cell imaging of 2D human enteroid-derived monolayers

Human enteroid-derived monolayers were generated on Matrigel-coated polymer coverslips (µ-Slide 8-well high; Ibidi; #80806) as previously described [31]. In brief, 7 days after passaging, enteroids were extracted from Matrigel domes using Gentle Cell Dissociation Reagent (StemCell), washed in DMEM/F12/1.5% BSA and dissociated into single cells using TrypLE (Gibco; #12605010) and vigorous pipetting. The single cell suspension was washed again with DMEM/F12/1.5% BSA, re-suspended in OGM supplemented with 10µM Y-27632 (Sigma-Aldrich; #Y0503-1MG) and seeded out at 750’000 cells/cm2 in wells pre-coated with a 1:40 Matrigel dilution in DPBS (Gibco; #14190250). The Y-27632 ROCK inhibitor was removed 2 days after seeding. Monolayers were used for infection 3 days after seeding. Shigella wt or ΔpcnB strains carrying the intracellular reporter puhpT-GFP were grown ON in LB containing 50µg/ml Ampicillin at 30°C and subcultured at 37°C in LB containing 50µg/ml Ampicillin. Equal numbers of bacteria were added to each well and spun down at 700g for 10min. Monolayers were incubated at 37°C, 5% CO2 for 40 min to allow bacterial invasion. Subsequently, monolayers were washed three times with DMEM/F12/400 µg/mL gentamicin, before replacing the medium with fresh OGM containing 50µg/ml Gentamicin (Sigma, #G1914) and 0.75 µM Draq7 (Invitrogen, #D15106). Imaging was performed on a custom-built microscope based on an Eclipse Ti2 body (Nikon), using a 40x/0.6 Plan Apo Lambda air objectives (Nikon) and a Prime 95B 25mm camera (Photometrics). The imaging chamber was maintained at 37°C, 5% CO2 in a moisturized atmosphere. Bright-field images were acquired using differential interference contrast (DIC), and fluorescence was imaged using the 475/34, or 648/20 excitation channels of light engine Spectra-X (Lumencor) and emission collected through quadruple bandpass filters (89402 & 89403; Chroma). Invasion foci in enteroid-derived monolayers were manually selected and images were acquired every 30 min. Image analysis was conducted in Fiji [85] by applying rolling ball background subtraction (radius 50 pixels) to the fluorescent channels and quantifying the area and mean fluorescence intensity (MFI) above a set threshold.

Immunostaining and fluorescent microscopy

Shigella wt or ΔpcnB strains were grown ON in LB at 30°C and subcultured in LB at 37°C prior to infection of enteroid-derived monolayers, as described above. At 12h p.i. monolayers were fixed with 2% paraformaldehyde (Pierce, #28906) for 15 min at 37°C and washed three times with PBS. Subsequently, monolayers were permeabilized with 0.5% Triton-X-100 (Sigma-Aldrich, #T8787) for 10 minutes at room temperature (RT) and blocked with 3% BSA (Sigma-Aldrich, #A9418) at 4°C ON. Monolayers were stained with anti-Shigella antibody (1:10) (Shigella polyclonal antibody, Invitrogen, #PA17245) for 2h at 37°C, followed by incubation with the secondary antibody (1:200) (Goat-α-rabbit-IgG(H + L)-AF488, Invitrogen, #A11034) for 2h at RT. Finally, monolayers were stained with 4′,6-diamidino-2-phenylindole (DAPI; 1:1,000) (Sigma, #D9542) and F-actin (1:100) (phalloidin-Alexa Fluor 647, Molecular Probes, #A22287) for 45 minutes at room temperature, before washing with PBS and imaging. Imaging was performed on a custom-built microscope based on an Eclipse Ti2 body (Nikon), using 60 × /0.7 and 40 × /0.6 Plan Apo Lambda air objectives (Nikon) and a and a Prime 95B 25mm camera (Photometrics).