The small non-coding RNA rli106 contributes to the environmental adaptation and pathogenicity of Listeria monocytogenes

Listeria monocytogenes (LM), an important Gram-positive foodborne human-animal pathogen of the Listeria genus, is widely distributed in nature and can be transmitted by a variety of routes (2, 10). Infection with LM in humans and animals can manifest clinically as gastroenteritis, meningitis, abortion and sepsis (24, 28). Especially once neonates, pregnant females and immunocompromised individuals are infected, the pathogen can cause the death of the patient. The associated disease, listeriosis, has a mortality rate of 20% to 30% in these groups (13, 19, 21, 25). Usually, LM can contaminate meat, eggs and dairy products during animal food production and processing, causing serious public health and safety problems (13, 22).

It has been revealed that small non-coding RNA (sRNA) sequences can bind to target messenger RNA (mRNA) and regulate expression at the post-transcriptional level of bacterial genes which play important roles in response to environmental changes and virulence regulation (1, 3, 12, 15). So far, more than 150 sRNA species have been identified in LM using bioinformatics and molecular biology techniques. However, the biological functions of most of them are still unclear (18, 26). One, rli106, is a newly identified sRNA species in the LM genome, which displays significant differences in intracellular and extracellular expression following LM infection in host cells (5). To date, the regulatory role of rli106 remains unknown.

The main purpose of this study was to investigate the biological function of the rli106 sRNA in regard to the environmental adaptation and pathogenicity of LM. In this study, we constructed rli106 deletion and complementation strains using a homologous recombination technique. By determining and comparing the differences in the environmental adaptability and pathogenicity of rli106 deletion, complementation and parental strains, the regulatory roles of rli106 in environmental adaptability and pathogenicity mechanisms were revealed. The target gene of rli106 was also predicted and verified using a co-expression system based on E. coli and Western blot analysis, which provided new insights into the molecular mechanism of sRNA mediation of environmental adaptation and pathogenicity in Listeria monocytogenes.

Primer design. The primers for the deletion and complementation strains were designed according to the genome sequence of the LM EGD-e parental strain registered in GenBank (accession number AL591824.1) using Primer 5.0 software (Premier Biosoft International, San Francisco, CA, USA). Enzyme-recognising sites such as BamH I, Hind III and Kpn I (TaKaRa Bio, Kusatsu, Japan) were introduced into the 5ʹ end of these primers (Table 1).

Construction of deletion mutant and complementation strains. The LM EGD-e parental strain (kindly donated by Prof. W. Goebel of the University of Würzburg, Germany) was cultured in brain heart infusion broth (BHI) (Sangon, Shanghai, China). Briefly, LM genomic DNA was extracted through the steps specified in the Bacterial Genomic DNA Extraction Kit (Omega, Norcross, GA, USA), and the rli106 gene was amplified using rli106 forward (F) and rli106 reverse (R) primers. The PCR product was recovered and ligated with a pMD19-T simple vector (TaKaRa Bio) to generate pMD19-T-rli106 for sequencing. Next, the upstream and downstream homologous arms of the rli106 gene were amplified with P1-P2 and P3-P4 primer pairs, respectively, and the deletion fragment Δrli106 was generated by splicing by overlap extension PCR. The Δrli106 fragment was cloned into the pMD19-T simple vector (TaKaRa Bio) to obtain the pMD19-T-Δrli106. The recombinant shuttle pKSV7-Δrli106 plasmid was generated by inserting the Δrli106 fragment into the shuttle pKSV7 plasmid. After that, pKSV7-Δrli106 was electrotransformed into LM EGD-e competent cells at 2,500 V for 5.0 ms, and the positive clones were screened by PCR using the P5-P6 primer pair.

The positive clones were cultured for 10 generations at 42°C in BHI medium containing 10 μg/mL chloramphenicol, and passaged for 20 generations in BHI liquid medium without chloramphenicol to generate an LM-Δrli106-deletion mutant strain. To construct the rli106 gene complementation strain, rli106 was first cloned into a pHT304 vector to obtain the recombinant plasmid pHT304-rli106. In the next step, pHT304-rli106 was electrotransformed into LM-Δrli106 competent cells and cultured on plates containing erythromycin-resistant BHI (final concentration 50 μg/mL) to generate the CLM-Δrli106 complementation strain.

Growth assay of LM. LM EGD-e, LM-Δrli106 and CLM-Δrli106 were inoculated into BHI liquid medium at the ratio of 1 : 100 and cultured at 30°C, 37 °C and 42°C, respectively. The 600 nm optical density (OD_600nm) was measured at different times, and growth curves were plotted. At the same time, the bacterial broth was transferred to BHI liquid medium under different conditions of pH 4, pH 7, pH 9, 5% NaCl, 8% NaCl, 3.8% ethanol and 5mM H₂O₂. The growth curves were plotted to examine the effects of the rli106 gene deletion on LM’s response to environmental stressors under different conditions (18).

Assay of motility and biofilm formation of LM. Briefly, LM EGD-e, LM-Δrli106 and CLM-Δrli106 were inoculated into BHI solid medium and incubated overnight at 37°C. Then, the semi-solid BHI tube (containing 0.3% agar) was punctured with the inoculation needle containing the bacteria and cultured at 28°C for 36 h. In addition, 1 μL of the bacterial solution was spotted on a 0.3% semi-solid BHI plate and cultured at 28°C for 36 h, after which time the diameters of the bacterial colonies which formed on the plate were measured to access the motility of bacteria. Meanwhile, LM EGD-e, LM-Δrli106 and CLM-Δrli106 were inoculated into 96-well microplates and cultured at 37°C for 12 h, 24 h and 48 h, respectively. The plates were stained with a 1% crystalline violet solution according to Zetzmann et al. (28) and decolourised by adding 95% ethanol, and biofilm size was measured by reading at OD_570nm using an enzyme-linked detector (BioTek Instruments, Winooski, VT, USA). Before decolourisation, the microtitre plate was placed under an inverted microscope (Leica Microsystems, Jena, Germany) for observation and photographic recording.

Determination of adhesion, invasion and intracellular survival of LM. In brief, RAW264.7 mouse macrophages were cultured in six-well microplates, and the cells were inoculated with LM EGD-e, LM-Δrli106 and CLM-Δrli106 at a 10 : 1 ratio of multiples of infection. One hour after infection, the cells were washed three times with phosphate-buffered saline (PBS), and 500 μL of 0.1% Triton X-100 was added to lyse the cells for 10 min. The lysate was collected for bacterial counting, and the adhesion rate was calculated according to the reference described by Peng (18). Bacteria were allowed to infect RAW264.7 cells for 1 h, and then replaced with fresh Dulbecco’s modified Eagle’s medium containing 100 μg/mL gentamicin for another 1 h. The cells were washed with PBS three times and lysed, their bacterial contents were counted and then the invasion rate was calculated. Additionally, 100 μg/mL gentamicin was used to kill the bacteria on the surface of infected RAW264.7 cells. After 2, 4, 6, 8, 10 and 12 h of cell culturing, the cells were lysed and the bacteria in them were counted to determine the intracellular viability of LM EGD-e, LM-Δrli106 and CLM-Δrli106.

Determination of pathogenicity of LM. One hundred and twenty 6–8-week-old Kunming mice were randomly divided into four groups. The use of these mice was approved by the Research and Ethical Committee of Shihezi University. Each group was further divided into five subgroups of six mice, each of which was injected intraperitoneally either with 500 μL of LM EGD-e, LM-Δrli106 and CLM-Δrli106 bacterial solutions at one of the dilution gradients 105, 106, 107, 10⁸ or 10⁹ CFU/mL, or with sterile PBS (pH 7.2). After observation for seven consecutive days, the mice deaths were counted every day, the survival curves were plotted and the lethal dose, 50% (LD₅₀) was calculated using Karber’s method. In parallel, twenty-four 6–8-week-old Kunming mice were randomly divided into four groups. Each group was intraperitoneally injected with one of a sublethal dose of LM EGD-e, a sublethal dose of LM-Δrli106, a sublethal dose of CLM-Δrli106 or sterile PBS (pH 7.2). After infection, these mice were humanely sacrificed and their spleens and livers were collected aseptically every day to determine bacterial loads in the organs. These organs were fixed with formalin and tissue sections were prepared for histopathological observation.

Predictive analysis of target genes of rli106. Potential target genes were predicted by the online TargetRNA2 software (http://cs.wellesley.edu/TargetRNA2). The possible regulatory mode of rli106 on target genes was analysed.

Analysis and verification of interaction between rli106 and target gene mRNA. Briefly, rli106 sRNA and the target gene were amplified and cloned into the pUT18C and pMR-LacZ vectors to generate the recombinant pUT18C-rli106 and pMR-LacZ-target plasmids. The pUT18C vector and pUT18C-rli106 plasmid were electrotransferred into the BTH101 E. coli competent cells, and then the pMR-LacZ-target was electrotransferred into BTH101 containing the pUT18C or pUT18C-rli106 plasmid. The bacteria were cultured in Luria-Bertani (LB) liquid medium containing ampicillin (Amp, 100 μg/mL) and kanamycin (Kan, 100 μg/mL) for 4 h at 37°C, and then cultivated in LB plates containing 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-gal), Amp (100 μg/mL) and Kan (100 μg/mL). Then the colour and OD_420nm of the colonies of different strains were observed and recorded.

Quantitative RT-PCR. The mRNA level of the DegU gene relative to the internal reference 16S rRNA gene was measured using a quantitative RT-PCR (RT-qPCR). In brief, the cultures of LM EGD-e, LM-Δrli106 and CLM-Δrli106 were diluted 100-fold with BHI medium and cultured to OD_600nm ≈ 0.5. Total RNA was extracted using Trizol (Invitrogen, Carlsbad, CA, USA) and reverse transcribed into cDNA via the AMV reverse transcriptase kit (TaKaRa). The RT-qPCR was performed using the SYBR Premix Ex Taq kit (TaKaRa) on a LightCycler 480 PCR device (Roche, Basel, Switzerland) according to the manufacturers’ instructions. Each sample was assayed in three replicates and the mRNA level of the target gene was calculated using the 2^-ΔΔCT method (15, 18).

Western blot analysis. Briefly, LM EGD-e, LM-Δrli106 and CLM-Δrli106 were inoculated into BHI liquid medium and incubated overnight at 37 °C. Protein extracts were prepared by ultrasonic lysis for sodium dodecyl sulphate polyacrylamide gel electrophoresis. After these proteins were transferred, the membrane was blocked by a 5% BSA buffer at 4°C overnight. Western blot analysis was then performed using the mouse anti-target protein antibody as the primary antibody, and HRP-labelled rabbit anti-mouse IgG as the secondary antibody. The bands of Western blot analysis were scanned using ImageJ 1.8.0 software (23), and the effect of rli106 on the protein level of the target mRNA was quantitatively analysed.

Statistical analysis of data. GraphPad Prism 5.0 software (GraphPad Software, San Diego, CA, USA) was used to analyse the experimental data. Statistical analysis was performed on continuous variables using one-way analysis of variance to compare whether there were significant differences between the groups. Variable data were analysed using the chi-squared test. Differences were considered significant when P < 0.05, and extremely significant when P < 0.01.

The deletion mutant strain LM-Δrli106 and the complementation strain CLM-Δrli106 were successfully generated, which was confirmed by PCR and sequencing (Supplementary Figs S1, S2, S3 and S4).

The growth of the three strains did not differ significantly (P > 0.05) under the conditions of 30°C and 37°C. However, at 42°C, the growth of LM-Δrli106 slowed after 8 h when compared with that of LM EGD-e and CLM-Δrli106, indicating that LM-Δrli106 was less adaptable in high-temperature environments (Fig. 1A). Moreover, under the conditions of pH 9, 5% NaCl and 8% NaCl, 3.8% ethanol and 5 mM H₂O₂, the growth of LM-Δrli106 was less than that of LM EGD-e and CLM-Δrli106 (Fig. 1B–1F).

Fig. 1

As shown in Fig. 2, there were no significant differences (P > 0.05) in the motility of LM-Δrli106 when compared to that of LM EGD-e and CLM-Δrli106 in the semi-solid BHI test tube (Fig. 2A) and on the BHI plates (Fig. 2B and 2C), indicating that rli106 gene deletion did not affect the motility of LM.

Fig. 2

All strains were able to form biofilms over 12–24 h. However, the biofilm formation by LM-Δrli106 was significantly weaker than that by LM EGD-e and CLM-Δrli106 (P < 0.01) at 48 h (Fig. 3A and 3B).

Fig. 3

Compared with LM EGD-e and CLM-Δrli106, the adhesion and invasion capacity of LM-Δrli106 were notably decreased, with a highly significant difference (P < 0.01) (Fig. 4A and 4B). Bacterial loads in RAW264.7 cells were significantly lower in LM-Δrli106 (P < 0.01) when compared with those of LM EGD-e and CLM-Δrli106 (Fig. 4C).

Fig. 4

The LD₅₀ of LM-Δrli106 was significantly decreased when compared with those of LM EGD-e and CLM-Δrli106 (P < 0.05) (Supplementary Table 1). Survival curves were significantly longer in mice inoculated with LM-Δrli106 when compared with the curves of mice in the LM EGD-e and CLM-Δrli106 groups (P < 0.05) (Fig. 5A). On days 1, 2 and 5 post infection, the bacterial loads in the livers and spleens of mice infected with LM-Δrli106 were significantly lower than those of mice infected with LM EGD-e and CLM-Δrli106 (P < 0.01) (Fig. 5B and 5C).

Fig. 5

Compared with the control mice (PBS-injected group), the livers, spleens and kidneys of mice of the LM EGD-e, LM-Δrli106 and CLM-Δrli106 groups displayed different degrees of pathological changes. Cell degeneration and necrosis were present in the livers of the LM EGD-e and CLM-Δrli106 group mice. Pathological changes in the spleens were haemorrhage and lymphocyte infiltration. Also, granular degeneration and hyperplasia in the kidney were observed. In contrast, the pathological changes in LM-Δrli106-infected mice were significantly weaker (Fig. 6), indicating that deletion of the rli106 gene can weaken the pathogenicity of LM.

Fig. 6

The bases 12–32 of rli106 can complementarily pair with the bases 49–69 of DegU 5′ untranslated region (UTR), suggesting that rli106 may interact with DegU mRNA. This interaction possibly plays a biological role in regulating DegU mRNA (Fig. 7A). The results of the expression experiment in the two plasmid co-expression system showed that the lawns of E. coli co-transformed with pUT18C-rli106 and pMR-LacZ-DegU turned a deeper blue on X-gal plates than lawns co-transformed by the pUT18C and pMR-LacZ-DegU plasmids (Fig. 7B, 7C and 7D), which suggests that rli106 can positively regulate the expression of the DegU gene through the rli106–DegU mRNA interaction.

Fig. 7

Quantitative RT-PCR results revealed that the mRNA level of the DegU gene was significantly downregulated in the deletion strain compared to LM EGD-e and CLM-Δrli106 (P < 0.01), indicating that rli106 can modify the mRNA level of the DegU gene via base pairing (Fig. 7E).

Compared with the level in the LM EGD-e and CLM-Δrli106 strains, Western blot analysis showed that the protein level of DegU in the LM-Δrli106 strain was decreased (P < 0.01) (Fig. 7F and 7G), which suggests that rli106 can modulate the expression of the DegU gene at the post-transcriptional level via a positive regulatory mechanism.

Current studies have shown that sRNA species can bind to target mRNA to regulate gene expression at the post-transcriptional level and are considered to be key negative or positive regulators of gene expression (8, 9, 16, 17). Therefore, they may participate in the modulation of physiological processes in bacteria, including stress responses, population sensing, toxin-antitoxin systems or pathogenicity (5, 7, 14). Several sRNA species in LM, such as rliB, rli31, rli33-1, rli27, rli47 and rli50, have recently been shown to be involved in the regulation of bacterial growth and virulence. Among them, rli27 can regulate the expression of wall proteins by binding to the 5′-UTR of the lmo0514 gene (20), while rli47 may repress isoleucine biosynthesis through a direct interaction with the ilvA transcript in LM (14). In addition, existing studies have verified that sRNA species, as well as positive regulatory factor A and alternative general stress sigma factor σB, formed a highly complex regulatory network in LM, which enabled the bacteria to survive and propagate in vitro and in vivo (9, 18).

Listeria monocytogenes can adapt to various stressful environments and survive in food processing, storage and transportation. It has been proved that sRNA species play important regulatory roles in bacteria during environmental changes such as variations in temperature, nutrition, oxidative stress and pH. Some sRNA species were involved in transmitting the signal and regulating stress responses through two-component signal transduction systems. In this investigation, we demonstrated that deletion of the rli106 gene can suppress the adaptation and biofilm formation of LM under environmental stresses. It was shown that rli106 played important regulatory roles in LM’s resistance to environmental stress. Notably, a binding site of the cpxR transcription factor was predicted in the 5′-UTR sequence of the rli106 gene (Supplementary Fig. S5). It has been proved that cpxR is a regulatory response protein of two-component systems in bacteria, which contributes to membrane stability and virulence by regulating the expression of target genes (6, 27). Thus, it was reasonably hypothesised that cpxR might regulate rli106 to enhance the environmental adaptation of LM.

In the present study, the effects of rli106 on the virulence of LM were also analysed by macrophage infection and mouse infection assays. The results showed that in interaction with macrophages, LM-Δrli106 had significantly lower adhesion, invasion and intracellular proliferation ability, and displayed significantly lower virulence compared to LM EGD-e and CLM-Δrli106. These results indicated that rli106 plays an important role in the regulation of LM pathogenicity.

Using TargetRNA2 software, the target genes of sRNA rli106 were successfully predicted. To verify the regulatory relationship between sRNA rli106 and DegU gene mRNA, two plasmid co-expressing systems based on E. coli and Western blot analysis were employed to confirm the interaction between sRNA rli106 and the DegU gene. The results showed that sRNA rli106 could positively regulate the expression of the DegU target gene. We deduced that sRNA rli106 can protect DegU mRNA from degradation by binding to the sequence of the DegU target gene mRNA. This regulatory mode is similar to the mode described previously in several publications (6, 16, 17, 25, 27). Given that the DegU protein is a pleiotropic regulator involved in the regulation of virulence genes (4, 11), it can be inferred that sRNA rli106 can modulate environmental adaptation and pathogenicity through the DegU protein in LM.

In conclusion, this study demonstrated for the first time that rli106 sRNA contributed to the modulation of environmental adaptation and pathogenicity through regulating the expression of DegU mRNA, which provides new insights into the regulatory roles and mechanisms of sRNA mediation in LM.