The two-component system histidine kinase EnvZ contributes to Avian pathogenic Escherichia coli pathogenicity by regulating biofilm formation and stress responses

Abstract

EnvZ, the histidine kinase (HK) of OmpR/EnvZ, transduces osmotic signals in Escherichia coli K12 and affects the pathogenicity of Shigella flexneri and Vibrio cholera. Avian pathogenic E. coli (APEC) is an extra-intestinal pathogenic E. coli (ExPEC), causing acute and sudden death in poultry and leading to severe economic losses to the global poultry industry. How the functions of EnvZ correlate with APEC pathogenicity was still unknown. In this study, we successfully constructed the envZ mutant strain AE17ΔenvZ and the inactivation of envZ significantly reduced biofilms and altered red, dry, and rough (rdar) morphology. In addition, AE17ΔenvZ was significantly less resistant to acid, alkali, osmotic, and oxidative stress conditions. Deletion of envZ significantly enhanced sensitivity to specific pathogen-free (SPF) chicken serum and increased adhesion to chicken embryonic fibroblast DF-1 cells and elevated inflammatory cytokine IL-1β, IL6, and IL8 expression levels. Also, when compared with the WT strain, AE17ΔenvZ attenuated APEC pathogenicity in chickens. To explore the molecular mechanisms underpinning envZ in APEC17, we compared the WT and envZ-deletion strains using transcriptome analyses. RNA-Seq results identified 711 differentially expressed genes (DEGs) in the envZ mutant strain and DEGs were mainly enriched in outer membrane proteins, stress response systems, and TCSs. Quantitative real-time reverse transcription PCR (RT-qPCR) showed that EnvZ influenced the expression of biofilms and stress responses genes, including ompC, ompT, mlrA, basR, hdeA, hdeB, adiY, and uspB. We provided compelling evidence showing EnvZ contributed to APEC pathogenicity by regulating biofilms and stress response expression.

INTRODUCTION

Avian pathogenic Escherichia coli (APEC), one of the extra-intestinal pathogenic E. coli (ExPEC) with a broad host spectrum, is a serious threat to poultry industry development and one of the main causes of morbidity and mortality in birds (Moulin-Schouleur et al., 2007; Kathayat et al., 2021). In APEC, there are several two-component systems (TCSs), including UvrY/BarA, CpxR/A, PhoP/Q, OmpR/EnvZ, BasR/S, and QseB/C, which help bacteria adapt to complex external environments and exert pathogenic effects (Oshima et al., 2002; Gotoh et al., 2010; Cai et al., 2013; Breland et al., 2017; Tiwari et al., 2017). A typical TCS contains a histidine kinase (HK) to receive external input signals and a response regulator (RR) to transmit signals. The HK detects many environmental stimuli which are transferred to homologous RRs via autophosphorylation and catalytic phosphorylation (Tokishita et al., 1991; Mascher et al., 2006; Gotoh et al., 2010; Gushchin et al., 2017). Studies have indicated histidine kinases have been found to play an important role in the TCSs and the regulation of APEC pathogenesis. QseC, a bifunctional sensor kinase, controls the phosphorylation state of QseB and contributes to APEC virulence and innate immunity (Chaudhari and Kariyawasam, 2014). The UvrY/BarA HK BarA is involved in DF-1 cell adhesion and invasion, serum resistance, and APEC virulence (Herren et al., 2006). CpxA is a sensor kinase of the CpxR/A stress response system and is required for adherence, invasion, motility, and the production of type Ⅰ fimbriae and biofilms in APEC (Matter et al., 2018). Thus, TCS HKs are key regulators that mediate bacterial responses to environmental changes to establish pathogen infection.

EnvZ, the histidine kinase of OmpR/EnvZ TCS, consists of 450 amino acid residues and has 2 transmembrane helices: a periplasmic structural domain and a cytoplasmic structural domain (Tokishita et al., 1991; Khorchid et al., 2005). When stimulated, EnvZ senses changes in osmotic pressure through its cytoplasmic structural domain, causing autophosphorylation at His-243 and later transferring the phosphate group to OmpR, with OmpR-P binding to outer membrane protein genes and responding to osmotic stress (Tiwari et al., 2017). In E. coli K12, EnvZ binds to the inner membrane protein MzrA and responds to environmental signals (Gerken et al., 2009). In Shigella flexneri, EnvZ regulates vir expression and affects the invasion of Hela cells (Bernardini et al., 1990). In Vibrio cholerae, EnvZ is required for virulence during infant mouse intestinal colonization (Xi et al., 2020).

Since EnvZ has been well characterized as an important factor in other Gram-negative bacteria, in this study, we evaluated the biological and virulence properties of envZ in APEC. We explored the function of EnvZ in APEC pathogenesis and showed EnvZ participated in biofilm formation, curli production, responses to environmental stresses (including acid, alkali, osmotic, and oxidative), serum resistance, and APEC pathogenicity. Transcriptomic studies showed EnvZ affected bacterial virulence genes and helped regulate several important biological processes, including microbial metabolism in diverse environments, biofilm formation, and other TCSs. All in all, we proved that the TCS histidine kinase EnvZ plays a key role in APEC pathogenicity, and provides potential drug targets for the treatment and prevention of APEC infection.

MATERIALS AND METHODS

Bacterial Strains, Plasmids, and Growth Conditions

The APEC17 strain is a serotype O2 strain isolated from a case of avian colibacillosis in Anhui Province, China (Tu et al., 2016). All strains and plasmids are listed in Table 1. Bacterial strains were grown in Luria-Bertani (LB) medium (Sangon, Shanghai, China) at 28°C or 37°C. If required, antibiotics were added at the following concentrations: ampicillin (100 µg/mL) and chloramphenicol (30 µg/mL) (Sangon).

Table 1.

Strains and plasmids used in this study.

Strains	Relevant genotype	Source
AE17	APEC wild-type strain	Laboratory stock
AE17ΔenvZ	AE17 envZ-deletion mutant strain	This study
AE17C-envZ	Cmr, AE17ΔenvZ with the complement plasmid pSTV28-envZ	This study
Plasmids
pKD46	Ampr, expresses λ red recombinase, temperature-sensitive	Takara
pKD3	Ampr Cmr, cat gene, confers resistance to chloramphenicol template plasmid	Takara
pCP20	Ampr Cmr, temperature-sensitive plasmid	Takara
pSTV28	Cmr, cloning vector with pACYC184 replication starting point	Takara
pSTV28-envZ	Cmr, pSTV28 with envZ gene	This study

Abbreviations: Cm^r, chloramphenicol-resistant; Amp^r, ampicillin-resistant; APEC, Avian pathogenic E. coli.

Construction of the envZ Mutant and Complement Strains

The mutant strain AE17ΔenvZ was constructed from the wild-type (WT) AE17 using λ-red recombination techniques (Datsenko and Wanner, 2000). The primers used are listed in Table 2. Briefly, AE17 genomic DNA was used to amplify envZ upstream and downstream segments using envZ-up-F/R and envZ-down-F/R primers, respectively. The pKD3 plasmid was used to amplify a chloramphenicol resistance fragment with envZ-cm-F/R primers. Using the overlap PCR method and envZ-up-F and envZ-down-R primers, the 3 fragments (as templates) were used to amplify a fragment of envZ containing the chloramphenicol (Cm) resistance fragment. The fragment was then electroporated (200 Ω and 2,500 V) using a Gene Pulser MX cell (Bio-Rad, Hercules, CA) into AE17 competent cells containing the pKD46 plasmid. Then, the pCP20 plasmid was transformed into mutant strain AE17ΔenvZ cells to delete the Cm cassette. Finally, the mutant strain was identified using envZ-out-F/R and envZ-in-F/R primers and named AE17ΔenvZ.

Table 2.

Primers used in this study.

Primer name	Sequence (5′-3′)	Size/bp
AE17-F	GAGCGATCTTGGCTATACC	1,023
AE17-R	GTGAAGCTATCTAACAACGG
envZ-up-F	GCGGGCCACGCAACGAAAGTTAT	810
envZ-up-R	CCAGCCTACAACCATTAGCGGCACGGGATTAGGGC
envZ-cm-F	CGCTAATGGTTGTAGGCTGGAGCTGCTT	1,033
envZ-cm-R	TGAACTTCGCCATATGAATATCCTCCTTAG
envZ-down-F	TATTCATATGGCGAAGTTCAGCACCACCAGATA	736
envZ-down-R	CGAAGCGTCGCTAATGCAGAACA
envZ-in-F	TCCCGCTCATAGCCACTT	499
envZ-in-R	TTCATCAGGGCGATTTCTC
envZ-out-F	ATCAAACCCAGTACCACTTCCTCT	2,102
envZ-out-R	GCTCTCCCGCGATAAGCT
envZ-BamHⅠ-F	CGCGGATCCATGAGGCGATTGCGCTTCTC	1,353
envZ-PstⅠ-R	AAACTGCAGTTACCCTTCTTTTGTCATGC
pKD46-F	CATACCGTCCGTTCTTTCCTT	888
pKD46-R	TGATGATACCGCTGCCTTACT
pCP20-F	TTAGTGGTTGTAAAAACACCTGACC	409
pCP20-R	GTTAGCGTTGAAGAATTTAGCCC
pKD3-F	TGTAGGCTGGAGCTGCTT	1,013
pKD3-R	CATATGAATATCCTCCTTAGTTC
pSTV28-F	CAGGAAACAGCTATGAC	102
pSTV28-R	GTTTTCCCAGTCACGAC

Restriction sites are indicated by bolded and underlined nucleotides.

For the complemented strain, the envZ fragment from AE17 was amplified using envZ-BamHI-F/envZ-PstⅠ-R primers. The pSTV28 plasmid and envZ fragment were digested with restriction endonucleases BamHⅠ and PstⅠ (TaKaRa, Dalian, Liaoning, China) at 37°C for 1 h, ligated with T4 ligase (TaKaRa) at 4°C for 12 h, and transformed into E. coli DH5α cells (TaKaRa). The recombinant plasmid pSTV28-envZ-DH5α was constructed and electroporated into AE17ΔenvZ cells, confirmed using envZ-BamHI-F/envZ-PstⅠ-R primers, and named AE17C-envZ.

Growth Curves

Growth curves of AE17, AE17ΔenvZ, and AE17C-envZ were monitored as described previously (Fu et al., 2022). Briefly, all 3 strains were cultured to the logarithmic growth phase and transferred into LB medium at a 1:100 volume ratio. The cell density was monitored hourly using a spectrophotometer (Bio-Rad), and the growth curves of strains were plotted.

Biofilm Formation Assay

The ability of all 3 strains to generate biofilms was determined using crystal violet (CV) assays as described previously (Yu et al., 2020). The strains were separately cultured to a logarithmic growth phase, diluted in LB liquid medium at a 1:50 volume ratio, and added to the glass test tubes and 96-well polypropylene plates, with 3 replicates/group. Test tubes and plates were incubated at 28°C for 48 h and the supernatant was discarded. Biofilms were washed 3 times in PBS. After air-drying, 0.1% (w/v) crystal violet (Sangon) was added, biofilms stained for 20 min, and washed 3 times in PBS. After air-drying, 33% (w/v) acetic acid (Sangon) was added to wells to fully dissolve biofilms which were measured at OD_492nm using a MicroELISA auto reader (Thermo Scientific, Pittsburgh, PA).

Scanning Electron Microscopy

Bacterial biofilms were visualized by scanning electron microscopy (SEM) as described (Yin et al., 2019). Briefly, bacteria cultures were added to 6-well plates and incubated with crawls for 48 h at 28°C. Crawls were removed, washed 3 times in PBS, fixed in 2.5% (w/v) glutaraldehyde at 4°C for 2 h, washed in PBS, fixed in 3% (w/v) glutaraldehyde at 4°C for 6 h, and washed 3 times in PBS for 20 min. The crawls were treated with 30%, 50%, 70%, 80%, 95%, and 100% (w/v) anhydrous ethanol at 4°C for 20 min. Ethanol solutions were discarded and crawls were treated in acetone at 4°C for 20 min. Then, crawls were freeze-dried using a Critical Point Drier (Emitech, Ashford, UK) and microscopic biofilm morphology was observed using a Hitachi S-4800 SEM (Hitachi, Tokyo, Japan).

Rdar Colony Morphotype Assay

Bacterial rdar morphotype was performed as described previously (Li et al., 2019). Briefly, all strains were cultured in LB medium to logarithmic growth phase, centrifuged at 5,000 × g for 10 min at 4°C, and washed twice in PBS to adjust the OD_600nm to 3.0. Then, 2 μL of each culture was plated onto a Congo red agar plate and incubated at 28°C for 3 d. Recipe for Congo Red medium: 10 g peptone, 5 g yeast powder, 8 g agar powder, 0.04 g Congo Red (Sangon) and 0.02 g Comas Brilliant Blue (Sangon), dissolved in 1 L of water.

Bacterial Resistance to Environmental Stress Assay

The resistance to diverse environmental stress assay was determined as described (Han et al., 2014). Briefly, strains were cultured to logarithmic phase, centrifuged at 5,000 × g, washed twice in PBS, and cells resuspended in an equal volume of PBS. For the acid assay, 100 μL bacteria and 900 μL LB medium (pH 3) were incubated at 37°C for 30 min; for the alkali assay, 100 μL bacteria and 900 μL Tris-HCl (100 mmol/L, pH 10) were incubated at 37°C for 30 min; for the oxidative stress assay, 100 μL bacterial and an equal volume of 10 mM hydrogen peroxide were incubated at 37°C for 1 h; and for the osmotic pressure assay, 100 μL bacteria and an equal volume of NaCl (4.8 mol/L) were incubated at 37°C for 1 h. Surviving bacteria were counted on agar plates using a 10-fold gradient dilution approach.

Serum Bactericidal Assay

The serum bactericidal assay was performed as described previously (Zhang et al., 2021). The 3 strains were cultured to the logarithmic growth phase, centrifuged at 5,000 × g for 10 min, washed 3 times and resuspended in PBS. Specific pathogen-free (SPF) chicken serum (Beyotime, Shanghai, China) was diluted with PBS to 0%, 12.5%, 25%, 50%, or 100%. Heat-inactivated serum was used as a negative control. Then, a 10 µL bacterial culture was inoculated into 190 µL of different serum concentrations, mixed, and incubated at 37°C for 30 min. Cultures were diluted in a 10-fold gradient, inoculated onto LB plates, incubated and counted for the number of surviving bacteria.

Bacterial Adherence Assay

The bacterial adherence assay was performed as described previously (Tu et al., 2021). The 3 strains were grown to a logarithmic phase, centrifuged, washed in PBS, and resuspended with Dulbecco's modified Eagle's medium (DMEM) (Biological Industries, Kibbutz Beit Haemek, Israel). Chicken embryonic fibroblast DF-1 cells were co-cultured with bacteria for 1.5 h at 37°C under 5% CO₂ according to a multiplicity of infection (MOI) of 100, and the cells were washed 3 times with PBS to remove the nonadherent bacteria and cells were lysed in 0.5% TritonX-100 (Beyotime). Lysates were diluted in a 10-fold gradient to count bacteria on LB agar plates.

Chick Infections

The Institutional Animal Care and Use Committee (IACUC) guidelines of Anhui Agricultural University (Number: 2021-021) were followed during chicken studies. One-day-old Romany broiler chicks were purchased from Anhui Anqin Poultry Company Ltd. (Anhui Province, Hefei, China). The chicks live in a clean and tidy environment at around 35°C and can freely eat food and drink water. There were 8 chicks in each group. The strains were cultured to the logarithmic growth phase (10⁸ colony forming units (CFU)/mL), washed with PBS, and diluted to 10⁵ CFU/mL (Yi et al., 2019; Fu et al., 2022). After a week of feeding, the chicks were injected intramuscularly into the legs with 10⁵ CFUs of the bacterial solution, and negative controls were injected with an equal volume of PBS. Chick morbidity and mortality were observed daily for 7 d, and a survival curve was drawn to compare strain virulence.

To determine the effect of envZ on APEC colonization in vivo, the systemic infection experiment of the chick model was conducted as described previously (Zhang et al., 2021). Briefly, the wild-type strain AE17 and mutant strain AE17ΔenvZ and complemented strain AE17C-envZ were cultured to the logarithmic growth phase, washed and resuspended in PBS. Romany chickens were infected intramuscularly with a bacterial suspension containing 10⁶ CFUs and euthanized humanely at 24 h postinfection. And chickens were euthanized, dissected, and disinfected with alcohol on the surface of the organs, scissors and forceps were overcooked and the heart, liver, spleen, and lung were collected under aseptic conditions, weighed and homogenized. Serial dilutions of the homogenates were plated onto LB plates to count the bacterial numbers, and the ability of these APEC strains to colonize the liver, spleen, and blood was compared.

Library Preparation and Transcriptome Sequencing

RNA sequencing and library construction were performed at the Novogene Co., Ltd, Shenzhen, China. Total RNA was extracted from AE17 and AE17ΔenvZ strains using the RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions and RNA integrity was assessed using the RNA Nano 6000 Assay Kit of the Bioanalyzer 2100 system (Agilent Technologies, Santa Clara, CA). A total amount of 3 μg RNA per sample was used as input material for the RNA sample preparations. Sequencing libraries were generated using NEBNext Ultra Directional RNA Library Prep Kit for Illumina (NEB, Ipswich, MA), and index codes were added to attribute sequences to samples. Library fragments were purified with AMPure XP system (Beckman Coulter, Brea, CA) and library quality was assessed on the Agilent Bioanalyzer 2100 system (Alegent, Santa Clara, CA). Library preparations were sequenced on the Illumina Hiseq platform (Illumina, San Diego, CA) and paired-end reads were generated. The raw data (raw reads) of fastq format are removed from reads containing adapter, reads containing ploy-N and low-quality reads to obtain clean data (clean reads). All downstream analyses are performed based on high-quality clean data. Sequence reads were mapped onto E. coli APEC O2-211 complete genome (NZ_CP006834.2) and filtered sequences were analyzed using Bowtie2-2.2.3 (Langmead et al., 2009). The number of reads mapped to each gene was counted with HTSeq v0.6.1 (Trapnell et al., 2009). The reads per kb of fragments per kilobase of exon per million fragments (FPKM) for each gene were calculated based on the length of the gene and reads count mapped to this gene. Differential expression analysis was performed using the DESeq R package (Nadler et al., 1997). Genes with an adjusted P value <0.05 identified by DESeq were assigned as differentially expressed genes (DEGs). Gene ontology (GO) enrichment analysis of DEGs was performed using the Goseq R package. We used KOBAS software to test for the statistical enrichment of DEGs in Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways (Mao et al., 2005). Transcriptome sequencing reads were deposited at the National Center for Biotechnology Information gene expression database (https://www.ncbi.nlm.nih.gov) under the Sequence Read Archive (SRA) accession number: SUB94498265.

Quantitative Real-Time Reverse Transcription PCR

Quantitative real-time reverse transcription PCR (RT-qPCR) was used to assess the regulatory role of envZ in APEC on gene expression, including biofilm formation (ompC, ompT, mlrA, and basR) and stress response genes (hdeA, hdeB, adiY, and uspB). Additionally, to further analyze the mRNA levels of inflammatory cytokines (interleukin-1β (IL-1β), IL6, and IL8), DF-1 cells were infected with AE17, AE17ΔenvZ, and AE17C-envZ strains. The gene primers are listed in Table 3. Transcription levels were normalized to the housekeeping genes dnaE and β-actin. Total RNA was extracted using the EasyPure RNA kit (TransGen, Beijing, China) and reverse transcribed to cDNA using the EasyScript One-Step gDNA Removal and cDNA Synthesis SuperMix kit (TransGen) according to the manufacturer's instructions. cDNA was performed under the following conditions: 25°C for 10 min, 42°C for 15 min, and 85°C for 5 s. The RT-qPCR was performed using the TransStart Tip Green qPCR SuperMix kit (TransGen) on a 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA). RT-qPCR was performed under the following conditions: 94°C for 30 s; 40 cycles at 94°C for 5 s, at 60°C for 15 s, and at 72°C for 10 s; and 94°C for 15 s, 60°C for 60 s, and 94°C for 15 s. The specific amplification of each gene was confirmed using melting curve analysis and the relative fold change was calculated using the 2^−ΔΔCt method (Livak and Schmittgen, 2001).

Table 3.

qRT-PCR primers used in this study.

Primer name	Sequence (5′-3′)	Size/bp
dnaE-F	ATGTCGGAGGCGTAAGGCT	181
dnaE-R	TCCAGGGCGTCAGTAAACAA
ompC-F	GAGGTTGTTAGCGTCGTATT	159
ompC-R	TTGGCGGCATCTATCACTT
ompT-F	AGCCAGCAGCCACGATAGA	115
ompT-R	GAAGAAGGAGGCCGAAAAGTC
mlrA-F	ATTACCCCGCCCAGACACT	140
mlrA-R	CGAAGCCAGGCAAATAGAGA
mcbR-F	GGCAGTTACTCTGGCTGTT	104
mcbR-R	GTTCCATGTCGCCCTTT
adiY-F	AATATGAGCAACCAAAGTGT	118
adiY-R	GATAAGGGTGAATGGGAAAG
hdeB-F	CTGGTGTCATTGCTTTTGGA	134
hdeB-R	AGCGTTTATTTTTATGGGCG
hdeA-F	GCCTGAACGATAGCTGGGGT	136
hdeA-R	ATTTCCTGGCTGTGGACGAA
uspB-F	GACCACCACCAGACCACACA	157
uspB-R	TACCTCACATGGCCAACCC
β-action-F	TGAACTCCCTGATGGTCAGGTC	100
β-action-R	ACCACAGGACTCCATACCCAAG
IL-1β-F	CAATAAATACCTCCACCCCG	106
IL-1β-R	ATGTTTCATTACCGTCCCGT
IL6-F	AGACTTTACTTCGTTTATGG	166
IL6-R	ATTGTTACTCTCTGGTTCCT
IL8-F	CATCCACCCTAAATCCATTCAAG	141
IL8-R	GGCCATAAGTGCCTTTACGATC

Statistical Analyses

These assay data were analyzed in GraphPad Prism 7.0 software package (GraphPad Software, LaJolla, CA). Statistical comparisons between groups were made using paired t tests. Two-way ANOVA was used to analyze RT-qPCR results. A P < 0.05 value was considered statistically significant.

RESULTS

EnvZ Did Not Affect APEC Growth

The mutant AE17ΔenvZ and the complemented strain AE17C-envZ were successfully constructed using λ-red recombination techniques (Figures 1A and 1B). AE17, AE17ΔenvZ, and AE17C-envZ growth curves showed no significant differences, suggesting EnvZ did not affect APEC growth (Figure 1C).

Figure 1.

Identification and characteristics of the envZ gene mutant and complemented strains. (A) The schematic diagram of the strategy for deleting the envZ gene in AE17; (B) confirmation of the mutant strain AE17ΔenvZ and the complemented strain AE17C-envZ. M: 5000 DNA marker; Lane 1: PCR product (2 012 bp) amplified from AE17 with primers envZ-out-F/R; Lane 2: PCR product (976 bp) amplified from AE17ΔenvZ with primers envZ-out-F/R; Lane 3 and 5: PCR product (499 bp) amplified from AE17 with primers envZ-in-F/R; Lane 4: PCR product (0 bp) amplified from AE17ΔenvZ with primers envZ-in-F/R; Line 6: PCR product (499 bp) amplified from AE17C-envZ with primers envZ-in-F/R; Lane 7–9: negative control. (C) Growth curves of AE17, AE17ΔenvZ, and AE17C-envZ strains in LB medium at 37°C, and the OD_600nm was measured every hour. The mutant strain shows a similar growth rate with the wild-type and complemented strains.

Deletion of the EnvZ Gene Down-Regulated APEC Biofilm Formation

The biofilm formation abilities were determined using CV and SEM assays. AE17ΔenvZ biofilm formation was significantly lower when compared with the AE17 strain, whereas the complemented strain AE17C-envZ was restored to a level similar to that of the wild-type strain (Figures 2A and 2B). SEM showed that AE17 biofilms were thicker with tight connections between bacteria, whereas AE17ΔenvZ biofilms were thinner and connections were relatively loose, and complemented strain AE17C-envZ restored biofilm formation capacity (Figure 2C). These results indicate that EnvZ exerts a negative role in APEC biofilm formation.

Figure 2.

The ability of 3 strains to form biofilms was tested by crystal violet (CV) and scanning electron microscope (SEM) assays. (A, B) Analyze the biofilms ability of AE17, AE17∆envZ, and AE17C-envZ strains using CV assay (*P < 0.05, **P < 0.01). The biofilm formation of the mutant strain AE17∆envZ was reduced compared with the wild-type strain. (C) (a) The morphological observation of AE17 (× 10,000); (b) the morphological observation of AE17ΔenvZ (× 10,000); (c) the morphological observation of AE17C-envZ (× 10,000).

EnvZ Influences the APEC Morphotype

We investigated the colony morphology of the 3 strains on Congo Red plates at 28°C for 3 d (Figure 3). The AE17 strain displayed a rdar morphology indicating curli and cellulose production. The envZ mutant strain displayed a saw (smooth and white) morphology, indicating a lack of both curli and cellulose production. The complemented AE17C-envZ strain also displayed a rdar morphology. These results suggested envZ deletion attenuated the APEC rdar morphotype.

Figure 3.

Bacterial display altered colony morphotypes. The AE17, AE17∆envZ, and AE17C-envZ strains were investigated on Congo Red agar plate at 28°C for 3 d. (A) AE17 displays the rdar morphotype; (B) AE17∆envZ results in the saw morphotype; (C) AE17C-envZ also displays the rdar morphotype.

Deletion of the EnvZ Gene Decreased APEC Environmental Stress Resistance

EnvZ functions as a histidine kinase to monitor and transduce osmolarity, so we evaluated its effects on an APEC strain's ability to adapt to different environments. Under acid, alkali, osmotic pressure, and oxidative stress conditions, when compared with the AE17, the survival rate of AE17ΔenvZ was significantly reduced, whereas AE17C-envZ survival was restored to WT levels (Figure 4). Thus, EnvZ may help APEC strains adapt to extraintestinal environments and survive stressful conditions.

Figure 4.

Bacterial resistance to environmental stress. The survival ability of AE17, AE17∆envZ, and AE17C-envZ strains were tested under diverse environmental stress. (A) Acid (PH 3); (B) alkali (PH 10); (C) oxidative (10 μM H₂O₂); and (D) osmotic (4.8 mol/L NaCl) (*P < 0.05, ^⁎⁎P < 0.01, and ^⁎⁎⁎P < 0.001). Mutant strain shows decreased survival under diverse environmental stress conditions compared with wild-type and complemented strains.

Deletion of the EnvZ Gene Decreased Bacterial Serum Resistance

The AE17, AE17ΔenvZ, and AE17C-envZ strains were separately incubated with serum at concentrations of 0%, 12.5%, 25%, 50%, or 100%. When compared with the wild-type AE17, the AE17ΔenvZ mutant strain was significantly susceptible to SPF chicken serum, whereas the AE17C-envZ strain survival was not restored (Figure 5). These results indicated EnvZ was involved in APEC's serum resistance.

Figure 5.

Serum bactericidal assay. SPF chicken serum at concentrations of 0%, 12.5%, 25%, 50%, 100%, and heat-inactivated. The AE17, AE17∆envZ, and AE17C-envZ strains were incubated with different dilutions of SPF chicken serum (*P < 0.05, **P < 0.01). The mutant strain shows decreased survival in SPF chicken serum compared with the wild-type strain.

EnvZ Influences Adherence and Inflammatory Responses Expression in DF-1 Cells

The role of envZ in bacterial adhesion was investigated by infecting DF-1 chicken fibroblast cells assay. The ability of AE17ΔenvZ to adhere to DF-1 cells was significantly enhanced when compared with the AE17, whereas the adherence to DF-1 cells of AE17C-envZ was lower compared with that of mutant strain and restored to a level to that of the wild-type strain (Figure 6A). These results indicate that EnvZ plays an important role in APEC adherence to DF-1 cells. And the relative expression levels of IL-1β, IL-6, and IL-8 were significantly increased in AE17ΔenvZ after infecting cells (Figure 6B).

Figure 6.

Bacterial adherence and inflammatory cytokines to DF-1 cells. (A) The DF-1 cells were infected with APEC strains for 1.5 h and the ability of AE17, AE17ΔenvZ, and AE17C-envZ strains to adhere to DF-1 cells was counted. (B) The relative expression levels of the inflammatory factors IL-1β, IL-6, and IL-8 were tested by qRT-PCR (*P < 0.05, **P < 0.01, ***P < 0.001).

Deletion of EnvZ Gene Attenuates APEC Virulence and Colonization In Vivo

To explore how envZ affected APEC virulence in vivo, we compared the pathogenicity of AE17, AE17ΔenvZ, and AE17C-envZ strains in a 7-day-old chick systemic infection model. The mortality rate of the AE17 strain was 100% (8/8 chicks), AE17ΔenvZ strain was 62.5% (5/8 chicks), and AE17 C-envZ strain was 87.5% (7/8 chicks) (Figure 7A). Chicks infected with AE17ΔenvZ died later than those infected with AE17 and AE17C-envZ. The virulence of the mutant strain was attenuated compared to the wild-type and complemented strains. Thus, EnvZ was required for APEC virulence in vivo.

Figure 7.

EnvZ is required for the virulence and colonization of APEC during systemic infection. (A) Determination of bacterial virulence. Seven-day-old chickens were injected intramuscularly with AE17, AE17ΔenvZ and AE17C-envZ at 10⁵CFU. Negative controls were injected with PBS. Survival was monitored until 7 d postinfection. (B) Colonization of APEC strains in chickens during systemic infection. Seven-day-old chickens were infected intramuscularly with 10⁶CFU of each APEC strain. After 24 h postinfection, the mutant strain had significantly lower bacterial loads in the heart, liver, spleen, and lung compared to the wild-type and complemented strains. A nonparametric Mann–Whitney U test was conducted to determine statistical significance (*P < 0.05, **P < 0.01).

To further investigate whether envZ could influence APEC colonization during systemic infection, the number of bacteria in the heart, liver, spleen, and lung tissue of chicks was measured 24 h after postinfection. When inoculated intramuscularly, the amount of mutant strain was significantly reduced in the heart, liver, spleen, and lung compared to the wild-type strain. And the number of complement strains in these tissues recovered to levels similar to those of the wild-type strain (Figure 7B). These results suggest that EnvZ is necessary for effective APEC colonization during systemic infection in vivo.

Identification of DEGs

To further understand how EnvZ contributed to APEC biological processes and pathogenicity, we compared gene transcript level changes in AE17 and envZ mutant strains by RNA-Seq. Clean reads from AE17 and AE17ΔenvZ were compared to the APEC O2-211 genome; the comparisons were 89.92% and 88.44%, respectively. According to the criteria of P value ≤ 0.05, |log2 (fold change) |≥1, 711 DEGs were significantly regulated, of which 303 genes were significantly up-regulated and 408 genes were significantly down-regulated in the envZ mutant strain when compared to the AE17 (Figure 8A). These genes were mainly involved in biofilm formation, TCSs, Iron–Sulfur (Fe–S), phosphotransferase system (PTS), and bacterial flagella (Table 4).

Figure 8.

(A) Differential gene volcano plots. Volcano plots generated by EdgeR showing transcript expression profile in the comparison performed: AE17ΔenvZ and AE17 (AEW2 vs A). Significantly differentially expressed genes (DEGs) were defined as those with at least |log2 (FoldChange)|≥1 (level of magnitude, vertical lines) and q-value ≤ 0.05 (level of statistical significance, horizontal line). DEGs were illustrated in red (increased expression), green (decreased expression), and blue (not significant). (B) KEGG pathway classification analysis of DEGs. KOBAS software was used for KEGG pathway enrichment analyses. The abscissa shows the rich factor of DEGs and the ordinate indicates the top 20 of KEGG pathway enrichment for DEGs.

Table 4.

Partial differentially expressed genes (DEGs) in the envZ-deficient mutant strain.

Classification	Gene ID	Gene	Description	log2FC
TCS	b2370	evgS	Two-component system, NarL family, sensor histidine kinase EvgS	1.85
	b2369	evgA	Two-component system, NarL family, response regulator EvgA	1.61
	b4113	basR	BasR/S two-component system, OmpR family, response regulator BasR	−5.18
	b1235	rssB	Two-component system, response regulator RssB	−1.27
Stress response	b1296	yeeF	Predicted amino acid transporter	−4.76
	b3494	uspB	Universal stress protein UspB	−2.97
	b1051	msy	Acidic protein MsyB	−2.8
	b4116	adiY	AraC family transcription regulator protein AdiY, acid stress	−2.19
	b3509	hdeB	Acid stress chaperone HdeB	−2.17
	b3510	hdeA	Stress response protein; acid-resistance protein; acid stress chaperone HdeA	−1.29
Outer membrane	b1024	pgaA	Biofilm PGA synthesis protein YcdS; predicted outer membrane protein	4.55
	b1023	ycdR	Predicted enzyme associated with biofilm formation	4.45
	b0488	ybbJ	Inner membrane protein	2.02
	b2605	yfiB	Putative outer membrane lipoprotein	1.63
	b0929	ompF	Outer membrane pore protein OmpF	1.59
	b2215	ompC	Outer membrane porin protein OmpC	−5.3
	b4149	blc	Outer membrane lipoprotein Blc precursor	−4.72
	b4496	nmpC	Outer membrane porin protein NmpC	−3.82
	b3522	yhjD	Membrane protein; conserved hypothetical protein	−3.01
	b1285	yciR	c-di-GMP phosphodiesterase	−2.57
	b2013	yeeE	Inner membrane protein	−2.45
	b3506	slp	Outer membrane lipoprotein	−2.21
	b0565	ompT	Outer membrane protein 3b (a), protease VII	−2.16
	b0112	aroP	Aromatic amino acid transport protein AroP	−1.96
	b3688	yidQ	Outer membrane protein	−1.80
Fe–S cluster	b4291	fecA	Iron complex outer membrane receptor protein, putative iron compound receptor	7.37
	b1451	yncD	Iron complex outer membrane receptor protein	3.24
	b1684	sufA	Fe–S cluster assembly protein SufA	−4.45
	b0150	fhuA	Iron complex outer membrane receptor protein; ferrichrome outer membrane transporter	−2.81
	b0150	fhuA	Iron complex outer membrane receptor protein	−2.81
	b1683	sufB	Fe–S cluster assembly protein SufB; component of SufBCD complex	−2.47
	b1682	sufC	Fe–S cluster assembly ATP-binding protein; component of SufBCD complex, ATP-binding component of ABC superfamily	−2.46
	b1681	sufD	Fe–S cluster scaffold complex subunit SufD	−2.28
	b1680	sufS	L-cysteine desulfurase	−1.84
	b2531	iscR	Rrf2 family transcriptional regulator, iron–sulfur cluster assembly transcription factor; DNA-binding transcriptional repressor	−1.19
Flagellar	b1892	flhD	Flagellar transcriptional activator FlhD	−1.9
	b1891	flhC	Flagellar transcriptional activator FlhC	−1.3
	b1921	fliZ	Regulator of sigma S factor FliZ	−1.27
	b1940	fliH	Flagellar assembly protein FliH	−1.26
	b1922	fliA	RNA polymerase sigma 28 (sigma F) factor for flagellar operon FliA	−1.19
LPS	b0641	rlpB	LPS-assembly lipoprotein; minor lipoprotein	1.65
	b0054	imp	LPS-assembly protein; organic solvent tolerance protein precursor	1.27
	b3627	rafI	LPS alpha-1,3-glucosyl transferase	−6.72
	b3626	rfaJ	Lipopolysaccharide 1,2-glucosyltransferase	−6.40
	b3631	rfaG	Lipopolysaccharide core biosynthesis glucosyltransferase	−4.98
	b3622	waaL	O-antigen ligase; lipid A-core, surface polymer ligase	−4.76
	b3632	rfaQ	Lipopolysaccharide core biosynthesis glycosyltransferase	−4.59
Drug resistance	b2685	emrA	Multidrug efflux pump membrane fusion protein EmrA	1.31
	b2519	pbpC	Penicillin-binding protein	1.55
	b3028	mdaB	The modulator of drug activity; NADPH quinone reductase	1.39
	b4337	mdtM	Multidrug efflux pump/bile salt:H (+) antiporter/Na (+):H (+) antiporter/K (+):H (+) antiporter	−5.15
	b0543	emrE	Multidrug/betaine/choline efflux transporter EmrE	−2.61
PTS system	b2703	srlE	PTS system, glucitol/sorbitol-specific IIBC component	2.21
	b3947	ptsA	Putative PTS multi-phosphoryl transfer protein PtsA	2.20
	b2704	srlB	PTS system, glucitol/sorbitol-specific IIA component	2.00
	b2702	srlA	PTS system, glucitol/sorbitol-specific IIC2 component	1.19
	b2167	fruA	PTS system, fructose-specific IIBC component	−1.93
	b1737	celB	PTS system, cellobiose-specific IIC component	−1.59
	b3599	mtlA	Fused mannitol-specific PTS enzymes: IIA components/IIB components/IIC components	−1.61
	b1738	celA	PTS system, cellobiose-specific IIB component; PEP-dependent phosphotransferase enzyme IV	−1.65

In total, 269 DEGs were annotated in the GO database, and 1,163 GO functional annotations were generated. DEGs were involved in 2 main GO categories: biological processes and molecular functions. Most DEGs participated in binding, ion binding, small molecule binding, biological regulation, and organic cyclic compound binding (Figure S1; Table S1). KEGG annotations identified 661 DEGs, involving 78 signaling pathways, including metabolic pathways, microbial metabolism in diverse environments, biosynthesis of amino acids, sulfur metabolism, TCSs, flagellar assembly, and biofilm formation (Figures 8B and S2; Table S2).

Deletion of EnvZ Gene Down-Regulated the Expression of Biofilms and Stress Response Genes

To validate RNA-Seq results, RT-qPCR studies were undertaken to examine the transcript levels of biofilm formation-associated genes (ompC, ompT, mlrA, and basR) and stress response genes (hdeA, hdeB, adiY, and uspB). OmpC, the pore protein of the outer membrane of E. coli, contributes to Salmonella typhimurium and APEC pathogenicity (Chatfield et al., 1991; Hejair et al., 2017b). Outer membrane protein OmpT is a protease in E. coli and plays important role in APEC pathogenicity (Hejair et al., 2017a). MlrA is a MerR family transcriptional regulator and a key regulator of biofilm formation (Ogasawara et al., 2010; Lindenberg et al., 2013). BasR is a RR of BasSR, with BasSR having roles in APEC biofilms and virulence (Yu et al., 2020). The hdeA and hdeB genes encode the acid-activated periplasmic chaperones HdeA and HdeB, respectively, to protect the E. coli periplasmic proteome against rapid acid-mediated damage to the mammalian stomach (Dahl et al., 2015). AdiY is an AraC family transcription regulator protein and regulates acid stress in some strains (Krin et al., 2010). The universal stress protein UspB is an integral membrane protein and is required for resistance to different starvation and stress conditions (Farewell et al., 1998). The transcript levels of these genes were significantly reduced in the AE17ΔenvZ strain when compared with the AE17 strain and were restored in the AE17C-envZ strain (Figure 9). Therefore, EnvZ executed a positive regulation on the expression of biofilms and stress response genes.

Figure 9.

Relative expression of virulence genes was tested with qRT-PCR. These genes in the RNA-Seq results were selected for qRT-PCR verification. Data were normalized to the housekeeping gene dnaE. The X-axis represents different genes. The relative gene expression levels were calculated using the 2^−△△Ct method (**P < 0.01).

DISCUSSION

OmpR/EnvZ TCS is widely present in most Gram-negative or positive bacteria, detects and responds to extracellular osmolarity, and involves in bacterial pathogenesis, which can alter the expression of a variety of genes, including those encoding virulence factors, antibiotic biosynthesis, motility, and biofilm formation (Bernardini et al., 1990; Xi et al., 2020). EnvZ is an HK of the OmpR/EnvZ TCS and has a key role in transducing osmotic signals in E. coli K12 (Gotoh et al., 2010; Cai et al., 2013; Breland et al., 2017) and contributes to S. flexneri and V. cholerae pathogenicity (Bernardini et al., 1990; Xi et al., 2020), but the role of its histidine kinase EnvZ in APEC's pathogenic process has not been reported. In this study, we demonstrated that EnvZ had central roles in APEC, including biofilm formation, rdar morphology, responses to environmental stresses, serum resistance, adhesion, immune system, and APEC virulence in vivo, suggesting that EnvZ has an important impact on the pathogenicity of APEC. And the transcriptome sequencing revealed that EnvZ is involved in metabolic pathways, microbial metabolism in diverse environments, biosynthesis of amino acids, sulfur metabolism, TCSs, flagellar assembly, and biofilm formation.

In this study, EnvZ was also involved in the response to various external environmental changes in APEC. In the process of infecting the host, bacteria will encounter some extreme environments, especially strong acids, strong alkalis, high osmolarity, oxidative stress, etc. Some genes have been reported to be critical for APEC's resistance to environmental stress, such as ibeR, ybjX, and mcbR (Wang et al., 2015; Yu et al., 2019; Song et al., 2020). Understanding the resistance of bacteria to environmental stress helps further explore the mechanisms of bacterial survival and infection. In E. coli K12, EnvZ interacts with the inner membrane protein MzrA and responds to environmental signals (Gerken et al., 2009). In this study, transcriptome sequencing data showed EnvZ also regulated the transcript levels of outer membrane protein genes, such as ompC and ompT. Moreover, transcriptome data also showed that EnvZ regulated the expression levels of stress-resistance genes, such as hdeA and hdeB. HdeA and HdeB are reported to involve in the acid stress response in E. coli and protect the E. coli periplasmic proteome against rapid acid-mediated damage to the mammalian stomach (Dahl et al., 2015). Together, these results suggest that envZ decreased the resistance to environmental stress by the down-regulation of outer membrane proteins and stress response genes.

Biofilm formation is a common phenomenon in APEC infection with hosts, which facilitates resistance to antibiotics and disrupts host cell immune mechanisms to achieve sustained colonization in the host, so biofilms often cause chronic diseases (Skyberg et al., 2007). Many TCSs and their components contribute to APEC biofilm formation, such as PhoP/Q, CpxA, and OmpR (Matter et al., 2018; Yin et al., 2019; Fu et al., 2022). In this study, deletion of the histidine kinase envZ gene reduced biofilm formation, and also weakened curli and cellulose production in APEC. It is worth noting that curli fimbriates have important roles in early biofilm formation stages (Prigent-Combaret et al., 2000; Ogasawara et al., 2010). And the transcriptome sequencing showed that EnvZ affects the biofilm formation signaling pathway, and many biofilms and curli fimbriae-associated genes were down-regulated in the envZ-deficient mutant strain, including ompC, ompT, mlrA, and basR. In E. coli MG1665, the OmpR/EnvZ TCS was shown to affect curli production (Prigent-Combaret et al., 2001; Jubelin et al., 2005). MerR family transcriptional regulator MlrA contributes to biofilm formation and interacts with ydaM to regulate the expression of the curli fimbriae regulator CsgD in E. coli K12 (Ogasawara et al., 2010; Kojima and Nikaido, 2014). Our study supports that the biofilm formation and the production of curli fimbriae connect, but the regulation mechanism of EnvZ requires further study.

Innate immunity is the first line of defense against infection and tissue damage. Pattern recognition receptors, such as Toll-like receptors, recognize pathogens such as bacteria, trigger an immune response and activate intracellular signaling pathways leading to antimicrobial activity and the production of inflammatory cytokines, triggering a range of inflammatory responses (Ciornei et al., 2010; Wu et al., 2018; Chambers and Scheck, 2020). In Yersinia pestis, the OmpR/EnvZ TCS participates in the host's innate immune processes (Reboul et al., 2014). In this study, EnvZ is necessary for stimulating the host production of inflammatory cytokines during APEC systemic infection in vivo. In S. flexneri, the inactivation of envZ gene attenuated the invasion of the epithelial cell (Bernardini et al., 1990). In different bacteria, EnvZ exhibits different effects on host cell adhesion, suggesting that it may also regulate other key factors that still need to be further explored.

APEC is one of the main causes of bloodstream infection, which initially colonizes poultry via the respiratory tract and then spreads systemically (Dziva and Stevens, 2008). Serum resistance is an important virulence trait of APEC, and many bacterial traits contribute to E. coli's serum resistance. It is indicated that many virulence factors contribute to the serum resistance of APEC, including the production of a capsule, a lipopolysaccharide (LPS) layer, and some outer membrane proteins (OMPs) (Li et al., 2011; Ma et al., 2018). In this study, we demonstrated that the inactivation of envZ led to more susceptibility to SPF chicken serum, which might be a reason for decreased expression of outer membrane genes due to envZ gene deletion. Studies have shown that the histidine kinase of TCS plays an importance in APEC virulence, including PhoQ, QseC, etc. (Chaudhari and Kariyawasam, 2014; Tu et al., 2016). Furthermore, EnvZ has also been required for Acinetobacter baumannii and V. cholerae virulence in some infection models (Tipton and Rather, 2017; Xi et al., 2020). In this study, the inactivation of envZ significantly attenuated APEC virulence in a chicken systemic infection model and decreased the bacterial colonization in the heart, liver, spleen, and lung tissues of chicks, indicating that EnvZ influences the bacterial survival and pathogenicity of APEC during systemic infection.

In conclusion, we demonstrated that EnvZ was essential for biological properties and virulence, including biofilm formation, curli production, stress responses, serum resistance, adhesion, and APEC pathogenicity. Transcriptomic studies showed EnvZ was involved in metabolic pathways, microbial metabolism in diverse environments, and biofilm formation. Generally, EnvZ mediates stress responses, and regulates and influences the expression and function of bacterial virulence factors involved in APEC pathogenesis.