Hypoxia-induced HIF-1α promotes Listeria monocytogenes invasion into tilapia

ABSTRACT HIF-1α is a nuclear transcription factor, and its activity is tightly regulated by the level of available oxygen in cells. Here, we investigated the roles of HIF-1α in the invasion of Listeria monocytogenes into tilapia under hypoxic environments. We found that the expression levels of HIF-1α in examined tissues of hypoxic tilapia were significantly upregulated, indicating that the tissue cells have been in hypoxic conditions. After 24-h infection with L. monocytogenes, we found that bacterial burden counts increased significantly in all examined tissues of hypoxic fish. To explore why the bacterial count increased significantly in the tissues of hypoxic fish, we modulated HIF-1α expression through RNAi technology. The results indicated that c-Met expression levels were positively related to HIF-1α expression. Since c-Met is the receptor of InlB that plays critical roles in the adhesion and invasion of L. monocytogenes, the ∆InlB strain was used to further explore the reason for the significant increase in bacterial counts in hypoxic fish. As expected, the decrease in the adhesion ability of ∆InlB suggested that InlB mediates L. monocytogenes infection in tilapia. After being infected with ∆InlB strain, we found that the bacterial counts in hypoxic fish were not affected by hypoxic conditions or HIF-1α expression levels. These findings indicate that HIF-1α may promote the internalization of InlB by upregulating c-Met expression and therefore contributes to the invasion of L. monocytogenes into hypoxic tilapia. IMPORTANCE Listeria monocytogenes is a zoonotic food-borne bacterial pathogen with a solid pathogenicity for humans. After ingestion of highly contaminated food, L. monocytogenes is able to cross the intestine invading phagocytic and nonphagocytic cells and causes listeriosis. China is the world’s largest supplier of tilapia. The contamination rate of L. monocytogenes to tilapia products was as high as 2.81%, causing a severe threat to public health. This study revealed the underlying regulatory mechanisms of HIF-1α in the invasion of L. monocytogenes into tilapia under hypoxic environments. This study will be helpful for better understanding the molecular mechanisms of hypoxic environments in L. monocytogenes infection to tilapia. More importantly, our data will provide novel insights into the prevention and control of this pathogen in aquaculture.

subunit to directly regulate the transcription of specific genes involved in responses to hypoxia.Respiratory, metabolic, and immune responses may be affected when HIF-1α is activated (5)(6)(7).
In 2017, a survey found that 2.81% of tilapia exports were contaminated with Listeria monocytogenes, which is far higher than the amount of contamination at Salmonella (0.20%), Vibrio parahaemolyticus (0.80%), and Vibrio cholerae (0.51%), which seriously threatened human public health security (8).L. monocytogenes is a zoonotic food-borne bacterial pathogen with a solid pathogenicity for humans (9).After ingestion of highly contaminated food, L. monocytogenes is able to cross the intestine invading phago cytic and non-phagocytic cells and causes listeriosis (10).L. monocytogenes is a typical intracellular pathogen, and the primary prerequisite for its pathogenicity to the host is that it can successfully invade the host cell.Internalin A (InlA) and InlB, as the critical virulence factors, play vital roles in the process of adhesion and invasion of L. monocyto genes (11).InlA is a unique surface protein of L. monocytogenes, of which the primary function is to mediate bacterial invasion into host cells and internalize into phagosomes (12).This process is the prerequisite for establishing infection between L. monocytogenes and the host.InlA mediates bacterial entry into host cells by specifically binding to E-cadherin (13)(14)(15).E-cadherin is a specific receptor on the surface of host cells, essential for maintaining tight and interstitium junctions (16).In the hypoxic regions of breast cancer, cervical cancer, gastric cancer, and other tumor cells, the abnormal expression of E-cadherin can lead to the abnormal regulation of the related signaling pathway, which will affect the adhesion, invasion, metastasis, and survival of cells, and the abnormal expression of E-cadherin is often closely related to the overexpression of HIF-1α in the hypoxic environment of tumors (17,18).In the internalization of L. monocytogenes, InlB also plays an irreplaceable role.Similar to InlA, InlB can specifically bind to cell surface hepatocyte growth factor receptor (c-Met) to mediate the invasion of L. monocytogenes into most host cells (19,20).Cell surface c-Met is a specific receptor for InlB and one of the downstream genes in the HIF-1α signaling pathway.A study of liver cancer cells showed that the hypoxia environment in tumors could promote the upregulation of the c-Met transcription level, and the upregulation process is regulated by HIF-1α, and the expression level of the two is positively correlated (21).
In the process of tilapia farming, it is unknown whether the low oxygen environment affects the infection ability of L. monocytogenes to tilapia, and whether the differential expression of HIF-1α mediates the infection of L. monocytogenes to tilapia by regulating E-cadherin and c-Met, which is the receptor for the InlA and InlB.Therefore, this study focused on assessing the roles of HIF-1α in the infection of hypoxic tilapia with L. monocytogenes.

The fish and bacterial strains
Tilapia (Oreochromis niloticus) juveniles were obtained from the Tilapia Breeding Farm of Guangdong Province (China) and transported to the laboratory.Before experiment initiation, all tested fish were confirmed to be L. monocytogenes free following bacterio logical examination.Approval was obtained from the Institutional Animal Care and Use Committee of Zhejiang Agriculture and Forestry University prior to using the animals for research.L. monocytogenes strains, including the wild-strain EGD-e and gene deletion strains of InlA and InlB (∆InlA and ∆InlB), were preserved in our laboratory.All strains were cultivated in Brain Heart Infusion (BHI, Oxoid), shaking at 180 rpm at 37°C.

Hypoxia exposure experiments and sampling
The healthy tilapia (mean weight = 12.0 ± 3.0 g) were reared in the tanks with a water temperature of 26 ± 0.5°C and dissolved oxygen (DO) concentration at 6.8 ± 0.5 mg/L (the normoxic group) for acclimation of 7 d.Then, a group of tilapia was exposed to tanks constantly aerated with nitrogen gas and oxygen, and the desired DO concentration was set to 2.0 ± 0.5 mg/L (the hypoxic group).In normoxic and hypoxic groups, half of the fish were administered intraperitoneally with 100 µL of EGD-e strain (1.37 × 10 4 CFU/fish), and the other half were administered intraperitoneally with 100 µL phosphate-buffered saline (PBS) as the control group.The DO level in each tank was measured every 4 h using a dissolved oxygen detector (JPB-70A, China).The nitrite level was <0.5 mg/L, and the pH was 6.5 ± 0.5 in all of the tanks.After hypoxia treatment for 24 h, livers, spleens, and muscles were collected from each group of tilapia and used for detecting HIF-1α and its important related gene expression, antioxidant enzyme activity, tissue section observation, and L. monocytogenes burden count.

Liver section observation
After 24-h infection with PBS or EGD-e strain, livers were collected from normoxic and hypoxic fish and cut into a number of 1 mm 3 slice and fixed in 4% glutaraldehyde for 2 d at 4℃.Fixed samples were routinely dehydrated in a graded acetone series (30-100%), infiltrated with acetone and resin mixture, polymerized in the oven at 60℃ for 48 h, and sectioned.Thin sections were mounted on grids, and post-stained with 2% uranyl acetate for 15 min and washed with distilled water before viewing the samples.The tissue sections were prepared in the Servicebio Biotechnology Co, Ltd. (Wuhan, China) and the samples were viewed using transmission electron microscopy (TEM; HITACHI HT7700, Japan).

Antioxidant enzyme activity
After 24-h infection with PBS or EGD-e strain, livers were collected from normoxic and hypoxic fish and then ground into pieces with aseptically frosted glass slides in PBS for enzyme activity analysis.Superoxide dismutase (SOD) and catalase (CAT) activities were measured using assay kits (Nanjing Jiancheng Institute, China), according to the manufacturer's instructions.

Quantitative real-time PCR
To investigate the changes in HIF-1α and downstream gene expression levels, livers, spleens, and muscles were collected from normoxic and hypoxic fish and used to detect HIF-1α, c-Met, and E-cadherin gene expression by qRT-PCR.Total RNA was isolated from livers, spleens, and muscles using the TRIzol reagent (Takara).qRT-PCR amplification with a single dissociation curve peak was performed in a LightCycler 480 Real-Time System (Roche, Switzerland).All the primers are listed in Table 1, and β-actin was chosen as an internal standard.The data were analyzed using the 2 −∆∆Ct method (22).

Knockdown of HIF-1α
The primers with specific restriction sites for HindIII and BamHI were designed based on the known nucleotide sequence of HIF-1α (GenBank: KY415998.1).PCR products digested with specific restriction sites were sub-cloned into LITMUS 38i vector (NEB, MA, USA) to gain the recombinant plasmid of L38i-HIF-1α and then verified by DNA sequencing.The recombinant plasmid was transformed into HT115 (DE3) cells, and cells were cultured and incubated according to the previous methods (23).After adding IPTG (0.8 mmol/L) to induce expression, bacterial solution with persistent expression of HIF-1α protein was subsequently harvested by centrifugation (4,000 rpm, 10 min) at 4°C.The double-stranded RNAs for HIF-1α (dsHIF-1α) were purified using a mirVana miRNA Isolation Kit (Ambion, USA).The dsEGPF was used as the negative control.The injection dose of dsHIF-1α (49 µg/fish) was determined by a preliminary experiment.After hypoxia and dsHIF-1α treatment for 24 h, the expression levels of HIF-1α in livers, spleens, and muscles were detected by qRT-PCR.If the expression level of HIF-1α was knocked down significantly at 24 h post-hypoxia and dsHIF-1α treatment, the mRNA expression levels of E-cadherin and c-Met were subsequently determined.

Growth curve of bacterial strains
To evaluate the growth curves, EGD-e, ∆InlA, and ∆InlB strains were cultured by inoculating into BHI broth and incubated in a shaker bath at 180 rpm at 37°C for 12 h.During the incubation period, the light transmittance of the bacterial solution was measured every 1 h by spectrophotometer, and the growth curves were made according to the OD 600 value.

Oxidative stress tolerance assay
To evaluate the antioxidant stress ability of EGD-e, ∆InlA, and ∆InlB strains, H 2 O 2 was used as a direct oxidant.EGD-e, ∆InlA, and ∆InlB strains were cultured overnight in BHI broth.The bacterial suspension was 10-fold serially diluted in PBS, and 10 µL of each dilution was spotted onto BHI agar plates containing various concentrations of H 2 O 2 (10-15 mM).Following incubation at 37°C for 24 h, colony growth on each plate was assessed and imaged.

Bacterial adhesion ability in vitro
To evaluate the adhesion rate, tissue cells of livers, spleens, and muscles from healthy tilapia cultured in Dulbecco's Modified Eagle Medium containing 20% fetal bovine serum were infected with EGD-e, ∆InlA, and ∆InlB strains with multiplicity of infection at 10.After 30 min post-infection, tissue cells were washed with tissue culture medium and then isolated by centrifugation.Then cells were serially diluted 10-fold in PBS.A 100 µL sample of each dilution was plated on BHI and incubated at 37°C for over 24 h.The adhesion rates of L. monocytogenes to tissue cells were calculated according to the number of colonies.

Bacterial motility observation
To evaluate the motility ability of bacteria in different dissolved oxygen environments, EGD-e, ∆InlA, and ∆InlB strains were cultured at 30°C for 24 h under normoxic and anaerobic conditions in vitro, respectively.Motility ability was assessed by examining the migration of bacteria through agar from the center toward the periphery of the colony.

Name
Nucleotide sequence (5′ to 3′) Accession number each dilution was plated on BHI and incubated at 37°C for over 24 h.The bacterial count was calculated using the following formula: CUF/g of tissue = numbers of CFUs counted on plate ×dilution factor.To analyze the correlation between bacterial counts and HIF-1α expression, the dsHIF-1α treatment hypoxic fish were administered intraperitoneally with 100 µL of PBS, EGD-e (1.40 × 10 4 CFU/fish), ∆InlA (1.30 × 10 4 CFU/fish), and ∆InlB (1.38 × 10 4 CFU/fish), respectively.The hypoxic fish were administered intraperitoneally with the above strains as the control group.After 24 h of infection with different strains, livers, spleens, and muscles were collected from each group to calculate bacterial count as described above.

Statistical analysis
In this study, all data are shown as mean ± standard deviation (SD).The statistical analysis was performed using one-way ANOVA followed by Duncan's test.The comparison of relative gene expression between the two groups was performed using t-tests.The differences were considered significant at P < 0.05.

Effects of hypoxia on the expression of hypoxia-related genes
After hypoxia treatment for 24 h, expression profiling of HIF-1α, c-Met, and E-cadherin genes in different tissues of hypoxic fish was conducted by qRT-PCR (Fig. 1).The results showed that the expression levels of HIF-1α in all the examined tissues of hypoxic fish were significantly upregulated (P < 0.05).It is worth noting that the expression levels of c-Met were significantly upregulated in all the examined tissues of hypoxic fish (P < 0.05).However, the expression level of E-cadherin was only significantly upregulated in the liver of hypoxic fish, and they were significantly downregulated both in the spleen and muscle of hypoxic fish (P < 0.05).The significantly heightened expression of HIF-1α in important organs suggested that the cellular internal environment was under hypoxic conditions.The significant upregulation of c-Met may be correlated with the changes of HIF-1α expression level.

Effects of hypoxia on bacterial count in tissues
After infection with L. monocytogenes of EGD-e strain, tilapia were divided into two groups and reared in normoxic and hypoxic environments, respectively.After hypoxia treatment for 24 h, L. monocytogenes were detected in all tissues examined, and the bacterial counts in the livers, spleens, and muscles increased significantly in hypoxic fish compared to normoxic fish (P ˂ 0.05; Fig. 2).In addition, the highest bacterial count was in the liver tissue.These findings suggest that tilapia could be infected by L. monocyto genes either in normoxic environments or in hypoxic environments, and hypoxic environment may be more conducive to L. monocytogenes invasion into tilapia.

Liver section observation and enzyme activity
To explore why the bacterial count in the tissues of hypoxic fish increased significantly, we first observed the status of tissue cells by transmission electron microscopy and analyzed the activity of antioxidant enzymes.The tissue section showed that the mitochondria were swollen and the nucleus was atrophic in liver cells of hypoxic fish either infected with EGD-e strain or not.In hypoxic fish infected with EGD-e strain, the number of mitochondria in liver cells increased obviously compared to those treated with PBS (Fig. 3).In the liver of hypoxic fish, CAT activity increased significantly.However, there was no significant difference in CAT and SOD activities between normoxic fish and hypoxic fish infected with EGD-e strain (P ˃ 0.05; Fig. 4).These findings suggest that L. monocytogenes infection may stimulate mitochondria production and maintain the stable antioxidant enzyme activities under hypoxic conditions.This, in turn, creates a favorable intracellular environment for L. monocytogenes infection.

Effects of HIF-1α knockdown on the expression of hypoxia-related genes
To further explore whether the expression levels of c-Met are regulated by HIF-1α, the effect of dsHIF-1α on mRNA expression was determined by qRT-PCR.The effect of dsHIF-1α on HIF-1α gene expression in livers, spleen, and muscles at different times post-dsHIF-1α treatment was determined.The results showed that dsHIF-1α inhibited the expression of HIF-1α mRNA in livers and spleen from 12 to 36 h post-dsHIF-1α treatment.HIF-1α mRNA expressions in all the examined tissues of hypoxic fish were all significantly knocked down at 24 h post-dsHIF-1α treatment (P < 0.01; Fig. 5A through C).The relationship between HIF-1α expression and the expression of internalin receptor-related genes was also investigated by analyzing the effects of dsHIF-1α on the expression levels of the receptor-related gene in livers, spleen, and muscles of hypoxic fish.The results showed that the expression levels of c-Met in all the examined tissues were all significantly downregulated, and the expression level of E-cadherin in the liver was significantly upregulated at 24 h post-dsHIF-1α treatment (P < 0.05; Fig. 5D through F).These findings suggest that the expression level of the InlB receptor, c-Met, may be regulated by HIF-1α.

Biological characteristics of different strains
E-cadherin and c-Met are the receptors of InlA and InlB, respectively.To explore whether the infection ability of ∆InlA and ∆InlB to tilapia is affected by hypoxic conditions, biological characteristics of EGD-e, ∆InlA, and ∆InlB are compared and analyzed.The results showed that there was no obvious difference in the growth curves and antioxi dant stress abilities between different strains when the strains were cultured in normoxic conditions (Fig. 6A and C).There was no obvious difference in motility between different strains cultured in anaerobic conditions (Fig. 6D).However, the adhesion rates of ∆InlB strain decreased significantly in all the examined tissue cells compared to EGD-e strain (P < 0.05; Fig. 6B).

Effects of hypoxia on the bacterial count of different strains
After infection with EGD-e, ∆InlA, and ∆InlB strains, tilapia were reared in normoxic and hypoxic conditions.After hypoxia treatment for 24 h, livers, spleens, and muscles were collected to calculate the bacterial count (Fig. 7A through C).The results showed that the  bacterial counts in all the examined tissues of hypoxic fish increased significantly compared to normoxic fish after infection with EGD-e and ∆InlA strains (P < 0.05).However, there was no obvious difference in bacterial counts between normoxic fish and hypoxic fish after infection with ∆InlB strain (P ˃ 0.05).These results suggested that L. monocytogenes burden counts in tissues of tilapia infection with ∆InlB strain may not be affected by the hypoxic environments.

Effects of HIF-1α knockdown on the bacterial count of different strains
After 24 h post-treatment with dsHIF-1α, the correlation between bacterial load and HIF-1α expression level was analyzed by calculating the bacterial counts in hypoxic fish infected with EGD-e, ∆InlA, and ∆InlB strains (Fig. 7D through F).The results showed that the bacterial counts decreased significantly in all the examined tissues of hypoxic fish infected with EGD-e and ∆InlA strains after dsHIF-1α treatment (P < 0.05).However, there was no significant difference in bacterial counts between the PBS group and the dsHIF-1α treatment group in hypoxic fish infected with ∆InlB strain (P ˃ 0.05).These results suggested that the expression levels of HIF-1α in hypoxic fish may affect the invasion ability of L. monocytogenes by regulating the expression of the InlB receptor.

DISCUSSION
The adverse effects of hypoxia on energy metabolism, immune response, and respiratory function have been reported in previous studies (24)(25)(26).In hypoxic responses, HIF-1α activity is tightly regulated by the level of available oxygen in cells (27).The HIF-1α transcript level has been identified as a potentially helpful biomarker of aquatic animal exposure to hypoxic conditions (28).As a regulatory factor of nuclear transcription, HIF-1α participates in a lot of physiological or pathological processes by regulating the transcription of target genes.Under hypoxic conditions, the increased susceptibility of aquatic animals to pathogenic bacteria may be closely related to the change in HIF-1α transcript level (23,29).To evaluate whether the infection ability of L. monocytogenes to tilapia was affected by a hypoxic environment and what was the correlation mechanism between the changes in infection ability and HIF-1α transcript level, a series of experi ments were carried out involving hypoxic tilapia in the present study.After hypoxia treatment for 24 h, HIF-1α was expressed at significantly high levels in the livers, spleens, and muscles of tilapia, indicating that the tissue cells have been in hypoxic conditions.L. monocytogenes burden counts increased significantly in the livers, spleens, and muscles of tilapia under hypoxic conditions, and the highest bacterial counts were in the liver tissue.The liver has a robust metabolic function and plays a vital role in glycogen synthesis and decomposition.The highest bacterial counts in the liver of hypoxic fish may be closely related to the energy supply for bacterial invasion and proliferation by enhanced glycolysis ability of the liver under hypoxic conditions (30).After hypoxia treatment for 24 h, the atrophied nucleus, swollen mitochondria, and increased CAT activity indicated that oxidative stress had occurred in the liver of hypoxic fish (31,32).After infection of the EGD-e strain, no significant difference in the numbers of mitochondria and CAT activities between normoxic fish and hypoxic fish suggested that L. monocytogenes that invade cells may use unique infection mechanisms to maintain the stability of the intracellular environment of hypoxic host for its survival (33).
To further explore the correlation mechanism between significantly increased bacterial counts and the changes in HIF-1α transcript level in tilapia under hypoxic conditions, the downstream genes of the HIF-1α signaling pathway including E-cadherin and c-Met were examined by qRT-PCR.E-cadherin is important to maintain tight connections and clearance connections.Overexpression of HIF-1α can induce abnormal expression of the E-cadherin gene in the oxygen-deprived areas of tumor cells, which leads to the abnormal regulation of the related signaling pathway and affects intercellu lar adhesion, invasion, metastasis, and survival (16,18,29).In this study, whether the abnormal expression of E-cadherin promotes the proliferation of bacteria in hypoxic fish by affecting the adhesion between cells remains to be further studied.It is worth noting that the expression levels of HIF-1α and c-Met were positively correlated in the examined tissues of hypoxic fish.The results are consistent with the previous study that the hypoxia environment in the tumor can promote the upregulation of the c-Met expression level (23).The expression levels of c-Met were all significantly downregulated when the expression level of HIF-1α in the liver, spleen, and muscle of hypoxic fish was successfully inhibited by specific double-stranded RNAs for HIF-1α (dsHIF-1α), suggesting that HIF-1α might regulate the expression process of c-Met under hypoxic conditions.Therefore, we guess that the increased bacterial counts in hypoxic fish may be closely related to the increase in c-Met expression levels because the increased c-Met expression may significantly improve the binding ability of InlB to the c-Met receptor.
In this study, the ∆InlA and ∆InlB strains of L. monocytogenes were used to further explore the reason for the significant increase in bacterial counts in hypoxic fish.No obvious difference in the growth curves, antioxidant stress, and motility abilities among EGD-e, ∆InlA, and ∆InlB strains of L. monocytogenes suggested that InlA and InlB deletion do not affect the growth, antioxidant stress, and motility abilities of L. monocytogenes, which is a prerequisite for comparing whether the bacterial counts of different strains are affected by the hypoxic environments.After infection of tissue cells with EGD-e, ∆InlA, and ∆InlB strains, the adhesion rates of ∆InlB strains decreased significantly, suggesting that InlB may be the main factor mediating L. monocytogenes infection of tilapia.In addition, the bacterial counts of ∆InlB strains in examined tissues were not affected by hypoxic conditions and HIF-1α expression levels, and the bacterial counts of EGD-e and ∆InlA in hypoxic fish decreased significantly when dsHIF-1α inhibited the expression levels of HIF-1α.The results are similar to the study of L. monocytogenes infection to Hela cells, which showed that the infection ability of L. monocytogenes to Hela cells was significantly decreased after inhibition of HIF-1α expression (34).
In conclusion, this study sheds light on the relationship between dissolved oxygen levels and L. monocytogenes invasion into tilapia, particularly the role of HIF-1α in bacterial invasion into hypoxic fish.The findings suggest that HIF-1α may promote the internalization of InlB by upregulating c-Met expression and result in increased bacterial counts in the livers, spleen, and muscles of hypoxic tilapia.HIF-1α plays a key role in promoting the invasion of L. monocytogenes into tilapia under hypoxic environments.

FIG 3
FIG 3 TEM observation of the liver cells from the normoxic and hypoxic tilapia 24 h post-L.monocytogenes infection.

FIG 4
FIG4 The SOD (A) and CAT (B) activities in the liver of tilapia after hypoxia treatment for 24 h.Data are expressed as the mean ± SD (n = 3).**P < 0.01.

FIG 5
FIG 5 qRT-PCR analysis of HIF-1α and receptor-related genes transcription in various tilapia tissues.(A-C) HIF-1α expression levels after hypoxia and dsHIF-1α treatment at different time points.(D-F) The expression levels of HIF-1α and receptor-related genes after hypoxia and dsHIF-1α treatment for 24 h.Data are expressed as the mean ± SD (n = 3).**P < 0.01.

TABLE 1
Primers used in this study