Resistant gastric environment of Lactobacillus crispatus from stomach inhibits Helicobacter pylori colonization and attenuates gastric inflammation

Recent studies have shown that gastric-derived Lactobacillus can inhibit the colonization of H. pylori and attenuate gastric inflammation in conventional animals, but the resistant of Lactobacillus to gastric environment is still unknown. Here, we aimed to screen the candidate Lactobacillus that could adapt to the harsh gastric environment and inhibit the colonization of H. pylori. In vitro, the growth rate of seven Lactobacillus strains in different pH and bile salt concentration were tested, the size of inhibition zone and adhesion rate of H. pylori when Lactobacillus exist were measured. In gnotobiotic mice models, we examined the amount of colonization of L. crispatus and H. pylori by qRT-PCR and evaluated the inflammation in the gastric tissue by the content of MPO and H&E stain. In vitro experiments showed L. crispatus had a better growth rate than other six Lactobacilluses in pH 2.5 to 4.5; under the 0.2% bile salt concentration, other bacteria did not grow except for L. crispatus; L. crispatus yielded 24.2 mm of mean inhibitory zone diameters; the adhesion rate of H. pylori only reached 41.3% in H. pylori-L. crispatus group(HLG). In vivo, the amount of colonization of H. pylori in HLG is fifteen times less than that in H. pylori group (HG) (p < 0.05); the MPO value of HG was 1.4 times that of HLG; the gastric tissue inflammation of HLG was obviously lighter than HG. L. crispatus may be an adjunctive therapy for treating H. pylori- associated disease in clinic.


Background
Helicobacter pylori(H. pylori) is a pathogen, which may cause chronic gastritis, peptic ulcers, gastric adenocarcinoma, and gastric mucosa-associated lymphoid tissue lymphoma [ 1 , 2 ]. H. pylori infection has a general prevalence from 11% in industrialized countries to 95% in developing countries [ 3 ]. Several strategies have been proposed to eradicate H. pylori, including the standard first-line treatment (such as proton pump inhibitors, two antibiotics), quadruple, sequential, and concomitant treatments [ 4 ]. However, the eradication rates for H. pylori infection have decreased to less than 80% in many countries [ 5 ], and the antibiotic resistance have become more serious [ 6 ]. Therefore, alternative treatments are of interest.
Microecology theory provide an excited idea to resist the infection of H. pylori [ 7 ]. Lactobacilli are one of human gastric microbiota, and can adhere to the stomach wall and grow under harsh acidic conditions [ 8 ], theyare also probiotics that have positive effects for the health of humans and animals [ 9 ]. It has been reported that Lactobacillus can inhibit the adhesion and growth of H. pylori in in vivo studies [ 10 , 11 ]. In Mongolian gerbils model, Lactobacillus have an anti-infective effect against H. pylori [ 12 ]. According to Thiraworawong's reports, Gastric-derived L. plantarum XB7 can modulate gastric mucosal inflammation in Sprague-Dawley rat model [ 13 ]. In some reports, Lactobacillus ingestion could even decrease the risk of some cancer and tumors [ 14 , 15 ]. Meanwhile, Lactobacillus adjunct therapy have the potential to reduce severity of side effects related to H. pylori eradication therapy [ 13 ], increase the H. pylori eradication rate [ 16 ], and decrease the severity of host cell damage [ 17 ]. So Lactobacillus will be a promising microbial ecological agent that functions as an adjunctive therapy for treating H. pylori-associated disease.
The stomach is a harsh environment, its pH is usually around 3.0 and the bile salt concentration is 0.03% to 0.3% [ 18 ]. Lactobacillus can colonize and live in the stomach which depend mainly on its acid and bile salt resistant [ 19 , 20 ], acid and bile salt resistance are also important indicators for screening Lactobacillus as probiotics [ 21 ]. Although some studies have shown that Lactobacillus can inhibit the H. pylori growth and reduce inflammation, but they did not study the acid and bile salt resistant of Lactobacillus, which will be affected in further development of Lactobacillus as probiotics to play against H. pylori in human gastric.
In the present study, we chose six Lactobacillus strains from the stomach of 21 healthy person, and examined the ability of acid and bile salt resistant, the size of inhibition zone and the adhesion rate of H. pylori in vitro. Based on above results, the most promising candidate L. crispatus was chosen. We use the L. crispatus and H. pylori successfully constructed the model of mono-and di-associated gnotobiotic Kunming mice models.
Meanwhile, the amount of bacteria in stomach were detected, the degree of gastric mucosal inflammation were evaluated by the content of MPO in stomach tissue and the HE staining of gastric biopsy.

Acid and bile salt resistance of Lactobacillus
After four hours incubation, the growths of all the Lactobacillus strains were shown in Fig.   1. The net growth of L. crispatus reached more 170% at pH 2.5 and 3.0, the value of the L. crispatus were 1.5 times that of the L2 (p = 0.002, 0.000), meanwhile other strains were subject to different degrees of growth inhibition. At pH 4.0 and pH 4.5, L. crispatus also grow better than other strains, but the growth advantage was not obvious compared with pH 2.5 and 3.0. In general, L. crispatus has a stable resistance to acid under low pH relative to other strains.
In Fig. 2, growth rate of all strains fell with the increase of concentration of bile salts. At the bile salt concentration 0.2%, only L. crispatus were in growth, other strains' germination was stopped. When the concentration of bile salt is 0.1%, L. crispatus were grown better than other strains (p < 0.05), the net growth is 100%. The L6 and L7 did not grow at the concentration of bile salt 0.05-0.2%. According to the above results, L. crispatus also has better bile salt tolerance than other six strains.
Lactobacillus inhibit the growth and adhesion of H. pylori in vitro As shown in Table 1, all the Lactobacillus have a inhibitory effect on the growth of H. pylori by Oxford cup method in vitro, but there is a best effect that L. crispatus inhibited the growth of the H. pylori, the value of inhibition zone is 24.2mm, which is 1.8 fold that of L2 (p = 0.000), and 3.2 fold that of L7 (p = 0.000). There were statistically significant differences in the L3 compared with other strains at IZD (p = 0.000).
With the H. pylori's adhesion rate as the positive control, the adhesion rate is defined as 100%. In the interaction between H. pylori and Lactobacillus, all the Lactobacillus have a inhibitory effect on the adhesion of H. pylori to GES, at the same time, the adhesion rate of H. pylori in L-H group are higher than that in H-L group for all Lactobacillus (Table 2). In L-H and H-L group, the adhesion rate of H. pylori when L3 exsited are 51.2% and 41.3%, the adhesion rate of H. pylori are 10% lower than other groups (p = 0.000). In turn, the inhibition of H. pylori adhesion of L3 is strongest.
The quantitative of H. pylori and L. crispatus in stomach The colonization amount of H. pylori and L. crispatus in different gnotobiotic Kunming mice is shown in Table 3. The colonization amount of H. pylori in HG is 2.008 × 10 7 copies/g, it is 5.2 times that of LHG and 14.8 times that of HLG, there were statistically significant differences between them (p = 0.000, 0.000). At the same time, the amount of L. crispatus in LHG and HLG are more than 1.0× 10 6 copies/g. It shows that L. crispatus can suppresse the adhesion and colonization of H. pylori in vivo, the L. crispatus has more obvious inhibitory effect among orally administered L. crispatus after H. pylori infection, which is consistent with the results of adhesion experimental in vitro.

Evaluation of gastric inflammation
To estimate the degree of inflammation, the MPO experiment data of four groups are shown in Fig. 3. In each groups, the value of MPO is relatively concentrated, this suggests that the experiment has good uniformity. In HG, the value of MPO is 0.248 μg/g, and the HLG is only 0.173μg/g, the HG is 1.2 and 1.4 times of LHG and HLG, separately (p = 0.003, 0.000). It shows that L. crispatus obviously reduces the MPO level of gastric tissue and the degree of gastric tissue damage caused by MPO.
The gastric inflammation was also determined by the histopathological examination and histology scores. Histopathology in the LG was normal, there was moderate inflammation in the HG, while the LHG and HLG improved stomach inflammation ( Figure 4). The histology scores for gastric inflammation are summarized in Table 4

Discussion
Recent studies have revealed that Lactobacilli can inhibit the growth of H. pylori and reduce gastric inflammation, some of them are also derived human gastric, and these have been shown that Lactobacill will be a potential therapeutic agent for treating the infection of H. pylori. However, it has not been clairified that whether these strains can colonize the human stomach and survival. In this study, we examined the acid and bile salt resistance of seven Lactobacilli strains in vitro, however, only L. crispatus has a best acid and bile salt resistance. We then studied the growth and adhesion rate of H. pylori that inhibited by the seven Lactobacilli strains, all these strains have a resistance to H. pylori activity, but the activity of L. crispatus is outstanding. Combined with the above results, we further confirmed the anti-growth and anti-inflammatory properties of L. crispatus in the gnotobiotic Kunming mice models, the effect of anti-growth is consistent with the experimental results in vitro, furthermore, it significantly reduce the H. pyloriassociated gastritis at the same time. These data strongly suggest L. crispatus will be a probiotics agent that can prevent the infection and improve the inflammation of H. pylori in human stomach.
Lactobacillus secretes some antibacterial metabolic products, such as lactic acid, bacteriocin, peroxide, proteinase, exopolysaccharide [ 22 ], lactic acid inhibits the urease activity and viability of H. pylori among these.
According to Sunanliganon's report [ 10 ], low pH values are important for anti-H. pylori activity. At the same time, H. pylori infection and the subsequent increase of gastric pH value may result in a more permissive milieu for colonization with other bacteria, that their metabolic products associated with stomach related diseases [ 23 ]. In this study, the acid-resisting strain also produced a large amount of acid (data not shown), which is also very important that to maintain the normal environment of stomach and resist the H. pylori infection.
To make Lactobacillus inhabit the complex and changeable stomach and play the role of probiotics to maintain the normal pH and bacteria flora of stomach, acid and bile tolerance of its own is very important [ 21 ]. In this study, L2 and L3 have a certain degree of resistance to acid and bile salt, but in the bile salt concentration 0.2%, the growth of L2 was inhibited. At the same time, although L1 and L5 have a certain degree of resistance to low the bile salt concentration, but didn't grow under pH 3.5; the growth of L4, L6 and L7 were inhibited in low acid and bile salts. Those suggest that strains of acid and bile salt resistant properties are inconsistencies. In addition, all Lactobacillus strains inhibit the growth and adhesion of H. pylori in vitro, combined with acid and bile salt resistance, which shows that Lactobacillus has an inhibition effect for H. pylori, but it doesn't adapt to the complex stomach environment, which will also be affected to against the infection and inflammation of H. pylori in human gastric. It also suggests that the importance of acid and bile salt resistant properties in screening of Lactobacillus as a probiotics agent.
In order to verify the anti-inflammatory effect of Lactobacilli in vivo, we chose the gnotobiotic Kunming mice model. Through establish the mono-and di-associated gnotobiotic models, it clearly revealed the interaction between H. pylori and Lactobacillus in vivo. The colonization amount of H. pylori in LHG and HLG are significantly lower than that in HG, which suggest that L. crispatus play a role in reducing the engraftment of H. pylori in mice. At the same time, the better inhibitor effect that L. crispatus inoculated after H. pylori than L. crispatus inoculated before H. pylori in vitro and in vivo, is helpful for developing L. crispatus as probiotics to treat H. pylori infections. In addition, the amount of colonization of Lactobacillus is 1.12× 10 6 copies/g or more in LHG and HLG, it is consistent with the reports that the colonization amount of Lactobacillus must be more than 1.00× 10 6 CFU/mL which can play a role of probiotics [ 24 ].
In order to study the anti-inflammatory effect of Lactobacilli in gnotobiotic Kunming mice, we detected MPO and histopathology to evaluate the gastritis of stomach tissues. The H. pylori strain M13, which was derived from a Chinese clinical isolate and adapted to colonize the gastric mucous layer (GML) of mice [ 30 ]. Itwas recovered and cultivated on Skirrow's medium, and then inoculated in brain heart infusion broth supplemented with 5% foetal bovine serum and 1% glucose with shaking at 220 rpm and 37℃ under 87% N 2 , 5% O 2 and 8% CO 2 in three gas incubators.
After harvesting, the concentration of bacteria was regulated to 1.0 × 10 9 CFU/mL.

Acid and bile salt resistance of Lactobacillus strains in vitro
For the acid resistance experiment, the MRS broth's pH value was adjusted to 2.5, 3.0, 3.5, 4.0, 4.5, they were shaked at 220 rpm and 37℃ for 4h. In the bile salt resistance experiment, the bile salt concentration of MRS broth is 0.05%, 0.1% and 0.2%, the medium were shaked at 220 rpm and 37℃ for 4h, 8h and 12h, respectively. All the initial concentration is 1×10 8 CFU/mL, the total volume is 20 mL. Then the bacterium solution were centrifuged and then resuspended in 1mL, each of them was added to 96-well plates, For the adhesion assay, the experiment were divided into three groups: the control group, the H. pylori M13 were inoculated in microtiter plates and incubated for 2h; the H-L group, the H. pylori M13 were inoculated in microtiter plates and incubated for 60 minutes, then the Lactobacillus strains were added, followed cultivation for 1h; the L-H group, the Lactobacillus strains were added to wells first, and inoculated the H. pylori M13 in microtiter plates later on, then cultivated for 1 h, respectively. In these groups, gastric mucosal epithelial cells (GES-1) were firstly grown on microtiter plates to form a confluent monolayer after 1 day of culturing [ 32 ]. The microtiter plates were washed three times with normal saline, then the H. pylori and Lactobacillus strains suspensions (100 μL) was added to each well, their concentration were 1.0 × 10 7 and 1.0 × 10 8 CFU/mL, separately. After 120 minutes' incubation, nonadherent bacteria were washed off by normal saline for three times. The level of adherent bacteria was estimated by the urease assay [ 33 ], 100μL of urease test solution (7 mM phosphate buffer pH 6. Eight weeks after inoculation, mice were sacrificed by cervical dislocation. The stomachs were removed from the mice and cut into four parts. One part of the stomach was serial plated on Skirrow's medium and MRS agars, separately. The Skirrow agar plates were incubated for 2 days at 37℃ under N 2 87%, 5% O 2 and 8% CO 2 in three gas incubators; the MRS agars were incubated for 24h under the same conditions. One part of the stomach was homogenised with normal saline, the mixture was identified using the urease test. In addition, the mixture was also centrifuged for 5 min at 13000 × g, and the precipitate was

Statistical analysis
The data from three independent experiments and are expressed as means ± standard deviations (SD). The statistical analyses were performed using IBM SPSS Statistics version 19 software (IBM China Inc., Beijing, China), for statistical comparisons, the independentsample t test and χ 2 test or Fisher's exact test were performed, statistically significant differences between groups were defined as a P value of < 0.05.     Figure 1 Comparison of net growth of seven Lactobacillus strains in different pH MRS broth after four hours incubation. The initial concentration of cultivation is 1×108 CFU/mL, the results are expressed as the mean ± SD. The net growth were statistically significant differences in the L3 compared with the other strains at the same pH (*p < 0.01, **p < 0.05).   Histopathological examination of stomachs of gnotobiotic mice by H&E stained. a

List Of Abbreviations
LG showed no gastric inflammation; b HG showed a amount of neutrophil and mononuclear cell infiltration and the destroyed glands structure (arrows); c, d LHG and HLG showed small number of neutrophil and mononuclear cell infiltration (arrows). Original magnification, ×200; bars = 25μm.

Supplementary Files
This is a list of supplementary files associated with the primary manuscript. Click to download.
NC3Rs ARRIVE Guidelines Checklist.PDF