Effects of Infections of Helicobacter Pylori With Different Virulent Factors on Severity of Gastrointestinal Diseases

Background: Helicobacter Pylori (H. pylori) infection, one of the most common chronic bacterial infections, has been considered as a major cause of diseases such as lymphoma, gastritis, peptic ulcers, and stomach cancer. Here, we aimed to determine whether H. pylori strains with different virulence contribute to the gastrointestinal diseases differentially in clinical settings, which may provide future direction for eradication of H. pylori infection. Methods: We recruited 501 patients with gastrointestinal disorders for analysis of antibody types of H. pylori infection. Correlation analysis was done to determine the association of different virulence of H. pylori with patients’ baseline parameters and personal disease history. Next, subjects with each type of anti- H. pylori infection antibody were subjected to esophagogastro duodenoscopy (cid:0) EGD) and colonoscopy examinations. The pathological diagnosis was also conducted in endoscopic samples. Chi-squared test was employed to compare the differences in endoscopic assessments and pathological ndings among three types of H. pylori infection determined by the presence of antibodies to virulent factors. Results: There were 296 cases with Type I H. pylori infection, 120 cases with Type II H. pylori infection, and 85 cases without H. pylori infection (negative, Type III). No correlation was found between different virulence of H. pylori and participants’ baseline data (P > 0.05). EGD results showed that the incidences of peptic ulcer, bleeding and malignant lesions in Type I group were signicantly higher than that in Type II and Type III (P (cid:0) 0.05). Despite of increased trends of incidences of precancerous alterations and the malignance in Type I group compared with type II and III groups, there was no signicant difference (P > 0.05). In addition, coloscopic features were similar among three groups. On the other hand, infections of H. pylori with cytotoxin-associated gene A (CagA) and/or vacuolating cytotoxin A (VacA) virulent

In clinical practices, patients are often requested to eradicate H. pylori infection if the subject is tested positive using UBT. According to the 2019 Expert Consensus Opinion on Eradication, Prevention and Control of Gastric Cancer in Helicobacter Pylori China (Shanghai, 2019), the elimination of H. pylori helps to control the outcome of gastric cancer [5] . However, the role of different virulence of H. pylori in the development of gastrointestinal diseases remains unknown. Until now, based on antibodies against the bacterial virulence factors in serum, patients with H. pylori infection are classi ed into three types in accordance with the presence of CagA, VacA and Urease A/B (Ure A/B) or absence. We hypothesized that different virulent factors predict the severity and outcomes of gastrointestinal diseases in those patients with H. pylori infection. In this study, serum samples from 501 patients who had been hospitalized for gastrointestinal symptoms from September 2019 to June 2020 in the Department of gastroenterology at Puren Hospital in Wuhan were collected for evaluation of bacterial virulence. By doing so, we anticipate to solve the puzzle about the effects of different virulent factors of with H. pylori on gastrointestinal diseases.

Patients
We recruited consecutively 501 patients with gastrointestinal symptoms (abdominal pain, abdominal distension, acid re ux, etc.) in the Department of Gastroenterology at Puren Hospital from September 2019 to May 2020 for H. pylori antibody tests and other examinations. All patients were informed consent, including 260 males and 241 females, with a minimum age of 15 years and a maximum age of 90 years (average age of 57.02 ± 14.01). The baseline data of the patients included sex, age, smoking history, alcohol consumption history and family history of cancers.
Exclusion criteria: Patients with severe cardiovascular, hepatic, kidney dysfunction, immune de ciency disease, those with esophagogastric variceal hemorrhage, pregnant women or with mental illness, previous anti-H. pylori administration or those undergoing H. pylori eradication treatment were also excluded.
All subjects were briefed about this study, and had signed the informed consent form before the enrollment. The study protocol was also approved by the institutional review board (IRB) at the Puren Hospital.

Measurements of types of virulent factors of H. pylori infection
Types of H. pylori antibody in the patient serum were determined using a detection kit produced by Shenzhen Brautt Company, and Western blotting method. The manufacturer's instructions were followed. The kit classi ed the infections as Type I, Type II and Type III (negative) based on three types of antibody scenarios. Type I is positive for CagA and/or VacA antibody and Ure A/B antibody. Type II is positive for Ure A/B antibody, and negative for CagA and/or VacA antibody. Type III (Negative) is negative for CagA, VacA, and Ure A/B antibodies.

Esophagogastroduodenoscopy and colonoscopy
Of the 501 patients who underwent the antibody classi cation measurement, 384 patients received esophagogastroduodenoscopy (EGD) and 136 patients received colonoscopy. All participants signed the informed consent for their respective examinations. To those who would receive endoscopic therapy (such as: hemostasis with gastric and duodenal ulcer, excision of polyps), the informed consent for those operations were also obtained. EGD and colonoscopy were performed using the Olympus GIF-HQ290 and Olympus CF-H290L / I, respectively.
Endoscopists with more than 5 years of experience reviewed the endoscopic images and then made nal endoscopic diagnosis.

Pathological examination
Except bearing the risk of hemorrhage in certain sorts of lesion locations by sampling, lesions of all patients who underwent endoscopy were collected by endoscopists and then were diagnosed by pathologists. The slides were reviewed by three pathologists who have worked for more than 5 years. If two pathologists disagreed with the diagnosis, the third pathologist then reviewed the slides to reach a consensus.

Statistical Analysis
First, SPSS 23.0 software was used to perform correlation analysis to determine the association between different antibody types of H. pylori infection and age, sex, smoking, alcohol consumption and family history of cancer according to the baseline data of 501 patients involved. Secondly, we conducted the Chi-squared test to compare the effects of presence or absence of antibodies to the virulent factors of H. pylori in those patients on gastrointestinal diseases via endoscopy and pathological measurements. A two-sided P-value less than 0.05 was considered statistically signi cant.

Results
H. pylori antibody types were not correlated with age, sex and personal history Among 501 patients examined for the antibodies to the virulent factor of H. pylori infection, 296 cases were classi ed as Type I H. pylori infection, 120 cases were classi ed as Type II and 85 cases were classi ed as Type III (negative) H. pylori infection. Correlation analysis indicated that none of three groups (Type I, Type II and negative) showed any signi cant association with patients' baseline data (age, sex), and risk factors (e. g. smoking, alcohol consumption, family history of cancer) (P > 0.05) as shown in Table 1.  Table 2, the primary gastroscopic characteristics of type I group were gastric and duodenal ulcers with or without hemorrhage, which had a total rate of 38.9%. This number in type I group was signi cantly higher than that in type II and negative groups (P < 0.05).
More interestingly, the ratio of neoplasia ndings in type I group was also signi cantly higher than that in the other two groups (P < 0.05). On the other hand, the combined rate of polyps in esophagus, stomach and duodenum was comparatively added in type II and negative group, which accompanied with reduction of incidence of ulcers, bleeding and neoplasia ndings. There was no signi cant difference of any of these parameters among three groups as shown in Table 3 (P > 0.05). The percentage numbers of examined cases among three type groups are not different (P > 0.05). As shown in

Discussion
H. pylori is a group of gram-negative spirochetes widely colonized in the stomach. More than 50% of the world's population is currently infected with the bacteria, which have been identi ed by the International Cancer Organization as the leading cause of gastroduodenal ulcers, chronic gastritis and stomach cancer [6] . The genome of H. pylori has high genetic diversity which affects its invasion to the gastric mucosa of the host. The mechanism of its pathogenicity is related to the interactions among virulence factors, host genes and environmental factors. The genome of H. pylori encodes more than 1,500 proteins, of which more than 500 genes are speci c for H. pylori. The pathogenic ability of H. pylori strains is diverse, so the development and progression of host diseases caused by H. pylori strains vary. So far, the known virulence factors of H. pylori strains have included CagA, VacA, Urease A/B, DupA, OMPs and ICEA, of which CagA, VacA and Urease A/B are the best characterized and investigated ones [7][8][9] .
Based on the presence or absence of the major three virulence factors (CagA, VacA and Urease A/B), patients with H. pylori infection can be divided into three types. The data presented in this manuscript are from 501 participants. We observed that the incidence of type I infection is higher than that of Type II and negative group. The correlation analysis revealed that there is no association between different virulence of H. pylori and patients' baseline parameters and personal history including age, sex, history of smoking, alcohol consumption, and family history of cancer. In the future, we plan to recruit more participants to further evaluate whether any of the risk factors of hypertension, diabetes and hyperlipidemia is related to the different virulence of H. pylori infection.
It has been established that the highly virulent H. pylori strains comprise the cytotoxin-associated genes pathogenicity island (CagPAI), which is a 40 kb region containing 31 genes encoding the elements of a type IV secretion system, participating in CagA toxic activity and following a series of in ammatory responses [10] . Research has shown that infections with CagA positive strains are more virulent than the strains without this genotype in gastric colonization and proliferation [11] . In the process of infection, CagA is localized on the plasma membrane, in which it is phosphorylated at speci c Glu-Pro-Ile-Tyr-Ala (EPIYA)-motifs through Src and Abl kinases in host [12,13] .
The degree of toxicity with CagA is associated highly with numbers and types of the EPIYA-motifs at the C-terminal region. After translocation, CagA interacts with multiple host cell molecules and induces the dysregulation involved in the homeostatic signal transduction of gastric epithelial cells, triggering chronic pro-in ammatory responses involving apoptosis, disruption of cell polarity and promotion of genetic instability, through which carcinogenesis then takes place. Owing to such cancer-inducing traits, CagA therefore has been designated as the rst bacterial oncoprotein [14] .
The vacuolating cytotoxin A (VacA) is known for its capacity to induce the formation of vacuoles in eukaryotic cells.
VacA gene exists in all H. pylori strains, with different vacuolating ability, which is conferred by variations in ve VacA regions: s-region (s1 and s2), i-region (i1, i2, i3), m-region (m1 and m2), d-region (d1 and d2), and the recently identi ed c-region (c1 and c2). Different variants of VacA like s1, s2, m1 and others could bring about varied toxic effects from cellular vacuolation, oxidative stress to apoptosis induction [15,16] . According to results from Ogiwara H, the potential risk of developing peptic ulcers and cancer in patients carrying s1 or m1 is signi cantly higher than those carrying other toxins [17] . In Iran, for example, the d1 was identi ed as a novel oncogenic factor for gastric cancer [18] . Additionally d1 and i1 have displayed a synergistic effect on the progress of gastric cancer [19] . Once infected, VacA induces the vacuolation of epithelial cells, triggers the release of mitochondrial cytochrome C and then initiates apoptosis [20] . Moreover, it also confers the formation of membrane channels, through which cytochrome C is released and then binds to extracellular receptors. As a result, cellular in ammatory responses occur consequently [21,22] .
In our present study, we found that the probability of peptic ulcer, gastrointestinal bleeding and malignancy increased signi cantly in the type I patients carrying H. pylori with CagA/Vaga compared with type II and negative counterparts. At the same time, infections of H. pylori with UreA/B show primarily as relative mild gastroscopic manifestations like gastritis, re ux esophagitis, polypoid changes. In regard to mechanism, we reason that the main virulent factors CagA and VagA act on the gastric and duodenal mucosa, and trigger a series of in ammatory reactions, which lead to relatively severe gastroscopic outcomes such as peptic ulcer, bleeding, and even cancer.
UreA/B in Type II group is less virulent than CagA/VacA in type I group, which could only in uence vesicular transport and cell cycles [23] . As a result, gastroscopic characteristics are based on chronic in ammatory cell in ltration of the mild to moderate in ammatory response without ulcers and malignant changes. Although precancerous changes (atrophy, intestinal metaplasia, intraepithelial neoplasia) and cancers in type I group show a trend of increase compared with that in type II and negative groups, the difference among them is not signi cant.
The inadequate number of participants maybe a factor attributing to this nonsigni cant result. In future studies, we should further subdivide the pathological results based on severity of in ammation, and various precancerous changes from mild to severe. By analyzing histopathological results like that, a de nite link between the different virulence of H. pylori strains and pathologic abnormalities may be established accurately.
There have been increasing documents showing that the prevalence of colorectal ailments has been linked to the infection with H. pylori, whereas no consistent conclusion has been drawn by far [24][25][26][27] . Boyuk B et al investigated 341 subjects and showed that the progression of colorectal adenoma and carcinoma is independent of H. pylori infection [24] . From our colonoscopic results, the incidences of colorectal polyps, and carcinoma with different virulent factors of H. pylori infection are not different among the three groups. However, the pathological diagnosis revealed that rates of severe in ammation, adenomatous polyp and colorectal cancer in patients with CagA/VacA toxins are much higher than that with UreA/B and negative group. Infection with H. pylori causes the secretion of gastrin in the blood, which can act as a hormone to stimulate the growth of the colonic mucosa cells. H. pylori can cause hypergastrinemia only or in combination with changes in the normal gastrointestinal ora, suggesting an acceptable mechanism for its carcinogenicity [28,29] . The limitation of our study is the relatively small number of patients who received colonoscopy examinations. If we further group the colonoscopic ndings according to size, location and morphology of intestinal lesions, the exact relationship could be de ned.

Conclusions
In conclusion, the H. Pylori strains carrying CagA/VacA are more likely to cause severe gastrointestinal diseases. Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.