UreB immunodominant epitope‐specific CD8+ T‐cell responses were beneficial in reducing gastric symptoms in Helicobacter pylori‐infected individuals

Although Helicobacter pylori is recognized as an extracellular infection bacterium, it can lead to an increase in the number of CD8+ T cells after infection. At present, the characteristics of H. pylori antigen‐specific CD8+ T cells and the epitope response have not been elucidated. This study was focused on putative protective antigen UreB to detect specific CD8+ T‐cell responses in vitro and screen for predominant response epitopes.


| INTRODUC TI ON
Helicobacter pylori (H. pylori) can be survived in the strongly acidic, slightly aerobic environment of the human and animal stomach 1 and secrete a large number of antigenic components, ultimately inducing chronic gastritis, gastric and duodenal ulcers, and gastric cancer gradually. [2][3][4] However, H. pylori infection does not always result in the development of these gastric diseases, and a significant number of infected individuals exhibit asymptomatic carriage. It is unclear why some H. pylori-infected individuals develop disease while others do not. Diet, H. pylori virulence factor variations, environment, and host immune factors can all play a role. 5 The multiple interactions between host and gastric microbiome were deserved to be discussed. Of particular, interest is the ability of the host generating a specific immune response after H. pylori infection. [5][6][7] Therefore, the mechanism of the immune response to resistant H. pylori in body deserves to be explored in depth.
Urease B subunit (UreB) is the main components of urease, which is secreted in large quantities during natural infection to assist H. pylori colonization. Studies on vaccine had confirmed that UreB-specific CD4 + T cells secreted multiples of cytokines, which effectively reduced the amount of H. pylori colonization. 8,9 However, CD4 + T cells from natural host response could not prevent the long-term infection of H. pylori. 10 Therefore, the natural immune protective mechanism against H. pylori could not be only from the perspective of specific CD4 + T-cell response. 11 In recent years, some studies had focused on CD8 + T-cell immune response, attempting to provide a new vision for analyzing the immune response during H. pylori infection. The number of CD8 + T-cell infiltration in the gastric mucosa of a mouse model was significantly increased and triggered a severe local inflammatory response even atrophy of the gastric mucosa. [12][13][14] These conclusions had suggested that specific CD8 + T cells participated in the resistance to H. pylori infection but its particular functions were still controversial.
Given that the current studies mainly focused on the changes of CD8 + T-cell numbers in H. pylori-infected individuals, 12,13 the changes in the responses level and intensity of specific CD8 + T cells induced by H. pylori antigenic components have not been clearly described, especially the studies of antigen dominant peptide-specific CD8 + T cells have been rarely addressed. It is well known that the specific T-cell response is related to a few dominant peptides (epitopes) on the whole antigen molecule, that is, the phenomenon of "immunodominance." 15  were both positive, the subjects were considered to be infected with H. pylori, and when both negative, they were uninfected. To analyze overall or local response levels of CD8 + T cells, peripheral blood and gastric biopsy samples from H. pylori-positive and H. pylori-negative subjects were obtained. All gastric biopsy samples were collected from the gastric antrum mucosa and were examined by rapid urease test to differentiate H. pylori infection. For the bulk culture of peptide-specific T cells, peripheral blood samples from 506 H. pyloripositive subjects were collected. Among them, 74 individuals responded to the UreB peptides pools and the predominant response epitopes were screened through stimulation with synthetic series of peptides. We then performed HLA-restricted identification on these 74 subjects. Subsequently, the rest of 506 samples were cultured in vitro and detected the response to the dominant peptides by ICS. HLA information of the 74 individuals and the basic information (name, gender, age, disease diagnose and so on) of all the 506 individuals had been summarized and attached as Data S1.
The peripheral blood mononuclear cells (PBMCs) were isolated from the peripheral whole blood samples using Ficoll (TBDscience, Tianjin, China) density gradient centrifugation and cryopreserved in liquid nitrogen. The gastric tissues were formalin-fixed and paraffinembedded to made into paraffin pathological sections. Gastric biopsies were histologically evaluated for chronic gastritis diagnosis and pathological inflammation scores according to the criteria of the visual analog scales in Sydney classification and grading of gastritis. 20

| Immunofluorescence staining (IF)
Paraffin-embedded samples were cut into 4 μm sections for IF.
Briefly, after deparaffinization and hydration, tissue sections were

| Intracellular cytokine staining (ICS)
In the peptide screening assay, bulk cultured T cells were stimu-

B cells were isolated from PBMCs using EasySep™ Human CD19
Positive Selection Kit II (STEMCELL Technologies, Vancouver, Canada) according to the manufacturer's instructions.

| Statistical analyses
Statistical analysis was performed using Prism version 9.0 (GraphPad Software, San Diego, CA, USA). The Mann-Whitney U-test or unpaired t-test was generally used to analyze the differences between two unpaired groups. While the Wilcoxon's signed rank test or paired t-test was used for two paired samples. Differences between categorical variables were assessed through the Fisher's exact test.
The data were shown as the mean ± SD unless otherwise stated.
Two-sided p < .05 was considered significant.

| Significant increase of CD8 + T cells in gastric mucosa after H. pylori infection, compared with CD4 + T cells
We adopted flow cytometry to evaluate the percentage of CD4 + T and CD8 + T cells in peripheral blood from H. pylori uninfected and infected individuals. The results were shown in Figure 1A,B, the percentage of CD4 + T cells in H. pylori uninfected and infected individuals was not significantly different, neither was CD8 + T cells.
Meanwhile, we calculated the numbers of CD4 + T and CD8 + T cells in the gastric tissue. It showed that the densities of both CD4 + T and CD8 + T cells in gastric tissue of H. pylori-infected individuals were significantly higher than uninfected individuals ( Figure 1C,D, p < .0001). We further made pairwise comparison of CD4 + T and CD8 + T cells in the gastric tissue of infected individuals, and found that the density of CD8 + T cells was significantly higher than CD4 + T cells ( Figure 1E, p < .01). These data suggested that both CD4 + T and CD8 + T cells were involved in host immunity during H. pylori infection, while more CD8 + T than CD4 + T cells in gastric tissue. So we focused on the specific CD8 + T cells responses to H. pylori infection in this study.

| Characterization of HLA-I genotype distribution in Chinese Han population
By analyzing the HLA-I genotypes of the Chinese Han population in the allele database, we summarized the characteristics of top ten HLA-I genotypes, as shown in Figure 2A. HLA-A*1101, HLA-B*4001, and HLA-C*0702 were the most frequent genotypes, accounting for 22%, 11.5%, and 15.7%, respectively. Then, these three major HLA-I genotypes restricted peptides of H. pylori UreB protein were predicted using the IEDB database, and the top ten predicted non-overlapping peptides were synthesized. The synthesized peptides were in the range of 9-12mer ( Figure 2B).

| Significant specific CD8 + T-cell responses in peripheral blood of H. pylori-infected individuals
We combined the 30 restriction epitopes into three stimulatory peptide pools according to HLA-I genotype, as PoolA, PoolB, and PoolC, respectively, and then stimulated PBMC from H. pylori-infected individuals in vitro with the three specific peptide pools for 12 days.
After that, the specific CD8 + T-cell responses were detected by

| Identification of immunodominant epitopes in H. pylori-UreB-specific CD8 + T cells responders
Although we had identified the presence of UreB-Pools specific CD8 + T-cell responses in H. pylori-infected individuals, it was needed to further explore the dominant epitopes. Firstly, cultured PBMCs from Donor1 with PoolA responding were stimulated with 10 individual peptides. The results showed that the A-2 response was higher than the others. Similarly, we found that Donor 2 and Donor 3 had a dominant response to B-4 and C-1, respectively ( Figure 3A).

| UreB-C-1-specific CD8 + T-cell responses were associated with the reduction of gastric symptoms in H. pylori infection
According to the clinical, pathological and imaging diagnosis, the H. pylori-infected individuals were divided into two groups with gastric symptoms (Symptoms) and without gastric symptoms (No Symptoms). The "Symptoms" mainly included gastric polyps, chronic gastritis, peptic ulcer, and gastric cancer, while participants without above symptoms were selected into "No Symptoms." Then, we we further analyzed the gastric pathological inflammation scores in Symptoms group according to Updated Sydney system (1994). 20 The results displayed that the inflammation score of UreB-B-4 responders was significant lower than that of non-responders in subjects with gastric symptoms, and similar results were also found in UreB-C-1 responders ( Figure 6B, p < .05). It was manifested that the immune response to UreB-B-4 and UreB-C-1 was associated with alleviating gastric inflammation during H. pylori infection, and may play an immunoprotective role in H. pylori-infected individuals.

| DISCUSS ION
Considering that urease is an exogenous secreted protein, the UreB was mainly recognized by APCs and presented to CD4 + T cells. Our group has previously performed systematic screening and identification of the dominant epitopes of UreB-specific CD4 + T cells in humans 23 and mice, 8 and analyzed the function of the dominant Th1 epitopes in H. pylori infection. However, CD4 + T cells from natural host response could not prevent the long-term infection of H. pylori. 10 The specific CD8 + T cells had been participated in the response to H. pylori, but its particular functions remained unclear.
Therefore, we focus on CD8 + T cell-specific response characteristics of the UreB dominant epitopes and its relationship with H. pylorirelated disease symptoms in this study.  were carried out by Meissner et al. 27 Given that T cells in human gastric tissue were difficult to obtain clinically and gastric primed T cells were recruited to mucosal tissues after undergoing peripheral circulation, peripheral blood samples from H. pylori-infected patients were used to analyze the function of H. pylori-specific CD8 + T cells in this research. 28 However, considering that the delivery of APCs was not only related to the protein content, but also depended on the ability of protein processing and delivery process. It had been reported that various proteins had differentially generated MHC-I restricted epitope responses. An abundance of specific CD8 + T-cell epitope responses were derived from out-of-frame proteins, but the high quantity of N-protein was produced very few CD8 + T-cell epitopes. 33 So it could not be excluded that stronger immune response ability was stimulated by other antigens from H. pylori.
In our studies, we had predicted dominant epitopes of UreB on basis of the high frequency of HLA-I genotypes in Han population.
The Han nationality accounts for 92% of China's population; thus, it is widely representative of the main HLA-I genotype characteristics.
The length of all three dominant epitopes was 9mer, which has been considered as the minimal antigenic epitopes recognized by specific CD8 + T cells. 21,34 The prediction method has been widely applied for the screening and identification of dominant epitopes, as reported in previous studies. 35 In this study, we had screened and identified the immunodominant peptides of specific CD8 + T-cell responses through prediction methods combined with T cells bulk culture in vitro. However, due to the biases of the prediction algorithm, the peptide candidates may be incomprehensively and inaccurately predicted compared to other methods. The possible existence of other dominant response peptides could not be ruled out.
As far as we know, specific CD8 + T-cell responses are strictly regulated by HLA-I molecules. 18,19,36 During the clarification of the F I G U R E 6 UreB-specific CD8 + T-cell responses were associated with reduced risk of gastric symptoms in H. pylori infection. The clinical, pathological, and imaging diagnosis information of 506 H. pylori-infected individuals were collected and statistically analyzed. The symptoms included gastric polyps, gastritis, peptic ulcer, and gastric cancer. (A) A chi-square analysis of the relationship between the presence or absence of the gastric symptoms and the specific CD8 + T-cell responses to immunodominant epitope UreB-A-2, UreB-B-4, and UreB-C-1 in the H. pylori-infected individuals, respectively. (B) Comparison of gastric inflammation scores between UreB-A-2 (n = 16), UreB-B-4 (n = 7), or UreB-C-1 (n = 7) peptide-specific responders and non-responders (n = 118) in individuals with stomach symptoms, respectively. Data were presented as mean ± SD, ns, no significant difference, *p < .05.
HLA-restricted characteristics of the dominant epitopes, we found that epitopes A-2, B-4, and C-1 were separately presented by HLA-A*1101, HLA-B*4001, and HLA-C*0702. However, the individuals who carried the same major HLA-I genotype could present different epitopes, which contain both dominant epitopes and subdominant response epitopes. Chen et al. 37  Furthermore, current studies had confirmed that HLA-I molecules were closely related to gastric diseases. 18,19 The expression of HLA-B*51 was significantly in H. pylori-related gastritis and duodenal ulcer, and HLA-C*03 had a significantly higher risk of gastric cancer associated with H. pylori infection. 19 Our study indicated that the frequency of individuals with HLA-B*4001 and HLA-C*0702 genotypes in the Han population was nearly one-third (27.2%; Figure 2).
UreB-B-4 and UreB-C-1 specific CD8 + T-cell responses, which were restricted by HLA-B*4001 and HLA-C*0702, were associated with decreasing gastric inflammation degree, and may reduce the risk of appearance in gastric symptoms ( Figure 6). Although we had analyzed the gastric inflammatory condition in relation to the specific T-cell response status, further investigation is needed to clarify whether specific responses play a role in more severe symptoms. If the dominant peptides would be applied in vaccines, it may be beneficial for the prevention and treatment of gastric symptoms related to H. pylori infection. In addition, considering the quantity of H. pylori antigen components, more than 800, the specific CD8 + T-cell responses induced by UreB cannot fully represent the function of total CD8 + T cells. Thus, the presence of other antigen-specific CD8 + Tcell responses except UreB needs to be further verified.