| |

Received: 15 May 2019    Revised: 3 July 2019    Accepted: 9 July 2019

DOI: 10.1111/hel.12652

ORIGINAL ARTICLE

Vaccine‐induced gastric CD4+ tissue‐resident memory T

cells proliferate in situ to amplify immune response against
Helicobacter pylori insult

Ningyin Xu1,2 | Guojing Ruan1,2 | Wei Liu1,2 | Chupeng Hu1,2 | An Huang1,2 |

Zhiqin Zeng1,2 | Shuanghui Luo1,2 | Zhenxing Zhang1,2 | Menghui Fan1,2 | Feng Ye3 |
Tao Xi1,2  | Yingying Xing1,2

1
School of Life Science and
Technology, China Pharmaceutical Abstract
University, Nanjing, China Background: Tissue‐resident memory T cells accelerate the clearance of pathogens
2
Jiangsu Key Laboratory of Carcinogenesis
during recall response. However, whether CD4+ TRM cells themselves can provide
and Intervention, China Pharmaceutical
University, Nanjing, China gastric immunity is unclear.
Materials and methods: We established a parabiosis model between the enhanced
3
Department of Gastroenterology, the
First Affiliated Hospital of Nanjing Medical
University, Nanjing, China
green fluorescent protein and wild‐type mice that the circulation system was shared,
and the wild‐type partner was vaccinated with H pylori vaccine composed of CCF
Correspondence
Yingying Xing, School of Life Science
and silk fibroin in gastric subserous layer to induce gastric EGFP+ CD4+ TRM cells.
and Technology, China Pharmaceutical Antigen‐specific EGFP+ CD4+ T cells and proliferous TRM cells were analyzed by
University, Nanjing 210009, China.
Email: cpuskyxyy@126.com
flow cytometry. The colonization of H pylori was detected by quantitative real‐time
PCR. EGFP+ CD4+ TRM cells and the inflammation of the stomach were observed by
Funding information
This work was supported by National histology.
Key R&D Program of China (No.
Results: A parabiosis animal model was employed to identify the cells that introduced
2017YFD0400303), National Natural
Science Foundation of China (No. by vaccination in GSL. Antigen‐specific EGFP+ CD4+ T cells could be detected at day
81502970), and the Priority Academic
7 post‐vaccination. Thirty days later, EGFP+ CD4+ TRM cells were established with a
Program Development (PAPD) of Jiangsu
Higher Education Institutions and Six Talent phenotype of CD69+ CD103‐. Of note, we found that when circulating lymphocytes
Peaks Project in Jiangsu Province (No. 2018‐
were depleted by FTY720 administration, these TRM cells could proliferate in situ
WSW‐003).
and differentiate into effector Th1 cells after H pylori challenge. A decrease in H py-
lori colonization was observed in the vaccinated mice but not unvaccinated mice.
Further, we found that although FTY720 was administrated, mounted pro‐inflamma‐
tory myeloid cells still emerged in the stomach of the vaccinated mice, which might
contribute to the reduction of H pylori colonization.
Conclusions: Our study reveals that H pylori vaccine‐induced CD4+ TRM cells can
proliferate and differentiate in situ to enhance gastric local immunity during recall
response.

KEYWORDS
EGFP+ CD4+ TRM cells, Helicobacter pylori vaccine, local immunity, proliferate, recall response

Helicobacter. 2019;00:e12652. wileyonlinelibrary.com/journal/hel © 2019 John Wiley & Sons Ltd  |  1 of 12
https://doi.org/10.1111/hel.12652
2 of 12      |
1 |  I NTRO D U C TI O N
Helicobacter pylori (H pylori) is a spiral‐shaped, gram‐negative patho‐
genic bacterium that colonizes in the gastric mucosa of about half
of the world's population.1,2 Although the most infected population
remains asymptomatic, long‐term infection by H pylori is associated
with the prevalence of peptic ulcer, gastric cancer, and malignant
lymphoma.3,4 H pylori is highly adapted to the physical environment
of the stomach, and compared with the intestine, the native lym‐
phocytes in the stomach are fewer. Therefore, enhancing the gas‐
tric local immunity may be able to provide the superior protection
against H pylori infection.
Tissue‐resident memory T (TRM) cells are noncirculating
lymphocytes that are resident in peripheral tissues, that is, sen‐
sory ganglia, 5 brain parenchyma, 6 skin epithelia, 7 genitourinary
tract, and gastrointestinal tract. 8,9 In comparison with circulat‐
ing memory T cells such as central memory T (TCM) cells and ef‐
fector memory T (TEM) cells, TRM cells preferentially localize at
sites of pathogen entry and provide first‐line protection to de‐
fense peripheral infection.10,11 It suggests that introducing TRM
cells into the stomach may be a desirable method to improve
gastric local immunity. Liu. et al had demonstrated that the
mice were vaccinated with H pylori vaccine composed of CCF
(a recombinant H pylori subunit vaccine constructed by cholera
toxin B, multi‐epitopes from H pylori urease, and self‐adjuvant
regions from Salmonella typhimurium phase I flagellin FliC) and
aluminum adjuvant in the gastric subserous layer (GSL) formed
+
a detectable pool of vaccine‐induced CD4 TRM cells in gastric
mucosa. Rather than other ways (such as p.o. and s.c.), vaccina‐
tion in GSL induced rapid and long‐term protection against H py-
lori infection.12 However, the limitation was that Alum‐based
H pylori vaccination in GSL led to granuloma and vaccine‐in‐
duced CD4 + TRM cells were regional. Here, we use silk fibroin
(SF), a biologic material for controlled release drug delivery with
good biocompatibility, degradability, stability, and controllabil‐
ity. 13
SF‐loaded H pylori vaccine can decrease destroy of stom‐
ach, diffuse in the gap of GSL, and induce wide vaccine‐induced
+
CD4 TRM cells.
After vaccination, a subset of effector cells that enter the tissue are
the source of TRM cells. 8,14
However, TRM cells have a numerical dis‐
advantage in infected tissues, so it is important to determine whether
TRM cells can proliferate in response to pathogen re‐exposure. 9 and the skin was separated from the subcutaneous fascia to form
Previous investigations had reported that CD8+ TRM cells not only re‐
sisted pathogens by proliferating in situ after secondary immunization
but also precipitated in the recruitment of recirculating lymphocytes to
the site of TRM cell re‐activation, 15,16
although some reports revealed
that CD4+ TRM cells played a major role in mediating Chlamydia tracho-
matis infection in the reproductive tract, 17 +
and vaccine‐induced CD4
TRM cells also optimized immune protection responses by accelerat‐
ing multiple innate and adaptive immune mechanisms in lung influenza
virus.18 In contrast, the proliferation of CD4+ TRM cells in situ during
recall response had not been reported. Here, we aim to detect whether
XU et al.
the vaccine‐induced CD4+ TRM cells themselves provide gastric im‐
munity during recall response.
Parabiosis is a surgical union of two organisms allowing foreign
cells to recycle into the peripheral circulation of the host through
sharing blood circulation.19,20 It helps us distinguish vaccine‐induced
TRM cells and investigate the formation, proliferation, and differ‐
entiation of TRM cells preferably. 21 This study develops a parabi‐
osis strategy of enhanced green fluorescent protein (EGFP) and
wild‐type (WT) mice combining with GSL vaccination with H pylori
vaccine CCF plus SF in WT mice to induce CD4+ EGFP+ TRM cells in
the stomach. Here, our findings demonstrate that CD4+ TRM cells
can rapidly proliferate and differentiate in situ during H pylori recall
response, which enhances the local immunity of the stomach inde‐
pendently of circulating memory T cells.
2 | M ATE R I A L S A N D M E TH O DS
2.1 | Animals and statement of ethics
Eight‐week‐old female C57BL/6J mice and transgenic EGFP
mice (C57BL/CAG‐EGFP) were respectively obtained from the
Comparative Medicine Center of Yangzhou University and Model
Animal Research Center of Nanjing University and bred at the China
Pharmaceutical University Animal Experimental Center. All animal
experiments were approved by the Animal Ethical and Experimental
Committee of China Pharmaceutical University. All mice were
housed under a controlled temperature (27°C) and light‐dark cycle
of 12 hours. The numbers of mice used in each experimental group
are shown in Table S1.
2.2 | Parabiotic surgery
Eight‐week‐old female C57BL/6J mice and transgenic EGFP mice
were cohoused for 15 days prior to surgery to reduce mutual
harm. For the parabiosis procedure, mice were anesthetized with
50  mg/kg pentobarbital (Servicebio). Then, the right of the WT
mice and the left of the EGFP mice were completely shaved, start‐
ing from about 1 cm above the elbow joint and finally about 1 cm
below the knee joint, wiped with iodine 2‐3 times in the hair re‐
moval area, and placed on a body temperature heating pad. A skin
incision was made on the flank from the elbow to the knee joint,
a 0.5 cm free skin, followed by the separation of the entire inci‐
sion area. The tibia and ulna of each mice were sutured together
with 4‐0 monocryl (Shanghai Pudong Jinhuan Medical Products
Co., Ltd.). After joining the forelimbs and hindlimbs together, the
skin incision of two mice from the elbow to the knee was closed
with a 4‐0 absorbable suture (Shanghai Pudong Jinhuan Medical
Products Co., Ltd.). Immediately following the surgery, each para‐
biont was administered sterile normal saline to prevent dehy‐
dration, erythrocin to minimize infection, and buprenorphine to
manage pain.19
XU et al.
2.3 | Preparation of H pylori vaccine
The CCF (CTB‐UE‐CF) protein that constructed by cholera toxin B,
multi‐epitopes from H pylori urease, and self‐adjuvant regions from
S typhimurium phase I flagellin FliC was expressed by Escherichia
coli Rosetta (DE3) cells with pET‐28a‐CCF. The purification of CCF
was as described previously. 22,23 Briefly, the protein was first puri‐
fied by nickel affinity chromatography (GE Healthcare), followed
by anion‐exchange chromatography with DEAE Sepharose FF
(Amersham Pharmacia Biotech AB). Purity of CCF was confirmed
by Coomassie blue staining. Helicobacter pylori vaccine was pre‐
pared by 5 μL CCF solution (3.3 mg/mL) and 1.2 mg SF (Simatech)
as an adjuvant.
2.4 | Gastric subserous layer vaccination
50 mg/kg pentobarbital was injected i.p. in WT mice to anaesthe‐
tize, and then, the mice were placed on a body temperature heating
pad. The right abdomen was shaved before making a 1‐cm incision
above the stomach, and then, the stomach was localized. Then,
a dose of 10 μL H pylori vaccine (10  μL CCF plus 2.4  mg SF) was
prepared to inject into the GSL of the corpus with a 33 G needle
of Hamilton syringe. After that, 7‐0 absorbable sutures (Shanghai
Pudong Jinhuan Medical Products Co., Ltd.) were performed using
uninterrupted sutures for the peritoneum and interrupted sutures
for the skin incision.
2.5 | Preparation of single cell from gastric tissue
The whole part of the stomach was placed into 15 mL RPMI 1640
(Life Technologies Corporation) containing 10  mM HEPES, 10%
FBS, 4 mM EDTA, and 0.5 mM DTT, and centrifuged at 250 rpm for
30 minutes at 37°C to collect gastric epithelial lymphocytes. After
passing through a 70‐μm cell strainer, gastric tissues were minced
and the isolation of the remaining lymphocytes was performed by
adding another 15 mL RPMI 1640 described above for 20 minutes.
Supernatants were passed through a 70‐μm cell strainer, and then,
all the cells were merged. After washing and centrifugation, cells
were re‐suspended in PBS. The single‐cell suspension of the stom‐
ach was purified with 67%/44% Percoll gradients. After cells were
collected, they were washed with 1 mL RPMI 1640 containing 10%
FBS.12
2.6 | Preparation of single‐cell suspensions from
lymphoid organs and blood
The spleen and MLN were isolated and gently pushed through a
70‐μm cell strainer. After centrifugation, the cells of the spleen and
blood were suspended in 3 mL Erythrocyte Lysis Buffer (BioLegend)
and lysed for 5 minutes on ice to remove red blood cells. After cen‐
trifugation, the cells were washed with 5 mL PBS. The single cells
from the lymph nodes were collected straightway after centrifuga‐
tion and washing. 24
|
      3 of 12
2.7 | Antigen‐specific EGFP+ CD4+ T‐cell
analysis and TRM differentiation experiment
The purified single‐cell suspensions of the stomach, MLN, and spleen
were stimulated with 1 × 106 naïve splenocytes that were preloaded
with CCF in RPMI 1640 containing 10% FBS and 5 μg/mL Brefeldin
A (BFA; BioLegend) for 12 hours. After collection, cells were stained
for IFN‐γ and IL‐17 to detect antigen‐specific EGFP+ CD4+ T cells or
EGFP+ CD4+ effector T cells during recall response.
2.8 | FACS analysis
1 × 106 cells of SP, MLN, BL, and ST were used for flow cytometric
analysis, respectively. For cell surface antigen detection, single‐cell
suspensions were stained with anti‐CD45 (30‐F11), anti‐CD4 (GK1.5
or RM4‐5), anti‐CD3 (17A2), anti‐CD69 (H1.2F3), anti‐CD103 (2E7),
anti‐CD11b (M1/70), anti‐Ly6C (HK1.4), and anti‐Ly6G (1A8). For
IFN‐γ and IL‐17 intracellular cytokine staining, after staining surface
markers, in vitro re‐stimulated cells were fixed and permeabilized
with Intracellular Staining Fixation/Permeabilization Wash Buffer
(BioLegend) and stained intracellularly with anti–IFN‐γ (XMG1.2)
and anti–IL‐17 (9B10) antibodies. All the antibodies were purchased
from BioLegend or BD Pharmingen. Multifarious analyses were per‐
formed on BD Accuri C6 or MACSQuant flow cytometer.
2.9 | In vivo labeling of mouse cells with
BrdU and Ki67
To detect cellular proliferation by labeling cells with BrdU in vivo,
10 mg/mL solution of BrdU (Sigma‐Aldrich) in 200 μL sterile PBS was
injected i.p. in mice, and BrdU was diluted to 0.8 mg/mL in the drinking
water that was provided daily after infection. Staining of cells for BrdU
incorporation was performed according to the BrdU Flow Kit (BD)
staining protocol. In brief, single‐cell suspensions were generated from
the gastric tissue as described above. After staining surface markers,
cells were fixed and permeabilized with Cytofix/Cytoperm Buffer for
15 minutes at room temperature. Cells were washed with 1 mL Perm/
Wash Buffer. After washing with Perm/Wash Buffer and centrifuging,
cells were incubated with Cytoperm Permeabilization Buffer Plus for
10 minutes on ice. After washing and centrifuging, cells were treated
with DNase I (0.3 mg/mL in PBS) for 1 hour at 37°C to expose incor‐
porated BrdU, and thereafter, they were stained with anti‐BrdU (BD)
and anti‐Ki67 (SolA15) in Perm/Wash Buffer for 20 minutes on ice.
Before re‐suspending in PBS for analysis by flow cytometry, cells were
washed with Perm/Wash Buffer and centrifuged.25
2.10 | Immunofluorescent staining and
hematoxylin‐eosin staining
The longitudinal specimens (from pylorus to squamous mucosa)
were fixed with 4% paraformaldehyde before embedded in paraf‐
fin, and then stained with hematoxylin and eosin (HE) to achieve
hematoxylin‐eosin staining as described in previous study. 22 As
 |
4 of 12       XU et al.
for immunofluorescent staining, 20‐ or 10‐μm frozen sections
were cut and dried at room temperature. Before staining with
anti‐CD4 antibody, these sections were blocked with confining
liquid containing 10% FBS, 0.1% Triton X‐100, and 0.1 M PBS. The
slides were washed and counterstained with DAPI to visualize cell
nuclei, and Panoramic 250 Flash III Scanner (3DHistech) was used
to acquire the images.
2.11 | FTY720 treatment
Immunized mice were i.p. injected with 1 mg/kg FTY720 (MCE) daily
until death to block circulating T‐cell egress from the secondary lym‐
phatic organ.
2.12 | Bacteria cultures
Helicobacter pylori SS1 was grown on H pylori solid medium (Qingdao
Hope Bio‐Technology Co., Ltd.) with 10% FBS and 1% H pylori bac‐
teriostat (including 0.3 mg vancomycin, 0.5 mg TMP, 1 mg nalidixic
acid, and 0.2 mg amphotericin B) under microaerophilic conditions
at 37°C for 2‐3 days. The bacteria were harvested and re‐suspended
in H pylori fluid medium to adjust the concentration about 1 × 109
colony‐forming units (CFUs) per milliliter before challenge (deter‐
mined by turbidimetry).
2.13 | H pylori challenge experiment and the
bacterial colonization
30 days after the vaccination, the mice were challenged with 1 × 109
CFUs H pylori SS1 (determined by turbidimetry) by p.o. after giv‐
ing 200 μL of 0.2% sodium bicarbonate solution. Twenty milligram
gastric tissue was cut, and gastric DNA was extracted by TIANamp
Genomic DNA Kit according to the manufacturer's instructions. The
bacterial colonization was performed using ChamQ Universal SYBR
qPCR Master Mix and the 16S DNA primers (P1‐H pylori 16S DNA:
5′‐TTTGTTAGAGAAGATAATGACGGTATCTAAC‐3′, P2‐H pylori 16S
DNA: 5′‐CATAGGATTTCACACCTGACTGACTATC‐3′). The H pylori
colonization is calculated according to quantitative curve by quan‐
titative real‐time PCR.
2.14 | Pro‐inflammatory myeloid cell depletion
experiments
Immunized mice were i.p. injected with a dose of 250μg anti‐Gr‐1
(BE0075) every 2 days until death to deplete inflammatory neutro‐
phils and mononuclear cells.
2.15 | Statistical analysis
Statistical analyses used GraphPad Prism 5.0 software. The differ‐
ences between all the groups were analyzed by using the ANOVA or
t test. P < .05 was considered statistically significant.
3 | R E S U LT S
3.1 | Vaccinating in GSL of wild‐type partner of the
parabiosis mice induced a pool of EGFP+ CD4+ TRM
cells in stomach
In order to distinguish vaccine‐induced CD4+ TRM cells, we adopted
parabiotic strategy that surgically joined EGFP with WT mice and
allowed them to share the peripheral circulation (Figure 1A). Three
weeks later, we observed bridging blood vessels across the incision
between the pairs macroscopically (Figure 1B). Flow cytometry
analysis showed approximately half of all CD4+ T cells in the blood,
spleen, and mesenteric lymph nodes of WT mice were EGFP+ cells
(Figure 1C), which were also readily appreciated on the smears under
the fluorescence microscope (Figure 1D). Expectedly, compared with
peripheral circulation, CD4+ EGFP+ T cells were scarcely presented
in the stomach (Figure 1C). Collectively, these data indicated that
shared peripheral circulatory system had been established in parabi‐
osis mice, which was conducive to probe the following experiments.
Gastric lymphocytes in mice were scarce compared to the in‐
testine by flow cytometry (Figure S1). Thus, we introduced vac‐
cine‐induced TRM cells to reinforce local immunity by vaccination
in the corpus. In the parabiotic model, the WT mice were vaccinated
with SF‐loaded H pylori vaccine in GSL (Figure 2A). Seven days later,
52.6% of CD4+ EGFP+ T cells were detected in the stomach of WT
mice (Figure 2B). Moreover, flow cytometry revealed that 6.68% an‐
tigen‐specific CD4+ EGFP+ T cells were present in vaccinated stom‐
ach after 7 days of vaccination, but adjuvant alone did not cause
obvious antigen‐specific Th1/Th17 cellular response (Figure 2C).
These data suggested that vaccination could recruit CD4+ EGFP+
effector T cells into stomach.
With the clearance of infection, the majority of vaccine‐induced
T cells that expanded to contribute to pathogen control die, but some
TRM cells survive in the resting stage.7,26 In our study, after 30 days
of vaccination, we observed a small fraction of CD4+ EGFP+ effec‐
tor T cells recruited in acute phase formed TRM cells in the resting
stage(Figure 2D). Otherwise, 76.6% of TRM cells in the stomach ex‐
pressed CD69, and only a minority expressed CD103 (Figure 2E).
In combination, these results suggested that vaccinating H pylori
vaccine in GSL was able to induce detectable antigen‐specific Th1/
Th17 cells and could form CD69+ CD103‐ TRM cells in the resting stage.
3.2 | CD4+ TRM cells proliferate in situ during
recall response
We next examined whether the gastric CD4+ EGFP+ TRM cells prolif‐
erated in situ during recall response. At day 30 after primary immuni‐
zation, parabiosis mice were administered a sphingosine‐1‐phosphate
analog (FTY720) intraperitoneally (i.p.) before challenging with H py-
lori (p.o.) (Figure 3A). Compared with PBS‐treated mice, decreasing
CD4+ lymphocytes in FTY720‐treated mice was observed in the
blood but not gastric tissue (Figure 3B), which proved that a single
XU et al. |
      5 of 12

F I G U R E 1   Establishment of a parabiosis system in mice. A, WT and EGFP mice established shared circulation by parabiotic surgery. B,
Blood vessels (arrows) developed across the suture site where the skin of WT and EGFP mice was conjoined (n = 5 pairs). C, After 3 weeks,
flow cytometry analysis revealed that the percentages of EGFP+ cells in all CD4+ cells of peripheral blood (BL), spleen (SP), mesenteric lymph
node (MLN), and stomach (ST) of a WT parabiont conjoined to an EGFP mice (n = 5 pairs). Representative FACS plots are shown. D, BL, SP,
and MLN smear showed the presence of EGFP+ BL, SP, and MLN cells in WT mice (n = 5 pairs). Original magnification ×200. All indicated P
values were tested by the ANOVAs. Data were shown as mean ± SD

dose of FTY720 prevented CD4+ lymphocytes in MLNs and spleen EGFP+ TRM cells (Figure 3C). It suggested that CD4+ EGFP+ TRM
27
from entering into blood, but remaining abundant in the stomach. cells formed after vaccination proliferated in situ to improve the
To detect the proliferation of CD4+ EGFP+ TRM cells, 2 mg BrdU local immunity of the stomach during recall response.
was injected in pairs (i.p.) after H pylori infection, and drinking water
containing 0.8 mg/ml BrdU was used daily for proliferation mon‐
3.3 | CD4+ TRM cells differentiate into effector Th1
itoring (Figure 3A). Strikingly, at day 7 post‐H pylori infection, the
cells against H pylori
mice that had been eliminated the circulatory lymphocytes showed
BrdU incorporation, which indicated that they were undergoing ac‐ Given that CD4+ EGFP+ TRM cells could proliferate in situ during re‐
tive DNA synthesis and proliferating in situ. Furthermore, another call response, however, the differentiation of CD4+ EGFP+ TRM cells
+
proliferation marker Ki67 was also significantly upregulated in CD4 was different from CD8+ TRM cells. IFN‐γ and IL‐17 were effector
 |
6 of 12       XU et al.
XU et al. |
      7 of 12

F I G U R E 2   GSL vaccination with H pylori vaccine induces antigen‐specific immune responses and the formation of EGFP+ CD4+ TRM
cells. A, After establishment of a parabiosis system, WT partner was vaccinated with H pylori vaccine CCF plus SF in GSL and induced
obvious inflammation at day 7. B, EGFP+ CD4+ T cells of SP, MLN, and ST were detected by flow cytometry (n = 5 pairs). C, The antigen‐
specific Th1 and Th17 cells in ST were measured at day 7 post‐GSL vaccination (n = 5 pairs). Representative FACS plots are shown. D,
Immunofluorescent staining showed the distribution of EGFP+ CD4+ T cells in acute stage (day 7) and TRM cells in resting stage (day 30)
(n = 5 pairs). E, The phenotype of EGFP+ CD4+ TRM cells was analyzed by flow cytometry (n = 5 pairs). All indicated P values were tested
using the ANOVAs or t tests. Data were shown as mean ± SD

cytokine of H pylori. To detect the differentiation of CD4+ EGFP+ TRM

cells, we challenged the WT pairing of parabiotic mice with H pylori
and also depleted circulating lymphocytes with FTY720 as described
above (Figure 4A). Seven days later, the flow cytometry revealed that
+ +
6.49% of CD4 EGFP TRM cells had differentiated and the majority
of differentiated CD4+ EGFP+ TRM cells secreted IFN‐γ (Figure 4B),
indicating that CD4+ EGFP+ TRM cells had a propensity to differenti‐
ate into effector Th1 cells in situ during H pylori recall response.
3.4 | Re‐activated CD4+ TRM cells are sufficient to
recruit pro‐inflammatory myeloid cells
Finally, we sought to determine whether CD4+ TRM cells contrib‐
uted to control H pylori infection in vaccinated mice as our predic‐
tion. We challenged H pylori (p.o.) in FTY720‐treated vaccinated mice
(Figure 5A). After 7 days, we found that the colonization of H pylori
in vaccinated mice was significantly reduced relative to unvaccinated

F I G U R E 3   CD4+ TRM cells proliferate in situ during recall response. A, After 30 days of vaccination, parabiotic mice were given FTY720
intraperitoneally (i.p.) for 3 days before H pylori (Hp) challenge. After H pylori infection, 2 mg BrdU was injected in pairs (i.p.), and drinking
water containing 0.8 mg/ml BrdU was used daily until the mice were euthanized. The gastric lymphocytes were stained with anti‐BrdU and
anti‐Ki67. B, The percentages of CD4+ lymphocytes in BL, SP, MLN, and ST after FTY720 administration were analyzed by flow cytometry
(n = 5 pairs). C, Flow cytometry showed the percentages of proliferation marker BrdU and Ki67 in EGFP+ CD4+ T cells at 7 days after H pylori
challenge (n = 5 pairs). The significance of differences in all indicated P values was tested using the ANOVAs. Data were shown as mean ± SD
 |
8 of 12       XU et al.

F I G U R E 4   CD4+ TRM cells differentiate into effector Th1 cells during recall response. A, After 30 days of vaccination, parabiotic mice
were treated with FTY720 (i.p.) for 3 days before H pylori (Hp) challenge. B, During H pylori recall response, Th1 and Th17 cells secreted
IFN‐γ and IL‐17, respectively. The percentages of antigen‐specific Th1 and Th17 cells of EGFP+ CD4+ T cells in 1 × 106 ST cells were detected
by flow cytometry (n = 5 pairs). Representative FACS plots are shown. The significance of differences in all indicated P values was tested
using the ANOVAs. Data were shown as mean ± SD

mice. But there was no significant difference in eliminating circulat‐
ing T cells or not (Figure 5B). These data investigated that H pylori
vaccine improved the local immunity of the stomach.
+ +
Pro‐inflammatory subsets of myeloid‐derived cells (CD45 CD11b )
28
were expanded in the gastric mucosa during H pylori recall response.
We next performed a flow cytometry analysis to detect pro‐inflam‐
matory myeloid cells. In recall response, vaccinated mice were with
higher Ly6G+ neutrophils and Ly6C+ inflammatory mononuclear cells
+ +
on CD45 CD11b cells (Figure 5C). Furthermore, histologic evaluation
also showed that the protective inflammation was more pronounced in
vaccinated mice (Figure 5D). More importantly, after eliminating pro‐
inflammatory myeloid cells by anti‐Gr‐1 (i.p.) (Figure 6A,B), 29,30 inflam‐
matory neutrophils and mononuclear cells decreased significantly and
the colonization of H pylori was higher than PBS‐treated vaccinated
group (Figure 6C). These data indicated CD4+ TRM cells mediated in
situ immunity by recruiting pro‐inflammatory myeloid cells.
4 |  D I S CU S S I O N
Vaccination with H pylori vaccine is an attractive strategy that can
prevent H pylori secondary infection by giving rise to immunologic
memory after gastric infection subside. In addition to TCM cells and
TEM cells, TRM cells are the third subset of memory T cell that resident
in the peripheral tissues for many months in mice to rapidly clear path‐
ogens encountered previously, especially in the peripheral tissues.31,32
Numerous studies showed that CD8+ TRM cells offered superior pro‐
tection against peripheral infection. But it was less known about CD4+
TRM cells, which might attribute to CD4+ TRM cells, which are less
efficiently generated in present mouse models.8 Fortunately, our pre‐
vious study demonstrated that CD4+ TRM cells could efficiently gen‐
erate by vaccination in GSL and protect the stomach from the H pylori
infection, but it did not distinguish vaccine‐induced CD4+ TRM cells
so that it could not persuasively explore whether the CD4+ TRM cells
themselves can induce protective effect.12
Adoptive transfer is a method to introduce exogenous cells by
intravenous injection.33,34 Many investigations reported that they
transfer antigen‐specific exogenous effector cells into host and
these effector cells would proliferate in corresponding tissue after
immunization.11,15,16,31 However, it was too negligible in host to be
detected during resting stage if we transfer naive splenic lympho‐
cyte. Therefore, in this study, we established parabiotic pairs to
introduce EGFP+ cells to WT mice.19 We detected relatively consid‐
erable EGFP+ cells in the blood, spleen, mesenteric lymph nodes, and
XU et al. |
      9 of 12

F I G U R E 5   CD4+ TRM cells mediate immunity in situ by recruiting pro‐inflammatory myeloid cells. A, Mice were vaccinated with H pylori
vaccine in GSL. Thirty days later, vaccinated mice were challenged with H pylori. Daily i.p. injection of FTY720 was performed. B, The
colonization of H pylori in gastric tissue was examined by RT‐qPCR (n = 8). C, Ly6G+ neutrophils and Ly6C+ inflammatory mononuclear cells
on CD45+ CD11b+ cells were detected by flow cytometry (n = 8). Representative FACS plots are shown. D, Representative H&E staining in
gastric lumina is shown, black area (arrow) represented positive cells (n = 8). All indicated P values were tested using the ANOVAs. Data were
shown as mean ± SD

stomach of vaccinated WT mice. The parabiotic mice allowed us to
rigorously distinguish vaccine‐induced CD4+ TRM cells. In the first
place, we detected about half of all CD4+ T cells were EGFP+ cells in
peripheral circulation of WT mice. After GSL vaccination, these cells
+ +
differentiate into CD4 EGFP effector T cells in the stomach. After
+ +
the inflammation resolution, the CD4 EGFP effector T cells formed
EGFP+ CD69+ CD103‐ TRM cells. Finally, we focused on the local
immunity of CD4+ EGFP+ TRM cells in H pylori immune responses
independently of circulating memory T cells.
The current set of experiments extend the work of others who
have examined that the CD8+ TRM cells enhance the local immu‐
nity of tissues (such as skin, genital tract, and small intestine) by
proliferation in situ when antigens are encountered again.15,16,35-37
In line with previous studies of CD8+ TRM cells, we also found the
proliferation markers BrdU and Ki67 were significantly upregulated
in the gastric CD4+ EGFP+ TRM cells after FTY720 administration.
Importantly, we showed that CD4+ EGFP+ TRM cells mainly dif‐
ferentiated into Th1 cells during recall response. Consistent with
 |
10 of 12       XU et al.

F I G U R E 6   The colonization of H pylori in gastric tissue is increased in the absence of pro‐inflammatory myeloid cells. A, Mice were
vaccinated with H pylori vaccine in GSL. Thirty days later, anti‐Gr‐1 was injected by i.p. in vaccinated mice before H pylori challenge. Daily
i.p. injection of FTY720 was performed. B, After anti‐Gr‐1 administration, the percentages of Ly6G+ neutrophils and Ly6C+ mononuclear
cells in BL, SP, MLN, and ST of vaccinated mice were analyzed by flow cytometry (n = 8). C, The colonization of H pylori in gastric tissue was
examined by RT‐qPCR (n = 8). All indicated P values were tested using the ANOVAs. Data were shown as mean ± SD

our finding, in a Leishmania major‐immune mice model, CD4+ TRM
rapidly upregulated IFN‐γ during recall responses.38 Although our
+
study points strongly that proliferating CD4 TRM cells can scav‐
enge H pylori, we also acknowledge that circulating memory cells
are also effective in recall response. Previous investigations demon‐
strated TRM cells could respond to pathogens in situ and produce
cytokines including IFN‐γ to enhance the recruitment of circulating
T cells, which provided enhanced immune protection. 38,39 However,
in contrast to TRM cells, circulating memory T cells alone could not
control pathogen infection in peripheral tissues, such as in skin.31
In prophylactic mice model, a previous study highlighted the
role of post‐immunization inflammation in the vaccine‐induced re‐
call response.40 In addition, many investigations revealed that post‐
immunization by H pylori vaccine could give rise to inflammation of
41-44
the stomach. Our previous study also showed that H pylori vac‐
cine induced defense by expanding pro‐inflammatory myeloid‐de‐
28
rived cells during recall response. Remarkably, here, the H pylori
colonization in stomach was significantly reduced and the transient
protective inflammation was striking in the FTY720‐treated vac‐
+
cinated mice. It seemed that CD4 TRM cells indeed played an ef‐
fective role in H pylori elimination by recruiting pro‐inflammatory
myeloid cells during recall response. At least in the stomach, CD4+
TRM cells themselves can enhance gastric immunity, which pro‐
vides a support for the design of the future H pylori vaccines.
These results are clinically relevant to the vaccine strategy.
Different vaccines may induce different TRM cells to prevent vari‐
ous diseases. It is promising if vaccine can specifically generate TRM
cells in all peripheral tissue by convenient methods; for example,
ingestible self‐orienting system for oral delivery of vaccine prepa‐
ration may be able to generate specific TRM cells in gastrointestinal
tract.45 The development of vaccine strategies involves a TRM cell
population, which may shed new light on the development of H pylori
vaccines. However, although TRM cells formed by vaccination can
proliferate and differentiate in situ, the deeper mechanism remains
to be explored. Moreover, it is unfortunate that gastric TRM cells
are difficult to explore in vitro due to the numerical disadvantage
and cultivation difficulty. The study provides an experimental and
biologic rationale for their further evaluation.
5 | CO N C LU S I O N S
In summary, the parabiotic strategy revealed that EGFP+ CD4+ T
cells were recruited into the stomach of WT mice by H pylori vac‐
cination in GSL. After the elimination of antigen, EGFP+ CD4+ effec‐
tor T cells formed CD69+ CD103‐ TRM cells. During recall response,
EGFP+ CD4+ TRM cells rapidly proliferated and differentiated to
resist H pylori infection independently on the circulating memory T
cells, which improved the local immunity of the stomach.
AC K N OW L E D G E M E N T S
We extend our appreciation to Linqi Zhao for her careful drawing of
the mice in the figure of this manuscript.
XU et al. |
      11 of 12

C O N FL I C T O F I N T E R E S T 14. Lanzavecchia A, Sallusto F. Dynamics of T lymphocyte re‐

sponses: intermediates, effectors, and memory cells. Science.
The authors declare that they have no conflict of interest. 2000;290(5489):92‐97.
15. Beura LK, Mitchell JS, Thompson EA, et al. Intravital mucosal im‐
aging of CD8(+) resident memory T cells shows tissue‐autonomous
AU T H O R S ' C O N T R I B U T I O N S recall responses that amplify secondary memory. Nat Immunol.
2018;19(2):173‐182.
YX, TX, and WL conceived and designed the project. NX performed 16. Park SL, Zaid A, Hor JL, et al. Local proliferation maintains a sta‐
most experiments and data analysis. GR, CH, AH, ZZ, SL, ZZ, MF, and ble pool of tissue‐resident memory T cells after antiviral recall re‐
FY participated in some experiments. WL and NX wrote the paper. sponses. Nat Immunol. 2018;19(2):183‐191.
17. Stary G, Olive A, Radovic‐Moreno AF, et al. A mucosal vaccine
All authors contributed to discussion of the results, reviewed the
against Chlamydia trachomatis generates two waves of protective
manuscript, and approved the final article. memory T cells. Science. 2015;348(6241):aaa8205.
18. Perdomo C, Zedler U, Kuhl AA, et al. Mucosal BCG vaccination in‐
duces protective lung‐resident memory T cell populations against
ORCID tuberculosis. MBio. 2016;7(6):e01686‐16.
19. Kamran P, Sereti KI, Zhao P, Ali SR, Weissman IL, Ardehali R.
Tao Xi  https://orcid.org/0000-0001-8065-2639 Parabiosis in mice: a detailed protocol. J Vis Exp. 2013;(80).
Yingying Xing  https://orcid.org/0000-0001-5566-3096 20. Sung J, Sodhi CP, Voltaggio L, et al. The recruitment of extra‐in‐
testinal cells to the injured mucosa promotes healing in radiation
enteritis and chemical colitis in a mouse parabiosis model. Mucosal
Immunol. 2019;12(2):503‐517.
REFERENCES
21. Jiang X, Clark RA, Liu L, Wagers AJ, Fuhlbrigge RC, Kupper TS. Skin
infection generates non‐migratory memory CD8+ T(RM) cells pro‐
1. Talley NJ, Zinsmeister AR, Weaver A, et al. Gastric adenocar‐
viding global skin immunity. Nature. 2012;483(7388):227‐231.
cinoma and Helicobacter pylori infection. J Natl Cancer Inst.
22. Guo LE, Li X, Tang F, et al. Immunological features and the abil‐
1991;83(23):1734‐1739.
ity of inhibitory effects on enzymatic activity of an epitope vac‐
2. Sokic‐Milutinovic A, Alempijevic T, Milosavljevic T. Role of
cine composed of cholera toxin B subunit and B cell epitope from
Helicobacter pylori infection in gastric carcinogenesis: cur‐
Helicobacter pylori urease A subunit. Appl Microbiol Biotechnol.
rent knowledge and future directions. World J Gastroenterol.
2012;93(5):1937‐1945.
2015;21(41):11654‐11672.
23. Song H, Lv X, Yang J, et al. A novel chimeric flagellum fused with the
3. Parsonnet J, Friedman GD, Vandersteen DP, et al. Helicobacter
multi‐epitope vaccine CTB‐UE prevents Helicobacter pylori‐induced
pylori infection and the risk of gastric carcinoma. N Engl J Med.
gastric cancer in a BALB/c mouse model. Appl Microbiol Biotechnol.
1991;325(16):1127‐1131.
2015;99(22):9495‐9502.
4. Mbulaiteye SM, Hisada M, El‐Omar EM. Helicobacter pylori associ‐
24. Zeng Z, Liu W, Luo S, et al. Shape of gastrointestinal immunity
ated global gastric cancer burden. Front Biosci. 2009;14:1490‐1504.
with non‐genetically modified Lactococcus lactis particles requires
5. Gebhardt T, Wakim LM, Eidsmo L, Reading PC, Heath WR, Carbone
commensal bacteria and myeloid cells‐derived TGF‐beta1. Appl
FR. Memory T cells in nonlymphoid tissue that provide enhanced
Microbiol Biotechnol. 2019;103(9):3847‐3861.
local immunity during infection with herpes simplex virus. Nat
25. Khan TN, Mooster JL, Kilgore AM, Osborn JF, Nolz JC. Local anti‐
Immunol. 2009;10(5):524‐530.
gen in nonlymphoid tissue promotes resident memory CD8+ T cell
6. Wakim LM, Woodward‐Davis A, Bevan MJ. Memory T cells per‐
formation during viral infection. J Exp Med. 2016;213(6):951‐966.
sisting within the brain after local infection show functional ad‐
26. Bevan MJ. Memory T cells as an occupying force. Eur J Immunol.
aptations to their tissue of residence. Proc Natl Acad Sci USA.
2011;41(5):1192‐1195.
2010;107(42):17872‐17879.
27. Matloubian M, Lo CG, Cinamon G, et al. Lymphocyte egress from
7. Gebhardt T, Whitney PG, Zaid A, et al. Different patterns of pe‐
thymus and peripheral lymphoid organs is dependent on S1P recep‐
ripheral migration by memory CD4+ and CD8+ T cells. Nature.
tor 1. Nature. 2004;427(6972):355‐360.
2011;477(7363):216‐219.
28. Luo S, Liu W, Zeng Z, et al. Toxic adjuvants alter the function
8. Park CO, Kupper TS. The emerging role of resident memory T
and phenotype of dendritic cells to initiate adaptive immune re‐
cells in protective immunity and inflammatory disease. Nat Med.
sponses induced by oral Helicobacter pylori vaccines. Helicobacter.
2015;21(7):688‐697.
2018;23(6):e12536.
9. Mueller SN, Mackay LK. Tissue‐resident memory T cells: local spe‐
29. Brown CR, Blaho VA, Loiacono CM. Treatment of mice with
cialists in immune defence. Nat Rev Immunol. 2016;16(2):79‐89.
the neutrophil‐depleting antibody RB6‐8C5 results in early de‐
10. Gray JI, Westerhof LM, MacLeod M. The roles of resident, central
velopment of experimental lyme arthritis via the recruitment
and effector memory CD4 T‐cells in protective immunity following
of Gr‐1‐ polymorphonuclear leukocyte‐like cells. Infect Immun.
infection or vaccination. Immunology. 2018;154:574‐581.
2004;72(9):4956‐4965.
11. Ariotti S, Beltman JB, Chodaczek G, et al. Tissue‐resident memory
3 0. Czuprynski CJ, Brown JF, Maroushek N, Wagner RD, Steinberg
CD8+ T cells continuously patrol skin epithelia to quickly recognize
H. Administration of anti‐granulocyte mAb RB6‐8C5 impairs the
local antigen. Proc Natl Acad Sci USA. 2012;109(48):19739‐19744.
resistance of mice to Listeria monocytogenes infection. J Immunol.
12. Liu W, Zeng Z, Luo S, et al. Gastric subserous vaccination with
1994;152(4):1836‐1846.
Helicobacter pylori vaccine: an attempt to establish tissue‐resi‐
31. Mackay LK, Stock AT, Ma JZ, et al. Long‐lived epithelial immu‐
dent CD4+ memory T cells and induce prolonged protection. Front
nity by tissue‐resident memory T (TRM) cells in the absence of
Immunol. 2019;10:1115.
persisting local antigen presentation. Proc Natl Acad Sci USA.
13. Pritchard EM, Kaplan DL. Silk fibroin biomaterials for controlled re‐
2012;109(18):7037‐7042.
lease drug delivery. Expert Opin Drug Deliv. 2011;8(6):797‐811.
 |
12 of 12       XU et al.

32. Mackay LK, Rahimpour A, Ma JZ, et al. The developmental path‐
way for CD103(+)CD8+ tissue‐resident memory T cells of skin. Nat
Immunol. 2013;14(12):1294‐1301.
33. Chen ZK, Cobbold SP, Waldmann H, Metcalfe S. Amplification of
natural regulatory immune mechanisms for transplantation toler‐
ance. Transplantation. 1996;62(9):1200‐1206.
3 4. Villani V, Yamada K, Scalea JR, et al. Adoptive transfer of renal
allograft tolerance in a large animal model. Am J Transplant.
2016;16(1):317‐324.
35. Bachmann MF, Wolint P, Schwarz K, Oxenius A. Recall prolifera‐
tion potential of memory CD8+ T cells and antiviral protection. J
Immunol. 2005;175(7):4677‐4685.
36. Çuburu N, Graham BS, Buck CB, et al. Intravaginal immunization
with HPV vectors induces tissue‐resident CD8+ T cell responses. J
Clin Invest. 2012;122(12):4606‐4620.
37. Wakim LM, Waithman J, van Rooijen N, Heath WR, Carbone FR.
Dendritic cell‐induced memory T cell activation in nonlymphoid tis‐
sues. Science. 2008;319(5860):198‐202.
38. Glennie ND, Yeramilli VA, Beiting DP, Volk SW, Weaver CT, Scott
P. Skin‐resident memory CD4+ T cells enhance protection against
Leishmania major infection. J Exp Med. 2015;212(9):1405‐1414.
39. Schenkel JM, Fraser KA, Vezys V, Masopust D. Sensing and
alarm function of resident memory CD8(+) T cells. Nat Immunol.
2013;14(5):509‐513.
4 0. Zhang E, Kosinska AD, Ma Z, et al. Woodchuck hepatitis virus core
antigen‐based DNA and protein vaccines induce qualitatively dif‐
ferent immune responses that affect T cell recall responses and an‐
tiviral effects. Virology. 2015;475:56‐65.
41. Garhart CA, Heinzel FP, Czinn SJ, Nedrud JG. Vaccine‐induced re‐
duction of Helicobacter pylori colonization in mice is interleukin‐12
dependent but gamma interferon and inducible nitric oxide syn‐
thase independent. Infect Immun. 2003;71(2):910‐921.
42. Raghavan S, Fredriksson M, Svennerholm AM, Holmgren J, Suri‐
Payer E. Absence of CD4+CD25+ regulatory T cells is associated with
a loss of regulation leading to increased pathology in Helicobacter
pylori‐infected mice. Clin Exp Immunol. 2003;132(3):393‐400.
43. Matsumoto Y, Blanchard TG, Drakes ML, et al. Eradication of
Helicobacter pylori and resolution of gastritis in the gastric mucosa
of IL‐10‐deficient mice. Helicobacter. 2005;10(5):407‐415.
4 4. Michetti P, Corthésy‐Theulaz I, Davin C, et al. Immunization of
BALB/c mice against Helicobacter felis infection with Helicobacter
pylori urease. Gastroenterology. 1994;107(4):1002‐1011.
45. Abramson A, Caffarel‐Salvador E, Khang M, et al. An ingestible
self‐orienting system for oral delivery of macromolecules. Science.
2019;363(6427):611‐615.
S U P P O R T I N G I N FO R M AT I O N
Additional supporting information may be found online in the
Supporting Information section at the end of the article. 
How to cite this article: Xu N, Ruan G, Liu W, et al. Vaccine‐
induced gastric CD4+ tissue‐resident memory T cells
proliferate in situ to amplify immune response against
Helicobacter pylori insult. Helicobacter. 2019;00:e12652.
https​://doi.org/10.1111/hel.12652​