Helicobacter pylori, Protected from Antibiotics and Stresses Inside Candida albicans Vacuoles, Cause Gastritis in Mice

Due to (i) the simultaneous presence of Helicobacter pylori (ulcer-induced bacteria) and Candida albicans in the stomach and (ii) the possibility of prokaryotic–eukaryotic endosymbiosis (intravacuolar H. pylori in the yeast cells) under stresses, we tested this symbiosis in vitro and in vivo. To that end, intravacuolar H. pylori were induced by the co-incubation of C. albicans with H. pylori under several stresses (acidic pH, non-H. pylori-enrichment media, and aerobic environments); the results were detectable by direct microscopy (wet mount) and real-time polymerase chain reaction (PCR). Indeed, intravacuolar H. pylori were predominant under all stresses, especially the lower pH level (pH 2–3). Interestingly, the H. pylori (an amoxicillin-sensitive strain) inside C. albicans were protected from the antibiotic (amoxicillin), while extracellular H. pylori were neutralizable, as indicated by the culture. In parallel, the oral administration of intravacuolar H. pylori in mice caused H. pylori colonization in the stomach resulting in gastritis, as indicated by gastric histopathology and tissue cytokines, similar to the administration of free H. pylori (extra-Candida bacteria). In conclusion, Candida protected H. pylori from stresses and antibiotics, and the intravacuolar H. pylori were able to be released from the yeast cells, causing gastric inflammation with neutrophil accumulations.


Introduction
Peptic ulcer disease and gastritis are inflammations of the mucosal and muscular layers [1,2] of the stomach and proximal duodenum that lead to several findings, including stomachache, indigestion, nausea with loss of appetite, malnutrition, positive stool occult blood, and hematemesis (in severe cases) [2][3][4]. Indeed, peptic ulcer is a worldwide health care problem, with the highest incidence in developing countries [5]; it is caused both by Helicobacter pylori (H. pylori) infection (50% of cases) and by the chronic use of non-steroidal anti-inflammatory drugs (NSAIDs) (25% of patients) [6,7]. Notably, the possible fecal-oral and oral-oral transmissions of H. pylori that occur among populations with poor socioeconomic conditions lead to a higher incidence of gastritis in developing countries [5]. H. pylori (previously named Campylobacter pylori) is a gram-negative spiral (helical) bacterium [8] that causes several lesions in the human gastrointestinal (GI) tract, including asymptomatic gastritis (85%), chronic gastric inflammation (15%), and gastroduodenal ulcer disease (less than 1%) that are associated with gastric carcinoma and intestinal metaplasia [9,10]. Interestingly, H. pylori are transmittable from person to person not only through saliva but also by

Induction of Intravacuolar H. pylori in Candida Yeast Cells
The endosymbiosis of H. pylori in yeast cells was indicated by the presence of H. pylori in Candida cytosols, referred to as "intravacuolar H. pylori", as previously mentioned [45,46]. As such, H. pylori and C. albicans were co-incubated under different conditions, including the bacteria-Candida ratio, pH of the media, and duration of incubation); the result was the identification of H. pylori inside Candida yeast cells, as illustrated by bacteria-like bodies (BLBs) (dense black dots in a state of movement) inside the yeast cells [40,41,43,[47][48][49][50], using bright-field microscopy ( Figure 1). With continuous time-frame illustrations, the movements of the dense black dots could be observed (Figure 1), supporting the possible existence of BLBs inside the yeast cells. Interestingly, BLBs were detectable after the H. pylori-C. albicans co-incubation but not in the C. albicans culture alone (Figures 1 and 2A). The higher density of H. pylori in BLBs may be responsible for the dense, black-colored dots observed in the cytosol of the co-incubated culture (Figure 2A lower), whereas the lower density of Candida's organelles may explain the non-colored cytosol of the Candida culture without bacteria (Figure 2A middle). The highest abundance of intravacuolar H. pylori was demonstrated in a 1:100 ratio of H. pylori to Candida (1 × 10 8 and 1 × 10 6 CFU/mL of H. pylori and C. albicans, respectively) in Sabouraud dextrose broth (SB) media at pH 2 after 3 h incubation. There was a decrease in intravacuolar H. pylori in the sub-cultures starting from the 2nd generation ( Figure 2B,C). At pH 2, intravacuolar H. pylori were detectable for all H. pylori: Candida ratios after a 5 h incubation, but only at the 1:100 ratio after a 3 h incubation ( Figure 2B). At pH 3, intravacuolar H. pylori were detectable only after a 5 h incubation, and only for the ratios 1:1 and 1:10 ( Figure 2B). Meanwhile, at pH 4 and 5, the 1:10 ratio was only the condition that could induce intravacuolar H. pylori ( Figure 2B). Hence, the stomach (pH 1.5-3.5) is possibly the most suitable organ for the formation of intravacuolar H. pylori in Candida. This formation may depend on the abundance of Candida in the stomach and the gastric emptying time. The intravacuolar H. pylori were able to pass through the daughter yeast cells as there were similar abundances of H. pylori in the 1st and 2nd generations of the culture; however, H. pylori abundance was reduced in the 3rd and 4th generations ( Figure 2C). Notably, incubation by SB (a Candida-enrichment media) in aerobic conditions (conditions for enhancing Candida growth) was the environmental stress for H. pylori.
result was the identification of H. pylori inside Candida yeast cells, as illustrated by bacteria-like bodies (BLBs) (dense black dots in a state of movement) inside the yeast cells [40,41,43,[47][48][49][50], using bright-field microscopy ( Figure 1). With continuous time-frame illustrations, the movements of the dense black dots could be observed (Figure 1), supporting the possible existence of BLBs inside the yeast cells. Interestingly, BLBs were detectable after the H. pylori-C. albicans co-incubation but not in the C. albicans culture alone (Figures 1 and 2A). The higher density of H. pylori in BLBs may be responsible for the dense, black-colored dots observed in the cytosol of the co-incubated culture (Figure 2A lower), whereas the lower density of Candida's organelles may explain the non-colored cytosol of the Candida culture without bacteria (Figure 2A middle). The highest abundance of intravacuolar H. pylori was demonstrated in a 1:100 ratio of H. pylori to Candida (1 × 10 8 and 1 × 10 6 CFU/mL of H. pylori and C. albicans, respectively) in Sabouraud dextrose broth (SB) media at pH 2 after 3 h incubation. There was a decrease in intravacuolar H. pylori in the sub-cultures starting from the 2nd generation ( Figure 2B,C). At pH 2, intravacuolar H. pylori were detectable for all H. pylori: Candida ratios after a 5 h incubation, but only at the 1:100 ratio after a 3 h incubation ( Figure 2B). At pH 3, intravacuolar H. pylori were detectable only after a 5 h incubation, and only for the ratios 1:1 and 1:10 ( Figure 2B). Meanwhile, at pH 4 and 5, the 1:10 ratio was only the condition that could induce intravacuolar H. pylori ( Figure 2B). Hence, the stomach (pH 1.5-3.5) is possibly the most suitable organ for the formation of intravacuolar H. pylori in Candida. This formation may depend on the abundance of Candida in the stomach and the gastric emptying time. The intravacuolar H. pylori were able to pass through the daughter yeast cells as there were similar abundances of H. pylori in the 1st and 2nd generations of the culture; however, H. pylori abundance was reduced in the 3rd and 4th generations ( Figure 2C). Notably, incubation by SB (a Candida-enrichment media) in aerobic conditions (conditions for enhancing Candida growth) was the environmental stress for H. pylori.   Characteristics of intravacuolar H. pylori inside Candida yeast cells under a bright-field microscope, 10 fields/slide (100× magnification), compared with the controls H. pylori and C. albicans alone (A). The real-time polymerase chain reaction (PCR) based on CagA gene expression demonstrates intravacuolar H. pylori in different conditions (the H. pylori vs. Candida ratio, pH media, and duration of incubation) compared with control (B). Intravacuolar H. pylori in each generation after the sub-culture of Candida yeast cells (C) (n = 9/group). Independent triplicate experiments were performed. *, p < 0.05; ϕ, p < 0.05 vs. others; BLBs, bacteria-like bodies; ctrl, control; Gen, generation of the Candida culture.

Release of Intravacuolar H. pylori from Candida Yeast Cells
While the endosymbiosis of H. pylori in Candida yeast cells was inducible by several stresses (high oxygen levels and a non-H. pylori-enrichment media) (Figure 2A-C), we Candida ratio, pH media, and duration of incubation) compared with control (B). Intravacuolar H. pylori in each generation after the sub-culture of Candida yeast cells (C) (n = 9/group). Independent triplicate experiments were performed. *, p < 0.05; φ, p < 0.05 vs. others; BLBs, bacteria-like bodies; ctrl, control; Gen, generation of the Candida culture.

Release of Intravacuolar H. pylori from Candida Yeast Cells
While the endosymbiosis of H. pylori in Candida yeast cells was inducible by several stresses (high oxygen levels and a non-H. pylori-enrichment media) (Figure 2A-C), we further tested whether the intravacuolar H. pylori could be released. Without any further manipulations of the Candida containing intravacuolar H. pylori, the culture of these Candida yeast cells in urea-based agar demonstrated urease activity (i.e., turning the color of the media from yellow into pink) at 3 and 5 days after the culture ( Figure 3A) with H. pylori detectable by qPCR ( Figure 3B). However, the breakdown of the yeast cell wall (with sonication) facilitated the release of H. pylori from Candida. There was a higher abundance of H. pylori released from sonicated Candida yeast cells with intravacuolar H. pylori; bacterial abundance after a 3-day-culture in urea-based agar using the sonicated H. pylori-containing Candida cells was higher than it was in the non-sonicated samples, as determined by the color of the agar (urease test) and qPCR assay ( Figure 3C,D). further tested whether the intravacuolar H. pylori could be released. Without any further manipulations of the Candida containing intravacuolar H. pylori, the culture of these Candida yeast cells in urea-based agar demonstrated urease activity (i.e., turning the color of the media from yellow into pink) at 3 and 5 days after the culture ( Figure 3A) with H. pylori detectable by qPCR ( Figure 3B). However, the breakdown of the yeast cell wall (with sonication) facilitated the release of H. pylori from Candida. There was a higher abundance of H. pylori released from sonicated Candida yeast cells with intravacuolar H. pylori; bacterial abundance after a 3-day-culture in urea-based agar using the sonicated H. pylori-containing Candida cells was higher than it was in the non-sonicated samples, as determined by the color of the agar (urease test) and qPCR assay ( Figure 3C,D).

Intravacuolar H. pylori Was Protected from Antibiotics and Stress Conditions
Because a possible benefit of the H. pylori endosymbiosis inside C. albicans may be protection from antibiotics and stresses, the culture conditions of Candida using SB (the non-enrichment media of H. pylori) in aerobic conditions (a stress factor for H. pylori) with or without amoxicillin (a representative antibiotic) were tested against Candida containing H. pylori. Amoxicillin had no effect on Candida containing H. pylori as the antibiotic could not reduce the yeast colonies ( Figure 4A upper), even though the selected strain of H. pylori was sensitive to aerobic conditions (H. pylori in Columbia blood agar in aerobic conditions) ( Figure 4A middle) and to amoxicillin (no bacterial colony in Columbia blood agar in microaerophilic conditions) ( Figure 4A lower). Notably, H. pylori were sensitive to amoxicillin in either low or high concentrations ( Figure 4A lower). Likewise, the bacterial abundance of H. pylori based on CagA gene expression demonstrated the highest H. pylori abundance in Candida with intravacuolar H. pylori without amoxicillin, despite an aerobic environment (an environment toxic to H. pylori) ( Figure 4B). In Candida containing intravacuolar H. pylori with amoxicillin, H. pylori abundance was lower than in the non-antibiotic group, but there was a non-difference between high versus low concentrations of amoxicillin ( Figure 4B). Meanwhile, H. pylori were non-detectable in the samples of free H. pylori (H. pylori alone) under aerobic conditions and microaerophilic conditions with amoxicillin ( Figure 4B). The reduced H. pylori abundance in amoxicillin-incubated Candida containing H. pylori may reflect the effect of amoxicillin on spontaneously released H. pylori.

Intravacuolar H. pylori Was Protected from Antibiotics and Stress Conditions
Because a possible benefit of the H. pylori endosymbiosis inside C. albicans may be protection from antibiotics and stresses, the culture conditions of Candida using SB (the non-enrichment media of H. pylori) in aerobic conditions (a stress factor for H. pylori) with or without amoxicillin (a representative antibiotic) were tested against Candida containing H. pylori. Amoxicillin had no effect on Candida containing H. pylori as the antibiotic could not reduce the yeast colonies ( Figure 4A upper), even though the selected strain of H. pylori was sensitive to aerobic conditions (H. pylori in Columbia blood agar in aerobic conditions) ( Figure 4A middle) and to amoxicillin (no bacterial colony in Columbia blood agar in microaerophilic conditions) ( Figure 4A lower). Notably, H. pylori were sensitive to amoxicillin in either low or high concentrations ( Figure 4A lower). Likewise, the bacterial abundance of H. pylori based on CagA gene expression demonstrated the highest H. pylori abundance in Candida with intravacuolar H. pylori without amoxicillin, despite an aerobic environment (an environment toxic to H. pylori) ( Figure 4B). In Candida containing intravacuolar H. pylori with amoxicillin, H. pylori abundance was lower than in the non-antibiotic group, but there was a non-difference between high versus low concentrations of amoxicillin ( Figure 4B). Meanwhile, H. pylori were non-detectable in the samples of free H. pylori (H. pylori alone) under aerobic conditions and microaerophilic conditions with amoxicillin ( Figure 4B). The reduced H. pylori abundance in amoxicillin-incubated Candida containing H. pylori may reflect the effect of amoxicillin on spontaneously released H. pylori.

Candida Containing Intravacuolar H. pylori Caused Peptic Ulcer Disease
Although H. pylori contained inside Candida yeast cells have been previously reported [40][41][42], the clinical importance is still not clear. To test the clinical impacts of intravacuolar H. pylori in Candida yeast cells, a gastritis mouse model was conducted using the daily oral administration of vehicle control, H. pylori alone, C. albicans alone, and Candida containing intravacuolar H. pylori for 12 weeks ( Figure 5A, schema). Results showed that H. pylori administration induced transient weight loss, especially at 2-6 weeks of administration; then, body weight increased to a level similar to pre-administration weight ( Figure 5B). Meanwhile, there was no weight change in the other groups (control, Candida alone, and Candida containing H. pylori) during the 12 weeks of the experiment ( Figure 5B). Despite the improved body weight at 12 weeks after H. pylori administration, colonization by H. pylori was detected in the stomach, as indicated by urease activity (a pink color change in urea-based agar) and CagA gene expression in samples taken from the mouse stomach ( Figure 5C,D). Likewise, H. pylori were also detected in the stomach of mice after 12 weeks of the administration of Candida containing intravacuolar H. pylori ( Figure 5C,D). In parallel, H. pylori-administered mice demonstrated gastritis with inflammatory responses ( Figure 6A-E), as indicated by the prominent histological scores based on inflammatory cell infiltration and epithelial damage ( Figures 6A,B and 7A,B) and increased inflammatory cytokines (IL-6 and TNF-α, but not IL-10) in the stomach ( Figure 6C-E). On the other hand, the administration of Candida containing H. pylori caused less extreme histological scores and levels of tissue cytokines in the stomach than the administration of H. pylori alone after 12 weeks of the experiment ( Figure 6A,B). There were no gastritis lesions in mice with daily fasting with 1xPBS (control) or Candida administration, although the levels of stomach pro-inflammatory cytokines (IL-6 and TNF-α) were higher in fasting mice with PBS or C. albicans than in the non-fasting control mice (normal mice) ( Figure 6C-E); nevertheless, the levels observed in the fasting mice were lower than those of mice administered with either H. pylori or Candida containing H. pylori ( Figure 6C-E). There was an abundance of inflammatory cells in the gut (submucosa and mucosa) in mice with either H. pylori alone or Candida containing H. pylori compared to the absence of injury in other groups ( Figure 6A,B), despite a tendency toward more prominent submucosal neutrophils and epithelial mononuclear cells (Figure 7).     The histological pictures of control mice and those who received Candida administration alone are not shown due to the non-difference compared to the fasting control group. The injury scores of these pictures are displayed in Figure 6A,B. A more prominent submucosal accumulation of neutrophils (polymorphonuclear cells with eosinophilic color staining) (yellow arrows) and epithelial mononuclear cells (red arrows) was observed in the H. pylori group than in the intravacuolar H. pylori mice.

Discussion
Intravacuolar H. pylori inside C. albicans were protected from antibiotics (and environmental stresses) and could cause gastritis after the release of H. pylori from the yeast cells, illustrating the importance of natural bacteria-fungi symbiosis.

The Transfer of H. pylori Inside the Vacuoles of Candida Yeast Cells
H. pylori are vulnerable to a non-acidic micro-environment, and they cannot be isolated from food, water, or non-stomach specimens (saliva and feces). Indeed, the gastric epithelium and immune cells (macrophages and dendritic cells) are the only eukaryotic cells that host H. pylori (the facultative intracellular bacteria) [41]. As such, viable intracellular prokaryotes inside eukaryotes are a prokaryotic adaptation to a wide range of environmental stresses [41]. H. Pylori can be established not only in eukaryotic human cells but also in eukaryotic microorganisms (e.g., Candida spp.) as a form of "prokaryoticeukaryotic endosymbiosis" [41,45,47] in which the endosymbiotic bacteria are localized inside the membrane-bound vacuole [40,41,47,48]. Likewise, several examples of the endosymbiosis of endobacteria inside fungi have been described [51]. For example, Rhizopus microcarpus, a plant pathogen fungus, provide micronutrients, such as phosphate, to Burkholderia rhizoxinica that stay inside the fungi [51]. Perhaps H. pylori's specific ergosterol interacts with the yeast vacuole to allow the viability of "intravacuolar H. pylori" [41]. Because Candida yeast is exceptionally resistant to stressful conditions [52,53], the yeast cell can be used as an alternative host to protect H. pylori from stress outside the stomach, offer nourishment, and serve as a conduit for the bacterium's spread within human populations and in the environment [41]. Indeed, the mutual cooperation between Candida spp. and H. pylori was previously mentioned; the exosymbiosis between the extracellular form of Candida yeast and H. pylori results in the synergistic production of an interkingdom biofilm [45]. However, the endosymbiosis of intravacuolar H. pylori inside yeast cells may be even more complex than the synergy observed in biofilm production because it indicates an evolution in the symbiosis between the structures of these organisms.
Here, within the culture media of fungi that were too acidic for H. pylori, had fewer nutrients, or were non-microaerophilic, H. pylori could form bacteria-like bodies (BLBs) inside C. albicans vacuoles, as supported by a previous publication [46]. Notably, the differences between the BLBs and the non-specific components of the yeast cell's cytosol were the movement and the density of the particles, as indicated in Figures 1 and 2A. While C. albicans can easily adapt to acidic pH levels by secreting acids out of the cell after approximately 50 h of incubation [46], H. pylori are more vulnerable to extreme acidity, despite their ability to enhance pH using urease [54,55]. Moreover, fungi can also adapt to alkali environments; some pathogenic fungi secrete acids that are used for the induction of host tissue injury [56]. Interestingly, the H. pylori-specific gene (CagA) was detectable in C. albicans with intravacuolar H. pylori; this finding supports the likelihood of interaction between these two microorganisms. Due to the effectiveness of the protection, H. pylori's presence inside C. albicans may be another mechanism explaining the persistence of H. pylori in the human GI tract [41,47]. Because the CagA gene was detectable in subsequent generations of C. albicans, showing vertical transmission into the daughter cells where it continued to express proteins of H. pylori [41], the yeast cell may be both a protective vehicle and a transmission method. Despite the use of antibiotics, including amoxicillin, together with a proton pump inhibitor for H. pylori eradication [57][58][59], and despite the amoxicillin susceptibility of the selected H. pylori in our experiments, amoxicillin could not inhibit the intravacuolar H. pylori, as indicated both by a culture on urea-based agar and by CagA expression. However, Candida may not protect intravacuolar H. pylori against the function of other antibiotics as some antibiotics may be able to diffuse inside the yeast cells. Nevertheless, the protection against amoxicillin provides proof of the concept of antibiotic protection inside yeast cells. Our data suggest that the intravacuolar H. pylori inside C. albicans were protected from antimicrobials. Hence, the development of intravacuolar H. pylori was very helpful for H. pylori in terms of resistance to treatments, extra-stomach transfers, person-to-person transmission, and environmental spread. Likewise, H. pylori can transform from a spiral-shaped bacillary form into a coccoid form in stress environments, similar to other Gram-negative bacteria; in this form, they are nonculturable and referred to as VBNC (viable but nonculturable) bacteria, with fewer activities than the spiral form, but they may revert to active regrowth conditions [60]. While H. pylori's transformation into a coccoid form is a well-known mechanism to adapt to stresses before a regrowth that is often mentioned as an underlying mechanism for the detection of Helicobacter in the environment [60], intravacuolar H. pylori is possibly another strategy for survival in harsh environments. However, the putative infectious capability and the regrowth ability of this form of H. pylori are still controversial. Here, we also demonstrated that the development of intravacuolar H. pylori depended on pH and the abundance of Candida, as a higher fungal abundance at lower pH levels facilitated the intravacuolar form of Helicobacter. Hence, an evaluation of Candida abundance in the gut and the use of fungicidal drugs in the patients in whom H. pylori eradication has failed might be helpful. Moreover, our data also suggest that H. pylori inside Candida can be spontaneously released from the yeast cells on the 2nd and 3rd days of the symbiosis with biological activity intact (positive urease test). The prospect of additional studies on this topic is interesting.

Intravacuolar H. pylori Inside C. albicans as a Cause of Gastritis
Although the administration of C. albicans alone does not cause gastritis, the worsening of H. pylori-induced gastritis by Candida through synergistically elevated gastric inflammation has been previously described [13,43,45]. Likewise, Candida alone did not induce stomach injury in our study, but gastritis was demonstrated in the mice orally administered both with intravacuolar H. pylori (the bacteria inside Candida) and with H. pylori alone, as indicated by inflammatory lesions, CagA expression, a urea-based culture, and tissue cytokine levels in the stomach. Notably, the spontaneous bodyweight regains observed in the H. pylori mouse model [61,62] and the absence of weight gain in mice during the fasting experiments have been previously mentioned [63]. Our data suggest that Candida yeast is a reservoir of H. pylori and that the bacteria can be released outside the yeast cells and regrow in the stomach epithelium. Although the mechanisms of H. pylori's release from the yeast cells are still unknown, vesicular release has been identified as a new fungal secretory pathway that might play a role here [64]. As such, some yeast fungi, including C. albicans, Cryptococcus neoformans, and Saccharomyces cerevisiae, can produce vesicles containing RNA or microRNA (miRNA) (the non-coding RNA sequence with gene regulatory effects) [65][66][67] that can alter the gene expression of target cells [68]. In addition, the fungal cell wall has a dynamic structure with flexible viscoelastic qualities that allow the release of vesicles [69], as well as fungal-host communication with some specific environmental factors (pH, temperature, presence of serum, and lack of some amino acids) that trigger the production and release of vesicles [70]. Thus, these data suggest that the large vesicles of intravacuolar H. pylori possibly pass through the cell wall of C. albicans yeast cells. With the release of intravacuolar H. pylori, Helicobacter bacteria can induce gastritis through decreases in mucus and mucosal blood flow, prostaglandin synthesis, cell healing processes, and gastric emptying time, similar to regular H. pylori [7,8,12,71]. Notably, the abundance of H. pylori in the stomach of H. pylori-administered mice was higher than in mice with Candida containing intravacuolar H. pylori ( Figure 5C,D), suggesting that some intravacuolar Helicobacter could not be released from the Candida yeast cells. In contrast, Candida administration alone did not cause ulcers. The severity of the gastritis caused by H. pylori alone versus that caused by H. pylori inside Candida was not different, indicating a lack of synergy between H. pylori and the Candida hosting it; this finding differs from previous reports of enhanced pro-inflammation with the co-presence of H. pylori and Candida [13,43,45]. Perhaps the Candida yeasts that host H. pylori lose their pro-inflammatory properties after the release of H. pylori. We hypothesize that the release of H. pylori from C. albicans vesicles enhances the pro-inflammatory response of enterocytes ( Figure 8A,B) that activate the gastric inflammation of mucosa through Toll-like receptors, especially TLR4 and TLR2, with several adapter molecules (MyD88, IRAK1, IRAK4, and NF-κB) [72]. Hence, our data suggest that Candida yeast cells could be the eukaryotic hosts that protect the prokaryotic bacteria from antibiotics and stressful environments and that the hosted bacteria are ready to be spontaneously released or let out by the breakdown of the yeast cells and/or other mechanisms. More mechanistic studies on this topic would be interesting. of H. pylori from C. albicans vesicles enhances the pro-inflammatory response of enterocytes ( Figure 8A,B) that activate the gastric inflammation of mucosa through Toll-like receptors, especially TLR4 and TLR2, with several adapter molecules (MyD88, IRAK1, IRAK4, and NF-κB) [72]. Hence, our data suggest that Candida yeast cells could be the eukaryotic hosts that protect the prokaryotic bacteria from antibiotics and stressful environments and that the hosted bacteria are ready to be spontaneously released or let out by the breakdown of the yeast cells and/or other mechanisms. More mechanistic studies on this topic would be interesting.  For the clinical translation, the spontaneous development of intravacuolar H. pylori inside Candida yeast cells may depend on the ratio of the abundances of Candida and H. pylori in each patient; such development would possibly induce resistance to the standard treatment. Currently, the causes of treatment failure for H. pylori focus on (i) increased antibiotic resistance [73] through the genes that affect the alteration of proton pump inhibitor pharmacokinetics and/or the host cytochrome CYP2C19 and (ii) host factors, such as poor adherence, insufficient duration of therapy, and smoking [74][75][76]. On the other hand, our data support another underlying mechanism of treatment failure through an increased abundance of fungi in the stomach. Indeed, advanced age and the consumption of proton pump inhibitors correlate with fungal colonization in the stomach, which may be associated with H. pylori-positive status and dyspepsia [13]. Hence, reducing the abundance of Candida (anti-fungal treatments) together with H. pylori treatment (triple therapy) may improve the outcome of treatment and reduce the recurrent rate of gastritis, especially in cases of treatment failure. Unfortunately, the correlation between Candida abundance and the treatment resistance of H. pylori is still unknown. In contrast, an interkingdom correlation between Candida and H. pylori may be clinically possible because the generation of intravacuolar H. pylori was easily induced by the co-incubation of both organisms in a fungal culture media, and the intravacuolar H. pylori could pass through the daughter yeast cells, protected from antibiotics, and possibly caused resistance to the standard treatment. More studies are warranted.
To see whether damaged yeast cell walls can release more H. pylori than non-damaged yeast cells, the CagA gene in the second generation of Candida yeast cells with or without sonication was extracted by a Beadbeater machine for whole DNA, and the presence of the CagA gene was determined by a qPCR system (Thermo Fisher Scientific).

Antimicrobial Incubation and Stresses against Intravacuolar H. pylori
To explore the possible benefits of the endosymbiotic H. pylori inside C. albicans in terms of protection from antibiotics and environmental stresses, amoxicillin, and an aerobic (high-oxygen) condition, which is a stress factor for microaerophilic H. pylori (i.e., it has the ability to grow in 5-15% oxygen), were tested. As such, the second generation of Candida yeast cells was cultured into SB with or without amoxicillin (0.06 and 8 ug/mL) (Tianjin TEDA Steyuan Pharm Co., Ltd., Shijiazhuang, Hebei, China) in aerophilic conditions (21% oxygen) at 37 • C overnight. Then, 100 µL of each sample was plated onto SDA and incubated in aerophilic conditions at 37 • C overnight. After that, the whole DNA was extracted to identify CagA gene expression in each experimental group using a qPCR system (Thermo Fisher Scientific).

Animal and Peptic Ulcer Model
The animal study (SST 018/2562) was approved by the Institutional Animal Care and Use Committee of Chulalongkorn University's Faculty of Medicine following the animal care and use procedure of the National Institutes of Health (NIH). Male 8-weekold C57BL/6 mice weighing 20-25 g was purchased from Nomura Siam International, Pathumwan, Bangkok, Thailand. The mice were housed in a temperature-controlled environment (24 ± 2 • C), with 50% relative humidity and a 12 h light-dark cycle (light from 7:00 a.m. to 7:00 p.m.). All mice received food and water ad libitum. Animal procedures were performed in adherence with U.S. National Institutes of Health guidelines and followed the 8th Edition of the Guide for Care and Use of Experimental Animals, published by the National Research Council of the National Academies (2011; available at https://grants.nih.gov/grants/olaw/guide-for-the-care-and-use-of-laboratoryanimals. pdf, accessed on 12 November 2021), as well as the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines.
A peptic ulcer model of H. pylori infection was modified from a previous study [78]. Briefly, H. pylori at 3.7 × 10 6 CFU/mL, C. albicans at 1 × 10 8 CFU/mL, or C. albicans with intravacuolar H. pylori at 1 × 10 8 CFU/mL (with approximately 3.7 × 10 6 CFU/mL of intravacuolar H. pylori inside the C. albicans) in 1 mL of 1x phosphate buffer solution (PBS) was once daily orally administered using a stainless-steel feeding tube (18-gauge size and 1.5 inches in length, with a rounded tip attached to a 1 mL syringe). Notably, all groups of mice fasted for 5 h before microorganismal administration. All mice were observed and sacrificed 12 weeks after the beginning of the experiment. Then, the stomach was divided longitudinally through the greater and lesser curvature into several parts, washed with 1xPBS, weighed, and used for determining inflammatory responses and fungal-bacterial interactions.

Mouse Gastric Analysis
Mouse stomachs were divided into 4 small pieces for performing (i) a H. pylori culture, (ii) an analysis of CagA gene expression, (iii) the measurement of gastric cytokine levels, and (iv) a histopathology analysis (fixed in 10% formaldehyde). To determine H. pylori abundance in mouse stomachs, the gastric tissues (0.01 g per sample) were cultured on urea-based agar and incubated at 37 • C overnight before colony enumeration. In addition, the CagA gene was identified by qPCR (Thermo Fisher Scientific) from gastric tissues (1 mL PBS per g tissue) that were sonicated with the setting of pulse-on for 20 s and pulse-off for 5 s in 30 min on ice using the Sonics Vibra Cell machine (VCX 750) (Sonics & Materials Inc., Newtown, CT, USA) until a homogeneous solution was formed. The supernatant, after the centrifugation, was used for the detection of gastric cytokines (TNF-α, IL-6, and IL-10) and CagA expression by ELISA assays (Invitrogen, Waltham, MA, USA) [79,80] and PCR, respectively. For histology, the stomach tissues were rinsed with PBS, fixed in 10% (weight by volume (W/V)) formaldehyde, embedded in paraffin, and stained with hematoxylin and eosin (H and E) color in 5.0 mm thickness sections. The histological analyses of 200× magnification slides were performed by 2 observers who were blinded to the experiments; the analyses used semi-quantitative scores based on inflammatory cell infiltration (macrophages and neutrophils), mucosal damage, and ulceration on a scale of 0 to 4 as modified from a previous publication [81].

Statistical Analysis
All data were analyzed by Statistical Package for Social Sciences software (SPSS 22.0, SPSS Inc., Chicago, IL, USA) and Graph Pad Prism version 7.0 software (La Jolla, CA, USA). Results were presented as mean ± standard error (SE). The differences between multiple groups were examined for statistical significance by one-way analysis of variance (ANOVA) with Tukey's analysis. The survival analysis and time-point data were determined by the log-rank test and repeated measures ANOVA, respectively. A p-value < 0.05 was considered statistically significant.

Conclusions
In conclusion, intravacuolar H. pylori benefit from increased transmissibility, and the fungal host protects the bacteria from stressful micro-environments, including antibiotics. Our mouse model provides evidence that intravacuolar H. pylori were able to induce gastric infection, inflammatory cell infiltration, and tissue damage. Future studies on the role of intravacuolar H. pylori in yeast cells are needed to extend our understanding of intravacuolar H. pylori colonization in humans in order to optimize and individualize health strategies.