Helicobacter pylori infection activates the HIF-1α/ALS2/Rab5 signaling axis in gastric cells Authors: Daniela Herrera-Ramírez, Sofía Sanhueza, Héctor Tapia, Yáreni Ávalos-Guajardo, Ignacio Wichmann, Lucas Faundes, Sebastián Charlín, Álvaro Neira-Troncoso, Marcelo Garrido, Manuel Varas-Godoy, Carolina Bizama, Andrew F. G. Quest, Vicente A. Torres Published: 2026 ABSTRACT Helicobacter pylori (H. pylori) infection has been associated with gastric carcinogenesis by activating different signaling pathways. In gastric cancer cells, H. pylori promotes the stabilization of HIF-1α (hypoxia inducible factor 1α), a transcription factor associated with the expression of several genes involved in cellular adaptation. We previously showed that HIF-1α regulates endosomal trafficking via activation of the small GTPase Rab5, which in turn activates downstream signaling pathways involved in tumor progression, such as the WNT/β-catenin pathway. Here, we explored the possibility that H. pylori infection activates this signaling mechanism in gastric cells, uncovering the guanine exchange factor ALS2 as a novel target induced by this bacterium. H. pylori infection increased ALS2 levels in a HIF-1α-dependent manner, leading to Rab5 activation during bacterial infection, thereby allowing sustained β-catenin signaling via endosomal sequestration of GSK3β. Rab5-GTP pulldown assays in MKN74 and AGS gastric cancer cells showed that bacterial infection induced Rab5 activation, which depended on HIF-1α stabilization, as indicated by siRNA and pharmacological inhibition approaches. Interestingly, we found that H. pylori infection upregulated the Rab5 activator ALS2 at both the mRNA and protein levels in a HIF-1α-dependent manner. These observations were further confirmed in patient-derived tumor organoids, as bacterial infection recapitulated ALS2 induction ex vivo. Moreover, ALS2 was required for H. pylori-induced Rab5 activation, early endosome enlargement, and downstream β-catenin signaling, as shown by shRNA-mediated knockdown of endogenous ALS2. Taken together, these observations uncover a novel pathway induced by H. pylori in gastric cells, which involves the HIF-1α/ALS2/Rab5/β-catenin signaling axis.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12964-026-02978-x. Introduction Gastric cancer (GC) is the fifth most common cancer in both sexes and leads to more than one million new diagnoses every year worldwide [1]. GC is classified histologically into diffuse and intestinal subtypes, whereby the intestinal type represents the most frequently diagnosed gastric cancer type, characterized by cohesive tumor cells that resemble the intestinal mucosa [2]. The main risk factor for the development of this subtype is the infection with the bacterium Helicobacter pylori (H. pylori), which is considered a class I carcinogen [3, 4]. H. pylori is a Gram-negative, microaerophilic bacterium that possesses several virulence factors, such as urease, which allows the bacterium to survive in the highly acidic gastric environment by producing ammonia [5, 6]. Interestingly, more recent evidence shows that this enzyme can also function as a ligand for specific cell surface receptors [7]. Urease is a nickel-dependent metalloenzyme, which is formed by a dodecameric complex of the UreA/UreB subunits and represents approximately 10% of the total proteins synthesized by H. pylori [5, 8]. This enzyme is localized in the cytoplasm and associated with the bacterial surface, and has been shown to participate in gastric carcinogenesis by activating various signaling pathways [7, 9]. Moreover, Valenzuela et al. determined that H. pylori urease induces the stabilization of the hypoxia-inducible factor 1α (HIF-1α) in gastric epithelial cells [7]. Specifically, urease upon binding to Toll-like receptor 2 (TLR2) activates the PI3K/mTOR signaling pathway, thereby promoting the upregulation of HIF-1α [7, 10], which is a transcription factor involved in adaptive cellular responses to hypoxia. Following stabilization, HIF-1α induces the expression of a plethora of genes involved in metabolic reprogramming, adaptation, cell migration, and invasion [11, 12]. Among HIF-1α targets, the gene encoding the guanine nucleotide exchange factor (GEF) ALS2 was shown to be induced by hypoxia in a HIF-1α-dependent manner, leading to the activation of Rab5, an endosomal GTPase that promotes tumor cell migration, invasion, and metastasis [13]. Rab5 belongs to the Rab family of small GTPases, with more than 60 members in humans, which is responsible for regulating endosomal trafficking [14]. Rab5 activation is coordinately regulated by GEFs, which promote the exchange of GDP to GTP, and GAPs (GTPase-activating proteins), which stimulate the intrinsic GTPase activity of Rab proteins [14, 15]. The Rab-GEF ALS2 contains numerous domains involved in cell signaling, and specifically, the C-terminal region contains a VPS9 domain, which catalyzes the exchange of GDP for GTP in Rab5 [16]. We previously showed that ALS2 expression is upregulated in tumor biopsies and involved in tumor cell migration and invasion via activation of Rab5 [13]. Accordingly, Rab5 expression and/or activation are also increased in different types of cancer, and its presence has been associated with the metastatic potential and the degree of differentiation [17–19]. Mechanistically, Rab5 has been shown to promote β-catenin signaling by inducing endosomal sequestration of components of the β-catenin destruction complex, thereby allowing for nuclear accumulation of β-catenin and subsequent expression of target genes [20, 21]. Whether this signaling mechanism is activated by H. pylori infection is an intriguing possibility supported by the fact that Rab5 undergoes activation in hypoxia via HIF-1α stabilization [13, 18]. In this study, we report that H. pylori, by stabilizing HIF-1α, upregulates the Rab-GEF ALS2, which in turn activates Rab5 and consequently promotes β-catenin signaling. These observations provide insights into the mechanisms by which H. pylori induces the acquisition of malignancy-associated characteristics in gastric epithelial cells. Methods Cell culture The following gastric cell lines were used: MKN74, AGS, and GES-1. These cell lines were grown in RPMI cell culture medium, free of antibiotics, supplemented with 10% fetal bovine serum (FBS, BioInd 04–121-1A), and maintained at 37 °C in a 5% CO2 atmosphere. H. pylori infection H. pylori, strain 26695, was cultured on tryptone soy agar (TSA) plates (Difco 236,950), containing 5% horse serum (BioInd 04–004-1A), and the specific supplements Vitox (Oxoid SR0090A) and Dent (Oxoid SR0147E). The bacteria were grown at 37 °C in 5% CO2 for 24 h, then collected in 1 ml phosphate-buffered saline (PBS) and plated again on supplemented TSA agar for 24 h. Gastric cells were infected with H. pylori 26695 at varying multiplicities of infection (MOI) of 10, 50, or 100, as indicated. For MOI calculations, the optical density (O.D.) was measured at 560 nm, and the ratio 0.4 (O.D.) was considered equivalent to 3.0 × 108 colony-forming units, as previously described by our group [7]. Infected cells were maintained at 37 °C in a 5% CO2 atmosphere. HIF-1α inhibition and RNAi transfection MKN74 cells were cultured for 24 h and then incubated with 30 µM of the HIF-1α inhibitor SC-205346 (Santa Cruz Biotechnology 205,346). Dimethyl sulfoxide (DMSO, Sigma D2650), which corresponded to the vehicle, was used as a control [13]. Vehicle and inhibitors were added at the same time as the infection with H. pylori 26695. The treatment was maintained throughout the time of infection with the bacteria. Transfection of 50 nM of pcDNA3.1 plasmid encoding HIF-1α and the control was performed as described by Rivas et al. [13]. Once the transfection was completed, cells were infected with H. pylori 26695, as described previously. GFP/S34N transfection Transfections with the pEGFP-C1 plasmids encoding for GFP alone, GFP-Rab5/S34N (Rab5 mutant with high affinity for GDP), GFP-Rab5/Q79L (GTPase-deficient mutant of Rab5), or GFP-ALS2, were carried out as previously described by our group [13, 22, 23]. ALS2 knockdown MKN74 cells were seeded in 6-well plates. After 24 h, cells were transduced with lentiviral vectors containing different shRNA sequences against ALS2, referred to as pLenti #1 (shALS2 #1) and pLenti #2 (shALS2 #2), as well as a shRNA control (control, shLuc). Following 24 h of cell transduction, 2 µg/ml of puromycin (Gibco A11138-03) was added for selection. As a result, MKN74 shALS2 #1 and MKN74 shALS2 #2, as well as MKN74 shLuc (control) cells, were obtained for subsequent analyses. These cell lines were amplified using the RPMI culture medium supplemented with 10% FBS and incubated at 37 °C in the presence of 5% CO2. Indirect immunofluorescence Cells were seeded on coverslips and fixed with fixation buffer (4% paraformaldehyde diluted in PBS) at room temperature. Then, the cells were washed with PBS three times for 5 min. Cells were then permeabilized with 0.5% Triton X-100 and washed with PBS. Cells were blocked in a humid chamber with blocking solution (2% BSA, Rockland BSA-50, diluted in blocking solution) for 30 min. Subsequently, incubation was performed overnight at 4 °C, with the primary antibodies against Rab5 (Santa Cruz Biotechnology SC46692), EEA1 (ThermoFisher Scientific PAI063A), GSK3β (Cell Signaling Technology CS124565S), or β-catenin (Dako M3539), which were diluted in blocking solution. Afterwards, cells were washed with PBS and incubated in the dark with the corresponding secondary antibodies coupled to Alexa 488 (Invitrogen A11008), Alexa 568 (Invitrogen A11004), or both, for 1 h at 37 °C. In addition, cells were incubated in the dark with DAPI (4',6-diamidino-2-phenylindole, Invitrogen D1306) for nuclear staining. Finally, the coverslips were washed and mounted with Mowiol Mounting Medium (Dako S3023) for subsequent observation using a Confocal C2 plus microscope at 60X and 100X magnification. Image analysis was performed using the ImageJ program. TCF/LEF reporter assay Briefly, cells were seeded in 6-well plates and co-transfected with 1 μg (pTOP FLASH) and 0.25 μg of Renilla for 24 h. To measure luciferase/Renilla activity, the supplier's instructions were followed (dual-luciferase reporter assay system, Promega, Madison, WI, USA), as reported by us [24]. Proteinase K protection assay Cells were grown on 10 cm plates, washed with cold PBS, and then gently scraped on ice. Samples were then centrifuged for 5 min at 151 × g and 4 °C, the supernatants were discarded and the pellets were resuspended in 1 ml of permeabilization buffer (Digitonin 100X (Sigma D5628), 100 mM potassium phosphate, 5 mM MgCl2, 250 mM sucrose). Samples were then incubated for 5 min at room temperature, followed by 30 min on ice. Subsequently, samples were centrifuged for 5 min at 151 × g, and the supernatants were removed. Pellets were resuspended in 300 µL of buffer without Digitonin (100 mM potassium phosphate, 5 mM MgCl2, 250 mM sucrose) and then aliquoted as follows: (1) 100 µL left untreated, (2) 100 µL incubated with 1 µg/ml proteinase K, (3) 100 µL incubated with 1 µg/ml proteinase K plus 0.1% Triton X-100 (Winkler BM2020). Finally, 0.25 mM PMSF was added for 10 min to stop the reaction, and 50 µL of 4X loading buffer was then added. Samples were analyzed by Western blotting. Western blotting Protein extraction was performed as previously described by our group [13]. Subsequently, polyacrylamide gel electrophoresis was performed under reducing conditions. Between 50 µg and 100 µg of proteins were loaded. After protein transfer, the membranes were blocked in 5% gelatin, washed, and incubated with antibodies (diluted in blocking solution) against HIF-1α (BD Biosciences BD610959), ALS2 (Abcam AB170896), β-catenin (Dako M3539), Rab5 (Santa Cruz Biotechnology SC46692), EEA1 (ThermoFisher Scientific PAI063A), β-actin (Santa Cruz Biotechnology SC47778), and GSK3β (Cell Signaling Technology CS124565S). Then, the membranes were washed and incubated with anti-Rabbit HRP (Jackson Immuno Research 115–035-062) or anti-Mouse HRP (Jackson Immuno Research 115–035-045) secondary antibodies. Finally, membranes were revealed by chemiluminescence with an automated equipment (Image System ULTRALUM Discovery 12ic). Rab5 pull-down The pull-down protocol was performed as previously described [13]. Briefly, one million cells were seeded in 10 cm plates. After 72 h, cells were homogenized by adding 500 µL of lysis buffer (25 mM HEPES pH 7.4, 100 mM NaCl, 5 mM MgCl2, 1% NP 40, 10% glycerol, 1 nM DTT, 12.5 µg/ml leupeptin, 10 µg/ml antipain, 100 µg/ml benzamidine, 1 mM phenylmethanesufonylfluoride, 1 mM sodium orthovanadate and 10 mM sodium fluoride) to the plate on ice and scraping with a rubber-scraper. The extracts were pipetted up and down 6 times, and then centrifuged for 1 min at 2839 × g, 4 °C. Then, 20 µL for each condition were mixed with loading buffer (total Rab5). The remaining homogenate was incubated with 5 μg of the GST-R5BD fusion protein for 15 min at 4 °C in a rotating shaker. After incubation, samples were centrifuged for 1 min at 151 × g and 4 °C. The supernatant was discarded, and 700 µL of washing buffer (25 mM HEPES pH 7.4, 100 mM NaCl, 5 mM MgCl2, 10% glycerol, 1 nM DTT, plus protease and phosphatase inhibitors) were added. The washing and centrifugation processes were repeated once more, and then 20 µL of loading buffer was added to the pellets for subsequent Western blot analysis. Reverse transcription-quantitative (RT-qPCR) RNA from cells was isolated using TRIzol (Invitrogen, Life Technologies) according to the manufacturer’s instructions. The pellet was dissolved in 20 μl nuclease-free water. RNA was quantified using the Infinite® 200 PRO NanoQuant Plate (TECAN). The extracted RNA was treated with DNase (Promega, Madison, Wisconsin, USA) following the manufacturer’s protocol to obtain DNA-free RNA. Reverse transcription (RT) was performed with 1 μg of DNA-free RNA using the M-MLV reverse transcriptase (Promega, Madison, Wisconsin, USA). qPCRs were performed using a Brilliant II SYBR Green qPCR Master Mix (Agilent Technologies, Santa Clara, California, USA) following the manufacturer´s instructions. Briefly, 20 μl reaction volumes in 96-well plates were analyzed in the Mx3000P QPCR System (Agilent Technologies, Santa Clara, California, USA) using the following conditions: 10 min denaturation at 95 °C, followed by 40 cycles at 95 °C for 20 s, annealing temperature for 20 s, and at 72 °C for 20 s. The following pairs of primers were used for cDNA amplification: ALS2, 5′-TCGCCTAAAGGATGCCACCTATGA-3′, 5′-TCTCCATACCCATCTTCCAAGCCA-3′; β actin 5′-TGGCACCCAGCACAATGAAGA-3′, 5′-GAAGCATTTGCGGTGGACGAT-3′. Relative gene expression levels were calculated by the 2−(ΔΔCT) method [25], with β-actin as an internal control. All data were expressed relative to values obtained for non-infected cells. Gastric Cancer Patient-Derived Organoids Culture (GC-PDOs) Gastric cancer patient-derived organoids (GC-PDOs) were generated from an endoscopy biopsy sample obtained from a patient with advanced gastric cancer recruited at the Hospital Clínico and Centro de Especialidades Médicas of UC-Christus. This study was approved by the Institutional Review Board of the School of Medicine of Pontificia Universidad Católica de Chile (ID: 210620005, april 7, 2022). The GC-PDOs were established as previously described, with minor modifications [26]. Briefly, the dissociated cells were washed in 1 × DPBS and embedded in growth factor-reduced Matrigel (#356231, Corning, NY, USA) and cultured in gastric cancer organoid medium containing Intesticult Medium (#6000 STEMCELL Technologies), 0.5 μmol/L A83-01 (#SML0788; Sigma-Aldrich), 100 ng/mL, FGF10 (#100–26, PeproTech), 10 nmol/L Gastrin I (#G9145; Sigma-Aldrich), 1 mmol/L N-acetyl-L-cysteine (#A9165; Sigma-Aldrich) 10 mmol/L nicotinamide (#N0636; Sigma-Aldrich), 1 × B27 supplement (#17504–044; Gibco), 1 × N2 supplement (#175020–048; Gibco), 1 mg/mL Primocin (#ant-pm-1; InvivoGen), 10.5 μmol/L Y-27632 (#Y0503, Sigma-Aldrich), 0.5 μmol/L Prostaglandin E2 (#P0409-1ug; Sigma-Aldrich), 7.5 μg/mL Insulin (#I9278-5ML; Sigma-Aldrich), 10 μmol/L SB202190 (#S7067; Sigma-Aldrich) and 10 μg/mL Transferrin (#T8158; Sigma-Aldrich). Organoids were passaged via mechanical dissociation, with weekly passages at 1:3 ratios. Before incubation with H. pylori, the polarity of GC-PDOs was reversed using a previously reported protocol [27]. Apical-out organoids were then cultured in ultralow-attachment plates (Corning Costar, #3473) in the absence of Matrigel and incubated in gastric cancer culture medium at the MOI previously standardized with cell lines. For microscopy analyses of ALS2 and β-catenin, the reported n values correspond to technical replicates. They are defined as individual organoid structures derived from the same culture batch and analyzed independently. All the experiments were performed using organoids derived from the same preparation. Independent biological replicates were not included in this study. However, multiple organoid structures were analyzed per condition to ensure consistency of the observed phenotype within the culture. Gene expression analysis Public microarray gene expression data from the Gene Expression Omnibus (GEO) dataset GSE66229 (ACRG cohort; [28]) were downloaded using the GEOquery R package [29]. RNA sequencing (RNA-seq) data for stomach adenocarcinoma from The Cancer Genome Atlas (TCGA-STAD; [28]) were retrieved using TCGA-Assembler 2 [30]. These cohorts include primary human gastric cancer and adjacent mucosa, enabling evaluation of gene expression changes in vivo that complement mechanistic findings derived from in vitro models. The ACRG dataset contaned 400 samples (300 tumors, 100 tumor-adjacent normals), including 98 patient-matched tumor–normal pairs, whereas the TCGA-STAD dataset contained 407 samples (375 primary tumors, 32 tumor-adjacent normals), including 32 matched pairs. To ensure consistency of gene identifiers across platforms, probe annotations from microarray data and gene symbols from TCGA-STAD were mapped to stable gene identifiers, using ENTREZID as common anchors via the Homo.sapiens and AnnotationDbi R packages. ACRG microarray data were pre-processed using robust multi-array average (RMA), including background correction, quantile normalization and probe-level summarization. When multiple probes mapped to the same gene, expression values were collapsed using the avereps function from limma [31]. TCGA RNA-seq data were processed using the limma-voom framework: raw counts were normalized using the trimmed mean of M-values (TMM) method and transformed with voom to model the mean–variance relationship, yielding log2 counts per million (logCPM) values. Principal component analysis (PCA) was performed for each cohort to assess technical variation; no clear batch-driven structure was observed, and no batch covariates were included in downstream models. Differential expression analyses were performed independently for each dataset using the limma (microarray) or limma-voom (RNA-seq) frameworks with empirical Bayes moderation, comparing gastric cancer (GC) samples with non-tumor adjacent mucosa (NTAM). For both cohorts, analyses were conducted using paired and unpaired study designs. For paired analyses, only matched tumor–normal samples were included, and patient identity was modeled as a fixed effect (~ 0 + Condition + Patient), with contrasts defined on Condition. To evaluate robustness to study design and sample imbalance, complementary unpaired analyses were performed using an unblocked model (~ 0 + Condition). In this setting, tumors from matched pairs were excluded while all available NTAM samples were retained (ACRG: 202 tumors vs 100 NTAM; TCGA-STAD: 343 tumors vs 32 NTAM). No minimum fold-change threshold was imposed; statistical significance was defined as a Benjamini–Hochberg false discovery rate (FDR) ≤ 0.05. Expression of selected genes of interest (including HIF1A and ALS2) was specifically evaluated within this framework in both paired and unpaired analyses. Differential expression was computed separately across datasets and integrated the results using a random-effects model meta-analysis of differential expression results using MetaVolcanoR (10.18129/B9.bioc.MetaVolcanoR). Per-gene differential expression outputs from limma (including log2 fold-changes with 95% confidence intervals and associated statistics) were used as input for the meta-analysis. This approach preserves within-platform statistical structure while enabling cross-cohort integration. The results were visualized in R using ggplot2. Statistical analysis The results were expressed as the mean ± standard error of the mean (mean ± SEM). A p < 0.05 was considered statistically significant. The results were analyzed using the T-Test, and One-way ANOVA with post-hoc multiple comparison analysis, Dunn's test, or Tukey’s test, as indicated. Results H. pylori promotes nuclear localization of β-catenin and endosomal sequestration of GSK3β in gastric cells We recently showed that brief infection of gastric cells with H. pylori induces a modest, but non-significant fluctuation in total and nuclear β-catenin levels, which were associated with decreased TCF/LEF-dependent transcription [24]. Nevertheless, the long-term effects of H. pylori infection on β-catenin signaling remained unclear, which was intriguing, as the effects of H. pylori on cell signaling are time-dependent. To address this possibility, we assessed the effects of H. pylori on β-catenin signaling following 24 h of infection. To this end, MKN74 and AGS cells were infected with H. pylori, and the subcellular localization of β-catenin was evaluated after 24 h by indirect immunofluorescence analysis. The results obtained revealed an increase in the nuclear localization of β-catenin following bacterial infection in both cases (Fig. 1A-D). In accordance with these observations, H. pylori infection induced a sevenfold increase in β-catenin/TCF-LEF-dependent transcription in MKN74 cells (Fig. 1E) and an 18-fold increase in AGS cells (Fig. 1F), as shown in luciferase reporter assays.Fig. 1H. pylori promotes nuclear accumulation of β-catenin in MKN74 and AGS cells. A Immunofluorescence of β-catenin in MKN74 cells 24 h after infection with H. pylori (MOI 100). Left panels, cell nuclei are shown in blue (DAPI); middle panels, β-catenin and the magnified images are shown in green; right panels show the superposition (merge) of both images. Photographs were obtained at 100X magnification, Nikon C2 plus confocal microscope (n = 4). Scale bar 10 μm. B Quantification of β-catenin nuclear fluorescence intensity relative to the nuclear area in non-treated (NT) and H. pylori-infected cells (Hp) (t-test, n = 4). C Immunofluorescence of β-catenin in AGS cells 24 h after infection with H. pylori (MOI 50). Cell nuclei are shown in blue (DAPI) and β-catenin in red. Photographs were obtained at 100X magnification with a Nikon C2 plus confocal microscope (n = 5). Scale bar 10 μm. D Quantification of β-catenin nuclear fluorescence intensity relative to the nuclear area in NT and Hp (t-test, n = 5). E, F TCF/LEF promoter reporter assay in (E) MKN74 (t-test, n = 4) and (F) AGS cells (t-test, n = 3), left untreated (NT) or following bacterial infection (Hp) for 24 h. G Proteinase K protection assays were carried out in MKN74 cells left either untreated (NT) or after bacterial infection (Hp) for 24 h. Samples were analyzed by Western blotting with antibodies against GSK3β, and the bands for GSK3β were quantified. Data are shown as the relative levels with respect to input conditions (one-way ANOVA, Post hoc Tukey´s test, n = 3). H Proteinase K protection assays were carried out using AGS cells, as described in (G). Representative Western blot image and the quantification are shown (one-way ANOVA, Post hoc Tukey´s test, n = 3). *p < 0.05, ** p < 0.01, *** p < 0.001, ****p < 0.0001 H. pylori promotes nuclear accumulation of β-catenin in MKN74 and AGS cells. A Immunofluorescence of β-catenin in MKN74 cells 24 h after infection with H. pylori (MOI 100). Left panels, cell nuclei are shown in blue (DAPI); middle panels, β-catenin and the magnified images are shown in green; right panels show the superposition (merge) of both images. Photographs were obtained at 100X magnification, Nikon C2 plus confocal microscope (n = 4). Scale bar 10 μm. B Quantification of β-catenin nuclear fluorescence intensity relative to the nuclear area in non-treated (NT) and H. pylori-infected cells (Hp) (t-test, n = 4). C Immunofluorescence of β-catenin in AGS cells 24 h after infection with H. pylori (MOI 50). Cell nuclei are shown in blue (DAPI) and β-catenin in red. Photographs were obtained at 100X magnification with a Nikon C2 plus confocal microscope (n = 5). Scale bar 10 μm. D Quantification of β-catenin nuclear fluorescence intensity relative to the nuclear area in NT and Hp (t-test, n = 5). E, F TCF/LEF promoter reporter assay in (E) MKN74 (t-test, n = 4) and (F) AGS cells (t-test, n = 3), left untreated (NT) or following bacterial infection (Hp) for 24 h. G Proteinase K protection assays were carried out in MKN74 cells left either untreated (NT) or after bacterial infection (Hp) for 24 h. Samples were analyzed by Western blotting with antibodies against GSK3β, and the bands for GSK3β were quantified. Data are shown as the relative levels with respect to input conditions (one-way ANOVA, Post hoc Tukey´s test, n = 3). H Proteinase K protection assays were carried out using AGS cells, as described in (G). Representative Western blot image and the quantification are shown (one-way ANOVA, Post hoc Tukey´s test, n = 3). *p < 0.05, ** p < 0.01, *** p < 0.001, ****p < 0.0001 Since earlier studies have shown that the stabilization and nuclear re-localization of β-catenin requires endosomal sequestration of the so-called β-catenin destruction complex [20], we explored the possibility that H. pylori infection promotes the accumulation of a marker of this complex within endosomal compartments. In doing so, we observed that bacterial infection increased the endosomal localization of GSK3β (Glycogen Synthase Kinase-3β), a central component of the destruction complex, as judged by proteinase K protection assays in MKN74 (Fig. 1G) and AGS cells (Fig. 1H). In accordance with the latter observations, immunofluorescence analysis showed that H. pylori promoted the localization of GSK3β within Rab5-positive endosomal compartments in MKN74 cells (Fig. 2A and 2B). These observations were intriguing because previous reports indicated that Rab5 is required for endosomal sequestration of components of the destruction complex, thereby promoting nuclear accumulation of β-catenin [21]. Thus, we explored the possibility that H. pylori-dependent re-localization of β-catenin depended on Rab5. To evaluate this possibility, MKN74 cells were transfected with the inactive mutant Rab5/S34N (dominant negative) or GFP alone as a control, and subsequently infected with H. pylori. As anticipated, these experiments revealed that the inactive mutant of Rab5 precluded nuclear localization of β-catenin, when compared with the GFP control condition (Fig. 2C and D). Taken together, these results indicate that H. pylori promotes nuclear localization of β-catenin in a Rab5-dependent manner.Fig. 2H. pylori infection promotes endosomal localization of GSK3β via Rab5. A MKN74 cells were infected with H. pylori at MOI 50 for 24 h, fixed and processed for subsequent immunofluorescence analysis of Rab5 and GSK3β. Representative confocal images showing the immunofluorescence of Rab5 and GSK3β in untreated cells (left panels) and cells infected with H. pylori (right panels). The nuclei are shown in blue (DAPI), Rab5 in green (Alexa 488), GSK3β in red (Alexa 568). The right panels of each figure show the superposition of images (merge) and their magnifications. Images were obtained at 100X magnification with a Nikon C2 plus confocal microscope (n = 3. Scale bar of 10 μm). B Co-localization between Rab5 and GSK3β was assessed by determining the Mander’s coefficient (t-test, n = 3. *p < 0.05). C Immunofluorescence analysis of β-catenin in MKN74 cells transfected with the inactive mutant of Rab5 (S34N) or GFP as a control. Both were infected with the bacteria at MOI 50 for 24 h and immunofluorescence analysis was performed. On the left, the nuclei are shown in blue (DAPI), GFP or S34N in green (Alexa 488), and on the right, β-catenin in red (Alexa 594) and the superposition of all the images. Images were obtained at 60X magnification with a Nikon C2 plus confocal microscope (n = 3. Scale bar of 10 μm). D Quantification of β-catenin positive nuclei relative to the untreated condition (one-way ANOVA, Post hoc Tukey´s test, n = 3). *p < 0.05 H. pylori infection promotes endosomal localization of GSK3β via Rab5. A MKN74 cells were infected with H. pylori at MOI 50 for 24 h, fixed and processed for subsequent immunofluorescence analysis of Rab5 and GSK3β. Representative confocal images showing the immunofluorescence of Rab5 and GSK3β in untreated cells (left panels) and cells infected with H. pylori (right panels). The nuclei are shown in blue (DAPI), Rab5 in green (Alexa 488), GSK3β in red (Alexa 568). The right panels of each figure show the superposition of images (merge) and their magnifications. Images were obtained at 100X magnification with a Nikon C2 plus confocal microscope (n = 3. Scale bar of 10 μm). B Co-localization between Rab5 and GSK3β was assessed by determining the Mander’s coefficient (t-test, n = 3. *p < 0.05). C Immunofluorescence analysis of β-catenin in MKN74 cells transfected with the inactive mutant of Rab5 (S34N) or GFP as a control. Both were infected with the bacteria at MOI 50 for 24 h and immunofluorescence analysis was performed. On the left, the nuclei are shown in blue (DAPI), GFP or S34N in green (Alexa 488), and on the right, β-catenin in red (Alexa 594) and the superposition of all the images. Images were obtained at 60X magnification with a Nikon C2 plus confocal microscope (n = 3. Scale bar of 10 μm). D Quantification of β-catenin positive nuclei relative to the untreated condition (one-way ANOVA, Post hoc Tukey´s test, n = 3). *p < 0.05 H. pylori promotes Rab5 activation in gastric cells The dependency on Rab5 in H. pylori-induced events prompted us to investigate the activation status of this GTPase following bacterial infection. To this end, MKN74 cells were infected with H. pylori, and Rab5-GTP levels (active Rab5) were measured in pull-down assays, as previously reported [18]. Following bacterial infection at different MOI (10, 50, 100), a significant increase in active Rab5 levels was observed at MOI 10 and 50, with the highest activation at MOI 50 (Fig. 3A). Similar observations were made in AGS cells, whereby bacterial infection increased Rab5-GTP levels, to the greatest extent at MOI 100 (Fig. 3B). Interestingly, H. pylori-induced activation of Rab5 was not restricted to gastric cancer cells, since bacterial infection recapitulated Rab5 activation in non-transformed gastric GES1 cells (Fig. 3C). Taken together, these results indicate that H. pylori infection promotes Rab5 activity in gastric cells.Fig. 3H. pylori infection promotes the activation of Rab5 in gastric cells. A-C Rab5 activity was determined by Rab5 pulldown assays in MKN74 (A), AGS (B), and GES-1 cells (C), following infection with H. pylori (24 h) at different MOI, as indicated. Representative Western blot images are shown. Rab5-GTP levels were determined by scanning densitometry and normalized to total Rab5 (one-way ANOVA, Post hoc Dunnett´s test, n = 3). D Immunofluorescence analysis of Rab5 in MKN74 cells infected for 24 h with H. pylori (MOI 50). Nuclei are shown in blue (DAPI) and Rab5 in green. Representative images are shown (n = 3. Scale bar of 10 μm). E Quantification of the number of Rab5-positive particles, larger than 2 µm, and normalized to cell number, in non-treated (NT) and infected cells (Hp) (t-test, n = 3). F Immunofluorescence analysis of EEA1 in MKN74 cells infected for 24 h with H. pylori (MOI 50). Nuclei are shown in blue (DAPI) and EEA1 in green. Representative images are shown (n = 3. Scale bar of 10 μm). Images were obtained at 100X magnification with a Nikon C2 plus confocal microscope. G Quantification of the number of EEA1.+ particles, larger than 2 µm, and normalized to cell number, in non-treated (NT) and infected cells (Hp) (t-test, n = 3). H Total Rab5 levels were determined at different time points after infection (0–24 h) with H. pylori (MOI 50) in MKN74 cells. Images were obtained by Western blot analysis using antibodies against Rab5. I Quantification of total Rab5 levels following infection with the bacteria and normalized to the value calculated at 0 h (non-treated) (one-way ANOVA, Post hoc Dunnett´s test, n = 3). J Total Rab5 levels were determined at different times of infection (0–24 h) with H. pylori (MOI 100) in AGS cells. Representative Western blot images are shown. K Quantification of total Rab5 levels following infection with H. pylori (one-way ANOVA, Post hoc Dunnett´s test, n = 3). *p < 0.05, ** p < 0.01, *** p < 0.001, ****p < 0.0001 H. pylori infection promotes the activation of Rab5 in gastric cells. A-C Rab5 activity was determined by Rab5 pulldown assays in MKN74 (A), AGS (B), and GES-1 cells (C), following infection with H. pylori (24 h) at different MOI, as indicated. Representative Western blot images are shown. Rab5-GTP levels were determined by scanning densitometry and normalized to total Rab5 (one-way ANOVA, Post hoc Dunnett´s test, n = 3). D Immunofluorescence analysis of Rab5 in MKN74 cells infected for 24 h with H. pylori (MOI 50). Nuclei are shown in blue (DAPI) and Rab5 in green. Representative images are shown (n = 3. Scale bar of 10 μm). E Quantification of the number of Rab5-positive particles, larger than 2 µm, and normalized to cell number, in non-treated (NT) and infected cells (Hp) (t-test, n = 3). F Immunofluorescence analysis of EEA1 in MKN74 cells infected for 24 h with H. pylori (MOI 50). Nuclei are shown in blue (DAPI) and EEA1 in green. Representative images are shown (n = 3. Scale bar of 10 μm). Images were obtained at 100X magnification with a Nikon C2 plus confocal microscope. G Quantification of the number of EEA1.+ particles, larger than 2 µm, and normalized to cell number, in non-treated (NT) and infected cells (Hp) (t-test, n = 3). H Total Rab5 levels were determined at different time points after infection (0–24 h) with H. pylori (MOI 50) in MKN74 cells. Images were obtained by Western blot analysis using antibodies against Rab5. I Quantification of total Rab5 levels following infection with the bacteria and normalized to the value calculated at 0 h (non-treated) (one-way ANOVA, Post hoc Dunnett´s test, n = 3). J Total Rab5 levels were determined at different times of infection (0–24 h) with H. pylori (MOI 100) in AGS cells. Representative Western blot images are shown. K Quantification of total Rab5 levels following infection with H. pylori (one-way ANOVA, Post hoc Dunnett´s test, n = 3). *p < 0.05, ** p < 0.01, *** p < 0.001, ****p < 0.0001 The Rab5 activation results were further supported by an additional approach, based on the measurement of early endosome size [22]. These experiments showed that H. pylori infection in MKN74 cells increased the amount of Rab5- (Fig. 3D and 3E) and EEA1- (Early endosome antigen) positive particles (Fig. 3F and 3G) larger than > 2 μm2, which are associated with endosomal redistribution towards the cell periphery. It should be noted that H. pylori-dependent Rab5 activation and early endosome particle size were not attributable to fluctuations in total Rab5 levels, as observed in MKN74 (Fig. 3H and 3I) or AGS cells (Fig. 3J and 3K). H. pylori increases ALS2 levels, which are required for Rab5 activation and nuclear localization of β-catenin The finding that H. pylori infection triggers Rab5 activation with no fluctuations in total Rab5 protein levels suggested that an indirect mechanism might be responsible, involving either induction of a Rab5 activator or downregulation of a Rab5 inhibitor. We reported previously that Rab5 activation in cancer cells involves the induction of the guanine exchange factor (GEF) ALS2 [13]. Thus, we evaluated the effects of H. pylori infection on the expression of ALS2, noting that bacterial infection was followed by a substantial increase in ALS2 mRNA levels in both MKN74 and AGS cells (Fig. 4A and 4B). Accordingly, Western blot analysis showed a time-dependent increase in ALS2 protein levels in MKN74 cells following bacterial infection (Fig. 4C and 4D). Likewise, in AGS cells, H. pylori infection increased ALS2 protein levels by 3.5-fold compared to control cells (Fig. 4E and 4F).Fig. 4H. pylori infection promotes the activation of Rab5 via ALS2 in gastric cells. A ALS2 mRNA levels were determined by qRT-PCR in H. pylori-infected (Hp) and non-infected MKN74 cells (NT) at MOI 50 for 24 h (t-test, n = 4). B ALS2 mRNA levels were determined by qRT-PCR in untreated cells (NT) or H. pylori-infected (Hp) AGS cells at MOI 100 for 24 h (t-test, n = 4). C Representative Western blot of ALS2 levels in MKN74 cells after different time points (0 h to 24 h) of H. pylori infection. D Quantification of ALS2 levels after infection of MKN74 cells with the bacteria compared to time point 0 h (non-treated) (one-way ANOVA, Post hoc Dunnett´s test, n = 3). E Representative image of ALS2 levels in the AGS cell line at different times (0 h to 24 h) of H. pylori infection as determined by Western blotting. F Quantification of ALS2 levels after infection of AGS cells with the bacteria compared to time point 0 h (non-treated) (one-way ANOVA, Post hoc Dunnett´s test, n = 3). G Representative Western blot of ALS2 levels in MKN74 ALS2 knockdown cells (shALS2 #1 and shALS2 #2) and the corresponding control (shLuc). H Quantification of ALS2 levels in the silenced cell lines compared to the shLuc control condition (one-way ANOVA, Post hoc Dunnett´s test, n = 4). I Representative Western blot images of active and total Rab5 in MKN74 Control (shLuc) and ALS2 knockdown (shALS2 #2) cells infected or not with H. pylori (Hp and NT, respectively). J Quantification of active Rab5 levels, normalized to total Rab5, and shown for each of the conditions indicated in (I), compared to the shLuc NT condition (one-way ANOVA, Post hoc Tukey´s test, n = 3). K Immunofluorescence analysis of β-catenin in MKN74 control cells (shLuc) and ALS2 knockdown (shALS2 #2), non-treated (NT), or infected (Hp) for 24 h with H. pylori (MOI 50). Nuclei are shown in blue (DAPI) and β-catenin in green (n = 3, Scale bar of 10 μm). L Quantification of β-catenin nuclear fluorescence intensity relative to the nuclear area in shLuc and shALS2 #2, infected and non-treated MKN74 cells (one-way ANOVA, Post hoc Tukey´s test, n = 3). M TCF/LEF promoter reporter assay in shLuc and shALS2 #2 MKN74 cells, left untreated (NT) or following bacterial infection (Hp) for 24 h (one-way ANOVA, Post hoc Tukey´s test, n = 3). *p < 0.05, ** p < 0.01, *** p < 0.001, ****p < 0.0001 H. pylori infection promotes the activation of Rab5 via ALS2 in gastric cells. A ALS2 mRNA levels were determined by qRT-PCR in H. pylori-infected (Hp) and non-infected MKN74 cells (NT) at MOI 50 for 24 h (t-test, n = 4). B ALS2 mRNA levels were determined by qRT-PCR in untreated cells (NT) or H. pylori-infected (Hp) AGS cells at MOI 100 for 24 h (t-test, n = 4). C Representative Western blot of ALS2 levels in MKN74 cells after different time points (0 h to 24 h) of H. pylori infection. D Quantification of ALS2 levels after infection of MKN74 cells with the bacteria compared to time point 0 h (non-treated) (one-way ANOVA, Post hoc Dunnett´s test, n = 3). E Representative image of ALS2 levels in the AGS cell line at different times (0 h to 24 h) of H. pylori infection as determined by Western blotting. F Quantification of ALS2 levels after infection of AGS cells with the bacteria compared to time point 0 h (non-treated) (one-way ANOVA, Post hoc Dunnett´s test, n = 3). G Representative Western blot of ALS2 levels in MKN74 ALS2 knockdown cells (shALS2 #1 and shALS2 #2) and the corresponding control (shLuc). H Quantification of ALS2 levels in the silenced cell lines compared to the shLuc control condition (one-way ANOVA, Post hoc Dunnett´s test, n = 4). I Representative Western blot images of active and total Rab5 in MKN74 Control (shLuc) and ALS2 knockdown (shALS2 #2) cells infected or not with H. pylori (Hp and NT, respectively). J Quantification of active Rab5 levels, normalized to total Rab5, and shown for each of the conditions indicated in (I), compared to the shLuc NT condition (one-way ANOVA, Post hoc Tukey´s test, n = 3). K Immunofluorescence analysis of β-catenin in MKN74 control cells (shLuc) and ALS2 knockdown (shALS2 #2), non-treated (NT), or infected (Hp) for 24 h with H. pylori (MOI 50). Nuclei are shown in blue (DAPI) and β-catenin in green (n = 3, Scale bar of 10 μm). L Quantification of β-catenin nuclear fluorescence intensity relative to the nuclear area in shLuc and shALS2 #2, infected and non-treated MKN74 cells (one-way ANOVA, Post hoc Tukey´s test, n = 3). M TCF/LEF promoter reporter assay in shLuc and shALS2 #2 MKN74 cells, left untreated (NT) or following bacterial infection (Hp) for 24 h (one-way ANOVA, Post hoc Tukey´s test, n = 3). *p < 0.05, ** p < 0.01, *** p < 0.001, ****p < 0.0001 ALS2 is a well-known activator of Rab5, and hence, we explored whether ALS2 was required for H. pylori-dependent Rab5 activation by using lentiviral-based expression of shRNA sequences targeting endogenous ALS2 (shRNA sequences shALS2 #1 and shALS2 #2; shLuc served as a control). The extent of ALS2 knockdown was evaluated by Western blotting, observing that the shALS2 #2 sequence was more efficient in knocking down endogenous ALS2 in MKN74 cells (Fig. 4G and 4H). Using these cells, we then determined whether ALS2 was a Rab5-GEF in MKN74 cells, by measuring Rab5-GTP levels under basal conditions. We observed that shALS2 #2 and, to a lesser extent, shALS2 #1 decreased basal Rab5-GTP levels when compared with shLuc cells (Supplementary Fig. 1A and 1B). Thereafter, the requirement for ALS2 in H. pylori-mediated Rab5 activation was determined, observing that knockdown of ALS2 prevented H. pylori-induced activation of Rab5 (Fig. 4I and 4J). Interestingly, ALS2 was required for H. pylori-induced β-catenin signaling, since shRNA-mediated knockdown of ALS2 precluded both nuclear accumulation of this protein (Fig. 4K and 4L) and TCF/LEF-dependent transcription following bacterial infection (Fig. 4M). These observations were further confirmed in reconstitution experiments, since transfection of GFP-ALS2, but not GFP alone, restored H. pylori-induced re-localization of β-catenin to the nucleus in shRNA-ALS2-treated cells (Supplementary Fig. 2). Interestingly, transfection with constitutively active Rab5 (Rab5/Q79L mutant) induced per se nuclear localization of β-catenin, regardless of H. pylori presence (Supplementary Fig. 2). The latter is in agreement with the anticipated sequence of events, whereby the upregulation of ALS2 is expected to precede Rab5 activation. H. pylori-dependent activation of Rab5 and upregulation of ALS2 involves HIF-1α stabilization It has been reported that HIF-1α, on the one hand, is upregulated in H. pylori-infected cells [7, 10] and, on the other hand, induces the expression of ALS2 under hypoxic conditions [13]. Therefore, we explored the role of HIF-1α in ALS2 upregulation following H. pylori infection in MKN74 cells. Importantly, bacterial infection triggered a time-dependent increase of HIF-1α levels, which peaked after 8 h of infection (Fig. 5A and 5B). Similarly, in AGS cells H. pylori induced an increase in HIF-1α levels, reaching a peak after 12 h (Supplementary Fig. 3A and 3B). Then, we determined the requirement of HIF-1α for ALS2 upregulation following H. pylori infection by using 30 μM of a small molecule HIF-1α inhibitor (SC-205346, HIFI), as previously described [13, 24]. This inhibitor prevented HIF-1α stabilization following H. pylori infection, as shown by Western blotting (Supplementary Fig. 3C and 3D). Remarkably, HIF-1α inhibition precluded H. pylori-induced increases in ALS2 protein levels, as compared with the DMSO control condition (Fig. 5C and 5D). To further confirm the role of HIF-1α in H. pylori-dependent upregulation of ALS2, we also used a siRNA-based approach (siHIF-1α), and observed that downregulation of HIF-1α prevented the H. pylori-dependent increases in ALS2 levels, as compared with the control condition (siC) (Fig. 5E and 5F). Next, we determined whether HIF-1α stabilization was required for Rab5 activation. To this end, MKN74 cells were infected with H. pylori in the presence of the HIF-1α inhibitor SC-205346, or DMSO as a control. In the presence of control vehicle (DMSO), bacterial infection increased Rab5-GTP levels 1.5-fold; however, HIF-1α inhibition prevented Rab5 activation induced by H. pylori infection (Fig. 5G and 5H). In conclusion, these observations indicate that Rab5 activation by H. pylori is dependent on HIF-1α.Fig. 5H. pylori infection increases ALS2 levels in a HIF-1α-dependent manner. A HIF-1α levels MKN74 cells were measured following H. pylori infection at different time points (0 h to 24 h). Representative Western blot images are shown. B Quantification of HIF-1α levels after infection with the bacteria compared to 0 h (non-treated) (one-way ANOVA, Post hoc Dunnett´s test, n = 3). C Representative Western blot of ALS2 in MKN74 cells, either non-treated (NT) or infected with H. pylori (Hp) at MOI 50 for 24 h, either in the presence of 30 μM of the HIF-1α inhibitor SC-205346 (HIFI) or DMSO (control vehicle). D Quantification of ALS2 levels shown in (C) and shown relative to the non-treated, DMSO condition (one-way ANOVA, Post hoc Tukey´s test, n = 3). E Representative Western blot showing the ALS2 levels in MKN74 cells transfected with a pool of 3 siRNA sequences against HIF-1α or a control siRNA sequence (siC). F Quantification of ALS2 levels in siRNA HIF-1α and siRNA-control cells, and shown with respect to the non-treated siRNA-control cells (one-way ANOVA, Post hoc Tukey´s test, n = 3). G Rab5-GTP levels were measured by pulldown assay in MKN74 cells left untreated or infected with H. pylori at MOI 50 for 24 h (Hp), in the presence or absence of the HIF-1α inhibitor SC-205346 (HIFI). DMSO treatment was included as vehicle control condition. H Quantification of active Rab5 (Rab5-GTP) levels normalized to total Rab5, from experiments obtained in (G) (one-way ANOVA, Post hoc Tukey´s test, n = 3). *p < 0.05, ** p < 0.01 H. pylori infection increases ALS2 levels in a HIF-1α-dependent manner. A HIF-1α levels MKN74 cells were measured following H. pylori infection at different time points (0 h to 24 h). Representative Western blot images are shown. B Quantification of HIF-1α levels after infection with the bacteria compared to 0 h (non-treated) (one-way ANOVA, Post hoc Dunnett´s test, n = 3). C Representative Western blot of ALS2 in MKN74 cells, either non-treated (NT) or infected with H. pylori (Hp) at MOI 50 for 24 h, either in the presence of 30 μM of the HIF-1α inhibitor SC-205346 (HIFI) or DMSO (control vehicle). D Quantification of ALS2 levels shown in (C) and shown relative to the non-treated, DMSO condition (one-way ANOVA, Post hoc Tukey´s test, n = 3). E Representative Western blot showing the ALS2 levels in MKN74 cells transfected with a pool of 3 siRNA sequences against HIF-1α or a control siRNA sequence (siC). F Quantification of ALS2 levels in siRNA HIF-1α and siRNA-control cells, and shown with respect to the non-treated siRNA-control cells (one-way ANOVA, Post hoc Tukey´s test, n = 3). G Rab5-GTP levels were measured by pulldown assay in MKN74 cells left untreated or infected with H. pylori at MOI 50 for 24 h (Hp), in the presence or absence of the HIF-1α inhibitor SC-205346 (HIFI). DMSO treatment was included as vehicle control condition. H Quantification of active Rab5 (Rab5-GTP) levels normalized to total Rab5, from experiments obtained in (G) (one-way ANOVA, Post hoc Tukey´s test, n = 3). *p < 0.05, ** p < 0.01 To evaluate whether this signaling mechanism was detectable in a clinically relevant setting, gastric patient-derived organoids (GC-PDOs) derived from one patient were infected with H. pylori, and ALS2 levels were measured by indirect immunofluorescence analysis, finding that bacterial infection increased ALS2 presence in GC-PDOs (Fig. 6A and 6B). In addition, the nuclear localization of β-catenin was also increased in human organoids infected with the bacteria, further supporting the notion that H. pylori infection promotes β-catenin signaling in a clinically relevant context (Fig. 6C and 6D). These results indicate that H. pylori infection increases ALS2 levels in different structures present in the same sample of GC-PDO, promotes nuclear translocation of β-catenin, and increases total β-catenin protein level (Supplementary Fig. 4). Additionally, we performed a bioinformatics analysis of the expression of six genes, namely HIF1A and ALS2 (Fig. 7A and 7B), as well as the β-catenin target genes CCND1, MYC, MMP7, and PTGS2 (Fig. 7C-7F), by comparing mRNA levels found in human GC samples versus non-tumor adjacent mucosa (NTAM) across independent ACRG and TCGA-STAD cohorts. Differential expression analyses were conducted within each dataset using both paired (patient-matched) and unpaired models, and results were integrated using a random-effects meta-analysis. This analysis revealed increased expression of HIF1A, ALS2, CCND1, and MMP7 levels in GC relative to NTAM, while changes in MYC and PTGS2 expression levels were not statistically significant (Fig. 7). These findings are consistent with the notion that activation of hypoxia-associated and β-catenin-related transcriptional programs occur in gastric cancer.Fig. 6H. pylori infection increases ALS2 levels and induces nuclear translocation of β-catenin in human organoids. Human organoids were infected with H. pylori at MOI 100 for 24 h. A Representative confocal images showing the immunofluorescence of ALS2 in untreated organoids (left panels) and organoids infected with H. pylori (right panels). The nuclei are shown in blue (DAPI) and ALS2 in green. The panels in each figure show the magnifications of superposed images (merge) and for the ALS2 marker. B Quantification of ALS2 fluorescence intensity of non-treated (NT) or cells infected with H. pylori (Hp) relative to NT cells (t-test, n = 3, technical replicates). C Representative immunofluorescence images of β-catenin. Nuclei are shown in blue (DAPI) and β-catenin in red. The panels in each figure show the magnifications of superposed images (merge) and the magnified areas for the β-catenin marker. Images were obtained at 60X magnification with a Nikon C2 plus confocal microscope. D Quantification of the number of β-catenin-positive nuclei in untreated (NT) or H. pylori infected (Hp) cells relative to NT. (t-test, n = 3, technical replicates). Scale bar of 25 μm. ** p < 0.01Fig. 7H. pylori infection increases the expression of HIF1A, ALS2 and β-catenin target genes in human samples. Forest plot summarizing differential expression of (A) HIF1A, (B) ALS2 and β-catenin target genes ((C) CCND1, (D) MYC, (E) MMP7, and (F) PTGS2). The points represent log2 fold-change (logFC) estimates and horizontal bars indicate 95% confidence intervals derived from paired and unpaired analyses of the TCGA-STAD and ACRG (GSE66229) cohorts shown in gray. The combined effect from the random-effects model meta-analysis is shown in red. Differential expression reflects the comparison of gastric cancer (GC) and non-tumor adjacent mucosa (NTAM) H. pylori infection increases ALS2 levels and induces nuclear translocation of β-catenin in human organoids. Human organoids were infected with H. pylori at MOI 100 for 24 h. A Representative confocal images showing the immunofluorescence of ALS2 in untreated organoids (left panels) and organoids infected with H. pylori (right panels). The nuclei are shown in blue (DAPI) and ALS2 in green. The panels in each figure show the magnifications of superposed images (merge) and for the ALS2 marker. B Quantification of ALS2 fluorescence intensity of non-treated (NT) or cells infected with H. pylori (Hp) relative to NT cells (t-test, n = 3, technical replicates). C Representative immunofluorescence images of β-catenin. Nuclei are shown in blue (DAPI) and β-catenin in red. The panels in each figure show the magnifications of superposed images (merge) and the magnified areas for the β-catenin marker. Images were obtained at 60X magnification with a Nikon C2 plus confocal microscope. D Quantification of the number of β-catenin-positive nuclei in untreated (NT) or H. pylori infected (Hp) cells relative to NT. (t-test, n = 3, technical replicates). Scale bar of 25 μm. ** p < 0.01 H. pylori infection increases the expression of HIF1A, ALS2 and β-catenin target genes in human samples. Forest plot summarizing differential expression of (A) HIF1A, (B) ALS2 and β-catenin target genes ((C) CCND1, (D) MYC, (E) MMP7, and (F) PTGS2). The points represent log2 fold-change (logFC) estimates and horizontal bars indicate 95% confidence intervals derived from paired and unpaired analyses of the TCGA-STAD and ACRG (GSE66229) cohorts shown in gray. The combined effect from the random-effects model meta-analysis is shown in red. Differential expression reflects the comparison of gastric cancer (GC) and non-tumor adjacent mucosa (NTAM) Discussion This study provides insights into the molecular mechanisms by which H. pylori promotes the activation of a signaling pathway involving the stabilization of HIF-1α, which is required for H. pylori-dependent upregulation of the Rab guanine nucleotide exchange factor ALS2 and subsequent activation of Rab5. This, in turn, promotes nuclear translocation of β-catenin and transcription-dependent events by sequestering GSK3β at endosomal compartments. GSK3β is required for β-catenin phosphorylation, which, together with the kinase CK1α, creates a degron for subsequent degradation via the proteasome [32, 33]. Our data show that H. pylori infection promotes the localization of GSK3β to Rab5-positive endosomes and increases the resistance of GSK3β against proteinase K-mediated digestion, suggesting endosomal sequestration of this protein, as previously shown by Taelman and others [20, 34]. Although it is intriguing to speculate that the whole destruction complex undergoes endosomal sequestration following H. pylori infection, this possibility requires further experimental confirmation by evaluating the presence of other components of the destruction complex, including CK1α, APC, and Axin. The signaling axis triggered by H. pylori begins with HIF-1α-dependent upregulation of ALS2 at both the mRNA and protein levels, as shown by pharmacological inhibition and siRNA-mediated approaches that prevented HIF-1α stabilization. This is interesting because the ALS2 gene is known to contain functional hypoxia response elements (HRE), which are HIF-1α binding sequences, in a proximal region of its promoter [13]. Thus, it is possible that H. pylori infection promotes HIF-1α-dependent transcription of the ALS2 gene in gastric cancer cells, although more experiments are needed to confirm this transcriptional mechanism. Thereafter, elevated ALS2 levels account for increased Rab5 activity (i.e., Rab5-GTP levels) in H. pylori-infected cells, which in turn promotes nuclear localization of β-catenin, as discussed above. These signaling events appear to be sequential, which is further supported by experiments based on shRNA-mediated silencing of ALS2, as well as in reconstitution experiments with recombinant ALS2 and constitutively active Rab5 (Rab5/Q79L mutant). However, these data need to be interpreted with caution, because the experimental approach used here suffered from drawbacks associated with the multiplicity of treatments known to challenge cell integrity. These included shRNA treatment, GFP-transfection, and then bacterial infection, all of which individually affected cell recovery. For this reason, it was not possible to reach the appropriate number of biological replicates, and so raw data from different samples within the same biological replicate were compared. Implicit in this limitation is the need for additional alternative experiments in the future to further dissect in greater detail the sequence of signaling events triggered by H. pylori infection. The positive effects of H. pylori on Rab5-GTP levels were observed in different cell settings, ranging from moderately differentiated gastric adenocarcinoma (MKN74 and AGS cells), through non-tumor gastric cells (GES-1) [35–37], which is in agreement with the notion that Rab5 is involved at different stages of malignant transformation and tumor progression [18, 38]. However, the precise role of this small GTPase in gastric carcinogenesis remains controversial and requires further research [39, 40]. On the other hand, the finding that H. pylori infection increased the amount of Rab5- and EEA1-positive large particles supports the observations of Rab5 activation, which is in agreement with the notion that activated Rab5 promotes endosome fusion and maturation [41]. Interestingly, increased endosome size and number were accompanied by re-localization of these structures to the cell periphery, a phenomenon that has been previously associated with increased cell adhesion and spreading [18], focal adhesion turnover [22, 42] and thus cell migration and invasion [18, 19, 22]. Hence, whether H. pylori infection promotes tumor cell migration and invasion via activation of Rab5-dependent signaling remains an intriguing possibility that requires future research. In this study, the identification of ALS2 as a new target of H. pylori was validated in relevant clinical settings, since the infection of gastric cancer patient-derived organoids increased ALS2 levels, which correlated with nuclear translocation of β-catenin in these structures. We acknowledge that the use of a single GC-PDO line is a limitation regarding biological generalizability. However, the consistent observations across multiple organoid structures within the same preparation support the robustness of the measured effect at the technical level. Future studies incorporating additional GC-PDO lines from a diverse patient cohort will be essential to confirm the broader applicability of these findings across different molecular subtypes of gastric cancer. Bioinformatic analyses supported these observations, which were associated with the expression of β-catenin target genes in patient samples, including cyclin D1 and matrix metalloproteinase-7. These genes are relevant in the context of gastric cancer, since both have been associated with malignancy by promoting the proliferation and invasion of gastric cancer cells, respectively [43]. In conclusion, our findings indicate that H. pylori infection induces HIF-1α, which in turn is required for H. pylori-dependent upregulation of the GEF ALS2, thereby allowing for Rab5 activation. This cascade promotes endosomal sequestration of GSK3β and nuclear translocation of β-catenin (Fig. 8, proposed model). Furthermore, taking the results documented here and previous reports into account, we propose that the endosomal sequestration of GSK3β is a late event necessary for a complete and sustained WNT/β-catenin response [20, 34]. In the clinical context of H. pylori infection, this mechanism may contribute to persistent oncogenic signaling and thereby facilitate tumor progression in gastric adenocarcinoma.Fig. 8Proposed mechanism of H. pylori infection-mediated induction of β-catenin translocation. A Initially, H. pylori promotes the stabilization of HIF-1α, which increases the expression of the GEF ALS2 in gastric epithelial cells. B Subsequently, ALS2 activates Rab5 and induces endosomal sequestration of GSK3β, a component of the β-catenin destruction complex, thereby permitting the translocation of newly synthesized β-catenin to the nucleus, and the transcription of β-catenin target genes related to proliferation and invasion Proposed mechanism of H. pylori infection-mediated induction of β-catenin translocation. A Initially, H. pylori promotes the stabilization of HIF-1α, which increases the expression of the GEF ALS2 in gastric epithelial cells. B Subsequently, ALS2 activates Rab5 and induces endosomal sequestration of GSK3β, a component of the β-catenin destruction complex, thereby permitting the translocation of newly synthesized β-catenin to the nucleus, and the transcription of β-catenin target genes related to proliferation and invasion Supplementary Information Supplementary Material 1: Supplementary Figure 1. Rab5 activation in MKN74 ALS2 knockdown cells. A. Representative Western blot image showing Rab5 activation in MKN74 control cells (shLuc) or following ALS2 silencing (shALS2 #1 and shALS2 #2). B. Quantification of Rab5 activation in the shLuc, shALS2 #1 and shALS2 #2 cells. (one-way ANOVA, Post hoc Dunnett´s test, n=3). *p < 0.05, **p < 0.01. Supplementary Material 2: Supplementary Figure 2. Reconstitution experiments with ALS2 and constitutively active Rab5. A. Immunofluorescence analysis of β-catenin in either control (shLuc) or shRNA-ALS2-treated (shALS2 #2) MKN74 cells, transfected with either GFP alone, GFP-ALS2 or GFP-Rab5/Q79L. Cells were infected or not with H. pylori at MOI 50 for 24 h and immunofluorescence analysis was performed. On the left, the nuclei are shown in blue (DAPI), GFP, GFP-ALS2, or GFP-Rab5/Q79L in green (Alexa 488), and on the right, β-catenin in red (Alexa 594) together with the superposition of all the images. Images were obtained at 60X magnification with a Nikon C2 plus confocal microscope (Scale bar of 10 μm). B. Quantification of relative nuclear β-catenin levels. A total of 6 to 15 cells per condition indicated in (A) were measured and are shown in the graph. Supplementary Material 3: Supplementary Figure 3. HIF-1α levels in the AGS and MKN74 cells after different treatments. A. Representative Western blot of HIF-1α at different times (0 h to 24 h) of H. pylori infection in AGS cells. B. Quantification of HIF-1α levels after infection with the bacteria compared to time 0 h (no treatment), (one-way ANOVA, Post hoc Dunnett´s test, n=3). C. Western blot of HIF-1α levels in untreated (NT) and H. pylori-infected (Hp) MKN74 cells, in the presence of the HIF-1α inhibitor SC-205346 (HIFI) or DMSO as a vehicle control. The infection was performed at a MOI of 50 for 8 h, in the presence of 30 μM of the HIF-1α inhibitor SC-205346 or the corresponding DMSO vehicle control. D. Quantification of HIF-1α levels in each of the above conditions compared to the NT DMSO condition. (one-way ANOVA, Post hoc Tukey´s test, n=3). *p<0.05, ** p<0.01, *** p<0.001. Supplementary Material 4: Supplementary Figure 4. β-catenin total levels in gastric cancer patient-derived organoids (GC-PDOs). Organoids were infected with H. pylori at MOI 100 for 24 h. The figure shows the quantification of β-catenin fluorescence intensity relative to GC-PDO area in H. pylori infected (Hp) or non-treated GC-PDOs. (t-test, n=10, *** p<0.001). Supplementary Material 5. Original Western blot images. Supplementary Material 1: Supplementary Figure 1. Rab5 activation in MKN74 ALS2 knockdown cells. A. Representative Western blot image showing Rab5 activation in MKN74 control cells (shLuc) or following ALS2 silencing (shALS2 #1 and shALS2 #2). B. Quantification of Rab5 activation in the shLuc, shALS2 #1 and shALS2 #2 cells. (one-way ANOVA, Post hoc Dunnett´s test, n=3). *p < 0.05, **p < 0.01. Supplementary Material 2: Supplementary Figure 2. Reconstitution experiments with ALS2 and constitutively active Rab5. A. Immunofluorescence analysis of β-catenin in either control (shLuc) or shRNA-ALS2-treated (shALS2 #2) MKN74 cells, transfected with either GFP alone, GFP-ALS2 or GFP-Rab5/Q79L. Cells were infected or not with H. pylori at MOI 50 for 24 h and immunofluorescence analysis was performed. On the left, the nuclei are shown in blue (DAPI), GFP, GFP-ALS2, or GFP-Rab5/Q79L in green (Alexa 488), and on the right, β-catenin in red (Alexa 594) together with the superposition of all the images. Images were obtained at 60X magnification with a Nikon C2 plus confocal microscope (Scale bar of 10 μm). B. Quantification of relative nuclear β-catenin levels. A total of 6 to 15 cells per condition indicated in (A) were measured and are shown in the graph. Supplementary Material 3: Supplementary Figure 3. HIF-1α levels in the AGS and MKN74 cells after different treatments. A. Representative Western blot of HIF-1α at different times (0 h to 24 h) of H. pylori infection in AGS cells. B. Quantification of HIF-1α levels after infection with the bacteria compared to time 0 h (no treatment), (one-way ANOVA, Post hoc Dunnett´s test, n=3). C. Western blot of HIF-1α levels in untreated (NT) and H. pylori-infected (Hp) MKN74 cells, in the presence of the HIF-1α inhibitor SC-205346 (HIFI) or DMSO as a vehicle control. The infection was performed at a MOI of 50 for 8 h, in the presence of 30 μM of the HIF-1α inhibitor SC-205346 or the corresponding DMSO vehicle control. D. Quantification of HIF-1α levels in each of the above conditions compared to the NT DMSO condition. (one-way ANOVA, Post hoc Tukey´s test, n=3). *p<0.05, ** p<0.01, *** p<0.001. Supplementary Material 4: Supplementary Figure 4. β-catenin total levels in gastric cancer patient-derived organoids (GC-PDOs). Organoids were infected with H. pylori at MOI 100 for 24 h. The figure shows the quantification of β-catenin fluorescence intensity relative to GC-PDO area in H. pylori infected (Hp) or non-treated GC-PDOs. (t-test, n=10, *** p<0.001). Supplementary Material 5. Original Western blot images.