Low-dose 5-hydroxymethylfurfural mitigates radiation -induced intestinal toxicity via HIF2α-driven IL22/STAT3 signaling enhancement

Abstract

Background

Ionizing radiation (IR)-induced intestinal injury limits the efficacy of radiotherapy for abdominal/pelvic malignancies, and there are few effective preventive measures. 5-Hydroxymethylfurfural (5-HMF), a bioactive compound abundant in heat-processed foods and herbal decoctions, has shown therapeutic benefits in inflammatory diseases, yet its role in IR-induced intestinal damage remains unclear.

Methods

C57BL/6 male mice (n = 15~17/group) were administered intraperitoneal vehicle, low-dose (50 mg/kg/day), or high-dose (200 mg/kg/day) 5-HMF for 7 days prior to 8–10 Gy of whole-body irradiation (WBI) or total abdominal irradiation (TAI). Assessments included survival, weight recovery, intestinal permeability, crypt survival, as well as proliferation and apoptosis of intestinal stem cells (ISCs). Mechanistic investigations utilized RNA sequencing of crypts, organoid cultures, dual luciferase assays, and pharmacological hypoxia-inducible factor-2α (HIF-2α) inhibition (PT-2385) to explore the radioprotective mechanisms of 5-HMF.

Results

Compared with the control group or high-dose group, low-dose 5-HMF significantly improved survival, accelerated weight recovery, and enhanced crypt regeneration. It promoted ISCs proliferation while inhibiting apoptosis. Mechanistically, 5-HMF stabilized HIF-2α, which bound to hypoxia-response elements (HREs) in the Interleukin-22 receptor-1 (IL22R1) promoter, thereby upregulating IL22R1 expression and amplifying IL22-dependent signal transducer and activator of transcription 3 (STAT3) phosphorylation. Importantly, PT-2385-mediated HIF-2α inhibition abolished the effects of 5-HMF on IL22R1/STAT3 and its radioprotective role both in vivo and in organoids.

Conclusions

Low-dose 5-HMF protects against IR-induced intestinal injury by stabilizing HIF-2α to enhance IL22/STAT3 signaling and drive ISC-mediated epithelial regeneration, identifying it as a promising dietary-derived radioprotectant for such damage.

Supplementary information

The online version contains supplementary material available at 10.1186/s12967-026-07757-3.

Background

Radiotherapy (RT) is a mainstay treatment for abdominal and pelvic malignancies, including prevalent cancers such as colorectal, cervical, and pancreatic carcinomas. However, its effectiveness is severely limited by radiation-induced intestinal injury (RII), a major dose-limiting toxicity [1, 2]. The pathogenesis of acute RII involves extensive apoptosis of crypt epithelial cells, particularly crucial intestinal stem cells (ISCs). This damage results in villus blunting, loss of mucosal barrier integrity, and significant inflammation. Clinically, this condition manifests as debilitating symptoms, including severe diarrhea, malnutrition, and profound weight loss, and can lead to fatal complications such as sepsis. Despite its clinical importance, safe, effective, and practical agents specifically designed to prevent or mitigate intestinal damage during RT are lacking. Consequently, RII often forces clinicians to interrupt treatment schedules or reduce radiation doses below curative levels. This directly compromises local tumor control and overall survival rates for patients. This significant unmet clinical need highlights the urgent need for novel radioprotective strategies targeting the intestinal epithelium.

5-Hydroxymethylfurfural (5-HMF), a bioactive heterocyclic aldehyde (C₆H₆O₃), is a Maillard reaction product that is abundant in heat-processed foods and serves as a key constituent of traditional Chinese medicines (TCMs), including Rehmannia glutinosa Praeparata, Polygonum multiflorum, and black garlic extracts [3–5]. Its pharmacological significance is highlighted during TCM processing, where stachyose conversion in Rehmannia glutinosa increases the 5-HMF content fourfold, establishing it as both a quality marker and primary therapeutic agent [5]. Recent studies have reported that 5-HMF exerts multiple protective biological activities: it functions as a potent antioxidant by scavenging reactive oxygen species (ROS), exhibits anti-inflammatory effects through the suppression of proinflammatory mediators (e.g., nitric oxide, TNF-α) in macrophages, and confers hypoxia-protective properties via allosteric modulation of hemoglobin to enhance oxygen diffusion gradients [6–10]. These mechanisms collectively substantiate its protective effects against ischemia-induced neuronal damage [11], alcoholic liver injury [12], osteoarthritis-associated cartilage injury [13] and cardiovascular stress [4]. However, whether 5-HMF can protect against radiation-induced intestinal injury (RII) remains unexplored.

The intestinal epithelium is highly radiosensitive because of rapid cellular turnover [14]. ISCs, located at the base of the crypts, continuously differentiate, mature, and migrate upward to replenish the epithelial lining. These ISCs are indispensable for maintaining epithelial homeostasis and facilitating postradiation regeneration [15]. Research has demonstrated that ISCs are characterized by specific markers, including Lgr5 and Olfm4 [16]. Importantly, their regenerative capacity not only depends on conserved signaling pathways such as the Wnt, Bmp, Notch, Yap, and PPAR-δ pathways but is also modulated by immune cells that secrete inflammatory mediators, including cytokines and chemokines, which collectively orchestrate epithelial repair [17–21]. Among these mediators, interleukin-22 (IL-22), which is produced by group 3 innate lymphoid cells (ILC3s) and γδ T cells, confers significant radioprotection against RII. IL-22 binding to its heterodimeric receptor, including IL-10R2 and IL-22R1, activates signal transducer and activator of transcription 3 (STAT3), thereby promoting ISC proliferation [22–25].

Moreover, the intestinal mucosa naturally exists within a steep physiologic hypoxia gradient. Recent studies have demonstrated that increased HIF-2α signaling alleviates radiation-induced intestinal damage by inhibiting the activity of prolyl hydroxylase (PHD), which normally targets HIF proteins for degradation via hydroxylation [26, 27]. Although 5-HMF has been reported to stabilize HIF-1α in vivo through vitamin C sequestration and PHD inhibition [28], its effects on intestinal HIF isoforms such as HIF-2α and potential crosstalk with the IL-22/STAT3 pathway remain uncharacterized. In the present study, we demonstrated that a moderate dose of 5-HMF can promote ISC regeneration via HIF-2α-mediated augmentation of IL22/STAT3 signaling. This mechanism contributes to the mitigation of ionizing radiation-induced intestinal injury and an improvement in mouse survival rates. Collectively, these findings suggest that 5-HMF has potential as an effective radioprotective agent in the intestine.

Materials and methods

Chemicals and reagents

5-HMF was confirmed by high-performance liquid chromatography (HPLC) and purchased from Sigma‒Aldrich (St. Louis, MO, USA). Sterile saline was obtained from Beyotime Biotechnology (Shanghai, China), and PBS buffer was purchased from HyClone Company (New York, USA). BrdU and reagents related to intestinal organoid culture, including DMEM/F12 medium, mEGFP, Noggin, Rspondin-1, GlutaMAX, B27 supplement and N2 supplement, were purchased from Thermo Fisher Scientific (Waltham, MA, USA). FITC-dextran, penicillin, streptomycin, N-acetyl cysteine and HEPES were purchased from Sigma (St. Louis, MO, USA). Recombinant murine IL-22 protein (rmIL22) and CHIR99021 were purchased from R&D Systems (Minneapolis, MN, USA). The information for other 5-bromo-20-deoxyuridine (BrdU) reagents is summarized in the corresponding section of the article.

Animals and treatment

Male C57BL/6 mice, aged 6–8 weeks, were purchased from Chengdu Dossy Biological Technology Company (License No. SCXK (Chuan) 2024–0031) and housed under controlled conditions: temperature of 24 ± 1 °C, humidity of 40 ± 5%, and a 12-hour light/dark cycle. Experimental protocols involving animals were approved by the Laboratory Animal Center of Chongqing University (Approval ID: CQU-IACUC-RE-202501-009) in compliance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The mice were randomly assigned to three groups (n = 15~17 per group): 1) the vehicle control (sterile saline, 200 μL/d) group; 2) the 5-HMF low-dose group (5-HMF, 50 mg/kg/d); and 3) the 5-HMF high-dose group (5-HMF, 200 mg/kg/d). All treatments were administered via daily intraperitoneal injection for 7 consecutive days prior to X-ray irradiation. Following pretreatment, the mice received localized X-irradiation under anesthesia (2% isoflurane) via an X-RAD 320 irradiator (Precision X-ray Inc.) with the following dose regimens: 10 Gy whole-body irradiation (WBI, lethal dose), 8 Gy total abdominal irradiation (TAI, sublethal dose), and 10 Gy abdominal irradiation (TAI, lethal dose). Radiation was delivered at 1.8 Gy/min via an RS-2000 X-ray Biologic Irradiator. Throughout the study, all the mice had unrestricted access to food and water. At the end of the experiment, the animals were injected intraperitoneally with pentobarbital sodium (1%) for anesthesia.

FITC-dextran assay for intestinal permeability

Intestinal barrier integrity was evaluated on day 3 post-10 Gy TAI using fluorescein isothiocyanate-dextran (FITC-dextran; 4 kDa). Following a 4-hour fast period with access to water, the mice received 600 mg/kg FITC-dextran dissolved in sterile saline via oral gavage under light-protected conditions. After tracer circulation for 4 hours, the animals were anesthetized with sodium pentobarbital, and blood samples were obtained through retro-orbital puncture. The blood was immediately centrifuged at 3000 × g for 10 min in light-shielded tubes to isolate the serum. FITC-dextran fluorescence in serum samples was quantified via a BioTek Synergy Neo2 multimode plate reader (excitation: 490 ± 10 nm; emission: 520 ± 10 nm). Sample concentrations were determined against a standard curve of serially diluted FITC-dextran in naive mouse serum (0–1000 μg/mL).

Crypt insolation and intestinal organoid culture

The crypts in the mice in the above groups were isolated as described previously [29]. Briefly, the proximal two-thirds of the small intestine were harvested and rinsed with ice-cold Hank’s balanced salt solution (HBSS) without Ca²⁺/Mg²⁺. The intestine was longitudinally opened and divided into approximately 5 cm fragments. These fragments were then incubated in EDTA solution (2.5 mM) with gentle agitation at 4 °C for 30 minutes. The supernatant was subsequently aspirated, and the fragments were washed with 10 mL of cold HBSS and vortexed for 30 seconds with 3-second pulses. The resulting fractions were allowed to settle on ice for approximately 10 minutes and subsequently filtered through 70 μm pore cell strainers (BD Biosciences, Bedford, MA). The filtrate was then centrifuged at approximately 100 × g for 5 minutes, and the resulting pellet was resuspended in 1 mL of DMEM/F12. The isolated crypts were quantified and embedded in Matrigel at a concentration of 10 crypts/μL. Approximately 50 to 100 crypts were plated per well in a 48-well plate and cultivated in crypt culture DMEM/F12 medium containing 50 ng/ml mEGFP, 100 ng/ml Noggin, 10% human Rspondin-1, 2 mM GlutaMAX, 10 mM HEPES, 100 μg/ml penicillin and 100 μg/ml streptomycin, 1 mM N-acetyl cysteine, B27 supplement, and N2 supplement. The medium was changed every 2 days.

Organoid treatment and measurement

Organoids were treated with rmIL22 (10 ng/ml) or CHIR99021 (3 μM) in conjunction with medium changes. For the assessment of organoid growth, total organoid counts per well were counted via light microscopy to evaluate colony formation efficiency. Complexity was evaluated by counting the number of buds in each organoid. To evaluate organoid size, an MTT solution (final concentration of 0.9 mg/ml; Beyotime, China) was used to stain the organoids. Images were captured via a camera to circumvent the limited depth of field and field of view inherent to microscopes. Area measurements were manually defined via ImageJ software, with a focus on organoid perimeters.

Immunofluorescence and immunohistochemistry (IHC)

Immunofluorescence staining was performed following previous protocols [30]. In brief, the animals were euthanized through decapitation, and segments of the small intestine were excised. These tissue samples were then fixed in zinc formalin fixative (Sigma, St. Louis, MO) for a 24-hour period. Postfixation, the tissues were processed and paraffin embedded. Subsequently, 3 μm sections were obtained via a microtome and positioned onto glass slides. These sections were subjected to deparaffinization and hydration, followed by a 95 °C boiling step in citric acid antigen retrieval buffer (Maixin; MVS-0100; Fuzhou, China) for 15 minutes. Following the cooling phase, the sections were rinsed with PBS and then blocked with 5% BSA (Beyotime, China). The next step involved overnight incubation of the sections with appropriately diluted primary antibodies. The primary antibodies used in this study included anti-BrdU monoclonal antibodies (1:200, AB10015219, BD Biosciences), anti-Ki67 monoclonal antibodies (1:200, AB16667, Abcam), anti-Olfm4 antibodies (1:200, 39,141, CST), anti-Lysozyme C antibodies (1:300, sc-27958, Santa Cruz), and anti-chgA antibodies (1:300, sc-393941, Santa Cruz). The sections were subsequently rinsed with PBS containing Tween-20 (PBST) and subsequently incubated with suitable secondary antibodies at room temperature for 1.5 hours. The secondary antibodies used included Donkey Anti-Mouse IgG (H+L) secondary antibody conjugated with Alexa Fluor 555 (1:300, Thermo, USA), Alexa Fluor 488 AffiniPure Donkey Anti-Rabbit IgG (H+L) (1:300, Jackson, USA), and Alexa Fluor 647-conjugated AffiniPure Donkey Anti-Rat IgG (H+L) (1:300, Jackson, USA). The slides were then subjected to multiple rinses and counterstained with DAPI (Beyotime, China) for 5 minutes. Following rinses with PBS, the slides were mounted with a prolonged gold antifade agent and sealed with coverslips. Fluorescence imaging was accomplished via either a fluorescence microscope (Olympus, BX63, Japan) or a Zeiss LSM 780 confocal microscope. For immunohistochemistry, the same procedural steps were followed, except for the secondary antibody incubation. In this case, secondary antibodies conjugated with HRP were employed, and staining was executed using DAB substrate solution in accordance with the manufacturer’s instructions (rabbit polymer detection system PV-6001; ZSGB-BIO, Beijing, China).

Promoter analysis and dual luciferase assays

The mouse IL22R1 promoter sequence, spanning from −2000 to +133 nucleotides (including the first exon) from the +1 transcriptional start site, was synthesized via PCR using specific primers (Supplementary Table S1) and mouse genomic DNA as the template. The fragment was subsequently inserted into the pGL4.10 (luc2) vector (E6651, Promega, Madison, WI, USA) via the ClonExpress Ultra One Step Cloning Kit (C115-01, Vazyme, Nanjing, China) with KpnI and XhoI cleavage sites. All the mutant-promoter vectors were constructed in a similar manner. The coding sequences of mouse HIF-1 and HIF-2 were ligated into the pCMV6-Entry vector via the EcoRI and HindIII sites. HEK293 cells were cultured in DMEM supplemented with 10% FBS and 2 mM L-glutamine. Transfection of the cells was carried out via Lipofectamine™ 3000 transfection reagent (L3000075; Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s standard protocol. For example, in a 96-well opaque plate, 5 × 10³ cells/well were seeded and transfected with 80 ng of mHIF-1 plasmid, 80 ng of mHIF-2 plasmid, or GFP plasmid, in addition to 90 ng of the IL22R1-luciferase construct and 10 ng of Renilla luciferase. The transfected cells were then incubated for 48 hours in Opti-MEM (31985070, Gibco, Waltham, MA, USA). Dual luciferase assays were performed via the Dual-Glo Reagent (E2940, Promega, Madison, WI, USA) following the manufacturer’s instructions. Luminescence data were measured via a Cytation 3 luminometer (Biotek Instruments, Inc.), and the luciferase signal was normalized to the Renilla signal.

Immunoblotting analysis

For western blot analysis, a procedure established in a previous publication was used [31]. In brief, organoids cultured in vitro or crypts isolated from the small intestine were lysed in RIPA buffer (PI89900, Thermo) supplemented with a protease and phosphatase inhibitor cocktail (5872, CST, Danvers, MA, USA). After sonication, the protein content was determined via a bicinchoninic acid (BCA) assay. Lysates containing 30 μg of protein per well were loaded onto a 10% polyacrylamide gel for electrophoresis. Proteins were subsequently transferred onto PVDF membranes. The membranes were blocked with 2% nonfat milk at room temperature for one hour, followed by overnight incubation at 4 °C with primary antibodies. The primary antibodies used were anti-phospho-Stat3 (Y705) (D3A7) (9145, CST), anti-Stat3 (D3Z2G) (12640, CST), and anti-HIF-1α (D2U3T) (14179, CST) and anti-HIF-2α (E8E5Z) (71565, CST), anti-IL22R1 (SAB1407826, Sigma), anti-IL10R2 (HPA065647, Sigma), β-actin (93473, CST) and GAPDH (60004, Proteintech). After primary antibody incubation, the membranes were probed with an anti-rabbit HRP secondary antibody (7074, CST), and visualization was achieved via enhanced chemiluminescence (ECL) Western blotting substrate (32106, Thermo).

RT‒qPCR

Total RNA was extracted from crypts or organoids via TRIzol reagent (15596026CN, Invitrogen) according to the manufacturer’s instructions. cDNA synthesis was performed via a reverse transcription kit (Takara, Japan). Gene expression was quantified via real-time PCR with SYBR Green (Takara, Japan) on a CFX96 Touch system (Bio-Rad). The list of PCR primers is provided in Supplementary Table S1.

Trancriptomic analysis

For transcriptomic analysis, total RNA was extracted from intestinal crypts isolated from the duodenum of mice 2 days after exposure to 8 Gy total abdominal irradiation (TAI), following the standard instructions of Trizol reagent (Takara). Briefly, crypts from 10 mice per group were pooled and homogenized in 1 mL Trizol reagent and stored at −80 °C. Library preparation was performed using the NEBNext Ultra II RNA Library Prep Kit for Illumina (New England Biolabs). The process included mRNA enrichment with oligo(dT) magnetic beads, fragmentation, first- and second-strand cDNA synthesis, end repair, adapter ligation, and PCR amplification. The resulting libraries were sequenced on an Illumina NovaSeq 6000 platform to generate 150 bp paired-end reads, yielding approximately 40 million raw reads per sample. Raw sequencing reads were subjected to quality control using FastQC (v0.11.9) and were subsequently trimmed to remove adapters and low-quality bases using Trimmomatic (v0.39). The high-quality clean reads were then aligned to the mouse reference genome (GRCm39) using HISAT2 (v2.2.1). Differential expression analysis was conducted using the DESeq2 package (v1.30.1) in R. Genes with an adjusted p-value (False Discovery Rate, FDR) < 0.05 and an absolute log2 fold change (|log2FC|) > 1 (corresponding to a linear fold change > 2) were defined as statistically significant differentially expressed genes (DEGs). The complete list of expressed genes, including gene symbols, log2 fold changes, p-values, FDR, and functional annotations, is provided in Supplementary Table S2. The functions and signaling pathways of the identified DEGs were further investigated through Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses.

Statistical analysis

All the quantitative experimental results were subjected to tests for normal distribution and variance homogeneity via SPSS software version 17.0 prior to statistical analysis. Two-tailed Student’s t test was employed for significance determination when only two groups were compared. For comparisons among three or more groups, one-way ANOVA was conducted, followed by Tukey’s or Dunnett’s test. Statistical significance levels were set at P < 0.05 (*), P < 0.01 (**), and P < 0.001 (***). All the data are presented as the means ± SDs.

Results

Low-dose 5-HMF mitigates radiation-induced mortality and intestinal toxicity

To investigate the effects of various doses of 5-HMF on radiation-induced intestinal toxicity, C57BL/6J mice were divided into three groups daily: sterile saline (vehicle control), 50 mg/kg 5-HMF (low-dose), or 200 mg/kg 5-HMF (high-dose) for 7 days before irradiation (Fig. 1A). Following 10 Gy whole-body irradiation (WBI), low-dose 5-HMF-treated mice exhibited 26.6% survival on day 10, whereas all the vehicle- and high-dose groups died before day 9 (Fig. 1B). To assess the response to pelvic/abdominal radiation, the mice received 8 Gy total abdominal irradiation (TAI). Low-dose 5-HMF treatment initiated body weight recovery by day 5 postirradiation (dpi), which occurred 3–4 days earlier than those in the other two groups, with full restoration to pre-TAI weights at 10–12 dpi (Fig. 1C). At the lethal dose of 10 Gy TAI, all the vehicle-treated and high-dose-treated mice died by 8 dpi with severe weight loss, whereas the low-dose-treated 5-HMF-treated mice achieved 100% survival with weight recovery (Fig. 1D). Within 3–4 days post-10 Gy TAI, the mice developed diarrhea, as quantified by decreased stool counts. Manual fecal pellet collection revealed 65% and 47.9% reductions in the number of formed stools at 5 dpi in the control and high-dose groups, respectively, versus only 16–22% with low-dose 5-HMF (Fig. 1E). The epithelial integrity was evaluated via a FITC-dextran assay. Four hours after gavage, the mice given low-dose 5-HMF exhibited decreased FITC-dextran uptake in the bloodstream following 10 Gy TAI by 4 ~ 5-fold over those in the other two groups, indicating attenuated radiation-induced intestinal permeability (Fig. 1F). Overall, low-dose 5-HMF administration improved survival, accelerated weight recovery, preserved intestinal barrier function, and suppressed inflammatory responses, demonstrating radioprotective efficacy in vivo.

Fig. 1.

Low-dose 5-HMF mitigates radiation-induced mortality and intestinal toxicity. (A) scheme of the experimental program. Male C57BL/6 mice received daily intraperitoneal injections of vehicle control (n = 16), low-dose (50 mg/kg, n = 17), or high-dose (100 mg/kg, n = 15) 5-HMF for 7 consecutive days. On day 7, the mice were subjected to 8‒10 Gy TAI or WBI. Survival, body weight, and intestinal injury were monitored at the indicated time points postirradiation. (B) Kaplan‒Meier analysis of the mice exposed to WBI for 10 Gy, n = 15–17 mice/group. (C and D) body weights of mice subjected to 8 Gy or 10 Gy TAI at the indicated time points (n = 15/group). The mice were sacrificed if they exhibited any signs of distress. Dpi represents days post-irradiation, n = 15/group. (E) stool counts from individual mice in metabolic cages from day 0 to day 5 after receiving 10 Gy TAI (n = 6/group). (F) measurement of epithelial integrity by FITC-dextran (4 kDa) gavage at a dose of 0.5 mg/kg. The data are presented as the means±sds. T test for the corresponding panels. Statistical significance compared with the control group (sterile saline administration). *P < 0.05, **P < 0.01, and NS, no significant difference

Low-dose 5-HMF promotes ISC-mediated epithelial injury repair postradiation

To evaluate the potential of 5-HMF to enhance radiation-induced intestinal repair, we collected intestinal tissues from mice in the above three groups at 3 days after WBI. Ex vivo measurements of the stretched small intestine revealed significantly greater small intestinal length in the low-dose 5-HMF group than in both the vehicle control and high-dose 5-HMF groups, with a consistent 4–6 cm increase observed across biological replicates (Fig. 2A and B). Concurrently, severe hemorrhagic foci were evident along the intestinal serosa in 100% of the control and high-dose 5-HMF-treated mice, whereas the low-dose 5-HMF cohort displayed no macroscopic hemorrhage (Fig. 2A). Hematoxylin‒eosin (HE) staining and immunohistochemistry revealed that mice in the control and high-dose 5-HMF groups presented visibly sparse, shorter, and less robust villi throughout the small intestine (Fig. 2C). Additionally, compared with those in the low-dose 5-HMF group, there were significantly fewer crypts and fewer BrdU⁺ transient amplifying cells throughout the entire intestine, including the duodenum, jejunum, ileum, and colon (Fig. 2D–F). Intestinal regenerative capacity relies primarily on Lgr5⁺ ISCs located at the base of crypts. To investigate the role of ISCs in epithelial repair following radiation in mice with low-dose 5-HMF, we performed immunostaining for olfactomedin-4 (Olfm4, a marker for Lgr5⁺ ISCs), lysozyme (Lyz, a Paneth cell marker critical for maintaining the ISC-niche microenvironment), and chromogranin A (ChgA, an enteroendocrine cell marker). After 8 Gy TAI, mice treated with low-dose 5-HMF presented a 53% increase in Olfm4⁺ cells within the crypts, accompanied by a notable increase in ki67⁺ transient amplifying (TA) cells, whereas the populations of Lyz⁺ and ChgA⁺ cells remained relatively unchanged. Furthermore, compared with the control group, these mice presented visibly deeper and enlarged crypts, indicating that low-dose 5-HMF facilitates ISC expansion following low-dose X-ray irradiation (Fig. 3A and B). At a higher radiation dose (10 Gy WBI), TUNEL staining revealed extensive ISC apoptosis in control mice. In contrast, low-dose 5-HMF-treated mice presented increased numbers of BrdU⁺ and Olfm4⁺ crypt cells, suggesting enhanced ISC survival and proliferation (Fig. 3C, E and F). Ki67 staining further confirmed that this group had thicker, longer, and denser crypts and villi (Fig. 3D and G). Collectively, these results demonstrate that pretreatment with low-dose 5-HMF enhances ISC proliferation and promotes the survival and regeneration of intestinal crypts after radiation.

Fig. 2.

Low-dose 5-HMF promotes intestinal structural repair and epithelial proliferation after radiation. (A) Representative macroscopic images of the small intestine from the three treatment groups at 3 days after WBI. (B) comparison of small intestinal length among the three treatment groups. n = 12 mice/group. (C and D) Representative photomicrographs of H&E and BrdU staining of the jejunum. The scale bar is 100 μm. (E) quantification of BrdU⁺ proliferating cells per crypt in the jejunum. (F) quantification of crypt numbers across the entire intestinal tract, including the duodenum, jejunum, ileum, and colon. n = 30 microscopic fields from 3 mice per group. The scale bar is 100 μm. The data are presented as the means ± SDs. Statistical significance was determined relative to the low-dose group; n = 12/group; *P < 0.05, **P < 0.01, NS = not significant

Fig. 3.

Low-dose 5-HMF facilitates intestinal stem cell (ISC) expansion following irradiation. (A) Representative immunostaining of the proteins marked with ISC (Olfm4, green), Paneth cells (Lyz, white) and enteroendocrine cells (chgA, red) as well as the proliferation marker ki-67 in the jejunum of mice administered sterile saline or low-dose 5-HMF at 10 hours after receiving 8 Gy of TAI; the scale bar represents 50 μm. (B) quantification of OLFM4⁺, LYS⁺ and Ki67⁺ cells per crypt as well as crypt depth (μm). (C) Apoptotic cells and proliferative cells were stained with TUNEL and BrdU (red) after 10 Gy WBI, and the scale bar represents 50 μm. (D) Representative immunohistochemical images of “Swiss roll” intestinal tissues stained with an anti-Ki67 antibody and imaged under 10× and 20× objective lenses under a microscope; the scale bar represents 50 μm. (E and F) quantification of TUNEL⁺ cells and BrdU⁺ cells (13 crypts per mouse, 10 mice/group, n = 130). (G) quantification of Ki67⁺ cells per crypt. (13 crypts per mouse, 10 mice/group, n = 130). The data are presented as the mean±SD. Statistical significance compared with the control group. *P < 0.05, **P < 0.01, *P < 0.001 and NS, no significant difference

Low-dose 5-HMF-induced ISC proliferation depends on radiation-induced niche signaling

To investigate whether low-dose 5-HMF directly enhances ISC proliferation, we utilized an ex vivo culture system to evaluate the organoid-forming capacity of isolated intestinal crypts. Organoids derived from low-dose 5-HMF-treated mice were comparable in size, quantity, and complexity (bud number) to those derived from control mice, indicating that the hyperproliferative phenotype of the crypts in vitro was not significant (Fig. 4A–D). Unexpectedly, when these organoids were exposed to 5 Gy X-ray irradiation, no significant differences were detected between the two groups (Fig. 4E–H). Moreover, under basal conditions without irradiation, low-dose 5-HMF administration alone did not alter the crypt architecture compared with that of the controls (Fig. S1A–C). However, after exposure to a sublethal dose of 5 Gy TAI, which induced minimal intestinal damage, 5-HMF-treated mice exhibited more significant crypt expansion in the duodenum and jejunum 2 days post-irradiation (Fig. S1D–F) and increased proliferation of Id1⁺ transient amplifying (TA) cells throughout the intestinal segments (Fig. S2A, B). Collectively, these data demonstrate that the radioprotective effects of 5-HMF on ISCs require in vivo contextual signals and are mediated through the modulation of the stem cell niche during radiation challenge.

Fig. 4.

Effect of low-dose 5-HMF on intestinal organoid growth with or without irradiation. (A, E) Bright-field images of intestinal organoids derived from control or low-dose 5-HMF-treated mice without (A) or with (E) 5 Gy irradiation. (B, F) Representative images of organoids stained with MTT reagent in 96-well plates under nonirradiated (B) and irradiated (F) conditions (n = 3 biological replicates; 3 wells per replicate per group). (C, G) quantification of the organoid surface area (μm²). (D, H) viability assessment based on MTT staining (% positive wells; n = 6 wells per group). The data are presented as the means ± SDs; NS, not significant

Low-dose 5-HMF enhances ISC proliferation through IL22/STAT3 signaling

To further investigate the potential mechanism involved, we conducted transcriptome profiling of crypts isolated from the duodenum of mice 2 days after 8 Gy TAI. Unsupervised hierarchical clustering of the RNA-seq data, encompassing 4555 genes across three biological replicates per group, revealed strong intragroup consistency and clear separation between the control and low-dose 5-HMF-treated groups (Supplementary Figure S3A; Supplementary Table S2). Differential gene expression analysis was performed using stringent thresholds (p < 0.05, FDR < 0.05, and fold change > 2), identifying 379 significantly upregulated and 120 downregulated genes in the 5-HMF-treated group relative to the control group. The overall expression patterns are visually summarized in a volcano plot (Fig. 5A; Supplementary Table S2). And the GO analysis revealed significant enrichment in terms related to tissue repair and regenerative processes, including wound healing, tissue remodeling, regulation of epithelial cell proliferation, and regulation of developmental growth. Additionally, terms associated with cellular homeostasis such as transition metal ion transport and regulation of apoptotic signaling pathway were also prominently enriched (Fig. 5B; Supplementary Table S3). Furthermore, we next investigated the expression of genes associated with key signaling pathways known to regulate ISC proliferation, including Wnt, Notch, Yap, PPAR-δ, and IL-22/STAT3. Notably, among these pathways, only the IL-22/STAT3 pathway exhibited substantial transcriptional alterations in response to 5-HMF treatment. In contrast, the expression of major genes within the Wnt, Notch, Yap, and PPAR-δ pathways remained largely unchanged, which was subsequently confirmed via quantitative real-time PCR, reinforcing the specificity of IL-22/STAT3 activation by low-dose 5-HMF (Fig. 5C–F).

Fig. 5.

Transcriptomic profiling of intestinal crypts reveals 5-HMF-induced regenerative and signaling pathways after radiation. (A) volcano plot of differentially expressed genes (DEGs). Genes meeting the thresholds of |log₂(fold change)| > 1 (linear Fold change > 2) and false discovery rate (FDR) < 0.05 are highlighted in red (upregulated, n = 379) and blue (downregulated, n = 120). Grey dots represent non-significant genes. (B) GO enrichment analysis of DEGs. The bar graph shows selected significantly enriched biological processes. (C-F) expression of select genes associated with Wnt (C), Notch, Yap (D), PPARδ (E) and IL22/STAT3 (F) in the crypts of the mice treated with low-dose 5-HMF and sterile saline, as determined by qPCR. (n = 6 mice/group, 3 replicates). The data are presented as the means ± SDs; *p < 0.05, **p < 0.01, ***p < 0.001, and NS, no significant difference (two-tailed Student’s t test)

Previous studies have shown that IL22 is produced by Group 3 innate lymphoid cells (ILC3s) after intestinal injury and plays a crucial role in ISC proliferation and expansion, as well as promoting the growth of intestinal organoids after radiation. To further confirm the role of the IL22/STAT3 pathway in this context, we isolated crypts and cultured them with recombinant murine IL22 (rmIL22). After 3 days of culture, rmIL22 significantly increased the budding of organoids derived from low-dose 5-HMF-treated mice (Fig. 6A and B). Additionally, the crypts were cultured with CHIR, an agonist of the Wnt/β-catenin signaling pathway. MTT staining and quantitative analysis revealed that, in the presence of rmIL22, crypts from low-dose 5-HMF-treated mice generated substantially larger organoids than did those from mice treated with CHIR (Fig. 6C and D). Given the significance of STAT3 signaling in response to IL22 for ISC maintenance, we evaluated STAT3 in this model and found that rmIL22 increased the phosphorylation level of STAT3 (Y705) in organoids from mice treated with low-dose 5-HMF compared with those from the controls (Fig. 6E and F). Collectively, these data indicate that low-dose 5-HMF enhances the IL22–STAT3 signaling cascade in crypts after radiation, ultimately promoting ISC proliferation and regeneration.

Fig. 6.

Low-dose 5-HMF sensitizes intestinal crypts to IL-22‒STAT3 signaling and enhances organoid growth. (A, B) organoid formation assay. Crypts isolated from control or low-dose 5-HMF-treated mice were cultured with recombinant murine IL-22 (rmIL22; 50 ng/mL) for 3 days. Representative time-lapse images (A) and quantification of the organoid surface area (μm²) from day 1 to day 3 after cultivation (B) are shown. (C) comparison of organoid growth stimulated by rmIL22 and CHIR99021 (3 μM), a Wnt/β-catenin agonist. Representative images of MTT-stained organoids. (D) MTT viability analysis (% positive, n = 6 wells/group). (E, F) immunoblot analysis (E) and quantification (F) of STAT3 phosphorylation (Tyr705) in organoids treated with or without rmIL22 (50 ng/mL, 48 h). The data are presented as the mean±SD. Statistical significance compared with the regular diet group. *p < 0.05, **p < 0.01 and NS, no significant difference

IL22R1 expression is directly regulated by HIFs via hypoxia response elements (HREs)

To investigate how 5-HMF enhances the IL22–STAT3 signaling pathway in ISCs, we examined the expression of IL22 receptors, including IL22R1 and IL10R2. Compared with that in the control group, the expression of IL22R1 significantly increased at both the mRNA and protein levels in the crypts of the mice treated with both low- and high-dose 5-HMF. However, the other heterodimeric receptor, IL10R2, showed no significant changes (Fig. 7A and B). These findings suggest that 5-HMF upregulates IL22R1 expression at the transcriptional level. To further elucidate these mechanisms, we analyzed the regulatory region involved in the transcriptional control of the IL22R1 gene and discovered four occurrences of the sequence “ACGTG” ranging from −1526 bp to +80 bp within the gene body. This “ACGTG” sequence has been reported as a core element of hypoxia response elements (HREs), which are bound by hypoxia-inducible factors (HIFs). The ChIP-seq data of LOVO cells (a human colon cancer cell line) retrieved from the GEO database indicate a significant enrichment peak of HRE elements in the transcriptional regulatory region of IL22R1, suggesting that IL22R1 may be a downstream gene regulated by HIFs (Fig. 7C).

Fig. 7.

5-HMF upregulates IL-22 receptor expression through HIF-dependent transcriptional activation. (A, B) expression of IL-22 receptor subunits in intestinal crypts. (A) mRNA and (B) protein levels of IL22R1 and IL10R2 in mice treated with vehicle, low-dose (50 mg/kg), or high-dose (100 mg/kg) 5-HMF. (C) Bioinformatics analysis of the IL22R1 promoter region. The schematic illustrates the positions of four putative hypoxia response elements (HREs; consensus core sequence) within the murine IL22R1 locus. Supporting evidence from publicly available human ChIP-seq data (GEO database) is shown below. The track labeled “Batch4_CHROM1_LOVO_EPAS1_RABBIT” visualizes HIF-2α (encoded by EPAS1) binding sites on chromosome 1 in the LOVO human colon epithelial cell line (experimental batch 4; rabbit antibody). The prominent peak (red box) indicates significant enrichment of HIF-2α binding within the upstream regulatory region of the human IL22R1 gene, supporting its status as a direct transcriptional target under HIF-mediated regulation. (D) mouse IL22R1 promoter scheme showing distal (−1525~-1449 bp) and proximal (+36 ~ +74 bp) HERs identified by sequence analysis. (E) dual-luciferase reporter assay in HEK-293 cells with an IL22R1 promoter-luciferase construct containing a wild-type sequence (−2000 wt) or mutations of distal or/and proximal HERs with transactivation by HIF-1, HIF-2, or GFP-control plasmids (n = 3 transfections/group). Asterisks (*) indicate the comparison of HIF1 or HIF2 to their respective GFP controls. ** p < 0.01, *p < 0.001 and NS, no significant difference. The cross (†) indicates a comparison of the transactivation of the wild-type sequence to the sequences mutated by HIF1 (†) or by HIF2 (†). † p < 0.05. All the error bars represent the means±sds

To further determine whether the putative HRE motifs can be directly induced by HIFs, we designed luciferase reporter constructs of the mouse IL22R1 promoter, which spans from 2000 bp nucleotides upstream of the transcriptional start site to the first exon and contains 4 HRE consensus sequences that are closely associated with HIF ancillary sequences (HASs) and E-box motifs, which are necessary cis-elements for HRE activation. We transfected this construct into human embryonic kidney-derived adherent 293 cells (HEK293) along with an expression vector encoding GFP or constitutively active human HIF1α or HIF2α and performed dual luciferase reporter assays. We found that compared with the GFP control, both HIF1α and HIF2α significantly increased IL22R1 promoter transactivation by 3 ~ 5-fold. To understand whether any specific HRE motif is required for its activation by HIF1α or HIF2α, we performed mutational analysis. The results revealed that the two most distal HREs in the IL22R1 promoter had the greatest impact on HIF2α-induced transactivation, whereas all four HREs were necessary for HIF1α-induced transactivation, suggesting that the expression of IL22 can be differentially regulated by HIF1α and HIF2α in various scenarios (Fig. 7D and E).

HIF-2α specifically mediates the radioprotective effect of 5-HMF

To understand the potential links between 5-HMF and intestinal HIF signaling, we examined the protein levels of HIFs, including HIF-1α and −2α, in the intestinal crypts of mice treated with low- and high-dose 5-HMF. Compared with those of the controls, the intestinal epithelial crypts of the 5-HMF-treated mice presented a specific increase in HIF-2α protein expression, whereas no increase in HIF-1α was observed, indicating that 5-HMF selectively stabilizes the HIF-2α protein in the crypts (Fig. 8A). Under normoxia, HIF-α subunits undergo rapid ubiquitin‒proteasome degradation mediated by proline hydroxylases (PHDs). To investigate whether 5-HMF can exert competitive inhibitory effects on PHD, we performed molecular docking simulations between 5-HMF and the catalytic domain of human PHD2. Computational modeling positioned 5-HMF within the enzyme’s hydrophobic active-site pocket, forming potential interactions with residues Arg383, His313, His374, Asp315 and Tyr329. The lowest-energy binding conformation yielded a ΔG of −4.8 kcal/mol, suggesting possible complex formation and competitive inhibition (Fig. 8B).

Fig. 8.

5-HMF selectively stabilizes HIF-2α, and its radioprotective effect is abolished by HIF-2α inhibition. (A) Western blot analysis of HIF-1α and HIF-2α protein levels in the intestinal crypts of mice treated with vehicle (n = 15), low-dose (50 mg/kg, n = 15), or high-dose (100 mg/kg, n = 15) 5-HMF. (B) molecular docking simulation of 5-HMF binding to the catalytic domain of human PHD2. (C) schematic time point of HIF-2α inhibitor (PT-2385) administration (red arrow). (D) changes in the survival curves of mice treated with low-dose 5-HMF with or without the HIF-2α inhibitor PT-2385 (20 mg/kg, i.P., daily) after 10 Gy WBI. The adjacent table details the results of pairwise comparisons, listing the exact p-values between groups. (E) body weight changes in mice after 10 Gy TAI with or without PT-2385 treatment. Significant differences indicated in the figure primarily result from pairwise comparisons between the b group (low-dose 5-HMF treatment) and each of the other three experimental groups at corresponding time points. (F) Representative H&E staining of the duodenum/jejunum and ileum in this experiment after 10 Gy TAI for 2 days; the yellow arrowhead indicates the regenerated cryptsthe scale bar represents 50 μm. (G) quantification of crypt numbers in whole intestinal tissues. n = 10 fields under the microscope from three mice/group. All the data are presented as the means±sds. Statistical significance compared with the regular diet group. *p < 0.01, **p < 0.001, and NS, no significant difference

Additionally, low-dose 5-HMF-treated mice received daily intraperitoneal injections of the HIF-2α inhibitor PT-2385 at 24 hours after 10 Gy WBI, and the results revealed that the survival advantage conferred by low-dose 5-HMF was nullified by PT-2385 treatment (Fig. 8C and D). Similarly, following exposure to 10 Gy TAI, the inhibitor treatment resulted in a sharp decrease in mouse weight and death within 6 to 7 days (Fig. 8E). Histological analysis revealed that, regardless of whether the mice were administered 5-HMF or not, those in the inhibitor-treated group experienced more crypt loss and shorter villi from the duodenum to the ileum after 10 Gy TAI for 2 days. However, the mice treated with low-dose 5-HMF alone presented more proliferative crypts (Fig. 8F and G). Furthermore, analysis of IL22R1 expression in crypts revealed that HIF-2α inhibition abrogated the 5-HMF-induced upregulation of IL22R1 at the protein level (Fig. 9A and B). Consistently, PT-2385 treatment reversed the enhanced growth of organoids isolated from 5-HMF-treated mice when they were stimulated with recombinant murine IL-22 (rmIL-22) (Fig. 9C–D). Collectively, these findings strongly suggest that low-dose 5-HMF protects mice from irradiation-induced intestinal injury by promoting ISC-mediated epithelial regeneration, which is achieved through HIF-2α-mediated enhancement of IL22 signaling cascades (Fig. 9E).

Fig. 9.

HIF-2α is required for 5-HMF-induced IL22R1 expression and the functional potentiation of IL-22 signaling. (A, B) HIF-2α inhibition abolishes 5-HMF-induced upregulation of IL22R1 at the protein level in intestinal crypts from mice treated with low-dose 5-HMF with or without the HIF-2α inhibitor PT-2385. (C) MTT-stained organoids were isolated from the mice in this study and cultured in medium in the presence or absence of rmIL22 and/or PT-2385 (n = 3, 3 replicates/group). (D) quantification of the organoid surface area (μm²×10³) and MTT viability analysis (% positive, n = 6 wells/group). Statistical significance compared with the control group. *p < 0.05, **p < 0.01, *p < 0.001 and NS, no significant difference. (E) schematic model summarizing the proposed mechanism by which low-dose 5-HMF mitigates radiation-induced intestinal injury. 5-HMF inhibits prolyl hydroxylase domain (PHD) enzyme activity, leading to the selective stabilization of HIF-2α in intestinal crypt cells. HIF-2α translocates to the nucleus and binds to hypoxia‒response elements (HREs) in the promoter of IL22R1, where it is upregulated. Increased IL22R1 surface expression enhances responsiveness to IL-22, leading to activation of the STAT3 signaling pathway. This promotes the proliferation and survival of intestinal stem cells (ISCs), facilitating epithelial regeneration and ultimately protecting against radiation-induced intestinal damage

Discussion

Radiation-induced gastrointestinal toxicity remains a frequent dose-limiting complication in patients receiving abdominal or pelvic radiotherapy. Conventional radioprotective agents face challenges, including inherent toxicity, pharmacokinetic instability, and suboptimal administration routes. In particular, the only clinically approved radioprotectant, amifostine, is limited by its significant side effects, such as hypotension and nausea, and requires intravenous administration [32]. Alternatively, while recombinant IL-22 therapy has shown promise in preclinical models for promoting intestinal regeneration, its clinical translation faces hurdles related to protein stability, cost, and the risk of provoking systemic inflammation [33–35]. In contrast, herbal-derived compounds with dual food-medicine applications offer advantages through their low toxicity profiles and multitarget mechanisms of action. 5-Hydroxymethylfurfural (5-HMF), a bioactive constituent of traditional Chinese medicines (TCMs), such as Rehmannia glutinosa Praeparata, Polygonum multiflorum, and black garlic extracts, exemplifies this class [3–5]). Its efficacy in other pathological conditions, such as hypoxia-induced organ injury [13] and arthritis [6–10], further underscores its broad therapeutic potential. Our work demonstrated that moderate-dose 5-HMF mitigated radiation-induced intestinal injury by enhancing intestinal stem cell proliferation through a HIF-2α-driven IL-22/STAT3 signaling axis.

5- HMF is a six-carbon heterocyclic compound. The furan ring has the ability to attract electrons, which results in complex dose-dependent biological effects [36, 37]. Its beneficial effects have been well documented at specific thresholds: 100 mg/kg 5-HMF can protect the brains and kidneys of mice from hypoxia-induced injury [28]; 40 mg/kg 5-HMF can reduce the production of superoxide in skeletal muscle during altitude challenges [38]; and less than 30 mg/kg can alleviate arthritis-induced cartilage damage [13]. Conversely, previous toxicity trials on HMF showed that the oral LD50 values were 1910 mg/kg and 3100 mg/kg for mouse and rat models, respectively [39]. When rats are given a daily oral dose of 310 mg/kg HMF for 60 days, it can lead to impaired liver functions, including changes in serum protein and the albumin/globulin ratio and an increase in hepatic tributyrinase [40]. When B6C3F1 female mice were given 188 or 375 mg/kg HMF for approximately two years, the incidence of hepatocellular adenomas increased, indicating that long-term exposure to high levels of HMF has a carcinogenic effect [41]. Moreover, after HMF is consumed, it is transformed into SMF in the liver and then released into the bloodstream. The administration of 250 mg/kg SMF can cause damage to tubules and moderate toxicity to the liver [42]. For humans, the acceptable daily intake of 5-HMF ranges from 2–30 mg per person per day [39, 43]. Consistently, before the formal experiment, by analyzing the biochemical indicators of liver and kidney toxicity, we determined that the safe dosage threshold was no more than 200 mg/kg. In this study, we found that the administration of low-dose 5-HMF (50 mg/kg/d) effectively improved the survival rate of mice after WBI or TAI and reduced their weight loss and intestinal permeability, whereas high-dose 5-HMF (200 mg/kg/d) did not significantly improve survival. This may be because higher doses are likely to saturate the beneficial pathways and simultaneously trigger three detrimental effects: (1) glutathione depletion due to aldehyde reactivity, disrupting the precise redox balance required for prosurvival signaling and potentially inducing metabolic stress; (2) SMF generation exceeding the hepatic detoxification capacity; and (3) paradoxical pro-oxidant activity, leading to systemic off-target toxicity. Consequently, systemic toxicity, particularly renal OAT-mediated accumulation of SMF, may mask intestinal protection, which explains the narrow therapeutic window observed. The low dose of 50 mg/kg/day used in our study, which demonstrated efficacy, is substantially lower than the reported LD₅₀ values in rodents ( > 1900 mg/kg) and subchronic toxicity thresholds, suggesting a favorable safety margin for short-term radioprotective use. While the genotoxic potential of 5-HMF at very high doses via its metabolite SMF is a consideration, its ubiquitous presence in the human diet and the low effective dose relative to toxicological benchmarks support its clinical feasibility. Future pharmacokinetic studies will be crucial to define its human exposure profile and to firmly establish the therapeutic index for its application as a radioprotective agent.

Following exposure to radiation, varying degrees of villi blunting and fusion can occur, which contributes to villous epithelial attenuation and substantial loss of crypts. These changes disrupt the homeostasis and integrity of epithelial cells. Our findings underscore that the mice administered low-dose 5-HMF maintained an increased number of proliferative crypts and preserved the crypt–villus structure even after being subjected to 10 Gy WBI. Previous research has highlighted the importance of ISC repopulation and the regeneration of differentiated cells for the repair of intestinal epithelial damage caused by radiation [44]. Lgr5⁺ ISCs and Paneth cells located at the base of intestinal crypts play indispensable roles in maintaining intestinal homeostasis [45]. Notably, our results demonstrated that mice in the low-dose 5-HMF group presented an increased number of Lgr5⁺ ISCs, as identified by Olfm4 staining, and proliferative transient amplifying (TA) cells, marked by Ki67, upon exposure to nonlethal 8 Gy TAI. In contrast, the number of Paneth cells labeled with lysozyme (lyz) did not significantly change. These findings are consistent with our findings of an increase in the number of BrdU⁺/Lgr5⁺ double-positive ISCs and a decrease in the number of TUNEL⁺/Lgr5⁺ double-positive ISCs in the low-dose 5-HMF group after exposure to 10 Gy WBI. These findings suggest that the protective effect of low-dose 5-HMF against radiation-induced injury could be attributed to its capacity to increase the proliferation of the Lgr5⁺ ISC population, including progenitor cells involved in the regenerative process.

The utilization of intestinal organoids in a 3D culture setup provides a convenient platform for assessing stem cell proliferation. Regardless of whether radiation was involved, the growth of the intestinal organoids derived from the mice treated with low-dose 5-HMF did not significantly differ from that of the organoids derived from the control group. This inconsistency between the in vitro and in vivo results raises several questions. However, RNA-seq and gene expression analysis revealed the involvement of the IL22 signaling pathway in the proliferation of ISCs induced by low-dose 5-HMF in vivo. IL22, produced by tissue-resident ILC3s, is a critical factor in maintaining the integrity and barrier function of the intestinal epithelium across multiple experimental models of intestinal injury [24, 46]. Our in vitro findings demonstrated that compared with those from mice fed a regular chow diet or treated with CHIR99021 (CHIR), an agonist of the Wnt/β-catenin signaling pathway, crypts from mice fed low-dose 5-HMF, when cultured with recombinant murine IL22 (rmIL22), generated notably larger organoids. This observation reinforces the notion that 5-HMF promotes intestinal stem cell proliferation through the enhancement of the IL22 signaling cascade. Furthermore, the elevation in the phosphorylation level of Stat3 (Y705) serves as additional validation for the notion that low-dose 5-HMF promotes ISC proliferation by augmenting the IL22 signaling pathway. This mechanism effectively reconciles the apparent discordance between the in vivo and in vitro outcomes.

It is noteworthy that the activation of STAT3, which is central to the reparative process observed in our study, exemplifies a well-documented “double-edged sword” in pathophysiology. In many cancer types, persistent STAT3 signaling is a known driver of tumorigenesis, promoting cell proliferation, survival, and inflammation that fuels the tumor microenvironment [47, 48]. The stark contrast between its pro-tumorigenic role and the protective regeneration seen here underscores the profound context-dependency of STAT3 function. The critical determinants likely include the specific activating cytokine, for instance, IL-22 compared to others like IL-6; the cellular compartment involved, such as the intestinal epithelium versus immune or cancer cells; and the duration of activation, ranging from the transient signal required for repair to the constitutive activation observed in cancer. In our model, 5-HMF appears to engage a highly specific reparative axis that centers on the IL-22/IL-22R1 pathway. This axis preferentially activates STAT3 in the intestinal epithelium, thereby driving regeneration without instigating the broader pro-oncogenic programs associated with its chronic activation in other contexts.

IL22 is categorized within the IL-10 family of cytokines and exerts its effects through the IL-22 receptor (IL-22R1), which forms a complex with IL10R2 [49]. Our results revealed that both low- and high-dose 5-HMF led to an increase in the expression of IL22R1, whereas no significant changes in the expression of IL10R2 at either the mRNA or protein level were detected within the crypts of the mice. The presence of four repeated “ACGTG” sequences was observed within a genomic region ranging from −1526 bp to +80 bp around the transcriptional initiation site of IL22R1. These sequences have been demonstrated to be cis-elements that bind hypoxia inducible factors (HIFs) in our research and other studies [38]. Furthermore, 5-HMF has been reported to stabilize HIF-1α through the sequestration of vitamin C, which in turn inhibits PHD activity [27]. Similarly, our findings demonstrated that 5-HMF treatment induced HIF-2α expression specifically in intestinal crypts without affecting HIF-1α. The molecular docking results revealed that 5-HMF can enter the enzymatic active site of PHD and exhibit a certain binding capacity, suggesting possible competitive inhibition. These results indicate that 5-HMF exhibits isoform selectivity in regulating HIF signaling across different tissues/organs and various models. For example, DFO, a hypoxia-mimicking compound, can upregulate HIF-1α expression and enhance ISC proliferation and crypt class formation [50, 51]. Elevated HIF-2α levels caused by dimethyloxallyl glycine (DMOG), a PHD inhibitor, effectively attenuate radiation-induced intestinal injury and improve the survival rate of mice [52]. The utility of a HIF-2α-specific inhibitor, PT-2385, allowed us to delve deeper into this phenomenon. The inhibition of HIF-2α effectively abolished the radioprotective effects of low-dose 5-HMF in vivo and abrogated the 5-HMF-induced robust growth of organoids and upregulated the expression of IL22R1 in crypts.

Looking forward, our findings suggest promising avenues for combination therapy. The mechanism of 5-HMF, which centers on priming the intestine by upregulating the IL-22 receptor, positions it as an ideal candidate for synergy with recombinant IL-22 therapy. Such a combination could potentially lower the required dose of IL-22, mitigating its side effects while achieving enhanced and prolonged radioprotective efficacy. Furthermore, given the close interplay between the gut microbiota, mucosal immunity, and IL-22 production, exploring 5-HMF in conjunction with microbiota-based interventions represents another compelling strategy to foster a sustained regenerative microenvironment in the irradiated gut.

Conclusion

Taken together, our findings reveal the significant impact of pretreatment with a moderate dose of 5-HMF on alleviating radiation-induced intestinal damage in a mouse model. This study not only sheds light on the potential relationship between HIF-2α-driven IL-22/STAT3 signaling and the maintenance of epithelial homeostasis at both the cellular and molecular levels but also presents a dietary-derived therapeutic avenue for mitigating radiation-related injuries. Notably, despite the promising implications of this research, its translation to clinical practice requires further investigation. However, it is important to acknowledge that while this research offers a promising strategy, its immediate translation to human treatment requires additional investigation. The specific thresholds and duration of 5-HMF administration necessary to trigger the adaptive response outlined in this study necessitate further refinement through rigorous clinical trials. These trials will aid in establishing appropriate guidelines for safe and effective implementation in human contexts. Furthermore, the effects of potential HIF-2α stabilization on residual tumor growth subsequent to RT warrant careful exploration to comprehensively understand its potential benefits and limitations. In conclusion, this study advances our understanding of how targeted modulation of the hypoxia‒immune repair axis might serve as a protective measure against radiation-induced damage and offers a promising avenue, potentially alone or in combination with other modalities, for enhancing radiation resilience and potentially improving outcomes for individuals receiving radiation therapy.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

T. Zhang, J. He, and J. He conducted the majority of the experiments, including animal studies, organoid culture, and histological analyses. C. Xu, Y. Huang, and S. Huang provided essential experimental support in animal handling, in vitro assays, and molecular biology work. H. Jiang and L. Liu were responsible for data analysis and visualization. Y. Fan, H. Li, and X. Dong contributed critical resources and funding. The manuscript was written primarily by T. Zhang with input from all the authors, and the project was supervised by H. Li and X. Dong.

Funding

This research was funded by grants from the General Program of the Chongqing Natural Science Foundation (CSTB2023NSCQ-MSX0144), the Science and Technology Research Program of Chongqing Municipal Education Commission (KJQN202204015 and KJQN202402835), the Special Foundation of Chongqing Postdoctoral Project (2022CQBSHTB3064), and the Chongqing University of Science and Technology postgraduate innovation project (YKJCX2421611).

Data availability

The data used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The study protocol was conducted in accordance with the Declaration of Helsinki and approved by the Laboratory Animal Center of Chongqing University (Approval ID: CQU-IACUC-RE-202501-009) in compliance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Institutional review board statement

The animal study protocol was approved by the animal care committee at the Chongqing University Academy of Animal Sciences. (approval no. CQU-IACUC-RE-202501-009).

Consent to participate declaration

Not applicable.

Conflicts of interest

The authors declare that they have no competing interests.