Oncomicrobial vaccines mitigate tumor progression via precisely targeting oncomicrobes in mice

Graphical abstract

Dear Editor,

Colorectal cancer (CRC) is a leading global cancer, causing significant mortality and morbidity, particularly among younger individuals (Spaander et al., 2023). The gut microbiota, including certain bacterial species, has been identified as oncomicrobes and linked to CRC development (Holt, 2023). Oncomicrobes are microorganisms capable of causing cancer and can potentially influence the development and progression of tumors. Research demonstrates oncomicrobes, such as Fusobacterium nucleatum (Rubinstein et al., 2013), Campylobacter jejuni (He et al., 2019), and enterotoxigenic Bacteroides fragilis (ETBF) (Wu et al., 2009), contribute to CRC by causing genetic damage and modulating the immune system. Interventions to modify the gut microbiota, such as broad-spectra antibiotics and bacteriophage therapy, have shown potential in reducing oncomicrobes, which consequentially suppress CRC development. However, the emerging threat of antibiotic resistance and the need for targeted therapies highlight the need for novel approaches. For instance, therapeutic strategies could inhibit colibactin production by polyketide synthase (pks⁺) Escherichia coli using tungstate to reduce genotoxin-producing bacteria (Cougnoux et al., 2016). Fecal microbiota transplantation has shown promise in restoring microbial balance, but safety concerns and the narrow host range of phages limit their clinical use (Cullin et al., 2021). Vaccines targeting cancer-associated microbes hold promise for preventing and treating cancers, such as Gardasil (Merck & Co, USA), which is the first FDA-approved vaccine for primary human papillomavirus (HPV) and has demonstrated nearly 100% effectiveness in preventing cervical precancerous lesions in previously unexposed individuals (Garland et al., 2016). However, there is a lack of comprehensive investigation into the efficacy of oncomicrobial vaccines in oncomicrobes-associated CRC.

In this study, we used inactivated whole-cell vaccines targeting oncomicrobes as proof-of-concept and demonstrated that oncomicrobial vaccines could mitigate tumor progression. The vaccines specifically reduced the colonization of these oncomicrobes and suppressed tumor development, suggesting a potential for clinical application without disrupting the gut microbiota. Further research is necessary to translate these findings into effective CRC prevention and treatment strategies.

To validate the concept of using vaccines to prevent or treat oncomicrobes associated with CRC, we set up two animal models: preventive and therapeutic. In the preventive model, we used the oncomicrobe C. jejuni, which is a common exogenous gastrointestinal pathogen that causes bacterial gastroenteritis and various other gastrointestinal diseases and is associated with CRC and promotes tumorigenesis (He et al., 2019), and it is feasible to immunize host before C. jejuni infection. We prepared a formalin-fixed whole-cell C. jejuni vaccine. We tested this vaccine in the Apc^min/+ mice model, which carries an adenomatosis polyposis coli (Apc) mutation frequently mutated in human colon cancer and increased tumor formation and inflammation in the gut (Ren et al., 2019). Mice received two vaccinations with 10⁸ CFU of formalin-fixed C. jejuni and challenged with 10⁸ CFU of C. jejuni, alongside 2.5% DSS to accelerate colonic tumorigenesis (Fig. 1A). We assessed antibody levels against C. jejuni in serum and feces by Enzyme-Linked Immunosorbent Assay (ELISA). The vaccinated mice exhibited high levels of anti-C. jejuni antibodies in serum 2 weeks post-booster vaccination (Fig. 1B and 1C), with anti-C. jejuni total immunoglobulin (Ig) and IgG reached the optimal level a week after the booster vaccination (Fig. S1A and S1B). To investigate how much anti-C. jejuni IgG was secreted into the intestinal lumen, we measured anti-C. jejuni IgG levels in mouse feces. As shown in Fig. 1D, the vaccinated mice exhibited significantly higher levels of secreted IgG antibodies in feces than the mock group. Notably, anti-C. jejuni IgG was not detected in feces until 2 weeks after the booster vaccination (Fig. S1C). To assess the vaccine’s impact on C. jejuni colonization in the gut, we quantified C. jejuni in fecal samples by qPCR using specific primers that amplify the hipO gene. Campylobacter jejuni was detected in fecal samples of both C. jejuni and vaccine-C. jejuni mice at similar levels during the first week after the challenge but decreased in the vaccinated group beyond the first week, with significant differences observed at later time points (Fig. 1E). FISH analysis confirmed these findings on colon samples when mice were sacrificed (Fig. 1F). Thus, the vaccine elicited systemic humoral immunity responses and, although it did not prevent the initial colonization, it did accelerate the clearance of C. jejuni starting from 1 week after the challenge.

The above results indicate that the vaccine strategy effectively stimulates the mice to produce specific antibodies and eliminates cancer-promoting bacteria. Next, we investigated whether a reduction in the abundance of cancer-promoting bacteria would prevent and treat oncomicrobe-promoting tumorigenesis. As shown in Fig. 1G and 1H, C. jejuni-gavaged mice had significantly higher tumor incidence compared to the mock group, indicating the successful reproduction of this previously reported model (He et al., 2019). More importantly, the vaccine-C. jejuni mice carried significantly fewer tumors than the C. jejuni-gavaged ones and had no difference from the mock group (Fig. 1H). Furthermore, the vaccine-C. jejuni mice exhibited a lower inflammation level in the gut compared to the C. jejuni-gavaged group, as evidenced by the histopathology staining (Fig. 1I–K), as well as the levels of T lymphocyte infiltration, apoptosis, and cell proliferating (Fig. S2A and S2B). To monitor the impact of the vaccine on the safety of the mice, we measured the body weight of the mice during the experiment. The results showed that the vaccine had no significant effect on the body weight of the mice (Fig. S3). These results demonstrated that the vaccine could prevent C. jejuni-enhanced tumorigenesis by accelerating the clearance of C. jejuni without obvious safety concerns.

Campylobacter jejuni infection, associated with gut barrier disruption DNA damage, contributes to CRC development (He et al., 2019). The bacterium’s invasion triggers inflammation and promotes tumor growth by affecting gene expression. RNA-seq analysis showed that genes affected by C. jejuni infection in colon tumors and para-cancerous tissues of mice were also partially restored after vaccine treatment (Figs. S4 and S5; Tables S2–5).

Encouraged by the efficacy of the oncomicrobial vaccine in the preventive model, we further investigated this concept in a therapeutic model, which utilized ETBF as a testing oncomicrobe, a gut bacterium found in 100% of late-stage CRC patients, which promotes inflammation and tumorigenesis (Boleij et al., 2015). Since ETBF is prevalent in the human gut, especially among CRC patients, it is unlikely to be vaccinated before its primary infection. Therefore, assessment of vaccine effects after ETBF infection via a therapeutic model is more feasible. Apc^min/+ mice were treated with an antibiotic cocktail and gavaged with ETBF (ATCC 43858) followed by 2.5% DSS administration to induce colonic tumorigenesis. Formalin-fixed ETBF vaccination was administered on day 21, with a booster vaccination on day 35, and mice were sacrificed on day 70 (Fig. 1L). Antibody levels in serum were measured using ELISA. ETBF gavage increased anti-ETBF total Ig and IgG compared to the control (Fig. S6A and S6B). Primary vaccination with ETBF led to significantly higher anti-ETBF total Ig, IgG, and IgM levels in the ETBF-vaccine group compared to the ETBF-gavaged group (Fig. S6C–E), and anti-ETBF total Ig and IgG remained at relatively high levels at the end point of the experiment (Fig. S6F–J). Subcutaneous vaccination boosted the immune response, with significantly higher anti-ETBF total Ig and IgG levels after the booster (Fig. 1M and 1N). Anti-ETBF IgA levels were higher 2 weeks after ETBF gavage, and the ETBF-vaccine group showed higher total Ig, IgG, and IgA levels after primary immunization (Fig. S7A–F). Booster vaccination further increased antibody levels in mouse feces levels (Figs. 1O, 1P, S8A, and S8B), with IgA and IgG titers being >2-fold and ~8-fold higher after the booster, respectively (Fig. S8C–E). These results indicate a robust immune response and anti-ETBF antibodies in the serum and gut lumen.

To evaluate the vaccine’s effect on ETBF colonization in mice, we measured ETBF abundance in feces collected at time points after primary vaccination (from day 26 to 40) by qPCR using primers that amplify the toxin gene. We observed no significant difference in ETBF abundance between the ETBF-vaccine group and the ETBF group at 5 and 14 days after the primary vaccination; however, 3 and 5 days after the booster, the ETBF-vaccine group had significantly less ETBF abundance (Fig. 1Q–T). This was further confirmed by FISH analysis at the end of the experiment (Fig. S8F). The vaccine effectively reduced the abundance of ETBF in the DSS model, indicating the therapeutic potential of the vaccine strategy against gut pathobionts that had colonized the host. Bacteroides is a major genus in the gut microbiota of mice, and it is important to assess the impact of the ETBF vaccine on the levels of Bacteroidetes at the phylum level and Bacteroides levels at the genus level. The ETBF vaccine shows minimal impact on Bacteroidetes and Bacteroides abundance in mouse gut microbiota (Fig. S9A and S9B). Nontoxic B. fragilis (NTBF), as a commensal bacterium, has the extraordinary ability to regulate steady-state mucosal immunity and promote the development of systemic immunity (Sears et al., 2014). The study has shown that Apc^min/+ mice treated with NTBF, a non-oncogenic bacterium, do not affect tumor formation (Wu et al., 2009). The ETBF vaccine increased the abundance of NTBF in mice that received both ETBF infection and vaccination compared to those infected with ETBF only (Fig. S9C). These findings suggest that restraining ETBF colonization can restore NTBF abundance, aligning with previous observations of ETBF reducing NTBF abundance (Wagner et al., 2016).

Regarding the therapeutic effect of ETBF vaccines in mice, the ETBF-vaccine group exhibited a significant reduction in tumor numbers compared to the ETBF group (Fig. 1U and 1V). Additionally, since intestinal inflammation induced by ETBF often leads to a shortened intestinal lumen, we measured the length of the mouse colon to assess the extent of inflammation caused by ETBF. Mice infected with ETBF and subsequently vaccinated showed significantly greater colon length than unvaccinated infected mice (Fig. 1W), indicating that the ETBF vaccine effectively mitigated intestinal inflammation. Furthermore, the ETBF-vaccine mice had a less inflammatory gut than the ETBF group, as evidenced by the histopathology staining and scores (Fig. 1X and 1Y). In addition, reduced nuclear staining and milder inflammatory cell infiltration in the colon tissue of the ETBF-vaccine mice were observed compared to the ETBF mice (Fig. S10A and S10B). Additionally, throughout the experiment, especially the vaccine treatment showed no significant effect on the body weight of the mice, which preliminarily demonstrates the safety of the vaccine in mice (Fig. S11). Taken together, these results demonstrated that the vaccine effectively reduces ETBF-promoted tumorigenesis and inflammation in mice.

Vaccines trigger the immune system to recognize and respond to antigens, generating antibodies that specifically target pathogens to prevent infection. Antibodies can neutralize bacteria or their toxins and help clear bacterial infections and they also activate T cells to identify and destroy infected cells. Nonetheless, it is a well-established fact that conventional bacterial vaccines generally fail to activate certain immune cells, including T and B cells. However, the pathogenic bacterium ETBF fosters an immunosuppressive milieu through diverse mechanisms. We postulated that this suppression could be overturned by a vaccine designed to clear ETBF. To assess the ETBF vaccine’s effects on lymphocyte subsets in mouse spleen and tumor tissues, we conducted flow cytometry (Fig. S12A–C). The vaccine and ETBF infection increased CD45⁺ and CD8⁺ T cells but decreased CD4⁺ T cells compared to the mock group in the spleen of mice (Fig. S12B). However, administration of the vaccine after ETBF infection did not significantly affect the immune cell population in the spleen (Fig. S12B). In tumor tissues, following ETBF infection, vaccination significantly increased the number of CD8⁺ T cells compared to the vaccine group (Fig. S12C), indicating a potential break from immunosuppression. Regarding B. fragilis enterotoxin, it disrupts gut barrier function and increases inflammation by cleaving E-cadherin and promoting pro-inflammatory cytokine IL-7A production and β-catenin signaling (Wu et al., 2009). ELISA tests of the cytokines in the mouse serum showed that ETBF infection did increase levels of IL-17A in the pro-inflammatory cytokine compared with the uninfected group, the vaccine treatment after ETBF infection reduced IL-17A level to some extent (Fig. S12D), indicating the vaccine may mitigate the pro-inflammatory response caused by ETBF.

The vaccine, designed to target a specific microbe, prompted our interest in its effect on gut microbiota composition. In the preventive model, 16S rRNA gene sequencing on fecal samples revealed no significant alterations in the composition of the gut microbiota between the vaccinated group and the mock group throughout the experiments (days 14–83), as indicated by alpha diversity (Fig. 2A) and beta diversity (Figs. 2B, 2C, and S13). We further analyzed the bacterial abundance of Campylobacter in the 16S rRNA data. The results showed that administering the vaccine effectively decreased the colonization of Campylobacter, although there was no statistical significance in the data (Fig. 2D–E). These results demonstrated that the vaccination prevented the disturbance caused by C. jejuni infection on the diversity of gut commensals.

Metabolites derived from gut microbiota play a pivotal role in connecting the microbiome to cancer progression by altering the tumor microenvironment and regulating key signaling pathways (Yang et al., 2023). To explore the effects of a C. jejuni vaccination on tumorigenesis, we conducted a metabolomics analysis of fecal samples at the end of the study. The analysis revealed that the vaccine did not separate vaccine group and non-vaccine group samples in the PLS-DA model based on multivariate statistical analysis (Fig. 2F). However, there were differences in specific metabolites between vaccinated and non-vaccinated mice (VIP > 1, P < 0.05), with the vaccine-C. jejuni group showed increased levels of antitumor metabolites and decreased levels of tumor-promoting metabolites (Fig. S14A and S14B). The vaccination effectively mitigates the impact of C. jejuni on the intestinal metabolome, highlighting its potential as an effective intervention.

To probe the vaccine’s effect on the intestinal microbiota’s composition and structure within the therapeutic model, metagenomic sequencing was conducted on mouse fecal samples. The results revealed no significant difference in alpha diversity (Shannon Index) between the mock and vaccine groups (Fig. 2G), and beta diversity analysis also found no significant difference (Figs. 2H and S15A; P = 0.084), indicating that the vaccine did not disturb the normal gut microbiota (Fig. S15). Taxonomy composition comparison revealed that ETBF infection increased the number of B. fragilis species compared with the other groups (Fig. S16A), but the vaccine significantly reduced B. fragilis abundance in the ETBF-vaccine group compared to the ETBF group (Figs. 2I and S16A), aligning with previous qPCR results. In addition, the highest abundance of other species of Bacteroides, such as Bacteroides caccae, Bacteroides ovatus, and Bacteroides salyersiae, remained unchanged (Fig. S16B–D), suggesting that the ETBF vaccine did not have a significant impact on the normal intestinal flora of Apc^min/+ mice.

Bacteroidetes play a crucial role in maintaining intestinal health, including nutrient metabolism, microbial synthesis, and immune regulation (Wexler, 2007). Therefore, we urgently need to know whether the ETBF vaccine affects the interactions B. fragilis between and different microorganisms. To further investigate the impact of ETBF infection and the vaccine on the intestinal microbiota of Apc^min/+ mice, we conducted a correlation network analysis of the microbial community at the species level. The ETBF vaccine preserved the coexistence dynamics between intestinal bacterial species (Fig. S17A and S17B) and some of the functional pathways associated with B. fragilis were attenuated in vaccine-treated mice following ETBF infection compared with mice infected with ETBF alone (Fig. S17C and S17D). Interestingly, in mice infected with ETBF, some of the B. fragilis-associated functional pathways showed a significant negative correlation with species of Lachnospiraceae. However, in mice treated with the vaccine after ETBF infection, these functional pathways showed a positive correlation with species of Lachnospiraceae (Fig. S17C and S17D). The above results suggest that the ETBF vaccine may influence the B. fragilis-related functional pathways, inhibiting ETBF growth and suppressing tumor development. In contrast, vaccine treatment mitigates the effects of ETBF infection on the rest of the microbiota.

Correlation analysis of the species-level gut microbiota revealed the complex relationship between ETBF infection, the ETBF vaccine, and the gut microbial species in Apc^min/+ mice (Fig. 2J and 2K). In ETBF mice, B. fragilis positively correlated with Rikenellaceae bacterium and negatively correlated with Lachnospiraceae bacterium and Clostridiaceae bacterium (Fig. 2J), while in ETBF-vaccinated mice, B. fragilis only negatively correlated with Muribaculaceae and positively correlated with L. bacterium (Fig. 2K). Lachnospiraceae, typically beneficial in healthy gut microbiota, is involved in carbohydrate metabolism (Takeuchi et al., 2023). These findings indicate that the ETBF vaccine reversed the dysbiosis caused by ETBF infection and increased the abundance of beneficial bacteria, such as L. bacterium.

In conclusion, we assessed the effectiveness of bacterial vaccines in the prevention and treatment of CRC induced by oncomicrobes, specifically C. jejuni and ETBF. Our findings demonstrate that the vaccine treatment successfully decreased the abundance of specific oncomicrobes, spares gut commensal, and effectively controlled tumor growth in both mouse models. Furthermore, if a vaccine proves to be safe and efficient in ameliorating the occurrence and progression of CRC, it would be worthwhile to explore additional applications. For instance, individuals who are under surveillance for colorectal cancer due to their family history could potentially benefit from receiving cancer-promoting bacterial vaccination.

References

[1] Boleij A, Hechenbleikner EM, Goodwin AC et al The Bacteroides fragilis toxin gene is prevalent in the colon mucosa of colorectal cancer patients. Clin Infect Dis 2015;60:208–215.

[2] Cougnoux A, Delmas J, Gibold L et al Small-molecule inhibitors prevent the genotoxic and protumoural effects induced by colibactin-producing bacteria. Gut 2016;65:278–285.

[3] Cullin N, Azevedo Antunes C, Straussman R et al Microbiome and cancer. Cancer Cell 2021;39:1317–1341.

[4] Garland SM, Kjaer SK, Muñoz N et al Impact and effectiveness of the quadrivalent human papillomavirus vaccine: a systematic review of 10 years of real-world experience. Clin Infect Dis 2016;63:519–527.

[5] He Z, Gharaibeh RZ, Newsome RC et al Campylobacter jejuni promotes colorectal tumorigenesis through the action of cytolethal distending toxin. Gut 2019;68:289–300.

[6] Holt RA. Oncomicrobial vaccines: the potential for a Fusobacterium nucleatum vaccine to improve colorectal cancer outcomes. Cell Host Microbe 2023;31:141–145.

[7] Ren J, Sui H, Fang F et al The application of Apc^(Min/+) mouse model in colorectal tumor researches. J Cancer Res Clin Oncol 2019;145:1111–1122.

[8] Rubinstein MR, Wang X, Liu W et al Fusobacterium nucleatum promotes colorectal carcinogenesis by modulating E-cadherin/β-catenin signaling via its FadA adhesin. Cell Host Microbe 2013;14:195–206.

[9] Sears CL, Geis AL, Housseau F. Bacteroides fragilis subverts mucosal biology: from symbiont to colon carcinogenesis. J Clin Invest 2014;124:4166–4172.

[10] Spaander MCW, Zauber AG, Syngal S et al Young-onset colorectal cancer. Nat Rev Dis Primers 2023;9:21.

[11] Takeuchi T, Kameyama K, Miyauchi E et al Fatty acid overproduction by gut commensal microbiota exacerbates obesity. Cell Metab 2023;35:361–375.e9.

[12] Wagner VE, Dey N, Guruge J et al Effects of a gut pathobiont in a gnotobiotic mouse model of childhood undernutrition. Sci Transl Med 2016;8:366ra164.

[13] Wexler HM. Bacteroides: the good, the bad, and the nitty-gritty. Clin Microbiol Rev 2007;20:593–621.

[14] Wu S, Rhee KJ, Albesiano E et al A human colonic commensal promotes colon tumorigenesis via activation of T helper type 17 T cell responses. Nat Med 2009;15:1016–1022.

[15] Yang Q, Wang B, Zheng Q et al A review of gut microbiota-derived metabolites in tumor progression and cancer therapy. Adv Sci (Weinh) 2023;10:e2207366.

RIGHTS & PERMISSIONS

The Author(s) 2025. Published by Oxford University Press on behalf of Higher Education Press.