Bifidobacterium Strain-Specific Enhances the Efficacy of Cancer Therapeutics in Tumor-Bearing Mice

Simple Summary The efficacy of cancer therapeutics depends on several factors, including the tumor genome, epigenome, and transcriptome. In addition, the tumor microenvironment, which consists mainly of immune cells, can influence cancer treatment outcomes. Hence, effectively leveraging host immunity is an important aspect of cancer treatment strategies. The human gut microbiome is involved in the regulation of the immune responses and affects the efficacy of chemotherapeutic and immunotherapeutic agents, including oxaliplatin, cyclophosphamide, and immune checkpoint inhibitors. This study reveals an additional dimension to the Bifidobacterium strain-specific determination of anti-cancer therapeutic efficacy using flow cytometry and transcriptome analysis of bacterial strain-fed mice and bacterial whole transcriptome analysis. We hope that our work will contribute to leveraging the gut microbiome to improve anti-cancer therapies. Abstract Colorectal cancer (CRC) is among the leading causes of cancer-related death in the world. The development of CRC is associated with smoking, diet, and microbial exposure. Previous studies have shown that dysbiosis of the gut microbiome affects cancer development, because it leads to inflammation and genotoxicity. Supplementation with specific microbiota induces anti-tumor effects by enhancing of anti-tumor immunity. Here, we observed that supplementation with either of two B. breve strains reduces tumor growth in MC38 colon carcinoma-bearing mice. Interestingly, only one B. breve strain boosted the efficacy of cancer therapeutics, including oxaliplatin and PD-1 blockade. Extensive immune profiling and transcriptomic analysis revealed that the boosting B. breve strain augments lymphocyte-mediated anti-cancer immunity. Our results suggest that supplementation with B. breve strains could potentially be used as a strategy to enhance the efficacy of CRC therapeutics.


Introduction
It is increasingly reported that the gut microbiota plays a critical role in host metabolism and immunity [1], and that the gut microbiome composition changes with age, diet, and diseases [2,3]. Colorectal cancer (CRC) is among the leading causes of cancer death in men and women in the United States of America (USA). The development of CRC is related to genetic and epigenetic alterations, inflammation, and microbial exposure [4][5][6]. Specific gut microbiota, such as Fusobacterium nucleatum, enterotoxigenic Bacteroides fragilis, and Escherichia coli are correlated with the development of intestinal adenomas and carcinomas, because they damage the DNA and affect immune and intestinal barrier functions [7][8][9]. Several studies have demonstrated that supplementation with gut microbiota species, including Enterococcus hirae, Lactobacillus johnsonii, and Bifidobacterium species reinforces

Cell Culture
MC38 cells were purchased from Kerafast (Cat# EZH204, Winston-Salem, NC, USA) and authenticated using DNA fingerprint analysis. All cells were free of mycoplasma contamination. Cells were cultured in Dulbecco's Modified Eagle Medium (DMEM, GIBCO) with 10% fetal bovine serum (FBS, GIBCO) and 100 units/mL penicillin (GIBCO) and streptomycin (GIBCO). All cells were maintained in a humidified incubator at 37 • C in a 5% CO 2 atmosphere.

Bacteria
B. breve JCM92 (B. bre JCM92) were purchased from the Japan Culture Collection of Microorganisms (JCM). B. breve Bb03 (B. bre Bb03) were purchased from DuPont (Wilmington, NC USA). To verify bacterial viability, we cultured aliquoted cells on BL (Blood Liver) agar plates at 37 • C in anaerobic conditions (O 2 < 1 ppm, 5% CO 2 , 5% N 2 ) for 48 h and counted viable colonies (viable cells). For in vivo experiments, each B. breve strain was cultured in BL medium (Kisanbio, Seoul, Korea) at 37 • C in an anaerobic condition (O 2 < 1 ppm, 5% CO 2 , 5% N 2 ) for 48 h. After culture, the bacteria were lyophilized for preservation. Next, we measured colony forming unit (CFU) counts using a PBS serial dilution method (CORNING, 21-040-CVR). For each sampling point, an aliquot of culture was diluted to 10 −6 , 10 −7 , and 10 −8 , and 100 µL was spread on BL agar plates. After counting CFUs, each B. breve strain was divided into daily administration doses and stored at 4 • C. Each B. breve strain was treated at 1 × 10 9 CFU orally once daily. For bacterial whole transcriptome analysis, each B. breve strain was harvested in the exponential phase. Then, we cultured two B. breve strains in BL agar media at 37 • C in anaerobic conditions (O 2 < 1 ppm, 5% CO 2 , 5% N 2 ) for 48 h. After colony isolation from the BL agar plate, the colony was cultured in BL broth media at 37 • C for 48 h under anaerobic conditions (O 2 < 1 ppm, 5% CO 2 , 5% N 2 ).

Mice RNA Sequencing and Data Analysis
Intestines were harvested at day 22 after MC38 tumor cell inoculation, and an RNeasy Mini Kit (Qiagen) was used to extract RNA. After RNA extraction, 151-bp paired-end libraries were constructed from 1 µg RNA using a TruSeq RNA Sample Prep Kit v2 (Illumina, San Diego, CA). Whole-RNA sequencing was performed on an Illumina HiSeq instrument. RNA-seq reads were aligned to the mouse reference genome (GRCm38) using the STAR aligner [19]. Quantification of gene expression and differential expression analysis were performed using RSEM [20] and the edgeR package [21]. The ClueGO plug-in (v2.5.4, http://www.ici.upmc.fr/cluego/, 1 January 2009) in Cytoscape software (v3.3.0, http:// cytoscape.org/, 1 January 2003) was used to analyze gene ontology (GO) [22]. Functionallyrelated GO terms for biological processes in Mus musculus (version: 27 February 2019) were grouped based on a kappa score > 0.4 and a network specificity of 5-10 using GO term fusion. Statistical significance was calculated using a two-sided hypergeometric test, and the false discovery rate was corrected using the Bonferroni step-down method.

Bacterial RNA Sequencing and Data Analysis
Bacterial RNA was extracted using the ZymoBIOMICS RNA Miniprep Kit (Zymo Research, Irvine, CA). Sequencing and library construction were performed on the Illumina Hiseq 2500 with 101 bp paired-end. Ribosomal RNA was removed using the Ribo-Zero™  [19]. Then, the mapped reads were used to calculate read counts of genes using cufflinks [23], and the gene list was inputted into Cytoscape plug-in ClueGO v2.5.4 [22] to annotate functionally grouped networks. Functionally related GO terms for biological processes in Escherichia coli (version: November 18, 2016) were grouped based on a kappa score greater than 0.4 with a network specificity of 4-10. Statistical significance was calculated using two-sided hypergeometric tests, and the false discovery rate was corrected using the Bonferroni step down method.

Statistical Analysis
Statistical analyses were performed using GraphPad Prism 8.4.3. Differences between groups were analyzed using one-way or two-way ANOVA with Tukey's post-test. Differences between groups were considered significant if p < 0.05. Statistical details are provided in the figure legends.

Mice Supplemented with B. breve Strains Exhibit Anti-Tumor Effects
To examine the anti-tumor effects of Bifidobacterium spp. on colon cancer, we administered Bifidobacterium strains alone (B. breve JCM1192, B. bre JCM92; B. breve Bb03, B. bre Bb03) with oxaliplatin to MC38 colon carcinoma-bearing mice ( Figure 1A and Table S1). Oral administration of B. breve in tumor-bearing mice significantly decreased tumor growth, compared to mice treated with the vehicle (phosphate-buffered saline; PBS) ( Figure 1B). To determine the relationship between B. breve strain supplementation and cancer therapeutics, we treated colon carcinoma-bearing mice with each B. breve strain alone and in combination with oxaliplatin ( Figure 1A). Oxaliplatin is a third-generation platinumderivative chemotherapeutic agent and is an important metastatic CRC medication [24,25]. Interestingly, when mice were treated with oxaliplatin and B. breve strains, only mice treated with B. bre JCM92 showed significantly decreased tumor growth compared to mice treated with oxaliplatin alone ( Figure 1B). To understand why only B. bre JCM92 enhanced the effect of oxaliplatin, we performed RNA sequencing of B. bre JCM92 (boosting strain) and B. bre Bb03 (non-boosting strain). We observed that 88 genes were upregulated in B. bre JCM92 compared to B. bre Bb03 (Table S2, Fold Change > 2). ClueGO analysis showed that 27 terms, including protein processing, intracellular protein transport, cellular amino acid metabolic process, ribonucleoside metabolic process, and peptide metabolic process were significantly enriched in B. bre JCM92 ( Figure 1C and Table S3).

B. bre JCM92 Boosts the Efficacy of Oxaliplatin by Enhancing Anti-Tumor Immunity
To investigate the mechanism by which Bifidobacterium strains induce anti-tumor effects, we performed immune cell profiling of spleen and tumor tissue 15 days after tumor transplantation. Mice treated with B. bre JCM92 and B. bre Bb03 had increased CD4 + T, CD8 + T, and effector CD8 + T, and NK cells. These mice also had an increased ratio of CD4 + /regulatory T cells (Tregs), CD8 + /Tregs, and effector CD8 + /Tregs. However, the mice had decreased Treg counts in splenocytes (Figure 2A,B). When we compared immune cell profiling between oxaliplatin alone and oxaliplatin + B. breve-treated groups, B. bre JCM92 enhanced the effect of oxaliplatin by increasing CD4 + T, CD8 + T, NK cells and the CD4 + /Treg, CD8 + /Treg, and effector CD8 + /Treg ratios ( Figure 2A). However, B. bre Bb03 did not boost the anti-tumor effect of oxaliplatin ( Figure 2B).

B. bre JCM92 Boosts the Efficacy of Oxaliplatin by Enhancing Anti-Tumor Immunity
To investigate the mechanism by which Bifidobacterium strains induce anti-tumor effects, we performed immune cell profiling of spleen and tumor tissue 15 days after tumor transplantation. Mice treated with B. bre JCM92 and B. bre Bb03 had increased CD4 + T, CD8 + T, and effector CD8 + T, and NK cells. These mice also had an increased ratio of CD4 + /regulatory T cells (Tregs), CD8 + /Tregs, and effector CD8 + /Tregs. However, the mice had decreased Treg counts in splenocytes (Figure 2A,B). When we compared immune cell profiling between oxaliplatin alone and oxaliplatin + B. breve-treated groups, B. bre JCM92 enhanced the effect of oxaliplatin by increasing CD4 + T, CD8 + T, NK cells and the CD4 + /Treg, CD8 + /Treg, and effector CD8 + /Treg ratios ( Figure 2A). However, B. bre Bb03 did not boost the anti-tumor effect of oxaliplatin ( Figure 2B). Immune cell profiling of tumor-infiltrating lymphocytes (TILs) revealed that B. bre JCM92 and B. bre Bb03 significantly increased the CD4 + /Treg, CD8 + /Treg, and the effector CD8 + /Treg ratio ( Figure 3A,B). Furthermore, oxaliplatin combined with B. bre JCM92 significantly increased CD8 + T cells and the CD4 + /Treg and CD8 + /Treg ratios ( Figure 3A,B). Next, we examined intra-tumor cytokine expression using qPCR. Interferon-gamma (IFNγ) and interleukin-2 (IL-2) expression in tumor tissue was much higher in mice treated with oxaliplatin with or without B. breve strains (B. bre JCM92 and B. bre Bb03) compared to IFN-γ and IL-2 expression in vehicle-treated mice. Notably, mice treated with combined oxaliplatin and B. bre JCM92 exhibited significantly higher IFN-γ and IL-2 expression than mice treated with oxaliplatin alone ( Figure 3C). In contrast, interleukin 10 (IL-10) expression was much lower in the mice treated with oxaliplatin with or without B. breve strains than in mice treated with the vehicle ( Figure 3C). These results demonstrate that B. breve monotherapy induces anti-tumor effects by augmenting anti-tumor immunity, but only B. bre JCM92 enhances the anti-tumor effects of oxaliplatin. Cancers 2021, 13, x Immune cell profiling of tumor-infiltrating lymphocytes (TILs) revealed tha JCM92 and B. bre Bb03 significantly increased the CD4 + /Treg, CD8 + /Treg, and the CD8 + /Treg ratio ( Figure 3A,B). Furthermore, oxaliplatin combined with B. bre JCM nificantly increased CD8 + T cells and the CD4 + /Treg and CD8 + /Treg ratios (Figur Next, we examined intra-tumor cytokine expression using qPCR. Interferon-gamm γ) and interleukin-2 (IL-2) expression in tumor tissue was much higher in mice with oxaliplatin with or without B. breve strains (B. bre JCM92 and B. bre Bb03) co to IFN-γ and IL-2 expression in vehicle-treated mice. Notably, mice treated with co oxaliplatin and B. bre JCM92 exhibited significantly higher IFN-γ and IL-2 expressi mice treated with oxaliplatin alone ( Figure 3C). In contrast, interleukin 10 (IL-10) sion was much lower in the mice treated with oxaliplatin with or without B. breve than in mice treated with the vehicle ( Figure 3C). These results demonstrate that monotherapy induces anti-tumor effects by augmenting anti-tumor immunity, b B. bre JCM92 enhances the anti-tumor effects of oxaliplatin. p-values were determined by one-way ANOVA using Tukey's post-test. * p < 0.05, ** p < 0.001, *** p < 0.001, and ns, not significant.

B. bre JCM92 Boosts the Efficacy of PD-1 Blockade by Enhancing Anti-Tumor Immunity
Several studies have revealed that the gut microbiota enhances the response of immune checkpoint inhibitors that target cytotoxic T lymphocyte antigen-4 (CTLA-4) and programmed cell death protein-1 (PD-1) [12,26]. Therefore, we tested whether Bifidobacterium strains boost the efficacy of PD-1 blockade ( Figure 4A). Similar to previous results, mice treated with a PD-1 blockade and B. bre JCM92 exhibited significantly decreased tumor growth compared to mice treated with PD-1 blockade alone ( Figure 4A).
Splenic immune cell profiling showed that combined PD-1 blockade and B. bre JCM92 significantly enhanced anti-tumor immunity by increasing CD8 + T, effector CD8 + T, and NK cells. Additionally, combined PD-1 blockade and B. bre JCM92 increased the CD8 + /Treg and effector CD8 + /Treg ratios compared to PD-1 blockade alone ( Figure 4B). In tumor tissue, PD-1 blockade combined with B. bre JCM92 significantly increased the CD8 + T cell level and the CD8 + /Treg ratio compared to PD-1 blockade alone ( Figure 5A). However, immune cell profiling of the spleen and tumor tissue revealed that B. bre Bb03 did not enhance the anti-tumor immunity of PD-1 blockade (Figures 4C and 5B). Intra-tumoral cytokine expression revealed that mice treated with PD-1 blockade and B. bre JCM92 had much higher IFN-γ and IL-2 expression compared to mice treated with PD-1 blockade alone ( Figure 5C). Although IFN-γ expression was significantly enhanced in mice treated with PD-1 blockade combined with B. bre Bb03, the combination of PD-1 blockade and B. bre JCM92 resulted in higher IFN-γ expression ( Figure 5C). IL-10 expression was significantly decreased in mice treated with PD-1 blockade and either B. breve strain ( Figure 5C).  Several studies have revealed that the gut microbiota enhances the response of immune checkpoint inhibitors that target cytotoxic T lymphocyte antigen-4 (CTLA-4) and programmed cell death protein-1 (PD-1) [12,26]. Therefore, we tested whether Bifidobacte-   Splenic immune cell profiling showed that combined PD-1 blockade and B. bre JCM92 significantly enhanced anti-tumor immunity by increasing CD8 + T, effector CD8 + T, and NK cells. Additionally, combined PD-1 blockade and B. bre JCM92 increased the CD8 + /Treg and effector CD8 + /Treg ratios compared to PD-1 blockade alone ( Figure 4B). In tumor tissue, PD-1 blockade combined with B. bre JCM92 significantly increased the CD8 + T cell level and the CD8 + /Treg ratio compared to PD-1 blockade alone ( Figure 5A). However, immune cell profiling of the spleen and tumor tissue revealed that B. bre Bb03 did not enhance the anti-tumor immunity of PD-1 blockade ( Figure 4C and Figure 5B). Intratumoral cytokine expression revealed that mice treated with PD-1 blockade and B. bre JCM92 had much higher IFN-γ and IL-2 expression compared to mice treated with PD-1 blockade alone ( Figure 5C). Although IFN-γ expression was significantly enhanced in mice treated with PD-1 blockade combined with B. bre Bb03, the combination of PD-1 blockade and B. bre JCM92 resulted in higher IFN-γ expression ( Figure 5C). IL-10 expression was significantly decreased in mice treated with PD-1 blockade and either B. breve strain ( Figure 5C). were measured using qPCR. Data are expressed as the means ± SEM. p-values were determined by one-way ANOVA using Tukey's post-test within the PD-1 blockade-treated groups. * p < 0.05, ** p < 0.01, *** p < 0.001, and ns, not significant.

Transcriptome Analysis of Intestinal Tissue Reveals That B. bre JCM92 Has the Ability to Boost the Effects of Cancer Therapeutics
To understand the mechanism by which the administration of the gut microbiome influences the intestine and anti-tumor immune responses, we performed whole transcriptome sequencing on the intestine of mice treated with cancer therapeutics and B. bre JCM92 and B. bre Bb03. Hierarchical clustering analysis revealed that intestinal gene-expression patterns were distinct between mice treated with B. bre JCM92 and B. bre Bb03 ( Figure 6A). We analyzed differentially expressed genes (DEGs) between the enhanced (oxaliplatin + B. bre JCM92 and PD-1 blockade + B. bre JCM92) and unenhanced groups (oxaliplatin + B. bre Bb03 and PD-1 blockade + B. bre Bb03; p < 0.001; Figure 6B). Cytoscape network visualization exhibited that DEGs from mice treated with cancer therapeutics and B. breve strains (B. bre JCM92 and B. bre Bb03) are biologically related ( Figure 6C). ClueGO analysis revealed that the term "Regulation of cytokines" was significantly enriched in mice treated with cancer therapeutics combined with B. bre JCM92 compared to mice treated with cancer therapeutics and B. bre Bb03 ( Figure 6D and Table S4). Next, we performed gene set enrichment analysis (GSEA) between the boosted group and non-boosted group. The significantly enriched gene signatures in mice treated with the cancer therapeutics (oxaliplatin and PD-1 blockade) and B. bre JCM92 compared to mice treated with cancer therapeutics and B. bre Bb03 included gene sets involving E2F targets, tumor necrosis factor-α (TNF-α) signaling via nuclear factor kappa B (NFκB), allograft rejection, and inflammatory response ( Figure 6E and Figure S1). Consistent with tumor cytokine expression, mice treated with cancer therapeutics and B. bre JCM92 exhibited higher IL-2, STAT5 signaling, and IFN-γ response compared to mice treated with oxaliplatin or PD-1 blockade and B. bre Bb03 ( Figure 6F and Figure S2). Transcription factors regulate transcription initiation by recognizing specific binding sites, and transcription factor dysregulation is associated with human cancers [27][28][29]. To understand the regulatory mechanism of the gut microbiome, motif analysis of the intestinal transcriptome was performed. Nuclear transcription factor Y subunit beta and nuclear transcription factor Y subunit alpha genes were significantly associated with the upregulated genes in mice treated with the combination of cancer therapeutics and B. bre JCM92 (boosted group; Figure 6G). Nuclear factor Y binds to the interleukin-4 (IL-4) promoter and regulates IL-4 gene expression [30,31]. IL-4 regulates lymphocyte expansion and function [32] and exhibits anti-tumor activity by promoting Gr-1 + granulocyte and CD8 + T cell maturation [33]. These results reveal that supplementation of B. breve affects intestinal gene expression and that the changes are associated with anti-tumor immune responses.

Discussion
The gut microbiome affects immune system development and differentiation by regulating innate immune responses, including innate lymphoid cells and gene expression [34]. Several diseases, such as cancer, obesity, and inflammatory bowel diseases are correlated with intestinal dysbiosis [34,35]. Previous studies suggest that supplementation with probiotics influences host homeostasis and has a therapeutic effect on immune-mediated diseases [36,37]. Bifidobacterium spp. are among the most frequently used as probiotics and have favorable effects in several diseases, including ulcerative colitis and enteropathogenic E. coli O157 infection [38,39].
Here, we found that Bifidobacterium strains confer anti-tumor properties. Although two B. breve strains increased anti-tumor immunity, only one B. breve strain boosted cancer therapeutic efficacy. Immune cell profiling of spleen and tumor tissue revealed that B. breve strains alone augmented anti-tumor immunity by increasing CD8 + T and effector CD8 + T cell numbers, and by increasing the CD8 + /Treg and effector CD8 + /Treg ratio. However, only B. bre JCM92 boosted the efficacy of cancer therapeutics.
Interestingly, immune cell profiling exhibited that spleens from mice treated with B. bre JCM92 and PD-1 blockade had significantly increased CD4 + and CD8 + T cells compared to PD-1 monotherapy ( Figure 4B), whereas tumors from mice treated with B. bre JCM92 and PD-1 blockade only showed enhanced CD8 + T cells compared to PD-1 monotherapy ( Figure 5A). We hypothesize that the CD4 + T cells activate CD8 + cells rather than directly killing tumor cells. Most CD4 + T cells could not recognize cancer cells directly due to a lack of MHCII (Major Histocompatibility complex II) in most solid cancer cells (except melanoma and breast cancer [40]), whereas CD4 + and CD8 + T cells are activated in spleens under immune activation and inflammatory conditions [41]. For those reasons, increased CD4 + and CD8 + T were observed from splenocytes analysis, while only CD8 + T cells were increased in the tumor environment.
Using bacterial transcriptome analysis, we observed that the boosting B. bre JCM92 expressed genes were significantly enriched in the ribonucleoside metabolic process, cellular amino acid metabolic process, and amino sugar biosynthetic process pathways compared to non-boosting B. bre Bb03. Previous studies showed that amino acid degrading enzymes have anti-tumor effects in preclinical and clinical trials [36,37]. The difference in gene expression between the two B. breve strains may be related to the boosting effects of B. bre JCM92. However, the exact anti-cancer molecular mechanisms and metabolic processes in B. bre JCM92 need to be further investigated. When bacteria are orally administered, the intestine is the primary colonization site [42,43]. Previous studies also showed that the gut microbiome composition is an important factor in determining the response to cancer therapeutics, especially immunotherapy. For instance, Bifidobacterium spp., Ruminococcaceae and Faecalibacterium are significantly enriched in melanoma patients who respond to anti-PD-1 treatment [44,45]. In addition, Akkermansia muciniphila abundance is associated with anti-PD-1 treatment response in epithelial tumors [46]. Therefore, we hypothesize that the boosting effects of B. breve appear by altering intestinal gene expression. Intestinal transcriptome analysis showed that mice treated with cancer therapeutics and B. bre JCM92 showed significantly enriched E2F targets, TNF-α signaling via NFκB, inflammatory responses, and IFN-γ responses. The E2F family of transcription factors plays an important role as transcriptional regulators of cell cycle-dependent gene expression [47], and also modulate apoptosis, metabolism, and angiogenesis in cancer [48,49]. E2F expression in cancer shows conflicting results in both tumor suppression and progression [50,51]. However, the relationship between E2F expression in the intestine and tumors has not yet been investigated. TNF-α is produced by various cells, including macrophages, NK cells, and T cells, and is released in response to peptidoglycan, lipopolysaccharide, and bacterial components [52,53]. Furthermore, TNF-α is involved in apoptosis, cell survival, and inflammation, and is a potential target for cancer therapy [52,54]. Previous studies have shown that TNF-α enhances antitumor effects when used in combination with other cancer therapeutics and cytokines, including adriamycin, actinomycin D, and IFN-γ [55,56].
Similarly to the intestinal transcriptome analysis, intra-tumoral IFN-γ expression was significantly enhanced in mice treated with B. bre JCM92 (boosting) and cancer therapeutics compared to mice treated with B. bre Bb03 (non-boosting) and cancer therapeutics. IFN-γ is a major player in anti-tumor immunity, and exerts anti-proliferative and pro-apoptotic effects in tumor cells [57][58][59][60], recruits immune cells into the tumor microenvironment [61,62], and enhances the tumoricidal activities of innate and adaptive immune cells [63][64][65]. This cytokine also exhibits anti-angiogenic activity via its anti-proliferative and apoptotic effects on endothelial cells [66,67]. Mutations in IFN-γ signaling genes were identified in tumors resistant to cancer immunotherapy, highlighting the essential role for this protein in chemotherapy responses [68,69]. These results demonstrate that the administration of B. bre JCM92 increases anti-tumor immunity by modulating cytokines, including TNF-α and IFN-γ, through the gut-systemic-tumor pathway. However, IL-10 expression was different between intestine and tumors. This difference demonstrates that the anti-tumor effects of B. breve are not same in all tissues. We hypothesize that other molecules, including metabolites, underlie these differences. Investigating the underlying mechanisms distinguishing intestine and tumor tissue after B. breve treatment needs to be further explored.

Conclusions
In summary, we identified two B. breve strains that show anti-tumor effects by enhancing anti-tumor immunity. Notably, only specific Bifidobacterium strains boosted the efficacy of cancer therapeutics. Further study is needed to examine whether supplementation with Bifidobacterium strains can be used as supportive anti-tumor drugs to boost the efficacy of cancer therapeutics in clinical trials.
Supplementary Materials: The following are available online at https://www.mdpi.com/2072-669 4/13/5/957/s1, Figure S1: GSEA of intestine RNA sequencing data; Figure S2: GSEA of intestine RNA sequencing data; Table S1: Bifidobacterium strains used in this study; Table S2: Upregulated genes in B. bre JCM92 compared to B. bre Bb03; Table S3: Gene ontology of DEGs in B. bre JCM92 compared to B. bre Bb03; Table S4: Gene ontology of DEGs in mice treated with cancer therapeutics and B. bre JCM92 compared to mice treated with cancer therapeutics and B. bre Bb03.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding author, H.P., upon reasonable request.