Whey-based diet containing medium chain triglycerides modulates the gut microbiota and protects the intestinal mucosa from chemotherapy while maintaining therapy efficacy

Abstract

Cytotoxicity (i.e. cell death) is the core mechanism by which chemotherapy induces its anti-cancer effects. Unfortunately, this same mechanism underpins the collateral damage it causes to healthy tissues. The gastrointestinal tract is highly susceptible to chemotherapy's cytotoxicity, resulting in ulcerative lesions (termed gastrointestinal mucositis, GI-M) that impair the functional capacity of the gut leading to diarrhea, anorexia, malnutrition and weight loss, which negatively impact physical/psychological wellbeing and treatment adherence. Preventing these side effects has proven challenging given the overlapping mechanisms that dictate chemotherapy efficacy and toxicity. Here, we report on a novel dietary intervention that, due to its localized gastrointestinal effects, is able to protect the intestinal mucosal from unwanted toxicity without impairing the anti-tumor effects of chemotherapy. The test diet (containing extensively hydrolyzed whey protein and medium chain triglycerides (MCTs)), was investigated in both tumor-naïve and tumor-bearing models to evaluate its effect on GI-M and chemo-efficacy, respectively. In both models, methotrexate was used as the representative chemotherapeutic agent and the diet was provided ad libitum for 14 days prior to treatment. GI-M was measured using the validated biomarker plasma citrulline, and chemo-efficacy defined by tumor burden (cm³/g body weight). The test diet significantly attenuated GI-M (P = 0.03), with associated reductions in diarrhea (P < 0.0001), weight loss (P < 0.05), daily activity (P < 0.02) and maintenance of body composition (P < 0.02). Moreover, the test diet showed significant impact on gut microbiota by increasing diversity and resilience, whilst also altering microbial composition and function (indicated by cecal short and brained chain fatty acids). The test diet did not impair the efficacy of methotrexate against mammary adenocarcinoma (tumor) cells. In line with the first model, the test diet minimized intestinal injury (P = 0.001) and diarrhea (P < 0.0001). These data support translational initiatives to determine the clinical feasibility, utility and efficacy of this diet to improve chemotherapy treatment outcomes.

Fig. 1. An extensively hydrolyzed whey-protein diet rich in MCTs prevents weight loss, mucosal injury and anorexia caused by methotrexate.

Weight gain was comparable in rats during the induction phase, prior to MTX (A), with differences in weight only observed after MTX (B). Plasma citrulline, a biomarker of mucosal injury, decreased in both MTX treated groups (C), however the depth and duration were decreased in rats receiving the test diet resulting in a significantly lower AUC (D). Food intake, a marker of anorexia, was maintained in rats consuming the test diet (E). Water intake increased when consuming the test diet (F). All experiments were performed in N = 9 tumor-naïve rats treated with 45 mg/kg MTX or vehicle control. Data are shown as mean ± SEM and were analyzed using mixed models (repeated data) or a one-way ANOVA.

Fig. 2. Minimizing the severity of gastrointestinal mucositis maintained body composition.

Body composition was assessed longitudinally in all rats to determine body fat (A, B), body fluid (C, D) and lean muscle mass (E, F). When administered in rats consuming the control diet, MTX caused a decrease in all parameters. Consumption of the test diet mitigated these effects. At necropsy, atrophy of the tibialis anterior (G) and soleus (H) were decreased in the methotrexate+control diet group, but not the test diet group. All experiments were performed in N = 9 tumor-naïve rats per group. Data are shown as mean ± SEM and were analyzed using mixed models (repeated data) or a one-way ANOVA. Data are shown as mean ± SEM and were analyzed using mixed-models (repeated data) or a one-way ANOVA.

Fig. 3. Minimizing the severity of gastrointestinal mucositis with the test diet maintained physical activity.

Graphs (A–C) depict day and night activity, calculated from sensors mounted above cages connected to MED-PC® IV software collected from the day of MTX treatment. D shows normalized (night) activity over the entire time course, with methotrexate causing a profound decrease in daily activity from day 2–5 in the control diet. This was prevented by the test diet. All experiments were performed in N = 9 tumor-naïve rats per group. Data are shown as mean ± SEM and were analyzed using mixed models.

Fig. 4. The gut (fecal) microbiome is modulated by the test diet before and after methotrexate.

16 S rRNA gene sequencing revealed an increase in alpha diversity (A) and altered the relative abundance of Peptostreptococcaeceae (B), Muribaculaceae (C) and Ruminococcaceae (D). Microbial changes were most evident at day -1 (before MTX) and day 10. All analyses were performed using repeated fecal samples collected from N = 9 tumor-naïve rats per group.

Fig. 5. The test diet increases the production of both short- and branched-brain fatty acids.

Fatty acids were analyzed in cecal contents collected at day 10. All experiments were performed in N = 9 tumor-naïve rats per group. Data are shown as mean ± SEM and were analyzed using mixed models.

Fig. 6. An extensively hydrolyzed protein diet enriched with MCTs does not influence tumor growth or methotrexate efficacy, whilst maintaining mucoprotective effects.

Tumor burden shown longitudinally (A) and at necropsy (B) was not affected by dietary intervention. The number of tumors present at necropsy was significantly higher in the control diet compared to the test diet (C, P < 0.0001). Mucosal atrophy (D) and diarrhea (E, F) were minimized, but not prevented, by the test diet. All experiments were performed in N = 8 tumor-bearing rats per group. Data are shown as mean ± SEM and were analyzed using mixed models (A), a one-way ANOVA (B) or Chi Squared test (C, E, F).