Probiotics Mitigate High-cholesterol Diet-driven Fatty Liver and Pancreatic Cancer by Restoring Macrophage Homeostasis

Abstract

Background & aims: High dietary cholesterol is a known risk factor for metabolic dysfunction-associated steatotic liver disease (MASLD) and its associated hepatic carcinogenesis; however, its effect on pancreatic ductal adenocarcinoma (PDAC) is yet to be investigated. The current study explored the mechanistic association between high dietary cholesterol, MASLD, and PDAC. Importantly, we aimed to evaluate the effect of a multi-strain probiotic formulation on hypercholesterolemia-driven MASLD and PDAC.

Methods: In this study, wild-type (C57BL/6) and KC (Pdx-1 Cre; Kras^LSL-G12D) mice were fed either with regular diet or high cholesterol and cholic acid diet (HCCD) and received an oral probiotic consortium LR+F15 (Lactobacillus rhamnosus GG and Lactiplantibacillus plantarum ILSF15). Additionally, we also used a syngeneic orthotopic murine PDAC model to evaluate the efficacy of this probiotic consortium. For mechanistic studies, pancreas, liver, intestine, peri-pancreatic fats, peritoneal cells/lavage, and blood were evaluated for metabolic, inflammatory, and malignant changes through histology, enzyme-linked immunosorbent assay, flow cytometry, and quantitative reverse transcription polymerase chain reaction.

Results: HCCD induced nonobese MASLD and PDAC progression, which was eased upon probiotics intervention. Importantly, it also increased the survival of the HCCD-fed KC mice. The probiotic intervention protected against HCCD-induced leaky gut, gut microbiota translocation, and inflammatory milieu in different tissues. Interestingly, HCCD significantly increased the population of pro-inflammatory/pro-tumorigenic peritoneal macrophages, which got normalized upon probiotic administration.

Conclusions: The probiotic formulation LR+F15 significantly suppressed HCCD-induced MASLD and PDAC progression partly through suppressing leaky gut and normalizing peritoneal macrophages' inflammatory properties. These findings encourage evaluation of the potential benefits of this probiotic consortium in combination with the existing therapies against PDAC in the future.

Keywords: Cholesterol; LPM; Leaky Gut; MASLD; PDAC; PM; Probiotics.

Graphical abstract

Figure 1

HCCD induced nonobese MASLD and leaky gut in WT mice. (A) The schematic representation of the experiment. Graphs showing (B) body weight gain during the experimental weeks, (C) serum triglyceride levels, (D) total serum cholesterol, (E) serum AST levels, (F) serum adiponectin, and (G) serum leptin levels. (H) GTT curves and their respective AUC, (I) ITT curves and their respective AUC of mice on RD or HCCD. (J) Representative digital image of the gallbladder of mice fed with RD and HCCD, with gallbladder stones in the HCCD group. (K) Gross image (scale: 1 cm), H&E (scale: 20 μm), and oil red O-stained image of liver tissue (Scale: 50 μm). (L) The liver weight (g)/body weight (g) graph. (M) The serum FITC-dextran levels in the mice of different groups. Data represented as mean ± SEM. An unpaired Student’s t-test was used to test the statistical significance. ∗P < .05, ∗∗P < .01, ∗∗∗P < .001 and ∗∗∗∗P < .0001. n = 10 (5 male and 5 female), (H) n = 5 (2 male and 3 female), (I) n = 6 (3 male and 3 female) animals were used per group for the experiments.

Figure 2

HCCD enhanced the progression of pancreatic pre-neoplastic lesions in KC mice. (A) Schematic representation of the experimental setup. (B) Gross image of pancreas (left; scale: 1 cm) and pancreas (g)-to-body weight (g) ratio (right). Representative microscopic images of pancreas stained with (C) H&E and Alcian blue (scale: 100 μm), (D–E) IF co-staining with CK19 and amylase (scale: 50 μm). Graph depicting (F) the percentage of the total neoplastic area of the whole pancreatic tissue, and (G) the percentage of the types of neoplastic areas. (H) Representative microscopic images of the pancreas subjected to Masson’s trichrome (left; scale: 100 μm) and IF for α-SMA (right; scale: 75 μm). Graphs showing the quantification for (I) percentage of collagen-positive area, and (J) number of α-SMA-positive cells per field of view. (K) Representative images and (L) the graph showing the quantification of the Ki-67-positive cells (scale: 20 μm). (M) Representative images of organoids from RD and HCCD groups at Days 1, 5, and 6 (scale: 200 μm). (N) The graph represents the growth kinetics of the organoids. (O) The survival plot of the animals fed RD and HCCD for 12 weeks. Data represented as mean ± SEM, and statistical significance was tested by unpaired Student’s t-test represented as ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001. (B–C): n = 10 (5 male and 5 female); (D–E): n = 8 (4 male and 4 female); (F–L): n = 10 (5 male and 5 female)(M–N): n = 5 (3 male and 2 female); and for (O): n = 18 (9 male and 9 female) animals were used per experimental group.

Figure 3

HCCD-induced pancreatic cancer progression persisted even after the withdrawal of the diet. (A) Microscopic digital images of the pancreas of WT animals fed RD or HCCD, stained with H&E, CD3, and F4/80 (scale: 50 μm). (B) The pancreas (g)-to-body weight (g) ratio of KC mice, 20 weeks after the experiment commenced. (C–E) Histological analysis of the pancreas of KC mice after withdrawal of the HCCD showed the presence of higher-grade neoplastic lesions and PDAC (scale: 20 μm). Data represented as mean ± SEM, and an unpaired Student’s t-test with ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001 was performed to test the statistical significance. n = 10 (5 male and 5 female) animals were used per group for the experiments.

Figure 4

HCCD instigated the syngeneic orthotopic pancreatic tumor growth and metastasis. (A) Experimental setup: 12 weeks after RD and HCCD feeding, the animals were orthotopically injected with 2.5 × 10⁵ UN-KC-6141 cells and were fed with RD for another 2 weeks. (B) The graph shows the percentage of animals with metastasis to various organs. (C) The digital gross image of the whole tumor (left; scale: 1 cm) and its respective H&E image (right; scale: 100 μm). (D) The graph represents the tumor weights. (E–F) The representative IHC staining image of Ki-67 and the graph showing the number of Ki-67-positive cells per field of view (scale: 20 μm). (G) Graph representing spleen weight. (H–J) The representative histopathological image of the liver of tumor-bearing mice, and graphs showing higher liver damage and liver weight in the HCCD group (scale: 50 μm). Data represented as mean ± SEM, and statistical significance was calculated using an unpaired Student’s t-test, represented as ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001. n = 5 (3 male and 2 female) animals were used per group for the experiments.

Figure 5

HCCD induced inflammation in the pancreas and pancreatic tumors. (A) Flow cytometric analysis of total CD45-positive lymphocytes from the pancreas of WT mice fed with RD or HCCD. (B) Graph showing the expression of various pro- and anti-inflammatory genes in the pancreas of WT mice. (C–D) Flow cytometric analysis of CD45 and F4/80-positive cells in the pancreas of KC mice. (E–G) IF staining and the graph representing the CD3 and F4/80-positive cells in the pancreas of KC mice (scale: 50 μm). (H) Representative image showing COX-2 expression in the pancreas of KC mice through IHC staining (scale: 20 μm). (I) Relative expression of different inflammation-associated genes in the pancreas of KC mice. (J–L) Representative microscopic IF images and graph showing the number of CD3 and F4/80-positive cells per field of view, in the orthotopic tumor tissues (scale: 100 μm). Data represented as mean ± SEM. Unpaired Student’s t-test and 1-way ANOVA with Tukey’s analysis were used for testing the statistical significance. ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001. (A–B) and (E–I): n = 10 (5 male and 5 female), for (C–D): n = 6 (3 male and 3 female), and for (K–L): n = 5 (2 male and 3 female) animals were used per group for the experiments.

Figure 6

HCCD triggered inflammation in the pancreas-associated organs of WT mice. Representative microscopic images of liver tissue sections of WT mice stained for (A) COX-2 (left; scale: 20 μm) and F4/80 (right; scale: 50 μm). (B) Graph indicating the number of F4/80-positive cells infiltrating the liver tissues. (C) The relative mRNA expression of various inflammation-associated genes in the liver samples. (D–E) Flow cytometric analysis depicting the percentage of CD45 and F4/80-positive cells in the liver. (F–G) Representative microscopic images and graph showing the number of F4/80-positive macrophages in the liver of the orthotopic pancreatic tumor-bearing mice (scale: 50 μm). (H) The qRT-PCR analysis of signature inflammatory genes in the liver of the mice bearing the pancreatic tumor. (I) Representative microscopic images of PPF stained with H&E (left, scale: 50 μm) and IF of F4/80 (right; scale: 50 μm). (J) The graph represents the number of F4/80-positive macrophages per field of view in the PPF. Data represented as mean ± SEM, and statistical significance was tested by unpaired Student’s t-test and 1-way ANOVA with Tukey’s multiple comparison. ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001. (A–C) and (F–J): n = 5 (2 male and 3 female), and for (D–E): n = 6 (3male and 3 female) animals were used per experimental group.

Figure 7

The multi-strain probiotic formulation alleviated HCCD-persuaded liver damage. (A) Schematic representation of the experiment setup. The probiotics, individually or in the formulation, were administered orally daily. The probiotic treatment commenced 1 week before the start of experimental diet feeding and continued until the experimental endpoint. (B) The liver (g) to body weight (g) ratio graph. Graph showing the levels of (C) total serum cholesterol, (D) serum AST, and (E) serum triglyceride. (F) The representative gross image (scale: 1 cm), H&E (scale: 20 μm), and Oil red O (scale: 100 μm) staining image of the liver of different groups. (G) Graph depicting the liver damage. (H–I) qRT-PCR analysis of genes involved in cholesterol metabolism pathways in the liver. Data represented as mean ± SEM, and 1-way ANOVA with Tukey’s multiple comparison was used to calculate the statistical significance, with ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001. n = 5 (3 male and 2 female) animals were used per group for the experiments. For the liver score graph, 2-way ANOVA with Sidak’s multiple comparison test was performed.

Figure 8

The probiotic formulation reduced the leaky gut and bacterial translocation caused by HCCD. (A) The H&E images of the ileum tissues (black arrowhead indicating the epithelium; scale: 20 μm). (B–C) Representative TEM images of the ileum tissues showing the tight junctions and the gap between 2 epithelial cells (yellow arrow, scale: 2 μm), graph depicting the gap between the epithelial cells among different groups. (D) The relative mRNA expression of intestinal integrity-maintaining proteins. (E) The densitometric analysis of the Western blot of occludin in the ileum tissue. (F) The graph shows the quantification of serum FITC-dextran levels. Graph representing bacterial 16S rRNA gene copy number/ng of DNA from (G) liver and (H) abdominal fat tissue. Data represented as mean ± SEM, and 1-way ANOVA with Tukey’s multiple comparison test was used for statistical analysis, represented as ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001. n = 5 (3 male and 2 female), (C) n = 3 (2 male and 1 female), (G–H) n = 6 (3 male and 3 female) animals were used per group for the experiments.

Figure 9

The probiotic combination subdued the HCCD-induced progression of pancreatic cancer and improved the survival of KC mice. (A) Graphical representation of the experimental setup. (B) The representative H&E images of pancreatic tissue from different groups (scale: 50 μm). (C–D) Graphs representing the percentage of neoplastic area and its types in different experimental groups. (E–F) Co-IF analysis of pancreatic amylase and CK19 depicting the ductal to acinar cell ratio (scale: 50 μm). (G) The representative images for pancreatic tissues stained with Masson’s trichome (upper; scale: 50 μm) and IF for α-SMA (down; scale: 50 μm). The graph represents the quantification of (H) total collagen-positive area and (I) the number of α-SMA-positive cells per field of view. (J) The representative microscopic images and (K) quantification of Ki-67-stained pancreatic tissue samples (scale: 20 μm). (L) The survival percentage of KC mice (n = 15, comprising 7 males and 8 females per experimental group). Data are represented as mean ± SEM, and 1-way ANOVA with Tukey’s multiple comparison with ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001 was used for statistical analysis. n = 10 animals (5 male and 5 female) were used per group for the experiments.

Figure 10

The probiotic combination suppressed the HCCD-persuaded pancreatic tumor growth and metastasis. (A) The schematic representation of the experiment. (B) Representative digital images of whole tumors (scale: 1 cm), graph showing (C) tumor weight and (D) percentage of animals bearing metastasis in various tissues. (E) The representative microscopic images of tumor tissues showing H&E (left) and IHC for Ki-67 (right) (scale: 20 μm). (F) The graph shows the number of Ki-67-positive cells per field of view. (G–I) The flow cytometric analysis represents the percentage of CD45 and F4/80-positive cells in the tumor tissues. (J–L) The IF staining and quantification of CD3 and F4/80-positive cells per field (scale: 75 μm). Data represented as mean ± SEM, and 1-way analysis of variance with Tukey’s multiple comparison represented as ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001 was used for statistical analysis. n = 5 (3 male and 2 female) animals were used per experimental group.

Figure 11

The probiotic cocktail decreased the tumor burden in therapeutic mode. (A) Representation of the experiment, where 12 weeks after the experimental diets, the animals were orthotopically injected with UN-KC-6141 cells. Animals from all experimental groups were fed with RD afterwards, and the probiotic treatment was also provided after the orthotopic implantation of cancer cells. (B) The gross image of the whole tumor (scale: 1 cm), (C) the tumor weight, and (D) the percentage of animals with metastasis at various tissues. (E) The experimental setup: WT animals were orthotopically injected with UN-KC-6141 cells, and they were randomly divided into two groups: RD and RD-LR+F15. The animals were co-administered the probiotic mixture after the orthotopic implantation until the end of the experiment. Twenty-one days after cancer cell injection, the tumors were harvested. (F) The gross digital image of the whole tumor (scale: 1 cm). The graph represents (G) the tumor weight, and (H) the percentage of animals with metastasis in different tissues. Data represented as mean ± SEM, and 1-way ANOVA with Tukey’s multiple comparison or 2-tailed unpaired Student t-test was used to calculate the statistical significance with ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001. n = 5 (2 male and 3 female) animals were used per group for the experiments.

Figure 12

The probiotic combination subsided HCCD-induced pancreatic inflammation. The relative mRNA expression of different pro- and anti-inflammatory cytokines in the pancreas of (A) WT mice, and (B) KC mice. (C–F) Graphs showing the quantification of inflammation-associated cytokines expressed per mg protein of the pancreatic tissues of KC mice. (G–H) The flow cytometric analysis shows the percentage of CD45-positive cells in the pancreas of WT mice. (I–L) The contour plots and graphs show the percentage of CD45-positive lymphocytes and F4/80-positive macrophages present in the pancreas of KC mice, analyzed by flow cytometry. (M–O) The representative microscopic images and graphs show CD3- and F4/80-positive cells per field of view in the pancreas of KC mice (scale: 50 μm). The data are represented as mean ± SEM, and statistical significance was assessed using 1-way ANOVA with Tukey’s post-hoc correction and 2-way ANOVA with Bonferroni multiple comparison, with ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001, with n = 6/5 (3 male and 3/2 female) animals used per group for the experiments.

Figure 13

The probiotic blend suppresses the inflammation in the liver and PPF. (A) The relative mRNA expression of different pro- and anti-inflammatory cytokines in the liver of WT mice. (B–E) Graphs showing the quantification of cytokine expressions per mg protein of liver tissue of WT mice, measured by ELISA. (F) Representative IHC staining of COX-2 in the WT animals’ liver tissue (scale: 20 μm). (G–J) The representative flow cytometry contour plot and graphs depicting the percentage of CD45-positive cells and F4/80-positive macrophages in the liver of WT mice. (K–L) IF quantification for F4/80-positive macrophages per field of view of the liver tissues of WT mice (scale: 50 μm). (M) The representative H&E and IF image of F4/80-positive cells in the PPF of WT mice (scale: 50 μm). Data represented as mean ± SEM, and 1-way ANOVA with Tukey’s post-hoc correction and 2-way ANOVA with Bonferroni multiple comparison were used to calculate the statistical significance, represented as ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001. n = 6/5 (3 male and 3/2 female) animals were used per group for the experiments.

Figure 14

The probiotic cocktail alleviated the inflammation in the peritoneum. (A–C) Graphs showing the expression of IL-1β, IL-10, and TNF-α in the PL of WT mice through ELISA. (D) Representative images showing the migration of GFP-positive peritoneal leukocytes to the pancreas and PPF in KC mice (scale: 1 mm). (E–F) The graphs show the amount of NO produced with and without LPS stimulation by the PMs harvested from WT mice. (G–H) The flow cytometric analysis of the peritoneal cells from WT mice shows the percentage of the LPM population. Data represented as mean ± SEM, and statistical significance was tested using 1-way ANOVA with Tukey’s post-hoc analysis, with ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001. (A–C): n = 6 (3 males and 3 females), (E–F): n = 5 (3 males and 2 females), and for (G): n = 10 (5 males and 5 females) animals were used per group for the experiments.

Figure 15

Depletion of PMs and its effect on HCCD-induced pancreatic tumor progression. (A) Graphical representation of the experimental setup. (B) Flow cytometry analysis depicts that intraperitoneal clodronate treatment depleted the PM population, specifically the LPMs. (C) Representative gross image of tumors (scale: 1 cm). Graph depicting the (D) tumor weight and (E) the percentage of metastasis in various tissues. Data are presented as mean ± SEM, and 1-way ANOVA with Tukey’s post-hoc correction was used to test the statistical significance represented as ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, and ∗∗∗∗P < .0001. n = 5 animals (2 males and 3 females) were used per experimental group.