Metagenomic Study Reveals Phage-Bacterial Interactome Dynamics in Gut and Oral Microbiota in Pancreatic Diseases

Abstract

Individuals with pancreatic-related health conditions usually show lower diversity and different composition of bacterial and viral species between the gut and oral microbiomes compared to healthy individuals. We performed a thorough microbiome analysis, using deep shotgun sequencing of stool and saliva samples obtained from patients with chronic pancreatitis (CP), pancreatic ductal adenocarcinoma (PDAC), and healthy controls (HCs).We observed similar microbiota composition at the species level in both the gut and oral samples in PDAC patients compared to HCs, among which the most distinctive finding was that the abundance of oral-originated Fusobacterium nucleatum species did not differ between the oral and the gut samples. Moreover, comparing PDAC patients with HCs, Klebsiella oxytoca was significantly more abundant in the stool samples of PDAC patients, while Streptococcus spp. showed higher abundance in both the oral and stool samples of PDAC patients. Finally, the most important finding was the distinctive gut phage-bacterial interactome pattern among PDAC patients. CrAssphages, particularly Blohavirus, showed mutual exclusion with K. oxytoca species, while Burzaovirus showed co-occurrence with Enterobacteriaceae spp., which have been shown to be capable of inducing DNA damage in human pancreatic cells ex vivo. The interactome findings warrant further mechanistic studies, as our findings may provide new insights into developing microbiota-based diagnostic and therapeutic methods for pancreatic diseases.

Keywords: bacteria; chronic pancreatitis; gut microbiome; metagenome analysis; oral microbiome; pancreatic cancer; pancreatic diseases; phage–bacterial interactome; virus.

Figure 1

Diversity indexes for stool samples by diagnosis group. (A) Beta diversity principal coordinates analysis (PCoA) derived from Bray–Curtis distances among stool samples by group (cancer (PDAC), chronic pancreatitis (CP), and healthy controls (HCs)) with Bonferroni adjusted p-value for pairwise Adonis comparisons; (B) Alpha diversity indexes (Observed, Shannon, Inverse Simpson) for stool samples by study group (cancer (PDAC), chronic pancreatitis (CP), and healthy controls (HCs)).

Figure 2

Diversity indexes for saliva samples by diagnosis group. (A) Beta diversity principal coordinates analysis (PCoA) derived from Bray–Curtis distances among saliva samples by group (cancer (PDAC), chronic pancreatitis (CP), and healthy controls (HCs)) with Bonferroni adjusted p-value for pairwise Adonis comparisons; (B) Alpha diversity indexes (Observed, Shannon, Inverse Simpson) for saliva samples by study group (cancer (PDAC), chronic pancreatitis (CP), and healthy controls (HCs)).

Figure 3

A graph of the 20% most prevalent bacterial and viral taxa in the stool samples among the groups. (A) A graph of the 20% most prevalent bacterial taxa in the stool samples agglomerated at the family level, according to the diagnosis group (cancer (PDAC), chronic pancreatitis (CP), and healthy controls (HCs)); (B) a graph of the 20% most prevalent viral taxa in the stool samples agglomerated at the genus level, according to the diagnosis group (cancer (PDAC), chronic pancreatitis (CP), and healthy controls (HCs)).

Figure 4

A graph of the 20% most prevalent bacterial and viral taxa in the saliva samples among the groups. (A) A graph of the 20% most prevalent bacterial taxa in the saliva samples agglomerated at the family level, according to the diagnosis group (cancer (PDAC), chronic pancreatitis (CP), and healthy controls (HCs)); (B) a graph of the 20% most prevalent viral taxa in the saliva samples agglomerated at the genus level, according to the diagnosis group (cancer (PDAC), chronic pancreatitis (CP), and healthy controls (HCs)).

Figure 5

Gut microbial interaction network in the healthy controls (HCs). Green interactions represent co-occurrence relationships. The color gradient represents the relative abundance of the species in the samples (from low to high abundance) and the width represents the prevalence.

Figure 6

Gut microbial interaction network in chronic pancreatitis (CP) patients. Green interactions represent co-occurrence relationships, while red interactions represent mutual exclusion. The color gradient represents the relative abundance of the species in the samples (from low to high abundance) and the width represents the prevalence.

Figure 7

Gut microbial interaction network in pancreatic cancer (PDAC) patients. Green interactions represent co-occurrence relationships, while red interactions represent mutual exclusion. The color gradient represents the relative abundance of the species in the samples (from low to high abundance) and the width represents the prevalence.

Figure 8

Oral microbial interaction network in the healthy controls (HCs). Green interactions represent co-occurrence relationships. The color gradient represents the relative abundance of the species in the samples (from low to high abundance) and the width represents the prevalence.

Figure 9

Oral microbial interaction network in chronic pancreatitis (CP) patients. Green interactions represent co-occurrence relationships, while red interactions represent mutual exclusion. The color gradient represents the relative abundance of the species in the samples (from low to high abundance) and the width represents the prevalence.

Figure 10

Oral microbial interaction network in pancreatic cancer (PDAC) patients. Green interactions represent co-occurrence relationships. The color gradient represents the relative abundance of the species in the samples (from low to high abundance) and the width represents the prevalence.