Influence of the MUC1 Cell Surface Mucin on Gastric Mucosal Gene Expression Profiles in Response to Helicobacter pylori Infection in Mice

Abstract

The cell surface mucin MUC1 is an important host factor limiting Helicobacter pylori (H. pylori) pathogenesis in both humans and mice by providing a protective barrier and modulating mucosal epithelial and leukocyte responses. The aim of this study was to establish the time-course of molecular events in MUC1-modulated gene expression profiles in response to H. pylori infection in wild type (WT) and MUC1-deficient mice using microarray-determined mRNA expression, gene network analysis and Ingenuity Pathway Analysis (IPA). A time-course over the first 72 h of infection showed significantly higher mucosal loads of bacteria at 8 h of infection in Muc1^-/- mice compared with WT, confirming its importance in the early stages of infection (P = 0.0003). Microarray analysis revealed 266 differentially expressed genes at one or more time-points over 72 h in the gastric mucosa of Muc1^-/- mice compared with WT control using a threshold of 2-fold change. The SPINK1 pancreatic cancer canonical pathway was strongly inhibited in Muc1^-/- mice compared with WT at sham and 8 h infection (P = 6.08E-14 and P = 2.25 E-19, respectively) but potently activated at 24 and 72 h post-infection (P = 1.38E-22 and P = 5.87E-13, respectively). The changes in this pathway are reflective of higher expression of genes mediating digestion and absorption of lipids, carbohydrates, and proteins at sham and 8 h infection in the absence of MUC1, but that this transcriptional signature is highly down regulated as infection progresses in the absence of MUC1. Uninfected Muc1^-/- gastric tissue was highly enriched for expression of factors involved in lipid metabolism and 8 h infection further activated this network compared with WT. As infection progressed, a network of antimicrobial and anti-inflammatory response genes was more highly activated in Muc1^-/- than WT mice. Key target genes identified by time-course microarrays were independently validated using RT-qPCR. These results highlight the dynamic interplay between the host and H. pylori, and the role of MUC1 in host defense, and provide a general picture of changes in cellular gene expression modulated by MUC1 in a time-dependent manner in response to H. pylori infection.

Keywords: Helicobacter pylori; MUC1; SPINK1; gene expression; infection; lipid metabolism.

Figure 1

H. pylori colonization levels are elevated in Muc1^−/− mice compared with wild-type (WT) controls. WT or Muc1^−/− mice were infected with a single challenge of H pylori-SS1. At each time point, stomachs were removed, and bacterial colonization was determined by colony-forming assay. Statistics: Muc1^−/− vs. wild-type controls at each time point (Ordinary one-way ANOVA corrected for multiple comparisons by controlling the False Discovery Rate (FDR) using “Two-stage step-up method of Benjamini, Krieger, and Yekutieli;” n = 3, ***p < 0.001).

Figure 2

Differentially expressed (DE) genes in the gastric tissues across different time points of H. pylori infection. Control uninfected mice were sham-infected with 0.1 mL BHI and sampled after 8, 24, and 72 h mock-infection. Equal quantities of RNA from mice of the same genotype from the three time points were pooled as a single sham control for each of 8, 24, and 72 H. pylori infection. (A) DE genes between Muc1^−/− vs. wild type (WT) mice. (B) DE genes occurring in WT gastric tissues in response to H. pylori infection over the course of infection. (C) DE genes occurring in Muc1^−/− gastric tissues in response to H. pylori infection over the course of infection.

Figure 3

IPA regulator effect networks analysis of differentially expressed (DE) genes between Muc1^−/− vs. wild type (WT) gastric tissues before infection (The highest Consistency Score of regulator effect networks is shown). Upstream regulators are located at the top of the network, target genes are in the middle of network and predicated disease or function in the bottom of network. Log ratio fold changes of DE genes in our data set is showed underneath each gene (red color symbol indicates increased expression). The concept of Regulator Effects in IPA is that The Regulator Effects algorithm connects upstream regulators, dataset molecules and downstream functions or diseases affected in your dataset to generate a hypothesis that can explain how the activation or inhibition of an upstream regulator affects the downstream target molecule expression and the impact of the molecular expression on functions and diseases. The algorithm goes through one or more iterations to merge upstream and downstream results from the Upstream Regulator. The networks are merged only if the overlap of targets has possible statistical significance (Fisher's Exact Test p-value of <0.05). For each network, a Consistency Score is calculated that rewards for paths from regulator->target->disease or function that are consistent. Higher scoring hypotheses are those with more consistent causal paths represented by a high Consistency Score.

Figure 4

IPA molecular networks analysis of differentially expressed (DE) genes between Muc1^−/− vs. wild type gastric tissues before infection. Log ratio fold changes of DE genes in our data set is shown underneath each gene (- number or green symbol indicates decreased expression and otherwise/red indicates increased expression).

Figure 5

Ingenuity Pathway Analysis (IPA) showing the identified top canonical pathways of differentially expressed genes between Muc1^−/− vs. wild type (WT) gastric tissues in sham or in response to H. pylori infection over time. Significance P-values were calculated based on the Fisher's right tailed exact test. The-log (p-value) are shown on the top x-axis of the bar chart. The orange and blue colored bars indicate predicted pathway activation, or predicted inhibition, respectively (z-score). White bars are those with a z-score at or very close to 0. Gray bars indicate pathways, where no prediction can currently be made. IPA applies a-log (p-value) cutoff of 1.3 (threshold). SPC, SPINK1 Pancreatic Cancer Pathway; UC, Urea Cycle; NRTAZ, Neuroprotective Role of THOP1 in Alzheimer's Disease; MODYS, Maturity Onset Diabetes of Young Signaling; SCM, Superpathway of Citrulline Metabolism; MDI, Melatonin Degradation I; NDIII, Nicotine Degradation III; SMD, Superpathway of Melatonin Degradation; LXR/RXR, LXR/RXR Activation; FXR/RXR, FXR/RXR Activation; BD, Bupropion Degradation; AD, Acetone Degradation I (to Methylglyoxal); EB, Estrogen Biosynthesis; SGC, SPINK1 General Cancer Pathway; AICPRR, Activation of IRF by Cytosolic Pattern Recognition Receptors; INFS, Interferon Signaling; TD, Triacylglycerol Degradation; CS, Coagulation System; APRS, Acute Phase Response Signaling; RB, Retinol Biosynthesis.

Figure 6

Ingenuity Pathway Analysis (IPA) top regulator effect networks analysis of differentially expressed (DE) genes between Muc1^−/− vs. wild type gastric tissues at 8 h H. pylori infection. Log ratio fold changes of DE genes is shown underneath each gene (red color symbol indicates increased expression). Prediction legend is the same as for Figure 3.

Figure 7

Ingenuity Pathway Analysis regulator effect networks analysis of differentially expressed (DE) genes between Muc1^−/− vs. wild type gastric tissues at 8 h H. pylori infection showed PPARG up-regulated transport of molecule pathway. Log ratio fold changes of DE genes is shown underneath each gene (red color symbol indicates increased expression). Prediction legend is the same as for Figure 3.

Figure 8

Ingenuity Pathway Analysis regulator effect networks analysis of differentially expressed (DE) genes between Muc1^−/− vs. wild type gastric tissues at 72 h H. pylori infection. Log ratio fold changes of DE genes in our data set is shown underneath each gene (red color symbol indicates increased expression). Prediction legend is the same as for Figure 3.

Figure 9

RT qPCR validation of transcripts selected as the most highly differentially expressed genes between Muc1^−/− vs. wild type (WT) gastric tissues. Statistics: Ordinary one-way ANOVA corrected for multiple comparisons by controlling the False Discovery Rate (FDR) using “Two-stage step-up method of Benjamini, Krieger, and Yekutieli;” n = 3, * vs. WT at each time point, *P < 0.05, **P < 0.01, ***P < 0.001).