The colonic microbiome is essential in human health and disease. Clostridium difficile-associated diarrhea (CDAD), a highly morbid form of infectious diarrhea, is caused by antibiotics which perturb the microbiome and allow C. difficile to proliferate. Proton pump inhibitors (PPIs) are powerful suppressors of gastric acid and among the most common medicines in the United States. Dozens of observational studies show that longterm PPI use is associated with CDAD. However, the mechanism by which PPIs cause CDAD is unknown. We believe that PPIs cause CDAD by inducing alterations in the human colonic microbiome. We will confirm or refute the hypothesized mechanism for the association between PPIs and CDAD using an unblinded, single-armed study design. We will use pyrosequencing of the hypervariable V4 region of the bacterial 16S ribosomal subunit gene in human fecal samples to describe the colonic flora. We will collect fecal samples from volunteers before and after PPIs given for different durations and test the microbiome to determine 1) whether PPIs diminish overall diversity, 2) whether PPIs diminish relative abundance of Bacteroidetes, 3) whether increased duration of PPIs affects diversity, and 4) whether there is recovery of diversity after completing a defined course of PPIs. We believe that PPIs will cause a pattern of diminished overall microbiome diversity and reduced anaerobes - the same pattern seen after use of antibiotics. Furthermore, we believe that increased PPI duration will further diminish diversity and that the microbiome will return to pre-PPI levels of diversity after PPIs are stopped. These results will facilitate biologically-based clinical interventions to reduce rates of CDAD among patients who require acid suppression.

Study Design We will recruit 12 adult volunteers for a crossover study with a total duration of 12 weeks. Subjects will be observed off of PPIs for 4 weeks and then will be placed on PPIs for 4 weeks. Subsequently, subjects will be randomized to receive an additional 4 weeks of PPIs or no therapy. Stool samples will be collected at 4 separate time points.

Study Outcomes and Statistical Analyses The primary outcome will be change in overall diversity of fecal flora after 4 weeks of PPIs compared to 4 weeks of no acid suppression. Additional outcomes to be assessed include the effect of PPIs on the relative abundance of Bacteroidetes at week 4 and change in the diversity of fecal flora at week 8.

Inclusion Criteria:

• 18 or more years old
• Able to give informed consent

Exclusion Criteria:

• Use of systemic antibiotics within the past year
• Use of acid suppression medications (PPIs or H2-receptor antagonists) within the past year (antacids permitted if more than one month from date of enrollment)
• History of chronic gastrointestinal mucosal disease (e.g. inflammatory bowel disease, celiac disease, microscopic colitis)
• Any clinically significant or uncontrolled major morbidity, including but not limited to serious cardiac or respiratory disease or uncontrolled HIV
• Abnormal bowel frequency (minimum once every 2 days, maximum 3 times per day)
• Use of clopidogrel or medications with potential significant interaction with PPIs
• Osteoperosis or history of non-traumatic bone fracture
• History of adverse reactions to PPIs
• Initiation of any new medication within the month prior to enrollment
• Pregnancy
• Inability to give informed consent

Refer to the published study results for summary data.

Healthy volunteers.

Up to 6 months post-intervention.