Studies on bacterial transmission pathways in a high endemic area, with a focus on Helicobacter pylori

Abstract

Even though half of the world’s population is infected with Helicobacter pylori, which causes gastritis, peptic ulcer and gastric cancer, the transmission routes of these bacteria remain unknown despite extensive epidemiological studies. Enterotoxigenic Escherichia coli (ETEC) and Vibrio cholerae are two of the most common causes of acute watery diarrhea in developing countries. The main aim of this thesis was to study transmission pathways of these bacteria, with a focus on H. pylori, through analyses of clinical and water samples from Dhaka, Bangladesh, an area with high prevalence of gastrointestinal diseases. To determine the bacterial numbers in clinical and water samples we developed highly sensitive quantitative real-time PCR assays targeting specific and conserved virulence genes of H. pylori (cagA, flaA, glmM, hpaA, ureA and vacA), ETEC (eltA and estB) and V. cholerae (ctxB and tcpA). The assays were used for quantification of bacterial DNA and reverse-transcribed gene transcripts. Twenty-six of 39 (67 %) drinking and environmental water samples from a poor area in Dhaka were positive by real-time PCR for ETEC, whereas all 75 drinking and environmental water and 21 drinking water biofilms from the same location were negative for H. pylori, suggesting that ETEC may be waterborne while H. pylori is not. H. pylori transmission during epidemics of gastroenteritis was then explored by analyzing vomitus and stool samples collected from diarrhea patients admitted to the ICDDR, B hospital in Dhaka. All samples were positive for V. cholerae, with higher numbers in stool (median 2.5 x 106 genomes) than vomitus (median 2.7 x 104 genomes) and a strong correlation between DNA real-time PCR and quantitative culture. Analyses for H. pylori showed that 23 of 26 (88 %, median genome number = 4.35 x 105 ml-1) vomitus and 17 of 23 (74 %, median genome number = 7.33 x 102 ml-1) stool samples were positive in real-time PCR, but H. pylori could not be isolated by culture. The results indicate that high numbers of H. pylori are shed in vomitus during acute gastroenteric disease and indicate that H. pylori may be transmitted by this route. To establish possible infectivity of these bacteria, the gene expression of H. pylori in vomitus, stool and biopsies from infected individuals and in in vitro cultures was analyzed. Vomitus, biopsies and in vitro cultures showed high expression of cagA, flaA and ureA and lower expression of hpaA and vacA, whereas no expression was detected in diarrheal stool. Expression analyses of the same genes in a C57Bl/6 H. pylori strain SS1 infection mouse model showed a similar relative transcription pattern as in biopsies, in vitro cultures and vomitus and that expression is up-regulated during exponential growth. In conclusion, our results suggest that H. pylori may be disseminated through vomitus during outbreaks of gastrointestinal infections in Bangladesh and that waterborne transmission is less likely whereas waterborne transmission of ETEC may occur. Furthermore, the studies indicate that experimental murine infection and vomitus from H. pylori infected subjects may be suitable models of H. pylori virulence gene expression in vivo.