A Cross-Sectional Study of Helicobacter Infection Among Laboratory Animals and Animal Research Workers

Abstract

Occupational risks for Helicobacter pylori infection are not well-understood, but associations with infection have been found among certain health care workers, such as gastroenterologists, endoscopy staff, and those who work in institutions for the developmentally disabled.1,2 Some non–pylori Helicobacter species cause significant disease, such as chronic hepatitis and hepatocellular carcinoma in mice, as well as gastric disease similar to that caused by H. pylori in humans. Some non–pylori Helicobacter species, particularly species that infect cats and dogs, such as Helicobacter heilmannii and Helicobacter felis, and Helicobacter suis, common in pigs, are also known to infect humans.3,4 There are indications that zoonotic transmission of certain of these organisms is associated with human disease.5 During surveillance of the facility under investigation in this study, non–pylori Helicobacter infections were detected among mice. We hypothesized that transmission of Helicobacter pylori, non–pylori Helicobacter species, or both could occur in animal research facilities. METHODS This study was a cross-sectional study of laboratory workers, using questionnaires and stool sample polymerase chain reaction (PCR) testing to look for evidence of Helicobacter infection in workers and laboratory mice from the same facility. The study took place in a previously described university animal research laboratory.6 All nonpregnant full-time day-shift employees of the animal research facility were eligible for inclusion if they answered “Yes” to the screening question, “Do you work with animals, their fluids, tissues, or cages?” Fourteen mouse stool samples were provided for testing by the animal laboratory. Cages housed an average of three mice each; the sampling unit was the cage, not individual rodents. Samples were collected from cages that, on the basis of prior surveillance, were expected to be positive for Helicobacter. Six to 12 mouse stool pellets were taken from each cage. Human and animal stool samples were taken within 1 month of each other. A universal Helicobacter PCR was validated in controls, using Helicobacter typhlonius, Helicobacter rodentium, Helicobacter hepaticus, and Helicobacter bilis DNA (courtesy of Dr Robert Livingston, University of Missouri). The positive control consisted of H. pylori DNA (American Type Culture Collection, Manassas, VA) added to a human fecal sample. Positive and negative controls were tested with a universal primer, using a hot start AmpliTaq Gold (Invitrogen, Grand Island, NY). Universal Helicobacter primers C87 and C98 were as described previously.7 Details of PCR cycling protocol and gel verification of PCR products are available from the authors. A multiplex PCR was validated for the differentiation of five specific Helicobacter species known to infect mice: H. typhlonius, H. rodentium, H. hepaticus, Helicobacter muridarum, and H. bilis. Primer pairs specific to the species listed earlier were adapted from Feng et al.8,9 The multiplex primers were tested in reactions that contained one of the positive controls for H. bilis, H. hepaticus, H. rodentium, and H. typhlonius. These reactions yielded amplicons with the expected band sizes of 453 bp, 705 bp, 191 bp, and 122 base pairs, respectively. Multiplex reactions had primer dimers that occasionally obscured the low-molecular-weight band corresponding to the species H. typhlonius. A positive control for H. muridarum was not available; H. muridarum results were independently validated by an external commercial laboratory (IDEXX RADIL, Columbia, MO). Two multiplex sets of Helicobacter primers were developed. Mix A consisted of H. bilis and H. rodentium primers. Mix B consisted of H. hepaticus, H. muridarum, and H. typhlonius primers. Mouse and human stool samples were first tested using the universal Helicobacter PCR. Mouse fecal sample results were also analyzed independently by a commercial laboratory (IDEXX RADIL). Using the QIAmp®DNA Stool Mini Kit, DNA was extracted from the mouse and human samples. Polymerase chain reaction was run on the mouse and human stool samples by using two sets of universal Helicobacter primers, from two different sites of the 16s gene: HS1-HS210 and C97-C98.7 Negative controls tested negative. Samples positive by universal Helicobacter PCR then underwent multiplex PCR testing, for the five species listed earlier. This study was approved by the University of Illinois institutional review board. Written informed consent was obtained from all participants. RESULTS Of 14 mouse stool samples, 10 (71%) were positive for Helicobacter by universal Helicobacter PCR. Six samples (43%) were positive for H. rodentium alone, 2 (14%) were positive for both H. rodentium and H. typhlonius, and 1 (7%) was positive for both H. rodentium and H. hepaticus. Twenty-eight laboratory animal workers participated; 19 men and 9 women; mean age, 42.1; and range, 24 to 66 years. Half were African American. Most (71.4%) had daily occupational exposure to animals, their tissues, fluid, or cages. Two workers reported having been diagnosed with stomach ulcer, but none recalled having been diagnosed with H. pylori. No human stool samples were positive for Helicobacter. The human stool sample spiked with Helicobacter DNA did test positive by universal Helicobacter PCR. DISCUSSION Thousands of veterinarians, laboratory animal technicians, and laboratory animal caretakers work in hundreds of academic, pharmaceutical, and governmental animal research facilities in the United States. Accurate estimates of the number of people with laboratory animal contact are not available, but according to the US Bureau of Labor Statistics, 73,200 people work as veterinary assistants and laboratory animal caretakers (http://www.bls.gov/ooh/healthcare/veterinary-assistants-and-laboratory-animal-caretakers.htm). This estimate includes people who work in animal hospitals and clinics but excludes veterinarians. Approximately 13,000 people are members of the American Association for Laboratory Animal Science (http://www.aalas.org/). The membership is likely an underestimate of workers with exposure to laboratory animals, since an unknown fraction of laboratory animal workers are members of the association. We found that mice housed in an animal laboratory were infected by numerous Helicobacter species. No evidence of Helicobacter infection was found in humans. To the best of our knowledge, this is the first study to evaluate Helicobacter infection among laboratory animal research workers. Sen and colleagues11 evaluated the sensitivity and specificity of stool PCR for the diagnosis of H. pylori infection, using gastric biopsy as the criterion standard. They reported a 65% sensitivity and a 75% specificity of stool PCR. Although we used the same DNA extraction methods as Sen and colleagues, we used genus-level Helicobacter PCR as a screening test, reserving species-specific analyses for those that testing positive at the genus level. Because our methods are not identical, our sensitivity may have been greater or less than those of Sen and colleagues. Though the PCR may have had imperfect sensitivity, it did detect the positive control. This cross-sectional survey provided no evidence of transmission of Helicobacter infection between laboratory animals and people with occupational exposure to those animals. It is not known whether these results are because of good work practices, the dose and type of exposure to Helicobacter required for transmission, limited sensitivity of analytic methods, or limited sample size. Continued use of standard worker protections in animal research facilities seems obviously prudent. If animal laboratory populations are free of such infections, which is important given that they may compromise research results, any risk of transmission to workers would be eliminated. Linda Rosul, BA University of Illinois at Chicago School of Public Health Division of Epidemiology and Biostatistics Chicago, IL James Artwohl, DVM University of Illinois at Chicago Biological Resources Laboratory Chicago, IL William Hendrickson, PhD University of Illinois at Chicago College of Medicine Department of Microbiology and Immunology Chicago, IL Samuel Dorevitch, MD, MPH University of Illinois at Chicago School of Public Health Division of Epidemiology and Biostatistics, and Division of Environmental and Occupational Health Sciences Chicago, IL