Houseflies Are an Unlikely Reservoir or Vector for Helicobacter pylori

Abstract

The route of transmission of Helicobacter pylori from individual to individual remains undefined. It has recently been reported that the domestic housefly, Musca domestica, when fed pure cultures of H. pylori, was able to harbor the organism in its midgut for up to 30 h (P. Grubel, S. Hoffman, F. K. Chong, N. A. Barstein, C. Mepani, and D. R. Cave, J. Clin. Microbiol. 35:1300–1303, 1997). Our investigation examined whether houseflies could acquire H. pylori from fresh human feces. Domestic houseflies (40 flies/group) were exposed for 24 h to feces from an H. pylori-positive volunteer, feces from an H. pylori-negative volunteer, or feces from an H. pylori-negative volunteer to which a known amount of viable H. pylori had been added. At various intervals, flies were sacrificed and the midguts were excised, homogenized, and plated in duplicate onto selective horse blood agar plates. All plates were incubated under microaerobic conditions at 37°C for 14 days. Emergent colonies presumptive of H. pylori were picked and tested biochemically to confirm the identity as H. pylori. H. pylori was not recovered from houseflies fed human feces either naturally infected or artificially infected with H. pylori. These results suggest that the domestic housefly is not a vector for transmission or a reservoir for H. pylori infection.

It is recognized that gastritis, gastric ulcer, duodenal ulcer, gastric carcinoma, and primary gastric B-cell lymphoma are all associated with Helicobacter pylori infection (2, 3, 7, 11, 20, 35). The rate at which a population acquires H. pylori infection is greater in developing than in industrialized countries (1, 9, 11, 24, 26, 29, 30). Risk factors for an increased prevalence of H. pylori seropositivity include lower socioeconomic strata (31), overcrowded living conditions (27), large sibship size (32), substandard sanitation (28), the presence of regurgitated gastric contents in the environment (23), the presence of children in the family (25), the presence of a single parent in the household (29), and the consumption of contaminated drinking water (5, 19, 22). H. pylori infection clusters in families with children (4, 25), but the fact that the genetically unrelated spouse is at high risk of infection further confirms the importance of environmental factors in H. pylori infection. While the natural habitat of H. pylori is the human stomach, transmission has been linked to consumption of water in Peru and consumption of vegetables grown in regions where human feces is used as fertilizer (18). These conditions may predispose to infestation by flies and other pests. Two physiological factors make the fly an ideal vector and reservoir for H. pylori infection. First, the mid-midgut of flies is acidic (much like the human stomach) and could possibly be selective for H. pylori. Second, flies can only ingest liquefied meals and must regurgitate their enzyme-laden gastric contents onto food to facilitate feeding (6, 13, 33). The postulated mechanism by which flies disseminate infection involves the selective concentration of H. pylori in the fly’s midgut following ingestion of fecal material and then regurgitation of H. pylori-contaminated digestive fluids or defecation onto foodstuffs which are subsequently consumed by humans (6, 13, 34). Grubel et al. (14, 15) showed that flies fed pure cultures of H. pylori harbored viable bacteria in their midguts for as long as 30 h after the initial exposure. While these results demonstrated the prolonged retention of H. pylori in the fly’s alimentary tract, the study design failed to replicate adequately the “natural” conditions consistent with poor sanitation. Data from studies of fecal cultures to recover H. pylori suggest that this organism is present at significantly lower concentrations in nature than in a lawn of bacterial growth on an agar plate (19, 21, 22). However, the fly as a vector for the spread of infectious disease has already been established, as in the transmission of blinding chlamydial infection, cholera, shigellosis, and salmonellosis (8, 12, 13). We designed our study to ascertain whether the domestic housefly, Musca domestica, could acquire H. pylori from fresh human fecal specimens.

One hundred fifty domestic housefly (M. domestica) pupae (American Biological Supply Company, Gainesville, Fla.) were allowed to emerge at room temperature for 5 days and fed sterilized sugar water ad libitum until used. Fresh fecal specimens were obtained from both H. pylori-infected and noninfected volunteers. The infected volunteer had asymptomatic gastritis and had a positive ¹³C-urea breath test (Meretek, Houston, Tex.) and a positive HM-CAP enzyme-linked immunosorbent assay immunoglobulin G antibody test (Enteric Products, Inc., Westbury, N.Y.). The H. pylori-negative volunteer had negative results for all diagnostic tests (¹³C-urea breath test, culture, histology, and HM-CAP enzyme-linked immunosorbent assay immunoglobulin G antibody test). The H. pylori used for the inoculum was recovered from a frozen stock and plated on brain heart infusion agar (Difco Laboratories, Detroit, Mich.) plates containing 0.25% yeast extract (Difco) and 5% horse blood (Cocalico Biologicals, Inc., Reamstown, Pa.) (HBA). The plates were incubated at 37°C for 3 days under 12% CO₂. A bacterial suspension equivalent in density to a no. 5 McFarland standard (approximately 1.5 × 10⁹ CFU/ml) was prepared in sterile saline. Three milliliters of the bacterial suspension was then added to approximately 50 g of fresh stool from the H. pylori-negative volunteer. The flies were divided into three groups and placed in separate caged enclosures. Group 1 was exposed to feces from the H. pylori-positive volunteer, group 2 was exposed to feces from the H. pylori-negative volunteer, and group 3 was exposed to feces from the H. pylori-negative volunteer to which viable H. pylori was added. The exposure period for all groups was 24 h. At various intervals postexposure (0, 6, 12, 18, 24, 36, and 48 h), five flies from each group were sacrificed. The flies were immersed in 70% ethanol prior to dissection to clean the external surfaces of any adherent bacteria. The midguts of the flies were excised and placed in cold cysteine medium until cultured. The midgut tissue was homogenized and then plated in duplicate on selective HBA plates containing 10-mg/liter nalidixic acid, 5-mg/liter trimethoprim, 3-mg/liter vancomycin, and 2-mg/liter amphotericin B (16). All plates were incubated at 37°C under 12% CO₂ for up to 14 days. Emergent colonies were picked and tested to identify H. pylori. Criteria used to confirm the identity of the selected isolates as H. pylori included colony morphology on plated media, Gram stain and cellular morphology, and positive biochemical reactions to catalase, urease, and oxidase tests. Both the CLOtest (Tri-Med Specialties, Inc., Lenexa, Kans.) and hpfast (GI Supply, Camp Hill, Pa.) rapid urease tests (RUTs) were used to detect H. pylori in fly midgut tissue homogenates. Testing was only performed on tissues collected up to 24 h after exposure to the fecal samples. All testing was performed as instructed by the manufacturer. Results were based on reading after incubation at 25°C for 24 h. A change in the color of the gel from yellow to either red (CLOtest) or green (hpfast) was indicative of a positive result. Any change in the pH indicator after 24 h of incubation was read as a negative RUT result.

After we confirmed that M. domestica will not light on human stool, we removed all other sources of water. The flies were then observed to light on and consume stool. H. pylori were not recovered from the midgut tissue from any of the flies at any time point (Table 1). Although other contaminating organisms were recovered (gram-positive cocci and bacilli), none of the colonies resembling H. pylori and selected for subculture were confirmed as H. pylori following subsequent testing using biochemical and morphologic criteria.

TABLE 1.

Microbiological recovery of H. pylori from the alimentary tracts of domestic houseflies after exposure to human stool for 24 h

Stool sample source	No. of flies positive/no. tested at post exposure time (h):
	0	6	12	18	24	36	48
H. pylori-positive volunteer	0/5	0/5	0/5	0/5	0/5	0/5	0/5
H. pylori-negative volunteer	0/5	0/5	0/5	0/5	0/5	0/5	0/5
H. pylori-negative volunteera	0/5	0/5	0/5	0/5	0/5	0/5	0/5

^aSample seeded with H. pylori. 

Both the CLOtest and hpfast RUTs detected urease-producing bacteria in the remnant midgut tissues (Table 2). However, all alkaline responses occurred after 24 h of incubation at room temperature, suggesting the presence of non-H. pylori urease-positive organisms, and correlated with the presence of Proteus species in the fecal specimen used as the negative control. None of the tissues from flies exposed to feces from the H. pylori-positive volunteer produced an alkaline RUT result.

TABLE 2.

RUT results of housefly alimentary tract tissue after exposure to human stool for 24 h

Stool sample source	No. of flies positive/no. testeda at postexposure time (h):
	0	6	12	18	24	36	48
H. pylori-positive volunteer	0/5	0/5	0/5	0/5	0/5	NDb	ND
H. pylori-negative volunteer	0/5	0/5	0/5	0/5	0/5	ND	ND
H. pylori-negative volunteerc	0/5	0/5	0/5	0/5	0/5	ND	ND

^aBoldface entries indicate the presence of Proteus species in the stool specimen and groups giving an alkaline response in the RUT after >24 h of incubation (negative test result). 

^bND, not done. 

^cSample seeded with H. pylori. 

M. domestica is a classical eusyanthropic fly species; i.e., it is trophically linked to the human habitat (12, 13). Originally, M. domestica was coprophagous and was adapted to feeding on the excreta of ungulate mammals. Its presence in areas of decaying organic matter and its repute as a vector for disease transmission made M. domestica a likely candidate as a vector for transmitting H. pylori infection. However, its extended association with humankind has broadened its food range, and as a result, synanthropic populations of this species have become trophically adapted, instead, to nonfecal human waste such as kitchen offal, decaying organic material, and decomposing proteins of animals. The original trophic adaptation of M. domestica to feces has gradually disappeared and is now preserved only in the subtropical asynanthropic populations of M. domestica, e.g., M. domestica vicina (12, 13). Evidence for this trophic shift was shown in our study when the recently emerged M. domestica would not alight on the fecal specimens unless all other sources of nutrient were removed from the cage enclosures.

Our conclusions conflict with those of Grubel et al. (14, 15), who suggested that M. domestica could be a vector for the spread of infection and also serve as a reservoir for H. pylori. The fact that we were unable to recover H. pylori from houseflies that were exposed to stool containing approximately 9 × 10⁷ CFU of H. pylori per g of stool indicates that even higher levels of viable organisms must be present in nature to ensure positive recovery from flies. Previous data have shown that such high levels of viable H. pylori may not be present in the extragastric environment (5, 17, 19, 21, 34, 36). Therefore, it appears unlikely that the domestic housefly is a vector for transmission of H. pylori infection or is an extragastric reservoir of H. pylori.