|Home | About | Journals | Submit | Contact Us | Français|
Infections with enterohepatic Helicobacter species (EHS) can change the results of animal experiments. However, there is little information about the prevalence of EHS in noncommercial animal facilities. The aim of this study was to investigate the prevalence and the spread of EHS in specific-pathogen-free (SPF) mice. Fecal samples of 40 mouse lines were analyzed for members of the family Helicobacteraceae using a group-specific PCR targeting the 16S rRNA gene. Additional experiments were carried out to evaluate the spread of EHS among mice harbored in different caging systems. Helicobacter species were detected in 87.5% of the mouse lines tested. Five different Helicobacter species were identified: H. ganmani, H. hepaticus, H. typhlonicus, and the putative Helicobacter species represented by the isolates hamster B and MIT 98-5357. Helicobacter infection did not spread between animals in neighboring cages when individually ventilated cages were used; in contrast, when the mice were reared in open-air cages, EHS were found to spread from cage to cage. However, the spread was prevented by adding polycarbonate filter tops to the cages. When Helicobacter-negative and infected mice shared the same cage, transmission of the infection occurred in 100% within 2 weeks. Furthermore, we found that mice from commercial breeding facilities may carry undetected Helicobacter infections. Taken together, we show that infection with EHS may frequently occur and spread easily in mice reared under SPF conditions despite extensive safety precautions. Moreover, there is a high prevalence of rather uncommon Helicobacter species that may be a consequence of the current routine procedures used for health screening of SPF mice.
Recently, several novel enterohepatic Helicobacter species (EHS) have been detected in laboratory mice. These are H. muridarum (17), H. hepaticus (6), H. bilis (9), H. rodentium (31), H. typhlonicus (12), H. trogontum (22), H. ganmani (28), and most recently H. mastomycrinus (32). Moreover, additional EHS have been detected which have not yet been definitely validated as novel Helicobacter species (34) (GenBank AJ007931).
The exact pathogenic potential of these novel EHS has not yet been defined, but at least some of these bacteria can lead to severe pathologies in the infected animals: H. hepaticus was found to be associated with inflammatory bowel disease (IBD) (10, 16, 36), chronic hepatitis (10, 18, 37), and liver tumors (6, 7, 37) in various mouse strains. H. bilis causes inflammatory bowel disease in severe combined immunodeficient (SCID) mice and multiple-drug resistance-deficient mice (11, 20, 33) and results in hepatitis in infected Swiss Webster mice (8). In C57L/J mice, a mouse model with high susceptibility for gallstone formation, H. bilis as well as various other EHS significantly increases the incidence of cholesterol gallstones (21).
In susceptible mice, infections with H. typhlonius (12), H. ganmani (41), or H. muridarum (15) are associated with inflammatory bowel disease. Monoinfection with H. rodentium has not been shown to be associated with disease; however, the germ augments inflammatory bowel disease in immunodeficient mice coinfected with H. hepaticus (23). The Helicobacter species represented by isolate MIT 98-5357 induces cholangiohepatitis and IBD in A/J and Tac:ICR:HascidfRF mice (34). The pathogenic potential of H. mastomyrinus is unknown (32).
Beside the direct pathological effects of EHS, it has been shown that infections of mice by EHS may significantly influence the results of animal experiments (4, 5, 14, 41). However, investigators may not be aware that their laboratory mice are Helicobacter infected because the pathology of EHS is host dependent and infection may be subclinical (39). It is therefore important to use sufficient screening methods to detect infected animals. Usually, diagnostic PCR is performed for this purpose because microbiological culture and serology have been demonstrated to have significant disadvantages (30, 38, 40).
Basically, two types of PCR assays were developed: species-specific PCR assays that allow the identification of single Helicobacter species (27, 30) and Helicobacter group-specific PCR assays which can detect several Helicobacter species in a single assay (2, 25, 27). In samples that are positive in a group-specific PCR a further step is required to identify the Helicobacter species. DNA sequencing and restriction fragment length polymorphism (RFLP) analysis are commonly used for this purpose (2, 27), although denaturing gradient gel electrophoresis has also been successfully used (25). Here, we used genus-specific PCR and consecutive DNA sequencing combined with RFLP analysis to evaluate the prevalence of Helicobacter infections and their spread in the SPF facility of our central animal laboratory.
All together, 40 mouse lines harbored in the SPF facility of the central animal laboratory of our university were tested for Helicobacter infections. There were 37 inbred mice strains and three outbred mice strains. The genetic backgrounds of the inbred strains were C57BL/6 (n = 25), BALB/c (n = 2), C3H (n = 1), CPB (n = 1), DBA (n = 1), SJL (n = 1), and mixed C57BL/6;J129 (n = 7). The outbred mice strains were BlackSwiss (n = 1), CBA;BALB/c (n = 1), and NMRI (n = 1).
The SPF tract is protected by restricted animal room entry conditions. Staff had to pass an air-shower at admission; cages and bedding (Lignocell, type HBK 1500-3000; J. Rettenmaier & Söhne GmbH & Co., Holzmühle, Germany) had been sterilized by autoclaving before being placed in the animal rooms. The animal rooms in the facility were maintained at 71 to 74°F and 45 to 55% humidity, with 16 air changes/hour and a cycle of 12 hour of light and 12 hour of dark. Mice were reared in different caging systems: open air cages in open racks, open-air cages in ventilated cabinets, and individually ventilated cages. With mice reared in individually ventilated cages, all procedures were performed in a laminar flow hood.
Between working with different mouse lines, gloves were changed and the laminar flow hood was decontaminated. Mice in individually ventilated cages were fed irradiated rodent diet 2018 (Harlan Winkelmann GmbH, Borchen, Germany). Mice in open-air cages or ventilated cabinets received nonirradiated rodent diet 2018. Tap water was used as drinking water for all animals. Routine health monitoring was carried out by Gesellschaft für innovative Mikroökologie mbH, Michendorf, Germany, in accordance with the Federation of European Laboratory Animal Science Associations recommendations (24).
Fecal samples were initially frozen in liquid nitrogen and stored at −20°C until analysis. DNA was extracted with the DNA stool minikit (QIAGEN, Hilden, Germany) according to the manufacturer's instructions. Analysis for Helicobacteraceae was performed by using a group-specific PCR assay with the primers C97-20 (5′-GGCTATGACGGGTATCCGGC-3′; He1icobacter py1ori 16S rRNA positions 260 to 279) and H3A-20 (5′-GCCGTGCAGCACCTGTTTTC-3′; anneals to positions 1007 to 1026 of H. pylori 16S rRNA) as previously described (2).
All PCRs were performed in a 40-μl volume. The reaction mixture contained 1.5 mM MgCl2, 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 0.001% (wt/vol) gelatin, 200 μM of each deoxynucleotide triphosphate, 2 U of AmpliTaq DNA polymerase (PE Applied Biosystems, Weiterstadt, Germany), and 20 pmol of each primer. PCR amplification with C97-20 and H3A-20 was performed under the following conditions: 5 min of preincubation at 90°C, followed by 45 cycles of 30 seconds at 95°C, 30 seconds at 65°C, and 90 seconds at 72°C. A final elongation step was performed at 72°C for 10 min. Ten-microliter aliquots of the PCR mixture were analyzed on an ethidium bromide-stained 1% agarose gel. The detection limit of the PCR method is 0.1 pg of Helicobacter DNA (2).
Positive samples were further analyzed using DNA sequencing and RFLP analysis to identify the individual Helicobacter species. DNA sequencing and consecutive biocomputing were performed as previously described (1, 3). RFLP analysis was performed using restriction enzymes BfaI and BstUI. In all samples that were tested negative, a second PCR amplification with the same sample after spiking it with H. pylori-derived DNA was performed in order to rule out false-negative results due to PCR inhibitors.
Several experiments were carried out to evaluate the spread of EHS within the animal facility. Therefore, Helicobacter-negative C57BL/6 mice were exposed to Helicobacter-infected mice in various caging systems: study 1 employed 20 mice that were reared in individually ventilated cages. Neighboring mice in the same rack were Helicobacter infected. Cages of uninfected and infected mice were changed in the same laminar flow hood. After 12 months the surviving mice as well as a random sample of the offspring were tested for Helicobacteraceae. In study 2, 31 pairs of uninfected and infected mice were examined. The mice in this study were maintained as in study 1 until after 2 weeks the uninfected animals were tested for Helicobacter infection. In study 3 and study 4, a total of 26 mice were used. These uninfected mice were taken from individually ventilated cages to open-air cages, and 12 mice were placed in the same rack next to infected mice in open-air cages (rack 1); the remaining 14 mice were also placed next to infected mice in open-air cages, but in a different rack (rack 2). After 2 months the animals were tested for Helicobacter infections. In study 5, 26 mice were placed in the same rack next to infected mice in open-air cages as was done in study 3 and study 4; however, the cages were equipped with a polycarbonate filter top. After 2 months these animals were tested for Helicobacter infections.
Almost nothing is known about the prevalence and transmission of EHS in mouse colonies. In order to evaluate the prevalence of Helicobacter infections in mouse strains harbored in the SPF facility of our university, 40 mouse lines permanently living in nine colony rooms were tested using a group-specific PCR that is able to detect all Helicobacter species currently known (2). As Table Table11 indicates, Helicobacter species were detected in 87.5% (35 of 40) of the mouse strains that were tested. Direct sequencing combined with RFLP analysis of PCR amplicons revealed that 27 mouse strains carried a single Helicobacter species, while eight mouse strains were infected by at least two different Helicobacter species. Three out of these eight coinfected animals carried a mixture of H. ganmani and H. hepaticus. In five other mouse strains the species mixture could not be resolved by direct sequencing and RFLP analysis. Analysis of the monoinfected mouse strains revealed H. ganmani (n = 15), the Helicobacter species represented by isolate MIT 98-5357 (n = 6), H. hepaticus (n = 3), the Helicobacter species represented by isolate hamster B (n = 2), and H. typhlonicus (n = 1). All 16S rRNA gene sequences in this study were 100% homologous to the published sequences of the 16S rRNA genes of the corresponding EHS.
Generally, introduction of new animals to the facility is associated with a risk of infection. Entry criteria for new mice into our SPF facility were a health report with no abnormality detected within the last 3 months; otherwise, embryo transfer was carried out before introducing the new mouse strain into the SPF tract. We hypothesized that despite these safety precautions Helicobacter infections could be introduced into the SPF tract together with new animals.
In order to test this hypothesis, five mouse strains from different commercial breeding facilities were tested with respect to Helicobacter infections directly upon purchase (Table (Table2).2). According to the accompanying health-monitoring reports, four of these five mouse strains were certified as Helicobacter free and one mouse strain was reported to be Helicobacter positive. In three mouse strains which were certified as Helicobacter negative the absence of Helicobacter species was confirmed. However, in one mouse strain that was outlined as Helicobacter free, H. hepaticus was detectable in stool samples of 10 out of 14 mice. In the fifth mouse strain, where the health report showed positive test results for H. bilis, H. hepaticus, and Helicobacter species, we detected a coinfection of H. hepaticus and H. ganmani. H. bilis could not be detected in our PCR assay. The results show that mice from commercial breeding facilities may carry occult Helicobacter infections and therefore have to be considered the starting point of an EHS endemia.
To investigate whether Helicobacter infections can spread within the animal facility, we performed the following experiments (see also Table Table3).3). In study 1, 20 Helicobacter-free mice were reared and maintained exclusively in individually ventilated cages. Among these 20 mice, 11 mice were still alive after an observation period of 12 months. These 11 mice as well as all of the tested offspring (n = 35) remained Helicobacter negative, although the same laminar flow hood was used to change cages of uninfected and infected mice and infected animals were harbored in the same rack in neighboring cages. Thus, mice harbored in individually ventilated cages were protected against transmission of EHS from neighboring mice. In contrast, study 2 showed that when noninfected animals were put in the same individually ventilated cage together with Helicobacter-infected animals, Helicobacter infection was transmitted to the previously uninfected animals in 100% of the cases (n = 31) within 14 days.
In study 3 and study 4, we investigated whether transmission of EHS between neighboring open-air cages might also be possible. Therefore, we transferred Helicobacter-free mice from individually ventilated cages into open-air cages and moved them to another colony room, where the cages were placed in two different racks next to cages with Helicobacter-infected mice. After 2 months, five out of 12 mice in study 3 and three out of 14 mice in study 4 were infected with Helicobacter species. All EHS-positive mice from study 3 which were placed in rack 1 became infected with H. typhlonicus, while mice from study 4 which were placed in rack 2 became infected with H. hepaticus.
In study 5, uninfected mice (n = 26) were also placed next to infected mice in open-air cages; however, the cages of the uninfected mice had been equipped with a polycarbonate filter top. All of the mice in filter-top-protected cages remained Helicobacter free for at least 2 months.
There is little information concerning Helicobacter status in noncommercial animal facilities. Data from Japan showed that 27 out of 47 (57.5%) sentinel mice obtained during 1998 and 1999 from universities, breeding companies, pharmaceutical companies, and national research institutions were contaminated with Helicobacter species (13). A Korean study revealed Helicobacter infection in only one out of 24 mouse strains that were housed in SPF facilities (29). A European study reported a prevalence of Helicobacter infection of 88.9% (eight of nine strains) in mice (25). Here, we found a comparable high prevalence rate of 87.5% in mice reared in different caging systems in an SPF facility.
Interestingly, none of the mice in this study was infected by H. bilis and only six animals harbored H. hepaticus. In earlier studies H. bilis and H. hepaticus were identified as the most common Helicobacter species in laboratory mice (27, 30). This observation led to the addition of H. bilis- and H. hepaticus-specific PCR assays to the recommended screening protocol for SPF rodents (25). The relatively low prevalence of both bacteria is presumably due to the use of these screening procedures.
The most frequent Helicobacter species that we found in this study were H. ganmani and the Helicobacter species represented by isolate MIT 98-5357. H. ganmani and the Helicobacter species represented by isolate MIT 98-5357, as well as the Helicobacter species represented by isolate hamster B, H. typhlonicus, and the other Helicobacter species we detected, are widely uncharacterized and thus no species-specific PCR assays are available for health screening of mice. Therefore, these rather uncommon Helicobacter species may be missed by the routine procedures used for health screening of SPF mice and therefore explain the relatively high prevalence of these rare species in Helicobacter-infected mice. Future studies will show whether this is a general trend or specific for the SPF facility tested. Moreover, our study clearly shows that the Helicobacter species represented by isolate hamster B, which was recently detected in Syrian hamsters (35), is also able to infect mice.
In order to avoid the spread of Helicobacter infections in an animal facility, it was important to elucidate the transmission route. Recently, it was shown that H. hepaticus can be transmitted by contaminated bedding (19). Therefore, accidental bedding contamination could be a possible reason for Helicobacter transmission. However, this possibility was considered unlikely based on the high hygiene standard in our facility. Thus, it was necessary to examine the possibility of alternative transmission routes.
One important observation is that we identified mice from commercial breeding facilities as potential infection sources. One mouse strain of a commercial vendor carried a Helicobacter infection, although the accompanying health report attested that the animals were Helicobacter negative as tested by PCR (Table (Table2).2). Since animals from commercial SPF facilities are usually not taken into quarantine rooms, unrecognized infections in these animals can introduce Helicobacter infections into the animal facility. It is therefore advisable to keep animals from commercial vendors in quarantine rooms until the health status of every shipment is confirmed.
As study 1 showed, Helicobacter infections can be effectively prevented by harboring mice in individually ventilated cages even if infected mice were reared in neighboring cages and if the same laminar flow hood was used to change the cages of uninfected and infected animals. However, since individually ventilated cages are expensive, immunocompetent wild-type mice are frequently harbored in open-air cages. Our studies 3 and 4 showed that mice reared in open-air cages were very likely to get infected, but for a certain time period not all mice are infected. This fact is important because Helicobacter infections in wild-type mice often lack clinical signs (39). Therefore, it is not possible to select diseased sentinel mice for Helicobacter screening. By selecting sentinel mice at random, the chance of detecting EHS-infected animals is decreased and as a consequence Helicobacter-infected colonies may remain unrecognized. If such wild-type animals with unrecognized Helicobacter infections are then used as sentinels or for backcross purposes, these animals are transferred into individually ventilated cages together with Helicobacter-free knockout animals. As study 2 indicated, cage mates then get infected within 2 weeks in 100% of the cases.
We have often observed that the mice in open-air cages distributed stool pellets and contaminated bedding around their cages onto the floor and into neighboring cages. This can be a significant amount of material, which is easy to observe by inspecting the floor next to racks with open-air cages after a weekend if the floor is not cleaned every day. Mice are coprophages. The high rate of Helicobacter transmission between neighboring open-air cages suggests that stool pellets which were transferred in this way between cages are ingested by neighboring mice, which in turn get infected. As study 5 showed, the use of filter tops on conventional cages to prevent the spread of stool and contaminated bedding to neighboring cages was sufficient to block the transmission of EHS. These findings strongly support a fecal-oral spread of EHS in mice.
Taken together, infections with EHS can be introduced into an SPF facility and in turn may spread among animals. In order to maintain SPF animals free of Helicobacter infections, we recommend the following measures: use of a group-specific PCR assay to test animals regularly for Helicobacteraceae; keeping mice which are introduced into the facility in individually ventilated cages until the Helicobacter status of every shipment is confirmed and, whenever feasible, keeping mice in individually ventilated cages or cages with polycarbonate filter tops; and checking mice which are placed as sentinels or mice used for backcrossing experiments to ensure that they are Helicobacter negative.