Introduction

Helicobacter spp. are gram-negative spiral to curved rod-shaped bacteria spread via the fecal-oral route that colonize the gastrointestinal tracts of multiple species.^11,29 Helicobacter species that infect mice include H. bilis, ganmani, hepaticus, muridarum, mastomyrinus, rappini, rodentium, and typhlonius.³⁰ Helicobacter spp. infections are used experimentally in mouse models to mimic human inflammatory bowel disease (IBD) and can also cause IBD in natural infections of various mouse strains.^10,23,28 Helicobacter spp. infections in immunocompetent mice rarely cause overt disease but can confound research through effects on the immune and reproductive systems and impacts on tumor growth.^{3,7,13,14,16,21,24-27} Infections in immunocompromised mouse strains can lead to colitis, typhlitis, rectal prolapse, and diarrhea.^{10,12,19,28,29} Due to the increased use of immunocompromised mouse strains, which may experience clinical disease, as well as confounding effects on research, many research institutions have initiated efforts to exclude Helicobacter spp. from their facilities.⁵ This has resulted in a need for quick and accurate detection of Helicobacter spp. infections in mouse colonies. Traditional surveillance for Helicobacter spp. in laboratory mouse colonies has been achieved through PCR on fecal pellets or serology of live sentinel mice exposed to dirty bedding.^18,30 More recently, PCR on in-rack or in-cage filters exposed to air or dirty bedding from colony animals has been validated for Helicobacter spp. detection.^15,18,20 These types of environmental health monitoring (EHM) have eliminated or reduced the need for live animal sentinels, which allows for replacement of animals with nonanimal alternatives in accordance with the 3Rs of humane animal research.²²

Previous work has validated PCR on in-cage filter media as a viable method to screen mouse colonies for Helicobacter spp.^15,18,20 However, there are no published studies investigating the effect of time of testing after exposure to Helicobacter spp.-positive bedding on the detection rate of this test. For the purposes of this paper, detection rate is defined as the number of cages positive via PCR divided by the total number of positive cages. At our institution, bedding from each cage on a rack is added to a sentinel cage containing in-cage filter media at each change out, which occurs every 3 wk. The filters are tested quarterly. We hypothesized that filters could become saturated with non-Helicobacter spp. nucleic acids throughout the quarter. This would theoretically limit the ability of Helicobacter spp. nucleic acids to bind to the filter media and subsequently decrease the PCR detection rate. If Helicobacter spp.-positive bedding had been added late in the quarter, PCR would have decreased detection rate. In this study, we investigated the effect of time of adding Helicobacter spp.-positive bedding on PCR detection rate. Positive bedding was added at 3 different time points: 3 wk (group A), 6 wk (group B), or 9 wk (group C). Helicobacter spp.-negative dirty bedding was added at all other time points, and filter media from each cage were tested at 12 wk. We hypothesized that group C would have a decreased detection rate of Helicobacter spp. PCR compared with groups A and B.

In addition, many of the previously published studies investigated detection rate of PCR on in-cage filter paper in populations with high prevalence or endemic Helicobacter spp. infection.^8,15,20 There is little evidence regarding detection rate of the test in a low prevalence population. The 1:9 Helicobacter spp. positive-to-negative bedding ratio used in this experiment results in dilution of Helicobacter spp. nucleic acids and likely leads to the filter paper being exposed to fewer Helicobacter spp. copy numbers compared with studies using 100% positive bedding. This could represent scenarios such as low prevalence of Helicobacter spp. infection on a rack, dirty bedding added from the rack containing low copy numbers due to sample selection (that is, sample not containing fecal pellets) or other factors leading to dilution of nucleic acids. At our institution, racks in rooms with active experimental manipulation of animals may contain multiple principal investigators, each with an average of one row of cages on the rack. Therefore, we added approximately 9 tablespoons of positive bedding and 71 tablespoons of negative bedding to each sentinel cage to represent approximately one row of positive cages per side of an animal housing rack. By adding Helicobacter spp.-positive bedding at 3-, 6-, and 9-wk time points, we sought to determine if we could detect Helicobacter spp.-positive cages added to a rack at different time points in relation to the time of filter media collection for PCR. This could represent scenarios such as positive animals housed on a rack for a short period of time, acute use of positive animals, or positive animals added to a rack shortly before filter paper collection, in the case of the 9-wk time point.

The present study seeks to determine the effect of time of testing on Helicobacter spp. PCR on in-cage filter paper exposed to low levels of Helicobacter spp. nucleic acids, which are both significant knowledge gaps in the field of EHM. Addressing this lack of information could improve overall animal welfare by ameliorating Helicobacter spp. screening practices, which can be pathogenic in immunocompromised mice. Finally, this study has the potential to increase the reproducibility of research through improved detection of a significant confounding factor, subclinical Helicobacter spp. infection in mice.

Materials and Methods

Soiled bedding collection and cage handling procedures.

The cages in this project were bedded and handled in accordance with the University of Oklahoma Health Sciences Center standard operating procedure for sentinel health monitoring. In our facility, PCR monitoring for Helicobacter spp. and other pathogens is done on a quarterly basis using filter media (Allentown, Allentown, PA) placed in sentinel cages with approximately one tablespoon of dirty bedding (Biofresh; Animal Specialties and Provisions, Quakertown, PA) added from each cage on the rack at each cage change (Figure 1). Per institutional standard operating procedures, cage change outs occur every 3 wk.

View Figure

• Download as Powerpoint
• Download Figure

For this project, dirty bedding from mouse IVC in a known Helicobacter spp.-positive room and dirty bedding from mouse IVCs in a Helicobacter spp.-negative barrier facility were collected. Both facilities housed a variety of mouse strains, sexes, and ages on a variety of experimental and breeding protocols. In rooms housing animals belonging to multiple investigators, each investigator may have a small number of cages, such as one row on a rack. Fully occupied IVC racks in our facility contain 8 to 10 cages per row, with an average of 80 cage spots per side of the rack. The median number of cages per row is 9. To determine if screening would be able to detect one positive row of cages on a rack, 9 tablespoons of Helicobacter-positive bedding and 71 tablespoons of negative bedding were added at the 3-, 6-, or 9-wk time points to 14 empty disposable cages (Allentown, Allentown, PA) per time point (Figure 2). This approximates 9 Helicobacter spp.-positive cages and 71 Helicobacter spp.-negative cages on a rack. Negative bedding was added every 3 wk to cages not scheduled to receive positive bedding. Fourteen negative control cages received 80 tablespoons of negative bedding and 14 positive control cages received 80 tablespoons of positive bedding at each time point.

View Figure

• Download as Powerpoint
• Download Figure

In accordance with institutional standard operating procedures, at each time point, bedding from the previous time point was dumped and any remaining dust was removed using a Kimtech Science Kimwipes (Kimberly-Clark, Roswell, GA). After dirty bedding was added, cages were thoroughly mixed using a plastic knife that was sanitized between cages with Peroxigard (Oakville, ON). Cages were then shaken 3 to 5 times with the lids closed to ensure the in-cage filter media were completely exposed to bedding. Cages were maintained on an IVC rack (Allentown, Allentown, PA) for the duration of the study. To minimize the risk of cross-contamination, cages containing only negative bedding were handled before cages with positive bedding. At 12 wk, filters were removed from the cages and submitted to a third-party testing company (IDEXX BioAnalytics, Columbia, MO) for PCR testing for Helicobacter spp.

Results

Of the 14 cages in group A, 2 were positive for Helicobacter spp. on PCR at 12 wk. Two cages out of 14 in group B tested positive and 3 cages in group C tested positive. All 14 positive control cages tested positive, and all 14 negative control cages tested negative (Figure 3). The total genus-wide Helicobacter spp. copy numbers detected for positive and negative control groups were 67,657 and 0, respectively (Table 1). The total genus-wide Helicobacter spp. copy number detected for group A was 824, for group B was 374, and for group C was 545 (Table 1). Helicobacter typhlonius, H. mastomyrinus, H. bilis, H. hepaticus, and H. ganmani were detected in one or more samples and H. rodentium was not detected in any of the samples (Table 1). Overall detection rate of Helicobacter spp. PCR on group A, B, and C cages was 16.7% (7/42) (Table 2). The detection rate for groups A and B was 14.3% (2/14) (Table 2). The detection rate for group C was 21.4% (3/14) (Table 2).

View Figure

• Download as Powerpoint
• Download Figure

There was no statistically significant difference between the number of positive and negative cages between groups A and B, B and C, or A and C (P = 1.0 for all comparisons). This suggests that the length of time (3, 6, or 9 wk) between introduction of infected bedding and PCR testing does not impact the detection rate of Helicobacter spp. PCR on in-cage filter media. There was no correlation between detection rate and Helicobacter spp. PCR copy number (r = −0.14, P = 0.9) for groups A, B, or C.

Discussion

Because the rate of cage positivity is not correlated with PCR copy number (r = −0.14, P = 0.9), this suggests that there was no relationship between time point of exposure to Helicobacter spp.-positive bedding (3, 6, or 9 wk) and copy number. It is therefore unlikely that Helicobacter spp.-positive bedding added later in the quarter would have a higher risk of going undetected compared with positive bedding added at earlier time points.

Overall detection rate among the 3 experimental groups was 16.7% and group A, B, and C detection rates were 14.3%, 14.3%, and 21.4%, respectively (Table 2). Previous studies on detection rates of Helicobacter spp. PCR on in-cage filter paper have used higher amounts of positive bedding, which resulted in 100% detection of Helicobacter spp.^15,20 The results of these previous studies are consistent with the 100% detection rate in the control group in the present study. An approximately 1:9 ratio of Helicobacter spp. positive-to-negative bedding was used for groups A, B, and C in the present study, whereas one previous study used a 1:4 ratio of dirty Helicobacter spp.-positive bedding to unused bedding²⁰ and another study used only positive bedding.¹⁵ These results suggest that situations such as low prevalence of Helicobacter spp.-infected mice, infected mice present on a rack for short time periods, or addition of infected mice to a rack shortly (≤3 wk) before filter paper collection may result in decreased detection rates of Helicobacter spp. using PCR on in-cage filter paper.

To combat the low detection rate of Helicobacter spp. PCR on in-cage filter paper in populations with low prevalence or other sources of dilution of Helicobacter spp. nucleic acids, a multifaceted approach should be used for Helicobacter spp. detection in mouse colonies.⁶ Combining a direct colony sampling method, such as fecal pellet Helicobacter spp. PCR, with in-cage filter paper or another EHM method, has the potential to increase Helicobacter spp. detection rates. PCR on fecal pellets has 100% diagnostic sensitivity and feces can be noninvasively collected from animals or animal cages.² However, relying solely on direct testing for colony health monitoring has several drawbacks, including increased staff labor and an increased risk of missing positive animals with random sampling tactics.

One limitation of the present study was the inclusion of bedding from cages with food-grinding behavior. Food grinding in mice is a behavior in which the mice fragment feed into small particles, often referred to as “orts,” which then accumulate in the cage bottom.^4,17 Accumulation of orts results in an increase in particulate matter inside the cage. If in-cage filter media are exposed to bedding containing orts, the orts could potentially coat the filter media and prevent other nucleic acids from adhering to the filter, possibly resulting in a decreased PCR detection rate. In addition, the presence of orts in dirty bedding could further dilute the Helicobacter spp. nucleic acid needed for positive detection. Future work is necessary to elucidate the effects of food grinding on the detection rate of Helicobacter spp. PCR.

In addition, the present study was limited to the evaluation of Helicobacter spp. PCR on in-cage filter paper in disposable cages. Caging type may influence the PCR detection rate for pathogen detection. Different forms of EHM, such as swabs and exhaust dust testing, have also been shown to result in varying detection rates^1,8,18 As a result, the detection rate of Helicobacter spp. PCR may vary depending on the caging type and collection methods used. When designing a pathogen surveillance program, animal facility personnel may consider validating their testing methods internally, as a methodology from the literature may not correlate with an individual facility’s caging and testing methods and could significantly affect testing results.

In conclusion, our study revealed low detection rates of Helicobacter spp. using PCR on in-cage filter paper when the filter paper was exposed to low concentration of Helicobacter spp. infected dirty bedding at the 3-, 6-, and 9-wk time points. Factors contributing to the decreased detection rate could include a lower Helicobacter spp. positive bedding-to-negative bedding ratio, short (3 wk) duration of filter paper exposure to infected bedding, presence of food grinding, and caging and collection media types. Multimodal EHM programs may help increase overall Helicobacter spp. detection rates. Further studies are needed to determine minimum duration of Helicobacter spp. infected bedding exposure to in-cage filter paper, minimum detectable concentration of infected bedding, and threshold of nucleic acid adsorption by the filter media to fully understand the potential limitations of the use of this EHM method to detect Helicobacter spp. in research facilities.