Soiled-bedding Sentinel Detection of Murine Norovirus 4

Abstract

According to serologic surveys, murine norovirus (MNV) is the most prevalent viral pathogen infecting mice used in biomedical research. However, the use of sentinel mice to detect MNV-infected mouse populations has not been evaluated thoroughly. To this end, an experimental method of soiled bedding transfer was created to mimic a quarterly sentinel monitoring program. Soiled bedding (15 or 30 cm³) from ICR mice experimentally infected with MNV4 was transferred weekly to cages of pair-housed 4-wk-old ICR mice. After 12 wk, both mice in 80% (4 of 5) of cages receiving either 15 or 30 cm³ of soiled bedding were seropositive for MNV and were shedding virus in feces. To evaluate the stability of MNV RNA in mouse feces, fecal pellets from MNV-infected sentinel mice were stored at room temperature for as long as 14 d. After storage, all fecal samples tested positive for MNV by RT-PCR. To determine whether fecal samples could be pooled for MNV detection, 1 MNV-positive fecal pellet was combined with either 9 or 19 MNV-negative fecal pellets. All pooled fecal samples were positive for MNV by RT-PCR at both dilutions. These data indicate that although MNV-infected mouse populations can be detected by exposing sentinel mice to MNV-contaminated bedding, detection failures can occur. In addition, there was high agreement in the MNV infection status of cohoused sentinel mice. These data also demonstrate that MNV is readily detectable in pooled fecal samples and in mouse feces stored at room temperature for 2 wk.

Since the discovery of murine norovirus (MNV) in 2003⁸ and the recent development of highly sensitive and specific serologic and molecular diagnostic assays for the detection of infected mice,⁷ MNV has been identified as the most prevalent viral infectious agent among laboratory mice.^7,10,17 A 2006 report from our laboratory described 3 novel field strains of MNV (MNV2, MNV3, and MNV4), which differ markedly from the prototypic MNV1 strain by causing persistent infection and prolonged fecal shedding in immunocompetent mice.⁶ These properties make MNV an ideal agent for potential detection by soiled bedding sentinels. MNV infection can result in clinical signs of weight loss, hunched posture, and ruffled fur and even death in mice with deficiencies in aspects of the innate immune system;^8,15,16 however, no clinical signs have been reported to occur in immunocompetent mice or immunocompromised mice with intact innate immunity. Although cellular tropisms and histologic lesions have been described in mice infected with MNV,^8,15,16 reports are now starting to emerge addressing the extent of the effect of MNV on research results.⁴

Detection of MNV-infected mice is crucial for studies in which associated disease and pathologic lesions may complicate data interpretation, and is fundamental in maintaining laboratory mouse colonies free of MNV. Although the use of sentinel mice for detecting MNV-infected mouse populations has been reported,^12,15 the effectiveness of this practice has not been evaluated. Sentinel monitoring programs often combine the use of sentinel exposure to soiled bedding, contact sentinels, and direct sampling of colony animals. Monitoring of sentinels exposed to soiled bedding often plays a primary role in mouse colony health monitoring programs because it provides an economical means of monitoring large, dynamic populations of mice with varying concurrent health statuses.

This study evaluated the effectiveness of testing sentinel mice exposed to soiled bedding for detection of MNV infection in laboratory mouse colonies. Over a 12-wk period, outbred sentinel mice were tested every 2 wk for antibodies to MNV by use of a MNV multiplex fluorescent immunoassay (MFI) and for the presence of MNV in excreted fecal pellets by MNV RT-PCR assay. In addition, stored fecal samples and pooled fecal samples were evaluated to access viral RNA stability within feces and the ability to combine fecal pellets for diagnostic testing.

Materials and Methods

Propagation of MNV4.

MNV4 cultures were generated as described previously.⁶ Briefly, MNV4-inoculated RAW 264.7 cells were cultured in suspension spinner flasks (MagnaFlex flasks, Wheaton Science Products, Millville, NJ) with DMEM (HyClone, Logan UT) supplemented with 10% low-endotoxin fetal bovine serum (Cambrex, East Rutherford, NJ), 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, and 10 μg/ml ciprofloxacin and incubated at 37 °C with 5% CO₂.

Mice.

Groups of female Hsd:ICR(CD1) mice (Harlan Sprague-Dawley, Indianapolis, IN) were obtained that were documented to be free of ectoparasites, endoparasites, Mycoplasma pulmonis, Helicobacter spp., known enteric and respiratory bacterial pathogens, and antibodies to mouse hepatitis virus, Sendai virus, pneumonia virus of mice, reovirus 3, Theiler murine encephalomyelitis virus, Ectromelia virus, polyoma virus, lymphocytic choriomeningitis virus, mouse adenovirus, minute virus of mice, mouse parvovirus, mouse rotavirus, mouse cytomegalovirus, mouse thymic virus, murine norovirus, Encephalitozoon cuniculi, and Clostridium piliforme. For studies conducted, all mice were housed in auto claved polysulfone filter-top static isolation caging (Alternative Design Manufacturing and Supply, Siloam Springs, AR) on corncob bedding (Bed-o'Cobs, The Andersons, Maumee, OH) with nesting material (Nestlet, Ancore, Bellmore, NY). Mice were provided free access to autoclaved, acidified water and irradiated chow (5053 PicoLab Rodent Diet 20, Purina LabDiet, St Louis, MO) with a 7-d cage-change cycle. Cage size, animal density, and corncob bedding volume varied depending on the experimental procedure. The macroenvironment of the animal holding cubicle was maintained at a room temperature between 21.1 to 23.9 °C with 30% to 70% humidity with at least 12 fresh-air changes per hour and a controlled 14:10-h light:dark cycle. All mouse manipulations were performed in a class II biological safety cabinet. All animal studies were approved by the University of Missouri Animal Care and Use Committee.

Sample collection.

Whole blood samples were collected from the lateral saphenous vein every 2 wk from experimental groups, inoculated mice, and control mice. Serum was separated from whole blood by centrifugation, and frozen at –20 °C until evaluated. Pooled fecal samples were collected from the cage bedding, gathered at the end of the 7-d cage change cycle. Single fecal pellets were collected directly from the anus and were stored at –80 °C until processing.

Bedding production.

For the production of MNV-contaminated bedding, 4-wk-old female ICR mice (n = 10) were inoculated by oral gavage with 1 × 10⁶ pfu of MNV4 in RAW 264.7 cell lysate in a volume of 200 μl. MNV-inoculated mice were group-housed in a single autoclaved 10.25 × 18.75 × 6-in. (147 in²) polysulfone filter-top static isolation cage filled with 2000 cm³ of corncob bedding. Weekly, MNV-inoculated mice were transferred to clean autoclaved cages, and the MNV-contaminated bedding from the dirty cage was used for experiments. Every 2 wk, a pool of 5 to 10 fecal pellets from the MNV-contaminated bedding was tested for MNV by RT-PCR.

For the production of MNV-free soiled bedding, 5 cages of 4-wk-old MNV-negative control ICR mice were pair-housed in autoclaved 7.25 × 11.5 × 5-in. (65 in²) polysulfone filter-top static isolation cages filled with 1000 cm³ of corncob bedding. Weekly, MNV-negative control mice were transferred to clean autoclaved cages, and the MNV-free soiled bedding from the soiled cages was divided and used for experiments. To confirm the MNV-negative status of control mice, whole blood and fecal samples were collected from each pair-housed mouse at the initiation of the study and every 2 wk thereafter until conclusion of the study.

Study of effect of age on MNV kinetics.

Female mice (ages, 4 wk [n = 10] and 8 wk [n = 11]) were segregated by age and group-housed at either 10 or 11 mice per cage in 10.25 × 18.75 × 6-in. (147 in²) autoclaved polysulfone filter-top static isolation cages filled with 2000 cm³ of clean corncob bedding. Weekly for 12 wk, 50 cm³ of MNV-contaminated bedding was transferred into each cage in coordination with receiving a cage change containing clean corncob bedding. To monitor for seroconversion to MNV and fecal shedding of MNV, whole blood and fecal samples were collected every 2 wk for the duration of the 12-wk study.

Soiled bedding sentinel study.

Four-week-old ICR sentinel mice were pair-housed in 7.25 × 11.5 × 5-in. (65 in²) polysulfone filter-top static isolation cages (Figure 1). Weekly, mice were transferred to a new cage that contained 15 cm³ (low dose) or 30 cm³ (high dose) of MNV-contaminated soiled bedding mixed with 500 cm³ of MNV-free soiled bedding (n = 5 cages per group). Whole blood and fecal samples were collected from each sentinel mouse every 2 wk for the duration of the 12-wk study.

Figure 1.

Schematic representation of bedding volume distribution for soiled-bedding sentinel detection of MNV. Solid arrows represent the transfer of MNV-contaminated soiled bedding. Dashed arrows represent the transfer of MNV-free soiled bedding. Large and small cubes represent 10.25 × 18.75 × 6-in. (147 in²) and 7.25 × 11.5 × 5-in. (65 in²) polysulfone filter-top static isolation cages, respectively.

MNV serology.

Sera were tested for antibodies to MNV by MFI as described previously⁷ with minor modifications. Modifications included using MNV4 whole virus as the MFI antigen, 5 μl of serum diluted 1:5 in sterile PBS, and a change in the threshold values used to interpret MFI data. Sera with MFI values of less than 550 were classified as negative for MNV antibodies, and those with MFI values greater than 1120 were classified as positive for MNV. These threshold values were previously established by use of a receiver operating characteristic curve to provide a 98% test sensitivity and specificity (data not shown). Sera with MFI values from 550 to 1120 were classified as intermediate and of equivocal MNV status; these sera were further tested by an MNV indirect fluorescent antibody assay, and classified as positive or negative for MNV based on the outcome of this test as previously described.⁷

RNA extraction and fecal RT-PCR.

RNA was extracted from fecal homogenates (MagAttract RNA Tissue Mini M48 Kit, Qiagen, Valencia, CA) according to the manufacturer's recommended protocol and as described previously.⁷ Briefly, approximately 20 mg of fecal material or that equivalent to 1 fecal pellet was homogenized in buffer RLT by using a 5-mm sterile stainless steel ball and a TissueLyser (Qiagen) at 30 Hz for 10 s. Fecal homogenates were centrifuged for 5 min at 300 × g. The supernatants were used for magnetic silica-based RNA purification on the BioRobot M48 Workstation (Qiagen). MNV RT-PCR was performed by using MNV primers as previously described.⁶

Stability of MNV RNA in feces.

To assess the stability of MNV RNA in feces for detection by RT-PCR, fecal pellets were collected from sentinel mice that were naturally infected by exposure to MNV-contaminated soiled bedding and had been shedding MNV in their feces for at least 8 wk. Individual fecal pellets were placed immediately into 3 ml sterile cell-culture tubes with venting caps (Fisher Scientific, Hampton, NH) by using sterile forceps. The tubes were stored at ambient temperature for 1, 3, 5, 7, 10 (n = 5 per time point), and 14 d (n = 15). At each time point, fecal pellets were stored at –80 °C until the completion of the 14-d study, when 5 pellets from each time point were evaluated individually by MNV RT-PCR.

Pooled fecal sample detection of MNV.

To determine whether pooled fecal samples can be used for routine diagnostic evaluation of feces for the presence of MNV, fecal pellets were collected from sentinel mice as stated previously. Immediately after collection, fecal sample pooling was conducted by combining a fecal pellet from an MNV-infected mouse with either 9 or 19 fecal pellets from MNV-free control mice, resulting in a 1:10 or 1:20 dilution ratio. Six replicates of each dilution were created and tested by RT-PCR. In addition, to assess whether stored fecal pellets can be pooled for MNV detection by RT-PCR, fecal pellets that had been held at room temperature for 14 d were diluted 1:10 with fecal pellets from MNV-free control mice and tested by RT-PCR. A total of 10 replicates were performed. Fecal samples were mixed manually in sterile 2-ml microcentrifuge tubes (Fisher Scientific, Hampton, NH) each containing 0.8 ml of buffer RLT (Qiagen) by using a sterile 1/8-in. diameter wooden dowel (Fisher Scientific) until homogeneous. Approximately 20 mg of the fecal homogenate, equivalent to 1 fecal pellet, then was transferred to a fresh 2-ml microcentrifuge tube, from which RNA was extracted as described previously.⁷

Statistical analysis.

The Fisher exact test was performed by using SigmaStat 3.1 software (Systat Software, Point Richmond, CA) to detect differences in MNV seroconversion or fecal shedding among experimental groups exposed to MNV-contaminated soiled bedding and unexposed control groups. The threshold for significance was a P value of less than 0.05 for all statistical analyses.

Results

Soiled bedding.

To confirm MNV contamination of soiled bedding used in experiments and persistent fecal shedding of MNV from inoculated mice, a pool of 5 to 10 fecal pellets were collected every 2 wk from the soiled bedding and tested for MNV by RT-PCR. All pooled fecal samples collected from the soiled bedding tested positive for MNV (data not shown). In addition, at the end of the 12-wk study, fecal pellets and serum were collected individually from MNV-inoculated mice, and all tested positive for MNV and antibodies to MNV, respectively (data not shown). From MNV-negative control mice, all fecal samples and serum collected at 2-wk intervals from each control mouse used to generate the MNV-free soiled bedding were tested and found to be negative for MNV and antibodies to MNV, respectively (data not shown). These data confirmed that MNV inoculated mice were infected and persistently shed virus throughout the 12-wk experiment and that control mice remained free of MNV infection.

Age-optimized horizontal transmission of MNV in ICR mice.

Four- and 8-wk-old ICR mice exposed weekly to 50 cm³ of MNV-contaminated bedding were used to assess whether mouse age influenced susceptibility to MNV infection. Mice were monitored every 2 wk for the development of antibodies to MNV and for the presence of MNV in the feces. At 2 wk after the start of soiled bedding exposure, 70% (7 of 10) of the 4-wk-old mice and 45% (5 of 11) of the 8-wk-old mice were positive for MNV serologically, with 100% (21 of 21) of both the 4-wk-old mice and the 8-wk-old mice being serologically positive after 4 and 6 wk, respectively (Figure 2 A). The magnitude of the seroreactivity of all mice, as determined by MFI, continued to increase until the end of the 12-wk study (data not shown).

Figure 2.

Percentage of 4- and 8-wk-old ICR mice testing positive for (A) antibodies to MNV by MFI and indirect fluorescent antibody assay and (B) MNV in feces by RT-PCR, after weekly exposure to 50 cm³ of MNV-contaminated bedding. The 2 age groups represent the ages of the mice when first exposed to MNV-contaminated bedding. All MNV-negative control mice tested at the same time points were fecal-negative for MNV and negative for antibodies to MNV.

Two weeks after the initial exposure to soiled bedding, 100% (10 of 10) of the 4-wk-old mice and 73% (8 of 11) of the 8-wk-old mice were fecal-positive for MNV. After 4 wk of exposure to soiled bedding and for the duration of the 12-wk study, all (21 of 21) 4- and 8-wk-old mice were fecal-positive for MNV (Figure 2 B). No significant difference was noted in the number of mice that seroconverted to MNV or shed MNV in their feces between the 4- and 8-wk-old groups at any time point. However, because 100% of the 4-wk-old mice had seroconverted and shed virus in the feces at the earlier time point, 4-wk-old mice were used as sentinels for the soiled bedding study.

Soiled bedding sentinel detection of MNV.

To mimic a quarterly (every 12 wk) sentinel monitoring program based on soiled bedding transfer, 4-wk-old ICR sentinel mice were pair-housed in filter-top static isolation caging. Weekly, mice were transferred to a new cage that contained 15 cm³ (low dose) or 30 cm³ (high dose) of MNV-contaminated soiled bedding mixed with 500 cm³ of MNV-free soiled bedding (n = 5 cages per group). Mice were monitored every 2 wk for MNV seroconversion and MNV fecal shedding. The earliest time point at which antibodies to MNV were detected in sentinel mice after the start of MNV-contaminated soiled bedding transfer was 2 wk, observed in 1 mouse from 1 cage in the low-dose sentinel group. From weeks 6 to 10, the low-dose sentinel group led the high-dose group in total number of seropositive cages by 1 cage (Figure 3 A) the termination of the study, both the high- and low-dose sentinel groups each had 80% (4 of 5) of sentinel mice cages serologically positive for MNV. With the exception of the earliest time point, if antibodies to MNV were detected in 1 sentinel mouse, the corresponding cage mate was also serologically positive.

Figure 3.

Number of cages of pair-housed 4-wk-old ICR mice testing positive for (A) antibodies to MNV and (B) MNV in feces by RT-PCR, after weekly exposure to MNV-contaminated soiled bedding. All MNV-negative control mice tested at the same time points were negative for MNV infection by serology and fecal RT-PCR. ^#, One half of a cage reported as positive, with 1 of 2 sentinel mice seropositive for MNV. *, statistically significant difference between sentinel cages and control cages at the same time point (P ≤ 0.05).

MNV fecal shedding in sentinel mice paralleled, but occurred earlier than, seroconversion. (Figure 3 B). MNV was detected in the feces of 20% (1 of 5) of sentinel cages from both the low- and high-dose groups at 2 wk after exposure. Initially MNV detection increased more rapidly in the low-dose sentinel group between 4 and 8 wk after exposure, and at the 12-wk time point, 80% (4 of 5) sentinel cages were positive for MNV by fecal RT-PCR. In all cases, if 1 sentinel mouse was shedding MNV in the feces, the cagemate sentinel also was shedding virus. At each time point, the number of sentinel cages testing positive for antibodies to MNV or viral shedding did not differ significantly between low- and high-dose groups. However, when compared with control mice, statistically significant differences in seroconversion were observed at 10 and 12 wk in the low-dose group and at 12 wk in the high-dose group. Similarly, when compared with control mice, statistically significant differences in fecal shedding occurred at 8, 10, and 12 wk in the low-dose group and at 10 and 12 wk in the high-dose group (P = 0.048 for all comparisons).

Stability of MNV RNA in feces.

To determine the stability of MNV RNA in feces for detection by RT-PCR, fecal samples collected from MNV-infected sentinel mice were placed in vented-top cell culture tubes immediately after collection and stored at ambient temperature for 1, 3, 5, 7, 10, and 14 d. The fecal samples from all time points (n = 5 replicates per time point) were positive for MNV by RT-PCR (data not shown), indicating that for diagnostic testing purposes, MNV RNA is stable in feces at room temperature for up to 2 wk.

MNV RNA detection in pooled feces.

To determine the feasibility of testing pooled fecal samples for MNV by RT-PCR, fecal samples collected from MNV-infected sentinel mice and MNV-free mice were combined in ratios of 1:10 and 1:20. Pooled samples were homogenized and tested for MNV by RT-PCR. All pooled fecal samples (n = 6 replicates per dilution) tested positive for MNV by RT-PCR (data not shown). In addition, when fecal pellets from MNV-infected mice were held for 14 d at ambient temperature and then diluted 1:10 with 9 fecal pellets from MNV-uninfected mice, all 10 replicates tested positive for MNV by RT-PCR (data not shown).

Discussion

Serologic data indicate that MNV is the most prevalent viral infection in research mice in North America.⁷ Because MNV-infected mice have the potential to alter research outcomes, many research institutions have begun monitoring mouse colonies for MNV infection.^4,8,12,15 In this study, we mimicked a quarterly sentinel monitoring program to evaluate the effectiveness of detecting MNV-infected mice by using outbred sentinels exposed to MNV4-contaminated bedding. MNV4 was used as a representative of MNV isolates that produce persistent infections in immunocompetent mice typical of the majority of MNV strains encountered in the field.^6,14

MNV was transmitted with equal efficiency to sentinel mice exposed to 15 or 30 cm³ of soiled bedding from MNV-infected mice, with 80% (4 of 5) of sentinel cages in both bedding dose groups becoming positive during the 12-wk study. However, 20% (1 of 5) of the sentinel cages exposed to either volume of contaminated bedding failed to detect MNV infection. Failure of soiled bedding sentinel monitoring to detect infectious agents is often ascribed to the biologic and physical characteristics of the agent, such as a short period of environmental shedding, intermittent shedding, or poor environmental stability. During the course of the 12-wk experiment, pooled fecal samples confirmed the presence of MNV in fecal pellets in the soiled bedding. Although intermittent fecal shedding of MNV by MNV-inoculated sentinel mice is possible, our data indicate the contrary. At no point during the study was a negative fecal pellet collected from an individual mouse that had previously shed virus in the feces, indicating persistent fecal shedding of MNV in infected mice. Further, long-term persistence of infectious MNV has been demonstrated at ambient temperature within fecal material.³ These properties suggest that MNV is an ideal candidate for detection by soiled bedding sentinel monitoring, yet failure of some sentinels to detect infected bedding did occur.

Dilution of the MNV-contaminated bedding may have contributed to the detection limitation of the sentinel cages that failed to detect infected bedding. In our study, to simulate a sentinel-monitoring scenario where the prevalence of MNV infection in a colony is low, the majority of the soiled bedding provided to the sentinel mice was free of MNV. Exposing sentinel mice to 15 or 30 cm³ of MNV-contaminated bedding diluted with 500 cm³ of MNV-free soiled bedding mimics the scenario of detecting MNV infection when only 1 of 34 or 1 of 18 cages contain MNV-infected mice, respectively. These volumes of contaminated bedding were selected because they represent a range of volumes used in soiled bedding sentinel monitoring systems reported by others^12,13 and used at our own institution. However, in our study, exposing sentinel mice to more MNV-contaminated bedding did not increase sentinel detection of MNV. Paradoxically, sentinel mice receiving 15 cm³ of MNV-contaminated soiled bedding began shedding MNV in their feces more rapidly than did those in cages receiving 30 cm³ from weeks 4 to 8. These results are counterintuitive, because one would expect that a larger amount of MNV-contaminated soiled bedding exposure would cause a more rapid rate of sentinel cage infection. A suggested reason for these unexpected findings is that although 2 standardized doses of bedding were transferred to each sentinel group, we did not enumerate fecal pellets in the 2 doses. Some transfers of dirty bedding may have contained variable numbers of fecal pellets, such that the low dose group actually received more mouse feces. Another potential explanation is that if the primary mode of transmission of MNV is though coprophagic ingestion of virus-infected feces, by random chance, mice in the low-dose group may have ingested MNV-contaminated feces at earlier time points than did mice in the high-dose group.

Although transmission of MNV to sentinel mouse cages exposed to MNV-contaminated bedding occurred at varying rates and did not occur for 20% of sentinel cages, at all time points examined, there was 100% agreement in the MNV fecal shedding status of pair-housed sentinel mice. At only 1 time point (week 2) did 1 sentinel mouse from a pair-housed sentinel cage differ in MNV seroconversion status. Possible explanations for this observation include simultaneous infection of sentinel mice or infection of 1 sentinel mouse followed by efficient intracage horizontal transmission to the cohoused sentinel. Failure of both pair-housed sentinels to become infected or seroconvert poses a considerable challenge in sentinel monitoring programs, as conflicting data can be obtained when both mice are tested. Moreover, if only 1 sentinel mouse of a pair is sampled for diagnostic testing and the second mouse retained only for confirmatory testing, infections within a colony may be missed. However, our data suggest that for MNV, a high percentage of sentinel pairs will match in both serologic and fecal shedding status within a 2-wk interval.

As previously reported, mice experimentally infected with MNV4 persistently shed virus in the feces for the duration of an 8-wk study.⁶ In a subset of MNV-infected sentinel mice retained at the end of our study, fecal shedding continued for 6 mo, at which time the mice were euthanized (data not shown). Because of the prolonged fecal shedding of the majority of field isolates of MNV, fecal RT-PCR provides a rapid and accurate ante-mortem means of detecting both acute and chronic MNV infections in mice in a quarterly sentinel monitoring program. Although 2 wk was the shortest interval evaluated in our study, data from all mice infected with MNV in this study indicate that fecal shedding preceded seroconversion by 2 wk, as anticipated from previous observations.¹⁷ In addition, seroconversion was achieved in all sentinel mice that shed MNV in feces. These findings are also consistent with a report using 4- to 6-wk-old Swiss Webster mice for soiled-bedding sentinel monitoring of MNV.¹²

The use of young sentinel mice can lead to enhanced detection of infectious agents. For example, 12-wk-old outbred ICR mice were less susceptible to infection and seroconversion to mouse parvovirus than were 4- and 8-wk-old ICR mice.² In our study, 100% of 4-wk-old mice seroconverted to MNV and shed MNV in their feces at the earlier time points, but this finding was not statistically different from that for 8-wk-old mice. Therefore, either 4- or 8-wk-old mice are acceptable for use in a quarterly soiled-bedding sentinel monitoring program for MNV detection.

To aid in fecal sample collection for diagnostic detection of MNV, we examined the ability to detect MNV by RT-PCR in feces maintained at room temperature as long as 14 d. The 14-d point was selected as the final time point as it represents an accepted maximal allowable time between cage sanitations.¹¹ Because MNV was readily detected in all fecal samples at all time points up to and including 14 d, this characteristic provides considerable latitude in the collection, storage, and transport of fecal samples for detection of MNV by RT-PCR. This finding also contradicts the general recommendation of storing and shipping samples for detection of RNA viruses at or below –80 °C to prevent rapid RNA degradation and false-negative test results.⁵

Fecal sample pooling for PCR detection of rodent Helicobacter species at dilutions of up to 1:10 are currently common practice at some diagnostics laboratories.¹ For MNV, our data indicate that RT-PCR can detect a single MNV-positive fecal pellet in a pool of as many as 20 mouse fecal pellets. However, in our experience, efficiently homogenizing 20 fecal pellets is difficult, potentially reducing test sensitivity or reproducibility. Therefore, we recommend pooling of no more than 10 fecal pellets for MNV RT-PCR detection. The high environmental stability of MNV RNA in feces and the ability to pool feces for MNV RT-PCR analysis reduce the stringency of sample storage and provide cost-effective means of screening large numbers of mice while continuing to provide accurate diagnostic test results.

This study provides the first data demonstrating the effectiveness of soiled-bedding sentinel detection of MNV-infected mice, which provides an important tool for management of MNV infections in mouse research colonies. These data also show that failure to detect MNV-infected mouse populations by using dirty bedding sentinels can occur. From these data, we suggest the use of young immunocompetent mice between 4 and 8 wk of age for soiled bedding sentinel detection of MNV. This suggestion harmonizes with current recommendations for sentinel detection of other mouse pathogens, such as mouse parvovirus.^2,9 Our findings also demonstrate consistent rates of MNV transmission to sentinel cages with 2 different doses of soiled bedding and high concordance of MNV infection among naïve pair-housed sentinel cage mates exposed to MNV-contaminated bedding. In addition, the ability to store fecal pellets at room temperature and pool fecal samples for MNV RT-PCR analysis will help reduce testing costs and diagnostic limitations on sample transport and storage.