Rodent Pathogen Detection via Testing of Soiled Nesting Material
Murine pathogens affect laboratory animal health and research outcomes, and the prevention of pathogen incursion or the elimination of pathogen outbreaks is paramount. To this end, sensitive methods for pathogen detection are continually being developed and improved. Environmental health monitoring has become a popular and sensitive method for pathogen detection. Published methods for environmental sampling include the collection and testing of exhaust air filters, exhaust air duct swabs, and swabs or filter media placement in empty cages with soiled bedding. Our study tested soiled, cotton nesting material (Nestlet™) in occupied cages for the detection of nucleic acid from certain, high-prevalence, murine pathogens. Nesting material from cages housing mice positive for mouse norovirus, Helicobacter spp., and Rodentibacter heylii consistently tested positive for these agents. In addition, nesting material from cages housing naïve mice to which soiled bedding from the infected cages was transferred tested positive for these agents more often than testing the mice directly. This study concluded that testing of particulate material (for example, dust) from soiled nesting material is a sensitive detection method for certain, high-prevalence murine pathogens.
Introduction
Historically, rodent health surveillance has been performed using soiled bedding sentinels (SBS). Sentinel rodents, not part of research colonies but housed in proximity, are exposed to feces and soiled bedding material at regular intervals from a specified population. These sentinels are then periodically tested for a serological response to certain pathogens, the presence of pathogen nucleic acids via PCR, the presence of parasites via direct exams, culture of anatomic sites or samples (for example, feces), and possibly histologic evaluation. However, the advent of enclosed individual rodent cages (static microisolation or IVC) improved biocontainment to the cage level and limited the organisms transmissible to SBS to primarily those shed in the feces.1,2
Environmental testing for infectious agents by PCR has become common including screening for SARS-CoV-2 and highly pathogenic avian influenza.3,4 Numerous publications have demonstrated the sensitivity of environmental health monitoring (EHM) in laboratory animal colonies via testing of soiled particulate (dust) material for murine pathogens compared with samples collected from SBS,1,5–18 and many institutions have chosen to eliminate SBS and only perform EHM. The terminology used for the various sampling locations or collection media is confusing, and a recent publication by LaFollette et al.19 attempts to standardize these terms. However, the definitions in that manuscript19 have not been debated in our field and have not been universally accepted.
Exhaust dust testing (EDT) from IVCs depends on the location of air filtration within the exhaust air flow and the exhaust air flow design.2,8 IVC racks without cage-level filtration and horizontal exhaust air return plenums (for example, those from Allentown and Lab Products) can usually be sampled at the end of each horizontal plenum, thereby limiting the number of cages being tested to that on a given row, or be sampled at a terminal vertical plenum, allowing whole rack collection. IVC racks without cage-level filtration and vertical exhaust air return plenums (for example, those from Tecniplast) do not easily allow testing of individual columns and must be sampled at the end of the rack’s return air flow thereby sampling material from all cages on the rack. Since IVC racks are typically not cleaned as often as the cages, the dust collected from these systems usually spans a much longer period of time, which would increase the likelihood of pathogens being present and detected. IVC systems with cage-level filtration (for example, those from Innovive and Thoren) or static microisolation cages can only be sampled at the cage level with or without SBS.
Sentinel-free soiled bedding (SFSB) approaches to EHM programs,19 in which soiled bedding from colony animals are transferred to an empty cage, have reported a myriad of particulate collection methods and materials, including air filter media or sticky/flocked swabs.1,5,6,19–23 Once collection materials have been added, the cage is then periodically shaken or stirred to create dust particles that are captured by the added collection media and subsequently tested. At NIEHS, we perform a hybrid testing program that includes SBS and EHM. We prefer to use a multimodal approach to health monitoring including serology, PCR, direct parasitic exams, bacterial/fungal culturing, and gross necropsies. In a paper published in January 2023,24 12 of 111 (12 of 111 [11%]) institutions surveyed reported using a sentinel-free, EHM program, and 45 of 111 (41%) reported using a combination of SBS and EHM, similar to ours.
Our primary rodent housing utilizes the Tecniplast IVC system (GM500 Sealsafe mouse cages; Tecniplast USA, West Chester, PA) that uses rack-level filtration, and we perform EDT from the rack’s exhaust plenum. We also have a small number of Innovive IVC racks that use disposable Innocage (Innovive, San Diego, CA) cages with cage-level filtration. With the Innovive cages, we previously collected the filter media from the cage lid as a source of EDT from SBS cages and found we could detect certain high-prevalence pathogens including MNV, Helicobacter spp., and Rodentibacter pneumotropicus/heylii (data not shown), which were 3 of the limited pathogens allowed in our remote housing facility. We had not housed animals with other pathogens, so we were never able to test the efficacy of this method with other pathogens. Given the many recent publications reporting the addition of particulate collection media or swabs to sentinel-free cages, we wanted to determine if addition of similar particulate collection materials to cages with SBS provides an alternate approach. In an initial assessment not described in detail here, we evaluated the filter media present in a commercial air dust filter, designed for placement within the air handling unit of the IVC system (Interceptor; Tecniplast, West Chester, PA) after removal from the cardboard carrier and placement directly into the cage with the SBS. The mice quickly shredded the filter material, and it could not be sufficiently collected for testing. As an alternative, we tested soiled, cotton nesting material (Nestlet; Ancare, Inc., Bellmore, NY), routinely used in all cages at NIEHS, for the presence of pathogen nucleic acid. Our results demonstrate that the soiled Nestlet material is an acceptable collection media that can be used for testing for MNV, rodent Helicobacter spp., and R. heylii. This form of EHM can potentially enhance a multimodal approach to rodent pathogen detection.
Materials and Methods
Ethical review.
All animal procedures were approved by the NIEHS Animal Care and Use Committee. NIEHS maintains PHS assurance and is registered with the USDA. NIEHS is fully accredited by AAALAC, International.
Experimental animals.
Female SW (Tac:SW) mice from our in-house, gnotobiotic breeding colony were used in the study. This SW colony had been housed in our isolators, colonized with Altered Schaedler Flora,25 and used as our SBS for approximately 9 y. The colony was tested quarterly via aerobic bacteriological cultures of feces and interior isolator surface swabs, and also underwent annual testing of colony mice via a combination of serology, fecal PCR (Helicobacter spp. and pinworms), anal test tape microscopic analysis for pinworms, fecal flotation, direct pelt exam, and microscopic examination of fur plucks for ectoparasites, OP swab culture for Rodentibacter spp., and gross necropsy. Immediately before the start of the study, mice from this colony had tested negative for the following pathogens via a combination of serology, PCR, and bacterial culture: enzootic diarrhea of infant mice virus (EDIM), ectromelia virus, Encephalitozoon cuniculi, Filobacter rodentium, Hantavirus, lactate dehydrogenase elevating virus, lymphocytic choriomeningitis virus, minute virus of mice, mouse adenovirus types 1 and 2, mouse cytomegalovirus, mouse hepatitis virus, mouse norovirus, mouse parvovirus types 1 through 5, mouse pneumotropic virus (K virus), mouse polyoma virus, mouse thymic virus, murine ectoparasites (Myocoptes musculinis, Myobia musculi, Radfordia affinis), murine Helicobacter spp., murine pinworms (Syphacia muris, Syphacia obvelata, Aspiculuris tetraptera), Mycoplasma pulmonis, pneumonia virus of mice, reovirus 3, Rodentibacter pneumotropicus/heylii, Sendai virus, and Theiler murine encephalomyelitis virus.
The study mice were removed from the isolator and housed (5/cage) in IVCs (Tecniplast USA; West Chester, PA) with autoclaved, hardwood bedding (Sani-Chip; PJ Murphy Forest Products, Montville, NJ) and enrichment (Nestlet; Ancare, Bellmore, NY; and Bed-r’Nest; The Andersons, Maumee, OH) and provided autoclaved NIH-31 diet26 (Zeigler Bros, Gardners, PA) and deionized, reverse-osmosis filtered water ad libitum. Housing conditions were consistent with the standards described in the Guide for the Care and Use of Laboratory Animals27 and maintained at 72 ± 2 °F (22 ± 1 °C), relative humidity between 30% and 70%, and a 12:12-h light:dark cycle. All mice were evaluated at least once daily by trained facility staff. During the study, mice were housed in Innovive IVCs and provided the same bedding, enrichment, feed, and water. Cages were changed every 14 d in both systems. The Tecniplast cages were changed in a laminar flow, animal change station (Model 9CS5ET4TYFL; Tecniplast USA, West Chester, PA). The Innovive study cages were all changed in a Class II, A2—Biologic Safety Cabinet (NuAire, Plymouth, MN).
Initially, 3 naïve, female, SW mice were cohoused in Innovive cages for 14 d with mice confirmed to be infected with mouse norovirus (MNV), Helicobacter typhlonius, Helicobacter mastomyrinus, Helicobacter ganmani, and Rodentibacter heylii. The infected mice had tested positive via serology and fecal PCR (MNV, Helicobacter spp., and R. heylii) by a commercial testing laboratory (IDEXX BioAnalytics, Westbrook, ME) and in-house oropharyngeal culture (R. heylii). After the 14 d of cohousing, the 3 mice were individually housed in clean, Innovive cages, tested for confirmation of the 3 pathogens via serology (MNV) and fecal PCR (MNV, Helicobacter spp., and R. heylii) by a commercial testing laboratory (IDEXX BioAnalytics, Westbrook, ME) and by in-house oropharyngeal culture (R. heylii), and served as the founders for group 3 (infected).
The study design consisted of 3 groups of Swiss-Webster mice:
- 1.
Naïve group:
- a.
4 cages of 3 mice/cage
- b.
No dirty bedding transfer for the duration of the study (weeks 0 to 12)
- 2.
Sentinel group:
- a.
2 cohorts of 4 cages with 3 mice/cage: cohort 1 testing weeks 0 to 12; cohort 2 was tested on week 8 before their first dirty bedding transfer and then tested on weeks 10 and 12
- b.
Dirty bedding transfer from group 3 at each 14-day cage change beginning week 0 (cohort 1) and week 8 (cohort 2)
- 3.
Infected group:
- a.
3 cages of 1 mouse/cage for weeks 0 to 6. 2 mice/cage for weeks 8 to 12
The experimental unit was the cage, and the sole outcome measured in this study was pathogen assay results (positive/negative) for the 3 pathogens tested, MNV, Helicobacter spp., and R. heylii. The sample size was initially determined by available mice for the study and recent publications on the same topic.5,6,8,21 Post hoc analysis determined that the positive rate detected by the fecal PCR method was 12%, while that detected by Nestlet was 85%. A sample size of 8 (the sample size at time 10 and 12 wk for group 2 sentinels) provides 80% power to detect such a difference for a one-sided test. The study did not have any inclusion or exclusion criteria, and no mice were excluded from the results.
All group 1 (naïve) cages were always changed first, followed by group 2 (sentinel), and finally group 3 (infected). At each cage change, all clean study cages received a new Nestlet and Bed-r’Nest (a paper nesting material). While changing group 3 cages, soiled bedding was collected from all 3 cages, mixed, and then equally distributed to each group 2 (sentinel) cage. We avoided collecting Nestlet material as part of the soiled bedding transferred to group 2 (sentinel). This transfer occurred at each cage change except week 6. We suspected that transmission of Helicobacter from group 3 (infected) to group 2 (sentinel) mice was low, and the Helicobacter detected in the group 2 Nestlet material was primarily from the feces transferred from group 3 (infected). On week 6, no soiled bedding was transferred at cage change (which included addition of a new Nestlet), and the samples collected before the week 8 cage change would only detect pathogens shed by the mice within the cage. Soiled bedding transfer from group 3 (infected) to group 2 (sentinel) cages was resumed on week 8 and continued to the end of the study.
Experimental procedures.
Biologic sample collection.
All samples were collected between 4 and 24 h before the scheduled cage change and transfer of dirty bedding from group 3 (infected) to group 2 (sentinel). In doing so, the soiled Nestlet material collected would have been present in the cage for 13 to 14 d (Figure 1). Week 0 was defined as the sample collection from all cages 4 to 24 h before the first dirty bedding transfer. Thereafter, samples were collected every 14 d for a total of 12 additional weeks. Samples were also collected from group 3 cages starting 6 wk prior and groups 1 and 2 cages 4 wk before the start of dirty bedding transfer to confirm the presence or absence of all 3 pathogens (MNV, Helicobacter spp., and R. heylii).
Citation: Journal of the American Association for Laboratory Animal Science 64, 5; 10.30802/AALAS-JAALAS-25-037
Whole blood was collected every 14 d from a manually restrained mouse via lancet puncture of the submandibular vein onto filter paper. One mouse per cage was tested from all 3 groups at each time point. The dried blood spots were sent to one of 3 commercial laboratories (IDEXX BioAnalytics, Westbrook, ME; Charles River Laboratories, Wilmington, MA; or VRL Animal Health Diagnostics, Gaithersburg, MD) and tested for antibodies against MNV.
Feces and soiled Nestlet material were collected individually from all study cages for evaluation of total nucleic acid (TNA) isolation every 14 d. Feces were collected directly from the mouse restrained for blood collection but also from the cage bottom to have adequate material for testing and backup. Both sample types were stored separately at −70 °C for 1 to 3 d before TNA isolation. On weeks −2 (group 3 only), 0, 2, 4, 10, and 12, feces were also sent individually to a commercial laboratory for PCR testing for murine Helicobacter spp., MNV, and/or R. heylii. Speciation of rodent Helicobacter present in samples was only performed when tested by the outside laboratory.
Oropharyngeal (OP) swabs were collected from one mouse/cage from all study cages and plated to Trypticase soy agar (BD Difco, Franklin Lakes, NJ) with 5% sterile sheep blood (blood agar plates [BAPs]; Carolina Biological Supply, Burlington, NC) and incubated at 37 °C aerobically for 48 h. The BAPs were prepared by our in-house media core facility using the manufacturer’s instructions.
Total nucleic acid isolation from feces and soiled Nestlet material.
Total nucleic acid was isolated from feces (80 to 100 mg) and soiled Nestlet material (approximately 100 mg) using the Qiagen RNeasy PowerFecal Pro Kit (Qiagen, Santa Clarita, CA). Fecal TNA was isolated following the kit protocol with the exception of the DNAse 1 treatment step. Soiled Nestlet TNA was isolated with an initial incubation of approximately 100 mg material in a sterile 2.0 mL, screw-top tube with 650-µL kit lysis buffer (Solution CD1; Qiagen, Santa Clarita, CA) and 100 µL phenol-chloroform-isoamyl alcohol. The tubes were incubated at 55 °C and constantly mixed for approximately 15 h (overnight). After the incubation, the tubes were centrifuged at 13,000 rpm for 1 min, and the supernatant was removed to a clean collection tube. The remaining Nestlet material was transferred to an Investigator Lyse and Spin Basket (Qiagen, Santa Clarita, CA) and centrifuged at 13,000 rpm for 1 min. The flow-through solution was transferred to the same collection tube. Thereafter, the kit isolation protocol was followed with the exception of the DNase 1 treatment step.
PCR and sequence analysis.
Reverse transcription of fecal (approximately 200 to 300 ng) and Nestlet (approximately 100 to 150 ng) TNA was performed using the Omniscript RT kit (Qiagen, Santa Clarita, CA) and random hexamer primers at 37 °C for 1 h in a T100 Thermal Cycler (Bio-Rad Laboratories, Hercules, CA).
Realtime PCR was performed using the CFX96 Realtime System (Bio-Rad Laboratories, Hercules, CA) on cDNA (MNV) or TNA [Helicobacter spp., R. heylii, or mouse 18S rRNA (template quality target)] using the probe and primer sets listed in Table 1. PCR cycle conditions were 95 °C for 3 min followed by 40 cycles of 95 °C for 10 s, 55 °C for 30 s, and 72 °C for 30 s. Results were analyzed using the CFX Maestro software (Version 2.3; Bio-Rad Laboratories, Hercules, CA).
| Target | Oligo name | Sequence (5'-3') | Reference |
|---|---|---|---|
| Mouse norovirus (MNV): ORF1-ORF2 junction | MNV-F2 (forward primer) | ATGGTR*GTCCCACGCCAC | Hanaki et al.46 (primers only) |
| MNV-R2 (reverse primer) | TGCGCCATCACTCATCC | ||
| MNV-P2 (probe) | [HEX]-TTGGGACAATGGATGCTGAGACCC-[BHQ1] | ||
| Helicobacter spp.: 16S rRNA gene, partial | q-hE-406 (forward primer) | ACCAAGGCTATGACGGGTATC | Huijsdens et al.47 |
| q-hE-626 (reverse primer) | CGGAGTTAGCCGGTGCTTATT | ||
| P-hE-535-555 (probe) | [6FAM]-AACCTTCATCCTCCACGCGGC-[TAM] | ||
| Rodentibacter pneumotropicus/heylii: inclusion body protein A (ibp-4) | R-ibp4 (forward primer) | GATGTGGGTGTCTCTGTAG | Buchheister et al.48 |
| R-ibp4 (reverse primer) | CCATCCGR*CTCGTTTCATC | ||
| P-R-ibp4 (probe) | [6FAM]-ACCAACAGGTAGCGTAACA-[BHQ1] | ||
| Mouse-specific 18S rRNA gene, partial | 18SrRNA(F) (forward primer) | CGGCTACCACATCCAAGGAA | Agenès et al. 49 |
| 18SrRNA(R) (reverse primer) | GCTGGAATTACCGCGGCT | ||
| 18SrRNA(P) (probe) | [6FAM]-GTGGGTTATGGTCAG-[BHQ1] |
*R, Guanine or Adenine (purine).
Suspect Rodentibacter spp. colonies from the OP swab BAPs were subcultured for isolation on fresh BAPs and incubated at 37 °C aerobically for 24 h. Isolated bacterial colonies from the pure cultures were removed from the BAP using an inoculation needle and suspending in 200 µL of sterile phosphate-buffered saline followed by 200 µL of lysis buffer from the DNeasy Blood and Tissue kit (Qiagen, Santa Clarita, CA). Genomic bacterial DNA was isolated per manufacturer’s kit instructions. The isolated genomic DNA samples were confirmed to be R. heylii by 16S rRNA PCR using the universal bacterial primers 27F (5′-AGAGTTTGATCMTGGCTC AG-3′, M = C/A) and 1492R (5′-GGTTACCTTGTTACGACTT-3′)28 in a T100 Thermal Cycler (Bio-Rad). PCR cycle conditions were 95 °C for 3 min, followed by 35 cycles of 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 1 min with a final 5 min extension period at 72 °C. The 16S rRNA gene amplicons were cleaned using a QIAquick PCR Purification Kit (Qiagen, Santa Clarita, CA). Sanger sequencing was performed by Genewiz Inc. Sequencing results were trimmed and assembled using CLC Main Workbench 8 (Qiagen, Santa Clarita, CA) using the default settings. Assembled 16S contigs were identified by uploading sequences into the National Center for Biotechnology Information’s (NCBI) Basic Local Alignment Search Tool (BLAST) using the 16S ribosomal RNA sequences (Bacteria and Archaea)’ database.
Statistical methods.
R (version 4.3.2) was used in the data analysis.29 We used generalized linear mixed models (GLMM) for binomial data to compare testing positive rates between different testing approaches based on results from weeks 2, 6, 8, 10, and 12 in group 2 (sentinel) samples. R package glmmTMB was used to fit GLMM. To account for the correlation between measurements of the same cage we used a random intercept for the cage and assumed a first-degree autoregressive [AR(1)] structure for the random effects. A P value ≤ 0.05 was considered significant.
Results
Animal health.
All mice remained healthy with no clinical signs noted throughout the study duration.
MNV testing.
PCR testing results of soiled Nestlet, feces, and serological testing results for MNV are summarized in Table 2. All group 3 (infected) samples tested were MNV positive at all time points except for 1 out of 3 (33.3%) serum samples on week 0; and 1 out of 2 (50%) Nestlet samples on week 4. All group 1 (naïve) samples tested were MNV negative at all time points.
| MNV serology and PCR testing results |
|---|
 |
| Helicobacter serology and PCR testing results |
 |
| Rodentibacter heylii culture and PCR testing results |
 |
−, Negative result; +, positive result.
All group 2 (sentinel) samples, which were collected before the first soiled bedding transfer, were MNV negative at week 0. By week 2, all group 2 (sentinel) Nestlet and fecal samples (100%) tested MNV positive by PCR, and 2 of 3 (66.6%) serum samples were MNV positive. In weeks 4, 6, and 8, all (100%) group 2 (sentinel) Nestlet, fecal and serum samples were MNV positive. In week 10, 7 out of 8 (87.5%) Nestlet, fecal, and serum samples were MNV positive. In the final week of the study (week 12), all (100%) Nestlet and fecal samples and 7 out of 8 (87.5%) serum samples were MNV positive. The GLMM model did not show any difference in positive testing rate between Nestlet and fecal samples (P value = 1) or sera (P value = 0.2).
Helicobacter testing.
PCR testing of soiled Nestlet and feces for Helicobacter spp. are summarized in Table 2. All group 3 (infected) Nestlet and fecal samples tested were Helicobacter positive at all time points, including feces tested by an outside laboratory (weeks 0, 2, 4, 10, and 12). All group 1 (naïve) Nestlet and fecal samples were Helicobacter negative at all time points.
All group 2 (sentinel) Nestlet and fecal samples, which were collected prior to the first soiled bedding transfer, were Helicobacter negative at week 0. By week 2, all group 2 (sentinel) Nestlet samples (100%) were Helicobacter positive by in-house PCR testing and remained 100% PCR positive through weeks 4 and 6. On week 8, only 2 out of 4 (50%) of the Nestlet samples were PCR positive; however, no soiled bedding was transferred at the week 6 transfer from group 3 (infected) to the sentinel (group 2) cages. Soiled bedding was again transferred on weeks 8 and 10, and all (8 out of 8 or 100%) of the Nestlet samples were PCR positive at week 10, whereas 5 out of 8 Nestlet samples (62.5%) were PCR positive at week 12. Fecal samples from group 2 (sentinel) mice were PCR negative on weeks 2 and 4 using our in-house PCR assay and via testing performed by a commercial testing laboratory. One out of 4 (25%) fecal samples were PCR positive on weeks 6 and 8, 3 out of 4 (75%) were PCR positive on week 10, and 4 out of 8 (50%) were PCR positive on week 12 using our assay. Two out of 8 (25%) of these fecal samples were PCR positive at 10 wk, and 1 out of 8 (12.5%) fecal samples were PCR positive at week 12 when tested by an outside laboratory. The GLMM model showed significantly higher positive results for Nestlet samples compared with testing of fecal samples by either the in-house PCR (P value = 1.6 × 10−4) or fecal PCR testing conducted by an outside laboratory (P value = 1.5 × 10−4).
Speciation of Helicobacter was performed when fecal samples were tested by a commercial testing laboratory. Prior to the start of the study, the original colony to which the first group of SW mice (group 3) were cohoused had tested PCR positive for H. mastomyrinis, H. typhlonius, and H. ganmani. Feces from these original group 3 SW mice were tested 2 wk before the start of the study (week −2) and H. mastomyrinis, H. typhlonius were identified, but H. ganmani was not detected. Both species were again detected from group 3 feces on weeks 0 and 2. By week 4 until the end of the study (week 12), only H. typhlonius was detected in the feces of group 3 (infected) or group 2 (sentinel; weeks 6 to 12) feces.
Rodentibacter testing.
Results for PCR testing of soiled Nestlet samples, fecal samples, and oropharyngeal cultures are summarized in Table 2. All group 3 (infected) Nestlet samples (100%) were PCR positive for R. heylii at all time points. All group 3 (infected) fecal samples (100%) were PCR positive for R. heylii using our PCR assay at weeks 0, 2, 6, 8, and 10, whereas 1 out of 2 (50%) fecal samples were PCR positive on week 4, and the single fecal sample from week 12 tested PCR negative. Group 3 feces were also tested by a commercial testing laboratory at weeks 0, 2, and 4. Two (2) out of 3 fecal samples (66.7%) were PCR positive on week 0, and single samples collected on each of weeks 2 and 4 were PCR positive at a commercial testing laboratory. All group 3 OP swabs (100%) were culture positive for R. heylii on weeks 0, 2, 4, and 6. On weeks 8 and 10, the single OP swab taken from a group 3 cage was culture negative for R. heylii, but the OP swab taken on week 12 was culture positive for R. heylii. All group 1 (naïve) Nestlet, fecal, and OP swabs were negative for R. heylii at all time points.
All group 2 (sentinel) Nestlet, fecal, and OP swabs were PCR or culture negative for R. heylii at weeks 0, 2, 4, 8, 10, and 12. One of 4 (25%) Nestlet and fecal samples were PCR positive for R. heylii on week 6; however, the OP swab cultures were negative. Due to the low positive rate, we did not fit the GLMM model.
When we performed 16S rRNA PCR and sequencing of presumed R. heylii isolates, our NCBI BLAST search often listed Rodentibacter caecimuris as the closest sequence match and not R. heylii, R. caecimuris, and R. heylii are considered heterotypic synonyms,30 meaning they are the same organism.
Mouse 18S rRNA testing.
The mouse 18S rRNA gene was used as our PCR template control target, since we expected the presence of mouse cells in the samples collected from all cages for PCR testing. We detected the mouse 18S rRNA target in all samples tested by PCR indicating there were no compounds in the samples that inhibited the PCR assays (data not shown).
Discussion
Rodent pathogens affect animal health and research outcomes, and accurate detection relies on sensitive testing modalities. In recent years, environmental testing for infectious agents has been widely reported and is becoming a popular addition to the animal health surveillance programs at many research institutions. Many of these publications have demonstrated that EHM is equal or superior to the sensitivity of testing SBS.5–7,16,19,23,24,31–34 These reports describe several different methods and materials used for collecting soiled particulate material. In this report, we tested soiled nesting material (Nestlet) as an alternative source for the detection of rodent pathogen nucleic acids and compared it to historical testing sources such as serology, fecal PCR, and microbial cultures. Our results demonstrate the efficacy of this material for the detection of MNV, murine Helicobacter spp., and R. heylii.
MNV is the most prevalent viral infection in laboratory mice35 and is known to alter physiologic responses.36,37 MNV primarily infects the rodent gut and is chronically shed in the feces in high copy numbers. Our results demonstrated that MNV was readily transferred by soiled bedding and infected the group 2 (sentinel) mice; and the soiled Nestlet sample was suitable for the detection of MNV with a sensitivity equal to fecal testing and serological testing of both known infected mice (group 3) and sentinels (group 2). We must note that MNV nucleic acid detected by PCR of group 2 (sentinel) fecal samples cannot be definitively assumed to have come directly from the mice, since feces were collected directly from one mouse/cage but also from the cage floor. Thus, it is possible that we could have collected MNV-containing feces that had been transferred from the group 3 (infected) cages 14 d prior; however, given the volume of feces present in the group 2 (sentinel) cages after 13 to 14 d, the likelihood of collecting fecal pellets transferred from group 3 (infected) would be very low. Given that the mice were serologically positive for MNV, we feel confident that they were, indeed, infected with MNV.
Helicobacter are gram-negative bacteria that preferentially infect the gastrointestinal tract but can also be found in the hepatobiliary tract of mammals.38 Helicobacter species found most often in mice and rats include H. hepaticus, H. bilis, H. typhlonius, H. mastomyrinus, H. muridarum, H. ganmani, and H. trogontum.38 Infected rodents chronically shed Helicobacter in the feces. H. hepaticus and H. mastomyrinus are readily transmissible through soiled bedding; whereas, other Helicobacter spp., including H. bilis, H. ganmani, and H. typhlonius, are poorly detected by soiled-bedding sentinels.7,38 Most Helicobacter infections in laboratory rodents are subclinical; however, they have also been reported to have widespread gastrointestinal, hepatic, and reproductive effects in mice.38 In our study, we could readily detect Helicobacter in the Nestlet material and feces of the group 3 (infected) mice. Detection of Helicobacter in the group 2 (sentinel) Nestlet material was significantly higher than feces, but fecal detection in this group was low. Only 1 to 2 fecal samples from this group tested PCR positive for Helicobacter. We believe this discrepancy between the detection rate in Nestlet material compared with feces was due to transmission to the group 2 (sentinel) mice, and that the detection in the Nestlet material was more likely from the soiled bedding transferred from the group 3 cages. To test this idea, no soiled bedding was transferred from group 3 (infected) cages to group 2 (sentinel) on week 6. When samples were collected on week 8, the number of PCR-positive Nestlet samples dropped from 4 of 4 (100%) to 2 of 4 (50%). Soiled bedding transfer resumed on week 8, and the number of PCR-positive Nestlet samples returned to 100% at week 10. We must note that the fecal PCR detection of Helicobacter from the group 2 (sentinel) cages cannot be stated as coming directly from the mice with complete certainty since feces were collected directly from one mouse/cage but also from the cage floor. We could have collected Helicobacter-containing feces that had been transferred from the group 3 (infected) cages 14 d prior; however, given the volume of feces present in the group 2 (sentinel) cages after 13 to 14 d, the likelihood of collecting fecal pellets transferred from group 3 (infected) would be very low.
Rodentibacter pneumotropicus (formerly Pasteurella pneumotropica—serovar Jawets) and R. heylii (formerly P.asteurella—serovar Heyl) are member of the Pasteurellaceae family and considered the most common pathogenic bacteria isolated from laboratory rodents worldwide.39,40 Both Rodentibacter species infect the oropharyngeal cavity and respiratory tract of rodents and are shed in oral/nasal secretions and feces but in low copy number and intermittently. Both species are usually nonpathogenic in immunocompetent rodents; however, severe disease and mortality have been reported in immunodeficient rodents.41,42 These bacteria do not survive long outside a host, especially in low moisture environments43 such as an IVC, hampering the transmission to SBS. In our study, we could detect R. heylii in the Nestlet samples, fecal samples, and OP swab cultures of the group 3 (infected) mice with high accuracy; however, transmission of R. heylii to the group 2 sentinels was low and could not be detected well using any source material, as others have reported.34
We have demonstrated that soiled Nestlet material is a viable source material for the detection of high-prevalence rodent pathogens (MNV, Helicobacter, and Rodentibacter). Since we use Nestlets as part of the enrichment provided to all mice, the collection of soiled material for testing is readily available from any cage within our mouse population and does not require additional particulate collection material and minimal disturbance of occupied cages. We plan to include the PCR testing of soiled Nestlet material as part of our comprehensive animal health surveillance program. Our current study tested Nestlet material that was only present for 14 d. Maintaining the same soiled Nestlet material in a SBS cage for longer period may increase our chances of pathogen detection. As promoted by the recent publication by LaFollette et al.,19 we agree that EHM is superior to SBS for the detection of many rodent pathogens by PCR. However, most of the studies reviewed in this paper looked at high prevalence pathogens. As pointed out in one of these publications, we still need more data on the sensitivity of EHM for low prevalence pathogens,31 especially those included on recommended exclusion lists.44 Thankfully, a recent publication has started the discussion on revising exclusion lists.45 For now, we must rely on currently accepted testing methodologies for many of these pathogens.
Nestlet material used for testing.
Contributor Notes