How To Clean A Mouse Cage

J Am Assoc Lab Anim Sci. 2018 Mar; 57(2): 138–142.

Published online 2018 Mar.

Evaluation of Extended Sanitation Interval for Cage Top Components in Individually Ventilated Mouse Cages

Brianne LS Ball

¹Office of Animal Resources, University of Iowa, and

Kathleen M Donovan

¹Office of Animal Resources, University of Iowa, and

Steven Clegg

²Department of Microbiology and Immunology, Roy J and Lucille A Carver College of Medicine, University of Iowa, Iowa City, Iowa

James T Sheets

¹Office of Animal Resources, University of Iowa, and

Received 2017 Sep 26; Revised 2017 Oct 24; Accepted 2017 Nov 2.

This article has been cited by other articles in PMC.

Abstract

Sanitation frequency of mouse cage components can be determined through verification of microenvironment, including microbiologic load and air quality within the cage. Here we demonstrate a consistent microbiologic load on wire IVC lids that were used for as long as 8 continuous weeks to house 4 or 5 mice and significant decreases in the microbial load on filter tops at 4, 6, and 8 wk compared with 2 wk. In addition, air quality, represented by intracage ammonia concentration at the time of bedding change, did not differ between 2-, 4-, and 6-wk time points in cages containing same-sex groups of 4 or 5 male or female adult mice. We propose that the lack of significant differences represents justification for an extended sanitation frequency of as long as 6 wk for cage top components in mouse IVC housing and represents a performance standard that might be reproduced by similar facilities to determine appropriate sanitation frequencies for mouse caging components.

The frequency of changing and sanitation of rodent cage components to produce an appropriate microenvironment for the housed animals is a subject of debate.^{⁵,⁶,¹²,¹³,¹⁵,¹⁶} The Guide for the Care and Use of Laboratory Animals states that caging components, such as cage tops, should be sanitized at least once every 2 wk but that decreased sanitation frequency might be justifiable when the cage microenvironment is not compromised.^⁷,¹³ Furthermore, regulatory and accrediting agencies have accepted that institutions may involve site-specific and data-driven approaches to support alterations in sanitation frequencies.^¹,¹⁰ In most contemporary rodent management programs, bedding and cage bottoms can become soiled rapidly and require changing on 1- or 2-wk intervals, even in the context of IVC, whereas the sanitation of components such as feeders, wire lids, and noncontact filter lids must be sufficiently frequent to ensure animal welfare. However, frequent changing and sanitation can be disruptive to the animals, increasing stress responses and introducing variables into studies.^{⁵,¹¹,¹⁵} More frequent changes (and the resultant cleaning) also lead to more rapid deterioration of caging components and unnecessarily increase the workload and expense of animal care.^⁴,⁵,¹²

According to the Guide, extended sanitation intervals can be justified with verification of appropriate microenvironment; this verification might include measurements of pollutants such as ammonia and carbon dioxide, evaluation of microbiologic load, observation of the animals' behavior and appearance, and the condition of bedding and cage surfaces.^⁷ Studies evaluating sanitation frequency have used luciferase testing swabs and microbial culture plates to determine microbiologic load and to evaluate ammonia levels in static mouse cages.^¹⁶ Breakdown of the urea in urine can result in ammonia buildup and unhealthy air quality within the cage environment over time. IVC are designed to have continuous air flow to maintain good air quality throughout the period between mouse cage changes. Intracage ammonia levels have been measured through diverse methods in various studies, including the commercially available colorimetric paper we used to evaluate waste gas levels in the current study.^²

In this study, we evaluated the animals' microenvironment by assessing both microbial load on cage top surfaces and air quality within cages. The tested hypothesis was whether the continual use of cage top components (wire lid and filter top) would, over time, result in an increase in bacterial contamination of the cage tops. In addition, this study examined whether filters became clogged during prolonged cage-top sanitation intervals, thus resulting in reduced airflow and increased ammonia levels within the cages.

Materials and Methods

Animals.

All experiments were approved by the University of Iowa IACUC and were performed in accordance with the Guide for the Care and Use of Laboratory Animals.⁷ Mice were housed in IVC (Thoren Caging Systems Hazleton, PA) in an AAALAC-accredited nonbarrier animal facility of the University of Iowa. Standard husbandry in this facility involved cage changes and handling in room air (that is, no ventilated changing station), irradiated rodent chow (diet 7913, Envigo Teklad, Madison WI), and automatic delivery of filtered water, with a 12:12-h light:dark cycle, 10 to 15 room air changes hourly, and a room temperature of 21.1 ± 1.1 °C. Humidity is not actively managed due to the age of the facility but is monitored continuously and ranged between 15% and 55% during the study period. The study populations included both male and female socially housed mice (age, 4 to 24 mo), with cage change frequencies as described later. Mice used in the microbiology and welfare assessments were 160 female NIH Swiss (NCI Cr:NIH(S); age, 6 wk) purchased from Charles River Laboratories (Frederick, MD). Mice used in the ammonia monitor study were retired colony mice with various genetic modifications on C57BL/6J and BALB/c backgrounds (ongoing inhouse breeding, n = 102 mice in 24 cages). Animals were housed in a nonbarrier with quarterly health surveillance via dirty-bedding sentinels. Sentinels were consistently negative for the following excluded agents: mouse parvovirus, minute virus of mice, pneumonia virus of mice, reovirus 3, epizootic diarrhea of infant mice virus, lymphocytic choriomeningitis virus, mouse hepatitis virus, ectromelia, Sendai virus, Theiler encephalomyelitis virus, mouse adenoviruses types 1 and 2, polyoma virus, and Mycoplasma pulmonis. Mouse norovirus and Helicobacter spp. are considered accepted agents in this facility.

Husbandry.

Mice were house 4 or 5 per cage in IVC (size 9 Maxi-Miser, Thoren; Figure 1), with automatic water valves, wire lids with integrated steel food hoppers, and a cage ventilation rate of 0.5 to 1 cage-volume exchange per minute (30 to 60 cage-volume air changes hourly). Paper-based bedding (SoftZorb Enrichment Blend, Nepco, Warrensburg, NY) was used, with a measured 6 cups of bedding placed in each cage. Cages were mechanically washed with alkaline cagewash detergent (Cage Klenz 100, STERIS, Mentor, OH) and water heated to 180 °C but were not autoclaved between uses. The mice were placed in clean, bedded cages and housed undisturbed on the housing rack for each 14-d interval, with daily welfare checks performed without removal of the cage from the rack. Every 2 wk, cages were removed from the housing rack and the filter tops lifted and sampled directly as described later, without direct contact with glove or animal to the inner surface of the filter top. Each filter top was sampled and set on a clean surface, the wire bars were lifted and sampled, and the mice were moved to the clean cage bottom, the steel cage top and filter were placed on the new cage, and the unit was returned to the housing rack.

An external file that holds a picture, illustration, etc. Object name is jaalas2018000138f01.jpg

Design and air movement in the IVC used in the current study. These IVC have HEPA-filtered air supplied through sealed shelf plenums into the cage, through air supply ports above the cage filter top. Air enters through the rear of the cage and exits at the front of the plenum.^¹⁷ Water valves are located at the back of the shelf and enter the cage approximately 3.5 in. above the plastic floor through a manufacturer-installed grommet. This grommet is covered by a hinged door when not in use and is the access point for ammonia testing.

Microbiology.

A standardized group size of 4 NIH/Swiss male or female mice per cage was used, with a total of 40 test cages and 4 control cages. Unoccupied cages containing bedding and feed were placed at the top, midpoint, and bottom levels of the housing rack to serve as controls. Every 14 d (when the cage bottom and bedding were changed), 4 areas of the cage tops were swabbed with saline-moistened sterile cotton-tipped applicators. Samples (4 cm × 4 cm each) were obtained from 2 sites of the filter top: sample A, directly under the exhaust port; and sample B, in the opposite quadrant away from both inflow and exhaust ports (Figure 1). In addition, 4-cm segments of 5 wire bars in the cage lid were sampled: sample C was collected at the lowest face of the feed hopper (parallel to the cage bottom), and sample D was collected directly under the exhaust port (Figure 2). These swabs were applied immediately to 100-mm culture plates containing trypticase soy agar (Fisher Scientific, Waltham MA) in a spread-plate pattern. After 24 h of incubation at 37 °C, colonies were counted to determine total aerobic microbial growth.

An external file that holds a picture, illustration, etc. Object name is jaalas2018000138f02.jpg

Locations of microbiologic sampling points. Two 4 × 4-cm squares of the inner face of the filter top (sites A and B) and two 4-cm lengths of 5 bars of the wire lid (sites C and D) were sampled.

Ammonia monitoring.

For this experiment, University of Iowa researchers donated retired colony mice, with various genetic modifications on C57BL/6J and BALB/c backgrounds and without significant phenotypic effects on housing or behavior. Mice (age, 8 to 24 mo) were housed 4 or 5 per cage; 4 cages of 5 male mice and 20 cages of 4 or 5 female mice were used. To evaluate whether filters clogged or deteriorated with use, resulting in altered air quality, 10 cages were outfitted with new filters and 10 cages with filters randomly selected from those in active use in the facility. The ammonia level within the unopened cage was tested immediately after its removal from the IVC rack, prior to changing of the cage bottom (dirtiest cage conditions). Test strips (Hydrion Ammonia Test Paper, Micro Essential Laboratory, Brooklyn, NY), in which color change occurred at 5, 10, 20, 50, and 100 ppm, were used to evaluate ammonia levels. Test paper was moistened with sterile water according to the assay directions and inserted through the grommet hole at the end of the cage by using 12-in. steel forceps. The paper was held in the center of the cage for 15 s, after which the paper was withdrawn and immediately compared with the color chart to determine the ammonia concentration (in ppm) of the intracage air environment. A positive control test was performed at each assay time point by applying 10 μL of household ammonia to the paper bedding in an empty, freshly bedded cage; placing a filter top over the cage; and allowing 60 s for diffusion. After 60 s, the filter paper was inserted, and a prompt color change indicated an ammonia concentration of 10 ppm above the center of the cage.

Each cage bottom, including bedding, was changed at 14-d intervals. The same filter top and wire lid were used continuously for more than 6 wk, as described earlier.

Blinded welfare assessments.

In addition to daily welfare checks by animal care staff, during which any abnormal appearance or activity was reported to veterinary staff, a clinical veterinarian blinded to cage status evaluated each cage and scored the animals/ appearance and behavior as normal or abnormal, with a detailed description of any abnormality observed. Any appearance of abnormal mouse activity, illness, pain, or distress would have resulted in the removal of that cage from the study; observation of welfare concerns in multiple cages would have resulted in early termination of the experiment. The 40 cages of mice described for the microbiology evaluation were used for the welfare assessments.

Statistical methods.

After a microbiology pilot study was completed, a biostatistician was consulted regarding a power analysis to produce statistically significant results. For a power of 0.80 with an α value of 0.05, the minimal number of cages required was estimated to be 36 (we added 4 more cages to ensure sufficient power). For each test location, microbial load was compared over the 4 time points (2, 4, 6, and 8 wk) by using a Friedman test. Microbial load within each cage was ranked at each time point, and the mean rank was compared among the 4 time points. An overall test was performed to evaluate any effect of time; if this result was significant, then posthoc pairwise comparison was performed, with P values adjusted according to the Bonferroni method. A P value of 0.05 or less was considered statistically significant. All analysis was performed using SAS version 9.4 (SAS Institute Inc., Cary, NC).

Results

Microbiology.

Colony counts at 24 h of aerobic incubation were 0 or 1 for all control cages (empty of animals) and 0 to 140 cfu for occupied cages (Figure 3). After 48 h of incubation, the confluence of bacterial colonies on plates with previous growth resulted in an inability to enumerate the data further. The presence of microbial growth on cage surfaces was expected; the purpose of this study was to identify the point at which a significant increase in microbial levels could be observed after successive weeks without sanitation of the cage top components. The data were compiled and scored by using a Friedman test, which indicated no significant difference (P > 0.05) between microbial loads at either location on the wire lid or food hopper (contact surfaces) and a significant decrease (P ≤ 0.05) in aerobic microbial load between 2 wk and all later time points (4, 6, and 8 wk) at both sampling locations on the filter top.

An external file that holds a picture, illustration, etc. Object name is jaalas2018000138f03.jpg

Microbial load (no. of cfu) on cage top components. The median (point) and middle quartiles (bars) of each data set are indicated for each sample location (sites A, B, C, and D) across the 4 time points (2, 4, 6, and 8 wk). Significant differences were identified only for sample A, between the 2-wk time point and each of the other time points. P values as measured by Friedman testing are indicated where significance was identified.

Welfare assessments.

During 8 wk of continuous housing, no welfare concerns or health reports requiring medical attention were reported in any of the animals on study. The welfare assessments by a blinded clinical veterinarian indicated no abnormal appearance or behavior of the mice was observed, except for 4 of the 40 cage, in which evidence of barbering or alopecia on one or more of the mice in the cage was noted. All mice in these cages were otherwise bright and alert with normal activity, and the incidence of alopecia was not inconsistent with the reported rate for this activity.^⁸,¹⁴

Ammonia monitoring.

The threshold of detection for the assay used was 5 ppm, with additional color changes at 10, 20, 50, and 100 ppm of ammonia in the air tested. After 2 wk of housing, testing of all 24 cages resulted in either no or very slight color change (less than 5 ppm), thus setting a baseline expectation for the normal status in the cage at bedding change. The color change after 4 and 6 wk of continuous housing with the same filter top and wire lid showed no difference compared with that for the 2-wk time point (that is, all cages were measured at less than 5 ppm). In addition, there was no measured difference in intracage ammonia levels according to mouse sex or filter status (circulated compared with unused).

Discussion

The goal of this study was to determine whether extending the time between changes of cage top components resulted in detectable differences in microbial load, animal behavior, or intracage air quality over time. Although cage bottoms and bedding can become soiled within a matter of days of animal housing, the wire lids and filter tops retained low microbiologic loads under continuous housing for as long as 4 times the typical interval (that is, 8 wk compared with 2 wk). The load on the wire lids was repeatedly low across all times tested, consistent with the inert nature of a metal surface. Over prolonged periods, buildup of organic debris might result in high microbial loads or biofilm formation; however, a change interval of as long as 8 wk for steel tops is justified in light of the lack of significant increase in microbiologic load on these cage components.

The microbial load on the cage filters was generally quite low, a finding that can readily be explained by the fact that these are noncontact surfaces for the animals; only dust, dander, and bedding particles should reach the filter top, given the design of the IVC system. Contrary to the wire lid, which mice contact directly as they move about the cage and procure food from the feed hopper, the filter top should not be exposed to fecal matter, feed pellets, or skin and mucosal flora. Surprisingly, the number of colonies measured on the filter surface decreased significantly between the 2-wk measurement and all subsequent time points. Because this difference was unexpected, concerns arose that the filter surfaces were clogging due to extended usage and that air flow through the ventilated cage might be impaired, as described in other publications.^² Although the microbial load did not increase with increased time housed, the concerns regarding air quality required further exploration.

Various studies have evaluated the effect of intracage ammonia concentration, with observations of histopathologic lesions starting at ammonia levels as low as 50 ppm.^{³,⁹,¹⁸,¹⁹} An 8-h time-weighted average of 25 ppm has been established as a safe upper limit for human exposure to ammonia, and this value has been used as a threshold for mouse exposure in some studies.^⁶,¹⁸ In the current study, we planned for 25 ppm ammonia as an upper limit for continued housing; however, cage ventilation appeared to be sufficient to prevent measurement of more than a trace amount of ammonia across all time points. The decrease in microbial counts after the 14-d time point, along with a subjective visual assessment of the filter tops, suggests the development of a debris mat or material buildup in the filter at the exhaust port location. However, the measurements indicate that any buildup was insufficient to adversely influence the intracage ammonia concentrations and support the intended function of IVC systems to reduce waste gas concentrations continually.

Taken together, the animal behavior observations, microbiologic quantification, and ammonia concentration measurements suggest that the frequency of sanitation for the steel cage lids and filter tops of the tested IVC system can be prolonged to at least 6 wk without compromising the microenvironment within the cage. Further assessment is needed to establish a performance standard for intervals longer than 6 to 8 wk and for other IVC systems housing adult mice. The procedures we outline here are modest in cost and likely can readily be reproduced by other animal care programs to evaluate whether a reduced sanitation frequency for caging components is appropriate. It is beyond the scope of this paper to evaluate all available IVC systems; however, similar evaluations of microbial load and air quality would be appropriate for demonstrating performance standards in many mouse and rat housing systems. Furthermore, additional studies might be conducted to evaluate the effects of sanitation interval for cage components under specialized husbandry conditions, such as those for breeding animals or study models such as diabetes mellitus or renal disease, and to validate the methods in other caging systems or programs of animal care.

Acknowledgments

We thank the Department of Microbiology (Carver College of Medicine) for providing microbiologic culture supplies; Dr M Bridget Zimmerman (University of Iowa Biostatistics Core Alliance) for statistical assistance; and the animal caretakers of the Office of Animal Resources for their daily commitment to animal care and welfare.

References

2. Creamer MA, Petty J, Martin T, Bergdall V, Hickman-Davis JM. 2014. Implications of natural occlusion of ventilated racks on ammonia and sanitation practices. J Am Assoc Lab Anim Sci 53:174–179. [PMC free article] [PubMed] [Google Scholar]

3. Ferrecchia CE, Jensen K, Van Andel R. 2014. Intracage ammonia levels in static and individually ventilated cages housing C57BL/6 mice on 4 bedding substrates. J Am Assoc Lab Anim Sci 53: 146–151. [PMC free article] [PubMed] [Google Scholar]

4. Fox JG, Barthold S, Davisson M, Newcomer C, Quimby F, Smith A. 2007. The mouse in biomedical research. Burlington (MA): Elsevier. [Google Scholar]

5. Godbey T, Gray G, Jeffery D. 2011. Cage-change interval preference in mice. Lab Anim (NY) 40:225–230. [PubMed] [Google Scholar]

6. Horn MJ, Hudson SV, Bostrom LA, Cooper DM. 2012. Effects of cage density, sanitation frequency, and bedding type on animal wellbeing and health and cage environment in mice and rats. J Am Assoc Lab Anim Sci 51:781–788. [PMC free article] [PubMed] [Google Scholar]

7. Institute for Laboratory Animal Research 2011. Guide for the care and use of laboratory animals. 8th ed Washington (DC): National Academies Press. [Google Scholar]

8. Garner JP, Weisker SM, Dufour B, Mench JA. 2004. Barbering (fur and whisker trimming) by laboratory mice as a model of human trichotillomania andobsessive-compulsive spectrum disorders. Comp Med 54:216–224. [PubMed] [Google Scholar]

9. Mexas AM, Brice AK, Caro AC, Hillanbrand TS, Gaertner DJ. 2015. Nasal histopathology and intracage ammonia levels in female groups and breeding mice housed in static isolation cages. J Am Assoc Lab Anim Sci 54:478–486. [PMC free article] [PubMed] [Google Scholar]

10. Office of Laboratory Animal Welfare. [Internet] 2017. Frequently asked questions. PHS policy on humane care and use of laboratory animals. Office of Laboratory Animal Welfare. [Cited 10 September 2017]. Available at: https://grants.nih.gov/grants/olaw/faqs.htm#674

11. Rasmussen S, Miller MM, Filipski SB, Tolwani RJ. 2011. Cage change influences serum corticosterone and anxiety-like behaviors in the mouse. J Am Assoc Lab Anim Sci 50:479–483. [PMC free article] [PubMed] [Google Scholar]

12. Reeb-Whitaker CK, Paigen B, Beamer WG, Bronson RT, Churchill GA, Schweitzer IB, Myers DD. 2001. The impact of reduced frequency of cage changes on the health of mice housed in ventilated cages. Lab Anim 35:58–73. [PubMed] [Google Scholar]

13. Reeb C, Jones R, Bearg D, Bedigan H, Myers D, Paigen B. 1998. Microenvironment in ventilated animal cages with differing ventilation rates, mice populations, and frequency of bedding changes. Contemp Top Lab Anim Sci 37:43–49. [PubMed] [Google Scholar]

14. Reinhardt V. 2005. Hair pulling: a review. Lab Anim 39:361–369. [PubMed] [Google Scholar]

15. Rosenbaum MD, VandeWoude S, Johnson TE. 2009. Effects of cage-change frequency and bedding volume on mice and their microenvironment. J Am Assoc Lab Anim Sci 48:763–773. [PMC free article] [PubMed] [Google Scholar]

16. Schondelmeyer CW, Dillehay DL, Webb SK, Huerkamp MJ, Mook DM, Pullium JK. 2006. Investigation of appropriate sanitization frequency for rodent caging accessories: evidence supporting less-frequent cleaning. J Am Assoc Lab Anim Sci 45:40–43. [PubMed] [Google Scholar]

18. Vogelweid CM, Zapien KA, Honigford MJ, Li L, Li H, Marshall H. 2011. Effects of a 28-d cage-change interval on intracage ammonia levels, nasal histology, and perceived welfare of CD1 mice. J Am Assoc Lab Anim Sci 50:868–878. [PMC free article] [PubMed] [Google Scholar]

19. Washington IM, Payton ME. 2016. Ammonia levels and urine-spot characteristics as cage-change indicators for high-density individually ventilated mouse cages. J Am Assoc Lab Anim Sci 55:260–267. [PMC free article] [PubMed] [Google Scholar]