Extended Sanitization Frequency for Bottles: Using Volume as an Alternative to Time-Scheduled Replacements for Mice in Disposable Individually Ventilated Caging Systems
The Guide for the Care and Use of Laboratory Animals specifies that sipper tubes require weekly sanitization, although reduced frequency for sterile, disposable caging components may be justified with performance-based assessments; therefore, extending bottle duration beyond 1 week, and replacing instead at a lower volume limit, may be acceptable if it does not compromise mouse health or water quality. Weekly bottle change for sterile, disposable bottles results in excessive waste of both water and plastic—bottles are typically more than half full at 7 days. A volume-based replacement schedule is a visual alternative to weekly changes for systems using sterile, disposable bottles, reducing the number of bottles processed and facilitating operations by allowing spot changes without the need to date or track bottles. Furthermore, the gravity-drip design of Innovive’s Aquavive bottle minimizes contamination, suggesting that water quality and cleanliness may be maintained for longer durations. To assess the impact of a volume-based replacement schedule on mouse health and water quality, C57BL/6NCrl mice at varying housing densities were monitored for changes in animal weight and mean water consumption. At bottle replacement, packed cell volume was collected from individual mice as a marker of hydration status, and water quality was assessed by visual inspection, culture for Escherichia coli and coliform bacteria, and total microbial count. Baseline measurements were collected over 7 days, followed by a volume-based replacement period, where bottles remained on the cage until they reached 100 mL (13-47 days). No significant differences were observed in animal body weight, body condition, water consumption, or packed cell volume, regardless of cage density or the amount of time the bottle was deployed. Microbial analysis showed no bacterial growth in any bottle, and visual inspection showed no turbidity or cloudiness. Based on this information, our institution allowed prefilled disposable, acidified water bottles to be maintained well beyond 7 days, using a volume-based replacement at 100 mL.
Introduction
The adoption of individually vented cage (IVC) systems and recyclable, irradiated caging components has many advantages. IVC systems allow individual cages to receive a supply of HEPA-filtered air at a set air exchange rate, theoretically ensuring that each cage is isolated from other cages on the same rack.1 Due to decreased ammonia and CO2 buildup in IVC systems compared with static microisolation cages, a growing number of institutions have investigated extending cage change frequency for cage bottoms and accessories2–5; however, few institutions have investigated extended replacement frequency for sterile, disposable water bottles.4
The standardization of microenvironmental parameters is essential to animal welfare and research reproducibility.6,7 Standards for the sanitization frequency of cage components are outlined in the Guide for the Care and Use of Animals (the Guide).7 Although the Guide recommends weekly sanitization for bottles and sipper tubes, it acknowledges that adjustments may be acceptable for IVC systems. The Guide states that “decreased sanitization frequency [for ventilated caging systems] may be justified if the microenvironment in the cages, under the conditions of use …, is not compromised.”7 Verification of microenvironmental conditions includes measurement of pollutants such as ammonia and CO2, microbiologic load, observations of animals’ behavior and appearance, and the condition of bedding and cage surfaces.7
This study adds to the limited body of literature surrounding the use of disposable, prefilled bottles, and it investigated whether weekly bottle replacement was necessary and appropriate, given the abundant plastic and water waste associated with these bottles. In addition to generating physical waste, bottle replacement is labor-intensive and ergonomically challenging: bottle replacement must be tracked and scheduled, caps must be removed, and remaining water emptied to recycle the bottle. Bottles may be spot-changed before their scheduled replacement dates due to variability in water consumption or water loss events, such as flooding. These bottles then need to be disposed of at the next scheduled change, resulting in abundant waste, or dated and tracked not to exceed the allowed timeframe between changes. Previous work with disposable, prefilled, and chlorinated bottles demonstrated viability of leaving bottles for up to 30 days in singly housed mouse cages, to align with a monthly cage accessory changing schedule.4
A limited, preliminary internal investigation examined bottle change extension to 14 days to align with an IACUC-approved exception for cage change. For 3 cages of group-housed C57BL/6NCrl female mice (n = 15) and 3 cages of singly housed, male transgenic mice on a C57BL/6NTac background (n = 3), no significant differences were reported in water consumption; all mice maintained or increased body weight; and no microbial growth was noted via internal culturing. This pilot suggested that water bottle change extension to 14 days did not adversely affect animal welfare or water quality, including in potentially sensitive transgenic strains. Bottle change at 14 days, however, resulted in operational inefficiencies, where some bottles, particularly those in densely populated cages, reached the allowed lower limit of 100 mL before their scheduled 14-day replacement. Bottles then had to be dated and tracked, resulting in daily spot changes and a misalignment with our cage change schedule. As a follow-up, this study aimed to evaluate whether it was possible to extend bottle duration on the basis of volume, thereby completely eliminating scheduled bottle replacements.
Aquavive mouse water bottles provide 300 mL of prefilled water, acidified with HCl to a pH range of 2.5 to 3.0. Acidified water is used at our institution due to its known antibacterial effects8 and because mice in our facility are predominantly supplied from vendors who use acidified water. Based on published work, an adult C57BL/6NCrl mouse on an IVC system consumes ∼3 to 8 mL of water daily,9,10 suggesting that a volume of 300 mL should last between 8 and 20 days for a cage of 5 mice (8 mL/d × 5 mice = 40 mL/d × 7.5 days = 300 mL; 3 mL/d × 5 mice = 15 mL/d × 20 days = 300 mL). The lowest acceptable limit for bottle replacement at our institution is 100 mL, which represents an ∼4-day supply for group-housed mice, based on a midrange consumption rate of 5 mL/d at the highest housing density (5 mL/d × 5 mice = 25 mL × 4 days = 100 mL). Other institutions have similar limits.4 The 100 mL volume is clearly marked on the bottle’s design, allowing a quick and accurate assessment of bottle volume at the same time as daily cage checks. An extension in duration of disposable, acidified water bottles, on the basis of volume, for both group- and singly housed mice, may be appropriate if water remains palatable and free of contaminants, and mice remain in good health.
Materials and Methods
Ethical review.
All work described was reviewed and approved by the site’s IACUC, under an approved IACUC protocol. Eisai is an AAALAC-accredited institution.
Study design.
Mice were housed at varying densities according to institutional standards. Differing densities were selected to represent a range of housing densities in a typical housing room, since the number of mice in a cage directly affects the rate of water consumption. Experimental groups included cages with female mice at high and low densities, and singly housed males.
This study spanned 2 observation periods, first to establish baseline trends using the Guide’s recommended 7-day bottle replacement schedule, and second to allow the bottles to extend beyond 7 days, until the volume reached 100 mL. The same mice were observed across these 2 observation periods, separated by a 2-week period to allow for a return to baseline following study manipulations. This allowed each group to serve as its own control. High-density females were housed in fewer cages, whose volume-based observation period was repeated after the 2-week holding period. This allowed the total number of cages per group to remain the same, while reducing the number of animals used.
Water quality and animal health were evaluated during these periods at varying frequencies to ensure that water remained palatable and free of contaminants, and that mice exhibited normal activity and weight gain trends.
Sample size.
Experimental groups consisted of 3 cages of 5 female mice (n = 15 total) in the high-density female (HDF) group, 6 cages of 2 female mice (n = 12 total) in the low-density female (LDF) group, and 6 cages of singly housed male mice (n = 6 total). A priori sample size calculations were not conducted, but post hoc analysis was calculated using R Studio 4.4.0.
Inclusion and exclusion criteria.
Naive animals in good health were included in the study. During the length of the study, exclusion criteria included any adverse clinical event or evidence of pain and distress according to institutional policies (total body condition score, 20% weight loss). No mice were excluded.
Randomization.
Neither animals nor procedure order was randomized for this study.
Blinding.
All participants were aware of group allocation along different experimental stages; no blinding was used.
Outcome measures.
The following parameters were assessed: water volume loss measured over time, animal weight in grams measured over time, packed cell volume of blood (PCV, or hematocrit), and microbial burden and assessments of visual contamination of the water in bottles at bottle replacement.
Statistical methods.
All statistical analyses were performed using R version 4.4.0. Statistical significance was defined as P < 0.05 with t tests and regression linear models with Bonferroni correction for multiple testing. Results were reported with mean and SD. Differences in model slope were tested by bootstrapping confidence intervals for a difference in slope to account for unequal sample size and variance, with P values estimated from the resulting F tests (ANOVA) with the boot package.11 Normality assumptions were tested with the Shapiro–Wilk test, and homogeneity of variance with a Levine test in the car package. Wilcoxon rank-sum tests were used to test the differences in packed cell volume, as the HDF measurements had a nonnormal distribution. Sample size was determined based on literature and feasibility constraints in consultation with the IACUC, and post hoc power analyses for the number of mice and observations were conducted assuming a power of 0.8 and α of 0.5 with the pwr package12 for tests, and bootstrapping power calculations for Wilcoxon tests. Visualizations were produced with ggplot213 and ggstatsplot.14
Experimental animals.
A total of 33 C57BL/6NCrl mice (Charles River Laboratories, Raleigh, NC) were used, and all were naive with respect to previous procedures or manipulations. Housing density varied to reflect a typical distribution where cage density is variable within accepted limits due to research attrition, veterinary guidance, mortality, and incompatibility (such as breeding males). Males were selected to be the lowest density group, because fighting typically results in singly housed cages, whereas females are cohoused whenever possible.
All animals were housed in an AAALAC-accredited facility in accordance with standards as described in the Guide. Mice were housed in 100% polyethylene terephthalate– and bisphenol A–free plastic IVC systems prefilled with corncob bedding (M-BTM-C8 Innocage mouse bottom with corncob bedding; Innovive, San Diego, CA). Water was provided in the form of a 300-mL prefilled, acidified bottle (Aquavive; Innovive, San Diego, CA). Two forms of enrichment were provided, in the form of a nesting pouch (80EPI EnviroPak irradiated; ScottPharma, Marlboro, MA) and a paper cellulose hut (SKBI Shepherd Shack backless irradiated; ScottPharma, Marlboro, MA). Cages were maintained on a 12-hour light/12-hour dark cycle. Temperature (68-79 °F [20-26 °C]) and relative humidity (30%-70%) of rooms housing mice remained within acceptable limits. Mice were maintained with ad libitum access to acidified water and rodent chow (2918 Teklad Global 18% protein rodent diet; Inotiv, West Lafayette, IN). Cage bottoms were changed every 2 weeks, and cage lids and accessories were changed monthly, in accordance with IACUC policies and as approved exceptions to the Guide.
Facility colony health surveillance was conducted quarterly using pooled fecal and bedding samples from each cage change on sentinel-free soiled bedding media (PathogenBinder collection media; Charles River Laboratories, Wilmington, MA), which was sent to Charles River Laboratories (Wilmington, MA) for infectious disease PCR diagnostic analysis (PathogenBinder pilot mouse surveillance plus PRIA infectious disease PCR; Charles River Research Animal Diagnostic Services, Wilmington, MA). Excluded agents included the following: Sendai virus, murine hepatitis virus, mice minute virus, mouse parvovirus 1 and 2, mouse theilovirus, murine norovirus, murine rotavirus, mousepox (ectromelia), epizootic diarrhea of infant mice, lymphocytic choriomeningitis virus, Mycoplasma pulmonis, pneumonia virus of mice, reovirus, Hantaan virus, Filobacterium rodentium, Myobia, Myocoptes, Radfordia spp., Aspiculuris, and Syphacia spp. Proteus mirabilis was present in the facility, among a different mouse line.
Mice were euthanized via CO2 asphyxiation followed by cervical dislocation, or transferred to other approved protocols at the conclusion of this study. Euthanasia was carried out in accordance with institutional standard operating procedures and the AVMA Guidelines for the Euthanasia of Animals (2020).
Experimental procedures.
Water consumption.
As a proxy for water consumption, water volume loss was measured by weighing the water bottles every 2-5 days until the end of the assessment period (7 days for baseline, or until the volume reached 100 mL, or ∼130 g, for volume-based bottle replacement). A final, empty bottle weight was obtained, although the decision to replace bottle volume was visually approximated. Volume loss for each experimental group was compared with their own baseline using bootstrapping confidence intervals for a difference in slope.
Animal health.
Animals were weighed weekly, in the morning, to track body condition and compare growth trends with published data, assuming that animals abstaining from drinking unpalatable water would show inconsistent or irregular growth trends and, potentially, weight loss. Clinical observations were noted, monitored, and treated as appropriate. Barbering and other hair and whisker loss are common among group-housed C57BL/6 mice,15–17 and they were not noted unless intervention and treatment were required. Weight gain or loss from each observation period was reported in comparison to initial weight.
Hydration status via PCV.
Animal hydration status, inferring water palatability (and therefore, consumption), was assessed at the conclusion of the observation period (7 days for baseline, or up to 47 days for volume-based bottle replacement). Animals abstaining from drinking unpalatable water would likely have become dehydrated. PCV was used to measure hydration status, with an increase in PCV indicating dehydration. Approximately 40 µL of whole blood was sampled from the lateral tain vein into 40-mm microcapillary tubes, and due to the need to generate duplicate samples, up to 120 µL was collected at each collection. Sealed samples were spun in a centrifuge (081383 Covetrus Pro ZipCombo centrifuge; Covetrus, Portland, ME) at 6,900 g for 3 minutes, according to the manufacturer’s guidelines, and read against a lineal hematocrit reader card. All blood collection protocols followed institutional guidelines.
Microbial analysis.
At the conclusion of the observation period, and after obtaining final bottle weights, 1 mL of water from each bottle was incubated for 24 hours at 35 °C in a sealed Peel Plate EC (Charm Sciences, Lawrence, MA) for Escherichia coli/coliform. All bacterial colony growth was noted and counted.
Bioburden analysis.
At the conclusion of the observation period, and after obtaining final bottle weights, water was pooled by experimental group to submit two 100-mL samples to Charles River Laboratories (Wilmington, MA) for analysis via their microbial bioburden water test (BT-ENV-3). Collection materials were provided by Charles River Laboratories. Baseline bottles (changed at 7 days) were prepped for analysis at the time of collection; however, during the volume-based replacement period, bottles within each group reached their endpoints at different timepoints. In this case, bottles were wrapped with parafilm and set aside in a biosafety cabinet or within a closed box. Samples were not processed until all bottles within a group reached the lower volume limit.
Results
Bottle duration.
The bottle change time was extended beyond the Guide’s recommended 1-week sanitization period by 1 week for high-density cages containing 5 mice, and by up to 6 weeks for singly housed mice (Table 1).
| Group | No. of mice per cage | No. of days until bottle replacement |
|---|---|---|
| HDF | 5 | 13-14 |
| LDF | 2 | 26-29 |
| SHM | 1 | 41-47 |
Bottles were replaced at ∼2 wk (14 d) for HDF cages containing 5 mice, ∼4 wk (28 d) for LDF cages containing 2 mice, and ∼6 wk (42 d) for cages of SHM.
Abbreviations: HDF, high-density female; LDF, low-density female; SHM, singly housed male.
Water consumption.
Water bottle volume loss, as a proxy for water consumption, declined at a consistent rate across all experimental groups and across both 7-day baseline and volume-based observation periods (Figure 1). Slopes were compared per experimental group across observation periods, and no significant difference in rate of consumption was found as bottle change duration extended (Figure 1, Table S1). Rate of water volume loss translated to an average of 3.3 ± 0.4 mL per mouse per day for females and 4.4 ± 0.8 mL for males (Figure 1, Tables S1 and S2).
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-25-116
Animal health.
No clinical concerns were reported throughout the study duration. Body weight evaluations showed that mean weight generally increased during each observation period (Figure 2). Comparisons of the slope of body weight increase found no significant differences between mean weights of animals consuming bottles across 7 days and animals consuming bottles across longer durations (Figure 2, Table S3). All animals remained within 10% of their baseline weight at the conclusion of each observation period.
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-25-116
Hydration status via PCV.
PCV values for all groups were within normal limits. Mean PCV was not significantly different between 7-day bottle replacement and volume-based bottle replacement for high-density cages (Shapiro–Wilk test for normality, P = 0.048; Wilcoxon signed-rank test, PHDF = 0.096; Tables S4-S6). A significant decrease in PCV was recorded for low-density groups (Wilcoxon signed-rank test, PLDF= 0.045). Mean hematocrit for singly housed males did not differ between observation periods, but post hoc analysis suggested that the sample size was too small to view effect (mean7 d = mean41-47 d = 53.3%; Wilcoxon signed-rank test, PSHM = 1.0). No group had increased mean PCV at the end of volume-based bottle replacement periods (up to 47 days) versus day 7 (Figure 3, Table S7).
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-25-116
Microbial and bioburden analyses.
Bacterial culture of water from the bottles did not show growth of E. coli, coliform, aerobic bacteria, or fungi at any timepoint evaluated between 7 and 47 days. No visible contaminants, cloudiness, or turbidity were noted in any bottle, regardless of duration on the cage or mouse density within the cage.
Discussion
Taken together, these findings indicate that extending the use of a disposable water bottle beyond a weekly replacement frequency had no adverse effects on water quality (including microbial count and palatability) or animal welfare. Water in these gravity-drip bottles is held by surface tension, and mice obtain water by licking the bottle opening. In this nonmetabolic system, water may also be lost due to other activities, such as cage manipulation by personnel during daily health checks and cage change. All cages were on the same cage change schedule and were handled consistently throughout this study to minimize interference. Water volume declined at a steady, predictable rate until 100 mL, suggesting that water was consumed at an even rate; and within each group, this rate did not differ from baseline (Figure 1, ANOVA, PHDF = 0.594, PLDF = 0.892, PSHM = 0.689). The change in volume equated to 3.3 mL per mouse per day for females and 4.4 mL per mouse per day for males, consistent with other published data for adult C57BL/6NCrl mice in IVC systems.9,10 Animals did not preferentially drink water at faster rates from newer bottles, or reduce drinking from older bottles; however, only one bottle remained available to animals at a time. Follow-up studies may examine preferential consumption between older and newer bottles if both bottles are offered simultaneously.
Body weight is a well-described indicator of animal hydration and wellbeing; water intake in normally hydrated animals is proportional to body weight.10,18 A normally hydrated mouse would be expected to maintain or increase weight8; CD-1 mice in an IVC system under dehydration parameters were reported to lose 12%-18% of their baseline weight over 48 hours.18 Our study showed increases in body weight across all observational periods, consistent with strain-specific growth trends,19 suggesting that mice remained healthy and hydrated.
Hydration status was further evaluated using PCV, commonly used to indicate anemia, polycythemia, and dehydration.20 PCV measures the ratio of red blood cells within whole blood, and dehydrated animals will have a higher reading than normally hydrated animals. Animals who abstain from drinking potentially unpalatable water would be expected to demonstrate an increase in PCV within hours of abstaining. Published work investigating CD-1 mice under dehydration parameters found that PCV readings increased after 12, 24, and 48 hours with restricted water access.18 The published PCV range is 37.2%-58.0% for an adult female C57BL/6NCrl mouse, and 37.3%-62.0% for an adult male mouse.19 All readings in the present study were within this published range, except for one mouse in the LDF experimental group, who had a slightly elevated value (60% up from 52%; LDF, volume-based; Figure 3). No clinical signs of dehydration were observed in that individual. In general, mean PCV did not increase from baseline for any group, suggesting that the mice were not more dehydrated at later timepoints (Figure 3). PCV trended toward lower values at later timepoints (eg, LDF; Figure 3), possibly indicating a slight increase in hydration between observation periods. This trend, while significant for the LDF group, may also result from natural fluctuation in hydration or as a result of low sample size, as mean PCV values remained within normal limits. Regardless, trends toward stable or decreasing PCV values support the hypothesis that mice do not abstain from drinking water from older bottles. This, combined with increasing weight trends and steady per-cage consumption trends, suggest that baseline health was maintained until the bottles were replaced at 100 mL (up to 47 days).
All bottle contents remained free of E. coli, coliform, aerobic bacteria, and fungi for up to 47 days. This is consistent with other published data, where no microbial bioburden was reported in chlorinated water out to 30 days.4 That work also described culture of the waterspout and found 2 organisms that were ubiquitous to the environment or present in the normal skin flora of their mice; as such, the organisms were not cause for concern.4 Waterspout culture was not included in our study, and it may be considered for future analysis. The lack of microbial growth or visible contamination within the bottle, however, reasonably support that water maintains potability up to 47 days (∼6 weeks).
The results of this study lead us to conclude that extended bottle duration from 1 to 6 weeks, based on volume, had no adverse effect on body weight, water consumption, or hydration status. A volume-based replacement schedule allows bottles to be changed quickly, by visual assessment, and co-occurring with daily cage checks. Further, changing bottles based on the volume reduces the amount of water and plastic waste and alleviates some of the ergonomic stress associated with emptying bottles that are nearly full. This study investigated only disposable, 300-mL bottles prefilled with acidified water. Other watering systems using nonchemically treated water, automatic watering systems, or bottles with larger volumes require follow-up testing. We also did not measure pH over time, which may be relevant to studies or facilities that must maintain a specific pH value of water throughout the course of study.
These findings are significant because the acceptable duration of use of sterile, prefilled bottles is not well described. The impact of extended bottle change duration is quantifiable. In cages with group-housed mice, bottles lasted 13-14 days, representing a 100% reduction in wasted bottles and water compared with standard 7-day bottle changes. The savings are even greater for lower density caging, where bottles for singly housed mice lasted 42-47 days, representing a 600% reduction in waste. With current pricing, for a facility with 1,000 cages of group-housed mice, this represents a potential annual savings of $31,720 (26,000 bottles at $1.22 per bottle). Plastic waste could be reduced by as much as 26,000 bottles and caps annually, and water waste by 2,060 gallons. Facilities, particularly those using disposable systems, have an obligation to consider the environmental impact as long as animal welfare is not compromised, through performance-based testing.
Water Bottle Volume in Each Cage Type by Experimental Condition. Bottle volume is reported in milliliters. (A) HDF group. (B) LDF group. (C) SHM group. Purple lines show water volume over a 7-d (control) observation period; additional colors show the experimental observation period. HDF, high-density female; LDF, low-density female; SHM, singly housed male.
Change in Body Weight, Expressed as a Percentage of Initial Weight, in Each Group Following Bottle Replacement. (A) HDF group. (B) LDF group. (C) SHM group. Purple lines show weight change over a 7-d (baseline) observation period; additional colors show the experimental observation period. Fitted regression lines estimate rate of change. HDF, high-density female; LDF, low-density female; SHM, singly housed male.
Change in PCV Between Observation Periods, With Typical Reference Ranges Plotted in Light Yellow (Minimum Female = 37.2, Minimum Male = 37.3, Maximum Female = 58.0, Maximum Male = 62.0). Boxplot lower and upper hinges represent the first and third quartiles of data, while the midline is the median. Upper and lower whiskers extend no farther than 1.5× the interquartile range. Mean values were as follows: HDF control = 52.5, HDF experimental = 50.6, LDF control = 52.9, LDF experimental = 51.2, SHM control = 53.3, SHM experimental = 53.3. PCV, packed cell volume of blood; HDF, high-density female; LDF, low-density female; SHM, singly housed male.
Contributor Notes
This article contains supplemental materials online.