Investigation of Food Marking and Contamination Behaviors Informs Feeding Practice in Research Colonies of Laboratory Opossums
The laboratory opossum, Monodelphis domestica, serves as a critical marsupial model in biomedical research. Proper feeding approaches are essential for promoting animal growth and wellbeing. In this study, we systematically evaluated food scattering and potential food contamination from feces across 4 feeding methods: direct placement of food pellets on bedding and using 3 different types of containers. We conducted timed daily observations of food scattering and marking behaviors in 22 animals, capturing images by photograph at specific intervals over the course of a week. Body weight was measured before and after the trial. Our findings revealed that the containers did not prevent food scattering behaviors, as evidenced by comparable survival curves for food scattering across all methods (P > 0.05, log-rank test). Although the paper tray and ceramic dish delayed the occurrence of food marking by feces, indicated by a significant extension in the time to marking events (P = 0.009 and P < 0.001, respectively), these containers introduced new animal welfare concerns. The paper tray increased bleeding incidents in digits and paw pads nearly 8-fold (P = 0.0002), presumably due to sharp edges. The ceramic dish was associated with urine marking, and small but statistically significant weight loss (0.7%, P < 0.05). By 144 h, all cages showed food contamination regardless of the feeding method. The results suggest that containers provide minimal benefit in preventing food contamination, and some types of containers may pose health risks. Therefore, we propose that placing food pellets directly on the bedding, a practice used for 45 y of laboratory opossum maintenance, is acceptable for promoting optimal health and operational efficiency for this species. Our results fill a significant gap in care practices and offer insights into optimal colony management for this important research model.
Introduction
The laboratory opossum, Monodelphis domestica, was established as a research model in 1979 and has been successfully maintained in the laboratory setting since then.31 This small, nocturnal mammal is native to South America and possesses many unique characteristics that have allowed its successful establishment as a laboratory animal.1,32,33 The average size of an adult laboratory opossum is 50 to 100 g for females and 80 to 150 g for males. Monodelphis has a short gestation period, lasting only 13.5 d, and readily breed year-round in captivity.7 Monodelphis pups are born at an immature developmental stage, comparable to 11.5 d of mouse gestation and 6 wk of human gestation.3 The typical litter size averages 7 to 8 pups, but can range from 1 to 13 pups.33 The pups are weaned at 2 mo of age, and they are typically individually housed by 4 mo of age due to approaching sexual maturity and an associated increase in aggressive behavior toward conspecifics.8,33 Monodelphis is an important laboratory model for various research disciplines.9 Due to their divergence from eutherian mammals approximately 160 million years ago,18 metatherian mammals, or marsupials, possess unique characteristics that allow them to be considered as “alternative mammals.”21,24 Despite the characteristics that differentiate marsupials (metatherians) from eutherian mammals, most of the physiologic functions of these 2 infraclasses remain conserved. The differences as well as the similarities between the infraclasses enable the laboratory opossum to serve as a useful and, in some cases, unique model for various genetic, physiologic, and disease processes.2,15,17,19,22,25–27,29,33–35,37
There are currently 20 genetic stocks and inbred strains of laboratory opossums, all derived from the 38 animals imported to the United States between 1978 and 1993. During these years, there were 7 separate importations of Monodelphis, with animals sourced from 4 locations in Brazil and one location in Bolivia. Each animal was assigned to a population numbered 1 to 5 based on the location from which they were imported.32 The Auburn University College of Veterinary Medicine (AUCVM) established a breeding colony of PBP and LSD strains of laboratory opossums in 2018. The PBP strain is a random-bred line derived from population 2, imported from Piraua, Brazil between 1984 and 1988. The LSD strain is an inbred line derived from population 1, imported from Exu, Brazil in 1978.32 The inbreeding coefficient of this strain is 0.99, and the fecundity of this strain is robust.38
Monodelphis domestica is a USDA-regulated species and conducts annual inspections to ensure compliance with expectations.30 Due to behavioral differences between Monodelphis and rodents, appropriate husbandry methods for Monodelphis are different from those of rodents. For laboratory mice, feed pellets are provided on top of the wire bars, and animals reach them when needed. However, this approach does not work for opossums. According to the Handbook on the Care and Management of Laboratory and Other Research Animals, it is recommended that opossum food be placed directly on the bedding without any food container.33 Other research colonies have used metal bowls and glass dishes for food presentation.14,23 At the AUCVM research colony, the IACUC-approved feeding method is to provide the animal with ¼ cup (or 25 g) of fresh food pellets on clean bedding in one corner of the cage, with additional fresh pellets added as needed. However, after the addition of fresh pellets, the animals typically remove the pellets from their original location and disperse them throughout the bedding of the cage1,16,36 and/or mark their food with excrement. Following an annual USDA inspection of the AUCVM Monodelphis colony in 2023, a concern was raised that the opossums did not have ad libitum access to uncontaminated food. Based on previous observations made by Division of Laboratory Animal Health animal care staff and Wang laboratory personnel during routine animal care and maintenance, the food scattering and food marking behaviors are the cause of this concern. To evaluate whether providing a food container can alleviate the food contamination issue, we formally evaluated 3 types of containers provided by the Division of Laboratory Animal Health. Our results inform choice of feeding methods that accommodate both animal welfare and operational efficiency for the animal husbandry.
Materials and Methods
Ethics statement.
The AUCVM Monodelphis domestica colony is maintained in a facility accredited by AAALAC, International that follows guidelines set forth in the IACUC-approved standard operating procedures for the breeding colony. The food scattering and marking behavior observation research was approved by the IACUC of Auburn University (PRN 2024 to 5400).
Animals and husbandry.
Adult animals are individually housed due to their solitary nature and tendency to display aggressive behavior toward conspecifics. Each opossum is housed in a polypropylene or polycarbonate rodent box of standard size (43 × 22 × 13 cm [length × width × height]). This standard shoebox cage is used for both maintenance and breeding. Each cage is fitted with stainless steel wire bars and a microisolation lid. Each cage contains 3 to 4 cm (depth) of Sani-Chips bedding (Envigo Bioproducts, Madison, WI), nesting material (crinkle paper), a plastic tunnel, and a glass or polycarbonate water bottle with a sipper tube of at least 3.5 cm in length. The plastic tunnels provided for enrichment are amber or red in color. This color selection was made to provide a reduced light environment, as Monodelphis is a nocturnal species and has photic preference for reduced light conditions.28 The tunnel also provides an enclosed space for nesting and allows for climbing activities. Furthermore, the wire bars fitted on the top of the cage serve not only to hold the water bottle, but they allow for climbing activities, as well. Cages are changed every 2 wk, as weekly cage changes are distressing for the opossums.1 The food source is a specialized commercial diet for gray, short-tailed opossums (https://www.labdiet.com) with a balanced nutritional profile based on previous research.5,6 Under the current IACUC-approved standard operating procedures, the opossums are fed ¼ cup (or 25 g) of food in a paper tray. The water source is Auburn Municipal Water (Auburn, AL), or tap water, which is tested annually by the City of Auburn. The lights are on an automatic cycle of 14 h light/10 h dark. Room humidity is maintained at 35% to 60% with the help of supplemental humidifiers placed in the animal rooms.
Experimental design.
Twenty-two adult laboratory opossums were enrolled in this study, representing both sexes, a random-bred stock and an inbred strain, and 3 age groups (Table 1). Fourteen opossums were of the PBP stock, and 8 opossums were of the LSD strain. An equal number of males and females were included in the study (n = 11 males, n = 11 females). The 3 age groups were 6 to 12 mo (n = 8 opossums), 13 to 18 mo (n = 8 opossums), and opossums aged 19 mo and older (n = 6 opossums). The average lifespan of the laboratory opossum is about 3 y, at which point, if not sooner, their health tends to decline. For this reason, few opossums of advanced age were available for this study.
| Age group 1 (6–12 mo) | Age group 2 (13–18 mo) | Age group 3 (19+ mo) |
|---|---|---|
| 2 PBP males | 2 PBP males | 3 PBP males |
| 2 PBP females | 2 PBP females | 3 PBP females |
| 2 LSD males | 2 LSD males | |
| 2 LSD females | 2 LSD females |
Four feeding methods were tested for efficacy in providing ad libitum access to uncontaminated food. Among these 4 methods are the original IACUC-approved method using no container, a polystyrene weighing boat (VWR International, Radnor, PA), the current IACUC-approved method using a paper tray (Stock Your Home, Brooklyn, NY), and a ceramic dish (Dowan, Charlotte, NC). The disposable polystyrene weighing boat was made of 100% virgin antistatic polystyrene and measured 9 × 9 × 2 cm (length × width × height). The disposable paper tray was made of premium paperboard and measured 13 × 9.5 × 3.2 cm (length × width × height). The round, ceramic dish had a glazed finish and measured 4.5 cm in height and 9 cm in diameter.
Behavioral observations and analysis.
Twelve observation timepoints were selected in advance of the experiment: t = 0, 0.5, 1, 2, 4, 24, 48, 72, 96, 120, 144, and 168 h. A week-long observation period was selected as half of the standard amount of time between cage changes. To standardize observation data for analysis, 3 behavioral outcome options were preselected to represent the behaviors relevant to this study. “None” was selected if no or minimal change to the food presentation was observed. “Food scattering” was selected if the food was removed from the original location and dispersed throughout the bedding. “Food scattering” was also selected if the animal added a significant amount of bedding or nesting material into the container/the food’s original location, covering the food (a “significant amount” was considered any amount in excess of what is expected to occur by chance). “Food marking” was selected if the animal defecated on top of the food or if fecal matter was observed to be in contact with the food. “Food marking” was also selected if the animal urinated on the food, although this event is difficult to ascertain unless it is observed directly.
Several approaches were implemented to control the semiobjective nature of the experiment. All observations and documentation were conducted by the same laboratory personnel all of whom have extensive experience in opossum care and breeding. Photographs of each cage were taken at t = 24 h, t = 72 h, and t = 168 h. At the end of the study, the photographs were analyzed and assigned a food contamination score using a numerical scale with ratings of 1 to 5. To limit bias, all photos from the same time point, regardless of feeding method, were assessed concurrently. The various scores were defined based on prior informal observations made during routine animal care. A score of 1 indicates minimal to no change to the food presentation. A score of 5 indicates drastic change to the food presentation, with the food being heavily contaminated.
Body weight measurements.
Each animal was weighed to the nearest gram on a digital scale (Etekcity, Anaheim, CA) at the beginning and end of the observation period for each container. The trends observed with weight loss or gain are useful for inferring the behavioral response and preference of the opossums for each feeding approach.
Statistical analysis.
Comparison of bleeding incidence was performed using a Fisher exact test using R version 4.3.3.12 Comparisons of changes in body weight were performed using the Wilcoxon signed-rank test with continuity correction implemented in R. Kaplan–Meier survival analysis13 was performed using the R package ‘survival’ based on observed survival times inferred from the presence in the censored data of the 4 different feeding methods. Empirical survival curves were plotted for 4 feeding methods. Statistical significance of the survival curve comparisons was determined by the log-rank test.20
Results
A significant increase in digit bleeding incidence after implementing paper tray as the food container.
In response to the USDA’s concerns about ensuring access to uncontaminated food, we modified the food delivery approach for opossums in the maintenance cages. Previously, the practice involved placing food directly on the bedding in one corner of the cage (Figure 1A). We since transitioned to using a paper tray to hold the food (Figure 1B–D), which provided a practical method for monitoring the cleanliness of the food. In the event of contamination, the disposable paper tray could easily be replaced to ensure the availability of clean food. Shortly after the implementation of the paper tray feeding method, a significant increase in the number of bleeding incidents was observed by the Division of Laboratory Animal Health veterinary care staff (P = 0.0002, Fisher exact test; Figure 2A). Many of the postimplementation bleeding reports involved observation of dried blood without an identifiable cause (such as fighting in breeding cage), and the digits and/or paw pads were often identified as the source of the bleeding (Figure 2B). The incidence of digit/paw pad bleeding increased by 7.5-fold within 5 mo following the introduction of the paper tray container (5.0 reports per month), compared with the period before its addition (0.7 reports per month; Table S1). The sharp edge of the paper tray is a plausible cause of the injuries during the feeding activity (Figure 2B).
Citation: Journal of the American Association for Laboratory Animal Science 64, 1; 10.30802/AALAS-JAALAS-24-060
Citation: Journal of the American Association for Laboratory Animal Science 64, 1; 10.30802/AALAS-JAALAS-24-060
Opossum food scattering behavior occurs daily regardless of food container type.
To address the USDA’s concern and the adverse effects observed after switching to the paper tray as the food container (Figure 2), we designed a controlled study to characterize food scattering and marking behaviors (Figure 3A, B and Table 1). This study aimed to determine whether the use of the paper tray or 2 alternative containers (Figure 3C) could reduce food contamination, and to quantify their relative effectiveness in achieving this goal. Four feeding approaches were tested: no container, a paper tray, a weighing boat, and a ceramic dish (Figure 3D). Typically, food scattering behavior, characterized by hiding food within the bedding, was observed on day 1 after cage change (Figure 4A). We performed survival analyses on the food scattering behavior using observation data from 0.5 h to 7 d after a cage change (Data S1; see Materials and Methods). Our findings showed no significant difference in the survival probability of food scattering events among all pairwise comparisons of the 4 feeding approaches (Table 2). Compared with the use of no container, the ceramic dish (P = 0.24), paper tray (P = 0.45), and weighing boat (P = 0.5) offer insignificant to no benefit (log-rank test; Figure 4B).
Citation: Journal of the American Association for Laboratory Animal Science 64, 1; 10.30802/AALAS-JAALAS-24-060
Citation: Journal of the American Association for Laboratory Animal Science 64, 1; 10.30802/AALAS-JAALAS-24-060
| Behavior phenotype | Feeding method 1 | Feeding method 2 | Log-rank test P value |
|---|---|---|---|
| Food scattering or marking | No container | Ceramic dish | 0.0023 |
| No container | Paper tray | 0.026 | |
| No container | Weighing boat | 0.038 | |
| Ceramic dish | Paper tray | 0.58 | |
| Ceramic dish | Weighing boat | 0.16 | |
| Paper tray | Weighing boat | 0.56 | |
| Food scattering only | No container | Ceramic dish | 0.24 |
| No container | Paper tray | 0.45 | |
| No container | Weighing boat | 0.5 | |
| Ceramic dish | Paper tray | 1 | |
| Ceramic dish | Weighing boat | 0.25 | |
| Paper tray | Weighing boat | 0.5 | |
| Marking only | No container | Ceramic dish | 0.00001 |
| No container | Paper tray | 0.0088 | |
| No container | Weighing boat | 0.3 | |
| Ceramic dish | Paper tray | 0.041 | |
| Ceramic dish | Weighing boat | 0.00001 | |
| Paper tray | Weighing boat | 0.04 |
Ceramic dish and paper tray can delay the occurrence of fecal marking of opossum food, but they cannot prevent the food marking behavior.
We then investigated the behavior of food marking with feces, independent of food scattering (Figure 3B). When no container was used, most of the cages were marked by feces within day 1 soon after the opossums defecated (Figure 4A). The cages with weighing boats became contaminated with feces as quickly as for direct placement of feed on the bedding (no container) (P = 0.30, log-rank test; Table 2), with more than 50% of cages marked with feces after 1 d (Figure 5B). By the 120-h timepoint, all cages using the weighing boat or no container exhibited food marked with fecal matter. In contrast, the paper tray container (Figure 5C) offered a significant advantage over both the use of no container (P = 0.01) and the weighing boat (P = 0.04, log-rank test; Table 2). For the ceramic dish, it took 2 d after a cage change for more than 50% of the cages to become contaminated with feces (Figure 5D); the ceramic dish was the most effective in delaying fecal contamination compared with use of no container (P = 0.00001), the weighing boat (P = 0.00001), or the paper tray (P = 0.041, log-rank test; Table 2). However, the ceramic dish was frequently marked by urine (Figure 5E). In addition, fecal piles began to accumulate immediately outside the ceramic dish (Figure 5F), suggesting that the physical separation provided by the dish prevented the opossums from defecating inside it, rather than causing a decrease in the frequency of food marking behavior. In summary, we found statistical significance in the survival time to marking when a paper tray or ceramic dish was used compared to use of a weighing dish or no container (Figure 5G), but we did not conclude that the former methods were effective in preventing food contamination by feces. Thus, while provision of feed in either the paper tray and ceramic dish delayed the onset of fecal contamination of the food, the outcome was ultimately the same. By the end of the 144-h trial period, all cages had become contaminated either through food scattering and/or food marking behaviors.
Citation: Journal of the American Association for Laboratory Animal Science 64, 1; 10.30802/AALAS-JAALAS-24-060
Survival probability from food contamination across 4 tested feeding methods.
We define ‘lack of access to uncontaminated food’ as situations where food is hidden in the bedding or marked with feces, and we analyzed the survival time from when the animal was placed in a clean cage until the food became contaminated. Pairwise comparison revealed that the estimated survival probability, which is the inverse probability of a food scattering or food marking event occurring, was significantly increased by using any container compared with no container (ceramic dish, P = 0.002; paper tray, P = 0.026; weighing boat, P = 0.038; Table 2 and Figure 6). When evaluating the 3 containers, there was no significant difference in the occurrence of contamination based on which container was used (P > 0.05, log-rank tests; Table 2 and Figure 6). The most notable difference between feeding approaches is observed within the first 24 h of the trial period, after which the situation stabilizes (Figure 6). By t = 144 h, all cages were observed to have some form of food contamination.
Citation: Journal of the American Association for Laboratory Animal Science 64, 1; 10.30802/AALAS-JAALAS-24-060
Effect of sex and strain on food marking behavior.
We hypothesized that males and females would exhibit significant differences in the frequency of food marking, but the hypothesis was refuted (P = 0.71, log-rank tests; Figure S2). When the impact of genetic strain (LSD compared with PBP) was examined, we observed that food in cages of PBP animals was marked significantly faster than that of LSD animals (P = 0.0076, log-rank tests; Figure S2). LSD is a highly inbred strain with an inbreeding coefficient of greater than 0.99, whereas PBP is random bred and has a higher level of genetic diversity.38 This finding demonstrates that genetic factors significantly affect food marking behavior in opossums.
Weight change patterns observed in response to various containers.
Weight measurements, which were collected at the beginning and end of each trial period, were analyzed to detect significant changes in body weight (Table S3). The animals gained weight after the 7-d trial for no container (0.64 ± 0.39 g, average weight change ± SE) and weighing boat (0.73 ± 0.43 g); no weight change difference was detected between these 2 feeding approaches (P = 0.96, Wilcoxon signed-rank test). In contrast, opossums showed significant weight loss when fed from the paper tray (−0.59 ± 0.44 g, average weight change ± SE) or ceramic dish (−0.59 ± 0.43 g), compared with not using any container (P < 0.05, Wilcoxon signed-rank test). Since the absolute weight loss is less than 1% of the total body weight, we conclude that this weight fluctuation is unlikely to be due to malnourishment. However, we did observe a small but statistically significant weight loss in this experiment for 2 types of containers.
Discussion
In the laboratory setting, opossums urinate and defecate on the bedding, which can result in contamination when food is placed directly on the bedding. To address this concern, we evaluated 3 types of food containers intended to create a physical barrier between the food and any urine or feces. Despite this strategy, our results indicated that none of the container types we tested was successful in preventing fecal and urine contamination of the food. The most effective method for delaying food contamination was the ceramic dish. We observed feces accumulating near the dish, and the opossums made considerable effort to defecate inside the ceramic dish. However, they occasionally failed due to physical restrictions and depth of the dish. These observations suggest that the food marking behavior is instinctive and is likely a normal part of their feeding routine in nature.
In terms of ability to delay food contamination, presentation of food using the ceramic dish was found to be the most effective. The paper tray was less effective, while the weighing boat had little or no effect. The difference in contamination occurrence between methods was most apparent in the first 24 h and diminished over the subsequent days. By t = 96 h, all cages with the ceramic dish or no container were contaminated. By t = 120 h, all cages with the weighing boat had food contamination, and by t = 144 h, all cages with the paper tray had been contaminated. The observation period lasted for 1 wk, which is only half of the typical duration between cage changes. Of note, Increased frequency of cage changes is not recommended, as that practice could cause significant distress to the opossums, a nocturnal and solitary species that relies heavily on its olfactory system. For these reasons, we conclude that food contamination cannot be effectively prevented by any of the container types evaluated in this study.
The paper tray was initially considered the best alternative feeding approach; it was implemented due to its practicality in food quality monitoring and easy replacement. However, the transition to use of the paper tray led to a nearly 8-fold increase in bleeding incidents, as well as weight loss. Therefore, the paper tray feeding method is contraindicated, because it is largely ineffective, is associated with an increased occurrence of cuts on the digits and paw pads, and causes weight loss.
The polystyrene weighing boat has advantages such as being disposable, sterile, and readily available as a standard laboratory supply. However, the polystyrene boat is thin, rigid, and prone to chipping, which has elicited negative responses from some opossums, leading them to chew on the container (Figure S1). This behavior creates hazardous sharp edges due to the container’s material properties. Although it has not been tested over a long duration, the sharp edges could potentially cause cuts to exposed skin. In addition, the small pieces that chip off the boat may pose a risk of gastrointestinal damage if swallowed. Monodelphis is prone to developing gastrointestinal pathologies, specifically intestinal blockages (due to fur) and rectal prolapse1,5,11; introducing a new hazardous variable could place the animals at further risk. Use of the weighing boat for presentation of feed is not recommended due to its inefficacy in reducing contamination and its potential to cause health issues.
The ceramic dish appears to be the best option for reducing contamination by food marking. It is reusable and can be sterilized by autoclaving, making it a more sustainable choice, albeit more labor intensive. The food appeared to be cleaner overall with the use of this container; however, each ceramic dish was emptied at the end of the trial period, at which point it was observed that food had caked at the bottom of many containers, presumably due to urine marking. Because the container is heavy, the opossums were not able to move it from its original location in the front corner of the cage. For this reason, it was presumably urine that caused the food to become caked. The nonporous nature of the ceramic dish means that liquids cannot drain away, which exacerbates the food contamination.
The reduction in food marking in groups with a paper tray or a ceramic dish was accompanied by a statistically significant reduction in body weight and, presumably, food consumption. In only a week, animals fed from these containers lost a mean of 0.6% of body weight, whereas animals fed with no container or a weighing boat gained 0.7% of body weight. Because the paper tray or ceramic dish delayed the occurrence of fecal marking, one plausible interpretation would be that fecal marking enhances food consumption and may be an important behavioral characteristic for the wellbeing not only of opossums in a laboratory setting, but also of opossums in their natural environment. In this experiment, we observed that the opossums repeatedly attempted to defecate in the ceramic dish but failed to do so, in some cases, due to physical constraints imposed by the ceramic dish. We suspect that in the wild, opossums engage in food marking behavior to create a sense of security while consuming their food. We are not claiming that the use of food containers will cause malnutrition; however, we did observe weight loss in some of the groups where containers were tested in this experiment. Therefore, long-term studies are needed to formally evaluate the effect of containers on weight gain or loss.
The laboratory opossum, Monodelphis domestica, is a nocturnal, solitary species that is primarily active at night. Monodelphis species are unusual marsupials in that they are pouchless; newborns attach to the mother’s nipples and are carried on the mother’s belly without any external protection. The pups are exceedingly robust and resistant to microbial infections. Since the colony’s establishment in 2018, the AUCVM has not observed a single case of Monodelphis pup infection. Monodelphis possesses robust innate and adaptive immune systems, which can defend against pathogens. Notably, they have an enriched amount of immunoglobulins supplied through the mother’s milk and saliva, which protect the pups during grooming activities when the mother licks them.10,19 In addition, Monodelphis has been found to express a diverse array of antimicrobial peptides as part of the innate immune response. These small peptides are capable of targeting a wide range of pathogens, including bacteria, viruses, and fungi.4 Furthermore, the previous feeding practice was used for the last 6 y at AUCVM, and the laboratory opossum colony is thriving in terms of health and breeding, with the guidance from other well-established colonies and support from the veterinary staff. There may be a discrepancy between what humans perceive as clean and what is actually clean to animals. The concept of cleanliness can vary significantly depending on the species, and it is influenced by their natural behaviors and adaptations. The zealousness of some veterinarians and colony managers to maintain ‘cleanliness’ via the use of food containers may detract from the wellbeing of opossums by inhibiting natural food marking and, consequently, food consumption. Our results suggest that any food container which inhibits marking (and keeps the food clean) is likely also to cause a reduction in food consumption.
Our evaluation of food contamination tested 3 types of containers in a research colony setting. We do not recommend paper-based or plastic-based containers because of the digital bleeding and plastic chipping that we observed. Hard material-based containers that can be heat-sterilized, such as the ceramic dish tested in this study, as well as metal bowls and glass dishes used by other research groups, have permitted the successful breeding and maintenance of the opossums 14,23 without the risks posed by paper-based or plastic-based containers. However, because the use of hard material-based containers may (depending on the size and shape) inhibit food marking, leading to reduced food consumption, we recommend placing the food directly on the bedding without any container, as the optimal feeding approach. After Monodelphis was first established as a laboratory model for biomedical research at the Texas Biomedical Research Institute in 1979, Dr. John L. VandeBerg developed and maintained 20 genetic strains using the feeding method of placing food pellets at a corner of the cage. Dr. Paul Samollow, Professor Emeritus at Texas A&M University College of Veterinary Medicine, successfully maintained his opossum colony for more than 20 y using this same method. Both colonies have been clinically healthy since their establishment. Because placement of feed directly onto bedding does not negatively impact the health of opossums, and because use of any of the 3 types of containers that we tested results in potential adverse physical effects including, for 2 types of containers, potential weight loss, we recommend feeding without a container, a method that has been proven to be successful for promoting health and reproduction of this species in the laboratory for 45 y.
Monodelphis maintenance cage setup before and after USDA inspection. (A) Setup of an opossum maintenance cage, with opossum food placed directly on the bedding at one corner of the cage (approved by IACUC). (B) Modification of the opossum maintenance cage in response to USDA concerns, by adding a paper tray as the food container. (C and D) Top view of the opossum maintenance cage demonstrating the paper tray with opossum chow inside, without (C) and with (D) the wire bar top.
Increase of Monodelphis bleeding incidence after switching to paper tray as feeding container. (A) Plot of the number of bleeding reports between January 2023 and February 2024. The total incidence of bleeding is represented by the blue line/dots. Bleeding from known causes, such as aggression/fighting following the pairing of breeders, is indicated by the orange line/dots. Bleeding from unknown causes, such as dried blood found in the cage, is shown in green. (B) Photo showing an opossum placing its paw on the edge of a paper tray.
Design of food contamination and marking behavior experiments to test 4 feeding containers. (A) Behavior experiment design. A total of 22 animals were enrolled in this study and housed individually. Following each cage change, food contamination, by scattering and marking behaviors, was observed at multiple time points: 0, 0.5, 1, 2, 4, 24, 48, 72, 96, 120, 144, and 168 h. (B) Cartoon illustrations depict 2 ways of food contamination: food scattering (top panel), the opossum moves the food pellets and hides them under the bedding, where they can come into contact with feces; food marking (bottom panel), the opossum defecates or urinates directly onto the food pellets or inside the food containers. (C) Top view and side view of 3 food types of food containers tested: paper tray, weighing boat, and ceramic dish (from left to right). (D) Cage setting for the 4 treatment groups: 1) food placed on the bedding, 2) food placed inside the weighing boat, 3) food placed inside the paper tray, and 4) food placed inside the ceramic dish.
Characterization of food scattering behavior in opossum maintenance cages under 4 different feeding approaches. (A) Two photographs of the same opossum, taken consecutively, demonstrate examples of opossum food scattering/hiding behaviors. (B) Estimated survival probability (inverse probability of food scattering events) using ceramic dish (green), paper tray (blue), weighing boat (orange), and no container (red). Statistical significance was determined using a log rank test (ns, not significant with P > 0.05). The shaded bands in the graph represent 95% confidence intervals.
Characterization of food marking behavior in opossum maintenance cages under 4 different feeding approaches. (A–D) Examples of fecal marking contaminations using no container (A), weighing boat (B), paper tray (C), and ceramic dish (D). Contamination levels 2 and 4 are shown. (E and F) Observations of food marking in the ceramic dish container revealed caked food at the bottom of the dish (E) caused by urine marking, and accumulation of fecal piles near the dish (F). (G) Estimated survival probability (inverse probability of food-marking events) using ceramic dish (green), paper tray (blue), weighing boat (orange), and no container (red). Statistical significance was determined using a log rank test (*, P ≤ 0.05; †, P ≤ 0.01; ‡, P ≤ 0.001; §, P ≤ 0.0001; +, P ≤ 0.005; ×, P ≤ 0.0005). The shaded bands represent 95% confidence intervals.
Survival curves demonstrating time to food contamination by food scattering or marking under 4 different feeding approaches. Estimated survival probability (inverse probability of food scattering or marking events) using ceramic dish (green), paper tray (blue), weighing boat (orange), and no container (red). Statistical significance was determined using a log rank test (*, P ≤ 0.05; †, P ≤ 0.01; ‡, P ≤ 0.001; §, P ≤ 0.0001; +, P ≤ 0.005; ×, P ≤ 0.0005). The shaded bands represent 95% confidence intervals.
Contributor Notes
This article contains supplemental materials online.