Establishment of the body condition score for adult female Xenopus laevis

The assessment of animals’ health and nutritional status using a Body Condition Score (BCS) has become a common and reliable tool in lab-animal science. It enables a simple, semi-objective, and non-invasive assessment (palpation of osteal prominences and subcutaneous fat tissue) in routine examination of an animal. In mammals, the BCS classification contains 5 levels: A low score describes a poor nutritional condition (BCS 1–2). A BCS of 3 to 4 is considered optimum, whereas a high score (BCS = 5) is associated with obesity. While BCS are published for most common laboratory mammals, these assessment criteria are not directly applicable to clawed frogs (Xenopus laevis) due to their intracoelomic fat body instead of subcutaneous fat tissue. Therefore, this assessment tool is still missing for Xenopus laevis. The present study aimed to establish a species-specific BCS for clawed frogs in terms of housing refinement in lab-animal facilities. Accordingly, 62 adult female Xenopus laevis were weighed and sized. Further, the body contour was defined, classified, and assigned to BCS groups. A BCS 5 was associated with a mean body weight of 193.3 g (± 27.6 g), whereas a BCS 4 ranged at 163.1 g (±16.0 g). Animals with a BCS = 3 had an average body weight of 114.7 g (±16.7 g). A BCS = 2 was determined in 3 animals (103 g, 110 g, and 111 g). One animal had a BCS = 1 (83 g), equivalent to a humane endpoint. In conclusion, individual examination using the presented visual BCS provides a quick and easy assessment of the nutritional status and overall health of adult female Xenopus laevis. Due to their ectothermic nature and the associated special metabolic situation, it can be assumed that a BCS ≥3 is to be preferred for female Xenopus laevis. In addition, BCS assessment may indicate underlying subclinical health problems that require further diagnostic investigation.


Introduction
Xenopus laevis, the Smooth or South African clawed frog, belongs to the genus Xenopus and the family Pipidae and is classified in the order Anura. There are numerous Xenopus species a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 In general hygienic and species-appropriate husbandry conditions, adequate feeding, qualified personnel, and correct handling of the animals are assumed to be the essential prerequisites for a stable state of health and body condition of the animals. However, for this species, species-specific criteria are still needed to determine, assess, and evaluate possible severity in animal experiments. It must be possible to determine the severity of an individual specimen at the actual time and during the experiment, considering its species-specific anatomy, physiology, and behavior. In principle, the preparation of this severity assessment is required by law as a prospective assessment already in the experimental design (e.g., Annex VIII EU Directive 2010/63 or § 7 para. 1, no. 2, § 8 section. 1, no. 7 German Animal Welfare Act, § 17) [26,27].
Here, animal experiments commonly use species and protocol-specific score sheets as assessment tools. In Germany, recommendations for severity assessment in animal experiments and explicitly for clawed frogs are made by the Committee for Animal Welfare Officers of the GV-SOLAS [28].
In general, different criteria are required for the assessment of an individual [29], e.g.: • assess the animal with as little stress as possible • identify and rank the expected signs of severity using predefined criteria • can be applied by various personnel in the same way • be able to record and depict the chronological course of severity that occur • enable predictions of potential, expected, and subsequent severity • enable documentation of progress controls in the case of treatment • Definition of termination criteria or humane endpoints.
At the same time, the quantitative, qualitative, and temporal criteria to be investigated must be: • adapted to the experimental design (here, e.g., surgical intervention with wound suture or injection for egg laying) • chosen to recognize severity caused by changes in individual behavior or organ systems (e.g., skin changes, lethargy, emaciation).
For a proper severity assessment of Xenopus laevis, compliance with husbandry and hygiene regulations is indispensable. Likewise, a basic knowledge of physiology, anatomy and the biological activity pattern of the test animal needs to be assessed [30]. Here, the body condition score is another standard tool for severity or health assessment. This score determines the nutritional status based on palpable bone landmarks and subcutaneous fat depots and classifies animals into 5 nutritional levels, ranging from emaciated up to obese.
In other laboratory animal species, the use of those species-specific body condition scores (BCS) is already established (zebrafish, mouse, rat, rabbit, cats, dog, sheep, pig, NHP) [31][32][33][34][35][36][37][38][39][40], but there is still no BCS for clawed frogs. It should be noted that Xenopus do not have palpable subcutaneous adipose tissue, as their only fat body is intracoelomic [41]. Thus, standard examination criteria are difficult to transfer to the frog due to its particular anatomy. The score described here represents a simple approach for the experimental refinement of these animals within procedures.
The study aimed to establish a scoring system to assess frog condition using non-invasive and objective assessment criteria. This score can be used as an additional tool in the scoring process and routine animal check-ups, e.g., as an individual health status indicator. Further, we investigated whether the classification to a BCS group depends on the body weight or the length of a Xenopus laevis.
Due to the experimental approaches of egg and oocyte harvesting by the research groups working with the animals assessed here, the present study used female adult Xenopus laevis animals to establish the body condition score.

Animals and husbandry
The animals were bred by the commercial breeder Xenopus 1 (Xenopus-I Inc, Michigan, USA) as lab-breed animals for experimental purposes. They were ordered with a body size of at least 12 cm in length (nose to cloaca). This assumed that all animals were sexually mature and of adult age and could be used for egg-laying protocols after acclimation at the experimen- The animals used were kept in three different husbandry systems: 1. individual water tanks (length x width x height (LxWxH): 115x55x36 cm), no water conditioning; average water temperature 12˚C ±3˚C, n = 20 animals.
2. circulating, a semi-closed system with 4 tanks arranged in parallel (Aqua Schwarz GmbH, Göttingen, Germany; LxWxH: 99x50x21 cm); conditioning of water by mechanical filtration systems and UV radiation; average water temperature 20˚C ± 1˚C, n = 20 animals 3. circulating, a semi-closed system with 17 tanks arranged in parallel (Aqua Schwarz, Göttingen, Germany; LxWxH: 100x50x25 cm); conditioning of water by mechanical filtration systems, UV radiation, demineralization, salting, and activated carbon filtration; average water temperature 20˚C ± 1˚C, n = 22 animals.
All animals were marked for individual identification with numbered metal tags (mouse ear tag, ITEM: 56779, Stoelting Co., Ireland) inserted into the webbing of the lateral outer toes of the right hind leg.
A total of 62 female adult Xenopus laevis were assessed within this study. The examined animals were randomly selected, whereas at least 20 of each husbandry system were used. All animals were fed every working day with a formulated dry pellet diet (Ssniff Spezialdiäten GmbH, Soest, Germany, Alleinfuttermittel für Krallenfrösche, V7106-030, 5 mm extrudate, 3 pellets per animal). Feed that was not immediately ingested was removed after 30 minutes. Within one day (24h), 3-5% of the total water volume was replaced by fresh water by continuous rinsing. The day-night cycle was 12 hours each. The monitoring of the tank system and the animal population, as well as the corresponding documentation of the water quality, took place daily.

Body Condition Scoring (BCS)
The presented BCS is based on established concepts of body condition scores of other species [31-34, 38, 39, 42], including Xenopus laevis-specific physical characteristics of its body. The classification is made in 5 ascending score levels (Figs 1 and 2), whereby BCS 1 is associated with a humane endpoint due to emaciation (Fig 3), and BCS 5 is regarded as a very well-conditioned physical state. The shape of the upper, lower, and lateral body silhouette, muscling and fleshing of the hind limbs, and visibility/protrusion of the thoracic and pelvic bone points are considered in the evaluation. The following assessment criteria are distinguished and defined for the respective BCS levels (S1 Fig).

Body weight and metrics
To assess the animals, they were individually removed from their housing tank with a fishing net, placed in an empty bowl, and weighed. Then the body dimensions were determined with a flexible ruler (nose-cloaca length, NCL). Hereafter, the bowl was filled with water and scored and photographed from above and from the side. Here, the water was taken from the animals' housing tank to avoid distress for the animal.

Statistics
Statistical analyses were performed in R (v4.0.3) [43]. Data were tested against the hypothesis of normal distribution using the Shapiro-Wilk test and were visually inspected with a QQplot. Normally distributed body weight data were analyzed in linear regression with treatment contrasts (BCS = 5 as the intercept level). BCS-class differences were represented as relative coefficient differences (planned contrasts) and were reported as estimates (β) with the corresponding standard errors (SE) as well as the p-values. Between BCS-class differences were analyzed with a one-way analysis of variance, followed by Tukey-Kramer post hoc tests to adjust for multiple comparisons. Post hoc test results were expressed as marginal means. In the nose-cloaca length (NCL) analysis, the body weight was analyzed as a fixed effect and for its interaction with BCS (bw:BCS). Both models were compared in a likelihood-ratio test using the lmtest package [44]. The following p-values were considered significant at the levels: p<0.05 (*), p<0.01 (**), and p<0.001 (***). Data were visualized with the ggplot2 package [45].

Results
A total of 62 adult female Xenopus laevis were assessed within this study. While BCS 1 is accompanied by a humane endpoint and should not occur in an ideal healthy animal population, this score was observed once. The BCS 2 group comprised 3 frogs, and the BCS 3 group included 16 animals. In comparison, BCS groups associated with good up to optimal nutritional status were found in 18 (BCS4) and 24 (BCS5) animals. The absolute body weights of the animals ranged from 83 g to 235 g. The allocation to the respective BCS groups is shown in Table 1. The distribution of assessed BCS within the different housing conditions and their effects on BCS are provided in the supplemental materials (S1 File).
The frog nose-cloaca length (NCL) was modeled as a function of the body weight in linear regression. Fig 4B shows the linear regression line of the model with its 95% confidence band. The model was significant (F(1,59) = 119.9, p<0.001, R adj = 0.665), and the average NCL was estimated as 8.82 cm. In addition, the continuous body weight variable contributed   Fig 4B, ranging from pink (BCS = 1) to dark green (BCS = 5). Lower BCSs appeared to relate to shorter body lengths and lower weights. However, the between-class transitions were diffuse. Therefore, the linear model was expanded by the bw:BCS interaction to estimate the BCS's independence from body weight (Table 3) The model with the interaction term was significant (F(4,56) = 40.46, p<0.001, R adj = 0.725) and showed a significant body weight coefficient (β bw = 0.025, SE = 0.45, p<0.001). Additionally, the interactions for bw:BCS4 (β bw:BCS4 = 0.002, SE = 0.001, p = 0.046) and bw: BCS3 (β bw:BCS3 = 0.004, SE = 0.001, p<0.001) were significant, indicating interdependence from both variables, body weight and the BCS. The interaction of bw:BCS2, however, was not significant at the p�0.05 threshold but at the p�0.1 level (β bw:BCS2 = 0.005, SE = 0.003, p<0.0921), still providing evidence for the interaction of body weight and BCS with lower sample sizes (n BCS2 = 3). The significant bw:BCS interaction term requires both variables, body weight, and BCS, to model the NCL.
Finally, the body weight and interaction models were compared in a likelihood-ratio test to evaluate whether the more complex model adds any benefit. The model with the bw:BCS interaction is more complex with 6 degrees of freedom, compared to the model with body weight only (df = 3). Adding the interaction term leads to a significantly better-performing model (Χ 2 = 15.21, p = 0.002), so the hypothesis of restricting the model to a more straightforward design can be rejected.

Discussion
The first Body Condition Score (BCS) for adult female Xenopus laevis was developed and introduced in this study. This BCS was intended to be used for severity assessment in animal husbandry and as an additional parameter in the preparation of protocol-specific score sheets [46] and health examination of adult female Xenopus laevis.
With increasing age (�20 years) [14], however, egg quality decreases [15], so that on the one hand, the limited productive life of the animals in the trials is justified. In addition, the animals are used only a limited number of times for egg production to not exceed a moderate degree of severity. Therefore, biological age is probably not generally reached in the experimental use of clawed frogs. However, since only body length and not the age of the animals are decisive for the purchase by the breeder, no predictions can be made regarding the impact of age on body weight and BCS in the present study.
Hence, the BCS represents the physical status in relation to the individual's body size, independent of other factors. In previously published body condition scores of other species, high score values are usually associated with obesity and thus possible pathological overconditioning of the animals [47,48]. These body condition score systems distinguish between emaciated and obese animals and therefore define a mid-level as well-conditioned [31,32] or even as optimum [34,40,49]. Our data indicate that in Xenopus laevis the BCS 4 and 5 are above a physiological optimum but are not associated with a pathological or avoidable condition and thus represent the desirable condition. This is justified by the fact that due to their particular anatomy and adaptive (ectothermic) metabolism, the animals should be physically resilient to withstand experimental severity (e.g., induced laying of eggs, surgical ovariectomy) [41,50,51]. However, care must be taken to distinguish appropriate disease symptoms that cause a biased body shape from the expected condition, e.g., lymph sac edema (Fig 5). The lymphatic sac edema is a pathological clinical phenomenon in clawed frogs [52], which must be distinguished from healthy, obese animals. It should be taken into account that in the dorsal view, the body shape of animals of BCS 5 may be similar to the body shape of frogs with edema of the dorsal lymph sac. This assumption is justified by the fact that the body composition, ratio, and anatomical position of fat and muscle mass differ from that of mammals. There is no palpable subcutaneous fat tissue which allows conclusions to be drawn about the general nutritional status [41]. Instead, in this species, a large digitiform fat body exists inside the body cavity (coelomic cavity) and can, therefore, neither be assessed visually nor palpated during the examination. At this point, however, it is impossible to determine if the dimension of the fat body correlates with the actual body weight and which size the fat bodies of the animals of different BCS groups have. This would require imaging techniques (e.g., CT or ultrasound) or at least surgical resection of the fat body, which is impossible in the context of the experiments presented here. Another assessment criterion for the physical health condition is measuring the actual body weight and dimensions of the individuals. Both factors can fluctuate significantly due to the intervention (egg laying, surgery, health status, etc.), so this parameter should be evaluated as far as possible during the assessment and always only in combination with the BCS. Depending on the method used to obtain the oocytes/eggs (induction by injection or surgical removal of the ovary), this may affect the animal's body weight. However, during the BCS assessment, these interventions do not affect body length or hindlimb fleshiness (Figs 1 and 2). Since this intervention was performed in all animals evaluated here, this systematic error can thus be neglected. The absolute body weights of the animals (ranging from 83 g to 235 g) showed that the different allocations of the BCS are associated with varying classes of weight in each case (Table 1). Nevertheless, individual animals belong to the lower or upper limit range or could be classified in a different body condition group merely due to their weight (Fig 3).
Considering the determined values, 33% of the animals show a BCS <3, which does not correspond to the desired optimum. This can be explained by the factors mentioned above (time since the last spawning, husbandry conditions, metabolic activity, and age). In case there are hardly any animals with BCS 4 or 5 in husbandry, an optimization of the housing conditions and a detailed evaluation of the animal population (health monitoring, age, and convalescence phases after interventions) are urgently recommended.
In the performing facility, animals with a BCS 2 are separated, fed, and their weight is monitored strictly (3 times per week). Animals showing a BCS 1 are excluded from any experiment (humane endpoint).
However, assessing a clawed frog with the presented body condition score must always be carried out from two optical perspectives: A lateral view must supplement the information from the dorsal view. This is because breathing and the motion of the hind legs can affect the body's contour. While the external body shape can be easily distorted by inhalation (ventilation of the lungs increases the convexity of the dorsal contour), the extension of the hind limb elongatess the body shape and its body shape appears less pear-shaped. Therefore, the animals are assessed freely-moving, motionless, and relaxed in a water basin to assess BSC values.
It is essential to know that neither the absolute body weight of an animal nor its body dimensions are sufficient to assess the condition of a clawed frog as a single evaluation criterion. These must always be obtained and interpreted in relation to the BCS of the individual.

Conclusion
It could be demonstrated that the different allocations of the BCS are associated with varying classes of body weight in each case (Table 1). Nevertheless, individual animals belong to the lower or upper limit range or could be categorized in a different body condition group merely due to their weight (Fig 4). The Body Condition Score is therefore intended as a simple, objective tool for personnel involved in experiments and animal caretaker staff to assess conditional changes in Xenopus laevis in chronic experiments (e.g., with a focus on oocyte collection) or even in mere animal husbandry, and can classify the physical condition of the animals in general. In addition, this assessment tool tries to define humane endpoints for this underestimated species based on images, to avoid unnecessary animal suffering (experimental refinement) (Fig 3). Since the interaction term of bw:BCS is significant and improves the model, we found evidence for the usefulness of measuring both the body weight and the BCS. Therefore, it is essential to consider body weight, body shape, and condition when assessing Xenopus laevis. However, what effect the age of the animals has on the occurrence of the respective BCS groups and what influence the BCS might have on the quality of oocytes and eggs cannot be addressed at this time. Therefore, the BCS is seen as a purely descriptive tool for daily animal care and routine examination of individual animals within laboratory animal husbandry. It also remains to be investigated to what extent the BCS for female Xenopus laevis can be transferred to male animals.