Development of mycobacterial growth inhibition assays for early evaluation and gating of novel TB vaccine candidates
Rachel Tanner, University of Oxford
Tuberculosis (TB) continues to pose a serious global health threat, with an estimated 9 million new cases and 1.5 million deaths annually. The current vaccine, BCG, is not always protective – particularly in areas where TB is most prevalent. Developing a new vaccine is extremely difficult because it is not clear which parts of the immune response should be measured to demonstrate protection. Without any alternative, candidate vaccines are currently tested using animal models such as mice and non-human primates. Large numbers of animals are required and must be infected with M.tb (the causative agent of TB): a procedure classified as ‘Moderate’ in severity by the Home Office. As TB disease progresses, animals may experience loss of body weight, fever and respiratory distress.
Rachel Tanner, under the supervision of Professor Helen McShane, University of Oxford, was awarded a 3Rs Liaison Group Studentship in 2012 to carry out a PhD project concerned with the optimisation and evaluation of an in vitro alternative to infecting animals with M.tb to test TB vaccines.
Blood or cells can be infected with M.tb (or BCG as a safer surrogate) in a test tube, and the amount by which the blood or cells inhibit growth of the bacteria quantified. If inhibition is measured before and after vaccination, vaccine protection may be estimated. The first step involved ensuring the assay had acceptable levels of reproducibility, and many different parameters were compared and tweaked to offer the best chance of detecting differences. Secondly, proof-of-concept was required to show that, where a vaccine was known to be protective, a corresponding response could be demonstrated with the assay. This was achieved using samples from BCG-vaccination studies in UK adults, macaques and mice. Finally, the assay underwent biological validation (to demonstrate that it is a meaningful representation of what would happen in live animals or people). Encouragingly, a correlation was observed between the in vitro response and in vivo protection from challenge in the mouse and macaque models. There was also concordance between the assay and epidemiological data or in vivo outcome in humans.
This work has generated considerable interest including inclusion of the assay in TB biomarker consortia packages, workshops and transfer of the technology to other laboratories worldwide. With further development, the in vitro assay represents a potential correlate of protection that may be game-changing in preclinical vaccine testing, reducing the number of animals used in challenge experiments and accelerating the development of an effective vaccine.
Dr Tanner was awarded her PhD in 2015.
Published papers arising from Rachel’s work
Tanner R, O’Shea MK, White AD, Müller J, Harrington-Kandt R, Matsumiya M, Dennis MJ, Parizotto EA, Harris S, Stylianou E, Naranbhai V, Bettencourt P, Drakesmith H, Sharpe S, Fletcher HA, McShane H. 2017. The influence of haemoglobin and iron on in vitro mycobacterial growth inhibition assays. Nature Scientific Reports 7, Article number: 43478. Doi:10.1038/srep43478
Tanner R, O’Shea MK, Fletcher HA, McShane H. 2016. In vitro mycobacterial growth inhibition assays: A tool for the assessment of protective immunity and evaluation of tuberculosis vaccine efficacy. Vaccine. 34(39): 4656-4665. https://doi.org/10.1016/j.vaccine.2016.07.058
Tanner R, and McShane H. 2016. Review Article: Replacing, reducing and refining the use of animals in tuberculosis vaccine research. ALTEX Online, first published 26 Sept 2016. http://dx.doi.org/10.14573/altex.1607281.
Harris S, White A, Stockdale L, Tanner R, Dennis M, Sibley L, Sarfas C, Meyer J, Peter J, Satti I, Manjaly Z, Hamidi A, McShane H, Sharpe H. “Optimisation of a non-human primate BCG challenge model for selection of candidate tuberculosis vaccines.” (submitted to Journal of Infectious Diseases).
Pepponi I, Katri B, Tanner R, Villarreal-Ramos B, Vordermeier M, McShane H. “A mycobacterial growth inhibition assay (MGIA) for bovine TB vaccine development.” (submitted to Transboundary and Emerging Diseases).
Employing the social amoeba, Dictyostelium, as a first pass screen in drug development
Steven Robery, Royal Holloway College, University of London
A major problem when developing new drugs for human (or animal) use is that potential drugs can cause nausea or vomiting (emesis) as a side effect. In fact, emesis is reported as a possible side effect in approximately one third of medicines. Whilst the physiological mechanisms behind the emetic response are well characterized, the diverse range of stimuli that can generate the response indicate the molecular mechanisms involved are poorly understood.
Rats, mice, ferrets, dogs, house musk shrews and non-human primates are commonly used as models in emetic research. In clinical (human) studies nausea and vomiting are perceived as unpleasant and it is very likely that animals experience similarly unpleasant feelings when undergoing emetic testing.
In an effort to reduce the number of animals required for testing emetic drugs, in 2009 the UFAW 3Rs Liaison Group supported a project by Mr Steven Robery (under the supervision of Robin Williams and Paul Andrews) at Royal Holloway, University of London. Steven examined the possibility of using the soil-living amoeba Dictyostelium discoideum as a preliminary screen for emetic liability in drug development in order to reduce or replace animal use in this area.
A diverse range of emetic compounds were investigated and the effect on Dictyostelium discoideum behaviour was monitored. Steven found that a small number of known emetic compounds strongly inhibited cell migration in a concentration-dependent and reversible manner, and thus provided specific families of emetic/aversive compounds that were suitable for more detailed investigation. These active compounds included a range of bitter compounds, including phenylthiourea and naringenin, and the hot tastant capsaicin.
Figure 1: Analysis of Dictyostelium cell behaviour.
Cells moving under a chemotactic gradient were analysed using ImagePro software to determine cell velocity (μm/sec); cell aspect (shape measured as a ratio between the length of cells across each axis); cell angle (degrees-where cell migration was measured in comparison to the y-axis); and cell tracking (where the coordinates of cell pathways were illustrated following normalisation to (0,0) at 5 minutes) in order to illustrate changes in migration before and after compound addition.
Analysis of the bitter tastant phenylthiourea, identified a poorly characterised receptor for this compound in Dictyostelium. This receptor was analysed in Dictyostelium as a potential cellular mechanism for bitter tastant receptor activity. This study then translated the discovery of this receptor to a human context, by identifying a poorly characterised human receptor (part of the inhibitory neurotransmitter receptor for GABA) that is responsible for bitter tastant function. Analysis of another bitter tastant, naringenin (a flavonoid), also identified a poorly characterised ion channel responsible for naringenin function using Dictyostelium as an animal replacement model. The discovery of this channel as a target for naringenin was then characterised in mammalian cell lines, to confirm the functionality of this interaction in mammalian models.
Overall, Steven concluded that because only seven of the twenty-nine compounds with known emetic effects in mammals affected Dictyostelium behaviour, its use as a general model for predicting emetic liability is limited. However, his research identified four different sub-groups of compounds that blocked chemotaxis, including a range of bitter tastants. He concluded that Dictyostelium may therefore provide a useful model in the analysis of these compounds in structure-activity relationships or translational studies to identify novel active vanilloids or bitter tastants. Using Dictyostelium, compounds structurally related to bitter tastants such as denatonium benzoate may also help to identify other bitter and non-bitter analogues, which may target specific bitter taste receptors, and in the characterisation of bitter tastants potency in translational studies. The project led to further industry collaboration to potentially reduce the use of sentient animals (ferrets, dogs and non-human primates) in procedures classified as of moderate severity under the Animals (Scientific Procedures) Act 1986.
Dr Robery was awarded his PhD in 2013.
Published papers arising from Steven’s work
Waheed A, Ludtmann MHR, Pakes N, Robery S, Kuspa A, Dinh C, Baines D, Williams RSB and Carew MA. 2014. Naringenin inhibits the growth of Dictyostelium and MDCK-derived cysts in a TRPP2 (polycystin-2)-dependent manner. British Journal of Pharmacology. 171(10): 2659-2670. https://dx.doi.org/10.1111/bph.12443.
Robery S, Tyson R, Dinh C, Kuspa A, Noegel AA, Bretschneider T, Andrews PL & Williams RS. 2013. A novel human receptor involved in bitter tastant detection identified using the model organism Dictyostelium discoideum. Journal of Cell Science. 126: 5465-5476. http://dx.doi.org/10.1242/jcs.136440.
Robery S, Mukanowa J, Percie du Sert N, Andrews PLR. & Williams RSB. 2011. Investigating the effect of emetic compounds on chemotaxis in Dictyostelium identifies a non-sentient model for bitter and hot tastant research. PLoS One 6(9): e24439. http://dx.doi.org/10.1371/journal.pone.0024439.
Refining research procedures by assessing distress in laboratory rodents
Claire Richardson, Newcastle University
In 2007 Claire Richardson was awarded a 3Rs Liaison Group Studentship to develop methods to improve objective methods of assessing the welfare and emotional state of laboratory mice, the most commonly used laboratory animal. Ms Richardson undertook her project at Newcastle University under the supervision of Professor Paul Flecknell.
Determining the affective state of animals, or how they ‘feel’, is essential if we are to make meaningful judgements about their welfare and to assess the value of measures aimed at improving their welfare. Emotional states such as fear and anxiety are typically assessed in laboratory rodents by measuring the responses of individual animals in standardized behavioural tests, such as the elevated plus maze. Although these tests are widely used, individual testing of animals can be stressful and repeated testing of the same animal can be problematic and time consuming.
Claire was interested in investigating how mouse behaviour could be measured through using non-invasive automated technology, in particular Claire focused on the Intellicage® – a transponder-based automated home-cage that can collect and record certain behavioural data (e.g. activity levels, nosepokes, drinking). The capacity to non-invasively study rodents in a social setting would have scientific, animal welfare and economic advantages.
Claire undertook a series of three experiments: 1) developing a task to examine cognitive biases in an automated homecage, 2) examining measures from the automated homecage that may relate to standardised measures of well-being, and 3) studying analgesic (pain-relief) seeking behaviour in mice with cancer.
During the first series of experiments, anxiety-like behaviour from 32 individual mice from two commonly used strains of inbred wild-type laboratory mice was assessed using standardized behavioural tests. The same animals were then examined whilst group-housed using a novel assessment technique: response to unpredictable, mild airpuffs in an automated homecage. Claire found that individual mouse behaviour in the emergence test was positively correlated with individual response to airpuffs, e.g. individuals with the greatest latency to leave a shelter in the emergence test showed the greatest response to airpuffs. The findings of this study suggest that examining reactivity to unpredictable, mild airpuffs may be a useful technique to non-invasively assess fear and anxiety in socially housed animals.
In the second series of experiments mice with three types of cancer (bladder, subcutaneous and intramuscular) were studied in the automated homecage. Although mice are frequently used as models in cancer research, relatively little is known about which studies are painful and when pain/distress is most likely to occur. Subtle behavioural changes detected by the homecage preceded clinical signs of disease in mice with bladder and intramuscular cancer. These findings have implications for the implementation of humane endpoints in cancer research. If exploratory changes in behaviour reliably precede clinical disease it may be possible to humanely kill animals prior to the onset of pain or distress without compromising the scientific outcomes of many studies.
With the aim of developing methods for the objective assessment of cancer pain, in addition to examining the spontaneous behaviour of mice with cancer within in automated homecage, analgesic self-administration and place preference were also studied. This work provided evidence that quickly developing tumours were likely to be associated with acutely painful periods. Mice with quickly developing tumours were significantly more likely to approach a homecage location associated with morphine compared to control animals. The strength of the animals with cancer peak preference for the morphine location was also positively correlated with the speed of the tumour growth.
Claire concluded that this project provided evidence of a variety of ways in which an automated system can be used to monitor the welfare of laboratory mice. A further advantage is that this intensive monitoring takes place non-invasively in socially housed animals within enriched homecages.
Dr Richardson was awarded her PhD in 2012
Published papers arising from Claire’s work
Richardson CA. 2015. The power of automated behavioural homecage technologies in characterizing disease progression in laboratory mice: A review. Applied Animal Behaviour Science. 163: 19-27. https://doi.org/10.1016/j.applanim.2014.11.018.
Coulter CA, Flecknell PA, Leach MC, Richardson CA. 2011. Reported analgesic administration to rabbits undergoing experimental surgical procedures. BMC Veterinary Research. 7: 12. DOI: 10.1186/1746-6148-7-12.
Miller AL, Richardson CA. 2011. Rodent Analgesia. Veterinary Clinics of North America: Exotic Animal Practice, Analgesia and Pain Management. 14(1): 81-92. https://doi.org/10.1016/j.cvex.2010.09.004.
Leach MC, Coulter CA, Richardson CA, Flecknell PA. 2011. Are We Looking in the Wrong Place? Implications for Behavioural-Based Pain Assessment in Rabbits (Oryctolagus cuniculi) and Beyond? PLoS ONE 6(3): e13347. https://doi.org/10.1371/journal.pone.0013347
Coulter CA, Flecknell PA, Richardson CA. 2009. Reported analgesic administration to rabbits, pigs, sheep, dogs and non-human primates undergoing experimental surgical procedures. Laboratory Animals. 43(3): 232-238. DOI: https://doi.org/10.1258/la.2008.008021.
Stokes EL, Flecknell PA, Richardson CA. 2009. Reported analgesic and anaesthetic administration to rodents undergoing experimental surgical procedures. Laboratory Animals. 43(2): 149-154. DOI: 10.1258/la.2008.008020.
Leach MC, Allweiler S, Richardson C, Roughan JV, Narbe R, Flecknell PA. 2009. Behavioural effects of ovariohysterectomy and oral administration of meloxicam in laboratory housed rabbits. Research in Veterinary Science. 87(2): 336-347. DOI: 10.1016/j.rvsc.2009.02.001
Stress, sex and memory in laboratory rats
Anjanette Harris, University of Edinburgh
Ms Anjanette Harris was awarded a 3Rs Liaison Group Scholarship in 2005 for her project looking into stress, sex and memory in laboratory rats. Anjanette undertook her project at the University of Edinburgh and was supervised by Dr Susan Healy and Dr Rick D’Eath.
In 2004, of the 2.85 million scientific procedures carried out, 16% were on rats (mice were the most commonly used animal species and were the animal model of choice in 67% of experimental procedures). Typically, rats are housed in small, barren, ‘standardised housing’ which is used to reduce environmentally-induced variation both across and within laboratories. Additionally, rats may be housed in isolation, which can lead to behavioural and physiological changes indicative of impaired welfare. Importantly, factors that influence laboratory rat welfare may also undermine the reliability of experimental data. For example, rodents housed under stressful conditions may be cognitively impaired, which can invalidate research on learning and memory, which uses these animals.
Much of our understanding of spatial cognition (e.g. the cues animals use to navigate and the neurobiological basis of spatial cognition) comes from studying learning and memory in laboratory rats. Anjanette was interested in exploring whether there are sex differences in rats undergoing spatial cognition tests, since, in a number of mammalian species, males outperform females in tests of spatial ability. In particular, Ms Harris was interested in studying whether a variation in stress levels may be the cause of any differences as there is some evidence to suggest that females respond more poorly to acute stress, such as is imposed by a test situation, and yet their spatial performance may be unchanged or enhanced by chronic stress. Male cognitive abilities, on the other hand, may be adversely affected under conditions of chronic stress.
Anjanette’s research focussed on the impact of housing conditions (isolation, enrichment) on spatial cognition in male and female rats. Specifically, Anjanette sought to answer four questions: 1) does isolation housing impair cognition in rats? And are the sexes affected differently? 2) Is isolation without visual contact more stressful than isolation with visual contact? 3) Does environmental enrichment reduce stress during testing? And 4) Is it stressful to remove social enrichment? And can physical enrichment alone ameliorate stress caused by the removal of social enrichment? If the sexes do differ in the way they respond to housing, then it may be beneficial for males and females to be housed in different ways.
Ms Harris tested cognition using a Morris water maze (MWM; see Figure 1), which is one of the most frequently used spatial tests. Anjanette aimed to determine whether chronic isolation stress acts in concert with acute stress (associated with the MWM) and whether males and females are affected to the same degree. She hoped to determine if housing conditions could cause sufficient stress so as to affect the outcome of a cognitive experiment i.e. looking for sex differences in spatial cognition. Anjanette used behavioural measures of stress e.g. thigmotaxis (wall-hugging), to assess acute stress during testing, and bar biting in the home cage, to assess stress associated with different housing conditions. Body weight and food intake were also monitored as additional measures of welfare (both typically increase in isolated rats).
Figure 1: The Morris water maze (MWM) consists of a circular pool (approx 2m in diameter) filled with opaque luke-warm water. Located under the surface of the water is an escape platform (8cm in diameter). A rat is released into the maze at the edge of the tank and swims around until it encounters the platform. Upon re-release the rat uses extra-maze cues posters on the wall etc) to locate the platform. The most commonly used measure of performance in an MWM is the time taken to reach the platform.
Anjanette found that there were no significant effects of isolation housing on behaviour in the home cage or on cognitive performance in the MWM in six experiments. Furthermore, visual contact between neighbouring cages and the holding room seemed to be more important to laboratory rats than social housing. Crucially, Anjanette considered that isolation housing did not explain the inconsistencies in reports of sex differences in spatial ability in the literature. However, stress due to the test situation could explain why males performed better than females in the MWM in her experiments. Indeed, since thigmotaxis confounds MWM performance, Anjanette proposed that measuring thigmotaxis may be a non-invasive refinement of MWM procedures and that MWM data should not be presented without consideration of thigmotaxis. Anjanette suggested that environmental enrichment (social and physical) enhanced cognitive performance in the MWM because it reduced thigmotaxis during testing. However, it was unclear which component of enrichment (physical or social) had the greatest beneficial impact on welfare or if the effects were the same for different strains, ages and sex of rat. These questions would need answering before encouraging widespread use of enrichment as a tool to improve welfare outwith the home cage.
Dr Harris was awarded her PhD in 2009.
Published papers arising from Anjanette’s work
Harris AP, D’Eath RB & Healy SD. 2010. A cage without a view increases stress and impairs cognitive performance in rats. Animal Welfare. 19(3): 235-241.
Harris AP, D’Eath RB & Healy SD. 2009. Environmental enrichment enhances spatial cognition in rats by reducing thigmotaxis (wall hugging) during testing. Animal Behaviour. 77(6): 1459-1464. http://dx.doi.org/10.1016/j.anbehav.2009.02.019.
Healy SD, Bacon IE, Haggis O. Harris AP & Kelley LA. 2009. Explanations for variation in cognitive ability: behavioural ecology meets comparative cognition. Behavioural Processes. 80(3): 288-294. http://dx.doi.org/10.1016/j.beproc.2008.10.002.
Harris AP, D’Eath RB & Healy SD. 2008. Sex differences in spatial cognition are not caused by isolation housing. Behaviour. 145(6): 757-778. http://dx.doi.org/10.1163/156853908783929142.
Harris AP, D’Eath RB & Healy SD. 2008. Sex differences, or not, in spatial cognition: acute stress is the key. Animal Behaviour. 76(5): 1579-1589. http://dx.doi.org/10.1016/j.anbehav.2008.07.016.
Assessing the practicalities and welfare implications of training laboratory primates
Verity Bowell, Stirling University
Assessing the practicalities and welfare implications of training laboratory primates
Photo credit: Jean McKinley
Verity Bowell became the third UFAW 3Rs Scholar in 2003. Verity was interested in primate welfare, specifically primates used in laboratory research, and how their welfare may be improved through positive reinforcement training. Verity undertook her PhD work at the University of Stirling, under the supervision of Dr Hannah Buchanan-Smith.
Ms Bowell noted that in 2001 there were nearly 4000 licensed scientific procedures carried out on primates in the UK (this figure was 3600 in 2015 according to Home Office Statistics). Procedures such as capture and blood collection have been, and still are, carried out using force and restraint, which is stressful for the animals involved. This not only compromises welfare, but also affects the results of the data collected due to physiological changes. One means of decreasing the aversiveness of laboratory procedures is to train animals using positive reinforcement training (PRT). PRT may be used with laboratory primates to encourage cooperation of the animal with laboratory staff by training animals to, for example, offer a limb for blood collection, give urine samples on request, or to co-operate in capture.
Verity found that although laboratories were increasingly interested in using positive reinforcement training for laboratory-housed primates, there remained a reluctance to put into practice training programmes. Much of the reticence seemed to stem from lack of expertise in the running of training programmes, and a perception that training required a large time investment, with concurrent staff costs.
The aim of Verity’s PhD work was to provide practical recommendations for the use of training programmes in laboratories, providing primate users and care-staff with background information needed to successfully implement training programmes whilst improving the welfare of the animals in their care.
Training was carried out with two species, cynomolgus macaques (Macaca fascicularis) – also known as long-tailed or crab-eating macaques – and common marmosets (Callithrix jacchus) in three different research laboratories to ensure practicability was as wide ranging as possible.
Figure 1: Marmoset holding target whilst in transport box
Training success and the time investment required were closely related to the primate’s temperament, most notably an individual’s willingness to interact with humans, in both common marmosets and cynomolgus macaques. Age and sex however had no effect on an individual’s trainability. The training of common marmosets was more successful than that with cynomolgus macaques, possibly due to differences in early experience and socialisation. Positive reinforcement training helped both species to cope with the stress of cage change or cleaning, with the monkeys showing less anxiety-related behaviour following the training programme than before.
Involving two trainers in the training process did not affect the speed at which common marmosets learned to cooperate with transport box training, but behavioural observations showed that initial training sessions with a new trainer led to animals experiencing some anxiety. This, however, was relatively transient. Whilst the training of common marmosets to cooperate with hand capture was possible, there seemed little benefit in doing so as the monkeys did not show a reduced behavioural or physiological stress response to trained capture compared to hand capture prior to training. However, strong evidence was found that following both training and positive human interactions the marmosets coped better with capture, and stress was reduced.
On the basis of her findings, Ms Bowell recommended that an increased use of early socialisation would benefit laboratory housed primates, and would also help improve the success of training. Further, the time investment required showed that training is practicable in the laboratory for both species, and that positive reinforcement training is an important way of improving their welfare, likely through reducing boredom and fear.
Dr Bowell obtained her PhD in 2010.
Published papers arising from Verity’s work
Prescott MJ, Bowell VA and Buchanan-Smith HM. 2005. Training laboratory housed non-human primates, Part 2: Resources for developing and implementing training programmes. Animal Technology & Welfare, 4: 133-148.
Effects of cage size, space allowance, environmental enrichment and their interactions on behaviour, stress, immune function and welfare of laboratory mice
Effects of cage size, space allowance, environmental enrichment and their interactions on behaviour, stress, immune function and welfare of laboratory mice
In 2002 the second 3Rs Liaison Group studentship was awarded to Ms Kerry Westwood, under the supervision of Professor Mike Mendl, Professor Christine Nicol and Dr Chris Sherwin at the University of Bristol. Kerry’s research project was entitled: ‘Effects of cage size, space allowance, environmental enrichment and their interactions on behaviour, stress, immune function and welfare of laboratory mice’.
Mice have historically, and continue to be, the most commonly used animal in research. Of the 2.73 million scientific procedures started in 2002, 63% were on mice. Making improvements to the cage environment of mice used in a laboratory therefore has the potential to improve the welfare of a great many animals.
The first part of Ms Westwood’s study examined the effects of varying cage size (340 and 960cm2) and space allowance (60, 100 and 167cm2, per mouse) on behavioural, physiological, immunological, morphological and histopathological indicators of welfare in two strains of laboratory mice (one inbred and one outbred) between 5.5 and 11 weeks of age. Cage sizes, space allowances and mouse strains were chosen according to the Council of Europe proposals for rodent housing, and on the basis of a survey of current practice carried out at 10 commercial laboratories as part of the project.
The inbred mice yielded data on all three space allowances and both cage sizes in a 3 x 2 factorial design, while the outbred mice were too heavy to be housed at the lowest space allowance and so were studied using a 2 x 2 factorial design. Increasing both cage size and space allowance appeared to induce behavioural changes which may indicate better welfare (e.g. lower stereotypy, higher levels of activity and non-agonistic social interaction per individual). There were few suggestions of adverse effects, especially for the outbred animals which seemed to be more socially active than the inbred, and may thus have benefited more from the larger groups that accompany increased cage size, as evidenced by a slight reduction in their faecal corticosterone levels in larger cages. The inbred strain studied here appeared less socially active, and may have benefited less from the larger cage size in which their faecal corticosterone levels were higher, and their body weight lower. However, it is possible that these changes were related to their higher levels of activity in larger cages. There was also no evidence that they experienced extra aggression and related social stress in large cages. Furthermore, increasing space allowance (and hence decreasing group size) also appeared to lead to a decrease in corticosterone and an accompanying increase in growth rate in this strain. The clear strain differences in response to different cage sizes and space allowance emphasise that different strains may be affected in different ways by the same housing environment. It should also be noted that the cage sizes and space allowances studied were relevant to current commercial practice, and it is possible that the use of much larger cages or greater space allowances may have had led to more pronounced effects on the welfare indicators measured.
The second part of the project investigated the effects of providing an enrichment object (a ‘hammock feeder’ designed on the basis of results from another study) on ICR(CD-1) male and female mice housed in single sex groups of three at a 100cm2 per mouse space allowance in small (340cm2) cages between 3.5 and 9 weeks of age. Many of the welfare indicators used in the first study were also measured in this study. The hammock feeder encouraged more physical and social activity, but it is not clear that this lead to an enhancement of welfare, as bar chewing, aggression and stereotypies increased as the study progressed. High intensities or increases in any of these behaviours would generally be taken to indicate that welfare could be improved. The hammock feeder also had only a minor influence on physiological indicators of welfare. It may be that the effects of the small cage over-rode any welfare benefits that the enrichment could offer, or that earlier (e.g. pre-weaning) provision of the hammock feeder may have been more effective. Females used the hammock feeder more than males, confirming previous reports of sex differences in the use of other enrichment objects by mice. A sex biased difference in the use of enrichments could have important consequences for the design and successful implementation of environmental enrichments.
Dr Westwood obtained her PhD in 2007.
Development of laboratory rabbit housing: meeting animal and industry needs
Shirley Seaman, University of Edinburgh
The first 3Rs Liaison Group Studentship to be awarded was in 1999 to Ms Shirley Seaman. Ms Seaman was interested in refining the housing of laboratory rabbits and carried out her project at Edinburgh University, under the supervision of Dr Natalie Waran, Dr Mike Appleby and Dr Rick D’Eath.
In the year 2000, approximately 39,700 scientific procedures were carried out on laboratory rabbits, this equates to around 1.5% of the 2.71 million procedures started that year. Where rabbits are used in research, it is important to refine their care where possible to improve welfare and the Home Office recommends that female rabbits should, wherever practicable, be housed in groups to satisfy their need for social contact. However, for experimental reasons, this is not always possible and rabbits may be housed singly. Where rabbits are housed singly, they have been found to perform stereotypic behaviours, such as bar-biting and fur pulling which can either lead to or be a result of poor welfare.
Ms Seaman was interested in assessing whether providing caged rabbits with minimal tactile and visual contact with conspecific rabbits in adjacent cages may be beneficial to welfare, and also the motivation of caged rabbits to gain access to a platform (rabbits housed in cages without a platform/bolt-hole have been found to be more restless and more easily affected by their environment than rabbits housed with a platform).
To investigate social contact and platform use, Shirley carried out a number of short and long-term motivational tests on singly and socially (pair) housed rabbits. In the short-term motivational tests to assess social contact, rabbits were given the opportunity to push through a weighted push-door in a runway (see Figure 1) to gain access to 1 minute of visual and minimal tactile contact through a mesh panel with an unfamiliar rabbit, a familiar rabbit, or nothing (control). The weight on the door was increased in stages to determine the maximum weights the rabbits would push through – it is thought that the greater an animal will ‘work’ to gain access to a resource, the more important that resource is to that animal.
For the long-term motivational test, rabbits were housed in plus-shaped sets of apparatus for the duration of the experiment (Figure 2). The ‘arms’ of the plus were termed resource cages and the central area was termed the home cage. Four resources were available in the resource cages: food, social contact through a mesh panel, a cage platform and an empty space. Food and an empty space were expected to be of high and low value respectively; the importance of social contact and a platform could therefore be compared to the importance of these resources. Each resource cage could only be entered through a one-way weighted push-door. Weights were increased every two days. The maximum weights pushed through for each resource were recorded, as were the number of times rabbits entered each resource cage at each weight and the total and mean duration of visits to each resource cage.
Additionally, the cage environment of ten socially (pair) caged and eleven singly caged rabbits was manipulated to investigate how platform use was affected by social stimuli from conspecifics. Singly caged rabbits were housed both with and without a platform, and both with and without visual and minimal tactile contact with the rabbit in the adjacent cage (through a mesh panel). Socially caged rabbits were housed in their pairs, next to their cage-mate and next to an unfamiliar rabbit, both with and without visual and minimal tactile contact.
Ms Seaman found that both singly and socially housed rabbits worked to gain visual and minimal tactile contact with both a familiar and unfamiliar conspecific. Singly housed rabbits pushed through heavier weights than the socially housed rabbits, suggesting that singly housed rabbits were more motivated for contact than previously socially housed rabbits.
Olfactory cues were found to be important during the short-term test on singly housed rabbits – several of the rabbits did not push through the push-door in the control when there was no rabbit at the end of the runway. However, when further trials were carried out with an unfamiliar rabbit at the end of the runway, these rabbits pushed through the push-door again. This suggests that if rabbits can smell and hear other rabbits, they are motivated to gain visual and minimal tactile contact with them. Olfactory cues were also found to be important in the experiment with the socially housed rabbits. When the previously pair housed rabbits were housed in olfactory contact with their cage-mate they were less motivated to gain visual and minimal tactile contact than they were when they were housed out of olfactory contact. This suggests that olfactory contact with a familiar rabbit is important, and it may at least partly compensate for the lack of visual and minimal tactile contact.
During the long-term motivational test, food and social contact were found to be of equal importance. Rabbits were highly motivated to gain limited social contact with a conspecific. The rabbits made frequent short visits to the food and social contact cages and continued to make frequent visits as the weight on the doors increased. The rabbits did not alter their visits to make fewer longer visits to the social contact cage as the work effort to gain contact increased indicating that it was the frequency of visits to the social contact cage that was important, rather than the total duration of time spent in close proximity to another rabbit.
In the social contact cage, rabbits spent over a third of their time out of direct visual contact with the other rabbit suggesting that whilst rabbits are motivated to be in close proximity to another rabbit, they do not necessarily want to be in direct visual contact all the time. In the platform cage the rabbits spent almost all of their time in front of the platform, rather than on or under it, showing that rabbits were motivated to be in close proximity to the platform.
Shirley concluded that laboratory rabbits are motivated to gain visual and minimal tactile contact with conspecifics and that social contact was found to be of equal importance to food. Therefore, where it is not possible to house female rabbits in groups, providing limited social contact is likely to be beneficial. Rabbits were also found to be motivated for a cage platform, although it appeared to be proximity to a bolt-hole that was most important.
Dr Seaman gained her PhD in 2002.
Published papers arising from Shirley’s work
Seaman SC, Waran NK, Mason G & D’Eath RB. 2008. Animal economics: assessing the motivation of female laboratory rabbits to reach a platform, social contact and food. Animal Behaviour. 75(1): 31-42. http://dx.doi.org/10.1016/j.anbehav.2006.09.031.