Veterinary · Clinical · also Laboratory Animal Medicine
Biology, housing and health monitoring of laboratory animals, their common diseases, the 3Rs, ethics committees and experimental techniques.
In the exams: EAEVE Day One competence on the 3Rs; ACLAM and ECLAM specialist colleges
Laboratory animal science is the discipline that makes research on animals humane, lawful and reproducible. It joins the biology of the species used in research, the design of the facilities that house them, the control of infections that would ruin both welfare and data, the ethics and law that decide whether a project may go ahead, and the techniques by which animals are handled, dosed, sampled, anaesthetised and humanely killed. Laboratory animal medicine is its clinical arm, where the patient is often a room or a colony rather than one animal.
Almost every research establishment in the world must have a veterinarian with authority over animal care: the designated veterinarian under EU Directive 2010/63/EU, the attending veterinarian under US law, the Named Veterinary Surgeon in the United Kingdom, and the veterinarian on every Institutional Animal Ethics Committee in India. That veterinarian judges pain, humane endpoints, colony health status and whether a protocol has been refined as far as it can be.
This manual assumes the basics taught in livestock production management (species data, breeding systems, pheromone effects, an outline of the 3Rs). It goes deeper, to postgraduate and specialist level: species peculiarities that change experiments, genetic nomenclature, health-monitoring statistics, the three named diseases (Tyzzer's disease, murine respiratory mycoplasmosis and pasteurellosis), experimental design for reduction, the main legal frameworks compared, and the techniques themselves. Drug doses are deliberately not given.
An experiment measures the response of the animal's physiology, so every peculiarity of the species becomes part of the result. The house mouse (Mus musculus) and the Norway rat (Rattus norvegicus) make up most animals used in research, so their biology is examined in the most detail.
Both are small, nocturnal myomorph rodents with a very high metabolic rate per unit of mass and a large surface area relative to volume. They lose heat quickly, which is why they become hypothermic so fast under anaesthesia and why room and cage temperature alter food intake, metabolism and drug responses. Mice have no functional sweat glands; they regulate heat through blood flow to the tail and ears, behaviour and nest building. At standard laboratory temperatures most mice live below their thermoneutral zone, now recognised as a confounder in metabolic, immune and cancer research.
Neither species can vomit, so pre-anaesthetic fasting is not needed and emesis studies use the ferret or dog instead. Their incisors are open-rooted and grow throughout life while the molars do not, so malocclusion shows as overgrown incisors. Both are obligate nasal breathers, so upper respiratory disease soon becomes respiratory distress.
The adult mouse weighs roughly 20 to 40 g depending on strain and sex. A gestation of about 19 to 21 days, puberty at about six weeks and an easily manipulated genome explain its dominance. Females have five pairs of mammary glands, three thoracic and two inguinal.
Normal mouse urine contains much protein, mainly the major urinary proteins, lipocalins that carry pheromones. Proteinuria on a dipstick is therefore normal and is not by itself evidence of kidney disease; the same proteins are the main cause of laboratory animal allergy in staff. The mouse leukogram is lymphocyte-predominant, and young mice have a transient adrenal X-zone that regresses at puberty in males and in the first pregnancy in females.
Behaviour affects both welfare and data. Males of many strains fight when group-housed, especially when regrouped as adults, and wounds on the rump, tail and genitals are a common reason for removal from a study. Barbering, in which a dominant mouse plucks the whiskers and facial hair of cage-mates, is an abnormal behaviour; the mouse with intact whiskers is usually the barber. Mice hear well into the ultrasound range, so equipment silent to people can still disturb them.
The rat is larger (adults commonly 250 to 500 g or more), calmer when handled well and easier to instrument, so it is favoured for physiology, pharmacokinetics, toxicology, behaviour and surgical models. Females have six pairs of mammary glands.
The rat has no gall bladder; bile flows continuously into the duodenum through a bile duct that passes through the pancreas, which matters in bile-duct cannulation and biliary excretion studies. The Harderian gland behind the eye secretes a porphyrin-rich red pigment. A stressed or ill rat shows red staining around the eyes and nose (chromodacryorrhoea), which fluoresces under ultraviolet light and is easily mistaken for blood. It is a non-specific sign of illness, often respiratory disease or pain.
Two kidney conditions confound toxicology. Alpha-2u-globulin nephropathy is a male-rat-specific accumulation of protein in proximal tubules after exposure to certain chemicals; regulators treat it as not relevant to human risk. Chronic progressive nephropathy is a common age-related disease, more severe in males, and diet affects its rate.
Ringtail is a husbandry disease of pre-weaned rats: annular constrictions of the tail that can lead to necrosis, associated with low relative humidity. It is prevented by keeping humidity in range.
The Syrian or golden hamster (Mesocricetus auratus) has the shortest gestation of the common laboratory rodents, about 16 days. Its large cheek pouches are relatively immunologically privileged, which once made them a site for tumour transplantation and microcirculation studies. Males have prominent flank glands used for scent marking. Adults are solitary and females are often aggressive to males outside oestrus. A startled hamster bites, so it is woken before handling. Proliferative ileitis ('wet tail'), caused by Lawsonia intracellularis, is the classic disease of weanlings.
The Mongolian gerbil (Meriones unguiculatus) is a desert species that drinks little and produces very concentrated urine. It has a ventral midline scent gland, larger in males. Many gerbils have an incomplete circle of Willis, so occluding one common carotid artery causes unilateral cerebral ischaemia; this made the gerbil a stroke model. Some lines have spontaneous epileptiform seizures triggered by handling or a new environment. Gerbils are highly susceptible to Tyzzer's disease. The tail skin slips off (degloving) if a gerbil is held by the tip, so it is lifted at the base of the tail or cupped.
The guinea pig (Cavia porcellus) is a hystricomorph rodent. It cannot synthesise vitamin C because it lacks L-gulonolactone oxidase, so its diet must supply it; deficiency causes scurvy with swollen joints, gum haemorrhage, poor wound healing and reluctance to move. Gestation is long (roughly 59 to 72 days) and the young are precocial. If a sow is not bred before about six to seven months, the pubic symphysis fuses and dystocia is likely. Its blood contains Kurloff cells, mononuclear cells with a large cytoplasmic inclusion that increase under oestrogen.
Its sensitivity to anaphylaxis and strong delayed hypersensitivity made it the classic animal for skin sensitisation and tuberculin work, and its accessible cochlea suits hearing research. Like the hamster and rabbit, it depends on a Gram-positive-dominated hindgut flora. Oral antibiotics with a mainly Gram-positive spectrum, such as penicillins, ampicillin, lincosamides and macrolides, can allow clostridial overgrowth and fatal enterotoxaemia. Guinea pigs also suffer Bordetella bronchiseptica pneumonia, which rabbits often carry without signs, so the two species are not housed together.
The rabbit (Oryctolagus cuniculus) is a lagomorph, not a rodent; it has a small second pair of upper incisors (peg teeth). It is an induced ovulator with a gestation of about 31 days and a hindgut fermenter that re-ingests caecotrophs, so a high-fibre diet is essential for gut motility and tooth wear. Rabbits absorb dietary calcium in proportion to intake and excrete the excess in urine, so normal urine is often cloudy and excess calcium predisposes to urolithiasis. Some rabbits have serum atropinesterase, so atropine may be ineffective.
The skeleton is light relative to the powerful hind-limb muscles. A rabbit that kicks with its hindquarters unsupported can fracture or luxate the lumbar spine, causing hind-limb paralysis. Handling therefore always supports the hindquarters, and rabbits are never lifted by the ears. The marginal ear vein and central ear artery are used for injection and sampling. Rabbits breath-hold when exposed to irritant inhalant agents, which makes mask induction risky.
Research uses include polyclonal antibody production, atherosclerosis, ophthalmology and developmental toxicology, where the rabbit is a standard non-rodent species. The rabbit pyrogen test and the Draize eye and skin tests are being replaced by in vitro methods.
The zebrafish (Danio rerio) is now among the most used vertebrates: many transparent embryos develop outside the mother and suit genetic screens and imaging. Under EU law fish are protected once the larva is independently feeding. Xenopus laevis supplies oocytes for molecular and electrophysiological work. The dog (mainly the beagle) and non-human primates (mainly macaques and marmosets) are used chiefly in regulatory toxicology and are the most tightly regulated species. Macaques can carry Macacine herpesvirus 1 (B virus), which can cause fatal encephalitis in people after bites or scratches, so first-aid and medical protocols are compulsory. Pigs and minipigs serve skin, cardiovascular and xenotransplantation research; ferrets serve influenza research because they are susceptible to human influenza viruses. Sheep serve fetal physiology and orthopaedics, and chick embryos the chorioallantoic membrane assay. Invertebrates such as Drosophila and Caenorhabditis elegans are unprotected by most laws (cephalopods are the EU exception) and replace vertebrates for many genetic questions.
An inbred strain is produced by at least 20 consecutive generations of brother-sister (or parent-offspring) mating, with every animal traceable to a single pair at generation 20 or later. After 20 generations of sib mating the coefficient of inbreeding is about 0.986, so roughly 98.6% of loci are expected to be homozygous. Inbred animals are nearly identical, which reduces between-animal variation and so the numbers needed, and allows tissue transplantation without rejection.
Outbred stocks, such as Swiss-derived mice and Wistar, Sprague-Dawley and Long-Evans rats, are bred to avoid inbreeding, usually by rotational mating in large colonies. Two colonies with the same stock name kept apart for years may differ markedly. The belief that they are always more 'representative' than inbred strains is contested.
An F1 hybrid is the first-generation cross of two inbred strains: heterozygous wherever the parents differ, but all F1 animals are genetically identical and show hybrid vigour. Recombinant inbred strains, made by inbreeding the offspring of two inbred strains for 20 or more generations, form panels such as the BXD lines used to map complex traits.
Names follow international rules kept by the committees on mouse and rat genetic nomenclature. Inbred strains are written in capitals, with long-established exceptions such as BALB/c. A substrain arises when branches are bred apart long enough to diverge and is shown by a slash and a laboratory code: C57BL/6J and C57BL/6N are substrains kept at different institutions. The difference matters: C57BL/6J carries a deletion in the nicotinamide nucleotide transhydrogenase gene (Nnt) that affects glucose handling, and many C57BL/6N colonies carry the rd8 retinal degeneration allele of Crb1. Reporting only 'C57BL/6' is inadequate.
Outbred stocks are named laboratory code, colon, stock name. F1 hybrids are named female parent first: B6D2F1 is a C57BL/6 dam crossed with a DBA/2 sire. Congenic strains, made by backcrossing a donor gene into an inbred background for at least 10 generations (N10), are named background.donor-gene; speed congenics use marker-assisted selection to get there in fewer generations.
Gene symbols are italicised; mouse genes have an initial capital (Foxn1), human genes are all capitals (FOXN1), and proteins are not italicised. Alleles are superscripts (Foxn1 superscript nu is nude). Targeted mutations made in embryonic stem cells carry the code tm, alleles made with CRISPR or other nucleases carry em, and random transgene insertions are written Tg.
Genetically altered animals include transgenics, knockouts, knock-ins and conditional alleles, in which a gene is deleted only in certain cells or at certain times, typically by crossing a line with loxP-flanked ('floxed') alleles to a line expressing Cre recombinase from a tissue-specific promoter. Under EU law, creating and maintaining a line with a harmful phenotype is itself a regulated procedure.
Immunodeficient mice are central to cancer and infection research. The nude mouse (Foxn1 nu/nu) lacks a thymus and mature T cells. The SCID mouse (Prkdc scid) lacks functional B and T cells but some become 'leaky' with age. Rag1 or Rag2 knockouts lack mature B and T cells without leakiness. Strains that also lack natural killer cell function, such as NOD scid mice with a null interleukin-2 receptor common gamma chain, accept human cells best. All need barrier housing.
Classic spontaneous models include the ob/ob mouse (leptin deficient), the db/db mouse (leptin receptor defect), the NOD mouse (type 1 diabetes), the mdx mouse (Duchenne muscular dystrophy), the spontaneously hypertensive rat, the Zucker fatty rat, the Gunn rat (unconjugated hyperbilirubinaemia) and the Brattleboro rat (central diabetes insipidus).
Genetic quality is protected by defined foundation colonies, monitoring with marker panels such as single-nucleotide polymorphisms, careful records and cryopreservation of embryos or sperm, which limits drift and avoids keeping live colonies of lines not in use.
A facility protects the animals from infection and stress, the people from allergens and zoonoses, and the data from uncontrolled variables. It is zoned by microbiological risk. Barrier units hold specified-pathogen-free or immunodeficient animals; everything entering is sterilised or disinfected, and staff enter through changing rooms or air showers. Quarantine rooms hold incoming animals of unknown status until their health is confirmed. Containment suites hold animals infected with hazardous agents, graded by animal biosafety level in many countries.
People, animals, cages and waste flow from clean to dirty, never the reverse, either through separate clean and dirty corridors or a single corridor with strict procedures. Air pressure reinforces this: rooms holding high-status animals are positive to the corridor, and quarantine and infected rooms negative. Cage washing is split into a dirty side and a clean side, often linked by a pass-through tunnel washer and a double-door autoclave validated with biological indicators containing heat-resistant spores.
Surfaces should be smooth, impervious and easy to disinfect. Rooms have no uncontrolled daylight, and air-handling plant is placed so that its noise and vibration do not reach animal rooms. Procedure rooms close to housing rooms avoid carrying animals between buildings, a welfare and a biosecurity gain.
The macroenvironment is the room; the microenvironment is the cage, which is what the animal experiences. In a static filter-top cage, temperature, humidity, carbon dioxide and ammonia can all be higher than in the room. Ammonia forms when urease-producing faecal bacteria break down urinary urea; it rises with humidity, temperature, stocking density and the interval between cage changes, irritates the respiratory epithelium and worsens infections such as murine respiratory mycoplasmosis.
Individually ventilated cages (IVCs) supply filtered air to each cage and filter the exhaust. They lower cage ammonia and humidity, allow longer intervals between changes, and isolate each cage microbiologically when cages are opened only in a cabinet. A rack run at positive pressure protects the animals; one run at negative pressure protects staff and the room, as in allergen control or containment. Isolators give complete separation for germ-free animals.
The US Guide for the Care and Use of Laboratory Animals (eighth edition, 2011) recommends about 20 to 26 degrees C for common rodents and about 16 to 22 degrees C for rabbits, relative humidity of about 30 to 70%, and as a general guide 10 to 15 air changes per hour for rooms with static cages, as performance targets rather than fixed laws. Annex III of Directive 2010/63/EU gives the European values. Light runs on a fixed cycle with no interruption of the dark phase, because light at night disrupts circadian rhythms; albino rodents are prone to light-induced retinal degeneration, so cage-level light is kept low.
Cage space is set by law in the EU (Annex III) and by recommendation in the US Guide, which gives floor space by weight class; for example at least about 15 square inches (about 97 square centimetres) per mouse over 25 g. These are minima that professional judgement adjusts.
Solid-bottomed cages with bedding are the default for rodents because they allow nesting and burrowing; grid floors need justification. Untreated softwood shavings such as cedar and pine release aromatic hydrocarbons that induce hepatic microsomal enzymes and can alter drug metabolism, so hardwood, corncob or paper bedding is preferred. Food, bedding and water for barriers are autoclaved or irradiated; autoclavable diets carry extra heat-labile vitamins to allow for losses.
Natural-ingredient diets vary slightly between batches and contain phytoestrogens; purified diets are used when intake must be controlled exactly. Drinking water may be acidified or chlorinated to limit bacteria such as Pseudomonas aeruginosa, an opportunist in immunodeficient mice; the treatment must be recorded because it can affect the gut microbiota.
Enrichment lets animals perform strongly motivated behaviour. For mice, nesting material is the most valuable single item; rats benefit from shelters and taller cages that allow rearing. Social species are housed in compatible groups unless there is a justified reason not to, and enrichment is kept consistent and reported.
Infections matter even without disease, because subclinical infection changes immune responses, tumour growth, behaviour and physiology. Mouse hepatitis virus alters macrophage and T-cell function, and mouse parvoviruses can suppress immune responses and contaminate cell lines. Monitoring documents which agents are present or absent so that results can be interpreted and colonies protected.
Germ-free (axenic) animals have no detectable micro-organisms and are derived by caesarean section or embryo transfer into isolators. Gnotobiotic animals have a fully known flora. Specified-pathogen-free (SPF) animals are free of a defined list of agents, so 'SPF' means nothing unless the list and the testing are stated. Conventional animals have an undefined flora.
Agents commonly listed for mice include mouse hepatitis virus, mouse norovirus, minute virus of mice and mouse parvovirus, Theiler's murine encephalomyelitis virus, mouse rotavirus, Sendai virus, pneumonia virus of mice, ectromelia virus and lymphocytic choriomeningitis virus; bacteria such as Mycoplasma pulmonis, Clostridium piliforme, Helicobacter species, Citrobacter rodentium, Salmonella, Streptobacillus moniliformis, Rodentibacter species and Corynebacterium bovis; and pinworms (Syphacia, Aspiculuris) and fur mites (Myobia, Myocoptes, Radfordia). Rat lists add rat coronaviruses including sialodacryoadenitis virus, rat parvoviruses, Seoul hantavirus and the cilia-associated respiratory bacillus, now Filobacterium rodentium. Mouse norovirus and Helicobacter are among the agents found most often today.
Most programmes follow the FELASA recommendations, which specify agents, methods and frequency, typically with key agents screened at least every three months. Results are summarised in a health report that travels with shipped animals.
In IVC systems most programmes use soiled-bedding sentinels: immunocompetent animals whose cage regularly receives used bedding from colony cages. Agents that spread poorly through bedding, such as Mycoplasma pulmonis and fur mites, are often missed. Exhaust air dust testing, in which material from the rack's exhaust plenum is tested by PCR, detects several of these better and can reduce sentinel numbers. Immunodeficient animals make poor serology sentinels because they may not produce antibody.
Serology (ELISA or immunofluorescence) detects past exposure in immunocompetent animals; PCR detects agents in faeces, swabs, tissues and environmental samples and suits immunodeficient colonies; culture serves many bacteria. The perianal tape test finds Syphacia, whose females lay eggs around the anus; faecal flotation finds Aspiculuris; pelage examination finds mites. Cell lines, tumours and sera are screened by PCR, which replaced the mouse and rat antibody production tests.
The classic sample-size formula assumes a large population, random sampling and a perfect test: n = log(1 - C) / log(1 - p), where C is the confidence and p the prevalence. For 95% confidence and 30% prevalence, n = log(0.05)/log(0.7), about 8.4, so 9 animals; for 10% prevalence, about 28.4, so 29. Lower prevalence or poorer test sensitivity raises the number.
Infections enter mainly with animals from other sources, biological materials, people and supplies. Animals from approved sources with acceptable health reports may enter directly; others are quarantined and tested, or rederived. Embryo transfer, in which washed embryos with an intact zona pellucida go into clean recipient females, is generally preferred to caesarean rederivation with foster mothers. Neither is absolute: agents that cross the placenta or infect the embryo, such as lymphocytic choriomeningitis virus, require testing of offspring and recipients.
Inside the facility, risk falls with sterilised supplies, opening IVC cages only in a cabinet, dedicated clothing, working from higher-status to lower-status rooms within a day, and control of wild rodents and insects.
Laboratory animal allergy, caused mainly by rodent and rabbit urinary and salivary proteins, is the commonest occupational disease in animal facilities and can progress to asthma; IVCs, ventilated changing stations, respiratory protection and health surveillance control it. Zoonoses include lymphocytic choriomeningitis (from mice, hamsters and contaminated tumours), rat-bite fever (Streptobacillus moniliformis), Seoul hantavirus from rats, dermatophytosis, Salmonella, and B virus and tuberculosis where macaques are kept.
Tyzzer's disease is caused by Clostridium piliforme, formerly Bacillus piliformis, first described by Ernest Tyzzer in 1917 in Japanese waltzing mice. It is a long, slender, spore-forming bacterium that stains Gram-negative although it is classed with the clostridia. It is obligately intracellular and cannot be grown on cell-free media; it is propagated in embryonated eggs, cell cultures or animals.
The host range is wide. Gerbils are highly susceptible, and disease is important in rabbits, rats, mice, hamsters and guinea pigs. Outside the laboratory it causes acute disease in foals in their first weeks of life, and sporadic cases in cats, dogs, calves and wildlife.
Infection is by ingesting spores shed in faeces. Spores survive long periods in bedding and the environment and resist alcohol and many common disinfectants, so a contaminated colony can stay infected. Many infections are subclinical. Disease follows when the host is young (often around weaning), stressed, immunosuppressed or kept in poor hygiene; overcrowding, transport, experimental immunosuppression and dietary change are triggers. Corticosteroid treatment was historically used as a provocation test to reveal latent infection.
Disease is often peracute: animals are found dead or die after a short illness with lethargy, ruffled coat, hunched posture, dehydration and watery diarrhoea staining the perineum. Weanling rabbits develop profuse diarrhoea with high mortality.
The pathology is a classic triad of intestine, liver and heart. The organism invades the epithelium and smooth muscle of the distal ileum, caecum and colon, causing necrotising enterocolitis with oedema and haemorrhage. In rats the terminal ileum can become grossly dilated, flaccid and hyperaemic, a lesion called megaloileitis. Via the portal circulation the organism reaches the liver, causing multiple pale foci of coagulative necrosis; multifocal hepatic necrosis is the most consistent gross lesion. Pale streaks of myocardial necrosis occur in some animals, especially gerbils and rabbits.
Histologically, bundles of long, slender bacilli lie inside hepatocytes at the margins of necrotic foci, and inside enterocytes and intestinal smooth muscle, arranged in criss-cross or parallel bundles often likened to a pile of sticks. They are hard to see with haematoxylin and eosin and are shown with silver stains (Warthin-Starry), Giemsa or periodic acid-Schiff.
A presumptive diagnosis rests on enterocolitis with multifocal hepatic necrosis in a susceptible species, confirmed by demonstrating the intracellular bacilli with special stains or by PCR on liver or intestine. At colony level, serology detects exposure in immunocompetent animals and PCR on faeces or environmental samples is increasingly used. Failure to grow bacteria from lesions does not rule it out, because the agent does not grow on ordinary media.
Differentials in mice include mouse hepatitis virus (also hepatic necrosis), salmonellosis, Helicobacter hepatitis and Citrobacter rodentium colonic hyperplasia; in rabbits, coccidiosis, enteropathogenic Escherichia coli and clostridial enterotoxaemia; in hamsters, proliferative ileitis.
Treatment of individuals is generally unrewarding. Antibiotics may reduce deaths in an outbreak but do not eliminate spores. Control is by removing affected and in-contact animals, rederiving valuable lines, decontaminating rooms and equipment with sporicidal agents such as chlorine-based disinfectants or peracetic acid, or with heat, and sourcing animals from colonies shown free by monitoring. Good sanitation, avoiding overcrowding and reducing stress lower the risk of outbreaks where the agent is present. Infection can confound hepatic, intestinal and immune studies even when deaths are few.
Murine respiratory mycoplasmosis (MRM) is a chronic respiratory and genital disease of rats and mice caused by Mycoplasma pulmonis. Mycoplasmas lack a cell wall, so beta-lactam antibiotics have no effect. M. pulmonis attaches to the ciliated epithelium of the nasal passages, middle ear, trachea and bronchi, impairs mucociliary clearance and provokes a persistent inflammatory and lymphoid response. It also colonises the genital tract.
Transmission is mainly by aerosol and direct contact, and from dam to young; venereal and in utero transmission also occur, which matters for rederivation. Infection is lifelong, and the rat is generally more severely affected than the mouse.
MRM is the classic multifactorial colony disease. Severity rises with intracage ammonia, which damages the respiratory epithelium, and with concurrent infection: Sendai virus, rat coronaviruses including sialodacryoadenitis virus, and the cilia-associated respiratory bacillus (Filobacterium rodentium), which colonises the same ciliated surfaces. Host genetics matter: Lewis rats are more susceptible than F344 rats, and C3H mice develop more severe disease than C57BL mice.
Many infected animals show no signs for a long time. Early signs in rats are snuffling and audible respiratory noise ('chattering'), sneezing, nasal discharge and porphyrin staining around the eyes and nose. Later come weight loss, rough coat, hunched posture, laboured breathing and head tilt from otitis media. Reduced fertility, smaller litters and neonatal losses reflect genital infection.
Gross lesions include mucopurulent exudate in the nasal passages and middle ears and cranioventral pulmonary consolidation. In chronic cases the bronchi dilate and fill with exudate (bronchiectasis), forming raised, pale, abscess-like nodules that give the lung a cobblestone appearance. Females may have oophoritis, salpingitis, endometritis and pyometra.
Histologically there is suppurative rhinitis and otitis media, chronic bronchitis and bronchiolitis with neutrophils in airway lumens, marked peribronchial and perivascular lymphoid hyperplasia with plasma cells, squamous metaplasia of airway epithelium and bronchiectasis. Syncytia, which point to a viral cause such as Sendai virus, are not features of uncomplicated MRM. Because these lesions overlap with those of Sendai virus and the cilia-associated respiratory bacillus, laboratory confirmation is needed.
Colony diagnosis uses serology (ELISA) on immunocompetent animals, PCR on nasopharyngeal, tracheal or genital swabs or lavage, and culture on special mycoplasma media. Seroconversion can be slow, one reason PCR on respiratory samples and exhaust air dust is widely used. Soiled-bedding sentinels often fail to detect M. pulmonis because it transmits poorly through bedding.
MRM matters to research because it causes chronic illness and death, especially in ageing studies; its inflammatory lesions interfere with inhalation, respiratory, immunological and carcinogenicity studies; and it reduces breeding performance.
Tetracyclines, fluoroquinolones or macrolides can reduce clinical signs in valuable animals but do not eliminate infection, and treatment changes the animal as a research subject. Control is therefore by exclusion: stock from colonies shown free of M. pulmonis, quarantine and testing of incoming animals, barrier housing with IVCs, and rederivation by embryo transfer with testing of offspring. In an infected colony, low ammonia (good ventilation, frequent cage changes) and exclusion of viral co-infections markedly reduce severity.
Pasteurellosis, caused by the Gram-negative coccobacillus Pasteurella multocida, is the most important bacterial disease of rabbits. Many rabbits carry it in the nasal cavity and sinuses without signs; stress, poor ventilation, high ammonia, transport, parturition and concurrent disease turn carriage into disease. Transmission is by contact, aerosol and fomites, and from does to young.
The form depends on where the organism spreads. Rhinitis (snuffles) is commonest: serous then mucopurulent nasal discharge, sneezing and matted fur on the inner forelegs. Spread along the nasolacrimal duct gives conjunctivitis and dacryocystitis; up the Eustachian tube, otitis media that may extend to otitis interna with head tilt (torticollis, 'wry neck'); down the airway, bronchopneumonia and pleuritis. Haematogenous spread causes subcutaneous and internal abscesses of thick, creamy pus, pyometra in does and orchitis in bucks. Septicaemia can kill rapidly with few signs.
In research, infection causes deaths and chronic disease, interferes with respiratory, ocular and immunological studies, and means rabbits bought for long studies may fall ill midway. Pasteurella-free colonies are the preferred source.
Individual diagnosis is by culture of a deep nasal swab (a superficial swab can miss the organism), culture of pus or tissue, or PCR. Serology (ELISA) screens colonies and selects seronegative stock, but a positive result shows exposure rather than current disease. Necropsy lesions include rhinitis, pus in the tympanic bullae, cranioventral bronchopneumonia, fibrinous pleuritis and pericarditis, and abscesses.
The key differential for head tilt is Encephalitozoon cuniculi, a microsporidian common in rabbits that causes granulomatous encephalitis and nephritis. Pasteurella otitis is shown by exudate in the bulla on imaging or necropsy, while E. cuniculi produces granulomatous brain and kidney lesions and is supported by serology. Other respiratory differentials include Bordetella bronchiseptica (often harmless in rabbits but a pathogen in guinea pigs), Staphylococcus aureus and, where it occurs, myxomatosis.
Antibiotics chosen by culture and sensitivity can control signs, but deep infection in sinuses and middle ear is seldom eliminated, and choice is constrained by the rabbit's susceptibility to antibiotic-associated enterotoxaemia; oral penicillins, ampicillin and lincosamides are avoided. Any dose must come from the product label and an approved formulary.
In research the aim is to keep Pasteurella out. Rabbits come from suppliers whose colonies are monitored and free, are quarantined and tested on arrival, and are housed with good ventilation, low ammonia, low stocking density and minimal stress. Infected breeding colonies can be cleaned by test-and-cull programmes using serology and nasal culture, or by caesarean rederivation with hand-rearing or fostering.
Rodents have related organisms. The bacterium long called Pasteurella pneumotropica was reclassified in 2017 into the genus Rodentibacter, as Rodentibacter pneumotropicus and Rodentibacter heylii. These are common opportunists of the upper respiratory tract, conjunctiva and genital tract of mice and rats. In immunocompetent animals they rarely cause disease alone, but in immunodeficient mice they cause conjunctivitis, periorbital and preputial abscesses, mastitis, metritis and pneumonia. Many barrier facilities exclude them, especially where immunodeficient lines are kept. Diagnosis is by culture or PCR of oral, pharyngeal or genital swabs; serology is less reliable.
The 3Rs were set out by William Russell and Rex Burch in The Principles of Humane Experimental Technique (1959). Replacement means methods that avoid or replace the use of animals; reduction means obtaining comparable information from fewer animals, or more information from the same number; refinement means minimising pain, suffering, distress or lasting harm and improving welfare throughout an animal's life in research. Russell and Burch distinguished absolute replacement, which uses no animal material (computer models, human cells and tissues), from relative replacement, which still uses animals or their tissues but avoids procedures on living sentient animals, for example cells from humanely killed animals.
Directive 2010/63/EU makes applying the 3Rs a legal duty (Article 4) and states the final goal of full replacement once scientifically possible. National and international bodies promote them: the NC3Rs in the United Kingdom, the EU Reference Laboratory for alternatives (EURL ECVAM), ICCVAM in the United States, and the OECD, whose validated test guidelines regulators accept. The 3Rs operate with a harm-benefit analysis, and within a culture of care in which staff are trained, supported and encouraged to raise welfare concerns.
Replacement has had its largest successes in regulatory testing. The classical LD50 test was deleted from OECD guidelines in 2002 and replaced by the fixed dose procedure, the acute toxic class method and the up-and-down procedure, which use fewer animals and less severe endpoints. Reconstructed human epidermis models replace rabbit skin corrosion and irritation tests. A tiered strategy of in vitro and ex vivo methods has largely replaced the Draize rabbit eye test. For skin sensitisation, the murine local lymph node assay first replaced guinea pig tests, and defined approaches combining in chemico and in vitro assays now replace animals for many substances.
In biologicals, the rabbit pyrogen test is being replaced by the bacterial endotoxin test and the monocyte activation test, which also detects non-endotoxin pyrogens; the European Pharmacopoeia has moved to phase out the rabbit test. Monoclonal antibody production by the painful mouse ascites method has largely given way to in vitro culture, and PCR replaced the mouse and rat antibody production tests.
In research, replacement includes cell and organoid culture, organ-on-chip systems, computational models, human volunteer studies and human tissues; teaching uses models, simulators and ethically sourced cadavers. The EU completed its ban on cosmetic testing on animals in 2013, and India banned it in 2014.
Reduction is mostly good design and statistics. Too few animals is as unethical as too many, because an underpowered study wastes every animal used. Key elements are a clear hypothesis and primary outcome, the correct experimental unit (the cage, not the animal, when treatment is given in a whole cage's food or water), randomisation, blinded outcome assessment and a planned analysis.
For two group means at a two-sided significance level of 0.05 and 80% power, n per group is approximately 2 (1.96 + 0.84) squared x sigma squared / delta squared, where sigma is the standard deviation and delta the difference worth detecting. When delta equals sigma, n = 2 x 7.84 = 15.7, so 16 per group. Halving the variation cuts numbers sharply, which is why defined animals and controlled environments are reduction measures.
When variation is unknown, Mead's resource equation gives a rough guide: E = N - T - B, where N is total degrees of freedom (animals minus one), T treatment degrees of freedom (groups minus one) and B block degrees of freedom. E between about 10 and 20 is adequate. Four groups of six with no blocking give N = 23, T = 3, E = 20.
Other reduction tools are factorial and randomised block designs, longitudinal imaging of the same animals, shared tissues and controls, pilot studies, systematic reviews, efficient breeding of genetically altered lines and cryopreservation. The ARRIVE guidelines (version 2.0, 2020, with an 'Essential 10') and the PREPARE planning guidelines improve the value of every animal used.
Refinement covers an animal's whole life: housing and enrichment, social housing, habituation and training, non-aversive handling, the least invasive procedure, anaesthesia and analgesia, trained staff, post-operative care, the earliest humane endpoints the science allows, and humane killing.
Some refinements are now standard. Mice picked up by the tail show more anxiety and avoidance than mice lifted in a tunnel or cupped hands. Positive reinforcement training lets primates and dogs cooperate with sampling and dosing. Microsampling reduces blood removed and can supply toxicokinetic data from main study animals. Telemetry gives continuous data from freely moving animals instead of repeated restraint. Ear punches used for identification can supply genotyping tissue, avoiding a second procedure.
Severity links refinement to regulation. Under EU law every procedure is prospectively classed as non-recovery, mild, moderate or severe, and the actual severity each animal experienced is recorded and reported. Score sheets combining body weight, appearance, behaviour, clinical signs and procedure-specific measures, often with grimace scales, decide when to add care or analgesia and when an endpoint has been reached.
Directive 2010/63/EU applied from 1 January 2013, replacing Directive 86/609/EEC. It protects live non-human vertebrates, including independently feeding larval forms and mammalian fetuses from the last third of normal development, and live cephalopods. It covers procedures that may cause pain, suffering, distress or lasting harm at or above the level of a needle insertion in accordance with good veterinary practice, and the creation and maintenance of genetically altered lines with a harmful phenotype.
Breeders, suppliers and users must be authorised and inspected. Each needs a designated veterinarian, persons responsible for welfare and for staff competence, and an animal-welfare body that advises on welfare and the 3Rs and follows project outcomes but does not authorise projects. Projects are authorised by the competent authority after evaluation of objectives, application of the 3Rs, severity classification and harm-benefit analysis. Projects using non-human primates or involving severe procedures must undergo retrospective assessment. Non-technical project summaries are published and annual statistics include actual severity.
Great apes may be used only exceptionally under a safeguard clause; non-human primates must come from captive-bred or self-sustaining colonies; listed species including mice, rats, rabbits, dogs, cats and zebrafish must be purpose-bred. Annex III sets housing minima, Annex IV killing methods and Annex VIII severity examples. Reuse requires, among other conditions, veterinary advice and that earlier procedures were no more than moderate in actual severity. The UK Animals (Scientific Procedures) Act 1986, amended in 2012, uses establishment, project and personal licences, named persons including the Named Veterinary Surgeon and the Named Animal Care and Welfare Officer, and an Animal Welfare and Ethical Review Body.
The Animal Welfare Act, enforced by USDA APHIS through the Animal Welfare Regulations, covers warm-blooded animals but excludes purpose-bred rats of the genus Rattus, mice of the genus Mus and birds bred for research, as well as farm animals in agricultural research. The Public Health Service Policy, administered by the Office of Laboratory Animal Welfare, covers all live vertebrates in PHS-funded work and requires an institutional Assurance and adherence to the Guide. Accreditation by AAALAC International is voluntary.
The Institutional Animal Care and Use Committee (IACUC) under PHS Policy has at least five members: a veterinarian with programme responsibility, a practising scientist experienced in animal research, a member whose primary concerns are non-scientific, and a member not otherwise affiliated with the institution. The AWA minimum is three, including a veterinarian and an unaffiliated member. The IACUC reviews the programme and inspects all facilities at least every six months, approves, requires modifications to or withholds approval of protocols, reviews concerns and may suspend activities. The attending veterinarian must have authority to ensure adequate veterinary care.
Protocol review covers rationale, consideration of alternatives (USDA expects a written search for alternatives to painful procedures), species and numbers, procedures, pain relief, endpoints, euthanasia and training. USDA pain categories are B (held, not used), C (no more than momentary pain, no drugs needed), D (pain or distress relieved by drugs) and E (pain or distress not relieved, requiring written scientific justification). PHS Policy requires full re-review at least every three years; the AWA requires annual review.
In India, animal experiments are regulated under Chapter IV of the Prevention of Cruelty to Animals Act, 1960, which empowers the central government to constitute a committee for the control and supervision of experiments on animals. Long known as CPCSEA (Committee for the Purpose of Control and Supervision of Experiments on Animals), it now operates as CCSEA under the Ministry of Fisheries, Animal Husbandry and Dairying, through the Breeding of and Experiments on Animals (Control and Supervision) Rules, 1998, as amended.
Every establishment that breeds animals for experiments or conducts experiments must register with the committee and constitute an Institutional Animal Ethics Committee (IAEC). Membership includes biological scientists from different disciplines, a veterinarian involved in animal care, the scientist in charge of the animal facility, a scientist from outside the institution, a socially aware non-scientist and a nominee of the central committee. The IAEC approves projects on small animals and inspects facilities; projects on large animals and certain other categories need the central committee's approval on the IAEC's recommendation. Establishments keep records of animals bred, acquired and used and report to the committee, whose guidelines also address housing, staff qualifications, the 3Rs and rehabilitation of large animals after experiments.
India bans testing cosmetics on animals and importing animal-tested cosmetics. Rules change by notification, so read the committee's current documents.
Canada relies on the Canadian Council on Animal Care, whose guidelines and certification condition public funding. Australia uses the Australian code for the care and use of animals for scientific purposes under state law; animal ethics committees need at least one member in each of four categories: a veterinarian (A), a researcher (B), a person committed to animal welfare (C) and an independent lay person (D). The WOAH Terrestrial Animal Health Code includes principles for the use of animals in research and education.
Every committee asks the same questions. Is the objective justified, and could it be reached without animals? Is the species the least sentient that will answer the question? Is the number justified statistically? Are procedures refined, with anaesthesia, analgesia and humane endpoints? Is severity correctly classed? Are staff competent? Do benefits outweigh harms? Committees work best when lay and welfare members participate fully and approved projects are followed up by post-approval monitoring.
Poor handling changes heart rate, corticosterone, glucose and behaviour, so good handling is both refinement and good science. Mice are lifted in a tunnel or cupped hands and restrained by the scruff for procedures. Rats are lifted round the thorax behind the forelegs with the hindquarters supported. Hamsters are woken first and scooped or scruffed with a generous fold of skin. Gerbils are cupped or held at the base of the tail, never the tip. Guinea pigs are lifted with one hand under the thorax and the other under the hindquarters. Rabbits always have the hindquarters supported and may be wrapped in a towel or restrainer for ear procedures; the immobility some show when laid on their backs is a fear response, not relaxation.
Identification options include ear punches or notches (which can supply genotyping tissue), ear tags, tattoos on the tail, ear or, in neonates, toes, subcutaneous radiofrequency transponders, and temporary marks. Toe clipping is permitted in some jurisdictions only in neonates when no other method is suitable.
The route suits the aim, the substance and the animal's welfare. Oral gavage uses a ball-tipped or flexible tube passed into the oesophagus; risks are perforation and dosing into the trachea, so training is essential, and voluntary intake in a palatable vehicle is a refinement where possible. Subcutaneous injection into loose skin of the scruff or flank is easy. Intraperitoneal injection in rodents is given in the lower right abdominal quadrant, avoiding the caecum, bladder and midline; misplacement into gut, bladder or fat is common, so the route is less reliable than often assumed. Intramuscular injection is limited to small volumes in the thigh muscles, avoiding the sciatic nerve. Intravenous injection uses the lateral tail veins of mice and rats, dilated by warming, and the marginal ear vein of rabbits.
Volumes follow published good-practice limits; the table of Diehl and colleagues (2001) is the standard reference, giving for mice about 10 mL/kg for oral gavage and about 5 mL/kg for an intravenous bolus. A 30 g mouse given 10 mL/kg receives 0.3 mL; to deliver a test substance at 50 mg/kg in that volume, the solution must contain 5 mg/mL. Substances should be sterile and near physiological pH, and needles are used once.
Circulating blood volume in mice and rats is roughly 55 to 80 mL/kg; about 72 mL/kg for the mouse and 64 mL/kg for the rat are commonly used figures. Good-practice guidance suggests that a single sample of about 7.5% of circulating volume needs about one week for recovery, 10% about two weeks and 15% about four weeks, with cumulative limits for repeated samples. For a 25 g mouse at 72 mL/kg, circulating volume is 0.025 x 72 = 1.8 mL, so a 10% sample is 0.18 mL.
Recovery sites in rodents include the lateral saphenous vein, lateral tail vein and submandibular (facial) vein. Retro-orbital sampling needs anaesthesia, risks eye injury and is restricted in many establishments. Cardiac puncture and collection from the vena cava or aorta are terminal procedures under deep anaesthesia. Rabbits are sampled from the marginal ear vein or central ear artery; guinea pigs, which have few accessible veins, from the saphenous vein.
Microsampling with sensitive assays reduces harm and can provide pharmacokinetic data from the main study animals rather than satellite groups.
Small laboratory species are harder to anaesthetise safely than dogs and cats: they are small, metabolically fast, lose heat quickly and are difficult to monitor and intubate. Pre-anaesthetic fasting is unnecessary in rodents and rabbits because they cannot vomit, and risks hypoglycaemia. Core principles are heat support from induction to full recovery, eye lubricant, monitoring depth by pedal withdrawal reflex and respiratory pattern, temperature monitoring and warmed fluids for long procedures.
Inhalant isoflurane or sevoflurane from a precision vaporiser, via induction chamber and nose cone, is the most controllable method for rodents and gives rapid recovery; waste gas must be scavenged. Injectable combinations such as ketamine with an alpha-2 agonist are common, and the alpha-2 component can be reversed, but responses vary by strain, sex and age and depth cannot be adjusted once given. Rabbits often breath-hold when given inhalants by mask, so many protocols sedate them first.
Analgesia is expected for any procedure likely to cause more than momentary pain unless withholding it is scientifically justified and approved. Opioids, NSAIDs and local anaesthetics are used, preferably together (multimodal) and before the painful stimulus (pre-emptive). No doses are given here: confirm against the product label and your national formulary. Pain is assessed by behaviour, weight, food intake, nest-building, burrowing and validated grimace scales for mice, rats and rabbits.
Survival surgery follows aseptic technique adapted to small animals: clean clothing, mask, sterile gloves and instruments, and a clipped, prepared site. For a series of rodent surgeries, the tips-only technique, in which only the sterile instrument tips touch the field, with re-sterilisation of tips between animals (for example in a hot-bead sterilizer), is accepted where the institution approves it. Regulators distinguish major survival surgery, which penetrates and exposes a body cavity or causes substantial impairment, from minor surgery; multiple major survival surgeries on one animal need specific justification. Non-survival surgery entirely under anaesthesia is classed in the EU as non-recovery. Post-operative care covers warmth, fluids, analgesia, wound checks, weight and behaviour monitoring against a score sheet, and timely removal of skin clips or sutures. Implants such as telemetry transmitters and pumps should be a small proportion of body weight. Surgical competence must be documented, and practice on models and cadavers before live animals is itself a refinement.
A humane endpoint is the earliest indicator of severe pain, distress, suffering or impending death that can be used to end a procedure or kill the animal, in place of death or severe suffering as the endpoint. It is defined before the study, observable, and written into the protocol with the observation frequency and the action to take. Death as a planned endpoint is generally unacceptable and, where it seems unavoidable as in some regulatory tests, needs specific justification.
Common criteria include body-weight loss beyond a set percentage of baseline or of age-matched controls (15 to 20% is often used, but the right figure depends on the model), a low body condition score (weight alone misleads when tumours or ascites add mass), hypothermia, hunched posture, piloerection, reduced mobility or inability to reach food and water, laboured breathing, neurological signs and persistent pain. Tumour studies set limits on size, ulceration and location. Pilot studies in infection and toxicology models find predictive signs such as a fall in body temperature before death.
Score sheets combine criteria into a cumulative score with defined actions: increase monitoring, give supportive care, consult the veterinarian, or kill the animal. The veterinarian has the authority and duty to end a procedure on welfare grounds.
Euthanasia is a death with minimal pain and distress: rapid loss of consciousness followed by cardiac or respiratory arrest and loss of brain function. Accepted methods are set out in the AVMA Guidelines for the Euthanasia of Animals (2020 edition) and, in the EU, Annex IV of Directive 2010/63/EU, with species, size and age limits and sedation requirements for some physical methods. The method must also suit the science, since some alter tissues or blood chemistry.
For rodents, methods include an overdose of injectable barbiturate or other anaesthetic, inhalant overdose, carbon dioxide, and physical methods such as cervical dislocation and decapitation by trained people. AVMA guidance for CO2 is a gradual fill at a displacement rate of 30% to 70% of the chamber volume per minute, not a pre-filled chamber; for a 20-litre chamber that is 6 to 14 L/min. CO2 is aversive at higher concentrations, and whether it is the most humane option is debated. Neonatal rodents resist hypoxia and need longer exposure or another method. Rabbits and larger species are usually given a barbiturate overdose, with sedation where needed.
Death is confirmed before disposal by observing cessation of circulation and breathing or by a secondary method such as exsanguination, bilateral thoracotomy or cervical dislocation.
A veterinary student needs the 3Rs, the outline of the law in their own country, and safe handling of mice, rats and rabbits, because these are Day One competences and because students increasingly do research projects. Know the species facts that are tested (gestation lengths, vitamin C in guinea pigs, rabbit spinal fracture, rat gall bladder), the definitions of germ-free, gnotobiotic and SPF, and the three named diseases with their agents and classic lesions. Learn how to recognise pain in a rodent using a grimace scale, and why the cage, not the mouse, can be the experimental unit. If you join a research project, read the approved protocol before you touch an animal and make sure your training is recorded.
A veterinarian five years into laboratory animal work is usually the clinical veterinarian for several facilities. The daily tasks are rounds with the care staff, deciding whether an animal has reached a humane endpoint, advising researchers on anaesthesia and analgesia, reviewing protocols before the ethics committee meets, and reading health-monitoring reports. When a sentinel tests positive for mouse norovirus or Helicobacter, you decide whether the agent is excluded in that unit, which cages to test further, and whether to cull, quarantine or rederive. You train staff in handling, dosing and surgery, sign off competence, and investigate welfare incidents. You must be able to explain to a scientist, without jargon, why a group size or a procedure needs to change.
A senior laboratory animal veterinarian with 25 years' experience is often the designated or attending veterinarian for an institution and sits on, or chairs, the ethics committee. The decisions are programme-level: which agents to exclude in each barrier, whether to move to exhaust air dust monitoring, how to design or refurbish a facility (pressure regimes, IVC strategy, cage-wash flow), how to handle an outbreak that threatens irreplaceable genetically altered lines, and how to prepare for inspection or accreditation. Such a person sets institutional policies on endpoints, euthanasia methods, analgesia and handling, builds a culture of care, supports staff who kill animals, and represents the institution to regulators and, increasingly, to the public through openness agreements.
A PhD scientist with 25 years in laboratory animal science works on the evidence that the rules rest on. Research questions include validating non-aversive handling and humane killing methods, developing and validating welfare indicators and grimace scales, measuring how housing temperature, enrichment or microbiota change experimental outcomes, improving health-monitoring sensitivity and sentinel-free designs, and developing and validating replacement methods for regulatory use. Such scientists teach experimental design and statistics to reduce animal numbers, run systematic reviews of animal studies to expose poor reproducibility, advise national committees on the revision of laws and guidelines, and supervise residents preparing for ACLAM or ECLAM examinations.
This manual gives no drug doses for anaesthesia, analgesia or treatment; use your establishment's approved formulary. Numeric housing and space values are quoted only from the US Guide and in general terms; the binding EU Annex III tables and other national minima are not reproduced and must be read in the original. Non-rodent species (dogs, non-human primates, pigs, fish, amphibians, cephalopods) are covered only in outline, and their specific diseases, housing and regulation are not. The detailed pathology of the many other agents on health-monitoring lists (mouse hepatitis virus, parvoviruses, norovirus, Helicobacter and others) is listed but not described. Exact membership rules, forms and procedures of the Indian committee (CCSEA, formerly CPCSEA) change by notification and are given only in general terms; confirm them from the committee's current documents. Laws in Japan, China, Brazil and many other countries are mentioned only in passing.
Original practice questions written to the syllabus and pattern of the exams named in each question's examTags. They are not past papers and do not reproduce any board's questions.
These notes are free to read here, with no account.
What the Laisha app adds:
Free for ten days, then a subscription. We would rather write that here than have you find it later.