In-Vivo Animal Studies & Study Design
A compound that has demonstrated potent, selective, and mechanistically coherent activity in vitro must next be evaluated within the physiological complexity of a living organism, where absorption, distribution, metabolism, excretion, immune interaction, and organ-level compensation can all influence — and sometimes completely reverse — the effect predicted from isolated cell or enzyme systems. Phase 4 addresses the ethical and regulatory framework governing animal research in India, the selection of appropriate animal models, dosing methodology, pharmacokinetic study design, and the principal disease models used to demonstrate in-vivo efficacy.
In-vivo studies evaluate drug effects within living animal systems, bridging the gap between reductionist in-vitro findings and human clinical trials by generating data on efficacy, pharmacokinetics, biodistribution, and preliminary safety within an intact, homeostatically regulated biological system.
In India, every animal study — irrespective of whether it is intended for regulatory submission or purely academic publication — requires prior institutional CPCSEA registration and Institutional Animal Ethics Committee (IAEC) protocol approval before any procedure is performed.
An institution wishing to conduct animal experiments must first register with the CPCSEA, a statutory body operating under India's Ministry of Environment, Forest and Climate Change; this registration is renewed every three years and is subject to periodic inspection by a CPCSEA-appointed inspector. Each registered institution must constitute an Institutional Animal Ethics Committee comprising a chairperson drawn from outside the institution, a biological scientist, a veterinarian, a scientist member from a different institution, a scientist member from the same institution, a legal expert or a nominee of the state animal welfare board, and a socially aware nominee representing community interests — a deliberately plural composition intended to prevent purely internal or purely scientific self-approval.
Every proposed study must be submitted to the IAEC as a complete protocol specifying the scientific justification for the chosen species, the number of animals to be used (justified statistically, as discussed in Phase 6), the experimental procedures, the anaesthesia and analgesia plan, and clearly defined humane endpoints. Underlying all such protocols is the internationally accepted 3Rs principle: Replacement (using non-animal alternatives such as in-vitro or in-silico methods wherever scientifically valid), Reduction (using the minimum number of animals statistically consistent with a reliable result), and Refinement (minimising pain, distress, and suffering through improved technique, analgesia, and humane endpoints). Justification against each of the 3Rs is a mandatory component of every IAEC protocol submission.
The choice of animal model is governed by the physiological similarity of the species to humans for the process under study, practical considerations of cost and handling, and precedent within the relevant regulatory guideline.
The Wistar rat (150–250 g body weight) is the most widely used general-purpose species for CNS, cardiovascular, metabolic, anti-inflammatory, and analgesic pharmacology, owing to its well-characterised physiology and manageable size. The Sprague-Dawley rat (250–350 g) is particularly favoured for toxicology, reproductive studies, and obesity models, and is the strain preferred by the US-FDA for many regulatory toxicology submissions. The Swiss albino mouse (20–30 g) is the traditional model for acute toxicity (LD50) determination as well as anticancer, infectious disease, and immunology research, while the C57BL/6 mouse (20–25 g), a genetically well-defined inbred strain, is the standard background for diet-induced obesity, neurological, and knockout (genetically modified) models.
The guinea pig (400–700 g) is used in respiratory pharmacology, anaphylaxis, and auditory research owing to its distinctive immune and airway physiology. The rabbit (1.5–3.0 kg) is preferred for cardiovascular studies, ophthalmic irritation testing (the Draize test), and pyrogen testing, given its large ear vessels and sensitivity to pyrogenic substances. Zebrafish (Danio rerio, 0.5–1.5 g) have gained prominence for high-throughput CNS, cardiovascular, toxicological, and teratogenicity screening owing to their transparent embryos and rapid external development. Drosophila (fruit fly, microgram body mass) serves as a genetically tractable model for neuropharmacology, ageing research, and exploratory screening, offering exceptionally rapid generational turnover at minimal cost.
The welfare of laboratory animals is not merely an ethical obligation but a direct determinant of data quality, since a stressed, poorly housed, or inadequately cared-for animal introduces uncontrolled physiological variability that can obscure or mimic a genuine drug effect. CPCSEA guidelines, aligned with international standards such as the Guide for the Care and Use of Laboratory Animals, specify housing conditions in detail: ambient temperature is conventionally maintained at 22 ± 2°C, relative humidity at 50–60%, and a 12-hour light/12-hour dark cycle is maintained to preserve normal circadian physiology, which itself influences numerous pharmacological endpoints including drug metabolism and behavioural test performance. Cage density limits (minimum floor area per animal, which varies by species and body weight) prevent overcrowding-related stress and fighting, while environmental enrichment — nesting material, chew items, or shelters — is increasingly required as a matter of good practice, since its absence can itself produce stereotypic behaviours that confound behavioural pharmacology endpoints.
Animal husbandry extends to feeding (ad libitum access to a nutritionally standardised pelleted diet, with fasting protocols applied only where specifically required by the experimental design, such as before oral gavage or fasting blood glucose measurement), water (typically provided ad libitum via automated or bottle systems, with water intake itself sometimes serving as a study endpoint), and routine health monitoring, including sentinel-animal screening for pathogens that could otherwise confound immunological or inflammatory endpoints. A dedicated veterinarian, as required under the IAEC structure described above, oversees animal health and is empowered to mandate early euthanasia on welfare grounds independent of the study's scientific objectives, directly operationalising the Refinement principle of the 3Rs.
The oral (per os) route, administered by gavage using a graduated feeding needle, permits a maximum volume of approximately 10 mL/kg in the rat; correct needle placement must be confirmed before dosing, and animals are conventionally fasted for 12–16 hours beforehand to standardise gastric emptying. The intraperitoneal (i.p.) route, also limited to approximately 10 mL/kg and delivered via a 27-gauge needle into the lower left abdominal quadrant to avoid the bladder and gastrointestinal tract, is the most common parenteral route in mice owing to its technical simplicity. The intravenous (i.v.) route, limited to approximately 5 mL/kg administered slowly, uses the tail vein in rats and mice or the marginal ear vein in rabbits, with gentle warming of the limb used to promote vasodilation and ease of injection. The subcutaneous (s.c.) route, also limited to approximately 5 mL/kg, is typically administered into the scruff of the neck and produces a depot effect suited to chronic dosing regimens. The intramuscular (i.m.) route is limited to approximately 0.5 mL per injection site in the thigh muscle; because it is comparatively painful, its use is generally limited to vaccine and adjuvant studies. Topical and transdermal administration requires shaving the application area, defining it precisely in square centimetres, and may be studied ex-vivo using Franz diffusion cells, with tape-stripping used to assess stratum corneum penetration. The intranasal route, delivering 5–10 microlitres per nostril to an anaesthetised animal with the head tilted back, is used for both CNS-targeted delivery and vaccine studies.
Because metabolic rate does not scale linearly with body weight, a simple per-kilogram conversion of a human dose to an animal dose is scientifically invalid. The Body Surface Area (BSA) method, recommended in the FDA's 2005 Guidance for Industry on estimating the maximum safe starting dose in initial clinical trials, instead scales dose according to the ratio of species-specific Km factors: Animal dose (mg/kg) = Human dose (mg/kg) × (Human Km ÷ Animal Km). Km factors are approximately 37 for humans, 6 for rats, 3 for mice, 12 for rabbits, 20 for dogs, and 12 for monkeys. As a worked example, a human dose of 10 mg/kg converts to an approximate rat dose of 10 × (37/6) ≈ 61.7 mg/kg. An alternative approach, allometric scaling, uses body weight raised to the power 0.75 to account for the non-linear relationship between body size and metabolic rate across species.
A typical animal pharmacokinetic study proceeds through a defined sequence: dose selection, encompassing both single-dose and multiple-dose designs; selection of the administration route matching the intended clinical route; serial blood sampling, typically spanning 0.25 to 24 hours post-dose to capture the full absorption, distribution, and elimination phases; plasma separation, performed using EDTA anticoagulant and maintained at 4°C to minimise ex-vivo degradation; bioanalysis using a validated chromatographic method, most commonly HPLC or LC-MS/MS, of the kind described in ICH-compliant method validation; non-compartmental analysis (NCA), typically performed using dedicated software such as WinNonlin, to derive parameters including maximum plasma concentration (Cmax), time to maximum concentration (Tmax), area under the curve (AUC), half-life (t½), and clearance; and finally compilation of a formal pharmacokinetic report summarising these derived parameters.
A sound experimental design is what allows an in-vivo study to yield a statistically and biologically defensible conclusion from the minimum ethically justifiable number of animals. Three design principles, originally formalised by Ronald Fisher for agricultural field trials but directly applicable to laboratory pharmacology, underlie virtually every well-designed animal study. Randomisation — the random allocation of animals to treatment groups — prevents systematic bias arising from, for example, consistently assigning the first animals removed from a cage (which may be the calmest or most easily handled) to a particular group. Replication — the use of a biologically meaningful number of animals per group, informed by the power calculation introduced in Phase 6 — ensures that the observed effect reflects a genuine population-level phenomenon rather than the idiosyncrasy of one or two individual animals. Blocking (or stratification) — deliberately balancing known sources of variability, such as body weight, litter, or cage, evenly across treatment groups — reduces unexplained variance and increases the statistical power of the eventual analysis.
A further critical design consideration is blinding: wherever feasible, the individual performing behavioural scoring, tissue evaluation, or histopathological grading should remain unaware of each animal's treatment group allocation, since even well-intentioned observers show measurable, unconscious bias favouring the expected result when blinding is not employed. Modern reporting guidelines for animal research, most notably the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines, formalise these expectations and are increasingly required by journals as a condition of publication, directly reinforcing the publication and thesis checklist presented in Phase 7.
Careful biological sample collection is essential to the validity of downstream pharmacokinetic, biochemical, and histopathological analysis. Blood is typically collected by retro-orbital, tail-vein, or cardiac-puncture techniques depending on volume requirements and whether the procedure is terminal, using appropriate anticoagulants (commonly EDTA) and minimising haemolysis through gentle handling and prompt centrifugation. Organs designated for histopathology or biochemical assay are harvested immediately following euthanasia, weighed (both as absolute organ weight and relative to total body weight, since organ-to-body-weight ratios are a sensitive indicator of toxicity), and either snap-frozen for biochemical analysis or fixed in 10% neutral-buffered formalin for subsequent histopathological processing, a topic examined further in Phase 5.
Demonstrating in-vivo efficacy requires a disease model that reproduces the relevant human pathophysiology with sufficient fidelity and reproducibility to allow a genuine drug effect to be distinguished from background biological variability. Over several decades, pharmacologists have developed and validated a large catalogue of such models, spanning virtually every major organ system and disease category. This section presents approximately forty of the most widely used and pedagogically important pharmacological animal models, organised by disease category. For each model, the underlying disease background, the species and strain typically employed, the method and mechanism of induction, the experimental procedure and parameters measured, the standard reference drug, and the principal applications, advantages, limitations, and recent advances are described in theoretical form. Together, these models constitute the practical vocabulary of in-vivo pharmacology and are essential knowledge for GPAT, NIPER, and dissertation-level examination as well as for actual bench research.
Central nervous system disease models are amongst the most extensively used in pharmacological research because behavioural and electrophysiological endpoints, unlike many peripheral disease markers, can often be measured non-invasively and repeatedly in the same animal, improving statistical power while reducing overall animal usage.
Epilepsy is a chronic neurological disorder characterised by recurrent, unprovoked seizures arising from abnormal, synchronous neuronal discharge. Two complementary acute seizure models are used to screen candidate anticonvulsants and probe distinct seizure mechanisms. In the Maximal Electroshock (MES) model, typically performed in Wistar rats or Swiss albino mice, a supramaximal electrical stimulus (approximately 50 mA for 0.2 seconds, delivered via corneal or ear electrodes) induces tonic hindlimb extension analogous to human generalised tonic-clonic seizures; abolition of the tonic extensor phase indicates anticonvulsant activity, and the model is considered predictive of drugs effective against generalised seizures acting via sodium-channel blockade or glutamatergic inhibition. In the pentylenetetrazole (PTZ) model, a chemoconvulsant dose of 80–90 mg/kg administered intraperitoneally antagonises GABA-A receptor-mediated inhibition, producing myoclonic jerks and clonic seizures; protection in this model is considered predictive of drugs enhancing GABAergic transmission, such as benzodiazepines and valproate. Efficacy in both models is quantified as the percentage of animals protected from the defined seizure endpoint, and standard reference drugs include phenytoin (MES-predictive) and diazepam or sodium valproate (PTZ-predictive). Limitations include the acute, single-seizure nature of both models, which does not reproduce the epileptogenesis process underlying chronic human epilepsy; genetic and kindling models (for example, the pilocarpine or kainate status-epilepticus models) are increasingly used as complementary chronic alternatives in advanced research.
Depression is modelled behaviourally using tests of 'behavioural despair', in which rodents subjected to an inescapable stressor progressively adopt an immobile posture interpreted as analogous to the psychomotor retardation and hopelessness of human depression. In the Forced Swim Test, rats or mice are placed in a cylinder of water from which escape is impossible, and immobility time is recorded over a 5–6 minute observation period following an initial habituation swim; increased immobility indicates a depressive-like phenotype, while reduced immobility following drug treatment indicates antidepressant-like activity. The Tail Suspension Test applies an analogous logic to mice suspended by the tail for approximately 6 minutes, avoiding the confound of differential swimming ability across genotypes or treatments. Imipramine and fluoxetine are standard reference antidepressants in both assays. Both tests are valued for their speed, low cost, and sensitivity to clinically effective antidepressant drug classes, but are criticised for reflecting acute pharmacological reactivity rather than a genuine model of the chronic, multifactorial aetiology of clinical depression; chronic mild stress and chronic social defeat stress paradigms are increasingly used as more construct-valid alternatives in contemporary neuropsychopharmacology research.
Anxiety disorders are modelled by exploiting rodents' natural conflict between the drive to explore a novel environment and their innate aversion to open, elevated, and brightly lit spaces. The Elevated Plus Maze apparatus consists of two open arms and two enclosed arms arranged in a plus shape and elevated approximately 50 cm above the floor; an anxious animal spends proportionally less time in, and makes fewer entries into, the open arms. Anxiolytic drug activity is indicated by an increase in the percentage of time spent in, and entries into, the open arms, without a confounding non-specific increase in total locomotor activity (assessed by total arm entries). Diazepam is the standard positive control. The Open Field Test provides a complementary measure, in which reduced central-zone exploration and increased thigmotaxis (wall-hugging behaviour) indicate an anxious phenotype. Both tests are rapid, inexpensive, and require no chemical or surgical induction, but are sensitive to procedural variables such as handling stress, testing-room lighting, and prior test experience, all of which must be rigorously standardised across a study.
Alzheimer's disease is a progressive neurodegenerative disorder characterised by cholinergic neuronal loss, amyloid-beta plaque deposition, and progressive memory impairment. The scopolamine-induced amnesia model exploits the central role of cholinergic transmission in memory consolidation: administration of the muscarinic antagonist scopolamine (typically 1 mg/kg intraperitoneally) in Wistar rats or Swiss mice produces reversible, pharmacologically induced amnesia that mimics the cholinergic deficit of Alzheimer's disease without requiring genetically modified animals. Cognitive impairment is quantified using the Morris Water Maze (escape latency to a submerged platform, and time spent in the target quadrant during a subsequent probe trial), the passive avoidance task (latency to re-enter a chamber previously associated with a mild foot shock), and biochemical acetylcholinesterase (AChE) activity assay in brain homogenate. Piracetam and donepezil, both clinically used cognitive enhancers, serve as standard reference drugs. The principal advantage of this model is its speed and low cost relative to genetically modified amyloid- or tau-transgenic mouse lines, which better recapitulate the underlying neuropathology but require considerably longer study durations and specialised breeding colonies; contemporary Alzheimer's research increasingly combines the scopolamine screening model for rapid throughput with transgenic models for definitive mechanistic and disease-modifying studies.
Parkinson's disease results from progressive degeneration of dopaminergic neurons in the substantia nigra pars compacta, producing bradykinesia, rigidity, and resting tremor. Two chemical lesioning approaches are standard. Stereotaxic unilateral intracerebral injection of 6-hydroxydopamine (6-OHDA) into the substantia nigra or medial forebrain bundle of rats produces a selective, largely unilateral dopaminergic lesion; the resulting motor asymmetry is quantified using apomorphine- or amphetamine-induced rotational behaviour (rotations away from or towards the lesioned side, respectively), striatal dopamine content assay, and tyrosine hydroxylase (TH) immunohistochemistry as a marker of surviving dopaminergic neurons. Systemic administration of MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine), typically in mice, is metabolised to the active toxin MPP+ by glial monoamine oxidase-B and selectively taken up by dopaminergic neurons via the dopamine transporter, producing bilateral nigrostriatal degeneration that more closely parallels the bilateral presentation of human Parkinson's disease. Levodopa and dopamine agonists such as bromocriptine serve as reference standards for symptomatic reversal. Both models successfully reproduce dopaminergic degeneration but do not reproduce the Lewy body alpha-synuclein pathology characteristic of the human disease, motivating the increasing use of alpha-synuclein overexpression and pre-formed fibril seeding models in translational Parkinson's research.
Schizophrenia is modelled pharmacologically by exploiting the two principal neurochemical hypotheses of the disorder. The amphetamine-induced hyperlocomotion model, based on the dopamine hypothesis, uses systemic amphetamine administration to produce excessive dopaminergic signalling, resulting in stereotyped behaviour and hyperlocomotion in rodents that models the positive symptoms of schizophrenia; attenuation of this hyperlocomotion by a test compound is interpreted as antipsychotic-like activity, with haloperidol or clozapine as reference standards. The ketamine-induced model, based on the glutamate (NMDA receptor hypofunction) hypothesis, uses sub-anaesthetic doses of the NMDA receptor antagonist ketamine to produce both positive-symptom-like hyperlocomotion and negative/cognitive-symptom-like deficits in social interaction and working memory, offering somewhat broader construct validity than the amphetamine model alone. Prepulse inhibition of the acoustic startle reflex, a sensorimotor gating measure disrupted in both models and in human schizophrenia patients, provides a further translationally relevant endpoint. Both models are pharmacologically induced and acute, and therefore do not capture the neurodevelopmental origin of schizophrenia; neurodevelopmental models such as prenatal maternal immune activation are used in more specialised research settings.
Neuropathic pain arises from injury or dysfunction of the somatosensory nervous system itself, rather than from ongoing tissue damage, and is characterised clinically by allodynia (pain from normally non-painful stimuli) and hyperalgesia (exaggerated pain from normally painful stimuli). The Chronic Constriction Injury model, performed in rats, involves surgically placing four loose ligatures around the common sciatic nerve, producing a controlled, partial nerve injury that induces robust and reproducible mechanical and thermal hypersensitivity in the ipsilateral hind paw within one week of surgery. Mechanical allodynia is quantified using von Frey filaments (the force required to elicit paw withdrawal), and thermal hyperalgesia is quantified using the Hargreaves radiant heat test. Gabapentin, pregabalin, and amitriptyline serve as standard reference analgesics, reflecting the distinct pharmacology of neuropathic (as opposed to nociceptive or inflammatory) pain. The CCI model is valued for closely reproducing the time course and quality of human neuropathic pain following nerve injury, though the surgical skill required and the inherent variability in ligature tightness between operators necessitate careful surgeon training and, ideally, blinded outcome assessment.
Beyond its application in Alzheimer's disease research described above, the Morris Water Maze is used more broadly as a general assay of spatial learning and memory, applicable to cognitive-enhancer screening, neurotoxicology, and ageing research. The apparatus consists of a circular pool filled with opacified water, containing a submerged, invisible escape platform whose location the animal must learn using extra-maze spatial cues over repeated training trials; the primary endpoints are escape latency (time to locate the platform) during training and time spent in the target quadrant during a probe trial, reflecting the strength of the spatial memory formed. Because the task depends on hippocampal function, the Morris Water Maze is particularly sensitive to compounds or lesions affecting hippocampal long-term potentiation. Advantages include its extensive historical validation and sensitivity; limitations include the physical stress of forced swimming, which can itself confound results in animals with impaired stress-coping ability, and the requirement for intact vision and swimming ability, which restricts its use in certain genetically or surgically modified animal lines.
The Open Field Test is among the most widely used general-purpose behavioural assays in pharmacology, serving simultaneously as a locomotor-activity assay, an anxiety-related measure, and a screen for sedative or stimulant drug effects. An animal is placed in a large, open, brightly lit arena, and its behaviour is recorded (manually or via automated video-tracking software such as ANY-maze or EthoVision) over a fixed observation period, typically five to ten minutes. Key parameters include total distance travelled and rearing frequency (locomotor activity), time spent in the central versus peripheral zone (anxiety-related thigmotaxis), and grooming frequency and duration. Because many CNS-active drugs alter locomotor activity, the Open Field Test is frequently run as a companion assay alongside more specific behavioural tests (such as the elevated plus maze or forced swim test) to rule out the possibility that an apparent anxiolytic or antidepressant effect is in fact a non-specific sedative or stimulant artefact. Its principal advantages are speed, minimal equipment requirement, and applicability across essentially any rodent strain or genotype.
Cardiovascular disease models allow the efficacy of candidate antihypertensive, anti-ischaemic, and cardioprotective agents to be evaluated using readily quantifiable physiological endpoints such as blood pressure, infarct size, and cardiac function.
Hypertension, a chronic elevation of systemic arterial blood pressure, is modelled using both chemically induced and genetic approaches. In the deoxycorticosterone acetate (DOCA)-salt model, uninephrectomised rats receive DOCA (25 mg/kg subcutaneously, twice weekly) combined with 1% saline as drinking water, producing mineralocorticoid-driven sodium and water retention and consequent volume-dependent hypertension over two to four weeks. The Spontaneously Hypertensive Rat (SHR), a genetically inbred strain that develops hypertension without any experimental intervention, models the polygenic, essential (primary) hypertension that accounts for the majority of human cases. Blood pressure in both models is measured non-invasively by tail-cuff plethysmography or, for greater precision, by radiotelemetry, with captopril or losartan serving as reference antihypertensive standards. The DOCA-salt model is particularly useful for studying mineralocorticoid- and volume-dependent hypertension mechanisms, while the SHR strain is preferred for studying the broader pathophysiology and end-organ (cardiac and renal) consequences of chronic essential hypertension, including left ventricular hypertrophy.
Myocardial infarction results from prolonged coronary ischaemia causing irreversible cardiomyocyte necrosis. The isoproterenol-induced model exploits the observation that supraphysiological doses of the beta-adrenergic agonist isoproterenol (typically 85–150 mg/kg administered subcutaneously on two consecutive days in rats) produce myocardial necrosis through a combination of excessive oxygen demand, coronary vasospasm, and direct catecholamine-mediated calcium overload toxicity, without requiring surgical coronary occlusion. Cardiac injury is assessed through serum cardiac biomarkers (creatine kinase-MB, cardiac troponin, and lactate dehydrogenase), electrocardiographic changes (ST-segment elevation), and histopathological evidence of myocardial necrosis and inflammatory infiltrate. Standard cardioprotective reference agents vary by mechanism under investigation but commonly include beta-blockers or antioxidant compounds, reflecting the oxidative-stress component of isoproterenol-induced injury. The principal advantage of this model is its technical simplicity relative to surgical infarction models; its main limitation is that the mechanism of injury (catecholamine toxicity) differs mechanistically from the coronary thrombotic occlusion that underlies most human myocardial infarction.
Chronic heart failure following myocardial infarction is modelled surgically by permanent ligation of the left anterior descending coronary artery in rats or mice, producing a discrete infarct followed, over subsequent weeks, by progressive ventricular remodelling, dilation, and contractile dysfunction that closely mirrors the pathophysiological sequence of post-infarction heart failure in humans. Cardiac function is assessed longitudinally using echocardiography (ejection fraction, fractional shortening, and chamber dimensions), invasive haemodynamic measurement (via a pressure-volume catheter for load-independent contractility indices), and terminal histopathological assessment of infarct size and interstitial fibrosis. ACE inhibitors, angiotensin receptor blockers, and beta-blockers, the cornerstone pharmacological classes of human heart failure therapy, serve as standard reference comparators. This model offers excellent translational relevance to the most common clinical aetiology of heart failure but demands substantial surgical expertise, longer study durations (typically four to eight weeks to allow remodelling to develop), and specialised, often costly, imaging or haemodynamic equipment.
Metabolic disease models are central to antidiabetic, anti-obesity, and lipid-lowering drug development, and increasingly incorporate dietary as well as chemical induction methods to better reflect the multifactorial aetiology of human metabolic disease.
Diabetes mellitus is characterised by chronic hyperglycaemia resulting from insufficient insulin secretion, insulin resistance, or both. The streptozotocin-induced model exploits the selective uptake of STZ by pancreatic beta cells via the GLUT2 glucose transporter, followed by DNA alkylation and beta-cell necrosis; a single intraperitoneal dose (commonly 45–60 mg/kg, freshly prepared in citrate buffer at pH 4.5 and administered within minutes of preparation owing to the compound's instability in solution) produces a Type 1-like insulin-deficient diabetic state in rats within 48–72 hours, confirmed by fasting blood glucose exceeding 250 mg/dL. A combined high-fat-diet-plus-low-dose-STZ protocol instead produces a Type 2-like model incorporating both insulin resistance and partial beta-cell dysfunction, more closely reflecting the pathophysiology of the more common human disease form. Disease progression and drug efficacy are monitored via fasting blood glucose, the oral glucose tolerance test (OGTT), glycated haemoglobin (HbA1c), and serum insulin, with metformin, glibenclamide, or insulin serving as standard reference comparators depending on the mechanism under study. This model's principal advantage is its rapid onset and low cost; its principal limitation is that classical high-dose STZ diabetes does not reproduce the autoimmune beta-cell destruction underlying human Type 1 diabetes, for which the non-obese diabetic (NOD) mouse is a more mechanistically faithful, if considerably slower and more expensive, alternative.
Diet-induced obesity is modelled by feeding rodents, most commonly C57BL/6 mice owing to their well-documented diet-responsiveness, a palatable high-fat diet (typically providing 45–60% of calories from fat) ad libitum for eight to sixteen weeks, producing progressive weight gain, adiposity, and the metabolic sequelae of human dietary obesity, including insulin resistance, dyslipidaemia, and hepatic steatosis. Efficacy of candidate anti-obesity agents is assessed through body weight trajectory, body composition (by dual-energy X-ray absorptiometry or nuclear magnetic resonance), food intake, and the metabolic parameters described above. Orlistat and, increasingly, GLP-1 receptor agonists such as liraglutide serve as clinically relevant reference comparators. The principal strength of this model is its aetiological similarity to the dietary and lifestyle origins of most human obesity; its main limitation is the considerable inter-individual variability in weight gain typically observed even among genetically identical animals fed an identical diet, necessitating larger group sizes than many other metabolic models to achieve adequate statistical power.
Hyperlipidaemia, an elevation of circulating cholesterol and/or triglycerides, is modelled acutely using Triton WR-1339 (a non-ionic detergent administered intravenously or intraperitoneally at approximately 100–400 mg/kg), which transiently blocks lipoprotein lipase activity and produces a rapid, reproducible rise in serum triglycerides and cholesterol within 24 hours, suitable for rapid screening of hypolipidaemic candidates. Chronic hyperlipidaemia is alternatively modelled through prolonged high-fat, high-cholesterol dietary feeding, which more closely reproduces the gradual dyslipidaemia of human diet-related hyperlipidaemia and is often combined with atherosclerosis endpoint assessment (aortic lipid staining) in more advanced study designs. Serum total cholesterol, LDL-cholesterol, HDL-cholesterol, and triglycerides are the standard biochemical endpoints, with statins (for example atorvastatin) or fibrates serving as reference hypolipidaemic comparators. The Triton model offers speed and reproducibility for initial screening, while the dietary model offers superior translational relevance for candidates intended for chronic human use.
Postmenopausal osteoporosis results from the loss of the bone-protective effects of oestrogen following the menopausal decline in ovarian hormone production. The ovariectomised rat model surgically removes both ovaries, producing an oestrogen-deficient state that closely parallels human menopause and results in progressive trabecular bone loss over eight to twelve weeks, most prominently in the proximal tibia and lumbar vertebrae. Bone status is assessed using dual-energy X-ray absorptiometry (bone mineral density), micro-computed tomography (trabecular microarchitecture), biomechanical testing (femoral neck or vertebral compression strength), and biochemical markers of bone turnover (serum osteocalcin as a formation marker, and urinary deoxypyridinoline or serum C-telopeptide as resorption markers). Standard reference comparators include oestrogen replacement therapy, bisphosphonates (alendronate), and selective oestrogen receptor modulators (raloxifene). This model is considered the gold-standard preclinical model for postmenopausal osteoporosis research and is explicitly recommended by regulatory guidance for osteoporosis drug development, though the extended study duration required to observe measurable bone loss represents a practical limitation.
Inflammatory and immune-mediated disease models range from acute, hours-long assays suited to high-throughput screening of anti-inflammatory candidates to chronic, weeks-long models capturing the autoimmune and tissue-destructive character of diseases such as rheumatoid arthritis.
Rheumatoid arthritis is a chronic autoimmune disease characterised by synovial inflammation and progressive joint destruction. The adjuvant-induced arthritis model, typically performed in Wistar or Lewis rats, involves a single sub-plantar or intradermal injection of Complete Freund's Adjuvant (heat-killed Mycobacterium tuberculosis suspended in mineral oil), which triggers a delayed-type hypersensitivity response producing polyarthritis affecting both the injected and, characteristically, distal untreated paws within ten to fourteen days, reflecting a systemic autoimmune process rather than a purely local inflammatory reaction. Disease severity is quantified using paw volume (plethysmometry), an arthritic scoring system grading redness, swelling, and joint rigidity across all four limbs, radiographic assessment of bone erosion, and histopathological synovial inflammation scoring. Methotrexate, the cornerstone disease-modifying antirheumatic drug in human rheumatoid arthritis therapy, together with NSAIDs such as indomethacin, serve as standard reference comparators. This model's principal strength is its capture of the systemic, immune-mediated character of rheumatoid arthritis; collagen-induced arthritis, using type II collagen immunisation, is a mechanistically complementary alternative model increasingly favoured for its closer resemblance to the autoantibody-driven pathology of the human disease.
Acute inflammation is most commonly modelled using sub-plantar injection of 0.1 mL of 1% carrageenan (a sulphated polysaccharide extracted from red seaweed) into the hind paw of rats, which triggers a well-characterised, biphasic inflammatory response: an early phase (0–2 hours) mediated by histamine, serotonin, and bradykinin, followed by a later phase (2–6 hours) mediated predominantly by prostaglandins and neutrophil infiltration. Paw swelling is quantified by plethysmometry (water or mercury displacement) at hourly intervals up to four to six hours post-injection, and percentage inhibition of oedema relative to vehicle-treated controls is calculated for each treatment group. Indomethacin or diclofenac serve as standard NSAID reference comparators. This model's speed (a complete experiment within a single day), reproducibility, and mechanistic tractability (the biphasic time course allows differentiation between early mediator-blocking and late prostaglandin-blocking drug mechanisms) make it the most widely used first-line screen for candidate anti-inflammatory agents, although it captures only acute, and not chronic or autoimmune, inflammatory processes.
Nociceptive (pain) pharmacology is assessed using thermal reflex-withdrawal assays that measure the latency between a defined noxious thermal stimulus and a reflexive withdrawal response. In the tail-flick test, radiant heat or a focused heat source is applied to the tail, and the latency to tail withdrawal is recorded, reflecting a predominantly spinal reflex circuit. In the hot-plate test, an animal is placed on a surface maintained at a fixed noxious temperature (commonly 52–55°C), and the latency to a supraspinally integrated response (paw licking or jumping) is recorded, providing complementary information about centrally mediated analgesia. Increased latency following drug treatment indicates analgesic activity, with morphine as the standard opioid reference and a defined cut-off time enforced in both assays to prevent thermal tissue injury. Chemical nociceptive models, including the formalin test (biphasic paw-licking response to subcutaneous formalin injection, distinguishing acute nociceptive and inflammatory pain phases) and the acetic-acid writhing test (counting abdominal writhing responses following intraperitoneal acetic acid, sensitive to both opioid and NSAID analgesics), provide further mechanistically distinct assessment of analgesic efficacy.
Fever is a regulated elevation of core body temperature mediated by hypothalamic prostaglandin E2 synthesis in response to circulating pyrogenic cytokines. The yeast-induced pyrexia model produces this response experimentally by subcutaneous injection of a 15–20% aqueous suspension of brewer's yeast, which triggers an inflammatory cytokine cascade and a measurable rise in rectal temperature peaking approximately 18–19 hours post-injection. Antipyretic efficacy is assessed by comparing rectal temperature (measured with a digital or thermocouple thermometer) in treated versus vehicle-control animals at defined time points following drug administration at the temperature peak, with paracetamol or aspirin as standard reference antipyretics. The model's principal strength is its mechanistic parallel to genuine cytokine-mediated fever, as opposed to models of hyperthermia produced by direct heat exposure, which do not involve the same central prostaglandin-mediated regulatory pathway and are therefore not appropriate substitutes for antipyretic drug screening.
Osteoarthritis is a degenerative joint disease characterised by progressive articular cartilage breakdown, subchondral bone remodelling, and chronic joint pain. The monoiodoacetate model produces this pathology by a single intra-articular injection of sodium monoiodoacetate into the knee joint, which inhibits chondrocyte glycolysis (via inhibition of glyceraldehyde-3-phosphate dehydrogenase), causing chondrocyte death and cartilage degeneration that closely mimics the histopathological features of human osteoarthritis within two to three weeks. Disease severity and drug efficacy are assessed using weight-bearing asymmetry (via an incapacitance meter comparing load distribution between the injected and contralateral limb), von Frey mechanical allodynia testing, and histopathological cartilage (Mankin) scoring. NSAIDs and, increasingly, nerve growth factor-targeted biologics serve as reference comparators, reflecting both the inflammatory and neuropathic components of osteoarthritic pain. The model's speed and reproducibility make it the most widely used preclinical osteoarthritis model, although the acute chemical induction mechanism differs from the slow, mechanically driven degeneration underlying most human osteoarthritis, motivating the use of surgical (destabilisation of the medial meniscus) models in more mechanistically focused research.
Multiple sclerosis and related CNS autoimmune demyelinating diseases are modelled using Experimental Autoimmune Encephalomyelitis, induced by immunising susceptible mouse strains (commonly C57BL/6) or rats with myelin-derived peptides (such as myelin oligodendrocyte glycoprotein, MOG35-55) emulsified in Complete Freund's Adjuvant, together with pertussis toxin to enhance blood-brain-barrier permeability and immune cell infiltration into the CNS. Immunised animals develop an ascending paralysis over one to three weeks, scored using a standardised 0–5 clinical scale ranging from a limp tail to complete hindlimb and forelimb paralysis, accompanied by histopathological CNS demyelination and inflammatory infiltrate. Corticosteroids and, increasingly, disease-modifying biologics targeting specific immune cell populations (such as natalizumab, an alpha-4 integrin antibody) serve as reference comparators. EAE is the principal preclinical model underlying the development of virtually all currently approved multiple sclerosis therapeutics, though its relapsing-remitting or chronic-progressive character varies considerably between mouse strains and immunisation protocols, requiring careful strain selection matched to the specific disease phenotype under investigation.
Drug-induced immunosuppression is modelled using cyclophosphamide, an alkylating agent that preferentially depletes rapidly dividing immune cell populations, particularly in the bone marrow and lymphoid organs. Administration of cyclophosphamide (commonly 50–200 mg/kg intraperitoneally, as a single dose or short repeated regimen) produces a reproducible, dose-dependent leucopenia and immune dysfunction that is used both to study the pathophysiology of immunosuppression itself and, more commonly, as a standard experimental background against which candidate immunostimulant or immunomodulatory agents (including many phytochemical and Ayurvedic formulations of research interest) are evaluated for their ability to restore immune parameters. Endpoints include total and differential white blood cell counts, relative spleen and thymus weight, humoral immune response (antibody titre following a test antigen such as sheep red blood cells), and cell-mediated immune response (delayed-type hypersensitivity reaction). Levamisole is a commonly used reference immunostimulant. This model is valued for its simplicity and reproducibility, and for directly modelling a clinically important adverse effect of cancer chemotherapy.
Sepsis, a dysregulated systemic inflammatory response to infection carrying high clinical mortality, is modelled surgically using the caecal ligation and puncture procedure, in which the caecum is exteriorised, ligated below the ileocaecal valve, and punctured to allow controlled leakage of faecal material into the peritoneal cavity, producing a polymicrobial infection and systemic inflammatory response that closely mirrors the clinical progression of human intra-abdominal sepsis. A simpler, non-surgical alternative, the lipopolysaccharide (LPS)-induced endotoxaemia model, produces a systemic inflammatory response by intraperitoneal LPS injection alone, offering greater procedural reproducibility at the cost of omitting the genuine polymicrobial infectious component. Endpoints in both models include survival rate over 72–96 hours, circulating pro-inflammatory cytokines (TNF-alpha, IL-6), organ-function biomarkers reflecting multi-organ dysfunction, and bacterial colony counts in blood and peritoneal fluid. Broad-spectrum antibiotics combined with fluid resuscitation represent the clinical standard of care and serve as the benchmark against which novel sepsis-directed immunomodulatory therapies are compared; the CLP model is widely regarded as the preclinical gold standard for sepsis research given its close mechanistic parallel to human disease.
Cancer pharmacology models range from chemically induced carcinogenesis models, which recapitulate the multistep process of tumourigenesis over an extended time course, to transplantable tumour (xenograft) models, which provide a rapid, reproducible platform for evaluating candidate anticancer agents against an established tumour.
Chemically induced carcinogenesis models use a defined carcinogenic initiator, most commonly 7,12-dimethylbenz[a]anthracene (DMBA), to induce tumour formation in a specific target organ, allowing study of the complete multistep carcinogenic process from initiation through promotion to malignant progression. In the widely used DMBA-induced mammary carcinogenesis model, a single or repeated oral or subcutaneous dose of DMBA is administered to young female Sprague-Dawley rats during a defined window of mammary gland susceptibility, producing palpable mammary tumours within eight to sixteen weeks that share substantial histopathological and hormone-responsiveness similarity with human breast cancer. Tumour incidence, latency, multiplicity, and volume (calculated from calliper measurements) are the principal endpoints, alongside terminal histopathological grading.