Animal Models and Their Value in Predicting Drug Efficacy and Toxicity

Journal Articles

Amrita Ahluwalia

Vinci B, Murphy E, Iori E, et al. An in vitro model of metabolism connecting adipose tissue, endothelial cells and hepatocytes in a multicompartmental bioreactor. Biotechnol. J. 2011.

Vozzi G, Lenzi T, Montemurro F, et al. A novel method to produce immobilised biomolecular concentration gradients to study cell activities: design and modelling. Mol. Biotechnol. 2011.

Vozzi F, Mazzei D, Vinci B, et al. A flexible bioreactor system for constructing in vitro tissue and organ models. Biotechnol. Bioeng. 2011;108(9):2129-2140.

Nick Andrews

Andrews N, O'Neill MF. It might be a big family but the pain remains-last chance saloon for selective 5-HT receptor ligands? Curr. Opin. Pharmacol. 2011;11(1):39-44.

Andrews N, Loomis S, Blake R, et al. Effect of gabapentin-like compounds on development and maintenance of morphine-induced conditioned place preference. Psychopharmacology (Berl.) 2001;157(4):381-387.

Andrews N, Legg E, Lisak D, et al. Spontaneous burrowing behaviour in the rat is reduced by peripheral nerve injury or inflammation associated pain. Eur. J. Pain 2011.

Jimmy D. Bell

Thomas EL, Parkinson JR, Frost GS, et al. The missing risk: MRI and MRS phenotyping of abdominal adiposity and ectopic fat. Obesity (Silver Spring). 2011.

Thomas EL, Parkinson JR, Hyde MJ, et al. Aberrant adiposity and ectopic lipid deposition characterize the adult phenotype of the preterm infant. Pediatr. Res. 2011;70(5):507-512.

Watson RA, Pride NB, Thomas EL, Ind PW, Bell JD. Relation between trunk fat volume and reduction of total lung capacity in obsese men. J. Appl. Physiol. 2011.

Maria Gabriela Belvisi

Belvisi MG, Dubuis E, Birrell MA. Transient receptor potential A1 channels: insights into cough and airway inflammatory disease. Chest 2011;140(4):1040-1047.

Eltom S, Stevenson CS, Rastrick J, et al. P2X7 receptor and caspase 1 activation are central to airway inflammation observed after exposure to tobacco smoke. PLoS ONE 2011;6(9):e24097.

Grace MS, Dubuis E, Birrell MA, Belvisi MG. TRP channel antagonists as potential antitussives. Lung 2011.

B. Taylor Bennett

Bailey MR, Rich BA, Bennett BT. Crisis planning to manage risks posed by animal rights extremists. ILAR J. 2010;51(2):138-148.

Bennett BT. Commentary on animal rights stirs more debate. J. Am. Vet. Med. Assoc. 2006;229(7):1080, author reply 1080-1081.

Halliday LC, Artwohl JE, Bunte RM, Ramakrishnan V, Bennett BT. Effects of Freund's complete adjuvant on the physiology, histology, and activity of New Zealand White Rabbits. Contemp. Top Lab Anim. Sci. 2004;43(1):8-13.

Odd-Geir Berge

Angeby-Möller K, Berge O, Hamers FPT. Using the CatWalk method to assess weight-bearing and pain behaviour in walking rats with ankle joint monoarthritis induced by carrageenan: effects of morphine and rofecoxib. J. Neurosci. Methods 2008;174(1):1-9.

Berge O. Predictive validity of behavioural animal models for chronic pain. Br. J. Pharmacol. 2011;164(4):1195-1206.

Krekels EHJ, Angesjö M, Sjögren I, et al. Pharmacokinetic-pharmacodynamic modeling of the inhibitory effects of naproxen on the time-courses of inflammatory pain, fever, and the ex vivo synthesis of TXB2 and PGE2 in rats. Pharm. Res. 2011;28(7):1561-1576.

Susan Brain

Bezerra MM, Brain SD, Girão VCC, et al. Neutrophils-derived peroxynitrite contributes to acute hyperalgesia and cell influx in zymosan arthritis. Naunyn Schmiedebergs Arch. Pharmacol. 2007;374(4):265-273.

Sharpley AL, Rawlings NB, Brain S, McTavish SFB, Cowen PJ. Does agomelatine block 5-HT2C receptors in humans? Psychopharmacology (Berl.) 2011;213(2-3):653-655.

Szabó A, Czirják L, Sándor Z, et al. Investigation of sensory neurogenic components in a bleomycin-induced scleroderma model using transient receptor potential vanilloid 1 receptor- and calcitonin gene-related peptide-knockout mice. Arthritis Rheum. 2008;58(1):292-301.

Camron D. Bryant

Bryant CD, Roberts KW, Culbertson CS, et al. Pavlovian conditioning of multiple opioid-like responses in mice. Drug Alcohol Depend. 2009;103(1-2):74-83.

Bryant CD, Chang HP, Zhang J, et al. A major QTL on chromosome 11 influences psychostimulant and opioid sensitivity in mice. Genes Brain Behav. 2009;8(8):795-805.

Hofford RS, Hodgson SR, Roberts KW, et al. Extracellular signal-regulated kinase activation in the amygdala mediates elevated plus maze behavior during opioid withdrawal. Behav. Pharmacol. 2009;20(7):576-583.

Julia C. Buckingham

Dufton N, Hannon R, Brancaleone V, et al. Anti-inflammatory role of the murine formyl-peptide receptor 2: ligand-specific effects on leukocyte responses and experimental inflammation. J. Immunol. 2010;184(5):2611-2619.

John CD, Gavins FNE, Buss NAPS, Cover PO, Buckingham JC. Annexin A1 and the formyl peptide receptor family: neuroendocrine and metabolic aspects. Curr. Opin. Pharmacol. 2008;8(6):765-776.

Solito E, McArthur S, Christian H, et al. Annexin A1 in the brain—undiscovered roles? Trends Pharmacol. Sci. 2008;29(3):135-142.

Gary A. Churchill

Aitman TJ, Boone C, Churchill GA, et al. The future of model organisms in human disease research. Nat. Rev. Genet. 2011;12(8):575-582.

Philip VM, Sokoloff G, Ackert-Bicknell CL, et al. Genetic analysis in the Collaborative Cross breeding population. Genome Res. 2011;21(8):1223-1238.

Yang H, Wang JR, Didion JP, et al. Subspecific origin and haplotype diversity in the laboratory mouse. Nat. Genet. 2011;43(7):648-655.

Julian R. Davis

Harper CV, Featherstone K, Semprini S, et al. Dynamic organisation of prolactin gene expression in living pituitary tissue. J. Cell. Sci. 2010;123(Pt 3):424-430.

Harper CV, Finkenstädt B, Woodcock DJ, et al. Dynamic analysis of stochastic transcription cycles. PLoS Biol. 2011;9(4):e1000607.

Semprini S, Friedrichsen S, Harper CV, et al. Real-time visualization of human prolactin alternate promoter usage in vivo using a double-transgenic rat model. Mol. Endocrinol. 2009;23(4):529-538.

Sandra Engle

Berna M, Ott L, Engle S, et al. Quantification of NTproBNP in rat serum using immunoprecipitation and LC/MS/MS: a biomarker of drug-induced cardiac hypertrophy. Anal. Chem. 2008;80(3):561-566.

Miller RA, Bookstein F, Van der Meulen J, et al. Candidate biomarkers of aging: age-sensitive indices of immune and muscle function covary in genetically heterogeneous mice. J. Gerontol. A Biol. Sci. Med. Sci. 1997;52(1):B39-47.

Wang J, Engle S, Zhang Y. A new in vitro system for activating the cell cycle checkpoint. Cell Cycle 2011;10(3):500-506.

Eric B. Fauman

Delaney AM, Printen JA, Chen H, Fauman EB, Dudley DT. Identification of a novel mitogen-activated protein kinase kinase activation domain recognized by the inhibitor PD 184352. Mol. Cell. Biol. 2002;22(21):7593-7602.

Fauman EB, Hopkins AL, Groom CR. Structural bioinformatics in drug discovery. Methods Biochem. Anal. 2003;44:477-497.

Fauman EB, Rai BK, Huang ES. Structure-based druggability assessment—identifying suitable targets for small molecule therapeutics. Curr. Opin. Chem. Biol. 2011;15(4):463-468.

Garret A. FitzGerald

FitzGerald GA. NCATS purrs: emerging signs of form and function. Sci. Transl. Med. 2011;3(83):83ed2.

Markosyan N, Chen EP, Ndong VN, et al. Deletion of cyclooxygenase 2 in mouse mammary epithelial cells delays breast cancer onset through augmentation of type 1 immune responses in tumors. Carcinogenesis 2011;32(10):1441-1449.

Ricciotti E, FitzGerald GA. Prostaglandins and inflammation. Arterioscler. Thromb. Vasc. Biol. 2011;31(5):986-1000.

Felicity N. E. Gavins

Gavins FNE, Russell J, Senchenkova EL, et al. Mechanisms of enhanced thrombus formation in cerebral microvessels of mice expressing hemoglobin-S. Blood 2011;117(15):4125-4133.

John CD, Gavins FNE, Buss NAPS, Cover PO, Buckingham JC. Annexin A1 and the formyl peptide receptor family: neuroendocrine and metabolic aspects. Curr. Opin. Pharmacol. 2008;8(6):765-776.

Solito E, McArthur S, Christian H, et al. Annexin A1 in the brain—undiscovered roles? Trends Pharmacol. Sci. 2008;29(3):135-142.

Aron Geurts

Dwinell MR, Lazar J, Geurts AM. The emerging role for rat models in gene discovery. Mamm. Genome 2011;22(7-8):466-475.

Moreno C, Hoffman M, Stodola TJ, et al. Creation and characterization of a renin knockout rat. Hypertension 2011;57(3):614-619.

Xu D, Guo H, Xu X, et al. Exacerbated pulmonary arterial hypertension and right ventricular hypertrophy in animals with loss of function of extracellular superoxide dismutase. Hypertension 2011;58(2):303-309.

Dale L. Greiner

Diiorio P, Jurczyk A, Yang C, et al. Hyperglycemia-induced proliferation of adult human beta cells engrafted into spontaneously diabetic immunodeficient NOD-Rag1null IL2rγnull Ins2Akita mice. Pancreas 2011;40(7):1147-1149.

Dorrell C, Grompe MT, Pan FC, et al. Isolation of mouse pancreatic alpha, beta, duct and acinar populations with cell surface markers. Mol. Cell. Endocrinol. 2011;339(1-2):144-150.

Schleifman EB, Bindra R, Leif J, et al. Targeted disruption of the CCR5 gene in human hematopoietic stem cells stimulated by peptide nucleic acids. Chem. Biol. 2011;18(9):1189-1198.

Sian Harding

Abdul Kadir SHS, Ali NN, Mioulane M, et al. Embryonic stem cell-derived cardiomyocytes as a model to study fetal arrhythmia related to maternal disease. J. Cell. Mol. Med. 2009;13(9B):3730-3741.

Brito-Martins M, Harding SE, Ali NN. β1- and β2-adrenoceptor responses in cardiomyocytes derived from human embryonic stem cells: comparison with failing and non-failing adult human heart. Br. J. Pharmacol. 2008;153(4):751-759.

Földes G, Mioulane M, Wright JS, et al. Modulation of human embryonic stem cell-derived cardiomyocyte growth: a testbed for studying human cardiac hypertrophy? J. Mol. Cell. Cardiol. 2011;50(2):367-376.

Thomas Hartung

Hartung T. Comparative analysis of the revised Directive 2010/63/EU for the protection of laboratory animals with its predecessor 86/609/EEC—a t4 report. ALTEX 2010;27(4):285-303.

Hartung T, McBride M. Food for thought ... on mapping the human toxome. ALTEX 2011;28(2):83-93.

Hartung T, Sabbioni E. Alternative in vitro assays in nanomaterial toxicology. Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 2011.

Kenneth L. Hastings

Hastings KL. Prospects for developmental immunotoxicity guidance and an update on ICH S8. J. Immunotoxicol. 2005;2(4):217-220.

Piccotti JR, Lebrec HN, Evans E, et al. Summary of a workshop on nonclinical and clinical immunotoxicity assessment of immunomodulatory drugs. J. Immunotoxicol. 2009;6(1):1-10.

Weaver JL, Tsutsui N, Hisada S, et al. Development of the ICH guidelines for immunotoxicology evaluation of pharmaceuticals using a survey of industry practices. J. Immunotoxicol. 2005;2(3):171-180.

Michael D. Hayward

Hayward MD, Low MJ. The contribution of endogenous opioids to food reward is dependent on sex and background strain. Neuroscience 2007;144(1):17-25.

Hayward MD, Jones BK, Saparov A, et al. An extensive phenotypic characterization of the hTNFalpha transgenic mice. BMC Physiol. 2007;7:13.

Thorogood D, Armstead IP, Turner LB, Humphreys MO, Hayward MD. Identification and mode of action of self-compatibility loci in Lolium perenne. L. Heredity 2005;94(3):356-363.

Simon Howell

Barrett JC, Clayton DG, Concannon P, et al. Genome-wide association study and meta-analysis find that over 40 loci affect risk of type 1 diabetes. Nat. Genet. 2009;41(6):703-707.

Hall MA, King NMP, Perdue LH, et al. Biobanking, consent, and commercialization in international genetics research: the Type 1 Diabetes Genetics Consortium. Clin. Trials 2010;7(1 Suppl):S33-45.

Hilner JE, Perdue LH, Sides EG, et al. Designing and implementing sample and data collection for an international genetics study: the Type 1 Diabetes Genetics Consortium (T1DGC). Clin. Trials 2010;7(1 Suppl):S5-S32.

Ann Jacqueline Hunter

Pugh PL, Ahmed SF, Smith MI, Upton N, Hunter AJ. A behavioural characterisation of the FVB/N mouse strain. Behav. Brain Res. 2004;155(2):283-289.

Rastan S, Hough T, Kierman A, et al. Towards a mutant map of the mouse—new models of neurological, behavioural, deafness, bone, renal and blood disorders. Genetica 2004;122(1):47-49.

Toye AA, Moir L, Hugill A, et al. A new mouse model of type 2 diabetes, produced by N-ethyl-nitrosourea mutagenesis, is the result of a missense mutation in the glucokinase gene. Diabetes 2004;53(6):1577-1583.

Ravi Iyengar

Berger SI, Iyengar R. Role of systems pharmacology in understanding drug adverse events. Wiley Interdiscip. Rev. Syst. Biol. Med. 2011;3(2):129-135.

He JC, Chuang PY, Ma'ayan A, Iyengar R. Systems biology of kidney diseases. Kidney Int. 2011.

Sobie EA, Lee Y, Jenkins SL, Iyengar R. Systems biology—biomedical modeling. Sci. Signal. 2011;4(190):tr2.

Mike Kastello

Kastello M. ACLAM position statement on animal experimentation. Comp. Med. 2003;53(5):472.

Klein H, Hasselschwert D, Handt L, Kastello M. A pharmacokinetic study of enrofloxacin and its active metabolite ciprofloxacin after oral and intramuscular dosing of enrofloxacin in rhesus monkeys (Macaca mulatta). J. Med. Primatol. 2008;37(4):177-183.

Julie Keeble

Alawi K, Keeble J. The paradoxical role of the transient receptor potential vanilloid 1 receptor in inflammation. Pharmacol. Ther. 2010;125(2):181-195.

Bezerra MM, Brain SD, Girão VCC, et al. Neutrophils-derived peroxynitrite contributes to acute hyperalgesia and cell influx in zymosan arthritis. Naunyn Schmiedebergs Arch. Pharmacol. 2007;374(4):265-273.

Liang L, Tam CW, Pozsgai G, et al. Protection of angiotensin II-induced vascular hypertrophy in vascular smooth muscle-targeted receptor activity-modifying protein 2 transgenic mice. Hypertension 2009;54(6):1254-1261.

Peter Kohl

Brook BS, Kohl P, King JR. Towards the virtual physiological human: mathematical and computational case studies. Philos. Transact. A Math. Phys. Eng. Sci. 2011;369(1954):4145-4148.

Hales PW, Burton RAB, Bollensdorff C, et al. Progressive changes in T₁, T₂, and left-ventricular histo-architecture in the fixed and embedded rat heart. NMR Biomed. 2011;24(7):836-843.

Kohl P, Hunter PJ, Winslow RL. Model interactions: 'It is the simple, which is so difficult'. Prog. Biophys. Mol. Biol. 2011;107(1):1-3.

Marc Lalande

Chamberlain SJ, Chen P, Ng KY, et al. Induced pluripotent stem cell models of the genomic imprinting disorders Angelman and Prader-Willi syndromes. Proc. Natl. Acad. Sci. USA 2010;107(41):17668-17673.

Leung KN, Chamberlain SJ, Lalande M, LaSalle JM. Neuronal chromatin dynamics of imprinting in development and disease. J. Cell. Biochem. 2011;112(2):365-373.

Martins-Taylor K, Nisler BS, Taapken SM, et al. Recurrent copy number variations in human induced pluripotent stem cells. Nat. Biotechnol. 2011;29(6):488-491.

Margaret S. Landi

Fillman-Holliday D, Landi MS. Animal care best practices for regulatory testing. ILAR J. 2002;43 Suppl:S49-58.

Rock FM, Landi MS, Meunier LD, et al. Diagnosis of a case of Mycobacterium tuberculosis in a cynomolgus (Macaca fascicularis) monkey colony by polymerase chain reaction and enzyme-linked immunosorbent assay. Lab. Anim. Sci. 1995;45(3):315-319.

Rock FM, Landi MS, Hughes HC, Gagnon RC. Effects of caging type and group size on selected physiologic variables in rats. Contemp. Top Lab Anim. Sci. 1997;36(2):69-72.

Jon Levine

Alvarez P, Ferrari LF, Levine JD. Muscle pain in models of chemotherapy-induced and alcohol-induced peripheral neuropathy. Ann. Neurol. 2011;70(1):101-109.

Ferrari LF, Chum A, Bogen O, Reichling DB, Levine JD. Role of Drp1, a key mitochondrial fission protein, in neuropathic pain. J. Neurosci. 2011;31(31):11404-11410.

Green PG, Chen X, Alvarez P, Ferrari LF, Levine JD. Early-life stress produces muscle hyperalgesia and nociceptor sensitization in the adult rat. Pain 2011;152(11):2549-2556.

Kent Lloyd

Barros CS, Calabrese B, Chamero P, et al. Impaired maturation of dendritic spines without disorganization of cortical cell layers in mice lacking NRG1/ErbB signaling in the central nervous system. Proc. Natl. Acad. Sci. USA 2009;106(11):4507-4512.

Fazzari P, Paternain AV, Valiente M, et al. Control of cortical GABA circuitry development by Nrg1 and ErbB4 signalling. Nature 2010;464(7293):1376-1380.

Villablanca A, Lubahn D, Shelby L, Lloyd K, Barthold S. Susceptibility to early atherosclerosis in male mice is mediated by estrogen receptor alpha. Arterioscler. Thromb. Vasc. Biol. 2004;24(6):1055-1061.

Mark Lythgoe

Price AN, Cheung KK, Lim SY, et al. Rapid assessment of myocardial infarct size in rodents using multi-slice inversion recovery late gadolinium enhancement CMR at 9.4T. J. Cardiovasc. Magn. Reson. 2011;13:44.

Riegler J, Lau KD, Garcia-Prieto A, et al. Magnetic cell delivery for peripheral arterial disease: A theoretical framework. Med. Phys. 2011;38(7):3932-3943.

Smart N, Bollini S, Dubé KN, et al. De novo cardiomyocytes from within the activated adult heart after injury. Nature 2011;474(7353):640-644.

Gary Mirams

Cooper J, Mirams GR, Niederer SA. High-throughput functional curation of cellular electrophysiology models. Prog. Biophys. Mol. Biol. 2011;107(1):11-20.

Fletcher K, Shah RR, Thomas A, et al. Novel approaches to assessing cardiac safety—proceedings of a workshop: regulators, industry and academia discuss the future of in silico cardiac modelling to predict the proarrhythmic safety of drugs. Drug Saf. 2011;34(5):439-443.

Quinn TA, Granite S, Allessie MA, et al. Minimum Information about a Cardiac Electrophysiology Experiment (MICEE): Standardised reporting for model reproducibility, interoperability, and data sharing. Prog. Biophys. Mol. Biol. 2011;107(1):4-10.

Jane Mitchell

Harrington LS, Lundberg MH, Waight M, Rozario A, Mitchell JA. Reduced endothelial dependent vasodilation in vessels from TLR4(−/−) mice is associated with increased superoxide generation. Biochem. Biophys. Res. Commun. 2011;408(4):511-515.

Kirkby NS, Leadbeater PDM, Chan MV, et al. Antiplatelet effects of aspirin vary with level of P2Y₁₂ receptor blockade supplied by either ticagrelor or prasugrel. J. Thromb. Haemost. 2011;9(10):2103-2105.

Leadbeater PDM, Kirkby NS, Thomas S, et al. Aspirin has little additional anti-platelet effect in healthy volunteers receiving prasugrel. J. Thromb. Haemost. 2011;9(10):2050-2056.

Alysson Renato Muotri

Beltrão-Braga PICB, Pignatari GC, Maiorka PC, et al. Feeder-free derivation of induced pluripotent stem cells from human immature dental pulp stem cells. Cell Transplant. 2011.

Mitne-Neto M, Machado-Costa M, Marchetto MCN, et al. Downregulation of VAPB expression in motor neurons derived from induced pluripotent stem cells of ALS8 patients. Hum. Mol. Genet. 2011;20(18):3642-3652.

Muotri AR, Marchetto MCN, Coufal NG, et al. L1 retrotransposition in neurons is modulated by MeCP2. Nature 2010;468(7322):443-446.

Andrew Murphy

Kang K, Schmahl J, Lee J, et al. Mouse ghrelin-O-acyltransferase (GOAT) plays a critical role in bile acid reabsorption. FASEB J. 2011.

Okamoto H, Latres E, Liu R, et al. Genetic deletion of Trb3, the mammalian Drosophila tribbles homolog, displays normal hepatic insulin signaling and glucose homeostasis. Diabetes 2007;56(5):1350-1356.

Wortley KE, Garcia K, Okamoto H, et al. Peptide YY regulates bone turnover in rodents. Gastroenterology 2007;133(5):1534-1543.

Kevin G. Murphy

Addison ML, Minnion JS, Shillito JC, et al. A role for metalloendopeptidases in the breakdown of the gut hormone, PYY3-36. Endocrinology 2011.

Amin A, Dhillo WS, Murphy KG. The central effects of thyroid hormones on appetite. J. Thyroid Res. 2011;2011:306510.

Beale KEL, Murphy KG, Harrison EK, et al. Accurate measurement of body weight and food intake in environmentally enriched male Wistar rats. Obesity (Silver Spring) 2011;19(8):1715-1721.

Eric M. Ostertag

Babushok DV, Ostertag EM, Courtney CE, Choi JM, Kazazian HH. L1 integration in a transgenic mouse model. Genome Res. 2006;16(2):240-250.

Kano H, Godoy I, Courtney C, et al. L1 retrotransposition occurs mainly in embryogenesis and creates somatic mosaicism. Genes Dev. 2009;23(11):1303-1312.

Ostertag EM, Madison BB, Kano H. Mutagenesis in rodents using the L1 retrotransposon. Genome Biol. 2007;8 Suppl 1:S16.

Clive P. Page

Calzetta L, Spina D, Cazzola M, et al. Pharmacological characterization of adenosine receptors on isolated human bronchi. Am. J. Respir. Cell Mol. Biol. 2011.

Cazzola M, Calzetta L, Page CP, et al. Protein prenylation contributes to the effects of LPS on EFS-induced responses in human isolated bronchi. Am. J. Respir. Cell Mol. Biol. 2011;45(4):704-710.

Keir SD, Spina D, Douglas G, Herd C, Page CP. Airway responsiveness in an allergic rabbit model. J. Pharmacol. Toxicol. Methods 2011;64(2):187-195.

Roger Pedersen

Brown S, Teo A, Pauklin S, et al. Activin/Nodal signaling controls divergent transcriptional networks in human embryonic stem cells and in endoderm progenitors. Stem Cells 2011;29(8):1176-1185.

Hemberger M, Pedersen R. Stem cells. Epigenome disruptors. Science 2010;330(6004):598-599.

Vallier L, Pedersen R. Differentiation of human embryonic stem cells in adherent and in chemically defined culture conditions. Curr. Protoc. Stem Cell Biol. 2008; Chapter 1: Unit 1D.4.1-1D.4.7.

Tony M. Plant

Plant TM. Undifferentiated primate spermatogonia and their endocrine control. Trends Endocrinol. Metab. 2010;21(8):488-495.

Ramaswamy S, Seminara SB, Ali B, et al. Neurokinin B stimulates GnRH release in the male monkey (Macaca mulatta) and is colocalized with kisspeptin in the arcuate nucleus. Endocrinology 2010;151(9):4494-4503.

Ramaswamy S, Seminara SB, Plant TM. Evidence from the agonadal juvenile male rhesus monkey (Macaca mulatta) for the view that the action of neurokinin B to trigger gonadotropin-releasing hormone release is upstream from the kisspeptin receptor. Neuroendocrinology 2011.

Alexander Ploss

Billerbeck E, Barry WT, Mu K, et al. Development of human CD4+FoxP3+ regulatory T cells in human stem cell factor-, granulocyte-macrophage colony-stimulating factor-, and interleukin-3-expressing NOD-SCID IL2Rγ(null) humanized mice. Blood 2011;117(11):3076-3086.

Dorner M, Horwitz JA, Robbins JB, et al. A genetically humanized mouse model for hepatitis C virus infection. Nature 2011;474(7350):208-211.

Marukian S, Andrus L, Sheahan TP, et al. Hepatitis C virus induces interferon-λ and interferon-stimulated genes in primary liver cultures. Hepatology 2011.

Leonard D. Shultz

Diiorio P, Jurczyk A, Yang C, et al. Hyperglycemia-induced proliferation of adult human beta cells engrafted into spontaneously diabetic immunodeficient NOD-Rag1null IL2rγnull Ins2Akita mice. Pancreas 2011;40(7):1147-1149.

Dorrell C, Grompe MT, Pan FC, et al. Isolation of mouse pancreatic alpha, beta, duct and acinar populations with cell surface markers. Mol. Cell. Endocrinol. 2011;339(1-2):144-150.

Schleifman EB, Bindra R, Leif J, et al. Targeted disruption of the CCR5 gene in human hematopoietic stem cells stimulated by peptide nucleic acids. Chem. Biol. 2011;18(9):1189-1198.

Donald B. Stedman

Chapin RE, Stedman DB. Endless possibilities: stem cells and the vision for toxicology testing in the 21st century. Toxicol. Sci. 2009;112(1):17-22.

Marx-Stoelting P, Adriaens E, Ahr H, et al. A review of the implementation of the embryonic stem cell test (EST). The report and recommendations of an ECVAM/ReProTect Workshop. Altern. Lab Anim. 2009;37(3):313-328.

Paquette JA, Kumpf SW, Streck RD, et al. Assessment of the Embryonic Stem Cell Test and application and use in the pharmaceutical industry. Birth Defects Res. B Dev. Reprod. Toxicol. 2008;83(2):104-111.

Steven L. Stice

Dodla MC, Young A, Venable A, et al. Differing lectin binding profiles among human embryonic stem cells and derivatives aid in the isolation of neural progenitor cells. PLoS ONE 2011;6(8):e23266.

Lu Y, West FD, Jordan BJ, et al. Avian induced pluripotent stem cells derived using human reprogramming factors. Stem Cells Dev. 2011.

Simpson DL, Boyd NL, Kaushal S, Stice SL, Dudley SC. Use of human embryonic stem cell derived-mesenchymal cells for cardiac repair. Biotechnol. Bioeng. 2011.

William S. Stokes

Haseman JK, Strickland J, Allen D, et al. Safety assessment of allergic contact dermatitis hazards: an analysis supporting reduced animal use for the murine local lymph node assay. Regul. Toxicol. Pharmacol. 2011;59(1):191-196.

Haseman JK, Allen DG, Lipscomb EA, Truax JF, Stokes WS. Using fewer animals to identify chemical eye hazards: revised criteria necessary to maintain equivalent hazard classification. Regul. Toxicol. Pharmacol. 2011;61(1):98-104.

Stokes WS, Wind M. Validation of innovative technologies and strategies for regulatory safety assessment methods: challenges and opportunities. ALTEX 2010;27(3):87-95.

Eric M. Walters

Men H, Zhao C, Si W, et al. Birth of piglets from in vitro-produced, zona-intact porcine embryos vitrified in a closed system. Theriogenology 2011;76(2):280-289.

Whyte JJ, Zhao J, Wells KD, et al. Gene targeting with zinc finger nucleases to produce cloned eGFP knockout pigs. Mol. Reprod. Dev. 2011;78(1):2.

Whyte JJ, Samuel M, Mahan E, et al. Vascular endothelium-specific overexpression of human catalase in cloned pigs. Transgenic Res. 2011;20(5):989-1001.

Dominic Wells

Kaneb HM, Sharp PS, Rahmani-Kondori N, Wells DJ. Metformin treatment has no beneficial effect in a dose-response survival study in the SOD1(G93A) mouse model of ALS and is harmful in female mice. PLoS ONE 2011;6(9):e24189.

Muntoni F, Torelli S, Wells DJ, Brown SC. Muscular dystrophies due to glycosylation defects: diagnosis and therapeutic strategies. Curr. Opin. Neurol. 2011;24(5):437-442.

Muses S, Morgan JE, Wells DJ. A new extensively characterised conditionally immortal muscle cell-line for investigating therapeutic strategies in muscular dystrophies. PLoS ONE 2011;6(9):e24826.

Garth Whiteside

Anand P, Whiteside G, Fowler CJ, Hohmann AG. Targeting CB2 receptors and the endocannabinoid system for the treatment of pain. Brain Res. Rev. 2009;60(1):255-266.

Hummel M, Strassle B, Miller S, Kaftan E, Whiteside G. Anatomical localization and expression pattern for the NMDA-2D receptor subunit in a rat model of neuropathic pain. Neuroscience 2008;155(2):492-502.

Whiteside G, Doyle CA, Hunt SP, Munglani R. Differential time course of neuronal and glial apoptosis in neonatal rat dorsal root ganglia after sciatic nerve axotomy. Eur. J. Neurosci. 1998;10(11):3400-3408.

Brian P. Zambrowicz

Tang T, Li L, Tang J, et al. A mouse knockout library for secreted and transmembrane proteins. Nat. Biotechnol. 2010;28(7):749-755.

Vogel P, Read R, Hansen GM, et al. Situs inversus in Dpcd/Poll−/−, Nme7−/−, and Pkd1l1−/− mice. Vet. Pathol. 2010;47(1):120-131.

Vogel P, Read RW, Hansen GM, et al. Congenital hydrocephalus in genetically engineered mice. Vet. Pathol. 2011.

Sponsors

Presented by

Bronze Sponsors

GlaxoSmithKline
Pfizer
Sanofi-Aventis
The Global Medical Excellence Cluster (GMEC)
Wellcome Trust

Academy Friends

Bristol-Myers Squibb
British Pharmacological Society
Charles River
Imperial College London
King's College London
National Swine Research and Resource Center
Sigma Advanced Genetic Engineering (SAGE Labs)
Taconic
The Jackson Laboratory
The Physiological Society

Grant Support

Funding for this conference was made possible by the Office of the Director of the National Institute of Environmental Health Sciences and grant number R13RR032638 from the National Center for Research Resources, National Institute of General Medical Sciences, and National Institute of Diabetes and Digestive and Kidney Diseases. The views expressed in written conference materials or publications and by speakers and moderators do not necessarily reflect the official policies of the Department of Health and Human Services, nor does mention by trade names, commercial practices, or organizations imply endorsement by the U.S. Government.

Promotional Partners

Americans for Medical Progress

Animals Scientific Procedures Inspectorate

Bioinformatics Journal (Oxford University Press)

Biotechnology and Biological Sciences Research Council

Biotechnology Industry Organization (BIO)

British Pharmacological Society

Charles River - Cerebricon

Clinical Immunology Society

Edinburgh Neuroscience at the University of Edinburgh

Federation of European Neurosciences

Federation of Laboratory Animal Science Associations

Foundation for Biomedical Research (FBR)

The Hastings Center

International Neuroethics Society

International Society for Computational Biology

Journal of Experimental Medicine (Rockefeller University Press)

Laboratory Animal Management Association (LAMA)

National Association for Biomedical Research (NABR)

Nature Medicine

New York Biotechnology Association (NYBA)

Nucleic Acids Research Journal (Oxford University Press)

Since the earliest days of medical research, scientists have used model organisms to understand human biology. From ancient Greek analyses of comparative anatomy, to J.S. Haldane's studies on decompression sickness, to the modern pharmaceutical development pipeline, animals have provided handy surrogates for measuring all types of biological phenomena.

For just as long, researchers have understood that no animal model is a perfect representation of humans. But while vivisection of prisoners may have been acceptable to Aristotle's contemporaries, modern biomedical scientists must navigate an ethical mine field even when working on animals and especially when working with human subjects.

At the New York Academy of Sciences conference on Animal Models and Their Value in Predicting Drug Efficacy and Toxicity, held September 15 – 16, 2011, researchers from around the world discussed the ways animal experiments inform—and sometimes misinform—the vast research effort that now underpins the regulation of drugs and toxic chemicals.

The meeting began with a keynote presentation by Jackie Hunter who provided a broad overview of the problems facing researchers. While pointing out the numerous shortcomings of animal research, Hunter emphasized that such research remains at the heart of preclinical drug development and chemical toxicity testing: "We are concerned because we are not able to have models that are as predictive as we had hoped, but let's not forget that ... actually animal models have been very valuable in coming up with new medicines for a range of conditions and disorders."

After a set of concurrent workshop sessions, attendees reconvened for a joint discussion on regulations and best practices. Though scientists are always anxious to dig right into an experiment, any animal studies first require careful ethical review, in order to evaluate the work against a changing backdrop of rules, guidelines, and ethical norms. Research policy experts from both the U.S. and the European Union reviewed the current regulatory framework, and then an interactive panel discussion allowed audience members to share their own insights on ethical animal experimentation.

At another joint session the meeting's focus shifted to new animal models, especially ones developed with the latest techniques in genetic engineering and cell culture. One of the most exciting developments in this area is the rapid advance in embryonic stem cell research, and the resulting potential for growing genetically engineered organs of one species inside bodies of another species. That work could lead to much more human-like laboratory models, but it also raises its own set of ethical concerns. "I think that's an experiment that would, if it worked ... potentially produce developing fetuses or at least in vitro developing embryos with human tissues in a native human context," said Roger Pedersen of the University of Cambridge.

The conference's first day ended with another set of parallel workshops. The group then spent most of the second day in an extended session on new technologies for animal studies. Presentations covered a wide range of approaches, including several talks that emphasized the potential for novel imaging and analytical techniques that could reduce the number of animals that are necessary for an experiment, while simultaneously providing higher-quality data. Other speakers talked about entirely computerized strategies that use sophisticated algorithms to simulate human biology without needing animals at all. While both approaches are clearly advancing, the talks and the subsequent panel discussion emphasized that the field is still in its infancy, and that animal models will remain an essential part of research for the foreseeable future.

Following lunch and a poster session, the group split into another set of parallel workshops, and then reconvened for a final panel discussion. Besides a general review of the issues discussed in earlier sessions, presenters talked about some of the other thorny problems still facing the field, such as the tendency to bury negative results and the difficulty of determining what resources are available in potential collaborators' institutions. Attendees also suggested possible solutions that could help optimize animal studies across disciplines.

Speakers:
Ann Jacqueline Hunter, OI Pharma Partners, Ltd., Cambridge, UK
Brian P. Zambrowicz, Lexicon Pharmaceuticals, Inc.
Kevin G. Murphy, Imperial College London, London, UK
Jason Kim, University of Massachusetts Medical School, Mouse Metabolic Phenotyping Center
Alan Daugherty, University of Kentucky
Amrita Ahluwalia, Barts & The London School of Medicine, Queen Mary University of London, London, UK
Jane Mitchell, Imperial College London, London, UK
Garret A. FitzGerald, University of Pennsylvania School of Medicine
Julie Keeble, King's College London, London, UK
Nick Andrews, Pfizer Global Research and Development
Jon Levine, University of California, San Francisco School of Medicine
Garth Whiteside, Purdue Pharma
Dominic Wells, The Royal Veterinary College, London, UK
B. Taylor Bennett, Foundation for Biomedical Research
William S. Stokes, National Toxicology Program Interagency Center for the Evaluation of Alternative Toxicological Methods, National Institute of Environmental Health Sciences, NIH

Highlights

• Rising costs and declining productivity in pharmaceutical development have researchers seeking better animal models.
• Several new research models are now available for studying obesity, cardiovascular disease, and pain.
• Several new models for safety testing have significantly reduced animal use, eliminated pain and distress, and replaced animal use in a few situations.
• Regulations on animal research vary widely between countries.
• Good animal care is crucial for scientific reproducibility.

The clog in the pipeline

Ann Jacqueline Hunter, of OI Pharma Partners, Ltd., kicked off the meeting with a keynote presentation that surveyed the major problems in animal models of disease. Basic scientists use such models regularly, but their biggest and most economically important use is in industry, particularly in pharmaceutical development. In that context, the news is not good.

For several years, industry watchers have been pointing to a major problem with the pharmaceutical business model: even as companies have spent increasing amounts of money on research and development, they have seen decreasing returns on that investment. Hunter explained the crux of the problem: "Productivity in terms of new chemical or new molecular entities approved has been pretty steady over the past few decades. What has changed substantially is the cost ... and frankly this has reached a level that is not sustainable, so we have to do something about it."

More specifically, drugs have been failing at a higher rate in phase II trials, which is the point at which researchers first test efficacy in humans. The problem, according to Hunter, is that current preclinical animal models often do a poor job predicting a drug's efficacy in human disease. As a result, pharmaceutical companies may be taking the wrong compounds into clinical trials and abandoning potentially good ones prematurely.

To address this problem, Hunter advocates fundamentally changing the way drug developers, toxicologists, and regulatory agencies operate. Instead of moving progressively from simple cultured cell models to imperfect animal models and then into clinical trials, future drug developers may need to focus more on the underlying mechanisms at work in a disease and optimize drugs to affect that mechanism. "I do think that [adjustment] demands a change of mindset for all the stakeholders involved," said Hunter.

Modeling metabolic disease, heart disease, and pain

The group then split up for three parallel sessions. In the metabolic disease workshop, three speakers discussed different ways to study conditions such as diabetes and obesity. The content of the presentations ranged from case studies of particular drug development efforts to an overview of basic researchers' efforts to speed up metabolic phenotyping.

The cardiovascular disease workshop focused on another major field of pharmaceutical effort, with four presentations covering the complex pathologies of cardiac and circulatory diseases. Different animal models of cardiovascular biology often produce conflicting results, but new techniques may help address some of this variability.

Pain was the topic of the third workshop. In particular, researchers discussed the difficulty of measuring pain in laboratory animals. In the four presentations in this session, researchers discussed new models for understanding pain and for testing analgesics.

Good mousekeeping

In the meeting's second joint session, scientists closely involved in policymaking described the current regulatory situation for animal research in the U.S. and in Europe. Dominic Wells of the Royal Veterinary College discussed the difficulty of coordinating such policies across the European Union member states, all of which already have their own regulations in place.

British rules for animal experimentation date all the way back to 1876, when experiments by some 19th century biologists caused a public outcry. "That's where the term vivisection really applies, because a lot of that work was done without anesthesia ... and therefore was a matter of great public concern," said Wells.

Harmonizing animal regulations across the diverse countries of the European Union will require a massive effort. (Image courtesy of Dominic Wells)

Over a century later, the UK replaced that law with the Animal Scientific Procedures Act of 1986, which established a rigorous system of licensing and placed a heavy emphasis on "the three Rs": reduction, refinement, and replacement of animal models in research. In 2010 the European Union passed a pair of directives that seek to harmonize regulations across the continent.

While Wells supports the new emphasis on harmonization, he also advocates keeping the laboratory animal problem in perspective: "In the UK we use something on the order of three and a half million animals in research annually, but we eat two and a half billion fish and other animals, therefore we eat seven hundred times the number of animals that are used in research."

B. Taylor Bennett, from the National Association for Biomedical Research, described the overall regulatory regime for animal research in the United States. The main legal structure for this work is the Animal Welfare Act (AWA), passed in 1966 and amended half a dozen times since. The AWA places responsibility for experimental animals with the Department of Agriculture, but exempts rats, mice, and birds from those rules. However, the Public Health Service covers those species, establishing guidelines and requiring detailed compliance plans for experiments on all vertebrates. The non-governmental Association for Assessment and Accreditation of Laboratory Animal Care also maintains a progressive set of standards that many funding agencies require researchers to meet.

Taking good care of laboratory animals is not just a matter of law; it is also good science. "By assuring animal welfare, what you've done is minimize non-experimental variability. That's a very important issue," said Bennett. He also highlighted the distinction between an animal's phenotype, which describes a trait that persists over its lifetime, and its dramatype, which is its performance on a particular physiological measurement at a specific point in time. Good animal care helps reduce the variability of dramatypes from one experiment to the next.

New analytical techniques allow researchers to end animal tests sooner, producing clearer results and more humane assays. (Image courtesy of William Stokes)

The session's final speaker was William Stokes, from the NIH's National Institute of Environmental Health Sciences, who focused on recent advances in alternative methods for safety testing as a result of efforts by the U.S. National Toxicology Program's Interagency Center for the Evaluation of Alternative Toxicological Methods (NICEATM) and the U.S. Interagency Coordinating Committee on the Validation of Alternative Methods (ICCVAM) . The Center's and ICCVAM’s purpose, by law, is to reduce, refine, and replace animal models in toxicology testing. So far, it has made good progress. Since 1999 NICEATM and its collaborating Federal partners have contributed to over 43 alternative safety testing methods that have been accepted or endorsed by U.S. and international agencies.

As an example of a successful project, Stokes pointed to the testing procedures for botulinum toxin, which is used as a therapeutic and cosmetic drug worldwide. Traditionally, the protein's manufacturer had to determine the potency of each batch using the standard LD50 mouse test, which estimates the dose that kills 50% of the animals receiving it. Stokes estimates that test consumes about 600,000 laboratory mice annually worldwide. But after a decade of work, researchers developed a cellular potency assay that regulators are starting to accept as a replacement for the LD50 mouse potency test.

Stokes also described how systems biology approaches are being applied to better understand how chemicals and drugs alter normal molecular pathways and gene expression pathways to cause toxic effects. Understanding how perturbations to these pathways lead to toxicity is expected to yield sensitive biomarkers that can be applied to improve both animal and cellular models for safety assessments.

The session concluded with a panel discussion. After a brief series of opening remarks, the speakers fielded questions and comments from the audience. The discussion ranged from speculation about the impact of feed choices on experimental results to the problem of modeling idiosyncratic toxicity.

Speakers:
Roger Pedersen, University of Cambridge, Cambridge, UK
Aron Geurts, Medical College of Wisconsin
Steven L. Stice, University of Georgia
Andrew Murphy, Regeneron Pharmaceuticals, Inc.
Kent Lloyd, University of California, Davis
Eric M. Ostertag, Transposagen Biopharmaceuticals, Inc.
Gary A. Churchill, The Jackson Laboratory
Michael D. Hayward, Taconic
Camron D. Bryant, University of Chicago
Maria Gabriela Belvisi, Imperial College London, London, UK
Felicity N. E. Gavins, Imperial College London, London, UK
Susan Brain, King's College London, London, UK
Odd-Geir Berge, AstraZeneca R&D
Clive P. Page, King's College London, London, UK

Highlights

• Pluripotent stem cells may allow researchers to grow human-like tissues in laboratory animals.
• The availability of new genetic tools has many scientists returning to rat models, which may be better than mice for some studies.
• Two new techniques allow investigators to make precise gene deletions in multiple species.
• A novel technology that can replace entire genomic loci could revolutionize immunology research.
• Scientists need to pay closer attention to the genetic backgrounds of their animal models.

Not your father's rat

After lunch, the discussion moved to the latest advances in genetically modifying animals for drug discovery and development. Roger Pedersen of the University of Cambridge started the next joint session with results from one of the hottest new technologies in the field: pluripotent stem cells. These cells can develop into any of several tissue types, depending on their environment. Manipulating human pluripotent cells in vitro is a common tactic for cellular studies, but could these cells also form human-like organs inside laboratory animals?

Perhaps, said Pedersen. Citing several recent studies, he described the steps necessary to grow tissues of one species inside another species. When researchers put rat embryonic stem cells into mouse blastocyst-stage embryos, the resulting animal has a mixture of rat and mouse cells in many tissues. In a mouse strain that has been genetically modified to eliminate pancreatic development, the entire pancreas in the chimeric animals is made from rat cells. "You have to ablate the niche of its normal components, which would be the mouse pancreas tissue," said Pedersen.

Chimeric mice show that it is possible to grow organs of one species in another. (Image courtesy of Roger Pedersen)

While that model is informative, Pedersen explained that putting human-like organs in an animal may be a somewhat taller order. Several teams of researchers are now using different strategies to generate human pluripotent stem cells, with varying degrees of success.

The chief problem is that murine stem cells behave like "naive" early embryonic cells, while the human stem cells produced to date seem to display the "primed" phenotype of later embryonic cells, which have begun to differentiate along particular lineages and therefore may be unable to form a full range of tissues outside their normal context. "We don't yet have any functional information about the human type of pluripotency in a normal tissue context, we only have the information from in vitro, which is not a normal tissue context, and from the teratomas, which is not functional," said Pedersen. Teratomas are tumors that develop much like embryos, but they do not grow fully functional organs of their own.

Though researchers will have to wait a while to get rats with human livers, other promising new rat models are available now. Aron Geurts of the Medical College of Wisconsin described the field's recent return to this classic animal system.

Historically, rats were the dominant animal model for pharmaceutical and toxicological research, but mice overtook them around 2000, largely because of the relative ease of modifying mouse genetics. Now that relationship is changing. "There's been just an enormous explosion of [rat] models, and it's really because of the development of new technologies," said Geurts.

One of those new technologies is the deletion of rat genes with zinc finger nucleases, a method Geurts and his colleagues pioneered. Zinc finger nucleases consist of a DNA binding domain fused to an endonuclease. By designing the DNA binding portion correctly, researchers can target the nuclease to cleave a specific site in an animal's genome and to disrupt the gene using simple embryo microinjection techniques. The method allows investigators to delete or replace genes with extraordinary precision. An analogous technique, called TAL Effector-nuclease, uses a similar approach.

After proving that the strategy works, Geurts and his colleagues applied it to disrupt the rat orthologs of human genes implicated in hypertension and renal failure. Genome-wide association studies in large human populations have identified over 1,000 candidate genes in this phenomenon, but few of those leads had been tested in animals. Doing so with traditional gene deletion techniques in rat embryonic stem cells would be prohibitively expensive and slow.

Using zinc finger nucleases, though, Geurts's team has already deleted or modified over 100 of the candidate genes, and performed rapid phenotype analysis to examine blood pressure and renal disease traits in the resulting rats. "The fact that we've been able to knock out 96% of the genes that we've gone after with the zinc finger nucleases I think speaks to the reproducibility and general applicability of this technology," said Geurts. The technique also works well in other common laboratory animals, such as zebrafish, mice, rabbits, and pigs.

Pluripotent pigs

Researchers have developed new genetic engineering techniques for pigs. (Image courtesy of Steven Stice)

While pigs are not an animal model at the front of most researchers' minds, Steven Stice of the University of Georgia argued persuasively that they should be. Classic models such as rats and mice are compact and inexpensive, but these tiny rodents provide only a rough approximation of the biology of large omnivores such as humans. Pigs, however, are a lot like people. Besides having similar sizes and diets, pigs develop many similar diseases, including atherosclerosis and white matter ischemia.

Scientists initially had trouble getting porcine embryonic stem cells to behave well in the lab, which limited the pig's utility as a disease model. Stice and his colleagues have helped remedy that problem. Using the same mixture of growth factors that has worked well to reprogram human cells, the researchers created lines of porcine pluripotent stem cells.

The new cells seemed to establish themselves easily when injected into pig embryos. "To our surprise really, we did see a high level of chimerism. There were only two fetuses that were negative, and you can see when we took tissue from the various organ systems, we were able to see different levels of chimerism rates within [the chimeric fetuses]," said Stice.

When the researchers bred the chimeric animals, some carried markers from the original stem cells, indicating that the manipulated cells had propagated through the germline. Unfortunately, only two such fetuses developed, and one was stillborn while the other died five days after birth. "So yes, we made germline chimeras. Is it as robust as we'd like it to be? No, it's not there yet," said Stice. The team is pressing ahead to refine the technique, and Stice hopes to use it to develop new models for testing drugs, especially against circulatory diseases such as stroke.

The big swap

Andrew Murphy of Regeneron Pharmaceuticals, Inc. gave the final presentation in the session, and described a family of technologies for making large gene deletions and for generating human antibodies in mice. For gene deletions, Murphy and his colleagues have developed the VelociGene system, which uses bacterial artificial chromosomes that can carry huge pieces of DNA into mouse embryonic stem cells.

By inserting large regions homologous to any target region in the mouse genome, researchers can use the system to make precise gene deletions or insertions, and to replace entire genetic loci. The system is also amenable to automation, so Murphy's team can now generate large numbers of mouse strains with deletions or modifications in different genes.

Having established that technique, the investigators turned their attention to antibodies. Antibody drugs are a mainstay of the biotechnology industry, but because they are generally developed in mice, drug developers must take pains to "humanize" the antibodies' constant regions to avoid triggering deleterious immune responses in patients. In a tour de force of gene replacement, Murphy's team removed the entire antibody loci in mice, and replaced them with the human loci. "In every single measure ... we didn't see any difference between the ability of these mice to make antibodies and [the ability of] wild-type mice, so now we can make fully human antibodies in mice as efficiently as you can make a mouse antibody in mice," said Murphy.

Following the talks, a panel discussion covered the feasibility of using rat-human chimeras in drug studies, the need to characterize background disease rates in newly generated models, and the ethical challenges of developing any type of chimeric animal using human cells.

For the remainder of the afternoon, the conference split into two workshop sessions. In one, attendees heard a series of talks about screening animal phenotypes and the importance of evaluating genetic backgrounds. The presentations described a variety of resources for researchers working with rats and mice, and the impact of strain choices on different phenotypes. The other workshop focused on the link between vascular inflammation and pain. Presenters covered models for lung disease and inflammatory pain. Both sessions included interactive panel discussions with extensive audience participation.

Speakers:
Mark Lythgoe, University College London, London, UK
Julian R. Davis, The University of Manchester, Manchester, UK
Jimmy D. Bell, Imperial College London, London, UK
Thomas Hartung, Johns Hopkins University Center for Alternatives to Animal Testing
Eric B. Fauman, Pfizer Worldwide Research

Highlights

• Imaging technologies such as MRI can provide unprecedented detail about experimental animals' phenotypes.
• Live cell imaging allows investigators to track gene regulation in real time.
• High-resolution animal and human imaging studies reveal that fat distribution is a crucial factor in obesity.
• Current animal models for toxicology testing are woefully inadequate.
• Data mining can uncover useful correlations between animal and human phenotypes.

Phenotyping animal models by imaging

The meeting's second day featured an extended session on new technologies for animal research. Mark Lythgoe of University College London provided an overview of the cutting edge of biomedical imaging, which plays a growing role in modern animal model studies.

For imaging facilities like the one Lythgoe runs, the deluge of new genetically-modified models is simultaneously invigorating and mind-boggling. "To try to phenotype all these animals is no small problem, and I just like the challenge of this," said Lythgoe, adding that "there's definitely been a push to see if imaging is a new platform to phenotype all the animals."

The problem is both huge and tiny. Researchers need to screen hundreds of animals to get robust phenotypic data, but the creatures and the structures inside them are typically minuscule. In one project, for example, Lythgoe's team has been looking for holes in the hearts of developing mouse embryos. Each embryo is about the size of a thumbnail, and the hole the diameter of a human hair. Pushing their imaging techniques to the limits, the investigators were eventually able to identify these tiny defects in batches of 40 embryos at once.

High resolution imaging shows the area of myocardial infarction (top row) and a 3D rendering of the major blood vessels (bottom left) in a mouse's heart. Imaging and computer reconstructions reveal (lower right image) the location of stem-cell-derived tissue in red. (Images courtesy of Mark Lythgoe)

That created a new problem, though. "If every embryo has a thousand [virtual] slices, in order to look through 40 embryos, it's literally weeks of work...," said Lythgoe. To address that issue, the team developed a new computer algorithm that automates the process.

Besides high-throughput phenotyping, some technologies, such as magnetic resonance imaging (MRI), might be useful for targeted treatments. In one study, Lythgoe and his colleagues loaded stem cells with nanoparticles of iron oxide, and then used the powerful magnetic field of their MRI machine to guide the cells to particular sites in a mouse. "This is a new technique called magnetic resonance targeting, and it actually doesn't take that much to modify a clinical MRI system from an interventional tool into a therapeutic tool," said Lythgoe.

Julian Davis of the University of Manchester picked up the imaging theme with a presentation on his group's work imaging live cells inside rat pituitary glands. Pituitary tumors are relatively common in humans, and they lead to secretory disorders, but many aspects of this gland's function are still obscure. To probe the problem, Davis and his colleagues monitored live rat pituitary cells expressing a reporter gene from the promoter for prolactin, a hormone normally secreted in the pituitary gland.

The cells' prolactin expression pattern was bizarre: rather than each cell expressing steady levels of the reporter, gene expression fluctuated wildly across the culture, and within a given cell over time. To explain the result, the scientists hypothesized that pituitary gene expression is stochastic, with each cell having a particular probability of expressing the gene at any given point in time. Stimuli that favor hormone production skew the odds toward more cells expressing the gene overall, but each individual cell's expression can still vary widely.

It was a good theory, but the team needed more convincing data to test it. "The snag was this was all cell lines, and cell lines are not much like real cells. What we wanted to do is to say what happens in the real tissue," said Davis. To do that, he and his colleagues performed live cell imaging in the pituitary glands of rats carrying a human prolactin gene locus. The experiment confirmed the stochastic expression pattern seen in vitro.

Subsequent in vivo and ex vivo imaging experiments revealed additional details. While gene expression in individual cells varies substantially, the pituitary gland seems to coordinate its overall activity across space and time, with cells at the margins of the gland all activating and repressing the gene simultaneously. Fetal pituitary tissue lacks this coordination, which seems to be implemented immediately after birth.

They carry it well: Fat distribution and obesity

Turning the focus to human studies, Jimmy Bell of Imperial College London presented his team's work on fat distribution and obesity. Like many large research projects, Bell's began with a few seemingly simple questions. "How much fat do my volunteers have, where do they have it, and why? And I thought 'I can answer the first two before lunch, and then I'll write a grant for the third one,'" quipped Bell.

Patients with identical body mass indices show significant differences in fat distribution. (Image courtesy of Jimmy Bell)

Things turned out to be considerably more complicated than Bell had anticipated. The first problem was that a cursory glance at epidemiological data revealed a fundamental paradox: body mass index (BMI) often, but not always, correlates with comorbidities such as insulin resistance and diabetes. Sumo wrestlers have BMIs as high as 50, qualifying them as morbidly obese by any definition, but they show no signs of insulin resistance, and patients who lack body fat completely because of a genetic lipodystrophy are profoundly unhealthy.

Using MRI to determine the fat distribution within each subject, Bell's team discovered that the location of fat is at least as important as its quantity. Patients with high levels of visceral fat had unhealthy metabolic profiles, while those with little visceral fat were healthier, regardless of their BMIs. Excess fat around the liver seems to be particularly dangerous.

Animal models suggested some of the factors driving these differences. Mice on high-fat diets showed some weight gain compared to those on standard chow, but adding fructose in addition to high fat altered the animals' metabolic profiles significantly, making them appear less healthy. Allowing the animals to exercise also makes a major difference, with those that spend more time running in a wheel showing fewer symptoms of metabolic trouble.

The researchers took their animal results back to the clinic, with an interventional study in healthy women who had no previous exercise habits. When the volunteers adopted a regular exercise routine, their fat distribution and metabolic profiles became considerably healthier, even though their weights and waist-to-hip ratios remained the same. Bell and his colleagues were pleased with the results, the volunteers somewhat less so. "They all became metabolically thin, and they all looked at me and they said 'But I still look the same,'" said Bell, indicating that researchers should be aware of different goals motivating participation in intervention studies.

A burdensome tox

About half of all chemicals tested in mice appear to be carcinogenic, undermining the credibility of the assay. (Image courtesy of Thomas Hartung)

While sophisticated imaging and genetic techniques are advancing rapidly in many fields, the toxicological tests that determine whether a drug or other new chemical is safe are still mired in the past. "Toxicology is the only field where we use protocols which are 40 to 80 years old and have essentially not changed," said Thomas Hartung of Johns Hopkins University.

Holding onto traditional tests would be appropriate if they worked well, but Hartung argued persuasively that they do not. To share an example, he pointed to results showing that a standard cancer bioassay in rats and mice is about 53% reliable in predicting actual human effects of a chemical. Worse, about 50% of all chemicals test positive as carcinogens, regardless of their source or identity, suggesting that the test may be no better than a coin toss. As an example, Hartung highlighted the components in coffee, some of which appear to be carcinogenic based on current standard tests. However, epidemiological analyses have found that moderate coffee consumption may reduce the risk of cancer in humans, demonstrating that the toxicological tests were at best misleading. With $10 trillion worth of products riding on the outcomes of these assays across numerous industries, Hartung says there must be a better way.

To find that better way, he and his colleagues are turning to modern "omic" efforts, which seek to determine the expression and metabolic activities of all of the genes and proteins in a cell or animal simultaneously. These studies should help toxicologists define the problem more precisely. "The basic idea is there will not be an endless number of pathways of toxicities, there's a certain number ... but there's not an endless number," said Hartung. Toxins can induce apoptotic or necrotic metabolic pathways in cells, for example, but the number of genes and proteins involved in those pathways is finite. To facilitate finding such pathways, Hartung and his colleagues have established a consortium called the Pathways of Toxicity Mapping Center, which is now working to map the possible pathways of toxin activity in order to build more informative testing protocols.

Mad mice: Mouse models of psychiatric disorders

Eric Fauman of Pfizer Worldwide Research spoke next, and took up the complicated problem of correlating animal phenotypes with human ones. Laboratory mice are a standard model for understanding human disease, but researchers are often simply guessing what phenotype they should be looking for. For example, if drug developers want to find new antipsychotic agents, it is not obvious what effect those drugs should have on mice. "What would be a way that you would assess whether a mouse is having delusions?" asked Fauman.

Rather than guess, he and his colleagues have pursued a statistical approach. Data from decades of mouse experiments and human epidemiological studies are now easily available, so the researchers tried to develop statistical algorithms to mine both data sets and find correlations between mouse and human phenotypes.

Eventually, the team settled on a method called reciprocal best matches. The algorithm assembles sets of phenotypes for each species, and refines the sets until each one is the most likely homolog of the other one based on the available data. For example, abnormal triglyceride levels in mice correlate strongly with a type 2 diabetes phenotype in humans; researchers working on diabetes drugs should therefore focus on compounds that lower the levels of triglycerides in their experimental animals. Schizophrenia, meanwhile, seems to match learning and memory in mice, suggesting that a simple behavioral test might detect the murine neuronal processes that correlate with human delusions.

Speakers:
Ravi Iyengar, Mount Sinai School of Medicine
Peter Kohl, Imperial College London, London, UK
Gary Mirams, University of Oxford, Oxford, UK
Kenneth L. Hastings, Sanofi-Aventis
Donald B Stedman, Pfizer Research & Development
Eric M. Walters, National Swine Research and Resource Center, University of Missouri–Columbia
Tony M. Plant, University of Pittsburgh
Alysson Renato Muotri, University of California, San Diego
Marc Lalande, University of Connecticut Stem Cell Institute
Sian Harding, Imperial College London, London, UK
Leonard D. Shultz, The Jackson Laboratory
Dale L. Greiner, University of Massachusetts Medical School
Alexander Ploss, The Rockefeller University
Simon Howell, Kings College London, London, UK

Highlights

• Careful study of massive data sets of protein interactions improves drug toxicity predictions significantly.
• Systems biologists have built sophisticated new models of heart function, which can be used to predict drug activity in the heart.
• Miniature pigs may model human biology better than rodents or nonhuman primates for many studies.
• Animal researchers need to adopt new strategies to develop the next generation of personalized medicines.
• Scientists should make a greater effort to publish results from failed preclinical and clinical studies.

Working in the data mine

The new technologies session focused part of its discussion on large data sets, which figure prominently in the work of Ravi Iyengar from Mount Sinai School of Medicine. By mapping known interactions between FDA-approved drugs and their targets, researchers have assembled a growing "interactome," showing the networks connecting different drug targets. While the results are extremely complex, some encouraging trends emerge. "Drugs interact with targets that don't talk to too many other things," said Iyengar, adding that this should simplify the problem of identifying potential side-effects of new compounds.

The network of possible drug interactions is huge, but actual drugs target specific "neighborhoods" within it. (Image courtesy of Ravi Iyengar)

Looking specifically at arrhythmia, a potentially life-threatening side effect that is notoriously difficult to detect in preclinical studies, Iyengar and his colleagues tried to build a computer model of the phenomenon. The model revealed a "neighborhood" of interacting proteins that are likely to be involved in arrhythmia.

Having constructed the molecular model, the team tested it with real clinical data. Over 100 FDA-approved drugs have been associated with a specific arrhythmia, called long QT syndrome, in humans. Looking at the targets of those drugs, Iyengar found that about 68% of the hit molecules map to the arrhythmia neighborhood. That suggests the researchers are on the right track, but Iyengar hopes to improve the model to identify even more potential side-effects.

Iyengar is also incorporating data from genome-wide association studies (GWAS), which have revealed long lists of genes that could be involved in drug side effects. The interactome map seems to improve the predictive power of these GWAS results considerably. "Putting the genes in context allows us to make predictions that are much stronger than just sets of individual genes by themselves," he said.

A heart of silicon: modeling in silico

Peter Kohl of Imperial College London is also interested in cardiac function, which he hopes to understand using a systems biology approach. After pointing out that the term "systems biology" lacks a consistent definition, Kohl explained that his strategy is to identify the individual parts of a biological system, and then to try to build integrated models of those parts to simulate and predict experimental outcomes.

Using the electrical potential of cardiomyocytes, researchers were able to produce 2- and 3-dimensional models of the heart, and were able to simulate the response of an electrocardiograph (ECG) to different stimuli. While impressive, that model is incomplete. Because of limitations in algorithms and computing power, the model fails to account for the fibroblast cells that actually outnumber the heart's myocytes. "If we are building models, we need to build them on real data, but even if we have the data we're often not able to integrate that fully," said Kohl.

A sophisticated computer model of the heart lets scientists map forces to the cellular level. (Image courtesy of Peter Kohl, his collaborator Peter Hunter, and the Auckland Bioengineering Institute, University of Auckland, New Zealand)

Even with these limitations, Kohl's team has been able to build progressively more sophisticated heart simulations and to make meaningful predictions from them. In one experiment the scientists identified a variety of previously unknown forces propagating through their simulated heart. Taking that result into the lab, they measured mechanical stresses on contracting cardiomyocytes and determined that mechanical force alone can induce a change in cellular calcium flux, leading to an electrical signal. That is exactly the reverse of the traditional electrically-activated movement that cardiac researchers normally think about. "That is something we didn't know two years ago and that we could not possibly put into models, so that also highlights that models cannot really explore reality of which we have no idea," said Kohl. Instead, he advocates recursively adding new results to the model to make subsequent predictions, which can then inform more laboratory experiments.

Gary Mirams of the University of Oxford addressed a similar problem, but he focused specifically on a type of arrhythmia called torsade de pointes (TdP). This rare but life-threatening ventricular arrhythmia has led to the withdrawal of numerous medicines from the market, prompting drug developers to nickname TdP "pharmageddon." Because the phenomenon is so rare, it is notoriously difficult to predict from pre-market drug trial data.

In a joint academic-industry collaboration, Mirams and other researchers are trying to build better predictive models for TdP, with the goal of weeding out bad drugs as early as possible in development. The project, called preDiCT, is extremely ambitious. "We're trying to do the impossible, [to] go from the earliest stage ion channel data, [using] computational models to predict all of these different levels of cardiac safety tests...through to the actual TdP risk," said Mirams.

To do that, Mirams and his colleagues have been testing the effects of different drugs using computerized heart models developed by teams at Oxford University. Combining those results with other data, the team can predict the TdP potential of known drugs much more accurately than previous screening tests did.

Petite pigs predict pathology

Kenneth Hastings of Sanofi-Aventis finished the session with a discussion of miniature pigs as animal models for toxicology. While full-size pigs provide a useful approximation of human physiology and size, Hastings argued that their smaller cousins can offer many of the same benefits at lower cost.

There are about 40 breeds of minipigs, but researchers usually use one of a few strains weighing between 13 and 20 kilograms. They offer human-like physiology with about the same maintenance costs as dogs. Nonetheless, researchers have been relatively slow to swap out other animal models for minipigs. "The problem is that it's just a less well-established laboratory species," said Hastings.

Pressure to reduce, replace, and refine the use of animal models is starting to change that. By using a more human-like species earlier in preclinical studies, toxicologists can cut down on animal use while simultaneously getting more predictive results. Indeed, for some experiments pigs could even be better models than non-human primates. Pig skin, for example, is very similar to human skin. "For dermal studies, the minipig...might be a valuable model," said Hastings.

To highlight the model's potential, Hastings pointed to Dovonex (calcipotriene), a topical treatment for psoriasis. In preclinical studies, researchers found that the compound caused severe gut irritation in dogs even at very low doses. In pigs, however, the drug was quite safe, and clinical trials found similar safety in humans.

In an interactive panel discussion after the talks, audience members talked about the limitations of systems biology approaches, regulatory authorities' willingness or reluctance to accept non-animal models in place of animals, and the need to balance the need for robust toxicological tests with the drive to get useful chemicals and drugs to market quickly.

Wrapping up

Following another networking lunch and poster session, the meeting split into three parallel tracks. In the workshop on alternatives to rodent disease models, attendees heard about the advantages and challenges of working with zebrafish and nonhuman primates, and also learned more about pigs. The session on stem cells as disease models featured data about the ways these cells are informing several fields, including neural development, cardiac repair, and genetic diseases such as Prader–Willi syndrome. Finally, a workshop on humanized animal models covered the use of "humanized" mice—genetically engineered mice carrying specific human genes—which could provide more predictive results for preclinical studies.

The conference closed with a final joint session, which was structured as an extended, audience-driven panel discussion. Attendees generally agreed that animal research needs a new approach, particularly in drug development and toxicology. "Personalized medicine is going to be the flavor of the next 10 to 20 years, individual patients with individual therapies, and here we are in the animal model side using inbred strains of mice of doubtful provenance and doubtful value in predicting efficacy," said Simon Howell of Kings College London.

Besides developing better models, researchers also need to be more forthcoming when their experiments don't work. After acknowledging the tendency to bury negative results, several speakers brainstormed ways to get data from failed animal and clinical studies into the public domain, to prevent others from repeating the same mistakes. "There's an obligation to reveal this hidden information, an obligation to the patients who participated in the trials. It's a matter of public trust," said Garret FitzGerald.

Though many of the problems with current animal models may defy quick solutions, attendees expressed general optimism about the field's current direction. In particular, the ability to understand more about a chemical's mechanism of action in cultured cells could eventually shorten the drug and toxicology testing process without compromising safety. As Sian Harding of Imperial College London explained, the field appears to be moving toward using new mechanistic insights and computer models to move directly from cell culture to small-scale human studies, with an ultimate aim of eliminating the troublesome animal models entirely.