Translational Animal Models in Drug Discovery and Development

Atherosclerosis is a potentially fatal disease of the arteries affecting everyone, yet there are relatively few animal models that enable research into the events leading up to the rupture of an atherosclerotic plaque (the underlying cause of the majority of fatal thrombotic events). The apolipoprotein E knockout (ApoE^-/-) mouse has been used for over a decade now, because when fed a high-fat diet it develops lesions in the brachiocephalic artery that spontaneously rupture at a known time point. Critics argue that the ApoE^-/- mouse does not exactly replicate human atherosclerotic plaque rupture, yet this model gives us valuable insight into the mechanisms and processes leading up to this clinically significant event. In this article, atherosclerosis shall be discussed, followed by some examples of animal models of atherosclerosis and plaque rupture used before the development of the ApoE^-/- mouse model. Differences between mice and humans, and also the reasons that the ApoE^-/- mouse model is of great benefit to the field of atherosclerotic plaque rupture are discussed, followed by recent translational applications of the model.

Heart failure (HF) is a major cause of morbidity and mortality. While the current therapies for inhibition of the rennin-angiotensin-aldosterone and sympathetic adrenergic systems have provided clear benefits to HF patients, the disease still progresses in most patients and development of novel therapeutics is still needed. Moreover, concerns about cardiac safety in drug development have taken center stage following recent reports of adverse cardiovascular effects with the use of PPARγ- modulators for the treatment of type-2 diabetes. Issues related to drug efficacy and safety call for appropriate preclinical models of HF that mirror the complex pathophysiological conditions in patients. In particular, cardiovascular drug safety is conventionally assessed pre-clinically in young healthy animals, whereas patients subjected to treatment with drugs such as PPARγ-modulators often have multiple risk factors for cardiovascular disease. While a number of animal models have been developed to explore the pathophysiology of HF and to develop novel therapies, only a few reports address use of these models to assess cardiac drug safety. This chapter focuses on the use of experimental models of HF to assess drug-induced cardiovascular liability by means of biochemical, pharmacogenomic, physiological and imaging biomarkers. Rosiglitazone, a PPARγ activator that has cardiovascular liability in humans, is used as an example for translational investigation in HF models.

Stroke is a leading cause of death and disability worldwide. Effective intervention for acute stroke remains to be a significant unmet medical need. In the past several decades, a large number of therapeutic-pharmacological agents have been evaluated in preclinical models with many advancing to late stages of clinical study. However, all potential interventions (with the exception of the approval of early thrombolysis using recombinant tissue plasminogen activator; tPA) have failed. To address this translational gap, a series of meetings by the Stroke Therapy Academic Industry Roundtable (STAIR) has conducted and has made recommendations over the past decade in order to provide guidance to the stroke research community in the development of novel stroke therapeutic interventions. In spite of these efforts, large clinical studies have continued to fail. This chapter describes the most commonly used preclinical stroke models that have been applied to novel drug discovery, including critiques and translational perspectives of their advantages and limitations. Furthermore, novel biomarkers are used in stroke intervention research, including the monitoring of salvageable brain tissue as a therapeutic index of potential protection. For example, the measurement of the ischemic penumbra is discussed as an important means to more accurately assess and translate the efficacy of therapeutic agents from animal models to man.

In vertebrates, formation of a clot at a site of blood vessel injury requires a group of plasma proteins (coagulation factors) that regulate generation of the protease α- thrombin. α-Thrombin mediates a number of key activities during clot formation, including conversion of soluble plasma fibrinogen into an insoluble fibrin mesh, activation of platelets, and stimulation of vascular endothelial cells. While α-thrombin is required for life, dysregulated generation of this protease can contribute to life-threatening thrombotic disorders and consumptive coagulopathies. The coagulation factors were originally identified and characterized using plasma clotting assays. While deficiency of any single coagulation factor results in an abnormal result in one or more of these in vitro assays, the severity of the associated bleeding disorder can range from fatal hemorrhage to complete absence of symptoms. Genes for coagulation factors have been manipulated by a variety of “knockout”, “knock-in”, and transgenic strategies in mice. In this chapter, we review insights gained into hemostasis, thrombosis and wound healing through studies involving mouse models of coagulation deficiencies. The chapter is divided into sections focusing on deficiencies associated with lethal phenotypes (prothrombin, tissue factor, tissue factor pathway inhibitor, factor V, factor VII, factor X and γ-glutamyl carboxylase), deficiencies associated with severe bleeding disorders that do not compromise viability (fibrinogen, factor VIII, factor IX and factor XIII), and those not associated with abnormal hemostasis (factor XI, factor XII and high molecular weight kininogen) and the translational value of each model in drug discovery and development is discussed.

Experimental and translational breast cancer research is currently hampered by the limited number of in vivo models available that accurately represent the full spectrum of human breast disease. At least six unique classes of human breast cancers have been identified based on gene expression analyses. While immortalized human breast cancer cell lines as well as mouse in vivo models have been developed to represent many of these classes, in vivo xenograft models are rapidly becoming a viable preclinical experimental platform for both testing of experimental therapeutics and the mechanistic studies of tumor cell regulators. These models include viral transductioninduced, human cell line xenografts, and more recently, primary tumor xenografts established directly from patients. Each model type is endowed with its own set of advantages and limitations. However, as a collection, xenograft models of human breast cancer represent powerful tools for preclinical analysis.

Tumor-transplanted rodents are primarily adopted to study cancer progression in vivo and the inhibitory potential of the immune system. Genetically engineered, cancer-prone mice, however, more closely mimic several features of human cancer. Inbred BALB/c female mice transgenic for the rat neu (Her-2/neu, ErbB-2) oncogene (BALB-neuT mice) constitute an intensively studied mammary cancer model. Key features of this model include that (1) each of their ten mammary glands develops an independent carcinoma that slowly progresses from microscopic lesions to invasive tumors; (2) multiple in situ carcinomas are accompanied by greater dissemination of neoplastic cells into the bone marrow; (3) lung metastases are evident in the later stages; (4) over-expression of the protein product of the transgenic neu oncogene in the newborn thymus induces the deletion of T cell clones reacting with high-affinity against it, while the step-wise progression of mammary lesions triggers negative regulation mechanisms that suppress antibody- and perforin-mediated immune surveillance mechanisms. Thus, boosts of innate immunity delay cancer progression and vaccines administered when only early microscopic lesions are present provide lifelong protection, whereas their efficacy tails off when they are administered to mice with more advanced lesions. Because this model mimics some of the most critical features of human disease, it has been successfully used to investigate a number of therapeutic agents, including the role of adaptor proteins (P130 cap), signal transducers (STAT3), phosphoinositide 3-kinase (PI3K), and oncogenic stress sensing kinases (MKK7) in neu-driven carcinogenesis, and assessment of the efficacy of braki therapy in the control of mammary cancer.

Animal models of asthma play a central role in mechanistic studies into the pathophysiology of asthma. However, asthma is a uniquely human disease that remains poorly understood, and which has so far resisted attempts to reproduce all its complexities faithfully in an animal. Nevertheless, the key asthma phenotypic features of airway hyperresponsiveness and inflammation have been recapitulated in a number of animal species, most commonly through immune sensitization and challenge with a foreign protein in order to produce an allergic inflammatory reaction in the lungs. The mouse has become the most commonly used species to model asthma in this fashion because of its obvious advantages related to cost, gestation period, and the wide range of biological manipulations it can be subjected to. Until recently, the small size of the mouse made assessing its lung physiology a challenge, but modern imaging methodologies coupled with the forced oscillation technique for measuring lung mechanical function now make it possible to phenotype mice in detail. This has significantly advanced our understanding of the immunological, genetic and physiological mechanisms of asthma pathogenesis, particularly as they relate to airways hyperresponsiveness.

Animal models of chronic obstructive lung disease (COPD) are complicated by the fact that COPD encompasses 4 different anatomic entities: emphysema, small airway remodeling, pulmonary hypertension, increased mucin secretory alterations (the pathology found in the clinical process, chronic bronchitis) and one clinical entity without distinct pathological association (acute exacerbations). Models using chronic cigarette smoke exposure can reproduce mild forms of some, but not all, of these lesions; however, cigarette smoke exposure never causes the disabling/lethal GOLD stage III or IV lesions associated with human COPD. Further complications of chronic smoke models relate to species differences, and, for mice, strain differences, as well as developmental abnormalities caused by genetic manipulation, abnormalities that can mimic changes of COPD. Numerous drug/genetic modification models using chronic smoke exposure appear to protect partially or completely against the changes of COPD, but trials of these same agents in humans have been disappointing, perhaps because animals are typically treated from day 1 of smoke exposure whereas treatment in humans is always a late intervention; this conclusion suggests that animal models should use late interventions and humans should be treated much earlier in the course of their disease.

Inflammatory bowel disease (IBD), which includes Crohn’s disease and ulcerative colitis, is a chronic intestinal disorder that affects nearly 1.4 million Americans. Although the precise cause of IBD remains unknown, genetic susceptibility, the intestinal microbiota and immune dysfunction are all thought to contribute to disease pathogenesis. Standard therapies are effective in maintaining remission in a subset of IBD patients; however, many patients are non-responsive to the available treatment options. Several of the novel therapeutic targets currently in development or clinical trials were identified using mouse models of IBD. The difficulty in determining which preclinical targets will be successful in patients lies in the inherent complexities underlying the pathogenesis of IBD. Furthermore, the secondary immunological effects that preclude widespread clinical use of many of these drugs, such as risk of infection or serious adverse side effects, are difficult to recapitulate in mice. This considered, mouse models of IBD still provide the best tools to assess the individual contribution of different genes, cell types and bacteria to disease pathogenesis. This chapter explores many of the available mouse models of IBD and the contribution of these models to the identification of current and future targets for therapeutic development.

Models of rheumatoid arthritis (RA) in laboratory animals are important tools for research into pathogenic mechanisms and the development of effective and safe therapies. Rodent models (rats and mice) are the most widely used and have provided important information about the pathogenic mechanisms operating in the disease. However, the evolutionary distance between rodents and humans hampers the translation of scientific principles into effective therapies. This has in part resulted in a high attrition rate of new drugs to get approved for patient use and a concomitant dramatic increase in the cost for the development of these new drugs. The impact of the genetic distance between the species is especially seen for treatments based on human specific biological molecules, which are usually species-specific and are now successfully used in the treatment of RA. Non-human primates may help to bridge the evolutionary gap between rodent models and the patient because of their phylogenetic proximity. Here, we review two non-human primate models of inflammatory arthritis, specifically the rhesus monkey (Macaca mulatta) and the common marmoset (Callithrix jacchus).