Animal Models of Diabetic Retinopathy

Abstract

Diabetic retinopathy (DR) is a major cause of visual impairment worldwide. The precise pathogenesis of this diabetic complication remains ill-defined and this is reflected in the limited options for preventing development and progression of this disease. The value of animal models to understand and treat human disease is well recognised and this chapter focuses on the range of in vivo model systems that are available for studying DR. These models have been developed over many decades and utilised to aid our understanding of what causes DR, about how microvascular and neural lesions develop and to provide evidence for key cellular and molecular mechanisms that drive this pathology. A wide range of animal models of DR are currently available, each with advantages and disadvantages that need to be understood and evaluated for their scientific and clinical value. As transgenic and imaging technology improves, more models will be developed and they will continue to play a critical role in the development of new therapeutic approaches to DR by providing robust, preclinical evidence prior to clinical trial.

Commentary

Noemi Lois

e-mail: n.lois@qub.ac.uk

Centre for Experimental Medicine,

Queen’s University,

Belfast, UK

Despite of early detection, by means of diabetic retinopathy (DR) screening programmes, improved metabolic control and new therapies, DR remains a leading cause of visual impairment and blindness worldwide. In DR sight loss occurs as a result of the development of diabetic macular edema (DME), ischaemic maculopathy and/ or proliferative diabetic retinopathy (PDR). Imperfect treatments are available for these complications once established; there is no therapeutic intervention to prevent their development. Thus, the search to improve the care of people with DR must continue.

Animal models of disease are extremely valuable, if mimicking adequately the disease in humans, to investigate pathogenic pathways involved and, subsequently, design new therapeutic strategies. It is essential for researchers and clinicians to understand well the nature of the disease and the mechanisms involved in its development in these animal models to interpret correctly any results generated and their relevance to human disease. In this chapter, Drs Chen and Stitt (1) provide us with a thorough review of the experimental animal models of DR available, pointing out their advantages and shortcomings.

It is apparent that animal models reproduce early features of DR. But there are additional requirements to take into consideration. For instance, the two histopathological lesions essentially unique to DR which, in addition, are the earliest observed, namely the diffuse thickening of the capillary basement membranes (somewhat different to that occurring in ageing) and the selective loss of pericytes, do occur in animal models and are often seen after 6-9 months duration of diabetes. The pathogenic routes involved in the occurrence of these early features can be, hence, investigated; therapeutic measures can be also sought. However, these features cannot be recognised by the in vivo imaging technologies available. To determine change in these parameters result of a potential intervention, terminal studies and, subsequently, the use of high number of animals would be required. Furthermore and even more importantly, translating findings into humans in potential phase 1 interventional trials would be challenging, as the endpoints evaluated in the pre-clinical studies (i.e. the thickening of the basement membrane and pericyte loss) would not be detectable in humans by current means. Thus, although the animal model may mimic closely the change observed in humans, the end points evaluated are not translatable to clinical studies.

Abnormalities occurring as a result of the above and other histopathological changes in early DR, for example increased vascular permeability and areas of non-perfusion, may be suitably imaged. However, at present time, quantitative evaluation on these endpoints remains challenging. Moreover, current imaging techniques to evaluate them are invasive, requiring an administration of a dye intravenously. Although fluorescein angiography is a test routinely used in Ophthalmology clinics, its invasive nature would still limit its use in human investigations.

Functional studies to determine early functional abnormalities in animal models of DR have been scarce. These may be more difficult to undertake but have the potential value of being objective and more likely to be translatable to human studies. Furthermore, they may be more significant to people with DR as it is the functional loss, rather than the anatomical change, what matters to those affected by the disease and especially considering that structural changes do not relate always to reduce or loss of function.

Considerations may be given by basic scientists to the above matters when designing pre-clinical studies using animal models of DR. Endpoints that can be obtained by using reproducible, user-friendly, non-invasive technologies should be favoured.

Lastly, and also as discussed by Chen and Stitt in their comprehensive review, there is no adequate animal model of PDR. PDR may develop and progress in humans very quickly. Large oscillations in glucose levels, a rapid and tight glucose control after a period of chronically high levels of glycemia, kidney failure, etc, may precipitate its fast development and progression. It is possible that, in order to achieve development of PDR in experimental animal models these events may need to be mimicked; combination of insults (for example, high glucose levels and blood pressure within the context of a genetically predisposed strain) may be also required.

It is clear that work must continue. Basic scientists and clinicians should work together in the search for improved in vivo models and endpoints for research into DR with the final goal of improving the quality of life of people with DR.