Proceedings of The Physiological Society

Experimental Models (Exeter, UK) (2018) Proc Physiol Soc 40, PC09

Poster Communications

Of Mice and Men to "Animal Farm": Selecting animal models based on homologous protein primary sequences?

H. A. Widmer1, S. F. Cruickshank1, I. C. Rowe1, A. Strath1

1. School of Pharmacy & Life Science, Robert Gordon University, Aberdeen, Aberdeenshire, United Kingdom.

The choice of rodents as primary models for the investigation and better understanding of human diseases has been challenged (1-2). Poor results in clinical trials of promising lead compounds have prompted calls to refocus at least some of translational research onto larger animals with closer molecular matches and disease-relevant physiology to human (3). 27-28 million cattle, sheep and pigs were slaughtered in the UK in 2014 (4) while large farm animals represented 1,3% of the 4 million total animals used in scientific procedures over the same time period - a fraction of the numbers slaughtered for food (5). Rodents represented 83% of all animals used in regulated experimental procedures. There is an opportunity to make better use of our resources and reduce our reliance on rodent models through the choice of large animal tissue where evidence exists to support its use. The molecular match of potential large animal target proteins to human homologs could form part of that evidence-based decision making process in choosing appropriate animal models. We have undertaken a preliminary study and present data obtained by comparing the primary sequence of ten distinct proteins in rodents (rat & mouse) and in large farm animals (pigs, cattle & sheep) with their homologous human sequences. The ten proteins were: kv2.1; kv1.5; BKCa; TREK-1; Kir6.2; vasopressin receptor V1a; TGFb1; CTLA-4; IL23; and CD55; and have roles ranging from regulating membrane potential to monitoring cellular metabolism or modulating inflammatory/immune responses. We used three separate approaches: protein primary sequence alignment; phylogenetic analysis; and similar protein searches (UniProt UniRef) to establish species ranking with respect to their match to human. All three approaches yielded similar patterns with farm animal homologs closer-sometimes dramatically so- to the human homolog than either of the rodent forms in 7 out of 9 proteins. CD55's poor homologous sequence alignment resulted in its exclusion from further study. The ranking of the two remaining proteins, kir6.2 and avpr1a, appeared to be approach-dependant- with both sequence alignment and phylogenetic analysis finding the rodent kir6.2 to be closer to the human than any of the farm animal homologs while phylogenetic analysis found only the rodent vasopressin receptor to be closer to human. As for the similar protein search method, a rodent homolog was never found to be closer to human than any of the farm animal homologs. We propose that bioinformatics approaches, as described here, can help inform decision making in the selection of appropriate model species by giving consideration to the molecular identity of the target proteins. Better use of abattoir tissues may be an ethical and effective route to improvements in translational and pharmaceutical research.

Where applicable, experiments conform with Society ethical requirements