Michael H. Court, BVSc, PhD, DACVA
Pharmacogenomics in Human Medicine
Pharmacogenomics is the study of how genetic variation influences drug disposition and effect. The anticipated benefits of pharmacogenomics for human health are substantial, with the most commonly touted practical outcome being the "personalization" of drug therapy.1 Specifically, pharmacogenomics has the potential to provide a means of identifying individuals that are more likely to respond to a particular drug (efficacy pharmacogenomics), and also those individuals that are more (or less) likely to have an adverse effect from a specific drug (safety pharmacogenomics). The greatest benefit would be derived for drugs with a narrow therapeutic index (i.e., the margin between toxic and effective dose), providing a rational means of selecting a safe and effective starting dose. While many predictive pharmacogenomic biomarkers have been identified, the adoption of pharmacogenomic testing into routine practice in human medicine has been relatively slow, in part because of a lack of studies that convincingly demonstrate the benefits of adding such testing to a therapeutic regimen. Several proof-of-principle studies are currently underway with the aim of demonstrating the benefits of using CYP2C9 (cytochrome P450 2C9) and VKORC (vitamin K oxidoreductase complex) gene variants to predict a safe and effect starting dose of the anticoagulant warfarin for the prevention of thrombosis and stroke.
Pharmacogenomics in Veterinary Medicine
Compared with human pharmacogenomics, veterinary pharmacogenomics is in its infancy, although the recent availability of complete genome sequences for a number of species of veterinary importance will likely contribute to growth of this discipline.2 To date, the only pharmacogenomic biomarker of proven clinical importance is the MDR1-1Δ polymorphism in dogs. This polymorphism is a 4-base pair deletion (Δ=delta) in the canine MDR1 gene resulting in protein truncation with complete inactivation of the encoded drug transporter, P-glycoprotein (or P-gp). This membrane transporter is expressed in gut, liver, kidney, and blood-brain barrier and is critical to limiting systemic and brain exposure to a large variety of drugs of clinical importance. Although this polymorphism was first described in collie dogs that showed hypersensitivity to the neurotoxic effects of ivermectin, not all collies are affected (in one study at least 22% of collies do not have the MDR1-1Δ mutation), and at least 9 other dog breeds are affected with allele frequencies ranging from 4 to 42%. Importantly, a number of drugs other than ivermectin have shown enhanced adverse side effects in homozygous deficient dogs, including loperamide3, digoxin and mexiletine4, and a range of cancer chemotherapeutic agents.5 Although it has been speculated that many other drugs are likely to be substrates for P-gp in dogs with the potential for adverse effects in P-gp deficient dogs, this contention was based on data from other species and has yet to be substantiated. The presence of MDR1-1Δ can be determined with a relatively simple genetic test (PCR) available through the University of Washington. Given the therapeutic advantages of knowing the P-gp status of a patient, it is likely that assay for this polymorphism will be adopted by veterinary clinicians and informed pet owners particularly for "at risk" breeds of dog.
Polymorphisms have also been identified in the canine TPMT (thiopurine methyltransferase) gene that could potentially impact myelotoxicity associated with use of thiopurines such as azathioprine.6 However, the validity of these potential biomarkers for prediction of such toxicity in dogs has yet to be established.7 Genetic variants in a number of canine CYP genes have also been discovered that appear to influence drug clearance, but again the effect on drug efficacy and toxicity have not been evaluated.8-10 In horses, adverse neurological reactions to procaine penicillin antibiotics have been associated with low plasma activities of esterase, the enzyme that metabolizes procaine.11 Although as yet unproven, this interesting observation could be explained by a genetic polymorphism in the equine esterase gene.
Species Differences in Drug Effects
In addition to the differences in drug response that exist between individual animals of a particular species, veterinarians must routinely contend with sometimes considerable differences in drug disposition and response that occur between different species of animal. A good example of this is the exaggerated responses to drugs with a simple phenol structure, such as acetylsalicylic acid and acetaminophen, which is observed in cats, but not in dogs or most other species. This species difference results from inactivation (in all cats) of the gene encoding the major phenol detoxification enzyme, UDP-glucuronosyltransferase 1A6 (UGT1A6).12,13 Consequently, phenolic drugs are cleared by glucuronidation very slowly in cats compared with other species. A second example is the complete lack of genes encoding N-acetyltransferases (NAT) in dogs and the presence of only one functional NAT gene in cats; contrasting with most other species which have at least 2 NAT genes encoding different NAT isoforms.14,15
Although there is much empirical information describing species differences for existing drugs, the molecular mechanisms and evolutionary basis for these differences are poorly understood such that it is not yet possible (for example) to predict the response of a particular species to a drug based on known responses in other species. This has importance in terms of the development of drugs for use in humans and domestic animals, and also for use of drugs in zoo and wildlife practice.
One approach to enable a better understanding of species differences in drug response is through comparative pharmacogenomics. In its simplest form, this involves DNA sequencing and alignment of genes known to be involved in drug response (called "pharmacogenes") from different species, including those with known drug effect, and those species in which the effect is unknown. Until recently there were relatively few species (human and mouse) in which the entire genome sequence was known, requiring generation of sequence data by individual laboratories. However, this process has been greatly facilitated in the last few years through the availability of complete genome sequence information for several species of veterinary importance, including dog, sheep, cow, and chicken, and horse (cat is scheduled for release soon). In addition, the Broad Institute at MIT (involved in sequencing the dog, cat and horse genomes) has embarked on an ambitious mammalian comparative genomics research program with plans to sequence the genomes of 24 species, representing a diverse array of mammals, including elephant, rabbit, llama, and dolphin, among others (see http://www.broad.mit.edu/mammals/).
One of the primary purposes of comparative pharmacogenomics is to identify regions of DNA within a gene of interest that have been highly conserved throughout evolutionary time and so may be critical to gene function. Conserved DNA sequence is commonly located within the protein coding region and often define protein functional domains, but also may be within noncoding regions important for gene regulation. Comparative pharmacogenomics is also a common technique to identify gene regions that are likely to contain genetic polymorphisms of functional importance. Finally, phylogenetic computational methods can be used to study the molecular evolution of pharmacogenes.
We have been using a comparative pharmacogenomic approach in our laboratory to understand the genesis of deficient drug glucuronidation in cats. Through collaborations with several zoos and universities, and with Dr Stephen O'Brien's Laboratory of Genomic Diversity at NIH, we have obtained DNA samples representing all families within the mammalian order Carnivora, including most of the species within the Felidae (cat) family. Comparative sequence analysis of the UGT1A6 gene indicated that the UGT1A6 defects first appeared approximately 11 million years ago, such that all extant Felidae species have a defective UGT1A6 gene (like the domestic cat). In contrast, out of nearly 50 other species sampled, only Brown Hyena and Northern Elephant seal showed UGT1A6 defects. Interestingly, Brown Hyena showed only a single mutation that was identical to one shared by all Felidae, raising the possibility that this may be a founding mutation, existing prior to Hyenidae-Felidae divergence and inherited in Brown Hyena as a polymorphism. In contrast, Northern Elephant seal showed multiple unique UGT1A6 defects consistent with an independent origin.
Diet may be an important factor influencing evolution of the drug metabolizing enzymes. Specifically, we hypothesized that inactivation of the UGT1A6 gene should only occur in species with diets containing minimal amounts of plant-derived phenols. An exhaustive literature review of the natural diet of sampled species determined that all species with defective UGT1A6 were hypercarnivores (diet contains less than 30% plant matter). However, some hypercarnivore species showed no UGT1A6 defects indicating that factors in addition to diet, such as genetic bottlenecks (reported for Brown Hyena, Northern Elephant seal, and Felidae populations), may be required for UGT1A6 inactivation. These findings provide the first evidence supporting a role for diet in the evolution of the drug metabolizing enzymes, and have substantial implications for the management and restoration of endangered hypercarnivore mammalian species.
1. Court MH. J Clin Pharmacol 2007; 47(9):1087-103.
2. Mealey KL. Vet Clin North Am Small Anim Pract 2006; 36(5):961-73, v.
3. Sartor LL, et al. J Vet Intern Med 2004; 18(1):117-8.
4. Henik RA, et al. J Vet Intern Med 2006; 20(2):415-7.
5. Mealey KL, et al. J Am Vet Med Assoc 2003; 223(10):1453-5, 1434.
6. Salavaggione OE, et al. Pharmacogenetics 2002; 12(9):713-24.
7. Rodriguez DB, et al. J Vet Intern Med 2004; 18(3):339-45.
8. Blaisdell J, et al. Drug Metab Dispos 1998; 26(3):278-83.
9. Mise M, et al. Drug Metab Dispos 2004; 32(2):240-5.
10. Paulson SK, et al. Drug Metab Dispos 1999; 27(10):1133-42.
11. Olsen L, et al. J Vet Pharmacol Ther 2007; 30(3):201-7.
12. Court MH, Greenblatt DJ. Biochem Pharmacol 1997; 53(7):1041-7.
13. Court MH, Greenblatt DJ. Pharmacogenetics 2000; 10(4):355-69.
14. Trepanier LA, et al. Biochem Pharmacol 1997; 54(1):73-80.
15. Trepanier LA, et al. Pharmacogenetics 1998; 8(2):169-79.