Introduction
The ABCB1 (MDR1) gene encodes the drug transporter P-glycoprotein.1 As a key component of the blood-brain barrier, P-glycoprotein prevents substrate drugs from entering the CNS; it also facilitates biliary excretion of substrate drugs.2,3 A 4 bp deletion mutation of the ABCB1 gene in dogs, ABCB1-1∆, results in a frameshift that generates premature stop codons, thereby disrupting protein synthesis.4 The first stop codon encountered during translation actually occurs before much of the protein can be synthesized. Truncation at this stop codon in exon 4 near the amino terminus prevents synthesis of P-glycoprotein's 4 key functional domains: 2 substrate-binding sites and 2 ATP-binding domains. Subsequently, a P-glycoprotein null phenotype is created in animals homozygous for the mutation. Animals lacking P-glycoprotein function are not able to pump substrate drugs like macrocyclic lactones and loperamide out of the brain and therefore are highly susceptible to neurologic toxicosis associated with these drugs.
Like the ABCB1-1∆ mutated gene in dogs, the MDR1 gene mutation in cats, ABCB11930_1931del TC, also generates premature stop codons.5 The first encountered stop codon occurs in exon 15, allowing synthesis of only 1 ATP-binding domain and 1 substrate-binding domain. Because all 4 functional domains are considered essential for P-glycoprotein function, cats homozygous for ABCB11930_1931del TC are expected to display the multidrug sensitivity phenotype including enhanced susceptibility to the neurologic effects of macrocyclic lactones, although only 1 case5 has been reported to date.
The Veterinary Clinical Pharmacology Laboratory at Washington State University recently received a temporal cluster of requests to genotype cats for ABCB11930_1931del TC. Samples from 5 cats that had experienced adverse neurologic effects following exposure to macrocyclic lactones were submitted over the course of 5 months. On the basis of this cluster of requests and subsequent results, we questioned the reported5 frequency of cats homozygous for this mutation (4.6%). Therefore, the purposes of the study reported here were to determine whether ABCB11930_1931del TC predisposed these 5 cats to macrocyclic-lactone toxicosis and to determine the frequency of ABCB11930_1931del TC in DNA samples banked from 1,006 cats.
Materials and Methods
Animals
Retrospective case series
This study was approved by Washington State University's Institutional Animal Care and Use Committee. Samples were submitted for ABCB11930_1931del TC genotyping from 5 cats that were reported to have had neurologic signs after exposure to a macrocyclic lactone. The submitting entity was contacted to verify the cat's age, sex, breed, and body weight; the drug administered, dose, and route of administration; and whether the owner or veterinarian had administered the drug, what adverse clinical signs were observed and their timing in relation to macrocyclic-lactone exposure, whether other drugs had been administered, and whether exposure to additional potential toxicants was possible.
These exposures occurred independently and at distinct geographic locations (4 in the United States and 1 in New Zealand), with no cats from the same household. Each cat was examined and treated by a different veterinarian. Four of the exposures were intentional, involving topically applied eprinomectin-containing productsa,b that were approved for use in cats in the United States and New Zealand. Both products were administered in accordance with the label. The fifth cat consumed up to 6 ivermectin-containing heartworm preventive chewsc (three 136-µg chews and three 272-µg chews), which possibly resulted in a maximum ivermectin dose of 532 µg of ivermectin/ kg (1,170 µg/lb).
Prospective study
The DNA samples from 1,006 cats banked at the DNA bank at the College of Veterinary Medicine, Washington State University were analyzed for the ABCB11930_1931del TC mutation.
DNA extraction
For 2 of the 5 cats that experienced neurologic adverse effects and for all banked samples, samples were collected as buccal swabs and DNA was extracted by alkaline lysis as previously described.6
For 2 of the other cats that experienced neurologic adverse effects, postmortem samples of fresh frozen brain or liver were submitted for genotyping; genomic DNA was extracted from the tissue samples by use of a proprietary kit.d For the cat from New Zealand, DNA was extracted there and then submitted to the Veterinary Clinical Pharmacology Laboratory for analysis.
ABCB11930_1931del TC genotyping
Data from a previous sequencing study5 were used to develop a PCR-fragment analysis assay that exploits the 2-bp difference of the ABCB11930_1931del TC allele of exon 15. The 6-FAM–labeled PCR products of 152 and 154 bp (5′6-FAM–reverse primer) were separated on an analyzer,e and the data were analyzed with proprietary software.f Genotype controls were 690-bp PCR products of feline ABCB1 exon 15 and flanking DNA amplified from previously sequenced DNA that represented the 3 potential genotypes as follows: wild-type and heterozygous and homozygous for ABCB11930_1931del TC. To facilitate screening the larger set of DNA-banked samples, a probe-based real-time PCR-genotyping assay was developed by use of independent, nonoverlapping primers from the fragment analysis assay. The three 690-bp PCR assay–generated genotype controls representing the wild-type or the heterozygous or homozygous ABCB11930_1931del TC mutation were used in every analyzed batch for quality control.
Results
Retrospective case series
Cat 1, a 2-year-old 4.1-kg (9.0-lb) neutered male domestic shorthair, had been presented to an emergency hospital for progressive neurologic signs approximately 24 hours after the owner had applied an eprinomectin-containing topical producta according to the label instructions. Other cats in the home had also received the product but remained clinically normal. All cats had been separated for 2 hours after the product had been applied to prevent accidental oral ingestion. No other macrocyclic lactones had been administered. The clinical signs described by the owner included drooling, vomiting, tremors, shaking, and seizure activity. Medical records from the emergency hospital indicated that the cat had had mydriasis, seizures, and tremors and had been open-mouth breathing. The cat had been treated with diazepam, IV fluid therapy, and methocarbamol. The cat's neurologic status had improved over several days, and it had been discharged from the hospital. During a phone call with the owner 3 weeks after the exposure, the owner had indicated that the cat had fully recovered. This had been the second time that the owner had applied the product on the cat. The owner had not noticed any adverse events after the first application approximately 1 month previously. Geno-typing for ABCB11930_1931del TC indicated that the cat was homozygous for the mutation.
Cat 2, an 8-month-old 3.2-kg (7.0-lb) sexually intact female domestic shorthair, had its first topical dose of an eprinomectin-containing producta per the label instructions. The owner had applied the product topically between the cat's shoulder blades but suspected that the cat had ingested some of the product. The following day, the owner had described the cat as being jittery. Three days later, the cat had been presented to a veterinarian who had described the cat as being ataxic, having mydriasis, and experiencing dyspnea and intention tremors. The cat had been treated with methocarbamol and was monitored at home. Results of a CBC and serum biochemical analysis had been within reference limits. During a phone call with the owner several weeks after the exposure, the owner had indicated that the cat had fully recovered. Geno-typing for ABCB11930_1931del TC indicated that the cat was homozygous for the mutation.
Cat 3, a 16-month-old 7.3-kg (16.1-lb) neutered male domestic shorthair, had the label dose of an eprinomectin-containing topical producta applied by a veterinarian between the cat's shoulder blades during a wellness appointment. The following day, the owner had returned to the veterinary hospital because of the cat's abnormal behavior and dilated pupils. The owner had described the cat as jumpy and reluctant to being touched. The veterinarian had described the patient as looking terrified. The cat had been hospitalized, sedated with diazepam, and administered dexamethasone sodium phosphate and maropitant. The following day, the cat had begun to twitch and seemed to be uncomfortable, so methocarbamol, gabapentin, and butorphanol had been administered. The following day, the cat's signs had worsened. The veterinarian had conveyed that the cat had developed blindness and become fractious, so it had then been euthanized. On the basis of recommendations from a consulting veterinarian, an immunofluorescent assay for antibodies to feline enteric coronavirus had been performed, and the result was positive at > 1:6,400. No neurologic lesions supporting a diagnosis of feline infectious peritonitis had been found on postmortem examination. Genotyping for ABCB11930_1931del TC indicated that the cat was homozygous for the mutation.
Cat 4, a 12-month-old spayed female domestic shorthair, had vomited shortly after the owner had topically applied the label dose of an eprinomectin-containing product.b The cat had not exhibited other abnormal clinical signs. The cat had been allowed to roam outside and was found by the owner 2 days later and immediately taken to a veterinarian. The cat had been dehydrated, had bilateral mydriasis, and had demonstrated vestibular (circling to the left) and behavioral changes (aggression). Serum biochemical analysis had indicated hypokalemia and hyperglobulinemia. Supportive care including enteral nutrition administered through a feeding tube had been provided. An MRI of the brain had been interpreted to be normal. The cat had then been euthanized and a necropsy performed. The important findings had been moderately extensive demyelination in the ventral and lateral corticospinal tracts of the white matter that extended from the thalamus to the lumbar spinal region and retinopathy, mainly characterized by degeneration in the nerve fiber and ganglion cell layers. Genotyping for ABCB11930_1931del TC indicated that the cat had been homozygous for the wild-type ABCB1 allele.
Cat 5, a 5-month-old 2.3-kg (5.1-lb) neutered male domestic shorthair, had apparently gained access to an ivermectin-containing chewable heartworm preventivec for dogs while its owner had been at work. When the owner had returned home, they had noticed that the cat had difficulty walking. They also had found 2 boxes of the product that had appeared to have been torn apart by the cat. One box had contained chews formulated to provide appropriate dosing of ivermectin for 11.8- to 22.7-kg (26- to 50-lb) dogs and the other chews were formulated to provide appropriate dosing for 11.8-to 45.4 kg (51- to 100-lb) dogs (total dose up to 532 µg of ivermectin/kg). Each package would have had only 3 chews remaining, on the basis of the dosing schedule. The cat had been taken to a veterinary clinic where it had been described as ataxic, tremoring, and lethargic. Supportive treatment had included parenteral fluid administration. After 2 days with no improvement in the cat's condition, the owner had elected to euthanize the cat. Genotyping for ABCB11930_1931del TC indicated that the cat had been homozygous for the mutation.
Prospective study
The ABCB11930_1931del TC genotype for each of the 1,006 DNA-banked feline samples was determined. The distribution of genotypes from these samples was as follows: 0 homozygous for ABCB11930_1931del TC, 47 heterozygous for ABCB11930_1931del TC, and 959 homozygous for the wild-type ABCB1 allele. Thus, the frequency of the ABCB11930_1931del TC mutation was 4.7% (47/1,006). Among the 47 cats of various breeds (purebred and mixed [eg, domestic shorthair]) with the mutant ABCB1 allele, only 3 were purebred (Ragdoll, Russian Blue, and Siamese; Table 1). The other 44 cats harboring the ABCB11930_1931del TC mutation were not purebred.
Breed distribution of ABCB1 genotypes (homozygous for wildtype ABCB1 allele [WT/WT; n = 959], homozygous for ABCB11930_1931del TC [Mutant/Mutant; 0], and heterozygous for ABCB11930_1931del TC [WT/Mutant; 47]) of 1,006 DNA-banked feline samples at the DNA bank at the College of Veterinary Medicine, Washington State University.
| Breed | WT/WT | WT/Mutant | Mutant/Mutant |
|---|---|---|---|
| American Shorthair | 2 | — | — |
| Bambino | 1 | — | — |
| Bengal | 13 | — | — |
| Birman | 6 | — | — |
| Bombay | 1 | — | — |
| British Shorthair | 1 | — | — |
| Burmese | 2 | — | — |
| Calico | 4 | — | — |
| Chartreaux | 1 | — | — |
| Devon Rex | 2 | — | — |
| Domestic longhair | 125 | 7 | — |
| Domestic medium hair | 80 | 5 | — |
| Domestic shorthair | 581 | 30 | — |
| Exotic | 1 | — | — |
| Highland Lynx | 1 | — | — |
| Himalayan | 4 | — | — |
| Maine Coon | 20 | — | — |
| Manx | 3 | — | — |
| Manxamese | 1 | — | — |
| Norwegian Forest Cat | 3 | — | — |
| Ocicat | 2 | — | — |
| Oriental Shorthair | 2 | — | — |
| Other* | 21 | 2 | — |
| Persian | 9 | — | — |
| Pixie Bob | 1 | — | — |
| Ragdoll | 9 | 1 | — |
| Russian Blue | 7 | 1 | — |
| Savannah | 1 | — | — |
| Scottish Fold | 1 | — | — |
| Siamese | 26 | 1 | — |
| Siberian Forest Cat | 2 | — | — |
| Singapura | 2 | — | — |
| Snowshoe | 1 | — | — |
| Sphinx | 21 | — | — |
| Tonkinese | 1 | — | — |
| Turkish Van | 1 | — | — |
Other includes cats listed as mix (Siamese mix or Burmese mix) or unknown breed.
— = Homozygosity or heterozygosity not identified for breed.
Discussion
The first objective of the study reported here was to determine whether the ABCB11930_1931del TC gene mutation was associated with neurologic signs observed in each of 5 cats that had been exposed to a macrocyclic lactone–containing product. These products contained eprinomectin or ivermectin. Four of the 5 cats were homozygous for ABCB11930_1931del TC, whereas 1 was homozygous for the wild-type ABCB1 allele. The latter cat had exhibited unilateral vestibular signs inconsistent with macrocyclic-lactone toxicosis,7 so not identifying the mutation in this cat was not surprising. The other 4 cats had exhibited generalized CNS abnormalities consistent with macrocyclic-lactone toxicosis and were all homozygous for ABCB11930_1931del TC, thereby providing evidence that the ABCB1 mutation in cats creates a P-glycoprotein–deficient state similar to that in dogs with the ABCB1-1∆ (MDR) mutation. Thus, macrocycliclactone sensitivity caused by a deletion mutation in the ABCB1 gene can occur in cats.
Because 4 of 5 cats with ABCB11930_1931del TC were identified, albeit in a phenotyped population, we questioned the low (4.6%) frequency estimate reported5 previously involving 100 cats. Therefore, we genotyped an additional 1,006 banked feline DNA samples. Interestingly, the overall frequency of cats having 1 mutant allele (4.7%) in the present study was nearly identical to the frequency (4.6%) in that previous study.5 However, we were surprised that our genotyping yielded no cats that were homozygous for ABCB11930_1931del TC, which suggested that this genotype is rare in the general cat population (only 1 cat homozygous for the mutation in the previous study5). Whether ABCB11930_1931del TC is present in higher frequencies among specific cat breeds remains to be determined. Most cats (44/47) with ABCB11930_1931del TC were not purebred and were instead listed as domestic short-, medium-, or long-haired cats. The 3 purebred cats with ABCB11930_1931del TC were a Ragdoll, Russian Blue, and Siamese. However, numbers of these and other breeds were insufficient to make definitive statements about the presence of the mutation in specific cat breeds. Additional studies should be undertaken to determine the frequency of ABCB11930_1931del TC in specific cat breeds.
Although the present study was not designed to definitively determine whether ABCB11930_1931del TC is associated with avermectin sensitivity, it provided strong supportive evidence that it was. Given that < 1/1,000 cats in the present study were homozygous for ABCB11930_1931del TC, and on the basis of the results, the odds that 4 cats would be homozygous for ABCB11930_1931del TC by random sampling would be 1 X 10−16. We recommend that drug treatment protocols for cats with the ABCB11930_1931del TC geno-type follow protocols similar to those for dogs with the ABCB1-1∆ mutation.1,8
One of the macrocyclic lactone–containing productsc that had caused neurologic toxicosis in one of the cats reported here was labeled for use in dogs. This exposure had been unintended, and the cat had gained access to the packaged products and consumed up to 532 µg of ivermectin/kg. This oral dose is not sufficient to cause neurologic signs in an animal with functional P-glycoprotein, but it exceeds the dose that would cause toxicosis in dogs with the ABCB1-1∆ mutation.1,7
Two relatively new eprinomectin-containing productsa,b had caused neurologic signs in 3 of the cats reported here after the products had been applied topically in accordance with label instructions. These products are intended to deliver a dose of approximately 500 µg of eprinomectin/kg (227 µg/lb), which is well below the LD50 for rats and mice of 70 mg/kg (32 mg/lb) and 55 mg/kg (25 mg/lb), respectively. However, animals with defective P-glycoprotein function are susceptible to macrocyclic lactone–induced toxicosis at much lower doses because macrocytic lactones accumulate in the brain. Unfortunately, no antidote for macrocyclic-lactone toxicosis exists, so treatment consists of providing supportive care and, if the product was administered topically, bathing the animal to prevent further absorption. Intravenous lipid emulsion has also been described as an effective treatment.9 Importantly, however, dogs homozygous for the ABCB1-1∆ mutation do not seem to respond to this treatment, and therefore, cats homozygous for ABCB11930_1931del TC would also not be expected to respond.10
The present study explored the pharmacogenetic relationship between the ABCB11930_1931del TC mutation and eprinomectin-containing products labeled for use in cats. Results suggested a strong relationship between homozygosity for ABCB11930_1931del TC and toxicosis. Veterinarians should be aware of this potential genetically mediated adverse drug reaction and that full recovery may be achievable with appropriate supportive care.9
Acknowledgments
Funded in part by the Ott Endowment.
The authors declare that there were no conflicts of interest.
The authors thank the cat owners who contributed samples from their cats to the DNA bank at the College of Veterinary Medicine, Washington State University.
Footnotes
Centragard, Boehringer Ingelheim, Duluth, Ga.
Broadline, Boehringer Ingelheim, North Ryde, NSW, Australia.
Heartgard Plus, Boehringer Ingelheim, Duluth, Ga.
Quick-gDNA FFPE Kit D3067d, Zymo Research, Irvine, Calif.
Applied Biosystems AB3730 genetic analyzer, Thermo Fisher Scientific, Waltham, Mass.
Applied Biosystems GeneMapper 5 software, Thermo Fisher Scientific, Waltham, Mass.
Abbreviations
| bp | Base pairs |
References
- 2. ↑
Mealey KL, Greene S, Bagley R, et al. P-glycoprotein contributes to the blood-brain, but not blood-cerebrospinal fluid, barrier in a spontaneous canine p-glycoprotein knockout model. Drug Metab Dispos 2008;36:1073–1079.
- 3. ↑
Coelho JC, Tucker R, Mattoon J, et al. Biliary excretion of technetium-99m-sestamibi in wild-type dogs and in dogs with intrinsic (ABCB1-1∆ mutation) and extrinsic (ketoconazole treated) P-glycoprotein deficiency. J Vet Pharmacol Ther 2009;32:417–421.
- 4. ↑
Mealey KL, Bentjen SA, Gay JM, et al. Ivermectin sensitivity in Collies is associated with a deletion mutation of the mdr1 gene. Pharmacogenetics 2001;11:727–733.
- 5. ↑
Mealey KL, Burke NS. Identification of a nonsense mutation in feline ABCB1. J Vet Pharmacol Ther 2015;38:429–433.
- 6. ↑
Richards B, Skoletsky J, Shuber AP, et al. Multiplex PCR amplification from the CFTR gene using DNA prepared from buccal brushes/swabs. Hum Mol Genet 1993;2:159–163.
- 7. ↑
Merola VM, Eubig PA. Toxicology of avermectins and milbemycins (macrocyclic lactones) and the role of P-glycoprotein in dogs and cats. Vet Clin North Am Small Anim Pract 2018;48:991–1012.
- 9. ↑
Jourdan G, Boyer G, Raymond-Letron I, et al. Intravenous lipid emulsion therapy in 20 cats accidentally overdosed with ivermectin. J Vet Emerg Crit Care (San Antonio) 2015;25:667–671.
- 10. ↑
Wright HM, Chen AV, Talcott PA, et al. Intravenous fat emulsion as treatment for ivermectin toxicosis in three dogs homozygous for the ABCB1-1∆ gene mutation. J Vet Emerg Crit Care (San Antonio) 2011;21:666–672.
