P-glycoprotein (P-gp) is an efflux transporter belonging to the family of ATP-binding cassette (ABC) transporters and is encoded by the multidrug resistance gene MDR1 (syn. ABCB1, ATP-binding cassette transporter subfamily B member 1) (1). In vertebrates, it is highly expressed in the blood–brain barrier and in various tissues such as the intestine, liver, and kidney. Here, P-gp restricts the entry of many drugs and xenobiotics into the central nervous system (CNS), limits their uptake from the gastrointestinal tract, and enhances the excretion via liver and kidney (2–4). The substrate spectrum of P-gp is broad and includes many compounds, e.g., anticancer agents, immunosuppressants, opioids, antiparasitic and antimicrobial drugs, cardiac drugs, steroid hormones, and many more (5–7). Antiparasitic drugs from the macrocyclic lactone class including ivermectin, doramectin, and moxidectin are also typical and high-affinity P-gp substrates (8–11). Drug sensitivities toward macrocyclic lactones were first described in the early 1980s and were mostly referred to Collie dogs that received ivermectin (12). Later the term “ivermectin-sensitive Collie” was used to describe this drug-sensitive phenotype (13, 14), which is characterized by ataxia, depression, excessive salivation, tremors, recumbency, apparent blindness, and mydriasis. The hypothesis of a genetic defect in the canine Mdr1 as a possible reason for this drug sensitivity was formulated some years later following observations in Mdr1 knockout laboratory mice. When these mice were treated with therapeutic doses of ivermectin for mite infection, severe neurological toxicity occurred while the Mdr1 wild-type mice were completely unaffected (15). Today, it is well-established that the increased drug sensitivity in Collies and related dog breeds is caused by nt230(del4) 4-bp gene deletion mutation (syn. ABCB1-1Δ) in the canine Mdr1 gene. This mutation leads to a premature stop codon and subsequently to a non-functional protein (16), resulting in an increased penetration of P-gp substrates into the CNS.

Interestingly, signs of intoxication seen in ivermectin-sensitive Collies are consistent with the neurological signs observed in a goat of the current report: a group of goats was treated subcutaneously with 0.3 mg/kg of a doramectin product approved for deworming in cattle and sheep. One of these goats, a 1-year-old female Thuringian goat (26 kg body weight), developed neurological signs at about 12 h after treatment. According to the animal owner, symptoms included ataxia, salivation, tremor, apathy, blindness, disorientation, mydriasis, and recumbency. The goat showed no water or food intake, swallowing reflex was absent, and rumen motility was decreased.

So far, no data are available from the literature describing neurological signs in goats after treatment with doramectin. The Federal Office of Consumer Protection and Food Safety (Germany) conducted a research in the European pharmacovigilance database following an official data request by the authors of this report. According to these search results, only a few cases of suspected adverse events have been reported in goats after administration of doramectin and ivermectin worldwide (search period from January 1, 2000 to December 31, 2020). However, considering these few reports on doramectin, the terms ataxia, lethargy, and apathy are mentioned frequently.

As the adverse reactions of these suspected drug-sensitive goats as well as those of the affected goat of the current case phenocopy quite well the ivermectin sensitivity of the Collie (with Mdr1 mutation), we hypothesized a putative Mdr1 mutation in these animals. Therefore, we performed diagnostic sequencing of the full-length Mdr1 transcript of the suspected drug-sensitive Thuringian goat and of two related (sire and half-sibling) non-sensitive goats to allow differential diagnosis.

Mutations and polymorphisms in the MDR1/Mdr1 gene have been reported in humans (17), mice (18), dogs (16), and cats (19). In humans, numerous MDR1 SNPs have been identified, but there is no clear consensus regarding their functional and/or clinical consequence (20). In mice, drug sensitivity has been shown in Mdr1 knockout mice against ivermectin (15, 21). The same applies to dogs, in which the nt230(del4) 4-bp deletion mutation in the Mdr1 gene dramatically increases drug sensitivity to macrocyclic lactones, such as ivermectin, doramectin, or moxidectin (7, 11, 16). Typical symptoms of neurological toxicity induced by macrocyclic lactones in Mdr1 mutant dogs include ataxia, CNS depression, excessive salivation, tremor, apparent blindness, and even death.

In the current report, a Thuringian goat showed similar neurological signs after subcutaneous injection of 0.3 mg/kg doramectin as well. To date, neither any report about drug sensitivity to doramectin in goats, nor any mutations in the Mdr1 gene of goats have been published. However, a research in the European pharmacovigilance database that was conducted by the Federal Office of Consumer Protection and Food Safety (Germany) revealed global occurrence of few cases of suspected adverse events in goats following the administration of ivermectin or doramectin. In these cases of adverse event reports after administration of doramectin, the following terms are commonly mentioned: ataxia, lethargy, and apathy. The adverse drug reactions of these suspected drug-sensitive goats phenocopy quite well the drug sensitivity of Mdr1 mutant dogs against macrocyclic lactones (7). Based on this, it was hypothesized that the suspected doramectin-sensitive Thuringian goat of the present report might have defective expression or function of the Mdr1 drug efflux transporter in the blood–brain barrier. Therefore, we focused on searching for missense mutations or sequence variations in the mRNA coding sequence and flanking 3′- and 5′-UTR regions. However, it was not possible to analyze Mdr1 mRNA and protein expression at the blood–brain barrier as this would have required sampling from the brain. The Mdr1 transcript was PCR amplified and sequenced from the suspected drug-sensitive Thuringian goat as well as from two closely related animals (sire and half-sibling), which also received the doramectin treatment, but did not show any adverse events. The only sequence variation that could explain the different phenotypes was a heterozygous SNP in the 3′-UTR of the Mdr1 transcript, being c.3875C>A. It can only be speculated that this SNP might have affected Mdr1 expression that finally would have resulted in defective of reduced Mdr1 expression at the blood–brain barrier. This can, however, not be evidenced in the current case and needs further investigation.

A variety of possibilities exists, how alterations of 3′-UTRs can affect protein expression levels, i.e., changes of mRNA stability, changes in poly-A-tailing motifs, accessory-protein binding, as well as changes of microRNA binding sites. To address this question, we in silico screened the c.3875C>A sequence variation for potential changes of binding motives with various algorithms (RBPDB: http://rbpdb.ccbr.utoronto.ca and Scan for Motifs: http://crispr.otago.ac.nz/sfm/sfm_main.pl) but could not find any plausible motif. This might be due to the fact that available databases are mainly human specific or pan-mammalian, so their usefulness for goat-UTRs is limited and may yield false-negative results. To underline the possible impact of a 3′-UTR mutation, we want to hint to the recent work of Martinez et al. (22). They showed that CYP2B11 3′-UTR sequence variations, which are frequently present in dog breeds such as the Greyhound, can lead to decreased CYP2B11 protein abundance. To verify if there might be a similar causality for caprine Mdr1 expression for specific goat breeds or goats in general, an elaborate prospective study like the one mentioned above would be needed, to provide sufficient genetic and especially tissue samples to be able to prove or reject this hypothesis. Therefore, it finally remains unclear whether or not the c.3875C>A SNP located in the 3′-UTR impacted P-gp expression of the suspected drug-sensitive goat and thus is responsible for the observed differences in drug sensitivity. Furthermore, other mutations in the non-coding region or epigenetic modifications altering the Mdr1 gene expression cannot be excluded. Here, for example, promotor region mutations would also be possible. In the current case, however, a severe defect of the promotor seems to be unlikely since sequencing revealed amplification of both alleles. In total, further research with a larger number of affected animals and additional sample acquisition for measuring P-gp expression, especially in the brain, would be necessary in the future to verify a potential association of the c.3875C>A mutation with increased drug sensitivity to avermectins in the goat.

In the present case, besides pharmacogenetic and epigenetic causes, other reasons responsible for the neurological signs of the suspected drug-sensitive goat must be considered as well. According to the animal owner, neurological signs included ataxia, apathy, tremor, salivation, mydriasis, and recumbency, although the goat was able to stand and walk on its own when the animal was placed in an upright position. Obviously, the goat was blind and disorientated; moreover, swallowing reflex was absent and rumen motility was decreased. Of note, these symptoms occurred in close time with the doramectin treatment of the goat. Symptomatic treatment included subcutaneous infusions of saline (WDT) and Amynin (Merial) solutions, injections of Konstigmin 2.5 mg/ml (Vetoquinol) and Vitamin-B-Komplex pro inj. (Serumwerk), as well as Vetalgin 500 mg/ml (MSD) for several days. Additionally, the goat received ColoSan (SaluVet) and Pansenstimulans (WDT) orally when swallowing reflex was recovered. Symptoms lasted for about 1 week and then improved gradually. In total, symptoms and time course of intoxication are very similar to ivermectin and doramectin intoxications reported for Mdr1 mutant dogs (23, 24). However, there are reports of dogs treated with macrocyclic lactones that showed similar clinical symptoms but did not have the nt230(del4) Mdr1 mutation (25–27). Some of these cases include accidental ingestion and overdosing. In the current case, a doramectin product approved for cattle and sheep was administered to goats, implying an extra-label use. According to the product information, the manufacturer's dose recommendation for cattle and sheep is 0.2 mg/kg doramectin for intramuscular administration (accessed March 2021). This dose has also been administered to goats in some pharmacokinetic, milk kinetic, and efficacy studies with a subcutaneous route of administration (28–33). In the current case, 0.3 mg/kg doramectin were administered subcutaneously. However, this dose was well-tolerated by several other goats of the group treated with doramectin at the same time. Nevertheless, it is well-known that inter-individual factors, such as percent of body fat, age, diet, or disease, have an influence on the pharmacokinetics or pharmacodynamics of a drug and may therefore lead to variations in drug response (34). It should also be mentioned that the Thuringian goat has recovered almost completely, but according to the owner, blindness lasted for several months, which seems to be unusual for an intoxication with macrocyclic lactone drugs (25, 27). Therefore, other possible reasons for the occurrence of neurological adverse events should be considered. A variety of differential diagnoses exist for neurological disorders in goats, such as polioencephalomalacia/cerebrocortical necrosis, listeriosis, hepatic encephalopathy, further metabolic diseases, as well as many others. Blindness often occurs as a secondary finding in some of these diseases. Poisonings with toxic plants or chemicals may also lead to neurological signs associated with blindness (35, 36). In this regard, it should be noted that the goat was found on a surrounding pasture after doramectin treatment, which had been limed the day before.

In conclusion, this report provides the first sequence of the caprine Mdr1 based on blood-derived mRNA. Diagnostic Mdr1 sequencing revealed no missense mutations in the Mdr1 coding sequence of the goat showing neurological signs after treatment with doramectin, but we identified a heterozygous variation located in the 3′-UTR (c.3875C>A) where the sequence of the suspected drug-sensitive goat differed from the two related non-sensitive goats that were treated with doramectin as well. It finally remains unclear if this mutation is responsible for the observed differences in potential drug sensitivities among the three goats after treatment with doramectin. In addition, it is unknown if other mutations in the non-coding region or epigenetic mechanisms exist that may have altered Mdr1 gene expression in the affected goat. In this case, however, besides pharmacogenetic and epigenetic causes, other reasons responsible for the neurological signs cannot be excluded.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethical review and approval was not required for the animal study because blood samples were received for diagnostic sequencing of the Mdr1 transcript due to suspected Mdr1-related drug sensitivity. For this reason, ethics approval was not necessary in agreement with the institutional animal welfare officer of the Justus Liebig University Giessen. Written informed consent was obtained from the animal owner for publication of the data. Written informed consent was obtained from the owners for the participation of their animals in this study.

DN, SM, MH, and JG conceived the diagnostic sequencing, analyzed and interpreted the data, and critically edited and revised the manuscript. DN performed the sequencing and drafted the first manuscript. DN and SM prepared figures and tables. All authors contributed to the article and approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.