Untitled

Journal of Veterinary Emergency and Critical Care 17(1) 2007, pp 61–66 Comparison of the hypothalamic^pituitary^adrenalaxis in MDR1-1D and MDR1 wildtype dogs Katrina L. Mealey, DVM, PhD, DACVIM, DACVCP, John M. Gay, DVM, PhD, DACVPM,Linda G. Martin, DVM, MS, DACVECC and Denise K. Waiting, LVT Objective: To evaluate the hypothalamic–pituitary–adrenal (HPA) axis in MDR1-1D (dogs with the MDR1mutation associated with ivermectin sensitivity) and MDR1 wildtype dogs.
Design: Prospective study.
Setting: Institutional vivarium.
Animals: Seven healthy Collie dogs.
Measurements: MDR1 genotyping was used for allocation of dogs to 1 of 2 groups: dogs homozygous for thewildtype MDR1 allele (MDR1 wildtype) and those homozygous for the MDR1-1D mutation (MDR1 mutant).
Blood samples were obtained for determination of cortisol and adrenocorticotropin hormone (ACTH)concentrations under basal conditions, before and after ACTH administration, and before and afterdexamethasone administration.
Main results: Significant differences were identified between the MDR1 mutant and MDR1 wildtype groups.
Basal plasma cortisol concentrations and cortisol concentrations after ACTH administration were significantlylower in MDR1 mutant dogs as compared with MDR1 wildtype dogs. Plasma ACTH concentrations afterdexamethasone administration were significantly lower in MDR1 mutant dogs as compared with MDR1wildtype dogs.
Conclusions: Results suggest that P-glycoprotein (P-gp) plays a role in regulation of the HPA axis.
Furthermore, it appears that the HPA axis in MDR1 mutant dogs that lack P-gp is suppressed compared withMDR1 wildtype dogs. This finding may explain some clinical observations in breeds known to harbor theMDR1 mutation including Collies, Shelties, Australian Shepherds, and others. There is a clinical impressionthat many of these dogs have worse outcomes in response to stress and, at times, respond poorly toappropriate therapy. HPA axis suppression, secondary to the MDR1 mutation, could result in a relativeadrenal insufficiency (RAI) state during times of stress or illness. Further studies are required to determine therelationship between the MDR1 genotype and RAI.
(J Vet Emerg Crit Care 2007; 17(1): 61–66) doi: 10.1111/j.1476-4431.2006.00196.x Keywords: blood–brain barrier, collie, cortisol, P-glycoprotein, relative adrenal insufficiency capillary lumen. P-gp appears to be highly homolo-gous, but not identical, among mammalian species.
P-glycoprotein (P-gp), the product of the MDR1 or Drugs that are known substrates for canine P-gp in- ABCB1 gene, is a crucial component of the blood–brain clude ivermectin, loperamide, vincristine, vinblastine, barrier, protecting the brain from many potentially toxic and doxorubicin.2–5 Over 50 drugs have been shown to xenobiotics.1 P-gp functions as an ATP-dependent drug be substrates for human and murine P-gp6,7 and, be- transporter that is expressed on the luminal membrane cause of the high degree of homology of P-gp between of brain capillary endothelial cells where it transports a species, it is suspected that these same drugs are subst- variety of substrates from the brain tissue back into the rates for canine P-gp. Individuals that lack P-gp, asoccurs in herding breed dogs with the MDR1-1D From the Department of Veterinary Clinical Sciences, College of VeterinaryMedicine, Washington State University, Pullman, WA.
mutation, are highly susceptible to neurotoxicosis afterroutine doses of P-gp substrate drugs such as ivermec- Address correspondence and reprint requests to:Katrina L. Mealey, Department of Veterinary Clinical Sciences, College tin and loperamide.8–11 This susceptibility underscores of Veterinary Medicine, Washington State University, Pullman, the importance of P-gp in limiting exposure of the brain WA 99164-6610.
E-mail: [email protected] & Veterinary Emergency and Critical Care Society 2006 Exogenous substances are not the only substrates for P-gp. In rodents, endogenous hormones including cor-ticosterone are substrates for P-gp suggesting thatP-gp may have a role in regulating their plasma con-centrations.12 Results from recent studies support this contention. In rodent studies, P-gp was shown to re-strict access of corticosterone and cortisol to the brain.12Furthermore, abcb1ab ( À / À ) double knockout micethat lack P-gp have a suppressed hypothalamic–pitu-itary–adrenal (HPA) axis compared with wildtype mice.13 Collectively, these results suggest that P-gpnormally limits the concentration of cortisol and corti- costerone at the hypothalamus and pituitary bluntingfeedback inhibition of the HPA axis. If this premise is correct, then higher concentrations of cortisol and cor-ticosterone would be expected to reach the hypo-thalamus and pituitary in animals that lack P-gp re-sulting in greater feedback inhibition of the HPA axis and low endogenous cortisol levels (Figure 1). Conse- quently, dogs such as herding breed dogs with theMDR1-1D mutation would be expected to be predis-posed to relative adrenal insufficiency (RAI).
RAI is characterized by inadequate production of cortisol in relation to an increased physiological de- mand during periods of stress such as critical ill- ness.14,15 Human patients with RAI have a reducedcapacity to cope with critical illness and these patients Figure 1: Illustration of the hypothalamic–pituitary–adrenal have been shown to have a poorer outcome than pa- (HPA) axis in wildtype dogs (A) and MDR1 mutant dogs (B).
tients with a normal HPA axis.14 Interestingly, some In wildtype dogs, P-glycoprotein (P-gp) is present at the blood– veterinarians have described Collies as ‘wimpy’ or as brain barrier and limits entry of cortisol into the brain further ‘not participating in their own recovery’ because some limiting cortisol’s feedback inhibition of corticotrophin releasing individuals of this breed have had poorer outcomes or hormone (CRH) secretion and adrenocorticotropin hormone have not responded as well as dogs of other breeds (ACTH). In MDR1 mutant dogs, P-gp is not present at the with similar illnesses.a These anecdotal observations blood–brain barrier, therefore, greater concentrations of cortisol are consistent with the hypothesis that dogs with the are present within the brain enabling cortisol to exert feedback MDR1-1D mutation have a blunted HPA axis compared inhibition of CRH, ACTH, and, ultimately, cortisol secretion.
Solid lines indicate stimulatory effects (with bolder lines indi- with MDR1 wildtype dogs. Therefore, the purpose of cating greater stimulation) while dashed lines indicate inhibi- this study was to investigate HPA system regulation in tory effects (with bolder lines indicating greater inhibition).
MDR1-1D and MDR1 wildtype dogs utilizing dexameth-asone suppression and adrenocorticotropin hormone 2 females) were included in the study. Genotyping was performed at a commercial laboratory.b For each dog, baseline ACTH values were measured on All aspects of this study were approved by the Insti- 3 occasions and baseline cortisol values were measured tutional Animal Care and Use Committee. Seven Collie on 5 occasions. ACTH stimulation and dexamethasone dogs were studied ranging in age from 1 to 8 years of suppression studies were performed once. Studies were age. Dogs were determined to be healthy on the basis of performed such that collection of basal (prestimulation results from physical examination, a complete blood or presuppression) samples occurred at approximately count, serum biochemistry panel, and urinalysis. Three 08:30 hours ( Æ 30 minutes). Blood was collected by dogs that were homozygous for the MDR1 wildtype jugular venipuncture. Blood was injected into EDTA genotype (1 male, 2 females) and 4 dogs that were tubes for cortisol determination. For ACTH determina- homozygous for the MDR1-1D mutation (2 males, tion, blood was injected into prechilled EDTA tubes & Veterinary Emergency and Critical Care Society 2006, doi: 10.1111/j.1476-4431.2006.00196.x with a protease inhibitor, 100 mL aprotinin, added. Plas-ma was harvested by centrifugation at 4 1C. Plasma wastransferred to prechilled cryovials and frozen at À 80 1C until analyzed. Cortisol and ACTH concentra- tions were determined by a commercial laboratoryc us-ing methodology validated for the dog as previouslydescribed.16 A washout period of at least 2 weeks wasallowed between each study.
ACTH stimulation testingACTH (1 mg/kg) was administered intravenously (IV)(cephalic vein). Blood was collected before (time 0) and1 hour after ACTH administration for determination ofplasma cortisol concentrations.
Figure 2: Box and whisker plot of basal plasma cortisol samples in MDR1 mutant dogs (mutant; n 5 4 dogs, 3 samples per dog) Three different doses of dexamethasone (0.05, 0.01, and and MDR1 wildtype dogs (control; n 5 3 dogs, 3 samples per 0.001 mg/kg) were used. Dexamethasone was injected dog). Basal plasma cortisol concentrations are significantly low- IV (cephalic vein). Blood was collected before (time 0), er in MDR1 mutant dogs (nP 5 0.044) as compared with MDR1 4, and 8 hours after dexamethasone injection for deter- wildtype dogs. Box, interquartile range (IQR); line, median; mination of cortisol and ACTH concentrations.
whiskers, 1.5 Â IQR; dots, data points greater than 1.5 Â IQR.
vs. 226 Æ 12.7 nmol/L SEM, F1, 5 5 7.02, P 5 0.045), but Data were entered into a spreadsheetd and statistical the degree of response (i.e., slope) was not different analyses were performed using a commercial statistics program.e Because of repeated sampling over fixed For dexamethasone suppression testing, statistically times after multiple levels of stimulation, repeated significant differences in plasma cortisol concentrations measures ANOVA models with nested subjects as ran- between MDR1 mutant and wildtype dogs in response dom factors were employed. Standardized residuals to dexamethasone administration were not detected were plotted to detect departures from normality.
at any dexamethasone dose (Figure 4). At the 4- and Greenhouse–Geisser epsilon values were calculated 8-hour time points after 0.05 mg/kg of dexamethasone for the repeated measures to determine if F-test de-grees of freedom required reduction for deviations fromcompound symmetry and were adjusted accordingly. Acritical value of 0.05 was used as the threshold of sta-tistical significance and results are presented withstandard error of the mean (SEM) as an indication ofdata variability.
For each dog, baseline ACTH values were measured on3 occasions and baseline cortisol values on 5 occasions.
A statistically significant association was observed be-tween P 5 0.044), but not for ACTH (P 5 0.24). Mean basalcortisol concentrations in MDR1 mutant dogs were Figure 3: Box and whisker plot of plasma cortisol concentra- 39.3 Æ 6.18 versus 61.2 Æ 5.35 nmol/L SEM for MDR1 tions in MDR1 wildtype dogs (control; n 5 3 dogs) and MDR1 wildtype dogs. Figure 2 shows a box and whisker plot mutant dogs (mutant; n 5 3 dogs) immediately before (pre) and of basal cortisol concentrations in MDR1 mutant and 1 hour after (post)-intravenous administration of 1 mg/kg syn- MDR1 wildtype dogs. For ACTH stimulation testing, thetic adrenocorticotropin hormone (ACTH). The mean plasma MDR1 mutant dogs displayed significantly lower over- cortisol concentration after ACTH administration was signifi- all mean plasma cortisol concentrations after ACTH cantly lower in MDR1 mutant dogs (n) than in MDR1 wildtype administration than MDR1 wildtype dogs (181.5 Æ 11.0 & Veterinary Emergency and Critical Care Society 2006, doi: 10.1111/j.1476-4431.2006.00196.x Cortisol Concentration (nmol 10
ACTH Concentration (pg
Time (hours)
Time (hours)
Figure 4: Mean (SEM) plasma cortisol concentrations in MDR1wildtype dogs (wild; n 5 3) and MDR1 mutant dogs (mutant; Figure 5: Mean (SEM) plasma adrenocorticotropin hormone n 5 4) before (time 0), 4, and 8 hours after dexamethasone ad- (ACTH) concentrations in MDR1 wildtype dogs (wild; n 5 3) ministration. Dexamethasone was administered at 3 different and MDR1 mutant dogs (mutant; n 5 4) before (time 0), 4, and 8 doses with at least a 2-week washout period between dosing.
hours after dexamethasone administration. A significant inter- Statistically significant differences in plasma cortisol concentra- action (P 5 0.037) was detected between genotype (MDR1 sta- tions between MDR1 mutant and wildtype dogs in response to tus) and ACTH plasma concentrations. Dexamethasone was dexamethasone administration were not detected.
administered at 2 different doses with at least a 2-week washoutperiod between dosing.
was administered, plasma ACTH concentrations fellbelow the limit of quantitation and so statistical com-parisons were not performed. However, a significant tion by allowing higher brain concentrations of endog- between genotype (MDR1 status) and ACTH plasma releasing hormone (CRH) secretion. Studies were per- concentrations. Means for the MDR1 mutant dogs formed in the abcb1ab double knockout mice to test this receiving 0.01 mg/kg dexamethasone dose were signif- hypothesis. Compared with wildtype mice, abcb1ab icantly lower (12.67 Æ 1.47 pg/mL SEM) than MDR1 ( À / À ) double knockout mice have lower plasma cor- wildtype dogs (22.89 Æ 1.70 pg/mL SEM) receiving that tisol concentrations under basal conditions and under dose of dexamethasone (Figure 5). At the lowest dexa- conditions of stress.13 The knockout mice also have methasone dose (0.001 mg/kg), a statistically significant downregulated CRH mRNA expression in the hypo- difference in plasma ACTH concentrations was not thalamic paraventricular nucleus compared with wild- type mice suggesting a sustained suppression of theHPA system.13 Because MDR1 mutant dogs are pheno-typically similar to abcb1ab ( À / À ) double knockout mice with respect to brain penetration of other P-gp P-gp functions as an ATP-dependent drug efflux pump substrates (e.g., ivermectin, loperamide), it was sus- that is capable of transporting substrate drugs against a pected that MDR1 mutant dogs might show evidence of concentration gradient from the intracellular to the ex- tracellular space.17 It is expressed on the luminal aspect The present data suggest that lack of P-gp function at of brain capillary endothelial cells where it functions to the blood–brain barrier in MDR1 mutant dogs results in limit brain penetration of substrate xenobiotics includ- altered activity and regulation of the HPA axis. Similar ing the endogenous hormones, corticosterone, and co- to results from rodent studies, mean basal cortisol con- rtisol.1,17,18 Studies in rodents have shown that P-gp centrations in MDR1 mutant dogs (39.3 Æ 6.18 nmol/L limits access of corticosterone to the brain.12 Further SEM) were significantly lower than in MDR1 wildtype studies were performed in genetically engineered dogs (61.2 Æ 5.35 nmol/L). It is presumed that this ap- abcb1ab double knockout mice that are phenotypically parent suppression of systemic cortisol concentrations similar to MDR1 mutant dogs since both lack P-gp. Be- in MDR1 mutant dogs results from greater cortisol cause P-gp normally limits access of endogenous cor- feedback on hypothalamic paraventricular neurons in ticosteroids to the brain, it was hypothesized that lack these dogs as compared with wildtype dogs that have of P-gp might profoundly influence HPA axis regula- normal P-gp function. The fact that MDR1 mutant dogs & Veterinary Emergency and Critical Care Society 2006, doi: 10.1111/j.1476-4431.2006.00196.x had lower plasma cortisol concentrations, even after as acute illness, resulting in RAI or an RAI-like ACTH stimulation, suggests that there is chronic sup- pression of cortisol production. This observation raises RAI is a syndrome characterized by insufficient pro- questions as to the ability of MDR1 mutant dogs to duction of cortisol in relation to an increased demand during periods of severe stress, particularly in critical Results of dexamethasone suppression tests from the illnesses such as sepsis or septic shock.14,19 The syn- present study differed somewhat from results reported drome is presumed to be associated with altered func- for abcb1ab ( À / À ) double knockout mice. In both tion of the HPA axis. Insufficient stress hormone MDR1 mutant dogs and abcb1ab ( À / À ) double knock- synthesis appears to be a transient phenomenon in hu- out mice, plasma concentrations of ACTH were signif- man RAI patients since life-long replacement of cor- icantly lower than in their wildtype counterparts.
ticosteroids (as would be essential in patients with true However, while plasma concentrations of cortisol were hypoadrenocorticism) is not necessary.19 From a clinical suppressed at lower dexamethasone doses in abcb1ab perspective, it is extremely important to recognize sep- ( À / À ) double knockout mice as compared with wild- tic patients with RAI because these patients tend to type mice, there was not a statistically significant dif- carry a worse outcome if untreated. Treatment, which consists of low (‘physiologic’) doses of corticosteroids, dexamethasone administration in MDR1 mutant versus appears to reduce morbidity and mortality rates, par- ticularly in septic patients.20 Although RAI is well doc- It is not clear why ACTH concentrations following umented in critically ill human patients,20–22 relatively dexamethasone suppression differ in MDR1 mutant less is known about adrenal dysfunction in critically ill dogs as compared with wildtype dogs while cortisol veterinary patients. However, research models of sepsis concentrations do not. One can speculate that differ- and hemorrhagic shock suggest the presence of RAI in ences in plasma cortisol concentrations may have been animals.21,22 There are also a small number of clinical observed if plasma samples had been obtained at later studies that suggest RAI or RAI-like syndromes in times. The higher ACTH concentrations in MDR1 wild- certain populations of veterinary patients.f,g,h type dogs may have resulted in greater plasma cortisol Interestingly, herding breed dogs such as Collies and concentrations after the last sample time point (i.e., 8 Shelties appear to have a reputation for not handling hours after dexamethasone administration). It is also illness well. For example, 1 oncologist stated in an on- possible that greater subject numbers would have dem- line discussion that ‘Shelties are indeed ‘‘wimps’’ and onstrated a significant difference in plasma cortisol they tend to join Collies in this category.’c Another on- concentrations. There was also greater variability for cologist commented that ‘Collies sometimes do not plasma cortisol concentrations than for plasma ACTH participate in their own recovery.’c Indeed, the obser- concentrations. Whether this difference may have been vation that led to the study reported here involved due to variability among the subjects or variability what appeared to be an RAI-like phenomenon in an within the cortisol assay as compared with the ACTH MDR1 mutant Collie that had undergone a prolonged, assay is not known. However, despite the variability of but relatively benign, surgical procedure.
plasma cortisol concentrations, the number of subjects From this study, it appears that P-gp is an important was adequate to demonstrate significant differences in component of the HPA axis in dogs. Dogs lacking P-gp HPA axis regulation for several other parameters tested (i.e., MDR1 mutant dogs) appear to have a chronically including resting cortisol concentrations, post-ACTH suppressed HPA axis compared with normal dogs.
cortisol concentrations, and greater suppression of Further studies are necessary to determine the rela- ACTH secretion after dexamethasone administration.
tionship between MDR1 genotype and RAI. However, Collectively, results of this study provide evidence it might be prudent for veterinarians to consider testing that the HPA axis in MDR1 mutant dogs is suppressed for RAI (ACTH stimulation) in critically ill canine pa- as compared with wildtype dogs. Similar to abcb1ab tients of the herding breed group, since the MDR1 mu- ( À / À ) double knockout mice, MDR1 mutant dogs tation has been identified in Collies, Shelties, Old lack P-gp function. Consequently, there is increased English Sheepdogs, Australian Shepherds, and oth- penetration of corticosteroids into the central nervous ers.23,24 If these dogs have clinical signs consistent with system, resulting in more pronounced negative feed- RAI, then ‘physiologic’ doses of corticosteroids may be back inhibition of stress hormone secretion. While this considered while awaiting test results. Treatment can be excessive negative feedback inhibition in MDR1 mutant gradually discontinued in those dogs that appear to dogs does not appear to interfere with basal physiologic respond to treatment and have normal results in re- function in these dogs, it is reasonable to speculate that sponse to ACTH stimulation and can be continued in HPA axis function may be inadequate in situations such those dogs that fail to respond appropriately. Because & Veterinary Emergency and Critical Care Society 2006, doi: 10.1111/j.1476-4431.2006.00196.x MDR1 genotyping is commercially available,a many 7. Seelig A. A general pattern for substrate recognition by P-glyco- protein. Eur J Biochem 1998; 251:252–261.
owners of Collies and other herding breeds know their 8. Schinkel AH, Smit JJ, van Tellingen O, et al. Disruption of the dog’s genotype. It is important to obtain this informa- mouse MDR1a P-glycoprotein gene leads to a deficiency in the tion and consider the implications of the MDR1 mutant blood–brain barrier and to increased sensitivity to drugs. Cell1994; 20(77):491–502.
genotype when treating these patients.
9. 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.
10. Jonker JW, Wagenaar E, van Deemter L, et al. Role of blood–brain barrier P-glycoprotein in limiting brain accumulation and sedative This study was funded by a grant from the Collie side-effects of asimadoline, a peripherally acting analgaesic drug.
11. Sartor LL, Bentjen SA, Trepanier L, et al. Loperamide toxicity in a collie with the MDR1 mutation associated with ivermectin sensi-tivity. J Vet Intern Med 2004; 18:117–118.
12. Uhr M, Holsboer F, Muller MB. Penetration of endogenous steroid hormones corticosterone, cortisol, aldosterone and progesterone Veterinary Information Network; www.vin.com/members/search.
into the brain is enhanced in mice deficient for both MDR1a and Veterinary Clinical Pharmacology Laboratory, College of Veterinary MDR1b P-glycoproteins. J Neuroendocrinol 2002; 14:753–759.
Medicine, Washington State University, Pullman, WA; www.vetmed.
13. Muller MB, Keck ME, Binder EB, et al. ABCB1 (MDR1)-type P- glycoproteins at the blood–brain barrier modulate the activity of Endocrine Diagnostic Service, Department of Anatomy, Physiology and the hypothalamic–pituitary–adrenocortical system: implications Pharmacology, College of Veterinary Medicine, Auburn University, AL.
Excel 2002, Microsoft Corporation, Redmond, WA.
for affective disorder. Neuropsychopharmacology 2003; 28:1991– NCSS Number Cruncher Statistical Systems, Kaysville, UT; NCSS.com.
Prittie JE, Barton LJ, Peterson ME, et al. Hypothalmo-pituitary–adrenal 14. Beishuizen A, Thijs LG. Relative adrenal failure in intensive care: (HPA) axis function in critically ill cats. In: Proceedings of the 9th an identifiable problem requiring treatment? Best Pract Res Clin International Veterinary Emergency and Critical Care Symposium, New Endocrinol Metab 2001; 15:513–531.
Orleans, September 9–13, 2003, p. 771.
15. Marik PE, Zaloga GP. Adrenal insufficiency in the critically ill: a Farrelly J, Hohenhaus AE, Peterson ME, et al. Evaluation of pituitary– new look at an old problem. Chest 2002; 122:1784–1796.
adrenal function in cats with lymphoma. In: Proceedings of the 19th 16. Brockus CW, Dillon AR, Kemppainen RJ. Effect of alternate-day Annual Veterinary Cancer Society Conference, Wood’s Hole, November activity in dogs. Am J Vet Res 1999; 60:698–702.
Boozer A, Behrend EN, Whitley EM, et al. Hypothalamic–pituitary– 17. Schinkel AH, Wagenaar E, Mol CA, et al. P-glycoprotein in the adrenal axis function in dogs with neoplasia. J Vet Intern Med 2004; blood–brain barrier of mice influences the brain penetration and pharmacological activity of many drugs. J Clin Invest 1996;97:2517–2524.
18. Cordon-Cardo C, O’Brien JP, Casals D, et al. Multidrug-resistance gene (P-glycoprotein) is expressed by endothelial cells at blood–brain barrier sites. Proc Natl Acad Sci USA 1989; 86:695–698.
1. van Asperen J, Mayer U, van Tellingen O, et al. The functional role 19. Martin LG, Groman RP. Relative adrenal insufficiency in critical of P-glycoprotein in the blood–brain barrier. J Pharm Sci 1997; illness. J Vet Emerg Crit Care 2004; 14:149–157.
20. Ligtenberg JJ, Zijlstra JG. The relative adrenal insufficiency 2. Mealey KL, Barhoumi R, Rogers K, et al. Doxorubicin induced syndrome revisited: which patients will benefit from low-dose expression of P-glycoprotein in a canine osteosarcoma cell line.
steroids? Curr Opin Crit Care 2004; 10:456–460.
21. Wang P, Ba ZF, Jarrar D, et al. Mechanism of adrenal insufficiency 3. Mealey KL, Bentjen SA, Gay JM, et al. Dexamethasone treatment following trauma and severe hemorrhage: role of hepatic 11beta- of a canine, but not human, tumour cell line increases chemore- hydroxysteroid dehydrogenase. Arch Surg 1999; 134:394–401.
sistance independent of P-glycoprotein and multidrug resistance- 22. Koo DJ, Jackman D, Chaudry IH, et al. Adrenal insufficiency dur- related protein expression. Vet Comp Oncol 2003; 1:67–75.
ing the late stage of polymicrobial sepsis. Crit Care Med 2001; 4. Page RL, Hughes CS, Huyan S, et al. Modulation of P-glycopro- tein-mediated doxorubicin resistance in canine cell lines. Antican- 23. Neff MW, Robertson KR, Wong AK, et al. Breed distribution and history of canine MDR1-1{Delta}, a pharmacogenetic mutation that 5. Mealey KL, Northrup NC, Bentjen SA. Increased toxicity of P- marks the emergence of breeds from the collie lineage. Proc Natl glycoprotein-substrate chemotherapeutic agents in a dog with the Acad Sci USA 2004; 100:11725–11730.
MDR1 deletion mutation associated with ivermectin sensitivity.
24. Nelson OL, Carsten E, Bentjen SA, et al. Ivermectin toxicity in an J Am Vet Med Assoc 2003; 223:1453–1455, 1434.
Australian Shepherd dog with the MDR1 mutation associated 6. Ford JM, Hait WN. Pharmacologic circumvention of multidrug with ivermectin sensitivity in Collies. J Vet Intern Med 2003; resistance. Cytotechnology 1993; 12:171–212.
& Veterinary Emergency and Critical Care Society 2006, doi: 10.1111/j.1476-4431.2006.00196.x

Source: http://www.mdr1-defekt.de/texte/PublicationMealey.pdf