Hypercholesterolaemia in briards in the United Kingdom

Research in Veterinary Science 1993, 54, 80-85

Hypercholesterolaemia in briards in the United Kingdom P. WATSON, K. W. SIMPSON, P. G. C. BEDFORD, Royal Veterinary College, University of

London, Hawkshead Lane, North Mymms, Hatfield, Hertfordshire AL9 7TA

believed to originate from lipid material from rod outer segment membranes (Boulton et al 1989). Abnormal plasma lipids have been documented in some human retinitis pigmentosa patients and these may be involved in the aetiopathogenesis of the ocular lesions (Converse et al 1985, Jahn et al 1987)o Riis and Aguirre (1983) described a condition in 24 juvenile briards in the USA involving the accumulation of lipid droplets in the RPE. This condition is referred to as briard lipid retinopathyo Affected dogs had a mild hypeflipidaemia. An apparently similar problem to this lipid retinopathy has been described in five of a litter of nine briard pups in Sweden (Narfstrom et al 1989) although investigations into plasma biochemistry were not reported. A pilot study to investigate potential disturbances in lipid metabolism in canine RPED was undertaken in briards in the UK, a breed in which RPED commonly occurs (Bedford 1984, Lightfoot 1988). This study revealed that 84 of 90 clinically healthy briards randomly selected from 300 briards examined for RPEDhad significantly higher fasting plasma cholesterol concentrations than RETINAL pigment epithelial dystrophy (RPED) a control group of 30 dogs of other and mixed is a degenerative disease of the canine retina breeds. The plasma was not lipaemic and triglycwhich is seen in several breeds of dog in the eride concentrations were within the normal range United Kingdom but, surprisingly, is seldom for all animals. Ophthalmoscopic findings condiagnosed in other parts of the world. The disease sistent with a diagnosis of RPED were identified is characterised ophthalmoscopically by multiple in 17 of the 90 briards but there was no significant light-brown pigment spots within the tapetal fun- difference in plasma cholesterol concentrations dus which represent the accumulation of abnor- between these dogs and the ophthalmoscopically mal, autofluorescent pigment granules within normal briards. No evidence of corneal lipid cells of the retinal pigment epithelium (RPE) deposition was observed in any animal. The (Aguirre and Laties 1976, Lightfoot 1988). Little results of the preliminary plasma biochemistry is known about the nature of this pigment in investigations suggest that briards may have an dogs or the reasons for its development. abnormality of lipid metabolism and the present However, similar 'lipopigments' which develop study was undertaken to explore the nature of in a number of human retinal disorders are this hypercholesterolaemia.

Elevated fasting plasma cholesterol concentrations were identified in clinically healthy briards. Biochemical investigations revealed no other major abnormalities. Plasma lipoprotein electrophoresis demonstrated a marked increase in the intensity of the c~2 band (compared with control dogs) which was reduced by dextran sulphate-magnesium chloride or sodium phosphotungstate-magnesium chloride precipitation of apo B and apo E containing lipoproteins in the plasma. The study has identified a hyperlipidaemia in briards characterised by increased cholesterol but normal triglyceride concentrations. The absence of obvious metabolic changes associated with secondary hypercholesterolaenfia, suggests the breed may have a primary abnormality in cholesterol metabolism. The increased density of the precipitable lipoprotein which migrates to the c~2 band suggests that the hypercholesterolaemia may be due to an abnormal accumulation of high density lipoprotein (HDL) possibly HDLc. The possibility that abnormality in lipid metabolism might play a role in the development of retinal pigment epithelial dystrophy in briards is currently being investigated.

Hypercholesterolaemia in briards

Materials and methods

Samples for urinalysis, haematology, clinical biochemistry (including triglyceride and thyroxine [T4] concentrations) and cholesterol determination were collected from 15 clinically healthy briards, 12 dogs of various breeds matched for age and sex, and a five-year-old female briard with clinical signs of hypothyroidism (lethargy and symmetrical alopecia). Details of age, sex, diet, medical history and fundus features were recorded for each dog. All dogs had been fasted for 12 to 18 hours before sampling. Blood samples for clinical biochemistry and cholesterol determination were collected into heparin, spun immediately and the plasma transferred to the laboratory on ice, but not frozen. Blood samples for haematology and lipoprotein electrophoresis were collected into K-EDTA. All blood samples were inspected after centrifugation for visible lipaemia and again after standing at 4°C for 24 hours. They were stored at 4°C and plasma cholesterol concentrations were measured within 24 hours of sample collection. All other analyses were performed within 72 hours of collection. Urinalysis included a full sediment examination, specific gravity measurement and dipstick evaluation (BM-Test-8; Boehringer). Routine plasma haematological and biochemical analyses were carried out by a commercial veterinary laboratory. Plasma cholesterol concentrations were measured using a commercially available kit (Sigma Diagnostics) which was calibrated against a certified calibrator and were cross-checked by a commercial laboratory. Lipoprotein electrophoresis of plasma samples from the 16 briards and from eight control dogs was performed on an agarose gel using the Paragon system (Beckman). Electrophoresis gels were scanned at 500 nm on a densitometer (Appraise, Beckman). Lipoprotein electrophoresis was repeated following lipoprotein precipitation with dextran sulphate-magnesium chloride. 0.1 ml of a solution of dextran sulphate 50,000 (20 g litre 1), sodium azide (1 g litre-1) and magnesium chloride (200 g litre -1) or 0.1 ml of a solution of sodium phosphotungstate (40 g litre-l) and magnesium chloride (100 g litre -1) (Sigma reagents) was added to 1 mi of test plasma. This was vortexed, left to stand at room temperature for 20 minutes and then centrifuged at 1500 g for 20 minutes. Residual cholesterol concentra-

81

tions were measured in the supernatants and lipoprotein electrophoresis was performed within two hours. Combined thyroid stimulating hormone/ adrenocorticotrophic hormone (TSH/ ACTH) stimulation tests were performed on three of the 15 healthy briards and on the briard bitch with clinical signs of hypothyroidism. TSH (Sigma) at 0-1 iu/kg -1 to a maximum of 5 iu and tetracosactrin (Synacthen; Ciba) at 250 ~tgper dog were injected intravenously. A control sample was taken before injection and samples were collected at one (for cortisol) and six (for T4) hours after injection and analysed by a commercial veterinary laboratory. Normal plasma T4 concentrations were considered by the laboratory to be 13 to 52 nmol litre-1 (basal) and 26 to 104 nmol litre -1 or 1.5 times basal level (after TSH stimulation). Normal resting plasma cortisol concentrations were considered as 20 to 250 nmol litre-1 and less than 660 nmol litre -1 was considered normal one hour after stimulation with ACTH (Serono). Results

Visual inspection of plasma immediately after centrifugation and after 24 hours of refrigeration revealed no evidence of lipaemia in any sample except that from the briard bitch with clinical signs of hypothyroidism. Results of urinalysis, haematology and clinical biochemistry were within normal ranges apart from a significant (P<0.01) increase of plasma cholesterol (mmol litre-1 mean + sE) in the briards (8.0 + 0.5), compared with the 12 control dogs (4-1 + 0-3). Plasma triglyceride concentrations were within the normal range (0.57 to 1.14 mmol litre -1) for these animals. All 15 clinically healthy briards had normal resting total T4 concentrations and the three clinically healthy briards responded normally to the combined TSH/ACTH stimulation test. The briard bitch with clinical signs of hypothyroidism had elevated plasma cholesterol (21 mmol litre-1) and triglyceride (1.7 mmol litre-1) .concentrations. The TSH stimulation test result (resting T4 7.3 nmol litre-1; six hour T4 7.0 nmol litre= 1) was consistent with a diagnosis of hypothyroidism. The result of the ACTH stimulation test was normal. The results of lipoprotein electrophoresis of normal canine plasma in this study (Fig 1) showed

P. Watson, K. Ire". Simpson, P. G. C Bedford

82

PRE"~

o 1

2

3

4

5

6

~

~

,~1

o

ml

0

I~

~

a1

7

FIG 1: Photograph of representative electrophoresis gel. Lanes 1, 3, 4 normal briards. Lane 2, hypothyroid briard. Lanes 5, 6 and 7 control dogs. 0 represents electrophoresis origin

four primary bands ([3, pre-[3, a2 and ~ 1 ) and demonstrated a marked increase in the intensity of the o~2 band in healthy briards compared with the controls. Representative densitometer tracings for the briards and the normal dogs are shown in Fig 2. Minor differences were observed in the lipoprotein profiles for dogs of differing diet, age and sex for both briards and control dogs. These differences were minimised by careful age, sex and diet matching of dogs. The increase in intensity of the c~2 band in the briard samples was reduced equally following either dextran sulphate-magnesium chloride (Fig 3) or sodium phosphotungstate-magnesium chloride precipitation. Residual cholesterol concentrations in the supematants following dextran sulphate-magnesium chloride precipitation, expressed as a percentage of the original concentration were 90.3

j

A

ilia

I]

o~ 51

.

~

_]

FIG 2: Densitometer scans of electrephoresis gels. A, control dog, B, C, normal briards, D, hypothyroid briard

0

~xI

FIG 3: Densitometer scans of electrophoresis gels. Control dog before (A) and after (C) precipitation with dextran sulphate-magnesium chloride. Normal briard before (B) and after (D) precipitation with dextran sulphate-magnesium chloride

+ 0-7 for the 12 control dogs compared with 71.8 + 2.0 (mean + SE) for the 15 briards. Discussion

Increases in serum or plasma cholesterol occur most commonly in dogs secondary to a range of metabolic disorders which include diabetes mellitus, hypothyroidism, hyperadrenocorticism, protein losing nephropathies, cholestatic liver disease and pancreatitis (Rogers et a11975). Primary or idiopathic hypercholesterolaemia is rare and is diagnosed by excluding known metabolic causes of hypercholesterolaemia. The absence of abnormal clinical, haematological and serum biochemical findings in the 15 healthy briards excluded metabolic disorders commonly associated with hyperlipidaemia. Normal T4 concentrations in all the dogs and normal T S H / A C T H stimulation tests in three of them suggest that occult hypothyroidism or hyperadrenocorticism are unlikely to be responsible for the hypercholesterolaemia. The results of these tests suggest that briards may have a primary hyperlipidaemia. A number of authors have shown that normal canine plasma lipids and lipoproteins can be affected by age, diet and sex hormones (Crispin et al 1992) and it has been suggested that there may be differences between pet and laboratory populations of dogs (Schiller et al 1964). The present study minimised these variations by the accurate matching of control and briard samples.

Hypercholesterolaemia in briards

83 Cholesterol is mainly transported in the body described in the current study had normal plasma in its esterified form combined with apoproteins triglyceride concentrations and this may explain and other lipids to form soluble lipoproteins. the difference between lipoprotein electrophoresis In dogs and cats the predominant fasting plasma profiles for these dogs and those of the dogs lipoproteins are the high density lipoproteins described by Rogers et al (1975). There was no (HDL). In these species HDLS are the major evidence on lipoprotein electrophoresis of the cholesterol carriers and in the dog they are sub- shift of all lipoproteins to lower density forms divided into four main classes (HDL1, HDL2, which has been described in dogs maintained on HDL3 and HDLe) of which the first and last may cholesterol rich diets (Bauer 1991). in fact be the s a m e . HDL 2 and HDL 3 a r e assoLipoproteins can be precipitated from plasma ciated with apoprotein A (apo A) while the HDL c by a range of agents, depending on the apoprois thought to be associated with apoprotein E teins which they contain. F o r human plasma (apo E) (J. E. Bauer, personal communication). lipoproteins, dextran sulphate-magnesium chloThis contrasts markedly with man where the ride will precipitate those containing apo B and major carrier lipoprotein is low density lipopro- apo E, but not those containing apo A (Gibson tein (LDL). et al 1984). This precipitant is now used as the The results of lipoprotein electrophoresis of standard in many medical laboratories. Sodium normal canine plasma in this study are consistent phosphotungstate-magnesium chloride has been with those from previous studies in the dog used for canine plasma to precipitate VLDLS and (Rogers et al 1975). The high density lipoproteins, LDLS (Crispin et al 1992) although its effect on whose major apoprotein is apoprotein A migrate HDL c has not been reported. The decrease in the electrophoretically to the ct band. HDL 2 and HDL 3 intensity of the ct2 band after precipitation of which carry most of the cholesterol in the normal lipoproteins containing apo B and apo E but not dog cannot be separated by electrophoresis but those containing apo A suggests that the abnorcan be separated by ultracentrifugation (Crispin mal band may represent an accumulation of et al 1992). HDL1, which migrates electrophoret- apolipoprotein B or E. Further characterisation ically to the t~2 band, is the predominant choles- by ultracentrifugation and apoprotein electerol carrier in the recently fed hypothyroid dog trophoresis will be necessary to confirm this sug(Mahley and Weisgraber 1974). Despite the gestion. Apart from the high incidence of RPEDreported n o m e n c l a t u r e , HDL 1 is probably not a true HDL and cannot be separated from LDL on ultracen- for the breed, briards do not appear to be pretrifugation because their size and density overlap disposed to any particular medical disorders. (Rogers et al 1975). It has also been shown that There was no evidence of the corneal lipid depocanine plasma contains an HDL and a very low sition which has been described in dogs with density lipoprotein (VLDL)molecule which both hypothyroidism (Crispin and Barnett 1978) and contain apoprotein E. This last HDL, usually post mortem examinations of a number of briards t e r m e d HDLc, seems to be induced in dogs fed of varying ages by one of the authors (P.W., high cholesterol diets and it has been suggested unpublished observations) have not revealed evithat it may be a form of HDL 1 (Mahley et al dence ofatherosclerosis or other diseases. In man 1974, Mahley and Innerarity 1977) or of HDL 2 elevated plasma LDL cholesterol is thought to predispose patients to the development of (Zerbe 1986) overloaded with cholesterol. The apparent marked increase in the density atherosclerosis while elevated HDL cholesterol of the t~2 band on lipoprotein electrophoresis of concentrations are thought to have a protective the briard samples suggests that these animals effect (Gordon and Rifkind 1989). At present there is no direct evidence that have an abnormal accumulation of one or more plasma apolipoproteins. The lipoprotein profiles hypercholesterolaemia is directly related to the obtained for the briards were unlike those pre- development of RPED in briard dogs. However, viously reported for dogs with secondary or pri- circumstantial evidence suggests that the two mary hyperlipidaemias (Rogers et al 1975). problems may be linked. The very high incidence Primary hyperlipidaemias in the dog and cat are of clinical ReED (approximately 33 per cent of usually associated with increased plasma trigly- briards over two years old) reported by Bedford ceride concentrations (Zerbe 1986). The briards (1984) has been supported by histological studies

84

P. Watson, K. W. Simpson, P. G. C Bedford

(Lightfoot 1988) which show that almost all aged briards have evidence of RPE lipopigment accumulation. These factors have led to the hypothesis that all briards in the U K may be in some way predisposed to the development of RPED. Previous authors have suggested that R P E D m i g h t be inherited as a dominant trait with variable gene penetrance (Barnett 1969) or as an autosomal recessive trait (P. G. C. Bedford and M. D. Willis, unpublished data). There has been widespread debate over the aetiopathological significance of hyperlipidaemia in human patients with retinitis pigmentosa (Converse et al 1985). Of particular significance to the present study are the findings of Jahn et al (1987) which demonstrated an increased prevalence of an apolipoprotein E isoform in a group of patients with retinitis pigmentosa. This may be of relevance to the finding that briards appear to have an increase in apo E containing HDL. It is noteworthy that briards in the UK with RPED and briards in the USA with briard lipid retinopathy (Riis and Aguirre 1983) both accumulate abnormal, apparently lipid, material within the R P E and have a systemic hyperlipidaemia. It was reported, however, that the briards in the USA primarily had elevations in the LDL fraction. Treatment for hypercholesterolaemia in this breed has not been considered necessary since it has not been shown that the anomaly is in itself detrimental to the general health of the dogs. Abnormality in plasma lipoproteins may have a major impact upon a wide range of physiological processes since they play a central role in many aspects of mammalian metabolism. A number of these are being examined currently and studies involving ultracentrifugation and apoprotein electrophoresis are continuing in order to pinpoint the precise nature of the lipoprotein abnormality o This study has demonstrated a hyperlipidaemia in briards that is characterised by increased plasma cholesterol but normal triglyceride concentrations. The absence of obvious metabolic derangements associated with secondary hypercholesterolaemia, suggests that briards in the UK may have a primary abnormality in cholesterol metabolism. The increased intensity of a precipitable lipoprotein which migrates to the o~2band suggests that the hypercholesterolaemia is due to an abnormal accumulation of HDL, possibly H D L c.

Acknowledgements The work was supported by funding from Nestec Ltd, Switzerland. The authors are grateful to the British Briard Club and Association and their members for their generous cooperation. They also wish to thank Dr Bill Richmond and his staff at St Mary's Hospital, London and Dr Ron Riis of Cornell University, New York.

References AGUIRRE, G. D. & LATIES, A. (1976) Pigment epithelial dystrophy in the dog. Experimental Eye Research 23, 247-256 BARNETI', K. C. (1969) Genetic anomalies of the posterior segment of the canine eye. Journal of Small Animal Practice 10, 451-455 BAUER, J. E. (1991) Proceedings of 9th ACWMForum. New Orleans. pp511-512 BEDFORD, P. G. C. (1984) Retinal pigment epithelial dystrophy (cPRA): A study of the disease in the briard. Journal of Small Animal Practice 25, 129-138 BOULTON, M., MCKECHNIE, N. M., BREDA, J., BAYLY, M. & MARSHALL, J. (1989) The formation of autofluorescent granules in cultured human RYE.Investigative Ophthalmology and Visual Science 30, 82-89 CONVERSE, C. A., MCLACHLAN, T., HAMMER, H. M., PACKARD, C. J. ~t; SHEPHERD, J. (1985) Hyperlipidaemia in retinitis pigmentosa. In Retinal Degeneration: Experimental and Clinical Studies. Ed M. M. LaVial, J. G. Hollyfield and R. E. Anderson. Alan R. Liss (Distributed by J. Wiley & Sons, Chichester) pp63-74 CRISPIN, S. M. & BARNETT, K. C. (1978) Arcus lipoides corneae secondary to hypothyroidism in the alsatian. Journal of Small Animal Practice 19, 127-142 CRISPIN, S. M., BOLTON, C. H. & DOWNS, L. G. (1992) Plasma lipid and lipoprotein profile of working and pet border collies. Veterinary Record 130, 185-186 GIBSON, J. C., RUBINSTEIN, A. & BROWN, W. V. (1984) Precipitation of apo E containing lipoproteins by precipitation reagents for apolipoprotein B. Clinical Chemistry 30, 1784-1788 GORDON, D. J. & RIFKIND, B. M. (1989) High density lipoprotein - - the clinical implications of recent studies. New England Journal of Medicine 321, 1311-1316 JAHN, C. E., OETTE, K., ESSER, A., BERGMANN, K. yon & LEISS, O. (1987) Increased prevalence of apolipoprotein E2 in patients with retiimtis pigmentosa. Ophthalmic Research 19, 285288 LIGHTFOOT, R.M. (1988) Retinal pigment epithelial dystrophy in the dog. PhD thesis, University of London MAHLEY, R. W. & INNERARITY, T. L. (1977) Interaction of canine and swine lipoproteins with the low density lipoprotein receptor of fibroblasts as correlated with heparin/manganese precipitability. Journal of Biological Chemistry 252, 3980-3986 MAHLEY, R. W. & WEISGRABER, K. H. (1974) Canine lipoproreins and atherosclerosis I Isolation and characterisation of plasma lipoproteins from control dogs. Circulation Research 35, 713-721 MAHLEY, R. W., WEISGRABER, K. H. & INNERARITY, T. (1974) Canine lipoproteins and atherosclerosis II. Characterisation of the plasma lipoproteins associated with atherogenic and nonatherogenic hyperlipidaemia. Circulation Research 35, 722-733 NARFSTROM, K., WRIGSTAD, A. & NILSSON, S. E. G. (1989) The briard dog: a new animal model of congenital stationary night blindness. British Journal of Ophthalmology 73, 750-756 RIIS, R.C. & AGUIRRE, G. D. (1983) The briard problem. Transactions of X1V Annual Scientific Programme of American College of Veterinary Ophthalmologists. pp62-75 ROGERS, W. A., DONOVAN, E. F. & KOCIBA, G. J. (1975)

Hypercholesterolaemia in briards Lipids and lipoproteins in normal dogs and in dogs with secondary hypedipoproteinaemia. Journalofthe American VeterinaryMedical Association 166, 1092-1100 SCHILLER, I., BERGLUND, N. E., TERRY, J. R., REICHLIN, R., TRUEHEART, R. E. & COX, G. E. (1964) Hypercholesterolemia in pet dogs. Archives of Pathology 77, 389-392

85

ZERBE, C. A. (1986) Canine hyperlipaemias. In Current Veterinary Therapy IX. Ed R. W. Kirk. Philadelphia, W. B. Sannders. pp 10451053

Received April 3, 1992 Accepted July 13, 1992