Abstract
A proportion of Burmese cats in Australia have an exaggerated post-prandial triglyceride (TG) response after an oral fat tolerance test (OFTT). The aim of this study was to determine (a) whether Burmese cats with presumed lipid aqueous (PLA) had exaggerated post-prandial triglyceridaemia, (b) if Burmese cats with exaggerated post-prandial triglyceridaemia (‘affected’ cats) had decreased lipoprotein lipase (LPL) activity and (c) whether affected cats were more insulin resistant than normal Burmese cats. Of cats with a history of PLA, 4/5 were shown to be lipid intolerant (4 h TG>4.5 mmol/l). Four affected Burmese cats were age, gender and body condition matched to four normal Burmese cats. Serum TG, insulin, non-esterified fatty acids (NEFA), lipoprotein and apolipoprotein concentrations were determined 2 weeks after commencing a standardised high-protein diet, with an OFTT performed 4 weeks later. Affected Burmese cats did not have significantly different fasting insulin, fructosamine, NEFA, apolipoprotein or lipoprotein concentrations compared to control cats. During the OFTT, affected cats had significantly higher 4 h and 6 h serum TG and NEFA concentrations than normal cats. There was a trend for lower LPL activity, higher insulin concentrations (at 4 and 6 h) and higher insulin area under the curve (AUC) during the OFTT in affected Burmese cats compared to controls, although these results failed to reach significance, probably due to the small number of cats studied. Further investigations using larger numbers of cats should focus on reduced LPL activity and insulin resistance as potential causes of delayed TG clearance.
Previously we demonstrated that 28% of randomly selected Burmese cats in Australia exhibited marked post-prandial hypertri-glyceridaemia after an oral fat tolerance test (OFTT) (defined as ‘affected’ cats). 1 This previous study was instigated because Burmese cats are at increased risk for developing lipid aqueous, 1,2 a condition thought to be due to a combination of anterior uveitis 2 and an underlying disorder of lipid metabolism that results in increased serum triglyceride (TG) concentrations. Australian Burmese cats also commonly develop type 2 diabetes, 3–5 although it is unknown whether or not this is a consequence of the same underlying dyslipidaemia.
The degree of post-prandial lipaemia in many species is influenced by both environmental and genetic factors affecting the synthesis and catabolism of chylomicrons (CMs) and very low-density lipoproteins (VLDL). A reduction in lipoprotein lipase (LPL) activity and/or certain genetic variations in apolipoprotein CII (an LPL activator), apolipoprotein E (found on CM and VLDL) 6 and apolipoprotein B concentrations (found mainly on low-density lipoprotein [LDL]) 7 can lead to higher serum TG peak concentrations and delayed TG clearance. Additionally, multiple factors including a high carbohydrate diet, physical inactivity and obesity can also increase post-prandial lipaemia in humans, both directly, and indirectly, resulting in insulin resistance. Insulin resistance may then increase non-esterified fatty acids (NEFA) and VLDL concentrations, as well as reducing LPL activity.8
As many factors may affect post-prandial TG concentrations, this study designed to identify the cause of the post-prandial lipaemia in a representative sample of affected Burmese cats from our previous study. The first specific aim was to ascertain if Burmese cats with naturally occurring lipid aqueous had elevated TG concentrations after a high fat meal. The next was to determine whether Burmese cats had fasting lipoprotein and apolipoprotein concentrations and LPL activity that would explain their exaggerated TG response to an OFTT. These cats were studied in comparison to control Burmese cats matched for age, gender, body condition score (BCS) and diet. Finally, we considered whether a derangement in lipid metabolism could produce insulin resistance, thereby predisposing them to the development of diabetes.
Materials and Methods
Burmese cats with naturally occurring lipid aqueous
Clinical data was obtained from a cohort of Burmese cats with lipid aqueous, by contacting veterinary ophthalmologists and alerting feline practitioners across Australia for referral of cases via a survey in the ‘control and therapy’ series of the Centre for Veterinary Education (June 2008). Data on age of presentation for lipid aqueous, gender, diet, concurrent disease (eg, uveitis) and whether relapse occurred, was collated. All cats had been neutered. Where possible, a 16 h fasting blood sample was collected to determine fasting TG concentrations and an OFTT was performed as described previously.1
Concerning cats with lipid aqueous, no studies were conducted to confirm that the lactescent appearance of the aqueous was due to the presence of lipid. There was no pathological data to confirm the presence of transient uveitis in these cats. The diagnosis of uveitis and the presence of lipid in the aqueous were based on clinical examination alone. For this reason, we use the term presumed lipid aqueous (PLA) subsequently throughout this manuscript.
Burmese cats and the dietary trial
Four affected Burmese cats were age, gender and BCS-matched with control Burmese cats. Cats were characterised as affected or normal based on a preliminary OFTT using the protocol previously described. 1 Body condition score was determined using a nine-point scale (1–3=thin, 4–5=ideal, 6–9=overweight). 9 Feline body mass index (BMI) was calculated using a published formula ([(9th rib (cm)÷0.7062)−LIM (cm)/0.9156]−LIM (cm), where LIM=length from middle of patella to hock), 10 whereby cats with BMI scores>30% were considered overweight. All cats were fed the same diet (Royal Canin Feline Diabetic food), consisting of 18.8% carbohydrate, 49% protein and 12.9% fat on a dry-matter basis and fed twice daily meals for 4 weeks, with amounts that maintained a constant body weight. All cats were deemed healthy based on prior medical history and routine physical examination. The study design was approved by The University of Sydney Animal Ethics Committee.
Experimental design
Body weight and BCS were determined prior to commencing the diet, after 2 weeks, then again at 4 weeks. After a 2-week dietary ‘acclimatisation’ period, a 15 h fasting blood sample (with water freely available during the fasting period) was collected to determine the fasting lipoprotein and apolipoprotein profile, serum insulin concentration, fructosamine concentration and full haematological and biochemical analyses. After a further 2 weeks on this diet, an OFTT was performed (ie, 4 weeks after commencing the diet). Each cat was fasted for at least 15 h prior to blood collection. For the OFTT, baseline serum lipid, insulin and NEFA concentrations were determined at t=0 h. Each cat then received 1.2 g fat/kg body weight using a standardised canned diet (Hill's Prescription Diet canine/feline a/d), consisting of 15.4% carbohydrate, 44.2% protein and 30.4% fat on a dry-matter basis. Blood was collected for determination of serum lipids, insulin and NEFA concentrations at t=4 h and 8 h post-prandially. Six of the eight cats had additional blood collected at 6 h: LPL and hepatic lipase (HL) activities were determined in three normal and three affected Burmese cats at the end of the dietary trial, ie, after cats had been on the same diet for 4 weeks.
Blood collection and handling
Blood was collected from the external jugular vein or cephalic vein using a 23-gauge needle and syringe, or a 23-gauge butterfly needle and syringe, respectively. A total of 4 ml of blood was collected in serum tubes for biochemical analyses, NEFA, fructosamine and insulin concentrations. Following centrifugation for 10 min at 2500×g, serum was inspected for the presence of lipaemia, separated and processed within 12 h of collection. Biochemical analyses were performed and the remaining serum frozen at −80°C for later processing for insulin, fructosamine and NEFA concentrations. A small aliquot of serum was stored at 4°C overnight to determine the presence or absence of CMs. Blood for lipoprotein and apolipoprotein analyses was placed in tubes containing ethylene-diamine-tetra-acetic acid (EDTA). For LPL and HL analyses, 2 ml blood was collected in lithium heparin tubes, at t=0 h and t=10 min after intravenous administration of 70 IU heparin/kg. All tubes and samples were placed on ice and plasma separated by centrifugation at 4°C and then frozen at −80°C prior to shipping to Glasgow on dry ice (Division of Cardiovascular and Medical Sciences, University of Glasgow, G31 2ER, Scotland).
Serum biochemistry, insulin, NEFA and fructosamine
Measurement of all serum analytes was performed on an automated Hitachi-Roche analyser using standard methods. Serum glucose concentrations were determined using a hexokinase method with a Gluco-quant enzyme kit (Roche Diagnostics/Hitachi). Serum insulin was determined by a commercial test kit (Coat-A-Count, DPC), as EDTA plasma can produce falsely elevated results. 11 Serum NEFA and fructosamine were also determined by commercially available kits (NEFA-c Wako and Cobas Roche, respectively).
Lipoprotein and apolipoprotein concentrations
Serum TG and cholesterol concentrations were determined by glycerol-phosphate oxidase (GPO-PAP, Boehringer Mannheim), and cholesterol oxidase and peroxidase (CHOD-PAP, Boehringer Mannheim) kits. Methods for lipoprotein separation were modified from a previous study. 12 Briefly, potassium bromide was added to 1.4 ml plasma and stained with Coomassie Blue, to achieve a density of 1.24 g/ml.13 Separation of VLDL, LDL and high-density lipoprotein (HDL) was performed using a Beckman TL-100 ultracentrifuge in a TLA 100.4 rotor. Plasma was spun at 100,000 rpm (500,000×g) at 15°C for 180 min. Tubes were sliced using a Beckman tube slicer; VLDL-TG and cholesterol were measured in the top fraction, LDL and HDL layers were identified and each removed by aspiration via the tube wall using a 23 g needle and 3 ml syringe. Triglyceride and cholesterol concentrations were measured in both LDL and HDL layers. Apolipoprotein B, CII and CIII were measured using commercial test kits (Apo B-HA, Apo CII-HA and CIII-HA, Wako Chemicals).
LPL and HL activity
Post-heparin lipolytic activity was measured using 14C-labelled TG and gum arabic as a substrate. Free fatty acids released were counted using liquid scintillation. Results were expressed in micromoles of fatty acids released per millilitre of post-heparin plasma per hour of incubation (μmol of FA/ml/h). Total lipase activity and HL activity were determined, and LPL was calculated based on the subtraction of HL from total lipase activity. Detailed methods are described elsewhere.14
Statistical analysis
Differences in biochemical analytes, lipoproteins, fructosamine, insulin and NEFA concentrations between affected and normal Burmese cats were analysed with paired t-tests. Comparisons between affected Burmese and normal Burmese cats were performed using analysis of co-variance. Homeostasis model assessment-estimated insulin resistance (HOMA-IR) was calculated as HOMA-IR=fasting insulin (μIU/ml)×fasting glucose (mmol/l)/22.5. 15 For the OFTT data, serum TG, cholesterol, glucose, NEFA and insulin concentrations were plotted against time, and a residual maximum likelihood (REML) procedure performed. The area under the curve (AUC) and maximum change for each analyte were compared in affected and normal Burmese cats. Skewed data was log-transformed for data analysis, then back-transformed for presentation purposes. Densitometer percentage readings were compared using paired t-tests. Values are expressed as mean±standard deviation (SD) and/or median and inter-quartile range (IQR). P values of ≤0.05 were considered significant. All data analyses were performed using a statistical software package (Minitab 15.1.1, State College, PA, USA). REML analysis was performed using GenStat software (VSN International).
Results
Burmese cats with naturally occurring lipid aqueous
Six Burmese cats which had previously had PLA were recruited, with the owners of five permitting an OFTT to be performed, and only two owners permitting post-heparin LPL activity measurements. All cats were less than 12 months-of-age when they presented with PLA (mean 7.8±1.6 months), five were female and one was male. Five of the six cats had more than one episode of PLA, in the same or contralateral eye. Clinical signs resolved in most cases within 24 h with all resolved within 3 days, and in most cases after topical steroid therapy (dexamethasone 0.1%; Maxidex, Alcon Laboratories). All cats were being fed a combination of commercial kitten dry and wet foods at the time PLA developed, with the exception of one cat, which was consuming low fat commercial pet mince and tinned food.
Results of the OFTT for the five cats tested are illustrated in Fig 1. Following heparin administration, LPL activity in two of these cats was lower than the laboratory reference interval (4.9 and 8.4 μmolFA/ml/h: reference range 11.6–19.4 μmolFA/ml/h). Results for LPL and HL activities are found in Table 1
Fig 1.
Scatterplot of 0 h and 4 h serum TG concentrations during an OFTT, comparing affected and normal Burmese, to Burmese cats with a history of presumed lipid aqueous. There were no significant differences in TG concentration between affected Burmese cats and those with lipid aqueous (P=0.44), but both were significantly different to normal Burmese cats (P<0.05).
Table 1.
Post-heparin lipoprotein lipase and hepatic lipase activity in affected and normal Burmese cats and Burmese cats with presumed lipid aqueous. Results shown as mean±SD.
| Cats | Lipoprotein lipase 11.6–14.94† FA/ml/h | Hepatic lipase | P value* |
|---|---|---|---|
| Normal Burmese (n=3) | 9.2±3.4 | 21.6±8.8 | – |
| Affected Burmese (n=3) | 4.2±1.4 | 12.9±0.9 | 0.14 |
| Lipid aqueous (n=2) | 6.7±2.5 | 21.2±1.8 | 0.25 |
∗P value≤0.05 (if affected cats or cats with lipid aqueous had significantly different LPL activity compared to normal cats).
†Approximate feline reference range.24
Matched Burmese cats
The mean (±SD) age, weight, BCS and BMI of the eight cats were as follows: 5.1±2.5 years, 4.5±0.7 kg, 5.4±0.3 and 25±4%. There were equal numbers of male and female cats in each group. Of the normal Burmese cats, two were chocolate and two were brown, whereas affected Burmese cats comprised two brown, one chocolate and one red cat.
Serum biochemistry
Results from the biochemical analyses performed after the 2-week conditioning period are presented in Table 2. Affected cats had higher serum alkaline phosphatase (ALP) and lipase activities compared to normal cats, although differences were slight and clinically unimportant, (ie, values were well within the laboratory reference intervals). Affected cats had lower fructosamine concentrations than normal Burmese cats although the difference was slight and not significant (P=0.14).
Table 2.
Biochemical analysis, lipoprotein and apolipoprotein analysis performed after a 2-week conditioning period of the trial diet, comparing normal Burmese and affected Burmese cats. Results shown as mean±SD.
| Reference range/units | Normal Burmese (n=4) | Affected Burmese (n=4) | P value* | |
|---|---|---|---|---|
| Fasting TG | 0.1–0.8 mmol/l | 0.35±0.06 | 0.60±0.27 | 0.12 |
| Fasting cholesterol | 2.4–5.2 mmol/l | 3.42±0.88 | 3.90±0.71 | 0.58 |
| Fasting NEFA | <2000 μmol/l | 502±191 | 369±134 | 0.30 |
| Fasting glucose | 3.9–8.3 mmol/l | 5.08±0.94 | 5.50±0.46 | 0.57 |
| Fasting insulin | <7.22 mIU/ml | 5.33±1.55 | 6.63±2.69 | 0.45 |
| HOMA-IR | <1.18 | 1.55±0.93 | 1.93±0.83 | 0.61 |
| Fructosamine | <300 μmol/l | 255±17 | 226±19 | 0.14 |
| Alanine aminotransferase | 1–80 U/l | 53±16 | 76±29 | 0.16 |
| Alkaline phosphatase | <81 U/l | 18±2.8 | 24±4.1 | 0.04 |
| Amylase | <2400 U/l | 882±26 | 1042±186 | 0.20 |
| Lipase | 1–32 U/l | 22±3.2 | 29±4.4 | 0.05 |
| Fasting apoB† | 0.5–1.0 g/l | 0.08±0.02 | 0.05±0.01 | 0.12 |
| Fasting apo CII† | 1.9–4.1 g/l | 0.31±0.20 | 0.28±0.17 | 0.88 |
| Fasting apo CIII† | 5.6–10.2 g/l | 0.12±0.06 | 0.18±0.03 | 0.55 |
| VLDL-TG† | mmol/l | 0.08±0.04 | 0.20±0.03 | 0.23 |
| VLDL-c† | mmol/l | 0.07±0.05 | 0.04±0.12 | 0.50 |
| LDL-TG† | mmol/l | 0.07±0.05 | 0.06±0.02 | 0.63 |
| LDL-c† | mmol/l | 0.70±0.49 | 0.84±0.36 | 0.68 |
| HDL-TG† | mmol/l | 0.01±0.01 | 0.03±0.03 | 0.45 |
| HDL-c† | mmol/l | 1.77±0.30 | 1.95±0.40 | 0.52 |
∗P value≤0.05 (affected cats are significantly different to normal cats).
†Human reference range.
‡No published reference range for cats.
Lipids, lipoprotein and apolipoprotein concentrations
Affected cats had higher fasting serum TG concentrations compared to normal cats at both 2 and 4 weeks after commencing the high-protein diet (P=0.12, P=0.24, respectively), although the differences were not significant. There were no significant differences in fasting lipoprotein and apolipoprotein concentrations between affected and normal cats (Table 2). There was a trend towards higher VLDL-TG concentrations (0.20 vs 0.08 mmol/l; P=0.23) and lower apolipoprotein B concentrations (0.05 vs 0.08 g/l; P=0.12) compared to normal cats. However, these differences were not significant.
Serial changes in serum TG and cholesterol concentrations during the 4-week OFTT are presented in Fig 2. As expected, affected cats had significantly higher (P<0.05) t=4 h and 6 h mean TG concentrations (4 h: 7.94 mmol/l; 6 h: 7.81 mmol/l) than normal cats (4 h: 2.02 mmol/l; 6 h: 1.10 mmol/l). The AUC for serum TG and cholesterol for the OFTTs were significantly higher in affected Burmese cats (TG P=0.03; cholesterol P=0.05), although fasting serum cholesterol concentrations were not significantly different at each time-point between the two groups.
Fig 2.
Individual TG and cholesterol curves during an OFTT after 4 weeks. Affected Burmese cats are shown as red lines, normal Burmese cats as blue lines. Four and 6 h TG concentrations were significantly higher in affected cats (P<0.05).
Critically, there were no significant differences in 0 h and 4 h in serum TG concentrations between the original screening OFTT in both the affected cats and the normal cats, to that obtained after 4 weeks on a high-protein, low carbohydrate diet (Table 3).
Table 3.
Results of the OFTT performed after 4 weeks, comparing normal Burmese and affected Burmese cats. Mean 0 h and 4 h TG performed prior to enrolling in this study also shown, ie, the screening OFTT. Results shown as mean±SD.
| Normal Burmese (n=4) | Affected Burmese (n=4) | P value* | |
|---|---|---|---|
| Mean 0 h TG (screening) | 0.38±0.15 | 0.56±0.17 | 0.22 |
| Mean 0 h TG (4 weeks)† | 0.38±0.09 | 0.87±0.71 | 0.24 |
| Mean 4 h TG (screening) | 1.98±1.52 | 7.85±3.37 | 0.02 |
| Mean 4 h TG (4 weeks)† | 2.02±0.20 | 7.94±3.30 | 0.03 |
| Fasting cholesterol (mmol/l) | 3.55±0.84 | 3.73±0.66 | 0.82 |
| Fasting NEFA (μmol/l) | 553±417 | 481±204 | 0.77 |
| Fasting glucose (mmol/l) | 5.10±0.94 | 5.45±0.47 | 0.58 |
| Fasting insulin (mIU/ml) | 6.35±2.80 | 7.25±3.42 | 0.70 |
| TG AUC (mmol/l) | 5.40±0.78 | 37.5±16.4 | 0.03 |
| Cholesterol AUC (mmol/l) | 17.6±3.54 | 24.7±4.31 | 0.05 |
| NEFA AUC (μmol/l) | 1999±953 | 4011±840 | 0.02 |
| Glucose AUC (mmol/l) | 24.7±7.96 | 29.6±2.86 | 0.30 |
| Insulin AUC (mIU/ml) | 30.4±8.85 | 58.3±20.2 | 0.06 |
AUC=area under the curve.
∗P value≤0.05.
†No significant difference between screening and 4-week OFTT.
LPL and HL activity
Lipoprotein lipase and HL activities were measured in six Burmese cats (three affected and three normal cats) and compared with results from two Burmese cats with PLA. There was a trend for affected cats to have lower LPL activity (2.6, 4.6 and 5.5 μmolFA/ml/h) compared to normal cats (5.5, 10.1 and 12.2 μmolFA/ml/h), although the difference was not significant (P=0.14). The two Burmese cats with PLA had comparable LPL activity (4.9 and 8.4 μmolFA/ml/h) to affected and control Burmese cats. Hepatic lipase activity was not significantly different between any of the groups of Burmese cats.
Serum insulin, NEFA and fructosamine concentrations
The mean fasting serum insulin concentration and the HOMA-IR ratio both exceeded the reference interval in affected cats, although values did not significantly differ from control Burmese cats. There was no significant difference in fasting NEFA concentrations between the two groups (P=0.30). Affected Burmese cats had lower serum fructosamine concentrations than control cats, although the difference was not significant (P=0.14).
Serial changes in serum insulin, NEFA and glucose concentrations were plotted in Fig 3. The t=4 h and 6 h NEFA were significantly higher (P<0.05) in affected cats (4 h: 660 μmol/l; 6 h: 779 μmol/l) than control Burmese cats (4 h: 402 μmol/l; 6 h: 318 μmol/l). Serum insulin was higher at t=4 h and 6 h in affected cats (4 h: 9.32 mIU/ml; 6 h: 14.0 mIU/ml) compared to control cats (4 h: 5.90 mIU/ml; 6 h: 5.25 mIU/ml) although the difference was not significant. The AUC for NEFA was significantly higher in affected Burmese cats (NEFA P=0.02), and there was a trend for affected cats to have higher insulin AUC (P=0.06).
Fig 3.
Individual NEFA, insulin and glucose curves during an OFTT after 4 weeks. Affected Burmese cats are shown as red lines, normal Burmese cats as blue lines. Four and 6 h NEFA concentrations were significantly higher in affected cats (P<0.05).
Discussion
This study continues preliminary work concerning the TG response during an OFTT in Burmese and non-Burmese cats. 1 The earlier study identified a subset of Burmese cats with a substantial post-prandial triglyceridaemia (TG>6.0 mmol/l) 4 h after eating. No trends were identified concerning age, gender, coat colour, dietary intake or BCS in these affected cats. As disorders of lipid metabolism causing hypertriglyceridaemia have been associated with glucose intolerance and diabetes in humans, 16,17 it was originally postulated that cats with a history of PLA may have been predisposed to the development of type 2 diabetes. 1,2
The current investigation first recruited a representative number of cats with spontaneous PLA who underwent an OFTT. Four of the five cats were lipid intolerant (4 h TG>4.5 mmol/l) at the time of testing, as might have been expected. The next part of the study examined a small, carefully selected cohort of affected and normal Burmese cats matched for age, gender, condition and diet. This enabled us to ascertain whether there was a difference in serum lipoproteins, apolipoproteins, LPL activity or insulin resistance markers that might explain the observed difference in TG response to an OFTT.
Prior to commencing this study, all Burmese cats were fed various veterinary formulations of dry and/or wet foods, with a relatively high glycaemic index and low protein content compared to fresh meat. Studies in humans have demonstrated that high carbohydrate diets are more likely to produce post-prandial lipaemia, 18–20 therefore, the standardised diet chosen in this study was lower in carbohydrate and higher in protein, to determine whether this might reduce post-prandial triglyceridaemia in affected cats. However, on average, there was no significant difference when comparing the initial screening OFTT to the one performed at 4 weeks; one of the four affected cats had a higher 4 h TG concentration at the 4 week OFTT, whereas post-prandial triglyceridaemia was lower in the remaining three cats. Of the four normal cats, three had similar 4 h TG concentrations between both OFTTs; one cat had higher 4 h TG concentrations at 4 weeks. These results suggest that either (a) this change in diet composition had minimal effect in relation to OFTT testing or (b) a change in dietary composition requires a longer period to impact on fat tolerance.
The best-characterised inherited dyslipidaemia in cats to date is an autosomal recessive disorder resulting in an LPL deficiency. 21 Individuals homozygous for this condition develop severe fasting and post-prandial triglyceridaemia and the diagnosis is confirmed by the presence of exceedingly low post-heparin LPL activity. Heterozygous individuals have a more subtle defect, and are often only distinguishable from normal cats after conducting an OFTT. 22 One of the hypotheses of the current study, was that affected Burmese cats had reduced LPL activity as the explanation of their delayed TG clearance. In the current study, there was a trend towards lower LPL activity in affected Burmese cats, with apolipoprotein CII (an activator of LPL) and apolipoprotein CIII (an LPL inhibitor), having similar activities in both groups of cats.
The dyslipidaemia which underlies lipid aqueous is likely an inherited disorder. From the limited pedigree analysis performed in this study (Figs 4 and 5), it is probably not due to a dominant gene, as both parents of a Burmese cat with PLA (cat 17) (see Fig. 5), had neither a history of this disorder nor an abnormal OFTT. It is clear that further work is needed to determine the phenotypes from more sires and dams of affected Burmese cats and those with PLA.
Fig 4.
Genetic relationship between affected and normal Burmese cats. Number=number assigned for the pedigree analysis, □=male, ○=female. Open symbols are normal cats. Filled symbols are affected cats. Symbols with one line through are cats that were not tested. Hatched symbols are cats classed as intermediate. 1 REP are cats repeated on the pedigree diagram.
Fig 5.
Genetic relationship between affected and normal Burmese cats, and those with presumed lipid aqueous and LPL deficiency. Number=number assigned for the pedigree analysis where ×=unknown, □=male, ○=female. Open symbols are normal cats. Filled symbols are affected cats. Symbols with one line through are cats that were not tested. Hatched symbols are cats classed as intermediate. 1 Cross hatched symbols are cats with lipid aqueous. Half-filled symbol is a cat with LPL deficiency (155); note that this cat is the offspring from cat 146 and 134. REP are cats repeated on the pedigree diagram.
Interpretation of LPL activity is challenging, due to inconsistencies in methodology between studies. Furthermore, a ‘normal’ reference interval has not been accepted for cats. The current study used similar methodology and the same research laboratory as the only comparable study published to date. 23 Watson found normal cats had LPL activities between 11.6 and 14.94 μmolFA/ml/h, cats homozygous for a defective LPL gene had activities of approximately 5.85 μmolFA/ml/h, and relatives (likely heterozygous) had LPL activity between 6.68 and 9.01 μmolFA/ml/h. 23 It is, therefore, possible that reduced LPL activity contributed to delayed TG clearance in affected Burmese cats. Greater numbers of LPL values for both affected Burmese and normal cats are required to confirm this.
Unfortunately, results from apolipoprotein E phenotyping did not eventuate, likely due to insufficient cross-reactivity between feline and rabbit anti-human antibodies. As variation in this apolipoprotein can influence post-prandial lipaemia, further work is required to develop a feline apolipoprotein E assay to evaluate its role in feline lipid metabolism. In people, more than 90% of apolipoprotein B is found in LDL lipoproteins, the remainder carried in VLDL (and CM, if present). 24 Demacker demonstrated this may also be true for feline plasma. 25 In the current study, affected cats had apolipoprotein B concentrations that were not significantly different from normal Burmese cats. A previous study in cats demonstrated lower reactivity against anti-human apolipoprotein AI, B and E25; this may explain the lower apolipoprotein B concentrations and the lack of results from apolipoprotein E phenotyping in this study cohort.
Although insulin concentrations increase in patients with peripheral insulin resistance, the normal post-prandial actions of insulin are reduced. This in turn leads to increased hepatic VLDL-TG production and fatty acid release from adipose tissue, and decreased adipose LPL activity. 8 Lederer identified a cohort of apparently normal Burmese cats with higher fasting glucose concentrations and lower glucose tolerance, suggesting reduced insulin sensitivity in a population of Burmese cats. 26 In the current study, inherited insulin resistance may explain why the 4 h and 6 h insulin and NEFA concentrations in the majority of affected Burmese cats were higher than in normal Burmese cats. Fasting insulin concentrations and the HOMA-IR were not significantly different between the two groups of cats in the current study, although fasting insulin was significantly higher in affected cats when considered together with results from the previous investigation. Persistently elevated NEFA concentrations may eventually lead to peripheral and hepatic insulin resistance in humans, as fatty acids are toxic to pancreatic β-cells. 17 One could, therefore, hypothesise that Burmese cats with an inherited disorder of lipid metabolism could develop secondary insulin resistance due to persistently elevated TG and NEFA concentrations. However, in a recent study, glucose-infused cats but not lipid-infused cats displayed β -cell dysfunction resulting in decreased insulin secretion during an IVGTT.27
This study is not without limitations. Firstly, serum glucose, insulin and NEFA concentrations were not measured more frequently. Glucose and insulin concentrations peak at approximately 1–2 h after a meal, and NEFA concentrations reach their nadir at a similar time-point, then subsequently increase. 28 The main aim in this study was to compare the difference in post-prandial lipaemia between the two groups of cats, so time-points of greatest interest were 4, 6 and 8 h. Future studies could be specifically aimed at performing an IVGTT to determine glucose, insulin and NEFA concentrations at short-term time-points for these analytes. Secondly, the sample size in this study was less than ideal. Many of the affected Burmese cats screened originally were not available for further studies because owners declined further participation. One affected cat developed severe diarrhoea after starting the new diet and was thus excluded. An original aim was to measure LPL activity in many more Burmese and control cats; however, the majority of owners declined to have their cat subjected to an injection of heparin. However, despite relatively low patient numbers, we believe this study provides useful preliminary information to frame future investigations on TG metabolism and insulin resistance in Burmese cats.
In conclusion, the results from this study confirm that affected Burmese cats have significantly higher post-prandial TG response compared to age, gender and BCS-matched normal Burmese cats, and that this is largely a genetic phenomenon and unrelated to the recent dietary history. From the limited analysis of lipoprotein/apolipoprotein concentrations, LPL activity, NEFA and insulin concentrations, there is no compelling single explanation to why some Burmese cats have a delayed TG clearance after an OFTT. Affected Burmese cats did on average have lower post-heparin LPL activity and higher post-prandial serum NEFA and insulin concentrations compared to normal Burmese cats; however, larger study numbers are required to examine these effects more closely. As the TG response during an OFTT was similar in Burmese cats that had PLA, and in affected Burmese cats, it is possible that both groups have the same genetic defect.
Acknowledgements
This study was financially supported by the Waltham Foundation, UK. Thanks to Grace Stewart and the laboratory staff at the Glasgow Royal Infirmary, Glasgow; to Professor Roland Stocker and Dr Cacang Suarna at the Vascular Research Laboratory for technical advice; to Francesca Volpato and Jennifer Burns at the Protein and Lipid Department, Royal Prince Alfred Hospital, for performing the NEFA and fructosamine assays; to Dr Navneet Dhand at the Faculty of Veterinary Science, The University of Sydney, for statistical input; and Richard Malik is supported by the Valentine Charlton Bequest of the Centre for Veterinary Education at the University of Sydney.
References
- 1.Kluger E.K., Hardman C., Govendir M., et al. Triglyceride response following an oral fat tolerance test in Burmese cats, other pedigree cats and domestic crossbred cats, J Feline Med Surg 11, 2009, 82–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hardman C., O’Brien CR, Stanley RG. Lipid aqueous as a sign of hyperlipidaemia in Burmese cats. Annual American College of Veterinary Ophthalmologists; 1999. [Google Scholar]
- 3.Swinney G.. Diabetes mellitus in the cat: a retrospective study 1984–1994 at Sydney University Teaching Hospital. Australian Veterinary Association Conference Proceedings, Melbourne, 1995: 84.
- 4.Rand J.S., Bobbermien L.M., Hendrikz J.K., Copland M. Over representation of Burmese cats with diabetes mellitus, Aust Vet J 75, 1997, 402–405. [DOI] [PubMed] [Google Scholar]
- 5.Baral RM, Rand JS, Catt MJ, Farrow HA. Prevalence of feline diabetes mellitus in a feline private practice. 21st Annual Forum of the American College of Veterinary Internal Medicine Conference; Charlotte NC, United States, 2003.
- 6.Brummer D., Evans D., Berg D., Greten H., Beisiegel U., Mann W.A. Expression of type III hyperlipoproteinaemia in patients homozygous for apolipoprotein E-2 is modulated by lipoprotein lipase and postprandial hyperinsulinemia, J Mol Med 76, 1998, 355–364. [DOI] [PubMed] [Google Scholar]
- 7.Lopez-Miranda J., Ordovas J.M., Ostos J.A., et al. Dietary fat clearance in normal subjects is modulated by genetic variation at the apolipoprotein B gene locus, Arterioscler Thromb Vasc Biol 17, 1997, 1765–1773. [DOI] [PubMed] [Google Scholar]
- 8.Jeppesen J., Hollenbeck C.B., Zhou M.Y., et al. Relation between insulin resistance, hyperinsulinemia, postheparin plasma lipoprotein lipase activity and postprandial lipemia, Arterioscler Thromb Vasc Biol 15, 1995, 320–324. [DOI] [PubMed] [Google Scholar]
- 9.LaFlamme D. Development and validation of a body condition score system for cats: a clinical tool, Feline Pract 25, 1997, 13–18. [Google Scholar]
- 10.Elliott D.A. Disorders of metabolism. Nelson R.W., Couto C.G. Small animal internal medicine, 3rd edn, 2003, Mosby: Oxford, 818–819. [Google Scholar]
- 11.Lutz T.A., Rand J.S. Comparison of five commercial radioimmunoassay kits for the measurement of feline insulin, Res Vet Sci 55, 1993, 64–69. [DOI] [PubMed] [Google Scholar]
- 12.Martin GJ. The clinical management of feline diabetes. PhD thesis, The University of Queensland, Australia, 2002.
- 13.Sattler W., Mohr D., Stocker R. Rapid isolation of lipoproteins and assessment of their peroxidation by high performance liquid chromatography postcolumn chemiluminescence, Methods Enzymol 233, 1994, 469–489. [DOI] [PubMed] [Google Scholar]
- 14.Watson T., Butterwick R.F., McConnell M., Markwell P.J. Development of methods for analysing plasma lipoprotein concentrations and associated enzyme activities and their use to measure the effects of pregnancy and lactation in cats, Am J Vet Res 56, 1995, 289–297. [PubMed] [Google Scholar]
- 15.Appleton D.J., Rand J.D., Sunvold G.D. Basal plasma insulin and homeostasis model assessment (HOMA) are indicators of insulin sensitivity in cats, J Feline Med Surg 7, 2005, 183–193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Mingrone G., Henriksen F.L., Greco A.V., et al. Triglyceride-induced diabetes associated with familial lipoprotein lipase deficiency, Diabetes 48, 1999, 1258–1263. [DOI] [PubMed] [Google Scholar]
- 17.Frayn K.N. Insulin resistance and lipid metabolism, Curr Opin Lipidol 4, 1993, 197–204. [Google Scholar]
- 18.Jeppesen J., Chen Y.D., Zhou M.Y., Wang T., Reaven G.M. Effect of variations in oral fat and carbohydrate load on postprandial lipemia, Am J Clin Nutr 62, 1995, 1201–1205. [DOI] [PubMed] [Google Scholar]
- 19.Parks E., Krauss R., Christiansen M., Neese R., Hellerstein M. Effects of a low-fat, high-carbohydrate diet on VLDL–triglyceride assembly, production and clearance, J Clin Invest 104, 1999, 1087–1096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Parks E., Hellerstein M. Carbohydrate-induced hypertriacylglycerolemia: historical perspective and review of biological mechanisms, Am J Clin Nutr 71, 2000, 412–433. [DOI] [PubMed] [Google Scholar]
- 21.Jones B.R., Johnstone A.C., Cahill J.I., Hancock W.S. Peripheral neuropathy in cats with inherited primary hyperchylomicronaemia, Vet Rec 119, 1986, 268–272. [DOI] [PubMed] [Google Scholar]
- 22.Ginzinger D.G., Clee S.M., Dallongeville J., et al. Lipid and lipoprotein analysis of cats with lipoprotein lipase deficiency, Eur J Clin Invest 29, 1999, 17–26. [DOI] [PubMed] [Google Scholar]
- 23.Watson T.D.G., Gaffney D., Mooney C.T., Thompson H., Packard C.J., Shepherd J. Inherited hyperchylomicronaemia in the cat: Lipoprotein lipase function and gene structure, J Small Anim Pract 33, 1992, 207–217. [Google Scholar]
- 24.Bhatnagar D., Durrington P.N. Measurement and clinical significance of apolipoproteins A-I and B. Rifai N., Warnick G., Dominiczak M.H. Handbook of lipoprotein testing, 2000, AACC Press: Washington, 287–310. [Google Scholar]
- 25.Demacker P.N.M., van Heijst P.J., Hak-Lemmers H.L.M., Stalenhoef F.H. A study of the lipid transport system in the cat, Felix domesticus, Atherosclerosis 66, 1987, 113–123. [DOI] [PubMed] [Google Scholar]
- 26.Lederer R., Rand J., Morton J. Fasting glucose concentrations are higher and glucose tolerance is lower in Burmese cats compared to matched non-Burmese cats, J Vet Int Med 19, 2005, 462–463. [Google Scholar]
- 27.Zini E., Osto M., Franchini M., et al. Hyperglycaemia but not hyperlipidaemia causes beta cell dysfunction and beta cell loss in the domestic cat, Diabetologia 52, 2009, 336–346. [DOI] [PubMed] [Google Scholar]
- 28.Jackson K.G., Wolstencroft E.J., Bateman P.A., Yaqoob P., Williams C.M. Acute effects of meal fatty acids on postprandial NEFA, glucose and apo E response: implications for insulin sensitivity and lipoprotein regulation?, Br J Nutr 93, 2005, 693–700. [DOI] [PubMed] [Google Scholar]