Apo E Allele Frequency in Primary Endogenous Hypertriglyceridemia (Type IV) with and without Hyperapobetalipoproteinemia

Transcription

1 Apo E Allele Frequency in Primary Endogenous Hypertriglyceridemia (Type IV) with and without Hyperapobetalipoproteinemia Suzanne Lussier-Cacan, Daniel Bouthillier, and Jean Davignon Apolipoprotein E polymorphism is responsible for the existence in the population of six apo E phenotypes determined by three alleles acting at a single gene locus. We have previously reported an enrichment in the E2 allele and the E2-bearing phenotypes in an unselected sample of subjects with primary hyperlipidemia consisting mainly of endogenous hypertriglyceridemia (Type IV). A study was carried out on 214 Type IV hypertriglyceridemic subjects to determine whether there was the same distribution in subjects with hyperapobetalipoproteinemia as in those without. The study showed that the relative enrichment in the e2 allele was associated only with Type IV subjects without hyperapobetalipoproteinemia. Since hyperapobetalipoproteinemia is a presumed marker for familial combined hyperlipidemia (FCHL), this finding may provide further evidence that FCHL and familial hypertriglyceridemia, both associated with a Type IV lipoprotein pattern, are truly separate disease entities. (Arteriosclerosis 5: , November/December 1985) There are three major isoforms of apolipoprotein E (apo E), a protein constituent of plasma very low density lipoprotein (VLDL): E2, E3, and E4, which are under the control of three independent alleles (e2, e3, e4) acting at a single gene locus. As a result, six major apo E phenotypes occur in the population: three homozygotes, E4/4, E3/3 (the most frequent), and E2/2, and three heterozygotes, E4/3, E3/2 and E4/2. 1 The phenotype E2/2 is associated with Type III hyperlipidemia, 23 while a higher prevalence of apo E4 has been reported in Type V hyperlipoproteinemia. 4 In the course of our own studies on the polymorphism of apo E, we noted that the s2 allele predisposes to higher triglyceride levels and lower low density lipoprotein (LDL) cholesterol concentrations, whereas the e4 allele is associated with higher From the Department of Lipid Metabolism and Atherosclerosis Research, Clinical Research Institute of Montreal and the Faculty of Medicine, University of Montreal, Montreal, Quebec, Canada. This work was supported by the Quebec Heart Foundation, Health and Welfare Canada (Grant ), and the Medical Research Council of Canada (Grant MT-5427). Daniel Bouthillier is the recipient of a studentship of the Fonds de la Recherche en Sante du Quebec. Address for reprints: Dr. Suzanne Lussier-Cacan, Clinical Research Institute of Montreal, 110 Pine Avenue West, Montreal, Quebec, Canada H2W 1R7. Received January 25, 1985; revision received June 12, cholesterol and LDL cholesterol levels. 5 " 7 Of particular significance was the observation that the e4 allele raises LDL cholesterol in normolipidemic individuals. 6 ' 7 Two independent studies 8 ' 9 have reported an increased frequency of the allele e4 in "hypercholesterolemia" defined as cholesterol levels above 280 mg/dl 8 or 260 mg/dl 9 and normal triglycerides. We further reported 6 an enrichment in the apo E2-bearing phenotypes in an unselected hyperlipidemic population consisting mainly (76%) of individuals with the Type IV lipoprotein phenotype (endogenous hypertriglyceridemia, hyperprebetalipoproteinemia). Endogenous hypertriglyceridemia represents a group of heterogeneous disease entities with different etiologies and a varied degree of associated risk for atherosclerosis. 10 Two major hereditary disease entities harboring a Type IV phenotype may be considered, one'with an associated elevation of LDLapo B levels and believed to represent familial combined hyperlipidemia (FCHL) which is a multiple lipoprotein-phenotype disease; the other, familial hypertriglyceridemia (FHTG), in which LDL-apo B levels are normal. 11 " 13 In this report, we examine the apo E allele frequency distribution of Type IV hypertriglyceridemic individuals to see whether the enrichment in the e2 allele we previously observed is found equally in patients with and without hyperapobetalipoproteinemia.

2 640 ARTERIOSCLEROSIS VOL 5, No 6, NOVEMBER/DECEMBER 1985 Subjects Methods Subjects with endogenous hypertriglyceridemia were selected from among the new untreated patients referred to our lipid clinic between November 1982 and November Informed consent was obtained and the study protocol was approved by the institutional Ethics Committee. Included in the study were 214 individuals fulfilling the following criteria: presence after a 12-hour fast of triglyceride levels between 200 and 1000 mg/dl, a prebetalipoprotein on paper electrophoresis, VLDL cholesterol >35 mg/dl and LDL-C <190 mg/dl, absence of chylomicrons and (3-VLDL in the fasting state, and absence of secondary causes of hyperiipidemia. Type IV patients were then separated into two groups according to their LDL-apo B levels. Group 1 had LDL-apo B >125 mg/dl (above the 95th percentile of our control population) and Group 2, ==104 mg/dl (below the 90th percentile of our controls). Group 1 was designated as having hyperapobetalipoproteinemia according to the criteria defined by Sniderman et al. 14 (LDL-C <200 mg/dl, LDL-apo B >2 SD above normal mean). Patients with intermediate levels (n = 49, LDL-apo B >104 <125 mg/dl) were excluded from the study. Normolipidemic controls were 102 civil servants from Ottawa, Canada, who were participating in a health survey program. 6 ' 7 Their cholesterol and triglyceride levels were below 240 mg/dl and 150 mg/dl respectively. The unselected hyperlipidemic sample (102 consecutive patients referred to our lipid clinic) which led to our original observations 8 was included in this study for comparison. Procedures In the morning after the patients had fasted for 12 to 14 hours, blood was collected from an antecubital vein into Vacutainer tubes (Becton-Dickinson) containing NajEDTA (1 mg/ml). Sodium azide (0.02%) was added to the separated plasma. Cholesterol 15 and triglycerides 16 were measured enzymatically on an automated analyzer (ABA-100 bichromatic analyzer, Abbott Laboratories, Pasadena, California). The plasma lipoprotein profile was examined by paper electrophoresis according to the method of Lees and Hatch. 17 Lipoproteins were separated for determination of VLDL, LDL, and HDL cholesterol by ultracentrifugation at a density = g/ml and heparin-manganese precipitation. 18 Total and LDL-apo B concentrations were measured by electroimmunoassay (EIA) on total plasma and on the d >1.006 ultracentrifugal fraction, essentially by the technique of Reardon et al. 19 VLDL-apo B was obtained by subtracting the LDL-apo B from the total plasma value. An LDL fraction isolated between densities and g/ml was used for standardization with the addition of lipoprotein-deficient serum (d >1.21 g/ml) as suggested by Rosseneu et al. 20 In each assay, a commercial serum of known apo B concentration (Behringwerke, Marburg, West Germany) was used as a reference standard. The apo E phenotype was obtained by isoelectric focusing of the soluble apoproteins of VLDL on a single polyacrylamide gel. 5 Results Mean values (± SD) for plasma lipid, lipoprotein cholesterol, and apo B levels are given in Table 1 for normal controls and for the different groups of subjects with hyperiipidemia. The two groups of Type IV individuals (Group 1 with and Group 2 without hyperapo B) were different from the point of view of total triglyceride and cholesterol levels, Group 1 having higher cholesterol, and Group 2, higher triglycerides. The cholesterol difference was accounted for by LDL cholesterol which, in Group 1, was considerably higher than in Group 2 ( + 73%). The difference in LDL-apo B levels was of a similar magnitude. There Table 1. Plasma Liplds, Lipoproteins, and Apollpoprotein B in Normollpldemic and Hyperlipidemic Subjects Normal Unselected. Type IV hyperlipidemics* subjects hyperlipidemics Group 1 Group 2 Group 1a Group 2a Number (males) Age (yrs, mean ± so) Total cholesterol (mg/dl) Total triglycerides (mg/dl) VLDL-cholesterol (mg/dl) VLDL-apo B (mg/dl) LDL-cholesterol (mg/dl) LDL-apo B (mg/dl) HDL-cholesterol (mg/dl) 102 (55) 36.1 ± ±29 74 ±30 24±13 12±10 103±27 83 ±21 47± (70) 45.9 ± ±83 506± ±24 ND* 132±65 ND 33±10 93 (65) 47.0± ±37 384±176t 82±36 68±38 152±20t 144±16t 32±9 72 (57) 43.7 ± ±36f 501±235t 94 ±41 82±43 88±22t 82±14t 29±9 72 (47) 49.4 ± ±31f 306 ±79 68±21 54±23 157±18f 146±15f 33±8 38 (30) 45.6 ± ±28t 309 ±89 65±21 51 ±24 98±22t 85±15f 33±9 'Groups 1 and 1a = LDL-apo B>125; Groups 2 and 2a = LDL-apo B<104. Subjects with triglycerides >500 mg/dl were excluded from Groups 1a and 2a. tgroups 1 vs 2 and 1a vs 2a were significantly different at p < determined.

3 APO E IN TYPE IV HYPERTRIGLYCERIDEMIA Lussier-Cacan et al. 641 Table 2. Apo E Phenotype and Allele Distribution In Normollpldemic and Hyperlipldemic Subjects Type IV hyperlipidemics* Normal Unselected subjects hyperlipidemics Group 1 Group 2 Group 1a Group 2a Phenotypes Apo E-4/4 Apo E-3/3 ApoE-2/2 ApoE-4/3 Apo E-3/2 ApoE-4/2 Total n (%) Alleles e4 e3 e2 4 (3.9) 63 (61.7) 2 (2.0) 21 (20.6) 10 (9.8) 2 (2.0) 102 (100.0) t 50 (49.0) 3 (2.9) 24 (23.5) 18 (17.7) 7 (6.9) 102 (100.0) (2.1) 53 (57.0) 21 (22.6) 11 (11.8) 6 (6.5) 93 (100.0) (5.55) 36 (50.0) 10 (13.9) 18 (25.0) 4 (5.55) 72 (100.0) f "Groups 1 and 1a = LDL-apo B >125; Groups 2 and 2a = LDL-apo B <104. tgroup 2 vs normals = p < 0.04; Group 2a vs Group 1a and normals = p < (1.4) 42 (58.3) 16 (22.2) 8 (11.1) 5 (7.0) 72 (100.0) (5.3) 16 (42.1) 7 (18.4) 12 (31.6) 1 (2.6) 38 (100.0) f was more fluctuation in VLDL cholesterol within both groups, but HDL cholesterol levels were not significantly different. Thus, we have two heterogeneous samples that nonetheless meet the same basic criteria for Type IV hyperlipidemia. Another major difference between these two Type IV groups was noted in the apo E phenotype distribution (Table 2). There was enrichment in the e2 allele at the expense of the e3 allele in Group 2 (Group 2 vs normals, x 2 = 11-39, p <0.04) as was previously observed in our unselected hyperlipidemics, 6 and this was evident at all triglyceride levels (not shown). Conversely, the allele distribution of Group 1 was the same as in normal individuals. Because Group 2 included more individuals with marked hypertriglyceridemia (>500 mg/dl) than Group 1, we decided to look only at individuals whose triglyceride levels were between 200 and 500 mg/dl. These subgroups are identified as Groups 1 a and 2a in the tables. The two reduced samples had identical triglycerides and VLDL and differed only in their LDL cholesterol and apo B levels (Table 1). The apo E allele distribution difference noted above was also observed in Group 2a and was statistically significant when compared both to Group 1a and to the normal group of individuals (x 2, p <0.05; Table 2). The heterogeneity found in Groups 2 and 2a was due to an excess of the genotype E3/2. Discussion Apo E plays a role in the catabolism of triglyceriderich lipoproteins, probably by its interaction with a hepatic receptor. 21 In subjects with Type III hyperlipoproteinemia who have a defective remnant catabolism, 22 the E2/2 phenotype is usually found. In this condition, apo E2-bearing triglyceride-rich lipoproteins are catabolized more slowly and accumulate in the plasma as (3-VLDL. 23 Even though in our experience (n = 60, unpublished observations) and in that of others, 3 86% to 88% of patients diagnosed as having Type III dysbetalipoproteinemia harbor the E2/2 phenotype, it has been reported that most E2/2 subjects are normolipidemic. Since the receptorbinding defect of apo E2 has been demonstrated in hypo- and normolipidemic, as well as in hyperlipidemic E2/2 subjects, 24 it is surmized that one or more additional factors are necessary for hyperlipidemia to develop in this phenotype. Indeed, one of these factors may be another hyperlipidemia gene as suggested by Utermann et al. 25 The presence of the e2 allele, even in heterozygotes, seems to predispose subjects to hypertriglyceridemia. This was first reported by Utermann et al., 26 who noted that individuals with the apo E-ND phenotype (one 82 allele) had triglyceride levels intermediate between that of normals and of e2 homozygotes. We previously found a higher prevalence of the 82 allele in a hyperlipidemic population that consisted mostly of hypertriglyceridemic Type IV subjects. 6 Consistent with these observations is the possibility that we inherit a predisposition to hypertriglyceridemia that is revealed when the system is overloaded by ecological or hormonal stresses (excessive caloric or alcohol intake, use of oral contraceptives, pregnancy, or hypothyroidism) or other genetic factors (diabetes, increased VLDL production, or decreased VLDL catabolism). In this study, the primary nature of the hyperlipidemia is assumed on the basis that all secondary causes have been excluded by a thorough clinical and laboratory evaluation at the initial visits. Although a familial history was obtained in all subjects, no systematic study of first-degree relatives was carried out and the presence of occasional cases of nongenetic hyperlipidemia cannot be excluded. Our results show that the 82 allele appears to interact with a genetic predisposition in individuals having a Type IV lipoprotein profile alone, but not in those with hyperapobetalipoproteinemia. This suggests that we are dealing with two separate disease entities. It is not possible with the available information to deter-

4 642 ARTERIOSCLEROSIS VOL 5, No 6, NOVEMBER/DECEMBER 1985 mine whether the former group represents only familial hypertriglyceridemia (FHTG) and the second only familial combined hyperlipidemia (FCHL), two disorders that have specific abnormalities. Indeed, the major metabolic defect ascribed to FHTG is increased hepatic synthesis of triglycerides with the accumulation of triglyceride-rich VLDL in plasma, resulting from an apparent saturation of catabolic processes In FCHL, there is increased synthesis of apo B with an overproduction of both VLDL- and LDL-apoB Structural abnormalities of VLDL may also influence its catabolism and determine the amount of density <1.019 g/ml apo B that proceeds to LDL. As has been demonstrated recently, large VLDL particles in hypertriglyceridemia are not a good substrate for LDL formation and appear to be removed by an alternative pathway Concurrently, reduced amounts of abnormal LDL are formed. In our study, VLDL with apo E2 represent another structural abnormality resulting in impaired LDL formation. It appears that the additional stress imposed by poorly catabolized E2-bearing lipoproteins in subjects with the E2 allele is more likely to produce hyperlipidemia in FHTG. Apo E may also play a significant role in FCHL by allowing the increased production of VLDL to proceed through the pathway to LDL resulting in higher LDL-cholesterol and -apo B levels. Further, we were able to confirm the findings of Sniderman et al. 14 that a substantial proportion of hypertriglyceridemic subjects also have elevated LDL-apo B levels. The higher incidence of atherosclerotic disease they reported in this group emphasizes the importance of measuring apo B levels in hypertriglyceridemia. Finally, as we have recently reported, 7 the apo E locus accounts for more than 15% of the genetic variability of plasma cholesterol in a "normal" population. Additional studies are warranted to determine the impact of apo E polymorphism on the expression of various forms of dyslipidemia. Acknowledgments We thank Lucie Boulet, Ann Chamberiand, Louis-Jacques Fortin, Josee Gingras, and Michel Tremblay for technical assistance and Rejeanne Ouellet for typing the manuscript. We are grateful to Charles F. Sing for his help with the statistics. References 1. Zannla VI, Breslow JL Human very low density lipoprotein E Isoproteln polymorphism is explained by genetic variation and post-translatlonal modification. Biochemistry 1981 ;20: Utermann G, Hees M, Stelnmetz A. Polymorphism of apolipoprotein E and occurrence of dysbetalipoproteinaemia in man. Nature 1977;269: Breslow JL, Zannls VI, SanGlacomo TR, Third JLCH, Tracy T, Glueck CJ. Studies of familial type III hyperlipoproteinemia using as a genetic marker the apo E phenotype E2/2. J Lipki Res 1982;23: Ghlselll G, Schaefer EJ, Zech LA, Gregg RE, Brewer HB Jr. Increased prevalence of apolipoprotein E4 in type V hyperlipoproteinemia. J Clin Invest 1982;70: Bouthllller D, Sing CF, Davlgnon J. Apolipoprotein E pheno typing with a single gel method application to the study of informative matings. J Lipid Res 1983;24: Davlgnon J, Sing CF, Lussler-Cacan S, Bouthllller D. Xanthelasma, latent dyslipoproteinemia and atherosclerosis: contribution of apo E polymorphism. In: De Gennes JL, PolonovsW J, Paoletti R, eds. Latent dyslipoproteinemias and atherosclerosis. New York: Raven Press, 1984; Sing CF, Davlgnon J. Role of the apolipoprotein E polymorphism In determining normal plasma lipid and lipoprotein variation. Am J Hum Genet 1985;37: Utermann G, Kindermann I, Kaffarnlk H, Stelnmetz A. Apolipoprotein E phenotypes and hyperlipidemia. Hum Genet 1984;65: Assmann G, Schmltz G, Menzel H-J, Schutte H. Apolipoprotein E polymorphism and hyperlipidemia. Clin Chem 1984; 30: Havel RJ, Goldstein JL, Brown MS. Lipoproteins and lipid transport. In: Bondy PK, Rosenberg LE, eds. Metabolic control and disease. Philadelphia: WB Saunders, 1980: Goldstein JL, Schrott HG, Hazzard WR, Blerman EL, Motulsky AG. Hyperlipidemia in coronary heart disease. II. Genetic analysts of lipid levels in 176 families and delineation of a new inherited disorder, combined hyperlipidemia. J Clin Invest 1973;52: Sniderman AD, Worfson C, Teng B, Franklin FA, Bachorlk PS, Kwtterovlch PO Jr. Association of hyperapobetalipoprotelnemla with endogenous hypertriglyceridemia and atherosclerosis. Ann Intern Med 1982:97: Brunzell JD, Albers JJ, Chart A, Grundy SM, Groszek E, McDonald GB. Plasma lipoproteins in familial combined hyperlipidemia and monogenlc familial hypertriglyceridemia. J Upid Res 1983:24: Sniderman A, Shapiro S, Marpole D, Skinner B, Teng B, Kwtterovlch PO Jr. Association of coronary atherosclerosis with hyperapobetalipoprotelnemia [increased protein but normal cholesterol levels in human plasma low density (6) lipoproteins]. Proc Natl Acad Sci (USA) 1980;77: Allaln CB, Poon LS, Chan FCS, Richmond W, Fu PC. Enzymatic determination of total serum cholesterol. Clin Chem 1974:20:470-^ Sampson EJ, Demers LM, Krleg AF. Faster enzymatic procedure for serum triglycerides. Clin Chem 1975:21: Lees RS, Hatch FT. Sharper separation of lipoprotein species by paper electrophoresis in albumin-containing buffer. J Lab Clin Med 1963:61: Upid Research Clinics Program. Manual of Laboratory Operations. Vol 1: Upid and lipoprotein analysis. DHEW publication no (NIH) Washington DC: U.S. Government Printing Office 1974: Reardon MF, Poapst ME, Uffelman KD, Stelner G. Improved method for quantitation of B apoprotein in plasma lipoproteins by electroimmunoassay. Clin Chem : Rosseneu M, Vinalmont N, Vercaemst R, Dekeersgleter W, Belpalre F. Standardization of immunoassays for the quantitation of plasma apo B protein. Anal Biochem 1981; 116: Mahley RW. Apolipoprotein E and cholesterol metabolism. Win Wochenschr 1983:61: ChaH A, Brunzell JD, Albers JJ, Hazzard WR. Type III hyperfipoproteinaemia ('remnant removal disease"). Lancet 1977:1: Gregg RE, Zech LA, Schaefer EJ, Brewer HB Jr. Type III hyperiipoprotelnemia: defective metabolism of an abnormal apolipoprotein E. Science 1981:211: Rail SC Jr, Welsgraber KH, Innerarrty TL, Mahley RW. Identical structural and receptor binding defects in apolipo-

5 APO E IN TYPE IV HYPERTRIGLYCERIDEMIA Lussier-Cacan et al. 643 protein E2 in hypo-, normo- and hypercholesterolemic dysbetalipoproteinemia. J Clin Invest 1983;71: Utermann G, Vogelberg KH, Steinmetz A, et al. Polymor- 29 phism of apolipoprotein E. II- Genetics of hyperlipoproteinemia type III. Clin Genet 1979;15: Utermann G, Pruin N, Steinmetz A. Polymorphism of apolipoprotein E. III. Effect of a single polymorphic gene locus on 30. plasma lipid levels in man. Clin Genet 1979;15: Chait A, Albers JJ, Brunzell JD. Very low density lipoprotein overproduction in genetic forms of hypertriglyceridaemia. Eur J Clin Invest 1980;10: Kissebah AH, Alfarsi S, Adams PW. Integrated regulation of very low density lipoprotein triglyceride and apolipoprotein B kinetics in man: normolipemic subjects, familial hypertriglyceridemia and familial combined hyperlipidemia. Metabolism 1981;30: Kissebah AH, Alfarsi S, Evans DJ. Low density lipoprotein metabolism in familial combined hyperlipidemia. Mechanism of the multiple lipoprotein phenotypic expression. Arteriosclerosis 1984;4: Packard CJ, Munro A, Lorimer AR, Gotto AM, Shepherd J. Metabolism of apolipoprotein B in large triglyceride-rich very low density lipoproteins of normal and hypertriglyceridemic subjects. J Clin Invest 1984;74: Oschry Y, Olivecrona T, Deckelbaum RJ, Eisenberg S. Is hypertriglyceridemic very low density lipoprotein a precursor of normal low density lipoprotein? J Lipid Res 1985:26: Index Terms: apolipoprotein E polymorphism hyperapobetalipoproteinemia apo E allele endogenous hypertriglyceridemia Correction The Editors of Arteriosclerosis regret the error in this abstract, which appeared on page 550a of the September/ October issue {Arteriosclerosis 5:550a, 1985) and reprint it in its correct form. Hepatic Uptake of LDL in Rats Induced by Lactosaminated Fab Antibodies F. Bernini, S. R. Tanenbaum, B. C. Sherrill, A. M. Gotto, L C. Smith. Baylor College of Medicine and Methodist Hospital, Houston, TX Proteins modified by conjugation with lactose residues often exhibit increased hepatic uptake mediated by the galactose or asialoglycoprotein receptor. As reagents to promote plasma clearance and hepatic uptake, we prepared lactosaminated Fab fragments of sheep IgG specific for human LDL. Fab fragments of anti-ldl IgG retained their affinity after derivatization. Rats were injected first with 125 i-ldl and ihen after 10 minutes with 1 mg of either lactosaminated Fab (lac-fab) or Fab. Ten minutes later the radioactivity in the serum of rats treated with lac-fab was reduced to 47.7% (±3.4%) of the injected 125 I-LDL, with an increase of hepatic localization from 3% to 30%. After 60 minutes 78.5% (±5.6%) of 125 I-LDL was cleared from the circulation. In the rats treated with Fab or saline, 85% of 125 I-LDL remained in the circulation with only 7% present in the liver. These processes were completely inhibited by simultaneous injection of asialofetuin. These findings suggest that lac-fab is an effective means to induce the catabolism of LDL by the liver.