Clinical Chemistry 45, No. 6, 1999
10.
11.
12.
13.
Muskiet FAJ. Hemoglobin S and C: reference values for glycohemoglobin in heterozygous, double-heterozygous and homozygous subjects, as established by 13 methods. Clin Chim Acta 1994;231:161–71. Martina WV, Martijn EG, van der Molen M, Schermer JG, Muskiet FAJ. b-N-Terminal glycohemoglobin in subjects with common hemoglobinopathies: relation with fructosamine and mean erythrocyte age. Clin Chem 1993;39:2259 – 65. Holownia P, Bishop E, Newman DJ, John GW, Price CP. Adaptation of latex-enhanced assay for percent glycohemoglobin to a Dade Dimension® analyzer. Clin Chem 1997;43:76 – 84. Konigsberg W, Goldstein J, Hill RJ. The structure of human hemoglobin VII: the digestion of the b chain of human hemoglobin with pepsin. J Biol Chem 1963;238:2028 –33. Barbedo MMR, McCurdy PR. Red cell life span in sickle cell trait. Acta Haematol 1974;51:339 – 43.
Simple Precipitation-based Method for the Screening of Type III Hyperlipoproteinemia, Matthias Nauck,1 Lioba Glatt,1 Winfried Ma¨rz,1 Giso Feussner,2 and Heinrich Wieland1 (1 University Hospital of Freiburg, Department of Clinical Chemistry, Hugstetter Strasse 55, 79106 Freiburg i. Br., Germany, and 2 University Hospital of Heidelberg, Department of Endocrinology, D-69115 Heidelberg, Germany; * author for correspondence: fax 49-761-270 3444, e-mail
[email protected]) Familial type III hyperlipoproteinemia (HLP) is characterized by the accumulation of cholesterol-rich remnants (b-VLDL). The differential diagnosis of type III HLP is clinically important because patients with type III HLP develop premature coronary artery disease (CAD) and peripheral atherosclerosis and because type III HLP responds well to dietary treatment and fibric acid derivatives (1, 2 ). Pathogenetically, type III HLP is related to dysfunctional isoforms of apolipoprotein (apo) E. At the APOE gene locus, three common alleles exist, designated e2, e3, and e4 (3–5 ). apoE2 is defective in its binding to lipoprotein receptors (6, 7 ). Because of the impaired catabolism of chylomicron and VLDL remnants, individuals homozygous for apoE2 reveal detectable amounts of b-VLDL in their plasma. b-VLDLs are atypical lipoproteins with a density ,1.006 kg/L and b-mobility on agarose gel electrophoresis. Compared with normal VLDL, b-VLDLs are cholesterol-enriched; compared with normal LDL, they are relatively enriched in triglycerides. More than 90% of patients with type III HLP are homozygous for apoE2, but only ;1 in 20 individuals carrying the E2/2 phenotype finally develops type III HLP (8 ). Those homozygous carriers of apoE2 having small amounts of b-VLDL in their plasma not sufficient to produce overt hyperlipidemia have been classified as suffering from normolipidemic dysbetalipoproteinemia. The term type III HLP, in contrast, is applied to hyperlipemic individuals only. Clinical characteristics such as palmar, tendon, and/or tubero-eruptive xanthomas do not occur in all individuals with type III HLP. To establish biochemically the diagnosis of type III HLP, the following criteria have been applied in this study: (a) presence of increased cholesterol and triglycerides at 2500 mg/L or more, (b) an increased
909
ratio of VLDL-cholesterol (VLDL-C) to VLDL-triglycerides (VLDL-TG; .0.4), and (c) an increased ratio of VLDL-C to total triglycerides (.0.3) (9 –11 ). We studied in total 1317 sera from women and men, ages 20 – 65 years. Among the participants with apoE phenotypes other than E2/2 (n 5 1288) were 468 CAD patients recruited at the University Hospital, Freiburg or at the Benedikt-Kreutz-Klinik, Bad Krozingen, and 820 clinically healthy individuals, recruited at the Rheintalklinik, Bad Krozingen or employees of the BASF, Ludwigshafen. Twenty-nine apoE2 homozygotes were studied, including 21 samples obtained at the University Hospital of Heidelberg. In 8 of the 12 individuals with type III HLP, signs of atherosclerosis were present (66%). Two of the 17 apoE2 homozygotes without manifest type III HLP showed CAD (12%). Informed consent was obtained from each individual participating in this study; all procedures followed were in accordance with the Helsinki Declaration of 1975, as revised in 1983. Blood was drawn after an overnight fast and allowed to clot at room temperature; serum was obtained by centrifugation at 1500g for 15 min. All analyses were performed within 2 days after blood collection at the University Hospital of Freiburg. Cholesterol and triglycerides, without blanking for glycerol, were determined enzymatically (Boehringer Mannheim) using a Hitachi type 747 (total cholesterol and triglycerides) or a Wako 30R analyzer (supernates of precipitation reactions), respectively. The total CVs for the methods were ,2%. LDL-like lipoproteins were isolated using 100 mL of serum and 1000 mL of dextran sulfate (DS)/MgCl2 precipitation reagent (Quantolip®; Immuno GmbH) and incubated for at least 10 min. The resulting precipitate was separated by a 5-min centrifugation, and the cholesterol and triglyceride content of the soluble lipoproteins (mainly VLDL and HDL) was measured in the supernate. Cholesterol and triglycerides associated with the precipitated lipoproteins (LDL-CDS and LDL-TGDS, respectively) were calculated as total cholesterol or triglycerides minus supernate cholesterol or triglycerides, respectively. A combined ultracentrifugation (UC) and precipitation assay was used as the comparison method (12, 13 ). In this method, the recoveries of the bottom fraction (LDL plus HDL) was between 97% and 102%, whereas the recoveries of the top fraction (VLDL) was between 75% and 103%. All lipid measurements of one sample—lipids and lipoprotein fractions—were performed in the same analytical runs. The between-day CVs for LDL-CDS, LDL-TGDS, LDL-CUC, and HDL-CUC were below 3% and 4%, respectively. apo E phenotyping was performed by isoelectric focusing on agarose and immunofixation (14, 15 ). Regression analyses were performed using the method of Passing and Bablok (16 ). We first wished to compare the results for LDL-C obtained with the DS precipitation method with those obtained with UC. In this comparison, we excluded 45 samples in which a recently developed algorithm indicated that the DS precipitation was incomplete because of
910
Technical Briefs
fatty acid concentrations exceeding 2 mmol/L (17–19 ). Alternatively, samples can be diluted with bovine serum albumin before precipitation, which will overcome the effect of free fatty acids, as described for lipoprotein electrophoresis [Nauck et al., unpublished results, and Ref. (20 )]. In the remaining 1243 samples with apoE phenotypes other than apoE2/2, LDL-CDS showed a good agreement to LDL-CUC (y 5 1.07x 2 154 mg/L; r 5 0.969) (19 ). Recent work demonstrated that DS/MgCl2 precipitation co-precipitates VLDL (19 ). In apoE2 homozygotes with normolipidemic dysbetalipoproteinemia, the two methods corresponded with each other (r 5 0.953; y 5 0.79x 1 304 mg/L), indicating that co-precipitation of b-VLDL does not markedly affect apparent LDL-CDS values. In individuals with type III HLP, however, virtually no correlation between LDL-CDS and LDL-CUC was observed (r 5 20.126). The LDL-CDS was on average twice as high as LDL-CUC, suggesting the amounts of b-VLDL. The accumulation of triglyceride-enriched b-VLDL can be detected by analyzing the triglyceride content of the DS precipitate. We compared the ratio of LDL-TGDS to LDLCDS between individuals with type III HLP and other types of HLP. Eight samples with type III HLP were matched with 16 samples with other types of HLP according to total cholesterol and total triglycerides. The ratio of LDL-TGDS to LDL-CDS was significantly higher in type III HLP, but this did not allow unequivocal diagnosis of type III HLP because this ratio may be increased in other types of HLP as well (Table 1). In Fig. 1A, absolute values of LDL-TGDS are plotted vs LDL-CDS, including values for 18 samples from apoE2 homozygotes. All samples from individuals with type III HLP (n 5 8) had increased LDL-TGDS .950 mg/L and LDL-CDS .1750 mg/L, whereas all samples from subjects with normolipidemic dysbetalipoproteinemia did not meet these criteria. Twenty-one samples of 1243 samples with phenotypes other than apoE2/2 (1.69%) exceeded the two threshold concentrations. The proportion of actual type III HLP among the suspicious samples thus was 28% (8 of 29). When calculated on the basis of all samples, the sensitivity to detect type III HLP by applying the combined thresholds of LDL-TGDS .950 and LDL-CDS .1750 mg/L was thus 100%, and the specificity was 98.3%. The specificity dropped to 92% if only samples with LDL-CDS .1750 mg/L were considered. To verify Table 1. Lipid values for patients with type III HLP (n 5 8) and other hyperlipoproteinemias (n 5 16). Total triglycerides,a mg/L Total cholesterol, mg/L HDL-C, mg/L LDL-TGDS/LDL-CDS
Type III
No type III
P
4333 6 903 3440 6 626 410 6 48 0.82 6 0.19
4300 6 855 3070 6 361 430 6 103 0.55 6 0.14
NSb NS ,0.001 ,0.001
a For total cholesterol and total triglyceride concentrations, the data were matched for each patient with deviations of ,500 mg/L. b NS, not significant.
Fig. 1. Screening for type III HLP using LDL-TGDS and LDL-CDS. (A), 21 of 1243 samples showed LDL-TGDS .950 mg/L and LDL-CDS .1750 mg/L. All patients with type III HLP were detected. (B), Four type III HLP patients were detected by the screening procedure, whereas eight homozygotes for apoE2 without type III HLP were correctly classified as unsuspicious.
these results, we analyzed another 11 completely independent serum samples from other individuals homozygous for apoE2 with and without type III HLP (Fig. 1B). All individuals were classified correctly when we applied the above threshold values to these samples. According to these results, a sample should be considered suspicious of type III HLP if (a) both total cholesterol and triglycerides exceed 2500 mg/L, (b) the LDL-CDS concentration is . 1750, and (c) the LDL-TGDS concentration is .950 mg/L. However, only ;30% of the suspicious cases really had type III HLP, whereas in the remaining cases other types
Clinical Chemistry 45, No. 6, 1999
of HLP existed. Because our diagnostic criteria have been chosen in favor of sensitivity rather than specificity, we recommend confirming the diagnosis of type III HLP in suspicious samples using apoE phenotyping and/or apoE genotyping along with preparative UC and/or lipoprotein electrophoresis. In conclusion, the advantage of the proposed procedure is that the screening for increased LDL-C concentrations includes the option to detect high concentrations of atherogenic LDL-like particles with an abnormal composition by simultaneous determination of triglycerides and cholesterol after precipitation with DS and Mg21 ions. Without this information, type III HLP will frequently be overlooked. Our procedure is inexpensive and can be performed in every laboratory because it needs no specialized equipment. We thank Brigitte Haas, Gabi Herr, Isolde Friedrich, Sabine von Karger, Brigitte Kreisel, Sibylle Rall, and Susanne Schuler for excellent technical assistance. We thank Prof. Dr. H. Just, Prof. Dr. H. Roskam, Dr. Fro¨lich, and Prof. Dr. Dr. A. Zober for their cooperation in the recruitment of patients and healthy individuals. References 1. Morganroth J, Levy RI, Fredrickson DS. The biochemical, clinical and genetic features of type III hyperlipoproteinemia. Ann Intern Med 1975;82:158 –74. 2. Brewer HB Jr, Zech LA, Gregg RW, Schwart D, Scha¨fer J. Type III hyperlipoproteinemia: diagnosis, molecular defects, pathology and treatment. Ann Intern Med 1983;98:623– 40. 3. Mahley RW. Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science 1988;240:622–30. 4. Davignon J, Gregg RE, Sing CF. Apolipoprotein E polymorphism and atherosclerosis. Arteriosclerosis 1988;8:1–21. 5. Weisgraber KH, Rall SC, Mahley RW. Human E apoprotein heterogeneity. Cysteine arginine interchanges in the amino acid sequence of the apo E isoforms. J Biol Chem 1981;256:9077– 83.
911
6. Schneider WJ, Kovanen PT, Brown MS, Goldstein JL, Utermann G, Weber W, et al. Familial dysbetalipoproteinemia. Abnormal binding of mutant apoprotein E to low density lipoprotein receptors of human fibroblasts and membranes from liver and adrenals of rats, rabbits and cows. J Clin Investig 1981;68:1075– 85. 7. Lalazar A, Weisgraber KH, Rall SC, Giladi H Jr, Innerarity TL, Lavanon AZ, et al. Site-specific mutagenesis of human apolipoprotein E. Receptor binding activity of variants with single amino acid substitutions. J Biol Chem 1988;263:3542–5. 8. Utermann G, Hees M, Steinmetz A. Polymorphism of apolipoprotein E and occurrence of dysbetalipoproteinemia in man. Nature 1991;269:604 –7. 9. Hazzard WR, Porte D, Bierman EL. Abnormal lipid composition of very low density lipoproteins in diagnosis of broad-beta disease (type III hyperlipoproteinemia). Metabolism 1972;21:1009 –19. 10. Fredrickson DS, Morganroth J, Levy RI. Type III hyperlipoproteinemia. An analysis of two contemporary definitions. Ann Intern Med 1975;82:150 –7. 11. Ma¨rz W, Feussner G, Siekmeier R, Donnerhak B, Schaaf L, Ruzicka V, Gross W. Apolipoprotein E to B ratio. a marker for type III hyperlipoproteinemia. Eur J Clin Chem Clin Biochem 1993;31:743–7. 12. National Heart and Lung Institute. Manual of laboratory operation. Lipid Research Clinics Program. DHEW No. (NIH) 75-628. Bethesda, MD: National Heart and Lung Institute, 1974:1–74. 13. Wanner C, Ho¨rl WH, Luley C, Wieland H. Effects of HMG-CoA reductase inhibitors in hypercholesterolemic patients on hemodialysis. Kidney Int 1991;39:754 –9. 14. Luley C, Baumstark MW, Wieland H. Rapid apolipoprotein E phenotyping by immunofixation in agarose. J Lipid Res 1991;5:880 –3. 15. Luley CH, Haas B, Bu¨hrer B, Wieland H. Improvement of apolipoprotein E phenotyping by isoelectric focusing/immunofixation [Letter]. Clin Chem 1992;38:168. 16. Passing H, Bablok W. Comparison of several regression procedures for method comparison studies and determination of sample size. Part II. J Clin Chem Biochem 1984;22:431– 45. 17. Armstrong V, Seidel D. Evaluation of a commercial kit for the determination of LDL cholesterol in serum based on precipitation of LDL with dextran sulfate. A¨rzt Lab 1985;31:325–33. 18. Siekmeier R, Ma¨rz W, Gross W. Precipitation of LDL with sulfated polyanions: three methods compared. Clin Chim Acta 1988;177:221–30. 19. Nauck M, Friedrich I, Wieland H. Measurement of LDL and VLDL cholesterol with precipitation techniques. A comparison with the ultracentrifugation method. Clin Lab 1994;40:167–76. 20. Nauck M, Ma¨rz W, Wieland H. Adding albumin normalizes electrophoretic mobility of lipoproteins in sera with high concentrations of free fatty acids. Clin Chem 1996;42:1283–5.
