Flow cytometry with a monoclonal antibody to the low density lipoprotein receptor compared with gene mutation detection in diagnosis of heterozygous familial hypercholesterolemia

¹ Department of Internal Medicine and Cardiology, Aarhus Amtssygehus University Hospital, DK-8000 Aarhus C, Denmark.

² Department of Cardiology, Skejby Sygehus University Hospital, DK-8200 Aarhus N, Denmark.
^a Address correspondence to this author at: Department of Internal Medicine and Cardiology, Aarhus Amtssygehus University Hospital, Tage Hansens Gade 2, DK-8000 Aarhus C, Denmark. Fax 45 89 49 76 19; e-mail rau{at}dadLnet.dk.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
We used a fluorescence flow cytometry assay with a monoclonal low density lipoprotein (LDL) receptor-specific antibody to detect LDL receptor expression on blood T lymphocytes and monocytes. We prepared peripheral blood mononuclear cells from patients with genetically verified LDL receptor-defective (Trp⁶⁶-Gly mutation, n = 17) or receptor-negative (Trp²³-stop mutation, n = 17) heterozygous familial hypercholesterolemia (FH) and from healthy individuals (n = 24). The cells were stimulated to express the maximum amount of LDL receptor by preincubation in lipoprotein-deficient medium. A dual-labeling technique allowed flow cytometric analysis of LDL receptor expression on cells identified by fluorescently conjugated surface marker antibodies. Knowing the LDL receptor gene mutation of the FH patients allowed us to compare the diagnostic capability of this functional assay with the DNA diagnosis and to validate the assay with molecular genetics instead of clinical indices of heterozygous FH. T lymphocytes expressed more LDL receptors and gave better diagnostic results than monocytes, and cells from patients with either the Trp⁶⁶-Gly or the Trp²³-stop mutation had variable but significantly reduced LDL receptor expression. The data indicate that this fluorescence flow cytometry assay is unsuitable for diagnosis of individual cases of heterozygous FH but that it may be useful for functionally characterizing mutations in the LDL receptor gene.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Familial hypercholesterolemia (FH)¹ is a monogenic, autosomal dominant disorder caused by mutations in the gene encoding the LDL receptor. More than 300 different mutations are identified in the LDL receptor gene. Some mutations are frequent and some are rare, but heterozygous FH affects ~1 in 500 individuals in most populations (1)(2). In Denmark, two mutations, Trp-Gly and Trp-stop, account for nearly one-third of FH patients (3). Current gene mutation detection methods allow molecular diagnosis of individual cases of FH (4)(5), but they cannot functionally characterize mutations in the LDL receptor gene to determine their impact on LDL catabolism.

The clearance of LDL from plasma is primarily controlled by the LDL receptor pathway. LDL receptors are expressed by all cell types studied (1). The expression in vivo, however, varies with cellular cholesterol requirements. Hepatocytes have high LDL receptor activity and are responsible for nearly 70% of the receptor-mediated uptake of LDL (6)(7). Studies of LDL receptor activity ex vivo have classically been based on the uptake of I-labeled lipoproteins into cultured human fibroblasts (8)(9). Faster and less painful measurements of LDL receptor function have been done on peripheral blood mononuclear cells (PBMCs) using radioactive probes (10)(11). Flow cytometry has been used to measure LDL receptor activity on stimulated human monocytes or lymphocytes with fluorescently conjugated LDL and LDL receptor-specific antibodies (12)(13)(14)(15)(16). A major problem has been that these methods of characterizing LDL receptor activity in patients with clinical FH produce results overlapping with those of healthy individuals (13)(14)(16).

We report here a fluorescence flow cytometry (FFC) assay that uses a monoclonal LDL receptor-specific antibody to determine the LDL receptor expression on stimulated PBMCs. The method was tested on T lymphocytes and monocytes from patients with genetically verified heterozygous FH and from healthy individuals. We could, therefore, assess the ability of FFC to predict the results of DNA analysis. We did not fully characterize functionally the LDL receptor with this assay because this requires not only measurements of LDL receptor expression but also examination of binding and internalization of LDL by the receptor.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
subjects and lipids
Blood samples were drawn from 17 patients with the Trp-Gly mutation, 17 patients with the Trp-stop mutation, and 24 healthy individuals without dyslipidemia or a family history of coronary heart disease. Lipid measurements are given in Table 1 . HDL-cholesterol was 0.8 mmol/L in four patients (three in the receptor-defective group and one in the receptor-negative group). All other individuals had plasma HDL-cholesterol 0.9 mmol/L. All patients were normotriglyceridemic (2.5 mmol/L), and none were treated with lipid-lowering drugs before blood sampling. This study was in accordance with the Helsinki Declaration of 1975, as revised in 1983.

mononuclear cell isolation and preincubation
Blood, collected into tubes containing 1.5 g/L K₂-EDTA, was cooled to 20 °C and diluted 1:1 with Hank's buffered saline solution. PBMCs were prepared under sterile conditions, using a modified version of the method of Böyum (17). Ficoll-Paque (Pharmacia Biotech) and diluted blood were layered in a centrifuge tube and centrifuged for 40 min at 400g and 20 °C. The interface containing the PBMCs was isolated, and the cells were washed three times in Hank's and resuspended in RPMI-1640 (Gibco BRL) with L-glutamine (290 mg/L), penicillin (100 000 U/L), streptomycin (100 mg/L), and 100 mL/L human lipoprotein-deficient serum (HLPDS) to a final concentration of 10 cells/mL. The PBMCs were preincubated for 46 h at 37 °C in a humidified carbon dioxide incubator.

cell labeling
Tissue culture flasks were placed in ice water for 60 min in the dark to reduce cell adhesion. PBMCs were removed by flushing with ice-cold Hank's (4 °C) and washed twice in ice-cold Hank's with 20 mL/L HLPDS. The cell number was adjusted to 0.3 x 10 cells/mL, and aliquots of 100 µL of cell suspension were pipetted into polypropylene tubes and placed in ice water. Cells were incubated with 1.5 µg of monoclonal mouse anti-human LDL receptor-specific antibody, clone C7 (Amersham Life Science), for 30 min in the dark at 4 °C. After the cells were washed twice in ice-cold Hank's with 20 mL/L HLPDS, they were incubated with 3 µL of fluorescein isothiocyanate (FITC, Dako) for 30 min in the dark at 4 °C. To determine the lineage of specific differentiation antigens, cells were incubated with 1 µL of R-phycoerythrin (RPE)-conjugated monoclonal antibody: CD3-RPE or IgG₁ isotype-RPE for T lymphocytes and CD14-RPE or IgG_2a isotype-RPE for monocytes (all from Dako). Cell suspensions serving as controls were labeled by the same procedure but with only FITC and CD3-RPE or CD14-RPE to determine the nonspecific binding of FITC to CD3-positive or CD14-positive cells, respectively.

flow cytometry measurements
The measurements were performed in a FACSort flow cytometer (Becton Dickinson) equipped with a 15 mW, 488 nm, air-cooled argon laser and linked to a Macintosh Quadra 650 computer with CellQuest software (Becton Dickinson). Forward scatter (FSC) and side scatter (SSC) were adjusted to exclude debris and dead cells. FITC emission was measured at 530 nm ± 15 nm (FL1), and RPE emission was measured at 585 nm ± 21 nm (FL2). Compensation was set using FITC-conjugated C7 (C7-FITC)-labeled cells (FL2-FL1) and CD3-RPE- or CD14-RPE-labeled cells (FL1-FL2). The acquisition number was set at 20 000. A morphological gate including lymphocyte-like or monocyte-like cells was defined in a FSC-SSC dot plot. A CD3- or CD14-positive gate was defined in a FSC-FL2 dot plot, using an IgG₁ isotype-RPE or IgG_2a isotype-RPE to determine the 1% cutoff limit for CD3 or CD14 positivity, respectively. Only cells recorded in these gates were included in the measurements of FL1. Fluorescence signals for the C7-FITC-labeled and FITC-labeled cells were recorded on a logarithmically transformed intensity scale in a histogram. A relative quantification of C7 bound to the T lymphocytes or to monocytes was expressed as the ratio of the median fluorescence of C7-FITC-labeled cells to the median fluorescence of FITC-labeled cells, the C7-FITC/FITC ratio. All ratios were based on measurements performed in triplicate.

statistics
The significance level was set at 5%. We used the Kruskal-Wallis unpaired rank sum test to test for differences between ratios for the three groups of individuals. To compare ratios for the groups two-by-two, we used the nonparametric Mann-Whitney U-test.

To evaluate the precision of the FFC assay, we estimated the random analytical error given by within-run and between-run imprecision and biological variation by making repeated measurements on blood samples from one healthy individual. We used the general linear model procedure in SPSS (Ver. 7.5) to estimate the ANOVA type I variance components in a model with "Day" and "Sample" as random effects and with "Sample" nested within "Day". Coefficients of variation were based on the grand means of C7-FITC/FITC ratios.

The best cutoff point for the C7-FITC/FITC ratio to discriminate between unaffected (non-FH) and reduced LDL receptor expression (heterozygous FH) was found using ROC curves (18) to give the best possible sensitivity and specificity for the FFC assay.

Knowing the true value of the frequency of occurrence of heterozygous FH (f) in the study population [as determined by DNA analysis (3)] and recognizing the probability for true heterozygous FH detection (p_TP, equal to sensitivity) and the probability for false heterozygous FH detection (p_FP, 1 - p_FP equals specificity), we calculated a predictive value of true heterozygous FH detection (PV_TP,i.e., the fraction of occasions that a positive test result identifies a heterozygous FH patient), a predictive value of true non-FH detection (PV_TN, i.e., the fraction of occasions that a negative test result identifies a non-FH individual), and an efficiency (19) for a given cutoff point for the C7-FITC/FITC ratio:

PV_TP = (p_TP x f): (p_TP x f p_FP x (1 - f))

PV_TN = ((1 - p_FP)(1 - f)): ((1 - p_FP)(1 - f) (1 - p_TP) x f)

Efficiency = p_TP x f (1 - p_FP)(1 - f)

To evaluate the accuracy of the FFC assay, we estimated the systematic analytical error by calculating the probability of independence between the observed occurrence of heterozygous FH in the study population as determined by FFC and the expected occurrence as determined by DNA analysis (3), using the test (19).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
A cell volume (FSC) vs density (SSC) dot plot showed a distinct lymphocyte population separated from a population of larger and more dense monocytes (Fig. 1 A). Labeling with CD3-RPE or CD14-RPE gave a >100-fold increase in the mean RPE (FL2) signal compared with the nonspecific binding of IgG₁ isotype-RPE or IgG_2a isotype-RPE (Fig. 1B ). Combining the morphologic lymphocyte or monocyte gate and the CD3- or CD14-positivity gate gave a >99% pure CD3- or CD14-positive cell population, respectively.

View larger version (22K): [in this window] [in a new window]   Figure 1. (A) A cell volume (FSC) vs density (SSC) flow cytometry dot plot of PBMCs. FSC and SSC were adjusted to exclude debris and dead cells. The plot shows a distinct lymphocyte population separated from a population of larger and more dense monocytes. (B) Labeling with T lymphocyte- or monocyte-specific RPE-conjugated monoclonal antibody gave a >100-fold increase in median FL2 signal compared with the nonspecific binding of IgG1 isotype-RPE or IgG2a isotype-RPE. A CD3- or CD14-positive gate was defined to determine the 1% cutoff limit for CD3 or CD14 positivity, respectively.

The C7-FITC/FITC ratios for the three groups of individuals were scattered for both T lymphocytes (Fig. 2 A) and monocytes (Fig. 2B ). The difference between the three groups of ratios was highly significant for T lymphocytes (P <0.0001) and for monocytes (P = 0.001). Ratios for the healthy group were significantly higher than ratios for the receptor-defective group on basis of the T-lymphocyte experiments only (P = 0.04) and for the receptor-negative group on the basis of experiments performed with T lymphocytes (P <0.0001) as well as monocytes (P = 0.0005). Ratios for the receptor-negative group were significantly lower than for the receptor-defective group, using both T lymphocytes (P = 0.002) and monocytes (P = 0.008).

View larger version (8K): [in this window] [in a new window]   Figure 2. A relative quantification of C7 bound to the peripheral blood mononuclear cells expressed as the ratio of the median fluorescence signal of C7-FITC-labeled cells to the median fluorescence signal of FITC-labeled cells (the C7-FITC/FITC ratio). Ratios for the receptor-negative and receptor-defective heterozygous FH patients and for healthy individuals are scattered and overlap for experiments with T lymphocytes (A) and monocytes (B).

We did analysis on one healthy individual sampled twice on the same day and once weekly for 6 weeks. The estimates of day-to-day, sample-to-sample (within the same day), residual, and total imprecision and CV values are given in Table 2 . Together, the day-to-day and sample-to-sample imprecision represent the sum of between-run imprecision and biological variance for the assay. The residual imprecision represents the within-run imprecision. T lymphocytes gave larger total imprecision than monocytes because of a relatively large day-to-day imprecision. The residual imprecision for T lymphocytes, on the other hand, was smaller than for monocytes, indicating a more robust method per se.

Table 3 gives the values for frequency of heterozygous FH in the study population, best cutoff point for the C7-FITC/FITC ratio to discriminate between non-FH and heterozygous FH, sensitivity, specificity, PV_TP, PV_TN, and efficiency for the assay.

The probability of independence between the observed occurrence and the expected occurrence of heterozygous FH is given in Table 4 .

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Knowing the actual LDL receptor gene mutation in the included FH patients enabled us to compare the diagnostic results of the FFC assay with the true DNA diagnosis and to validate the assay on the basis of molecular genetics instead of clinical indices of FH. To our knowledge, this report is the first to address this problem. With the cutoff point for the C7-FITC/FITC ratio to discriminate between heterozygous FH and non-FH set at 2.6 for T lymphocytes and at 2.3–2.8 for monocytes, we were able to detect a significantly reduced LDL receptor expression on PBMCs for both the LDL receptor-defective Trp-Gly mutation (class III or V) and the LDL receptor-negative Trp-stop mutation (class I). The FFC assay gave reproducible measurements, but the predictive values for true heterozygous FH detection, true non-FH detection, efficiency, and accuracy were relatively poor and do not allow the assay to be used for individual diagnosis of heterozygous FH.

Our FFC assay discriminated incompletely between patients with genetically verified receptor-defective or receptor-negative FH and healthy individuals. This is consistent with the results of other reports trying to separate patients with clinical FH from healthy individuals, using flow cytometry and fluorescently labeled LDL (13)(14)(16).

The monoclonal antibody (C7) used in our FFC assay is a commercial mouse IgG_2b immunoglobulin specific to LDL receptors of human and bovine origin. C7 binds to the LDL receptor in amounts equimolar to the binding of LDL, i.e., each LDL receptor has one C7-binding site per LDL-binding site (20). Studies of the binding of LDL and C7 have revealed, however, that the two binding sites are not identical, because binding of C7 is not inhibited even if the receptor is fully occupied by prior binding of LDL. On the other hand, prior binding of C7 to the LDL receptor reduces later LDL binding by as much as 80% at 4 °C (20). This suggests that C7 is bound so closely to the LDL-binding site that it can cause a steric hindrance or conformational change of the LDL binding site.

Flow cytometry has several advantages over other assays for characterization of the LDL receptor. Because of the construction of the flow cytometer, the FFC assay makes it possible to accumulate information about size and granularity of a single cell and simultaneously to measure the fluorescence signal of several probes attached to the cell. This contrasts with radioligand assays, for example, where measurements are an average signal for a cell suspension. Furthermore, the use of gates in the acquisition and analysis process of the FFC assay can exclude fluorescence signals of the medium and cellular debris and include or exclude various cell populations on the basis of cell morphology and cell surface marker labeling. Thus, in our assay, the dual-labeling technique allows measurements of the fluorescence signal from the LDL receptor-specific antibody on a gated population of cells identified by specific fluorescently conjugated surface marker antibodies.

Classically, LDL receptor activity has been studied in cultured human fibroblasts, using radioligand assays (9). However, by density gradient centrifugation it is possible to prepare relatively pure and large populations of human PBMCs from fresh blood samples (17). Furthermore, experiments have indicated that circulating human PBMCs reflect the LDL receptor status in other cells in the human body (21)(22).

LDL receptor activity has been shown by some authors to be greater in freshly isolated monocytes than in lymphocytes (11)(23), a feature also found in stimulated subfractions of PBMCs. Schmitz et al. (13) found that monocytes, preincubated in lipoprotein-deficient serum for 48 h, express more LDL receptor activity than phytohemagglutinin-stimulated lymphocytes. One recent study, however, found that the ability to measure fluorescence in T lymphocytes improved the discrimination between FH and non-FH individuals when compared with values obtained from the whole PBMC population (16).

These results are consistent with ours. In our study of LDL receptor expression, we found first that T lymphocytes expressed more LDL receptors than did monocytes. We also found that the best differentiation between groups of patients and healthy individuals was obtained in experiments with T lymphocytes. In addition, T lymphocytes gave a uniform cutoff point for the C7-FITC/FITC ratio for the best discrimination between healthy and groups of heterozygous FH patients. Finally, T lymphocytes generally gave higher values for true heterozygous FH detection, true non-FH detection, efficiency, and accuracy and gave a smaller within-run imprecision for the FFC assay than did monocytes.

Our work therefore suggests that the best results regarding LDL receptor expression are obtained with T lymphocytes stimulated in lipoprotein-deficient medium for 46 h rather than with monocytes stimulated in the same manner. We have no experience with phytohemagglutinin-stimulated cells. Because of the variance in LDL receptor expression between lymphocytes and monocytes, it is important in all studies of LDL receptor activity in human PBMCs to separate lymphocytes and monocytes, or at least to carefully take into account the percentage of monocytes in a mixed-PBMC population to avoid irregularity in results caused by variation in the composition of the PBMC pool.

In conclusion, we suggest that our FFC assay using fluorescently labeled C7 bound to LDL-receptor sites of human T lymphocytes stimulated for 46 h in lipoprotein-deficient medium can be used to evaluate LDL-receptor expression in the human body. Measurements of LDL receptor expression are necessary to fully characterize functionally mutations in the LDL receptor gene. Our experience suggests that valid results with this assay can be obtained if performed in at least seven individuals with the same mutation in the LDL receptor gene. Because of the relatively poor values for true heterozygous FH detection, true non-FH detection, efficiency, and accuracy for this assay, the assay is not suitable for diagnosis of individual cases of heterozygous FH.

	Acknowledgments

 
We thank Lars Ulrik Gerdes and Mogens Lytken Larsen for advice concerning the statistical analysis and Pia Buchtrup Hornbek and Gitte Glistrup Nielsen for excellent technical assistance. This study was supported by the Danish Heart Foundation (Grants 96136022380, 96136122381, 962221a22418, 962221b22419, 97136522498, and 97245822552), the Institute of Experimental Clinical Research, University of Aarhus, Aarhus, Denmark, the Kirsten Anthonius' Mindelegat, the Ebba Celinders Legat, and the Danish Medical Association Research Fund.

	Footnotes

 
¹ Nonstandard abbreviations: FH, familial hypercholesterolemia; PBMC, peripheral blood mononuclear cell; FFC, fluorescence flow cytometry; HLPDS, human lipoprotein-deficient serum; FITC, fluorescein isothiocyanate; RPE, R-phycoerythrin; FSC, forward scatter; SSC, side scatter; f, frequency of occurrence of heterozygous FH; p_FP, probability for false heterozygous FH detection; p_TP, probability for true heterozygous FH detection; PV_TN, predictive value of a true non-FH detection; and PV_TP, predictive value of a true heterozygous FH detection.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Goldstein JL, Hobbs HH, Brown MS. Familial hypercholesterolemia. Scriver CR Beaudet AL Sly WS Valle D eds. The metabolic basis of inherited disease 7th ed. 1995:1981-2030 McGraw-Hill New York. .
2. Andersen GE, Lous P, Friis-Hansen B. Screening for hyperlipoproteinemia in 10,000 Danish newborns. Follow-up studies in 522 children with elevated cord serum VLDL-LDL-cholesterol. Acta Paediatr Scand 1979;68:541-545. [Web of Science][Medline] [Order article via Infotrieve]
3. Jensen HK, Jensen LG, Hansen PS, Faergeman O, Gregersen N. The Trp²³-Stop and Trp⁶⁶-Gly mutations in the LDL receptor gene: common causes of familial hypercholesterolemia in Denmark. Atherosclerosis 1996;120:57-65. [Web of Science][Medline] [Order article via Infotrieve]
4. Hobbs HH, Brown MS, Goldstein JL. Molecular genetics of the LDL receptor gene in familial hypercholesterolemia. Hum Mutat 1992;1:445-466. [Medline] [Order article via Infotrieve]
5. Jensen HK, Jensen LG, Hansen PS, Faergeman O, Gregersen N. High sensitivity of the single-strand conformation polymorphism method for detecting sequence variations in the low-density lipoprotein receptor gene validated by DNA sequencing. Clin Chem 1996;42:1140-1146. [Abstract/Free Full Text]
6. Pittman RC, Carew TE, Attie AD, Witztum JL, Watanabe Y, Steinberg D. Receptor-dependent and receptor-independent degradation of low density lipoprotein in normal rabbits and in receptor-deficient mutant rabbits. J Biol Chem 1982;257:7994-8000. [Abstract/Free Full Text]
7. Spady DK, Bilheimer DW, Dietschy JM. Rates of receptor-dependent and -independent low density lipoprotein uptake in the hamster. Proc Natl Acad Sci U S A 1983;80:3499-3503. [Abstract/Free Full Text]
8. Brown MS, Dana SE, Goldstein JL. Regulation of 3-hydroxy-3-methylglutaryl coenzyme A reductase activity in human fibroblasts by lipoproteins. Proc Natl Acad Sci U S A 1973;70:2162-2166. [Abstract/Free Full Text]
9. Goldstein JL, Brown MS. Binding and degradation of low density lipoproteins by cultured human fibroblasts. Comparison of cells from a normal subject and from a patient with homozygous familial hypercholesterolemia. J Biol Chem 1974;249:5153-5162. [Abstract/Free Full Text]
10. Bilheimer DW, Ho YK, Brown MS, Anderson RG, Goldstein JL. Genetics of the low density lipoprotein receptor. Diminished receptor activity in lymphocytes from heterozygotes with familial hypercholesterolemia. J Clin Invest 1978;61:678-696.
11. Chait A, Henze K, Mazzone T, Jensen M, Hammond W. Low density lipoprotein receptor activity in freshly isolated human blood monocytes and lymphocytes. Metabolism 1982;31:721-727. [Web of Science][Medline] [Order article via Infotrieve]
12. Traill KN, Böck G, Winter U, Hilchenbach M, Jürgens G, Wick G. Simple method for comparing large numbers of flow cytometry histograms exemplified by analysis of the CD4 (T₄) antigen and LDL receptor on human peripheral blood lymphocytes. J Histochem Cytochem 1986;34:1217-1221. [Abstract]
13. Schmitz G, Brüning T, Kovacs E, Barlage S. Fluorescence flow cytometry of human leukocytes in the detection of LDL receptor defects in the differential diagnosis of hypercholesterolemia. Arterioscler Thromb 1993;13:1053-1065. [Abstract/Free Full Text]
14. Løhne K, Urdal P, Leren TP, Tonstad S, Ose L. Standardization of a flow cytometric method for measurement of low-density lipoprotein receptor activity on blood mononuclear cells. Cytometry 1995;20:290-295. [Web of Science][Medline] [Order article via Infotrieve]
15. Maczek C, Recheis H, Böck G, Stulnig T, Jürgens G, Wick G. Comparison of low density lipoprotein uptake by different human lymphocyte subsets: a new method using double-fluorescence staining. J Lipid Res 1996;37:1363-1371. [Abstract]
16. Verhoeye FR, Descamps O, Husson B, Hondekijn JC, Ronveaux-Dupal MF, Lontie JF, Heller FR. An improved method for detection of low density lipoprotein receptor defects in human T lymphocytes. J Lipid Res 1996;37:1377-1384. [Abstract]
17. Böyum A. Isolation of mononuclear cells and granulocytes from human blood. Isolation of mononuclear cells by one centrifugation, and of granulocytes by combining centrifugation and sedimentation at 1 g. Scand J Clin Lab Invest 1968;Suppl 97:77-89.
18. Christensen M. Some concepts and principles of clinical test evaluation. Clinical test evaluation. Basic concepts. Scand J Clin Lab Invest 1992;Suppl 208:13-29.
19. Westgard JO, Groth T, de Verdier CH. Principles for developing improved quality control procedures. Scand J Clin Lab Invest 1984;Suppl 172:19-41.
20. Beisiegel U, Schneider WJ, Goldstein JL, Anderson RG, Brown MS. Monoclonal antibodies to the low density lipoprotein receptor as probes for study of receptor-mediated endocytosis and the genetics of familial hypercholesterolemia. J Biol Chem 1981;256:11923-11931. [Abstract/Free Full Text]
21. Ho YK, Brown S, Bilheimer DW, Goldstein JL. Regulation of low density lipoprotein receptor activity in freshly isolated human lymphocytes. J Clin Invest 1976;58:1465-1474.
22. Cuthbert JA, East CA, Bilheimer DW, Lipsky PE. Detection of familial hypercholesterolemia by assaying functional low-density-lipoprotein receptors on lymphocytes. N Engl J Med 1986;314:879-883. [Abstract]
23. Fogelman AM, Seager J, Hokom M, Edwards PA. Separation of and cholesterol synthesis by human lymphocytes and monocytes. J Lipid Res 1979;20:379-388. [Abstract]