|
|
||||||||
Articles |
Laboratory of Chemistry, College of Liberal Arts and Sciences, Tokyo Medical and Dental University, 2-8-30 Kohnodai, Ichikawa-shi, Chiba 272, Japan.
1
School of Allied Health Science, Tokyo Medical and Dental University, 1-5-45, Yushima, Bunkyo-ku, Tokyo 113, Japan.
a Author for correspondence. Fax 81-473-74-0630; e-mail okazaki.cul{at}cul.tmd.ac.jp
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Recently, we reported a simple, rapid, and accurate analysis of cholesterol in lipoproteins particularly suitable for HDL-cholesterol (HDL-C) assay by using an HPLC with an improved gel permeation column and eluent for lipoprotein separation (TSKgel Lipopropak and TSKeluent LP-1, Tosoh) (10)(11)(12)(13)(14).1 With this new HPLC method, we have now evaluated the values of HDL-C measured by precipitation and direct methods, which are popular in Japanese clinical laboratories. Further, the effect of MgCl2 concentration in the precipitation reagent on fractionation of HDL was examined by analysis of the supernatants by using the HPLC method.
| Materials and Methods |
|---|
|
|
|---|
precipitation method for hdl-c
A commercial precipitation reagent (Determiner HDL, Kyowa
Medex) composed of 7.2 g/L sodium phosphotungstate, 0.3 g/L dextran
sulfate, and 44 mmol/L MgCl2 was used as a standard
fractionation reagent for HDL-C assay. We also prepared a series of
reagents with the same compositions as commercial reagents except
MgCl2 at the following concentrations: 7.3, 12.2, 14.7,
17.1, 19.6, 22, 24.4, 29.3, and 44 mmol/L. Equal volumes of serum and
precipitation reagent were mixed, and centrifuged at 1500g
for 10 min. The cholesterol values in supernatants were measured by an
automated chemical analyzer (Hitachi 7150) with a certified reference
serum [509AIJ, total cholesterol (TC) = 5.741 mmol/L, Kyowa Medex] as
a calibrator.
direct method for hdl-c
A commercial reagent kit (Determiner HDL-C, Kyowa Medex) for
direct HDL-C assay (15) that contained cholesterol
esterase and cholesterol oxidase modified by polyethylene glycol,
sulfated
-cyclodextrin, dextran sulfate, and MgCl2 was
used in this experiment. The concentration of HDL-C was measured by the
automated chemical analyzer with a certified reference serum (115AEK,
HDL-C = 1.361 mmol/L, Kyowa Medex) as a calibrator.
hplc method for hdl-c
A routine HPLC apparatus (CCP&8010, Tosoh) with two connected
columns (TSKgel Lipopropak, 7.5 mm in diameter, 300 mm in length) made
specially for lipoprotein analysis, together with the TSKeluent LP-1
(Tosoh), also made for lipoprotein analysis, were used at a flow rate
of 0.6 mL/min. Each 20 µL of serum diluted with an equal volume of
buffer (20 mmol/L HEPES, 0.15 mol/L NaCl, pH 7.5) or the supernatant of
the mixture of an equal volume of serum and precipitation reagent was
applied to the HPLC. The effluent from the columns was mixed with
reagent 1 [30 mmol/L 3-(N-morpholino)propanesulfonic acid
(MOPS) buffer (pH 6.75), containing 3000 U/L ascorbic acid oxidase,
1000 U/L cholesterol oxidase, 2500 U/L peroxidase, 30 mmol/L
N-ethyl-N-(3-methyl
phenyl)-N'-succinylethylenediamine] at a flow rate of 0.225
mL/min and reagent 2 [30 mmol/L MOPS buffer (pH 6.75) containing 5000
U/L cholesterol oxidase, 25 000 U/L peroxidase, 2.2 mmol/L
4-aminoantipyrine] at a flow rate of 0.075 mL/min in the reaction tube
(Teflon, 0.4 mm in diameter and 7.5 m in length) at a temperature
of 45 °C and the amount of a colored substance produced was measured
at 550 nm. TC and HDL-C were calculated automatically from the total
area and HDL area, respectively, on the HPLC pattern monitored by
cholesterol with a certified reference serum (WCHL 952 M, TC =
4.896 mmol/L, Health Care Technology Foundation, Standard Reference
Center, Kawasaki, Japan) as a calibrator. The linearity and detection
limit for the HDL-C assay with the HPLC method were examined by
applying 20 µL of diluted (20 mmol/L HEPES, 0.15 mol/L NaCl, pH 7.5)
serum samples. The linearity was confirmed in the HDL-C concentration
range between 0.074 and 2.42 mmol/L, and the detection limit was as low
as 0.5 mol/L. Identification of the HDL peak was done by comparing the
elution pattern of the HDL fraction separated by ultracentrifugation.
statistics
Quantitative variables are reported as mean ± SD. Paired
Student's t-test was used to determine the
significance levels of the differences. Relations between variables
were evaluated by Spearman's correlation. P
<0.05 was considered significant.
| Results |
|---|
|
|
|---|
|
comparison of hdl-c values assayed by the three methods
Within-run precision (CV) of the HPLC, precipitation, and direct
methods with a sample with a normal TG value (TC = 5.33 mmol/L,
TG = 0.97 mmol/L) were 0.55%, 1.95%, and 0.56%, respectively
(Table 1
).
|
Serum HDL-C of 74 subjects (TC = 5.36 ± 1.43 mmol/L, TG
= 1.64 ± 2.35 mmol/L) was measured by the three methods. The
average values of HDL-C by the three methods are summarized separately
for the three groups classified by the TG values (Table 2
).
|
The values of HDL-C both by precipitation and direct methods showed
good correlations with those by the HPLC method, as shown in Fig. 2
A. However, the paired t-test (n =
74) revealed that the values by the precipitation method were
significantly (P <0.001) lower than those by
HPLC. Those by the direct method, on the contrary, were slightly but
significantly higher (P <0.02) than those by the
HPLC method. Moreover, the differences of HDL-C between the
precipitation method and the HPLC method (HDL-C) showed a significant
negative correlation (r = -0.755,
P <0.001) with HDL-C values by the HPLC method,
but in the case of the direct method, no significant correlation
(r = 0.218) was observed, as shown in Fig. 2B
.
|
effect of mgcl2 concentrations on separation of
hdl
The effect of MgCl2 in the precipitation reagent
on the separation of HDL was examined by using a series of reagents
with different MgCl2 concentrations. As shown in Fig. 1
, peak height for the lipoproteins larger than HDL (peaks 1, 2, and 3) in
the elution patterns for each serum sample (pattern a) gradually
decreased in supernatants proportional to the increase in
MgCl2 concentration, from pattern b, 7.3 mmol/L
MgCl2 reagent to pattern g, 44 mmol/L (commercial)
MgCl2 reagent. Moreover, MgCl2 concentrations
enough to precipitate non-HDL fractions depended on TG values of each
subject. As for all subjects, including those with high TG values,
peaks of non-HDL lipoproteins were not observed in the supernatants at
22 mmol/L or 44 mmol/L (commercial) MgCl2 reagents (Fig. 1
).
Percentages of cholesterol values in the HDL and non-HDL fractions in
supernatants relative to the respective values in the original serum
determined by the HPLC method are plotted against concentration of
MgCl2 (mmol/L) in precipitation reagents for two
groups of subjects with normal and high TG values. As for the normal TG
group (TG = 0.84 ± 0.14 mmol/L, n = 5), non-HDL
lipoproteins disappeared at a concentration of 7.3 mmol/L (Fig. 3
A). However, in the high TG group (9.18 ± 0.37 mmol/L,
n = 5), >70% of non-HDL cholesterol in the supernatant at the
same concentration of MgCl2 and non-HDL fractions decreased
rapidly to zero as the concentration of MgCl2 was increased
to 22 mmol/L, corresponding to 50% of that contained in commercial
reagent (Fig. 3B
).
|
The cholesterol values in HDL fractions in the supernatants of sera
with normal and high TG values were almost the same, between 7.3 and
24.4 mmol/L MgCl2 concentrations. The values in both
groups, however, decreased gradually as the concentration of
MgCl2 exceeded 24.4 mmol/L (Fig. 3A
, B).
recovery of hdl fraction in supernatants
The concentration of cholesterol in the HDL and non-HDL fractions
in the supernatants by 44 mmol/L (commercial) and 22 mmol/L
MgCl2 reagents were measured by HPLC method for serum
samples from 55 subjects (TC = 5.83 ± 1.18 mmol/L, TG =
3.12 ± 3.63 mmol/L); those in non-HDL fractions (n = 55)
were negligible (0.0024 ± 0.0068 and 0.0099 ± 0.0165
mmol/L, respectively). On the contrary, the cholesterol value in the
HDL fraction was higher in the supernatants by 22 mmol/L
MgCl2 reagent (1.45 ± 0.86 mmol/L) than those by 44
mmol/L (commercial) reagent (1.58 ± 1.00 mmol/L). The recovery of
the HDL fraction in the supernatant was calculated as a percentage of
cholesterol values in the HDL fraction of the supernatant to those in
the HDL fraction of original serum, both obtained by the HPLC method.
The results obtained by both reagents are plotted against HDL-C values
of original serum measured by the HPLC method (Fig. 4
). The average recovery of HDL fraction in the supernatant by 44
mmol/L (commercial) reagent was 91.5% ± 4.9%, and that by 22 mmol/L
MgCl2 reagent 97.9% ± 3.6%.
|
A significant negative correlation (r =
-0.45, P <0.001) was found between the
percentage recovery of the HDL fraction in the supernatants by
commercial reagent and HDL-C values of original serum (Fig. 4
). On the
contrary, no such correlation was found in the supernatants treated
with 22 mmol/L MgCl2 reagent.
| Discussion |
|---|
|
|
|---|
We reported (10)(11)(12) that HPLC with an improved gel permeation column and new eluent has become a reliable determination method for HDL-C assay, with high precision (CV <0.5%) due to complete separation of the HDL peak from non-HDL lipoproteins, as well as elimination of nonspecific adsorption of lipoproteins to column materials. The elimination of nonspecific adsorption was confirmed by >99.8% recovery of lipoproteins from the column (10). The ultracentrifugation method is time consuming and also requires correction of the data on the basis of percentage recovery and contamination by apo B lipoproteins (17) to obtain accurate values of HDL-C, whereas the HPLC method is free from such factors.
Previously, we reported (13)(14) that the HDL-C values assayed by the precipitation method with different commercial reagent kits available in Japan were always lower than those by the HPLC method. In line with this assertion, Chiba et al. (18) reported that the precipitation reagent containing dextran sulfate, sodium phosphotungstate, and MgCl2 readily precipitates apo E-rich HDL as compared with 130 g/L polyethylene 6000. We realized that one of the reasons for lower values by commercial reagent as opposed to the HPLC method was attributable to apo E-rich HDL, often found in large-size HDL, which is readily precipitated. We could not, however, explain all the reasons for the lower values obtained with the precipitation method. From the present examinations of the supernatants obtained by the precipitation method, we speculate that the commercial reagents precipitate part of HDL with normal size, even in routine samples, because of the high concentration of MgCl2, which was necessary to precipitate non-HDL lipoproteins completely in samples with high TG values.
The effects of concentration of metal ions in the reagents for precipitation methods on the HDL-C values were already pointed out by Warnick et al. (17) by comparison with the ultracentrifugation method, and the present results confirm their results.
Unlike the precipitation method, the values of HDL-C measured by the direct method were very close to those by the HPLC method, despite a significant difference by statistics. Okamoto et al. (19) reported, using agarose gel chromatography, that the apo A-I peak was completely measured by the direct method, whereas partly by the precipitation method. This finding was also confirmed by our present work. The difference between HPLC and the direct method is suspected to be caused by the difference in calibrator, and for this the direct method has much advantage with respect to accuracy compared with the precipitation method.
We believe that the present HPLC method is most reliable for HDL-C assay among the methods available at present. The HPLC method is especially useful as a comparative method to evaluate other methods, because it gives not only quantitative information with excellent precision but also qualitative information through the elution patterns.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
-cyclodextrin. Clin Chem 1995;41:717-723.
The following articles in journals at HighWire Press have cited this article:
![]() |
M. Okazaki, S. Usui, A. Fukui, I. Kubota, and H. Tomoike Component Analysis of HPLC Profiles of Unique Lipoprotein Subclass Cholesterols for Detection of Coronary Artery Disease Clin. Chem., November 1, 2006; 52(11): 2049 - 2053. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Inoue, Y. Arai, H. Kurihara, T. Kita, and T. Sawamura Overexpression of Lectin-Like Oxidized Low-Density Lipoprotein Receptor-1 Induces Intramyocardial Vasculopathy in Apolipoprotein E-Null Mice Circ. Res., July 22, 2005; 97(2): 176 - 184. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Okazaki, S. Usui, M. Ishigami, N. Sakai, T. Nakamura, Y. Matsuzawa, and S. Yamashita Identification of Unique Lipoprotein Subclasses for Visceral Obesity by Component Analysis of Cholesterol Profile in High-Performance Liquid Chromatography Arterioscler Thromb Vasc Biol, March 1, 2005; 25(3): 578 - 584. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. R. Warnick, M. Nauck, and N. Rifai Evolution of Methods for Measurement of HDL-Cholesterol: From Ultracentrifugation to Homogeneous Assays Clin. Chem., September 1, 2001; 47(9): 1579 - 1596. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Usui, M. Nakamura, K. Jitsukata, M. Nara, S. Hosaki, and M. Okazaki Assessment of Between-Instrument Variations in a HPLC Method for Serum Lipoproteins and Its Traceability to Reference Methods for Total Cholesterol and HDL-Cholesterol Clin. Chem., January 1, 2000; 46(1): 63 - 72. [Abstract] [Full Text] [PDF] |
||||
![]() |
M.-L. Arranz-Pena, J. Tasende-Mata, and F. J. Martin-Gil Comparison of two homogeneous assays with a precipitation method and an ultracentrifugation method for the measurement of HDL-cholesterol Clin. Chem., December 1, 1998; 44(12): 2499 - 2505. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M a Simo, I. Castellano, N. Ferre, J. Joven, and J. Camps Evaluation of a homogeneous assay for high-density lipoprotein cholesterol: limitations in patients with cardiovascular, renal, and hepatic disorders Clin. Chem., June 1, 1998; 44(6): 1233 - 1241. [Abstract] [Full Text] [PDF] |
||||
![]() |
M.-J. Lin, C. Hoke, and B. Ettinger Evaluation of Homogeneous High-Density Lipoprotein Cholesterol Assay on a BM/Hitachi 747-200 Analyzer Clin. Chem., May 1, 1998; 44(5): 1050 - 1052. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |