Clinical Chemistry 45: 1974-1980, 1999;
(Clinical Chemistry. 1999;45:1974-1980.)
© 1999 American Association for Clinical Chemistry, Inc.
Dynamic Reaction in a Homogeneous HDL-Cholesterol Assay Visualized by Electron Microscopy
Akira Kondo1,a,
Yoshinori Muranaka2,
Isao Ohta2 and
Takashi Kanno1
1
Department of Laboratory Medicine and
2
Central Laboratory for Ultrastructure Research, Hamamatsu University School of Medicine, 3600 Handa-cho, Hamamatsu City, 431-3192 Japan.
a Author for correspondence. Fax 81-53-435-2794; e-mail akikondo{at}hama-med.ac.jp
 |
Abstract
|
|---|
Background: Measurement of HDL-cholesterol (HDL-C) by
homogeneous assays with automated analyzers is replacing
precipitation methods. However, in this reaction-type assay,
interactions between the reagents and lipoproteins remain unknown.
Methods: Electron microscopy was used to investigate the
reactions in a homogeneous HDL-C assay. Negative staining with 10
g/L uranyl acetate was performed for lipoprotein visualization
by electron microscopy. Observations of the interactions between
lipoproteins and the reagents of a polyanion-polymer/detergent assay
were achieved by cooling the reaction mixture in ice water. This
treatment also allowed observation of the time course of the reaction.
Results: In the first-reagent reaction (polyanion-polymer), every
lipoprotein aggregated almost completely. In the second-reagent
reaction (enzymes and detergent), only HDL in the lipoprotein
aggregates was selectively resolved and reacted enzymatically. Reagent
1 contains two important substances: polyanion and synthetic polymer.
Using x-ray microanalysis, we confirmed that aggregation of
lipoproteins in the first reaction occurred through interaction with
the phosphotungstate of the polyanion.
Conclusion: Electron microscopy morphologically revealed the
dynamic reaction in a homogeneous HDL-C assay.
|
Introduction
|
|---|
Many disorders involve qualitative and quantitative abnormalities
of lipoprotein metabolism in sera. Reduced serum HDL leads to the
development of risk for coronary heart disease (1). Singh et
al. (2) reported that diabetes mellitus causes normal
LDL to change qualitatively of small, dense LDL. These
lipoproteins are analyzed by determining their physicochemical
properties because the lipoprotein classes differ in density, particle
size, and charge.
Because of the risk of coronary heart disease, it is very important to
measure HDL-cholesterol (HDL-C) in human sera. HDL-C has been measured
mainly by chemical precipitation and enzymatic detection in combination
(3)(4)(5)(6)(7). That is, HDL-C generally is quantified as the
cholesterol remaining in the supernate after chemical precipitation and
sedimentation of other lipoproteins. The mechanism by which HDL-C
interacts with precipitation reagents is not yet completely clear.
HDL-C measurement by homogeneous assay is replacing the methods
described above (8)(9)(10)(11) because this assay is applicable to
automated analyzers. Routine screening of large populations has become
possible as well. However, in this reaction-type assay, the phenomena
that occur between the lipoproteins and the reagents remain unknown. In
this study, we used a morphological approach via electron microscopy to
elucidate the interaction between various lipoproteins and the reagents
of a homogeneous HDL-C assay.
|
Materials and Methods
|
|---|
reagents
The polyanion-polymer/detergent assay as a homogeneous HDL-C assay
kit was purchased from Daiichi Pure Chemicals. This kit is
constructed with a first reagent (reagent 1) and a second reagent
(reagent 2). Monoclonal antibody against lipoprotein(a) [Lp(a)] was
also obtained from Daiichi. Formyl-Cellulofine, used for affinity
chromatography, was purchased from Seikagaku. The agarose
electrophoresis gel film was from Helena Laboratories.
lipoprotein preparation
Serum lipoproteins were isolated by sequential ultracentrifugation
from pooled human serum. Sequential ultracentrifugation was performed
in a Beckman model L5-65 ultracentrifuge with a SW41TI swinging-bucket
rotor, using Beckman Ultra-Clear 1.4 x 8.9 cm 12-mL centrifuge tubes.
The following method was based on the same principles as the technique
reported by Hatch and Lees (12) with some modifications.
Fresh pooled serum was centrifuged at 51 000 g for 30 min
at 15 °C. The chylomicron-containing fraction (4 mL) was
removed from the supernatant. KBr solution (4 mL; d =
1.006 kg/L) was layered on top of the fraction, which was then
recentrifuged at 200 000g for 18 h at 15 °C. The
VLDL-containing top layer (1 mL) was collected by aspiration.
The infranate (6 mL) was mixed with 3 mL of KBr solution
(d = 1.045 kg/L) adjusted to d = 1.019
kg/L. The preparative fraction (8 mL) was placed in a tube, and 4 mL of
KBr solution (d = 1.019 kg/L) was layered on top of the
fraction. Similarly, after recentrifugation at 200 000g for
20 h at 15 °C, the supernatant (6 mL) was discarded. A 6-mL
aliquot of the bottom fraction was adjusted to d =
1.063 kg/L by mixing with 3 mL of KBr solution (d =
1.151 kg/L). Subsequently, 4 mL of KBr solution (d =
1.063 kg/L) was layered on the top 8 mL of this preparative fraction,
which was recentrifuged under the same conditions as the third
centrifugation. Recovery of LDL (1 mL) from the top of the tube was
performed. The infranate (4 mL) was adjusted to d =
1.21 kg/L by the addition of an equal volume of KBr solution
(d = 1.357 kg/L) and layered with 4 mL of KBr solution
(d = 1.21 kg/L). Separation of HDL was then carried out
by centrifugation at 200 000g for 40 h at 15 °C.
The supernatant was recovered as the HDL fraction. All salt solutions
contained 1 mmol/L EDTA. All recovered fractions of VLDL, LDL, and HDL
were dialyzed extensively against phosphate-buffered saline [25 mmol/L
phosphate buffer (pH 7.4) containing 150 mmol/L NaCl]. After
dialysis, each fraction was filtered through a 0.2 µm filter. In the
case of HDL fraction, contaminated Lp(a) was excluded by immunoaffinity
chromatography.
immunoaffinity chromatography
Monoclonal antibody against Lp(a) was covalently coupled to
Formyl-Cellulofine at a rate of ~7 mg/g of wet gel according to the
manufacturer's instructions. The immunosorbent (1 mL) was packed into
a column and equilibrated with five column volumes of
phosphate-buffered saline. Of the HDL fraction dialyzed, 1 mL was
applied to the column. Pure HDL was recovered as a nonbinding protein.
The purity of lipoprotein fractions and the absence of Lp(a)
contamination in the HDL fraction were evaluated by agarose gel
electrophoresis.
measurement in an automated analyzer
This homogeneous assay kit for the direct measurement of HDL-C was
suited for use in a Hitachi 7250 automated analyzer. Consequently,
lipoprotein fractions (3 µL) and reagent 1 (300 µL) were mixed and
incubated for 5 min at 37 °C. Reagent 2 (100 µL) was then added,
and the mixture was incubated for 5 min, during which the absorbance
was monitored every 12 s at 660 and 546 nm during the reaction.
negative stain electron microscopy
Before observation by electron microscopy, a portion of each
lipoprotein fraction was reacted with either reagent 1 alone or with
reagents 1 plus 2 in a homogeneous HDL-C assay kit. For example, when
the first reaction was monitored, each lipoprotein fraction (6 µL) at
certain concentrations was mixed with reagent 1 (600 µL) in a test
tube and incubated for 4.5 min at 37 °C, after which the reaction
tube was transferred to ice water and left for 0.5 min. This reaction
mixture was provided as a sample for negative staining. For analysis of
the second reaction, after the first reaction for 5 min at 37 °C,
the next step was started by the addition of reagent 2 (200 µL) to
the tube. Under the same conditions as the first reaction, the next
reaction was incubated for various time periods, and then the reaction
tube was transferred to ice water and left for 0.5 min. As a control
reaction, this same procedure was carried out in 150 mmol/L NaCl
solution exchanged for either reagent 1 or reagent 2. These reaction
samples, cooled by ice water, were then subjected to the method below.
A drop of the cooled sample was placed immediately onto a
Formvar/carbon-coated grid and immobilized for 0.5 min. Excess fluid
was removed with filter paper, and the specimen on the grid was dried
at room temperature for ~1 min. Negative staining with a drop of 10
g/L uranyl acetate in distilled water was performed. After 0.5
to 2 min, the drop of staining solution was again removed with filter
paper. The specimen was placed immediately into the microscope specimen
chamber and observed with a JEM-1220 electron microscopy at 80 kV.
x-ray microanalysis
For x-ray microanalysis of the reaction product, the specimen was
negatively stained under the same conditions as above. The analysis of
the reaction sample was performed with a JEM-200CX equipped with a
7025J (Kevex) for x-ray microanalysis by energy dispersion
spectrometry. The microscope was operated in transmission electron
microscopy (TEM) mode at 160 kV.
other analytical methods
The cholesterol concentrations in lipoprotein fractions were
determined enzymatically. Protein contents in lipoprotein fractions
were estimated using the method of Lowry et al. (13).
|
Results
|
|---|
Using each lipoprotein fraction at the same cholesterol
concentration as that observed under electron microscopy, we confirmed
that the reaction provided by the homogeneous HDL-C assay kit was
specific for HDL (Fig. 1
). The assay conditions that applied to the automated analyzer
almost coincided with those for the electron microscopic analysis as
described below.
reaction between each lipoprotein and reagent 1
We first examined the response of each lipoprotein to reagent 1 in
the kit. When each lipoprotein was incubated in 150 mmol/L NaCl
solution as a control reaction for the indicated time at 37 °C, none
changed in appearance (Fig. 2
, a-1 to a-3). This means that there were no artificial changes
that occurred in the lipoproteins via the experimental procedure for
observation by electron microscopy. In the case of reaction with
reagent 1, HDL particles coexisted in both aggregated and nonaggregated
forms (Fig. 2b
-1). That is, some clumps were >0.2 µm in diameter and
some were monomeric. Both LDL and VLDL responded to reagent 1 by
extensive aggregation (Fig. 2
, b-2 and b-3). Most clumps were >1 µm
in diameter, and many were larger than those from HDL. The clumps
were surprisingly similar in size considering that they were
constructed from hundreds of LDL particles.
View larger version (138K):
[in this window]
[in a new window]
|
Figure 2. The first reaction, using each lipoprotein and either 150
mmol/L NaCl (a-1 to a-3) or reagent 1
(b-1 to b-3), observed by electron
microscopy of negatively stained preparations.
The lipoprotein cholesterol concentrations were as follows: 445 mg/L
for HDL (a-1 and b-1), 1748 mg/L for LDL
(a-2 and b-2), and 1101 mg/L for
VLDL (a-3 and b-3); for an
explanation, see Materials and Methods. For electron
microscopic examination, cooled samples of HDL were diluted fourfold
with cooled 150 mmol/L NaCl or reagent 1 before placement on a grid,
but other lipoproteins were not diluted. The specimens were observed
with a JEM-1220 electron microscope.
|
|
reaction between each lipoprotein and reagent 2
A control of the second reaction was carried out by exchanging
reagent 2 for 150 mmol/L NaCl solution after the first reaction (Fig. 3
, a-1 to a-3). Every lipoprotein aggregate (HDL, LDL, and
VLDL) formed in the first stage was maintained during the second stage
with the NaCl solution. The responses of both LDL and VLDL to the
action of reagent 2 were similar to those of control reactions (Fig. 3
, b-2 and b-3). The HDL response differed from the responses of the other
lipoproteins: When clumps of HDL produced in the first reaction were
mixed with reagent 2 under the conditions indicated, HDL disappeared
from the grid (Fig. 3b
-1).
View larger version (117K):
[in this window]
[in a new window]
|
Figure 3. The second reaction, using each lipoprotein and either 150
mmol/L NaCl (a-1 to a-3) or reagent 1
(b-1 to b-3), observed by electron
microscopy of negatively stained preparations.
The lipoprotein cholesterol concentrations were as follows: 445 mg/L
for HDL (a-1 and b-1), 1748 mg/L for LDL
(a-2 and b-2), and 1101 mg/L for
VLDL (a-3 and b-3); for an
explanation, see Materials and Methods. For electron
microscopic examination, cooled samples of HDL were diluted fourfold
with cooled 150 mmol/L NaCl or reagents 1 plus 2 (3:1) before placement
on a grid, but other lipoproteins were not diluted. The specimens were
observed by electron microscopy.
|
|
interactions between reagent 1, hdl, and vldl
Before studying the interaction between reagent 1, HDL, and VLDL,
we confirmed that there was no interaction between HDL and VLDL in 150
mmol/L NaCl solution under the above-mentioned conditions (Fig. 4
a). Because these lipoproteins did not react with each other,
the first reaction observed under electron microscopy was that between
these lipoproteins and reagent 1 (Fig. 4b
). Both HDL and VLDL assembled
almost completely into HDL-VLDL aggregates that were >1 µm in
diameter. The surfaces of these clumps were covered by HDL particles
such that an overwhelming number of HDL particles (cholesterol, 339
mg/L in sample solution) in comparison with VLDL particles (551
mg/L) existed in the reaction mixture. All classes of lipoproteins
could aggregate during the first reaction.
View larger version (105K):
[in this window]
[in a new window]
|
Figure 4. The interaction between either 150 mmol/L NaCl
(a) or reagent 1 (b), HDL, and VLDL,
observed by electron microscopy of negatively stained preparations.
Concentrations of cholesterol in HDL and VLDL were 339 and 551
mg/L, respectively.
|
|
interactions between reagent 2, hdl, and vldl
Complexes from the first reaction were maintained during
incubation with 150 mmol/L NaCl under the same conditions described for
the second reaction (data not shown), indicating that reagent 2 caused
the changes in the complexes. The time course for the second reaction
was investigated by electron microscopy (Fig. 5
) after reagent 2 instead of NaCl was added to the
complexes. The complexes began to degrade into small clumps immediately
after removal from the ice water, even before incubation at 37 °C
(Fig. 5a
). Furthermore, degradation of the clumps was considerable at
0.5 min (Fig. 5b
). After 2.5 min, the surfaces of the VLDL particles
inside the clumps began to be exposed (arrowheads in Fig. 5c
). The
aggregated HDL-VLDL complexes had disappeared from the grid by the end
of the longest incubation period (Fig. 5d
), leaving only
aggregates of VLDL particles. Therefore, among all the clumps of
lipoproteins that formed in the first reaction, only HDL particles
resolved during the second reaction.
View larger version (201K):
[in this window]
[in a new window]
|
Figure 5. Monitoring of the interactions between reagent 2, HDL, and
VLDL for various time periods by electron microscopy of negatively
stained preparations.
Concentrations of cholesterol in HDL and VLDL were 339 and 551
mg/L, respectively. After the first reaction, reagent 2 was
added to the reaction mixture. The level of degradation of aggregates
at 0 (a), 0.5 (b), 2.5
(c), and 4.5 (d) min during the second
reaction are shown by electron micrographs. Arrowheads
in c show the exposed surfaces of VLDL particles inside
clumps.
|
|
x-ray microanalysis of the first reaction
Specifications in the HDL-C assay kit indicate that reagent 1
includes both a polyanion and a synthetic polymer; phosphotungstate is
used as the polyanion. We clarified a role of polyanion in the reaction
by x-ray microanalysis.
Aggregates of VLDL particles formed by reagent 1 were subjected to
x-ray microanalysis (Fig. 6
). Three kinds of atoms were found. One was copper derived from
the grid. The second was uranium from the negative staining reagent.
The third was tungsten, probably from reagent 1. Thus, we confirmed
that phosphotungstate in reagent 1 is contained in aggregates of VLDL.
Neither magnesium nor phosphorus could be detected, probably because
they were below the detection limits.
View larger version (96K):
[in this window]
[in a new window]
|
Figure 6. Resolution of the first-reaction products by x-ray
microanalysis.
Aggregates generated from VLDL (1101 mg/L cholesterol) in reagent 1
were provided for x-ray microanalysis. (a), area without
VLDL particles used as a control for b;
(b) aggregates of VLDL particles. The x-ray
microanalysis charts for both areas a (upper
right) and b (lower right) show
copper (Cu) and uranium (U), which were
derived from a grid and negative staining reagent, respectively, and
which were frequently detected, whereas tungsten (W) was
detected only in the chart of b.
|
|
|
Discussion
|
|---|
HDL-C measurement methods have been changing from the
precipitation type to the above type of a homogeneous assay
(8)(9)(10)(11). However, the phenomena involving
the interactions that occur between lipoproteins and the
re-agents have been unclear. To investigate these phenomena, we used
electron microscopy to visualize the dynamic reaction in a homogeneous
assay for HDL-C.
When we observed the lipoprotein fractions by electron microscopy, the
outlines of lipoproteins were wider than usual. Lipoproteins incubated
at 37 °C might become labile because the transition temperature of
lipoproteins is nearly at 37 °C (14). To overcome this
problem, we cooled the reaction mixture in ice water immediately after
the 37 °C incubation. With this step, the outlines of the particles
appeared unequivocally in the electron microscope. Furthermore, we
found that cooling the reaction mixture stops the reaction. As a
result, it became possible to observe the time course of the reaction.
According to the manufacturer's specifications, there are two
effective substances in reagent 1, the polyanion and the synthetic
polymer. The instructions state specifically that LDL-polymer-polyanion
and HDL-polymer complexes are formed by the addition of polymer and
polyanion in the first reaction and that the HDL structure is broken
down by a detergent in the second reaction. Our results confirmed this.
Almost every lipoprotein aggregated completely in the first reaction.
X-ray microanalysis revealed that phosphotungstate (the polyanion in
reagent 1) acts as an aggregating reagent. We assume that lipoprotein
complexes formed in the first reaction resulted from the action of the
polyanion in reagent 1. However, exactly what roles the polyanion and
synthetic polymer in reagent 1 play in the selective reaction remains
unknown.
We also confirmed that only the HDL in the aggregates selectively
resolved in the second reaction and that it reacted enzymatically, as
indicated in the manufacturer's specifications. In detail, the
detergent in reagent 2 selectively resolved HDL on the surface of the
aggregates, in which HDL was denuded from the inside (Fig. 5
). This
mechanism was shown to involve the combination with HDL and VLDL.
Because this combination has more distinct particle sizes than with HDL
and LDL, it is easy to morphologically show the interaction between HDL
and the other lipoproteins. Consequently, it was suggested that the
detergent in reagent 2 gradually permeated through localized HDL of
clumps to their core and completely degraded all HDL. However, the
detergent in reagent 2 alone could not exclusively react with HDL in
all lipoproteins (data not shown). For a complete reaction, it
was necessary to pretreat the lipoproteins with reagent 1.
Electron microscopy provided useful information for understanding the
reaction in this homogeneous HDL-C assay. More recently, a homogeneous
method for the quantification of LDL-cholesterol has been reported
(15). Analysis by electron microscopy would also be useful
for investigating the reaction in this LDL-cholesterol assay.
|
Acknowledgments
|
|---|
The study was made possible by a grant from the Diagnostics
Research Laboratories, Daiichi Pure Chemicals Co. We thank Youko
Kumakiri for excellent technical support.
|
References
|
|---|
-
Gordon T, Castelli WP, Hjortland MC, Kannel WB, Dawber TR. High density lipoproteins a protective factor against coronary heart disease. The Framingham Study. Am J Med 1977;62:707-714.
[Web of Science][Medline]
[Order article via Infotrieve]
-
Singh ATK, Rainwater DL, Haffner SM, Van deBerg JL, Shelledy WR, Jr, Moore PH, et al. Effect of diabetes on lipoprotein size. Arterioscler Thromb Vasc Biol 1995;15:1805-1811.
[Abstract/Free Full Text]
-
Sirowej H, Krieger R, Kattermann R. Carbohydrate content of lipoproteins after sequential precipitation with phosphotungstate magnesium chloride. J Clin Chem Biochem 1982;20:875-881.
-
Burstein M, Scholnick HR, Morfin R. Rapid method for the isolation of lipoproteins from human serum by precipitation with polyanions. J Lipid Res 1970;11:583-595.
[Abstract]
-
Warnick GR, Benderson J, Albers JJ. Dextran sulfate-Mg2+ precipitation procedure for quantitation of high-density lipoprotein cholesterol. Clin Chem 1982;28:1379-1388.
[Free Full Text]
-
Viikari J. Precipitation of plasma lipoproteins by PEG-6000 and its evaluation with electrophoresis and ultracentrifugation. Scand J Clin Lab Investig 1976;36:265-268.
[Web of Science][Medline]
[Order article via Infotrieve]
-
Warnick GR, Albers JJ. A comprehensive evaluation of the heparin–manganese precipitation procedure for estimating high density lipoprotein cholesterol. J Lipid Res 1978;19:65-76.
[Abstract]
-
Sugiuchi H, Uji Y, Okabe H, Irie T, Uekama K, Kayahara N, et al. Direct measurement of high-density lipoprotein cholesterol in serum with polyethylene glycol-modified enzymes and sulfated
-cyclodextrin. Clin Chem 1995;41:717-723.
[Abstract/Free Full Text]
-
Harris N, Galpchian V, Thomas J, Iannotti E, Law T, Rifai N. Three generations of high-density lipoprotein cholesterol assays compared with ultracentrifugation/dextran sulfate-Mg2+ method. Clin Chem 1997;43:816-823.
[Abstract/Free Full Text]
-
Simó JM, Castellano I, Ferré N, Joven J, Camps J. Evaluation of a homogeneous assay for high-density lipoprotein cholesterol: limitations in patients with cardiovascular, renal, and hepatic disorders. Clin Chem 1998;44:1233-1241.
[Abstract/Free Full Text]
-
Hino K, Nakamura M, Nakanishi K, Manabe M. A new method for the homogeneous assay of serum HDL-Cholesterol [Abstract]. Clin Chem 1996;42:S299.
-
Hatch FT, Lees RS. Practical methods for plasma lipoprotein analysis. Adv Lipid Res 1968;6:1-68.
[Medline]
[Order article via Infotrieve]
-
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265-275.
[Free Full Text]
-
Deckelbaum RJ, Shipley GG, Small DM, Lees RS, George PK. Thermal transitions in human plasma low density lipoproteins. Science 1975;190:392-394.
[Abstract/Free Full Text]
-
Rifai N, Iannotti E, DeAngelis K, Law T. Analytical and clinical performance of a homogeneous enzymatic LDL-cholesterol assay compared with the ultracentrifugation-dextran sulfate-Mg2+ method. Clin Chem 1998;44:1242-1250.
[Abstract/Free Full Text]
The following articles in journals at HighWire Press have cited this article:
|
|
|
|
 
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]
|
|
|
Top
Abstract
Introduction
), LDL (
), and VLDL (
)
were 445, 1748, and 1101 mg/L, respectively.