Clinical Chemistry 45: 2158-2163, 1999;
(Clinical Chemistry. 1999;45:2158-2163.)
© 1999 American Association for Clinical Chemistry, Inc.
Endpoint Colorimetric Method for Assaying Total Cholesterol in Serum with Cholesterol Dehydrogenase
Yuzo Kayamoria,
Hiroyuki Hatsuyama,
Tadayoshi Tsujioka,
Masato Nasu and
Yoshiaki Katayama
Department of Clinical Chemistry, National Cardiovascular Center Hospital 5-7-1, Fujishirodai, Suita, Osaka 5658565, Japan.
a Author for correspondence. Fax 81-6-6833-9865; e-mail ykayamor{at}hsp.ncvc.go.jp
 |
Abstract
|
|---|
Background: Various methods are available to measure serum
cholesterol concentrations. Of these, the cholesterol ester hydrolase
(CEH)-cholesterol oxidase-peroxidase chromogenic method is widely used.
However, this method has the disadvantage of interference by reducing
substances. We developed and evaluated an endpoint assay for serum
cholesterol, based on a CEH-cholesterol dehydrogenase (CDH)-ultraviolet
method.
Methods: Cholesterol esters are first hydrolyzed to free
cholesterol by CEH. The free cholesterol is then reduced by CDH to
cholest-4-ene-3-one with the simultaneous production of ß-NADH from
ß-NAD+. At equilibrium, the CDH reaction gives incomplete
conversion of cholesterol to cholest-4-ene-3-one. To overcome this
disadvantage, we added hydrazine monohydrate to the reaction mixture to
remove cholest-4-ene-3-one, which allowed the reaction to proceed to
completion and gave stoichiometric production of ß-NADH from the
reaction of ß-NAD+ with cholesterol.
Results: We tested whether the amount of cholesterol added was
equivalent to the absorbance change of NADH at 340 nm with six aqueous
samples. Recoveries were 97.1–100.3%. The reaction was linear up to
20.28 mmol/L. The mean within-day (n = 20) and between-day (n
= 10) imprecision (CV) was 0.29–0.43% and 0.22–0.61%, respectively.
No interference by bilirubin, hemoglobin, ascorbic acid, and other
reducing agents was observed. The equation obtained in comparison with
the modified Abell-Levy-Brodie-Kendall method was:
y = 0.992x - 0.0058 mmol/L;
r = 0.997; Sy|x = 0.117
mmol/L; n = 50.
Conclusion: This method is an accurate, reliable method for serum
cholesterol analysis and is amenable to automation.
|
Introduction
|
|---|
Various methods have been devised to measure cholesterol in serum.
The Abell-Levy-Brodie-Kendall
(ALBK)1
saponification method (1)(2) is a
precise method, but it is not applicable to many kinds of samples and
is difficult to automate. The cholesterol ester hydrolase-cholesterol
oxidase-peroxidase (CEH-CO-POD) chromogenic method of Allain et al.
(3) is widely used in clinical laboratories but is subject
to interferences from reducing agents such as ascorbic acid, bilirubin,
and reduced glutathione (4). Enzymatic procedures using
NAD(P+)-specific cholesterol dehydrogenase (CDH)
instead of CO have been described (5)(6), but to
date, application of these methods in an analytical setting has been
limited.
Here we report a new enzymatic cholesterol assay that uses CDH from
Nocardia sp. (7)(8). At equilibrium,
the CDH reaction gives incomplete conversion of cholesterol to
cholest-4-ene-3-one. To overcome this disadvantage, we added hydrazine
monohydrate to the reaction mixture to remove cholest-4-ene-3-one,
which allowed the reaction to proceed to completion and gave
stoichiometric production of ß-NADH from the reaction of
ß-NAD+ with cholesterol. The reaction sequence
is shown in Fig. 1
. The data presented suggest that the method is accurate,
simple, and amenable to automation.
|
Materials and Methods
|
|---|
reagents
CEH (EC3.1.1.13; Pseudomonas sp.) and CDH
(Nocardia sp.) were obtained from Amano
Pharmaceutical. Hemoglobin was prepared from lysed human
erythrocytes. Intralipid was obtained from Kabi Pharmacia AB.
Tris, "Preciset" cholesterol, and Preciset glucose
were from Boehringer Mannheim. NAD+ and lactate
dehydrogenase were from Oriental Yeast Co. Hydrazine
monohydrate, Triton X-100, HEPES buffer, the chemicals used in the
interference study except Intralipid and ditaurobilirubin, control sera
(Wako-liquid I and II), reagent kit "Glucose Wako", and reagent kit
"L-type Wako cholesterol", and all other ALBK chemicals were from
Wako Pure Chemical Industries. Ditaurobilirubin, a chemically
synthesized bilirubin conjugate, "High Level Check Lipid"
(lyophilized serum), and "Cholesterol Calibrator" were from
International Reagents. Standard Reference Material No. 911a was
obtained from NIST.
samples
Blood samples were collected without anticoagulant from
hospitalized patients. This study was approved by the committee for
ethical standards of the National Cardiovascular Center Hospital.
instruments
We used a TBA-80FR·NEO biochemical analyzer (Toshiba) for the
present assay and the other enzymatic assays in our comparison study.
For optimization studies, we used a Shimadzu UV-2100 spectrophotometer
with a constant temperature cell holder.
procedures
Optimization studies were performed for each of the components of
the cholesterol assay. A human serum pool and/or commercial control
serum were used in the optimization studies. When we had ascertained
the optimum concentration of a particular ingredient, it was maintained
at that concentration while the next ingredient was optimized. For
manual testing of the CDH reaction, 2.00 mL of reagent 1 and 0.06 mL of
specimen were pipetted into the cuvette. The mixture was preincubated
at 37 ° for 5 min, and the absorbance was read at 340 nm. The
reaction was started by the addition of 0.7 mL of reagent 2 and
incubated at 37 °C. After 10 min, the absorbance was again measured
at 340 nm. For the blank, the same volume of reagent 2 without CDH was
used. The cholesterol concentration was calculated from the change in
absorbance for the standard and the sample after subtraction of
the blank reading. The molar absorption coefficient of ß-NADH at 340
nm was calculated as 6.22 x 103
L · mol-1 · cm-1,
with the Glucose Wako reagent kit (glucokinase-glucose-6-phosphate
dehydrogenase-ultraviolet method with Preciset glucose as samples)
(9).
In the automated CDH assay for total cholesterol, 8 µL of serum was
incubated with 270 µL of reagent 1 for 5 min at 37 °C, and the
absorbance at 340/572 nm was measured. The reaction was started
immediately by the addition of 90 µL of reagent 2. After 4 min, the
absorbance at 340/572 nm was again measured. The cholesterol
concentrations of samples were determined from a calibration curve
constructed with Cholesterol Calibrator and/or a factor calculated from
the measured molar absorptivity of the reaction product, ß-NADH
(10). For comparison studies, we determined total
cholesterol concentrations by the modified ALBK method (2)
and by the
CEH-CO-POD-N-ethyl-N-(2-hydroxy-3-sulfopropyl)-3,5-dimethoxyaniline
(DAOS) method, using a TBA-80FR·NEO biochemical analyzer according to
the manufacturer’s instructions.
statistical analysis
The mean, SD, and CV were calculated using Microsoft Excel 97
(Microsoft). Linear regression analysis by the least-squares method was
performed using the StatView® statistics program
(Abacus Concepts).
|
Results
|
|---|
optimization
Optimization curves for several components are shown in Fig. 2
. In the assay, we used a two-reagent system, reagents 1
and 2, with the following compositions. Reagent 1 contained 71.4 mmol/L
Tris-HCl (pH 8.0), 4.3 mmol/L NAD+, 1.4 kU/L CEH,
80 mmol/L hydrazine monohydrate, 1 g/L bovine serum albumin, 3 g/L
sodium cholate, and 5 mL/L Triton X-100. Reagent 2 contained
14.3 mmol/L HEPES (pH 7.5), 4.3 kU/L CDH, 1 g/L bovine serum albumin, 3
g/L sodium cholate, and 5 mL/L Triton X-100.
View larger version (16K):
[in this window]
[in a new window]
|
Figure 2. Optimization studies.
(A), effects of pH and hydrazine monohydrate on
quantitative analysis of the CEH-CDH method. •, with
hydrazine (80 mmol/L);  , without hydrazine. High Level Check Lipid
was used as the sample and diluted sequentially with deionized water.
(B), effect of pH with hydrazine monohydrate (80 mmol/L;
•) and without hydrazine monohydrate (). High Level Check Lipid
diluted 1:1 with deionized water was used as the sample.
(C), effect of hydrazine monohydrate concentration on
the absorbance of the reaction at pH 8.0. High Level Check Lipid
diluted 1:1 with deionized water was used as the sample.
|
|
The addition of hydrazine monohydrate (
30 mmol/L) to reagent 1 drove
the CDH reaction to completion over a pH range of 7.6–9.0. The
reaction at pH 7.3 showed incomplete conversion of cholesterol to
cholest-4-ene-3-one. In contrast, the reaction in the absence of
hydrazine did not progress quantitatively at any pH.
The 5 mL/L Triton X-100 was incorporated as an activator for
both CEH and CDH. The 3 g/L sodium cholate was used as an additional
activator of CEH. The 1 g/L bovine serum albumin was added to reagents
1 and 2 as a stabilizer. For reagent 2, pH 7.5 was selected to avoid
loss of CDH. The final pH in the reaction mixture was 8.0, which was
optimal for overall reaction. Reagents 1 and 2 were stable for at least
2 weeks at 2–8 °C.
cdh reaction
To assess whether the CDH reaction in a complete reagent that
includes CEH can progress to completion, we determined whether
the amount of cholesterol added was equivalent to the expected
absorbance change of ß-NADH obtained at 340 nm. Six samples of known
cholesterol concentration (1.29, 2.59, 3.88, 5.17, 7.76, and 10.35
mmol/L; Preciset cholesterol calibrator) were assayed manually
by absorbance change at 340 nm on a Shimadzu UV-2100. The absorbance
changes obtained at 340 nm were 0.1734, 0.3509, 0.5234, 0.6936, 1.0265,
and 1.3587, respectively. The recovery, expressed as a percentage of
the expected absorbance change, was 97.1–100.3%, with a mean of
98.9%. Accuracy tests with Standard Reference Material No. 911a
aqueous standard (5.17 mmol/L, dissolved in isopropyl alcohol) also
showed similar recovery.
performance
Time courses.
Typical time courses of the reaction on a
TBA-80FR·NEO analyzer are shown in Fig. 3
. Two control sera and a human serum pool were used as samples.
All samples reached plateau within 3 min. Based on these results, we
used the change in absorbance between 4 and 5 min after the addition of
reagent 2 to assay cholesterol with the TBA-80FR·NEO analyzer.
Linearity studies.
The linearity of the assay was evaluated by
preparing sequential dilutions of High Level Check Lipid (20.28 mmol/L
cholesterol) and a human serum pool (5.32 mmol/L cholesterol). Both low
and high cholesterol concentrations gave a linear response, which was
fitted to the following linear regression equations (95% confidence
intervals in parentheses): y = 0.533 (0.531–0.535)
x - 0.014 (-0.028 to 0.00036) mmol/L;
r = 0.999; Sy|x = 0.0091 mmol/L
for the low cholesterol concentration; and y = 2.02
(1.98–2.06) x + 0.287 (0.063–0.511) mmol/L;
r = 0.999; Sy|x = 0.142 mmol/L for
the high cholesterol concentration (Fig. 4
).
View larger version (18K):
[in this window]
[in a new window]
|
Figure 4. Linearity curve for the CEH-CDH method.
Commercial control serum ( ) and pooled human serum ( ), diluted
sequentially with deionized water.
|
|
Imprecision.
Within- and between-day imprecision was
determined by performing measurements on two commercial control sera
(Wako-liquid) and one pooled serum. Between-day testing was carried out
on 10 days over a 2-week period. Calibration was performed each testing
day. The within- and between-day imprecision (CV), as determined by
replicate analyses of two commercial control sera (n = 20) and one
pooled serum (n = 10) with different cholesterol concentrations,
was 0.29–0.43% and 0.22–0.61%, respectively (Table 1
).
Interfering substances.
We evaluated the effect of several
potential interfering substances in this method. Bilirubin dissolved
with 0.1 mol/L NaOH, ascorbate dissolved with 10 mmol/L
glycine-HCl buffer (pH 3.0), lactate dehydrogenase diluted with
physiological saline, and other interferents dissolved with deionized
water were added to pooled serum (1:9, by volume). Each sample
was measured in 10 replicate analyses. As shown in Table 2
, we found that interference 0.08 mmol/L cholesterol resulted
from the addition of any of these substances to a serum pool.
Hemoglobin at 0.31 mmol/L positively interfered, adding 0.12 mmol/L
cholesterol at the upper 95% confidence limit. However, this
hemoglobin concentration occurs rarely in physiological samples.
Intermethod comparison data.
Fifty samples (mean ± SD
triglyceride concentration, 18.83 ± 15.61 mmol/L) were assayed by
a modified ALBK method and the automated enzymatic color test method
(Fig. 5
). The values obtained in comparison with the modified ALBK
method were fitted to a linear regression model using a least-squares
method. The equation for the CEH-CDH method compared with the ALBK
method was: y = 0.992x - 0.0058
mmol/L, where x is the mean ± SD ALBK cholesterol
concentration (5.50 ± 1.49 mmol/L); r = 0.997;
Sy|x = 0.117 mmol/L; slope = 0.992 (95%
confidence interval, 0.969–1.014); intercept = -0.0058 (95%
confidence interval, -0.134 to 0.123). In this comparison,
y = 5.45 ± 1.49 mmol/L (mean ± SD
cholesterol concentration).
View larger version (17K):
[in this window]
[in a new window]
|
Figure 5. Correlation studies.
(Left), ALBK method vs automated CEH-CDH-ultraviolet
method. (Right), automated CEH-CO-POD-DAOS method vs
automated CEH-CDH-ultraviolet method.
|
|
The equation for the CEH-CDH method compared with the CEH-CO-POD-DAOS
method was: y = 0.989x - 0.048 mmol/L,
where x is the mean ± SD colorimetric cholesterol
concentration for the CEH-CO-POD-DAOS method (5.35 ± 1.47
mmol/L); r = 0.999; Sy|x = 0.065
mmol/L; slope = 0.989 (95% confidence interval, 0.977–1.001);
intercept = -0.048 (95% confidence interval, -0.117 to 0.022).
There was good agreement between the results from the present method
and each of the two reference methods.
|
Discussion
|
|---|
We report a new method for cholesterol determination using CDH
from Nocardia sp. The characteristics of the enzyme used
have been reported previously (11)(12). The
substrate specificity of the CDH used in this report is 100% for
cholesterol, 52% for ß-sitosterol, 50% for ergosterol, 30% for
stigmasterol, and 14% for pregnenolone. Other physiological hormones,
dehydroepiandrosterone, and testosterone are not substrates. The
substrate specificity of CDH is similar or superior to that of CO,
which is used in current commercial cholesterol assays
(12)(13).
The optimal pH for CDH is >10.0. The
Km values for cholesterol and
ß-NAD+ are 0.15 and 0.23 mmol/L, respectively.
The reaction is reversible. For this reason, a simple endpoint assay
for cholesterol using CDH is impractical. Flegg (14)
reported an assay system for serum cholesterol that detected the
cholest-4-ene-3-one produced by the CDH reaction at 240 nm. However,
this method required a long incubation time, as long as 2 h, and
was not amenable to automation. We added hydrazine monohydrate to the
reaction mixture to trap cholest-4-ene-3-one, similar to the
phenylhydrazine reaction used to produce osazone from aldose
(15). Although the optimal pH of CDH is >10.0, the addition
of hydrazine makes it possible to carry out the analysis at pH 8.0. The
addition of hydrazine to the reaction drives the reaction to
completion, and the endpoint is achieved within 3 min. The
CEH-CDH method for determining serum cholesterol has several advantages
over traditional methods: rapidity, simplicity, and no interference
from various reductants, bilirubin, ascorbic acid, or reduced
glutathione. Moreover, because the amount of ß-NADH formed is
equivalent to the amount of cholesterol, the concentration of
cholesterol in serum can be estimated from a factor calculated from the
measured molar absorptivity of ß-NADH calibrated by that in the
Standard Reference Material. This method has excellent accuracy,
imprecision, linearity, and correlation with the ALBK and
CHE-CO-POD-DAOS methods.
In conclusion, this new CDH method is an accurate, simple, and
automatable method for the quantitative analysis of cholesterol.
|
Acknowledgments
|
|---|
We thank K. Kishi, M. Ikeda, and Y. Watazu of International
Reagents Corporation, Kobe, Japan for providing helpful technical
advice.
|
Footnotes
|
|---|
1 Nonstandard abbreviations: ALBK, Abell-Levy-Brodie-Kendall; CEH, cholesterol ester hydrolase; CO, cholesterol oxidase; POD, peroxidase; CDH, cholesterol dehydrogenase; and DAOS, N-ethyl-N-(2-hydroxy-3-sulfopropyl)-3,5-dimethoxyaniline. 
|
References
|
|---|
-
Abell LL, Levy BB, Brodie BB, Kendall FE. A simplified method for the estimation of total cholesterol in serum and demonstration of its specificity. J Biol Chem 1952;195:357-366.
[Free Full Text]
-
Dunkan IW, Mather A, Cooper GR. The procedure for the proposed cholesterol reference method, Clinical Chemistry Division, Centers for Disease Control 1982 Centers for Disease Control Atlanta, GA. .
-
Allain CC, Poon LS, Chan CSG, Richmond W, Fu PC. Enzymatic determination of total serum cholesterol. Clin Chem 1974;20:470-475.
[Abstract]
-
Siedel J, Rollinger W, Röschlau P, Ziegenhorn J. Total cholesterol, end-point and kinetic method. Bergmeyer HU eds. 3rd ed. Methods of enzymatic analysis 1985;Vol. 13:139-148 Verlag Chemie New York. .
-
Boehringer Mannheim GmbH, assignee. Total or
bound cholesterol determination using NAD or NADP-dependent cholesterol
dehydrogenase from liver or from an anaerobic microorganism. German
patent application, Deutsche Offenlegungsschrift 2649749, 1976..
-
Battelle-Institut e.V., assignee. Verfahren zur
Bestimmunug von Gesamtcholesterin. German patent application, Deutsche
Offenlegungsschrift 3032377, 1981..
-
Akiba T, inventor. An enzymatic measurement of
cholesterol with NAD(P)-dependent cholesterol dehydrogenases and the
reagents for the assay. Japanese patent 83-892000, 1983..
-
Horinouchi S, Ishizuka H, Beppu T. Cloning, nucleotide sequence, and transcriptional analysis of the NAD(P)-dependent cholesterol dehydrogenase gene from a Nocadia sp. and its hyperexpression in Streptomyces spp. Appl Environ Microbiol 1991;57:1386-1393.
[Abstract/Free Full Text]
-
McComb RB, Bond LW, Burnett RW, Keech RC, Bowers GN. Determination of the molar absorptivity of NADH. Clin Chem 1976;22:144-150.
-
Wako Pure Chemical Industries. Procedure of measurement
of K factor. Osaka, Japan: Wako Pure Chemical Industries, 1997..
-
Akiba T, inventor. The method for preparation of
NAD(P)H-dependent cholesterol dehydrogenases. Japanese patent 90-18064,
1990..
-
Amano. AMANO ENZYMES for Diagnostics 1998/1999. Nagoya,
Japan: Amano Pharmaceutical, 1998..
-
Toyobo. TOYOBO ENZYME. Osaka, Japan: Toyobo, 1996..
-
Flegg HM. An investigation of the determination of serum cholesterol by an enzymatic method. Ann Clin Biochem 1973;10:79-84.
-
Mayer PA. Carbohydrates. In: Harper HA, Rodwell VW, Mayer
PA, eds. Review of physiological chemistry, 16th ed. Maruzen Asian
Edition. Los Altos, CA: Lange Medical Publications, 1977:93–107..
Top
Abstract
Introduction
, Cholesterol Calibrator; 
, pooled serum; 
,
control serum. See text for additional reaction conditions.