CLIN. CHEM. 29/6, 1031-1033 (1983)
Pseudocholinesterase/High-Density Lipoprotein Cholesterol Ratio in Serum
of Normal Persons and of Hyperlipoproteinemics
Rajendra Jam,’ K. Madhavan Kutty,2’3’4 Shao-Nan Huang,2and Kenneth Kean2’3
The proposed complementary risk factor, pseudocholinesterase/high-density
lipoprotein cholesterol ratio, was significantly higherin patients with type lIb and IV hyperlipoproteinemias then in controls. In contrast, the established risk factor,
total cholesterol/high-density
lipoprotein cholesterol
ratio,
was significantly higher in patients with type ha and IV
hyperlipoproteinemias.
Discriminant analysis indicated that
prediction of risk for coronary heart disease on the basis of
lipoprotein phenotypes can be improved by about 20% when
both the above factors are assessed
concurrently. On the
basis of earlier studies in humans and animals, we also
suggest that the proposed risk factor may provide a better
understanding
of events leading to enhanced risk for coronary heart disease as a consequence
of nutrition and of
abnormal metabolism of lipids and lipoproteins.
AdditIonal
cholesterol
Keyphrases: risk factor
.
lipoprofein phenotypes
-
heart disease
total
discriminant analysis
The biological function of serum pseudocholinesterase
(PChE; EC 3.1.1.8) has not been clearly established.
A
recent review (1) describes several possible physiological
roles for it in health and disease. Most importantly,
PChE
may be involved in lipid (2) and lipoprotein metabolism (3).
Its activity is increased in patients with hyperlipoproteinemia, obesity (4, 5) and diabetes (6).
A positive correlation
between high risk for coronary
heart disease and increased concentrations
of low-density
lipoprotein (LDL) has been well documented
(7). In contrast,
it has been suggested that an increased
concentration
of
high-density
lipoprotein (HDL) may confer a low risk for
heart disease. A more recent study (8), however, suggests
that the ratio between LDL cholesterol and HDL cholesterol
is a better indicator of risk rather than either of them alone,
on the premise that LDL transports
the cholesterol
produced
in the liver to the extra-hepatic
tissues and HDL brings it
back into the liver for further
metabolism.
Currently,
many clinical chemistry
laboratories,
including ours, report
the risk factor by calculating
the ratio
between total cholesterol
and HDL cholesterol.
We therefore
term this ratio the “established
risk factor,” or ERF. In our
previous study (9) we calculated the ratio between serum
of Mathematics, Faculty of Science, and
of Pathology, Faculty of Medicine, Memorial University of Newfoundland, and
‘Department
2Department
3The Dr. Charles
A. Janeway
Child Health
Centre,
St. John’s,
Newfoundland.
Author
to whom
Charles A. Janeway
land, Canada. AlA
Presented in part
Anaheim, CA, 1982
correspondence should be addressed, at The Dr.
Child Health Centre, St. John’s, NewfoundiRS.
at the 34th national meeting of the AACC,
(Clin Chem 26: 1607, 1982; abstract 291).
5Nonstandard abbreviations: PChE, pseudocholinesterase:
LDL,
low-density lipoprotein; VLDL, very-low-density lipoprotein; HDL,
high-density lipoprotein; ERF, established
risk factor; CRF, complementary risk factor.
Received Aug. 12, 1982; accepted Mar. 16, 1983.
and HDL cholesterol and termed it the “complementary risk factor” (CRF). This new risk factor correlated well
with ERF in individuals
classified as low, medium, and high
risk. In the same study we also found that CRF might be
more closely associated
with triglyceride-dependent
hyperlipoproteinemias.
We undertook
this current investigation
to
obtain more details of the distribution
of CRF in various
types of hyperlipoproteinemias.
PChE
Materials and Methods
Blood was sampled
from each individual
after an overnight fast of 14-16 h. Serum was separated from the clot
and used for all the analyses. Serum PChE was measured
with acetylthiocholine
as the substrate (10). The enzyme
activity in international
(IUB) units (U), defined as the
activity of enzyme that converts 1 mol of substrate in 1
mm at 30 #{176}C,
is here multiplied
by 1000. Serum triglycerides (11) and total cholesterol (12) were measured by enzymatic procedures.
Total LDL was measured by turbidimetry
(13), with use of heparmn and calcium chloride. HDL cholesterol was measured
in the supernatant
fluid after precipitation of the serum with heparmn-manganese
chloride (14).
Results
This study
included
395 adult
patients,
classified
as types
Ha, Hb, and IV hyperlipoproteinemias
on the basis of the
concentrations
of cholesterol (>2500 mg/L), triglycerides
(>1850 mg/L), and total LDL in their serum.
We also
evaluated the intensity of lipoprotemn bands after separation
by polyacrylamide
gel electrophoresis
as recommended
by
the World Health Organization
(15).
The following were measured for each individual: PChE;
HDL cholesterol; triglycerides;
total cholesterol; and total
LDL (i.e., LDL + VLDL). We then calculated the following
ratios: ERF = total cholesterollHDL
cholesterol and CRF
PChE/HDL
cholesterol.
The data, as shown in Table 1, were all transformed into
=
their
natural
logarithms
for statistical
analysis.
The Kolmo-
gorov-Smirnov
goodness-of-fit test was used to check the
distribution of the transformed data. All the above variables
have a gaussian distribution at the 5% level of significance.
One-way analysis
of variance
of the transformed data of
the Table 1, presented in Table 2, demonstrated
that all the
above variables differ significantly in the hyperlipoproteinemic patients from the controls at 5% level of significance
except for HDL cholesterol.
Tukey’s HSD test (16) shows
that CRF is not significantly
different between hyperlipoproteinemia
types lib and N; ERF between ha and hlb; total
cholesterol between Ha and Hb and also between N and
normal; triglycerides
between JIb and N; total LDL between Ha and N; HDL cholesterol between Ha and normal
and also between Hb and normal; and PChE between Ha
and normal.
We treated CRF and ERF to discriminant
analysis, to
study the extent to which different
lipoprotein
phenotypes
overlapped or diverged
from one another. The percentage
of
relative
contribution
CLINICAL
of the variables
CHEMISTRY,
to each discriminant
Vol. 29, No. 6, 1983
1031
Table 1. Risk-Factor Ratios, Lipid Concentrations, and PChE Activity (Mean ± SE) in Sera from Patients
with Hyperlipoproteinemia and from Normal Persons
Hyperilpoprotelnemla
lie (n
63.08
5.25
CRF
ERF
Total cholesterol, mg/L
Triglycerides, mg/L
Total LDL, mg/L
HDL cholesterol,
PChE, U xl000
mg/L
66)
=
±
±
3051.2
±
1115.2
5970.3
620.455
±
±
±
±
3708
lIb
2.83
0.24
70.1
29.8
171.2
20.81
114.9
Table 2. Analysis of Variance of Variables Studied
Mean
Sources of
variation
Variable
CRF
ERF
Total cholesterol
HDL cholesterol
Triglycerides
Total LDL
PChE
Between types
Within typeS
Between types
Within types
Between types
Within typeS
Between types
Within types
Between types
Within types
Between types
Within types
Between types
Within types
sum of
squares
7.4.43
0.1662
92.6475
2.4186
3.1921
0.0868
0.9001
0.0638
26.1357
0.1687
8.717
0.1699
1.7695
0.0493
.ratio
8
38.307
36.77 1
14
1037
154.956
51.295
35.89
function I and H showed the following distribution.
Percentage of among-groups
variance,
51.8 and 48.2%; ERF, 53.6
and 34.6%; and CRF, 46.4 and 65.4%, respectively.
The
discriminant
functions
were
sufficient
to discriminate
subjects with different
lipoprotein
phenotypes.
The
relative
importance
of the discriminant
function
compared
with
other variables
indicates that either CRF or ERF alone can
predict the risk for coronary
heart disease in 45% of the
cases.However, when these factors are combined the predic-
tive value is increased
to 65%.
Discussion
Previously
we proposed (9) that CRF may be an additional
valuable index in predicting
the risk for coronary heart
disease associated with hypertriglyceridemia
and hyper#{149}
pre-beta-lipoproteinemia.
The next question is whether an
increase in serum triglycerides with a concomitant increase
in VLDL has any role in the development of atherosclerosis.
Studies (17) have demonstrated
a significant increase
in
plasma triglyceride and VLDL in many patients with ischemic heart disease,
and it has been suggested
that the
VLDL remnants formed after the action of lipoprotein lipase
are capable of inducing atherosclerosis.
Another study (18)
also demonstrated
a severe narrowing
of coronary arteries
in patients with type N hyperlipoproteinemia.
Gianturco
et
al. (19), in a very recent
study of the receptor-mediated
uptake of VLDL from patients
with hypertriglyceridemia
by
normal
human
fibroblasts,
suggested
that
hypertriglyceri-
demic VLDL can be rapidly accumulated
and degraded by
cells via the LDL-receptor
pathway when the receptor is
expressed,
or alternatively
by nonspecific routes when the
receptor is not expressed.
Therefore
they hypothesized
that
hypertriglyceridemic
VLDL can be taken up by cells that
possess LDL receptors,
and that this may be a link between
1032
CLINICALCHEMISTRY, Vol. 29, No.6, 1983
(n = 113)
86.65
5.74
2996.9
2659.8
7870
557.345
4506
type
±
±
IV (n = 125)
86.04
4.57
2.71
0.18
±
±
Normals (n = 91)
2.41
58.62
±
0.11
3.49
±
±
44.9
2193.6
±
31.8
2087.7
±
±
±
±
±
91.3
2462.9
172
12.96
72
6125.6
509.52
4128
±
±
±
±
95.1
179.8
11.81
75.9
968.1
3859.2
611.98
±
±
±
3351
±
2.21
0.11
33.9
32.36
102.1
16.43
76.5
the metabolism of triglyceride-rich
particles and the pathogenesis of atherosclerosis.
The production
of VLDL is regulated
by the rate of
synthesis of apopeptide and triglyceride
in the liver. The
rate of hepatic synthesis of triglyceride in turn depends on
the rate of either the formation of fatty acids from carbohydrate or their
mobilization
from adipose tissue. Ingested
carbohydrates
are rapidly converted
to lipid with consequent hypertriglyceridemia
in patients with types ifi, N, or
V hyperlipoproteinemia
(20). However, it later was shown
(21) that hypertriglyceridemia
perlipoprotemnemia
is due
associated
with typefailure
ifi hy-of
to a receptor-mediated
intermediate
density lipoprotein catabolism,
in all likelihood attributable
to a lack of the isomorphic
forms of
apolipoprotein
E (probably EIH/E1V).
Similar
response
for
carbohydrates
has also been observed
in diabetic and obese
patients
(22). Increased
PChE activity
in serum has also
been demonstrated
in diabetic (6) and obese patients (4). We
have extended those observations to obese and diabetic mice
(23), which have 100-150% greater
serum PChE activity
than do lean normal mice. However, lean mice, when fed a
high-carbohydrate
diet, also showed an increased
activity of
PChE in serum as compared with mice fed a regular chow.
Two reasons prompted
us to use the ratio of PChE/HDL
cholesterol in this and the previous study (9). First, many
patients with hyperlipoproteinemia
show an increase
in
serum PChE activity and a decrease in HDL. Also, in a
patient who accidentally
ingested
Parathion,
an organophosphate anticholinesterase,
we observed
a marked
decrease in serum PChE activity concomitant with a reduction
in pro-beta- and beta-lipoproteins,
and increase in alphalipoprotein.
During
the recovery phase (i.e., within 72 h),
the enzyme activity increased
in parallel with increases
in
pre-betaand beta-lipoproteins
and a decrease
in alphalipoprotein. Similarly,
rats treated
with neostigmine,
a
carbainate
anticholinesterase,
showed decreased
incorporation of [3Hllysine into LDL and increased
incorporation
into HDL (3).
Statistical
analysis
of our data indicates that only 45% of
the risk for coronary heart disease can be predicted on the
basis of lipoprotein
proffle if ERF or CRF alone is used.
However, when both factors are combined the prediction can
be improved by about 20%. Among the hyperlipoproteinemic patients we examined, about 40% are classified
as type
N hyperlipoproteinemia
(Table 1). Serum cholesterol may
not be significantly
increased
in these patients; as a result,
ERF will be a poor predictor
of risk. In contrast, CRF
appears to be a good predictor of risk in patients with abovenormal VLDL and triglycerides.
In type Ha hyperlipoproteinemia,
on the other hand, ERF is a better
marker
than
CRF. In the mixed type of hyperlipoproteinemia
(Type JIb),
both factors are equally
good. Under these circumstances,
we believe that using ERF and CRF concurrently
will cover
a wider spectrum
of their abnormal
of the population who are at risk because
lipid and lipoprotein metabolism.
We thank Ms. J. Besso and J. Hughes for their technical
assistance and Mrs. B. English for secretarial help. This prqject was
supported by a National Health Research Development
grant,
Health and Welfare Canada.
References
1. Kutty KM. Review: Biological function of cholinesterase.
Clin
Biochem 13, 239-243 (1980).
2. Clitherow JW, Mitchard M, Harper NJ. The possible biological
function of pseudocholinesterase.
Nature (London) 199, 1000-1001
(1963).
3. Kutty KM, Redheendran
R, Murphy D. Serum cholinesterase,
function in lipoprotein metabolism. Experientia 33,420-421(1977).
4. Cucuianu M, Opincaru A, Tapalaga D. Similar behavior of
lecithin:cholesterol
acetyltransferase
and pseudocholinesterase
in
liver disease and hyperlipoproteinemia.
Clin Chim Acta 85, 73-79
(1978).
5. Chu Ml, Fontaine P, Kutty KM, et al. Cholinesterase
in serum
and low density lipoprotein of hyperlipidemic
patients. Clin Chim
Acta 85, 55-59 (1978).
6. Antopol W, Tuchman T, Schifin W. Choline-esterase
activity in
human sera, with special reference to hyperthyroidism.
Proc Soc
Exp Biol Med 36, 46-49 (1973).
7. Castelli WP, Doyle JT, Gordon T, et al. HDL Cholesterol and
other lipids in coronary heart disease. The Cooperative Lipoprotein
Phenotyping Study. Circulation 55, 767-772 (1977).
8. Gordon T, Castelli WP, Hjortland MC, et al. High density
lipoprotein as a protective factor against coronary heart disease.
The Framingham Study. Am J Med 62, 707-7 14 (1977).
9. Kutty KM, Jam R, Huang SN, et al. Serum pseudocholinesterase: High density lipoprotein cholesterol ratio as an index risk for
cardiovascular
disease. Clin Chim Acts 115, 53-61 (1981).
10. Ellman GL, Courtney KD, Andres V Jr, et al. A new and rapid
colorimetric determination of acetylcholinesterase
activity. Biochem Pharmacol 7,88-96(1961).
11. Wahlefeld AW. Triglycerides
determination
after enzymatic
hydrolysis. In Methods of Enzymatic Analysis, HU Bergmeyer, Ed.,
Academic Press, New York, NY, 1974, p 183.
12. Klose S, Greif H, Hagen A. Comparison of two newly developed
enzymatic cholesterol color tests on AutoAnalyzer
systems with
other cholesterol tests. Clin Chem 21, 942 (1975). Abstract.
13. Burstein M, Samaille J. Sw la precipitation selective des betalipoproteins du serum par l’heparin et les heparinodides de syntheses en presence du Cl2Ca1. J Physiol (Paris) 49, 83-86 (1957).
14. Friedewald WJ, Levy RI, Fredrickson DS. Estimation of the
concentration
of low-density lipoprotein
cholesterol in plasma,
without use of preparative ultracentrifuge.
Clin Clam 18,499-503
(1972).
15. World Health Organization Bulletin, pp 891-908 (1971).
16. Kirk RE. Experimental Design: Procedures for the Behavioral
Sciences, Brooks Cole, Belmont, CA, 1968.
17. Grundy MS. Hypertriglyceridemia:
Mechanisms, clinical significance, and treatment. Med Clin North Am 66, 519-535 (1982).
18. Cabin HS, Roberts WC. Hyperlipoproteinemia
and coronary
heart disease: A quantitative
study demonstrating
more severe
coronary narrowing in patients with type IV hyperlipoproteinemia
than in those with type II or known normal lipoprotein patterns.
Lab Invest 45, 8A (1981).
19. Gianturco SH, Brown FB, Gotto AM Jr. et al. Receptormediated uptake of hypertriglyceridemic
very low density lipoprotein by normal human fibroblasts. J Lipid Res 23, 984-993 (1982).
20. Levy RI, Glueck CJ. Hypertriglyceridemia,
diabetes mellitus
and coronary vessel disease. Arch Intern Med 123, 220-228 (1969).
21. Utermann G, Jaeschke M, Menzel J. Familial hyperlipoproteinemia type ifi: Deficiency of the specific apolipoprotein (ape Eifi) in very-low-density
lipoprotein. FEBS Lett 56, 352-355 (1975).
22. Yalow RS, Gluck SM, Roth J. et al. Plasma insulin and growth
hormone levels in obesity and diabetes. Ann NYAcad Sci 131,357373 (1965).
23. Kutty KM, Huang SN, Kean K. Pseudocholinesterase
in obesity: Hypercaloric diet induced changes in experimental obese mice.
Experientia 37, 1141-1142 (1981).
CLINICAL CHEMISTRY, Vol. 29, No. 6, 1983 1033