|
|
||||||||
Molecular Diagnostics and Genetics |
Departments of1
Pharmacology and 2
Medicine, University of California at San Diego School of Medicine, and VA San Diego Healthcare System, La Jolla, CA.
3 The Queensland Institute of Medical Research, Brisbane, Australia.
4 Section of Molecular Epidemiology, Leiden University Medical Center, Leiden, The Netherlands.
5 Department of Clinical Biochemistry, Royal Prince Alfred Hospital, Sydney, Australia.
aAddress correspondence to this author at: Department of Clinical Biochemistry, Royal Prince Alfred Hospital, Camperdown, NSW 2050, Australia. Fax 61-2-9515-7931; e-mail John.Whitfield{at}email.cs.nsw.gov.au.
| Abstract |
|---|
|
|
|---|
Methods and Results: Cholinesterase activity was lower in women than in men before the age of 50, but increased to activity values similar to those in males after that age. There were highly significant correlations with variables associated with the metabolic syndrome: plasma triglyceride, HDL- and LDL-cholesterol, apolipoprotein B and E, urate, and insulin concentrations;
-glutamyltransferase and aspartate and alanine aminotransferase activities; body mass index; and blood pressure. The heritability of plasma cholinesterase activity was 65%. Linkage analysis with data from the dizygotic twin pairs showed suggestive linkage on chromosome 3 at the location of the cholinesterase (BCHE) gene and also on chromosome 5.
Conclusions: Our results confirm and extend the connection between cholinesterase, cardiovascular risk factors, and metabolic syndrome. They establish a substantial heritability for plasma cholinesterase activity that might be attributable to variation near the structural gene and at an independent locus.
| Introduction |
|---|
|
|
|---|
Although most of the evidence points to plasma cholinesterase acting as a marker (rather than a cause) of cardiovascular risk, metabolic syndrome, or diabetes, its proposed role in triglyceride metabolism might mean that natural variations in cholinesterase activity contribute to variations in risk. In addition to associations with cardiovascular risk factors, the metabolic syndrome, and possibly, type 2 diabetes (11)(12), there are conflicting reports of a causative role for the low-activity K variant in Alzheimer disease(13)(14)(15)(16)(17)(18)(19)(20). The biomedical importance of plasma cholinesterase is therefore wider than the pharmacogenetic phenomenon of delayed metabolism of succinylcholine or other drugs. The gene for plasma cholinesterase (BCHE) is on chromosome 3, at bp 166973394167037952, and many comparatively rare genetic variants leading to low activity are now known(21). In addition to the known gene variants leading to low activity, a proportion of individuals show an additional cholinesterase band on electrophoresis and increased enzyme activity. This occurs with a frequency of 8% to 10% among Europeans(22), and is ascribed to the effects of another gene, cholinesterase (serum), 2 (CHE2), whose location is uncertain(23). A recent report(24) has shown that the increased activity is not attributable to increased cholinesterase protein concentration, but rather to increased specific activity.
Variations in plasma cholinesterase activity are therefore associated with variations in risk factors for cardiovascular and metabolic disease and are at least partly under genetic control. We studied variations in plasma cholinesterase activity in a sample of adult twins to assess the covariation with cardiovascular and metabolic disease risk factors, the magnitude of genetic effects on variation within the population, and the location of genes that determine or modify cholinesterase activity.
| Participants and Methods |
|---|
|
|
|---|
We collected blood samples from 1134 men and 2241 women. Plasma and serum were separated and stored at 70 °C until analyzed. Immediately before blood collection, participants completed a brief questionnaire reporting their alcohol consumption over the previous week. They also reported the time of their last meal, and the time of blood collection was noted. At the same visit, their height and weight were measured. Body mass index was calculated from weight and height as weight (kg)/[height (m)]2. Systolic and diastolic blood pressures were measured, with the participants sitting, by use of an automated blood pressure recorder (Dynamap 845 Vital Signs Monitor; Critikon Inc.). The mean of 2 results taken at 1-min intervals was calculated. Blood pressure results were available for 1666 of the participants.
Plasma samples were analyzed for cholinesterase activity by measurement of the absorbance increase at 412 nm on addition of the substrate acetylthiocholine at a final concentration of 0.5 mmol/L, according to the colorimetric method of Ellman et al. (26). 5,5'-Dithio-bis(2-nitrobenzoic acid) at a final concentration of 0.3 mmol/L was used as the chromogenic indicator of thiocholine formation. Samples were either measured in 1-cm optical-path cuvettes in a spectrophotometer (Response; Gilford Instrument Laboratories) for 5 min or in a 0.73-cm optical-path microtiter plate reader (Safire; Tecan Systems Inc.) for 3 min and 35 s. In either case, activity was recorded as a change in the absorbance at 412 nm in 1 min per microliter of serum. Butyrylcholinesterase activity was determined by including the specific acetylcholinesterase inhibitor BW284c51 (1 µmol/L final concentration) in the reaction mixture. Activity was expressed in international units as µmoles of acetylthiocholine hydrolyzed per milliliter of sample per minute.
Serum
-glutamyltransferase (GGT), aspartate aminotransferase (AST), alanine aminotransferase (ALT), cholesterol, glucose, triglycerides, and urate were measured by Boehringer Mannheim reagents and methods on a Hitachi 747 analyzer. HDL-C was measured by precipitation of non-HDL lipoproteins with dextran/MgSO4 followed by enzymatic cholesterol assay. Apolipoproteins A-I, A-II, B, and E were measured by immunonephelometry on a Behring nephelometer using Behring reagents. Plasma insulin was measured by RIA (Diagnostic Products Corporation).
Several of the measured variables were log-transformed because their frequency distributions were skewed. All references to serum GGT, AST, ALT, triglycerides, and insulin and to the quantity of alcohol consumed per week are to the log-transformed values unless specified otherwise. LDL-C was calculated from the total cholesterol, HDL-C, and triglyceride values by the Friedewald equation if triglycerides were <4.5 mmol/L. If the serum triglyceride concentration was above this limit, LDL-C was treated as missing. The samples were not taken in the fasting state, but participants reported the time of their last meal, and the triglyceride, glucose, and insulin results were adjusted for the elapsed time between the last meal and blood collection.
Exploratory analysis was carried out with SPSS, Ver. 13 (SPSS Inc.). Because the participants were twins and therefore not genetically independent, the effective number of individuals for any characteristic with substantial heritability would be less than the actual number of participants, and the significance (but not the magnitude) of correlations may be overestimated. More detailed examination of the effects of covariates and the sources of variation in cholinesterase was performed with the Mx program, Ver. 1.50 (27), which is designed for analysis of twin and family data and overcomes this problem.
We performed a genome-wide linkage analysis for loci affecting plasma cholinesterase activity on the dizygotic twin pairs. DNA was extracted from blood or buccal swabs according to standard procedures (28). Genotype data were assembled from 4 genome scans that had been done previously for other projects by the Mammalian Genotyping Service (Marshfield, WI), Leiden University Medical Centre (Leiden, The Netherlands)(29), Sequana Inc., and Gemini plc (United Kingdom). Pedigree structures for each scan were examined to identify inconsistencies between the genotypic data and pedigree relationships. Once any discrepancies were resolved, data for the 4 scans were merged and then checked again for pedigree errors. The combined genome scan data included 458 markers that were typed in 2 or more scans, which were included separately on the genetic map for the scan, separated by a very small distance [0.001 centimorgans (cM)]. The consistency of genotype information among these 458 markers was checked via cross-tabulations of allele calls between different scans. Markers with genotypic data inconsistent between different genome scans were removed from further analysis. Map positions were in Kosambi cM, estimated via locally weighted linear regression from the National Center for Biotechnology Information build 34.3 physical map positions and from published deCODE and Marshfield genetic maps. The procedures for combining and checking the genotype data and for the linkage analysis are described in Ref.(30).
Trait-specific empirical genome-wide suggestive and significant thresholds were calculated through use of 1000 gene-dropping simulations as described by Abecasis et al. (31). Details are given in Ref.(30). The empirical genome-wide thresholds for suggestive or significant linkage(32) were defined as the thresholds for which we observed, on average, 1 or 0.05 peaks per simulation with a logarithm of odds (LOD) score at or above the threshold, respectively. After the initial simulation results, which produced unusually high significance thresholds, we applied winsorization to reduce the impact of outliers on the linkage analysis(33). This was done by setting values for all cholinesterase residuals greater than 3 SD above or below the mean to values equivalent to 3 SD above or below the mean, respectively. Linkage analysis and the results of simulations to determine the genome-wide empirical P value are reported for the winsorized dataset.
| Results |
|---|
|
|
|---|
Exploratory analysis of the correlations between cholinesterase activity, adjusted for variation over time, and other variables gave the results shown in Table 1
. There were multiple significant correlations with known cardiovascular risk factors and variables associated with the metabolic syndrome. There was no significant correlation, in either men or women, with alcohol intake or smoking.
|
The pairwise correlations by zygosity after adjustment for method variation, sex, and age are shown in Table 2 of the online Data Supplement, together with the results of testing models of genetic and environmental sources of variation. We found that 65% (95% confidence interval, 50%75%) of the variation in plasma cholinesterase activity was attributable to additive genetic effects. Although the model including only additive genetic and nonshared environmental sources of variation fitted the data satisfactorily, a small shared environmental effect cannot be excluded.
The results of linkage analysis on 368 dizygotic twin pairs with genome-scan data are shown in Fig. 1
. Two peaks with LOD scores of 3.0 or greater were found, on chromosomes 3 and 5. The empirical significance thresholds determined by simulation on these data were 3.7 (for 1 occurrence in every 20 simulations, genome-wide; P = 0.05) and 1.8 (for an average of 1 occurrence per simulation), however; therefore, both peaks must be considered suggestive by the criteria of Lander and Kruglyak (32) (expected to occur less than once per genome scan) rather than significant. The peak on chromosome 3 (peak LOD score, 3.00; empirical genome-wide significance, P = 0.241) at GATA3H01 (172.3 cM, or 168.7 Mb from the p-terminal end of the chromosome) coincided with the location of the BCHE gene (167.0 Mb). The peak on chromosome 5 (peak LOD score, 3.34; empirical genome-wide significance, P = 0.135) was at GATA12G02 (106.1 cM, or 91.0 Mb), and the 1-LOD interval ran from 98.4 to 123.1 cM (82.0114.1 Mb). No other linkage peaks exceeded the suggestive threshold of LOD score (>1.8).
|
| Discussion |
|---|
|
|
|---|
As noted above, several previous investigators have found significant relationships between cholinesterase activity and triglycerides, HDL-C, and LDL-C (5)(6)(8)(9). A recent report(10) has extended this to a wider range of variables associated with the metabolic syndrome. Our results agree with this, and it is now clear that plasma cholinesterase clusters with a wider range of characteristics, including body mass index, apolipoprotein concentrations, insulin, liver enzymes, and blood pressure. Although many of these associations can be related to lipid or lipoprotein metabolism, the association with blood pressure does not easily fit into a concept of cholinesterase as an esterase involved in the metabolism of triglycerides and VLDL. There seems to be a broader involvement of cholinesterase with the metabolic syndrome, extending to normal variation between people in their blood pressure and (in the extreme case) to hypertension. The associations with AST, ALT, and GGT activities, which are known to be associated with insulin resistance(35) and increased in the metabolic syndrome(36), probably reflect an association between cholinesterase activity and the metabolic syndrome, of which fatty liver is a feature.
One important issue arising from these findings, and previous similar ones, is whether higher cholinesterase activity is a cause or a consequence of dyslipidemia and metabolic syndrome. Several published reports of animal (7)(37)(38) or human(9) studies are relevant, but unfortunately, the results do not give a clear answer. Interventions that primarily increase or decrease lipids tend to have the same effect on cholinesterase, but inhibition of cholinesterase activity in vivo has been shown to decrease lipid concentrations. For example, mice with streptozotocin-induced diabetes showed increased serum LDL, triglycerides, and cholinesterase(7), which decreased with insulin treatment, suggesting that the insulin-deficient state led to changes in cholinesterase activity. However, in the same study, inhibition of cholinesterase activity with tetraisopropylpyrophosphoramide led to a decrease in serum LDL and triglyceride concentrations. These results appear to place cholinesterase in the chain of events leading to changes in lipid values, rather than being a consequence. If this is the case, and particularly if this reasoning also applies to the other features of the metabolic syndrome that showed significant correlations with cholinesterase activity, then the sources of variation in cholinesterase activity between people take on an added significance. Blood pressure, as well, might be influenced causally by cholinesterase because the cholinesterase substrate acetylcholine causes vasodilation when infused into the human vasculature, triggering nitric oxide release via endothelial muscarinic cholinergic receptors(39). Thus, an excess of cholinesterase activity in the metabolic syndrome could also adversely affect endothelial function and ultimately increase blood pressure.
The pattern of within-pair twin correlations by zygosity and the model-fitting results, shown in Table 2 of the online Data Supplement, suggest contributions to variation from additive genetic effects and nonshared environmental effects. Some shared environmental effects, possibly related to batch effects on the measurement of cholinesterase activity, cannot be excluded but are estimated at only 7% of the total variance. The genetic effects are substantially greater, at 65% (95% confidence interval, 50%75%). Therefore, the major source of variation is genetic, and the location of the relevant genes can be assessed from the linkage analysis carried out on a subset of the dizygotic pairs.
This linkage analysis revealed 2 suggestive peaks, on chromosomes 3 and 5. The localization of a gene or genes whose variation affects plasma cholinesterase activity to chromosome 3 is to be expected, as this is the location of the BCHE gene itself. Nevertheless, it is gratifying to be able to identify linkage in the appropriate region, and this shows that linkage analysis for genes affecting quantitative traits can be done with comparatively small numbers of sibling pairs. Several variants of the BCHE gene that affect cholinesterase activity are already known; these include the common K variant, which produces an
20% decrease in activity, and the much rarer variants with major effects. Given the high heritability of cholinesterase activity and the linkage peak on chromosome 3 at the BCHE locus, it is likely that there are other sequence variations in or near the BCHE gene that affect activity, and a search using single-nucleotide variations and haplotype analysis will help to define them.
The chromosome 5 linkage peak is less readily explained. There are
70 genes or possible genes in the region indicated by this peak, but none has obvious relevance to plasma cholinesterase activity. A search through this region with current techniques would be time-consuming and expensive and not justified until the linkage result in this region has been replicated. One possibility that should be considered is that this region of chromosome 5 contains a gene for a protein that binds to cholinesterase and increases its activity (24); efforts to identify the nature of this protein by conventional biochemical means have been unsuccessful to date. Another possibility is that genes in this region affect the risk of insulin resistance and metabolic syndrome and that the linkage for cholinesterase activity is a consequence of this. However, other linkage studies on the metabolic syndrome(40)(41) do not support linkage to this region of chromosome 5, nor do our own results on metabolic syndrome components using this cohort of persons [Ref.(42) and our unpublished data].
In summary, our results emphasize the relevance of cholinesterase activity to cardiovascular risk and extend knowledge of its sources of variation. Further characterization of the chromosome 3 BCHE locus might elucidate the effects of known variants against the overall heritability and the linkage peak. Such investigations might also determine whether genetic causes of low activity lead to lower values for the cardiovascular risk factors, a finding that would clarify the practical significance of plasma cholinesterase for cardiovascular risk and cardiovascular disease. The larger task of detailed examination of genes under the chromosome 5 peak must await identification of candidate genes in this region, or replication of our linkage result.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Nonstandard abbreviations: LDL-C and HDL-C, LDL- and HDL-cholesterol, respectively; GGT;
-glutamyltransferase; AST, aspartate aminotransferase; ALT, alanine aminotransferase; cM, centimorgan(s); and LOD, logarithm of odds. ![]()
| References |
|---|
|
|
|---|
-Glutamyl transpeptidase and the metabolic syndrome. J Intern Med 2000;248:230-238.[CrossRef][ISI][Medline]
[Order article via Infotrieve]The following articles in journals at HighWire Press have cited this article:
![]() |
J. B. Whitfield, V. Dy, P. A.F. Madden, A. C. Heath, N. G. Martin, and G. W. Montgomery Measuring Carbohydrate-Deficient Transferrin by Direct Immunoassay: Factors Affecting Diagnostic Sensitivity for Excessive Alcohol Intake Clin. Chem., July 1, 2008; 54(7): 1158 - 1165. [Abstract] [Full Text] [PDF] |
||||
![]() |
P.-a. B. Shih and D. T. O'Connor Hereditary Determinants of Human Hypertension: Strategies in the Setting of Genetic Complexity Hypertension, June 1, 2008; 51(6): 1456 - 1464. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |