| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Endocrinology and Metabolic Medicine, Imperial College School of Medicine, London, United Kingdom W2 1PG
Address all correspondence and requests for reprints to: Ian F. Godsland, Ph.D., Endocrinology and Metabolic Medicine, Imperial College School of Medicine, Norfolk Place, London, United Kingdom W2 1PG. E-mail: i.godsland{at}ic.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Several observations suggest that there might be links between insulin resistance and hyperhomocysteinemia. Homocysteine levels have been found to be raised in patients with type 2 diabetes, both in the fasting state (3) and after methionine loading (4), and are positively correlated with microalbuminuria (5). In rats made insulin resistant with a high fat sucrose diet, homocysteine levels rise, and this increase is associated with changes in critical enzymes of homocysteine metabolism (6). During a hyperinsulinemic euglycemic clamp, homocysteine levels fall in nondiabetics, but not in patients with type 2 diabetes mellitus (7). It has been suggested that stimulation of insulin-induced elimination of methionine, which is impaired in diabetics, might underlie these relationships (8).
There is, however, uncertainty over the relevance of these observations to risk factor interrelationships in humans, because endothelial damage or renal dysfunction was prevalent in several of the groups nominated to represent insulin-resistant states. Renal dysfunction in particular could independently increase homocysteine levels (9). Moreover, there is conflicting evidence about whether there is a more general relationship between insulin resistance and homocysteine levels in healthy humans. Giltay and colleagues measured insulin resistance as glucose utilization during a 2-h euglycemic hyperinsulinemic clamp in 24 healthy nonobese men and women and found significantly higher total homocysteine levels in the lowest tertile of insulin sensitivity (10). In contrast, Abbassi and colleagues found no such relationship when they explored associations between insulin resistance, measured by the insulin suppression test, and plasma total homocysteine levels in 55 healthy men and women (11). To better resolve this issue, we measured plasma total homocysteine in samples from 100 male participants in the second follow-up cohort of the Heart Disease and Diabetes Risk Indicators in a Screened Cohort Study (HDDRISC-2), each of whom have undergone measurement of insulin sensitivity by minimal model analysis of glucose and insulin concentrations during an iv glucose tolerance test (IVGTT). We have also investigated associations between total homocysteine and a range of other components of the insulin resistance syndrome, including fasting and glucose tolerance test glucose and insulin levels, fasting serum triglyceride, high density lipoprotein (HDL) cholesterol, HDL subfraction 2 (HDL2) cholesterol, uric acid concentrations, blood pressure, and body mass index (BMI).
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The HDDRISC is a cohort study of metabolic risk factors for the development of coronary heart disease and diabetes mellitus, with continuous recruitment between 1971 and 1999. The study derived from a company health program in which participants received a range of metabolic, clinical, and laboratory measurements at their first visit and, for those continuing in the program, at subsequent visits spaced at 2- to 3-yr intervals. Between 1986 and 1997, 689 participants in the study underwent measurement of insulin sensitivity by modeling analysis of glucose and insulin concentrations during an IVGTT. These form the second follow-up cohort of the HDDRISC, HDDRISC-2. The present investigation concerns 100 men from the HDDRISC-2 cohort for whom plasma samples remained in storage. A sample size of 100 is sufficient to detect a correlation coefficient of 0.20 as significant. Correlations less than this were considered unlikely to be of major importance. Samples were selected by working back in time from the most recently performed IVGTT. Full informed consent for the study was obtained in each case, and local ethics committee approval was given.
Procedures
Participants were instructed to consume more than 200 g/day carbohydrate in their diet for the previous 3 days as preparation for the IVGTT, to have fasted overnight (>12 h), and to have taken only water and refrained from cigarette smoking on the morning of the test. Height and weight were measured, and a clinical history was taken. Blood pressure was measured after resting for 15 min in a semirecumbent position. An indwelling cannula was then inserted into an antecubital vein in each arm. Prolonged venous stasis was avoided. Before glucose injection, blood samples were taken for the measurement of fasting plasma glucose and insulin (two samples in each case), serum lipids and lipoproteins, a biochemical profile including uric acid and creatinine concentrations, and a full blood count. All samples were kept on ice and processed within 1 h of being taken. An iv glucose injection was then given [0.5 g glucose/kg BW as a 50% (wt/vol) solution of dextrose, given over 3 min] via the cannula in the opposite arm from the sampling arm. Blood samples (10 mL) were obtained at 3, 5, 7, 10, 15, 20, 30, 45, 60, 75, 90, 120, 150, and 180 min for measurements of plasma glucose and insulin.
Laboratory measurements
Plasma total homocysteine was measured on samples stored at -20 C by gas chromatography-mass spectrometry by a modification of the methods of Stabler et al. (12, 13). d8-Homocystine (1.5 nmol; Cambridge Isotope Laboratories, Andover, MA) in 1.0 mL water was added as an internal standard to 100 µL plasma. Disulfide linkages were reduced by the addition of 50 µL 1% dithiothreitol in 1 mol/L sodium hydroxide and heating to 40 C for 30 min. The sample was applied to an anion exchange column (AG1-X8, acetate form), and the extraneous material was washed through with deionized water. Amino acids were eluted with 3 mL 0.1 mol/L HCl, and the eluates were collected in 3-mL Reactivials and dried under nitrogen. Homocysteine and d4-homocysteine were converted to their Tris-TBDMS derivatives by the addition of 50 µL N-methyl-N-tert-butyldimethylsilyltrifluoroacetamide containing 1% tert-butyldimethylchlorosilane and 100 µL acetonitrile, followed by heating at 90 C for 30 min. Analysis of homocysteine was carried out by gas chromatography-mass spectrometry on a DB-5ms capillary column (id, 15 m x 0.25 mm; film thickness, 0.25 µm). A 1-µL sample was injected splitless, and the column temperature was held for 1 min at 100 C, then raised to 200 C at 25 C/min, to 250 C at 5 C/min, then to 325 C at 25 C /min and held for 1 min. The injector temperature was 275 C. The [M-57]+ ion from the Tris-TBDMS derivative of homocysteine and d4-homocysteine was monitored at 420.2 and 424.2 amu, respectively. Quantification of total homocysteine was achieved by comparison of the peak area ratio (d0/d4) against that obtained from homocysteine standards over the concentration range 0–30 µmol/L. The assay coefficient of variation was 2.9%.
Plasma glucose concentrations were measured on the day of sampling by a glucose oxidase procedure (14). Plasma insulin concentrations were measured by RIA on samples stored at -20 C using the procedure of Albano et al.(15). Serum for measurement of plasma cholesterol, triglyceride, HDL cholesterol, and HDL3 cholesterol was stored at 2–6 C and assayed within 5 days of collection. Cholesterol and triglycerides were measured using a colorimetric method described by Siedel (16); HDL, HDL2, and HDL3 cholesterol were measured using standard precipitation techniques (17, 18); and low density lipoprotein (LDL) cholesterol was calculated using the Friedwald method (19). Quality control was monitored using commercially available lyophilized sera and by participation in relevant National Quality Assurance schemes (UK NEQAS for insulin and RIQAS for all other analytes). Over the 2 yr during which the participants in the present study were evaluated, within- and between-batch assay coefficients of variation were: plasma glucose, less than 3%; plasma insulin (10–110 µU/mL), less than 7%; serum cholesterol, less than 3%; serum triglycerides, less than 4%; serum HDL, less than 5%; and serum HDL2, less than 8%. Other biochemical measures were obtained using standard routine laboratory procedures.
Data analysis
Basal glucose and insulin concentrations were taken as the mean
of the two fasting measurements. Glucose tolerance was expressed as the
k value, the glucose elimination constant during the IVGTT. This was
taken as the slope of the regression line for the natural log of the
IVGTT glucose concentrations between 20 and 60 min. The glucose and
insulin responses during the IVGTT were taken as the increment in area
between the basal level and the concentration profile, with areas
calculated by the trapezium rule. Insulin sensitivity
(Si) was determined with the minimal model of
glucose disappearance (20), using programs written in
Fortran 77. The relatively high glucose dose (0.5 g/kg) employed
provides for a sufficient endogenous insulin response in nondiabetic
volunteers without recourse to additional augmentation of pancreatic
insulin secretion. This is apparent in the high rate of model
identification and the good correlation with measures of insulin
sensitivity derived from the euglycemic clamp (r = 0.92) that we
have obtained with this procedure (21, 22). Parameters of
posthepatic insulin delivery, the insulin elimination rate, phase 1
posthepatic insulin delivery (
1), and phase 2
posthepatic insulin delivery (2), were
estimated using the minimal model of ß-cell sensitivity of Toffolo
et al. (23).
Statistical analyses were carried out using SYSTAT statistical software (SYSTAT, Inc., Evanston, IL). Skewed distributions were detected for homocysteine, triglyceride, and insulin concentrations and for insulin sensitivity. Variables were summarized as medians and ranges. For subsequent statistical analysis using Pearson correlation and factor analysis, insulin sensitivity measurements were square root transformed (21), and homocysteine, triglyceride, and insulin concentration measurements were log transformed to normalize their distributions. Univariate correlations were explored using Pearson correlation coefficients. Clustered variation among the variables investigated was explored using factor analysis, as employed previously in the HDDRISC and other studies (24, 25, 26). In brief, factor analysis supposes that the existence of a large number of highly intercorrelated variables reflects variation in a more limited number of underlying variables or factors. A number of procedures are available for factor analysis. The one employed here and in our previous analysis (25) incorporates a principal components analysis by which independently varying linear combinations of measured variables are identified that account for the maximum proportions of variance in the dataset. Interpretation of these components is then assisted by a rotation procedure, in this analysis the so-called Varimax rotation was used, which maximizes the discrimination between those measured variables that relate to each component and those that do not. Measured variables are related to the resulting factors by loadings that are equivalent to the correlation coefficient between the variable and the factor. The variables with the highest loadings are then the measures that are likely to be of most importance in interpreting the nature of the underlying factor responsible for their covariation. In the present analysis the number of factors was limited to three, and only variables sharing a loading greater than 0.30 with a given factor were considered for interpretation. In selecting variables for factor analysis, all variables were included except diastolic blood pressure, LDL cholesterol, HDL3 cholesterol, and IVGTT incremental glucose area, variations in which were highly correlated with, respectively, systolic blood pressure, total cholesterol, HDL cholesterol, and the glucose elimination rate (k). Exclusion of these variables in the factor analysis avoided the emergence of uninformative factors solely associated with two highly correlated variables, for example total and LDL cholesterol.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. Age ranged from
31.4–61.6 yr, BMI from 20.6–36.5 kg/m2, and
total homocysteine from 7.5–30.6 µmol/L. Two men were being treated
for hyperlipidemia. The lipid-lowering agents used were simvastatin and
gemfibrozil, and total homocysteine levels in these men were 30.6 and
23.4 µmol/L, respectively. One man was being treated for
hyperuricemia with allopurinol (total homocysteine, 11.2 µmol/L) and
one for migraine with propranolol (total homocysteine, 11.0).
Otherwise, participants were healthy and were taking no drugs that
might be expected to affect the metabolic measures studied.
|
). No correlation
coefficient exceeded 0.20. When individuals were classified according
to quintiles of insulin sensitivity, median homocysteine concentrations
from the lowest to the highest quintile (n = 20 in each quintile)
were 11.9 (range, 8.0–24.5), 13.1 (range, 8.7–18.1), 11.9 (range,
8.7–15.3), 12.7 (range, 7.5–18.2), and 12.2 (range, 7.5–30.6). There
were no significant differences among the quintiles, nor was there a
significant trend. There were, nevertheless, significant univariate
correlations between the components of the insulin resistance syndrome
themselves (Table 3). These
intercorrelations were also apparent in the factor analysis (Table 4) in which the factor explaining the
greatest proportion of the variance (24.3%) included among its
components the IVGTT insulin response, insulin sensitivity,
triglyceride concentration, BMI, mean fasting insulin,
HDL2 cholesterol, and uric acid. A high
homocysteine level was included in a second factor that also comprised
low glucose effectiveness, a low IVGTT glucose elimination rate, high
mean fasting glucose, greater age, and higher systolic blood pressure.
This factors explained 14.8% of the variance.
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
If there is no relationship between homocysteine levels and insulin sensitivity in healthy individuals, what might be the basis for evidence linking the two in other studies? Insulin-resistant conditions in which elevated homocysteine levels or a significant positive correlation between homocysteine levels and insulin resistance have been reported include type 1 and type 2 diabetes (2), preeclampsia (28), and vascular disease (29). However, in all of these studies endothelial damage, renal dysfunction, or inflammation has either been detected or may be inferred. It is therefore possible that the correlation between homocysteine and insulin resistance is secondary to these disturbances. This is particularly apparent in the most widely studied of these conditions, diabetes mellitus, in which elevated homocysteine levels have been detected in the presence of macro- or microvascular disease, hypertension, microalbuminuria, or renal disease, but not in uncomplicated diabetes (4, 30). Diabetic patients with elevated homocysteine levels do appear to be at greater risk of developing vascular disease than those with normal homocysteinemia (31), so homocysteine might accelerate the progression of disease that is already present. Alternatively, elevated homocysteine levels could simply be acting as a marker for more severe underlying endothelial or renal damage or inflammatory processes in diabetics. In the present study no subjects with diabetes or known renal impairment were included, and plasma creatinine was measured in all subjects.
Apart from major inherited enzyme defects, vitamin deficiency has been identified as the principal cause of elevated homocysteine levels (32). Moderate increases in homocysteine may also be associated with a common point mutation in the methylene tetrahydrofolate reductase gene (33), which renders carriers susceptible to the effects of suboptimal folate intake (34). Other correlates of homocysteine concentrations have been described, including serum cholesterol, HDL cholesterol, triglycerides, and blood pressure, but studies have been inconsistent and, where such associations have been found, they have generally been weak (r < 0.20) (27). In contrast to some other studies, we found no association between age and total homocysteine in univariate analysis. This was in accord with the findings of Abbasi et al. (11) and might reflect the prevalence of relatively high cofactor intakes among the participants at all ages. Our volunteers were drawn from a relatively high socio-economic status grouping, and in the study by Abbasi et al., volunteers were drawn from a region of comparative prosperity (San Francisco, CA). It is noteworthy, however, that in factor analysis we found higher total homocysteine levels to be associated with greater age, higher basal glucose levels, and poorer glucose tolerance, specifically related to insulin-independent processes. This recalls the metabolic links between vitamin B6 deficiency and impaired tryptophan metabolism that result in increased gluconeogenesis and impaired glucose tolerance (35). Possibly there was a subgroup among the men we studied in which there was relatively poor nutrition, resulting in a factor cluster that included impaired glucose tolerance and age.
These findings focus attention on the possibility that cofactor deficiencies and inherited enzyme defects are the principal determinants of the variations in total homocysteine levels that have been linked with increased risk of cardiovascular disease. Further exploration of these issues may require the more stringent evaluation of homocysteine and methionine metabolism status provided by measurement of total homocysteine after methionine loading. However, it now seems likely that among generally healthy individuals, variation in fasting total homocysteine levels is not linked with insulin resistance or with the manifold disturbances of the metabolic syndrome.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received May 24, 2000.
Revised September 30, 2000.
Accepted October 9, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
,
IL-1 and 6) and homocysteine in android obesity and the phenomenon of
insulin resistance syndromes. Vnitrni Lekarstvi. 45:11–16.[Medline]
This article has been cited by other articles:
 |
|
 |
|
  Y. Song, N. R. Cook, C. M. Albert, M. Van Denburgh, and J. E. Manson Effect of Homocysteine-Lowering Treatment With Folic Acid and B Vitamins on Risk of Type 2 Diabetes in Women: A Randomized, Controlled Trial Diabetes, August 1, 2009; 58(8): 1921 - 1928. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Li, C. Jiang, G. Xu, N. Wang, Y. Zhu, C. Tang, and X. Wang Homocysteine Upregulates Resistin Production From Adipocytes In Vivo and In Vitro Diabetes, April 1, 2008; 57(4): 817 - 827. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. R Hajer, Y. van der Graaf, J. K Olijhoek, M. C Verhaar, F. L J Visseren, and for the SMART Study Group Levels of homocysteine are increased in metabolic syndrome patients but are not associated with an increased cardiovascular risk, in contrast to patients without the metabolic syndrome Heart, February 1, 2007; 93(2): 216 - 220. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. F. Godsland, O. F. Agbaje, and R. Hovorka Evaluation of nonlinear regression approaches to estimation of insulin sensitivity by the minimal model with reference to Bayesian hierarchical analysis Am J Physiol Endocrinol Metab, July 1, 2006; 291(1): E167 - E174. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Tessari, A. Coracina, E. Kiwanuka, M. Vedovato, M. Vettore, A. Valerio, M. Zaramella, and G. Garibotto Effects of Insulin on Methionine and Homocysteine Kinetics in Type 2 Diabetes With Nephropathy Diabetes, October 1, 2005; 54(10): 2968 - 2976. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Tessari, E. Kiwanuka, A. Coracina, M. Zaramella, M. Vettore, A. Valerio, and G. Garibotto Insulin in methionine and homocysteine kinetics in healthy humans: plasma vs. intracellular models Am J Physiol Endocrinol Metab, June 1, 2005; 288(6): E1270 - E1276. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Tyagi, W. Rodriguez, A. M. Patel, A. M. Roberts, J. C. Falcone, J. C. Passmore, J. T. Fleming, and I. G. Joshua Hyperhomocysteinemic Diabetic Cardiomyopathy: Oxidative Stress, Remodeling, and Endothelial-Myocyte Uncoupling Journal of Cardiovascular Pharmacology and Therapeutics, January 1, 2005; 10(1): 1 - 10. [Abstract] [PDF]
|
|
|
|
 |
|
 I. F. Godsland, D. Crook, A. J. Proudler, and J. C. Stevenson Hemostatic Risk Factors and Insulin Sensitivity, Regional Body Fat Distribution, and the Metabolic Syndrome J. Clin. Endocrinol. Metab., January 1, 2005; 90(1): 190 - 197. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S. Daskalopoulou, D. P. Mikhailidis, and M. Elisaf Prevention and Treatment of the Metabolic Syndrome Angiology, November 1, 2004; 55(6): 589 - 612. [Abstract] [PDF]
|
|
|
|
|
|
S. S. Daskalopoulou, D. P. Mikhailidis, and M. Elisaf Prevention and Treatment of the Metabolic Syndrome Angiology, November 1, 2004; 55(6): 589 - 612. [Abstract] [PDF]
|
|
|
|
 |
|
 M.-J. J. Pouwels, M. den Heijer, H. J. Blom, C. J. Tack, and A. R. Hermus Improved Insulin Sensitivity and Metabolic Control in Type 2 Diabetes Does Not Influence Plasma Homocysteine Diabetes Care, May 1, 2003; 26(5): 1637 - 1639. [Full Text] [PDF]
|
|
|
|
|
|
M. Chandalia, N. Abate, A. V. Cabo-Chan Jr., S. Devaraj, I. Jialal, and S. M. Grundy Hyperhomocysteinemia in Asian Indians Living in the United States J. Clin. Endocrinol. Metab., March 1, 2003; 88(3): 1089 - 1095. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Orio Jr., S. Palomba, S. Di Biase, A. Colao, L. Tauchmanova, S. Savastano, D. Labella, T. Russo, F. Zullo, and G. Lombardi Homocysteine Levels and C677T Polymorphism of Methylenetetrahydrofolate Reductase in Women with Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., February 1, 2003; 88(2): 673 - 679. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. S. Randeva, K. C. Lewandowski, J. Drzewoski, K. Brooke-Wavell, C. O'Callaghan, L. Czupryniak, E. W. Hillhouse, and G. M. Prelevic Exercise Decreases Plasma Total Homocysteine in Overweight Young Women with Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., October 1, 2002; 87(10): 4496 - 4501. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. B. Meigs, P. F. Jacques, J. Selhub, D. E. Singer, D. M. Nathan, N. Rifai, R. B. D'Agostino Sr., and P. W.F. Wilson Fasting Plasma Homocysteine Levels in the Insulin Resistance Syndrome: The Framingham Offspring Study Diabetes Care, August 1, 2001; 24(8): 1403 - 1410. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
