Oxidative stress occurs when there is a disturbance in the balance between the production of reactive oxygen species (free radicals) and antioxidant defences, which may lead to potential tissue damage (Ishibashi, 2006). Experimental findings showed that high levels of cholesterol led to increased production of reactive oxygen species (superoxide) within the endothelial microenvironment (Reilly et al., 2007). Enzymes proposed to be involved in increased superoxide production include xanthine oxidase, NO synthase and NAD(P)H oxidase. Superoxide rapidly reacts with NO to form peroxynitrite (ONO2¯), a highly reactive intermediate with cytotoxic properties (Vargas et al., 2007). The resulting oxidative stress plays a key role in the initiation of endothelial injury and perpetuation of the atherosclerotic process (Mather et al., 2001). Enhanced oxidative stress may directly induce endothelial dysfunction by decreased synthesis or release of NO by endothelial cells and by inactivating NO with superoxide in subendothelial space.
Oxygen free radicals not only bind to and inactivate NO; they also cause oxidation
of lipoproteins in the subintimal spaces. Accumulative evidence suggests that
oxidative modification of lipoproteins, in particular LDL, is one of the most
important and critical events in the pathogenesis of atherosclerosis and endothelial
dysfunction (Warnholtz et al., 2001). Studies from numerous laboratories
have shown that small, dense LDL is more susceptible to oxidation and predominance
of these small, dense LDL molecules correlates with greater risk of coronary
artery disease (Bays, 2003). Dyslipidaemia commonly occurs as part of a cluster
of cardiovascular risk factors, which includes hypertension, insulin resistance,
obesity and impaired homeostasis (NCEP, 2001). These risk factors
are collectively known as the metabolic syndrome and are common in type 2 diabetic
population and are undoubtedly responsible for much of the excess cardiovascular
mortality observed (Yuan et al., 2007). Diabetic dyslipidaemia is commonly
associated with increased triglycerides, decreased HDL-cholesterol and increased
small, dense LDL subparticles thus contributing towards endothelium dysfunction
occurs in response to cardiovascular risk factors.
Literature supports the paradigm of endothelial dysfunction as the common link between risk factors and atherosclerotic burden (Herman, 2006). Experimental and clinical studies have shown that hyperlipidaemia impairs endothelial function and promotes atherosclerosis (Warnholtz et al., 2001). In cholesterol-fed animals, endothelial dependent relaxation was found to be impaired and could be restored with cholesterol lowering therapy (Kamata et al., 1996). In humans, hypercholesterolaemia was associated with impaired endothelial dependent relaxation, both in atherosclerotic coronary arteries and in angiographically smooth coronary arteries (kamata et al., 1996). The deleterious effects of cholesterol on endothelial dependent relaxation are probably related to a direct relation between cholesterol and bioavailability of NO. Flavahan (1992) showed that diminished bioavailability of NO caused by hypercholesterolaemia did not result from impairment in the synthesis or release of NO, but most probably resulted from an impairment of a signal transduction process in the endothelium.
According to the response-to-injury hypothesis, hyperlipidaemia may be injurious to the endothelium. Another possible explanation for the association between cholesterol and endothelial dependent relaxation is increased oxidative stress. Even before the appearance of any ultrastructural change in the vessel wall, chronic elevations in serum cholesterol are often associated with an impaired endothelium dependent vasodilation. This early hallmark of endothelial dysfunction is thought to result from an imbalance between the influence of vasodilators and vasoconstrictors involved in the regulation of vascular tone (Frisbee et al., 2007).
N-α-tosyl L-arginine methyl ester (TAME)-esterase has been reported to be a component of the kinin-kallikrein system, involved in the sequence of events leading to the activation of the system (Subratty and Gunny, 2004).
A study carried out on Type 2 diabetic patients showed that TAME-esterase activity was significantly elevated in diabetic patients. Furthermore this was associated with increased triglycerides and total cholesterol levels, thereby suggesting a possible contribution of TAME-esterase in a furthering of diabetic dyslipidaemia leading to endothelial dysfunction (Subratty et al., 2003). Results so far tend to show that the enzyme TAME-esterase could be a biological marker associated with endothelial dysfunction and the onset of vascular diseases such as hypertension, cardiovascular disease (Subratty and Gunny, 2005).
This study was therefore undertaken to determine the level of oxidative stress and TAME-esterase activity in dyslipidaemic patients and to evaluate possible associations between total peroxides level as marker of oxidative stress, TAME-esterase activity and lipid disorders.
MATERIALS AND METHODS
A cross-sectional case control study design was used to explore possible association between oxidative stress, TAME-esterase activity and dyslipidaemia. Participants were recruited from existing population and family databases available at the SSR Centre for Medical Studies and Research, University of Mauritius. Participants were selected on the basis of information available from their files. Non-smokers, normotensive and subjects presenting with a plasma LDL-cholesterol >160 mg dL-1 (>4.2 mmol L-1) constituted the dyslipidaemic group. However the participants forming part of the control group were also non-smokers, normotensive but with a plasma LDL-cholesterol <130 mg dL-1 (<3.4 mmol L-1). Each participant gave a signed informed consent form upon joining this study.
A total of 20 mL of venous blood was collected for biochemical assays, which
also included TAME-esterase and oxidative stress assays. Oral Glucose Tolerance
Test (OGTT) was carried out for individuals who were not known to be diabetic
in order to evaluate the glucose metabolism status. However, before glucose
was given to the participant, a urinanalysis was performed using Boerhinger
Mannheim Uristik strips to detect presence of any glycosuria, which may be indicative
of an unknown diabetic status. OGTT was carried out on the same day. After the
initial blood collection in a fasting state, the participant was given 75 mL
of glucose dissolved in 250 mL of water. Venous blood was collected in fluoride
oxalate tubes 2 h after ingestion and plasma glucose was determined on this
sample using commercially available glucose oxidase kits.
Height was measured to the nearest centimetre (cm), with the individual
standing without shoes, feet close to a wall where a standiometer was fixed.
Subjects were weighed without shoes on a calibrated electronic scale (EXCELL
MHW platform scale). Weight was recorded to the nearest kilogram (kg). Body
Mass Index (BMI) was calculated as weight in kg/height in metres and given in
kg m-2. Waist and hip circumferences were measured with a measuring
tape, to the nearest centimetre. Waist-Hip ratios (WHR) were calculated as the
ratio waist/hip. Blood pressure measurements were performed in a sitting position,
after 15 min rest, using a standard mercury sphygmomanometer (ACCOSON, England),
with the average of 2 measurements of systolic and diastolic BP recorded from
A standard questionnaire was used to interview participants on their past medical history and present treatment. Participants were also asked about their behavioural habits such as smoking, physical exercise and alcohol consumption.
Determination of Plasma Glucose and Lipids
Plasma glucose, total cholesterol, triglycerides and HDL-cholesterol were
determined by enzymatic colorimetric assays using a multiparametric autoanalyser,
the COBAS Mira Plus (Roche Diagnostics Systems). The reagents used were purchased
from RANDOX Laboratories. Multiparametric calibrator and quality control sera
(normal and abnormal) were included for each assay.
For HDL-cholesterol determination, a direct method was used that did not require precipitation of the non-HDL cholesterol (chylomicrons, VLDL and LDL). LDL cholesterol was calculated in mmol L-1 using the Friedewald formula (Friedewald et al., 1972).
Determination of Serum Tame-Esterase Activity
Serum [TAME]-esterase activity was determined by a simple colorimetric micro
method (Seigelman et al., 1962). TAME was purchased from SIGMA chemical
Determination of Total Peroxides
The Biomedica-Oxystat assay was used to measure the peroxide concentration
in EDTA-plasma. The assay has a limit of detection of 7 μM L-1
with intra-assay and inter-assay coefficients of variation of 3.1 and 5.1%,
Normal and abnormal multiparametric sera (ABX Diagnostics N-control, P-control)
were used as internal quality controls in glucose and lipids assays.
Statistical analyses were performed using the Epi-Info 6.4 (CDC, Atlanta, USA) statistical software. Comparisons between mean values in groups with normally distributed data were done by ANOVA. For data with skewed distribution or when variance between groups was not homogenous, Kruskal-Wallis tests were applied. Associations between the different variables were examined by determination of Pearsons correlation coefficient. Data are expressed as mean±SD for normally distributed data and as median (interquartile range) for skewed data. p-value = 0.05 was considered as significant.
A total of 110 subjects (58 men and 52 women), mean age 47.5±8.3 years, were studied, grouped according to their LDL-cholesterol level. Based on the LDL-cholesterol values, qualitative traits were assigned to individuals, using the NCEP ATP III (2001) criteria. The optimal LDL subgroup consisted of 65 individuals (34 men, 31 women), the borderline LDL subgroup consisted of 23 individuals (8 men, 15 women) and the high LDL subgroup consisted of 21 individuals (16 men, 5 women). Findings (Table 1) show that there was no significant age difference between the different subgroups. Anthropometric variables were similar in all subgroups. Fasting glucose was slightly higher in the high-LDL subgroup but this difference did not reach statistical significance (p = 0.167). Post-load glycaemia (2 h glucose) was significantly higher in the borderline-high LDL subgroup compared to optimal-LDL subgroup (p = 0.042). Subjects in the borderline-high LDL subgroup had significantly higher plasma triglycerides in comparison with those in the optimal subgroup (p = 0.005). Total cholesterol and LDL-cholesterol were significantly higher in borderline-high and high-LDL subgroups (p<0.001) as compared to the optimal subgroup. HDL-cholesterol was higher in borderline-high and high-LDL subgroups but did not reach statistical significance (borderline, p = 0.086; high, p = 0.107). No significant differences in serum TAME-esterase activity and total peroxides concentrations were observed between the different subgroups (p>0.05).
Variables that correlated with TAME-esterase activity in men were post-load
glycaemia and total peroxides. In women, TAME-esterase activity was correlated
with fasting glucose and a slight correlation was observed with post-load glycaemia.
No correlation was observed with lipid values in men. However, TAME-esterase
showed a positive correlation with total cholesterol and a slight correlation
with triglycerides and LDL-cholesterol in women. No consistent correlation was
found with anthropometric variables in both groups. TAME-esterase activity was
slightly correlated with age in both men and women.
||Comparison of quantitative variables including physical and
biochemical characteristics in different LDL subgroups
|Data are expressed as means±SD for normally distributed
data and as median and range for skewed data. Differences between group
means were tested using analysis of variance (ANOVA). *Determination of
total peroxides was carried out on 84 individuals only (optimal LDL, n =
40; borderline LDL, n = 23; high LDL, n = 21). ap<0.05; bp<0.01;
cp<0.001 v/s optimal LDL subgroup
TAME-esterase activity showed a weak positive correlation with total peroxide concentrations in men (p = 0.06). No correlation was found in women.
Total peroxides showed no consistent correlation with fasting glucose and 2 h glucose. Correlation with lipid values showed that total cholesterol and LDL-cholesterol were positively correlated in men whereas in women these two lipid variables were negatively correlated. As stated above, total peroxides were correlated with TAME-esterase in men but no correlation was found in women. WHR showed a consistent negative correlation in both groups. Total peroxides were negatively correlated with age in women only.
The aetiology of cardiovascular disease is complex and multifactorial, with numerous predisposing factors (Warnholtz et al., 2001). Evidence increasingly suggests that various proatherogenic stimuli lead to increased production of reactive oxygen species within the endothelial microenvironment and the resulting oxidative stress plays a key role in mediating the pathologic manifestations of endothelial dysfunction associated with atherosclerosis (Cines et al., 1998). All of the factors associated with metabolic syndrome are interrelated. Obesity and lack of exercise tend to lead to insulin resistance. Insulin resistance has a negative effect on lipid production, increasing very low-density lipoprotein and LDL-cholesterol and triglycerides levels in the bloodstream and decreasing HDL-cholesterol. This can lead to fatty plaque deposits in the arteries, which, over time, can lead to cardiovascular disease, blood clots and strokes. Insulin resistance also leads to increased insulin and glucose levels in the blood. Excess insulin increases sodium retention by the kidneys, which increases blood pressure and can lead to hypertension. Chronically elevated glucose levels in turn damage blood vessels and organs, such as the kidneys and may lead to diabetes.
In the present study, TAME-esterase activity was found to be significantly higher in patients with impaired glucose metabolism as compared to normoglycaemic individuals.
There is evidence reporting that activation of the kinin-kallikrein system can enhance insulin sensitivity (Uehara et al., 1994). TAME-esterase is a component of the kallikrein family of proteases and one possible explanation to the increased TAME-esterase activity observed among our study population presenting with an abnormal glucose metabolism could be due to hyperactivation of the kinin-kallikrein. This response is an attempt to improve insulin sensitivity in individuals with insulin resistance syndrome. Various studies have shown that bradykinin, a kinin derived from the cleavage of high molecular weight kininogens by kallikreins, may have an effect in enhancing insulin action and insulin signalling at the skeletal muscle level, resulting in increase glucose uptake (Dietze et al., 1996). Animal model studies as well as interventional studies involving individuals with insulin resistance syndrome showed that in presence of ACE inhibitors, there were sustained elevations of circulating bradykinin concentration and improved whole body insulin sensitivity (Dietze et al., 1996). However, Dumke et al. (2002) have reported the possibility that kallikreins may influence post contraction insulin action via a bradykinin-independent mechanism. Kallikreins can act on proteins other than kininogens and it was postulated that through this catalytic activity, kallikreins might play a role in enhanced insulin action.
Our results suggest an increase in synthesis of kallikreins in patients with insulin resistance could be a compensatory mechanism towards enhancing insulin sensitivity and increase in glucose uptake. It is therefore proposed that TAME-esterase; a component of the kinin-kallikrein system may play an important role in insulin resistance and endothelial dysfunction. Insulin resistance per se plays an important role in the development of postprandial lipoprotein abnormalities observed in type 2-diabetes, even when fasting triglyceride levels are normal. A likely possible mechanism would thus seem to be a reduced inhibition of lipolysis (Watts and Playford, 1998).
In conclusion present study would thus tend to support the hypothesis that TAME-esterase activity may play a role during mechanisms leading to insulin resistance and endothelial dysfunction.