Mechanisms of increased vascular superoxide production in human varicose veins

Abstract

Varicose veins is one of the most common diseases observed in developed countries. Recent studies have shown that oxidative stress is increased in varicose veins, particularly in insufficient vessels. However all up-to- date studies focused primarily on antioxidant enzymes and used indirect and unspecific methods of ROS determination such as malondialdehyde (MDA) determination. Thus, in the present study we aimed to measure, superoxide anion production and further analyze the enzymatic sources of superoxide production in varicose veins in comparison to human saphenous veins obtained from subjects undergoing CABG. Moreover we compared superoxide production in distal and proximal segments of veins (control or varicose). We find increased superoxide production in varicose veins, in particular in distal segments of the vessel. The present study identified NADPH oxidases and uncoupled nitric oxide synthase as major sources or superoxide anion in varicose veins, indicating that these enzymes could become valuable drug targets for pharmacological treatment and prevention of varicose venous disease. These results may also link endothelial dysfunction, inflammation and immune activation to chronic venous dysfunction.

Introduction

Varicose disease is an important cause of morbidity and public health burden. The disease affects up to 20% of people in the developed countries and the occurrence increases with age to exceed 65% in women and 50% in men after the age of 45. While varicose veins constitute an important cosmetic problems such as pigmentation, lipo-dermatosclerossis and lead to certain functional limitations (e.g. pain-related) in patient’s activity, they are also associated with serious cardiovascular risks including superficial thrombophlebitis, and increased risk of deep vein thrombosis as well as ulcerations and hemorrhage. In spite of the importance of this disease options for treatment and prevention of this disease are limited at present. The traditional, most common treatment for varicose vein disease has been vein stripping to remove affected veins. While novel surgical approaches are being developed the advances in understanding the mechanisms of the disease are very limited. This is in turn related to the fact that for decades, no new treatment or prevention strategies have been introduced.

The hallmark of varicose vein disease is the insufficiency of venous valves, however the mechanisms leading to such primary dysfunction are not known. Endothelial dysfunction is one of possible mechanisms. It has been suggested recently that oxidative stress is increased in varicose veins. Increased reactive oxygen species (ROS) production has been implicated in the genesis of endothelial dysfunction characterized by loss of protective nitric oxide (NO) bioavailability in numerous vascular disease states including atherosclerosis, diabetes, hypertension, smoking. Finally, several recent studies have linked endothelial dysfunction to the development of subsequent venous valve dysfunction, which underlies varicose vein development. Moreover oxidative stress, through it’s effects on matrix metalloproteinases could be an important contributor to venous remodeling. Reactive oxygen species cause oxidation of lipid membranes, proteins and up-regulation of inflammation. Moreover we have previously shown that vascular oxidative stress is a systemic phenomenon related primarily to clinical risk factors. While initial studies have focused on the role of superoxide production and oxidative stress in arteries, we have shown that it is also very prominent in human veins and may play a role in human pathology.

In spite of this evidence the role of reactive oxygen species in human varicose disease has not been well defined. Two recent studies have shown that oxidative stress is increased in varicose veins, particularly in insufficient vessels. However these investigations focused primarily on antioxidant enzymes and used indirect and unspecific methods of ROS determination such as malondialdehyde (MDA) determination. While MDA is a product of oxidation by ROS it’s generation in biological systems is complex and does not allow for identification of individual ROS species involved. Therefore it is critical to understand which reactive oxygen species are primarily involved in oxidative stress and further define it’s enzymatic mechanisms in varicose veins. Thus, in the present study we aimed to measure, superoxide anion production from varicose vein segments in comparison to non-varicose human saphenous veins obtained from subjects undergoing CABG. We also analyzed the enzymatic sources of superoxide production in varicose veins. Finally, we compared superoxide production in distal and proximal segments of veins (control or varicose) in order to gain insight into the role of different degrees of blood reflux and stagnation, venous pressure increase in the regulation of vascular oxidative stress in varicose veins.

Patients and Methods

Patients and Blood Vessels

Segments of proximal and distal human saphenous vein was obtained from subjects with varicose vein disease (VV; n=14) and from subjects undergoing bypass graft surgery (excess of vein obtained from proximal and distal vein was compared; HSV n=15). Subjects were matched for major risk factors for atherosclerosis, sex and age known in previous studies to affect vascular oxidative stress and superoxide production. Vessel segments were harvested using a no-touch technique, before surgical distension (HSV) or rapid removal of varicose veins (VV). Vessel fragments were immediately transferred to ice cold Krebs-HEPES buffer (99 mM NaCl; 4.7 mM KCl; 1.2 mM MgSO₄; 1mM KH₂PO_4; 1.9mM CaCl₂; 25 mM NaHCO₃; 11.1 mM Glucose and 20mM HEPES), delicately flushed and carefully dissected to remove excess adventitial tissue, using microsurgical instruments. All vessels were collected before topical administration of drugs such as papaverine. Patient characteristics and risk factor profile was typical for patients with systemic vascular diseases. Collection of tissue specimens was approved by the Local Research Ethics Committee and informed consent was obtained.

Vascular superoxide production

Superoxide production was measured by lucigenin-enhanced chemiluminescence (LGCL), using previously described and validated methods (6, 20). Briefly, intact vessel segments were equilibrated in Krebs-HEPES gassed with 95%O₂/5%CO₂ for 30 minutes at 37°C. Lucigenin-enhanced chemiluminescence from intact vessels was measured in buffer (2 ml) containing low concentration lucigenin (5 µM). Superoxide production was expressed as relative light units (RLU) per mg of dry weight of the vessel as described before.

Determination of sources of vascular superoxide production

In order to address the sources of superoxide release in varicose human saphenous veins superoxide release was measured in the presence various potential oxidase inhibitors including diphenyliodonium (DPI, 10 umol/L, flavin oxidase inhibitor), apocynin (Apoc; NADPH oxidase inhibitor; 300umol/L); oxypurinol (100 umol/L; xanthine oxidase inhibitor); NG-nitro-L-arginine methyl ester (L-NAME; nitric oxide synthase inhibitor; 100 umol/L) and rotenone (100umol/L; mitochondrial oxidase inhibitor). Superoxide production was expressed as relative light units (RLU)/second/mg vessel dry weight.

Statistical analysis

Results are expressed as means ± SEM with n equal to the number of patients. Statistical comparisons between the two groups were made using Students t-test for independent or dependent samples. P values <0.05 were considered statistically significant.

Results

Clinical characteristics of patients

Vessels were obtained from total of 29 patients (14 subjects undergoing varicose vein surgical stripping and 15 control non-varicose subjects undergoing CABG). Demographic and clinical characteristics, shown in Table 1 demonstrates that patients were matched for sex, age and major risk factors for atherosclerosis known to affect vascular oxidative stress.

Table 1. Major demographic and clinical characteristics of studied patients.

Values are mean±SEM or n (%). ACE denotes angiotensin converting enzyme.

	Non-VV N=15	VV N=14
Age (year;mean±SEM)	56±4.0	52±3.5
Sex (F:M)	11:4	11:3
Risk Factors:
Current Smoking	6 (40%)	7 (50%)
Hypertension	9 (60%)	8 (57%)
Diabetes Mellitus	2 (13%)	2 (14%)
Hypercholesterolemia	7 (47%)	6 (42%)
Medication:
ß-blockers	8 (53%)	7 (50%)
Aspirin	12 (80%)	13 (92%)
Lipid-lowering agents	10 (66%)	10 (71%)
Calcium antagonists	6 (40%)	5 (35%)
ACE inhibitors	9 (60%)	8 (57%)

Superoxide production from varicose and non-varicose veins

Basal superoxide production from both control and varicose veins was determined by LGCL from intact vessel rings obtained from either proximal (typically mid-thigh) or distal (typically calf) location of the vessel. Specificity for superoxide detection was confirmed by co-incubation with superoxide dismutase (SOD).

In additional experiments (n=5) both varicose and non-varicose vessels were assayed with and without incubation with polyethylene glycol (PEG)-conjugated SOD (500U/ml), which resulted in a very significant inhibition of LGCL signal (92±6 % inhibition in control HSV and 93±8% inhibition in varicose veins; data not shown).

Superoxide production was observed in all studied vessels. It was significantly higher in varicose veins than in control vessels (Figure 1). This increase was observed in both proximal and distal segments of the vessels. Detailed comparison of proximal and distal fragments of vessels revealed that distal segments of varicose veins but not of the non-varicose control vessels demonstrate almost 2-fold increase in basal superoxide production when compared to proximal fragments (Figure 1).

Figure 1.

Vascular superoxide anion production in human varicose veins. Superoxide production was measured in paired proximal and distal segments of control, non-varicose human saphenous vein from patients undergoing CABG (n=15) and from varicose veins obtained from patients undergoing varicose vein removal (n=14). Mesurements were performed using 5 umol.L lucigenin enhanced chemiluminescence. *- P<0.05 vs HSV; LGCL – lucigenin enhanced chemiluminescence; dw- dry weight

Relationships between proximal and distal superoxide production

To gain further insight into the differences in superoxide production between distal and proximal segments of studied veins we analyzed the relationship between absolute amounts of superoxide produced by distal and proximal vessel segments in either control or varicose veins. Interestingly, superoxide production was very significantly correlated between proximal and distal vascular segments in non-varicose veins but not in varicose veins (Figure 2).

Figure 2.

Relationships between superoxide anion production in proximal and distal segments of veins obtained from control subjects with atherosclerosis (left panel; n=15) and varicose vein patients (right panel; n=14).

Sources of vascular superoxide production in human varicose disease

To investigate enzymatic sources of superoxide production in varicose vessels, we measured superoxide production from distal vascular segments following pre-incubation with a range of potential oxidase inhibitors (Figure 3). Superoxide production was very significantly inhibited by diphenylene iodonium, an inhibitor of flavin containing oxidases such as NADPH oxidases. Similar degree of inhibition was observed in the presence of apocynin, which is considered a relatively specific inhibitor of NADPH oxidase. Oxypurinol, a xanthine oxidase inhibitor or rotenone – inhibitor of mitochondrial electron transport had minimal effects on superoxide production from varicose vein. However the response to inhibition of nitric oxide synthase (NOS) with L-NAME demonstrated very consistent statistically significant (by ca. 25%) inhibition in all studied vessels.

Figure 3.

Enzymatic sources of superoxide anion in human varices. Superoxide production was determined y LGCL (5 umol/L) in the absence and presence of various oxidase inhibitors. Distal varicose vein segments were incubated for 30 minutes before and during superoxide determination with: diphenyleneiodonium (DPI; 10 umol/L); apocynin (300 umol/L), oxypurinol (100 umol/L); NG-nitro-L-arginine methyl ester (L-NAME; 100 umol/L) and rotenone (100umol/L); * p<0.05 vs native

Discussion

We have compared superoxide production between varicose and non-varicose veins and used proximal and distal segments of these vessels as a model system to assess the effects of venous pressure increase in the regulation of vascular oxidative stress in human veins. Our study, for the first time demonstrates that varicose veins generate significantly more superoxide anions. Importantly, increased venous blood pressure related to blood reflux and stagnation observed in distal segments of VV is related to further increase in superoxide anion production. At the same time, in the conditions of non-varicose vein, with fully functional venous valve system, superoxide production is not increased in distal segments of HSV. To gain further insight into the nature (local or systemic) factors which regulate superoxide production in VV we studied the relationships between superoxide production in proximal and distal vessel segments. Interestingly in while in non-varicose HSVs there was a strong correlation of superoxide production between proximal and distal segments, such relationship was not observed in VVs. This may indicate that while in non varicose veins systemic factors are the most critical in the regulation of superoxide production, in VV local factors may play an important role, which exceeds the effect of systemic regulation of oxidative stress in humans. We have previously identified a systemic nature of both endothelial dysfunction and oxidative stress in patients with atherosclerosis. Both endothelial function and superoxide production, as well as major oxidase expression are all correlated not only between different areas of one vascular bed but also in distant vascular beds including venous and arterial systems. Multivariate analysis performed in these studies identified that major risk factors for atherosclerosis are major regulators of systemic vascular oxidative stress. In particular role of diabetes and hypercholesterolemia has been defined as major factors increasing superoxide production in veins in general. However resu;ts of the present study show that apart from systemic effects, local disease environment may be important in enhancing superoxide production and oxidative stress in varicose veins. Lack of correlation between superoxide production in distal and proximal segments of VV may indicate that factors such as different degrees of blood reflux and stagnation, variable venous pressure have stronger effects than systemic factors. Alternatively to pressure changes it is possible that certain locally released humoral factors such as chemokines, cytokines, or certain hypoxia related metabolites. This may further our understanding of the propagation of VV rather than solely be related to their initiation. Obviously, further direct studies, which in larger cohorts would compare vascular superoxide production with clinically assessed local pressure-related or other biological and chemical factors.

Our studies are in line with previous studies which have measured oxidative stress markers in varicose veins and shown a significant increase of MDA which was also related to venous insufficiency. Krzysciak and Kozka have nicely demonstrated increased oxididative stress in insufficient veins and shown that the degree of oxidative damage may correlate with clinical degree of venous disease.

Our study however adds to that knowledge by assessing superoxide production directly from vascular segments, rather than indirect lipid oxidation products. Interestingly, previous studies also demonstrated that superoxide dismutase activity is increased in insufficient or varicose veins.

While we have not described that in the present study, we were unable to define clear differences in either SOD activity or expression in human VV (data not shown). This is in line with our previous studies showing that Cu-Zn SOD nor MnSOD expression and activity were unchanged in human veins in coronary artery disease. We did not, however, assess ec-SOD levels, which could actually be the most important dismutase variant in varicose disease.

Reactive oxygen species, and in particular superoxide anions, may play numerous roles in the pathogenesis of chronic venous insufficiency and varicose disease.

In general current view of the initiation of varicose vein disease include increased destruction of collagen and matrix proteins which are initiated by endothelial dysfunction, characterized by loss of NO and PGI2 bioavailability and subsequently increased vascular inflammation, initiated by increased adherence of leukocytes to venous endothelium. ROS may regulate these processes on number of levels (Figure 4). Firstly, superoxide anion is the most important cause for biodegradation of nitric oxide in blood vessels. This rapid reaction, exceeding Km of any antioxidant enzymes, results in the formation of another strong oxidant, peroxinitrite. We have previously described this reaction and it’s consequences in normal human veins.

Figure 4.

Putative role of superoxide anion and vascular oxidases in varicose vein formation and propagation.

ROS may also directly damage cell lipid membranes. This leads to concomitant cellular injury of either endothelium or smooth muscle cells, as well as lipid rafts dysfunction which may result with vascular and venous valve dysfunction. ROS are also involved in the induction of inflammatory reactions, thus contributing to the role of immune reactions that may occur and enhance vascular damage in varicose disease. Finally, ROS are potent inducers of matrix metalloproteinases. They both increase the activity and expression of MMP- and MMP- and are able to inhibit the expression of Tissue inhibitors of matrix metalloproteinases (TIMPs). All of these factors may be critical for the initiation of venous dysfunction and may lead to decreased ve;ocity of bllod flow through the vein, which further exacerbates the pathologic changes This may lead to venous dysfunction with the development of decreased blood Blood retracts and accumulates in peripheral veins leading to venous pressure increase and vascular wall dilation(4). Vein tension and hypoxia activate vascular wall cells and infiltrating into vascular wall leukocytes(1) closing vicious circle of varicose veins formation. Many of the risk factors known to affect the development of chronic venous insufficiency like ischemia(2) or angiotensin II(3) lead to superoxide and other ROS production through oxidases activation. In the present study we show that O2-. Production was increased in the distal segments of VV which are characterized by increased hemodynamic pressure, which points to the role of ROS in the propagation of varicose disease, as it can be further enhanced by local hemodynamic conditions.

While the results shown in this study provide for the first time direct measure of superoxide production from varicose veins, the study has some limitations. Firstly it would have been very informative to compare further clinical characteristics of venous functions with measures of superoxide production. Such comparisons however would be particularly valuable in larger study populations, where multivariate analysis would have had been applied to unequivocally answer this question, which is so important from clinical point of view. We used lucigenin enhanced chemiluminescence to measure superoxide production. This approach has been criticized in the past, particularly, in relation to the use of high concentrations of lucigenin (above 20uM) which could artificially enhance superoxide detection through redox cycling. Thus in the present study we used low concentrations of lucigenin which are highly validated measures of superoxide production and are free from “redox cycling” problem.

In summary, in the present study we demonstrate for the first time that varicose veins produce increased amounts of superoxide anion than normal non-varicose veins, particularly in distal segments of the varices. NADPH oxidases are the primary sources of superoxide anion in the varicose veins, in line with earlier findings in normal, control veins from atherosclerotic subjects.

Our study taken together with previous descriptions of increased oxidative stress in human varicose disease indicates that antioxidant approaches could be valuable in treating venous dysfunction. However these approaches should rather be local and possibly directed on inhibiting NADPH oxidase activity in these veins. Further clinical studies are needed to confirm this interesting possibility.