Selenium supplementation improves antioxidant capacity in vitro and in vivo in patients with coronary artery disease:
The SElenium Therapy in Coronary Artery disease Patients (SETCAP) Study
Article Outline
Background
Selenium is a central determinant of antioxidative glutathione peroxidase 1 (GPx-1) expression and activity. The relevance of selenium supplementation on GPx-1 in coronary artery disease (CAD) needs to be established. We assessed the effect of selenium supplementation on GPx-1 in cell culture and on endothelial function in a prospective clinical trial.
Methods
Human coronary artery endothelial cells were incubated with 5.78 to 578 nmol/L sodium selenite, Se-methyl-selenocysteine hydrochloride, or seleno-l-methionine. Glutathione peroxidase 1 mRNA and protein expression and activity were measured. Coronary artery disease patients (n = 465) with impaired endothelial function (flow-mediated dilation [FMD] <8%) were randomly assigned to receive 200 or 500 μg sodium selenite daily or matching placebo during a 12-week period. We tested the effect on red blood cell GPx-1 activity and brachial artery FMD. Furthermore, differences in biomarkers of oxidative stress and inflammation were measured.
Results
Sodium selenite and Se-methyl-selenocysteine hydrochloride increased GPx-1 protein and activity in a dose-dependent manner (P < .0001). The intention-to-treat groups comprised 433 CAD patients. Glutathione peroxidase 1 activity increased from 37.0 U/gHb (31.3-41.7) to 41.1 U/gHb (35.2-48.4) (P < .0001) in the 200 μg and from 38.1 U/gHb (33.2-43.8) to 42.6 U/gHb (35.0-49.1) (P < .0001) in the 500 μg sodium selenite group treated for 12-weeks. No relevant changes were observed for FMD or biomarkers of oxidative stress and inflammation.
Conclusions
Sodium selenite supplementation increases GPx-1 activity in endothelial cells and in CAD patients. Future studies have to demonstrate whether long-term CAD outcome can be improved.
Oxidative stress is central to coronary artery disease (CAD). Glutathione peroxidases (GPx's) constitute the major antioxidant defense system1, 2 and GPx-1 is one of the most abundant isoforms within eukaryotic cells.3 In vivo studies in knockout and transgenic mice have found that GPx-1 can modulate vascular function.3, 4
In this context, GPx-1–deficient mice are significantly more susceptible to oxidative challenge5, 6 and have endothelial dysfunction and abnormal cardiac function after ischemia/reperfusion injury.7 Other studies suggest that increased expression of GPx-1 in transgenic mice decreases tissue damage after cerebral and myocardial ischemia/reperfusion.8
In clinical studies, high activity of GPx-1 protects from future cardiovascular events and modifies risk associated with conditions of heightened oxidative stress, such as smoking9 or elevated homocysteine levels.10 Consequently, increasing GPx-1 activity constitutes an attractive approach for cardiovascular risk modification.
Selenium is a key component of GPx-1 required for opti mal enzyme activity. In Europe, the bioavailability of selenium is generally low.11, 12 An inverse relationship between selenium concentration and incident cardiovascular disease and mortality was observed.13, 14, 15, 16 In patients with acute coronary syndrome and ischemic cardiomyopathy, selenium concentrations are transiently decreased.17, 18
However, epidemiologic data supporting an association between circulating selenium levels and cardiovascular risk are inconsistent to date, and whether selenium supplementation promotes cardiovascular risk protection remains controversial.19
The pathways by which selenium may exert its potential protective effects in vivo are manifold. A direct link to GPx-1 suppression or activation, however, has not yet been established. Using a translational approach, we performed cell culture experiments to study GPx-1 expression and activity in human coronary artery endothelial cells (HCAEC) treated with different sources of selenium and also prospectively investigated the effect of sodium selenite supplementation on GPx-1 activity and endothelial function in patients with CAD.
Methods
Cell culture and cell treatment
Human coronary artery endothelial cells were maintained in endothelial basal medium supplemented with the endothelial basal medium–bullet kit (Lonza, Walkersville, MD) and used up to passage 10. Near-confluent HCAEC were incubated with 5.78 to 578 nmol/L sodium selenite, Se-methyl-selenocysteine hydrochloride, or seleno-l-methionine (Sigma-Aldrich, St. Louis, MO) for 4 days in complete medium. RNA was isolated (Qiagen Sciences, Germantown, MD), complementary DNA synthesized (Clontech, Mountain View, CA), and human GPx-1 transcript analyzed using quantitative reverse-transcription-polymerase chain reaction. Glutathione peroxidase 1 protein levels were determined by Western blot analysis and GPx-1 activity by an indirect assay as described elsewhere.20 Laboratory details on the experiments are provided in Appendix A.
Clinical trial design
The double-blind, placebo-controlled, randomized, single center, 3-armed, clinical phase II trial SElenium Therapy in Coronary Artery Disease Patients (SETCAP) evaluated the effect of a 12-week 200 or 500 μg sodium selenite supplementation against placebo on noninvasive measurement of endothelial function and GPx-1 activity in patients with stable CAD. The study protocol is outlined in Figure 1, and detailed inclusion/exclusion criteria are provided in Table I. In brief, the treatment was scheduled to last 12 ± 1 weeks. Patients were invited for a control visit after 6 ± 1 weeks. Data on potential side effects were documented and compliance was assessed. The study was approved by the local ethics committee of the University of Mainz (Mainz, Germany). Participation was voluntary; patients were enrolled after written informed consent was obtained.
Figure 1.
Study design and patient sample of the SETCAP trial. ITT indicates intention to treat.
Table I. Inclusion and exclusion criteria of the SETCAP trial
| Inclusion criteria |
| (1) Documented CAD treated by a standard therapy in accordance with current guidelines |
| (2) FMD <8% |
| (3) Male and female patients ≥18 y old |
| (4) Written informed consent |
| Exclusion criteria |
| (1) Left ventricular ejection fraction <30% |
| (2) Uncontrolled hypertension (blood pressure >180/100 mm Hg) or systolic blood pressure <100 mm Hg |
| (3) Renal impairment (serum creatinine ≥2.5mg/dL) |
| (4) Known hepatic disease or γ-glutamyl transferase >3× upper limit of normal range |
| (5) Manifest inflammation or infectious disease with clinical symptoms |
| (6) Malignancy with treatment indication |
| (7) HIV infection |
| (8) Initiation of therapy with any of the following medications within the last 12 weeks: |
 -Aspirin |
| -Lipid-lowering therapy |
| -Calcium-channel blockers |
| -ACE inhibitors |
| -AT1 receptor antagonists |
| -Hormone replacement therapy |
| -Clopidogrel |
| -Vitamins or other antioxidant substances, the daily dose of which exceeds the recommendations of the German Society of Nutrition |
| -High-dose corticosteroid therapy or therapy with cytostatic substances within the last year |
| -Long-term medication with NSAIDs except for aspirin |
| -Concomitant therapy with selenium-containing substances exceeding 50 μg/d |
| (9) Current pregnancy or breastfeeding |
| (10) Participation in another clinical trial within the 4 wk before study entry |
| (11) Former participation in this clinical trial |
| (12) Concomitant participation in another clinical trial |
| (13) Planned participation in another clinical trial within 4 wk after termination of the current trial |
| (14) Alcohol, medication, or drug dependency |
| (15) Unreliability as a participant in a clinical trial as assessed by the clinician |
| Outcomes assessed during 12 wk of sodium selenite supplementation |
| Primary outcome |
| (1) Change in whole blood selenium concentration |
| (2) Change in GPx-1 activity |
| Secondary outcome |
| (1) Change in FMD (measured in percentage) |
| (2) Change in nitroglycerin-mediated dilation |
| (3) Change in inflammatory biomarkers |
| (4) Change in serum lipid parameters |
Laboratory methods
Blood samples were drawn under standardized conditions after a minimum of 12 hours fast. Blood samples were obtained at the screening visit, at 6 ± 1 weeks (selenium [Biosyn Laboratories, Arzneimittel GmbH, Fellbach, Germany], GPx-1 [Randox Laboratories Ltd., Crumlin Co., Antrim, United Kingdom], and the safety parameters), and at 12 ± 1 weeks. Safety markers (creatinine, liver enzymes) and biomarkers for secondary analyses (blood lipids, C-reactive protein [CRP], fibrinogen) were measured by routine methods or by commercially available assays (Cu/Zn superoxide-dismutase, immunoluminometric assay, (BRAHMS [Hennigsdorf, Germany]), and 8-isoprostanes, competitive enzyme immunoassay; Cayman Chemical, Tallinn, Estonia).
Noninvasive endothelial function testing comprising flow-mediated dilation (FMD) and nitro-mediated dilation of the brachial artery as well as echocardiographic studies were performed at enrollment and after 12 ± 1 weeks as described previously.21 Additional information is provided in Appendix A.
Statistical considerations
Continuous variables were described using median and first and third quartiles, and categorical variables using absolute and relative frequencies. To investigate the change in continuous variables during the study, differences were calculated as the value at the last visit (12 weeks) minus the value at the baseline visit. Statistical details are provided in Appendix A.
There were 3 hierarchically ordered null hypotheses. Null hypothesis 1 was that there is no difference in the change of FMD between the group receiving 500 μg sodium selenite per day and the group receiving placebo. Null hypotheses 2 and 3 were formulated analogously for the comparison of the group receiving 200 μg sodium selenite with the placebo group and the comparison of both groups receiving supplemental selenium. The hypotheses were tested on a significance level of 5% (2-sided) with a Wilcoxon test in a confirmatory manner. The following null hypothesis was tested in a confirmatory manner only if a null hypothesis could be rejected; otherwise, it was tested in an exploratory manner. Other P values have not been corrected for multiple testing and must be considered descriptive. For experimental data, the effect of different selenium compounds on GPx-1 was tested by analysis of variance. Statistical analyses were performed with SAS 9.1 (SAS Institute, Cary, NC).
Results
Selenium treatment modifies expression and activity of GPx-1
Sodium selenite treatment increased GPx-1 protein expression dose-dependently, achieving a 2.4-fold maximal increase at 289 nmol/L in HCAEC (Figure 2, A); this increase was associated with a 1.9-fold increase in GPx-1 activity (203 ± 9.2 vs 109 ± 6.2 mmol min−1 mg−1, P < .0001) (Figure 2, B). The highest concentration of sodium selenite used, 578 nmol/L, was associated with a decrease in GPx-1 expression and accompanied by increased cell death. Selenium supplementation did not affect mRNA expression (data not shown). Se-methyl-selenocysteine hydrochloride treatment significantly increased GPx-1 protein and activity compared to untreated HCAEC (P < .0001), achieving a maximal 1.6-fold increase at 57.8 nmol/L. Seleno-l-methionine treatment did not have a significant effect on expression or activity (Figure 2, C and D).
Figure 2.
A, Glutathione peroxidase 1 protein in HCAEC after 4 days of sodium selenite treatment. B, Glutathione peroxidase 1 activity levels in HCAEC after 4 days of sodium selenite treatment. *P < .0001, #P < .05 vs no treatment (n = 8). C, Glutathione peroxidase 1 protein levels in HCAEC after 4 days of Se-methyl-selenocysteine hydrochloride or seleno-L-methionine. D, Glutathione peroxidase 1 activity levels in HCAEC after 4 days of Se-methyl-selenocysteine hydrochloride or seleno-L-methionine. *P < .0001 vs no treatment (n = 9). Data are expressed as mean ± standard error of the mean.
Clinical study—patient characteristics
We screened 669 patients with documented CAD, 465 of whom met the inclusion criteria (Figure 1). Baseline characteristics of the study population are given in Table II. The median age was 66 years, 18.5% were women, and the proportion of traditional risk factors was similar to that in a general CAD population.9 Randomization seemed to be intact.
Table II. Baseline characteristics of the study population according to treatment arm
| Placebo (n = 144) | Sodium selenite 200 μg (n = 148) | Sodium selenite 500 μg (n = 141) | P | |
|---|---|---|---|---|
| Age (y) | 66 (61-70) | 66 (61-70) | 66 (58-72) | .99 |
| Male sex (%) | 115 (79.9) | 118 (79.7) | 120 (85.1) | .41 |
| Body mass index (kg/m2) | 28.1 (26.1-30.8) | 28.4 (25.7-30.7) | 27.8 (25.4-30.7) | .78 |
| Hypertension (%) | 124 (86.1) | 124 (83.8) | 116 (82.3) | .67 |
| Systolic blood pressure (mm Hg) | 140 (120-150) | 130 (120-150) | 140 (120-150) | .83 |
| Heart rate (beat/min) | 61 (56-67) | 61 (55-67) | 63 (58-70) | .12 |
| Smoking | .55 | |||
| Former (%) | 60 (42) | 56 (38) | 64 (45) | |
| Current (%) | 17 (12) | 21 (14) | 12 (9) | |
| Diabetes (%) | 30 (21) | 34 (23) | 29 (21) | .86 |
| Cardiac medication | ||||
| ACE inhibitors (%) | 97 (67) | 98 (66) | 74 (53) | .02 |
| β-Blockers (%) | 116 (81) | 124 (83) | 111 (79) | .54 |
| Angiotensin receptor antagonists (%) | 23 (16) | 25 (17) | 36 (26) | .08 |
| Metabolic parameters⁎ | ||||
| Total cholesterol (mg/dL) | 182 (163-204) | 179 (158-205) | 184 (162-210) | .33 |
| HDL cholesterol (mg/dL) | 53 (46-64) | 53 (45-62) | 58 (49-64) | .06 |
| LDL cholesterol (mg/dL) | 99 (78-118) | 92 (77-115) | 100 (77-117) | .71 |
| Triglycerides (mg/dL) | 136 (91-200) | 134 (99-188) | 142 (99-200) | .78 |
| Serum creatinine (mg/dL) | 1.0 (0.9-1.1) | 1.0 (0.9-1.2) | 1.0 (0.9-1.1) | .59 |
| Glucose (mg%) | 104 (97-117) | 104 (96-119) | 103 (96-119) | .97 |
| Inflammatory parameters | ||||
| Fibrinogen (mg/dL)⁎ | 288 (259-330) | 285 (249-338) | 287 (255-328) | .75 |
| CRP (mg/L) | 2.2 (0.5-4.5) | 2.0 (1.1-3.5) | 1.5 (0.5-3.0) | .1 |
| Selenium and GPx′s | ||||
| Selenium (μg/L)⁎ | 93.4 (87.0-105.9) | 97.4 (87.7-104.9) | 99.0 (89.3-109.1) | .11 |
| GPx-1 (U/gHb) | 34.9 (31.0-40.3) | 37.0 (31.3-41.7) | 38.1 (33.2-43.8) | .02 |
| Echocardiographic parameters | ||||
| EDV (mL) | 93 (70-109) | 96 (79-123) | 98 (81-120) | .02 |
| ESV (mL) | 39 (30-53) | 42 (32-60) | 44 (34-57) | .06 |
| LV ejection fraction (%) | 54 (48-61) | 54 (49-60) | 55 (49-60) | .95 |
| Measures of vascular function | ||||
| FMD (%) | 4.9 (3.2-6.7) | 5.5 (3.3-7.2) | 4.8 (2.6-6.3) | .1 |
| Nitroglycerin-mediated dilation (%) (n=354)† | 10.8 (6.8-14.2) | 11.2 (7.5-15.1) | 10.9 (6.4-14.3) | .46 |
| Baseline brachial artery diameter (FMD; mm) | 5.0 (4.5-5.4) | 4.8 (4.3-5.5) | 5.0 (4.7-5.4) | .27 |
| Hyperemic brachial artery diameter (FMD; mm) | 5.2 (4.7-5.6) | 5.0 (4.5-5.8) | 5.3 (4.9-5.6) | .45 |
| Baseline brachial artery diameter (NMD; mm) | 5.0 (4.5-5.4) | 4.8 (4.3-5.4) | 5.1 (4.7-5.4) | .05 |
| Post nitro brachial artery diameter (NMD; mm) | 5.5 (5.1-5.9) | 5.4 (4.9-6.0) | 5.6 (5.3-6.0) | .07 |
⁎Lipid variables and fibrinogen were not obtained in one patient in the placebo group. Baseline selenium concentrations were not available in 6 patients in the placebo group, 7 patients in the 200 μg group, and for 6 patients in the 500 μg group. Glutathione peroxidase 1 baseline activity was not valid in 2 patients in the 200 μg group and 1 patient in the 500 μg group. |
†Nitro-mediated dilation was only available in 354 patients. |
Among the patients who were assigned to the treatment groups, 93.9% took sodium selenite supplementation as advised. No relevant differences were observed for side effects, dropout rates, or safety parameters in the sodium selenite intake treatment groups compared to the placebo group.
Sodium selenite and selenium concentration and GPx-1 activity
Circulating selenium concentrations were higher after the 12-week supplementation compared to placebo (P < .0001). Baseline selenium concentrations showed a median of 96.7 μg/L (87.7-106.3). After 12 weeks, a dose-dependent increase in selenium levels was observed, with the highest selenium concentrations in the 500 μg group (141.0 μg/L [127.5-152.0]) (Table III). The increase in selenium levels in the groups treated with sodium selenite was greater in the first half of the trial period than in the second half.
Table III. Absolute values and difference of parameters according to visit for selenium, GPx-1, FMD, and NMD
| Placebo (n = 144) | Sodium selenite 200 μg (n = 148) | Sodium selenite 500 μg (n = 141) | P | |
|---|---|---|---|---|
| Selenium concentration | ||||
| SeleniumV0 | 93.4 (87.0-105.0) | 97.4 (87.7-104.9) | 99.0 (89.3-109.1) | .11 |
| SeleniumV1 | 95.2 (87.8-103.9) | 118.6 (111.5-126.2) | 137.5 (128.1-149.4) | <.0001 |
| SeleniumV2 | 95.9 (88.8-104.2) | 122.9 (113.5-131.1) | 141.0 (127.5-152.0) | <.0001 |
| Change in selenium concentration | ||||
| SeleniumV1-V0 | 0.2 (−5.8 to 5.1) | 22.8 (13.3-30.5) | 37.6 (28.8-48.1) | <.0001 |
| SeleniumV2-V1 | 0.4 (−4.9 to 7.2) | 3.3 (−4.3 to 9.5) | 4.1 (−4.6 to 11.9) | .13 |
| SeleniumV2-V0 | 0.2 (−6.6 to 9.5) | 24.2 (15.3-35.6) | 41.2 (29.9-51.3) | <.0001 |
| GPx-1 activity | ||||
| GPx-1V0 | 35.0 (31-40.3) | 37 (31.3-41.7) | 38.1 (33.2-43.8) | .03 |
| GPx-1V1 | 35.7 (30.9-42.1) | 39.1 (33.2-46.0) | 40.9 (35.8-46.9) | <.0001 |
| GPx-1V2 | 35.5 (30.8-41.2) | 41.1 (35.2-48.4) | 42.6 (35.0-49.1) | <.0001 |
| Change in GPx-1 activity | ||||
| GPx-1V1-V0 | 0 (−3.1 to 3.5) | 1.9 (−1.3 to 6.2) | 3.6 (−1.1 to 6.8) | .0004 |
| GPx-1V2-V1 | 0.2 (−3.4 to 4.2) | 1.9 (−2.2 to 7.1) | 1.0 (−3.0 to 6.0) | .02 |
| GPx-1V2-V0 | −0.1 (−2.7 to 3.9) | 3.6 (0.0-8.9) | 3.9 (−0.2 to 10.2) | <.0001 |
| FMD | ||||
| FMDV0 | 4.6 (3.2-6.7) | 5.5 (3.3-7.2) | 4.8 (2.6-6.3) | .1 |
| FMDV2 | 5.9 (4.2-7.9) | 6.1 (4.2-8.3) | 5.5 (3.6-7.8) | .23 |
| Change in FMD | ||||
| FMDV2-V0 | 1.1 (−0.6 to 3.4) | 0.9 (−1.0 to 3.6) | 0.9 (−1.4 to 3.3) | .71 |
| Nitroglycerin-mediated dilation | ||||
| NMDV0 | 10.8 (6.8-14.2) | 11.2 (7.5-15.1) | 10.9 (6.4-14.3) | .47 |
| NMDV2 | 10.8 (8.1-13.5) | 11.0 (7.5-14.9) | 11.4 (8.1-13.8 | .77 |
| Change in nitroglycerin-mediated dilation | ||||
| NMDV2-V0 | 0.1 (−2.5 to 3.1) | −0.4 (−2.6 to 2.9) | 0.6 (−2.5 to 4.0) | .79 |
Table III shows the primary and secondary outcome variables. Median GPx-1 activity was 36.4 U/gHb (31.6-41.9) in the overall cohort at baseline. In both treatment groups, GPx-1 activity increased after 6 and 12 weeks (P < .0001). Although the 6-week value was higher in subjects receiving 500 μg selenium compared with those receiving 200 μg, this difference did not reach a P value of <.05, and after 12 weeks, GPx-1 activity did not differ relevantly among groups (P = .624). Median increases in GPx-1 activity from baseline values were 1.9 U/gHb (after 6 weeks, P = .006 vs placebo) and 3.6 U/gHb (after 12 weeks, P < .0001 vs placebo) in the 200 μg supplementation group, and 3.6 U/gHb (after 6 weeks, P < .0001 vs placebo, P = .217 vs 200 μg group) and 3.9 U/gHb (after 12 weeks, P < .0001 vs placebo, P = .624 vs 200 μg group) in the 500 μg intervention group.
Impact of sodium selenite supplementation on FMD
As demonstrated in Table III, sodium selenite supplementation did not have a significant effect on FMD compared to placebo (P = .430).
Secondary outcomes
None of the inflammatory (CRP and fibrinogen), oxidant stress markers (8-isoprostanes and Cu/Zn superoxide dismutase), or blood lipids were affected by sodium selenite (Table IV).
Table IV. Absolute values and difference of parameters by visit
| Placebo (n = 144) | Sodium selenite 200 μg (n = 148) | Sodium selenite 500 μg (n = 141) | P | |
|---|---|---|---|---|
| Inflammatory markers | ||||
| CRP (mg/L) V0 | 2.2 (0.5-4.5) | 2.0 (1.1-3.5) | 1.5 (0.5-3.0) | .1 |
| CRP V1 | 1.7 (0.5-3.4) | 1.7 (0.5-4.1) | 1.5 (0.5-3.1) | .76 |
| CRP V2 | 2.0 (1.0-3.6) | 2.0 (0.5-4.2) | 1.5 (0.5-3.3) | .14 |
| Change in CRP concentration | ||||
| CRP V1-V0 | −0.1 (−1.1 to 0.2) | 0.0 (−0.8 to 0.8) | 0.0 (−0.6 to 0.7) | .04 |
| CRP V2-V1 | 0.0 (−0.6 to 0.7) | 0.0 (−0.5 to 1.2) | 0.0 (−0.6 to 0.6) | .45 |
| CRP V2-V0 | 0.0 (−1.3 to 0.7) | 0.0 (−0.65 to 0.8) | 0.0 (−0.6 to 0.7) | .34 |
| Fibrinogen (mg/dL) V0 | 288 (259-330) | 285 (249-338) | 287 (255-328) | .75 |
| Fibrinogen V2 | 310 (272-372) | 317 (263-375) | 310 (263-361) | .77 |
| Change in fibrinogen concentration | ||||
| Fibrinogen V2-V0 | 17 (−20 to 74) | 28 (−16 to 74) | 19 (−22 to 70) | .73 |
| Serum lipids | ||||
| HDL cholesterol (mg/dL) V0 | 53 (46-64) | 53 (45-62) | 58 (49-64) | .06 |
| HDL cholesterol V2 | 53 (46-65) | 54 (46-63) | 55(49-64) | .33 |
| Change in HDL cholesterol | ||||
| HDL cholesterol (mg/dL) V2-V0 | 0.0 (−3.0 to 4.0) | 1.0 (−3.0 to 4.0) | −2.0 (−4.5 to 3.0) | |
| LDL cholesterol V0 | 99 (78-118) | 92 (77-115) | 100 (77-117) | |
| LDL cholesterol V2 | 94 (74-118) | 92 (77-115) | 95 (72-123) | |
| Change in LDL cholesterol | ||||
| LDL cholesterol V2-V0 | −1 (−10 to 12) | 1 (−11 to 10) | 0 (−14 to 13) | .03 |
| Triglycerides (mg/dL) V0 | 136 (91-200) | 134 (99-188) | 142 (99-200) | .71 |
| Triglycerides V2 | 140 (98-197) | 135 (102-187) | 148 (111-221) | .92 |
| Change in triglycerides | ||||
| Triglycerides V2-V0 | 9 (−36 to 33) | 0.5 (−30 to 32) | 8.5 (−19 to 43) | |
| 8-Isoprostanes (pg/mL) V0 | 379 (211-596) | 436 (265-673) | 427 (255-749) | .12 |
| 8-Isoprostanes V2 | 534 (333-950) | 613 (339-1091) | 672 (378-1097) | .43 |
| Change in isoprostanes | ||||
| 8-Isoprostanes V2-V0 | 185 (−49 to 508) | 217 (−128 to 483) | 193 (−34 to 577) | .85 |
| Cu/Zn SOD nmol/L V0 | 4.70 (3.91-5.70) | 4.87 (4.15-5.62) | 4.77 (3.96-6.61) | .50 |
| Cu/Zn SOD V2 | 4.53 (3.84-5.11) | 4.66 (3.96-5.29) | 4.415 (3.560-5.350) | .27 |
| Change in Cu/Zn SOD | ||||
| Cu/Zn SOD V2-V0 | −0.26 (−0.94 to 0.52) | −0.11 (−1.00 to 0.59) | −0.245 (−1.020 to 0.490) | .89 |
Linear regression analysis showed an association of sodium selenite with GPx-1 activity (Table V). The change in GPx-1 activity was further influenced by age (estimate 0.097) and body mass index (estimate 0.257). None of the other classic cardiovascular risk factors or left ventricular ejection fraction was relevantly related to changes in GPx-1 activity.
Table V. Linear regression for change in GPx-1 activity, visit2-visit0 (n = 410)⁎
| Variable | Coefficient | 95% CI | P | |
|---|---|---|---|---|
| Lower | Upper | |||
| Change in selenium V2-V0 | 0.098 | 0.067 | 0.129 | <.0005 |
| Age | 0.097 | 0.007 | 0.187 | .035 |
| Gender (female) | 0.358 | −1.488 | 2.204 | .704 |
| Smoking | .785 | |||
| Formerly | 0.298 | −1.307 | 1.193 | .716 |
| Currently | 0.879 | −1.626 | 3.384 | .492 |
| Diabetes | −1.292 | −3.057 | 0.473 | .151 |
| Hypertension | 0.231 | −1.740 | 2.202 | .818 |
| Body mass index | 0.257 | 0.067 | 0.446 | .008 |
| LV ejection fraction | −0.048 | −0.129 | 0.034 | .251 |
⁎Complete parameters were available in 410 individuals. |
There was no relevant heterogeneity of GPx-1 increase among subgroups defined according to age, classic risk factors, or left ventricular ejection fraction. Specifically, there was no relevant difference in the magnitude of the GPx-1 increase among patients with low and high selenium and low and high GPx-1 activities, respectively, although the increase in GPx-1 activity was greater in individuals with lower circulating selenium concentrations or GPx-1 activity below the median. There was no effect modification on the GPx-1 increase after 12 weeks with respect to treatment with statins, β-blockers, and angiotensin receptor antagonists (data not shown).
Discussion
Based on in vitro cell culture and in vivo clinical trial data, we have demonstrated that sodium selenite supplementation leads to an increase in GPx-1 activity. Although the administration of 500 μg showed a steeper increase in GPx-1 activity during the first 6 weeks and higher blood selenium concentration than the 200 μg dose, both doses led to a plateau of GPx-1 activity during long-term administration with no relevant difference in activity at the end of the study. Selenium supplementation did not influence endothelial function relevantly.
In HCAEC, we were able to induce GPx-1 protein expression and activity with sodium selenite and Se-methyl-selenocysteine hydrochloride. In contrast to these results, seleno-l-methionine treatment did not have any effect on GPx-1 expression or activity, likely owing to the nearby substitution of seleno-l-methionine for methionine in the translation of non-selenoproteins, thereby limiting availability of seleno-l-methionine for selenoprotein synthesis. Consistent with its role in translation, GPx-1 messenger RNA expression was not affected by sodium selenite treatment.
Selenium availability is the limiting factor in the biosynthesis of selenoproteins such as GPx-1. In states of selenium deficiency, the production of selenoproteins is reduced in a hierarchical order so that biologically more important proteins are expressed, whereas less frequently used enzymes are not expressed. By contrast, abundance in dietary selenium leads to increased induction of selenoprotein expression. The increase of GPx-1 by higher selenium availability was confirmed in the current clinical study.
Of importance, a plateau in GPx-1 activity was observed with both concentrations after the 12-week period. Although a dose-dependent increase in GPx-1 activity with lower concentrations of supplemental selenium has been described previously in individuals with selenium deficiency,22 we now demonstrate a ceiling effect with supplementation. This observation strengthens the postulate that GPx-1 levels could be used to adjust selenium supplementation in populations with relative selenium deficiency.12, 23
Despite considerable changes in selenium concentration and GPx-1 activity, no relevant improvement in noninvasively measured endothelial or vascular function was observed. Overall, a trend toward improved function was found even in the placebo group, which may be observed in patients taking part in a controlled trial with regular follow-up visits. With the present results, we cannot determine whether brachial artery function is adequate to capture the impact of selenium concentration and GPx-1 activity on vascular function, or whether the effect of higher GPx-1 on vascular performance is not sufficient to influence vascular function in the presence of a variety of risk factors that adversely affect vascular function itself.
Factors associated with the magnitude of change in GPx-1 activity were age and body mass index. An effect of age on body selenium status, although minor compared to nutritional factors, has previously been reported.24, 25 An association of selenium concentration and body mass index has also been observed, most obviously in overweight and obese subjects where a relative deficiency in micronutrients such as selenium has been observed. This observation has been attributed to inadequate intake as well as alterations in selenium metabolism.23, 26 Published data suggest that the impact of these factors on the relative increase of selenium concentration and GPx-1 activity may be due to a relatively lower baseline selenium concentration with older age and higher body mass index. A reduced effect of selenium supplementation on GPx-1 activity in participants treated with angiotensin-converting enzyme (ACE) inhibitors could be explained by findings that ACE inhibitors per se exhibit antioxidant characteristics and enhance glutathione-dependent antioxidant defense.27 Thus, the additional effect achieved by selenium supplementation might be lower than in patients not taking ACE inhibitors.
Owing to its role in antioxidant defense, selenium has long been the focus of observational studies and interventional trials in cardiovascular disease.28, 29 The overall effect of selenium concentration in long-term observational studies on cardiovascular outcome revealed a slight reduction in cardiovascular risk,30 whereas clinical trials of selenium supplementation have yielded inconsistent results.19, 31, 32, 33, 34, 35 Most trials used a combination of vitamins and trace elements and did not focus on GPx-1 as the major target variable. In patients with CAD, plasma selenium and GPx concentrations have been shown to be low after myocardial infarction, demonstrating that antioxidant defenses are challenged under such conditions.36
In that atherosclerosis is an inflammatory disease, a decrease in plasma selenium in subjects with atherosclerosis has been explained by it being an acute-phase reactant.37 If so, supplementation should overcome this deficiency. Our data suggest that the need for and the efficacy of selenium supplementation could be reflected in GPx-1 activity. Because it seems feasible to enhance GPx-1 activity with selenium supplementation, long-term intervention trials will be necessary to determine whether a benefit in cardiovascular outcome can be achieved.
Limitations
With the present data, we show an elevation of erythrocyte GPx-1 with sodium selenite supplementation in comparison to placebo, but we cannot prove that this higher activity contributes to a reduced rate of cardiovascular events and mortality. One limitation of the in vivo studies is that GPx-1 is measured in red blood cells rather than in vascular tissue. Our cell culture data indicate selenium supplementation can augment GPx-1 activity in endothelial cells; however, an in vivo marker for this effect is lacking. Whether treatment duration may have been inadequate to affect endothelial function remains to be determined.
One can argue that the supplementation with inorganic selenium is inferior to organified selenium (selenomethione or methyl-selenocysteine) because inorganic selenium increases oxidant stress during its organification.38 This effect may offset the benefits on endothelial cells. Blood selenium concentrations can be more effectively increased by the same molar concentration of selenomethionine, the major dietary form, than by sodium selenite.39, 40 There is, however, evidence that selenomethionine does not provide the full amount of selenium for selenoprotein synthesis because much of it is shunted into normal protein synthesis routinely substituting for methionine. In support of this prior observation, we showed that, in vitro, seleno-l-methionine had no effect on GPx-1 activity in HCAEC; Se-methyl-selenocysteine hydrochloride, by contrast, was effective at increasing activity and expression of GPx-1.
In summary, our in vitro and in vivo findings consistently demonstrate that sodium selenite supplementation induces an increase in the activity of GPx-1, a selenoprotein prominent in antioxidant defense and cardiovascular protection. These data should spur further investigations to elucidate whether enhanced GPx-1 activity achieved by selenium supplementation has a protective role in cardiovascular disease during long-term follow-up.
Acknowledgments
We are indebted to Beatrix C. Mattes for her critical review of the article.
Appendix A
RNA Isolation
RNA was extracted with the RNeasy kit (Qiagen Sciences), incorporating an optional DNase I step to remove residual DNA. Samples were quantitated and checked for purity and quality by A260/A280 measurements and gel electrophoresis. Complementary DNA was synthesized from 0.4 μg of each total RNA sample with oligo(dT) primers using the Advantage RT-for-PCR Kit (Clontech).
Quantitative real-time RT-PCR
All procedures, including data analysis, were performed on the Applied Biosystems PRISM 7900 HT Sequence Detector (Applied Biosystems 7900HT Fast Real-Time PCR Systems, Foster City, CA) using the TaqMan Gene Expression Assay (Applied Biosystems) for the human GPx-1 transcript (catalogue no. HS00829989_gH). Polymerase chain reaction products were analyzed by a method that compared the amount of glyceraldehyde 3-phosphate dehydrogenase, as an endogenous control. Cycle parameters were 95°C for 15 minutes to activate Taq, followed by 40 cycles of 95°C for 15 seconds, 58°C for 1 minute, and 72°C for 1 minute.
Western blot
For determination of GPx-1 protein levels, cells were washed in phosphate-buffered saline (1%), scraped from the plates, and pelleted at 300 × g for 5 minutes. Cell pellets were resuspended in ice-cold buffer (50 mmol/L Tris-HCl, pH 7.5, 5 mmol/L ethylenediaminetetraacetic acid [EDTA], and 1 mmol/L dithiothreitol), lysed by the freeze-thaw method,41 and debris removed by centrifugation at 8000 × g for 20 min. Samples were stored at −80°C. Protein samples (40 μg) were electrophoresed on 4% to 15% SDS-polyacrylamide gels (BioRad Laboratories, Hercules, CA), and transferred to nitrocellulose membranes (Hybond, Amersham Biosciences, Pittsburg, PA). To determine GPx-1 protein expression, the membranes were probed with monoclonal anti–GPx-1 antibody (Marine Biological Laboratory, Woburn, MA). The membranes were then stripped and reprobed with a polyclonal rabbit anti–β-actin antibody (Sigma). A Versadoc system and the accompanying software (BioRad) were used to quantitate protein bands.
Glutathione peroxidase enzyme activity from cultured endothelial cells
Cellular GPx activity was determined by an indirect assay that links GPx-1–mediated oxidation of glutathione to the recycled reduction of oxidized glutathione to glutathione by glutathione reductase using nicotinamide adenine dinucleotide phosphate hydrogenase (NADPH) as a reductant.42 The reaction was carried out in buffer containing 50 mmol/L Tris·HCl, 5 mmol/L EDTA, 1 mmol/L glutathione, 0.4 U/mL glutathione reductase, and 0.2 mmol/L NADPH (pH 7.6), and initiated by the addition of tert-butyl-hydroperoxide (0.22 mmol/L final concentration). Enzyme activity was calculated from the change of absorbance at 340 nm over time, a measure of NADPH oxidation, using a molar extinction coefficient for NADPH of 6.220 mol L−1 cm−1. Enzyme activity was normalized to protein concentration (mU/mg).
Clinical trial design
Laboratory methodsFor blinded determination of selenium at baseline, after 6 weeks, and after 12 weeks, measurements were made in whole blood anticoagulated with EDTA in a central laboratory (Biosyn). Additional selenium measurements were performed in cases of selenium levels higher than 170 μg/L. Treatment would have been stopped and patients would have been withdrawn from the study at selenium concentrations higher than 230 μg/L, but no patients achieved this predefined level. The reference values for whole blood selenium in the normal population are 100 to 140 μg/L. Glutathione peroxidase 1 activity was measured from washed red blood cells obtained immediately from whole blood anticoagulated with EDTA. Hemolyzed cells were stored frozen for up to 1 week, which did not alter enzyme activity. Glutathione peroxidase 1 was measured as described previously43 with minor modifications (Randox). Intra- and interassay coefficients of variation were 6.7 and 9.9%, respectively. Serum creatinine, aspartate aminotransferase, alanine aminotransferase, γ-glutamyl transferase, and CRP, as well as lipids, were measured immediately using routine methods. Low-density lipoprotein was calculated by the Friedewald formula. Fibrinogen was determined as derived fibrinogen.
Endothelial functionNoninvasive endothelial function testing was performed at enrollment and after 12 ± 1 weeks as described previously.44 Briefly, endothelial function was measured as FMD and after administration of 8 mg sublingual nitroglycerin according to standardized procedures after a 12-hour fast and after a 20-minute rest period.45 Two-dimensional high-resolution ultrasonic imaging of the right brachial artery was performed on a Philips HDI 5000 CV ultrasound machine (Philips, Best, Netherlands) with a linear array broadband probe, L12–5 (38 mm). All ultrasound scans were saved digitally and subsequently analyzed on an off-line reading station using the commercially available brachial tools software (Medical Imaging Applications LLC, Iowa City, IA). Nitroglycerin-mediated dilation was not determined in patients with contraindications to nitroglycerin intake. Echocardiographic studies were performed on a Philips HDI 5000 (Philips) machine and comprised parasternal and apical routine M-mode and 2-dimensional image acquisition. Left ventricular ejection fraction was calculated by apical planimetry.
Image acquisition and analysis were performed blinded for selenium and GPx-1 concentrations. To eliminate interobserver variability, both enrolled and close out endothelial function tests of any given patient were performed by the same technician.
Statistical methodsPredefined analysis variables were change in GPx-1 activity (primary outcome) and absolute change of FMD (measured in percentage) from baseline to last visit (secondary outcome). Further analysis variables were the change in CRP, fibrinogen, and other biomarkers. In addition, GPx-1, selenium, and CRP were measured at the control visit (6 weeks), and for those variables, differences between value at control visit and the value at the baseline visit were calculated as the value at the last visit minus the value at the control visit. In addition, age, GPx-1, and selenium at baseline were dichotomized as below or above the median. The changes in FMD and GPx-1 were reported as median and quartiles in each category of dichotomized age, GPx-1, and selenium, as well as smoking habits.
There were 3 hierarchically ordered null hypotheses. Null hypothesis 1 was that there is no difference in the change of FMD between the group receiving 500 μg sodium selenite per day and the group receiving placebo. Null hypotheses 2 and 3 were formulated analogously for the comparison of the group receiving 200 μg sodium selenite with the placebo group and the comparison of both groups receiving supplemental selenium. The hypotheses were tested on a significance level of 5% (2-sided) with a Wilcoxon test in a confirmatory manner. The following null hypothesis was tested in a confirmatory manner only if a null hypothesis could be rejected; otherwise, it was tested in an exploratory manner. For analysis of cell culture on GPx-1 activity and mRNA levels, the effect of different selenium compounds on GPx-1 was tested by analysis of variance with pairwise post hoc analysis by Fisher's protected least-square difference test. Statistical analyses were performed with SAS 9.1.
References
- . Glutathione peroxidase. Basic Life Sci. 1988;49:663–668
- . Glutathione and its role in cellular functions. Free Radic Biol Med. 1999;27:916–921
- Overexpression of cellular glutathione peroxidase rescues homocyst(e)ine-induced endothelial dysfunction. Proc Natl Acad Sci U S A. 2001;98:12503–12508
- Heterozygous cellular glutathione peroxidase deficiency in the mouse: abnormalities in vascular and cardiac function and structure. Circulation. 2002;106:1154–1158
- Mice with a homozygous null mutation for the most abundant glutathione peroxidase, Gpx1, show increased susceptibility to the oxidative stress-inducing agents paraquat and hydrogen peroxide. J Biol Chem. 1998;273:22528–22536
- Knockout of cellular glutathione peroxidase gene renders mice susceptible to diquat-induced oxidative stress. Free Radic Biol Med. 1999;27:605–611
- Cellular glutathione peroxidase deficiency and endothelial dysfunction. Am J Physiol Heart Circ Physiol. 2002;282:H1255–H1261
- Alterations of expression and distribution of the Ca(2+)-storing proteins in endo/sarcoplasmic reticulum during differentiation of rat cardiomyocytes. J Mol Cell Cardiol. 1996;28:553–562
- Glutathione peroxidase 1 activity and cardiovascular events in patients with coronary artery disease. N Engl J Med. 2003;349:1605–1613
- Glutathione peroxidase-1 and homocysteine for cardiovascular risk prediction: results from the AtheroGene study. J Am Coll Cardiol. 2005;45:1631–1637
- . Reference values for serum selenium in various areas-evaluated according to the TRACY protocol. J Trace Elem Med Biol. 1996;10:77–87
- . The importance of selenium to human health. Lancet. 2000;356:233–241
- Association between cardiovascular death and myocardial infarction and serum selenium in a matched-pair longitudinal study. Lancet. 1982;2:175–179
- Serum fatty acids, apolipoproteins, selenium and vitamin antioxidants and the risk of death from coronary artery disease. Am J Cardiol. 1985;56:226–231
- Serum selenium and the risk of coronary heart disease and stroke. Am J Epidemiol. 1985;122:276–282
- Association between toenail selenium and risk of acute myocardial infarction in European men. The EURAMIC Study. European Antioxidant Myocardial Infarction and Breast Cancer. Am J Epidemiol. 1997;145:373–379
- Serum and urine selenium concentrations in patients with cardiovascular diseases and relationship to other nutritional indexes. Ann Nutr Metab. 1999;43:30–36
- Decreased selenium levels in acute myocardial infarction. Jama. 1989;261:1161–1164
- Effects of selenium supplementation on cardiovascular disease incidence and mortality: secondary analyses in a randomized clinical trial. Am J Epidemiol. 2006;163:694–699
- . Assays of glutathione peroxidase. Methods Enzymol. 1984;105:114–121
- Effect of tirofiban on percutaneous coronary intervention–induced endothelial dysfunction in patients with stable coronary artery disease. Am J Cardiol. 2005;95:20–23
- An estimation of selenium requirements for New Zealanders. Am J Clin Nutr. 1999;70:896–903
- Serum selenium determinants in French adults: the SU.VI.M.AX study. Br J Nutr. 2006;95:313–320
- Serum concentrations of beta-carotene, vitamins C and E, zinc and selenium are influenced by sex, age, diet, smoking status, alcohol consumption and corpulence in a general French adult population. Eur J Clin Nutr. 2005;59:1181–1190
- Serum selenium levels in the US population: Third National Health and Nutrition Examination Survey, 1988-1994. Biol Trace Elem Res. 2003;91:1–10
- Associations between body mass index and the prevalence of low micronutrient levels among US adults. MedGenMed. 2006;8:59
- Enalapril and captopril enhance glutathione-dependent antioxidant defenses in mouse tissues. Am J Physiol Regul Integr Comp Physiol. 2000;278:R572–R577
- . Selenium as a risk factor for cardiovascular diseases. J Cardiovasc Risk. 1996;3:42–47
- . Selenium-dependent enzymes in endothelial cell function. Antioxid Redox Signal. 2003;5:205–215
- Selenium and coronary heart disease: a meta-analysis. Am J Clin Nutr. 2006;84:762–773
- The SU.VI.MAX Study: a randomized, placebo-controlled trial of the health effects of antioxidant vitamins and minerals. Arch Intern Med. 2004;164:2335–2342
- An intervention trial to inhibit the progression of precancerous gastric lesions: compliance, serum micronutrients and S-allyl cysteine levels, and toxicity. Eur J Cancer Prev. 2001;10:257–263
- Simvastatin and niacin, antioxidant vitamins, or the combination for the prevention of coronary disease. N Engl J Med. 2001;345:1583–1592
- . Coenzyme Q10 and antioxidants in acute myocardial infarction. Mol Aspects Med. 1994;15(Suppl):S143–S147
- Effect of selenium supplementation after acute myocardial infarction. Res Commun Chem Pathol Pharmacol. 1989;65:249–252
- Selenium levels and glutathione peroxidase activities in patients with acute myocardial infarction. Acta Cardiol. 1999;54:271–276
- Changes in the concentrations of plasma selenium and selenoproteins after minor elective surgery: further evidence for a negative acute phase response?. Clin Chem. 1998;44(8 Pt 1):1764–1766
- . Methioninase and selenomethionine but not Se-methylselenocysteine generate methylselenol and superoxide in an in vitro chemiluminescent assay: implications for the nutritional carcinostatic activity of selenoamino acids. Biochem Pharmacol. 2004;67:547–554
- . On supplementing the selenium intake of New Zealanders. 1. Short experiments with large doses of selenite or selenomethionine. Br J Nutr. 1978;39:579–587
- Effect of prolonged supplementation with daily supplements of selenomethionine and sodium selenite on glutathione peroxidase activity in blood of New Zealand residents. Am J Clin Nutr. 1982;36:24–31
- . Effect of multiple freeze-thaw cycles of cytoplasm samples on the activity of antioxidant enzymes. J Pharmacol Toxicol Methods. 2005;52:302–305
- . Assays of glutathione peroxidase. Methods Enzymol. 1984;105:114–121
- . Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J Lab Clin Med. 1967;70:158–169
- Effect of tirofiban on percutaneous coronary intervention–induced endothelial dysfunction in patients with stable coronary artery disease. Am J Cardiol. 2005;95:20–23
- Guidelines for the ultrasound assessment of endothelial-dependent flow-mediated vasodilation of the brachial artery: a report of the International Brachial Artery Reactivity Task Force. J Am Coll Cardiol. 2002;39:257–265
The study was funded by an unrestricted grant from Biosyn Arzneimittel GmbH, Fellbach, Germany; by National Institutes of Health (NIH) (Bethesda, MD) grants HL 61795, HV 28178, and HL 81587 from the National Heart, Lung, and Blood Institute (J.L.) (Bethesda, MD); and by grants from the Deutsche Forschungsgemeinschaft (Bonn, Germany) LU 1452/1-1 (E.L.) and SCHN 1149/1-1 (R.S.).
PII: S0002-8703(08)00789-8
doi:10.1016/j.ahj.2008.09.004
© 2008 Mosby, Inc. All rights reserved.
