Serum Vitamin E and Oxidative Protein Modification in Hemodialysis: A Randomized Clinical Trial

Serum Vitamin E and Oxidative Protein Modification in Hemodialysis: A Randomized Clinical Trial Liang Lu, PhD,1 Penny Erhard, BS,2 Robert G. Salomon, PhD,1 and Miriam F. Weiss, MD3 Background: Patients with end-stage renal disease have increased circulating concentrations of oxidatively modified circulating proteins. Therefore, we examined the ability of vitamin E ␣ (␣tocopherol) to alter levels of these modified proteins. Study Design: Randomized clinical trial. Setting & Participants: 27 clinically stable patients treated by means of hemodialysis in 4 freestanding outpatient dialysis units. Intervention: Oral administration of 800 IU of vitamin E ␣ or placebo daily. Outcomes & Measurements: Plasma levels of ␣- and ␥-tocopherol and oxidative protein modifications reflecting 2 pathways for protein-oxidant damage. The advanced glycation end product pentosidine reflects glycoxidation. The lipid peroxidation products iso[4]-levuglandin E2, (E)-4-hydroxy-2nonenal, and (E)-4-oxo-2-nonenal are formed through covalent adduction. Results: Circulating levels of all oxidative protein modifications were increased in patients with end-stage renal disease. Supplementation with ␣-tocopherol caused ␣-tocopherol levels to rise (13.2 ⫾ 3.7 to 27.3 ⫾ 14 ␮g/mL), but ␥-tocopherol levels to decrease (4.1 ⫾ 1.6 to 3.5 ⫾ 1.1 ␮g/mL). Control values were unchanged. There was no effect on oxidative protein modifications (placebo versus treatment; mean for pentosidine, 15.6 ⫾ 11.4 (SD): 95% confidence interval (CI), 8.2 to 23.1 versus 21.3 ⫾ 9.0 pg/mg protein; 95% CI, 16.1 to 26.6; iso[4]-levuglandin E2, 8.31 ⫾ 2.55; 95% CI, 6.77 to 9.85 versus 8.46 ⫾ 2.37 nmol/mL; 95% CI, 7.09 to 9.84; (E)-4-hydroxy-2-nonenal, 0.51 ⫾ 0.11; 95% CI, 0.45 to 0.57 versus 0.51 ⫾ 0.08 nmol/mL; 95% CI, 0.46 to 0.56; (E)-4-oxo-2-nonenal, 189 ⫾ 44; 95% CI, 162 to 215 vs 227 ⫾ 72 pmol/mL; 95% CI, 183 to 271). Limitations: Sample size was adequate to show changes in ␣- and ␥-tocopherol levels in response to treatment. However, power was insufficient to show an effect on oxidative protein modifications. Conclusions: Intervention of oral supplementation with ␣-tocopherol did not result in changes in circulating oxidative protein modifications. A larger study may be required to show an effect in this clinical setting. Am J Kidney Dis 50:305-313. © 2007 by the National Kidney Foundation, Inc. INDEX WORDS: Vitamin E; ␣-tocopherol; ␥-tocopherol; pentosidine; isolevuglandin; hydroxynonenal; oxononenal; hemodialysis; end-stage renal disease.

I

n the early 1990s, epidemiological studies established an inverse relationship between supplemental vitamin E use and cardiovascular disease.1 By 2000, an estimated 23 million Americans used supplemental vitamin E. However, large prospective randomized controlled trials of treatment failed to show improved cardiovascular or other health outcomes.2,3 Thus, proponents of antioxidant therapy argued that greater benefit might be found in the setting of abnormal levels of oxidative stress, such as in patients with end-stage renal disease (ESRD) treated with dialysis.4 Excess oxidant production and decreased antioxidant defense5 have been recognized characteristics of the uremic state. In patients with ESRD, oral vitamin E supplementation decreased the susceptibility of low-density lipoprotein (LDL) to oxidation6 and prevented the oxidative stress associated with intravenous iron administration for treatment of patients with anemia.7 The Secondary Prevention with Antioxidants of Cardio-

vascular disease in End-stage renal disease trial (SPACE) showed a significant decrease in cardiovascular events in 196 dialysis patients with preexisting cardiovascular disease administered oral vitamin E supplements.8 These results emphasized the importance of understanding the basic mechanisms of action of antioxidants in the uremic milieu.

From the Departments of 1Chemistry, 2Medicine, and Pathology, Case Western Reserve University, Cleveland, OH. Received December 4, 2006. Accepted in revised form May 10, 2007. Originally published online as doi: 10.1053/j.ajkd.2007.05.001 on June 28, 2007. Address correspondence to Miriam F. Weiss, MD, Renal Replacement LLC, 5096 Dogwood Trail, Lyndhurst, OH 44124. E-mail: [email protected] © 2007 by the National Kidney Foundation, Inc. 0272-6386/07/5002-0016$32.00/0 doi:10.1053/j.ajkd.2007.05.001 3

American Journal of Kidney Diseases, Vol 50, No 2 (August), 2007: pp 305-313

305

306

Lu et al

Clinical observation that vitamin E was protective against morbid events in some,8,9 but not in all, settings was named the antioxidant paradox.4 Factors that lead to oxidative stress in living systems are complex. For example, the kinetics of reaction between vitamin E and oxidants is slower than the rate at which free radical species inactivate nitric oxide.10 Therefore, vitamin E would not be expected to correct abnormalities in flow-mediated vasodilation11 despite effectively decreasing LDL oxidation.6 In addition, vitamin E was shown to act as a pro-oxidant, particularly in the absence of vitamin C.12 Patients with ESRD have increased quantities of oxidatively modified circulating and tissue proteins, including advanced glycation end products (reflecting both oxidative and carbonyl stress),13 advanced oxidative protein products, oxidized LDL, (E)-4-hydroxy-2-nonenal (hydroxynonenol, HNE)–derived-2-pentylpyrroles, and isolevuglandin adducts,14,15 as well as oxidized lipids, including isoprostanes16,17 and oxysterols.17 The goal of this project was to study the effect of supplemental vitamin E on levels of oxidative protein modifications. We measured pentosidine, a glycoxidation-derived protein modification, as well as protein modifications derived from the lipid peroxidation products iso[4]-levuglandin E2 (iso[4]-LGE2), HNE, and (E)-4-oxo-2-nonenal (ONE) in a highly oxidative clinical state—ESRD treated by means of hemodialysis. We explored relationships between levels of these products at baseline and during the course of 6 months of treatment to develop hypotheses about the relationship between levels of oxidation products and factors known to influence oxidation in the clinical setting. METHODS Patients At the time this study was initiated, 286 patients with ESRD were treated by means of hemodialysis at Centers for Dialysis Care Cleveland, Ohio, under the care of nephrologists at University Hospitals of Cleveland. Charts were reviewed from 168 of these patients recommended by the primary dialysis nurse to be clinically stable. Inclusion criteria included uninfected patients with an upper-extremity polytetrafluoroethylene graft or arteriovenous fistula. Exclusion criteria were current use of vitamin E or anticoagulant therapy, lower-extremity vascular access, central venous catheter access, central venous stenosis or clotting disorders,

and use of anticoagulants other than heparin during hemodialysis. One hundred thirty-four patients were considered ineligible or declined to participate. Informed consent was obtained in compliance with the Institutional Review Board of University Hospitals of Cleveland from 34 patients. Patients were recruited between June 2002 and September 2003, with study closeout in March 2004. During that time, all patients received treatment using polysulfone high-flux dialyzers (F80; Fresenius, Walnut Creek, CA, or APS1050; Asahi, Tokyo, Japan). Patients left the study because of death, transplantation, or initiation of anticoagulant therapy.

Study Design Patients were assigned to treatment based on a randomnumber-generation protocol performed by the hospital pharmacy at the time of dispensing treatment or placebo. Participants were enrolled by the study investigators and assigned to treatment or placebo by the dispensing hospital pharmacy. Investigators and patients were blinded to treatment with vitamin E (␣-tocopherol, 800 IU) or identical placebo capsules to be administered daily until the end of the study. “Natural vitamin E” (RRR-␣-tocopherol, also known as d-␣-tocopherol) was the kind gift of Cognis Inc (LaGrange, IL). Clinical data were collected by means of interview and chart review. Clinical parameters and routine clinical chemistry and hematology testing were performed in a single reference clinical laboratory. Results were recorded by using chart review every 3 months throughout the study. Total administered doses of erythropoietin and intravenous iron were tallied at each 3-month period, as were significant clinical events, including hospitalizations and surgical interventions.

Assays At each time point, blood was collected in EDTA-metal– free tubes for plasma or plain glass tubes for serum. Samples for plasma were placed on ice and spun in the cold at 3,000 rpm for 15 minutes. Serum was allowed to clot before centrifugation. Plasma for analysis of ␣- and ␥-tocopherol was protected from light at each step. Plasma for enzyme-linked immunosorbent assay was stabilized before storage by the addition of butylated hydroxytoluene (10 ␮mol/L final concentration) and protease inhibitors, as described.14 After processing, serum and plasma aliquots were quench-frozen in liquid nitrogen and transported to the laboratory, where they were layered with argon and stored at ⫺80°C. Tocopherols were measured using a modification of the method of Sommerburg et al.18 Briefly, plasma was extracted using ethanol and hexane, dried under nitrogen, and reconstituted in methanol before high-performance liquid chromatography with a C18 column and UV detection at 292 nm. Standards for ␣- and ␥-tocopherol were purchased from Sigma (St Louis, MO) and prepared freshly for each assay. Intra-assay coefficients of variation were 0.039 for ␣-tocopherol and 0.041 for ␥-tocopherol, and interassay coefficients were 0.078. The normal range for ␣-tocopherol was 12.50 ⫾ 0.23 ␮g/mL, and that for ␥-tocopherol was 2.41 ⫾ 0.26 ␮g/mL. Pentosidine was measured by means of high-performance liquid chromatography using a modification of the

Vitamin E and Oxidation in Hemodialysis

307

Figure 1. Study consort flow diagram. Baseline data were collected for 34 subjects randomly assigned to treatment or placebo. Twenty-seven subjects were included in the analysis of longitudinal effects up to 6 months (rectangle).

method of Odetti, as previously published.19 Total protein was assayed in a microassay modification of the Bradford method using Coomassie brilliant blue. In previous studies, the range for pentosidine in healthy control subjects was 1.35 ⫾ 0.6 pmol/mg protein; the usual range in patients on hemodialysis therapy was 22.9 ⫾ 10.8 pmol/mg protein.13 Protein adducts of iso[4]LGE2, HNE (“pentylpyrrole”), and ONE were measured by means of enzyme-linked immunosorbent assay. Iso[4]LGE2–bovine serum albumin, iso[4]LGE2–human serum albumin, ONE–bovine serum albumin, and ONE–human serum albumin were synthesized as described previously, and specific antibodies were produced, purified, and characterized.20 The normal range for iso[4]LGE2–protein adduct was 4,087 ⫾ 47.9 pmol/mL; the usual range in patients on hemodialysis therapy was 8,344 ⫾ 2,296 pmol/mL.20 Intra-assay and interassay coefficients of variation for iso[4]LGE2–protein adduct were 4.65% and 7.11%, respectively. The normal range for HNE–protein adduct was 436 ⫾ 77 pmol/mL; the usual range in patients on hemodialysis therapy was 543 ⫾ 190 pmol/mL. Intraassay and interassay coefficients of variation were 5% and 6%, respectively. ONE–bovine serum albumin and anti– ONE-ribonuclease A antibody were kind gifts of Dr Lawrence Sayre. The normal range for ONE-ribonuclease A was 114.72 ⫾ 28.73 pmol/mL; the usual range in patients on hemodialysis therapy was 152.58 ⫾ 33.68 pmol/mL. The intra-assay coefficient of variation was 3.36%, and interassay coefficient was 6.92%.

Statistical Analyses Longitudinal analyses were performed in the 13 treatment and 14 placebo patients who progressed through the 6-month follow-up. A description of clinical events was published previously.21 A comparison of randomized groups for repeated measures was performed for each outcome variable. Tables express point estimates (mean ⫾ SD) and 95% confidence intervals (CIs) for each group at each time, as well as differences between groups and 95% CIs for differences. Comparisons of tocopherols and protein oxidation modifications by means of linear regression were performed in the 34 subjects who entered the study, before randomized treatment. Statistical significance is defined as P less than 0.05. All analyses were performed using JMP 5.12 (SAS Institute, Cary, NC).

RESULTS

Baseline Characteristics

Figure 1 is a flow diagram of the study consort. Thirty-four subjects were included in baseline data analyses, and 27 subjects were studied longitudinally for at least 6 months. Demographic characteristics are listed in Table 1. Although individual participation in the study ranged from 2.1 to 20.6 months (mean followup, 9.8 ⫾ 4.7 months), only the 27 individuals

Table 1. Patients’ Baseline Clinical Characteristics

No. of patients Sex (men/women) Race (African American/white) Diabetes (no/yes) Age (y) Time on dialysis at start of study (mo) Creatinine (mg/dL) Urea (mg/dL) Standardized Kt/V (dimensionless measure of dialysis adequacy) Body mass index (kg/m2)

Placebo Group

Treatment Group

13 8/5 10/3 5/8 53.0/50.5 (31.7-72.5) 52/38 (6-213) 10.6 ⫾ 3.6 (4.1-17.2) 54 ⫾ 20 (34-100) 1.38 ⫾ 0.31 (1.05-1.92) 33.8 ⫾ 9.7 (24.2-58.1)

14 4/10 14/0 4/10 54.7/53.8 (31.7-86.1) 45/31 (5-96) 9.8 ⫾ 2.2 (6.5-13.0) 44 ⫾ 15 (5-66) 1.45 ⫾ 0.36 (0.91-2.12) 31.2 ⫾ 8.0 (21.5-47.3)

Note: Values expressed as number of patients, mean/median (range), or mean ⫾ SD (range). All differences not significant. To convert serum creatinine in mg/dL to ␮mol/L, multiply by 88.4; urea nitrogen in mg/dL to mmol/L, multiply by 0.357.

308

Lu et al Table 2. Plasma ␣- and ␥-Tocopherol Levels in Placebo and Vitamin E–Treated Patients Prob ⬎ F Baseline

3 Months

6 Months

Time

␣-Tocopherol (␮g/mL) Placebo 12.67 12.79 12.75 0.0017 SD 3.84 3.29 3.99 95% CI 10.35-14.99 10.81-14.79 10.35-15.17 Treatment 13.24 27.02 27.27 SD 3.73 17.97 14.31 95% CI 11.09-15.39 16.65-37.39 19.00-35.53 Mean difference (95% CI) ⫺0.63 (2.19-⫺3.45) ⫺13.57 (⫺4.40-⫺22.75) ⫺14.51 (⫺6.03-⫺22.99) ␥-Tocopherol (␮g/mL) Placebo 3.74 3.54 3.51 0.0002 SD 2.32 1.73 1.62 95% CI 2.34-5.16 2.50-4.59 2.52-4.49 Treatment 4.09 2.21 1.45 SD 1.56 1.45 1.05 95% CI 3.19-5.00 1.37-3.05 0.84-2.05 Mean difference (95% CI) ⫺0.24 (1.05-⫺1.53) 1.17 (2.37-⫺0.02) 2.06 (3.16-⫺0.53)

Treatment

Both

0.0019

0.0019

0.0013

0.0013

Abbreviation: CI, confidence interval.

with complete data at 0, 3, and 6 months were included. During the course of the study, 17 subjects developed thrombotic events involving their hemodialysis vascular accesses, as previously described.22 There were no differences in numbers of events between the placebo and vitamin E treatment groups or numbers of events between patients with and without diabetes. During the course of the study, 1 patient received a transplant, 1 patient transferred out of the city, 4 patients experienced documented myocardial infarctions, and 5 patients died. Clinical laboratory measurements of total iron, iron-binding protein, and ferritin were not found to influence levels of glycoxidation or lipid peroxidation products. There was no dose effect related to the quantity of exogenous intravenous iron and erythropoietin administered as part of the routine treatment of anemia in patients on dialysis therapy (data not shown). Change in Tocopherol Levels

␣ -Tocopherol levels increased 2-fold in treated patients (Table 2). Conversely, ␥-tocopherol levels decreased to half of baseline in patients treated with oral vitamin E ␣ (␣tocopherol; Table 2). At baseline, there was a

positive correlation between ␣- and ␥-tocopherol levels (Fig 2A). A positive correlation continued in the placebo group. After treatment with vitamin E ␣, the correlation between plasma ␣- and ␥-tocopherol levels became negative, with the highest ␣-tocopherol levels associated with the lowest ␥-tocopherol levels (Table 2), shown at the 3-month follow-up (Fig 2B). There were no differences between patients with and without diabetes with respect to circulating oxidative protein modification levels at baseline or during the course of the study (data not shown). There was no difference in the pattern of response to treatment with vitamin E ␣ between patients with and without diabetes (data not shown). Absence of Change in Oxidative Protein Modifications

There were no significant changes in circulating levels of glycoxidation or lipid peroxidation products during the course of the study; differences between groups remained unchanged (Table 3). The data set showed wide CIs for pentosidine and ONE–protein adduct at every time. The groups were not evenly distributed for the glycoxidation product pentosi-

Vitamin E and Oxidation in Hemodialysis

309

Figure 2. Relationship between plasma vitamin E ␣ and ␥ levels before and after treatment. (A) Baseline R ⫽ 0.46, P ⫽ 0.0055. (B) The positive correlation between vitamin E ␣ and ␥ is present in the placebo group, left regression line (R ⫽ 0.42; P ⫽ 0.05). In patients treated with vitamin E, there is a negative correlation between the 2 metabolites, right regression line (R ⫽ 0.11; P ⫽ not significant). Gray symbols, placebo; black symbols, vitamin E treatment; open circles, baseline; squares, 3-month values.

dine; thus, a trend toward increasing pentosidine levels in the vitamin E treatment group did not attain significance. Baseline Correlations Among Parameters

At baseline in the 34 subjects who consented to participate, regression relationships between lipid peroxidation–derived oxidative protein modifications and both ␣- and ␥-tocopherol levels were positive, with the greatest correlation coefficients for iso[4]LGE2–protein adducts (R ⫽ 0.42; P ⫽ 0.03). Iso[4]LGE2–protein adduct level correlated positively with both ␣- and ␥-tocopherol levels at baseline, during placebo treatment, and during treatment with ␣-tocopherol. HNE– protein adduct level correlated positively with ␣-tocopherol level at baseline (R ⫽ 0.46; P ⬍ 0.01). In addition, at baseline, correlations among lipid oxidation products were highly significant. Conversely, there was no correlation between the glycoxidation product pentosidine and any lipid peroxidation protein adduct. The regression relationship between levels of the glycoxidation product pentosidine and ␥-tocopherol was negative, although of borderline significance (R ⫽ ⫺0.30; P ⬍ 0.07).

DISCUSSION

These data showed consistently increased levels of circulating oxidative protein modifications in patients with ESRD throughout the study. Supplementation with ␣-tocopherol resulted in a 2-fold increase in ␣-tocopherol levels and a concomitant halving of ␥-tocopherol levels, but no change in levels of lipid peroxidation or glycoxidation protein modifications. At baseline in the consented cohort as a whole, there were statistically significant positive correlations between levels of lipid peroxidation products and ␣- or ␥-tocopherol and a negative correlation between pentosidine and ␥-tocopherol levels. These correlations could not be shown at later study times in the longitudinal study. In patients with ESRD, ␣-tocopherol levels are in the normal range23 whereas ␥-tocopherol levels may be normal (as in this study),24 decreased,25 or increased.23 Patients in this study were administered 800 IU/d of natural RRR-␣tocopherol, a dose and form reported to be effective in preventing secondary coronary events in patients with ESRD8 and restenosis after percutaneous transluminal coronary angioplasty. The power of this study was inadequate to show an

310

Lu et al

Table 3. Plasma Glycoxidation and Protein-Lipid Oxidation Products in Placebo and Vitamin E–Treated Patients Prob ⬎ F Baseline

Pentosidine (pmol/mg protein) Placebo SD 95% CI Treatment SD 95% CI Mean difference (95% CI) Iso[4]levuglandin E2 (nmol/mL) Placebo SD 95% CI Treatment SD 95% CI Mean difference (95% CI) Hydroxynonenal (nmol/mL) Placebo SD 95% CI Treatment SD 95% CI Mean difference (95% CI) (E)-4-oxo-2-nonenal (pmol/mL) Placebo SD 95% CI Treatment SD 95% CI Mean difference (95% CI)

15.1 11.4 8.2-22.0 19.3 7.9 14.7-23.9 ⫺3.7 (2.9-⫺10.2)

8.43 2.68 6.81-10.05 8.55 1.73 7.55-9.55 0.165 (2.04-⫺1.71)

0.46 0.13 0.37-0.53 0.49 0.10 0.44-0.55 ⫺0.02 (0.07-⫺0.1)

186 62 149-224 236 68 196-276 ⫺49 (2-⫺101)

3 Months

6 Months

15.8 11.6 8.8-22.8 20.7 9.3 15.3-26.0

15.6 12.3 8.2-23.1 21.3 9.0 16.1-26.6

⫺3.2 (1.8-⫺11.2)

8.84 2.32 7.43-10.24 8.09 2.35 6.73-9.45

0.47 0.09 0.42-0.53 0.45 0.09 0.39-0.51

182 62 145-219 204 61 169-240

0.3151

0.6902

0.6902

0.9480

0.5100

0.5100

0.0201

0.2065

0.2065

0.4182

0.7763

0.7763

0.005 (0.08-⫺0.07)

189 44 162-215 227 72 183-271

⫺22 (26-⫺71)

Both

0.153 (1.81-⫺2.11)

0.51 0.11 0.45-0.57 0.51 0.08 0.46-0.56

⫺0.03 (0.11-⫺0.04)

Treatment

⫺5.7 (2.9-⫺14.3)

8.31 2.55 6.77-9.85 8.46 2.37 7.09-9.84

0.442 (2.34-⫺1.45)

Time

⫺38 (10-⫺87)

Abbreviation: CI, confidence interval.

effect of treatment on cardiovascular outcome. However, we previously reported an association between lower baseline ␣- and ␥-tocopherol levels and subsequent thrombotic events in the hemodialysis vascular access.22 Oral supplementation with ␣-tocopherol decreased circulating and tissue ␥-tocopherol levels in healthy subjects,26 the elderly, and patients with ESRD.23 Presumably, supplementation with ␣-tocopherol competitively inhibits ␣-tocopherol transfer protein of dietary ␥-tocopherol.

␣-Tocopherol is considered the most potent nonspecific lipid-soluble antioxidant in the body.27 However, only small differences are expected in the ability of the various tocopherols and tocotrienols to neutralize free radicals by hydrogen atom transfer. For example, rate constants for hydrogen atom transfer to a peroxyl radical from ␣ and ␥-tocopherol are 23.5 and 15.9 ⫻ 105 M⫺1S⫺1, respectively.28 Therefore, all forms of vitamin E would be expected to be similarly effective as chain-breaking antioxi-

Vitamin E and Oxidation in Hemodialysis

311

Figure 3. (A) Fenton and Fenton-like reactions. (B) Reduction in transition metal ions by tocopherols.

dants for inhibiting lipid peroxidation. If the dominant activity of E vitamins is to prevent lipid peroxidation by neutralizing free radicals through hydrogen atom transfer, levels of lipid peroxidation should decrease as those of the E vitamins increase. Two observations of the present study are not in accord with this expectation: (1) there were striking positive correlations between circulating levels of lipid peroxidation products and circulating levels of both ␣- or ␥-tocopherol at baseline, and (2) the 2-fold increase in circulating ␣-tocopherol level accomplished by supplementation was not accompanied by a significant decrease in levels of lipid peroxidation products. However, a larger sample size may be required to fully test this hypothesis. At baseline, positive correlations between levels of lipid peroxidation products and those of tocopherols suggest that E vitamins were exerting pro-oxidative effects in the chronically oxidizing environment of ESRD. Tocopherols may promote lipid autoxidation by recycling catalytic amounts of transition metal ions to low-valent metals that can initiate free radical oxidative chain reactions by inducing Fenton and Fentonlike reactions, as shown in Fig 3A. In vitro, tocopheryl quinone is produced by oxidation of ␣-tocopherol with a concomitant decrease in transition metal ions (Fig 3B) that then mediate reductive cleavage of hydroperoxides to alkoxy radicals.29 Thus, tocopherols can promote the generation of alkoxy radicals from lipid hydroper-

oxides and thereby initiate free radical chain reactions that generate more lipid hydroperoxides through autoxidation of lipids.30 Another more subtle pro-oxidant effect of ␣-tocopherol supplementation may result from the consequent decrease in ␥-tocopherol levels (Fig 1). Myeloperoxidase, which represents up to 5% of cell protein in inflammatory phagocytes, functions as a major catalyst for lipid peroxidation.31 We previously showed that myeloperoxidase promoted the generation of iso[4]LGE2–protein adducts in vivo in a mouse model of chronic inflammation.32 ␣-Tocopherol was ineffective against myeloperoxidase-mediated oxidation.33 Conversely, ␥-tocopherol may have unique specificity for protection against reactive nitrogen species.34 Additional complexity arises from other biological activities of vitamin E. In addition to its antioxidant functions, vitamin E is involved in cell signaling, gene expression, immune response, and apoptosis. ␥-Tocopherol is metabolized largely to 2,7,8-trimethyl-2-(␤-carboxyethyl)-6-hydroxychroman (␥-CEHC). Both ␥-tocopherol and ␥-CEHC, but not ␣-tocopherol, inhibit cyclooxygenase activity and thus are anti-inflammatory.35 ␥-Tocopherol is more potent than ␣-tocopherol in modulating inflammatory pathways.35 Oral supplementation with ␣-tocopherol results in a decrease in circulating ␥-tocopherol levels with a resultant increase in inflammation in patients with unstable atherosclerotic lesions.36

312

In patients with ESRD, chronic inflammation is a major source of increased oxidative stress.37 Untreated hemodialysis membranes have been shown to activate complement and the immune system during dialysis. In contrast to the present study, dialysis membranes coated with ␣-tocopherol decreased the peroxidation of red blood cell and plasma lipids.38 In addition, hemodialysis with ␣-tocopherol–coated dialysis membranes normalized endothelial response and corrected the overproduction of cytokines usually associated with dialysis, normalizing immune function.21,38 Dialysis with ␣-tocopherol–coated dialyzers decreased levels of advanced glycation end products39 and free-circulating HNE.40 In the present study, circulating levels of advanced glycation end products, ie, pentosidine, were related inversely to ␥-tocopherol levels at baseline, suggesting that inflammatory pathways that are modulated normally by ␥-tocopherol influence pentosidine formation. Evidence that ␣-tocopherol supplementation can increase inflammation has been shown in hemodialysis patients administered a 14-day course of this vitamin. Interleukin 6 levels increased in response to treatment, but C-reactive protein levels decreased when patients were administered supplements rich in ␥-tocopherol.23 Our results emphasize the contrast between ␣-tocopherol localized to the hemodialysis membrane and ingested as an oral dietary supplement.36 We postulate that generation of oxidative products in reaction with ␣-tocopherol can be readily cleared by the dialysis process when ␣-tocopherol is localized to the membrane, but remain in situ in tissue and the circulation in the presence of increased systemic levels of ␣-tocopherol. An important limitation of this work is the wide CIs for oxidative protein modification values. In addition, there is evidence of imbalance between the randomized groups, particularly for the glycoxidation product pentosidine, suggesting that sample size is inadequate to show significant differences in response to treatment with vitamin E. In conclusion, in the clinical setting of oxidative stress experienced by patients with ESRD treated by means of hemodialysis, the balance of oxidant/antioxidant systems is highly complex. Both chemical pathways and inflammatory components appear to be involved in interactions that

Lu et al

lead to the formation of glycoxidation and protein-bound lipid oxidation products. The definitive delineation of antioxidant or pro-oxidant effects of vitamin E on circulating oxidative protein modifications in patients with ESRD treated by means of hemodialysis awaits further study using a larger sample size. ACKNOWLEDGEMENTS Support: This clinical trial was supported by the Leonard B. Rosenberg Renal Research Foundation of the Center for Dialysis Care, Cleveland, OH; by grant ES11461 from the National Institute of Environmental Health Sciences (Dr Weiss); by grants DK57733 and DK45619 from the National Institute of Diabetes and Digestive and Kidney Diseases (Dr Weiss); and by grant GM21249 from the National Institute of General Medical Sciences (Dr Salomon). Financial Disclosure: None.

REFERENCES 1. Rimm EB, Stampfer MJ, Ascherio A, Giovannucci E, Colditz GA, Willett WC: Vitamin E consumption and the risk of coronary heart disease in men. N Engl J Med 328:1450-1456, 1993 2. Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto miocardico: Dietary supplementation with n-3 polyunsaturated fatty acids and vitamin E after myocardial infarction: Results of the GISSI-Prevenzione Trial. Lancet 354:447-455, 1999 3. Lonn E, Bosch J, Yusuf S, et al: Effects of long-term vitamin E supplementation on cardiovascular events and cancer: A randomized controlled trial. JAMA 293:13381347, 2005 4. Halliwell B: The antioxidant paradox. Lancet 355:11791180, 2000 5. Himmelfarb J, Stenvinkel P, Ikizler TA, Hakim RM: The elephant in uremia: Oxidant stress as a unifying concept of cardiovascular disease in uremia. Kidney Int 62:15241538, 2002 6. Islam KN, O’Byrne D, Devaraj S, Palmer B, Grundy SM, Jialal I: Alpha-tocopherol supplementation decreases the oxidative susceptibility of LDL in renal failure patients on dialysis therapy. Atherosclerosis 150:217-224, 2000 7. Roob JM, Khoschsorur G, Tiran A, Horina JH, Holzer H, Winklhofer-Roob B: Vitamin E attenuates oxidative stress induced by intravenous iron in patients on hemodialysis. J Am Soc Nephrol 11:539-549, 2000 8. Boaz M, Smetana S, Weinstein T, et al: Secondary Prevention with Antioxidants of Cardiovascular disease in End stage renal disease (SPACE): Randomised placebocontrolled trial. Lancet 356:1213-1218, 2000 9. Stephens N, Parsons A, Shofield P, et al: Randomised controlled trial of vitamin E in patients with coronary disease: Cambridge Heart Antioxidant Study (CHAOS). Lancet 347:781-786, 1996 10. Landmesser U, Harrison DG, Drexler H: Oxidant stress—A major cause of reduced endothelial nitric oxide availability in cardiovascular disease. Eur J Clin Pharmacol 62:S13-S19, 2006 (suppl 13)

Vitamin E and Oxidation in Hemodialysis 11. Ghiadoni L, Cupisti A, Huang Y, et al: Endothelial dysfunction and oxidative stress in chronic renal failure. J Nephrol 17:512-519, 2004 12. Witting PK, Upston JM, Stocker R: The molecular action of alpha-tocopherol in lipoprotein lipid peroxidation. Pro- and antioxidant activity of vitamin E in complex heterogeneous lipid emulsions. Subcell Biochem 30:345390, 1998 13. Weiss M, Erhard P, Kader-Attia F, et al: Mechanisms for the formation of advanced glycation end products (AGEs) in end stage renal disease (ESRD). Kidney Int 57:25712585, 2000 14. Salomon R, Kaur K, Batyreva E: Isolevuglandinprotein adducts in oxidized low density lipoprotein and human plasma: A strong connection with cardiovascular disease. Trends Cardiovasc Med 10:53-59, 2000 15. Salomon R, Kaur K, Podrez E, Hoff H, Krushinsky A, Sayre L: HNE-derived 2-pentylpyrroles are generated during oxidation of LDL, are more prevalent in blood plasma from patients with renal disease or atherosclerosis, and are present in atherosclerotic plaques. Chem Res Toxicol 13:557564, 2000 16. Ikizler T, Moroz J, Roberts L, et al: Plasma F2isprostanes levels are elevated in chronic hemodialysis patients. Clin Nephrol 58:190-197, 2002 17. Siems W, Quast S, Peter D, et al: Oxysterols are increased in plasma of end-stage renal disease patients. Kidney Blood Press Res 28:302-306, 2005 18. Sommerburg O, Zang L-Y, van Kuijk F: Simultaneous detection of carotenoids and vitamin E in human plasma. J Chromatogr B 695:209-215, 1997 19. Friedlander M, Wu Y, Elgawish A, Monnier V: Early and advanced glycosylation end products: Kinetics of formation and clearance in peritoneal dialysis. J Clin Invest 97:728-735, 1996 20. Salomon RG, Sha W, Brame C, et al: Protein adducts of iso[4]levuglandin E2, a product of the isoprostane pathway, in oxidized low density lipoprotein. J Biol Chem 274:20271-20280, 1999 21. Girndt M, Lengler S, Kaul H, Sester U, Sester M, Kohler H: Prospective crossover trial of the influence of vitamin-E coated dialyzer membranes on T-cell activation and cytokine induction. Am J Kidney Dis 35:95-104, 2000 22. Schwing W, Erhard P, Hollamon C, et al: Thrombotic events and markers of oxidation and inflammation in hemodialysis. Hemodialysis Int 8:338-343, 2004 23. Himmelfarb J, Kane J, McMonagle E, et al: Alpha and gamma tocopherol metabolism in healthy subjects and patients with end-stage renal disease. Kidney Int 64:978991, 2003 24. Smith KS, Lee CL, Ridlington JW, Leonard SW, Devaraj S, Traber MG: Vitamin E supplementation increases circulating vitamin E metabolites tenfold in end-stage renal disease patients. Lipids 38:813-819, 2003 25. Galli F, Buoncristiani U, Conte C, Aisa C, Floridi A: Vitamin E in uremia and dialysis patients. Ann N Y Acad Sci 1031:348-351, 2004 26. Handelman GJ, Epstein WL, Peerson J, Spiegelman D, Machlin LJ, Dratz EA: Human adipose alpha-tocopherol

313 and gamma-tocopherol kinetics during and after 1 y of alpha-tocopherol supplementation. Am J Clin Nutr 59:1025-1032, 1994 27. Saldeen T, Li D, Mehta JL: Differential effects of alpha- and gamma-tocopherol on low-density lipoprotein oxidation, superoxide activity, platelet aggregation and arterial thrombogenesis. J Am Coll Cardiol 34:1208-1215, 1999 28. Burton GW, Ingold KU: Autoxidation of biological molecules. 1. The antioxidant activity of vitamin E and related chain-breaking phenolic antioxidants in vitro. J Am Chem Soc 103:6472-6477, 1981 29. Yamauchi R, Ozaki K, Shimoyamada M, Kato K: Iron-catalyzed reaction products of alpha-tocopherol with 1-palmitoyl-2-linoleoyl-3-sn-phosphatidylcholine (13S)hydroperoxide. Chem Phys Lipids 114:193-201, 2002 30. Yamauchi R, Yagi Y, Kato K: Oxidation of alphatocopherol during the peroxidation of dilinoleoylphosphatidylcholine in liposomes. Biosci Biotechnol Biochem 60:616620, 1996 31. Zhang R, Brennan ML, Shen Z, et al: Myeloperoxidase functions as a major enzymatic catalyst for initiation of lipid peroxidation at sites of inflammation. J Biol Chem 277:46116-46122, 2002 32. Poliakov E, Brennan ML, Macpherson J, et al: Isolevuglandins, a novel class of isoprostenoid derivatives, function as integrated sensors of oxidant stress and are generated by myeloperoxidase in vivo. FASEB J 17:2209-2220, 2003 33. Brennan ML, Hazen SL: Amino acid and protein oxidation in cardiovascular disease. Amino Acids 25:365374, 2003 34. Christen S, Woodall A, Shigenaga M, SouthwellKeely P, Duncan M, Ames B: G-Tocopherol traps mutagenic electrophiles such as NOx and complements a-tocopherol: Physiological implications. Proc Natl Acad Sci U S A 94:3217-3222, 1997 35. Jiang Q, Elson-Schwab I, Courtemanche C, Ames BN: Gamma-tocopherol and its major metabolite, in contrast to alpha-tocopherol, inhibit cyclooxygenase activity in macrophages and epithelial cells. Proc Natl Acad Sci U S A 97:11494-11499, 2000 36. Devaraj S, Jialal I: Failure of vitamin E in clinical trials: Is gamma-tocopherol the answer? Nutr Rev 63:290293, 2005 37. Himmelfarb J, Hakim RM: Oxidative stress in uremia. Curr Opin Nephrol Hypertens 12:593-598, 2003 38. Wanner C, Bahner U, Mattern R, Lang D, PasslickDeetjen J: Effect of dialysis flux and membrane material on dyslipidaemia and inflammation in haemodialysis patients. Nephrol Dial Transplant 19:2570-2575, 2004 39. Baragetti I, Furiani S, Vettoretti S, et al: Role of vitamin E-coated membrane in reducing advanced glycation end products in hemodialysis patients: A pilot study. Blood Purif 24:369-376, 2006 40. Odetti P, Traverso N, Monacelli F, et al: Vitamin E-coated filter decreases levels of free 4-hydroxyl-2-nonenal during haemodialysis sessions. Free Radic Res 40:207-212, 2006