Hemodialysis International 2015; 19:429–438
Serum sulfatide abnormality is associated with increased oxidative stress in hemodialysis patients Huang YUZHE,1,2* Yuji KAMIJO,1,3* Koji HASHIMOTO,3 Makoto HARADA,1,3 Taro KANNO,3 Eiko SUGIYAMA,1,4 Mamoru KYOGASHIMA,5 Tomomasa OGUCHI,6 Takero NAKAJIMA,1 Yutaka KANNO,7 Toshifumi AOYAMA1 1 Department of Metabolic Regulation, Institute of Pathogenesis and Disease Prevention, Shinshu University Graduate School of Medicine, Matsumoto, Nagano, Japan; 2Department of Human Anatomy, Hebei Medical University, Shijiazhuang, Hebei, China; 3Department of Nephrology, Shinshu University School of Medicine, Matsumoto, Nagano, Japan; 4Department of Nutritional Science, Nagano Prefectural College, Nagano, Nagano, Japan; 5Department of Microbiology and Molecular Cell Biology, Nihon Pharmaceutical University, Ina, Saitama, Japan; 6Kidney Disease and Dialysis Center, Aizawa Hospital, Matsumoto, Nagano, Japan; 7Department of Nephrology, Kanno Dialysis and Vascular Access Clinic, Matsumoto, Nagano, Japan
Abstract Sulfatides are major glycosphingolipids of lipoproteins that influence atherosclerosis and blood coagulation. Our previous cross-sectional study of hemodialysis patients showed that serum sulfatide levels decreased markedly with increasing duration of hemodialysis treatment, which may contribute to the development of cardiovascular disease. However, this past study could not demonstrate the time-dependent change in serum sulfatide levels in each patient, and the underlying mechanism is unknown. To confirm the time-dependent aggravation of serum sulfatide abnormality, 95 stable hemodialysis outpatients were followed up for 3 years. To show the underlying mechanisms, we statistically analyzed correlations between serum sulfatide levels and clinical factors, including an oxidative stress marker, malondialdehyde. Serum sulfatides were quantified by mass spectrometry after conversion to lysosulfatides. Malondialdehyde was measured using a colorimetric assay. The results showed a time-dependent decrease in serum sulfatide levels associated with increased malondialdehyde levels, although the absolute level of serum malondialdehyde does not determine the baseline level of serum sulfatides. Multiple linear regression analysis showed a significant correlation only between the time-dependent change in serum sulfatide levels and the time-dependent change in serum malondialdehyde levels. This study demonstrated, for the first time, a time-dependent aggravation of serum sulfatide abnormality in hemodialysis patients, as well as the potential relationship between serum sulfatide abnormality and increasing oxidative stress. These findings suggest that oxidative stress might be an aggravating factor in serum sulfatide abnormality. As continuation of hemodialysis treatment hardly improves abnormal serum sulfatide levels or increased oxidative stress, development of novel therapeutic strategies may be important. Key words: Cardiovascular disease, hemodialysis, malondialdehyde, oxidative stress, serum sulfatide level
Correspondence to: Y. Kamijo, MD, PhD, Department of Nephrology, Shinshu University School of Medicine, 3-1-1 Asahi, Matsumoto 390-8621, Japan. E-mail: [email protected] *These authors contributed equally to this work. Disclosure of grants or other funding: None.
© 2015 International Society for Hemodialysis DOI:10.1111/hdi.12270
429
Yuzhe et al.
INTRODUCTION Sulfatides are amphiphilic molecules that are synthesized from galactosylceramides by sulfation at C3 of the galactosyl residue.1 Sulfatides are widely distributed in mammalian tissues, such as the brain, kidney, liver, and digestive tract, and also exist in the sera of various mammals as major glycosphingolipids of lipoproteins.1–4 The physiological roles of serum sulfatides are still unclear. Our previous study found obvious accumulation of sulfatides in the sclerotic aorta of Watanabe hereditary hyperlipidemic rabbits, which provide an animal model for human familial hypercholesterolemia, suggesting that sulfatides play a significant role in the progression of atherosclerosis.2,5 Other experimental studies reported that intravenous administration of exogenous sulfatides had significant effects on blood coagulation.6,7 Most studies reported that serum sulfatides had anticoagulant and antiplatelet effects, but some reported that sulfatides accelerated blood coagulation.7 The results of these studies suggest that abnormal serum sulfatide levels may play a role in the pathogenesis of cardiovascular disease (CVD). Chronic kidney disease patients undergoing hemodialysis (HD) therapy have a high incidence of CVD, which is the major cause of death in this population.8,9 Interestingly, our recent cross-sectional study found that serum sulfatide levels were remarkably low in HD patients compared with normal subjects, and that serum sulfatide levels were significantly lower in HD patients with CVD than in HD patients without CVD.10 These findings suggest that decreased serum sulfatide levels may contribute to the development of CVD in HD patients because of abnormal sulfatide function. The mechanisms underlying the decrease in serum sulfatide levels in HD patients are currently unknown. Our previous cross-sectional clinical study found an inverse correlation between the serum sulfatide level and the duration of HD, suggesting that alteration of the internal environment caused by long-term kidney failure gradually decreases the serum sulfatide level.10 However, this past study could not demonstrate the time-dependent change in serum sulfatide levels in each patient. To confirm the time-dependent change in serum sulfatide levels in each HD patients and to elucidate the underlying mechanisms, we conducted a 3-year observational study and investigated the clinical factors correlated with serum sulfatide levels. Serum sulfatides are produced and secreted mainly by the liver as the major component of lipoproteins.2 Our previous experimental studies using an acute kidney
430
injury model and an acute hepatic injury model found that increased systemic oxidative stress reduced the hepatic expression of an essential synthetic enzyme of sulfatides, cerebroside sulfotransferase (CST), followed by a decrease in serum sulfatide levels.11,12 Furthermore, we reported that kidney transplantation normalized the serum sulfatide abnormality and oxidative stress synchronously in kidney transplant recipients.13,14 These findings suggest a close relationship between serum sulfatide abnormality and oxidative stress. Many past studies reported that HD patients are exposed to high levels of oxidative stress, caused by increased production of reactive oxygen species because of their uremia and HD procedures, as well as by decreased antioxidant activity and decreased uptake of dietary antioxidants.15,16 This unhealthy internal condition in HD patients may suppress hepatic sulfatide synthesis, leading to a decrease in serum sulfatide levels. In the current study, we investigated whether there is a relationship between serum sulfatide abnormality and oxidative stress in HD patients. Because malondialdehyde (MDA) is one of the most well-known classical oxidative stress markers and the analytical method of MDA is well established, we selected and utilized the serum MDA levels as a representative oxidative stress marker, which is common with all of our studies.
MATERIALS AND METHODS Hemodialysis patients and study protocol To examine the general characteristics in HD patients, we sought to survey all outpatients undergoing maintenance HD therapy at affiliated hospitals of Shinshu University. Among them, 201 patients agreed to participate in the study and were enrolled. From this population, unstable patients with cancer (n = 3) and/or serum positivity of chronic infection (hepatitis C virus, n = 26; hepatitis B virus, n = 12; syphilis, n = 1), 1 patient who planned to shift to the kidney transplantation unit, and 2 patients who withdrew their consent were excluded at the beginning of the observation period (baseline number = 156). Additionally, to confirm the time-dependent change in serum sulfatide levels in HD patients over 3 years, 34 patients who died during the study period and 27 patients without clinical data and/or serum samples were excluded. Finally, this study investigated 95 patients. Serum samples were collected at the beginning of the observation period (baseline) and after 18 and 36 months, and were stored in a deep freezer. All serum samples were obtained just before the first HD session of the week.
Hemodialysis International 2015; 19:429–438
Sulfatide abnormality in HD patients
Serum sulfatide levels and serum MDA levels in the frozen samples were analyzed together with many other laboratory data. The study protocol complied with the guidelines of the 2004 revision of the Declaration of Helsinki and was approved by the ethics committee of Shinshu University (approval number 750). Written informed consent was obtained from all patients before inclusion in the study.
in serum sulfatide levels were analyzed by multivariate linear regression analysis. The multivariate analysis included all clinical follow-up parameters that changed significantly during the follow-up period. All statistical analyses were performed using SPSS version 18.0J (SPSS, Inc., Chicago, IL, USA). The level of statistical significance was set at P < 0.05.
Quantification of serum sulfatide levels
RESULTS
Sulfatides were extracted from the serum samples using the hexane/isopropanol method and were converted to lysosulfatides (LSs; sulfatides without fatty acids) by saponification with sodium hydroxide.17–19 The LS samples were analyzed by matrix-assisted laser desorption ionization-time of flight mass spectrometry using a Voyager Elite XL Biospectrometry Workstation (PerSeptive Biosystems, Framingham, MA, USA) as previously described.18 Seven molecular species of LS were detected, based on differences in sphingoid base structures: LS-sphingadienine (LS-d18:2), LS-(4E)-sphinganine (LS-d18:1), LS-sphinganine (LS-d18:0), LS-4D-hydroxysphinganine (LS-t18:0), LS-(4E)-eicosasphingenine (LS-d20:1), LS-eicosasphinganine (LS-d20:0), and LS-4Dhydroxyeicosasphinganine (LS-t20:0). The total serum sulfatide level was calculated as the sum of the levels of these 7 LS molecular species. When sulfatide levels were measured for new samples, the same standard serum sample and the chemically prepared internal standard, hydrogenated N-acetyl LS (LS-d18:0 NAc), were always used in order to verify the consistency of the data.
Baseline characteristics
Analysis of serum oxidative stress marker The serum level of the oxidative stress marker MDA was measured using a lipid peroxidation colorimetric assay kit purchased from Oxis International (Beverly Hills, CA, USA).
Statistical analysis Continuous variables were handled as nonparametric variables and are shown as the median (minimum– maximum). Time-dependent changes in paired continuous variables were analyzed using the Wilcoxon matched-pairs signed-rank test. Two independent groups of continuous variables were compared using the MannWhitney U test. Correlations between clinical factors and serum sulfatide levels were determined using Spearman’s correlation coefficient. The clinical parameters that were most strongly correlated with the time-dependent changes
Hemodialysis International 2015; 19:429–438
The baseline characteristics of the HD patients included in the study are shown in Table 1. Compared with the latest overview of regular dialysis treatment in Japan by the Japanese Society for Dialysis Therapy,20 HD patients in the current study had a lower median age (63 years vs. 67 years), higher proportion of males (67% vs. 63%), and lower prevalence of diabetes mellitus (31.6% vs. 37%) at study entry (Table 1). The median duration of maintenance HD treatment was 8.4 years. This study included only HD patients who could be followed up for 3 years, and excluded unstable patients and the dead during the follow-up, and therefore may have enrolled patients with relatively good physical condition compared with the general population of Japanese HD patients. Nevertheless, 30.5% of enrolled patients had a history of CVD at study entry. These HD patients were treated with various medicines; however, no patient used any antioxidant medicine, such as alpha-tocopherol, vitamin C, and N-acetylcysteine.
Cross-sectional analysis of correlations between clinical factors and serum sulfatide levels The mean serum sulfatide level of the 95 HD patients at study entry was 3.66 ± 1.00 nmol/mL. When compared with the normal serum sulfatide level in healthy control participants (8.21 ± 1.50 nmol/mL) reported in our previous study,10 the serum sulfatide level in HD patients appeared to be less than 50%. Consistent with our previous study,10 the mean serum sulfatide level was markedly lower in HD patients with a history of CVD than in HD patients without a history of CVD (3.14 ± 0.62 nmol/mL vs. 3.90 ± 1.05 nmol/mL, P < 0.01). To show the mechanisms underlying the decrease in serum sulfatide levels in HD patients, we analyzed correlations between baseline clinical parameters and serum sulfatide levels. Spearman’s rank correlation analysis and scatter diagrams showed an
431
Yuzhe et al.
63 (30–91) 64/31 29 (30.5%) 9 (9.5%) 4 (4.2%) 1 (1.1%) 3 (3.2%) 3 (3.2%) 17 (17.9%) 2 (2.1%) 47 (49.5%) 8.4 (0.5–27.7) 85 (89.5%) 10 (10.5%) 0 (0%) 30 (31.6%) 27 (28.4%) 10 (10.5%) 28 (29.5%) 22.6 (19.6–33.8) 142 (110–178) 88 (68–102) 51 (53.7%) 75 (78.9%) 17 (17.9%) 26 (27.4%) 0 (0%) 48 (50.5%) 82 (86.3%)
Continuous variables are shown as median (minimum–maximum). CVD = cardiovascular disease; HD = hemodialysis; RAS = renin-angiotensin system.
inverse correlation between serum sulfatide levels and the duration of HD, but no significant correlations between serum sulfatide levels and other clinical parameters including serum MDA levels (Table 2; Figure 1).
Analysis of clinical factors correlated with time-dependent changes in serum sulfatide levels Follow-up of the enrolled HD patients for 3 years confirmed a time-dependent decrease in serum sulfatide levels (Figure 2A). Interestingly, this follow-up analysis
432
8 7
Sulfades (nmol/mL)
Age (y) Sex (male/female) History of CVD, n (%) Angina pectoris Myocardial infarction Aortic aneurysm Peripheral vascular disease Transient ischemic attack Cerebral infarction Cerebral hemorrhage Smoking history, n (%) Duration of HD (y) Vascular access, n (%) Arteriovenous fistula Arteriovenous graft Catheter Cause of kidney failure, n (%) Diabetes mellitus Chronic glomerulonephritis Hypertension Other Body mass index (kg/m2) Blood pressure (mmHg) Systolic Diastolic Medication, n (%) RAS inhibitor Calcium channel blocker Beta blocker Antiplatelet agent Warfarin Vitamin D Erythropoietin
A
y = -0.0718x + 4.274 R² = 0.217
6 5 4 3 2 1 0 0
5
10
15
20
25
30
2.50
3.00
HD duraon (y)
B
8 7
Sulfades (nmol/mL)
Table 1 Baseline characteristics of maintenance hemodialysis patients (N = 95)
y = 0.1011x + 3.5368 R² = 0.0013
6 5 4 3 2 1 0 0.00
0.50
1.00
1.50
2.00
MDA (μmol/L) Figure 1 Serum sulfatide levels in hemodialysis (HD) patients. (A) Scatter plot showing serum sulfatide levels according to duration of HD. (B) Scatter plot showing crosssectional analysis of serum sulfatide levels and serum malondialdehyde (MDA) levels.
also showed a marked time-dependent increase in serum MDA levels (Figure 2B). The composition of the different molecular species of serum sulfatides in the 95 HD patients did not differ at each follow-up point, suggesting that all molecular species similarly decreased in a timedependent manner (Figure 2C). Therefore, we used sum data of the 7 sulfatides as representative data of sulfatides in the statistical analyses. The time-dependent decrease in serum sulfatide levels was not affected by sex, CVD history, smoking history, diabetes mellitus, type of vascu-
Hemodialysis International 2015; 19:429–438
Sulfatide abnormality in HD patients
Table 2 Spearman’s rank correlations between serum sulfatide levels and various parameters at baseline Parameter Serum sulfatides (nmol/mL) HD duration MDA (μmol/L) Age Body mass index (kg/m2) Systolic blood pressure Diastolic blood pressure Total protein (g/dL) Albumin (g/dL) Urea nitrogen (mg/dL) Creatinine (mg/dL) Sodium (mEq/L) Potassium (mEq/L) Chloride (mEq/L) Calcium (mg/dL) Inorganic phosphorus (mg/dL) Uric acid (mg/dL) Total bilirubin (mg/dL)
Spearman’s correlation
Parameter
Spearman’s correlation
1 −0.453*** 0.041 0.104 0.092 0.023 0.003 −0.040 0.047 0.016 −0.122 0.042 0.002 0.117 −0.194 0.032 −0.034 0.183
Aspartate amino transferase (IU/L) Alanine transaminase (IU/L) Alkaline phosphatase (IU/L) Lactate dehydrogenase (IU/L) γ-Glutamyl transpeptidase (IU/L) Amylase (IU/L) Total cholesterol (mg/dL) Low-density lipoprotein (mg/dL) High-density lipoprotein (mg/dL) Triglyceride (mg/dL) White blood cell count (102/μL) Hemoglobin (g/dL) Platelet count (104/μL) Glucose (mg/dL) C-reactive protein (mg/dL) Ferritin (ng/mL) Transferrin saturation (%) Intact PTH (pg/mL)
−0.097 −0.023 0.060 0.028 −0.160 −0.027 −0.021 −0.058 −0.104 0.069 0.009 0.113 −0.027 −0.051 −0.180 0.078 0.194 0.106
HD = hemodialysis; MDA = malondialdehyde; PTH = parathyroid hormone. ***P < 0.001.
lar access, dialysis method (conventional HD or online hemodialysis filtration), or various oral medications (Figure 3). To identify other factors potentially correlated with time-dependent changes in serum sulfatide levels, we examined the changes in serum levels of various parameters during the follow-up period (Table 3). Linear regression analysis was performed using all observational point data of candidate parameters including serum MDA levels, and found that the MDA level was the only factor significantly correlated with the time-dependent change in serum sulfatide level (Table 4). The inverse correlation between serum sulfatide level and serum MDA level was confirmed by a scatter diagram (Figure 4), suggesting that serum sulfatide levels may decrease as oxidative stress increases in a time-dependent manner.
DISCUSSION The current follow-up study of maintenance HD patients found that serum sulfatide levels decreased and serum MDA levels increased from the baseline values in each patient, in a time-dependent manner. It is thought that the HD procedure cannot directly remove serum sulfatides because sulfatides are incorporated in lipoproteins with high molecular weight.2 Indeed, our preliminary study measuring serum sulfatide levels before and after a single HD session did not show a reduction in sulfatide levels by
Hemodialysis International 2015; 19:429–438
HD therapy (data not shown). This finding suggests that the time-dependent reduction in serum sulfatide levels in HD patients might be induced by alteration of the internal environment. Our multivariate statistical analysis showed a significant correlation between the time-dependent changes in serum sulfatide levels and serum MDA levels. This finding suggests that time-dependent aggravation of abnormal serum sulfatide levels might be associated with increased oxidative stress, and that the internal environment of HD patients continues to gradually deteriorate. This progressively unhealthy situation may contribute to the poor prognosis in HD patients. Interestingly, the inverse correlation between serum sulfatide levels and oxidative stress could only be detected by follow-up study and not by cross-sectional analysis. To exclude the factor of HD duration, we also examined the correlation between serum sulfatide levels and serum MDA levels in 64 chronic kidney disease patients (stage 5) just before the start of HD. This preliminary crosssectional analysis also could not detect a significant correlation between these two parameters. These findings suggest that serum sulfatide levels are different in each patient even if the absolute level of serum MDA is the same, and that other unknown factors may also play a role in regulating the baseline level of serum sulfatides. Oxidative stress should be considered as a time-dependent aggravating factor that reduces serum sulfatide levels
433
Yuzhe et al.
A
B *** ### *** ***
***###
C LS-d 18:2 LS-d 18:1 LS-d 18:0 LS-t 18:0 LS-d 20:1 LS-d 20:0 LS-t 20:0
4%
5%
7% 10%
8% 5%
9%
6%
5% 4%
9%
54%
10%
0 mo
11%
9%
9%
11%
9%
53%
18 mo
11%
51%
36 mo
Figure 2 Time-dependent alterations in serum levels of sulfatide and an oxidative marker. (A) Serum sulfatide levels and (B) serum malondialdehyde (MDA) levels during 3 y of follow-up. ***P < 0.001 compared with baseline, ###P < 0.001 compared with 18 months. (C) The composition (means) of the different molecular species of serum sulfatides in 95 hemodialysis patients at baseline, 18 mo and 36 mo.
rather than an essential factor regulating the baseline level of serum sulfatides. The factors regulating baseline level of serum sulfatides are still unclear, but the proteins influencing sulfatide synthesis and/or metabolism should be considered. Sulfatides are biosynthesized by only two enzymes, CST and ceramide galactosyltransferase, and are degraded by galactosylceramidase and arylsulfatase A.4 Our recent experimental study found that expression of CST was regulated by a transcriptional factor, peroxisome proliferator-activated receptor α (PPARα).21 It is reported that PPARα becomes degraded with conditions such as kidney dysfunction.22 PPAR α ability in each patients and its degradation by kidney dysfunction might influence the CST expression and serum sulfatide levels. Furthermore, it is possible that alterations of other key enzymes such as
434
arylsulfatase A, ceramide galactosyltransferase, and galactosylceramidase affect serum sulfatide levels in HD patients. We currently have no information regarding alterations of the function of PPARα and sulfatide metabolizing enzymes in HD patients or in experimental models of chronic kidney disease, and further study is required in this area. The time-dependent increases in oxidative stress in HD patients have been controversial. As baseline levels of oxidative stress appear to differ among HD patients depending on their background characteristics, it seems that cross-sectional studies hardly detect the timedependent increase in oxidative stress. The current follow-up study found that oxidative stress increased progressively in HD patients, regardless of the baseline level,
Hemodialysis International 2015; 19:429–438
Sulfatide abnormality in HD patients
Figure 3 Time-dependent alterations in serum sulfatide levels in various subgroups. Blue, green, and blown box plot distributions indicate values at baseline, 18 mo, and 36 mo, respectively. **P < 0.01, ***P < 0.001 compared with baseline. ##P < 0.01, ### P < 0.001 compared with 18 mo. ¶P < 0.05, ¶¶P < 0.01 compared with the other group. (AVF = arteriovenous fistula; AVG = arteriovenous graft; CVD = cardiovascular disease; HDF = hemodialysis filtration; RAS = renin-angiotensin system). Hemodialysis International 2015; 19:429–438
435
Yuzhe et al.
Table 3 Changes in clinical parameters during the follow-up period Follow-up period (mo) Laboratory examination 2
Body mass index (kg/m ) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Total protein (g/dL) Albumin (g/dL) Urea nitrogen (mg/dL) Creatinine (mg/dL) Sodium (mEq/L) Potassium (mEq/L) Chloride (mEq/L) Calcium (mg/dL) Inorganic phosphorus (mg/dL) Uric acid (mg/dL) Total bilirubin (mg/dL) Aspartate aminotransferase (IU/L) Alanine transaminase (IU/L) Alkaline phosphatase (IU/L) Lactate dehydrogenase (IU/L) γ-Glutamyl transpeptidase (IU/L) Amylase (IU/L) Total cholesterol (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Triglycerides (mg/dL) White blood cell count (102/μL) Hemoglobin (g/dL) Platelet count (104/μL) Glucose (mg/dL) C-reactive protein (mg/dL) Ferritin (ng/mL) Transferrin saturation (%) Intact PTH (pg/mL)
0
18
36
22.6 (19.6–28.6) 142 (110–178) 88 (68–102) 6.3 (5.3–7.4) 3.8 (2.9–4.4) 64.3 (15.9–98.6) 10.9 (5.1–17.8) 138 (130–145) 4.8 (3.3–7.1) 105 (89–112) 9.3 (7.8–12.0) 4.8 (1.7–8.8) 7.4 (4.7–10.6) 0.3 (0.1–0.5) 10 (2–46) 8 (3–28) 204 (98–502) 164 (99–303) 19 (6–185) 106 (33–292) 141 (78–241) 42 (20–91) 81 (36–178) 88 (32–289) 51 (25–106) 10.4 (6.5–13.4) 16.4 (2.5–30.9) 128 (83–231) 0.1 (0.0–0.9) 101 (6–566) 22 (6–61) 155 (13–700)
22.4 (18.8–28.4) 140 (102–182) 86 (64–106) 6.2 (5.0–7.3)*** 3.7 (2.4–4.3) 65.6 (12.5–90.7) 11.2 (4.5–17.7) 138 (132–143) 4.7 (2.9–6.1) 103 (89–112)*** 9.4 (7.4–11.9) 4.9 (2.0–7.2) 7.1 (3.1–11.4) 0.3 (0.1–1.4) 11 (2–29) 9 (3–25) 216 (96–518)* 167 (112–291) 19 (6–213) 111 (45–333) 143 (79–232) 42 (21–82) 77 (27–142) 95 (33–420) 51 (27–165) 10.5 (7.3–12.8) 16.9 (4.3–29.5)* 129 (61–211) 0.1 (0.0–1.1) 90 (8–1673) 23 (8–55) 170 (10–920)
22.5 (18.6–28.5) 141 (106–184) 87 (66–104) 6.4 (5.2–7.6)### 3.8 (2.4–4.9)# 59.2 (15.0–95.9)***,### 10.4 (3.4–15.6)**,## 139 (123–145) 4.6 (3.1–7.1)*,# 103 (92–113)** 9.3 (7.0–12.3) 4.3 (2.0–7.4)**,# 7.4 (4.7–10.4) 0.3 (0.1–1.0) 12 (3–39)***,## 9 (2–30) 259 (119–937)***,### 170 (105–295) 20 (5–162)# 114 (37–312)** 146 (86–215) 44 (22–88)***,### 77 (32–132) 90 (30–525) 57 (24–115)***,## 10.7 (7.8–13.9)* 18.5 (3.2–32.3)**,# 125 (81–223) 0.1 (0.0–1.0) 59 (6–1241) 22 (5–54) 206 (7–862)
Continuous variables are shown as median (minimum–maximum). Continuous variables were compared using the Wilcoxon matched-pairs signed-rank test. HDL = high-density lipoprotein; LDL = low-density lipoprotein; PTH = parathyroid hormone. *P < 0.05, **P < 0.01, ***P < 0.001 compared with baseline. # P < 0.05, ##P < 0.01, ###P < 0.001 compared with 18 mo.
suggesting that continuation of HD treatment hardly suppresses the increase in oxidative stress. Intervention to treat this unhealthy internal condition is therefore important. Many studies reported that the oxidative stress of HD patients could be ameliorated by administration of alphatocopherol, vitamin C, and N-acetylcysteine.16 However, neither normalization nor sufficient reduction of oxidative stress in HD patients has yet been established. Our previous clinical studies found that the serum MDA levels and serum sulfatide levels could be normalized by kidney
436
transplantation after a few years.13,14 As continuation of hemodialysis treatment hardly improves oxidative stress and serum sulfatide abnormality, development of novel therapeutic strategies, including promotion of kidney transplantation, may be important. This study has some limitations that should be considered. First, as organ biopsy is not performed in HD patients without clinical need, we could not measure the levels of sulfatide synthesis enzymes. Our proposed molecular mechanism for the decrease in serum sulfatide
Hemodialysis International 2015; 19:429–438
Sulfatide abnormality in HD patients
Table 4 Results of linear regression analyses
Univariate analysis MDA Multivariate analysis MDA
B
95% CI
β
P
−0.194
−0.352 ∼ −0.036
−0.142
0.016
−0.201
−0.367 ∼ −0.035
−0.147
0.018
Multivariate analysis included all parameters that changed significantly during the follow-up period: total protein, albumin, urea nitrogen, creatinine, potassium, chloride, inorganic phosphorus, aspartate aminotransferase, alkaline phosphatase, γ-glutamyl transpeptidase, amylase, high-density lipoprotein cholesterol, white blood cell count, hemoglobin, and platelet count. B = regression coefficient; β = standardized regression coefficient; MDA = malondialdehyde; 95% CI = 95% confidence interval of the regression coefficient.
levels is therefore speculative. Second, the observed inverse correlation between serum sulfatide levels and serum MDA levels does not prove direct causality. Third, we did not measure the other pro-oxidant and antioxidant markers; therefore, further investigation regarding timedependent alterations of these markers is necessary. Fourth, our results may have been influenced by confounding factors that were not accounted for. Fifth, this study included only HD patients who were followed up for 3 years, and patients who died during the study period were excluded. Sixth, this study did not determine whether decreased serum sulfatide levels affect prognosis, and this should be further investigated in future studies. Finally, this study seems to have a relatively small population size when compared with other clinical studies investigating common risk factors in HD patients; therefore, assessment of whether this was an adequate sample
8
Sulfades (nmol/mL)
7 6
y=-0.1941x + 3.8277 R2=0.0202
5 4 3 2
CONCLUSION The results of the current 3-year follow-up study demonstrated, for the first time to the best of the authors’ knowledge, the time-dependent aggravation of serum sulfatide abnormality in HD patients, accompanied by increased MDA levels. Although the absolute level of serum MDA does not determine the baseline level of serum sulfatides, time-dependent deterioration of the internal environment, such as increasing oxidative stress, might be an aggravating factor that results in decreased serum sulfatide levels. As continuation of HD treatment hardly corrects abnormal serum sulfatide levels or the increased oxidative stress, development of novel therapeutic strategies may be important.
ACKNOWLEDGMENTS
1 0
size is necessary. We examined a published study investigating serum sulfatide abnormality in kidney transplantation recipients in which there was a total of 81 samples from 17 recipients. In that study, nevertheless, we were able to detect a relationship between serum sulfatide abnormality and oxidative stress.14 In the current study, we investigated 285 samples from 95 HD patients and were able to detect a potential relationship between timedependent serum sulfatide abnormality and oxidative stress, a finding consistent with our past studies. Therefore, we believe that the sample size of the current study meets the level necessary to evaluate serum sulfatide abnormality in HD patients.
0
1
2
3
4
5
MDA (μmol/L) Figure 4 Scatter plots of serum sulfatide levels and serum malondialdehyde (MDA) levels, showing all the data from all patients. Blue diamonds, red squares, and green triangles indicate values at baseline, 18 mo, and 36 mo, respectively.
Hemodialysis International 2015; 19:429–438
This work was supported by GL Sciences (Tokyo, Japan) who supplied MonoTip C18, an essential material for measuring serum sulfatides. We would like to thank Ms Takayo Maruyama (Aizawa Hospital, Matsumoto, Japan) for collecting serum samples. Conflict of interest statement: The authors declare that they have no competing interests.
437
Yuzhe et al.
Manuscript received June 2014; revised August 2014. 12
REFERENCES 1 Ishizuka I. Chemistry and functional distribution of sulfoglycolipids. Prog Lipid Res. 1997; 36:245–319. 2 Hara A, Taketomi T. Occurrence of sulfatide as a major glycosphingolipid in WHHL rabbit serum lipoproteins. J Biochem. 1987; 102:83–92. 3 Zhu XH, Hara A, Taketomi T. The existence of galactosylceramide I3-sulfatide in serums of various mammals and its anticoagulant activity. J Biochem. 1991; 110:241– 245. 4 Takahashi T, Suzuki T. Role of sulfatide in normal and pathological cells and tissues. J Lipid Res. 2012; 53:1437– 1450. 5 Hara A, Taketomi T. Characterization and changes of glycosphingolipids in the aorta of the Watanabe hereditable hyperlipidemic rabbit. J Biochem. 1991; 109:904– 908. 6 Hara A, Uemura K, Taketomi T. Sulfatide prolongs bloodcoagulation time and bleeding time by forming a complex with fibrinogen. Glycoconj J. 1996; 13:187–194. 7 Kyogashima M, Onaya J, Hara A, Taketomi T. Contradictory functions of sulfatide in the blood coagulation system as coagulant and anticoagulant. Acta Biochim Pol. 1998; 45:493–499. 8 Sarnak MJ, Levey AS, Schoolwerth AC, et al. Kidney disease as a risk factor for development of cardiovascular disease: A statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation. 2003; 108:2154– 2169. 9 Locatelli F, Marcelli D, Conte F, et al. Cardiovascular disease in chronic renal failure: The challenge continues. Nephrol Dial Transplant. 2000; 15:69–80. 10 Hu R, Li G, Kamijo Y, et al. Serum sulfatides as a novel biomarker for cardiovascular disease in patients with end-stage renal failure. Glycoconj J. 2007; 24:565–571. 11 Zhang X, Nakajima T, Kamijo Y, et al. Acute kidney injury induced by protein-overload nephropathy downregulates gene expression of hepatic cerebroside sulfotransferase in mice, resulting in reduction of liver and
438
13
14
15
16
17 18
19
20
21
22
serum sulfatides. Biochem Biophys Res Commun. 2009; 390:1382–1388. Kanbe H, Kamijo Y, Nakajima T, et al. Chronic ethanol consumption decreases serum sulfatide levels by suppressing hepatic cerebroside sulfotransferase expression in mice. Arch Toxicol. 2014; 88:367–379. Wang L, Kamijo Y, Matsumoto A, et al. Kidney transplantation recovers the reduction level of serum sulfatide in ESRD patients via processes correlated to oxidative stress and platelet count. Glycoconj J. 2011; 28:125–135. Kamijo Y, Wang L, Matsumoto A, et al. Long-term improvement of oxidative stress via kidney transplantation ameliorates serum sulfatide levels. Clin Exp Nephrol. 2012; 16:959–967. Lekawanvijit S, Kompa AR, Wang BH, Kelly DJ, Krum H. Cardiorenal syndrome: The emerging role of proteinbound uremic toxins. Circ Res. 2012; 111:1470–1483. Coombes JS, Fassett RG. Antioxidant therapy in hemodialysis patients: A systematic review. Kidney Int. 2012; 81:233–246. Hara A, Radin NS. Lipid extraction of tissues with a low-toxicity solvent. Anal Biochem. 1978; 90:420–426. Li G, Hu R, Kamijo Y, et al. Establishment of a quantitative, qualitative, and high-throughput analysis of sulfatides from small amounts of sera by matrix-assisted laser desorption ionization-time of flight mass spectrometry. Anal Biochem. 2007; 362:1–7. Aoyama T, Yamano S, Waxman DJ, et al. Cytochrome P-450 hPCN3, a novel cytochrome P-450 IIIA gene product that is differentially expressed in adult human liver. cDNA and deduced amino acid sequence and distinct specificities of cDNA-expressed hPCN1 and hPCN3 for the metabolism of steroid hormones and cyclosporine. J Biol Chem. 1989; 264:10388–10395. Nakai S, Watanabe Y, Masakane I, et al. Overview of regular dialysis treatment in Japan (as of 31 December 2011). Ther Apher Dial. 2013; 17:567–611. Nakajima T, Kamijo Y, Yuzhe H, et al. Peroxisome proliferator-activated receptor α mediates enhancement of gene expression of cerebroside sulfotransferase in several murine organs. Glycoconj J. 2013; 30:553–560. Lopez-Hernandez FJ, Lopez-Novoa JM. Potential utility of PPARalpha activation in the prevention of ischemic and drug-induced acute renal damage. Kidney Int. 2009; 76:1022–1024.
Hemodialysis International 2015; 19:429–438
Serum sulfatide abnormality is associated with increased oxidative stress in hemodialysis patients Huang YUZHE,1,2* Yuji KAMIJO,1,3* Koji HASHIMOTO,3 Makoto HARADA,1,3 Taro KANNO,3 Eiko SUGIYAMA,1,4 Mamoru KYOGASHIMA,5 Tomomasa OGUCHI,6 Takero NAKAJIMA,1 Yutaka KANNO,7 Toshifumi AOYAMA1 1 Department of Metabolic Regulation, Institute of Pathogenesis and Disease Prevention, Shinshu University Graduate School of Medicine, Matsumoto, Nagano, Japan; 2Department of Human Anatomy, Hebei Medical University, Shijiazhuang, Hebei, China; 3Department of Nephrology, Shinshu University School of Medicine, Matsumoto, Nagano, Japan; 4Department of Nutritional Science, Nagano Prefectural College, Nagano, Nagano, Japan; 5Department of Microbiology and Molecular Cell Biology, Nihon Pharmaceutical University, Ina, Saitama, Japan; 6Kidney Disease and Dialysis Center, Aizawa Hospital, Matsumoto, Nagano, Japan; 7Department of Nephrology, Kanno Dialysis and Vascular Access Clinic, Matsumoto, Nagano, Japan
Abstract Sulfatides are major glycosphingolipids of lipoproteins that influence atherosclerosis and blood coagulation. Our previous cross-sectional study of hemodialysis patients showed that serum sulfatide levels decreased markedly with increasing duration of hemodialysis treatment, which may contribute to the development of cardiovascular disease. However, this past study could not demonstrate the time-dependent change in serum sulfatide levels in each patient, and the underlying mechanism is unknown. To confirm the time-dependent aggravation of serum sulfatide abnormality, 95 stable hemodialysis outpatients were followed up for 3 years. To show the underlying mechanisms, we statistically analyzed correlations between serum sulfatide levels and clinical factors, including an oxidative stress marker, malondialdehyde. Serum sulfatides were quantified by mass spectrometry after conversion to lysosulfatides. Malondialdehyde was measured using a colorimetric assay. The results showed a time-dependent decrease in serum sulfatide levels associated with increased malondialdehyde levels, although the absolute level of serum malondialdehyde does not determine the baseline level of serum sulfatides. Multiple linear regression analysis showed a significant correlation only between the time-dependent change in serum sulfatide levels and the time-dependent change in serum malondialdehyde levels. This study demonstrated, for the first time, a time-dependent aggravation of serum sulfatide abnormality in hemodialysis patients, as well as the potential relationship between serum sulfatide abnormality and increasing oxidative stress. These findings suggest that oxidative stress might be an aggravating factor in serum sulfatide abnormality. As continuation of hemodialysis treatment hardly improves abnormal serum sulfatide levels or increased oxidative stress, development of novel therapeutic strategies may be important. Key words: Cardiovascular disease, hemodialysis, malondialdehyde, oxidative stress, serum sulfatide level
Correspondence to: Y. Kamijo, MD, PhD, Department of Nephrology, Shinshu University School of Medicine, 3-1-1 Asahi, Matsumoto 390-8621, Japan. E-mail: [email protected] *These authors contributed equally to this work. Disclosure of grants or other funding: None.
© 2015 International Society for Hemodialysis DOI:10.1111/hdi.12270
429
Yuzhe et al.
INTRODUCTION Sulfatides are amphiphilic molecules that are synthesized from galactosylceramides by sulfation at C3 of the galactosyl residue.1 Sulfatides are widely distributed in mammalian tissues, such as the brain, kidney, liver, and digestive tract, and also exist in the sera of various mammals as major glycosphingolipids of lipoproteins.1–4 The physiological roles of serum sulfatides are still unclear. Our previous study found obvious accumulation of sulfatides in the sclerotic aorta of Watanabe hereditary hyperlipidemic rabbits, which provide an animal model for human familial hypercholesterolemia, suggesting that sulfatides play a significant role in the progression of atherosclerosis.2,5 Other experimental studies reported that intravenous administration of exogenous sulfatides had significant effects on blood coagulation.6,7 Most studies reported that serum sulfatides had anticoagulant and antiplatelet effects, but some reported that sulfatides accelerated blood coagulation.7 The results of these studies suggest that abnormal serum sulfatide levels may play a role in the pathogenesis of cardiovascular disease (CVD). Chronic kidney disease patients undergoing hemodialysis (HD) therapy have a high incidence of CVD, which is the major cause of death in this population.8,9 Interestingly, our recent cross-sectional study found that serum sulfatide levels were remarkably low in HD patients compared with normal subjects, and that serum sulfatide levels were significantly lower in HD patients with CVD than in HD patients without CVD.10 These findings suggest that decreased serum sulfatide levels may contribute to the development of CVD in HD patients because of abnormal sulfatide function. The mechanisms underlying the decrease in serum sulfatide levels in HD patients are currently unknown. Our previous cross-sectional clinical study found an inverse correlation between the serum sulfatide level and the duration of HD, suggesting that alteration of the internal environment caused by long-term kidney failure gradually decreases the serum sulfatide level.10 However, this past study could not demonstrate the time-dependent change in serum sulfatide levels in each patient. To confirm the time-dependent change in serum sulfatide levels in each HD patients and to elucidate the underlying mechanisms, we conducted a 3-year observational study and investigated the clinical factors correlated with serum sulfatide levels. Serum sulfatides are produced and secreted mainly by the liver as the major component of lipoproteins.2 Our previous experimental studies using an acute kidney
430
injury model and an acute hepatic injury model found that increased systemic oxidative stress reduced the hepatic expression of an essential synthetic enzyme of sulfatides, cerebroside sulfotransferase (CST), followed by a decrease in serum sulfatide levels.11,12 Furthermore, we reported that kidney transplantation normalized the serum sulfatide abnormality and oxidative stress synchronously in kidney transplant recipients.13,14 These findings suggest a close relationship between serum sulfatide abnormality and oxidative stress. Many past studies reported that HD patients are exposed to high levels of oxidative stress, caused by increased production of reactive oxygen species because of their uremia and HD procedures, as well as by decreased antioxidant activity and decreased uptake of dietary antioxidants.15,16 This unhealthy internal condition in HD patients may suppress hepatic sulfatide synthesis, leading to a decrease in serum sulfatide levels. In the current study, we investigated whether there is a relationship between serum sulfatide abnormality and oxidative stress in HD patients. Because malondialdehyde (MDA) is one of the most well-known classical oxidative stress markers and the analytical method of MDA is well established, we selected and utilized the serum MDA levels as a representative oxidative stress marker, which is common with all of our studies.
MATERIALS AND METHODS Hemodialysis patients and study protocol To examine the general characteristics in HD patients, we sought to survey all outpatients undergoing maintenance HD therapy at affiliated hospitals of Shinshu University. Among them, 201 patients agreed to participate in the study and were enrolled. From this population, unstable patients with cancer (n = 3) and/or serum positivity of chronic infection (hepatitis C virus, n = 26; hepatitis B virus, n = 12; syphilis, n = 1), 1 patient who planned to shift to the kidney transplantation unit, and 2 patients who withdrew their consent were excluded at the beginning of the observation period (baseline number = 156). Additionally, to confirm the time-dependent change in serum sulfatide levels in HD patients over 3 years, 34 patients who died during the study period and 27 patients without clinical data and/or serum samples were excluded. Finally, this study investigated 95 patients. Serum samples were collected at the beginning of the observation period (baseline) and after 18 and 36 months, and were stored in a deep freezer. All serum samples were obtained just before the first HD session of the week.
Hemodialysis International 2015; 19:429–438
Sulfatide abnormality in HD patients
Serum sulfatide levels and serum MDA levels in the frozen samples were analyzed together with many other laboratory data. The study protocol complied with the guidelines of the 2004 revision of the Declaration of Helsinki and was approved by the ethics committee of Shinshu University (approval number 750). Written informed consent was obtained from all patients before inclusion in the study.
in serum sulfatide levels were analyzed by multivariate linear regression analysis. The multivariate analysis included all clinical follow-up parameters that changed significantly during the follow-up period. All statistical analyses were performed using SPSS version 18.0J (SPSS, Inc., Chicago, IL, USA). The level of statistical significance was set at P < 0.05.
Quantification of serum sulfatide levels
RESULTS
Sulfatides were extracted from the serum samples using the hexane/isopropanol method and were converted to lysosulfatides (LSs; sulfatides without fatty acids) by saponification with sodium hydroxide.17–19 The LS samples were analyzed by matrix-assisted laser desorption ionization-time of flight mass spectrometry using a Voyager Elite XL Biospectrometry Workstation (PerSeptive Biosystems, Framingham, MA, USA) as previously described.18 Seven molecular species of LS were detected, based on differences in sphingoid base structures: LS-sphingadienine (LS-d18:2), LS-(4E)-sphinganine (LS-d18:1), LS-sphinganine (LS-d18:0), LS-4D-hydroxysphinganine (LS-t18:0), LS-(4E)-eicosasphingenine (LS-d20:1), LS-eicosasphinganine (LS-d20:0), and LS-4Dhydroxyeicosasphinganine (LS-t20:0). The total serum sulfatide level was calculated as the sum of the levels of these 7 LS molecular species. When sulfatide levels were measured for new samples, the same standard serum sample and the chemically prepared internal standard, hydrogenated N-acetyl LS (LS-d18:0 NAc), were always used in order to verify the consistency of the data.
Baseline characteristics
Analysis of serum oxidative stress marker The serum level of the oxidative stress marker MDA was measured using a lipid peroxidation colorimetric assay kit purchased from Oxis International (Beverly Hills, CA, USA).
Statistical analysis Continuous variables were handled as nonparametric variables and are shown as the median (minimum– maximum). Time-dependent changes in paired continuous variables were analyzed using the Wilcoxon matched-pairs signed-rank test. Two independent groups of continuous variables were compared using the MannWhitney U test. Correlations between clinical factors and serum sulfatide levels were determined using Spearman’s correlation coefficient. The clinical parameters that were most strongly correlated with the time-dependent changes
Hemodialysis International 2015; 19:429–438
The baseline characteristics of the HD patients included in the study are shown in Table 1. Compared with the latest overview of regular dialysis treatment in Japan by the Japanese Society for Dialysis Therapy,20 HD patients in the current study had a lower median age (63 years vs. 67 years), higher proportion of males (67% vs. 63%), and lower prevalence of diabetes mellitus (31.6% vs. 37%) at study entry (Table 1). The median duration of maintenance HD treatment was 8.4 years. This study included only HD patients who could be followed up for 3 years, and excluded unstable patients and the dead during the follow-up, and therefore may have enrolled patients with relatively good physical condition compared with the general population of Japanese HD patients. Nevertheless, 30.5% of enrolled patients had a history of CVD at study entry. These HD patients were treated with various medicines; however, no patient used any antioxidant medicine, such as alpha-tocopherol, vitamin C, and N-acetylcysteine.
Cross-sectional analysis of correlations between clinical factors and serum sulfatide levels The mean serum sulfatide level of the 95 HD patients at study entry was 3.66 ± 1.00 nmol/mL. When compared with the normal serum sulfatide level in healthy control participants (8.21 ± 1.50 nmol/mL) reported in our previous study,10 the serum sulfatide level in HD patients appeared to be less than 50%. Consistent with our previous study,10 the mean serum sulfatide level was markedly lower in HD patients with a history of CVD than in HD patients without a history of CVD (3.14 ± 0.62 nmol/mL vs. 3.90 ± 1.05 nmol/mL, P < 0.01). To show the mechanisms underlying the decrease in serum sulfatide levels in HD patients, we analyzed correlations between baseline clinical parameters and serum sulfatide levels. Spearman’s rank correlation analysis and scatter diagrams showed an
431
Yuzhe et al.
63 (30–91) 64/31 29 (30.5%) 9 (9.5%) 4 (4.2%) 1 (1.1%) 3 (3.2%) 3 (3.2%) 17 (17.9%) 2 (2.1%) 47 (49.5%) 8.4 (0.5–27.7) 85 (89.5%) 10 (10.5%) 0 (0%) 30 (31.6%) 27 (28.4%) 10 (10.5%) 28 (29.5%) 22.6 (19.6–33.8) 142 (110–178) 88 (68–102) 51 (53.7%) 75 (78.9%) 17 (17.9%) 26 (27.4%) 0 (0%) 48 (50.5%) 82 (86.3%)
Continuous variables are shown as median (minimum–maximum). CVD = cardiovascular disease; HD = hemodialysis; RAS = renin-angiotensin system.
inverse correlation between serum sulfatide levels and the duration of HD, but no significant correlations between serum sulfatide levels and other clinical parameters including serum MDA levels (Table 2; Figure 1).
Analysis of clinical factors correlated with time-dependent changes in serum sulfatide levels Follow-up of the enrolled HD patients for 3 years confirmed a time-dependent decrease in serum sulfatide levels (Figure 2A). Interestingly, this follow-up analysis
432
8 7
Sulfades (nmol/mL)
Age (y) Sex (male/female) History of CVD, n (%) Angina pectoris Myocardial infarction Aortic aneurysm Peripheral vascular disease Transient ischemic attack Cerebral infarction Cerebral hemorrhage Smoking history, n (%) Duration of HD (y) Vascular access, n (%) Arteriovenous fistula Arteriovenous graft Catheter Cause of kidney failure, n (%) Diabetes mellitus Chronic glomerulonephritis Hypertension Other Body mass index (kg/m2) Blood pressure (mmHg) Systolic Diastolic Medication, n (%) RAS inhibitor Calcium channel blocker Beta blocker Antiplatelet agent Warfarin Vitamin D Erythropoietin
A
y = -0.0718x + 4.274 R² = 0.217
6 5 4 3 2 1 0 0
5
10
15
20
25
30
2.50
3.00
HD duraon (y)
B
8 7
Sulfades (nmol/mL)
Table 1 Baseline characteristics of maintenance hemodialysis patients (N = 95)
y = 0.1011x + 3.5368 R² = 0.0013
6 5 4 3 2 1 0 0.00
0.50
1.00
1.50
2.00
MDA (μmol/L) Figure 1 Serum sulfatide levels in hemodialysis (HD) patients. (A) Scatter plot showing serum sulfatide levels according to duration of HD. (B) Scatter plot showing crosssectional analysis of serum sulfatide levels and serum malondialdehyde (MDA) levels.
also showed a marked time-dependent increase in serum MDA levels (Figure 2B). The composition of the different molecular species of serum sulfatides in the 95 HD patients did not differ at each follow-up point, suggesting that all molecular species similarly decreased in a timedependent manner (Figure 2C). Therefore, we used sum data of the 7 sulfatides as representative data of sulfatides in the statistical analyses. The time-dependent decrease in serum sulfatide levels was not affected by sex, CVD history, smoking history, diabetes mellitus, type of vascu-
Hemodialysis International 2015; 19:429–438
Sulfatide abnormality in HD patients
Table 2 Spearman’s rank correlations between serum sulfatide levels and various parameters at baseline Parameter Serum sulfatides (nmol/mL) HD duration MDA (μmol/L) Age Body mass index (kg/m2) Systolic blood pressure Diastolic blood pressure Total protein (g/dL) Albumin (g/dL) Urea nitrogen (mg/dL) Creatinine (mg/dL) Sodium (mEq/L) Potassium (mEq/L) Chloride (mEq/L) Calcium (mg/dL) Inorganic phosphorus (mg/dL) Uric acid (mg/dL) Total bilirubin (mg/dL)
Spearman’s correlation
Parameter
Spearman’s correlation
1 −0.453*** 0.041 0.104 0.092 0.023 0.003 −0.040 0.047 0.016 −0.122 0.042 0.002 0.117 −0.194 0.032 −0.034 0.183
Aspartate amino transferase (IU/L) Alanine transaminase (IU/L) Alkaline phosphatase (IU/L) Lactate dehydrogenase (IU/L) γ-Glutamyl transpeptidase (IU/L) Amylase (IU/L) Total cholesterol (mg/dL) Low-density lipoprotein (mg/dL) High-density lipoprotein (mg/dL) Triglyceride (mg/dL) White blood cell count (102/μL) Hemoglobin (g/dL) Platelet count (104/μL) Glucose (mg/dL) C-reactive protein (mg/dL) Ferritin (ng/mL) Transferrin saturation (%) Intact PTH (pg/mL)
−0.097 −0.023 0.060 0.028 −0.160 −0.027 −0.021 −0.058 −0.104 0.069 0.009 0.113 −0.027 −0.051 −0.180 0.078 0.194 0.106
HD = hemodialysis; MDA = malondialdehyde; PTH = parathyroid hormone. ***P < 0.001.
lar access, dialysis method (conventional HD or online hemodialysis filtration), or various oral medications (Figure 3). To identify other factors potentially correlated with time-dependent changes in serum sulfatide levels, we examined the changes in serum levels of various parameters during the follow-up period (Table 3). Linear regression analysis was performed using all observational point data of candidate parameters including serum MDA levels, and found that the MDA level was the only factor significantly correlated with the time-dependent change in serum sulfatide level (Table 4). The inverse correlation between serum sulfatide level and serum MDA level was confirmed by a scatter diagram (Figure 4), suggesting that serum sulfatide levels may decrease as oxidative stress increases in a time-dependent manner.
DISCUSSION The current follow-up study of maintenance HD patients found that serum sulfatide levels decreased and serum MDA levels increased from the baseline values in each patient, in a time-dependent manner. It is thought that the HD procedure cannot directly remove serum sulfatides because sulfatides are incorporated in lipoproteins with high molecular weight.2 Indeed, our preliminary study measuring serum sulfatide levels before and after a single HD session did not show a reduction in sulfatide levels by
Hemodialysis International 2015; 19:429–438
HD therapy (data not shown). This finding suggests that the time-dependent reduction in serum sulfatide levels in HD patients might be induced by alteration of the internal environment. Our multivariate statistical analysis showed a significant correlation between the time-dependent changes in serum sulfatide levels and serum MDA levels. This finding suggests that time-dependent aggravation of abnormal serum sulfatide levels might be associated with increased oxidative stress, and that the internal environment of HD patients continues to gradually deteriorate. This progressively unhealthy situation may contribute to the poor prognosis in HD patients. Interestingly, the inverse correlation between serum sulfatide levels and oxidative stress could only be detected by follow-up study and not by cross-sectional analysis. To exclude the factor of HD duration, we also examined the correlation between serum sulfatide levels and serum MDA levels in 64 chronic kidney disease patients (stage 5) just before the start of HD. This preliminary crosssectional analysis also could not detect a significant correlation between these two parameters. These findings suggest that serum sulfatide levels are different in each patient even if the absolute level of serum MDA is the same, and that other unknown factors may also play a role in regulating the baseline level of serum sulfatides. Oxidative stress should be considered as a time-dependent aggravating factor that reduces serum sulfatide levels
433
Yuzhe et al.
A
B *** ### *** ***
***###
C LS-d 18:2 LS-d 18:1 LS-d 18:0 LS-t 18:0 LS-d 20:1 LS-d 20:0 LS-t 20:0
4%
5%
7% 10%
8% 5%
9%
6%
5% 4%
9%
54%
10%
0 mo
11%
9%
9%
11%
9%
53%
18 mo
11%
51%
36 mo
Figure 2 Time-dependent alterations in serum levels of sulfatide and an oxidative marker. (A) Serum sulfatide levels and (B) serum malondialdehyde (MDA) levels during 3 y of follow-up. ***P < 0.001 compared with baseline, ###P < 0.001 compared with 18 months. (C) The composition (means) of the different molecular species of serum sulfatides in 95 hemodialysis patients at baseline, 18 mo and 36 mo.
rather than an essential factor regulating the baseline level of serum sulfatides. The factors regulating baseline level of serum sulfatides are still unclear, but the proteins influencing sulfatide synthesis and/or metabolism should be considered. Sulfatides are biosynthesized by only two enzymes, CST and ceramide galactosyltransferase, and are degraded by galactosylceramidase and arylsulfatase A.4 Our recent experimental study found that expression of CST was regulated by a transcriptional factor, peroxisome proliferator-activated receptor α (PPARα).21 It is reported that PPARα becomes degraded with conditions such as kidney dysfunction.22 PPAR α ability in each patients and its degradation by kidney dysfunction might influence the CST expression and serum sulfatide levels. Furthermore, it is possible that alterations of other key enzymes such as
434
arylsulfatase A, ceramide galactosyltransferase, and galactosylceramidase affect serum sulfatide levels in HD patients. We currently have no information regarding alterations of the function of PPARα and sulfatide metabolizing enzymes in HD patients or in experimental models of chronic kidney disease, and further study is required in this area. The time-dependent increases in oxidative stress in HD patients have been controversial. As baseline levels of oxidative stress appear to differ among HD patients depending on their background characteristics, it seems that cross-sectional studies hardly detect the timedependent increase in oxidative stress. The current follow-up study found that oxidative stress increased progressively in HD patients, regardless of the baseline level,
Hemodialysis International 2015; 19:429–438
Sulfatide abnormality in HD patients
Figure 3 Time-dependent alterations in serum sulfatide levels in various subgroups. Blue, green, and blown box plot distributions indicate values at baseline, 18 mo, and 36 mo, respectively. **P < 0.01, ***P < 0.001 compared with baseline. ##P < 0.01, ### P < 0.001 compared with 18 mo. ¶P < 0.05, ¶¶P < 0.01 compared with the other group. (AVF = arteriovenous fistula; AVG = arteriovenous graft; CVD = cardiovascular disease; HDF = hemodialysis filtration; RAS = renin-angiotensin system). Hemodialysis International 2015; 19:429–438
435
Yuzhe et al.
Table 3 Changes in clinical parameters during the follow-up period Follow-up period (mo) Laboratory examination 2
Body mass index (kg/m ) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Total protein (g/dL) Albumin (g/dL) Urea nitrogen (mg/dL) Creatinine (mg/dL) Sodium (mEq/L) Potassium (mEq/L) Chloride (mEq/L) Calcium (mg/dL) Inorganic phosphorus (mg/dL) Uric acid (mg/dL) Total bilirubin (mg/dL) Aspartate aminotransferase (IU/L) Alanine transaminase (IU/L) Alkaline phosphatase (IU/L) Lactate dehydrogenase (IU/L) γ-Glutamyl transpeptidase (IU/L) Amylase (IU/L) Total cholesterol (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Triglycerides (mg/dL) White blood cell count (102/μL) Hemoglobin (g/dL) Platelet count (104/μL) Glucose (mg/dL) C-reactive protein (mg/dL) Ferritin (ng/mL) Transferrin saturation (%) Intact PTH (pg/mL)
0
18
36
22.6 (19.6–28.6) 142 (110–178) 88 (68–102) 6.3 (5.3–7.4) 3.8 (2.9–4.4) 64.3 (15.9–98.6) 10.9 (5.1–17.8) 138 (130–145) 4.8 (3.3–7.1) 105 (89–112) 9.3 (7.8–12.0) 4.8 (1.7–8.8) 7.4 (4.7–10.6) 0.3 (0.1–0.5) 10 (2–46) 8 (3–28) 204 (98–502) 164 (99–303) 19 (6–185) 106 (33–292) 141 (78–241) 42 (20–91) 81 (36–178) 88 (32–289) 51 (25–106) 10.4 (6.5–13.4) 16.4 (2.5–30.9) 128 (83–231) 0.1 (0.0–0.9) 101 (6–566) 22 (6–61) 155 (13–700)
22.4 (18.8–28.4) 140 (102–182) 86 (64–106) 6.2 (5.0–7.3)*** 3.7 (2.4–4.3) 65.6 (12.5–90.7) 11.2 (4.5–17.7) 138 (132–143) 4.7 (2.9–6.1) 103 (89–112)*** 9.4 (7.4–11.9) 4.9 (2.0–7.2) 7.1 (3.1–11.4) 0.3 (0.1–1.4) 11 (2–29) 9 (3–25) 216 (96–518)* 167 (112–291) 19 (6–213) 111 (45–333) 143 (79–232) 42 (21–82) 77 (27–142) 95 (33–420) 51 (27–165) 10.5 (7.3–12.8) 16.9 (4.3–29.5)* 129 (61–211) 0.1 (0.0–1.1) 90 (8–1673) 23 (8–55) 170 (10–920)
22.5 (18.6–28.5) 141 (106–184) 87 (66–104) 6.4 (5.2–7.6)### 3.8 (2.4–4.9)# 59.2 (15.0–95.9)***,### 10.4 (3.4–15.6)**,## 139 (123–145) 4.6 (3.1–7.1)*,# 103 (92–113)** 9.3 (7.0–12.3) 4.3 (2.0–7.4)**,# 7.4 (4.7–10.4) 0.3 (0.1–1.0) 12 (3–39)***,## 9 (2–30) 259 (119–937)***,### 170 (105–295) 20 (5–162)# 114 (37–312)** 146 (86–215) 44 (22–88)***,### 77 (32–132) 90 (30–525) 57 (24–115)***,## 10.7 (7.8–13.9)* 18.5 (3.2–32.3)**,# 125 (81–223) 0.1 (0.0–1.0) 59 (6–1241) 22 (5–54) 206 (7–862)
Continuous variables are shown as median (minimum–maximum). Continuous variables were compared using the Wilcoxon matched-pairs signed-rank test. HDL = high-density lipoprotein; LDL = low-density lipoprotein; PTH = parathyroid hormone. *P < 0.05, **P < 0.01, ***P < 0.001 compared with baseline. # P < 0.05, ##P < 0.01, ###P < 0.001 compared with 18 mo.
suggesting that continuation of HD treatment hardly suppresses the increase in oxidative stress. Intervention to treat this unhealthy internal condition is therefore important. Many studies reported that the oxidative stress of HD patients could be ameliorated by administration of alphatocopherol, vitamin C, and N-acetylcysteine.16 However, neither normalization nor sufficient reduction of oxidative stress in HD patients has yet been established. Our previous clinical studies found that the serum MDA levels and serum sulfatide levels could be normalized by kidney
436
transplantation after a few years.13,14 As continuation of hemodialysis treatment hardly improves oxidative stress and serum sulfatide abnormality, development of novel therapeutic strategies, including promotion of kidney transplantation, may be important. This study has some limitations that should be considered. First, as organ biopsy is not performed in HD patients without clinical need, we could not measure the levels of sulfatide synthesis enzymes. Our proposed molecular mechanism for the decrease in serum sulfatide
Hemodialysis International 2015; 19:429–438
Sulfatide abnormality in HD patients
Table 4 Results of linear regression analyses
Univariate analysis MDA Multivariate analysis MDA
B
95% CI
β
P
−0.194
−0.352 ∼ −0.036
−0.142
0.016
−0.201
−0.367 ∼ −0.035
−0.147
0.018
Multivariate analysis included all parameters that changed significantly during the follow-up period: total protein, albumin, urea nitrogen, creatinine, potassium, chloride, inorganic phosphorus, aspartate aminotransferase, alkaline phosphatase, γ-glutamyl transpeptidase, amylase, high-density lipoprotein cholesterol, white blood cell count, hemoglobin, and platelet count. B = regression coefficient; β = standardized regression coefficient; MDA = malondialdehyde; 95% CI = 95% confidence interval of the regression coefficient.
levels is therefore speculative. Second, the observed inverse correlation between serum sulfatide levels and serum MDA levels does not prove direct causality. Third, we did not measure the other pro-oxidant and antioxidant markers; therefore, further investigation regarding timedependent alterations of these markers is necessary. Fourth, our results may have been influenced by confounding factors that were not accounted for. Fifth, this study included only HD patients who were followed up for 3 years, and patients who died during the study period were excluded. Sixth, this study did not determine whether decreased serum sulfatide levels affect prognosis, and this should be further investigated in future studies. Finally, this study seems to have a relatively small population size when compared with other clinical studies investigating common risk factors in HD patients; therefore, assessment of whether this was an adequate sample
8
Sulfades (nmol/mL)
7 6
y=-0.1941x + 3.8277 R2=0.0202
5 4 3 2
CONCLUSION The results of the current 3-year follow-up study demonstrated, for the first time to the best of the authors’ knowledge, the time-dependent aggravation of serum sulfatide abnormality in HD patients, accompanied by increased MDA levels. Although the absolute level of serum MDA does not determine the baseline level of serum sulfatides, time-dependent deterioration of the internal environment, such as increasing oxidative stress, might be an aggravating factor that results in decreased serum sulfatide levels. As continuation of HD treatment hardly corrects abnormal serum sulfatide levels or the increased oxidative stress, development of novel therapeutic strategies may be important.
ACKNOWLEDGMENTS
1 0
size is necessary. We examined a published study investigating serum sulfatide abnormality in kidney transplantation recipients in which there was a total of 81 samples from 17 recipients. In that study, nevertheless, we were able to detect a relationship between serum sulfatide abnormality and oxidative stress.14 In the current study, we investigated 285 samples from 95 HD patients and were able to detect a potential relationship between timedependent serum sulfatide abnormality and oxidative stress, a finding consistent with our past studies. Therefore, we believe that the sample size of the current study meets the level necessary to evaluate serum sulfatide abnormality in HD patients.
0
1
2
3
4
5
MDA (μmol/L) Figure 4 Scatter plots of serum sulfatide levels and serum malondialdehyde (MDA) levels, showing all the data from all patients. Blue diamonds, red squares, and green triangles indicate values at baseline, 18 mo, and 36 mo, respectively.
Hemodialysis International 2015; 19:429–438
This work was supported by GL Sciences (Tokyo, Japan) who supplied MonoTip C18, an essential material for measuring serum sulfatides. We would like to thank Ms Takayo Maruyama (Aizawa Hospital, Matsumoto, Japan) for collecting serum samples. Conflict of interest statement: The authors declare that they have no competing interests.
437
Yuzhe et al.
Manuscript received June 2014; revised August 2014. 12
REFERENCES 1 Ishizuka I. Chemistry and functional distribution of sulfoglycolipids. Prog Lipid Res. 1997; 36:245–319. 2 Hara A, Taketomi T. Occurrence of sulfatide as a major glycosphingolipid in WHHL rabbit serum lipoproteins. J Biochem. 1987; 102:83–92. 3 Zhu XH, Hara A, Taketomi T. The existence of galactosylceramide I3-sulfatide in serums of various mammals and its anticoagulant activity. J Biochem. 1991; 110:241– 245. 4 Takahashi T, Suzuki T. Role of sulfatide in normal and pathological cells and tissues. J Lipid Res. 2012; 53:1437– 1450. 5 Hara A, Taketomi T. Characterization and changes of glycosphingolipids in the aorta of the Watanabe hereditable hyperlipidemic rabbit. J Biochem. 1991; 109:904– 908. 6 Hara A, Uemura K, Taketomi T. Sulfatide prolongs bloodcoagulation time and bleeding time by forming a complex with fibrinogen. Glycoconj J. 1996; 13:187–194. 7 Kyogashima M, Onaya J, Hara A, Taketomi T. Contradictory functions of sulfatide in the blood coagulation system as coagulant and anticoagulant. Acta Biochim Pol. 1998; 45:493–499. 8 Sarnak MJ, Levey AS, Schoolwerth AC, et al. Kidney disease as a risk factor for development of cardiovascular disease: A statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation. 2003; 108:2154– 2169. 9 Locatelli F, Marcelli D, Conte F, et al. Cardiovascular disease in chronic renal failure: The challenge continues. Nephrol Dial Transplant. 2000; 15:69–80. 10 Hu R, Li G, Kamijo Y, et al. Serum sulfatides as a novel biomarker for cardiovascular disease in patients with end-stage renal failure. Glycoconj J. 2007; 24:565–571. 11 Zhang X, Nakajima T, Kamijo Y, et al. Acute kidney injury induced by protein-overload nephropathy downregulates gene expression of hepatic cerebroside sulfotransferase in mice, resulting in reduction of liver and
438
13
14
15
16
17 18
19
20
21
22
serum sulfatides. Biochem Biophys Res Commun. 2009; 390:1382–1388. Kanbe H, Kamijo Y, Nakajima T, et al. Chronic ethanol consumption decreases serum sulfatide levels by suppressing hepatic cerebroside sulfotransferase expression in mice. Arch Toxicol. 2014; 88:367–379. Wang L, Kamijo Y, Matsumoto A, et al. Kidney transplantation recovers the reduction level of serum sulfatide in ESRD patients via processes correlated to oxidative stress and platelet count. Glycoconj J. 2011; 28:125–135. Kamijo Y, Wang L, Matsumoto A, et al. Long-term improvement of oxidative stress via kidney transplantation ameliorates serum sulfatide levels. Clin Exp Nephrol. 2012; 16:959–967. Lekawanvijit S, Kompa AR, Wang BH, Kelly DJ, Krum H. Cardiorenal syndrome: The emerging role of proteinbound uremic toxins. Circ Res. 2012; 111:1470–1483. Coombes JS, Fassett RG. Antioxidant therapy in hemodialysis patients: A systematic review. Kidney Int. 2012; 81:233–246. Hara A, Radin NS. Lipid extraction of tissues with a low-toxicity solvent. Anal Biochem. 1978; 90:420–426. Li G, Hu R, Kamijo Y, et al. Establishment of a quantitative, qualitative, and high-throughput analysis of sulfatides from small amounts of sera by matrix-assisted laser desorption ionization-time of flight mass spectrometry. Anal Biochem. 2007; 362:1–7. Aoyama T, Yamano S, Waxman DJ, et al. Cytochrome P-450 hPCN3, a novel cytochrome P-450 IIIA gene product that is differentially expressed in adult human liver. cDNA and deduced amino acid sequence and distinct specificities of cDNA-expressed hPCN1 and hPCN3 for the metabolism of steroid hormones and cyclosporine. J Biol Chem. 1989; 264:10388–10395. Nakai S, Watanabe Y, Masakane I, et al. Overview of regular dialysis treatment in Japan (as of 31 December 2011). Ther Apher Dial. 2013; 17:567–611. Nakajima T, Kamijo Y, Yuzhe H, et al. Peroxisome proliferator-activated receptor α mediates enhancement of gene expression of cerebroside sulfotransferase in several murine organs. Glycoconj J. 2013; 30:553–560. Lopez-Hernandez FJ, Lopez-Novoa JM. Potential utility of PPARalpha activation in the prevention of ischemic and drug-induced acute renal damage. Kidney Int. 2009; 76:1022–1024.
Hemodialysis International 2015; 19:429–438
