Erythrocytes' antioxidative capacity as a potential marker of oxidative stress intensity in neuroinflammation

Journal of the Neurological Sciences 337 (2014) 8–13

Contents lists available at ScienceDirect

Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns

Erythrocytes' antioxidative capacity as a potential marker of oxidative stress intensity in neuroinflammation Srdjan Ljubisavljevic a,b,⁎, Ivana Stojanovic c, Tatjana Cvetkovic c, Slobodan Vojinovic a, Dragan Stojanov d, Dijana Stojanovic b, Nikola Stefanovic e, Dusica Pavlovic c a

Clinic of Neurology, Clinical Center Nis, Bul. Dr Zorana Djindjica 48, 18000 Nis, Serbia Institute for Pathophysiology, Faculty of Medicine, University of Nis, Bul. Dr Zorana Djindjica 81, 18000 Nis, Serbia Institute for Biochemistry, Faculty of Medicine, University of Nis, Bul. Dr Zorana Djindjica 81, 18000 Nis, Serbia d Center for Radiology, Clinical Center Nis, Bul. Dr Zorana Djindjica 48, 18000 Nis, Serbia e Department for Pharmacy, Faculty of Medicine, University of Nis, Bul. Dr Zorana Djindjica 81, 18000 Nis, Serbia b c

a r t i c l e

i n f o

Article history: Received 4 September 2013 Received in revised form 6 October 2013 Accepted 4 November 2013 Available online 13 November 2013 Keywords: Advanced oxidation protein products Malondialdehyde Superoxide dismutase Clinically isolated syndrome Relapsing–remitting multiple sclerosis

a b s t r a c t The study is designed to assess the oxidative stress intensity in erythrocytes obtained from patients in different clinical phenotypes of neuroinflammation, defined as clinically isolated syndrome (CIS) and relapsing–remitting multiple sclerosis (RRMS). Advanced oxidation protein products (AOPP), malondialdehyde (MDA) and superoxide dismutase (SOD) activity were measured and compared with patients' clinical severity (expanded disability status scale—EDSS), radiological findings (gadolinium enhancement lesion volume—Gd+) and disease duration (DD). AOPP, MDA values and SOD activity were significantly higher in both study patients than in the control group (p b 0.05). While AOPP and MDA approached higher values in RRMS, compared to the CIS group (p N 0.05, p b 0.05, respectively), SOD activity showed higher values in CIS than in RRMS patients (p b 0.05). Both study patients with higher EDSS, higher number of total radiological lesions and longer DD, had higher AOPP and MDA content (p b 0.05, p N 0.05). SOD activity was lower in both study patients with higher EDSS, higher number of total radiological lesions and longer DD (p b 0.05, p N 0.05). There were positive correlations between AOPP and DD and EDSS in CIS patients (p b 0.01), and MDA levels and DD, EDSS and Gd + in CIS, as well as with EDSS in RRMS patients (p b 0.01). There were negative correlations between SOD activity and DD and EDSS in both study patients (p b 0.01), as well as, between SOD activity and Gd+ in CIS patients (p b 0.01). The measured erythrocytes' biomarkers might represent one of the important biomarkers for the evaluation of the oxidative status of neuroinflammation and disease severity, especially in its early phase, defined as CIS. © 2013 Elsevier B.V. All rights reserved.

1. Introduction An imbalance in the oxidative/antioxidative status leads to oxidative stress, which is revealed as the main contributor involved in a number of diseases including neuroinflammation and its mediated disease such as multiple sclerosis (MS) [1]. The detrimental effects of oxidative stress are caused by reactive oxygen species (ROS), the inactivation and removal of which depend on the antioxidative defense system, which includes vitamins A, E, and C, beta-carotene, reduced glutathione (GSH), and several antioxidative enzymes, such as superoxide dismutase (SOD) [2]. Erythrocytes, the most abundant cells in blood, are constantly exposed to ROS produced in neuroinflammation during blood circulation. They may be important in regulating oxidant reactions thereby preventing ROS mediated CNS toxicity [3]. Although most ⁎ Corresponding author at: Faculty of Medicine, University of Nis, Bul. Dr Zorana Djindjica 81, 18000 Nis, Serbia. Tel.: +381 646727222; fax: +381 4570029. E-mail address: [email protected] (S. Ljubisavljevic). 0022-510X/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jns.2013.11.006

studies have shown the different changes of the redox ratio of erythrocytes as a significant parameter for oxidative stress as well as aging [4], and also in some autoimmune diseases [5], rare are those which correlate the named differences with disease pathogenesis i.e. with clinical and paraclinical features by way of different disease stages. Although MS is defined as an immune mediated disease of CNS, its pathogenesis is not fully understood yet. It has been considered that neuroinflammation is a great contributor of pathogenesis prevailing in the earliest phase of a disease characterized by demyelination, while at a later stage what prevails is mostly oxidative stress which mediates irreversible and neurodegenerative injuries of CNS. The respective roles of overlapping inflammatory and oxidative processes in MS pathogenesis vary over the course of the disease [6]. Most MS patients (85%) experience a relapsing–remitting (RR) course of the disease characterized by a relapse followed by a recovery period. Within years most of them evolve into a secondary progressive (SP) phase characterized by a steady increase in disability. For the most part, MS appears as a clinically isolated syndrome of CNS (CIS) which will be developed in defined MS

S. Ljubisavljevic et al. / Journal of the Neurological Sciences 337 (2014) 8–13

9

within the time. Only about 15% experience a primary progressive (PP) form defined by the accumulation of disability from the onset of the disease [7]. The present study has been done in order to clarify the possible correlations between imbalances in oxidant and antioxidant defense in erythrocytes, and clinical and paraclinical presentations of neuroinflammatory acute attacks, which were defined as the clinically isolated syndrome of CNS (CIS) and RRMS.

Table 1 Demographic data and basic hematologic parameters.

2. Patients and methods

Values are presented as medians (range) and means ± SD of healthy subjects, and CIS and RRMS patients. In all statistics calculations Mann Whitney and Kruskal–Wallis tests were done. a p b 0.05 CIS and RRMS vs CG. b p b 0.05 RRMS vs CIS.

This study was performed as a cross-sectional study. The study was approved by the Ethical Committee of the Faculty of Medicine, University of Nis and informed consent was obtained from each patient prior to entry into the study, according to the Declaration of Helsinki.

CG

CIS

RRMS

Number—female/male Age (years) Disease duration (months)

69–33/36 37 (23–55) /

50–35/15 37.5 (17–57) 2 (1–3)

57–45/12 40 (23–58) 84 (1–396)

Er (×1012/L) Hemoglobin (g/dL) Hematocrit (%)

4.79 ± 0.9 14 ± 0.85 42.3 ± 2.1

4.05 ± 0.75a 13.4 ± 1.04a 39.5 ± 4.3

3.55 ± 0.8a,b 12.7 ± 1.15a,b 36.5 ± 6.1a

2.4. Clinical assessment 2.1. Control patients Sixty nine (36 male, 33 female) patients, aged 23–55 years and nonsmokers, were involved in the control group (CG). They were admitted at the Clinic for Neurology, Clinical Center Nis, and underwent the complete diagnostic procedure for suspected demyelinating disease, without any objective abnormalities found at laboratory, MRI scan and CSF examination.

All CIS and RRMS patients' clinical presentations were assessed using Kurtzke's extended disability status scale (EDSS) [11]. Based upon the frequency distribution of EDSS all patients were divided by the median value for EDSS into those which had mild (≤3 for CIS and ≤5 for RRMS patients) and severe (N3 for CIS and N5 for RRMS patients) clinical disability [8]. 2.5. MRI assessment

2.2. CIS patients Fifty patients (15 male, 35 female), aged 17–57 years, admitted at the Clinic for Neurology, Clinical Center Nis, presented with an acute or sub-acute attack affecting different brain structures (cerebral hemispheres, brainstem, cerebellum, spinal cord, optic nerve or more than one functional system), which have been suggestive of MS. Neurological findings in details were prescribed in our recent published paper [8]. All clinical, laboratory and MRI investigations were performed less than 3 months after initial neurological presentations. Inclusion in the study was based only on the clinical expression and was not influenced by MRI. Due to exclusion of alternative diagnoses by the appropriate investigations performed, and since they didn't fulfill diagnostic criteria for MS [9], and had no previous history of possible demyelinating events, patients' condition was defined as a CIS. 2.3. RRMS patients Fifty seven patients (12 male, 45 female), aged 23–58 years, admitted at the Clinic for Neurology, Clinical Center Nis, and presenting with clinically definite MS as two separate attacks disseminated in time and place followed by clinical evidence of two separate lesions [9], were involved in the RRMS group. In all MS patients the disease was verified by clinical, laboratory and neuroimaging approaches. Neurological findings in details were prescribed in our recent published paper [8]. The interval between the previous and the present clinical episode was longer than 6 months. All of the MS defined patients were classified as having a relapsing–remitting (RR) form of MS, according to the Lublin and Reingold criteria [8,10]. The patients with other MS forms or any disease other than MS that would compromise organ function were excluded from the study, as were the patients who had received prior immunosuppressant, interferon or corticosteroid therapy within 6 months of the study entry. All of the patients in both groups, CIS and RRMS, have been non-smokers for at least one year. All CIS and MS patients were categorized according to their age, gender, duration of neurological onset at the admission at the Clinic, disease duration in RRMS patients, frequency of relapses in RRMS patients, affected CNS structure, MRI characteristics and frequency distribution on the clinical scales. Data are presented in Table 1.

Brain MRI was performed using a 1.5 T system (Avanto, Siemens, Erlangen, Germany). MRI protocol included the following conventional spin echo sequences: axial T1-weighted (repetition time [TR] = 500 ms, echo time [TE] = 78 ms, number of excitations [NEX] = 2) and T2-weighted (TR = 4700 ms, TE = 93 ms, NEX = 2) with 5-mm slice thickness and an intersection gap of 0.5 mm. The pixel size was 0.9 × 0.9 mm. Intravenous gadolinium contrast (Gadovist, Schering, Berlin, Germany) was administered in a dose of 0.1 mmol/kg of body weight. The number of hyperintense lesions seen on T2 images and the lesion load of Gd-enhancing lesions seen on T1-weighted images were calculated. The lesion loads were calculated as volumes. All MRI scans were reported by an experienced neuroradiologist unaware of the clinical findings and the classification of the patients. Considering total T2-weighted lesions (MRItl) all study patients were divided by the median value into those which had mild (≤ 9 for CIS and ≤ 40 for RRMS patients) and severe (N9 for CIS and N40 for RRMS patients) MRI changes [8]. 2.6. Biochemical assessment 2.6.1. Blood sampling Venous blood samples were collected into Venoject tubes with EDTA (0.47 mol/L K3-EDTA), from all study subjects after fasting for a duration of at least 12 h. Within 1 h after sampling, the blood samples were centrifuged at 3000 g for 10 min at 4 °C to separate plasma and erythrocytes. The buffy coat was removed and the remaining erythrocytes were drawn from the bottom, washed three times in cold saline (9.0 g/L NaCl), and hemolyzed by adding the 9 fold equivalent weight of ice-cold demineralized ultrapure water to yield a 10% hemolysate. The hemolysates were stored in a refrigerator at 4 °C for 15 min and erythrocyte membranes were removed and then hemolysates were frozen in aliquots at −80 °C for later analysis. 2.6.2. Determination of hematological parameters Erythrocyte count, and hematocrit (Hct), hemoglobin (Hb) and other hematological indexes were analyzed by an electronic hospital auto-analyzer, the automated Cell Counter, MEK-4100, Nihon Kohden, Japan. The hemoglobin concentration in lysates was determined with the aid of Drabkin's reagent [12].

10

S. Ljubisavljevic et al. / Journal of the Neurological Sciences 337 (2014) 8–13

2.6.3. Malondialdehyde measuring Concentration of MDA was determined by modified thiobarbituric acid (TBA) with the end product of lipid peroxidation— malondialdehyde [13]. The level of lipid peroxidation was measured in lysates of erythrocytes. The results are expressed as μmol MDA/g Hb.

2.6.4. Superoxide dismutase activity Superoxide dismutase activity in erythrocytes' lysates was measured according to method of Misra and Fridovich [14]. This method is based on the reaction autooxidation of adrenaline to adenochrome. The intermediate in this reaction is superoxide, which is scavenged by SOD. One SOD unit was defined as the enzyme amount causing 50% inhibition of the autooxidation of adrenaline. The activity was expressed as U/g Hb.

2.6.5. Advanced oxidation protein products Advanced oxidation protein products (AOPP), were determined in hemolysates mixed with H2O, acetic acid and potassium iodide. The absorbance of the reaction mixture was immediately recorded at 340 nm. The data were expressed as μm L−1 of chloramine equivalents and related lysate protein [15].

2.6.6. Chemicals Chemicals were purchased from Sigma (St. Louis, MO, USA). All used chemicals were of analytical grade. All drug solutions were prepared on the day of the experiment.

2.6.7. Statistics All statistical calculations were performed using appropriated nonparametric tests after verification of values distribution in each group. All comparisons between subgroups (in each study group and between both of them), divided based on gender, age, disease duration, relapse frequencies, EDSS score, and MRI findings (volume of Gdenhancement lesions and number of T2-weighted lesions), were performed using the Mann Whitney and Kruskal–Wallis tests, when appropriate. Spearman's rank correlation coefficient analysis was used to investigate the relationship between two comparable variables (AOPP, MDA levels, SOD activity and clinical, radiological and other patient's features), also, the linear regression analysis was used to verify the level of the examined relationship. All data are presented as medians with range throughout the text, or when it was appropriate as means ± SD. The p b 0.05 was considered as significant. All statistical calculations were done using “SPSS 13.0 for Windows” (SPSS Inc., USA).

3. Results The demographic and basic clinical characteristics in details are prescribed in our recent published paper [8]. Now, data are given in a brief version in Table 1. Erythrocyte count was decreased in both study patients, CIS and RRMS, compared to the values obtained in the control group subjects, p = 0.03 and p = 0.009, respectively, but decreasing was more prominent in RRMS than in CIS patients, p = 0.0018. The hemoglobin concentrations were also decreasing in CIS and RRMS patients compared to the control values, p = 0.041 and p = 0.045, respectively, with the same trend comparing hematocrit values between RRMS and control subjects, p = 0.022 (Table 1). 3.1. Erythrocytes' AOPP values in study patients Erythrocytes' AOPP values in the control subjects and study patients, divided regarding their EDSS, MRI characteristics and disease duration, are presented in Table 2. AOPP values in CIS and RRMS patients were significantly higher than those obtained in the control subjects (p = 0.001 and p = 0.0009, respectively). Comparing the total AOPP content in examined erythrocytes between CIS and RRMS patients, no statistical significance was observed (p = 0.12). In both CIS and RRMS groups, the patients with higher EDSS had higher AOPP content in erythrocytes, but these differences were significant only in CIS patients (p = 0.028 and p = 0.073, for CIS and RRMS, respectively). Regarding the total number of T2 weighted lesions all CIS and RRMS patients were divided into two subgroups based on the number of lesions related to its median (range) values (9 lesions for CIS and 40 lesion for RRMS) [8]. AOPP values were significantly higher in both study patients without observed significances (p = 0.071 for CIS, and p = 0.08, for RRMS patients). There were also differences in AOPP values in both CIS and RRMS patients, when they were divided based upon median range of disease duration (2 months for CIS and 84 months for RRMS patients). CIS patients with shorter disease duration showed lower AOPP values compared to those with longer disease duration (p = 0.013); the significance was not observed in RRMS patients (p = 0.081). 3.2. Erythrocytes' MDA values in study patients MDA values in erythrocytes in the control subjects and both study group patients, divided regarding their EDSS and MRI characteristics and disease duration, are presented in Table 3. The obtained MDA values in the CIS and RRMS patients were significantly higher than those obtained in the control subjects (p = 0.0012 and p = 0.0008, respectively). Generally, MDA levels were higher in RRMS than in CIS patients (p = 0.039). In both CIS and RRMS groups, the patients with higher EDSS had higher MDA levels in erythrocytes (p = 0.003 and

Table 2 The erythrocyte values of AOPP (μmol/g Hb in lysates) regarding EDSS, MRI and disease duration. CGb

Er AOPP/N

0.54 ± 0.08/69

Er AOPP/N

0.54 ± 0.08/69

Er AOPP/N

0.54 ± 0.08/69

CIS

RRMS

Total

EDSS ≤ 3

EDSS N 3

Total

0.98 ± 0.15/50a

0.85 ± 0.07/26d

1.21 ± 0.15/24c,d

1.02 ± 0.11/57a

Total

MRItl ≤ 9

MRItl N 9

Total

a

0.96 ± 0.09/16

0.92 ± 0.07/8

1.01 ± 0.1/8

Total

DD ≤ 2

DD N 2

0.98 ± 0.15/50a

0.9 ± 0.07/28c,d

1.1 ± 0.09/22c

EDSS ≤ 5 1 ± 0.09/30

EDSS N 5 1.07 ± 0.12/27

MRItl ≤ 40

MRItl N 40

1.02 ± 0.1/15

0.98 ± 0.11/8

1.06 ± 0.1/7

Total

DD ≤ 84

a

1.02 ± 0.11/57a

1 ± 0.1/32

DD N 84 1.05 ± 0.09/25

Values are medians (range) or means ± SD/number of patients. Clinical data were available for all CIS and RRMS patients, measuring expanded disability status scale (EDSS) in each patient. MRI brain scans data were available for 16 CIS and 15 RRMS patients. Mann Whitney and Kruskal–Wallis tests were done. a p b 0.05 CIS and/or RRMS vs CG. b p b 0.05 RRMS vs CIS. c p b 0.05 CIS (EDSS ≤ 3) vs CIS (EDSS N 3); CIS (MRItl ≤ 9) vs CIS (MRItl N9); CIS (DD ≤ 2) vs RRMS (DD N 2); RRMS (EDSS ≤ 5) vs RRMS (EDSS N 5); RRMS (MRTtl ≤ 40) vs RRMS (MRItl N 40); RRMS (DD ≤ 84) vs RRMS (DD N 84). d p b 0.05 between all study subgroups.

S. Ljubisavljevic et al. / Journal of the Neurological Sciences 337 (2014) 8–13

11

Table 3 The erythrocyte values of MDA (μmol/g Hb in lysates) regarding EDSS, MRI and disease duration. CG

CIS

RRMS EDSS ≤ 3

Total Er MDA/N

0.25 ± 0.08/69

Er MDA/N

0.25 ± 0.08/69

Er MDA/N

0.25 ± 0.08/69

a

0.62 ± 0.13/50

EDSS N 3 d

0.51 ± 0.1/26

EDSS ≤ 5

Total c,d

0.82 ± 0.19/24

a,b

0.79 ± 0.12/57

EDSS N 5 d

0.75 ± 0.09/30

0.89 ± 0.14/27 c,d

Total

MRItl ≤ 9

MRItl N 9

Total

MRItl ≤ 40

MRItl N 40

0.76 ± 0.18/16a

0.62 ± 0.11/8d

0.89 ± 0.14/8c,d

0.75 ± 0.11/15a

0.73 ± 0.14/8

0.77 ± 0.12/7

Total

DD ≤ 2

DD N 2

Total

DD ≤ 84

a

0.62 ± 0.13/50

d

c

0.6 ± 0.01/28

0.75 ± 0.13/22

a,b

0.75 ± 0.12/57

DD N 84 d

0.75 ± 0.12/32

0.76 ± 0.1/25

Values are medians (range) or means ± SD/number of patients. Clinical data were available for all CIS and RRMS patients, measuring expanded disability status scale (EDSS) in each patient. MRI brain scan data were available for 16 CIS and 15 RRMS patients. Mann Whitney and Kruskal–Wallis tests were done. a p b 0.05 CIS and/or RRMS vs CG. b p b 0.05 RRMS vs CIS. c p b 0.05 CIS (EDSS ≤ 3) vs CIS (EDSS N 3); CIS (MRItl ≤ 9) vs CIS (MRItl N9); CIS (DD ≤ 2) vs RRMS (DD N 2); RRMS (EDSS ≤ 5) vs RRMS (EDSS N 5); RRMS (MRTtl ≤ 40) vs RRMS (MRItl N 40); RRMS (DD ≤ 84) vs RRMS (DD N 84). d p b 0.05 between all study subgroups.

p = 0.065). MDA values were generally higher in all study patients who had a higher number of total T2-weighted lesions (p = 0.025 for CIS and p = 0.12, for RRMS patients). Also, MDA values were higher in erythrocytes obtained from patients with longer disease duration (p = 0.037 and p = 0.15 for CIS and RRMS patients, respectively).

3.3. Erythrocytes' SOD activity in study patients The values of the obtained SOD activities in erythrocytes are presented in Table 4. Erythrocytes' SOD activity in all study patients was significantly higher than in the control subjects (p = 0.002 for CIS, and p = 0.009 for RRMS patients). Differences in SOD activity between CIS and RRMS patients were significant (p = 0.02). In both CIS and RRMS groups, the patients with higher EDSS had lower SOD activity in erythrocytes (p = 0.001 and p = 0.023), respectively. Regarding the total number of T2 weighted lesions, SOD activity was higher in CIS and RRMS patients with lower, than in those with higher number of T2 weighted lesions, but these differences were significant only for erythrocytes' SOD activity in CIS patients (p = 0.009), not for RRMS patients (p = 0.061). There was higher enzyme activity in all patients with shorter disease duration (p = 0.0005 and p = 0.002), for CIS and RRMS patients, respectively. Erythrocyte oxidative stress biomarkers (AOPP and MDA) were in positive correlation with those obtained in plasma, while SOD activity in erythrocytes was in positive correlation regarding plasma activities, but in negative correlation regarding CSF obtained values (p b 0.05).

3.4. Erythrocytes' AOPP, MDA and SOD activity correlations with clinical, radiological and other patients' characteristics Correlation between erythrocytes' AOPP and MDA content and SOD activity, in both study groups, and their EDSS, MRI and disease duration are presented in Table 5. The obtained results significantly verify that there were positive correlations between AOPP and MDA levels and a negative correlation between SOD activity and clinical severity (EDSS) in CIS patients, r = 0.47 (p = 0.003), r = 0.39 (p = 0.007) and r = − 0.61 (p = 0.004), respectively. The same correlation was observed in RRMS patients but only between MDA values and SOD activity, r = 0.42 (p = 0.005) and r = −0.52 (p = 0.007), respectively, not for AOPP values, r = 0.14 (p = 0.083). Erythrocytes' MDA content in CIS patients showed positive correlation with Gd + enhancement lesion volume, r = 0.37 (p = 0.003) without a similar trend in RRMS patients r = 0.18 (p = 0.079). SOD activity in erythrocytes, showed negative correlation with named radiological features only in CIS, r = − 0.42 (p = 0.004), not in RRMS patients, r = − 0.14 (p = 0.064). Also, AOPP content in erythrocytes did not show significant correlation with the above mentioned radiological features, r = 0.18 (p = 0.067) and r = 0.15 (p = 0.072), for CIS and RRMS patients, respectively. In CIS patients, all examined biomarkers, AOPP, MDA and SOD activity, showed correlation with disease duration, r = 0.41 (p = 0.0025), r = −0.47 (p = 0.008) and r = − 0.53 (p = 0.009), respectively. In RRMS patients, SOD activity showed correlation with disease duration, r = − 0.41 (p = 0.009), without similar correlation regarding AOPP and MDA values and RRMS patients' disease duration, r = 0.12

Table 4 The erythrocyte values of SOD activity (U/g Hb in lysates) regarding EDSS, MRI and disease duration. CG

Er SOD/N

1407 ± 250/69

Er SOD/N

1407 ± 250/69

Er SOD/N

1407 ± 250/69

CIS

RRMS

Total

EDSS ≤ 3

EDSS N 3

Total

EDSS ≤ 5

EDSS N 5

2118 ± 357/50a

2431 ± 350/26d

1572 ± 266/24c,d

1887 ± 410/57a,b

2042 ± 312/30d

1777 ± 265/27 c,d

Total

MRItl ≤ 9

MRItl N 9

Total

MRItl ≤ 40

MRItl N 40

1902 ± 211/8

1799 ± 195/7

a

c,d

a

1789 ± 281/16

1979 ± 212/8

1600 ± 208/8

1865 ± 214/15

Total

DD ≤ 2

DD N 2

Total

DD ≤ 84

DD N 84

2118 ± 357/50a

2409 ± 320/28c

1560 ± 350/22c,d

1887 ± 410/57a,b

2121 ± 234/32c,d

1745 ± 204/25d

Values are medians (range) or means ± SD/number of patients. Clinical data were available for all CIS and RRMS patients, measuring expanded disability status scale (EDSS) in each patient. MRI brain scan data were available for 16 CIS and 15 RRMS patients. Mann Whitney and Kruskal–Wallis tests were done. a p b 0.05 CIS and/or RRMS vs CG. b p b 0.05 RRMS vs CIS. c p b 0.05 CIS (EDSS ≤ 3) vs CIS (EDSS N 3); CIS (MRItl ≤ 9) vs CIS (MRItl N9); CIS (DD ≤ 2) vs RRMS (DD N 2); RRMS (EDSS ≤ 5) vs RRMS (EDSS N 5); RRMS (MRTtl ≤ 40) vs RRMS (MRItl N 40); RRMS (DD ≤ 84) vs RRMS (DD N 84). d p b 0.05 between all study subgroups.

12

S. Ljubisavljevic et al. / Journal of the Neurological Sciences 337 (2014) 8–13

Table 5 Correlation of examined biomarkers and disease duration, EDSS and MRI findings. CIS

Er AOPP Er MDA Er SOD

RRMS

Disease duration

EDSS

Gd+ enhancement lesion volume

Disease duration

EDSS

Gd+ enhancement lesion volume

r = 0.41; p = 0.0025⁎ r = 0.47; p = 0.008⁎ r = −0.53; p = 0.009⁎

r = 0.47; p = 0.003⁎ r = 0.39; p = 0.007⁎ r = −0.61; p = 0.004⁎

r = 0.18; p = 0.067 r = 0.37; p = 0.003⁎ r = −0.42; p = 0.004⁎

r = 0.12; p = 0.085 r = 0.12; p = 0.079 r = −0.41; p = 0.009⁎

r = 0.14; p = 0.083 r = 0.42; p = 0.005⁎ r = −0.52; p = 0.007⁎

r = 0.15; p = 0.072 r = 0.18; p = 0.079 r = −0.14; p = 0.064

Spearman's rank correlation and linear regression analyses were performed. ⁎ Significant.

(p = 0.085) and r = 0.12 (p = 0.079), respectively. Data are presented in Table 5. There was a significant positive correlation between EDSS and T2 weighted lesion number in both study groups, r = 0.41 (p = 0.009), in the CIS, and r = 0.52 (p = 0.007), in the RRMS group. Also, EDSS showed strong positive correlation with disease duration, r = 0.51 (p = 0.002), in CIS, and r = 0.61 (p = 0.001), in RRMS patients. The positive correlations were also observed regarding T2 weighted lesion numbers in both CIS and RRMS groups and their disease duration, r = 0.48 (p = 0.008) and r = 0.68 (p = 0.003), respectively. We did not find a significant association between other CIS and RRMS patients' characteristics and their EDSS, MRI, disease duration and other features (data not shown). Also, there were no overall differences in AOPP and MDA levels, and SOD activity in erythrocytes, in both study groups, when they were divided based on other different criteria (age, gender, relapse frequency), and these data are not shown. 4. Discussion The results of the present study showed that oxidative stress was presented in erythrocytes of both examined clinical phenotypes of neuroinflammation. Similarly, the associations of oxidative stress and inflammation in a close correlation with redox dysfunction in erythrocyte have been reported in some proinflammatory conditions [16]. These data allow us to suppose that the development of oxidative stress under neuroinflammation is a result of a strong decrease in antioxidant defense in erythrocytes during disease development (Tables 2, 3, 4). It has been demonstrated that erythrocytes of MS patients are less stable against lysis with the conjunct for the development of the disease [17], and, that erythrocytes' membrane fluidity defects are accompanied with a strong inverse correlation with EDSS and closely interrelated with inflammation intensity [18,19] and lipid peroxidation process [20–22]. Consistent with our previous experimental and clinical findings [8,23], the data presented here demonstrate that CIS and RRMS patients had intensive erythrocyte lipid peroxidation with compensation by antioxidant mechanisms. The erythrocyte membrane is prone to lipid peroxidation under oxidative stress leading to formation of MDA. There is a study showing similar changes in the oxidative status in erythrocyte hemolysates obtained from patients with other autoimmune diseases, which suggests a significant increase in the level of lipid peroxidation, measured as MDA [5], as we found here. In the CNS, this prooxidative state is accompanied with an enhancement in the demyelization process leading to a severe degree of neurological injury [20-22,24,25], due to lipid peroxidation mediated chemical cross-linking and aggregation which leads to the formation of new compounds and modified structures, among them advanced oxidation protein products (AOPP) [26-28]. AOPP reflects an excess of ROS generation and measures of general protein damages [29]. An increase in AOPP level in erythrocytes, as well as in plasma and CSF [8], of our study patients, underlies the importance of the protein conformational changes in the pathogenesis of neuroinflammation. Recently, an increased level of protein modified products, such as protein carbonyls, in erythrocyte membrane during

oxidative stress condition has been shown [30]. Some previous findings [3,31] indicate that redox processes are involved in all stages of MS, particularly in the initiation of disease, which might be supported by the direct relation of MDA and AOPP erythrocyte values and EDSS and DD, observed only in CIS patients, which is mainly characteristic of neuroinflammation, but not in RRMS patients where neurodegeneration predominates. Collectively, these facts advocate that severity of disease might be enhanced by an imbalance between prooxidative and oxidative stresses in erythrocytes, especially in the earliest phase of CNS inflammation. In the present study, a marked reduction in the activity of erythrocyte SOD was observed in both study groups (Table 4). These data are similar with those previously published [28,32]. An enhancement in SOD gene expression in acute MS attacks has been demonstrated [33,34]. The SOD induction, more pronounced in CIS patients, might appear as a consequence of direct ROS effects, as an adaptive phenomenon, while, on the contrary, in RRMS, SOD showed a decrease in activity probably due to the irreversible inactivation by prolonged oxidative stress and its waste product [26]. As we found here, the correlation between SOD activity and EDSS, was also observed in a study designed to assess the named direction in SPMS patients [25]. Although, there are also results which have found a significant reduction of SOD-1 activity in erythrocytes of MS patients [35], contrary to data presented here, the enzyme activity was changed in relation to disease duration (Table 5). It has been demonstrated that the dose/concentration- and time-dependent effects of some antioxidants may provide erythrocytes protection against degenerative diseases, by increasing their antioxidative capacity [36], inducing an adaptive increase in the erythrocyte SOD activity [37]. Although there are no clinical studies investigating correlations between erythrocytes' MDA, AOPP and SOD values and MRI patients' findings, which might be used for those comparisons, authors are aware that correlations obtained here (Table 5), might be the consequence of a relatively small number of available MRI brain scans that have been analyzed in the present study. As ROS are increased under neuroinflammation and are constantly present in blood [8], it is likely that erythrocytes in circulation, may be more prone to being modified which may enhance their antioxidative activity , this way promoting systemic and CNS inflammatory process. In conclusion, erythrocytes possess a variety of antioxidative mechanisms which is extremely important to clearly and preciously learn more about redox regulation and oxidative stress in neuroinflammation. Since the measurement of biomarkers presented here is easy and fast to perform, and the results are very close to those obtained in plasma and CSF [8], they might represent one of the important biomarkers for the evaluation of the oxidative status of neuroinflammation and, based on data presented here, they are biomarkers of disease severity, especially in its early phase, defined as CIS. Investigations based on larger and different cohorts studied over longer periods of time, and also including more clinical phenotypes of neuroinflammation, are needed in order to precisely assess antioxidant approach adequacy, which might be given additionally with conventionally recommended treatments.

S. Ljubisavljevic et al. / Journal of the Neurological Sciences 337 (2014) 8–13

Conflict of interest No conflict of interest exists for any of the authors listed in the article. Acknowledgement This work was supported by the grant from the scientific project number 41018 financed by the Ministry of Education and Science, Republic of Serbia. We wish to thank our patients for blood donation and their agreement to participate in this study. References [1] Fiorini A, Koudriavtseva T, Bucaj E, Coccia R, Foppoli C, Giorgi A, et al. Involvement of oxidative stress in occurrence of relapses in multiple sclerosis: the spectrum of oxidatively modified serum proteins detected by proteomics and redox proteomics analysis. PLoS One 2013;8(6):e65184. [2] Sarsour EH, Louise Kalen A, Goswami P. Manganese superoxide dismutase regulates a redox cycle within the cell cycle. Antioxid Redox Signal 2013. http://dx.doi.org/ 10.1089/ars.2013.5303. [3] Xiang W, Weisbach V, Sticht H, Seebahn A, Bussmann J, Zimmermann R, et al. Oxidative stress-induced posttranslational modifications of human hemoglobin in erythrocytes. Arch Biochem Biophys 2013;529:34–44. [4] Rizvi SI, Maurya PK. Alterations in antioxidant enzymes during aging in humans. Mol Biotechnol 2007;37:58–61. [5] Shah D, Kiran R, Wanchu A, Bhatnagar A. Oxidative stress in systemic lupus erythematosus: relationship to Th1 cytokine and disease activity. Immunol Lett 2010;129:7–12. [6] Gonsette RE. Oxidative stress and excitotoxicity: a therapeutic issue in multiple sclerosis? Mult Scler 2008;14:22–34. [7] Inglese M. Multiple sclerosis: new insights and trends. AJNR Am J Neuroradiol 2006;27:954–7. [8] Ljubisavljevic S, Stojanovic I, Vojinovic S, Stojanov D, Stojanovic S, Cvetkovic T, et al. The patients with clinically isolated syndrome and relapsing remitting multiples sclerosis show different levels of advanced protein oxidation products and reduced thiols content in sera and CSF. Neurochem Int 2013;62(7):988–97. [9] Polman CH, Reingold SC, Banwell B, Clanet M, Cohen JA, Filippi M, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol 2011;69(2):292–302. [10] Lublin FD, Reingold SC. Defining the clinical course of multiple sclerosis: results of an international survey—National Multiple Sclerosis Society (USA) Advisory Committee on Clinical Trials of New Agents in Multiple Sclerosis. Neurology 1996;46:907–11. [11] Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983;33:1444–52. [12] Drabkin DL. The crystallographic and optical properties of the haemoglobin of man in comparison with those of other species. J Biol Chem 1946;164:703–23. [13] Jain SK, Levine NS, Duett J, Hollier B. Elevated lipid peroxidation levels in red blood cells of streptozotocin-treated diabetic rats. Metabolism 1990;39(9):971–5. [14] Misra HP, Fridovich J. The role of superoxide anion in the autoxidation of epinephrine and a simple assay superoxide dismutase. J Biol Chem 1972;247:3170–5. [15] Witko-Sarsat V, Friedlander M, Capeillere-Blandin C, Nguyen-Khoa T, Nguyen AT, Zingraff J, et al. Advanced oxidation protein products as a novel marker of oxidative stress in uremia. Kidney Int 1996;49:1304–13. [16] Tsuda K. Associations of oxidative stress and inflammation and their role in the regulation of membrane fluidity of red blood cells in hypertensive and normotensive men: an electron spin resonance investigation. Adv Biosci Biotechnol 2012;3:1020–7.

13

[17] Hon GM, Hassan MS, van Rensburg SJ, Abel S, Marais DW, van Jaarsveld P, et al. Erythrocyte membrane fatty acids in patients with multiple sclerosis. Mult Scler 2009;15(6):759–62. [18] de Freitas MV, de Oliveira MR, dos Santos DF, de Cássia Mascarenhas Netto R, Fenelon SB, Penha-Silva N. Influence of the use of statin on the stability of erythrocyte membranes in multiple sclerosis. J Membr Biol 2010;233(1–3):127–34. [19] Koch M, Ramsaransing GS, Fokkema MR, Heersema DJ, De Keyser J. Erythrocyte membrane fatty acids in benign and progressive forms of multiple sclerosis. J Neurol Sci 2006;244(1–2):123–6. [20] Koch M, Mostert J, Arutjunyan AV, Stepanov M, Teelken A, Heersema D, et al. Plasma lipid peroxidation and progression of disability in multiple sclerosis. Eur J Neurol 2007;14(5):529–33. [21] Ferretti G, Bacchetti T. Peroxidation of lipoproteins in multiple sclerosis. J Neurol Sci 2011;311(1–2):92–7. [22] Besler HT, Comoglu SS. Lipoprotein oxidation, plasma total antioxidant capacity and homocysteine level in patients with multiple sclerosis. Nutr Neurosci 2003;6(3):189–96. [23] Ljubisavljevic S, Stojanovic I, Pavlovic D, Milojkovic M, Sokolovic D, Stevanovic I, et al. Suppression of the lipid peroxidation process in the CNS reduces neurological expression of experimentally induced autoimmune encephalomyelitis. Folia Neuropathol 2013;51(1):51–7. [24] Pasichna EP, Morozova RP, Donchenko HV, Vinychuk SM, Kopchak OO. Lipid peroxidation and antioxidant defence enzyme activity in multiple sclerosis. Ukr Biokhim Zh 2007;79(5):165–74. [25] Miler E, Walczak A, Majsterek I, Kędziora J. Melatonin reduces oxidative stress in the erythrocytes of multiple sclerosis patients with secondary progressive clinical course. J Neuroimmunol 2013. http://dx.doi.org/10.1016/j.jneuroim.2013.02.012. [26] Erden Inal M, Kanbak G, Sunal E. Antioxidant enzyme activities and malondialdehyde levels related to aging. Clin Chim Acta 2001;305:75–80. [27] Vani R, Shiva CS, Devi SA. Oxidative stress in erythrocytes: a study on the effect of antioxidant mixtures during intermittent exposures to high altitude. Int J Biometeorol 2002;54(5):553–62. [28] Acar A, Ugur Cevik M, Evliyaoglu O, Uzar E, Tamam Y, Arıkanoglu A, et al. Evaluation of serum oxidant/antioxidant balance in multiple sclerosis. Acta Neurol Belg 2012;112(3):275–80. [29] Gallan PM, Carrascosa A, Gussinye M, Dominguez C. Biomarkers of diabetes associated oxidative stress and antioxidant status in young diabetic patients with or without subclinical complication. Free Radic Biol Med 2003;34:1563–74. [30] Pandey KB, Rizvi SI. Protective effect of resveratrol on formation of membrane protein carbonyls and lipid peroxidation in erythrocytes subjected to oxidative stress. Appl Physiol Nutr Metab 2009;34:1093–7. [31] Oliveira SR, Kallaur AP, Simao ANC, Kaminami Morimoto H, Lopez J, Panis C, et al. Oxidative stress in multiple sclerosis patients in clinical remission: association with the expanded disability status scale. J Neurol Sci 2012;321(1–2):49–53. [32] Tasset I, Agüera E, Sánchez-López F, Feijóo M, Giraldo AI, Cruz AH, et al. Peripheral oxidative stress in relapsing–remitting multiple sclerosis. Clin Biochem 2012;45(6):440–4. [33] Tajouri L, Mellick AS, Ashton KJ, Tannenberg AEG, Nagra RM, Tourtellotte WW, et al. Quantitative and qualitative changes in gene expression patterns characterize the activity of plaques in multiple sclerosis. Mol Brain Res 2003;119(2):170–83. [34] van Horssen J, Schreibelt G, Drexhage J, Hazes T, Dijkstra CD, van der Valk P, et al. Severe oxidative damage in multiple sclerosis lesions coincides with enhanced antioxidant enzyme expression. Free Radic Biol Med 2008;45(12):1729–37 [15]. [35] Zagórski T, Dudek I, Berkan L, Mazurek M, Kedziora J, Chmielewski H. Superoxide dismutase (SOD-1) activity in erythrocytes of patients with multiple sclerosis. Neurol Neurochir Pol 1991;25(6):725–30. [36] Pandey KB, Rizvi SI. Protective effect of resveratrol on markers of oxidative stress in human erythrocytes subjected to in vitro oxidative insult. Phytother Res 2010;24(S1):S11–4. [37] Rowiński R, Kozakiewicz M, Kędziora-Kornatowska K, Hübner-Woźniak E, Kędziora J. Markers of oxidative stress and erythrocyte antioxidant enzyme activity in older men and women with differing physical activity. Exp Gerontol 2013;48:1141–6.