E-Mail karger@karger.com

Original Paper

Neuroimmunomodulation 2014;21:226–233 DOI: 10.1159/000357002

Decreased Level of sRAGE in the Cerebrospinal Fluid of Multiple Sclerosis Patients at Clinical Onset

Anton Glasnović   b Hrvoje Cvija   a Maristela Stojić   b Ivana Tudorić-Đeno   c Sanja Ivčević   a Dominik Romić   d Nino Tičinović   e Vladimira Vuletić   b Ines Lazibat   b Danka Grčević   a

a   Department of Physiology and Immunology, University of Zagreb School of Medicine, and b   Department of Neurology, c   Department of Anesthesiology, Reanimatology and Intensive Care, d   Department of Neurosurgery and e   Clinical Department of Diagnostic and Interventional Radiology, Clinical Hospital ‘Dubrava’, Zagreb , Croatia

tients. The level of sRAGE in the CSF of MS patients was low-er (p = 0.021), with the ability to discriminate between MS patients and control subjects. Moreover, PBMC gene expres-sion for HMGB1 and S100A12 positively correlated with IL-6. Conclusions: Our study confirmed that the cytokine net-work is disturbed in PBL and CSF at MS clinical onset. The deregulated HMGB1/RAGE axis found in our study may pres-ent an early pathogenic event in MS, proposing sRAGE as a possible novel therapeutic strategy for MS treatment.

© 2014 S. Karger AG, Basel

Introduction

Multiple sclerosis (MS) is a chronic inflammatory neu-rodegenerative disease directed towards neural antigens. Focal lymphocytic infiltration and subsequent damage of the myelin and axons lead to significant disability with deterioration of the motor, sensible, autonomic and neu-rocognitive functions [1] . The key pathogenic event is an autoimmune reaction mediated by activated myelin-re-active CD4+ T lymphocytes in the Th1-type immune re-sponse, involving the action of interleukin (IL)-2 and in-terferon (IFN)-γ [2] . In addition, recent studies suggest

Key Words

Multiple sclerosis · Receptor for advanced glycation end products · HMGB1 · Cytokines · Cerebrospinal fluid · Inflammation

Abstract

Objectives: Receptor for advanced glycation end products (RAGE) ligands/RAGE interactions have been proposed to have a pathogenic role in neuroinflammatory disorders. Our study aimed to assess changes in high-mobility group box (HMGB)1 and its receptor RAGE in peripheral blood (PBL) and cerebrospinal fluid (CSF) of patients with multiple sclerosis (MS) at the disease onset compared with control subjects. Methods: PBL and CSF were collected from control subjects (n = 30) and MS patients (n = 27) at clinical onset. Soluble RAGE (sRAGE), HMGB1, S100 calcium-binding protein A12 (S100A12), interleukin (IL)-1β and tumor necrosis factor (TNF)-α were measured in the CSF and plasma by enzyme-linked immunosorbent assay. Gene expression in PBL mono-nuclear cells (PBMCs) was detected by quantitative PCR for RAGE, HMGB1, S100A12 and several proinflammatory/im-munoregulatory cytokines. Results: We found a significantly lower expression of IL-10 (p = 0.031) in the PBMCs of MS pa-

Received: July 29, 2013 Accepted after revision: October 30, 2013 Published online: March 1, 2014

Anton Glasnović Department of Neurology, Clinical Hospital ‘Dubrava’ Avenija Gojka Šuška 6 HR–10000 Zagreb (Croatia) E-Mail aglasnov   @   yahoo.co.uk

© 2014 S. Karger AG, Basel1021–7401/14/0215–0226$39.50/0

www.karger.com/nim

Dow

nloa

ded

by:

Mon

ash

Uni

vers

ity13

0.19

4.20

.173

- 9

/29/

2014

9:0

7:07

AM

sRAGE in MS Patients Neuroimmunomodulation 2014;21:226–233DOI: 10.1159/000357002

227

that Th17 lymphocytes play an important role in the de-velopment of MS and experimental autoimmune enceph-alomyelitis (EAE) [3, 4] .

The pathogenesis of MS involves disturbed cytokine production, including upregulation of proinflammatory cytokines [IL-6, IL-17, IL-1, tumor necrosis factor (TNF)-α, IL-22 and CCL20], and downregulation of anti-inflammatory cytokines (IL-3, IL-4 and IL-10) [2–6] . Pro-inflammatory cytokines accumulate in MS lesions, sug-gesting that their increased production may impair cellu-lar defense mechanisms, facilitating demyelination [2, 7, 8] . Also, there are other proinflammatory mediators re-lated to neuroinflammation, such as high-mobility group box (HMGB)1, which is detected in active lesions of MS and EAE, and correlates with the intensity of inflamma-tion [7] . HMGB1 is a nuclear protein that could be ac-tively released by stimulated monocytes and macrophages [9] , or passively released by necrotic cells [10] . It binds to several receptors, including the receptor for advanced gly-cation end products (RAGE), amplifying neuroinflam-matory processes in MS and EAE [7, 11–14] . Proinflam-matory cytokines and HMGB1, mutually induced, per-petuate inflammatory responses in the pathogenesis of several inflammatory and autoimmune diseases [15] .

Gene expression studies of brain tissue and peripheral blood (PBL) from MS patients have shown deregulated expression of cytokines, chemokines and transcription factors related to immune activation [2, 16] , but the rela-tion between the peripheral immune system and neuro-inflammation is still not well understood. In contrast to other studies, which mostly investigated the cytokine profile in the advanced stages of MS, we aimed to assess systemic and cerebrospinal fluid (CSF) changes in pa-tients with MS at the clinical onset of the disease. We fo-cused on the HMGB1/RAGE ligand-receptor axis and, in addition, determined the PBL gene expression profile of selected proinflammatory/immunoregulatory cytokines that may interfere with HMGB1 action, including IL-1α, IL-1β, IL-6, IL-10, S100 calcium-binding protein A12 (S100A12) and TNF-α, as well as the protein level of IL-1β and TNF-α in both plasma and CSF. Thus, our study may help identify early pathogenic mechanisms involved in the inflammatory response associated to MS.

Patients and Methods

MS Patients and Control Subjects PBL and CSF were collected from MS patients (n = 27) and

healthy controls (n = 30; table 1 ). Diagnosis of MS was established according to the revised McDonald criteria [17] and all patients

had the relapsing-remitting (RR)-MS form of the disease. Clinical severity of the disease for MS patients was evaluated using the ex-panded disability status scale (EDSS). MS samples were obtained from patients during the diagnostic procedure at the disease onset, before they had been subjected to any immunomodulatory or im-munosuppressive treatment, such as with corticosteroids or IFN. Control subjects were recruited among patients routinely under-going epidural anesthesia prior to lower extremity surgery, allow-ing us to obtain a small amount of CSF. All subjects with inflam-matory diseases were excluded, except for MS in the experimental group. Also, we recorded standard laboratory serum parameters, including C-reactive protein (CRP; mg/l), measured by high-sen-sitivity nephelometric assay, and erythrocyte sedimentation rate (ESR; mm/h), determined according to the Westergren method, as markers of systemic inflammation. None of the subjects had a his-tory of using medications that could interfere with the level of im-mune mediators evaluated by the study. We also excluded subjects on statin therapy due to the reported interactions between statins and the AGE/RAGE axis [2] . The study protocol was approved by the Ethical Committee of Clinical Hospital ‘Dubrava’ and the Uni-versity of Zagreb School of Medicine, and informed consent was obtained from all patients prior to the procedures. The study was conducted in accordance with the Declaration of Helsinki.

PBL and CSF Samples Paired PBL and CSF from MS and control groups were col-

lected for enzyme-linked immunosorbent assay (ELISA) and quantitative polymerase chain reaction (qPCR) analysis. CSF was collected in siliconized glass tubes and centrifuged immediately. PBL was sampled in ethylenediaminetetraacetic acid-containing tubes. PBL mononuclear cells (PBMCs) were separated by density gradient centrifugation using Histopaque (Sigma-Aldrich, St. Louis, Mo., USA). CSF and plasma were recovered and stored at –80   °   C until use. The proportion of viable cells, assessed by trypan blue exclusion, was more than 90%.

Gene Expression Analysis by Real-Time qPCR Total RNA was extracted from PBMCs using the 6,100 Nucle-

ic Acid PrepStation (Applied Biosystems, Life Technologies, Fos-ter City, Calif., USA) and converted into complementary DNA,

Table 1. Characteristics of MS patients and controls

Characteristic MS (n = 27) mean ± SD (median, IQR)

Control (n = 30)mean ± SD (median, IQR)

Age, years 37.7±12.9 (37.0, 28.0–49.5)

43.6±10.5 (42.5, 35.0–47.0)

CRP, mg/l 1.0±0.4 (0.8, 0.5–1.2)

1.1±0.7 (0.9, 0.5–1.6)

ESR, mm/h 8.3±5.2 (6.0, 5.0–11.5)

7.9±2.5 (7.0, 6.0–10.0)

EDSS (0–10 points) 3.0±0.7 (3.0, 2.5–3.5)

NA

Values are presented as mean ± SD with median and IQR in parentheses. NA = Not applicable.

Dow

nloa

ded

by:

Mon

ash

Uni

vers

ity13

0.19

4.20

.173

- 9

/29/

2014

9:0

7:07

AM

Glasnović    et al. Neuroimmunomodulation 2014;21:226–233DOI: 10.1159/000357002

228

cDNA, by reverse transcriptase (Applied Biosystems). The amount of cDNA corresponding to 20 ng of reversely transcribed RNA was amplified by qPCR in an ABI Prism 7500 Sequence Detection System (Applied Biosystems) instrument. Gene expression of the proinflammatory/immunoregulatory mediators RAGE, HMGB1, IL-1α, IL-1β, IL-6, IL-10, S100A12 and TNF-α was analyzed using the commercially available TaqMan Assays (Applied Biosystems; table 2 ). Each reaction was performed in duplicates in a 25-μl re-action volume. The relative quantities were calculated using a standard curve designed from 6 serial dilutions of the calibrator sample (control PBMCs). According to the standard curve, the relative amounts of RNA for target genes were calculated as the ratio of the quantity of the target gene normalized to GAPDH (glyceraldehyde 3-phosphate dehydrogenase) as the endogenous control.

Enzyme-Linked Immunosorbent Assay Concentrations of proinflammatory/immunoregulatory me-

diators in the CSF and plasma were determined by ELISA using the commercially available sets of chemicals according to the man-ufacturer’s instructions: TNF-α and IL-1β (Platinum Sensitive ELISA, eBioscience, San Diego, Calif., USA), HMGB1 (Shino-Test Corporation, IBL International, Tokyo, Japan), S100A12/EN-RAGE (CycLex Co., Medical & Biological Laboratories Co., Na-goya, Japan) and soluble (s)RAGE (Quantikine Immunoassay, R&D Systems, Minneapolis, Minn., USA). Optical density was de-termined using the microplate reader (Bio-Rad, Hercules, Calif., USA) set to an excitation wavelength of 450 nm.

Statistical Analysis Clinical and laboratory data for each group of patients were

presented as mean ± standard deviation (SD) and median with the interquartile range (IQR). Gene expression values of MS and con-trol samples were presented as median with IQR and compared using the nonparametric Mann-Whitney test. Values for inflam-matory mediators were correlated using Spearman’s rank correla-tion coefficient rho (ρ) with its 95% confidence interval (CI). The minimal false discovery rate level (Storey’s q) was calculated for multiple test correction [18] . The receiver-operating characteris-

tic, ROC, curve analysis was presented as the area under the curve, AUC, with its 95% CI and used to determine the efficacy of ob-tained values to discriminate between MS and the control group. Diagnostic efficacy was assessed using the sensitivity and specific-ity at a cut-off point. Statistical analysis was performed using the  software packages MedCalc (version 12.5, MedCalc Inc., Mariakerke, Belgium) and the R project for statistical computing (freely available at http://www.r-project.org/). For all experiments, results with p < 0.05 and q < 0.05 were considered statistically sig-nificant.

Results

Clinical Characteristics of MS Patients and Controls MS patients were clinically assessed at the disease on-

set and evaluated by their EDSS score ( table 1 ). Control and MS groups had standard laboratory serum parame-ters within the normal range (<5 mg/l for CRP and <15 mm/h for ESR; table 1 ).

Cytokine Gene Expression Profile in PBMCs of MS Patients In the first set of experiments we aimed to determine

the gene expression profile of proinflammatory/immu-noregulatory mediators (IL-1α, IL-1β, IL-6, IL-10 and TNF-α) associated with MS by comparing their expres-sion in PBMCs of MS patients with that in control sam-ples. We found a significantly lower expression of IL-10 (p = 0.031, q = 0.043) in MS patients, and no significant difference in the expression of IL-1α, IL-1β, IL-6 and TNF-α ( fig. 1 ).

RAGE Ligands/RAGE Axis in the CSF and Circulation The analysis of gene expression levels for RAGE,

HMGB1 and S100A12 of MS patients and control sub-jects revealed that there was no statistically significant dif-ference in PBMC gene expression between the groups ( fig. 2 a). The protein level of sRAGE was lower in the CSF of MS patients (p = 0.021, q = 0.023), whereas protein lev-els of HMGB1 and S100A12 in plasma and CSF, as well as sRAGE in plasma, were not significantly different be-tween the controls and MS patients ( fig. 2 b, c).

TNF-α and IL-1β in the CSF and Circulation Similar to the results for HMGB1 and S100A12, we

found no significant differences between MS patients and controls in the protein levels of IL-1β and TNF-α in plas-ma and CSF ( fig. 3 ). Nevertheless, the overall concentra-tions were very low in both plasma and CSF, close to the detection limit of the assay (<1 pg/ml).

Table 2. TaqMan assays used for qPCR analysis

Gene Gene symbol Assay ID

RAGE AGER Hs00542590_m1HMGB1 HMGB1 Hs01037385_s1S100A12 S100A12 Hs00942835_g1IL-1α IL1A Hs00174092_m1IL-1β IL1B Hs01555410_m1IL-6 IL6 Hs00985639_m1IL-10 IL10 Hs00961622_m1TNF-α TNF Hs99999043_m1GAPDH GAPDH Hs99999905_ml

Assays for qPCR analysis were commercially available and used in accordance with the manufacturer’s instructions (Applied Bio-systems).

Dow

nloa

ded

by:

Mon

ash

Uni

vers

ity13

0.19

4.20

.173

- 9

/29/

2014

9:0

7:07

AM

sRAGE in MS Patients Neuroimmunomodulation 2014;21:226–233DOI: 10.1159/000357002

229

Correlations of Proinflammatory Mediators By testing the association between proinflammatory

cytokines and HMGB1 or S100A12 systemic expression, we confirmed a significant positive correlation between PBMC gene expression for HMGB1 and S100A12 with IL-6 (ρ = 0.665, 95% CI 0.478–0.794, p < 0.001, q < 0.001 and ρ = 0.561, 95% CI 0.204–0.786, p = 0.007, q = 0.004, respectively; fig. 4 a, b). Finally, we evaluated the ability of sRAGE to discriminate between diagnostic samples from MS patients and control subjects by ROC curve analysis. MS patients at clinical onset and controls could be distin-guished based on the sRAGE level in the CSF (AUC = 0.788, 95% CI 0.584–0.922, p = 0.007; fig. 4 c), at a cut-off point ≤ 7.39 pg/ml, with a sensitivity of 72.22% and spec-ificity of 87.50%.

Discussion

Our study aimed to characterize the systemic and local profile of the HMGB1/RAGE axis in MS patients at the disease onset. We observed a lower level of sRAGE in the CSF of MS patients, confirming its role in the pathogen-esis of MS. Investigations linking peripheral and CSF im-mune responses are central to understanding the MS im-munopathogenesis, and to evaluating the importance of studies on the peripheral immune system compared with studies on CSF and MS lesions, which are still sparse due to the limited availability of local samples.

Cellular and molecular events triggering MS as well as regulatory mechanisms limiting the initiation and pro-gression of the central nervous system (CNS) inflamma-tion are still not well understood [19] . Our study is the first to investigate the level of HMGB1 and sRAGE in the plasma and CSF of MS patients at clinical onset. HMGB1, also known as amphoterin, is specifically related to the CNS, being highly expressed by developing neurons and playing an important role in its ontogenesis. In contrast to the study by Uzawa et al. [20] , we did not detect in-creased plasma or CSF levels of HMGB1 in our set of MS samples. However, this study included patients with neu-romyelitis optica (NMO), advanced-stage RR-MS and control patients with noninflammatory neurological dis-orders, and found a higher HMGB1 CSF level in NMO patients than in RR-MS patients and controls. Another study investigating autopsies from a heterogeneous group of MS patients observed an increased HMGB1 expression primarily in macrophages and microglia within the MS lesions [7] . The same study also confirmed a higher ex-pression of HMGB1 in PBMCs of RR-MS and secondary-progressive MS patients than in patients with noninflam-matory neurological diseases. Based on these results, we can speculate that HMGB1 is upregulated in MS at ad-vanced stages of the disease. HMGB1 release may in-crease in response to proinflammatory cytokine stimula-tion with the disease progression, forming a positive in-flammatory feedback loop.

In contrast to no detectable changes in HMGB1 level, we found a lower sRAGE level in the CSF of MS patients. HMGB1 engagement with the full-length membrane form of RAGE regulates immune responses. RAGE is implicated in a number of inflammatory diseases (in-cluding diabetes, atherosclerosis, arthritis and neurode-generative disorders). It is mainly expressed on immune cells, neurons, activated endothelial and vascular smooth muscle cells, triggering programs of acute and chronic inflammation [21] . Specifically, it is expressed on neu-

Fig. 1. Expressions of selected proinflammatory/immunoregulato-ry cytokines in PBMCs from control subjects and MS patients ana-lyzed by qPCR. Values are presented as RNA relative quantity (me-dian with IQR) for each group. Group-to-group comparisons were performed using the nonparametric Mann-Whitney test; results with p < 0.05 and q < 0.05 were considered statistically significant.

0

0.2

0.4

0.6

RNA

rela

tive

quan

tity

IL-1

0

2

4

6

8

10

RNA

rela

tive

quan

tity

IL-

0

1

2

3

RNA

rela

tive

quan

tity

TNF-

0

1.0

0.5

1.5

2.0

RNA

rela

tive

quan

tity

IL-6

0

1

2

3

RNA

rela

tive

quan

tity

IL-10

p = 0.031

ControlsMS

Dow

nloa

ded

by:

Mon

ash

Uni

vers

ity13

0.19

4.20

.173

- 9

/29/

2014

9:0

7:07

AM

Glasnović    et al. Neuroimmunomodulation 2014;21:226–233DOI: 10.1159/000357002

230

0

0.4

0.2

0.6

0.8

RNA

rela

tive

quan

tity

0

0.5

1.0

1.5

RNA

rela

tive

quan

tity

0

10

20

30

Conc

entra

tion

(ng/

ml)

S100

A12

0

0.5

1.0

1.5

Conc

entra

tion

(ng/

ml)

CSF

HM

GB1

0

1

2

3

Conc

entra

tion

(ng/

ml)

Plasma

0

1

2

3

4

RNA

rela

tive

quan

tity

PBMCs

0

20

40

60

80

Conc

entra

tion

(ng/

ml)

ControlsMS

a0

200

400

600

800Co

ncen

tratio

n (p

g/m

l)

b0

5

10

15

Conc

entra

tion

(pg/

ml)

RAGE

p = 0.021

c

Fig. 2. Gene expressions and protein levels of RAGE ligands/RAGE in CSF and PBL from control subjects and MS patients. HMGB1, S100A12 and RAGE were analyzed in PBMCs by qPCR ( a ), in plasma by ELISA ( b ) and in CSF by ELISA ( c ). Values are

presented as RNA relative quantity or cytokine concentration (me-dian with IQR) for each group. Group-to-group comparisons were performed using the nonparametric Mann-Whitney test; results with p < 0.05 and q < 0.05 were considered statistically significant.

0

1

2

3

Conc

entra

tion

(pg/

ml)

Plasma

0

1

2

3

Conc

entra

tion

(pg/

ml)

0

0.5

1.0

1.5

Conc

entra

tion

(pg/

ml)

CSF

IL-1

ControlsMS

0

1

2

3

Conc

entra

tion

(pg/

ml)

TNF-

Fig. 3. Protein levels of TNF-α and IL-1β in CSF and plasma from control subjects and MS patients. TNF-α and IL-1β were mea-sured by ELISA. Values are presented as cy-tokine concentration (median with IQR) for each group. Group-to-group compari-sons were performed using the nonpara-metric Mann-Whitney test.

Dow

nloa

ded

by:

Mon

ash

Uni

vers

ity13

0.19

4.20

.173

- 9

/29/

2014

9:0

7:07

AM

sRAGE in MS Patients Neuroimmunomodulation 2014;21:226–233DOI: 10.1159/000357002

231

rons, microglia and endothelial cells at the blood-brain barrier (BBB), regulating neuronal development and BBB integrity [22] . In contrast to the full-length form, sRAGE forms, including the cleaved form (cRAGE) shed by membrane metalloproteinase, and the endoge-nous secretory form (esRAGE), have been found circu-lating in plasma and tissues [22] . sRAGE acts as a decoy receptor, sequestering circulating ligands and competi-tively inhibiting RAGE activation, thus protecting sensi-tive cells from the deleterious effects of ligand/RAGE hyperactivity [22–24] . Since the distribution of sRAGE forms is tissue specific, with cRAGE being the predomi-nant form in serum and esRAGE secreted by endothe-lial cells, we hypothesized that a lower sRAGE level in CSF may be a consequence of endothelial dysfunction, well described in MS [25] .

In addition to HMGB1, other ligand/RAGE interac-tions (including S100 proteins, amyloid β-peptide and advanced glycation end products) have been involved in neuroinflammation and brain tissue injury in different inflammatory CNS disorders [19, 22, 26, 27] . Moreover, convergence of S100 proteins and HMGB1 onto their re-ceptor RAGE may be a rate-limiting step in RAGE-me-diated pathways [15] . Several studies proposed that on-going inflammation and oxidative stress within the CNS may, in turn, lead to the upregulation of RAGE or RAGE-ligands in neurons, microglia and BBB [22, 28] . Activa-tion of microglial cells is among the early events in lesion formation, causing accelerated production of inflamma-

tory cytokines and enhanced production of reactive oxy-gen species. Using cultured synovial fibroblasts and mac-rophages, it was shown that HMGB1 exhibits enhanced proinflammatory activity by binding to cytokines, spe-cifically IL-1β and TNF-α [29, 30] . Therefore, we pro-posed that inflammation may be potentiated by a com-plex formation of HMGB1 with IL-1β and TNF-α, found to be elevated in the CSF of RR-MS patients [5, 16] , but our study could not detect an increase in any of these molecules. The lack of difference compared with control samples indicates that the described pathogenic mecha-nism is not important for MS at the disease onset. An-other possibility is that HMGB1/cytokine interactions occur within the MS lesions, and are not reflected in the CSF.

Other proinflammatory cytokines also play an impor-tant role in the pathogenesis of neurodegenerative dis-eases such as MS. PBMCs and endothelial cells are well-known sources of cytokines, affecting the CNS by cross-ing the BBB, but the brain tissues also contribute to cytokine overproduction and chronic inflammation. Since both peripheral and brain-born cytokines are in-volved in the regulation of inflammatory responses through a complex interplay, their systemic expression may reflect the disease process. We found that PBMC ex-pression of IL-6 positively correlated with HMGB1 and S100A12, and that PBMC expression of IL-1α was higher, although not significantly (p = 0.086, q = 0.057), in MS patients compared with controls. Both IL-6 and IL-1α

Fig. 4. The cytokine profile in MS patients. Correlation between the gene expression of IL-6 and HMGB1 ( a ) or S100A12 in PBMCs ( b ). PBMC gene expressions were assessed by qPCR. Values were correlated using Spearman’s rank correlation coefficient ρ (95% CI for ρ); results with p < 0.05 and q < 0.05 were considered statisti-

cally significant. c The discriminatory ability of sRAGE protein levels in CSF between MS and control samples determined by ROC curves. Diagnostic efficacy for these values was assessed using the sensitivity and specificity at a cut-off point.

00 20 40 60 80 100

20

40

60

80

100

Sens

itivi

ty

c 100-specificity

CSF sRAGE

Cut-off

00 10 20 40

2

4

6

8

a

PBMC HMGB1

00 10 20 40

2

1

4

PBMCs S100A12

b

Dow

nloa

ded

by:

Mon

ash

Uni

vers

ity13

0.19

4.20

.173

- 9

/29/

2014

9:0

7:07

AM

Glasnović    et al. Neuroimmunomodulation 2014;21:226–233DOI: 10.1159/000357002

232

have been implicated in the pathogenesis of neuroinflam-mation in MS [5, 31–33] , and, in accordance with our re-sults, associated with early MS. On the other hand, we found that the expression of IL-10, a potent anti-inflam-matory cytokine, was lower in PBMCs of MS patients, concordant with several studies of neuroinflammatory disorders [5, 15, 33] . IL-10 could be induced by type I IFN in monocytes, macrophages and CD4+ T lymphocytes [19] , and represents an important negative feedback mechanism to downregulate uncontrolled production of proinflammatory cytokines (IL-1β, IL-6, IL-17, TNF-α), protecting brain homeostasis [5, 16, 33] .

To summarize, the control of inflammation and neu-roprotection should be the key point in designing MS therapeutic protocols since deregulated innate and adap-tive immune reactions may facilitate harmful inflamma-tory and autoimmune responses. We confirmed that the cytokine network was already disturbed in PBL and CSF at the clinical onset of the disease, showing a different profile compared with studies on advanced stages of MS [19, 25] . In our future investigations we plan to include other groups of MS patients, with different forms and stages of the disease, to further define the cytokine profile

specific for the disease onset. Lower sRAGE may be used to distinguish between MS at clinical onset and control patients, but only in the CSF. However, testing the dis-criminative ability of sRAGE between MS and other neu-roinflammatory diseases would be of even greater clinical importance. Other studies support a role for both cRAGE and esRAGE as biomarkers and endogenous protection factors against RAGE-mediated pathogenesis [22, 26, 34] . sRAGE levels have been found to be decreased in athero-sclerosis, diabetes, renal failure and the aging process [15, 21, 35] , representing a possible therapeutic target in chronic inflammatory diseases, including MS.

Acknowledgements

This work was supported by a grant from the Croatian Ministry of Science, Education and Sports (108-1080229-0142). We thank Prof. Željko Romić, Department of Laboratory Diagnostics, and the medical staff of the Department of Neurology and Department of Vascular Surgery, Clinical hospital ‘Dubrava’, for their profes-sional help. We also thank Ms. Katerina Zrinski-Petrović for her technical assistance, Ms. Antonia Paić for language editing and Prof. Ivan Krešimir Lukić and Prof. Mladen Petrovečki for statisti-cal editing.

References

1 Compston A, Coles A: Multiple sclerosis. Lancet 2008; 372: 1502–1517.

2 Sospedra M, Martin R: Immunology of mul-tiple sclerosis. Annu Rev Immunol 2005; 23: 683–747.

3 Muls N, Jnaoui K, Dang HA, Wauters A, Van Snick J, Sindic CJ, Van Pesch V: Upregulation of IL-17, but not of IL-9, in circulating cells of CIS and relapsing MS patients: impact of cor-ticosteroid therapy on the cytokine network. Neuroimmunol 2012; 243: 73–80.

4 Kürtüncü M, Tüzün E, Türkoğlu R, Petek-Balcı B, Içöz S, Pehlivan M, Birişik Ö, Ulusoy C, Shugaiv E, Akman-Demir G, Eraksoy M: Effect of short-term interferon-β treatment on cytokines in multiple sclerosis: significant modulation of IL-17 and IL-23. Cytokine 2012; 59: 400–402.

5 Reale M, De Angelis F, Di Nicola M, Capello E, Di Ioia M, Luca GD, Lugaresi A, Tata AM: Relation between proinflammatory cytokines and acetylcholine levels in relapsing-remit-ting multiple sclerosis patients. Int J Mol Sci 2012; 13: 12656–12664.

6 Kalinowska-Łyszczarz A, Szczuciński A, Paw-lak MA, Losy J: Clinical study on CXCL13, CCL17, CCL20 and IL-17 as immune cell mi-gration navigators in relapsing-remitting multiple sclerosis patients. J Neurol Sci 2011; 300: 81–85.

7 Andersson A, Covacu R, Sunnemark D, Danilov AI, Dal Bianco A, Khademi M, Wall-ström E, Lobell A, Brundin L, Lassmann H, Harris RA: Pivotal advance: HMGB1 expres-sion in active lesions of human and experi-mental multiple sclerosis. J Leukoc Biol 2008; 84: 1248–1255.

8 Musabak U, Demirkaya S, Genç G, Ilikci RS, Odabasi Z: Serum adiponectin, TNF-α, IL12p70, and IL-13 levels in multiple sclerosis and the effects of different therapy regimens. Neuroimmunomodulation 2011; 18: 57–66.

9 Gardella S, Andrei C, Ferrera D, Lotti LV, Tor-risi MR, Bianchi ME, Rubartelli A: The nucle-ar protein HMGB1 is secreted by monocytes via a non-classical, vesicle-mediated secretory pathway. EMBO Rep 2002; 3: 995–1001.

10 Scaffidi P, Misteli T, Bianchi ME: Release of chromatin protein HMGB1 by necrotic cells triggers inflammation. Nature 2002; 418: 191–195.

11 Huttunen HJ, Fages C, Rauvala H: Receptor for advanced glycation end products (RAGE)-mediated neurite outgrowth and activation of NF-κB require the cytoplasmic domain of the receptor but different downstream signaling pathways. J Biol Chem 1999; 274: 19919–19924.

12 Hori O, Brett J, Slattery T, Cao R, Zhang J, Chen JX, Nagashima M, Lundh ER, Vijay S,

Nitecki D, et al: The receptor for advanced gly-cation end products (RAGE) is a cellular bind-ing site for amphoterin: mediation of neurite outgrowth and co-expression of RAGE and amphoterin in the developing nervous system. J Biol Chem 1995; 270: 25752–25761.

13 Park JS, Svetkauskaite D, He Q, Kim JY, Stras-sheim D, Ishizaka A, Abraham E: Involve-ment of Toll-like receptors 2 and 4 in cellular activation by high mobility group box 1 pro-tein. J Biol Chem 2004; 279: 7370–7377.

14 Yu M, Wang H, Ding A, Golenbock DT, Latz E, Czura CJ, Fenton MJ, Tracey KJ, Yang H: HMGB1 signals through Toll-like receptor (TLR) 4 and TLR2. Shock 2006; 26: 174–179.

15 Fritz G: RAGE: a single receptor fits multiple ligands. Trends Biochem Sci 2011; 12: 625–632.

16 Romme Christensen J, Börnsen L, Hesse D, Krakauer M, Sørensen PS, Søndergaard HB, Sellebjerg F: Cellular sources of dysregulated cytokines in relapsing-remitting multiple sclerosis. J Neuroinflammation 2012; 9: 215.

17 Polman CH, Reingold SC, Banwell B, Clanet M, Cohen JA, Filippi M, Fujihara K, Havrdova E, Hutchinson M, Kappos L, Lublin FD, Mon-talban X, O’Connor P, Sandberg-Wollheim M, Thompson AJ, Waubant E, Weinshenker B, Wolinsky JS: Diagnostic criteria for multi-ple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol 2011; 69: 292–302.

Dow

nloa

ded

by:

Mon

ash

Uni

vers

ity13

0.19

4.20

.173

- 9

/29/

2014

9:0

7:07

AM

sRAGE in MS Patients Neuroimmunomodulation 2014;21:226–233DOI: 10.1159/000357002

233

18 Storey JD: A direct approach to false discov-ery rates. J R Stat Soc Ser B Stat Methodol 2002; 64: 479–498.

19 Zhang L, Yuan S, Cheng G, Guo B: Type I IFN promotes IL-10 production from T cells to suppress Th17 cells and Th17-associated au-toimmune inflammation. PLoS One 2011; 6:e28432.

20 Uzawa A, Mori M, Masuda S, Muto M, Kubawara S: CSF high-mobility group box 1 is associated with intrathecal inflammation and astrocytic damage in neuromyelitis opti-ca. J Neurol Neurosurg Psychiatry 2013; 84: 517–522.

21 Riehl A, Németh J, Angel P, Hess J: The recep-tor RAGE: bridging inflammation and can-cer. Cell Commun Signal 2009; 7: 12.

22 Han SH, Kim YH, Mook-Jung I: RAGE: the beneficial and deleterious effects by diverse mechanisms of actions. Mol Cells 2011; 31: 91–97.

23 Vazzana N, Santilli F, Cuccurullo C, Davì G: Soluble forms of RAGE in internal medicine. Intern Emerg Med 2009; 4: 389–401.

24 Kalousová M, Havrdová E, Mrázová K, Spacek P, Braun M, Uhrová J, Germanová A, Zima T: Advanced glycoxidation end prod-ucts in patients with multiple sclerosis. Prague Med Rep 2005; 106: 167–174.

25 Holman DW, Klein RS, Ransohoff RM: The blood-brain barrier, chemokines and multi-ple sclerosis. Biochim Biophys Acta 2011; 1812: 220–230.

26 Chan JK, Roth J, Oppenheim JJ, Tracey KJ, Vogl T, Feldmann M, Horwood N, Nancha-hal J: Alarmins: awaiting a clinical response. J Clin Invest 2012; 122: 2711–2719.

27 Muhammad S, Barakat W, Stoyanov S, Mu-rikinati S, Yang H, Tracey KJ, Bendszus M, Rossetti G, Nawroth PP, Bierhaus A, Schwan-inger M: The HMGB1 receptor RAGE medi-ates ischemic brain damage. J Neurosci 2008; 28: 12023–12031.

28 Sternberg Z, Ostrow P, Vaughan M, Chichel-li T, Munschauer F: AGE-RAGE in multiple sclerosis brain. Immunol Invest 2011; 40: 197–205.

29 Sha Y, Zmijewski J, Xu Z, Abraham E: HMGB1 develops enhanced proinflammato-ry activity by binding to cytokines. J Immunol 2008; 180: 2531–2537.

30 Wähämaa H, Schierbeck H, Hreggvidsdottir HS, Palmblad K, Aveberger AC, Andersson U, Harris HE: High mobility group box pro-tein 1 in complex with lipopolysaccharide or IL-1 promotes an increased inflammatory phenotype in synovial fibroblasts. Arthritis Res Ther 2011; 13:R136.

31 Harris VK, Sadiq SA: Disease biomarkers in multiple sclerosis: potential for use in thera-peutic decision making. Mol Diagn Ther 2009; 13: 225–244.

32 Kimura A, Kishimoto T: IL-6: regulator of Treg/Th17 balance. Eur J Immunol 2010; 40: 1830–1835.

33 Rubio-Perez JM, Morillas-Ruiz JM: A review: inflammatory process in Alzheimer’s disease, role of cytokines. Sci World J 2012; 2012: 756357.

34 Sims GP, Rowe DC, Rietdijk ST, Herbst R, Coyle AJ: HMGB1 and RAGE in inflamma-tion and cancer. Annu Rev Immunol 2010; 28: 367–388.

35 Maillard-Lefebvre H, Boulanger E, Daroux M, Gaxatte C, Hudson BI, Lambert M: Soluble receptor for advanced glycation end products: a new biomarker in diagnosis and prognosis of chronic inflammatory diseases. Rheuma-tology (Oxford) 2009; 48: 1190–1196.

Dow

nloa

ded

by:

Mon

ash

Uni

vers

ity13

0.19

4.20

.173

- 9

/29/

2014

9:0

7:07

AM