- Email: [email protected]
Elevated soluble syndecan-1 levels in neuromyelitis optica are associated with disease severity
Cytokine 111 (2018) 140–145
Contents lists available at ScienceDirect
Cytokine journal homepage: www.elsevier.com/locate/cytokine
Short communication
Elevated soluble syndecan-1 levels in neuromyelitis optica are associated with disease severity ⁎
Shanshan Peia, Dong Zhengb, Zhanhang Wangc, Xueqiang Hud, Suyue Pana, , Honghao Wanga,
T ⁎
a
Department of Neurology, Nanfang Hospital, Southern Medical University, Guangzhou, China Department of Neurology, The Affiliated Brain Hospital of Guangzhou Medical University, China c Department of Neurology, 39 Brain Hospital, China d Department of Neurology, Third Affiliated Hospital of Sun Yat-sen University, China b
A R T I C LE I N FO
A B S T R A C T
Keywords: Cerebrospinal fluid Soluble syndecan-1 Neuromyelitis optica Multiple sclerosis Disease severity
Syndecan-1 (SDC-1) is a transmembrane member that has a profound influence on the resolution of inflammation. Soluble syndecan-1 (sSDC-1) levels have been shown to increase in many inflammatory diseases. However, it remains unknown whether sSDC-1 concentration is elevated in neuromyelitis optica (NMO) and multiple sclerosis (MS) patients. The aims of this pilot study were to investigate the relationship between sSDC-1 and disease severity in NMO and MS and whether sSDC-1 has potential as an effective marker for disease severity. We measured sSDC-1 concentrations by using an enzyme-linked immunosorbent assay (ELISA). NMO patients had significantly higher CSF sSDC-1 levels than MS patients or controls. We also found a positive correlation between the increased CSF sSDC-1 levels and increased severity in NMO disease, but not in MS. In NMO, CSF sSDC-1 concentrations were positively correlated with CSF interleukin (IL)-6, IL-8 and IL-17. Overall, we showed levels of CSF sSDC-1 were higher in NMO patients and had a positive relationship with disease severity of NMO but not with MS. CSF sSDC-1 may be an effective marker of NMO disease severity.
1. Introduction
can be proteolytically released from the cell surface to become a soluble HSPG. This process can be induced by matrix metalloproteinases (MMPs) [6]. In NMO, proinflammatory cytokines such as tumor necrosis factor (TNF)-α, interleukin (IL)-1β, and IL-6 are upregulated [2,3,7] and induce MMPs [6–9]. Moreover, soluble SDC-1 (sSDC-1) is increased by shedding during inflammation and correspondingly membrane bound SDC-1 decreased [4]. The current observations suggest that syndecan-1 shedding has distinct regulatory functions in different inflammatory diseases. For example, syndecan-1 shedding functions as an anti-inflammatory molecule in allergic lung inflammation but an pro-inflammatory molecule in bleomycin-induced acute lung injury by inducing an CXC chemokine gradient mobilization that guides transepithelial efflux of neutrophils [8]. However, the role of sSDC-1 in NMO and MS remains unknown. This pilot study investigated the serum and cerebrospinal fluid (CSF) concentrations of sSDC-1 in NMO and MS patients and explored its role in the pathogenesis of NMO.
NMO and MS are two typical immune-mediated inflammatory demyelinating diseases of the central nervous system. NMO was regarded as a variant of MS until the water channel anti-aquaporin-4 (AQP4) antibodies were discovered. Moreover, differences between MS and NMO such as clinical characteristics, pathology, neuroimaging results, and response to some immunotherapies distinguish these diseases [1]. Many pro-inflammatory and anti-inflammatory cytokines and chemokines contribute to NMO pathogenesis and NMO severity can be regulated by the balance between these opposing factors [2,3]. Syndecan-1 (SDC-1) is a transmembrane HSPG that consists of three different domains (cytoplasmic, transmembrane, and extracellular) and predominantly expressed by epithelial cells and plasmacytes [4]. The main function of SDC-1 in adult mammals is to resolve inflammation. SDC-1 deficiency has been proved to be involved in many diseases with uncontrolled inflammation in vivo and vitro [5]. SDC-1 also takes part in wound healing, fibrosis, tumor biology, and infectious diseases [4]. Through ectodomain shedding, the syndecan extracellular domain
⁎
Corresponding authors at: Department of Neurology, Nanfang Hospital, Southern Medical University, 1838 North Guangzhou Avenue, Guangzhou, Guangdong 510515, China (H. Wang). Department of Neurology, Nanfang Hospital, Southern Medical University, 1838 North Guangzhou Avenue, Guangzhou, Guangdong 510515, China (S. Pan). E-mail addresses: [email protected] (S. Pan), [email protected] (H. Wang). https://doi.org/10.1016/j.cyto.2018.08.017 Received 5 May 2018; Received in revised form 27 July 2018; Accepted 15 August 2018 1043-4666/ © 2018 Elsevier Ltd. All rights reserved.
Cytokine 111 (2018) 140–145
S. Pei et al.
2.3. Study of serum and CSF samples
Table 1 Demographic and clinical features of patients and controls. NMO
MS
Control
No. of subjects (n) Gender (female/male) Age (years) Disease duration (years) EDSS scores
23 15/8 33.43(16–60) 2.24(1–4) 4.22(2–6.5)
12 5/7 34.58(15–52) 1.75(1–2) 2.25(1–3.5)
16 11/5 33.88(27–39) – –
NMO-IgG Positive Negative
19 4
– –
– –
sSDC-1 levels were measured via a commercially available ELISA kit (Diaclone, Besancon, France) according to the manufacturer's instructions. 2.4. Statistical analysis For CSF sSDC-1, serum sSDC-1, IL-6, IL-8 and IL-17, data were presented as mean ± standard deviation or median with range for age, disease duration and EDSS score. The Kruskal-Wallis tests and Spearman’s test were used for the comparison between groups and determining the correlations between sSDC-1 levels and proinflammatory cytokines as well as aquaporin-4 antibodies levels respectively. P-values < 0.05 were considered statistically significant. SPSS 20.0 software was used to perform statistical analyses. Receiver operating characteristic (ROC) curve were performed using GraphPad Prism 7.0 (GraphPad Software, La Jolla, CA, USA). Power analysis were analyzed using BMDP Statistical Software (1993), the following was found: With the proposed sample size of 23 NMO patients and 16 controls, the study will have power of 98.78% to yield a statistically significant result.
Disease duration (years) refers to years from disease onset to sampling. EDSS, Expanded Disability Status Scale; NMO, neuromyelitis optica; MS, multiple sclerosis.
2. Patients and methods 2.1. Patients and controls 23 NMO patients, 12 MS patients and 16 controls with noninflammatory neurological diseases were enrolled. This study was approved by the Ethics Committee of the Nan Fang Hospital of the Southern Medical University. Each enrolled participant signed informed consent. Table 1 shows the demographic and clinical features of the patients.
3. Results 3.1. CSF and serum sSDC-1 in MS and NMO patients and controls The mean plasma sSDC-1 (ng/ml) concentrations for NMO, MS and controls were 99.97 ± 87.25, 74.79 ± 25.98 and 59.61 ± 23.58, respectively. Differences in serum sSDC-1 levels between three groups did not reach statistical significance (Fig. 1A). CSF sSDC-1 levels were significantly higher in the NMO compared with MS and controls (MS, 13.44 ± 6.97 vs 8.19 ± 4.53 ng/ml, p = 0.013; controls,13.44 ± 6.97 vs 5.71 ± 1.79, p < 0.001). The CSF sSDC-1 levels were not significantly different between MS patients and controls (p = 0.189)
2.2. Preparation of blood and CSF samples After sampling, we immediately centrifuged all samples to eliminate cells and other insoluble materials and then distributed equal aliquots into polypropylene tubes for storage at −80 °C until assay.
Fig. 1. Comparison of serum and CSF sSDC1 levels (A, B). (A) No difference was found in serum sSDC-1 levels between NMO, MS and controls. (B) CSF sSDC-1 levels were higher in the NMO compared with MS and controls (MS, p < 0.001; CTLs, p = 0.013). Diagnostic values of sSDC-1 for the assessment of disease activity (C). The 95% confidence interval is from 0.8273 to 1.002. The area under the ROC curve is 0.9144. Comparison of CSF and serum sSDC-1 levels in active and remission periods in NMO (D). In NMO, both serum and CSF soluble SDC-1 levels were significantly higher in the relapse than in remission (CSF, p = 0.013; serum, p = 0.014).
141
Cytokine 111 (2018) 140–145
S. Pei et al.
C S F IL - 6 ( p g /m l)
A p = 0 .0 0 2 2 r= 0 .6 0 5
40 30 20 10 0 0
10
20
30
40
C S F s S D C -1 in N M O (n g /m l)
Fig. 2. Correlation between sSDC-1 and CSF cytokines in NMO and MS (A-F, I-N). In NMO, CSF sSDC-1 concentrations had a significant positive correlation with CSF IL-6 (A), IL-8 (B), and IL-17 (C) (IL-6, p = 0.0022; IL-8, p = 0.0067; IL-17, p = 0.0173). Correlation between sSDC-1 levels and EDSS in NMO and MS patients (G, H, O, P). Plasma sSDC-1 levels were not associated with EDSS scores in NMO (H) and MS (P) patients. CSF sSDC-1 levels in NMO patients were positively correlated with EDSS scores (G) but not in MS patients (O).
142
Cytokine 111 (2018) 140–145
S. Pei et al.
Fig. 2. (continued)
143
Cytokine 111 (2018) 140–145
S. Pei et al.
heparin treatment, a substitute for SDC1, in DSS colitis of SDC1-deficient mice [10]. Through ectodomain shedding, the syndecan extracellular domain can be proteolytically released and become a soluble HSPG [6]. In NMO, CSF cytokines derived from T helper (Th)-1 and Th-17 are increased, primarily TNF-α, IL-6, IL-17, and IL-1β [2,3], which induce MMPs[7,9], a known sheddases of syndecans [6,8,11]. Therefore, sSDC1 levels may be affected by the cytokine network relevant to NMO. As a component of glycocalyx, a gel-like layer on the luminal surface of the endothelium and exerts an barrier role in blood-brain barrier (BBB), the integrity of SDC-1 is important for protecting the integrity of the BBB [12]. It is confirmed that the BBB destruction is a key step in NMO pathogenesis [13]. Thus, SDC-1 shedding breaks the integrity and the protective function of SDC-1 and BBB. Based on the above theoretical basis, we assume that SDC-1 shedding involves in the pathogenesis of NMO. Similarly, previous studies also demonstrate increased levels of this HSPG in other autoimmune inflammatory diseases, such as Crohn’s disease, ulcerative colitis and systemic sclerosis [14,15] and lower sSDC-1 levels were shown in inflammatory bowel disease patients who responded effectively to disease-modifying anti-inflammatory medications compared with untreated patients [15], suggesting a potential role for sSDC-1 in the pathogenesis of autoimmune inflammatory disease. Furthermore, in this study, a positive correlation was found between sSDC-1 and disease severity, indicating a prognostic role of sSDC-1 in NMO. The development of sSDC-1 as a biomarker would permit a more reliable assessment of the BBB destruction in NMO, which has been confirmed to be a pathogenesis in NMO. A further research about sSDC1 in NMO is worthy. In conclusion, this pilot study found the CSF level of sSDC-1 was elevated in NMO patients compared with MS and controls and it had a positive relationship with disease severity, suggesting CSF sSDC-1 as a potential prognostic biomarker that controls underlying inflammation in the pathogenesis of NMO. Further detailed investigations are needed to elucidate the precise function of sSDC-1 in NMO.
(Fig. 1B). 3.2. CSF and serum sSDC-1 in active and remission periods in NMO Serum sSDC-1 in 11 NMO patients and CSF sSDC-1 in 10 NMO patients were measured in both relapse and remission stage. Both serum and CSF sSDC-1 levels were significantly higher in relapse than in remission (CSF p = 0.013; serum p = 0.014) (Fig. 1D). 3.3. CSF and serum sSDC-1 and disease activity For NMO, the median EDSS score was 4.2 (2.0–6.5), and for MS 2.3 (1.0–3.5). In both cohorts, no significant correlation was found between plasma sSDC-1 and EDSS score (Fig. 2H, P). NMO patients with higher EDSS scores showed higher CSF sSDC-1 concentrations during relapse (p < 0.0001) (Fig. 2G) but this result was not found in MS patients (Fig. 2O) (p = 0.563). 3.4. CSF proinflammatory cytokines, aquaporin-4 antibodies and their relationships with sSDC-1 levels Mean CSF IL-6 (pg/ml) of NMO, MS patients and controls were 10.62 ± 7.97, 5.69 ± 2.67 and 2.53 ± 0.43. Mean CSF IL-8 (pg/ml) of NMO, MS patients and controls were 236.90 ± 285.07, 87.69 ± 43.10 and 54.39 ± 16.87. Mean CSF IL-17 (pg/ml) of NMO, MS patients and controls were 9.73 ± 9.54, 5.75 ± 3.75 and 3.84 ± 1.86. Elevated serum IL-6, IL-8 and IL-17 levels were demonstrated in NMO compared with MS and controls (IL-6, p = 0.02, p < 0.001, respectively; IL-8, p = 0.017, p < 0.001, respectively; IL17, p = 0.041, p < 0.001, respectively). Higher IL-6, IL-8 and IL-17 levels were found in MS patients compared with controls (p < 0.001, p = 0.001, p = 0.01, respectively). In NMO, strong positive correlations were found between CSF sSDC-1 and IL-6 (p = 0.0022), IL-8 (p = 0.0067) and IL-17 (p = 0.0173) (Fig. 2A–C). But no positive relationship was found between serum sSDC-1 and cytokines except CSF IL-6 (p = 0.0270) (Fig. 2D). And no relationship was found between CSF aquaporin-4 antibodies and sSDC-1 in both relapse and remission stage (serum, p = 0.227, p = 0.280; CSF, p = 0.961, p = 0.906). The CSF and serum sSDC-1 concentrations also appeared to be associated with CSF cytokines concentrations in the MS group, though they did not quite reach statistical significance. (Fig. 2I–N).
5. Disclosure of conflict of interest No conflict of interest. Acknowledgments Supported by the National Natural Science Foundation of China (81673950). Guangdong provincial science and technology plan projects (2016A020215101, 2017A020215182). Natural Science Foundation of Guangdong Province of China (2016A030313828).
3.5. ROC curve of NMO and CTLs The 95% confidence interval of the ROC curve of NMO and CTLs was from 0.8273 to 1.002. The area under the ROC curve (AUC) was 0.9144, p < 0.001 (Fig. 1C).
References 4. Discussion [1] A.A. Argyriou, N. Makris, Neuromyelitis optica: a distinct demyelinating disease of the central nervous system, Acta Neurol. Scandinavica 118 (2008) 209–217. [2] K.C. Wang, C. Lee, S. Chen, J. Chen, C. Yang, S. Chen, C. Tsai, Distinct serum cytokine profiles in neuromyelitis optica and multiple sclerosis, J. Interferon Cytokine Res. 33 (2013) 58–64. [3] T. Matsushita, T. Tateishi, N. Isobe, T. Yonekawa, R. Yamasaki, D. Matsuse, H. Murai, J. Kira, Characteristic cerebrospinal fluid cytokine/chemokine profiles in neuromyelitis optica, Relapsing Remitting or Primary Progressive Multiple Sclerosis: e61835, 2013; 8. [4] Y.H. Teng, R.S. Aquino, P.W. Park, Molecular functions of syndecan-1 in disease, Matrix Biol. 31 (2012) 3–16. [5] A.H. Bartlett, K. Hayashida, P.W. Park, Molecular and cellular mechanisms of syndecans in tissue injury and inflammation, Mol. Cells 24 (2007) 153–166. [6] S. Brule, N. Charnaux, A. Sutton, D. Ledoux, T. Chaigneau, L. Saffar, L. Gattegno, The shedding of syndecan-4 and syndecan-1 from HeLa cells and human primary macrophages is accelerated by SDF-1/CXCL12 and mediated by the matrix metalloproteinase-9, Glycobiology 16 (2006) 488–501. [7] A. Sebestyén, M. Gallai, T. Knittel, T. Ambrust, G. Ramadori, I. Kovalszky, Cytokine regulation of syndecan expression in cells of liver origin, Cytokine 12 (2000) 1557–1560. [8] Q. Li, P.W. Park, C.L. Wilson, W.C. Parks, Matrilysin shedding of syndecan-1 regulates chemokine mobilization and transepithelial efflux of neutrophils in acute
This pilot study investigated the relationship between sSDC-1 and the inflammatory demyelinating diseases. Until now, this is the first study to examine sSDC-1 levels in NMO patients. We found higher CSF sSDC-1 concentrations in NMO compared with MS and controls but serum sSDC-1 concentrations were not significantly different in these three groups. This maybe because of the immune-mediated inflammatory demyelinating role of NMO and MS in central nervous system. The increased CSF sSDC-1 levels in NMO patients could be attributed to accelerated SDC1 shedding. Thus, CSF sSDC-1 levels played a reasonable diagnostic role in distinguishing NMO patients from MS patients and controls. More importantly, there was also a borderline positive correlation between CSF and plasma sSDC-1 levels and concentrations of CSF proinflammatory cytokines IL-6, IL-8 and IL-17, and EDSS scores in NMO patients. Syndecan-1 (SDC-1) is a transmembrane HSPG. Its main function is to resolve inflammation [4]. Floer et al. reported a protective role of 144
Cytokine 111 (2018) 140–145
S. Pei et al.
lung injury, Cell 111 (2002) 635–646. [9] X. Gan, B. Wong, S.D. Wright, T. Cai, Production of matrix metalloproteinase-9 in CaCO-2 cells in response to inflammatory stimuli, J. Interferon Cytokine Res. 21 (2001) 93–98. [10] M. Floer, M. Götte, M.K. Wild, J. Heidemann, E.S. Gassar, W. Domschke, L. Kiesel, A. Luegering, T. Kucharzik, Enoxaparin improves the course of dextran sodium sulfate-induced colitis in syndecan-1-deficient mice, Am. J. Pathol. 176 (2010) 146–157. [11] R. Day, Regulation of epithelial syndecan-1 expression by inflammatory cytokines, Cytokine 21 (2003) 224–233. [12] J. Zhu, X. Li, J. Yin, Y. Hu, Y. Gu, S. Pan, Glycocalyx degradation leads to blood–brain barrier dysfunction and brain edema after asphyxia cardiac arrest in
rats, J. Cereb. Blood Flow Metab. (2017) 0271678X1772606. [13] F. Shimizu, T. Kanda, Disruption of the blood-brain barrier in inflammatory neurological diseases, Brain Nerve 65 (2013) 165–176. [14] C. Çekiç, A. Kırcı, S. Vatansever, F. Aslan, H.E. Yılmaz, E. Alper, M. Arabul, E. Sarıtaş Yüksel, B. Ünsal, Serum syndecan-1 levels and its relationship to disease activity in patients with Crohn’s disease, Gastroenterol. Res. Practice 2015 (2015) 1–6. [15] D. Yablecovitch, A. Stein, M. Shabat-Simon, T. Naftali, G. Gabay, I. Laish, A. Oren, F.M. Konikoff, Soluble syndecan-1 levels are elevated in patients with inflammatory bowel disease, Dig. Dis. Sci. 60 (2015) 2419–2426.
145
Contents lists available at ScienceDirect
Cytokine journal homepage: www.elsevier.com/locate/cytokine
Short communication
Elevated soluble syndecan-1 levels in neuromyelitis optica are associated with disease severity ⁎
Shanshan Peia, Dong Zhengb, Zhanhang Wangc, Xueqiang Hud, Suyue Pana, , Honghao Wanga,
T ⁎
a
Department of Neurology, Nanfang Hospital, Southern Medical University, Guangzhou, China Department of Neurology, The Affiliated Brain Hospital of Guangzhou Medical University, China c Department of Neurology, 39 Brain Hospital, China d Department of Neurology, Third Affiliated Hospital of Sun Yat-sen University, China b
A R T I C LE I N FO
A B S T R A C T
Keywords: Cerebrospinal fluid Soluble syndecan-1 Neuromyelitis optica Multiple sclerosis Disease severity
Syndecan-1 (SDC-1) is a transmembrane member that has a profound influence on the resolution of inflammation. Soluble syndecan-1 (sSDC-1) levels have been shown to increase in many inflammatory diseases. However, it remains unknown whether sSDC-1 concentration is elevated in neuromyelitis optica (NMO) and multiple sclerosis (MS) patients. The aims of this pilot study were to investigate the relationship between sSDC-1 and disease severity in NMO and MS and whether sSDC-1 has potential as an effective marker for disease severity. We measured sSDC-1 concentrations by using an enzyme-linked immunosorbent assay (ELISA). NMO patients had significantly higher CSF sSDC-1 levels than MS patients or controls. We also found a positive correlation between the increased CSF sSDC-1 levels and increased severity in NMO disease, but not in MS. In NMO, CSF sSDC-1 concentrations were positively correlated with CSF interleukin (IL)-6, IL-8 and IL-17. Overall, we showed levels of CSF sSDC-1 were higher in NMO patients and had a positive relationship with disease severity of NMO but not with MS. CSF sSDC-1 may be an effective marker of NMO disease severity.
1. Introduction
can be proteolytically released from the cell surface to become a soluble HSPG. This process can be induced by matrix metalloproteinases (MMPs) [6]. In NMO, proinflammatory cytokines such as tumor necrosis factor (TNF)-α, interleukin (IL)-1β, and IL-6 are upregulated [2,3,7] and induce MMPs [6–9]. Moreover, soluble SDC-1 (sSDC-1) is increased by shedding during inflammation and correspondingly membrane bound SDC-1 decreased [4]. The current observations suggest that syndecan-1 shedding has distinct regulatory functions in different inflammatory diseases. For example, syndecan-1 shedding functions as an anti-inflammatory molecule in allergic lung inflammation but an pro-inflammatory molecule in bleomycin-induced acute lung injury by inducing an CXC chemokine gradient mobilization that guides transepithelial efflux of neutrophils [8]. However, the role of sSDC-1 in NMO and MS remains unknown. This pilot study investigated the serum and cerebrospinal fluid (CSF) concentrations of sSDC-1 in NMO and MS patients and explored its role in the pathogenesis of NMO.
NMO and MS are two typical immune-mediated inflammatory demyelinating diseases of the central nervous system. NMO was regarded as a variant of MS until the water channel anti-aquaporin-4 (AQP4) antibodies were discovered. Moreover, differences between MS and NMO such as clinical characteristics, pathology, neuroimaging results, and response to some immunotherapies distinguish these diseases [1]. Many pro-inflammatory and anti-inflammatory cytokines and chemokines contribute to NMO pathogenesis and NMO severity can be regulated by the balance between these opposing factors [2,3]. Syndecan-1 (SDC-1) is a transmembrane HSPG that consists of three different domains (cytoplasmic, transmembrane, and extracellular) and predominantly expressed by epithelial cells and plasmacytes [4]. The main function of SDC-1 in adult mammals is to resolve inflammation. SDC-1 deficiency has been proved to be involved in many diseases with uncontrolled inflammation in vivo and vitro [5]. SDC-1 also takes part in wound healing, fibrosis, tumor biology, and infectious diseases [4]. Through ectodomain shedding, the syndecan extracellular domain
⁎
Corresponding authors at: Department of Neurology, Nanfang Hospital, Southern Medical University, 1838 North Guangzhou Avenue, Guangzhou, Guangdong 510515, China (H. Wang). Department of Neurology, Nanfang Hospital, Southern Medical University, 1838 North Guangzhou Avenue, Guangzhou, Guangdong 510515, China (S. Pan). E-mail addresses: [email protected] (S. Pan), [email protected] (H. Wang). https://doi.org/10.1016/j.cyto.2018.08.017 Received 5 May 2018; Received in revised form 27 July 2018; Accepted 15 August 2018 1043-4666/ © 2018 Elsevier Ltd. All rights reserved.
Cytokine 111 (2018) 140–145
S. Pei et al.
2.3. Study of serum and CSF samples
Table 1 Demographic and clinical features of patients and controls. NMO
MS
Control
No. of subjects (n) Gender (female/male) Age (years) Disease duration (years) EDSS scores
23 15/8 33.43(16–60) 2.24(1–4) 4.22(2–6.5)
12 5/7 34.58(15–52) 1.75(1–2) 2.25(1–3.5)
16 11/5 33.88(27–39) – –
NMO-IgG Positive Negative
19 4
– –
– –
sSDC-1 levels were measured via a commercially available ELISA kit (Diaclone, Besancon, France) according to the manufacturer's instructions. 2.4. Statistical analysis For CSF sSDC-1, serum sSDC-1, IL-6, IL-8 and IL-17, data were presented as mean ± standard deviation or median with range for age, disease duration and EDSS score. The Kruskal-Wallis tests and Spearman’s test were used for the comparison between groups and determining the correlations between sSDC-1 levels and proinflammatory cytokines as well as aquaporin-4 antibodies levels respectively. P-values < 0.05 were considered statistically significant. SPSS 20.0 software was used to perform statistical analyses. Receiver operating characteristic (ROC) curve were performed using GraphPad Prism 7.0 (GraphPad Software, La Jolla, CA, USA). Power analysis were analyzed using BMDP Statistical Software (1993), the following was found: With the proposed sample size of 23 NMO patients and 16 controls, the study will have power of 98.78% to yield a statistically significant result.
Disease duration (years) refers to years from disease onset to sampling. EDSS, Expanded Disability Status Scale; NMO, neuromyelitis optica; MS, multiple sclerosis.
2. Patients and methods 2.1. Patients and controls 23 NMO patients, 12 MS patients and 16 controls with noninflammatory neurological diseases were enrolled. This study was approved by the Ethics Committee of the Nan Fang Hospital of the Southern Medical University. Each enrolled participant signed informed consent. Table 1 shows the demographic and clinical features of the patients.
3. Results 3.1. CSF and serum sSDC-1 in MS and NMO patients and controls The mean plasma sSDC-1 (ng/ml) concentrations for NMO, MS and controls were 99.97 ± 87.25, 74.79 ± 25.98 and 59.61 ± 23.58, respectively. Differences in serum sSDC-1 levels between three groups did not reach statistical significance (Fig. 1A). CSF sSDC-1 levels were significantly higher in the NMO compared with MS and controls (MS, 13.44 ± 6.97 vs 8.19 ± 4.53 ng/ml, p = 0.013; controls,13.44 ± 6.97 vs 5.71 ± 1.79, p < 0.001). The CSF sSDC-1 levels were not significantly different between MS patients and controls (p = 0.189)
2.2. Preparation of blood and CSF samples After sampling, we immediately centrifuged all samples to eliminate cells and other insoluble materials and then distributed equal aliquots into polypropylene tubes for storage at −80 °C until assay.
Fig. 1. Comparison of serum and CSF sSDC1 levels (A, B). (A) No difference was found in serum sSDC-1 levels between NMO, MS and controls. (B) CSF sSDC-1 levels were higher in the NMO compared with MS and controls (MS, p < 0.001; CTLs, p = 0.013). Diagnostic values of sSDC-1 for the assessment of disease activity (C). The 95% confidence interval is from 0.8273 to 1.002. The area under the ROC curve is 0.9144. Comparison of CSF and serum sSDC-1 levels in active and remission periods in NMO (D). In NMO, both serum and CSF soluble SDC-1 levels were significantly higher in the relapse than in remission (CSF, p = 0.013; serum, p = 0.014).
141
Cytokine 111 (2018) 140–145
S. Pei et al.
C S F IL - 6 ( p g /m l)
A p = 0 .0 0 2 2 r= 0 .6 0 5
40 30 20 10 0 0
10
20
30
40
C S F s S D C -1 in N M O (n g /m l)
Fig. 2. Correlation between sSDC-1 and CSF cytokines in NMO and MS (A-F, I-N). In NMO, CSF sSDC-1 concentrations had a significant positive correlation with CSF IL-6 (A), IL-8 (B), and IL-17 (C) (IL-6, p = 0.0022; IL-8, p = 0.0067; IL-17, p = 0.0173). Correlation between sSDC-1 levels and EDSS in NMO and MS patients (G, H, O, P). Plasma sSDC-1 levels were not associated with EDSS scores in NMO (H) and MS (P) patients. CSF sSDC-1 levels in NMO patients were positively correlated with EDSS scores (G) but not in MS patients (O).
142
Cytokine 111 (2018) 140–145
S. Pei et al.
Fig. 2. (continued)
143
Cytokine 111 (2018) 140–145
S. Pei et al.
heparin treatment, a substitute for SDC1, in DSS colitis of SDC1-deficient mice [10]. Through ectodomain shedding, the syndecan extracellular domain can be proteolytically released and become a soluble HSPG [6]. In NMO, CSF cytokines derived from T helper (Th)-1 and Th-17 are increased, primarily TNF-α, IL-6, IL-17, and IL-1β [2,3], which induce MMPs[7,9], a known sheddases of syndecans [6,8,11]. Therefore, sSDC1 levels may be affected by the cytokine network relevant to NMO. As a component of glycocalyx, a gel-like layer on the luminal surface of the endothelium and exerts an barrier role in blood-brain barrier (BBB), the integrity of SDC-1 is important for protecting the integrity of the BBB [12]. It is confirmed that the BBB destruction is a key step in NMO pathogenesis [13]. Thus, SDC-1 shedding breaks the integrity and the protective function of SDC-1 and BBB. Based on the above theoretical basis, we assume that SDC-1 shedding involves in the pathogenesis of NMO. Similarly, previous studies also demonstrate increased levels of this HSPG in other autoimmune inflammatory diseases, such as Crohn’s disease, ulcerative colitis and systemic sclerosis [14,15] and lower sSDC-1 levels were shown in inflammatory bowel disease patients who responded effectively to disease-modifying anti-inflammatory medications compared with untreated patients [15], suggesting a potential role for sSDC-1 in the pathogenesis of autoimmune inflammatory disease. Furthermore, in this study, a positive correlation was found between sSDC-1 and disease severity, indicating a prognostic role of sSDC-1 in NMO. The development of sSDC-1 as a biomarker would permit a more reliable assessment of the BBB destruction in NMO, which has been confirmed to be a pathogenesis in NMO. A further research about sSDC1 in NMO is worthy. In conclusion, this pilot study found the CSF level of sSDC-1 was elevated in NMO patients compared with MS and controls and it had a positive relationship with disease severity, suggesting CSF sSDC-1 as a potential prognostic biomarker that controls underlying inflammation in the pathogenesis of NMO. Further detailed investigations are needed to elucidate the precise function of sSDC-1 in NMO.
(Fig. 1B). 3.2. CSF and serum sSDC-1 in active and remission periods in NMO Serum sSDC-1 in 11 NMO patients and CSF sSDC-1 in 10 NMO patients were measured in both relapse and remission stage. Both serum and CSF sSDC-1 levels were significantly higher in relapse than in remission (CSF p = 0.013; serum p = 0.014) (Fig. 1D). 3.3. CSF and serum sSDC-1 and disease activity For NMO, the median EDSS score was 4.2 (2.0–6.5), and for MS 2.3 (1.0–3.5). In both cohorts, no significant correlation was found between plasma sSDC-1 and EDSS score (Fig. 2H, P). NMO patients with higher EDSS scores showed higher CSF sSDC-1 concentrations during relapse (p < 0.0001) (Fig. 2G) but this result was not found in MS patients (Fig. 2O) (p = 0.563). 3.4. CSF proinflammatory cytokines, aquaporin-4 antibodies and their relationships with sSDC-1 levels Mean CSF IL-6 (pg/ml) of NMO, MS patients and controls were 10.62 ± 7.97, 5.69 ± 2.67 and 2.53 ± 0.43. Mean CSF IL-8 (pg/ml) of NMO, MS patients and controls were 236.90 ± 285.07, 87.69 ± 43.10 and 54.39 ± 16.87. Mean CSF IL-17 (pg/ml) of NMO, MS patients and controls were 9.73 ± 9.54, 5.75 ± 3.75 and 3.84 ± 1.86. Elevated serum IL-6, IL-8 and IL-17 levels were demonstrated in NMO compared with MS and controls (IL-6, p = 0.02, p < 0.001, respectively; IL-8, p = 0.017, p < 0.001, respectively; IL17, p = 0.041, p < 0.001, respectively). Higher IL-6, IL-8 and IL-17 levels were found in MS patients compared with controls (p < 0.001, p = 0.001, p = 0.01, respectively). In NMO, strong positive correlations were found between CSF sSDC-1 and IL-6 (p = 0.0022), IL-8 (p = 0.0067) and IL-17 (p = 0.0173) (Fig. 2A–C). But no positive relationship was found between serum sSDC-1 and cytokines except CSF IL-6 (p = 0.0270) (Fig. 2D). And no relationship was found between CSF aquaporin-4 antibodies and sSDC-1 in both relapse and remission stage (serum, p = 0.227, p = 0.280; CSF, p = 0.961, p = 0.906). The CSF and serum sSDC-1 concentrations also appeared to be associated with CSF cytokines concentrations in the MS group, though they did not quite reach statistical significance. (Fig. 2I–N).
5. Disclosure of conflict of interest No conflict of interest. Acknowledgments Supported by the National Natural Science Foundation of China (81673950). Guangdong provincial science and technology plan projects (2016A020215101, 2017A020215182). Natural Science Foundation of Guangdong Province of China (2016A030313828).
3.5. ROC curve of NMO and CTLs The 95% confidence interval of the ROC curve of NMO and CTLs was from 0.8273 to 1.002. The area under the ROC curve (AUC) was 0.9144, p < 0.001 (Fig. 1C).
References 4. Discussion [1] A.A. Argyriou, N. Makris, Neuromyelitis optica: a distinct demyelinating disease of the central nervous system, Acta Neurol. Scandinavica 118 (2008) 209–217. [2] K.C. Wang, C. Lee, S. Chen, J. Chen, C. Yang, S. Chen, C. Tsai, Distinct serum cytokine profiles in neuromyelitis optica and multiple sclerosis, J. Interferon Cytokine Res. 33 (2013) 58–64. [3] T. Matsushita, T. Tateishi, N. Isobe, T. Yonekawa, R. Yamasaki, D. Matsuse, H. Murai, J. Kira, Characteristic cerebrospinal fluid cytokine/chemokine profiles in neuromyelitis optica, Relapsing Remitting or Primary Progressive Multiple Sclerosis: e61835, 2013; 8. [4] Y.H. Teng, R.S. Aquino, P.W. Park, Molecular functions of syndecan-1 in disease, Matrix Biol. 31 (2012) 3–16. [5] A.H. Bartlett, K. Hayashida, P.W. Park, Molecular and cellular mechanisms of syndecans in tissue injury and inflammation, Mol. Cells 24 (2007) 153–166. [6] S. Brule, N. Charnaux, A. Sutton, D. Ledoux, T. Chaigneau, L. Saffar, L. Gattegno, The shedding of syndecan-4 and syndecan-1 from HeLa cells and human primary macrophages is accelerated by SDF-1/CXCL12 and mediated by the matrix metalloproteinase-9, Glycobiology 16 (2006) 488–501. [7] A. Sebestyén, M. Gallai, T. Knittel, T. Ambrust, G. Ramadori, I. Kovalszky, Cytokine regulation of syndecan expression in cells of liver origin, Cytokine 12 (2000) 1557–1560. [8] Q. Li, P.W. Park, C.L. Wilson, W.C. Parks, Matrilysin shedding of syndecan-1 regulates chemokine mobilization and transepithelial efflux of neutrophils in acute
This pilot study investigated the relationship between sSDC-1 and the inflammatory demyelinating diseases. Until now, this is the first study to examine sSDC-1 levels in NMO patients. We found higher CSF sSDC-1 concentrations in NMO compared with MS and controls but serum sSDC-1 concentrations were not significantly different in these three groups. This maybe because of the immune-mediated inflammatory demyelinating role of NMO and MS in central nervous system. The increased CSF sSDC-1 levels in NMO patients could be attributed to accelerated SDC1 shedding. Thus, CSF sSDC-1 levels played a reasonable diagnostic role in distinguishing NMO patients from MS patients and controls. More importantly, there was also a borderline positive correlation between CSF and plasma sSDC-1 levels and concentrations of CSF proinflammatory cytokines IL-6, IL-8 and IL-17, and EDSS scores in NMO patients. Syndecan-1 (SDC-1) is a transmembrane HSPG. Its main function is to resolve inflammation [4]. Floer et al. reported a protective role of 144
Cytokine 111 (2018) 140–145
S. Pei et al.
lung injury, Cell 111 (2002) 635–646. [9] X. Gan, B. Wong, S.D. Wright, T. Cai, Production of matrix metalloproteinase-9 in CaCO-2 cells in response to inflammatory stimuli, J. Interferon Cytokine Res. 21 (2001) 93–98. [10] M. Floer, M. Götte, M.K. Wild, J. Heidemann, E.S. Gassar, W. Domschke, L. Kiesel, A. Luegering, T. Kucharzik, Enoxaparin improves the course of dextran sodium sulfate-induced colitis in syndecan-1-deficient mice, Am. J. Pathol. 176 (2010) 146–157. [11] R. Day, Regulation of epithelial syndecan-1 expression by inflammatory cytokines, Cytokine 21 (2003) 224–233. [12] J. Zhu, X. Li, J. Yin, Y. Hu, Y. Gu, S. Pan, Glycocalyx degradation leads to blood–brain barrier dysfunction and brain edema after asphyxia cardiac arrest in
rats, J. Cereb. Blood Flow Metab. (2017) 0271678X1772606. [13] F. Shimizu, T. Kanda, Disruption of the blood-brain barrier in inflammatory neurological diseases, Brain Nerve 65 (2013) 165–176. [14] C. Çekiç, A. Kırcı, S. Vatansever, F. Aslan, H.E. Yılmaz, E. Alper, M. Arabul, E. Sarıtaş Yüksel, B. Ünsal, Serum syndecan-1 levels and its relationship to disease activity in patients with Crohn’s disease, Gastroenterol. Res. Practice 2015 (2015) 1–6. [15] D. Yablecovitch, A. Stein, M. Shabat-Simon, T. Naftali, G. Gabay, I. Laish, A. Oren, F.M. Konikoff, Soluble syndecan-1 levels are elevated in patients with inflammatory bowel disease, Dig. Dis. Sci. 60 (2015) 2419–2426.
145