Distinct Serum Cytokine Profiles in Neuromyelitis Optica and Multiple Sclerosis

Abstract

Multiple sclerosis (MS) is the most common prototypic inflammatory demyelinating disease. Neuromyelitis optica (NMO) is another inflammatory demyelinating disease of the central nervous system that exhibits clinical symptoms mainly associated with optic neuritis and myelopathy. The inflammatory reaction in MS is associated with an upregulation of a variety of T helper 1 (Th1)- or Th17-mediated cytokines. However, NMO and MS are intertwined both clinically and pathologically, which complicates their diagnosis and treatment. The aim of this study was to evaluate the differences in serum cytokine levels in patients with NMO and MS. We collected peripheral serum from patients with these central nervous system demyelinating diseases for the study. A cytometric bead array was used to assess the cytokine levels using flow cytometry. We found more inflammatory [interleukin (IL)-2 and interferon-γ) and anti-inflammatory (IL-4 and IL-10) cytokines in NMO than in MS. The differences in the optimal cutoff points of serum cytokines, including IL-2 ≥5 pg/mL, can differentiate NMO from MS. In conclusion, patients with NMO had an increased Th1-mediated inflammatory response, but similar Th17-mediated inflammation changes compared to patients with MS. Serum cytokine studies can differentiate NMO cases from MS.

Introduction

M ultiple sclerosis (MS) is the most common prototypic inflammatory demyelinating disease (Hu and Lucchinetti 2009). It typically starts with sensory disturbances, unilateral optic neuritis, diplopia (internuclear ophthalmoplegia), limb weakness, gait ataxia, and neurogenic bladder and bowel symptoms (Noseworthy and others 2000). Neuromyelitis optica (NMO) exhibits clinical symptoms mainly associated with optic neuritis and myelopathy (Adorini and Trembleau 1997; Wingerchuk and others 2007). MS was incorrectly diagnosed in many patients with NMO (Adorini and Trembleau 1997; Wingerchuk and others 1999; de Seze and others 2003; Cree 2008). However, the clinical course of NMO has been shown to be more severe than that of MS (Adorini and Trembleau 1997; Cornelio and others 2009), and the presence of anti-aquaporin-4 antibodies (AQP4 Ab) or NMO-IgG is accentuated in the disease course of NMO (Adorini and Trembleau 1997; Wingerchuk and others 2006; Saiz and others 2007). Despite the reported differences, NMO and MS are still interwined both clinically and pathologically.

The primary effector T cell in the pathology of MS was originally thought to be a CD4+ T helper 1 (Th1) cell (McFarland and others 1992; Adorini and Trembleau 1997; Sinigaglia and others 1997). The inflammatory reaction in MS is indeed associated with upregulation of several Th1-mediated cytokines, including interleukin (IL)-2, interferon (IFN)-γ, and tumor necrosis factor (TNF)-α (Aktas and others 2009). In contrast, CD4+ T helper 2 (Th2) cells produce IL-4, IL-6, and IL-10 to mediate Ab production, and downregulate Th1 cellular responses (Adorini and Trembleau 1997). Thus, one of the major goals of IFN-β treatment for MS is to shift the Th1 response to the Th2 response to improve the patient's prognosis (Nicholson and Kuchroo 1996; Kozovska and others 1999). MS has also been considered by some as a CD4+ T helper 17 (Th17) cell-mediated disease recently, but the matter is still under debate (Matusevicius and others 1999). NMO with Ab and complement activation is considered as a humoral response (ie, Th2 cells) predominates (Correale and Fiol 2004), although a major controversy exists with regard to this particular point of view. It has been suggested that NMO is instead a Th1- or Th17-predominated disease (Horiuchi and others 2000; Wu and others 2000; Li and others 2011). To investigate this issue, we conducted a study to survey the cytokine changes that occur in both MS and NMO.

Materials and Methods

Patients

Eighty patients with central nervous system (CNS) demyelinating disorder with a history of more than 2 relapses and followed up at the MS Clinic in Taipei Veterans General Hospital, Taipei, Taiwan, from 2007 to 2008 were consecutively enrolled in our study. The final diagnosis was made according to the continuous 2-year clinical and serial imaging follow-up. At last, a total of 34 patients fulfilled the diagnostic criteria of NMO (Wingerchuk and others 2006). Briefly, optic neuritis and acute long-spinal-cord myelitis (>3 segments), but without traditional MS imaging patterns, constituted the diagnosis of NMO. Exclusion criteria included Sjögren's syndrome or systemic lupus erythematosus-related neurological disorder, recurrent myelitis, or recurrent optic disease. Twenty-four patients were found to have MS using the Poser criteria with 2 clinical attacks, and by not fulfilling the diagnosis of NMO and related disorders (Poser and others 1983). The remaining 20 patients were excluded because of an uncertain diagnosis of NMO or MS based on the clinical course, such as recurrent optic neuritis or myelitis. Thirty healthy controls were also enrolled. Blood samples were drawn at the initial enrollment and >1 month after any previous episode, that is, during disease remission. Since circulating adhesion molecules and inflammatory cytokines were greatly elevated during the acute phase of the demyelination process, whether in MS or NMO (Moreau and others 1996; Li and others 2011), we wanted to clarify the state of cytokine expression during remission, to provide distinguishable data for reference.

This study was approved by the Institutional Review Board of Taipei Veterans General Hospital. All patients were subjected to a thorough neurological examination, routine laboratory tests, and clinical evaluation at regular intervals in the clinic. Clinical characteristics of the patients, including current age, age at symptom onset, and annual relapse rate, were recorded and analyzed together with the results of the Expanded Disability Status Scale (EDSS) disability score at the 2-year follow-up.

Anti-AQP4 study method

Sera were harvested and stored at −20°C. Serum AQP4 Ab was detected using a cell-based assay with AQP4-transfected HEK293 cells, as previously described by Matsuoka and others (2007).

Cytometric bead array immunoassay of Th1, Th2, and Th17 cytokines

Serum samples were stored at −80°C for the cytometric bead array (CBA) analysis. The CBA technique is based on microparticles, which are dyed to 7 different fluorescence (FL) intensities, matched with Abs. The dye has a maximum emission wavelength of ∼650 nm (FL-3). The particles were matched by covalent linkage with Abs (Pharmingen, SanDiego, CA) against each of the 7 cytokines (IFN-γ, TNF-α, IL-2, IL-4, L-6, IL-10, and IL-17). The Ab particles served to capture each cytokine, and the cytokines were directly detected using an immunoassay with 7 different Abs matched with phycoerythrin (PE), which emits at 585 nm (FL-2). The PE-conjugated detector Ab was used to complete the sandwich, and the medium intensity in FL-2 was proportional to the concentration of the cytokine in the sample, which was quantified using a calibration curve. Seven standard curves (ranging from 0 to 5,000 pg/mL) were obtained from the set of calibrators, and 7 results were obtained for test samples.

Serum samples from the MS and NMO controls were incubated with each Ab-bead reagent and Ab-PE detector in the dark at room temperature for 160 min, briefly washed, and subjected to flow cytometry for data acquisition. Two-color cytometric analysis was carried out using an FACSCaliber flow cytometer (BD Biosciences) in the Medical Institute of the National Defense Medical Center, Taipei, Taiwan. Data were obtained and analyzed with CBA software.

Statistical interpretation

Due to the small sample size in the 2 disease groups, including MS (n=24) versus NMO (n=34) and NMO with IFN (n=20) versus NMO without IFN (n=14), continuous data were presented using the median and interquartile range (the range between the 25th and 75th percentile), and the differences between the 2 groups were tested with the nonparametric Mann–Whitney test. Categorical data were all presented using count and percentage, and the differences between the 2 groups were tested with the Fisher's exact test. The associations between the cytokines and the disease groups were described using odds ratios (ORs) with 95% confidence intervals (CI) from the univariate logistic regression analyses, and multivariate logistic regression analyses were conducted to evaluate the independent associations, controlling for EDSS score and annualized relapse rate. The area under the receiver operating characteristic (ROC) curve (AUC) was determined to evaluate the diagnostic value of the cytokines, and the optimal cutoff point of the ROC curves was determined using the Youden index (sensitivity+specificity −1). A P value<0.05 represented statistical significance for all statistical hypothesis tests. Statistical analyses were performed using SPSS 15.0 software (SPSS, Inc.).

Results

Baseline characteristics

The 24 patients with MS, 4 men and 20 women, had a median age of 42.0 years, and the 34 patients with NMO, 2 men and 32 women, a median age of 47.0 years. There was no significant difference in age or gender between the patients with MS and patients with NMO. The patients with NMO had significantly higher EDSS scores and annualized relapse rates than the patients with MS (median EDSS: 4.5 versus 2.5, P=0.001; annualized relapse rate: 1.0 versus 0.6, P<0.001). No significant difference was observed in the treatment of IFN between the 2 disease groups. In addition, 15 (44.1%) patients with NMO, but no patients with MS, were positive for AQP4 Ab (P<0.001) (Table 1). The 30 enrolled healthy controls included 1 man and 29 women, with a median age of 43.0 years, and none showed the existence of AQP4 Ab.

Table 1.

Patient Characteristics and Treatment

		MS (n = 24 )	NMO (n = 34 )	P-value
Age (year)		42.0 (32.5, 53.0)	47.0 (36.0, 58.0)	0.162
Gender	Male	4 (16.7%)	2 (5.9%)	0.220
	Female	20 (83.3%)	32 (94.1%)
Annualized relapse rate		0.6 (0.5, 0.8)	1.0 (0.8, 2.0)	<0.001^*
EDSS		2.5 (0.5, 4.3)	4.5 (2.5, 5.5)	0.001^*
IFN use	IFN	15 (62.5%)	20 (58.8%)	1.000
	No IFN	9 (37.5%)	14 (41.2%)
AQP4 antibody	Positive	0 (0.0%)	15 (44.1%)	<0.001^*
	Negative	24 (100.0%)	19 (55.9%)

Indicates a significant difference as compared to the MS patients.

MS, multiple sclerosis; NMO, neuromyelitis optica; IFN, interferon-β.

Serum concentration of cytokines

Patients with NMO had significantly higher levels of serum concentrations of IL-4 and IL-10 than patients with MS (median IL-4: 3.6 versus 0 pg/mL, P=0.013; median IL-10: 7.0 versus 2.9 pg/mL, P=0.038) (Fig. 1A). Patients with NMO also had significantly higher levels of serum concentrations of IL-2 and IFN-γ than patients with MS (median IL-2: 5.4 versus 0 pg/mL, P=0.006; median IFN-γ: 6.4 versus 0 pg/mL, P=0.019). No significant difference was observed in TNF-α and IL-17 levels (Fig. 1B). However, the cytokine surveys in both groups showed significantly higher levels than those of the healthy controls (Supplementary Table S1; Supplementary Data are available online at www.liebertpub.com/jir).

FIG. 1.

Analysis of serum cytokines of multiple sclerosis and neuromyelitis optica (NMO). (A) The serum concentrations of interleukin (IL)-4 and IL-10 were significantly higher in NMO. (B) The serum concentrations of IL-2 and interferon (IFN)-γ were significantly higher in NMO. *A significant difference between the 2 groups.

Associations between cytokines and diseases

The univariate logistic study showed that both gender and age were insignificant predictors for MS/NMO. The 4 ORs in the univariate logistic regression analyses of IL-4 (OR=1.14), IL-10 (OR=1.10), IL-2 (OR=1.24), and IFN-γ (OR=1.12) reached a statistically significant level. In the multivariate analyses, independent associations with the disease groups were observed for the IL-2 and IFN-γ cytokines, which implies that those with higher levels of IL-2 and IFN-γ were more likely to have NMO, rather than MS, at the same EDSS and annualized relapse rate, with ORs of 1.36 and 1.15. (Table 2).

Table 2.

Associations Between Cytokines and Diseases (MS and NMO)

	Univariate		Multivariate
	OR (95% CI)	P-value	OR ^a (95% CI)	P-value
Age	1.03 (0.99, 1.08)	0.163
Sex^b	3.2 (0.54, 19.1)	0.202
IL-4	1.14 (1.02, 1.28)	0.027^*	1.16 (0.97, 1.38)	0.102
IL-6	0.995 (0.94, 1.06)	0.872
IL-10	1.10 (1.00, 1.20)	0.044^*	1.12 (0.97, 1.29)	0.116
IL-2	1.24 (1.06, 1.45)	0.007^*	1.36 (1.04, 1.77)	0.024^*
TNF-α	1.04 (0.96, 1.14)	0.337
IFN-γ	1.12 (1.02, 1.24)	0.023^*	1.15 (1.01, 1.32)	0.040^*
IL-17	0.998 (0.98, 1.02)	0.811

With adjustments for annualized relapse rate and EDSS.

Female as reference group.

Indicates a significant odds ratio, implies those with higher levels of the corresponding variable were more likely to be NMO patients.

Diagnostic value of cytokines

The AUC for diagnosing NMO by IL-2 reached 0.701, with a 95% CI of 0.566–0.814, and that by IFN-γ was 0.674 with a 95% CI of 0.538–0.791, and that by IL-2 and IFN-γ together was 0.7000 with a 95% CI=0.551–0.854 (Fig. 2).

FIG. 2.

The area under the receiver operating characteristic (ROC) curves (AUC) for diagnosing NMO by cytokine. (A) IL-2 reached 0.701 with a 95% confidence interval (CI) of 0.566–0.814, (B) IFN-γ was 0.674 with a 95% CI of 0.538–0.791, and (C) IL-2 and IFN-γ together were 0.7000 with a 95% CI=0.551–0.854.

The optimal cutoff points of IL-2 ≥5 pg/mL and IFN-γ ≥6 pg/mL were determined by the Youden index. Sensitivity reached 52.9% when using the optimal cutoff points of IL-2 and IFN-γ (ie, 18 of the 34 patients with NMO would be detected), which was higher than the 44.1% when using the AQP4 Ab. Specificity was 91.7% and 75.0% when using the optimal cutoff points of IL-2 and IFN-γ (ie, 22 and 18 of the 24 patients with MS would be detected), which was lower than the 100% when using the AQP4 Ab. Positive predictive values (PPVs) of IL-2 ≥5 pg/mL and IFN-γ ≥6 pg/mL were lower than that of AQP4 Ab (90% and 75% versus 100%). The negative predictive value (NPV) of IL-2 <5 pg/mL was lower than that of AQP4 Ab (42.1% versus 44.2%), but the NPV of IFN-γ <6 pg/mL was higher than that of AQP4 Ab (47.1% versus 44.2%) (Table 3). Furthermore, the combination of IL-2 ≥5 pg/mL and IFN-γ ≥6 pg/mL as a cutoff point showed an 88.9% PPV, but a higher (55.0%) NPV, compared to the others.

Table 3.

Summary of Sensitivity and Specificity of IL-2, IFN-γ, and AQP-4

		NMO (n = 34 )	MS (n = 24 )	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)
IL-2^*	≥5 pg/ml: Positive (NMO)	18	2	52.9	91.7	90.0	42.1
	<5 pg/ml: Negative (MS)	16	22
IFN-γ^*	≥6 pg/ml: Positive (NMO)	18	6	52.9	75.0	75.0	47.1
	<6 pg/ml: Negative (MS)	16	18
IL-2	≥5 pg/ml and IFN-γ ≥6 pg/ml: Positive (NMO)	16	2	47.1	91.7	88.9	55.0
	Otherwise: Negative (MS)	18	22
AQP4	Positive (NMO)	15	0	44.1	100.0	100.0	44.2
	Negative (CMS)	19	24

The optical cutoff point was determined by the Youden index (sensitivity + specificity −1).

PPV, positive predictive value; NPV, negative predictive value.

Comparisons between the use and nonuse of IFN in patients with NMO

There is a lot of debate about the use of IFN treatment in NMO. Thus, our patients with NMO were divided into 2 groups, those with (n=20) and without (n=14) IFN treatment, to compare the clinical pictures and related cytokine changes. There was no significant difference in age, gender, or annualized relapse rate between the 2 groups. However, a higher EDSS was noted in the NMO-with-IFN treatment group. The patients with NMO using IFN had significantly higher levels of IL-17 cytokines (median IL-17: 28.8 versus 20.2 pg/mL), but lower levels of IL-2 cytokines (median IL-2: 3.0 versus 11.1 pg/mL) than those not using IFN (Table 4).

Table 4.

Comparisons Between IFN Use and Non-Use in NMO Patients

		IFN− (n = 14)	IFN+ (n = 20)	P-value
Age (year)		39.5 (32.3, 50.0)	57.0 (42.0, 58.0)	0.071
Gender	Male	1 (7.7%)	1 (5.0%)	0.661
	Female	13 (92.3%)	19 (95.0%)
Annualized relapse rate		1.3 (0.8, 2.1)	1.0 (0.7, 1.2)	0.158
EDSS		3.0 (2.4, 4.5)	5.0 (3.5, 6.5)	0.009^*
Cytokines
IL-4 (pg/ml)		8.4 (0.0, 17.9)	3.1 (0.0, 6.7)	0.104
IL-6 (pg/ml)		5.7 (0.0, 14.6)	3.8 (0.0, 6.6)	0.416
IL-10 (pg/ml)		9.5 (3.1, 21.1)	5.5 (3.4, 11.4)	0.204
IL-2 (pg/ml)		11.1 (0.0, 13.1)	3.0 (0.0, 6.7)	0.036^*
TNF-α (pg/ml)		5.3 (0.0, 14.7)	0.0 (0.0, 4.9)	0.056
INF-γ (pg/ml)		10.1 (0.0, 21.5)	4.1 (0.0, 9.2)	0.129
IL-17 (pg/ml)		20.2 (13.3, 28.1)	28.8 (27.0, 29.5)	0.007^*

Indicates a significant difference between the 2 groups.

IFN−, without IFN use; IFN+, with IFN use.

IL-6 correlated with EDSS in NMO

We found that the initial IL-6 expression correlated well with the later EDSS scores in the NMO group (Fig. 3).

FIG. 3.

The initial IL-6 expression correlated well with the later Expanded Disability Status Scale scores in the NMO group.

Discussion

In this study, we found increased inflammation in patients with NMO compared to patients with MS, based on a survey of serum cytokine levels. Mutual participation of Th1, Th2, and Th17 cytokines was found in both patient groups in the analysis. In general, proinflammatory (IL-2 and IFN-γ) and anti-inflammatory (IL-4 and IL-10) cytokines were increased in patients with NMO compared to patients with MS. Furthermore, we could use the optimal cutoff point of IL-2 ≥5 pg/mL to differentiate NMO cases from MS cases in the serum cytokine survey.

The levels of all study cytokines in each group were significantly higher than in the controls, indicating the mutual participation of Th1, Th2, and Th17 responses in NMO or MS. This supports past findings and suggests the complexity of autoimmune diseases with the involvement of multiple mechanisms (Amerio and others 2002), and also supports the involvement of IL-17 in the pathogenic mechanisms of NMO (Li and others 2011; Wang and others 2011a), which may be considered as a spectrum of MS. However, in our study, the Th17 response showed no significant change in these 2 diseases; therefore, we plan to perform future extensive studies, including a study of TH17-associated cytokines, to differentiate NMO from MS.

Inflammatory mechanisms play an important role in MS lesions. For example, the overexpression of several proinflammtory cytokines, such as IFN-γ, is reportedly related to attacks that occur in patients with MS (Frohman and others 2006; Ubogu and others 2006). A correlation between EDSS scores and IFN-γ levels was noted in a previous MS study (Petereit and others 2000). The observation regarding our patients with NMO is consistent with that of reports of clinically frequent recurrent disease and disabilities more severe than MS (Chan and others 2009). The increased IFN-γ levels in our patients with NMO accounted for their greater clinical severity compared to patients with MS. The increased inflammation in patients with NMO suggests that clinicians should use more potent anti-inflammatory drugs and/or more intense treatment for NMO than for MS.

NMO expressed more humoral/Th2 responses, compared to MS, with elevated IL-4 and IL-10 levels. Existing evidence shows that both T-cell subsets and B cells can induce autoimmunity in the CNS (Caspi and others 2008; McLaughlin and Wucherpfennig 2008). Abs to both myelin–oligodendrocyte glycoprotein and myelin basic protein in the serum of patients with MS (Wajgt and Gorny 1983; Bernard and de Rosbo 1991) and AQP4 Ab in NMO patients are particularly expressed (Wingerchuk 2006). The humoral response involved in NMO is not only documented from serum study, but also supported by the clinical evidence of a good response to plasma exchange (Watanabe and others 2007; Wang and others 2011b).

Some reports suggested that IFN was not effective in reducing the relapse number and the disability progression in NMO (Tanaka and others 2009; Uzawa and others 2010). IFN may even trigger severe exacerbation in NMO cases (Warabi and others 2007), and increase the production of AQP4 Ab (Palace and others 2010). Our study found increased EDSS scores in our patients with NMO using IFN, consistent with past reports. The patients with NMO with elevated IL-17 and using IFN had more disability, compatible with recent reports that a higher IL-17 level is associated with nonresponsiveness to IFN treatment. (Axtell and others 2010).

Furthermore, we found that the IL-6 concentration correlates even more with the EDSS score in NMO (Fig. 3). It has previously been shown that elevated IL-6 was found in the cerebrospinal fluid of patients with NMO (Hamnik and others 2008; Uzawa and others 2009). IL-6 is a proinflammatory cytokine secreted by Th2 cells that causes potent B cell stimulation (Diehl and Rincon 2002; Correale and Fiol 2004). These findings collectively imply that IL-6 may be a candidate biomarker for active CNS autoimmune inflammatory diseases, especially in NMO.

The immunopathogenesis of both MS and NMO has been shown to be heterogeneous. As such, the results of our ROC analysis herein indicate that the optimal cutoff points of IL-2 ≥5 pg/mL may represent a good biomarker for discriminating NMO from MS. Such a biomarker may provide physicians with an overall direction, even without using AQP4 Ab or NMO IgG, when diagnosing NMO.

However, there are some limitations to our study. First, the titer of AQP4 Ab was not investigated in our study; thus, we could not correlate the level of cytokines with the titer of AQP4 Ab, which may be of benefit in studying the role of AQP4 Ab in NMO. Second, we did not check the cytokine changes during the active attack period compared to usual, which could provide more clues about the dynamic change. Third, the sample size in this study was small, but did demonstrate significant inflammatory responses in NMO. In the future, we plan to collect more study subjects to seek out more clues to help differentiate these 2 diseases at the outset.

CNS inflammatory demyelinating disease represents a broad spectrum of disorders, including prototypic MS, NMO, Balo's concentric sclerosis, and several others. There are many symptomatic similarities among these disorders, leading to uncertainty in the clinical diagnosis and even diverseness in treatment. Our study has shed some light on the underlying difference between MS and NMO, and thereby may be of help in providing more effective treatment.

In conclusion, the NMO patient group showed increased inflammation, and an anti-inflammatory response, compared to the MS patient group. Furthermore, the serum cytokine study suggested that the presence of IL-2 ≥5 pg/mL in patients with these CNS demyelinating disorders can differentiate NMO cases from MS when clinically obscure at the initial. It is our hope that our present finding, in combination with the findings of inflammatory inhibitors, might facilitate greater clinical efficacy in treating NMO by modifying the underlying immune response.

Footnotes

Acknowledgments

This work was supported by Research Grants of the Cheng Hsin General Hospital, Taipei, Taiwan.

Patient Consent

We confirm that we obtained patient consent.

Provenance and Peer Review

Not commissioned; externally peer reviewed.

Author Disclosure Statement

We declare no competing financial interests exist.

References

Adorini

, Trembleau

. 1997. Immune deviation towards Th2 inhibits Th-1-mediated autoimmune diabetes. Biochem Soc Trans, 25,2:625–629.

Aktas

, Ciftci

, Bilgic

, Sezer

, Bozkanat

, Deniz

, Citici

, Deniz

. 2009. Peripheral immune response in pulmonary tuberculosis. Scand J Immunol, 70,3:300–308.

Amerio

, Frezzolini

, Abeni

, Teofoli

, Girardelli

, De Pita

, Puddu

. 2002. Increased IL-18 in patients with systemic lupus erythematosus: relations with Th-1, Th-2, pro-inflammatory cytokines and disease activity. IL-18 is a marker of disease activity but does not correlate with pro-inflammatory cytokines. Clin Exp Rheumatol, 20,4:535–538.

Axtell

, de Jong

, Boniface

, van der Voort

, Bhat

, De Sarno

, Naves

, Han

, Zhong

, Castellanos

, Mair

, Christakos

, Kolkowitz

, Katz

, Killestein

, Polman

, de Waal Malefyt

, Steinman

, Raman

. 2010. T helper type 1 and 17 cells determine efficacy of interferon-beta in multiple sclerosis and experimental encephalomyelitis. Nat Med, 16,4:406–412.

Bernard

, de Rosbo

. 1991. Immunopathological recognition of autoantigens in multiple sclerosis. Acta Neurol (Napoli), 13,2:171–178.

Caspi

, Silver

, Luger

, Tang

, Cortes

, Pennesi

, Mattapallil

, Chan

. 2008. Mouse models of experimental autoimmune uveitis. Ophthalmic Res, 40,3–4:169–174.

Chan

, Ramsden

, Yu

, Kwok

, Chu

, Ho

, Kwan

, Lee

, Lim

, Kung

, Ho

. 2009. Neuromyelitis optica-IgG in idiopathic inflammatory demyelinating disorders amongst Hong Kong Chinese. Eur J Neurol, 16,3:310–316.

Cornelio

, Braga

, Rosa

, Ayub

. 2009. Devic's neuromyelitis optica and pregnancy: distinction from multiple sclerosis is essential. Arch Gynecol Obstet, 280,3:475–477.

Correale

, Fiol

. 2004. Activation of humoral immunity and eosinophils in neuromyelitis optica. Neurology, 63,12:2363–2370.

10.

Cree

. 2008. Neuromyelitis optica: diagnosis, pathogenesis, and treatment. Curr Neurol Neurosci Rep, 8,5:427–433.

11.

de Seze

, Lebrun

, Stojkovic

, Ferriby

, Chatel

, Vermersch

. 2003. Is Devic's neuromyelitis optica a separate disease? A comparative study with multiple sclerosis. Mult Scler, 9,5:521–525.

12.

Diehl

, Rincon

. 2002. The two faces of IL-6 on Th1/Th2 differentiation. Mol Immunol, 39,9:531–536.

13.

Frohman

, Racke

, Raine

. 2006. Multiple sclerosis—the plaque and its pathogenesis. N Engl J Med, 354,9:942–955.

14.

Hamnik

, Hacein-Bey

, Biller

, Gruener

, Jay

. 2008. Neuromyelitis optica (NMO) antibody positivity in patients with transverse myelitis and no visual manifestations. Semin Ophthalmol, 23,3:191–200.

15.

Horiuchi

, Kawano

, Yamasaki

, Minohara

, Furue

, Taniwaki

, Miyazaki

, Kira

. 2000. Th1 dominance in HAM/TSP and the optico-spinal form of multiple sclerosis versus Th2 dominance in mite antigen-specific IgE myelitis. J Neurol Sci, 172,1:17–24.

16.

, Lucchinetti

. 2009. The pathological spectrum of CNS inflammatory demyelinating diseases. Semin Immunopathol, 31,4:439–453.

17.

Kozovska

, Hong

, Zang

, Li

, Rivera

, Killian

, Zhang

. 1999. Interferon beta induces T-helper 2 immune deviation in MS. Neurology, 53,8:1692–1697.

18.

, Wang

, Long

, Lu

, Hu

. 2011. Increased memory Th17 cells in patients with neuromyelitis optica and multiple sclerosis. J Neuroimmunol, 234,1–2:155–160.

19.

Matsuoka

, Matsushita

, Kawano

, Osoegawa

, Ochi

, Ishizu

, Minohara

, Kikuchi

, Mihara

, Ohyagi

, Kira

. 2007. Heterogeneity of aquaporin-4 autoimmunity and spinal cord lesions in multiple sclerosis in Japanese. Brain, 130,Pt 5:1206–1223.

20.

Matusevicius

, Kivisakk

, He

, Kostulas

, Ozenci

, Fredrikson

, Link

. 1999. Interleukin-17 mRNA expression in blood and CSF mononuclear cells is augmented in multiple sclerosis. Mult Scler, 5,2:101–104.

21.

McFarland

, Frank

, Albert

, Smith

, Martin

, Harris

, Patronas

, Maloni

, McFarlin

. 1992. Using gadolinium-enhanced magnetic resonance imaging lesions to monitor disease activity in multiple sclerosis. Ann Neurol, 32,6:758–766.

22.

McLaughlin

, Wucherpfennig

. 2008. B cells and autoantibodies in the pathogenesis of multiple sclerosis and related inflammatory demyelinating diseases. Adv Immunol, 98:121–149.

23.

Moreau

, Coles

, Wing

, Isaacs

, Hale

, Waldmann

, Compston

. 1996. Transient increase in symptoms associated with cytokine release in patients with multiple sclerosis. Brain, 119,Pt 1:225–237.

24.

Nicholson

, Kuchroo

. 1996. Manipulation of the Th1/Th2 balance in autoimmune disease. Curr Opin Immunol, 8,6:837–842.

25.

Noseworthy

, Lucchinetti

, Rodriguez

, Weinshenker

. 2000. Multiple sclerosis. N Engl J Med, 343,13:938–952.

26.

Palace

, Leite

, Nairne

, Vincent

. 2010. Interferon beta treatment in neuromyelitis optica: increase in relapses and aquaporin 4 antibody titers. Arch Neurol, 67,8:1016–1017.

27.

Petereit

, Richter

, Pukrop

, Bamborschke

. 2000. Interferon gamma production in blood lymphocytes correlates with disability score in multiple sclerosis patients. Mult Scler, 6,1:19–23.

28.

Poser

, Paty

, Scheinberg

, McDonald

, Davis

, Ebers

, Johnson

, Sibley

, Silberberg

, Tourtellotte

. 1983. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol, 13,3:227–231.

29.

Saiz

, Zuliani

, Blanco

, Tavolato

, Giometto

, Graus

. 2007. Revised diagnostic criteria for neuromyelitis optica (NMO). Application in a series of suspected patients. J Neurol, 254,9:1233–1237.

30.

Sinigaglia

, Sturniolo

, Raddrizzani

, Hammer

. 1997. Antigenic peptides and autoimmunity. APMIS, 105,2:81–88.

31.

Tanaka

, Tanaka

, Komori

. 2009. Interferon-beta(1b) treatment in neuromyelitis optica. Eur Neurol, 62,3:167–170.

32.

Ubogu

, Cossoy

, Ransohoff

. 2006. The expression and function of chemokines involved in CNS inflammation. Trends Pharmacol Sci, 27,1:48–55.

33.

Uzawa

, Mori

, Hayakawa

, Masuda

, Kuwabara

. 2010. Different responses to interferon beta-1b treatment in patients with neuromyelitis optica and multiple sclerosis. Eur J Neurol, 17,5:672–676.

34.

Uzawa

, Mori

, Ito

, Uchida

, Hayakawa

, Masuda

, Kuwabara

. 2009. Markedly increased CSF interleukin-6 levels in neuromyelitis optica, but not in multiple sclerosis. J Neurol, 256,12:2082–2084.

35.

Wajgt

, Gorny

. 1983. CSF antibodies to myelin basic protein and to myelin-associated glycoprotein in multiple sclerosis. Evidence of the intrathecal production of antibodies. Acta Neurol Scand, 68,5:337–543.

36.

Wang

, Dai

, Qiu

, Lu

, Peng

, Wang

, Bao

, Li

, Hu

. 2011a. Interleukin-17-secreting T cells in neuromyelitis optica and multiple sclerosis during relapse. J Clin Neurosci, 18,10:1313–1317.

37.

Wang

, Wang

, Lee

, Chen

, Tsai

. 2011b. The rescue effect of plasma exchange for neuromyelitis optica. J Clin Neurosci, 18,1:43–46.

38.

Warabi

, Matsumoto

, Hayashi

. 2007. Interferon beta-1b exacerbates multiple sclerosis with severe optic nerve and spinal cord demyelination. J Neurol Sci, 252,1:57–61.

39.

Watanabe

, Nakashima

, Misu

, Miyazawa

, Shiga

, Fujihara

, Itoyama

. 2007. Therapeutic efficacy of plasma exchange in NMO-IgG-positive patients with neuromyelitis optica. Mult Scler, 13,1:128–132.

40.

Wingerchuk

. 2006. Evidence for humoral autoimmunity in neuromyelitis optica. Neurol Res, 28,3:348–353.

41.

Wingerchuk

, Hogancamp

, O'Brien

, Weinshenker

. 1999. The clinical course of neuromyelitis optica (Devic's syndrome) Neurology, 53,5:1107–1114.

42.

Wingerchuk

, Lennon

, Lucchinetti

, Pittock

, Weinshenker

. 2007. The spectrum of neuromyelitis optica. Lancet Neurol, 6,9:805–815.

43.

Wingerchuk

, Lennon

, Pittock

, Lucchinetti

, Weinshenker

. 2006. Revised diagnostic criteria for neuromyelitis optica. Neurology, 66,10:1485–1489.

44.

, Osoegawa

, Yamasaki

, Kawano

, Ochi

, Horiuchi

, Minohara

, Ohyagi

, Yamada

, Kira

. 2000. Flow cytometric differentiation of Asian and Western types of multiple sclerosis, HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP) and hyperIgEaemic myelitis by analyses of memory CD4 positive T cell subsets and NK cell subsets. J Neurol Sci, 177,1:24–31.