A systematic analysis of the complement pathways in patients with neuromyelitis optica indicates alteration but no activation during remission
Introduction
Neuromyelitis optica is an autoimmune inflammatory demyelinating disease of the central nervous system, characterized by the coexistence of optic neuritis and transverse myelitis (Wingerchuk et al., 2006, Illes, 2010). While both clinical entities are required for a definite diagnosis, the spectrum also includes spatially limited forms, such as longitudinally extensive transverse myelitis (LETM), and recurrent or bilateral optic neuritis (RION/BON), aggregately termed as the NMO spectrum of diseases (NMO-SD) (Sellner et al., 2010).
Circulating antibodies against AQP4, the main water channel of the CNS can be found in the majority of patients with NMO (70–80%); the frequency of anti-AQP4-seropositivity is much lower among patients with NMO-SD (20–40%), defined as definite NMO-SD by the European guidelines (Sellner et al., 2010, Lennon et al., 2004, Collongues et al., 2010). Several lines of evidence suggest that systemic anti-AQP4 antibodies are pathogenic, and are responsible for the destruction of astrocytes expressing AQP4 in the blood-brain barrier (Jarius and Wildemann, 2010): (i) IgG isolated from patients with NMO (NMO-IgG) is able to destroy astrocytes transfected with AQP4 (Kalluri et al., 2010, Hinson et al., 2007); (ii) the systemic transfer of AQP4 in the presence of myelin-specific inflammation or the intrathecal/intracerebral transfer in the presence of human complement can elicit clinical and pathological signs of NMO in animals (Bradl et al., 2009, Kinoshita et al., 2009, Saadoun et al., 2010, Li et al., 2011, Asgari et al., 2013); (iii) serum concentration of anti-AQP4 increases before an NMO relapse (Takahashi et al., 2007, Jarius et al., 2008); (iv) the loss of AQP4 from NMO lesions co-exists with the deposition of IgG and loss of astrocytes (Lucchinetti et al., 2002, Roemer et al., 2007, Wingerchuk and Lucchinetti, 2007, Misu et al., 2007); (v) plasmapheresis reduces disease activity by decreasing plasma AQP4-IgG (Papadopoulos, 2012).
The current concept of the pathogenesis of NMO suggests that anti-AQP4 antibodies are first generated in the systemic immune compartment; this is indicated by the 500 times higher concentration of NMO-IgG in the plasma than in the CNS. Once they cross the blood-brain barrier, the antibodies cause astrocyte cytotoxicity. The consequent infiltration by granulocytes and macrophages contributes to the disruption of the blood brain barrier, to the inflammatory responses within the CNS, and finally, to myelin breakdown (Papadopoulos, 2012).
The activation of the complement system plays a central role in the death of astrocytes, and complement-mediated astrocyte damage may be the initial event in CNS inflammation in NMO (Papadopoulos, 2012). Because the majority of anti-AQP4 antibodies are of IgG1-isotype, it can activate complement upon binding to AQP4 present on the astrocytes; indeed, NMO-IgG is capable to destroy AQP4-transfected astrocytes in vitro (Kalluri et al., 2010). The loss of astrocytes from NMO lesions is accompanied by the perivascular deposition of activated complement and immunoglobulin (Lucchinetti et al., 2002, Roemer et al., 2007, Wingerchuk and Lucchinetti, 2007, Misu et al., 2007).
Azathioprine (AZA) is a first-line treatment for NMO/NM-SD (Sellner et al., 2010). As a pro-drug of 6-mercaptourine (6-MP), it is a purine analog, which blocks the de novo pathway of purine synthesis (Maltzman and Koretzky, 2003). Besides, 6-MP interacts with the small GTP-binding protein Rac-1, and blocks the upregulation of Bcl-xL mRNA/protein, which may explain its effect increasing lymphocyte apoptosis (Tiede et al., 2003). Azathioprine can induce long-lasting remission in patients with active myasthenia gravis, an IgG1 antibody-mediated disease of the peripheral nervous system, where the complement pathways also play central roles (Rozsa et al., 2009).
Despite of the presumed, major role of complement in the pathogenesis of NMO/NMO-SD, studies investigating the complement system are scarce and somewhat contradicting; a systematic analysis of the complement pathways has not been performed. A pioneering study indicated elevated CH50 levels in the sera of Japanese anti-AQP4-seropositive patients during relapse, compared to those with seronegative optico-spinal MS (Doi et al., 2009). Elevated levels of complement derivatives were detected in another study of sera obtained during relapse from patients with NMO (Tuzun et al., 2011). While CSF levels of C5a were also found to be elevated during relapse in an additional study, no differences were detected in the levels of serum C3a, C4a and C5a during relapse or remission (Kuroda et al., 2013). In a recent study, decreased level of CH50 was reported in sera of NMO patients during relapse with AQP4 antibodies compared to controls subjects with headache and trigeminal neuralgia (Chen et al., 2013).
The complement system is a component of innate immunity. It efficiently recognizes and eliminates intruding pathogens and altered self structures – it is indispensable in several inflammatory disorders (Ricklin et al., 2010). During activation by any of the three pathways, i.e. the classical, the alternative and the lectin pathway, native components disappear through consumption and complement activation products appear. The simultaneous use of three, non-interchangeable approaches and the pooled interpretation of the results may prove useful in exploring the in vivo events. In particular, the following should be measured: (1) the concentration of native complement proteins; (2) the concentrations of the complement products generated during in vivo activation; and (3) total complement activity, which represents the capacity of the complement system following in vitro activation of the serum sample with a potent activating agent. In case of the reduced synthesis of complement proteins or of enhanced in vivo complement activation, (residual) total complement activity and the blood levels of the native proteins both decrease. Only in vivo activation is followed by the appearance of complement activation products. The absolute concentration of the latter should be compared with those of the native proteins, because the continuous/chronic consumption and accelerated elimination of the native components from the circulation decreases the absolute levels of activation products.
So far, (i) systematic appraisal of the activation of all three complement pathways has not been performed in patients with NMO; (ii) all the previous studies analyzed sera, where intrinsic activation of the complement pathways may lead to false positive results; and (iii) the effects of azathioprine and remission on the activation of the complement pathways have not been fully addressed. Because anti-AQP4 antibodies are also present in the sera during remission and can activate complement, we systematically examined the three complement pathways in peripheral blood samples taken from patients with NMO during remission and compared to healthy controls. We also studied the complement activation products in CSF samples taken during NMO relapse, and compared to CSF taken from patients with multiple sclerosis (MS).
Although our results do not indicate substantial systemic complement activation in adequately controlled NMO, the data indicate definite differences compared to healthy controls, suggesting that the complement system is abnormally affected even during remission. The particular depression of the alternative pathway compared to the classical pathway may suggest that the remission state achieved by proper treatment may be related to the suppression of the alternative pathway.
Section snippets
Patients and controls
Twenty-five patients with anti-AQP4-seropositive NMO-SD were enrolled. Anti-AQP4 antibodies were determined by cell-based assays either commercially available (Euroimmune, Lübeck, Germany). The mean age of the patients was 40 years (range: 30–69 years), and the male to female ratio was 3:22. All patients received azathioprine for more than 6 months and all were in remission at the time of blood sampling. Twelve patients fulfilled the diagnostic criteria of definite NMO (Wingerchuk et al., 2006
Results
Here, we applied a complex approach. First, we studied the total complement activities of the three activation pathways, and then the concentrations of the individual complement proteins in the plasma; we also determined differences in the numbers of patients and controls out of the normal ranges. Next, we analyzed in vivo complement activation products in the plasma. We determined the relationship among the individual complement parameters in both NMO and controls. In patients with NMO, we
Discussion
NMO has only recently been considered an independent clinical entity; the initial set of its diagnostic criteria were drafted in 1999, and updated in 2006 (Wingerchuk et al., 2006). The pathogenesis involves an inflammatory process induced by a type 2 (antibody-mediated) hypersensitivity reaction and complement activation, supported by transfer experiments using animal models. The analysis of tissue specimens demonstrated the role of complement activation beyond doubt, and complement activation
Acknowledgements
This work was supported by the Hungarian National Research Fund, OTKA CK 80842 (to G. Füst) and OTKA 77892 (to Z. Illes) and the Hungarian Neuroimaging Foundation, TAMOP 4.2.2.A-11 (to Z. Illes).
References (40)
- et al.
Studies of hepatic synthesis in vivo of plasma proteins, including orosomucoid, transferrin, alpha 1-antitrypsin, C8, and factor B
Clinical Immunology and Immunopathology
(1980) - et al.
Complement-dependent pathogenicity of brain-specific antibodies in cerebrospinal fluid
Journal of Neuroimmunology
(2013) - et al.
Parameters of the classical complement pathway predict disease severity in hereditary angioedema
Clinical Immunology
(2011) - et al.
An ELISA technique for the measurement of C1q in cerebrospinal fluid
Journal of Immunological Methods
(1988) - et al.
Therapeutic inhibition of the alternative complement pathway attenuates chronic EAE
Molecular Immunology
(2013) - et al.
Neuromyelitis optica: passive transfer to rats by human immunoglobulin. Biochemical and biophysical research communications
(2009) - et al.
Increase of complement fragment C5a in cerebrospinal fluid during exacerbation of neuromyelitis optica
Journal of Neuroimmunology
(2013) - et al.
A serum autoantibody marker of neuromyelitis optica: distinction from multiple sclerosis
Lancet
(2004) Verkman AS. Aquaporin 4 and neuromyelitis optica
Lancet Neurology
(2012)- et al.
Complement in experimental autoimmune encephalomyelitis revisited: C3 is required for development of maximal disease
Molecular Immunology
(2007)
Conservation of the modular structure of complement factor I through vertebrate evolution
Developmental and Comparative Immunology
Enhanced complement consumption in neuromyelitis optica and Behcet's disease patients
Journal of Neuroimmunology
Human C′3: evidence for the liver as the primary site of synthesis
Science
Neuromyelitis optica: pathogenicity of patient immunoglobulin in vivo
Annals of Neurology
The complement and immunoglobulin levels in NMO patients
Neurological Sciences: Official Journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology
Neuromyelitis optica in France: a multicenter study of 125 patients
Neurology
Hypercomplementemia at relapse in patients with anti-aquaporin-4 antibody
Multiple Sclerosis
Acute-phase response in rat to carbon tetrachloride-azathioprine induced cirrhosis and partial hepatectomy of cirrhotic liver
Journal of Toxicology and Environmental Health
Pathogenic potential of IgG binding to water channel extracellular domain in neuromyelitis optica
Neurology
Pathogenesis, diagnosis and treatment of neuromyelitis optica: changing concept of an old disease
Clinical and Experimental Neuroimmunology
Cited by (19)
C3 and C4 complement levels in AQP4-IgG-positive NMOSD and in MOGAD
2021, Journal of NeuroimmunologyCitation Excerpt :This is compatible with attack-independent subclinical disease activity in AQP4-IgG+ NMOSD, which has likewise been evoked by studies of serum glial fibrillary acidic protein (Aktas et al., 2021; Schindler et al., 2021), cerebrospinal fluid findings (Jarius et al., 2011) and visual evoked potentials (Ringelstein et al., 2020) in AQP4-IgG+ NMOSD. In accordance with previous findings (Veszeli et al., 2014), there was no association of AQP4-IgG titers with C3 and C4 levels in patients with AQP4-IgG+ NMOSD. This may be due to the known heterogeneous ability of AQP4-IgG clones to activate complement, that is, individual properties of AQP4-IgG clones may be more important for complement activation and consumption than antibody titers (Soltys et al., 2019).
The clinical value of complement proteins in differentiating AQP4-IgG-positive from MOG-IgG-positive neuromyelitis optica spectrum disorders
2019, Multiple Sclerosis and Related DisordersCitation Excerpt :However, the degree of improvement after the first attack was not significantly different between the two groups. Multiple studies examining the roles of complement proteins in the pathology of NMOSD have indicated that complement activation drives the pathology of classical AQP4-IgG-positive NMOSD; this finding enables the reliable diagnosis of NMOSD (Veszeli et al., 2014) and its differentiation from MS (Hakobyan et al., 2017). However, the difference between NMOSD patients positive for AQP4-IgG or MOG-IgG has received relatively scant attention in the literature.
The link between morphology and complement in ocular disease
2017, Molecular ImmunologyCitation Excerpt :C3a levels have been seen to correlate with disease activity, neurological disability, and APQ4-IgG in NMO patients, supporting the potential value of therapeutic complement inhibition. The antibody and complement dependence is reflected in how the disease exacerbations are treated: After basic prednisolone treatment, intravenous IgG, plasmapheresis, or immune adsorption, rituximab and (recently) the complement inhibitors eculizumab and C1-INH have been shown to be efficacious (Etemadifar et al., 2017; Pilch et al., 2017; Veszeli et al., 2014). The link between complement system dysregulation and the morphological changes seen in AMD, glaucoma, uveitis, and NMO are abundant, and our knowledge of this topic is expanding.
A review of human diseases caused or exacerbated by aberrant complement activation
2017, Neurobiology of AgingCitation Excerpt :CSF levels of C5a and sC5b-9 were reported to be elevated in neuromyelitis optica indicating the activation of the MAC in the CNS (Kuroda et al., 2013; Wang et al., 2014a). Serum antibody levels of several complement components have also been reported to be affected, with the suggestion that inhibiting the alternative pathway of complement would be an effective therapeutic strategy (Nytrova et al., 2014; Veszeli et al., 2014). Support for this concept has been indicated by a small open label trial of eculizumab in which no relapses were recorded over a 12-month period (Pittock et al., 2013).
Complement activation in patients with neuromyelitis optica
2014, Journal of Neuroimmunology