Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Jan 10.
Published in final edited form as: Ann Neurol. 2018 Jan 10;83(1):40–51. doi: 10.1002/ana.25120

Mechanisms of Caspr2 antibodies in autoimmune encephalitis and neuromyotonia

Kristina R Patterson 1, Josep Dalmau 1,2,3, Eric Lancaster 1
PMCID: PMC5876120  NIHMSID: NIHMS930925  PMID: 29244234

Abstract

Objective

To determine the pathogenic mechanisms of autoantibodies to the cell adhesion molecule Caspr2 in acquired neuromyotonia and autoimmune encephalitis.

Methods

Caspr2 positive samples were confirmed using a cell-based assay, and their IgG subtypes were determined by ELISA and cell-based assay. A solid phase binding assay quantified the binding of Caspr2 to contactin-2 in the presence of Caspr2 autoantibodies. Living cultures of primary rat hippocampal neurons were incubated with Caspr2-positive or control sera and the distribution of Caspr2-positive immunofluorescent puncta and total surface Caspr2 were quantified. HEK cells transfected to express Caspr2 were incubated with Caspr2-positive or control samples, and cell-surface biotinylation and Western blot were used to assess total, internalized, and surface levels of Caspr2.

Results

We confirmed six samples with strong Caspr2 reactivity. IgG4 Caspr2 antibodies were present in all six cases. Caspr2 interacted with another cell adhesion molecule, contactin-2, with nanomolar affinity in the solid phase assay, and Caspr2 autoantibodies inhibited this interaction. Caspr2 autoantibodies did not affect the surface expression of Caspr2 in rat primary hippocampal neurons or transfected HEK cells.

Interpretation

Caspr2 autoantibodies inhibit the interaction of Caspr2 with contactin-2 but do not cause internalization of Caspr2. Functional blocking of cell adhesion molecule interactions represents a potential mechanism with therapeutic implications for IgG4 autoantibodies to cell adhesion molecules in neurological diseases.

Introduction

Contactin-associated protein-like 2 (Caspr2) is a cell adhesion molecule (CAM) of the neurexin family, and is expressed on central nervous system (CNS) and peripheral nervous system (PNS) axons.1 Autoantibodies to surface epitopes of Caspr2 have been associated with a form of acquired peripheral nerve hyperexcitability (acquired neuromyotonia or Isaacs’ syndrome) and autoimmune encephalitis.2–4 Some patients with anti-Caspr2 autoantibodies have both CNS and PNS involvement (Morvan syndrome), which is characterized by neuromyotonia, encephalitis, dysautonomia, insomnia, and/or neuropathic pain.4

Caspr2 is localized to the juxtaparanodal region of both CNS and PNS nodes of Ranvier1, which are regions of cellular specializations between myelinating glial cells and axons that enable saltatory conduction of action potentials.5,6 Nodes of Ranvier consist of three domains: the node, the paranode, and the juxtaparanode. At the juxtaparanodes, contactin-2 (also known as TAG-1) interacts with Caspr2, and this association is necessary for the clustering of voltage-gated potassium channels (VGKC).7,8

Antibody-mediated crosslinking and internalization is a common mechanism of autoantibodies targeting ionotropic receptors, including the N-methyl-D-aspartate receptor (NMDAR) and the a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) in autoimmune encephalitis.9–14 However, diseases caused by autoantibodies to CAMs may not share this mechanism. Caspr2 autoantibodies are frequently of the IgG4 subtype3,4,15, as are autoantibodies to other CAMs expressed on nodes of Ranvier, such as neurofascins and contactin.16–19 IgG4 has several unique characteristics not found in other IgG subtypes. IgG4 antibodies have low affinities for Fcγ receptors and are ineffective in activating complement. IgG4 have high affinities for their antigens as they undergo a higher number of somatic hypermutations. Finally, IgG4 undergo hemi-antibody exchange making each antibody functionally monovalent.17,20 Thus, cross-linking and internalization of the target antigen in these disorders is less likely. A different paradigm may be needed to understand the pathophysiology of IgG4 autoantibodies to CAMs. In this paper, we demonstrate that autoantibodies to Caspr2 do not cause its internalization, and thus do not function in a manner similar to autoantibodies against ionotropic receptors such as NMDAR. Rather, Caspr2 autoantibodies disrupt the binding of Caspr2 to contactin-2, potentially interfering with the clustering of VGKC at juxtaparanodes and resulting in hyperexcitability of peripheral nerves. This potential mechanism of Caspr2 autoantibodies serves as a paradigm for pathogenicity of IgG4 autoantibodies to other CAMs in the CNS and PNS.

Methods

Standard protocol approvals, registrations, and patient consents

Serum was collected from patients with clinical features of Caspr2 autoimmunity (e.g. encephalitis and/or peripheral nerve hyperexcitability) as well as control samples under a tissue bank protocol approved by an institutional review board (IRB) at the University of Pennsylvania. Controls from this tissue bank included samples from persons with anti-NMDAR encephalitis or persons with an inherited peripheral neuropathy of a known genetic cause (Charcot-Marie-Tooth disease; CMT). A separate protocol approved by the University of Pennsylvania IRB was used to access the samples from the tissue bank for testing and to access the clinical information. Informed consent was obtained from each subject.

Cell-based assay (CBA)

Serum samples were tested for Caspr2 antibodies using methods previously described. 2 Briefly, human embryonic kidney 293 (HEK293) cells were transiently transfected using Lipofectamine 3000 (Invitrogen) and a plasmid containing human Caspr2 in a Prk5 vector (gift from Dr. Elior Peles, Weizmann Institute of Science, Rehovot, Israel). Sera were heat inactivated by incubating at 55°C for 30 minutes then were applied (1:100) to live cells for 30 minutes at 37°C. Cells were washed, permeabilized, and stained with commercial rabbit antibody against Caspr2 at a dilution of 1:200 (Abcam cat. # ab33994), followed by fluorescently conjugated secondary antibodies. Images of transfected cells were taken with a Leica SP5 confocal microscope at 20x magnification. Six samples with strong Caspr2 reactivity were chosen for further experimentation. The Caspr2 antibody titer of these 6 samples was determined using serial dilutions of heat-inactivated sera applied to live HEK293 cells transfected to overexpress Caspr2 as above.

IgG subtyping of Caspr2 autoantibodies

The IgG subtype of the Caspr2 autoantibodies was determined using a solid phase binding assay with immobilized Caspr2. Recombinant peptide of the extracellular domain of human Caspr2 (R&D systems, 100 ng) diluted in binding buffer (20 mM Tris, pH 8.0, 100 mM NaCl, 5 mM CaCl2) was coated onto a 96-well plate at for 1 hour, blocked with blocking buffer (1% gelatin in binding buffer) for 2 hours, and then incubated with patient serum (1:200) in wash buffer (binding buffer + 0.05% Tween-20) for one hour. Wells were incubated with mouse anti-human IgG1, IgG2, IgG3, or IgG4 (Thermo Scientific, 1:250) for one hour followed by HRP conjugated anti-mouse IgG secondary antibody (Santa Cruz, 1:500) for 45 minutes. Plate was washed five times with wash buffer between steps. The reaction was developed with TMB substrate and stopped with H2SO4. Absorbance was read at 450 nm using an Envision Excite plate reader (PerkinElmer). All steps were performed at room temperature.

IgG subtyping was confirmed using a CBA for 3 out of 6 samples. (The remaining 3 samples were tested previously.21) HEK293 cells were transfected with Caspr2 as above and then treated with heat-inactivated sera (1:200) for 30 minutes at 37°C. Cells were washed, fixed, and stained with unconjugated mouse antibodies against IgG1 (Thermo Scientific, 1:200), IgG2 (Thermo Scientific, 1:1000), IgG3 (Thermo Scientific, 1:500), or IgG4 (Thermo Scientific, 1:75) for 1 hour at room temperature followed by a fluorescently conjugated secondary antibody. Cells were imaged as above.

Solid phase binding assays

Binding of Caspr2 to contactin-2 was determined using a protocol modified from Lu et al.22 Recombinant peptide of the extracellular domain of human Caspr2 (R&D systems, 200 ng) or the extracellular domain of human contactin-2 (Sino Biological, 200 ng) diluted in binding buffer (see above) was coated onto 96-well plates for 1 hour, blocked with blocking buffer for 2 hours, and then incubated with the converse peptide in wash buffer at concentrations ranging from 0 to 50 nM (in duplicate). Nonspecific binding of the peptides was assessed by incubation in pre-blocked wells without immobilized peptide, and the level of background binding was comparable to the background of the plate (data not shown). Wells were incubated with either anti-Caspr2 (R&D cat. # AF5245,1:100) or anti-contactin-2 (R&D cat. # AF1714, 1:1000) antibody for one hour followed by incubation with peroxidase-conjugated secondary antibody for 45 minutes. Plates were washed five times with wash buffer between steps. The reaction was developed with TMB substrate and stopped with H2SO4. Absorbance was read at 450 nm using an Envision Excite plate reader (PerkinElmer). All steps were performed at room temperature. The KD value was calculated by fitting the data to a one-site total binding equation (assumes total binding is a sum of specific binding plus a linear nonspecific component) using the non-linear regression model in GraphPad Prism.

To determine whether Caspr2 autoantibodies interfere with Caspr2 binding to contactin-2, human Caspr2 peptide was immobilized onto a 96-well plate, and blocked with blocking buffer as above. After the blocking step, patient sera with Caspr2 autoantibodies and control sera were applied at concentrations ranging from 1:10 to 1:100 for one hour (in duplicate). After washing the plate, wells were incubated with human contactin-2 (200 ng) in wash buffer for 1 hour. Wells were then incubated with anti-contactin-2 (R&D, 1:1000) for one hour followed by HRP-conjugated donkey anti-sheep IgG preadsorbed to human IgG (Santa Cruz, 1:2000). Samples were developed and read as outlined above. For all experimental conditions, control samples with immobilized Caspr2 and human sera (without contactin-2) were subtracted to account for a small amount of cross-reactivity of the secondary antibody to human IgG.

Immunofluorescence and quantification of cell-surface Caspr2

Rat hippocampal neuronal cultures were prepared and immunohistochemical experiments were performed to study surface expression of Caspr2 based on a modified protocol initially used to study internalization of NMDARs.10 Briefly, rat hippocampal neurons were cultured for 14–21 days on coverslips then treated for 24 hours at 37°C with heat-inactivated sera (1:100 or 1:25) containing Caspr2 autoantibodies versus sera from controls who had inherited neuropathy. To visualize Caspr2 on the cell surface, live neurons were incubated with a polyclonal antibody to the extracellular domain of Caspr2 (R&D systems cat. # AF5145, 1:200) for 30 minutes at 37°C. After removing the media and washing with phosphate-buffered saline (PBS), neurons were fixed with 4% paraformaldehyde, permeabilized with 0.3% Triton X-100, incubated with a monoclonal antibody to the C-terminal (intracellular) portion of Caspr2 (Abcam cat. #33994,1:200), washed and incubated with appropriate fluorescently conjugated secondary antibodies. A minimum of three replicates were performed for each experimental condition.

Images were captured on a Leica SP5 confocal microscope at 63x magnification with a step size of 0.34 um. Images were thresholded and Caspr2 puncta were quantified using SynPAnal software.23 For each image, 50 micron axonal segments (n≥8) were manually traced and the puncta (minimum 4 nm diameter) were automatically quantified for each segment. A minimum of three replicates with a minimum of four images per replicate were analyzed for each experimental condition for a minimum of 96 analyzed segments for each experimental condition. Cell-surface Caspr2 density, area, and fluorescence intensity are defined as the puncta stained by the polyclonal Caspr2 antibody to the extracellular domain that co-localize with the monoclonal (C-terminal) Caspr2 antibody to minimize nonspecific activity of the polyclonal antibody. Internalized Caspr2 was quantified as Caspr2 that was stained with the C-terminal antibody but was not stained with the antibody to the extracellular domain (total Caspr2 minus surface Caspr2).

Cell-surface biotinylation

HEK293 cells were cultured in 6-well plates and transfected to express human Caspr2. Cells were treated with anti-Caspr2 or control sera as above for 24 hours at 37°C. Biotinylation and isolation of surface Caspr2 was performed using PIERCE Cell Surface Protein Isolation Kit according to manufacturer protocol. Briefly, cells were quickly washed with ice-cold PBS, incubated with NHS-SS-Biotin (0.25 mg/mL) for 30 minutes at 4°C, and quenched. Cells were then scraped, transferred to a 15 mL conical tube, and centrifuged at 500 × g for 3 minutes. The pellet was lysed in RIPA buffer, sonicated for 5 seconds, and centrifuged at 10,000 × g for 2 minutes at 4°C. The supernatant was incubated with NeutrAvidin Agarose for 60 minutes. Sample was centrifuged for 1 minute at 1,000 × g. Supernatant was collected separately. Cell surface proteins were eluted from agarose with 1 mM DTT.

Total, internalized, and surface levels of Caspr2 were determined using SDS-PAGE/Western blot. Samples were run on 4–15% linear gradient gels (Bio-Rad) and transferred to nitrocellulose membrane. Membranes were blocked in 5% nonfat dried milk in PBS containing 0.05% Tween-20 then incubated with rabbit anti-Caspr2 (Abcam cat # ab137052,1:1000) or mouse anti-actin (Abcam cat # ab3280, 1:1000) overnight at 4°C. After rinsing, the membranes were incubated in peroxidase-conjugated secondary antibody for 1 h at room temperature. Reactivity was visualized using ECL substrate. Blots were developed with ChemiDoc Imaging System (Bio-Rad). Images were inverted and densitometry quantified using Fiji software (National Institutes of Health). Caspr2 expression was normalized to that of actin. Three replicates of each experimental condition were performed.

Statistics

SigmaStat software (Systat Software, Inc.) was used for all statistical tests. Comparisons were made using a t-test or one-way analysis of variance followed by Holm-Sidak post hoc analysis as indicated. Data were expressed as mean ± SEM, and significance was set at p-value of 0.05.

Results

Caspr2 autoantibodies are predominantly IgG4

Samples from 5 patients with acquired neuromyotonia and/or encephalitis that were previously determined to be Caspr2 antibody-positive were confirmed as positive by CBA. One additional case did not have reported neuromyotonia or encephalitis but presented with neuropathic pain. These six samples are designated C1-C6. Heat-inactivated sera from these cases were tested a 1:100 dilution on live HEK cells transfected to express Caspr2 (Figure 1). To determine the Caspr2 antibody titer of each sample, serial dilutions were performed. The clinical characteristics and Caspr2 antibody titers of these cases are outlined in Table 1.

Figure 1.

Figure 1

Open in a new tab

Table 1.

Clinical characteristics and antibody titers of Caspr2 positive subjects

Case Age Sex Symptoms Titer
C1 57 M encephalitis, ataxia, muscle spasms 1:1600
C2 74 M neuromyotonia, neuropathic pain, dysautonomia, muscle cramps <1:6400
C3 68 M limbic encephalitis, ataxia, insomnia 1:1600
C4 77 M neuromyotonia, encephalitis, neuropathic pain, dysautonomia, muscle cramps, insomnia 1:200
C5 71 M encephalitis, seizures, neuropathic pain 1:3200
C6 67 M neuropathic pain 1:3200
Open in a new tab

Previous work has demonstrated that Caspr2 autoantibodies are frequently of the IgG4 isotype.3,4,15 To determine the IgG composition of anti-Caspr2 autoantibodies in our case samples, we created a solid phase binding assay and tested the bound IgG with IgG subtype-specific secondary antibodies. All six case samples, but not control samples, contained IgG reactivity to Caspr2 in the assay. All six case samples contained IgG4 anti-Caspr2 autoantibodies, and IgG4 was the predominant subtype in five out of six sera (Table 2). The exception was sample C2 which had a mixture of IgG2 > IgG4 > IgG1. IgG subtyping was confirmed by CBA (Figure 2) for 3 out of 6 samples. Again, Caspr2 antibodies were predominantly of the IgG4 subtype. For the remaining samples, the presence of IgG4 antibodies by CBA has been previously reported.21

Table 2.

IgG subtypes of Caspr2 autoantibodies by ELISA

IgG1 IgG2 IgG3 IgG4
C1 − + − +++
C2 + +++++ − ++++
C3 − + − ++++
C4 − − − ++
C5 + + − ++
C6 − − − ++
CMT1 − − − −
CMT2 − − − −
CMT3 − − − −
Open in a new tab

Figure 2.

Figure 2

Open in a new tab

Caspr2 binds contactin-2 in vitro

Next we studied the interaction of Caspr2 with contactin-2 using a solid phase binding assay. Caspr2 was immobilized on 96-well plates. Then contactin-2 was added at concentrations ranging from 0 to 50 nM. We found contactin-2 binds immobilized Caspr2 with a KD of 15±5 nM (Figure 3A). Alternatively, when contactin-2 is immobilized, Caspr2 binds contactin-2 with a KD of 12±7 nM (Figure 3B). These values are comparable previously published results and confirm that Caspr2 and contactin-2 bind one another with nanomolar affinity.22

Figure 3.

Figure 3

Open in a new tab

Caspr2 autoantibodies inhibit the interaction of Caspr2 with Contactin-2

To test our hypothesis that Caspr2 autoantibodies inhibit Caspr2 binding to contactin-2, we immobilized Caspr2 on a 96-well plate, treated with the wells with varying dilutions of anti-Caspr2 case sera, and studied the ability of Caspr2 to bind contactin-2. To control for the effect of nonspecific binding of unrelated antibodies or serum elements, we also tested a control sera containing NMDAR antibodies. Results are expressed as fraction of contactin-2 binding compared to no treatment (Figure 3C). We found a dose-dependent decrease in contactin-2 binding in the presence of anti-Caspr2 sera, but not with anti-NMDAR sera, supporting our hypothesis that Caspr2 autoantibodies block the interaction of Caspr2 with contactin-2.

Subsequently, we tested our remaining anti-Caspr2 case samples for their ability to inhibit binding to contactin-2. Each anti-Caspr2 sample inhibited the interaction of Caspr2 with contactin-2, although to varying degrees, ranging from 33 ± 12 to 91 ± 2 percent inhibition, compared to control sera – either pooled CMT sera (3.6 ± 8.1) or an anti-NMDAR serum (8.5 ± 3.2) (Figure 3D). Interestingly, sample C2 showed the strongest inhibition even though it contains IgG2 > IgG4 anti-Caspr2 antibodies; however, sample C2 also showed the strongest reactivity to Caspr2 in general, so a higher titer of this antibody may explain the increased inhibition. These experiments demonstrate that Caspr2 autoantibodies consistently inhibit the interaction of Caspr2 with contactin-2.

Caspr2 autoantibodies do not affect surface expression of Caspr2

We used cultured rat hippocampal neurons to study surface expression levels of Caspr2. Immunostaining of live DIV 14–21 cultures for Caspr2 revealed punctate staining of axons in neurons (Figure 4A), in accord with predominantly axonal membrane localization found by Pinatel et al.15 To determine whether Caspr2 autoantibodies cause internalization of the antigen in a manner comparable to anti-NMDAR antibodies11, rat hippocampal neurons were treated with either anti-Caspr2 sera or control sera from CMT cases for 24 hours. This time point was chosen based on previous results which demonstrated that anti-NMDAR antibodies cause maximal internalization at 12 hours and this effect remains for at least 48 hours.11 Living hippocampal neurons were labeled with an antibody to the extracellular domain of Caspr2 (surface Caspr2), fixed and permeabilized, and then labeled with an antibody a cytoplasmic epitope of Caspr2 (total Caspr2). We found no difference in density of Caspr2 puncta in samples treated with Caspr2 anti-sera versus control (CMT) sera tested at a concentration of 1:100 (Figure 4E). In fact, the vast majority of Caspr2 was expressed on the cell surface in both treated and control samples as there was only a minor amount of total Caspr2 that did not co-localize with surface Caspr2 (data not shown). Neurons were also treated with anti-Caspr2 or control sera at a concentration of 1:25 (Figure 4F). Again no difference in Caspr2 puncta density was observed though there was considerable cell death in both the treatment and control groups at the higher sera concentration. Additionally, there was no difference in puncta intensity or puncta area in the treatment versus control samples at either sera concentration (data not shown).

Figure 4.

Figure 4

Open in a new tab

To confirm that Caspr2 autoantibodies do not cause internalization of Caspr2, cell-surface biotinylation experiments were performed. HEK293 cells transfected to express Caspr2 were treated with either anti-Caspr2 sera or control (CMT) sera for 24 hours, then treated with biotin conjugated to a sulfhydryl moiety, lysed with RIPA buffer, and mixed with avidin-conjugated resin to separate cell surface Caspr2 from Caspr2 located inside the cell. There was no difference in biotinylated (cell surface) Caspr2 in anti-Caspr2 sera treated samples versus control samples (Figure 5). There was also no difference in unbiotinylated (internalized) Caspr2 or total Caspr2 protein levels. Because only a fraction of surface Caspr2 was probably biotinylated, this approach may underestimate the amount of surface Caspr2 and overestimate the amount of internalized Caspr2, but this should be consistent across samples. These results demonstrate that anti-Caspr2 sera do not affect cell surface levels of Caspr2 in two model systems.

Figure 5.

Figure 5

Open in a new tab

Discussion

Our findings demonstrate a functional effect of Caspr2 autoantibodies and provide a potential pathogenic mechanism. We show that Caspr2 autoantibodies do not function like NMDAR and AMPAR autoantibodies, which result in the internalization of the targeted receptors; this is thought to be a primary mechanism in the pathophysiology of antibody-mediated autoimmune disorders.9–11,14 Caspr2 autoantibodies, in contrast, do not result in the internalization of Caspr2. This may be due to differences in the membrane trafficking of Caspr2 compared to ionotropic receptors and/or a lack of crosslinking by Caspr2 IgG4 antibodies, which are functionally monovalent. Notably, monovalent (Fab fragments) of autoantibodies to NMDAR and AMPAR also fail to internalized these receptors.10 Since the primary known function of Caspr2 is to act as CAM, we hypothesized that Caspr2 autoantibodies inhibit its interaction with its binding partner contactin-2, which we demonstrated directly.

All six anti-Caspr2 sera inhibited the interaction of Caspr2 with contactin-2, but the degree of inhibition varied. The reasons that some of the anti-Caspr2 autoantibodies inhibited binding more than others are likely two-fold. First, some of the autoantibodies reacted to Caspr2 more avidly than others. For instance, sample C2 had the strongest reactivity to Caspr2 and robustly inhibited binding to contactin-2. Conversely, sample C5 also had a very high titer and showed the least amount of inhibition, and sample C4 had a low titer but showed robust inhibition. These results demonstrate that the antibody titer cannot completely account for the degree of inhibition observed. Second, the epitopes of the autoantibodies also likely plays a role. We and others have previously showned that autoantibodies recognize multiple epitopes concentrated at the N-terminus of the Caspr2.15,24 For instance, the epitopes of C5 and C6 were previously mapped by Olsen et al.24 These samples (as well as most of the anti-Caspr2 samples tested) recognize the discoidin domain as well as other epitopes concentrated in the N-terminus of the protein. It is unclear how binding of IgG4 antibodies to these domains would inhibit binding to contactin-2; the regions for binding for contactin-2 have yet to be determined and antibodies to these regions will likely result in stronger inhibition. Limited amount of patient sample prevented epitope mapping of the remaining patient samples used in this paper.

Our work confirms that Caspr2 and contactin-2 bind one another with nanomolar affinity. Whether Caspr2 binds contactin-2 in cis and/or in trans is a matter of debate. Traka et al. previously published that contactin-2-Fc chimeric protein did not bind to cells transfected with Caspr2 alone whereas it interacted with cells co-transfected with Caspr2 and contactin-2,8 leading the authors to conclude that Caspr2 binds contactin-2 in cis and the Caspr2/contactin-2 complex binds contactin-2 in trans. More recent work demonstrated that Caspr2-Fc chimeric protein can bind contactin-2 transfected cells if Caspr2 is pre-clustered.15 The interaction of Caspr2 with contactin-2 is integral to the clustering of VGKC’s (mainly Kv1.1 and Kv1.2 subunits) at the juxtaparanode.7,8 These channels are necessary to stabilize conduction at nodes of Ranvier, prevent repetitive firing, and maintain resting potential.25 Knock-out mice for either Caspr2 or contactin-2 display diffuse localization of Kv1.1 and Kv1.2 along the internode.7,8 Kv1.1-null mice exhibit cooling-induced hyperexcitability in the neuromuscular transmission and hyperalgesia26–28 similar to symptoms seen in patients with Caspr2 autoantibodies (peripheral nerve hyperexcitability and neuropathic pain). Admittedly, Caspr2 null mice did not show any abnormalities in nerve conduction despite mislocalization of VGKCs.7 Whether Caspr2 autoantibodies result in mislocalization of VGKCs and what effects this may have on nerve conduction remain to be elucidated and will be a matter of future investigation. A schematic of our proposed mechanism of pathogenicity for autoantibodies to Caspr2 is presented in Figure 6.

Figure 6.

Figure 6

Open in a new tab

In addition to Caspr2 antibodies, autoantibodies to other CAMs are also predominantly of the IgG4 subtype.17 For instance, antibodies to neurofascin 155 (NF155), neurofascin 186 (NF186), and contactin-1 were discovered in a subset of patients with chronic and acute inflammatory demyelinating polyneuropathies (CIDP and AIDP, respectively) and are all predominantly IgG4.18,29–37 These proteins all play critical roles in conduction of action potentials at nodes of Ranvier. At nodes, NF186 interacts with extracellular matrix proteins such as gliomedin and mediates the clustering of voltage-gated sodium channels.38–40 At paranodes, NF155 is expressed by the terminal loops of myelin and associates with axonal CAMs contactin-1 and Caspr1 to form septate-like junctions, which prevent current leakage.41 The mechanism by which autoantibodies to NF155 and NF186 cause disease, assuming they do so, remains to be elucidated. Autoantibodies to contactin-1, on the other hand, were found to disrupt the interaction of contactin-1 with NF155/Caspr1 at the paranode.31,32

Other examples of IgG4 autoantibodies that associate with neurologic diseases are anti-leucine-rich glioma-inactivated1 (LGI1) which is associated with encephalitis and anti-muscle-specific tyrosine kinase (MuSK) antibodies found in a subset of patients with mysasthenia gravis.17 MuSK antibodies were found to prevent binding of MuSK to Lrp4 and Collagen Q which in turn inhibits Agrin-stimulated MuSK phosphorylation.42–44 LGI1 antibodies were discovered to inhibit binding to ADAM22 which, in turn, reduced synaptic AMPA receptors.45 Together, these data and our work suggest that blocking protein-protein interactions may be a common mechanism of IgG4 autoantibodies, particularly those targeting CAMs.

To date the largest cohort of anti-Caspr2 patients studied consisted of a retrospective analysis of 38 cases.4 Treatments utilized included steroids, IVIG, plasma exchange, azathioprine, cyclophosphamide, and/or rituximab. Full recovery occurred in 39% and a partial response was seen in 52% though 25% of patients had a clinical relapse. Due to the rarity of this disease, no head-to-head comparisons of the above treatments have been carried out. A potential therapeutic strategy for these diseases would be to reduce IgG4-producing B cells. In support of this, Rituximab, a monoclonal antibody against CD20, has been shown to be effective in the treatment of anti-MuSK myasthenia gravis and anti-LGI1 limbic encephalitis with long-term remission.46–48 Similarly, there are reported cases of good treatment response to Rituximab in otherwise treatment refractory anti-Caspr2 autoimmune encephalitis.2,49 Strategies directed at other components of the immune system (e.g. inhibiting complement or T cell therapies), however, may not be effective.

Acknowledgments

This work was supported by NIH/NINDS R25 NS065745 (KP), NIH/NINDS R01 NS077851 (JD), FIS PI14/00203 (JD), and NIH/NINDS K08 NS075142 (EL). Dr Lancaster received a Dana Foundation Award. The authors would like to thank Dr. Elior Peles for the Caspr2 cDNA construct, Dr. Steven S. Scherer for discussions and reading of this manuscript, Marjorie Maronski for her work culturing primary rat hippocampal neurons, and our colleagues for patient referrals.

Footnotes

Author Contributions

KP, JD, and EL contributed to study concept and design; KP and EL contributed to data acquisition and analysis; KP, JD, and EL contributed to drafting of the manuscript and figures.

Potential Conflicts of Interest

Dr Dalmau receives royalties from Euroimmun for the use of NMDA, GABAB receptor, GABAA receptor, DPPX and IgLON5 as autoantibody tests; he has received an unrestricted research grant from Euroimmun.

References

  • 1.Poliak S, Gollan L, Martinez R, et al. Caspr2, a new member of the neurexin superfamily, is localized at the juxtaparanodes of myelinated axons and associates with K+ channels. Neuron. 1999;24(4):1037–1047. doi: 10.1016/S0896-6273(00)81049-1. [DOI] [PubMed] [Google Scholar]
  • 2.Lancaster E, Huijbers MGM, Bar V, et al. Investigations of caspr2, an autoantigen of encephalitis and neuromyotonia. Ann Neurol. 2011 doi: 10.1002/ana.22297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Irani SR, Pettingill P, Kleopa Ka, et al. Morvan syndrome: Clinical and serological observations in 29 cases. Ann Neurol. 2012;72(2):241–255. doi: 10.1002/ana.23577. [DOI] [PubMed] [Google Scholar]
  • 4.Van Sonderen A, Ari??o H, Petit-Pedrol M, et al. The clinical spectrum of Caspr2 antibody-associated disease. Neurology. 2016;87(5):521–528. doi: 10.1212/WNL.0000000000002917. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Rasband MN, Trimmer JS. Developmental clustering of ion channels at and near the node of Ranvier. Dev Biol. 2001;236(1):5–16. doi: 10.1006/dbio.2001.0326. [DOI] [PubMed] [Google Scholar]
  • 6.Ogawa Y, Oses-Prieto J, Kim MY, et al. ADAM22, a Kv1 channel-interacting protein, recruits membrane-associated guanylate kinases to juxtaparanodes of myelinated axons. J Neurosci. 2010;30(3):1038–1048. doi: 10.1523/JNEUROSCI.4661-09.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Poliak S, Salomon D, Elhanany H, et al. Juxtaparanodal clustering of Shaker-like K+ channels in myelinated axons depends on Caspr2 and TAG-1. J Cell Biol. 2003 doi: 10.1083/jcb.200305018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Traka M, Goutebroze L, Denisenko N, et al. Association of TAG-1 with Caspr2 is essential for the molecular organization of juxtaparanodal regions of myelinated fibers. J Cell Biol. 2003 doi: 10.1083/jcb.200305078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Lindstrom JM. Acetylcholine receptors and myasthenia. Muscle and Nerve. 2000;23(4):453–477. doi: 10.1002/(SICI)1097-4598(200004)23:4&#x0003c;453::AID-MUS3&#x0003e;3.0.CO;2-O. [DOI] [PubMed] [Google Scholar]
  • 10.Hughes EG, Peng X, Gleichman AJ, et al. Cellular and synaptic mechanisms of anti-NMDA receptor encephalitis. J Neurosci. 2010;30(17):5866–5875. doi: 10.1523/JNEUROSCI.0167-10.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Moscato EH, Peng X, Jain A, Parsons TD, Dalmau J, Balice-Gordon RJ. Acute mechanisms underlying antibody effects in anti-N-methyl-D-aspartate receptor encephalitis. Ann Neurol. 2014 doi: 10.1002/ana.24195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Gleichman AJ, Panzer Ja, Baumann BH, Dalmau J, Lynch DR. Antigenic and mechanistic characterization of anti-AMPA receptor encephalitis. Ann Clin Transl Neurol. 2014;1(3):180–189. doi: 10.1002/acn3.43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Planagumà J, Leypoldt F, Mannara F, et al. Human N-methyl D-aspartate receptor antibodies alter memory and behaviour in mice. Brain. 2015;138(1):94–109. doi: 10.1093/brain/awu310. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Peng X, Hughes EG, Moscato EH, Parsons TD, Dalmau J, Balice-Gordon RJ. Cell Plast Induc by anti-AMPA Recept Enceph antibodies. Ann NeurolAnnals Neurol. 2014 doi: 10.1002/ana.24293. [DOI] [Google Scholar]
  • 15.Pinatel D, Hivert B, Boucraut J, et al. Inhibitory axons are targeted in hippocampal cell culture by anti-Caspr2 autoantibodies associated with limbic encephalitis. Front Cell Neurosci. 2015;9(July):265. doi: 10.3389/fncel.2015.00265. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Querol L, Nogales-Gadea G, Rojas-Garcia R, et al. Neurofascin IgG4 antibodies in CIDP associate with disabling tremor and poor response to IVIg. Neurology. 2014 doi: 10.1212/WNL0000000000000205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Huijbers MG, Querol La, Niks EH, et al. The expanding field of IgG4-mediated neurological autoimmune disorders. Eur J Neurol. 2015 doi: 10.1111/ene.12758. n/a-n/a. [DOI] [PubMed] [Google Scholar]
  • 18.Ogata H, Yamasaki R, Hiwatashi A, et al. Characterization of IgG4 anti-neurofascin 155 antibody-positive polyneuropathy. Ann Clin Transl Neurol. 2015;2(10):960–971. doi: 10.1002/acn3.248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Devaux J, Yumako M, Yumako F, et al. Neurofascin-155 IgG4 in chronic inflammatory demyelinating polyneuropathy. Neurology. 2016;86:800–807. doi: 10.1212/WNL.0000000000002418. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Aalberse RC, Schuurman J. IgG4 breaking the rules. Immunology. 2002;105(1):9–19. doi: 10.1046/j.0019-2805.2001.01341.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Van Sonderen A, Ariño H, Petit-Pedrol M, et al. The clinical spectrum of Caspr2 antibody-associated disease. Neurology. 2016;87(5):521–528. doi: 10.1212/WNL.0000000000002917. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Lu Z, Reddy MVVVS, Liu J, et al. Molecular architecture of contactin-Associated protein-like 2 (CNTNAP2) and its interaction with contactin 2 (CNTN2) J Biol Chem. 2016;291(46):24133–24147. doi: 10.1074/jbc.M116.748236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Danielson E, Lee SH, Fox MA. SynPAnal: Software for rapid quantification of the density and intensity of protein puncta from fluorescence microscopy images of neurons. PLoS One. 2014;9(12) doi: 10.1371/journal.pone.0115298. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Olsen AL, Lai Y, Dalmau J, Scherer SS, Lancaster E. Caspr2 autoantibodies target multiple epitopes. Neurol Neuroimmunol Neuroinflammation. 2015 doi: 10.1212/NXI.0000000000000127. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Horresh I, Poliak S, Grant S, Bredt D, Rasband MN, Peles E. Multiple molecular interactions determine the clustering of Caspr2 and Kv1 channels in myelinated axons. J Neurosci. 2008;28(52):14213–14222. doi: 10.1523/JNEUROSCI.3398-08.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Zhou L, Zhang CL, Messing a, Chiu SY. Temperature-sensitive neuromuscular transmission in Kv1.1 null mice: role of potassium channels under the myelin sheath in young nerves. J Neurosci. 1998;18(18):7200–7215. doi: 10.1523/JNEUROSCI.18-18-07200.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Zhou L, Messing a, Chiu SY. Determinants of excitability at transition zones in Kv1.1-deficient myelinated nerves. J Neurosci. 1999;19(14):5768–5781. doi: 10.1523/JNEUROSCI.19-14-05768.1999. http://www.ncbi.nlm.nih.gov/pubmed/10407018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Clark JD, Tempel BL. Hyperalgesia in mice lacking the Kv1.1 potassium channel gene. Neurosci Lett. 1998;251(2):121–124. doi: 10.1016/S0304-3940(98)00516-3. [DOI] [PubMed] [Google Scholar]
  • 29.Devaux JJ, Miura Y, Fukami Y, et al. Neurofascin-155 IgG4 in chronic inflammatory demyelinating polyneuropathy. Neurology. 2016;86(9):800–807. doi: 10.1212/WNL.0000000000002418. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Man JK, Malotka J, Kawakami N, et al. Neurofascin as a target for autoantibodies in peripheral neuropathies. Neurology. 2012;79(23):2241–2248. doi: 10.1212/WNL.0b013e31827689ad. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Labasque M, Hivert B, Nogales-Gadea G, Querol L, Illa I, Faivre-Sarrailh C. Specific contactin N-glycans are implicated in neurofascin binding and autoimmune targeting in peripheral neuropathies. J Biol Chem. 2014 doi: 10.1074/jbc.M113.528489. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Manso C, Querol L, Mekaouche M, Illa I, Devaux JJ. Contactin-1 IgG4 antibodies cause paranode dismantling and conduction defects. Brain. 2016;139(6):1700–1712. doi: 10.1093/brain/aww062. [DOI] [PubMed] [Google Scholar]
  • 33.Miura Y, Devaux JJ, Fukami Y, et al. Contactin 1 IgG4 associates to chronic inflammatory demyelinating polyneuropathy with sensory ataxia. Brain. 2015;138(6):1484–1491. doi: 10.1093/brain/awv054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Delmont E, Manso C, Querol L, et al. Autoantibodies to nodal isoforms of neurofascin in chronic inflammatory demyelinating polyneuropathy. Brain. 2017;140(7):1851–1858. doi: 10.1093/brain/awx124. [DOI] [PubMed] [Google Scholar]
  • 35.Kadoya M, Kaida K, Koike H, et al. IgG4 anti-neurofascin155 antibodies in chronic inflammatory demyelinating polyradiculoneuropathy: Clinical significance and diagnostic utility of a conventional assay. J Neuroimmunol. 2016;301:16–22. doi: 10.1016/j.jneuroim.2016.10.013. [DOI] [PubMed] [Google Scholar]
  • 36.Doppler K, Appeltshauser L, Wilhelmi K, et al. Destruction of paranodal architecture in inflammatory neuropathy with anti-contactin-1 autoantibodies. J Neurol Neurosurg Psychiatry. 2015;86(7):720–728. doi: 10.1136/jnnp-2014-309916. [DOI] [PubMed] [Google Scholar]
  • 37.Querol L, Nogales-Gadea G, Rojas-Garcia R, et al. Antibodies to contactin-1 in chronic inflammatory demyelinating polyneuropathy. Ann Neurol. 2013;73(3):370–380. doi: 10.1002/ana.23794. [DOI] [PubMed] [Google Scholar]
  • 38.Sherman DL, Tait S, Melrose S, et al. Neurofascins are required to establish axonal domains for saltatory conduction. Neuron. 2005;48(5):737–742. doi: 10.1016/j.neuron.2005.10.019. [DOI] [PubMed] [Google Scholar]
  • 39.Feinberg K, Eshed-Eisenbach Y, Frechter S, et al. A Glial Signal Consisting of Gliomedin and NrCAM Clusters Axonal Na+ Channels during the Formation of Nodes of Ranvier. Neuron. 2010;65(4):490–502. doi: 10.1016/j.neuron.2010.02.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Eshed Y, Feinberg K, Poliak S, et al. Gliomedin mediates Schwann cell-axon interaction and the molecular assembly of the nodes of Ranvier. Neuron. 2005;47(2):215–229. doi: 10.1016/j.neuron.2005.06.026. [DOI] [PubMed] [Google Scholar]
  • 41.Charles P, Tait S, Faivre-Sarrailh C, et al. Neurofascin is a glial receptor for the paranodin/Caspr-contactin axonal complex at the axoglial junction. Curr Biol. 2002;12(3):217–220. doi: 10.1016/S0960-9822(01)00680-7. [DOI] [PubMed] [Google Scholar]
  • 42.Huijbers MG, Zhang W, Klooster R, et al. MuSK IgG4 autoantibodies cause myasthenia gravis by inhibiting binding between MuSK and Lrp4. Proc Natl Acad Sci U S A. 2013;110(51):20783–20788. doi: 10.1073/pnas.1313944110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Kawakami Y, Ito M, Hirayama M, et al. Anti-MuSK autoantibodies block binding of collagen Q to MuSK. Neurology. 2011;77(20):1819–1826. doi: 10.1212/WNL.0b013e318237f660. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Otsuka K, Ito M, Ohkawara B, et al. Collagen Q and anti-MuSK autoantibody competitively suppress agrin/LRP4/MuSK signaling. Sci Rep. 2015;5(August):13928. doi: 10.1038/srep13928. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Ohkawa T, Fukata Y, Yamasaki M, et al. Autoantibodies to epilepsy-related LGI1 in limbic encephalitis neutralize LGI1-ADAM22 interaction and reduce synaptic AMPA receptors. J Neurosci. 2013;33(46):18161–18174. doi: 10.1523/JNEUROSCI.3506-13.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Illa I, Diaz-Manera J, Rojas-Garcia R, et al. Sustained response to Rituximab in anti-AChR and anti-MuSK positive Myasthenia Gravis patients. J Neuroimmunol. 2008;201–202(C):90–94. doi: 10.1016/j.jneuroim.2008.04.039. [DOI] [PubMed] [Google Scholar]
  • 47.Díaz-Manera J, Martínez-Hernández E, Querol L, et al. Long-lasting treatment effect of rituximab in MuSK myasthenia. Neurology. 2012;78(3):189–193. doi: 10.1212/WNL.0b013e3182407982. [DOI] [PubMed] [Google Scholar]
  • 48.Brown JWL, Martin PJ, Thorpe JW, et al. Long-term remission with rituximab in refractory leucine-rich glioma inactivated 1 antibody encephalitis. J Neuroimmunol. 2014;271(1–2):66–68. doi: 10.1016/j.jneuroim.2014.03.012. [DOI] [PubMed] [Google Scholar]
  • 49.Ong E, Viaccoz A, Ducray F, et al. Dramatic improvement after rituximab in a patient with paraneoplastic treatment-refractory Morvan syndrome associated with anti-CASPR2 antibodies. Eur J Neurol. 2013;20(7):96–97. doi: 10.1111/ene.12164. [DOI] [PubMed] [Google Scholar]

RESOURCES