ARTICLE OPEN ACCESS

Antindashpan-neurofascin IgG3 as a marker offulminant autoimmune neuropathyHelena Stengel Atay Vural MD PhD Anna-Michelle Brunder Annika Heinius Luise Appeltshauser MD

Bianca Fiebig Florian Giese MD Christian Dresel MD Aikaterini Papagianni MD Frank Birklein MD PhD

Joachim Weis MD Tessa Huchtemann MD Christian Schmidt MD Peter Kortvelyessy MD

Carmen Villmann PhD Edgar Meinl MD Claudia Sommer MD PhD Frank Leypoldt MD and

Kathrin Doppler MD

Neurol Neuroimmunol Neuroinflamm 20196e603 doi101212NXI0000000000000603

Correspondence

Dr Doppler

Doppler_Kukwde

AbstractObjectiveTo identify and characterize patients with autoantibodies against different neurofascin (NF)isoforms

MethodsScreening of a large cohort of patient sera for anti-NF autoantibodies by ELISA and furthercharacterization by cell-based assays epitope mapping and complement binding assays

ResultsTwo different clinical phenotypes became apparent in this study The well-known clinicalpicture of subacute-onset severe sensorimotor neuropathy with tremor that is known to beassociated with IgG4 autoantibodies against the paranodal isoform NF-155 was found in 2patients The second phenotype with a dramatic course of disease with tetraplegia and almostlocked-in syndrome was associated with IgG3 autoantibodies against nodal and paranodalisoforms of NF in 3 patients The epitope against which these autoantibodies were directed inthis second phenotype was the common Ig domain found in all 3 NF isoforms In contrastantindashNF-155 IgG4 were directed against the NF-155ndashspecific Fn3Fn4 domain The descriptionof a second phenotype of antindashNF-associated neuropathy is in line with some case reports ofsimilar patients that were published in the last year

ConclusionsOur results indicate that antindashpan-NF-associated neuropathy differs from antindashNF-155-associated neuropathy and epitope and subclass play a major role in the pathogenesis andseverity of antindashNF-associated neuropathy and should be determined to correctly classifypatients also in respect to possible differences in therapeutic response

These authors contributed equally

From the Department of Neurology (HS AMB LA BF AP CS KD) University Hospital Wurzburg Institute of Clinical Neuroimmunology (AV EM) Biomedical CenterUniversity Hospitals Ludwig-Maximilians-Universitat Munchen Planegg-Martinsried Universitatsklinikum Schleswig-Holstein Campus Kiel (AH FL) Neuroimmunology SectionInstitute of Clinical Chemistry KielLubeck Department of Neurology (FG) University Hospital Halle Department of Neurology (CD FB) University HospitalMainzMainz UniversityHospital Aachen (JW) Institute of Neuropathology Aachen Department of Neurology (TH PK) University Hospital Magdeburg Institute for Pharmacology and Toxicology (CS)Otto-von-Guericke University German Center for Neurodegenerative Diseases (PK) Magdeburg Institute for Clinical Neurobiology (CV) University Hospital Wurzburg Departmentof Neurology (FL) Universitatsklinikum Schleswig-Holstein Kiel Germany and Research Center for Translational Medicine (AV) Koccedil University Istanbul Turkey

Go to NeurologyorgNN for full disclosures Funding information is provided at the end of the article

The Article Processing Charge was funded by the authors

This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 40 (CC BY-NC-ND) which permits downloadingand sharing the work provided it is properly cited The work cannot be changed in any way or used commercially without permission from the journal

Copyright copy 2019 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology 1

The node of Ranvier has long been considered a potential siteof attack in inflammatory neuropathies With respect to theconcept of complement-mediated reversible conductionblock induced by antiganglioside autoantibodies the termldquoparanodopathynodopathyrdquo was introduced by Unciniet al12 In the past few years nonndashcomplement-fixing IgG4autoantibodies against paranodal proteins were described inpatients with immune neuropathies and were also comprisedunder that term34 Patients with autoantibodies against par-anodal proteins present with a distinct clinical phenotype ofsubacute-onset severe sensorimotor neuropathy with poorresponse to intravenous immunoglobulins but excellent re-sponse to rituximab5ndash10 Autoantibodies are mostly of theIgG4 subtype but IgG3 and IgG2 autoantibodies have alsobeen described possibly associated with early disease ormonophasic course910 In these cases complement de-position could be demonstrated11 Anti-neurofascin(NF)-155 appears to be the most prevalent paranodal auto-antibody and is strongly associated with tremor61213 NF-155is located at the paranodes whereas NF-186 and -140 arenodal isoforms the latter mainly expressed during embryonicdevelopment1415 Although all 3 isoforms share the Ig domainand some parts of the fibronectin domain16 most autoanti-bodies described so far were selectively directed against NF-15568 However 2 recent studies could also detect anti-NF-186140 autoantibodies mostly of the IgG4 subclass inpatients with chronic inflammatory demyelinating poly-radiculoneuropathy (CIDP)1718 Single cases of patients withautoantibodies against all isoforms and severe neuropathyhave been described1719 In contrast to antindashNF-155-associated neuropathy not much is known about neuropa-thy with antindashpan-NF autoantibodies

In the present study we aimed to identify and characterizepatients with autoantibodies against NF to potentially estab-lish antindashpan-NF-associated neuropathy as a distinct clinicalphenotype

MethodsPatientsThree different subcohorts of patients were included sera of182 patients with clinically suspected inflammatory neurop-athy that were sent to our department for testing of paranodalautoantibodies further referred to as ldquocomprehensive di-agnostic testing cohortrdquo sera of 306 patients with a differentialdiagnosis of inflammatory neuropathy who attended our de-partment for diagnostic workup (32 finally diagnosed asGuillain-Barre syndrome (GBS) 81 as CIDP fulfilling theEuropean Federation of Neurological Societies criteria20 and193 as noninflammatory neuropathy or suspected

inflammatory neuropathy not fulfilling the EFNS criterialdquoWurzburg cohortrdquo) and 181 sera of a patient cohort from theUniversity Hospitals of Kiel and Magdeburg that were col-lected between 2004 and 2016 (146 GBS 21 CIDP and 15Miller-Fisher syndromeBickerstaff encephalitis ldquoKielMagdeburg cohortrdquo) Diagnosis of GBS was assessed by theBrighton criteria21 CSF was obtained from diagnostic lumbarpuncture The study was approved by the Ethics Committeesof the Universities of Wurzburg and Kiel Demographic dataare summarized in table 1

Enzyme-linked immunosorbent assayAll patients were screened for anti-NF-155 autoantibodies byELISA and 252 patients of the Wurzburg cohort and allantindashNF-155-positive patients were also screened for anti-NF-186 All antindashNF-155-positive patients were tested foranti-NF-140 CSF was available from 3 seropositive patientsand was tested at a dilution of 120 Five patients withouta clinical history of neuropathy who had received lumbarpuncture for other diagnostic purposes served as controlsNF-155-186 ELISA was performed as previously describedusing the extracellular domains of human NF-155 and -1868

For anti-NF-140 ELISA wells were coated with recombinanthuman NF-140 protein (1 μgmL Sino Biological Inc Bei-jing China) Normal controls were run in each assay Thenormal value was set at 5 SDs above the mean of all healthycontrols that were tested in previous studies22 Titers weredetermined by measuring serial dilutions of antindashNF-positivesera Subclasses were detected using subclass-specifichorseradish peroxidase-conjugated secondary antibodies(anti-IgG1 11000 anti-IgG2 14000 anti-IgG3 13000and anti-IgG4 12000 Invitrogen Carlsbad CA)

Binding assays with murine teased fiberssympathetic ganglia cerebellar granularneurons and rat cerebellum sectionsAll sera were screened for paranodal and nodal autoantibodiesusing binding assays with murine teased fibers as previouslydescribed9 The preparations were incubated with patientsrsquosera diluted 1100 and 1500 overnight at 4degC following in-cubation with appropriate antindashhuman Cy3-conjugated sec-ondary antibodies (Dianova Hamburg Germany 1100)Cervical sympathetic ganglia were dissected from adult Lewisrats and 10-μm cryosections were cut Binding assays withpatientsrsquo and controlsrsquo sera were performed with a dilution of1100 and 1500 by incubating overnight Cy3-conjugatedanti-human secondary antibodies were added for 2 hours Allslides were assessed using a fluorescence microscope (ZeissAx10 Zeiss Oberkochen Germany) Cerebellar granularneurons were prepared from P5 mice the cerebellum wasdissected and cells were dissociated and suspended using

GlossaryCIDP = chronic inflammatory demyelinating polyradiculoneuropathy GBS = Guillain-Barre syndrome HEK = humanembryonic kidney MFI = mean fluorescence intensity MFS = Miller-Fisher syndrome NF = neurofascin

2 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 NeurologyorgNN

a papain dissociation system (Worthington BiochemicalCorporation Lakewood NJ) Cells were plated on cover slipsat a density of 200000 cells per well and fixed with 4paraformaldehyde after 3 days and binding assays with sera ofpatients and controls were performed at a dilution of 1500using an anti-human Cy3-conjugated secondary antibody(Dianova 1300) Antindashpan-NF served as a positive control

Binding assays on rat cerebellum sections were performedovernight at a serum dilution of 1200 or CSF dilution of 14 on6-μm sagittal adult rat brain cryosections optimized for de-tection of surface antigens as has been previously described23

Binding assays with NF-transfected humanembryonic kidney (HEK) 293 cellsBinding assays with HEK293 cells transiently transfected withplasmids of human NF-155 and -1868 were performed aspreviously described22 Sera were used at a dilution of 1200and 1500 Antindashpan-NF (11000) served as a control anti-body Serum of 1 patient (patient 6) with immunoreactivityagainst NF-155 by ELISA but no binding to NF-transfectedHEK293 cells at a dilution of 1200 and 1500 was additionallytested at a dilution of 140 Immunostaining was assessedusing a fluorescence microscope (Zeiss Ax10)

Preabsorption experimentsPreabsorption was performed by serial incubation of patientsrsquosera (1500) 1 hour each in 3 wells of a 24-well plate with NF-155-186ndashtransfected HEK293 cells as described before9 Seraincubated in wells with contactin-1ndashtransfected cells served ascontrols After incubation with transfected HEK293 cells serawere used for binding assays on teased fibers as described above

Epitope characterizationTE671 human rhabdomyosarcoma cells stably transfectedwith NF-155 NF-186 or 6 truncated forms of NF-155 andNF-186 were previously described8 Cells were incubatedwith patientsrsquo and controlsrsquo sera (1100) Detection of bindingwas performed as described8 with somemodifications SerumIgG binding was detected using Biotin-SP-conjugated

AffiniPure F(abrsquo)2 antibodies (Jackson ImmunoResearch)and AF647-conjugated streptavidin (Jackson ImmunoR-esearch) Mean fluorescence intensity (MFI) of the AF647channel was calculated after gating on enhanced green fluo-rescent protein (EGFP)high cells for NF truncations andempty EGFP vectorndashtransfected cells and delta MFI wascalculated from these values Patient 5 was only tested forFn3Fn4 as the sample was obtained at a later time point andNF-155 isoform specificity had already been shown by ELISA

Complement binding assayELISA plates were coated blocked and incubated withpatientsrsquo sera as described above The next day the wells wereincubated with C1q (10 μgmL Sigma-Aldrich St LouisMO) for 2 hours at room temperature following incubationwith horseradish peroxidasendashconjugated anti-C1q (LifeSpanBiosciences Seattle WA 1200) for 30 minutes Measure-ment of optical density was made as described above

Data availability statementAny data not published within the article will be shared byrequest from any qualified investigator in anonymized form

ResultsDetection of anti-NF autoantibodiesAnti-NF-155 autoantibodies were detectable by ELISA in seraof 6 patients (figure 1) Titers ranged from 1300 to 16000(table 2) Three of the 6 antindashNF-155-positive patients werealso positively tested for anti-NF-186 and -140 (patients 1ndash3table 2) Sera of 2 patients showed an optical density slightlyabove the cutoff value in the anti-NF-155 ELISA but a secondserum sample of these patients obtained during the course ofdisease was clearly negative Of note in none of the 252neuropathy cases tested for anti-NF-186 independent of NF-155 positivity did we observe isolated anti-NF-186 positivityAnti-NF-155 andor -186 autoantibodies were confirmed bybinding assays with NF-155ndash and NF-186ndashtransfectedHEK293 cells in 5 of these patients (patients 1ndash5 figure 1)Binding assays on murine teased fibers revealed a distinct

Table 1 Demographic data of patients of all subcohorts

Wurzburg cohort (n = 306) KielMagdeburg cohort (n = 181)Comprehensive diagnostic testing cohort(n = 182)

Median age(range)

62 (8ndash84) 59 (18ndash87) 59 (18ndash84)

Sex 218 male 88 female 91 male 90 female 136 male 46 female

Diagnosis

GBS 32 (22 level 1 6 level 2 3 level 3 and 1level 4)

146 (75 level 1 41 level 2 6 level 3 and 24level 4)

NA

CIDP 81 21 NA

Others 193 14 MFS 1 Bickerstaff NA

Abbreviations CIDP = chronic inflammatory demyelinating polyradiculoneuropathy GBS = Guillain-Barre syndrome MFS = Miller-Fisher syndrome

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 3

paranodal staining in the patients with anti-NF-155 auto-antibodies only (patients 4ndash6 figure 2 FndashH) and a nodaland paranodal staining in the 3 patients with antindashpan-NFautoantibodies (patients 1ndash3 figure 2 A C and D) Inpatient 1 paranodal staining was stronger than nodalstaining and in patients 2 and 3 nodal staining wasstronger (Figure 2A C-E) In patients 3 and 6 the fluo-rescence signal was weaker than in the other patients mostprobably due to the low titer (Figure 2E H) In patients 1and 2 paranodal binding became stronger during thecourse of disease (Figure 2B D)

Preincubation assays were performed to confirm paranodalnodal binding as anti-NF specific In the 2 patients who wereonly positive for anti-NF-155 (patients 4 and 5) paranodalbinding was lost after preincubation with NF-155ndashtransfectedcells not after incubation with NF-186ndashtransfected cells In 2of the patients with autoantibodies against both isoforms(patients 1 and 2) paranodal and nodal binding were lost afterincubation with NF-155-transfected cells but weaker para-nodal binding persisted after incubation with NF-186ndashtransfected cells suggesting the presence of at least 2 auto-antibodies with different epitopes (supplemental figure 1linkslwwcomNXIA132 and figure 3)

CSF was available from patients 2 3 and 6 No anti-NFautoantibodies were detectable in CSF by ELISA in patients 3and 6 In patient 2 no anti-NF autoantibodies were detectablein CSF by ELISA at the onset of disease (July 2014) or after 2months (September 2014) However 6 months after theonset of disease anti-NF autoantibodies were detectablemost probably due to disruption of the blood-brain barrier(CSFserum albumin index 372 times 10minus3) There was no evi-dence of intrathecal IgG production at this time point as nooligoclonal bands were detectable and the CSF IgG index waswithin the normal range (059)

Comparison of clinical phenotypes in patientswith pan-NF or NF-155ndashspecific antibodiesAll 6 antindashNF-positive patients had a clinical diagnosis ofCIDP except for patient 3 who had a monophasic course andwas classified as GBS Yet detailed analysis of clinical picturesshowed that antindashNF-positive patients segregated into 2 dis-tinct phenotypes

Patients 1ndash3 whowere seropositive for anti-NF-155 -NF-186and ndashNF-140 had a fulminant course of disease After GBS-like onset patients 1ndash3 quickly developed tetraplegia andsevere cranial nerve involvement No antiganglioside

Figure 1 Binding assays with NF-155ndash and NF-186ndashtransfected HEK293 cells (A-G) and ELISA (H)

Patients 1 2 and 3 show a clear binding to NF-155ndash and NF-186ndashtransfected cells (AndashC)whereas patients 4 and 5 only bind to NF-155ndashtransfected and not to NF-186ndashtransfected cells(F and G) Healthy control serum does not bindto the cells (E) and binding of commercialantindashpan-NF autoantibodies to NF-155ndash andNF-186ndashtransfected cells is shown in (D) NF =neurofascin OD = optical density

4 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 NeurologyorgNN

Table 2 Overview of assays and clinical data of antindashNF-positive patients

Patient 1 2 3 4 5 6

ELISA NF-155 -186 -140positive

NF-155 -186 -140positive

NF-155 -186 -140positive

Only NF-155positive

Only NF-155positive

Only NF-155positive

ELISA titer (onset) 14000 (NF-155) 12000 (NF-186)

11000 (NF-155) 1500(NF-186)

1300 (NF-155) 1200 (NF-186)

16000 (NF-155)

12500 (NF-155)

1500 (N-155)

IgG subclass (anti-NF-155 ELISA)

IgG3 IgG3 (course of diseaseweak IgG4)

Weak IgG3 IgG4gtIgG2 IgG4 IgG3

Cell-based assay NF-155 -186 positive NF-155 -186 positive NF-155 -186positive

NF-155positive

NF-155positive

Negative

Truncated NF (flowcytometry)

6Ig domain gt Fn3 6Ig domain gtgt Fn3 6Ig domain Fn3Fn4 Fn3Fn4 NA

IgG subclass (flowcytometry)

IgG3 IgG3 Weak IgG3 IgG4 IgG4 NA

Teased fiber binding Paranodal gtgt nodal Nodal gt paranodal Weak nodal andparanodal

Paranodal Paranodal Weakparanodal

Summary ofautoantibodyfinding

Definite IgG3antindashpan-NF Definite IgG3 antindashpan-NF

DefiniteIgG3 antindashpan-NF

Definite IgG4anti-NF-155

Definite IgG4anti-NF-155

Possible IgG3 anti-NF-155

Cohort Comprehensivediagnostictesting

Kiel Kiel Wurzburg Comprehensivediagnostictesting

Kiel

Diagnosis CIDP CIDP GBS CIDP CIDP CIDP

Agesex 71 male 63 male 74 female 28 male 18 male 49 male

CSF (cellcountprotein)

lt1μL907 mgL 2μL621 mgL lt1μL298mgL

4μL2011 mgL 3μL4400 mgL 3μL616 mgL

Autonomicsymptoms

Cardiacresuscitation

Severe blood pressureinstability paralytic ileus

No Orthostatichypotensionpollakiuriahyperhidrosis

No No

Cranial nerveinvolvement

Almostcompletelocked-in

Almost complete locked-in Dysphagia No No No

Respiratoryinsufficiency

Ventilated Ventilated No No No No

Motorsymptoms

Tetraplegia Tetraplegia Almosttetraplegia

Distal gt proximalparesis tremor

Mild distalparesis tremor

Severe distal gtproximal tetraparesis

Sensorysymptoms

Pallanesthesiadistalhypesthesia

Hypesthesia Hypesthesia Neuropathic pain Mild distalhypesthesia

Hypesthesia

Concomitantdiseases

Cervicalmyelopathy

Cervical myelopathy bullouspemphigoid

Chroniclymphaticleukemia

No No No

Response totreatment

None Glucocorticoids Plasmaexchange

Mildimprovement onIVIG PE andrituximab

Response to PEand rituximab

Mild response to IVIG

Outcome Died due tosepsis 10 moafter onset

Complete recovery of cranialnerves residual paresis of legsand arms (October 2018)

Completerecovery

Only minorimprovement

Improvement oftremor

Wheelchair user distalparesis andhypesthesia (February2009)

Abbreviations CIDP = chronic inflammatory demyelinating polyradiculoneuropathy GBS = Guillain-Barre syndrome IVIG = intravenous immunoglobulinsNA = not available NF = neurofascin PE = plasma exchange

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 5

antibodies were detectable Patient 3 quickly recovered withina few months Patients 1 and 2 needed mechanical ventilationfor several months with complete tetraplegia and were almostlocked-in Patient 1 died due to sepsis and patient 2 had a veryprotracted course with almost 2 years in tetraplegia and stillrecovering almost 5 years after onset of symptoms

Patients 4 and 5 were young men with anti-NF-155 auto-antibodies and presented with motor more than sensoryneuropathy associated with tremor as described in multiplereports613 Remarkably autonomic involvement wasa major problem in patient 4 In patient 5 action andpostural tremor and gait ataxia were the dominant symp-toms with only minor neuropathic complaints and no au-tonomic symptoms

In patient 6 anti-NF-155 autoantibodies could not be con-firmed by cell-based assay The patient had severe sensori-motor neuropathy without tremor

For detailed clinical data course of disease and response totreatment see Table 2 and Supplements (supplement courseof disease linkslwwcomNXIA137)

Different patterns of reactivity ofautoantibodies against truncations of NFTo find out whether the autoantibodies in patients 1ndash3 weredirected against different epitopes or target a common epi-tope of all NF isoforms and were thus responsible for thepositive reactivity to all 3 NF binding assays with truncatedNF mutants were performed Patients 1ndash3 showed immu-noreactivity against the Ig domains of NF that are part of allisoforms In addition patients 1 and 2 were found to beweakly immunoreactive to Fn3 a component of the NF-155isoform Immunoreactivity to Fn3 as a second NF-155ndashspecific epitope is in line with the results of the preincubationassays explaining persistence of weak paranodal binding afterpreincubation with NF-186ndashtransfected HEK293 cells Ad-ditional immunoreactivity to Fn3 as a NF-155ndashspecific epi-tope also accounts for higher titers of anti-NF-155 comparedwith anti-NF-186 Patients 4 and 5 were immunoreactiveagainst Fn3Fn4 a component of the fibronectin domain ofNF-155 that is isoform specific (figure 3)

IgG subclasses and C1q bindingELISA with subclass-specific secondary antibodies detectedIgG4 and IgG2 as the predominant subclasses in patient 4 IgG4in patient 5 and IgG3 in patients 1 2 and 6 In patient 1 IgG2was detectable at the onset of disease No specific subclass wasdetectable in patient 3 most probably due to the low titer andlower sensitivity of the subclass-specific ELISA (figure 4) Inpatient 2 a minor titer of IgG4 was transiently detectable duringthe course of disease at a time when IgG3 titers peaked atextraordinarily high titers (supplemental table linkslwwcomNXIA133) Results of subclasses were confirmed by flowcytometry Complement deposition induced by autoantibodybindingwas found in patients 1 2 and 4 but not in patients 3 and6 (both with low autoantibody titers and patient 6 withoutconfirmation of anti-NF-155 by cell-based assays) and patient 5(with IgG4 only) (figure 4) In patients 1 and 2 sera of differenttime points of disease with different autoantibody titers weremeasured and in total C1q deposition correlated with IgG3immunoreactivity (p = 0009 r = 072)

Autonomic and cranial nerve involvement inantindashNF-positive patients and binding tothe cerebellumAutonomic involvement was present in 3 of the patientsPatients 1 and 2 developed severe autonomic dysfunctionduring intensive care treatment Patient 1 was resuscitatedafter cardiac arrest Patient 4 had disabling orthostatic dys-function increased transpiration and pollakiuria As there isno obvious link between NF and the autonomic nervous

Figure 2 Microphotographs of binding assays of patientsrsquosera to murine teased fibers

Nodal binding is marked with arrows and paranodal binding with asterisksSerum of patient 1 clearly binds to the nodal and paranodal regions at theonset of disease (A) Paranodal binding becomes more prominent duringthe course of disease (B) Patient 2 binds to the nodes stronger than to theparanodes (C and D) but paranodal binding becomes stronger during thecourse of disease (D) Patient 3mainly binds to the nodal region (E) whereaspatients 4 and 5with anti-NF-155 autoantibodies bind to the paranodes only(F andG) Patient 6weakly binds to the paranodes (H) A normal control doesnot show any binding (I) and the commercial antindashpan-NF control antibodyclearly binds to the nodes and paranodes (J) Bar = 10 μm NF = neurofascin

6 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 NeurologyorgNN

system and postganglionic autonomic fibers are un-myelinated binding assays with patientsrsquo sera on sympa-thetic cervical ganglia were performed but we did notdetect any specific binding of patientsrsquo sera to autonomicneurons (data not shown) To assess binding of antindashNF-positive sera to preganglionic B-fibers we performedbinding assays with anterior root teased fibers of rats thatcontain preganglionic sympathetic B-fibers B-fibers aredistinguishable from A-alpha- and A-gamma-fibers by theirdiameter Binding of patientsrsquo sera to ventral root fibers ofall diameters (supplemental figure 3 linkslwwcomNXIA135) and also to A-beta fibers of the dorsal root wasdetected As former studies have proposed binding of anti-NF sera to the cerebellum as a possible correlate oftremor613 we also performed binding assays with cere-bellum sections that showed clear binding to the molecular

and granular layer in patients 1ndash4 and to the molecularlayer only in patient 5 and no cerebellar binding in patient6 Of note strong cerebellar binding was observed in pa-tient 2 using CSF which was not evident in patients 3 and 6(figure 5) Positive binding to cerebellar granular neuronswas also detectable in these patients (supplemental figure 3linkslwwcomNXIA135)

Cranial nerve involvement was detectable in the 3 patientswith pan-NF autoantibodies Patients 1 and 2 were almostlocked-in with complete facial and oculomotor palsy anddysphagia Patient 3 had dysphagia requiring a percutaneousendoscopic gastrostomy tube As concomitant glomerulone-phritis has been described in patients with paranodalautoantibodies91824 patientsrsquo records were checked for renalsymptoms There was no evidence of glomerulonephritis in

Figure 3 Epitope mapping of anti-NF autoantibodies by binding assays to truncated NF mutants and flow cytometry

TheNF isoforms consist of a common Ig6 domain and different isoform specific fibronectin domains (A right section) Reactivity of patient sera to NF-155 andNF-186 (A) and truncations of NF (B) by flow cytometry patients 1 2 and 3 show reactivity to both NF-155 (blue line) and NF-186 (orange line) compared withcontrol (empty RSV vector black filled) whereas patients 4 and 5 only react to NF-155 (A) Epitope analysis using NF truncations (B) demonstrated thatreactivity was against the 6Ig domain in 3 of the patients (patients 1ndash3) whowere positive for bothNF isoforms (green line) Patients 1 and 2 also reacted to theFn3 domain (red line) The Fn3Fn4 domain (yellow line) was recognized by the sera of patients 4 and 5 who reacted only to NF-155 Other truncated NFvariants (Fn1Fn2 = dark blue Fn4-Mucin-Fn5 = brown Mucin-Fn5 = purple GFP-only control = gray filled) were not recognized by any of the sera tested (sera1ndash4 and 6 tested for all mutants serum 5 for Fn3Fn4 only) NF = neurofascin

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 7

patients 2ndash6 but patient 1 had acute-on-chronic renal failurewith mild proteinuria and hematuria since the onset of dis-ease so concomitant glomerulonephritis is possible althoughit was not formally diagnosed

DiscussionIn the present study we identified 5 patients with definiteantindashNF-associated neuropathy 3 of them presenting witha distinct clinical phenotype of tetraplegia and cranial nerveinvolvement Our data suggest that the clinical picture is as-sociated with the IgG subclass and epitopemdashIgG4 autoanti-bodies against the Fn3Fn4 domain of NF-155 beingassociated with the well-known phenotype of severe senso-rimotor neuropathy with tremor and IgG3 autoantibodiesagainst the Ig domain of all isoforms leading to tetraplegia andcranial nerve involvement

We describe multiple clinically uniform cases of this moresevere form of antindashpan-NF-associated neuropathy and thedistinct clinical phenotype is strengthened by identification ofa common pattern of reactivity to the 6Ig domain that iscommon to all NF isoforms The notion of antindashpan-NF-associated neuropathy as a distinct subtype of anti-NF neu-ropathy is supported by reports of single cases similar to our 3

patients in the last years In a cohort by Burnor et al 1 patientwith autoantibodies against NF-155 -186 and -140 clinicallyvery much resembled our patients but had IgG4 subclassautoantibodies17 Only recently another patient with tetra-plegia respiratory failure almost locked-in syndrome andIgG3 autoantibodies against NF-186-140 and also NF-155was reported and ultrastructural analysis of nodal architecturerevealed occlusion of the nodal gap by myelin layers19

In a recent study 5 patients with anti-NF-186-140 auto-antibodies were compared with patients with anti-NF-155IgG autoantibodies18 AntindashNF-186-140-positive patientswere found to be clinically different from anti-NF-155patients and 2 antindashNF-186-140-positive patients pre-sented with cranial nerve involvement and respiratory failureone of them with predominance of the IgG3 subclass18 Incontrast to our study patients with pan-NF antibodies of theIgG4 subclass were also described in these studies so possiblythe epitope appears to be themajor determinant of the clinicalpicture and not the IgG subclass However these subclassesmay still contribute to the acuteness or other clinical aspectsof the disease This suggests that the unique clinical picture ofour patients 1ndash3 may be mostly explained by the uniqueepitopemdasha common domain of nodal NF-186-140 andparanodal NF-155mdashbut may also be associated with the IgG3

Figure 4 ELISA of total IgG and IgG subclasses (bars) and C1q deposition in the complement binding assay (line)

IgG3 can be identified as the dominant subclass in patients 1 2 and 6 IgG 2 and 4 in patient 4 and IgG4 in patient 5 Complement deposition can bemeasuredin patients 1 and 2 with IgG3 predominance and in patient 4 with IgG2 and IgG4 autoantibodies No complement deposition was detectable in patient 5 withpredominance of IgG4 and patients 3 and 6 with low titers of total IgG IgG = immunoglobulin G

8 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 NeurologyorgNN

subclass In contrast to IgG4 IgG3 autoantibodies activatecomplement and are considered to have a strong proin-flammatory effect25 Indeed C1q binding induced by antindashpan-NF autoantibodies could be shown in our patients by comple-ment binding assays Remarkably complement deposition wasalso detectable in one of the patients with the typical anti-NF-155 IgG4 phenotype most probably due to additional IgG2autoantibodies that were detectable in this case Complementdeposition may contribute to the more severe acute course ofdisease in this patient compared with the other anti-NF-155 IgGpatient in our cohort Complement-mediated conduction blockmay either be reversible as demonstrated by the case of patient 3with GBS or result in secondary axonal damage or nodal alter-ations (as demonstrated in an ultrastructural case description byVallat et al19) with a prolonged phase of recovery or lack ofrecovery as in patients 1 and 2 Pathophysiologic studies in-cluding passive transfer need to be conducted to further unravelthe connection between clinical phenotype and autoantibodiesin this subgroup of autoimmune neuropathy CNS axonaldamage and deterioration of clinical severity of experimentalallergic neuritis induced by autoantibodies against the commondomain of NF-155 and NF-186 were demonstrated by injectionof pan-NF antibodies to rats indicating a pathogenic effect ofautoantibodies against the common domain of NF826

Patient 6 who was antindashNF-155 positive (IgG3) by ELISA didnot present with the typical clinical phenotype of antindashNF-155-associated neuropathy Lack of binding in the cell-basedassay cerebellar binding studies and of complement

deposition and only weak nodalparanodal binding in teasedfibers may reflect a low affinity of autoantibodies to the in vivoconformation of NF or even a protein associated with NF inthis patient and may explain the mild phenotype

Whereas all previous case reports of patients with pan-NFautoantibodies only described nodal but not paranodalbinding on teased fibers1819 clear paranodal binding wasfound in all our antindashpan-NF patients Stronger paranodalbindingmay be explained by additional binding to Fn3 part ofthe NF-155ndashspecific fibronectin domain in 2 of our patientsRemarkably these 2 patients showed the most severe clinicalphenotype and a very long course of disease Thus multipleepitopes may be a predictor of disease severity as also de-scribed in other autoimmune diseases2728 The underlyingmechanism of multiple epitopes remains unclear intra-molecular epitope spreading could be an explanation and hasalso been described in other autoimmune diseases27 Indeedin patient 2 the Fn3 epitope was only detectable during thecourse of disease but not at disease onset Paranodal bindingin teased fiber assays increased respectively

Tetraplegia was the dominant symptom in all of our patientswith antindashpan-NF autoantibodies and in single cases de-scribed in the literature1729 We did not find any tremor inthese patients but tremor may be masked by tetraplegiaTremor in anti-NF-155 neuropathy has been attributed toautoantibody binding to cerebellar neurons6 and we couldshow cerebellar binding of anti-NF-155 and also antindashpan-NF

Figure 5 Binding assays with sera and CSF on rat cerebellum sections

CSF is depicted as triangular inset on the rightlower half of the panels In contrast to a healthycontrol that only shows some unspecific stainingto Purkinje cells due to tissue fixation optimizedfor detection of surface antigens (A) sera ofpatients 2 and 3 show distinct staining of thegranular and molecular layers (BndashD) Binding ofCSF antibodies to the cerebellum can only beshown in patient 2 and increases during thecourse of disease (B and D) not in the healthycontrol (A) or patient 3 (C) Bar = 100 μm CSF =cerebrospinal fluid CWM = cerebral white matterGCL = granular cell layer ML = molecular layerPCL = purkinje cell layer

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 9

serum Our study is the first to detect paranodal autoantibodiesinCSF Antindashpan-NF autoantibodies were detectable in theCSFof patient 2 as a potential correlate of cerebellar involvement Incomparison to the 5 patients with anti-NF-186-140 autoanti-bodies described by Delmont et al18 that also bound to the Igdomain the course of disease was more severe in our patientsand refractory to treatment Lack of tremor cranial nerve in-volvement and older age compared with anti-NF-155 patientsare features that were also found in the patients by Delmontet al18 Differences may be explained by IgG3 subclass in ourpatients and mostly IgG4 in Delmontrsquos patients

Remarkably 3 of our patients including a patient with onlyanti-NF-155 autoantibodies had severe autonomic in-volvement a feature that has not yet been investigated inantindashNF-associated neuropathy However autonomic symp-toms were also described in the single antindashpan-NF-positivepatient in the study by Burnor et al17 As postganglionic au-tonomic fibers are unmyelinated and we could not show anybinding to sympathetic ganglia this can most probably beexplained by radiculitis involving preganglionic autonomicfibers

The role of IgG subclasses in the pathogenicity of anti-NFautoantibodies and particularly association with therapeuticresponse needs to be further studied Detection of comple-ment binding in our study and by others argues forcomplement-mediated inflammation which extends theconcept of antindashNF-associated neuropathy as a pure IgG4-mediated disease Our data indicate that testing of IgG sub-classes should be performed when detecting autoantibodiesagainst paranodal proteins and testing of anti-NF autoanti-bodies should not be restricted to the NF-155 isoform

AcknowledgmentThe authors thank Barbara Reuter and Heike Rubsamen forexpert technical assistance The study was funded by a grant ofthe Interdisciplinary Center of Clinical Research to KD andCV AV was supported by research fellowships from theEuropean Academy of Neurology The Scientific andTechnological Research Council of Turkey (TUBITAK)and the Alexander von Humboldt Foundation EM receivedsupport from SFBTR128 and the Clinical CompetenceNetwork for Multiple Sclerosis

Study fundingNo targeted funding reported

DisclosureH Stengel A Vural AM Brunder A Heinius L Appelt-shauser B Fiebig F Giese C Dresel A Papagianni F Bir-klein J Weis T Huchtemann C Schmidt P Kortvelyessyand C Villmann report no relevant disclosures E Meinl re-ceived personal fees from Roche Novartis and Genzyme andgrants outside the submitted work from Novartis and Gen-zyme C Sommer received personal fees from Air LiquideAlnylam Astellas Baxalta CSL Behring Genzyme Grifols

Pfizer UCB and Kedrion F Leypoldt reports no relevantdisclosures K Doppler received personal fees from BaxterBaxalta and Grifols and a grant from Kedrion (outside thesubmitted work) Go to NeurologyorgNN for fulldisclosures

Publication historyReceived by Neurology Neuroimmunology amp NeuroinflammationApril 4 2019 Accepted in final form June 25 2019

Appendix Authors

Name Location Contribution

Helena Stengel University HospitalWurzburg

Data acquisition draftingrevising the manuscript andanalysis or interpretation ofdata

Atay Vural MDPhD

Biomedical Centerand UniversityHospital

Data acquisition draftingrevising the manuscript andanalysis or interpretation ofdata

Anna-MichelleBrunder

University HospitalWurzburg

Data acquisition

Annika Heinius UniversitatsklinikumSchleswig-HolsteinCampus Kiel

Data acquisition anddraftingrevising the manuscript

LuiseAppeltshauserMD

University HospitalWurzburg

Analysis or interpretation ofdata and acquisition of data

Bianca Fiebig University HospitalWurzburg

Analysis or interpretation ofdata and acquisition of data

Florian GieseMD

University HospitalHalle

Draftingrevising themanuscript

ChristianDresel MD

University HospitalMainz

Acquisition of data anddraftingrevising themanuscript

AikateriniPapagianni MD

University HospitalWurzburg

Draftingrevising themanuscript and acquisition ofdata

Frank BirkleinMD PhD

University HospitalMainz

Data acquisition anddraftingrevising the manuscript

Joachim WeisMD

RWTH AachenUniversity

Data acquisition draftingrevising the manuscriptanalysis or interpretation ofdata and contribution ofvital reagentstoolspatients

TessaHuchtemannMD

University HospitalMagdeburg

Data acquisition

ChristianSchmidt MD

Otto-von-GuerickeUniversityMagdeburg

Data acquisition

PeterKortvelyessyMD

University HospitalMagdeburg

Data acquisition

CarmenVillmann PhD

University HospitalWurzburg

Draftingrevising themanuscript study conceptor design and contributionof vital reagentstoolspatients

10 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 NeurologyorgNN

References1 Uncini A Susuki K Yuki N Nodo-paranodopathy beyond the demyelinating and

axonal classification in anti-ganglioside antibody-mediated neuropathies Clin Neu-rophysiol 20131241928ndash1934

2 Uncini A A common mechanism and a new categorization for anti-gangliosideantibody-mediated neuropathies Exp Neurol 2012235513ndash516

3 Uncini A Vallat JM Autoimmune nodo-paranodopathies of peripheral nerve theconcept is gaining ground J Neurol Neurosurg Psychiatry 201889627ndash635

4 Querol L Illa I Paranodal and other autoantibodies in chronic inflammatory neu-ropathies Curr Opin Neurol 201528474ndash479

5 Querol L Nogales-Gadea G Rojas-Garcia R et al Antibodies to contactin-1 inchronic inflammatory demyelinating polyneuropathy Ann Neurol 201373370ndash380

6 Querol L Nogales-Gadea G Rojas-Garcia R et al Neurofascin IgG4 antibodies inCIDP associate with disabling tremor and poor response to IVIg Neurology 201482879ndash886

7 Querol L Rojas-Garcia R Diaz-Manera J et al Rituximab in treatment-resistantCIDP with antibodies against paranodal proteins Neurol Neuroimmunol Neuro-inflamm 20152e149 doi101212NXI0000000000000603

8 Ng JK Malotka J Kawakami N et al Neurofascin as a target for autoantibodies inperipheral neuropathies Neurology 2012792241ndash2248

9 Doppler K Appeltshauser L Wilhelmi K et al Destruction of paranodal architecturein inflammatory neuropathy with anti-contactin-1 autoantibodies J Neurol Neuro-surg Psychiatry 201586720ndash728

10 Doppler K Appeltshauser L Villmann C et al Auto-antibodies to contactin-associated protein 1 (Caspr) in two patients with painful inflammatory neuropathyBrain 20161392617ndash2630

11 Appeltshauser L Weishaupt A Sommer C Doppler K Complement depositioninduced by binding of anti-contactin-1 auto-antibodies is modified by immunoglo-bulins Exp Neurol 201628784ndash90

12 Vural A Doppler K Meinl E Autoantibodies against the node of Ranvier in sero-positive chronic inflammatory demyelinating polyneuropathy diagnostic pathogenicand therapeutic relevance Front Immunol 201891029

13 Devaux JJ Miura Y Fukami Y et al Neurofascin-155 IgG4 in chronic inflammatorydemyelinating polyneuropathy Neurology 201686800ndash807

14 Kriebel M Wuchter J Trinks S Volkmer H Neurofascin a switch between neuronalplasticity and stability Int J Biochem Cell Biol 201244694ndash697

15 Zhang A Desmazieres A Zonta B et al Neurofascin 140 is an embryonic neuronalneurofascin isoform that promotes the assembly of the node of Ranvier J Neurosci2015352246ndash2254

16 Volkmer H Hassel B Wolff JM Frank R Rathjen FG Structure of the axonal surfacerecognition molecule neurofascin and its relationship to a neural subgroup of theimmunoglobulin superfamily J Cell Biol 1992118149ndash161

17 Burnor E Yang L Zhou H et al Neurofascin antibodies in autoimmune genetic andidiopathic neuropathies Neurology 201790e31ndash38

18 Delmont E Manso C Querol L et al Autoantibodies to nodal isoforms ofneurofascin in chronic inflammatory demyelinating polyneuropathy Brain 20171401851ndash1858

19 Vallat JM Mathis S Magy L et al Subacute nodopathy with conduction blocks andanti-neurofascin 140186 antibodies an ultrastructural study Brain 2018141e56

20 Van den Bergh P Hadden RD Bouche P et al European Federation of NeurologicalSocietiesPeripheral Nerve Society Guideline on management of chronic in-flammatory demyelinating polyradiculoneuropathy report of a joint task force of theEuropean Federation of Neurological Societies and the Peripheral Nerve SocietymdashFirst Revision J Peripher Nerv Syst 2010151ndash9

21 Sejvar JJ Kohl KS Gidudu J et al Guillain-Barre syndrome and Fisher syndrome casedefinitions and guidelines for collection analysis and presentation of immunizationsafety data Vaccine 201129599ndash612

22 Doppler K Appeltshauser L Kramer HH et al Contactin-1 and neurofascin-155-186 are not targets of auto-antibodies in multifocal motor neuropathy PLoS One201510e0134274

23 Dalmau J Lancaster E Martinez-Hernandez E Rosenfeld MR Balice-Gordon RClinical experience and laboratory investigations in patients with anti-NMDAR en-cephalitis Lancet Neurol 20111063ndash74

24 Hashimoto Y Ogata H Yamasaki R et al Chronic inflammatory demyelinatingpolyneuropathy with concurrent membranous Nephropathy an anti-paranode andpodocyte protein antibody study and literature Survey Front Neurol 20189997

25 Vidarsson G Dekkers G Rispens T IgG subclasses and allotypes from structure toeffector functions Front Immunol 20145520

26 Mathey EK Derfuss T Storch MK et al Neurofascin as a novel target forautoantibody-mediated axonal injury J Exp Med 20072042363ndash2372

27 Sinmaz N Nguyen T Tea F Dale RC Brilot F Mapping autoantigen epitopesmolecular insights into autoantibody-associated disorders of the nervous systemJ Neuroinflammation 201613219

28 Huijbers MG Lipka AF Plomp JJ Niks EH van der Maarel SM Verschuuren JJPathogenic immune mechanisms at the neuromuscular synapse the role of specificantibody-binding epitopes in myasthenia gravis J Intern Med 201427512ndash26

29 Mathis S Magy L Diallo L Boukhris S Vallat JM Amyloid neuropathy mimickingchronic inflammatory demyelinating polyneuropathy Muscle Nerve 20124526ndash31

Appendix (continued)

Name Location Contribution

Edgar MeinlMD

Institut fur KlinischeNeuroimmunologie

Draftingrevising themanuscript analysis orinterpretation of data andcontribution of vital reagentstoolspatients

ClaudiaSommer MDPhD

NeurologischeUniversitatsklinik

Draftingrevising themanuscript study concept ordesign and analysis orinterpretation of data

Frank LeypoldtMD

University HospitalSchleswig-HolsteinCampus Kiel

Data acquisition draftingrevising the manuscriptstudy concept or design andanalysis or interpretation ofdata

KathrinDoppler MD

University HospitalWurzburg

Data acquisition draftingrevising the manuscriptstudy concept or designanalysis or interpretation ofdata statistical analysis andstudy supervision

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 11

DOI 101212NXI000000000000060320196 Neurol Neuroimmunol Neuroinflamm

Helena Stengel Atay Vural Anna-Michelle Brunder et al pan-neurofascin IgG3 as a marker of fulminant autoimmune neuropathyminusAnti

This information is current as of August 16 2019

Academy of Neurology All rights reserved Online ISSN 2332-7812Copyright copy 2019 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the AmericanPublished since April 2014 it is an open-access online-only continuous publication journal Copyright

is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm

ServicesUpdated Information amp

httpnnneurologyorgcontent65e603fullhtmlincluding high resolution figures can be found at

References httpnnneurologyorgcontent65e603fullhtmlref-list-1

This article cites 29 articles 5 of which you can access for free at

Citations httpnnneurologyorgcontent65e603fullhtmlotherarticles

This article has been cited by 8 HighWire-hosted articles

Subspecialty Collections

httpnnneurologyorgcgicollectionguillainbarre_syndromeGuillain-Barre syndrome

nating_polyneuropathyhttpnnneurologyorgcgicollectionchronic_inflammatory_demyeliChronic inflammatory demyelinating polyneuropathy

httpnnneurologyorgcgicollectionautoimmune_diseasesAutoimmune diseases

httpnnneurologyorgcgicollectionall_clinical_neurologyAll Clinical Neurologyfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

Academy of Neurology All rights reserved Online ISSN 2332-7812Copyright copy 2019 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the AmericanPublished since April 2014 it is an open-access online-only continuous publication journal Copyright

is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm

ServicesUpdated Information amp

httpnnneurologyorgcontent65e603fullhtmlincluding high resolution figures can be found at

References httpnnneurologyorgcontent65e603fullhtmlref-list-1

This article cites 29 articles 5 of which you can access for free at

Citations httpnnneurologyorgcontent65e603fullhtmlotherarticles

This article has been cited by 8 HighWire-hosted articles

Subspecialty Collections

httpnnneurologyorgcgicollectionguillainbarre_syndromeGuillain-Barre syndrome

nating_polyneuropathyhttpnnneurologyorgcgicollectionchronic_inflammatory_demyeliChronic inflammatory demyelinating polyneuropathy

httpnnneurologyorgcgicollectionautoimmune_diseasesAutoimmune diseases

httpnnneurologyorgcgicollectionall_clinical_neurologyAll Clinical Neurologyfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

Academy of Neurology All rights reserved Online ISSN 2332-7812Copyright copy 2019 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the AmericanPublished since April 2014 it is an open-access online-only continuous publication journal Copyright

is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm