A new Purkinje cell antibody (anti-Ca) associated with subacute cerebellar ataxia: immunological characterization

Abstract

We report on a newly discovered serum and cerebrospinal fluid (CSF) reactivity to Purkinje cells (PCs) associated with subacute inflammatory cerebellar ataxia. The patient, a previously healthy 33-year-old lady, presented with severe limb and gait ataxia, dysarthria, and diplopia two weeks after she had recovered from a common cold. Immunohistochemical studies on mouse, rat, and monkey brain sections revealed binding of a high-titer (up to 1:10,000) IgG antibody to the cerebellar molecular layer, Purkinje cell (PC) layer, and white matter. The antibody is highly specific for PCs and binds to the cytoplasm as well as to the inner side of the membrane of PC somata, dendrites and axons. It is produced by B cell clones within the CNS, belongs to the IgG1 subclass, and activates complement in vitro. Western blotting of primate cerebellum extract revealed binding of CSF and serum IgG to an 80-97 kDa protein. Extensive control studies were performed to rule out a broad panel of previously described paraneoplastic and non-paraneoplastic antibodies known to be associated with cerebellar ataxia. Screening of >9000 human full length proteins by means of a protein array and additional confirmatory experiments revealed Rho GTPase activating protein 26 (ARHGAP26, GRAF, oligophrenin-1-like protein) as the target antigen. Preadsorption of the patient's serum with human ARHGAP26 but not preadsorption with other proteins resulted in complete loss of PC staining. Our findings suggest a role of autoimmunity against ARHGAP26 in the pathogenesis of subacute inflammatory cerebellar ataxia, and extend the panel of diagnostic markers for this devastating disease.

Figure 1

Magnetic resonance imaging of the cerebellum demonstrating marked cerebellar atrophy over the course of disease but no contrast enhancement. (A) T1-weighted sagittal sequence; image obtained at month 3 after onset. (B) Gadolinium-enhanced sagittal T1-weighted sequence; image obtained at month 17. Note the widening of the cerebellar sulci and the fourth ventricle.

Figure 2

Binding of CSF IgG to the molecular layer (ML), the Purkinje cell layer (PCL) and the white matter (WM) on a mouse cerebellum tissue section. An AlexaFluor^® 488 labeled goat anti-human IgG antibody was used to visualize bound patient IgG. GL = granular layer, P = pia mater.

Figure 3

Double labeling with an antibody to calbindin, a specific marker of Purkinje cells (PCs), proved that the cellular structures targeted by the patient's CSF IgG correspond to PC somata, PC dendrites (upper panel), and PC axons (lower panel). Anti-calbindin reactivity is depicted in red (AlexaFluor^® 568); the patient's antibody in green (AlexaFluor^® 488); and yellow color indicates overlay of the two antibodies. Nuclei are shown in blue (DAPI). Note that the patient's antibody spared the PC nucleus.

Figure 4

The patient's antibody bound selectively to Purkinje cells but not to astrocytes such as the Bergman glial cells (BGC) as demonstrated by double staining with an antibody to anti-glial fibrillary acidic protein (GFAP). The anti-GFAP antibody, staining astrocytes in the granular layer (GL) as well as the processes of the BGCs in the molecular layer (ML) and along the pia mater (P), is depicted in red (AF568); and the patient's antibody is labeled in green (AF488); yellow color would indicate overlay of the two antibodies, but is absent. Nuclei are shown in blue.

Figure 5

Double staining of the cerebellar cortex with an antibody to parvalbumin (green; AF488), a general marker of cerebellar neurons, demonstrates that the patient's antibody (red; AF568) binds neither to interneurons (arrow heads) in the molecular layer (ML) nor to granular cells in the granular layer.

Figure 6

Subclass analysis revealed that the antibody belongs mainly to the IgG1 subclass (panel A). In addition, antibodies of the IgA subclass were detectable (panel B). In line with the presence of IgG1 antibodies, fixing of complement C3b was found (panel C; the two upper images show negative control experiments with an isotype control and without complement, respectively). Only very weak IgG3 reactivity and no IgG2, IgG4, or IgM antibodies were detectable in serum or CSF (not shown).

Figure 7

A commercial western blot of primate cerebellum extract revealed binding of CSF IgG to a 80-97 kDa band, which does not correspond to bands found with antibodies to inositol-triphosphate receptor type I (IP3RI), protein kinase C gamma (PKCγ), glutamate receptor delta 2 (GluRδ2), metabotropic glutamate receptor 1α (mGluR1α), or Homer3.

Figure 8

High magnification revealed binding of IgG from the patient's CSF to membranes of PC dendrites and axons. Panel A shows transverse sections and panel B depicts longitudinal sections of distal PC dendrites (mouse tissue). Panel C shows three longitudinal sections of proximal PC dendrites (monkey tissue). Panel D consists of a composite of transverse and longitudinal sections of monkey PC axons. A goat anti-human IgG antibody labeled with AF488 was used to visualize bound IgG.

Figure 9

Double staining of cultured Purkinje cells (PC) with the patient's serum revealed binding of IgG to PC somata and to membranes of PC dendrites and axons. Calbindin is depicted in green (A; AF488); the patient's antibody in red (B; AF555); yellow color indicates overlay of the two antibodies (C). Binding was only observed after fixation with 4% paraformaldehyde and 0.05% triton-X (A-C), but not on living cells (D-E), suggesting an intracellular localization of the target antigen. Panel D shows fluorescence of a GFP-transfected living (i.e. unfixed) PC; Panel E demonstrates lack of binding of the patient' serum to the same cell.

Figure 10

Double staining of cerebellum tissue sections with a commercial antibody to p80-coilin identifies the dotted nuclear staining seen with the broad majority of cerebellar neurons following incubation with the patient's serum (but not CSF) as anti-coilin immunoreactivity. The patient's antibody is depicted in green (A; AF488); the commercial coilin antibody in red (B; AF568); yellow color indicates overlay of the two antibodies (C). The p80-coilin antibody recognized a ~80 kDa protein band (lane 1) in a western blot of primate cerebellum tissue, which was not identical to the 80-97 kDa band found with the patient's serum and CSF (lane 2).

Figure 11

No evidence for antibodies to Homer3 (A) or the metabotropic glutamate receptor 1 alpha (mGluR1α) (B) in the patient's CSF as demonstrated by double staining with commercial antibodies to these antigens. While the patient's CSF preferentially stained Purkinje cell dendrites on mouse (Panel A and B) and monkey (Panel A, inset) tissue, Homer3 and mGluR1a immunoreactivity was widely restricted to dendritic spines. IgG antibodies to Homer3 and mGluR1α are depicted in red (AF568); the patient's IgG is labeled in green; yellow color indicates overlay of the two antibodies. Nuclei are shown in blue (DAPI).

Figure 12

No evidence for antibodies to the glutamate receptor delta 2 (GluRδ2) (A) or the protein kinase C type gamma (PKCγ) (B) in the patient's CSF as demonstrated by double staining with commercial antibodies to these antigens. Note that the anti-GluRδ2 antibody as well as the anti-PKCγ antibody stained interneurons on mouse (A and B; arrows) and primate (B, left upper inset; arrows) cerebellar tissue, which were spared by the patient's antibody (cf. Figure 5). The anti-PKCγ antibody also bound to dendritic spines on monkey tissue that were not stained by the patient's antibody (B, right upper inset). In contrast to the patient's CSF, anti-PKCγ stained capillaries in primate brain (asterisk). Moreover, the cytoplasmic staining found with the anti-PKCγ antibody exceeded that caused by the patient's antibody and appeared to include the nuclear membrane (B, lower inset; left and right images represent different exposure times). IgG antibodies to GluRδ2 and PKCγ are depicted in red (AF568); the patient's IgG is labeled in green; yellow color indicates overlay of the two antibodies. Nuclei are shown in blue (DAPI).

Figure 13

Double labeling of a primate cerebellum tissue section with a commercial antibody to inositol-triphosphate receptor I (IP3RI), depicted in red (AF568), and the patient's CSF, shown in green (AF488). Nuclei are shown in blue (DAPI). An almost perfect overlay (yellow) of the two antibodies was found in the molecular layer and the Purkinje cell (PC) layer (Panels A-C), indicating that the patient's antibody targets an antigen expressed in close spatial proximity of IP3RI. Higher magnification (40×; D) suggests that the two antibodies might also bind to dendritic spines of PCs, though not all spines were stained by both antibodies. A fair, though less impressive, overlay was found in the white matter, where the two antibodies stained the same axons (E-G).

Figure 14

Probing of a commercial protein microarray revealed strong binding of the patient's sera to human ARHGAP26 (Panel A). In accordance with this finding, binding of IgG from the patient's serum (Lane P), but not from three healthy controls (Lane C1-C3), to a recombinant human full length ARHGAP26 protein was found (Panel B). Western blotting confirmed the presence of ARHGAP26 in primate cerebellar extract (Panel C, lane 2), and incubation with the patient's CSF resulted in a band running at the same height as the ARHGAP26 band (Panel C, lane 1); the significance of the additional band recognized by the commercial antibody is unknown. The commercial ARHGAP26 antibody bound to the Purkinje cell layer and the molecular layer (Panel D) and showed a good overlay with the patient's IgG depicted in yellow (Panel E). Finally, preadsorption of the patient's CSF with the ARHGAP26 protein (Panel H), but not preadsorption with a control protein (Panel G), resulted in complete loss of binding to cerebellar tissue sections in an indirect immunofluorescence assay; panel F shows binding of IgG from a non-preadsorbed aliquot of the same CSF sample (exposure time was 2 sec in all cases to detect also low fluorescence signals).