Autoantibodies against Purkinje cell antigens (PCA) are often linked to cerebellar ataxia. Known Purkinje cell autoantigens are Yo/CDR2/CDR2L, Tr/DNER, ARHGAP26, CAPVIII, Homer3 or ITPR1. Patients with central nervous system (CNS) syndromes, however, often present antibodies against cerebellar Purkinje cell somata or dendrites which do not react with any of the established PCA in monospecific tests. Our team identified diacylglycerol lipase alpha (DAGLA) as a new PCA targeted by sera of thirteen patients (P1-13) with similar immunoglobulin G (IgG) reactivity on cerebellar tissue. The three index patients with available clinical data presented with cerebellitis (P11), epilepsy and hippocampal sclerosis (P12), or brainstem encephalitis (P13).

In indirect immunofluorescence assays (IFA), the patient sera showed similar staining patterns in the molecular layer of rat and monkey cerebellum tissue sections, with reactivities against the dendrites but not the somata of the Purkinje cells. However, we did not find any of these sera reacting positively with 30 different recombinantly expressed established neural autoantigens, including the known PCA. Using immunoprecipitations (IP) with patient sera (P1-5) and lysates of cerebellum the target autoantigen could be enriched and was identified as Sn1-specific DAGLA by mass spectrometry. To verify that we indeed identified the target antigen, DAGLA was recombinantly expressed in HEK293 and IFAs were performed with the fixed cells. All patient sera with a typical anti-DAGLA pattern on cerebellar tissues (P1-13) reacted positive with DAGLA overexpressing HEK cells. Specificity was confirmed with sera from healthy controls (n=51) which did not show reactivities on cerebellar tissue sections, which were all negative in the IFA with HEK-DAGLA cells. We also observed that the cerebrospinal fluid (CSF) of P11 and P13 reacted strongly positive in this assay. In two other experiments we confirmed that anti-DAGLA autoantibodies indeed cause the pattern we observed with the patient sera on cerebellum. Firstly, immunolabeling of cerebellar sections of rat and monkey brain with a rabbit anti-DAGLA antibody revealed the same staining pattern as the patient sera. Secondly, recombinant DAGLA abolished the tissue reactivity of the samples in a neutralization experiment.

DAGLA is a cell membrane protein predominantly expressed on the dendritic surface of Purkinje cells. Therefore, we were interested to know whether the patient sera recognized an extracellular epitope. However, an epitope mapping experiment using six different fragments of the DAGLA protein in an ELISA revealed the main reactivity of patient sera with the c-terminal intracellular fragment.

Aiming to further analyze the clinical association of anti-DAGLA autoantibodies, we screened a panel of about 2200 consecutive sera from patients with various neurological symptoms using IFA with the recombinant protein. Here we identified 24 anti-DAGLA positive sera. Only one of them also showed the corresponding anti-DAGLA staining pattern on cerebellum. The clinical symptoms of anti-DAGLA RC-IFA positive patients identified in this study were diverse and different from our initial findings including neuropathies, multiple sclerosis, Parkinsonism, schizophrenia, stroke or depression.

Conclusively, we suggest that anti-DAGLA autoantibodies may only be relevant for diagnostics if they are detectable in CSF and/or if they can be confirmed by IFA with cerebellar tissue sections – similar to other autoantibodies against PCA. We plan to analyze a panel of consecutive CSF samples in the future to address this question.

REFERENCES

  1. S. Jarius, B. Wildemann: ‘Medusa head ataxia’: the expanding spectrum of Purkinje cell antibodies in autoimmune cerebellar ataxia. Part 1: Anti-mGluR1, anti-Homer-3, anti-Sj/ITPR1 and anti-CARP VIII (2015)
  2. S. Jarius, B. Wildemann: ‘Medusa head ataxia’: the expanding spectrum of Purkinje cell antibodies in autoimmune cerebellar ataxia. Part 2: Anti-PKC-gamma, anti-GluR-delta2, anti-Ca/ARHGAP26 and anti-VGCC (2015)
  3. S. Jarius, B. Wildemann: ‘Medusa head ataxia’: the expanding spectrum of Purkinje cell antibodies in autoimmune cerebellar ataxia. Part 3: Anti-Yo/CDR2, anti-Nb/AP3B2, PCA-2, anti-Tr/DNER, other antibodies, diagnostic pitfalls, summary and outlook (2015)