Data on isoaspartylation of neuronal ELAVL proteins

This article contains experimental data examining the propensity of neuronal ELAVL proteins to become isoaspartylated. The data are related to the article “Isoaspartylation appears to trigger small cell lung cancer-associated autoimmunity against neuronal protein ELAVL4” (M.A. Pulido, M.K. DerHartunian, Z. Qin, E.M. Chung, D.S. Kang, A.W. Woodham, J.A. Tsou, R. Klooster, O. Akbari, L. Wang, W.M. Kast, S.V. Liu, J.J.G.M. Verschuuren, D.W. Aswad, I.A. Laird-Offringa, 2016) [1], in which it was reported that the N-terminal region of recombinant human ELAVL4 protein, incubated under physiological conditions, acquires a type of highly immunogenic protein damage. Here, we present Western blot analysis data generated by using an affinity-purified polyclonal rabbit antibody (raised against an N-terminal ELAVL4 isoaspartyl-converted peptide) to probe recombinant protein fragments of the other three members of the ELAVL family: the highly homologous neuronal ELAVL2 (HuB) and ELAVL3 (HuC), and the much less homologous ubiquitously expressed ELAVL1 (HuR).


Subject area
Biology More specific subject area

Cancer immunology
Type of data Raw How data was acquired Western blot obtained on Biorad Fluor-S TM MultiImager.

Data format
Raw images cropped to fit into the figure were used.

Experimental factors
Recombinant N-terminal ELAVL proteins incubated under isoaspartylationinducing conditions for 0, 1, 3 or 7 days Experimental features Amino acid sequence alignment and Western blot analysis

Data source location
Los Angeles, CA, USA

Data accessibility
Data is presented in this article

Value of the data
The data encourages the investigation of isoaspartylation of other neuronal autoantigens. The data encourages the investigation of the oligomerization of autoantigens that might be prone to isoaspartylation and allows comparison to other antigens.
The data supports the generation of additional anti-isoaspartyl antisera by investigators, including antisera to other neuronal SCLC auto-antigens.

Data
The data shown consists of two parts. First, we provide the alignment of the N-terminal amino acid sequences of ELAVL proteins, indicating regions of homology between the proteins and a peptide used to generate a rabbit anti-isoaspartyl-ELAVL4 serum [1]. Secondly, we show Western blots documenting the reactivity of the affinity-purified rabbit antiserum with recombinant human ELAVL2, ELAVL3, ELAVL4, and ELAVL1 polypeptides (consisting of the N-terminal parts of the proteins, including the first RNA recognition motif).
(A) Amino acid sequence alignments of neuronal ELAVL proteins (ELAVL2-4) and the ubiquitouslyexpressed ELAVL1 protein indicate any regions of homology with the peptide used to generate the rabbit anti-isoaspartyl-ELAVL4 antiserum. The peptide sequence is underlined, amino acids identical with the peptide sequence are highlighted in yellow and conserved substitutions are highlighted in gray, with N for D substitutions noted in green. Canonical isoaspartylation sites (N or D followed by S, G or H) are bolded. The N-terminal portion of the RNA recognition motif (RRM) domain is marked by the black box. ELAVL1 was included as a negative control because it has little homology to the peptide used to generate the anti-isoaspartyl-ELAVL4 serum. (B) Western blot analysis of N-terminal regions of ELAVL proteins incubated under isoaspartyl-inducing conditions for 0, 1, 3 or 7 days (top panel). We have previously shown that the ELAVL4 aa 1-117 fragment, similar to the N-terminal fragment used to broadly characterize the ELAVL immune epitopes in numerous published studies (reviewed in [2]), becomes highly reactive with the affinity-purified anti-isoaspartyl-ELAVL4 rabbit antisera, and that both the monomer and oligomers that form during isoaspartyl conversion show reactivity to the sera [1]. Oligomerization is a known consequence of isoaspartylation [3][4][5][6]. Just like ELAVL4, ELAVL2 and 3 acquire cross-reactivity with the anti-isoAsp-ELAVL4 antiserum during the isoaspartyl conversion (Fig. 1B, top panel). Reactivity with ELAVL2 closely resembles that of ELAVL4, while ELAVL3 only shows cross-reactivity in the dimer form. This may be because the homology between the N-terminal regions of ELAVL4 and ELAVL2 is higher than that of ELAVL3. The dimer proteins may exhibit alternative epitopes or show altered binding kinetics favoring antibody binding to the dimer, which may explain why reactivity with the ELAVL3 dimer is seen while reactivity with the monomer is not observed. As the Coomassie gel shows (Fig. 1B, bottom panel), the oligomers (which are resistant to denaturation) are present in modest amounts but appear to be unusually reactive. This figure is related to Figure 2 in Pulido et al. [1].