Isoaspartate formation and irreversible aggregation of collapsin response mediator protein 2: implications for the etiology of epilepsy and age-related cognitive decline

Abstract

Collapsin response mediator protein 2 (CRMP2) functions in the genesis and activity of neuronal connections in mammalian brain. We previously reported that a protein coincident with CRMP2 on 2D-gels undergoes marked accumulation of abnormal L-isoaspartyl sites in brain extracts of mice missing the repair enzyme, protein L-isoaspartyl methyltransferase (PIMT). To conflrm and explore the signiflcance of isoaspartyl damage in CRMP2, we expressed and purifled recombinant mouse CRMP2 (rCRMP2). A polyclonal antibody made against the recombinant protein precipitated CRMP2 from brain extracts of PIMT-KO mice, but not from WT mice, suggesting that (1) the rCRMP2 antigen underwent signiflcant isoAsp formation in the process of antibody production and (2) the isoAsp form of CRMP2 is considerably more immunogenic than the native protein. In vitro aging of rCRMP2 at pH 7.4, 37°C for 0-28 days led to robust accumulation of isoAsp sites that were repairable by PIMT, and also induced a progressive accumulation of apparent dimers and higher-mass oligomers as judged by SDS-PAGE. A similar pattern of CRMP2 aggregation was observed in mice, with levels increasing throughout the lifespan. We conclude that CRMP2 is indeed a major target of PIMT-mediated protein repair in the brain; that isoAsp forms of CRMP2 are highly immunogeni; and that CRMP2 dysfunction makes a signiflcant contribution to neuropathology in the PIMT-KO mouse.

Keywords: aging; autoimmunity – epilepsy; protein aggregation; protein damage; protein repair.

Figure 1

Mechanism of isoaspartate formation and PIMT-catalyzed repair. Under physiological conditions, deamidation of asparagine residues or dehydration of aspartic acid residues results in the formation of a metastable intermediate succinimide which spontaneously hydrolyzes to form a mixture of normal L-aspartyl and atypical L-isoaspartyl linkages. PIMT, using AdoMet as a methyl donor, selectively methylates the isoaspartyl α-carboxyl group to form a highly labile methyl ester. Spontaneous demethylation occurs within minutes to reform the original succinimide, with release of methanol as a by-product. This succinimide is now the starting point for further cycles of repair, resulting in near complete conversion of the isoaspartyl β-linkages to normal aspartyl α-linkages. Dashed lines indicate degradation steps, while solid lines indicate repair steps.

Figure 2

Expression and puriflcation of recombinant murine CRMP2 (rCRMP2). (A) Diagram of the 6xHis-CRMP2 construct showing the full length (L), short (S) and thrombin-trimmed (Th) forms. Key residue numbers for the native protein are shown along the bottom edge. (B) SDS-PAGE of samples before and after puriflcation; E. coli extract (Lane 1), purifled rCRMP2 (Lane 2), and purifled rCRMP2 after treatment with thrombin.

Figure 3

A polyclonal antibody made against rCRMP2 selectively precipitates CRMP2 from brain extracts of PIMT KO mice. (A) Equal amounts of CRMP2 are seen in a Western blot of brain extracts from WT and KO mice. (B) Equal amounts of CRMP2 are also seen in blots of the unbound material after immunoprecipitation. (C) Western blot of brain extracts immunoprecipitated with anti-rCRMP2 for 2 hours at 4 °C shows a marked selectivity of the antibody for the extracts from PIMT KO mice. (D) Western blot of proteins immunoprecipitated for 2 hours at room temperature, shows a marked selectivity of the antibody with the detection of apparent degradation and aggregation products. Preparation of the antibodies used here is described in Materials and methods.

Figure 4

Formation of isoaspartates and loss of soluble rCRMP2 during in vitro aging. rCRMP2 was aged at pH 7.4, 37 °C for up to 28 days. The aged samples were centrifuged at 20,000 x g for 20 min. The collected supernatants were used to determine protein concentration and the determination of the isoaspartate content is described in Materials and methods.

Figure 5

SDS-PAGE of rCRMP2 reveals progressive oligomer formation coincident with isoaspartate formation during in vitro aging. All lanes contained 3.3 μg proteins from the supernatants (as described in Fig. 4) of the aged rCRMP2 samples. (A) Coomassie blue staining of aged rCRMP2 containing the S and L forms. (B) Autoradiogram after SDS-PAGE of aged rCRMP2 after ³H-methylation with PIMT prior to electrophoresis to label the isoAsp sites. (C) Same as B, except the aged samples subjected to the thrombin treatment. (D) SDS-PAGE of unaged and aged (18 days) rCRMP2 using a different set of mass standards to more accurately estimate the molecular weights of the high mass bands seen in the aged samples. Mass numbers directly adjacent to the right of the gel were calculated from a 2nd-order polynomial flt (R²= 0.998) to a plot of log mol. wt. vs reference MWs for the flve standards shown. Numbers in parenthesis indicate the expected mass for oligomers based on the average mass (60.5 kDa) estimated for the monomeric S and L forms of CRMP2.

Figure 6

Demonstration that PIMT reverses isoaspartyl damage to rCRMP2 incurred by in vitro aging. Samples of control (0 day) and aged (14 day) rCRMP2 were incubated for 7 hours in a repair buffer (pH 7.8) with (+) or without (−) the addition of PIMT and unlabeled AdoMet. After the extensive dialysis to remove the AdoMet, equal amounts of fresh PIMT were added to all samples, along with [³H-methyl]AdoMet, and incubated in an assessment buffer (pH 6.2, needed to stabilize the ³H-methyl esters) to label any remaining isoAsp sites. Samples were then subjected to SDS-PAGE, followed by Coomassie staining (left panel) and fluorography (right panel). A comparison of lanes 7 and 8 indicates the presence of PIMT and AdoMet in the repair buffer resulted in nearly complete elimination of isoAsp in CRMP2 and its oligomers. Note that lanes 2 and 4 have 2X the amount of PIMT than in lanes 1 and 3, due to the addition of PIMT during the repair reaction. A more detailed description of the repair procedure can be found in Carter and Aswad (2008).

Figure 7

Comparison of CRMP2 oligomer formation in the mouse brain extracts. (A) Western blot for CRMP2 in brain extracts from PIMT +/+ and −/− mice, aged 4 weeks or 2 years. CRMP2 dimers are seen in the 2 years mice, and the amount of dimer is unaffected by loss of PIMT. (B) A comparison of CRMP2 oligomer patterns in PIMT +/+ vs +/− male mice at 8 months and 2 years. (C) A plot of oligomer index values [band density values of (dimer + trimer)/(monomer + dimer +trimer)] obtained from blots as shown in panel B.

Figure 8

Sequence alignment of human and mouse CRMP2s, highlighted to show potential sites of isoAsp formation in relation to its secondary structure. Canonical hot spots for isoAsp formation are bold-faced in red. Regions of b-strands (green), a-helices (cyan), and turns (magenta), as seen in the Structural Features Viewer in UniProtKB for the mouse protein.