Changes in properties of serine 129 phosphorylated α-synuclein with progression of Lewy-type histopathology in human brains

Abstract

Modifications of α-synuclein resulting in changes in its conformation are considered to be key pathological events for Lewy body diseases (LBD), which include Parkinson's disease (PD) and dementia with Lewy bodies (DLB). We have previously described a histopathological Unified Staging System for LBD that classifies the spread of α-synuclein phosphorylated at serine 129 (pS129-α-synuclein) from olfactory bulb to brainstem or limbic regions, and finally neocortex. Lewy bodies and Lewy neurites are highly enriched in pS129-α-synuclein. Increased formation of pS129-α-synuclein changes its solubility properties enhancing its tendency to aggregate and disrupt normal function. As in vitro and animal studies have shown that inhibiting formation of pS129-α-synuclein can prevent toxic consequences, this has become one of the therapeutic targets for LBD. However, detailed biochemical descriptions of the changes in pS129-α-synuclein properties in diseased human brains are needed to further our understanding of how these might contribute to molecular pathogenesis. In this study, we used 130 separate brain samples from cingulate cortex (limbic cortex) and 131 from temporal cortex (neocortex) that had been staged according to our Unified Staging System to examine progressive changes in properties of pS129-α-synuclein with the formation of progressively more severe histological Lewy-type pathology. The brain samples from these staged cases had been separated into cytosol-enriched, membrane-enriched (detergent soluble) and insoluble (ureas/SDS soluble) fractions. We also characterized the nature and appearance of higher molecular weight forms of pS129-α-synuclein. The major species was the 16 kD monomeric form; this accumulated with increasing stage with a large increase in Stage IV samples. By comparing two brain regions, we showed higher accumulation of insoluble pS129-α-synuclein in cingulate cortex, where histological deposits occur first, than in temporal cortex in samples with advanced (stage IV) LB pathology.

Figure 1

Figure 1A: Fractionation scheme for brain tissue to identify changes in solubility of pS129- α-synuclein. A three stage extraction scheme was used for processing cingulate cortex (127 staged cases) and temporal cortex (134 staged cases). Fractions were considered as soluble (cytosol-enriched), detergent soluble (membrane enriched fraction) and insoluble (dissolved in 6M Urea/2%SDS with heat). Figure 1B: Assessment of purity of soluble and membrane fractions. To determine the level of contamination of the soluble (cytosol-enriched) (S) fractions with membrane (M) components, paired samples from selected fractionated temporal cortex Stage 0 and Stage IV cases were analyzed for pS129-α–synuclein (top panel), CD200 as a marker for membrane proteins (middle panel), and β-actin (lower panel). It can be seen that CD200 levels in soluble fractions were slight.

Figure 1

Figure 1A: Fractionation scheme for brain tissue to identify changes in solubility of pS129- α-synuclein. A three stage extraction scheme was used for processing cingulate cortex (127 staged cases) and temporal cortex (134 staged cases). Fractions were considered as soluble (cytosol-enriched), detergent soluble (membrane enriched fraction) and insoluble (dissolved in 6M Urea/2%SDS with heat). Figure 1B: Assessment of purity of soluble and membrane fractions. To determine the level of contamination of the soluble (cytosol-enriched) (S) fractions with membrane (M) components, paired samples from selected fractionated temporal cortex Stage 0 and Stage IV cases were analyzed for pS129-α–synuclein (top panel), CD200 as a marker for membrane proteins (middle panel), and β-actin (lower panel). It can be seen that CD200 levels in soluble fractions were slight.

Figure 2. Characterization of α-synuclein species in Staged temporal cortex samples showing phosphorylation dependent specificity of pS129-α–synuclein antibodies

A) Progression of increased pS129-α–synuclein immunoreactivity with increased LB staging. Randomly selected samples (temporal cortex RIPA extracted fraction) from each LB stage were analyzed using the pS129 α-synuclein antibodies (pS129-α–syn (Tokyo) and pS129-α–syn (Epitomics) and an antibody to total α-synuclein (syn-1). Panels show a large increase in intensity in the Stage IV sample with the pS129-α–syn specific antibodies with total α-synuclein being detected in all samples. Blots were reprobed with antibody to β-actin to show equivalent loadings. B) Loss of immunoreactivity with pS129-α– syn (Tokyo) following treatment with alkaline phosphatase. Samples were temporal cortex membrane-enriched fraction from four different Stage IV cases. One membrane (+ Alk phosph) was treated with alkaline phosphatase while the other was incubated with buffer (− Alk phosph). Dephosphorylation of separated proteins resulted in loss of immunoreactive bands. There was equivalent reactivity on alkaline phosphatase-treated and untreated membranes with antibody to β-actin. C) In vitro phosphorylation of recombinant α-synuclein with casein kinase 2 (CK2). Dilutions of CK2 treated α-synuclein (P α-syn), or control treated α-synuclein (NonP α-syn) probed with pS129-α–synuclein antibodies pS129-α-syn (Tokyo) and pS129 α-syn (Epitomics) and with antibody syn-1 which recognizes unmodified forms of α-synuclein.

Figure 2. Characterization of α-synuclein species in Staged temporal cortex samples showing phosphorylation dependent specificity of pS129-α–synuclein antibodies

A) Progression of increased pS129-α–synuclein immunoreactivity with increased LB staging. Randomly selected samples (temporal cortex RIPA extracted fraction) from each LB stage were analyzed using the pS129 α-synuclein antibodies (pS129-α–syn (Tokyo) and pS129-α–syn (Epitomics) and an antibody to total α-synuclein (syn-1). Panels show a large increase in intensity in the Stage IV sample with the pS129-α–syn specific antibodies with total α-synuclein being detected in all samples. Blots were reprobed with antibody to β-actin to show equivalent loadings. B) Loss of immunoreactivity with pS129-α– syn (Tokyo) following treatment with alkaline phosphatase. Samples were temporal cortex membrane-enriched fraction from four different Stage IV cases. One membrane (+ Alk phosph) was treated with alkaline phosphatase while the other was incubated with buffer (− Alk phosph). Dephosphorylation of separated proteins resulted in loss of immunoreactive bands. There was equivalent reactivity on alkaline phosphatase-treated and untreated membranes with antibody to β-actin. C) In vitro phosphorylation of recombinant α-synuclein with casein kinase 2 (CK2). Dilutions of CK2 treated α-synuclein (P α-syn), or control treated α-synuclein (NonP α-syn) probed with pS129-α–synuclein antibodies pS129-α-syn (Tokyo) and pS129 α-syn (Epitomics) and with antibody syn-1 which recognizes unmodified forms of α-synuclein.

Figure 2. Characterization of α-synuclein species in Staged temporal cortex samples showing phosphorylation dependent specificity of pS129-α–synuclein antibodies

A) Progression of increased pS129-α–synuclein immunoreactivity with increased LB staging. Randomly selected samples (temporal cortex RIPA extracted fraction) from each LB stage were analyzed using the pS129 α-synuclein antibodies (pS129-α–syn (Tokyo) and pS129-α–syn (Epitomics) and an antibody to total α-synuclein (syn-1). Panels show a large increase in intensity in the Stage IV sample with the pS129-α–syn specific antibodies with total α-synuclein being detected in all samples. Blots were reprobed with antibody to β-actin to show equivalent loadings. B) Loss of immunoreactivity with pS129-α– syn (Tokyo) following treatment with alkaline phosphatase. Samples were temporal cortex membrane-enriched fraction from four different Stage IV cases. One membrane (+ Alk phosph) was treated with alkaline phosphatase while the other was incubated with buffer (− Alk phosph). Dephosphorylation of separated proteins resulted in loss of immunoreactive bands. There was equivalent reactivity on alkaline phosphatase-treated and untreated membranes with antibody to β-actin. C) In vitro phosphorylation of recombinant α-synuclein with casein kinase 2 (CK2). Dilutions of CK2 treated α-synuclein (P α-syn), or control treated α-synuclein (NonP α-syn) probed with pS129-α–synuclein antibodies pS129-α-syn (Tokyo) and pS129 α-syn (Epitomics) and with antibody syn-1 which recognizes unmodified forms of α-synuclein.

Figure 3. Further characterization α-synuclein species in Staged temporal cortex samples using immunoprecipitation/western blot procedures

A and B) Identification of α-synuclein reactive bands immunoprecipitated with phospho-specific antibody pS129-α-syn (Tokyo) and detected using LB509 antibody to α-synuclein. Selected samples (n=4) from each stage were analyzed. In Stage IV samples (DLB cases), there were 5 distinct α-synuclein species and also higher molecular weight (mw > 50 –250 kD) aggregated α-synuclein. Panel B represents a longer exposure of panel A and also shows trace amounts of pS129-α-synuclein precipitated from certain Stage 0 –Stage III samples. C) Identification of ubiquitin-reactive bands immunoprecipitated with phosphorylation specific antibody pS129-α-syn (Tokyo) and detected using an antibody to ubiquitin. Three of the high molecular weight p129-α-synuclein bands (approximate molecular weight 36 kD, 42 kD and 48–50 kD) were immunoreactive for ubiquitin. Monomeric pS129-α-synuclein (18 kD) and 26 kD bands (panel A) were not reactive for ubiquitin. D) Identification of α-synuclein reactive bands immunoprecipitated with LB509, a non-phosphorylation dependent antibody to α-synuclein and detected using phosphorylation specific antibody pS129-α-syn (Tokyo). The pattern of Stage IV bands was essentially the same as the reverse analyses (panel A). E–H) Defining soluble oligomer forms of α-synuclein. Samples were analyzed for pS129-α-synuclein. E) non-phosphorylation dependent α-synuclein using LB509 (F); and non-phosphorylation dependent α-synuclein using syn-1 (G). Equivalent loading was demonstrated with β-actin (H). A major soluble oligomeric species (O) was detected using LB509 all of these staged samples.

Figure 3. Further characterization α-synuclein species in Staged temporal cortex samples using immunoprecipitation/western blot procedures

A and B) Identification of α-synuclein reactive bands immunoprecipitated with phospho-specific antibody pS129-α-syn (Tokyo) and detected using LB509 antibody to α-synuclein. Selected samples (n=4) from each stage were analyzed. In Stage IV samples (DLB cases), there were 5 distinct α-synuclein species and also higher molecular weight (mw > 50 –250 kD) aggregated α-synuclein. Panel B represents a longer exposure of panel A and also shows trace amounts of pS129-α-synuclein precipitated from certain Stage 0 –Stage III samples. C) Identification of ubiquitin-reactive bands immunoprecipitated with phosphorylation specific antibody pS129-α-syn (Tokyo) and detected using an antibody to ubiquitin. Three of the high molecular weight p129-α-synuclein bands (approximate molecular weight 36 kD, 42 kD and 48–50 kD) were immunoreactive for ubiquitin. Monomeric pS129-α-synuclein (18 kD) and 26 kD bands (panel A) were not reactive for ubiquitin. D) Identification of α-synuclein reactive bands immunoprecipitated with LB509, a non-phosphorylation dependent antibody to α-synuclein and detected using phosphorylation specific antibody pS129-α-syn (Tokyo). The pattern of Stage IV bands was essentially the same as the reverse analyses (panel A). E–H) Defining soluble oligomer forms of α-synuclein. Samples were analyzed for pS129-α-synuclein. E) non-phosphorylation dependent α-synuclein using LB509 (F); and non-phosphorylation dependent α-synuclein using syn-1 (G). Equivalent loading was demonstrated with β-actin (H). A major soluble oligomeric species (O) was detected using LB509 all of these staged samples.

Figure 4. Comparison of relative levels of pS129-α– synuclein with increased LB staging in soluble and membrane fractions of cingulate and temporal cortex

A) Representative western blots of staged temporal cortex samples separated into SOLUBLE and MEMBRANE samples showing different species of α-synuclein. Figure represents a composite from individual western blots to show a representation of bands across all blots analyzed. Membranes probed with pS129-α-syn (Tokyo) showed monomeric (M) α-synuclein in both fraction, a truncated (T) species in the soluble fraction and an oligomer form enriched in the soluble fraction (O). Higher molecular weight (HMW) pS129-α-synuclein bands (6 separate bands with molecular weights 24–50 kD) were found primarily in Stage IV samples. An abundant dimeric form of unphosphorylated α-synuclein (D) was present in all fractions. The band identified as (O) and that identified with (D) were not the same species and differed slightly in molecular weight. Figure shows LB staging for the selected samples. Similar patterns of bands in soluble and membrane fractions of cingulate samples were observed (not shown). B) and C) Changes in pS129-α-synuclein levels with increased staging; comparison between cingulate and temporal cortex samples. Large increases in levels of pS129-α–synuclein were apparent in both soluble (A) and membrane (B) fractions of Stage IV samples compared to Stage 0, but with progressive increase between Stage 0 and Stage III. Results represent normalized mean values ± standard error of mean (S.E.M.). D and E) Changes in higher molecular weight pS129-α-synuclein in soluble and membrane fractions from cingulate and temporal cortex samples with increased staging. Significant increases in all stage IV samples compared to Stage 0-III. Results represent normalized mean values ± standard error of mean (S.E.M.).

Figure 4. Comparison of relative levels of pS129-α– synuclein with increased LB staging in soluble and membrane fractions of cingulate and temporal cortex

A) Representative western blots of staged temporal cortex samples separated into SOLUBLE and MEMBRANE samples showing different species of α-synuclein. Figure represents a composite from individual western blots to show a representation of bands across all blots analyzed. Membranes probed with pS129-α-syn (Tokyo) showed monomeric (M) α-synuclein in both fraction, a truncated (T) species in the soluble fraction and an oligomer form enriched in the soluble fraction (O). Higher molecular weight (HMW) pS129-α-synuclein bands (6 separate bands with molecular weights 24–50 kD) were found primarily in Stage IV samples. An abundant dimeric form of unphosphorylated α-synuclein (D) was present in all fractions. The band identified as (O) and that identified with (D) were not the same species and differed slightly in molecular weight. Figure shows LB staging for the selected samples. Similar patterns of bands in soluble and membrane fractions of cingulate samples were observed (not shown). B) and C) Changes in pS129-α-synuclein levels with increased staging; comparison between cingulate and temporal cortex samples. Large increases in levels of pS129-α–synuclein were apparent in both soluble (A) and membrane (B) fractions of Stage IV samples compared to Stage 0, but with progressive increase between Stage 0 and Stage III. Results represent normalized mean values ± standard error of mean (S.E.M.). D and E) Changes in higher molecular weight pS129-α-synuclein in soluble and membrane fractions from cingulate and temporal cortex samples with increased staging. Significant increases in all stage IV samples compared to Stage 0-III. Results represent normalized mean values ± standard error of mean (S.E.M.).

Figure 4. Comparison of relative levels of pS129-α– synuclein with increased LB staging in soluble and membrane fractions of cingulate and temporal cortex

A) Representative western blots of staged temporal cortex samples separated into SOLUBLE and MEMBRANE samples showing different species of α-synuclein. Figure represents a composite from individual western blots to show a representation of bands across all blots analyzed. Membranes probed with pS129-α-syn (Tokyo) showed monomeric (M) α-synuclein in both fraction, a truncated (T) species in the soluble fraction and an oligomer form enriched in the soluble fraction (O). Higher molecular weight (HMW) pS129-α-synuclein bands (6 separate bands with molecular weights 24–50 kD) were found primarily in Stage IV samples. An abundant dimeric form of unphosphorylated α-synuclein (D) was present in all fractions. The band identified as (O) and that identified with (D) were not the same species and differed slightly in molecular weight. Figure shows LB staging for the selected samples. Similar patterns of bands in soluble and membrane fractions of cingulate samples were observed (not shown). B) and C) Changes in pS129-α-synuclein levels with increased staging; comparison between cingulate and temporal cortex samples. Large increases in levels of pS129-α–synuclein were apparent in both soluble (A) and membrane (B) fractions of Stage IV samples compared to Stage 0, but with progressive increase between Stage 0 and Stage III. Results represent normalized mean values ± standard error of mean (S.E.M.). D and E) Changes in higher molecular weight pS129-α-synuclein in soluble and membrane fractions from cingulate and temporal cortex samples with increased staging. Significant increases in all stage IV samples compared to Stage 0-III. Results represent normalized mean values ± standard error of mean (S.E.M.).

Figure 5. Comparison of relative levels of insoluble pS129-α-synuclein species in staged LB cingulate and temporal cortex samples

A and B) Significantly higher levels of insoluble monomeric (A) and higher molecular weight (B) pS129-α-synuclein in cingulate cortex samples compared to temporal cortex samples. Significantly higher levels of pS129-α-synuclein in cingulate stage IV samples (***P<0.001) and stage III samples compared to temporal cortex samples (** P<0.01) for monomeric (B) and higher molecular weight samples (C). Results represent normalized mean values ± standard error of mean (S.E.M.). C) Representative western blots showing patterns of pS129-α-synuclein immunoreactive bands for a selection of staged cingulate cortex insoluble fraction samples. Major bands being detected were monomeric pS129 α-synuclein (16–17 kD) and higher molecular weight (hmw) species ranging for 25–50 kD. Insoluble pS129-α-synuclein can be detected in some Stage I samples. D) and E) Significantly higher levels of insoluble monomeric pS129-α–synuclein in temporal cortex insoluble DLB compared to PD samples but not in cingulate cortex. Stage IV samples were separated into two groups based on neuropathological diagnosis of PD or DLB.

Figure 5. Comparison of relative levels of insoluble pS129-α-synuclein species in staged LB cingulate and temporal cortex samples

A and B) Significantly higher levels of insoluble monomeric (A) and higher molecular weight (B) pS129-α-synuclein in cingulate cortex samples compared to temporal cortex samples. Significantly higher levels of pS129-α-synuclein in cingulate stage IV samples (***P<0.001) and stage III samples compared to temporal cortex samples (** P<0.01) for monomeric (B) and higher molecular weight samples (C). Results represent normalized mean values ± standard error of mean (S.E.M.). C) Representative western blots showing patterns of pS129-α-synuclein immunoreactive bands for a selection of staged cingulate cortex insoluble fraction samples. Major bands being detected were monomeric pS129 α-synuclein (16–17 kD) and higher molecular weight (hmw) species ranging for 25–50 kD. Insoluble pS129-α-synuclein can be detected in some Stage I samples. D) and E) Significantly higher levels of insoluble monomeric pS129-α–synuclein in temporal cortex insoluble DLB compared to PD samples but not in cingulate cortex. Stage IV samples were separated into two groups based on neuropathological diagnosis of PD or DLB.

Figure 6. Correlation of pS129-α-synuclein levels with Lewy Type Density Scores in each fraction and brain region

Panels A–E show correlation plots of pS129-α-synuclein levels in cingulate cortex soluble enriched fraction (A), cingulate membrane enriched fraction (B) and cingulate insoluble fraction C) against cingulate Lewy Type Density scores, and temporal cortex soluble enriched fraction (D), temporal membrane enriched fraction (E) and temporal insoluble fraction (F) against corresponding temporal cortex Lewy Type Density scores. Blots show linear regression line. Statistical analysis using Spearman non-parametric analysis (r) with corresponding P values.

Figure 7. Selected regression plots of pS129-α-synuclein levels with different brain regions and fractions

Correlation analysis between selected brain regions and fractions (complete list of all analyses shown in Table 3). Panel A and B shows negative correlations between cingulate-membrane and cingulate soluble (A) and between cingulate membrane and temporal soluble fractions (B). Panels C, D, E and F show positive correlations between cingulate membrane and cingulate insoluble (C); cingulate membrane and temporal membrane (D); cingulate and temporal insoluble (E) and temporal membrane and temporal insoluble (F). Graphs show Pearson R correlation coefficients with corresponding P values to indicate degree of significance.