|Home | About | Journals | Submit | Contact Us | Français|
In Parkinson’s disease (PD) and other α-synucleinopathies, pre-fibrillar α-synuclein (αS) oligomer is implicated in the pathogenesis. However, toxic αS oligomers observed using in vitro systems are not generally seen associated α-synucleinopathy in vivo. Thus, pathologic significance of αS oligomers to αS neurotoxicity is unknown. Herein, we show that, αS that accumulate within endoplasmic reticulum/microsome(ER/M) forms toxic oligomers in mouse and human brain with the α-synucleinopathy. In the mouse model of α-synucleinopathy, αS oligomers initially form prior to the onset of disease and continue to accumulate with the disease progression. Significantly, treatment of αS transgenic mice with Salubrinal, an anti-ER stress compound that delays the onset of disease, reduces ER accumulation of αS oligomers. These results indicate that αS oligomers with toxic conformation accumulate in ER and αS oligomer dependent the ER-stress is pathologically relevant for PD.
Parkinson’s disease (PD) is the second most common neurodegenerative disease after Alzheimer’s disease (AD). While the etiology of PD is unknown in most cases, degenerating neuronal populations in PD exhibit α-synuclein (αS) abnormalities and mutations in αS gene cause familial PD, indicating that the αS abnormalities are mechanistically linked to pathogenesis of PD and other α-synucleinopathies (Vila and Przedborski, 2004; Obeso et al., 2010).
Neuropathological studies of human α-synucleinopathy cases establish αS aggregates, appearing as Lewy bodies and neurites, as prominent pathological features. Further, transgenic (Tg) mouse models of α-synucleinopathy with overt progressive neurodegeneration are associated with αS aggregates. These results have supported view that the αS fibrils are of pathogenic importance in PD and other α-synucleinopathies (Obeso et al., 2010). However, this view has been challenged due to inconsistent correlations between αS fibrils and neurodegeneration in several model systems (Volles and Lansbury, 2003). In addition, membrane disrupting and cytotoxic properties of various in vitro assembled αS oligomers and the emerging evidence for toxic Aβ oligomers in Alzheimer’s disease (AD) have fueled the view that pre-fibrillar soluble αS oligomers may be pathogenic in PD (Volles and Lansbury, 2003; Lansbury and Lashuel, 2006). However, in vivo relevance of the soluble αS oligomers in PD is uncertain due to the lack of convincing evidence for disease-linked accumulation of toxic soluble αS oligomers. Thus, while multiple pathogenic mechanisms are proposed for PD based on the soluble oligomers (Obeso et al., 2010), it is unknown if any of the proposed toxic mechanisms contribute to neurodegeneration associated with α-synucleinopathies.
In this report, we show that α-synucleinopathy in human and mouse is associated with preferential accumulation of toxic αS oligomers within the endoplasmic reticulum/microsomes (ER/M). In the Tg mouse model of α-synucleinopathy, accumulation of the toxic αS oligomers are temporally and spatially linked to the induction of chronic ER stress, neuropathology, and neurological abnormalities. Finally, Salubrinal, an anti-ER stress agent, attenuates disease manifestations and reduces accumulation of toxic αS oligomers in the αS Tg mouse model. We propose that toxic αS oligomers accumulate in vivo and contribute to neurodegeneration by causing chronic ER stress response.
Transgenic mice expressing high levels of WT or mutant (A53T and A30P) αS under the control of the mouse prion protein (PrP) promoter have been described previously (Lee et al., 2002; Martin et al., 2006; Wang et al., 2008). Mice expressing A53T αS [line G2-3(A53T)] develop fatal neurological disease at ~12 months of age. Pre-symptomatic mice were 10–14 months old mice free from any motor dysfunction. All animal study methods were approved in full by the Institutional Animal Care and Use Committee of the Johns Hopkins University and consistent with the requirements of the National Institutes of Health Officeof Laboratory Animal Welfare Policy.
For subcellular fractionation of ER membrane enriched microsomes we used published methods (Cox and Emili, 2006) with modifications (Colla et al., submitted). To further enrich for the ER content, the microsome preparation were purified using a discontinuous sucrose gradient (Croze and Morre, 1984). Pure mitochondria were obtained by discontinuous gradient fractionation of the crude mitochondria pellet (P10) (Vijayvergiya et al., 2005). Pure nuclei were isolated starting from the crude nuclei pellet (P1) (Cox and Emili, 2006) using a sucrose gradient (Colla et al., submitted).
The microsome fractions were treated with or without 50 μg/ml proteinase K (Roche) and 1% Triton x-100 for 20 min on ice. The reaction was stopped by addiction of 2mM (final concentration) of phenylmethylsulfonyl fluoride.
Immunoblot and dot-blot analysis of mice and human brain tissues were performed as previously described (Lee et al., 2002; Li et al., 2004; Li et al., 2005; Martin et al., 2006; Wang et al., 2008). For dot-blot analysis, lysates were spotted directly on the nitrocellulose membrane and let it dry completely. Immunoreactivity was visualized using chemiluminescence detection (Pierce, Rockford, IL) after incubations with the appropriate horseradish peroxidase-conjugated secondary antibody, using a CCD based Biorad Molecular Imager ChemiDoc XRS+ System (Biorad, Hercules, CA) or X-ray films. The immunoreactive band intensities were determined by densitometry or the Quantity One software (Biorad, Hercules, CA).
Following antibodies were used: grp78, grp94, (Stressgen, Ann Arbor, MI); Syn-1, cytochrome C (BD Transduction Laboratories, Franklin Lakes, NJ); A11 (Kayed et al., 2003); pser129-αS (Fujiwara et al. 2002); syn303 (Duda et al., 2000); FILA-1 (Lindersson et al., 2004).
Studies indicate that αS can functionally impact multiple organelles (Gosavi et al., 2002; Smith et al., 2005; Cooper et al., 2006; Martinez-Vicente et al., 2008; Winslow et al., 2010). Similarly, our analysis of αS species in various subcellular compartments [nuclear, mitochondrial, ER/M, and cytosol] reveals association of αS monomers and high MW αS (αS aggregates), including serine-129 phosphorylated (pS129) αS (Fujiwara et al., 2002), with ER/M (Colla et al., submitted). In the companion report (Colla et al., submitted), we also show that accumulation of αS in the ER is linked to pathogenic chronic ER stress.
Because of the pathogenic relevance of ER associated αS aggregates, the ER/M associated αS were examined in more detail. First, we examined if the ER associated αS becomes detergent insoluble with disease, as we have shown for total αS aggregates (Lee et al., 2002). The ER/M associated αS were fractionated into non-ionic detergent (0.5% Tx-100) soluble and insoluble fractions and subjected to the immunoblot analysis (Fig. 1A). The result show that the majority of SDS-resistant high MW αS variants in the ER/M from the sick A53TαS Tg mice, as well as associated αS monomers, pellet with detergent-insoluble (Tx-I) fractions (Fig. 1A). Surprisingly, the detergent soluble fractions (Tx-S) of ER/M from the sick Tg mice contain significant amounts of SDS-resistant high MW αS variants (Fig. 1A). Similar analysis using total spinal cord lysates do not show soluble high MW αS variants (Fig. 1B). We also performed a side-by-side immunobolot comparison of the Tx-I ER/M and the TX-Ptot fractions from the endstage A53TαS Tg mice (Fig. 1C). The results show that within each subcellular fraction, the patterns of αS variants are remarkably consistent between individuals. Further, despite some similarities, there were notable differences in the αS variants between the fractions. In particular, compared to the TX-Ptot, the ER/M Tx-I fractions contain notably lower levels of truncated αS variant (Fig. 1C, ΔC) and less complex set of high MW αS variants (Fig. 1C, arrowheads). While it the αS variants in the ER/M represent a subset of the TX-Ptot variants, it is also clear that these two fractions contain biochemically distinct set of αS variants. These results suggest that the ER/M associated αS variants may be biochemically distinct from the total detergent insoluble αS variants (TX-Ptot).
The soluble SDS-resistant αS variants in ER/M could be the elusive “toxic soluble oligomers”. However, simple mobility shifts of αS are not sufficient indicators of toxic conformation. Thus, the subcellular fractions were evaluated using antibodies to the toxic oligomers; A11, selective for all soluble toxic oligomers (Kayed et al., 2003), and FILA-1, selective for the toxic αS oligomer/fibril (Lindersson et al., 2004; Paleologou et al., 2009). Both antibodies can also inhibit αS oligomer-induced cytotoxicity in vitro (Kayed et al., 2003; Lindersson et al., 2004). As indicators of mature αS aggregates, we also used antibodies that detect oxidized/aggregated αS (Syn303) (Duda et al., 2000) and phospho-serine129 (pS129) αS (Fujiwara et al., 2002).
Initial dot blot analysis using FILA-1 shows that FILA-1 reacts with in vitro assembled αS fibrils (αS F) but does not bind to in vitro assembled tau fibrils (Tau F). Significantly, FILA-1 showed obvious reactivity to the ER/M fraction but not to the total insoluble αS (Tx-Ptot) from the A53TαS Tg mice (Fig. 1E). To determine if ER/M associated toxic αS oligomers accumulate with disease in vivo, we performed dot blot analysis of ER/M, cytosolic (C), and total triton insoluble fractions (TX-P) from the spinal cords (SpC) of A53TαS Tg and nTg using the above antibodies (Fig. 1D). As expected, the cytosolic fraction exhibited robust signal for αS but showed weak signals for all other antibodies. The TX-Ptot fractions reacted with antibodies characteristic for aggregated αS (Syn303, pS129), confirming the presence of αS aggregates, but did not react to anti-oligomer antibodies. In contrast, the ER/M from the aged presymptomatic (PreS) and diseased (sick) A53TαS Tg mice react with both A11 and FILA-1 (Fig. 1D). The A11 and FILA-1 reactions are not nonspecific cross-reactions to increased αS levels since the TX-P and cytosolic fractions, despite the high levels of αS, do not react to either A11 or FILA-1. Similarly, the ER/M fraction from the mature (6 mos) but pathology free A53TαS Tg mice did not react to A11 or FILA-1. The FILA-1 reactivity to ER/M fraction is selective for the presence of α-synucleinopathy as ER/M from aged WTαS (I2-2) and A30PαS (O2) Tg mice, are not FILA-1 reactive (Fig. 1E). Similarly, ER/M from cortex of end stage A53TαS Tg mice is not FILA-1 reactive (Fig. 1F). Collectively, these results show the disease specificity of toxic αS oligomer accumation.
The ER/M fractions from PreS were only reactive to anti-oligomer antibodies. However, the ER/M fractions from the sick A53TαS Tg mice also react to Syn303 and pS129 antibodies, indicating that these are mature aggregates (Fig. 1D). Similarly, analysis of PreS, early symptomatic, and end stage mice show that the ER/M acquires pS129αS reactivity with the onset of motoric symptoms (Fig. 1F).
In our companion report, we showed that ER/M associated αS monomers are located within the lumen of ER/M. However, similar analysis of ER/M from symptomatic A53TαS Tg mice show that high MW αS variants are partially sensitive to proteinase K (PK), resulting in protected αS monomer and proteolytic fragments of ~6–8kDa (Fig. 1G, H). Significantly, when the PK treated ER/M fractions were further separated into Tx-S (soluble) and Tx-I (insoluble) fractions, most of the αS monomer partitions with Tx-S while the 6kDa PK-derived αS fragments remain insoluble with the Tx-I (Fig. 1H).
These results raise the possibility that while αS monomers are located within the lumen, αS oligomers could be forming on the cytosolic face of the ER/M. To study this possibility, we examined whether the ER/M associated oligomers were sensitive to PK proteolysis. When the PK-treated ER/M were dot blotted for FILA-1, the FILA-1 reactive oligomers were PK-resistant in both presymptomatic and symptomatic Tg mice. Consistent with PK resistance of ER/M αS (Fig. 1G, H), FILA-1 epitope in the pre-symptomatic mice is resistant to PK (Fig. 1I). Despite the partial PK-proteolysis of high MW αS (Fig. 1G, H), FILA-1 epitope survive both detergent and PK-treatments (Fig. 1I). Based on the immunoblot analysis of the PK-digested fractions (Fig. 1G, H), we propose that the PK-resistant αS fragment at ~6 kDa could be responsible for the oligomer conformation. This fragment is similar to the previously identified NAC-region containing peptide truncated at both N- and C-terminus (Li et al., 2005). Collectively, the results suggest that αS oligomers initially forms within the ER/M lumen but becomes exposed to the cytosol with the formation of detergent stable αS oligomers, a hypothesis consistent with the membrane destabilization properties of the toxic oligomers (Volles and Lansbury, 2003; Lansbury and Lashuel, 2006).
Our data also indicate to the presence of at least four distinct αS pools in brain afflicted with α-synucleinopathy; soluble monomers, A11/FILA-1 negative αS aggregates, soluble A11/FILA-1 reactive (toxic) oligomers, and insoluble A11/FILA-1 reactive (toxic) oligomers/aggregates. Our studies also provide several additional insights. First, both A11 and FILA-1 reactivity is abolished if ER/M from PreS Tg mice are exposed to 0.5% TX-100 but when ER/M from symptomatic mice is used, only A11 reactivity is abolished by TX-100. Thus, it is likely that the subset of αS oligomers that react to both A11 and FILA-1 are labile to detergent exposure, such as in PreS Tg mice. However, because FILA-1 binds to both soluble oligomer and fibrils, FILA-1 epitope is associated with more mature αS fibrils/oligomers in symptomactic mice. In conjunction with temporal appearance of the pS129αS and syn303 reactivity, our results are consistent with the progressive conversion of the soluble αS oligomers to αS fibrils.
We also determined αS oligomers accumulate in human α-synucleinopathy. We examined human PD and control cases for the presence of toxic αS oligomers (Fig. 2). While we used postmortem tissues, distribution of organelle markers upon fractionation show that our ER/M fractions are enriched for ER markers (Calnexin), but relatively free of mitochondrial and cytosolic markers (Colla et al., submitted). Dot blot analysis ER/M fractions from human Brain Stem (BrSt), area most affected by α-synucleinopathy, show higher FILA-1 and A11 reactivity in PD cases (Fig. 2A). Consistent with the advanced neuropathology in the human cases, the ER/M fractions from PD cases also show increased pS129 αS reactivity (Fig. 2A) and high molecular mass αS species (Fig. 2A). In contrast, total detergent soluble (Tx-Stot) and insoluble (TX-Ptot) fractions from PD cases do not show increased levels of toxic oligomers (Fig. 2B, C). Thus, in both human and mouse cases of α-synucleinopathy, there is selective accumulation of toxic αS oligomers within the ER/M compartment. These results represent the first convincing demonstration of disease-associated in vivo accumulation of αS oligomers with toxic conformations.
Consistent with the toxic nature of the ER associated αS oligomers, we were able to document ER dysfunction and activation of ER stress/unfolded protein response (ERS/UPR) with disease in the A53TαS Tg mouse model (Colla et al., Submitted). It is significant that initial appearance of the toxic αS oligomers precede the onset of ERS and motoric dysfunction in the mice. In addition, treatment of the A53TαS Tg mouse model with Salubrinal, an anti-ERS compound (Boyce et al., 2005) that delays the onset of disease (Colla et al., submitted), leads to significantly reduced accumulation of toxic αS oligomers (Fig. 3). This is a selective effect on the ER/M αS, as the total αS levels do not change with Salubrinal treatment (Colla et al., submitted). Overall, the results provide temporal, spatial, and quantitative associations between ER/M accumulation of αS oligomer, ERS, and neurodegeneration.
In this report, we show that a small subset of αS localizes to the lumen of the ER/M (Colla et al. submitted) and accumulate as toxic αS oligomers. Initial appearance of the toxic αS oligomers precede onset of disease and increases with the progression of α-synucleinopathy. Moreover, treatment of mice with a known anti-ER stress agent, Salubrinal, attenuates disease manifestation and reduces levels of ER/M αS oligomers. Overall, our results indicate to pathogenic significance of the ER/M associated αS oligomers and resulting ER stress in PD.
In AD, in vivo accumulation of the toxic Aβ oligomers in disease conditions are well documented and the pathologic significance of toxic Aβ oligomers in AD is generally accepted (Lansbury and Lashuel, 2006). However, for PD and other α-synucleinopathies, that lack of evidence for in vivo accumulation of toxic αS oligomers have limited the significance of toxic αS oligomers in vivo. Thus, despite the in vitro studies showing cellular toxicity of αS oligomers (Kayed et al., 2003; Lindersson et al., 2004; Tsika et al., 2010), in vivo pathologic significances of αS oligomers were unclear. In particular, while FILA-1 and A11 reactive toxic αS oligomers were seen using in vitro systems, evidence for in vivo accumulations of these types of αS oligomers were limited (Paleologou et al., 2009). Thus, it is remarkable that we are able to show the accumulations of such oligomers in Tg mouse model and PD cases. The fact that accumulation of ER/M associated αS oligomers are temporally and quantitatively linked to onset of ER stress (Colla et al., submitted) and neurodegeneration supports the view that the ER/M αS oligomers are indeed pathogenic. In this regard, the non-ionic detergent soluble 53Å αS oligomers present in spinal cord of anther A53TαS Tg mouse model (Tsika et al., 2010) may be related to the αS oligomer identified here.
Based on our results, we hypothesize a series of events leading to neurotoxicity. Normally, with aging and other stressed conditions, low levels of αS in the ER forms immature soluble αS oligomers that are A11/FILA-1 positive. These initial oligomers within the ER lumen may seed further assembly and maturation of αS oligomers, leading to accumulation of detergent insoluble macro aggregates/fibrils. Significantly, phosphorylation of Ser-129 on αS seems to occur only with mature oligomers/aggregates, suggesting that this event is not required for the initiation of the toxic oligomer formation. Moreover, it is possible that the accumulation of αS oligomers in the ER/M could be related to the secretion αS and cell-to-cell transmission of αS toxicity (Desplats et al., 2009). With the maturation of the αS oligomers from soluble to insoluble states, we hypothesize that the integrity of ER membranes are compromised and exposes portions of ER lumen to cytosolic compartment, leading to chronic ERS (shown in our companion report; Colla et al., submitted). While αS oligomers appear prior to ERS, it is also possible that onset of ERS promotes further accumulation and maturation of the toxic αS oligomer. While the mechanistic particulars behind our findings will require further studies, our results show that, in α-synucleinopathy, toxic αS oligomer accumulation precede disease onset and mature with the disease progression. Thus, the results in this and our companion report support direct pathological links between αS oligomers, ERS, and α-synucleinopathy.
We thank Dr. Virginia Lee for kindly providing the Syn303 antibody. This work was supported by NIH Grants NS038065, NS0380377, NS055776 and ES017384 (MKL); and Lundbeck Foundation and European Community’s Seventh Framework Programme (PHJ).
Authors declare no conflict of interest