Burré J, Sharma M, Südhof TC.
α-Synuclein assembles into higher-order multimers upon membrane binding to promote SNARE complex formation.
Proc Natl Acad Sci U S A. 2014 Oct 7;111(40):E4274-83. Epub 2014 Sep 22
PubMed.
Wang et al. present a compelling new model for the physiological role of αSynuclein multimerization in synaptic vesicle recycling and thus neural function. Starting from the key question, “What is the form of αSyn at synapses, one of its normal locales?,” the study builds upon the recently discovered evidence that this amphipathic protein forms metastable, partially α-helical tetramers and related multimers under physiological conditions in healthy cells (Bartels et al., 2011; Dettmer et al., 2013; Westphal and Chandra, 2013; Gould et al., 2014). Wang et al. place the issue directly into a physiological context at the synapse, using a set of sophisticated in situ fluorescent assays. In their key experiment, they transduced αSyn -/- neurons with AAV particles encoding Venus YFP complementation pairs VN/VC-αSyn and visualized the entry of newly synthesized proteins into presynaptic boutons four to five hours later. They observed strong YFP signals (indicative of at least two αSyn molecules interacting) that co-localized with synaptic vesicle clusters. Then, the authors show that the stabilized αSyn multimers attenuate the activity-dependent dispersion of synaptic vesicles and suggest that αSyn multimers associate with synaptic vesicle clusters to restrict their trafficking.
The study is important in several ways. It is the first paper that suggests a physiological function for the recently recognized αSyn multimers in cells. The suggested role is generally in line with earlier studies suggesting such a function for αSyn but without addressing the assembly state of the protein. Moreover, it is the first paper that explicitly interprets oligomerization of αSyn, which has been reported in prior bimolecular fluorescence complementation studies, in light of the emerging concept of physiological multimers. Heretofore, investigators using fluorescence protein complementation to study αSyn have interpreted fluorescent signals as indicating the formation of toxic oligomers. Yet there has been doubt about this because the complementation depends on a defined position of the tags relative to αSyn and occurs very quickly (within hours) in healthy cultured cells, and even in primary neurons, as shown by Wang et al. for αSyn expressed at roughly endogenous levels (not overexpressed, as the authors emphasize). Moreover, the YFP complementation pairs (VN-αSyn and αSyn-VC) rapidly reconstitute fluorescent signals that are almost as strong as those achieved by expression of full-length Venus YFP-tagged αSyn, whereas the generation of toxic, β-sheet oligomers in acute cell culture is generally very hard to achieve. We therefore believe that αSyn fluorescent protein complementation is primarily useful to study the physiological propensity of this abundant neuronal protein to self-interact.
We expect that the findings of Wang et al. may lead to a general concept of how normal multimerization in the synuclein family (α, β, and γ) may regulate membrane homeostasis in cell types other than neurons, especially erythrocytes and their cellular precursors, which are rich in αSyn. Importantly, αSyn fluorescent protein complementation can be observed easily in cultured cells, as shown by Wang et al. (for HEK cells) and in earlier papers, including a cytosolic staining pattern consistent with soluble αSyn multimers (predominantly tetramers) that we (Bartels et al., 2011) and others (Westphal and Chandra, 2013) have observed upon multi-step purification of αSyn from red blood cells under non-denaturing conditions, and also upon differential centrifugation after live-cell crosslinking of proteins in neurons and other healthy cells (Dettmer et al., 2013). Future studies should address in detail the changes in αSyn conformation and assembly state that are associated (in neurons) with its transport to synapses, its binding to synaptic membranes/vesicles, and its release from these vesicles into the cytosol. We postulate that αSyn multimers can remain intact, at least for a time, upon dissociation from membranes, and the Wang et al. study is not incompatible with that idea.
The new study by Burré et al. is consistent with the principal finding of Wang et al. that αSyn multimers exist on membrane vesicles. Burré et al. also address the assembly size of αSyn on neuronal membranes in the context of the strong and direct Synaptobrevin-2/αSyn interaction they have proposed previously (Burré et al., 2010). Upon adding the crosslinking agents glutaraldehyde, dimethyl suberimidate (DMS), dimethyl adipimidate (DMA), and dimethyl pimelimidate (DMP) to mouse brain homogenates made in the absence of detergent, the authors find substantial amounts of αSyn in multimeric states—dimers, tetramers, and higher-order multimers (their Fig. 2). The same crosslinking only yielded monomers when the PBS-soluble (cytosolic) fraction was analyzed, prompting the authors to conclude that the multimers are entirely on membranes and not in the cytosol. Unfortunately, the relative abundance of αSyn in membrane vs. cytosol fractions in their starting total homogenates was not described. Given the well-documented cytosolic localization of ~90 percent of total cellular αSyn (as these authors themselves published [Burré et al., 2013, Fig. 1a]) and the abundance of αSyn multimers in total brain homogenates (their Fig. 2), the total absence of multimers they now report in the cytosol fraction is surprising (but see below). Also, their postulated strong and direct interaction of αSyn and synaptobrevin-2 was not probed in the crosslinked samples, although it would have been interesting to see here whether a trapping of the two proteins in a hetero-oligomer could be achieved. As a next step, they applied crosslinkers to acute brain slices, and >40 percent of αSyn was trapped at dimeric or tetrameric positions by glutaraldehyde, while the monomer pool was virtually completely depleted. This finding is only compatible with their conclusion that cytosolic αSyn is entirely monomeric if one assumes that very little αSyn in the brain is localized in the cytosol. Here, the authors missed the opportunity of defining the localization of the multimers by generating sequential extracts of the tissue (e.g., PBS->Triton->SDS; as described in Dettmer et al., 2013). (Parenthetically, we have recently done such an analysis using the well-established crosslinker disuccinimidyl glutarate [DSG] on a fresh, normal human brain biopsy [Dettmer et al., under review] and found abundant amounts of 60 kDa [tetrameric] αSyn [plus related minor 80 and 100 kDa multimers] in the PBS-soluble fraction, while the Triton fraction only contained monomers.) As their next step, and generally consistent with the αSyn fluorescence complementation done by Wang et al., Burré et al. used FRET to reveal an ordered assembly of the αSyn subunits in the oligomers. Then, using separation of membranes by a sucrose gradient, they find αSyn to be associated with docked vesicles at the presynaptic plasma membrane, in line with the finding of Wang and colleagues.
All in all, both of these interesting papers clearly report the existence of physiological αSyn multimers, whereas only a few years ago, αSyn assemblies—whether in cytoplasm or at membranes—had been interpreted as pathological, since the native form was believed to be an unfolded monomer. The strength of the study by Wang et al. is their almost exclusive use of intact-cell methods for their key experiments, which is the best way to address αSyn homeostasis in our experience. Crosslinking in vitro, i.e., in lysates, as extensively done by Burré et al. here, leads to recovery of principally monomeric αSyn (as we emphasized: see Dettmer et al., 2013; Fig. 5) because of the quantitative depolymerization of physiological multimers by cell lysis, which prevents the detection of the characteristic αSyn multimer pattern that we invariably see upon intact-cell crosslinking (Dettmer et al., 2013). This finding was ignored in the experiments by Burré. Crosslinking of brain slices has also been challenging for us, in part due to the opening of cells by the slicing as well as a risk of over-crosslinking in intact outer-cell layers and under-crosslinking in the inner parts. Instead of slices, we recommend a protocol that involves fine mincing of tissues, followed by multiple, gentle centrifugal washes to pellet the brain pieces and then applying crosslinker only to the tissue pieces that are left largely intact.
The literature has described αSyn as an amphipathic protein that can bind to membranes but is also found abundantly in the cytosol, and we believe that dynamic processes—frequent association with and dissociation from membranescharacterize αSyn homeostasis in an intact, healthy cell. (Parenthetically, Burré et al. again claim [see also Burré et al., 2013] that we postulated an entirely ‘stable’ tetramer in cytosol in our first study [Bartels et al., 2011], but that is incorrect; we postulated a metastable nature of the tetramers we observed, did not use the term stable, and explicitly proposed a search for small molecules that could stabilize them therapeutically [Bartels et al., 2011].) Emerging data, including the two new studies discussed here, point at physiological multimerization being a key part of αSyn biology. Further work is now needed to fully characterize the physiological multimers with regard to their exact subcellular localizations and half-life as well as factors involved in their assembly and disassembly.
In review, we described the tetrameric and cytosolic character of αSyn multimers (Bartels et al., 2011; Dettmer et al., 2013), W. Wang et al. their potential tetrameric and dynamic character in vitro, Westphal and Chandra their tetrameric but apparently "inactive" character, and now L. Wang et al. point out their active function on synapses, while Burré et al. characterize higher-n multimers on membranes, describing the monomeric protein as an inactive cytosolic form. Further work will bring these partially disparate observations together. Given the variety of conformations and assembly states observed to date and the potential multiple physiological functions of αSyn, it seems that both structure and functional activity are context dependent for αSyn. It might thus be inappropriate to assign only one “correct” structure, since a variety of physiological conformations are likely to be in a constant equilibrium, governed by the functional state of the cell. Given the α-helical structure of some of these species, as also discussed here by Burré et al., αSyn multimers should be protective against beta-sheet rich aggregates of αSyn that characterize the synucleinopathies. We therefore conclude that it is important to study physiological multimers, as their destabilization to monomers promises to be one of the earliest events in human synucleinopathies, suggesting entirely novel strategies for treatment.
References:
Wang L, Das U, Scott DA, Tang Y, McLean PJ, Roy S.
α-synuclein multimers cluster synaptic vesicles and attenuate recycling.
Curr Biol. 2014 Oct 6;24(19):2319-26. Epub 2014 Sep 25
PubMed.
Bartels T, Choi JG, Selkoe DJ.
α-Synuclein occurs physiologically as a helically folded tetramer that resists aggregation.
Nature. 2011 Aug 14;477(7362):107-10.
PubMed.
Dettmer U, Newman AJ, Luth ES, Bartels T, Selkoe D.
In vivo cross-linking reveals principally oligomeric forms of α-synuclein and β-synuclein in neurons and non-neural cells.
J Biol Chem. 2013 Mar 1;288(9):6371-85. Epub 2013 Jan 14
PubMed.
Westphal CH, Chandra SS.
Monomeric synucleins generate membrane curvature.
J Biol Chem. 2013 Jan 18;288(3):1829-40. Epub 2012 Nov 26
PubMed.
Gould N, Mor DE, Lightfoot R, Malkus K, Giasson B, Ischiropoulos H.
Evidence of native α-synuclein conformers in the human brain.
J Biol Chem. 2014 Mar 14;289(11):7929-34. Epub 2014 Jan 28
PubMed.
Burré J, Sharma M, Südhof TC.
α-Synuclein assembles into higher-order multimers upon membrane binding to promote SNARE complex formation.
Proc Natl Acad Sci U S A. 2014 Oct 7;111(40):E4274-83. Epub 2014 Sep 22
PubMed.
Burré J, Sharma M, Tsetsenis T, Buchman V, Etherton MR, Südhof TC.
Alpha-synuclein promotes SNARE-complex assembly in vivo and in vitro.
Science. 2010 Sep 24;329(5999):1663-7.
PubMed.
Burré J, Vivona S, Diao J, Sharma M, Brunger AT, Südhof TC.
Properties of native brain α-synuclein.
Nature. 2013 Jun 13;498(7453):E4-6; discussion E6-7.
PubMed.
Wang W, Perovic I, Chittuluru J, Kaganovich A, Nguyen LT, Liao J, Auclair JR, Johnson D, Landeru A, Simorellis AK, Ju S, Cookson MR, Asturias FJ, Agar JN, Webb BN, Kang C, Ringe D, Petsko GA, Pochapsky TC, Hoang QQ.
A soluble α-synuclein construct forms a dynamic tetramer.
Proc Natl Acad Sci U S A. 2011 Oct 25;108(43):17797-802. Epub 2011 Oct 17
PubMed.
Comments
U.K. Dementia Research Institute at University College London
Harvard Medical School, Brigham and Women’s Hospital
Co-Director, Brigham and Women's Hospital's Ann Romney Center for Neurologic Diseases
Wang et al. present a compelling new model for the physiological role of αSynuclein multimerization in synaptic vesicle recycling and thus neural function. Starting from the key question, “What is the form of αSyn at synapses, one of its normal locales?,” the study builds upon the recently discovered evidence that this amphipathic protein forms metastable, partially α-helical tetramers and related multimers under physiological conditions in healthy cells (Bartels et al., 2011; Dettmer et al., 2013; Westphal and Chandra, 2013; Gould et al., 2014). Wang et al. place the issue directly into a physiological context at the synapse, using a set of sophisticated in situ fluorescent assays. In their key experiment, they transduced αSyn -/- neurons with AAV particles encoding Venus YFP complementation pairs VN/VC-αSyn and visualized the entry of newly synthesized proteins into presynaptic boutons four to five hours later. They observed strong YFP signals (indicative of at least two αSyn molecules interacting) that co-localized with synaptic vesicle clusters. Then, the authors show that the stabilized αSyn multimers attenuate the activity-dependent dispersion of synaptic vesicles and suggest that αSyn multimers associate with synaptic vesicle clusters to restrict their trafficking.
The study is important in several ways. It is the first paper that suggests a physiological function for the recently recognized αSyn multimers in cells. The suggested role is generally in line with earlier studies suggesting such a function for αSyn but without addressing the assembly state of the protein. Moreover, it is the first paper that explicitly interprets oligomerization of αSyn, which has been reported in prior bimolecular fluorescence complementation studies, in light of the emerging concept of physiological multimers. Heretofore, investigators using fluorescence protein complementation to study αSyn have interpreted fluorescent signals as indicating the formation of toxic oligomers. Yet there has been doubt about this because the complementation depends on a defined position of the tags relative to αSyn and occurs very quickly (within hours) in healthy cultured cells, and even in primary neurons, as shown by Wang et al. for αSyn expressed at roughly endogenous levels (not overexpressed, as the authors emphasize). Moreover, the YFP complementation pairs (VN-αSyn and αSyn-VC) rapidly reconstitute fluorescent signals that are almost as strong as those achieved by expression of full-length Venus YFP-tagged αSyn, whereas the generation of toxic, β-sheet oligomers in acute cell culture is generally very hard to achieve. We therefore believe that αSyn fluorescent protein complementation is primarily useful to study the physiological propensity of this abundant neuronal protein to self-interact.
We expect that the findings of Wang et al. may lead to a general concept of how normal multimerization in the synuclein family (α, β, and γ) may regulate membrane homeostasis in cell types other than neurons, especially erythrocytes and their cellular precursors, which are rich in αSyn. Importantly, αSyn fluorescent protein complementation can be observed easily in cultured cells, as shown by Wang et al. (for HEK cells) and in earlier papers, including a cytosolic staining pattern consistent with soluble αSyn multimers (predominantly tetramers) that we (Bartels et al., 2011) and others (Westphal and Chandra, 2013) have observed upon multi-step purification of αSyn from red blood cells under non-denaturing conditions, and also upon differential centrifugation after live-cell crosslinking of proteins in neurons and other healthy cells (Dettmer et al., 2013). Future studies should address in detail the changes in αSyn conformation and assembly state that are associated (in neurons) with its transport to synapses, its binding to synaptic membranes/vesicles, and its release from these vesicles into the cytosol. We postulate that αSyn multimers can remain intact, at least for a time, upon dissociation from membranes, and the Wang et al. study is not incompatible with that idea.
The new study by Burré et al. is consistent with the principal finding of Wang et al. that αSyn multimers exist on membrane vesicles. Burré et al. also address the assembly size of αSyn on neuronal membranes in the context of the strong and direct Synaptobrevin-2/αSyn interaction they have proposed previously (Burré et al., 2010). Upon adding the crosslinking agents glutaraldehyde, dimethyl suberimidate (DMS), dimethyl adipimidate (DMA), and dimethyl pimelimidate (DMP) to mouse brain homogenates made in the absence of detergent, the authors find substantial amounts of αSyn in multimeric states—dimers, tetramers, and higher-order multimers (their Fig. 2). The same crosslinking only yielded monomers when the PBS-soluble (cytosolic) fraction was analyzed, prompting the authors to conclude that the multimers are entirely on membranes and not in the cytosol. Unfortunately, the relative abundance of αSyn in membrane vs. cytosol fractions in their starting total homogenates was not described. Given the well-documented cytosolic localization of ~90 percent of total cellular αSyn (as these authors themselves published [Burré et al., 2013, Fig. 1a]) and the abundance of αSyn multimers in total brain homogenates (their Fig. 2), the total absence of multimers they now report in the cytosol fraction is surprising (but see below). Also, their postulated strong and direct interaction of αSyn and synaptobrevin-2 was not probed in the crosslinked samples, although it would have been interesting to see here whether a trapping of the two proteins in a hetero-oligomer could be achieved. As a next step, they applied crosslinkers to acute brain slices, and >40 percent of αSyn was trapped at dimeric or tetrameric positions by glutaraldehyde, while the monomer pool was virtually completely depleted. This finding is only compatible with their conclusion that cytosolic αSyn is entirely monomeric if one assumes that very little αSyn in the brain is localized in the cytosol. Here, the authors missed the opportunity of defining the localization of the multimers by generating sequential extracts of the tissue (e.g., PBS->Triton->SDS; as described in Dettmer et al., 2013). (Parenthetically, we have recently done such an analysis using the well-established crosslinker disuccinimidyl glutarate [DSG] on a fresh, normal human brain biopsy [Dettmer et al., under review] and found abundant amounts of 60 kDa [tetrameric] αSyn [plus related minor 80 and 100 kDa multimers] in the PBS-soluble fraction, while the Triton fraction only contained monomers.) As their next step, and generally consistent with the αSyn fluorescence complementation done by Wang et al., Burré et al. used FRET to reveal an ordered assembly of the αSyn subunits in the oligomers. Then, using separation of membranes by a sucrose gradient, they find αSyn to be associated with docked vesicles at the presynaptic plasma membrane, in line with the finding of Wang and colleagues.
All in all, both of these interesting papers clearly report the existence of physiological αSyn multimers, whereas only a few years ago, αSyn assemblies—whether in cytoplasm or at membranes—had been interpreted as pathological, since the native form was believed to be an unfolded monomer. The strength of the study by Wang et al. is their almost exclusive use of intact-cell methods for their key experiments, which is the best way to address αSyn homeostasis in our experience. Crosslinking in vitro, i.e., in lysates, as extensively done by Burré et al. here, leads to recovery of principally monomeric αSyn (as we emphasized: see Dettmer et al., 2013; Fig. 5) because of the quantitative depolymerization of physiological multimers by cell lysis, which prevents the detection of the characteristic αSyn multimer pattern that we invariably see upon intact-cell crosslinking (Dettmer et al., 2013). This finding was ignored in the experiments by Burré. Crosslinking of brain slices has also been challenging for us, in part due to the opening of cells by the slicing as well as a risk of over-crosslinking in intact outer-cell layers and under-crosslinking in the inner parts. Instead of slices, we recommend a protocol that involves fine mincing of tissues, followed by multiple, gentle centrifugal washes to pellet the brain pieces and then applying crosslinker only to the tissue pieces that are left largely intact.
The literature has described αSyn as an amphipathic protein that can bind to membranes but is also found abundantly in the cytosol, and we believe that dynamic processes—frequent association with and dissociation from membranescharacterize αSyn homeostasis in an intact, healthy cell. (Parenthetically, Burré et al. again claim [see also Burré et al., 2013] that we postulated an entirely ‘stable’ tetramer in cytosol in our first study [Bartels et al., 2011], but that is incorrect; we postulated a metastable nature of the tetramers we observed, did not use the term stable, and explicitly proposed a search for small molecules that could stabilize them therapeutically [Bartels et al., 2011].) Emerging data, including the two new studies discussed here, point at physiological multimerization being a key part of αSyn biology. Further work is now needed to fully characterize the physiological multimers with regard to their exact subcellular localizations and half-life as well as factors involved in their assembly and disassembly.
In review, we described the tetrameric and cytosolic character of αSyn multimers (Bartels et al., 2011; Dettmer et al., 2013), W. Wang et al. their potential tetrameric and dynamic character in vitro, Westphal and Chandra their tetrameric but apparently "inactive" character, and now L. Wang et al. point out their active function on synapses, while Burré et al. characterize higher-n multimers on membranes, describing the monomeric protein as an inactive cytosolic form. Further work will bring these partially disparate observations together. Given the variety of conformations and assembly states observed to date and the potential multiple physiological functions of αSyn, it seems that both structure and functional activity are context dependent for αSyn. It might thus be inappropriate to assign only one “correct” structure, since a variety of physiological conformations are likely to be in a constant equilibrium, governed by the functional state of the cell. Given the α-helical structure of some of these species, as also discussed here by Burré et al., αSyn multimers should be protective against beta-sheet rich aggregates of αSyn that characterize the synucleinopathies. We therefore conclude that it is important to study physiological multimers, as their destabilization to monomers promises to be one of the earliest events in human synucleinopathies, suggesting entirely novel strategies for treatment.
References:
Wang L, Das U, Scott DA, Tang Y, McLean PJ, Roy S. α-synuclein multimers cluster synaptic vesicles and attenuate recycling. Curr Biol. 2014 Oct 6;24(19):2319-26. Epub 2014 Sep 25 PubMed.
Bartels T, Choi JG, Selkoe DJ. α-Synuclein occurs physiologically as a helically folded tetramer that resists aggregation. Nature. 2011 Aug 14;477(7362):107-10. PubMed.
Dettmer U, Newman AJ, Luth ES, Bartels T, Selkoe D. In vivo cross-linking reveals principally oligomeric forms of α-synuclein and β-synuclein in neurons and non-neural cells. J Biol Chem. 2013 Mar 1;288(9):6371-85. Epub 2013 Jan 14 PubMed.
Westphal CH, Chandra SS. Monomeric synucleins generate membrane curvature. J Biol Chem. 2013 Jan 18;288(3):1829-40. Epub 2012 Nov 26 PubMed.
Gould N, Mor DE, Lightfoot R, Malkus K, Giasson B, Ischiropoulos H. Evidence of native α-synuclein conformers in the human brain. J Biol Chem. 2014 Mar 14;289(11):7929-34. Epub 2014 Jan 28 PubMed.
Burré J, Sharma M, Südhof TC. α-Synuclein assembles into higher-order multimers upon membrane binding to promote SNARE complex formation. Proc Natl Acad Sci U S A. 2014 Oct 7;111(40):E4274-83. Epub 2014 Sep 22 PubMed.
Burré J, Sharma M, Tsetsenis T, Buchman V, Etherton MR, Südhof TC. Alpha-synuclein promotes SNARE-complex assembly in vivo and in vitro. Science. 2010 Sep 24;329(5999):1663-7. PubMed.
Burré J, Vivona S, Diao J, Sharma M, Brunger AT, Südhof TC. Properties of native brain α-synuclein. Nature. 2013 Jun 13;498(7453):E4-6; discussion E6-7. PubMed.
Wang W, Perovic I, Chittuluru J, Kaganovich A, Nguyen LT, Liao J, Auclair JR, Johnson D, Landeru A, Simorellis AK, Ju S, Cookson MR, Asturias FJ, Agar JN, Webb BN, Kang C, Ringe D, Petsko GA, Pochapsky TC, Hoang QQ. A soluble α-synuclein construct forms a dynamic tetramer. Proc Natl Acad Sci U S A. 2011 Oct 25;108(43):17797-802. Epub 2011 Oct 17 PubMed.
View all comments by Dennis SelkoeMake a Comment
To make a comment you must login or register.