| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
-synuclein oligomeric species by fluorescence lifetime imaging
* MassGeneral Institute for Neurodegenerative Disease, Alzheimer’s Disease Research Unit, Massachusetts General Hospital, Charlestown, Massachusetts, USA;
Department of Neurology, University of Regensburg, Regensburg, Germany
1Correspondence: MassGeneral Institute for Neurodegenerative Disease, Alzheimer’s Disease Research Unit, Massachusetts General Hospital, 114 16th St., Charlestown, MA 02129 USA. E-mail: pmclean{at}partners.org
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-synuclein molecules are believed to play a major role in neuronal dysfunction and loss in Parkinson’s disease (PD) and dementia with Lewy bodies. However, -synuclein oligomerization and aggregation have been detected only indirectly in cells using detergent extraction methods. Here, we show for the first time intracellular -synuclein oligomerization using fluorescence lifetime imaging (FLIM). Two forms of -synuclein homomeric interactions were detected: an antiparallel amino terminus-carboxyl terminus interaction between -synuclein molecules, and a close amino terminus-carboxy terminus interaction within single -synuclein molecules. Coexpression of the chaperone protein Hsp70, which can block -synuclein toxicity in several systems, causes -synuclein to adopt a different, open conformation, but Hsp70 does not alter -synuclein–-synuclein interactions. Thus, the neuroprotective effect of Hsp70 can be explained by its chaperone activity on -synuclein molecules, rather than alteration of -synuclein–-synuclein interactions.—Klucken, J., Outeiro, T. F., Nyugen, P., McLean, P. J., and Hyman, B. T. Detection of novel intracellular -synuclein oligomeric species by fluorescence lifetime imaging.
Key Words: Parkinson’s disease • Lewy body disease • chaperone • protein aggregation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This problem is especially critical in understanding the protein-folding phenomena that occur in neurodegenerative diseases. A consensus is emerging that natively unfolded molecules can misfold and form either toxic oligomeric species or microscopic aggregates, leading to neurodegeneration.
-Synuclein is one example of this general principle. -Synuclein is a small (140 amino acid) protein that has three major domains, an amphipathic -helical domain at the amino terminus, a relatively hydrophobic NAC domain, and an acidic carboxyl terminal domain. Although in solution it has been suggested to be in random conformation (1
, 2)
, -synuclein adopts a specific conformation in association with model lipids (3
4
5)
. -Synuclein is a major constituent of Lewy bodies in PD, and mutations of -synuclein are a cause of autosomal dominantly inherited PD. Moreover, recent observations suggest that triplication of the -synuclein gene is sufficient to lead to an autosomal dominant Parkinson syndrome as well (6)
. Similarly, overexpression of -synuclein has proven to be toxic in multiple experimental conditions, including drosophila, rodents, nonhuman primates, and human neural-derived cells in culture (7
8
9
10
11
12)
. Concurrent overexpression of the chaperone protein Hsp70 (or Hsc70) provides protection from this toxicity, consistent with the idea that misfolding of overexpressed -synuclein is central to the toxic mechanisms (13
14
15
16
17)
. However, it is unknown whether the effect of HSPs is on -synuclein conformation, on oligomeric forms of -synuclein, or on prevention or dissolution of macroscopic aggregates.
To explore this question further, we have developed a method that specifically detects -synuclein conformation as well as -synuclein interactions within cells, using a highly sensitive and specific assay of molecular proximity, fluorescence lifetime imaging microscopy (FLIM).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-synuclein have been described previously (11, 14–16, 18). Briefly, cDNA encoding the genes were cloned into pcDNA3.1 (Invitrogen, Carlsbad, CA, USA) expression vectors. N- and/or C-terminal hemagglutinin (HA) or V5 tags were generated by using annealed oligomers coding for HA or V5, and subcloned into WT -synuclein expressing pcDNA3.1 plasmid. Five differently tagged -synuclein constructs were used in this study (Fig. 1
). Human Hsp70 cDNA was kindly provided by J.-C. Plumier, MA General Hospital, and subcloned into pcDNA3.1 (Clontech, Palo Alto, CA, USA).
|
Cell culture, transfection, and immunocytochemistry
Human H4 neuroglioma cells [HTB-148; American Type Culture Collection (ATCC), Manassas, VA, USA] were maintained in OPTI-MEM (Life Technologies, Grand Island, NY, USA) supplemented with 10% FBS. H4 cells were passaged 24 h prior to transfection and plated in four-well chamber slides for immunocytochemistry (Labtek, Nalgen-Nunc, Naperville, IL, USA). Cells were transfected with equimolar ratios of plasmids using Superfect (Qiagen, Chatsworth, CA, USA), according to the manufacturer’s instructions. After 24 h, cells were washed with PBS and fixed with 4% paraformaldehyde for 10 min at room temperature. After washing with PBS, cells were permeabilized in TBS containing 0.1% triton X-100 for 20 min at room temperature. After blocking in 1.5% normal goat serum containing TBS for 1 h, cells were incubated with primary antibody (Ab) for 2 h at room temperature or overnight at 4°C (rat anti-HA 1:1000, 3F10, Roche Diagnostics GmbH, Mannheim, Germany; rabbit anti-V5 1:3000, AB9116, Abcam, Cambridge, MA, USA), followed by washing with PBS and secondary Ab incubation for 1 h (goat antirat IgG-Alexa 488, 1:300, Molecular Probes, Eugene, OR, USA; goat anti-rabbit IgG-Cy3 1:300, Jackson Immunoresearch, PA). After a final wash, slides were mounted with aqueous mounting solution (GVA, Zymed, San Francisco, CA, USA) and subjected to fluorescence microscopy and fluorescence lifetime imaging microscopy (FLIM).
Coimmunoprecipitation and immunoblotting
Twenty-four hours after transfection, H4 cells were washed with cold PBS, harvested by scraping in cold lysis buffer without detergents [50 mM tris/HCl, pH 7.4; 175 mM NaCl; 5 mM EDTA, pH 8.0, protease inhibitor cocktail (Roche, Basel, Switzerland)] and sonicated for 10 s. Lysates were cleared from debris by a 10,000 RPM centrifugation for 10 min at 4°C. Equal amounts of protein were incubated with Protein A/G agarose beads for 2 h at 4°C to preclear the lysate followed by centrifugation at 10,000 RPM for 10 min at 4°C. The supernatant was incubated with 5 µg of rat-anti-HA Ab overnight at 4°C. Protein A/G aggarose beads were added and incubated for 1 h at 4°C to precipitate immuncomplexes followed by centrifugation at 10,000 RPM for 10 min at 4°C. Bead-immunocomplexes were washed two times in lysis buffer and three times with washing buffer (25 mM tris/HCl, 150 mM NaCl, pH 7.2) containing either 0.01% Nonidet P-40 or no detergent. Immunocomplexes were subjected to SDS-PAGE using 10–20% tris-glycine gels (Novex, San Diego, CA, USA) for Western blot analysis. Protein was transferred to Immobilon-P membrane (Millipore, Bedford, MA, USA) and blocked in Lycor blocking buffer (Lycor, Lincoln, NE, USA) for 1 h prior to the addition of the primary Ab (anti-Hsp70 – SPA830, anti-Hsp90 – SPA812, or anti-Hdj-1 – SPA400; Stressgen, Victoria, BC, Canada, anti-V5 – AB9116, Abcam, Cambridge, MA, USA) at room temperature for 1–2 h or overnight at 4°C. The blots were washed three times in tris-buffered saline with 0.2% Tween (TBS-T, pH 7.4) and were incubated at room temperature for 1 h in fluorescent-labeled secondary antibodies (IRDye 800 anti-rabbit or anti-mouse, Rockland Immunochemicals, Gilbertsville, PA, USA, 1:3000 or Alexa-680 anti-rabbit or anti-mouse, Molecular Probes, Eugene, OR, USA; 1:3000). After being washed three times in TBS-T, immunoblots were processed and quantified using the Odyssey infrared imaging system (Lycor, Lincoln, NE, USA).
Fluorescence lifetime imaging microscopy and calculation
Recently, FLIM has been described as a technique for the analysis of protein proximity (19
20
21)
. The technique is based on the observation that fluorescence lifetimes of a donor fluorophore shorten if it is in close proximity (<10 nm) to a FRET acceptor. The decrease in lifetime is proportional to the distance between the fluorophores at R6. A mode-locked Ti-sapphire laser (Spectra-Physics, Fremont, CA, USA) emits a femtosecond pulse every 12 ns to excite the fluorophore. A high-speed Hamamatsu (Bridgewater, NJ, USA) detector and hardware/software (SPC-830 Becker and Hickl, Berlin, Germany) were used to measure fluorescence lifetimes on a pixel-by-pixel basis. Donor fluorophore (Alexa 488) lifetimes were fit to two-exponential decay. One component was fixed to the expected lifetime of Alexa 488 without an acceptor (Cy3) in close proximity (negative control – monofit) that was determined by fitting to one-exponential decay curve [mean lifetime monofit 2268±8 ps (SEM)]. To validate the two-component fit procedure, the same cells from the negative control mono-fit were subjected to two exponential decay curve fittings and revealed the negative control value for Alexa 488 lifetime of 2244 ps (±12 ps SEM), which did not differ from the mono-fit lifetime and was used in the experiment as calculated negative control. All combinations of HA-, V5-, or untagged -synuclein molecules were stained using the same Ab combination as described above. The normal lifetime of Alexa 488 fluorophore (negative control) was estimated from cells transfected with a 50–50 mix of HA-tagged -synuclein and untagged -synuclein containing plasmids and immunostained as described above. As a positive control, Alexa 488 lifetime was measured in the presence of a FRET acceptor (Cy3) in close proximity (19)
presenting the acceptor with a donkey anti-goat Cy3 labeled Ab, directed against the goat anti-mouse Alexa 488 secondary Ab used to visualize the anti-V5 monoclonal antibody (mAb). In the positive control, Alexa 488 lifetime is shortened to less than 1000 ps.
Toxicity assay
Toxicology was analyzed 24 h after transfection by measuring the release of adenylate kinase (AK) from damaged cells into the cultureTM media as described by the manufacture (ToxiLightTM, Cambrex, Bioscience Rockland, Inc., Rockland, ME, USA).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-synuclein molecule that had been tagged with small epitope tags on both the amino and carboxyl terminus, or singly tagged on either the amino or carboxyl terminus (Fig. 1A
). In all cases, the tagged molecules showed complete colocalization at the light level (Fig. 1A
). No aggregates are observed under these conditions. No differences between the constructs used were observed in the minimal degree of cellular toxicity induced by untagged -synuclein (15)
compared to control (empty vector). In particular, no increase in toxicity was observed for the tagged constructs of -synuclein (Fig. 1B
).
Fluorescence lifetime imaging is a method to monitor the presence of fluorescence resonance energy transfer (FRET), in which the proximity of two fluorophores is measured. If two fluorophores are within close proximity, the lifetime of the donor fluorophores is shortened in proportion to the relative distance between the fluorophores (19
, 20)
. This can be color-coded, so that pixels in which FRET is absent are pseudo-colored blue, whereas shortening of lifetimes are colored red (Fig. 2
). In the absence of an acceptor, the fluorescence lifetime of the donor Alexa 488 fluorophore (marking the amino terminus of -synuclein) is 
2200 ps (Fig. 2A
). When the carboxyl terminus of the molecule is labeled with an acceptor, Cy3, the lifetime of the donor fluorophores decreases in these experiments, (to 1400 ps), suggesting close proximity and a strong interaction between the amino and carboxyl termini of the molecule (Fig. 2B
). Since the 140 amino acid distance between the amino and carboxyl termini of -synuclein would likely be too large to support a FRET interaction if it was linearly configured, these data suggest that, within intact cells, the N and C terminal domains fold in a fashion to bring them relatively close to one another, and contrasts with observations of isolated -synuclein, which has a random coil conformation (22
23
24)
.
|
An alternative explanation for interaction of the amino and carboxyl termini in this experimental setting would be interaction of the amino terminus on one -synuclein molecule with the carboxyl terminus of -synuclein on another molecule. We reasoned that if this were the case, by cotransfecting doubly tagged -synuclein (HA-syn-V5) with untagged -synuclein we would dilute out "between" molecule interactions and diminish the observed FRET. Instead, when untagged -synuclein was cotransfected, we observed an even stronger interaction between the amino and carboxyl termini as reflected by a faster lifetime of the donor (1000 ps; Fig. 3
). We interpret this finding to mean that the original (1400 ps) signal was due to two components: a very strong interaction between the amino and carboxyl termini of the same molecule (demonstrated by an interaction on the order of 1000 ps) and possibly a weaker interaction between the amino terminus and carboxyl terminus of two different molecules.
|
To directly examine the postulated interaction between the amino terminus and carboxyl terminus of two different -synuclein molecules, we examined the interaction between N-terminally tagged -synuclein (HA-syn) and C-terminally tagged–-synuclein (syn-V5), where the tags were uniquely present on different molecules. Compared with the negative control (2200 ps), the FLIM assay revealed a close -synuclein–-synuclein interaction as shown by a diminution of the lifetime of donor fluorophore to 1500 ps (Fig. 3)
. This change in lifetime suggests that two -synuclein molecules are in quite close proximity. To further evaluate the specificity of this interaction, we also examined the interaction of -synuclein molecules using a 50:50 mix of HA-syn and V5-syn, or a 50:50 mix of syn-HA and syn-V5. In all cases, the -synuclein constructs completely colocalized at the light level (Fig. 1A-C
). However, in the latter two instances (N-terminus-N terminus or C-terminus-C-terminus) only a small change in lifetime could be detected, suggesting that neither the amino termini nor the carboxyl termini of -synuclein molecules come into close proximity (Fig. 3)
. These data show that the -synuclein interactions are specific for the amino-terminal domain and the carboxyl terminal domain and suggest that oligomerization occurs in an antiparallel fashion.
We next studied the biochemical characteristics of the synuclein-synuclein interaction observed by FRET. Coimmunoprecipitation of -synuclein, using either the HA- or V5-tag, failed to coimmunoprecipitate the corresponding form of -synuclein using Triton X-100, Nonidet P-40, or even mechanical homogenization of the cells. This suggests that the -synuclein–-synuclein interaction is dependent on being in an intact cell and is detergent-sensitive and thus unlikely to be detected by classical biochemical techniques (Fig. 4
).
|
We had previously shown that -synuclein overexpression leads to cell death that can be prevented by coexpression of Hsp70 in H4 neuroglioma cells, but the mechanism of Hsp70 neuroprotection is unknown. Similar chaperone-mediated protection has been observed in multiple systems, including drosophila (13
14
15
16
17)
. Interestingly, Hsp70 protects against toxicity without changing the morphological qualities of -synuclein aggregates in drosophila. We postulated that Hsp70 impacts -synuclein conformation rather than aggregation phenomena per se. We tested this hypothesis by evaluating the effect of coexpression of Hsp70 on the proximity of the amino and carboxyl terminus domains of -synuclein. The addition of Hsp70 by cotransfection altered the distance of the amino and carboxyl terminal domains within a molecule as represented by a change in lifetime from 1000 ps to greater than 1700 ps (Fig. 5
), suggesting a rather marked conformational change. By sharp contrast, cotransfection with Hsp70 did not affect the interaction of the amino terminus and carboxyl terminus with one another when the tags were on different molecules.
|
These results suggest that Hsp70 affects the folding of -synuclein, perhaps favoring a conformation in which the amino- and carboxy-termini are less closely packed. No change in subsequent -synuclein molecular interactions is detected. This result would be consistent with the observations that Hsp70 overexpression reduces -synuclein toxicity without changing the characteristics of the aggregates in flies and H4 cells, and also that Hsp70 diminishes Triton X-100 insoluble forms of -synuclein without changing macroscopic aggregates in -synuclein overexpressing mice (15)
.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-synuclein in intact cells. In vitro data suggest that aggregate formation initiates with the oligomerization of partially folded monomers (24
, 25)
, which is increased by oxidative modification of -synuclein (26)
. Furthermore, -synuclein oligomers may represent the toxic species in the neurodegenerative process (27
, 28)
, and it has been shown that oligomeric species of -synuclein are degraded by the lysosome (29)
. In this study, we can detect -synuclein oligomers within intact cells using FLIM. Our results detect a new amino-terminus carboxyterminus interaction between -synuclein molecules, which appears to be present only in an intact cellular environment and is extremely sensitive to detergents. We propose either an antiparallel interaction of at least two -synuclein molecules or a chain-like conformation of several -synuclein molecules with the amino terminus of one molecule in close proximity to the carboxy-terminus of a second molecule, that itself interacts with the amino terminus of a third -synuclein molecule, and so on.
In vitro studies using recombinant -synuclein proteins in artificial environments have proposed a model of -synuclein interactions with model lipid membranes (3
, 5)
. The conformation of -synuclein directly influences its membrane binding ability (30)
and vice versa (4
, 31)
. Even more, -synuclein regulates lipid composition of the membrane (32)
. Although the current data do not directly address synuclein-membrane interactions, taken together with these in vitro studies we propose a model in which membrane binding is associated with oligomer formation and that the observed molecule interaction occurs along membrane surfaces, with the amphipathic amino terminus and hydrophobic NAC domain providing interactions with the membrane. Because the membrane interaction of -synuclein is a crucial step for both the localization of -synuclein in synaptic membranes, which might have an impact on its physiological function (33
34
35)
and toxic modifications of -synuclein (36)
, future studies on -synuclein oligomer formation in intact cells will help understanding and deciphering functional and toxic aspects of -synuclein.
Moreover, there appears to be a tight interaction between the amino terminus and carboxyl terminus of an individual -synuclein molecule; this causes a closed conformation in which the highly acidic carboxylterminal domain comes into closer proximity with the amphipathic amino terminus. Cytosolic monomer exists mainly in randomly unfolded conformation but may aggregate and induce fibrill formation by enhancing a partial folded ß-sheet conformation (22
23
24)
. Recently, it was shown that native -synuclein adopts an ensemble of conformations that are stabilized by long-range interactions as analyzed by NMR techniques (37
38
39)
. It has been proposed from in vitro studies using recombinant -synuclein molecules that the N- and the C-termini shield the highly amyloidogenic, non-Aß component (NAC) domain from inducing aggregation of the protein (37
, 39
40
41
42)
. Alterations in the conformation of -synuclein may thus expose the NAC domain (43)
inducing -synuclein’s amyloidogenic character (44)
. Our FLIM assays are consistent with these ideas.
Hsp70 is known to protect against -synuclein toxicity (13
, 15)
, but its mechanism of action is unknown. Hsp70 is known to mediate protein refolding as a chaperone and also to act in concert with other proteins, including carboxy-terminal Hsp70 interacting protein (CHIP), as a mediator of protein degradation (14
, 45)
. Our previous studies suggest that Hsp70 can mediate degradation of -synuclein under some circumstances (14
, 15
, 18)
. The current studies using FRET-based techniques, however, support the idea that Hsp70 can also stabilize a specific conformation of -synuclein. We suggest that Hsp70 overexpression leads to a stabilization of -synuclein in an open conformation. In the model proposed (Fig. 6
), this conformational change would stabilize -synuclein–-synuclein intermolecular interactions, decreasing potentially toxic NAC-NAC interactions. We interpret this to be the underlying phenomena that leads to increased -synuclein solubility in mild detergents in the presence of Hsp70 (15)
. Biophysical methods examining isolated proteins similarly suggest a preferential direct interaction of Hsp70 with specific conformations of "prefibrillar" -synuclein (46)
, which may well correspond to the folded -synuclein conformation that gives rise to the strongest FRET signal in our experiments, and which is altered by coexpression of Hsp70.
|
In summary, we have developed an assay that can detect protein domain-protein domain interactions within -synuclein in intact cells. Under the conditions we explored, the cells do not form any aggregates that can be detected even by confocal microscopy. However, the FLIM assay can clearly detect submicroscopic oligomeric forms of -synuclein–-synuclein interacting molecules. The FLIM assay, as we have developed it, provides relative proximities rather than absolute distances, as it depends on indirect interactions of the fluorophores with the epitope tags. In addition, it is important to note that the conformations and interactions we observe in cells are likely modified by endogenous proteins, such as endogenous -synuclein itself, CHIP, and other -synuclein interacting proteins (45
, 47)
, as well as the local microenvironment and membranes, perhaps leading to results that differ from -synuclein studied in isolation or in model systems. Nonetheless, the FLIM data strongly suggest that a head-to-tail arrangement of -synuclein oligomerization can occur within cells, revealing a new metastable conformation of -synuclein. We can demonstrate that Hsp70 alters the conformation of -synuclein under these conditions, demonstrating in principle a molecular level readout for therapeutics that alter -synuclein folding. This assay may, therefore, prove useful in numerous circumstances where intracellular protein conformational changes, or protein self-aggregation in oligomeric forms, is postulated to be a critical element in pathophysiology, including accumulation of tau in neurofibrillary tangles in Alzheimer’s disease, prion proteins in prion diseases, and polyglutamine containing proteins in trinucleotide repeat diseases.
|
ACKNOWLEDGMENTS |
|---|
Received for publication November 4, 2005. Accepted for publication May 8, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. Jiang, L.-w. Ko, K. R. Jansen, T. E. Golde, and S.-H. Yen Using leucine zipper to facilitate {alpha}-synuclein assembly FASEB J, September 1, 2008; 22(9): 3165 - 3174. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. McLear, D. Lebrecht, A. Messer, and W. J. Wolfgang Combinational approach of intrabody with enhanced Hsp70 expression addresses multiple pathologies in a fly model of Huntington's disease FASEB J, June 1, 2008; 22(6): 2003 - 2011. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
