Molecular Biology Reports

, Volume 37, Issue 7, pp 3183–3192

Molecular interaction of α-synuclein with tubulin influences on the polymerization of microtubule in vitro and structure of microtubule in cells

Authors

  • R. M. Zhou
    • State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention
  • Y. X. Huang
    • State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention
  • X. L. Li
    • State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention
  • C. Chen
    • State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention
  • Q. Shi
    • State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention
  • G. R. Wang
    • State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention
  • C. Tian
    • State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention
  • Z. Y. Wang
    • State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention
  • Y. Y. Jing
    • State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention
  • C. Gao
    • State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention
    • State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention
Article

DOI: 10.1007/s11033-009-9899-2

Cite this article as:
Zhou, R.M., Huang, Y.X., Li, X.L. et al. Mol Biol Rep (2010) 37: 3183. doi:10.1007/s11033-009-9899-2

Abstract

Microtubule dynamics is essential for many vital cellular processes such as in intracellular transport, metabolism, and cell division. Evidences demonstrate that α-synuclein may associate with microtubular cytoskeleton and its major component, tubulin. In the present study, the molecular interaction between α-synuclein and tubulin was confirmed by GST pull-down assay and co-immunoprecipitation. The interacting regions within α-synuclein with tubulin were mapped at the residues 60–100 of α-synuclein that is critical for the binding activity with tubulin. Microtubule assembly assays and sedimentation tests demonstrated that α-synuclein influenced the polymerization of tubulin in vitro, revealing an interacting region-dependent feature. Confocal microscopy detected that exposures of α-synuclein proteins inhibited microtubule formation in the cultured cells, with a length-dependent phenomenon. Our data highlight a potential role of α-synuclein in regulating the microtubule dynamics in neurons. The association of α-synuclein with tubulin may further provide insight into the biological and pathophysiological function of synuclein.

Keywords

α-SynucleinTubulinMolecular interactionMicrotubule dynamicsSynucleinopathies

Introduction

α-Synuclein (SNCP) is a kind of presynaptic proteins, distributing widely in brain tissues [1]. Increasing evidences suggest that α-synuclein is a common pathogenic molecule in several neurodegenerative disorders, in which deposition of α-synuclein as fibrillary aggregates in neurons or glial cells is a hallmark lesion [2]. Those disorders include Parkinson’s disease [3], dementia with Lewy bodies [4, 5], Lewy body variant of Alzheimer’s disease [6] and multiple system atrophy [7], collectively referred to as synucleinopathies [8]. However, the physiological function of α-synuclein remains to be largely unclear. It was reported that α-synuclein is functional microtubule-associated protein (MAP), and that tubulin co-localizes with α-synuclein in Lewy bodies and influences the formation of α-synuclein aggregation [9, 10]. Thus, the association of α-synuclein with microtubular network may be related to physiological or pathologic activities of α-synuclein in the neurons.

Microtubules are cellular structures that play a central role in intracellular transport, metabolism, and cell division. A wide range of chemicals or proteins are able to interrupt microtubule function, which can be categorized as microtubule stabilizers and microtubule depolymerizers [11, 12]. Our previous studies have demonstrated that PrP, a cellular protein related with transmissible spongiform encephalopathy or prion disease, affects the formation of microtubule through interacting with tubulin [13, 14]. Interference with microtubule dynamics leads to mitotic arrest and initiation of apoptosis. Previous studies reported that tubulin plays a protagonist role in α-synuclein oligomer and fibril formation [9], and that α-synuclein can facilitate tubulin polymerization [10]. However, there was a different report that α-synuclein inhibits tubulin polymerization [15]. The discrepancy of the effectiveness of α-synuclein on microtubule indicates a need of more experimental data.

In this study, we propose the molecular evidence of recombinant human α-synuclein interacting with native tubulin. The residues 60–100 within α-synuclein are responsible for recognizing tubulin, except for the C-terminus of α-synuclein (residues 96–102 and 131–140) reported previously [10]. Transiently expressed α-synuclein co-locates well with the endogenous microtubules in HeLa cells. We also confirm that the full-length α-synuclein, as well as the segments for interacting with tubulin, have marked influence on the polymerization of tubulin in vitro and the structure of microtubule in cultured cells.

Materials and methods

Plasmid construction and protein purification

Total RNA of a human neuroblastoma cell line SH-SY5Y was extracted and the single-stranded cDNA was synthesized. With a PCR technique, the cDNA sequence encoding human α-synuclein (SNCP) was amplified with the primers Hu-SNCA-F (5′-GTCGACATGGATGTATTCATGAAAGG-3′, with a SalI site underlined) and Hu-SNCA-B (5′-GGATCCTTAGGCTTCAGGTTCGTAGTA-3′, with a BamHI site underlined). After verified by DNA sequencing, the full-length human SNCP cDNA was cloned into T-vector (pT-SNCP1-140) and subcloned into a GST-fusion protein expression vector pQE-30-GST, designated pGST-SNCP1-140. To construct the mammalian expressing plasmid encoding the full-length SNCP, SNCP sequence was released from pT-SNCP1-140 and subcloned into vector pcDNA3.1, generating recombinant plasmid pcDNA-SNCP1-140. Various lengths of sequences encoding truncated mutants of SNCP, including SNCP1-60, SNCP61-100, SNCP31-100 and SNCP1-100, were generated by PCR using the plasmid pT-SNCP1-140 as the templates. The PCR product was subsequently inserted into a prokaryotic expressing vector pGEX-2T, generating recombinant plasmids pGST-SNCP1-60, pGST-SNCP61-100, pGST-SNCP31-100 and pGST-SNCP1-100.

The recombinant prokaryotic proteins tagged with GST were bacterially expressed in E. coli M15 and purified with Glutathione Sepharose 4B Agarose (Amersham Pharmacia) according to the protocol described in our previous study [13]. Each purity of the protein was verified by SDS-PAGE and immunoblotting, showing to be >95%.

The commercial rabbit tubulin protein was purchased from Sigma. Native hamsters’ tubulin was purified from fresh prepared hamster brains according to the procedure of microtubules preparation kit (Sigma). Before experiments tubulin preparations were thawed from −80°C and centrifuged at 20,000×g at 4°C for 30 min. The obtained supernatants were used in subsequent tests.

Antibodies

Anti-α-tubulin monoclonal antibody (mAb) and anti-SNCP polyclonal antibody (pAb) were purchased from Santa Cruz Biotechnology. Anti-His, anti-GST mAb and anti-α-tubulin pAb were purchased from Tiangen Biotechnology, China.

Preparation of brain tissue homogenates

Normal hamsters’ brain tissues were homogenized in nine volumes (10% w/v) of lysis buffer (100 mM NaCl, 10 mM EDTA, 0.5% Nonidet P-40, 0.5% sodium deoxycholate in 10 mM Tris-Cl, pH 7.4). The brain homogenates were centrifuged at 20,000×g for 30 min, and the supernatants were collected for subsequent experiments.

Pull-down assay

To identify the potential interaction between SNCP and tubulin, 5 μM various SNCP proteins was incubated with 5 μM tubulin or 0.1 ml of the supernatant of normal hamster brain homogenates in 500 μl binding solution containing 20 mM Tris-Cl, 200 mM NaCl, 10 mM aprotinin, pH 8.0, at 4°C for 4 h, while equal amount of GST protein was used as control. Ten microliter of Glutathione Sepharose 4B beads were added to the reaction solution and incubated at 37°C for 30 min with end-over-end mixing. After centrifugation at 2,000 rpm for 2 min, the supernatants were discarded and beads were washed with 500 μl washing buffer (50 mM Tris-Cl, 300 mM NaCl, pH 8.0) for three times. The complexes were separated on 15% SDS-PAGE and transferred to nitrocellulose filter blocked with 5% nonfat milk in PBST (135 mM NaCl, 1.3 mM KCl, 3.2 mM Na2HPO4, 0.5 mM KH2PO4, 0.05% Tween 20, pH 7.4). Tubulin reactive band was detected with 1:4000 anti-α-tubulin mAb as the primary antibody and 1:4000 HRP-conjugated anti-mouse IgG (Santa Cruz Biotechnology) as the second one, using ECL method (Perkin–Elmer).

Co-immunoprecipitation

Five micromolar of various SNCP protein were separately incubated with 5 μM tubulin or 0.1 ml of the supernatant of normal hamster brain homogenates in 500 μl binding solution containing 20 mM Tris-Cl, 200 mM NaCl, 10 mM aprotinin, pH 8.0, at 4°C for 4 h, while equal amount of GST protein was employed as control in parallel. After incubated with anti-GST mAb (Tiangen Biotechnology) for 2 h, 10 μl Protein G Sepharose (Roche) was applied into the reactions and incubated for further 2 h. The Sepharose beads were precipitated at 2,000 rpm for 5 min and washed with 500 μl washing buffer (50 mM Tris-Cl, 300 mM NaCl, pH 8.0) for three times. The bound complexes were separated by 15% SDS-PAGE and transferred to nitrocellulose membranes. The bound tubulin was detected by anti-α-tubulin pAb in Western blot.

Microtubule assembly assay

Microtubule was assembled in vitro from the purified tubulin at 3 mg/ml in the polymerization buffer contained 80 mM PIPES (pH 6.9), 0.5 mM MgCl2, 1 mM GTP, 5% Glycerol, and the turbidity of the solution was monitored continuously after mixture based on the protocol described previously [13]. The assay ran for 30 min while measuring the turbidity every 5 min at 340 nm in a spectrophotometer.

Sedimentation experiments

Samples containing 3 mg/ml tubulin were incubated for 30 min and subsequently centrifuged for 10 min at 14,000×g. The pellets were resuspended in a volume of deionized water equal to the volume of the supernatants and analyzed by 15% SDS-PAGE.

Cell culture, transfection and immunofluorescence microscopy

HeLa cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal calf serum. Cells were transiently transfected with the pcDNA-SNCP1-140 vector using Lipofectamin 2000 Reagent (Qiagen) according to the manufacture’s protocol. One day after transfection, cells were fixed with 4% paraformaldehyde and 10% methanol in PBS for 30 min at room temperature. After blocking with 5% fetal bovine serum, 0.1% Triton X-100 in PBS for 30 min, cells were incubated with anti-α-synuclein (1:400), anti-α-tubulin (1:500) antibodies overnight at 4°C. After washing with PBS, cells were stained with Alexa Fluor 546 conjugated anti-rabbit IgG (1:500, Invitrogen) for α-synuclein or Alexa Fluor 488 conjugated anti-mouse IgG (1:500, Invitrogen) for α-tubulin for 1 h at room temperature. Fluorescently stained cells were analyzed using a confocal laser scanning microscope (Bio-Rad).

Formation of microtubule in the cultured cells

To see the influence on the formation of microtubule in the cultured cells, 10 μM of each SNCP was added to the cells for 24 h. In parallel, 10 μM GST protein and 10 μM colchicines were employed as control. After incubation, cells were extensively washed and fixed in 4% paraformaldehyde. In order to allow antibodies to penetrate the cell membrane, monoclonal tubulin antibodies and Alexa488 (green) Fluor-conjugated anti-rabbit IgG was added with 0.1% Triton X-100. Fluorescently stained cells were analyzed using a confocal laser scanning microscope (Bio-Rad).

Cell viability determination

HeLa cells were plated as normal in 96-well trays and maintained in DMSM containing 10% BSA. Various materials were applied directly to the wells according to different experiments. Viability was assessed by the conversion of 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT, Sigma, USA) to a formazan product. Absorbance was measured at 492 nm in a spectrophotometer.

Results

Agarose gel revealed that the product of human α-synuclein specific RT-PCR was about 420 bp (data not shown). To obtain the full-length and truncated human SNCP proteins, various recombinant SNCP were constructed (Fig. 1a) and purified from the transformed E. coli. SDS-PAGE showed that the purified recombinant GST-SNCP proteins and purified hamsters’ tubulin at the positions as expected, respectively (Fig. 1b, c).
https://static-content.springer.com/image/art%3A10.1007%2Fs11033-009-9899-2/MediaObjects/11033_2009_9899_Fig1_HTML.gif
Fig. 1

Protein purification of various recombinant SNCP and native tubulin. a Schematic structure of the full-length SNCP and various SNCP deletion mutants, including SNCP1-60, SNCP61-100, SNCP31-100 and SNCP1-100. b 15% SDS-PAGE assay of the various purified recombinant SNCP proteins. Lane 1: GST-SNCP1-60; lane 2: GST-SNCP61-100; lane 3: GST-SNCP31-100; lane 4: GST-SNCP1-100; lane 5: GST-SNCP1-140. Protein molecular markers are shown on the left. c 15% SDS-PAGE assay of the purified tubulin from hamsters. Protein molecular markers are shown on the right

Recombinant full-length α-synuclein interacts with tubulin in vitro

To address the possible interaction between the full-length α-synuclein and tubulin, the purified GST-SNCP1-140 fusion protein was mixed with the commercial tubulin or 10% hamsters’ brains homogenate in GST pull-down assay, while the recombinant GST protein was used as control. Western blot of the eluted products with anti-α-tubulin mAb demonstrated clear reactive bands in the preparation of GST-SNCP1-140 and the commercially purified rabbit’s tubulin (Fig. 2a, lane 2), as well as in that of GST-SNCP1-140 and 10% hamsters’ brains homogenate (Fig. 2b, lane 1), whereas no such band in the preparation of GST with the purified rabbit’s tubulin (Fig. 2a, lane 1) or with 10% hamsters’ brains homogenate (Fig. 2b, lane 2). To confirm the observations in GST pull-down assay, GST-SNCP1-140 and tubulin or 10% hamsters’ brains homogenate were mixed and the protein complexes were precipitated by anti-α-tubulin pAb (Fig. 2c) or anti-GST mAb (Fig. 2a, d). Western blot of the eluted products showed a reactive band in the reaction of GST-SNCP1-140 with the purified rabbit’s tubulin (Fig. 2a, lane 2, 5), as well as in that of GST-SNCP1-140 with 10% hamsters’ brains homogenate precipitated by anti-α-tubulin pAb (2C, lane 1) and by anti-GST mAb (2D, lane 1), while not in the mixtures of GST with purified rabbit’s tubulin (Fig. 2a, lane 6) and that of GST with 10% hamsters’ brains homogenate (Fig. 2c, lane 2; Fig. 2d, lane 2). It suggests that the full-length human α-synuclein can form complex with purified tubulin from rabbit brian or tubulin in hamster brain homogenate.
https://static-content.springer.com/image/art%3A10.1007%2Fs11033-009-9899-2/MediaObjects/11033_2009_9899_Fig2_HTML.gif
Fig. 2

Molecular interaction between recombinant SNCP1-140 and native tubulin. a GST pull-down assay and co-immunoprecipitation of GST-SNCP1-140 and the purified rabbit’s tubulin. Lanes 1–3 were GST pull-down assays and lanes 5–7 were co-immunoprecipitation. Lane 1: GST protein and the purified tubulin; lane 2: GST-SNCP1-140 and the purified tubulin; lane 3: The purified tubulin only. The SNCP–tubulin complexes were precipitated with glutathione agarose beads. The bound tubulin was detected by anti-α-tubulin pAb. Lane 4: same amount of purified tubulin was directly loaded as control of Western blot. Lane 5: GST-SNCP1-140 and the purified tubulin; lane 6: GST protein and the purified tubulin; lane 7: the purified tubulin only. The SNCP–tubulin complexes were precipitated by anti-GST mAb and the bound α-tubulin was detected by anti-α-tubulin pAb. b GST pull-down assays of GST-SNCP1-140 and 10% hamsters’ brains homogenate. Lane 1: GST-SNCP1-140 and 10% hamsters’ brains homogenate; lane 2: GST protein and 10% hamsters’ brains homogenate; lane 3: 10% hamsters’ brains homogenate only; lane 4: GST-SNCP1-140 only; lane 5: same amount of 10% hamsters’ brains homogenate was directly loaded as control of Western blot. The SNCP–tubulin complexes were precipitated with glutathione agarose beads. The bound tubulin was detected by anti-α-tubulin mAb. c Co-immunoprecipitation of GST-SNCP1-140 and 10% hamsters’ brains homogenate by anti-α-tubulin pAb. Lane 1: GST-SNCP1-140 and 10% hamsters’ brains homogenate; lane 2: GST protein and 10% hamsters’ brains homogenate; lane 3: 10% hamsters’ brains homogenate only; lane 4: same amount of 10% hamsters’ brains homogenate was directly loaded as control of Western blot; lane 5: GST-SNCP1-140 only. The SNCP–tubulin complexes were precipitated by anti-α-tubulin pAb and the bound GST-SNCP was detected by anti-GST mAb. d Co-immunoprecipitation of GST-SNCP1-140 and 10% hamsters’ brains homogenate by anti-GST mAb. Lane 1: GST-SNCP1-140 and 10% hamsters’ brains homogenate; lane 2: GST protein and 10% hamsters’ brains homogenate; lane 3: 10% hamsters’ brains homogenate only; lane 4: same amount of 10% hamsters’ brains homogenate was directly loaded as control of Western blot. The SNCP–tubulin complexes were precipitated by anti-GST mAb and the bound α-tubulin was detected by anti-α-tubulin pAb. Protein molecular markers are shown on the left

α-Synuclein co-locates with microtubule in transiently transfected HeLa cells

To test the association of α-synuclein with microtubules in cells, recombinant plasmid pcDNA-SNCP1-140 was transiently transfected into HeLa cells and co-locations of α-synuclein and microtubule were monitored by a double-labeling immunofluorescence method with confocal microscopy 24 h after transfection. Remarkable red signal was observed in the cells receiving pcDNA-SNCP1-140 after stained by polyclonal α-synuclein antibody (Fig. 3a, top right panel), while no signal in the cells receiving pcDNA3.1 (Fig. 3b, top right panel), which indicates an extremely low expression of endogenous α-synuclein in HeLa cells. Both cells receiving pcDNA-SNCP1-140 and pcDNA3.1 showed strong green fluoresces after stained by monoclonal anti-tubulin (Fig. 3a, b, top left panels), suggesting detectable endogenous expression of tubulin in HeLa cells. After merging the images of tubulin and α-synuclein, marked yellow fluoresces were observed in the cells transiently transfected with pcDNA-SNCP1-140 (Fig. 3a, bottom right panel). It indicates that the expressed α-synuclein is co-located with microtubules in cells.
https://static-content.springer.com/image/art%3A10.1007%2Fs11033-009-9899-2/MediaObjects/11033_2009_9899_Fig3_HTML.jpg
Fig. 3

Co-location of the expressed SNCP and cellular microtubule in HeLa cells under confocal microscopy. HeLa cells were transiently transfected with the plasmid pcDNA-SNCP1-140 (a) or pcDNA3.1 (b) and analyzed 24 h after transfection. The expression of cellular microtubule (top left), SNCP (top right) and cell nucleis (bottom left) were stained by individual antibodies or reagent, respectively. The merged images of expressed SNCP and cellular microtubule are shown in the panel of bottom right

The interacting domains of α-synuclein with tubulin are located at SNCP spanning residues 60–100

To map the possible position within α-synuclein interacting with tubulin, a series of truncated SNCPs were incubated with tubulin, respectively. GST pull-down assays identified clear protein complexes in the reactions of GST-SNCP61-100 (Fig. 4a, lane 5), GST-SNCP31-100 (lane 6) and GST-SNCP1-100 (lane 7), but not in that of GST-SNCP1-60 (lane 4). Co-immunoprecipitations of anti-α-tubulin pAb (Fig. 3b) and anti-α-synuclein mAb (Fig. 3c) revealed also the SNCP–tubulin protein complexes in the preparations of GST-SNCP61-100 (lane 5), GST-SNCP31-100 (lane 6) and GST-SNCP1-100 (lane 7), whereas not in that of GST-SNCP1-60 (lane 4). It implies that the regions within α-synuclein responsible for the interaction with tubulin are likely located at SNCP spanning residues 60–100.
https://static-content.springer.com/image/art%3A10.1007%2Fs11033-009-9899-2/MediaObjects/11033_2009_9899_Fig4_HTML.gif
Fig. 4

Analyses of the molecular interactions of various recombinant SNCP proteins with the purified tubulin. a GST pull-down assays. Lane 1: GST protein and tubulin; lane 2: tubulin only; lane 3: same amount of tubulin was directly loaded as control of Western blot; lane 4: GST-SNCP1-60 and tubulin; lane 5: GST-SNCP61-100 and tubulin; lane 6: GST-SNCP31-100 and tubulin; lane 7: GST-SNCP1-100 and tubulin. The SNCP–tubulin complexes were precipitated with glutathione agarose beads. The bound tubulin was detected by anti-α-tubulin mAb. b Co-immunoprecipitation by anti-α-tubulin pAb. Lane 1: GST protein and tubulin; lane 2: tubulin only; lane 3: same amount of GST-SNCP1-140 was directly loaded as control of Western blot; lane 4: GST-SNCP1-140 only; lane 5: GST-SNCP1-60 and tubulin; lane 6: GST-SNCP61-100 and tubulin; lane 7: GST-SNCP31-100 and tubulin; lane 8: GST-SNCP1-100 and tubulin. The SNCP–tubulin complexes were precipitated by anti-α-tubulin pAb and the bound GST-SNCP was detected by anti-GST mAb. c Co-immunoprecipitation by anti-GST mAb. Lane 1: His-GST protein and tubulin; lane 2: tubulin only; lane 3: Same amount of tubulin was directly loaded as control of Western blot; lane 4: GST-SNCP1-60 and tubulin; lane 5: GST-SNCP61-100 and tubulin; lane 6: GST-SNCP31-100 and tubulin; lane 7: GST-SNCP1-100 and tubulin. The SNCP–tubulin complexes were precipitated by anti-GST mAb and the bound α-tubulin was detected by anti-α-tubulin pAb. Protein molecular markers are shown on the left

α-Synuclein increases turbidity of tubulin solution and induces sedimentation of tubulin

To test whether the interaction between α-synuclein and tubulin influenced the assembling of microtubules from tubulin in vitro, a microtubule assembly assay was performed by standard light scattering measurements at 340 nm [13]. Various SNCP proteins were mixed with tubulin in a polymerization buffer containing GTP and magnesium ions, while the preparations with His-GST were used as control. Figure 5a revealed that the turbidity in the preparation of tubulin alone increased along with the incubation times, while addition of GST did not influence the curve. When the full-length SNCP was employed into the solution, the turbidity of the solution increased remarkably and rapidly at various concentrations (molar ratios to tubulin from 1:32 to 1:8), showing an obvious dose-dependent manner (Fig. 5a). Meanwhile, addition of α-synuclein at the highest concentration into the buffer without tubulin did not change the solution turbidity within 30 min (data not shown).
https://static-content.springer.com/image/art%3A10.1007%2Fs11033-009-9899-2/MediaObjects/11033_2009_9899_Fig5_HTML.gif
Fig. 5

The influences of the full-length and various truncated SNCP on the turbidity values of tubulin solutions. a Incubations of tubulin with various amounts of GST-SNCP1-140 (molar ratios of tubulin to SNCP ranged from 8:1 to 32:1), as well as fixed amount of GST (molar ratio of tubulin to His-GST was 8:1). b Incubations of tubulin with the full-length and various truncated SNCP (molar ratio of tubulin to various SNCP was 8:1). The concentration of tubulin was constant at 3 mg/ml in each assay. c Comparison of the turbidities of tubulin solutions after incubated with various SNCP (molar ratio of tubulin to PrP was 8:1) for 30 min. The average values were calculated from three independent tests and presented with as mean ± SD

To see the influence of different SNCP truncated peptides on the changes of turbidities of tubulin solution, various recombinant SNCP proteins were subjected into the solutions (molar ratio of tubulin to various SNCP was 8:1), while the preparations with GST were used as control. GST-SNCP1-140, GST-SNCP61-100, GST-SNCP31-100 and GST-SNCP1-100 showed significant influence on the solution turbidities compared with the control, while the solution turbidity of GST-SNCP1-60 have no significant change (Fig. 5b, c). It suggests that the region influencing the turbidity of tubulin solution is likely located at SNCP spanning residues 60–100.

To highlight the characteristics of tubulin in the solutions co-incubated with various SNCP a sedimentation test was performed. A relative higher amount of tubulin (3 mg/ml) that was believed for favoring microtubule formation was used and the molar ratio of tubulin and SNCP was fixed at 8:1. After centrifugation, a small portion of tubulin presented in pellet of the preparation containing tubulin alone (Fig. 6a, lane 7 and 8). Incubations with GST-SNCP1-60 and GST did not change the distribution of tubulin, while the input GST-SNCP1-60 and GST presented only in the supernatant fractions (Fig. 6a, lane 3 and 11, lane 1 and 9), implying that SNCP1-60 did not influence the dynamics of microtubule assembly as GST. As expected, incubations with GST-SNCP1-140 (lane 2 and 10), GST-SNCP61-100 (lane 4 and 12), GST-SNCP31-100 (lane 5 and 13) and GST-SNCP1-100 (lane 6 and 14) increased markedly the proportions of tubulin in the pellets, while almost all of the input SNCPs were detected in the pellet. Analyses of the ratios signal intensities between supernatant and pellet each preparation demonstrated apparent increases of the signal intensities in fractions of pellets of GST-SNCP1-140, GST-SNCP61-100, GST-SNCP31-100 and GST-SNCP1-100 (Fig. 6b), highly implying that SNCP spanning residues 60–100 affects the assembling process of microtubules in vitro.
https://static-content.springer.com/image/art%3A10.1007%2Fs11033-009-9899-2/MediaObjects/11033_2009_9899_Fig6_HTML.gif
Fig. 6

Evaluation of the sedimentation of tubulin in the presences of full-length and various truncated SNCP with SDS-PAGE. a Sedimentation of tubulin in the condition of favoring microtubule formation (3 mg/ml). Tubulin was incubated with GST (lane 1 and 9), GST-SNCP1-140 (lane 2 and 10), GST-SNCP1-60 (lane 3 and 11), GST-SNCP61-100 (lane 4 and 12), GST-SNCP31-100 (lane 5 and 13) and GST-SNCP1-100 (lane 6 and 14), at an 8:1 (tubulin: SNCP or His-GST) molar ratio. Tubulin alone (lane 7 and 8) was processed with the same protocol in the polymerization buffer as controls. S and P represented as supernatant and pellet after centrifuging, respectively. b Comparative analyses of sedimentation of tubulin in the presences of full-length and various truncated SNCP by densitometry. Tubulin signals on a were quantified with computer-assisted software Image TotalTech. The total signal intensities of tubulin in supernatant and pellet were summated and defined as 100%. The percentages of each sample were calculated and are given as ratios of the signal intensity. The average values were calculated from three independent tests and presented with as mean ± SD

α-Synuclein affects the formation of microtubules in the culture cells

To identify the possible role of α-synuclein on the structure of cellular microtubule, the cultured HeLa cells were exposed to various SNCP for 24 h and the integrity of the cellular microtubule network was examined by immunofluorescent staining. Confocal microscopy showed that the cellular microtubule network in the cells receiving GST (Fig. 7b) was quite similar as that in the mock cells (panel a), whereas severe disruption of microtubule structure were observed in the cells treated with colchicine, a microtubule-disrupting agent (panel c). Exposure of SNCP1-140 destroyed the fibril-like structure of microtubule instead of small dot fluoresces in cytoplasm (panel d). As expected, treatment of SNCP1-60 did not change the structure of cellular microtubule (panel e), while treatment of SNCP1-100 caused clear, but not as severe as that of SNCP1-140, disruption of microtubule (panel f). Interestingly, contrary to the remarkable influences on the turbidities of tubulin solutions, addition of SNCP31-100 (panel g) and SNCP61-100 (panel h) seemed not to affect the structures of microtubules under this experimental condition. These results indicate that introduction of α-synuclein into cells damages the fibril-like structure of microtubule, largely depending on its binding capacity with tubulin.
https://static-content.springer.com/image/art%3A10.1007%2Fs11033-009-9899-2/MediaObjects/11033_2009_9899_Fig7_HTML.gif
Fig. 7

Morphological analyses of the influences of full-length and various truncated SNCP on the structures of microtubule in HeLa cells by confocal microscopy. Cells were exposed to the different recombinant proteins for 24 h. a Cultured HeLa cell; b GST; c colchicines; d SNCP-FL; e SNCP1-60; f SNCP1-100; g SNCP31-100; and h SNCP61-100, respectively. The final concentration of colchicines, GST and various SNCP are all 10 μM

α-Synuclein affects the cell viability

To see the influence of α-synuclein on the cell viability, various recombinant SNCPs were introduced into the cultured HeLa cells and the cell viabilities were measured by MTT assays 12, 24, 36 and 48 h after exposure. Generally, the cell viabilities reached to the top at 24 h and maintained at high level 36 h after exposure. The OD values of the cells exposed with the full-length α-synuclein (SNCP1-140) were markedly lower than that of mock cells and the cells received GST in the assays of 12, 24 and 36 after exposure (Fig. 8). However, the other SNCPs did not induce significantly decreased OD values. In the preparations of 48 h, all SNCPs caused similar cell viabilities that were lower than GST and mock (Fig. 8). It seems that the cytotoxic effect of α-synuclein is length-dependent.
https://static-content.springer.com/image/art%3A10.1007%2Fs11033-009-9899-2/MediaObjects/11033_2009_9899_Fig8_HTML.gif
Fig. 8

MTT assays for the viabilities of cultured cells exposed to full-length and various truncated SNCP. Cells were measured 12 (a), 24 (b), 36 (c) and 48 (d) h post-exposure. The average OD values from three independent tests were showed vertically. The final concentrations of various recombinant SNCPs were 25 and 50 µg/ml, respectively

Discussion

In this study co-immunoprecipitation and pull-down assay provide reliable evidence again that α-synuclein interacts with both the native tubulin presenting in 10% hamsters’ brains homogenate and the purified tubulin from rabbit brains in vitro. α-synuclein is the major constituent of pathological intracellular inclusion bodies, which is a common feature of several neurodegenerative diseases. In Lewy body, aggregation and deposits of α-synuclein usually co-localize with tubulin [9, 16, 17]. The interaction data of α-synuclein and tubulin supply credible molecular basis for co-deposit of these two proteins in the pathological intracellular inclusion bodies. In addition, well co-localization of endogenous microtubule and transiently expressed α-synuclein in cells also implies that accumulation of α-synuclein in cytoplasm may easily bind to cellular microtubule.

The residues 61–95 of α-synuclein, also known as non-Aβ component region, is believed to by itself highly amyloidogenic, and residues 31–109 of α-synuclein may represent the core unit of α-synuclein filaments, which contributes to the structural stability of these filaments [18, 19]. Previous study has proposed that the potential binding sites of α-synuclein to tubulin locate at its C-terminal region, between residues 96 and 102, as well as 131 and 140 [10]. Our study confirms that the segment of α-synuclein spanning residues 60–100 also has remarkable binding capacity with native tubulin. Unfortunately, the expression of segment of SNCP101-140 is not successful, which prohibits us to carry out further assay. Existences of several binding sites for tubulin within α-synuclein may somehow reflect a relative strong interference between two proteins.

Although α-synuclein segments of residues 61–100 and 31–100 show comparable binding activities with tubulin and similar influences on the tubulin polymerization in vitro as the full-length protein, their capacities in destroying the structure of microtubule in the cultured cells are almost undetectable. However, the segment of residues 1–100 reveals clear disrupting activity on the structure of microtubule in cells, although its disrupting activity seems to be lower than the full-length α-synuclein. These results highlight a length-dependent manner for influence of structure of cellular microtubule induced by α-synuclein, which imply that besides the interacting sites other potential domains or structures in the context of whole α-synuclein protein contribute to its microtubule disruption. Certainly, the experimental odds can not be excluded, e.g. the different proteolysis pathways or speeds in cells. Possibly, the shorter peptides are easier to be degraded than the larger one.

Tubulin forms the monomeric subunit of microtubules that are responsible for eukaryotic cell structure in the formation of the cytoskeleton [20]. Microtubules can be stabilized by structural microtubule-associated proteins (MAPs), such as tau that stimulates microtubule assembly [11, 21]. Microtubule networks are critical elements in a variety of fundamental functions, including intracellular scaffolding, cell division, secretory processes, regulatory of cellular motility and transport. Typically, dimers of α- and β-tubulin continually undergo polymerization and de-polymerization by the addition and dissociation of tubulin subunits, thereby affording the cell’s ability to move or change functions [2224]. In neurodegenerative diseases, one of the common characteristics is deposit of abnormal pathologic proteins in brains, which may form plaques. Interestingly, many pathologic proteins and their precursors show the abilities to form protein complexes with tubulin and affect the structures of microtubule in cells, e.g. prion protein [13, 14, 25], amyloid precursor protein [26], tau [27, 28] and α-synuclein. Moreover, disfunction of microtubule has been repeatedly described in neurodegenerative diseases [29, 30]. Therefore, interfering with the normal function of tubulin and normal structure of microtubule of neurons due to deposits of pathologic proteins may be the essential hallmark in the pathogenesis of neurodegenerative diseases.

Acknowledgements

This work was supported by Chinese National Natural Science Foundation Grants 30771914 and 30800975, National Science and Technology Task Force Project (2006BAD06A13-2), Institution Technique R&D Grant (2008EG150300), National Basic Research Program of China (973 Program) (2007CB310505), China Mega-Project for Infectious Disease (2009ZX10004-101) and the SKLID development Grant (2008SKLID102 and 2008SKLID202).

Copyright information

© Springer Science+Business Media B.V. 2009