Synaptic vesicle endocytosis impaired by disruption of dynamin-SH3 domain interactions

Science

Volumn 276, Issue 5310, 1997, Pages 259-263

  Shupliakov, Oleg ^a   Löw, Peter ^a   Grabs, Detlev ^a   Gad, Helge ^a   Chen, Hong ^a   David, Carol ^a   Takei, Kohji ^a   De Camilli, Pietro ^a   Brodin, Lennart ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

CLATHRIN; DYNAMIN;

ARTICLE; ENDOCYTOSIS; MOLECULAR DYNAMICS; MOLECULAR INTERACTION; NEUROTRANSMITTER RELEASE; NONHUMAN; PRIORITY JOURNAL; SYNAPSE VESICLE;

AMINO ACID SEQUENCE; ANIMALS; BINDING SITES; CELL MEMBRANE; COATED PITS, CELL-MEMBRANE; DYNAMINS; ENDOCYTOSIS; GTP PHOSPHOHYDROLASES; HUMANS; LAMPREYS; MICROSCOPY, ELECTRON; MOLECULAR SEQUENCE DATA; NERVE TISSUE PROTEINS; PROLINE; RECOMBINANT FUSION PROTEINS; SRC HOMOLOGY DOMAINS; SYNAPSES; SYNAPTIC TRANSMISSION; SYNAPTIC VESICLES;

EID: 1842296915     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.276.5310.259     Document Type: Article

References (48)

1. P. De Camilli and K. Takei, Neuron 16, 481 (1996).
2. M. Robinson, Curr. Opin. Cell. Biol. 6, 538 (1994).
3. K. Takei, P. S. McPherson, S. L. Schmid, P. De Camilli, Nature 374, 186 (1995).
4. J. H. Koenig and K. Ikeda, J. Neurosci. 9, 3844 (1989).
5. M. S. Chen et al., Nature 351, 583 (1991); A. M. Van Der Bliek and E. M. Meyerowitz, ibid., p. 411.
6. M. S. Chen et al., Nature 351, 583 (1991); A. M. Van Der Bliek and E. M. Meyerowitz, ibid., p. 411.
7. I. Gout et al., Cell 75, 25 (1993); J. S. Herskovits, H. S. Shpetner, C. C. Burgess, R. B. Vallee, Proc. Natl. Acad. Sci. U.S.A. 90, 11468 (1993); K. Seedorf et al., J. Biol. Chem. 269, 16009 (1994).
8. I. Gout et al., Cell 75, 25 (1993); J. S. Herskovits, H. S. Shpetner, C. C. Burgess, R. B. Vallee, Proc. Natl. Acad. Sci. U.S.A. 90, 11468 (1993); K. Seedorf et al., J. Biol. Chem. 269, 16009 (1994).
9. I. Gout et al., Cell 75, 25 (1993); J. S. Herskovits, H. S. Shpetner, C. C. Burgess, R. B. Vallee, Proc. Natl. Acad. Sci. U.S.A. 90, 11468 (1993); K. Seedorf et al., J. Biol. Chem. 269, 16009 (1994).
10. B. Lichte, R. W. Veh, H. E. Meyer, M. W. Kilimann, EMBO J. 11, 2521 (1992).
11. C. David, P. McPherson, O. Mundigl, P. De Camilli, Proc. Natl. Acad. Sci. U.S.A. 93, 331 (1996).
12. Single-letter abbreviations for the amino acid residues are as follows: A, Ala; G, Gly; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; V, Val.
13. D. Grabs et al., J. Biol. Chem., in press.
14. L. H. Wang, T. C. Südhof, R. G. W. Andersson, ibid. 270, 10079 (1995).
15. C. David, M. Solimena, P. De Camilli, FEBS Lett. 351, 73 (1994).
16. L. Brodin, O. Shupliakov, J. Hellgren, V. A. Pieribone, R. H. Hill, J. Neurophysiol. 72, 592 (1994).
17. O. Shupliakov, L. Brodin, S. Cullheim, O. P. Ottersen, J. Storm-Mathisen, J. Neurosci. 7, 1111 (1992).
18. The giant reticulospinal axons in the lamprey spinal cord are unbranched and form glutamatergic synapses [AMPA (α-amino-3-hydroxy-5-methyl-4-isoxazole-propianate) and NMDA (N-methyl-D-aspartate) receptors] from their main trunk (diameter 30 to 50 μm) onto dendrites of spinal neurons (13, 14). GST fusion proteins were prepared as described (8, 10, 12) and labeled with a monofunctional Cy5 dye (Amersham) according to the manufacturer's instructions. The dye-to-protein molar ratio was between 1.5 and 4. The labeled proteins [0.5 to 1.5 mg/ml in 250 mM K acetate and 10 mM Hepes (pH 7.4)] were injected with pressure pulses (5 to 15 psi) of 200 ms duration through micropipettes (with a resistance of 50 to 70 megohm) into giant reticulospinal axons (13) with a resting membrane potential of at toast -60 mV. The protein-linked fluorescence was monitored with a charge-coupled device (CCD) detector cooled to -6O°C (Princeton Instruments). The injection microelectrode was removed before stimulation was applied through an extracellular electrode (13). The stimulation period was ended by replacing the physiological solution (21) with fixative containing 3% glutaraldehyde and 0.5% p-formaldehyde in 0.1 M phosphate buffer (pH 7.4).
19. The preparations were processed as described (21). Synapses were cut in serial ultrathin sections and examined with a Philips CM12 electronmicroscope.
20. V. A. Pieribone et al., Nature 375, 493 (1995).
21. The total number of synaptic vesicles in synapses of GST-amphSH3-injected axons was 254 ± 75 (unstimulated) and 148 ± 67 (0.2 Hz stimulation) (n = 4 and 6, respectively; P < 0.05, t test). Values represent the average number of vesicles per micrometer of active zone in the center section of synapses ± SD. This value shows a linear correlation with the total number of synaptic vesicles per synapse (21).
22. The active zones exhibited a disk like shape, as described for normal synapses (41). The number of synaptic vesicles within 0 to 50 nm from the active zone membrane (the first row of vesicles) was 14 ± 3 and 13 ± 4 for GST-amphSH3 and control synapses, respectively. The number of vesicles in the 0-to 150-nm region was 43 ± 4 and 51 ± 4 for GST-amphSH3 and control synapses, respectively (P < 0.05, comparison of the latter two values; n = 5; t test).
23. The total number of synaptic vesicles per micrometer of active zone for GST-amphSH3-injected axons stimulated at 5 Hz was 58 ± 32 (n = 6), P < 0.05 (t test) in comparison with GST-amphSH3-injected axons stimulated at 0.2 Hz (18); P < 0.001 in comparison with GST-amphSH3-injected, unstimulated axons (18).
24. O. Shupliakov, V. A. Pieribone, H. Gad, L. Brodin, Eur. J. Neurosci. 7, 1111 (1995).
25. The number of synaptic vesicles within 0 to 50 nm from the active zone membrane was 7 ± 3 and 14 ± 1, for GST-amphSH3-injected and control synapses, respectively (5 Hz stimulation; P < 0.001, n = 5, t test). The corresponding values for the 0-to 150-nm region were 20 ± 11 and 47 ± 5 (P < 0.001, n = 6; values represent synaptic vesicles per micrometer of active zone).
26. Lamprey spinal cords were homogenized (20% w/v) in a buffer containing 150 mM NaCl; 10 mM Hepes (pH 7.4); 5 mM EDTA; 4 μg/ml each of leupeptin, pepstatin, antipain and aprotinin; 10 mM benzamidine; and 0.4 mM phenylmethylsulfonyl fluoride. Triton X-100 (20% w/v) was added to the homogenate (for a final concentration of 1 % w/v), which was agitated for 1 hour on ice. Insoluble material was removed at 40,000g for 1 hour at 4°C. The extract was incubated with glutathione-Sepharose beads and then with Sepharose beads (50 to 200 μl) bound with GST fusion protein for 2 hours at 4°C. The beads were then washed extensively with the buffer containing 1% w/v Triton X-100. The proteins were eluted with SDS-polyacrylamide gel electrophoresis sample buffer, and protein separation and immunoblotting was done as described (8, 70).
27. The peptide PPPQVPSRPNRAPPG (9) corresponds to amino acids 828 to 842 of dynamin laa [J. M. Sontag et al., J. Biol. Chem. 269, 4547 (1994)]. In peptide competition assays, a final concentration of 300 μM was used. The peptide produced a similar inhibition of dynamin binding to GST-amphSH3 when rat brain extract was used (P. Low, L. Brodin, D. Grabs, P. De Camilli, unpublished observation).
28. P. S. McPherson et al., Nature 379, 353 (1996).
29. Lamprey spinal cord extract was affinity-purified (23) with GST fusion proteins containing either the full-length proline-rich domain of dynamin laa or the proline-rich domain lacking the last 31 amino acids in the COOH-terminus (10). Monoclonal antibodies to the COOH-terminal portion of human amphiphysin (12) were from A. Zhang and M. Butler. Similar results were obtained with polydonal antibodies to amphiphysin (M. Butler and P. De Camilli, unpublished observations).
30. The peptide (24) was mixed with injection solution (15) at a concentration of 20 mM with 5 μM Texas Red-conjugated dextran 3000 (Molecular Probes) added as an injection marker.
31. O. Shupliakov and L. Brodin, unpublished observation. See also (17) for other examples of control-injected synapses.
32. The electron-dense dynamin collars present in the shibire mutant are visible in specimens fixed and stained with the same methods used in the present study (4, 21).
33. Recruitment of rat brain dynamin to endocytic sites in mammalian cell membranes under in vitro conditions is blocked by GST-amphSH3 and the 15-oligomer dynamin peptide (24), whereas the recruitment of the clathrin coat, as shown by AP180, amphiphysin, or clathrin labeling, is not affected (H. Chen, K. Takei, P. De Camilli, unpublished observations).
34. The total EPSP amplitude measured 26 to 30 min after GST-amphSH3 had reached the synaptic region in the presynaptic axon remained within ±22% of the amplitude during a 3-to 5-min control period before the injection (mean difference + 4%; n = 5). Recordings were done as described (13, 17).
35. This observation and the lack of effect on the clustering of synaptic vesicles suggest that GST-amphSH3 does not perturb the function of proteins involved in the organization (17) or exocytosis of synaptic vesicles, in spite of the presence of proline-rich regions in many of these proteins [M. Linial, Neuroreport 5, 2009 (1994)].
36. For GST-amphSH3-injected axons, the total EPSP amplitude decreased on average 77% (range 68 to 84%; n = 5) when the initial control period (5 min, 0.2 Hz stimulation) was compared with the end of a 30-min 5-Hz stimulation period. For GST-amphSH3mut-injected axons, the corresponding decrease was 46% (range 36 to 56%; n = 4), and for GST-injected axons it was 43% (range 38 to 49%; n = 3). The EPSPs studied had only a small electrotonic component in relation to the chemical component, and this relation was similar between the groups. A small chemical EPSP could be detected even at the end of the 30-min stimulation period after GST-amphSH3 injection. This indicates that GST-amphSH3 produces a kinetic inhibition of synaptic vesicle recycling, rather than a permanent blockade.
37. J. T. Buchanan, L. E. Moore, R. H. Hill, P. Wallén, S. Grillner, Biol. Cybernet. 67, 123 (1992).
38. The "resting" dynamin pool, rather than being soluble in the cytoplasm, is accumulated at synaptic sites, presumably through interactions with components of the cytomatrix (3) [P. S. Estes et al., J. Neurosci. 16, 5443 (1996)]. This is consistent with our observation that GST-amphSH3, while inhibiting dynamin function, binds to and remains associated with the synaptic sites.
39. O. Shupliakov, P. Löw, H. Gad, L. Brodin, unpublished observation.
40. R. Fesce, F. Grohovaz, F. Valtorta, J. Meldolesi, Trends Cell Biol. 4, 1 (1994); A. W. Henkel and W. J. Betz, J. Neurosci. 15, 8246 (1995).
41. R. Fesce, F. Grohovaz, F. Valtorta, J. Meldolesi, Trends Cell Biol. 4, 1 (1994); A. W. Henkel and W. J. Betz, J. Neurosci. 15, 8246 (1995).
42. J. E. Heuser, Cell Biol. Int. Rep. 13, 1063 (1989).
43. R. Scaife, I. Gout., M. D. Waterfield, R. L. Margolis, EMBO J. 13, 2574 (1994); Z. Wang and M. F. Moran, Science 272, 1935 (1996).
44. R. Scaife, I. Gout., M. D. Waterfield, R. L. Margolis, EMBO J. 13, 2574 (1994); Z. Wang and M. F. Moran, Science 272, 1935 (1996).
45. H. Shpetner, J. S. Herskovits, R. Vallee, J. Biol. Chem. 271, 13 (1996).
46. V. Gundersen et al., J. Neurosci. 15, 4417 (1995).
47. Three-dimensional (3D) reconstructions were obtained from electronmicrographs of serial ultrathin sections (35 and 18 sections, respectively, for GST-amphSH3 and GST-amphSH3mut) with the use of MacStereology software. Only the main contours of the plasma membrane were included.
48. We thank A. Zhang and M. Butler for help and suggestions; P. Greengard, S. Grillner, and R. Petterson for advice and comments on the manuscript; S. J. Redman for providing computer software; and M. Bredmyr and H. Axegren for technical assistance. Supported by grants from the Swedish Medical Research Council (project 11287) and Jeanssons Stiftelser to L.B.; by grants from the Donaghue Foundation, the Human Frontier Science Program, and NIH (CA46128) to P.D.C.; and by postdoctoral fellowships from the Deutscher Akademischer Austauschdienst (to D.G.) and the United States Army Medical Research and Development Command (to C.D.).