Structure of the yeast polarity protein Sro7 reveals
a SNARE regulatory mechanism
Douglas A. Hattendorf1, Anna Andreeva2, Akanksha Gangar2, Patrick J. Brennwald2& William I. Weis1
Polarized exocytosis requires coordination between the actin
cytoskeleton and the exocytic machinery responsible for fusion
of secretory vesicles at specific sites on the plasma membrane1.
Fusion requires formation of a complex between a vesicle-bound
Proteins in the lethal giant larvae protein family, including lethal
giant larvae and tomosyn in metazoans and Sro7 in yeast, interact
with Q-SNAREs and are emerging as key regulators of polarized
exocytosis3.The crystal structure ofSro7 reveals two seven-bladed
WD40 b-propellers followed by a 60-residue-long ‘tail’, which is
bound to the surface of the amino-terminal propeller. Deletion of
the Sro7 tail enables binding to the Qbc SNARE region of Sec9
and this interaction inhibits SNARE complex assembly. The N-
terminal domain of Sec9 provides a second, high-affinity Sro7
interaction that is unaffected by the tail. The results suggest that
Sro7 acts as an allosteric regulator of exocytosis through inter-
cate that lethal giant larvae and tomosyn have a two-b-propeller
foldsimilarto that of Sro7, but only tomosyn appearsto retain the
Growth of the bud during cell division in Saccharomyces cerevisiae
requires polarized vesicle fusion at the bud tip1. Sro7, a 1,033-residue
protein, is part of the cellular machinery that facilitates this process4.
proteins, including SNAREs, exocyst subunits and the Rab GTPase
Sec4 (refs 4–6). In addition, Sro7 suppresses mutations in Rho3 and
Cdc42, two Rho GTPases necessary for exocytosis and polarization
of the actin cytoskeleton, and in Myo2, a myosin involved in vesicle
transport7–9. Genetic analysis places Sro7 function downstream of the
in polarized bud growth remains unclear.
Like Sro7, lethal giant larvae (LGL) and tomosyn regulate targeted
functions in the presynaptic nerve terminal to inhibit priming of syn-
Q-SNARE proteins, but with different specificities. Sro7 binds to Sec9
to syntaxin4, a Qa-SNARE13. Tomosyn has an R-SNARE motif at its
carboxy terminus that substitutes for synaptobrevin in the neuronal
,40amino acids at the extreme C terminus of Sro7 are predicted to
form an a-helix of unknown function15.
Wedeterminedthecrystalstructure ofaproteolytic fragment ofS.
cerevisiae Sro7, spanning residues 61–962, at 2.4A˚resolution. The
final model includes residues 62–582 and 598–951. The sequence
containing the predicted a-helix at the C terminus was protease-
sensitive and is absent from the crystallized fragment. The structure
sequence, followed by a 60-residue tail (Fig. 1a–d). The WD40
repeats fold into two seven-bladed b-propeller domains with a topo-
logy similar to Aip1 (refs 16, 17). Blades 1–7 form the N-terminal
domain, and blades 8–14 form the C-terminal domain. Each blade is
to D. The bottom surface of each propeller is formed by the A–B and
C–D loops, and the top surfaces are formed by the B–C and D–A
The b-propeller domains are arranged to resemble a twisted, open
clamshell. A notable feature of the domain interface is the amphi-
pathic 7C–8A loop, which is buried between the two domains
(Fig. 1c). Its hydrophobic face packs against the N-terminal b-
propeller (Supplementary Fig. 1a), and its hydrophilic face interacts
with the C-terminal b-propeller through an extensive network of
direct and water-mediated hydrogen bonds (Supplementary Fig.
1b). Another feature of the interface is the 8AB loop, which is
anchored by a hydrophobic a-helix that binds to the C-terminal
b-propeller and which contains an extended 23-residue region that
wraps around the 1AB and 8CD loops at the bottom side of the
domain interface (Supplementary Fig. 1c). Leu490 and Leu491 in
the extended region act as a second anchor by packing into a hydro-
phobic pocket formed by residues in the N-terminal b-propeller and
the 7C–8A loop (Supplementary Fig. 1a).
TheSro7 tail(residues892–951)isbound tothebottom surfaceof
the N-terminal b-propeller (Fig. 1e). The binding interface is extens-
of the 8AB loop, which links the tail to the domain interface. The tail
spans the entire propeller once in an extended conformation, then
again as two short a-helices separated by a 10-residue linker. The
pocket for tail residues 898–901.
Sequence alignment of Sro7 with tomosyn and LGL shows signifi-
cant conservation in the 14 WD40 repeats and in many elements of
the domain interface (Fig. 1d and Supplementary Fig. 2). The meta-
zoan proteins have insertions in the 10D–11A loop that contain
phosphorylation sites implicated in their regulation18,19. Notably,
many residues in the Sro7 tail that interact with the N-terminal
propeller (Fig. 1e) are conserved in tomosyn, but not in LGL. This
divergence is also reflected in residues that form the binding site for
the tail: they are conserved in the yeast (Fig. 2a) and tomosyn sub-
certain positions, co-variation of residues in the tail and its binding
site in tomosyn seem to preserve the contacts observed in Sro7.
suggesting that the tail serves a regulatory rather than a structural
role. The effect of deleting the Sro7 tail on binding of Sec9 was
2Department of Cell and Developmental Biology, University of North Carolina, 538 Taylor Hall, Chapel Hill, North Carolina, 27599-7090, USA.
Vol 446|29 March 2007|doi:10.1038/nature05635
measured by isothermal titration calorimetry (ITC; Supplementary
Fig. 3 and Supplementary Table 2). An Sro7 variant lacking the tail,
Sro7(61–891), binds to Sec9 with 3.3-fold higher affinity (Kd5
800nM) and with a 2.3-fold more favourable enthalpy change than
either wild-type Sro7 or the crystallization construct Sro7(61–962),
which contains the tail. Thus, the tail has an autoinhibitory effect on
Sro7 binding to Sec9.
To understand this effect of the tail, we first defined the regions of
Sec9 that bind to Sro7 using a yeast two-hybrid assay. Sec9 has a 413-
residue N-terminal domain that is not required for growth20, followed
motifs, designated the Qbc-SNARE domain (Supplementary Fig. 4a).
SNARE domain shows a weaker interaction (Supplementary Fig. 4a).
Residues 255–343 in the N-terminal domain were necessary for the
high-affinity interaction, because their deletion reduced binding to
the level of the isolated Qbc domain. The region of Sec9 necessary
tary Fig. 4a) is distinct from the high-affinity Sro7 binding site. The
two-hybrid results were confirmed by co-immunoprecipitation (Sup-
plementary Fig. 4b). Deletion of residues 255–343 of Sec9 reduced the
was still associated with the Sec9 mutant.
On the basis of the results of the two-hybrid assay and additional
mapping experiments (see Supplementary Methods), a fragment of
the Sec9 N-terminal domain was purified to measure its binding to
Sro7 in vitro. Sec9(250–372) bound to Sro7(61–962) with a Kdof
3mM, identical to that of full-length Sec9 (Supplementary Fig. 3 and
Supplementary Table 2). Significantly, deletion of the Sro7 tail had
no effect on binding of Sec9(250–372) (Supplementary Fig. 3 and
Supplementary Table 2). To locate the binding site for the Sec9 N-
terminal domain fragment, we examined the conservation of Sro7
surface residues. In addition to the highly conserved tail-binding site
(Fig. 2a), there is a second cluster of conserved residues in a shallow
binding to Sec9(250–372), and three others (E562A, R664A and
(Supplementary Fig. 5 and Supplementary Table 2).
In this case, binding was only observed if the Sro7 tail was deleted
3), but the low affinity and small enthalpy change prevented a pre-
cise determination of the binding thermodynamics. Nonetheless,
two independent titrations indicated a Kdof approximately 50mM.
These results indicate that in wild-type Sro7, the tail masks the
S. cerevisiae Sro7
Mus musculus LGL1
Rattus norvegicus tomosyn
C-terminal propeller Tail
Figure 1 | Overview of the Sro7 structure. a–c, The N-terminal b-propeller
of the N-terminal b-propeller. WD40 blades 1–7, b-strands A–D of blade 6,
9 that is not visible in the structure (residues 583–597). c, Side-view of the
structure; the interface loops and the tail are labelled. d, Schematic of the
structure-based sequence alignment of the LGL family. Structural elements
are coloured as in a–c. The predicted C-terminal a-helix in Sro7 and the
R-SNARE of tomosyn are shown in red. Light green and light blue boxes
correspond to the WD40 blades as observed structurally, rather than the
standard definition of a WD40 repeat; that is, strand D of one blade and
strands A–C of the next blade. Note that strand 14D is located near the
Phosphorylation sites in tomosyn and LGL are marked by ‘P’. e, The Sro7
tail-binding site, with the structure oriented as in panel a. The b-propeller
domains are shown in surface representation and coloured by electrostatic
potential25in the range of –10kT/e (red) to 10kT/e (blue). The tail and the
8AB loop are shown as ribbons. Side chains in the tail that interact with the
N-terminal domain are shown as sticks. Side chains in the 8AB loop that
form part of the tail-binding site are also shown.
NATURE|Vol 446|29 March 2007
Qbc-SNARE bindingsite, eitherbydirectcompetition orthrough an
allostericmechanism. Thisinteraction isnotentirely specifictoSec9,
because the SNARE domain of the Qa-SNARE Sso1 (residues 193–
265) also binds Sro7(61–891), albeit more weakly (Supplementary
Sro7(61–891) did not bind to the pre-assembled SNARE complex
of GST–Sec9 Qbc-SNARE, the Qa-SNARE Sso1 and the R-SNARE
Snc2 (Fig. 3a). In a kinetic assay, Sro7(61–891) slowed SNARE
complex assembly fourfold relative to assembly reactions lacking
Sro7 or containing Sro7(61–962) (Fig. 3b and Supplementary Fig.
6b). Thus, Sro7(61–891) binds to the Sec9 Qbc-SNARE domain in a
manner that competes with SNARE complex formation. At the con-
centrations used in the kinetic assay, approximately 60% of the Sro7
and full-length Sec9 are expected to be in a complex, whereas almost
no complex would exist in the absence of the Sec9 N-terminal
domain. The Sec9 N-terminal domain may act as a high-affinity
Figure 2 | Surface representation of Sro7, with residues coloured from
100% identity. Because the Sec9 N-terminal domain is unique to the
Saccharomycotina subphylum of fungi, the sequences used were limited to
species in this subphylum (S. cerevisiae, Candida glabrata, Kluyveromyces
lactis, Ashbya gossypii, Yarrowia lipolytica, Candida albicans and
Debaryomyces hansenii). The tail is shown as a dark blue ribbon.
the N-terminal b-propeller, as in Fig. 1a, showing the conservation of the
tail-binding site. b, Structure rotated from a to show the second conserved
labelled on the surface.
GST–Sec9 Qbc –
Bgl2 internal (%)
200 600 1,000
Fraction Sso1 bound
Bgl2 internal (%)
Figure 3 | InteractionofSec9withSro7. a,GSTpull-downassaysofbinding
of Sro7 variants to the GST–Sec9 Qbc-SNARE domain. A gel stained with
Coomassie blue is shown. b, Fraction of Sso1(145–265) pulled down by
GST–Sec9 as a function of time in the absence of Sro7 (filled circles), or
presence of Sro7(61–962) (open squares) or Sro7(61–891) (filled squares).
Error bars, s.d.; n53. Data were fitted to a second-order rate equation (see
Supplementary Information). In the absence of Sro7, the rate constant is
7,30061,200M21s21, which is very similar to that previously determined
using a circular dichroism assay for assembly27. Sro7(61–962) had no effect
on assembly (rate constant of 7,2006700M21s21, whereas Sro7(61–891)
deletion mutants on growth. The top panel shows the ability of SRO7 alleles
to complement the cold-sensitive growth phenotype of the sro7D,sro77D
strain. The bottom panel examines dominant-negative effects of the SRO7
alleles in wild-type yeast. In each panel, growth of three independent
colonies at the permissive (37uC) and restrictive temperatures (25uC and
19uC) is shown. d, The effect of SRO7 alleles on secretion. Graphs show the
in either the sro7D,sro77D strain (top panel) or wild-type strain (bottom
panel). Error bars, s.d.; n53.
NATURE|Vol 446|29 March 2007
tether for the Qbc-SNARE domain, such that the regulatory inter-
action between Sro7 and the Qbc domain occurs at a biologically
significant concentration. Given the sequence homology among
Q-SNAREs21, a second, related role may be to increase the specificity
of Sro7 for the Qbc-SNARE over Sso1.
The Sro7 mutants that are defective in binding to the Sec9 N-
terminal domain had no observable phenotype when expressed in
yeast (data not shown), consistent with the fact that this domain is
which removes the tail, caused a loss-of-function phenotype in a
sro7D,sro77D strain and a dominant-negative phenotype in the
wild-type strain (Fig. 3c). In the complementation assay, the sro7-
D142 mutant was expressed at only 20% of wild-type Sro7 level
(Supplementary Fig. 7). However, the mutant does not complement
sro7D,sro77D even when expressed at a level higher than wild-type
Sro7, from the stronger tetracycline-repressible promoter (data not
shown). These phenotypes were specifically associated with removal
of the tail that is observed in the structure, because a smaller deletion
of the predicted a-helix at the C terminus (residues 987–1,028; sro7-
D42) was a much weaker loss-of-function allele with no dominant-
with defects in secretion, monitored by examining the accumulation
with the biochemical data, this suggests that the tail is an essential
component of Sro7 function during SNARE-mediated exocytosis.
The interaction of wild-type Sro7 with the Sec9 Qbc-SNARE
domain in vivo suggests that Sro7 exists in two conformational states
with different affinities for the tail. The equilibrium between these
states may be affected by other factors that regulate Sro7 activity.
architecture of Sro7 is suggestive of an allosteric mechanism whereby
rearrangements in the b-propeller domain interface that are assoc-
iated with ligand binding are propagated to the tail through inter-
action between the tail and the 8AB loop (Fig. 1e and Supplementary
Unlike other Qb and Qc SNARE proteins, Sec9 does not have a
transmembrane domain or palmitoylation sites to tether it to the
plasma membrane2. Sro7 is associated with the membrane at the
bud site4, so one role of Sro7 may be to localize Sec9 to the plasma
membrane. The mechanism of Sro7 membrane association is un-
known, but we suggest that it is coupled to displacement of the tail
from its binding site (Fig. 4). In this state, Sro7 could both recruit
Sec9 to the eventual site of membrane fusion and inhibit binding to
Sso1 and Snc2. A factor that stimulates rebinding of the tail would
release the Sec9 Qbc-SNARE domain, allowing SNARE complex
localized to the vesicle22,23. In this manner, release of Sec9for SNARE
complex formation could be coupled to the arrival of a secretory
vesicle at the site of exocytosis.
Crystallization and structure determination. Crystals of Sro7 purified from
an S. cerevisiae overexpression strain were grown at 22uC in 12% PEG3350,
Phases were determined by single isomorphous replacement with anomalous
scattering from a thimerosal derivative of Sro7(61–962). The final model
has Rworkand Rfreevalues of 21.1% and 26.1% at 2.4A˚(Table 1). Details of
the crystallographic analysis and statistics are presented as Supplementary
Sro7–Sec9 interaction. Yeast two-hybrid assays, immunoprecipitations and
secretion assays were performed essentially as described4,7. For phenotypic ana-
lysis, Sro7 constructs were expressed from the Sro7 promoter on a CEN-HIS3
plasmid (complementation assay) or from a tetracycline-repressible promoter24
(dominant-negative assay). For pull-down and ITC experiments, Sro7 con-
structs were purified from S. cerevisiae and Sec9 constructs were purified from
Escherichia coli. ITC was performed at 25uC by injecting Sec9 at 80 to 200mM
into the ITC cell containing Sro7at 4 to 10mM. N-terminal GST fusion proteins
immobilized on gluthathione-agarose were used for pull-down assays. The rate
of SNARE complex assembly was measured in a pull-down assay by incubating
2.5mM each of GST–Sec9, Sso1(145–265) and Snc2(1–92), in the presence or
absence of 2.5mM of Sro7(61–962) or Sro7(61–891), then quenching the reac-
tionatgiventime pointsby tenfolddilution intobuffer.Detailedproceduresare
presented in Supplementary Information.
Received 13 December 2006; accepted 29 January 2007.
1. Pruyne, D., Legesse-Miller, A., Gao, L., Dong, Y. & Bretscher, A. Mechanisms of
polarized growth and organelle segregation in yeast. Annu. Rev. Cell Dev. Biol. 20,
Jahn, R. & Scheller, R. H. SNARE—engines for membrane fusion. Nature Rev. Mol.
Cell Biol. 7, 631–643 (2006).
Wirtz-Peitz, F. & Knoblich, J. A. Lethal giant larvae take on a life of their own.
Trends Cell Biol. 16, 234–241 (2006).
Figure 4 | Proposed model for how Sro7 coordinates release of Sec9 with
arrival of a secretory vesicle. The N-terminal and SNARE domains of Sec9
structural data it is equally plausible that rebinding of the tail causes a
conformational change that displaces the Sec9 Qbc-SNARE domain from a
different site. See text for details.
Table 1 | X-ray data collection and refinement statistics
a, b, c (A˚)
a, b, c (u)
Number of reflections (work/free)
Number of atoms
Ligand or ion
Ligand or ion
Bond lengths (A˚)
Bond angles (u)
130.5, 130.5, 116.0
90, 90, 90
*The highest-resolution shell is shown in parentheses.
NATURE|Vol 446|29 March 2007
4. Lehman, K.,Rossi,G.,Adamo, J.E.&Brennwald, P.Yeasthomologuesoftomosyn Download full-text
and lethal giant larvae function in exocytosis and are associated with the plasma
membrane SNARE, Sec9. J. Cell Biol. 146, 125–140 (1999).
Zhang, X. etal. Lethal giant larvae proteins interact with the exocyst complex and
are involved in polarized exocytosis. J. Cell Biol. 170, 273–283 (2005).
Grosshans, B. L. et al. The yeast lgl family member Sro7p is an effector of the
secretory Rab GTPase Sec4p. J. Cell Biol. 172, 55–66 (2006).
Adamo, J. E., Rossi, G. & Brennwald, P. The Rho GTPase Rho3 has a direct role in
exocytosis that is distinct from its role in actin polarity. Mol. Biol. Cell 10,
Adamo, J. E. et al. Yeast Cdc42 functions at a late step in exocytosis,
Kagami, M., Toh-e, A. & Matsui, Y. Sro7p, a Saccharomyces cerevisiae counterpart
of the tumor suppressor l(2)gl protein, is related to myosins in function. Genetics
149, 1717–1727 (1998).
10. Vasioukhin, V. Lethal giant puzzle of Lgl. Dev. Neurosci. 28, 13–24 (2006).
11. Gracheva, E. O. et al. Tomosyn inhibits synaptic vesicle priming in Caenorhabditis
elegans. PLoS Biol. 4, e261 (2006).
12. McEwen, J. M., Madison, J. M., Dybbs, M. & Kaplan, J. M. Antagonistic
regulation of synaptic vesicle priming by Tomosyn and UNC-13. Neuron 51,
13. Musch, A. et al. Mammalian homolog of Drosophila tumor suppressor lethal (2)
kidney cells. Mol. Biol. Cell 13, 158–168 (2002).
14. Hatsuzawa, K., Lang, T., Fasshauer, D., Bruns, D. & Jahn, R. The R-SNARE motif
of tomosyn forms SNARE core complexes with syntaxin 1 and SNAP-25 and
down-regulates exocytosis. J. Biol. Chem. 278, 31159–31166 (2003).
15. Pobbati, A. V., Razeto, A., Boddener, M., Becker, S. & Fasshauer, D. Structural
basis for the inhibitory role of tomosyn in exocytosis. J. Biol. Chem. 279,
16. Mohri, K., Vorobiev, S., Fedorov, A. A., Almo, S. C. & Ono, S. Identification of
functional residues on Caenorhabditis elegans actin-interacting protein 1 (UNC-
78) for disassembly of actindepolymerizing factor/cofilin-bound actinfilaments.
J. Biol. Chem. 279, 31697–31707 (2004).
protein that regulates cofilin-mediated actin depolymerization. J. Biol. Chem. 278,
18. Betschinger, J., Eisenhaber, F. & Knoblich, J. A. Phosphorylation-induced
autoinhibition regulates the cytoskeletal protein Lethal (2) giant larvae. Curr. Biol.
15, 276–282 (2005).
19. Baba, T., Sakisaka, T., Mochida, S. & Takai, Y. PKA-catalyzed phosphorylation of
tomosyn and its implication in Ca21-dependent exocytosis of neurotransmitter.
J. Cell Biol. 170, 1113–1125 (2005).
20. Brennwald, P.etal.Sec9isaSNAP-25-likecomponent ofayeastSNAREcomplex
that may be the effector of Sec4 function in exocytosis. Cell 79, 245–258 (1994).
Proc. Natl Acad. Sci. USA 95, 15781–15786 (1998).
22. Goud, B., Salminen, A., Walworth, N. C. & Novick, P. J. A GTP-binding protein
required for secretion rapidly associates with secretory vesicles and the plasma
membrane in yeast. Cell 53, 753–768 (1988).
23. Boyd, C., Hughes, T., Pypaert, M. & Novick, P. Vesicles carry most exocyst
subunits to exocytic sites marked by the remaining two subunits, Sec3p and
Exo70p. J. Cell Biol. 167, 889–901 (2004).
24. Gari, E., Piedrafita, L., Aldea, M. & Herrero, E. A set of vectors with a tetracycline-
regulatable promoter system for modulated gene expression in Saccharomyces
cerevisiae. Yeast 13, 837–848 (1997).
25. Baker, N. A., Sept, D., Joseph, S., Holst, M. J. & McCammon, J. A. Electrostatics of
nanosystems: application to microtubules and the ribosome. Proc. Natl Acad. Sci.
USA 98, 10037–10041 (2001).
26. Landau, M. et al. ConSurf 2005: the projection of evolutionary conservation
scores of residues on protein structures. Nucleic Acids Res. 33, W299–W302
27. Nicholson, K. L. et al. Regulation of SNARE complex assembly by an N-terminal
domain of the t-SNARE Sso1p. Nature Struct. Biol. 5, 793–802 (1998).
Supplementary Information is linked to the online version of the paper at
Acknowledgements We thank L. Katz for help with the yeast two-hybrid analysis,
and S. Kaiser for assistance with mass spectrometry. Diffraction data were
measured at the Advanced Light Source and the Stanford Synchrotron Radiation
Laboratory. D.A.H. was supported by a fellowship from the American Cancer
Society. This work was supported by NIH grants to P.J.B. and W.I.W and a grant
from the G. Harold and Leila Y. Mathers Foundation to P.J.B.
Author Contributions Crystallographic and biochemical experiments were
designed by D.A.H. and W.I.W. and performed by D.A.H. Plasmid and strain
construction were designed and performed by D.A.H. and P.J.B. Yeast two-hybrid
analysis, secretion assays and analysis of mutant phenotypes were designed by
P.J.B. and performed by A.A. and A.G. D.A.H. and W.I.W. wrote the paper and all
authors made editorial comments.
Author Information Coordinates and structure factors have been deposited in the
Protein Data Bank under accession number 2OAJ. Reprints and permissions
information is available at www.nature.com/reprints. The authors declare no
competing financial interests. Correspondence and requests for materials should
be addressed to W.I.W. (firstname.lastname@example.org) or P.J.B.
NATURE|Vol 446|29 March 2007