PII: S0969-2126(02)00892-4


Previews
1471

The crystal structure of SurA is the foundation stone
to understanding SurA functions at a molecular level.
The urgent questions of how SurA recognizes, binds,
and releases its outer membrane protein substrates and
how it facilitates their maturation without the driving
force of ATP hydrolysis can now be directly addressed.
New, exciting insights are likely to emerge within the
near future. Moreover, the structure paves the way to
face the next big challenge—studying the protein dy-
namics involved in substrate binding and release.

Susanne Behrens
Department of Molecular Genetics

and Preparative Molecular BiologyViews of the Putative Polypeptide Binding Channel in SurA
Institute for Microbiology and GeneticsSurface rendering of SurA, highlighting the channel and putative
Georg-August University Göttingenpolypeptide binding surface (green) and the proline binding pockets

of the inactive (blue) and active (orange) PPIase domains (kindly Grisebachstrasse 8
provided by E. Bitto and D. McKay). The left SurA molecular is shown D-37077 Goettingen
in the same orientation as the first figure. Germany

Selected Reading

tide binding channel finds further support in crystal-
1. Hartl, U.F., and Hayer-Hartl, M. (2002). Science 295, 1852–1858.packing interactions. In the crystallographic unit, a short
2. Sauer, F.G., Knight, S.S., Waksman, G.J., and Hultgren, S.J.

helical peptide segment that sits piggyback on SurA (2000). Semin. Cell Dev. Biol. 11, 27–34.
binds to the flap-like structure of the adjacent SurA 3. Spiess, C., Beil, A., and Ehrmann, M. (1999). Cell 97, 339–347.
molecule. However, there are at least two caveats with 4. Krojer, T., Garrido-Franco, M., Huber, R., Ehrmann, M., and

Clausen, T. (2002). Nature 416, 455–459.this interaction identifying the substrate binding pocket.
5. Rouvière, P.E., and Gross, C.A. (1996). Genes Dev. 10, 3170–First, it spans only a fraction of the actual capacity of

3182.the crevice. More importantly, the bound peptide has
6. Missiakas, D., Betton, J.M., and Raina, S. (1996). Mol. Microbiol.

�-helical conformation whereas the known SurA sub- 21, 871–884.
strates, the outer membrane porins, are � stranded. Is 7. Behrens, S., Maier, R., de Cock, H., Schmid, F.X., and Gross,

C.A. (2001). EMBO J. 20, 285–294.the channel indeed the polypeptide binding site relevant
8. Bitto, E., and McKay, D. (2002). Structure 10, this issue, 1489–to the chaperone function in SurA, and if so, which parts

1498.of it actually form the binding surface? Is the observed
9. Lu, K.P., Hanes, S.D., and Hunter, T. (1996). Nature 380,

binding specificity of SurA [7] determined by primary 544–547.
polypeptide sequence, secondary structure, or possibly 10. Dolinski, K., Muir, S., Cardenas, M., and Heitman, J. (1997).

Proc. Natl. Acad. Sci. USA 94, 13093–13098.both?

proteins and the corresponding receptors) and cotrans-
lational transport of proteins into the endoplasmic retic-

Structure, Vol. 10, November, 2002, 2002 Elsevier Science Ltd. All rights reserved. PII S0969-2126(02)00892-4

Obg, a G Domain
with a Beautiful Extension ulum (ER). The remaining GNBPs form a large group of

different proteins [1]. The common property shared by
these is that they contain a more or less conserved
structural module, the G domain, which is usually in-

The structure of Obg, a protein involved in a compli-
volved in the switching of the protein between a GDP-

cated genetic network that regulates stress response
bound and a GTP-bound conformation [2–4]. With the

and sporulation in Bacillus subtilis, reveals a com- exception of translation factors, the signal recognition
pletely new type of guanine nucleotide binding protein particle and its receptor, most of the guanine nucleotide
and provides some hints about its function. binding proteins mentioned above are found only in eu-

karyotes. Recently, in large part fuelled by the large
Guanine nucleotide binding proteins (GNBPs) regulate sequencing projects going on worldwide, it has become
many different types of processes in both prokaryotes clear that there are a number of guanine nucleotide binding
and eukaryotes. They come in a large variety of different proteins that are conserved from bacteria to man and that
sequences and shapes. The most important and widely the functions of most of these are unknown [1].
studied groups are the family of protein biosynthesis One of these is the protein Obg, which was first dis-
factors acting on the ribosome, the heterotrimeric G covered in the Bacillus subtilis spo0 operon as a protein
proteins, the Ras superfamily of small 20–25 kDa pro- involved in a complicated genetic network that regulates

stress response and sporulation. Hence the name, Obg,teins, and the factors involved in signal recognition (SR



Structure
1472

which derives from spo0B-associated GTP binding pro- Some hints about the function of Obg come from
some lucky accidents. Although the protein was purifiedtein. The exact function of this protein is not known, and

many proposals have been put forward for its biological without the addition of nucleotides, one of the mono-
mers in the dimeric asymmetric unit contained, in addi-function. To approach the functional role from a struc-

tural perspective, the group of Chris Lima, in the course tion to the 5� phosphates sitting in the P loop, additional
density bound to ribose, which was modeled as ppG-of a structural genomics project, has now solved the

structure of the Bacillus subtilis Obg, which appears in 3�pp, a nucleotide that is part of a stress response and
accumulates to millimolar concentrations in the cell. In-this issue of Structure [5].

Unlike the G domain-only proteins of the Ras super- cidentally, the TGS domain, which was absent in the
present structure, is found in a number of proteinsfamily, Obg is a large protein that consists of three do-

mains, an N-terminal glycine-rich domain, a Ras-like G known to interact with ppGpp. While this seemed to
suggest a contribution of the TGS domain, the biochemi-domain, and a C-terminal TGS domain. The latter is a

small domain found in a number of proteins that have the cal experiments did not reveal such a contribution. Al-
though it is not completely clear whether the presence ofcommon property of being involved in bacterial stress

response. Unlike Ras or G� proteins, Obg, like a number ppGpp is functionally relevant, biochemical experiments
showed that ppGpp is having some effect on the rateof other GNBPs, shows micromolar affinity for GDP/

GTP, which results in fast exchange of the nucleotide of GTP hydrolysis. However no direct binding of the
nucleotide was observed, and the effects on the GTPaseand thus does not require a guanine nucleotide ex-

change protein (GEF), which, in Ras and G�, regulates were somewhat inconsistent. In any case the fact that
ppGpp, and not any other guanine nucleotide, was foundactivation. It has a very slow GTPase, with a rate on the

same order as that of the Ras proteins (�0.02 min�1), in the protein and was retained during the purification
procedure warrants further experiments to test the im-suggesting that there may be a GTPase-activating pro-

tein (GAP) or another mechanism that increases the rate portance of this observation for the role of Obg in stress
response. Nothing at all is known about the role of mam-of the reaction [6]. However, like guanine nucleotide

binding proteins such as hGBP, dynamin, and septins, malian Obg proteins. Since ppGpp is not involved in
it does not have a glutamine residue in the switch II eukaryotic stress response, we still have to look ahead
region, which has been shown to be crucial for the for some more biology on this seemingly important GTP
GTPase reaction of Ras and G� proteins, suggesting a binding proteins.
rather different mechanism of GTP hydrolysis. Obg is a completely new type of guanine nucleotide

The structure of Obg shows the well-known features binding protein. It is just another example of how a
of the G domain, which typically contains six � strands conserved module, the G domain, has been used for
and five � helices. Obg has one additional � strand in many different switching reactions. As with other gua-
the switch I region and one additional � helix in the nine nucleotide binding proteins, the structures of the
switch II region. The putative switch regions are involved nucleotide-free and the ppGpp-bound forms are only
in the interaction with the rest of the structure, such that the beginning. More structures are needed to fully un-
one could envision nucleotide-dependent changes in derstand the switching mechanism and what biological
the interface between the G domain and the rest of the processes it drives, but the present structure is a marvel-
molecule. While the structure was solved in the presence ous beginning.
of various nucleotides, no significant structural changes
were observed, although this could be due to con-

Alfred Wittinghoferstraints of the crystal packing.
Max-Planck-Institut f. molekulare PhysiologieThe most remarkable feature of the structure is the
44227 DortmundN-terminal part of the Obg, which is unique and is thus
Germanycalled the Obg fold. It is an elongated barrel, the lower

part of which consists of an eight-membered � sheet
Selected Reading

sitting on top of, and interfacing with, the G domain.
The glycine-rich part of the Obg fold consists of a six- 1. Leipe, D.D., Wolf, Y.I., Koonin, E.V., and Aravind, L. (2002). J.
helix bundle, where the helices share structural features Mol. Biol. 317, 41–72.

2. Kjeldgaard, M., Nyborg, J., and Clark, B.F. (1996). FASEB J. 10,with a type II polyproline helix. The six helices pack
1347–1368.together in parallel and antiparallel pairs. There is an

3. Sprang, S.R. (1997). Annu. Rev. Biochem. 66, 639–678.extensive main chain hydrogen bonding pattern be-
4. Vetter, I.R., and Wittinghofer, A. (2001). Science 294, 1299–1304.

tween the helices. Probably for steric reasons, most of
5. Buglino, J., Shen, V., Hakimian, P., and Lima, C.D. (2002). Struc-

the invariant glycines are in the interior of the fold, which ture 10, this issue, 1581–1592
lacks a normal hydrophobic core and is instead stabi- 6. Prakash, B., Renault, L., Praefcke, G.J.K., Herrmann, C., and

Wittinghofer, A. (2000). EMBO J. 19, 4555–4564.lized by side chain interaction on the outside of the fold.