Economou lab - SEC


Sec translocase
Secretory and membrane polypeptides are translocated across the plasma (Bacteria and Archaea) or the ER (Eucaryotes) membranes by the Sec translocase. These polypeptides are delivered to the translocase by specific pilots/chaperones like SecB and the SRP (signal recognition particle). Many Sec translocase subunits are essential for viability and conserved in evolution. We focus on the molecular mechanism of the bacterial Sec translocase. All of the subunits of this enzyme are known and catalysis has been reconstituted fully in vitro and this allows us to probe the enzymatic mechanism in molecular detail. The membrane subunits SecYEG and the peripheral SecA ATPase comprise the translocase core. SecY, SecE and SecA are essential for catalysis, while SecD, SecF, YajC optimize secretion and YidC is required primarily for the export of polytopic inner membrane proteins. Molecular, biochemical, biophysical and structural tools are combined to reveal novel fascinating features of the translocase nanomachine at work!

To understand how translocase works we focus on four questions:
I. SecA structure
SecA is the translocase motor and converts chemical ATP energy into mechanical work. Our structural studies of SecA involve low resolution analysis using small angle X-ray scattering (Shilton et al., 1998; Dempsey et al., 2002) and high resolution crystallography (Papanikolau et al., unpublished results) and NMR (collaboration with B. Kalodimos, U. Rutgers). In parallel, we have determined the domain structure of SecA using deletion and mutation analysis and biochemical reconstitution (Karamanou et al., 1999; Sianidis et al., 2001; Baud et al., 2002; Vrontou et al., 2004; Papanikou et al., 2004). SecA comprises an aminoterminal DEAD (Asp-Glu-Ala-Asp) motor domain that is homologous to similar domains of RNA and DNA helicases. The DEAD motor forms a mononucleotide fold built of two sub-domains (NBD and IRA2). The nine DEAD family helicase Motifs() shared by SecA and other helicases line the walls of the ATP binding crevice ()formed between NBD and IRA2 (Karamanou et al., 1999; Sianidis et al., 2001; Vrontou et al., 2004). Like other helicases SecA has two unique specificity appendages: a. substrate specificity domain (SSD) that physically "sprouts out" of the NBD sub-domain of the DEAD motor. This region is important for the interaction of SecA with signal peptides (Baud et al., 2002) and whole preproteins (Kimura et al., 1991). b. the C-domain binds to the DEAD motor in each SecA protomer (Karamanou et al., 1999) and is embraced by SSD (). In addition, the C-domain was proposed to provide a dimerization interface (Hirano et al., 1996 Karamanou et al., 1999). In the absence of the C-domain, the DEAD motor has a propensity to oligomerize (Dempsey et al. 2002).Deletion of the SecA termini does not affect dimerization or function (Karamanou et al., 2005) We recently determined the 3D structure of E.coli SecA using X-ray crystallography (Papanikolau et al., 2007) and NMR (Gelis et al., 2007).

 
II. SecA ATPase catalysis
We have determined the salient features of SecA ATP hydrolysis mechanism. The DEAD motor houses all the ATP binding and catalytic activities of the protein (Karamanou et al., 1999; Sianidis et al., 2001; Vrontou et al., 2004; Papanikou et al., 2004). The C-domain of the protein binds in trans to the DEAD motor and regulates catalysis through the IRA1 intramolecular switch (Karamanou et al., 1999; Sianidis et al., 2001; Vrontou et al., 2004). Detailed biochemical kinetics and fluorescence studies have revealed that ADP release from the enzyme imposes a rate limiting step (Sianidis et al., 2001) that is overcome by binding of preprotein ligands. The central regulatory component of the DEAD motor is the IRA2 region (Karamanou et al., unpublished) that is flexible and mobile and was proposed to regulate nucleotide binding and release through cycles of binding and distancing from NBD (Sianidis et al., 2001).
Our aim is to gain further insight into the order and the rates of the nucleotide-driven subreactions and how these are modulated by the ligands during catalysis. Furthermore, we wish to understand how these events lead to the conformational changes that underly movement of preproteins across the membrane.
III. SecA-preprotein interaction
Using synthetic signal peptides and mature preprotein regions, SecA mutants, cross-linking and surface plasmon resonance we determined the location of the preprotein binding site on SecA (Baud et al., 2002). Binding occurs on PBD (Preprotein Binding Domain (previously SSD: Substrate Specificity Domain) present in SecA but absent from other DEAD helicases. In addition we have developed inner membrane model proteins (collaboration with A.Kuhn, Stuttgart; Roos et al., 2001) as well as mature domain secretory model substrates (Baud et al., submitted). Defined residues of the SecA DEAD motor like the Helicase Motif III sense the presence of the bound preprotein and activate specifically the translocation ATPase (Papanikou et al., 2004). This interaction starts with alterations in the conformational state of the PBD that drive opening and closing of Gate1, a salt-bridge that controls DEAD motor activation ().PBD is directly sensning the preprotein signal peptide in a shallow elongated groove (Gelis et la., 2007).
Our aim is to understand how SecA senses preproteins and how these modulate the enzyme's nucleotide-related activities.

IV. SecA-SecY interaction
We established experiemental assays for the quantitation of kinetic parameters characterizing the SecA/SecY interaction. Furthermore we are developing surface labeling technologies using a number of reagents (Karamanou et al., 2008).
Our aim is to define the SecY-SecA interaphase and to est
ablish how this is modulated during catalysis.


OK so how does it work?

We previously proposed the "SecA-cycling hypothesis" to explain translocase function. This model involves repeated piston-like palindromic movements of SecA that lead to transfer of the secretory chain into and across the membrane in segements of 20-30 aminoacyl residues (Economou and Wickner, 1994; Economou et al., 1995). SecA was proposed to walk processively on its preprotein substrate similar to the motion of helicases on nucleic acids (Economou, 1998; Papanikou et al., 2007). We have now gained further insight! Combining our high resolution structural analyses and biochemical and mutational data we formulated a simple "2-lever" or "2-chopsticks" hypothesis (Vrontou and Economou, 2004) (). In this model: a. PBD and the C-domain interact together and with the preprotein substrate b. nucleotide cycling at the DEAD motor causes conformational changes to the DEAD motor c. The DEAD motor and the "specificity domains" establish conformational "cross-talk" d. nucleotide-driven DEAD motor confromational changes are transmitted to the "specificity domains" thereby changing their interaction with the preprotein and achieving its dislocation. e. binding of the DEAD motor to SecYEG ensures that these conformational events are meaningful and take place at the mouth of the preprotein channel.