Back to the OMM Exhibits

The Signal Recognition Particle Core of E. coli

© David Marcey, 2000

I. Introduction
II. Ffh M Domain and Srp RNA Domain IV Structure
III. RNA-Protein Interaction
IV. Signal Sequence Binding
V. References

Note:  This exhibit is best viewed if the cue buttons ( ) are pressed in sequence and if the viewer does not independently manipulate the molecule on the left.

In both bacteria and eukaryotic cells, proteins destined for secretion or for insertion into cellular membranes need to be targeted appropriately. Genes that encode such proteins specify a short, amino-terminal signal sequence that is required for the protein to find its way to the membrane. The signal sequence is proteolytically removed after its targeting role has been performed. In both groups of organisms, a signal recognition particle (SRP) is responsible for recognizing and binding to a signal sequence at the amino terminus of a growing, membrane-bound protein. The SRP then targets the ribosome that is synthesizing the protein to either the endoplasmic reticulum (eukaryotes) or to the plasma membrane (bacteria).
The role of the SRP is shown schematically at left for a eukaryotic cell.

SRPs in all organisms are ribonucleoproteins (RNPs). In E. coli, a single protein, Ffh, and a single 4.5S RNA make up the SRP. In eukaryotes, the SRP is more complex, containing a larger RNA component and multiple proteins, but a conserved RNP core is retained. For example, the human SRP core contains a homolog of Ffh, SRP54, plus a 7S RNA with homology to the E. coli 4.5S RNA. The bacterial Ffh-4.5S RNA complex thus represents an evolutionarily preserved core of the SRP.

At left is shown the E. coli Ffh protein with its M domain highlighted (Keenan, et al., 1998). The M domain is responsible for binding both the 4.5S SRP RNA and the nascent amino terminal signal peptide of membrane proteins. Ffh also contains a four-helix bundle N-terminal domain and a G domain responsible for interacting with the SRP receptor and possibly regulating SRP function through GTP hydrolysis during each cycle of signal sequence binding, protein translocation, and release (see above) .

An important breakthrough in understanding SRP core function was reported by Batey, et al. (2000). These authors describe the structure of a complex between the Ffh M domain and a portion of the 4.5S SRP RNA. The structure sheds light on important protein-RNA interactions, explains the role of the RNA component in signal sequence binding and SRP function, and is the subject of this exhibit. Before studying RNA-protein interactions and signal sequence binding, let's first examine important structural features of the RNA and protein separately.

A crystal structure (at 1.8 Angstrom resolution) of portions of the Ffh M domain complexed with domain IV of the 4.5S SRP RNA is shown at left (Batey, et al., 2000).

Domain IV of the RNA contains three loops that are conserved in SRP RNAs in all three kingdoms of life. The first, the tetraloop, lies at the tip of the domain, with the the other two, the symmetric and asymmetric loops, closely juxtaposed in the center of the domain . This juxtaposition, generated through unusual base pairing and stacking in the symmetric loop, creates a novel, flat surface in the minor groove for recognition by the M domain . Several of these non-canonical base interactions have important structural consequences, most notably:

• A G-G base pair pinches the opposite strands of the RNA backbone together. The G-G bonding is achieved partially through a potassium ion bridge.
• Dual A-C base pairs are stacked beneath the G-G, one pair of which is partially bridged by a second potassium ion.
• Base stacking of one of the Gs in the G-G pair and a C from the opposite strand cause the intervening A-C base pair to be thrust into the minor groove where their atoms can interact with backbone atoms of the M domain (see below).

An unpaired A of the asymmetric loop remains stacked in the interior of the helix, but the 4 residues on the opposite (5') side of the loop are extruded from the helix, with their bases facing outward. One of these 4 residues, A42, is unstacked and is nonconserved in other SRP RNAs, likely serving as a linker between the 3 stacked residues (A39, C40, C41) and the remaining helical RNA . A39 is universally conserved in SRPs from diverse organisms, and is positioned to interact with key residues in the symmetric loop and in the Ffh M domain . The novel arrangement of extruded bases creates a large cavity in the RNA helix that is filled with H₂Os (Hs not shown) and hydrated magnesium ions .

Turning now to the Ffh M domain, the structure reveals 5 a-helices, 4 of which (2, 2b, 3, 4) make up a helix-turn-helix (HTH) motif found in a variety of nucleic acid binding proteins . However, the mode of RNA recognition by Ffh is notably different than classic HTH binding. Canonical HTH binding involves insertion of a recognition helix (corresponding to helix 4 of the Ffh M domain) deep into the major groove of a double helix, where amino acid sidechain atoms can bond with nitrogenous base atoms. The Ffh M domain, in contrast, contacts the relatively flat surface of the juxtaposed symmetric and asymmetric loops of the RNA (see above) . The contacts utilize numerous backbone atoms of helices 2, 2b, and 3, another novel feature of Ffh-RNA recognition.

Interestingly, when the solution structures of free Ffh and domain IV of the 4.5S RNA are compared with structure of these molecules in the complex just described, it is clear that that upon binding, it is the RNA that undergoes significant conformational change, with the protein shape remaining relatively static. The extrusion of the 4 residues of the asymmetric loop is apparently induced upon binding to the M domain of Ffh, creating the large cavity filled with cations and solvent molecules. Walter, et al. (2000) speculate that the extensive network of covalently unrestrained water molecules, magnesium ions, and potassium ions may endow the the SRP with considerable conformational flexibility. This plasticity could be employed to achieve different conformational states associated with various steps in the targeting of signal sequence bearing proteins (see Introduction).

The RNA-protein interface contains extensive interactions between helices 2b and 3 of the Ffh M domain and the symmetric and asymmetric loops of domain IV of the 4.5S SRP RNA. These interactions include:

• Stacking of the universally conserved A39 of the asymmetric loop under arg398 of helix 3. Arg398 is invariant in Ffh SRP homologs in all of life's kingdoms, but, intriguingly, is not required for stable binding with SRP RNA (Kurita, et al., 1996). Perhaps the protruding arg398-A39 stack modulates GTPase activity of Ffh by allowing the arginine to facilitate GTP hydrolysis (Batey, et al., 2000).
• H bonding of A39 to arg401, another invariant residue of helix 3.
• An A39 (2'OH)-A63 (PO) H bond, helping to link the symmetric and asymmetric loops.
• A central salt bridge involving glu386 of helix 2b and arg398 & arg401 of helix 3. This salt bridge is cradled between the asymmetric and symmetric loops of the RNA. This arrangement strengthens the connection between the loops through H bonds between arg401 and A39 (asymmetric loop) and C62 (symmetric loop).

The RNA-protein interaction is also mediated by a complex network of H₂O and metal ion contacts. For example, the potassium ion that is coordinated to the G-G base pair in the symmetric loop also coordinates to the backbone carbonyl of gly405 of helix 3 .

Having considered SRP core Ffh-4.5S RNA interactions, let's now turn to the structural features of the complex that facilitate binding of signal sequence peptides.

At left is shown a model of the Ffh M domain (complete) bound to domain IV of the SRP 4.5S RNA. The tetra-, symmetric, and asymmetric RNA loops are highlighted. The proposed signal sequence binding site of the M domain is a prominent groove lined with hydrophobic residues (Keenan, et al., 1998).

Signal sequence peptides at the N termini of secreted or membrane proteins vary widely in sequence, but they are a-helical and typically comprise a hydrophobic region of 6-15 amino acids and a short terminal stretch of 2-5 positively charged residues. A short, a-helix representing an 18 amino acid signal sequence, can be modeled to fit in the hydrophobic groove of the M domain . The helix's hydrophobic stretch is buried in the hydrophobic groove of the M domain, and the positively charged terminal region projects towards the negatively charged backbone of the RNA, near the tetraloop .

Thus, the SRP core signal sequence binding site has both a protein and RNA component. This explains the heretofore mysterious role of RNA in one one of the most conserved ribonucleoproteins known.

Batey, R. T., Rambo, R. P., Lucast, L., Rha, B., Doudna, J. A. (2000). Structure of the Ribonucleoprotein Core of the Signal Recognition Particle. Science 287: 1232-1239.

Keenan RJ, Freymann DM, Walter P, Stroud RM. (1998). Structure of the Signal Sequence Binding Subunit of the Signal Recognition Particle. Cell 94: 181-91.

Kurita K, Honda K, Suzuma S, Takamatsu H, Nakamura K, Yamane KJ (1996). Identification of a region of Bacillus subtilis Ffh, a homologue of mammalian SRP54 protein, that is essential for binding to small cytoplasmic RNA. J. Biol. Chem. 271: 13140-13146.

Lodish, H., Berk, A., Matsudaira, P., Baltimore, D., and Darnell, J. (2000). Molecular Cell Biology, 4th ed. W.H. Freeman, New York.

Walter, P., Keenan, R., Schmitz, U. (2000). SRP-Where the RNA and Membrane Worlds Meet. Science 287: 1212-1213.

Back to Introduction