2.A.9 The Membrane Protein Insertase (YidC/Alb3/Oxa1) Family
Mitochondria import nuclearly-encoded proteins, made in the cell cytoplasm, into the mitochondrial matrix where their mitochondrial targeting sequences are removed by proteolysis. They then export some of these proteins as well some mitochondrially-encoded proteins to the inter membrane space, or they insert them into the inner membrane. This latter export pathway requires the membrane potential. At least for those proteins that contain their N-termini in the intermembrane space, export is mediated by the Oxa1p export machinery. Bacteria also export (to the periplasm) N-tails of membrane proteins synthesized without leader sequences by a Sec (Type IIPS)-independent mechanism. These export domains do not bear a net positive charge but are neutral or negatively charged. Insertion of a membrane protein via the Oxa1p export machinery follows the 'positive-inside' rule for membrane protein topology. Conserved negative charges in the transmembrane segments of subunit K of the NADH:ubiquinone oxidoreductase determine its dependence on YidC for membrane insertion (Price and Driessen, 2010). The functions and mechanisms of YidC homologues have been reviewed by Dalbey et al. 2014 and Hennon et al. 2015. The crystal structure of full-length E. coli YidC revealed that a hydrophilic groove, formed by five transmembrane helices, is a conserved structural feature of YidC, as compared to the previous YidC structure from Bacillus halodurans, which lacks a periplasmic domain. Structural mapping of the substrate- or Sec protein-contact sites suggested the importance of the groove for the YidC functions as a chaperone and an insertase (Kumazaki et al. 2014).
Homologues of the yeast Oxa1 protein are found in chloroplasts of plants and in a wide variety of bacteria. The chloroplast albino 3 (ALB3) protein appears to integrate the light harvesting chlorophyll-binding protein into thylakoid membranes using a pathway that is distinct from the chloroplast Sec translocation pathway. The matrix exposed C-terminal α-helical domain of Oxa1 can bind mitochondrial ribosomes to facilitate co-translational insertion of proteins into the mitochondrial membrane (Jia et al., 2003; Szyrach et al., 2003). The role of YidC homologues in membrane protein biogenesis is discussed by Wang and Dalbey (2011). The Pf3 phage coat protein requires the presence of the YidC membrane insertase, but not the Sec system for membrane insertion, whereas mutants that have a membrane-spanning region with increased hydrophobicity could spontaneously insert into the liposomes without YidC (Ernst et al., 2011).
Oxa1p homologues are found ubiquitously in all living organisms. While Gram-negative bacteria only have one homologue, several Gram-positive bacteria and archaea have two. Yeast encode two distant YidC homologues, Oxa1 and Oxa2. The latter can replace E. coli YidC for Sec-independent insertion of proteins (Bloois et al., 2007). Eukaryotes encode in their genomes between 1 and 6 paralogues (Yen et al., 2001). Mitochondria have two, one for co-translational, and one for post-translational insertion of membrane proteins. The former but not the latter has a ribosome binding domain (Preuss et al., 2005). Knock out mutants in the human mitochondrial Oxa1 (impairs biogenesis of the F-type ATPase and the NADH dehydrogenase complex I, but not complexes III and IV (Stiburek et al., 2007).
Oxa1 homologues exhibit 3 (Oxa1p of S. cerevisiae), 4 (the E. coli 60 kDa inner membrane protein, YidC), or 5 (the Pseudomonas putida 60kDa protein (spP25754)) putative TMSs. One TMS occurs at the N-termini of the bacterial proteins while the rest are in the C-terminal domain. Thus the P. putida protein (560 aas) exhibits putative TMSs at positions 7-23, 343-361, 371-394, 434-458 and 516-535. The E. coli protein exhibits putative TMSs at positions 6-26, 350-370, 420-440 and 499-515. The yeast Oxa1 protein lacks the first 150 residues of the bacterial proteins and thus lacks the N-terminal TMS. Its putative TMSs are at positions 119-139, 200-220 and 282-302. Co-translational membrane insertion of mitochondrially encoded proteins using the Oxa1 apparatus has been demonstrated (Ott and Herrmann, 2010). Oxa1 directly binds to mitochondrial ribosomes and, together with the inner membrane protein Mba1, aligns the polypeptide exit tunnel of the ribosome with the insertion site at the inner membrane.
The E. coli YidC protein has been shown to associate with the SecYEG complex (TC #3.A.5) (Scotti et al., 2000). It also forms a tetrameric complex with SecDF-YajC (Chen et al., 2002). It is essential for the insertion of several SecYEG-independent integral membrane proteins (Nouwen and Driessen, 2002; Serek et al., 2004) and facilitates insertion of some Sec-dependent membrane proteins (van der Laan et al., 2005). For example, YidC supports the folding of MalF into a stable conformation before it is incorporated into the maltose transport complex (Wagner et al., 2008). YidC is also required for the functional insertion of both cytochrome oxidase (3.D.4) and the F-type ATPase (3.A.2) of E. coli, and the same is true for Oxa1 in mitochondria (van der Laan et al., 2003). Subunits a and b of the E. coli F-type ATPase are inserted by a Sec/SRP/YidC-dependent) pmf-independent mechanism, but the c-subunit is inserted by a Sec/SRP-independent, YidC, pmf-dependent mechanism (Yi et al., 2004; Yi and Dalbey, 2005). YidC catalyzes c-subunit insertion prior to its oligomerization (Kol et al., 2006). Yuan et al. (2007) have shown that mutants in TMS3 of E. coli YidC lead to a cold sensitive phenotype and affect substrate protein affinity. The crystal structure of the major periplasmic domain of the E. coli YidC reveals a β-sandwich, one edge of which binds SecF (Oliver and Paetzel, 2007). The crystal structure of the periplasmic domain of the E. coli YidC reveals a conserved substrate binding domain (Ravaud et al., 2008). X-ray structures of YidC from Bacillus halodurans and Escherichia coli revealed a hydrophilic groove that is accessible from the lipid bilayer and the cytoplasm. Chen et al. 2017 explored the water accessibility within the conserved core region of the E. coli YidC. As expected from the structure, YidC possesses an aqueous membrane cavity localized to the membrane inner leaflet. The lipid-exposed transmembrane helices 3, 4, and 5 are short, leading to membrane thinning around YidC.
A single TMS protein with a highly hydrophobic transmembrane segment that inserts into the membrane by a YidC/Sec-independent mechanism becomes YidC-dependent if negatively charged residues are inserted into the translocated periplasmic domain or if the hydrophobicity of the transmembrane segment is reduced by substituting polar residues for nonpolar ones (Zhu et al. 2013). This suggests that charged residues in the translocated domain and the hydrophobicity within the transmembrane segment are important determinants of the insertion pathway. The addition of a positively charged residue to either the translocated region or the transmembrane region can switch the insertion requirements such that insertion requires both YidC and SecYEG.
In E. coli, YidC also inserts the phage coat protein Pf3 and the M13 procoat protein in vivo and in vitro independently of other proteins (Serek et al., 2004; Stiegler et al., 2011). However, it has little or no effect on the export of periplasmic secretory proteins (Samuelson et al., 2000). YidC exhibits different structural requirements for Sec-dependent versus Sec-independent membrane protein insertion and also for Sec-independent insertion of two different phage coat proteins (Chen et al., 2003). The two driving forces for membrane protein insertion seem to be (1) the PMF and (2) hydrophobic forces (van der Laan et al., 2005), and both can compensate for each other. A projection structure of YidC obtained by electron cryomicroscopy revealed that the E. coli YidC forms dimers in the membrane, and each monomer has an area of low density that may be part of the path transmembrane segments follow during their insertion (Lotz et al., 2008).
In Neurospora crassa, a tetrameric Oxa1p is likely (Nargang et al., 2002). Oxa1 complements YidC of E. coli for the insertion of Sec-independent proteins but cannot take over the Sec-associated function of YidC (van Bloois et al., 2005). The YidC Sec-independent function may be conserved and essential.
In Bacillus subtilis, there are two YidC homologues, SpoIIIJ and YqjG. Neither is essential for vegetative growth, but a double knockout mutant is lethal (Murakami et al., 2002). SpoIIIJ alone is essential for sporulation. Both proteins localize to the plasma membrane in vegetative cells but to the polar and engulfment septa in sporulating cells. These two proteins have different but overlapping functions. They function both in membrane protein biogenesis and in protein secretion (Tjalsma et al., 2003).
The structures of both the E. coli YidC (TC#2.A.9.3.1) and the Saccharomyces cerevisiae Oxa1 bound to the E. coli ribosome nascent chain complexes have been determined by cryo-electron microscopy (Kohler et al. 2009). Dimers of YidC and Oxa1 are localized above the exit of the ribosomal tunnel. Crosslinking experiments showed that the ribosome specifically stabilizes the dimeric state. Functionally important and conserved transmembrane helices of YidC and Oxa1 were localized at the dimer interface by cysteine crosslinking. Both Oxa1 and YidC dimers contact the ribosome at ribosomal protein L23 and conserved rRNA helices 59 and 24, similarly to what was observed for the nonhomologous SecYEG translocon. Dimers of the YidC and Oxa1 presumably form the insertion pores and share a common overall architecture with the SecY monomer.
Kumazaki et al. 2014 determined the crystal structure of YidC from Bacillus halodurans at 2.4 Å resolution. The structure revealed a novel fold, in which five conserved transmembrane helices form a positively charged hydrophilic groove that is open towards both the lipid bilayer and the cytoplasm, but closed on the extracellular side. Structure-based in vivo analyses revealed that a conserved arginine residue in the groove is important for the insertion of membrane proteins by YidC. They proposed an insertion mechanism for single-spanning membrane proteins, in which the hydrophilic environment generated by the groove, recruits the extracellular regions of substrates into the low- dielectric environment of the membrane.
YidC insertases interact with their substrates in a groove-like structure at an amphiphilic protein-lipid interface, thus allowing the transmembrane segments of the substrate to slide into the lipid bilayer. The high-resolution structures of YidC provide new mechanistic insights of how transmembrane proteins achieve the transition from an aqueous environment in the cytoplasm to the hydrophobic lipid bilayer environment of the membrane (Kuhn and Kiefer 2016). YidC prevents LacY from misfolding by stabilizing the unfolded state from which LacY inserts structural segments stepwise into the membrane until folding is completed. During stepwise insertion, YidC and the membrane together stabilize the transient folds. Remarkably, the order of insertion of structural segments is stochastic, indicating that LacY can fold along variable pathways toward the native structure (Serdiuk et al. 2016).
The reaction believed to be catalyzed by the Oxa1 apparatus is:
(1) protein (cytoplasm) → protein (membrane) or
(2) protein (or protein fragment) (in) → protein (or protein fragment) (out)