Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Nov 10.
Published in final edited form as: Mol Cell. 2012 Sep 27;48(3):398–408. doi: 10.1016/j.molcel.2012.08.023

Membrane Protein TM Segments Are Retained at the Translocon during Integration until the Nascent Chain Cues FRET-detected Release into the Lipid Phase

Bo Hou 1,2, Pen-Jen Lin 1, Arthur E Johnson 1,3,4
PMCID: PMC3496027  NIHMSID: NIHMS406731  PMID: 23022384

SUMMARY

Most membrane proteins are integrated cotranslationally into the ER membrane at the translocon where nonpolar nascent protein transmembrane segments (TMSs) are widely believed to partition directly into the nonpolar membrane interior. However, a FRET approach that monitors the separation between a fluorescent-labeled TMS and fluorescent phospholipids diffusing in the bulk lipid reveals that TMSs do not immediately enter the lipid phase of the membrane. Instead, TMSs are retained at the translocon by protein-protein interactions until their release into bulk lipid is triggered by translation termination or, in some cases, by the arrival of another nascent chain TMS at a translocon. Nascent chain status and structural elements therefore dictate the timing of TMS release into the lipid phase by altering TMS and flanking sequence interactions with translocons, ribosomes, and associated proteins, thereby controlling when successive TMSs assemble in the bilayer and TMS-delineated loops fold.

INTRODUCTION

In mammalian cells, multi-spanning polytopic membrane proteins (PMPs) are integrated into the endoplasmic reticulum (ER) membrane at protein complexes termed translocons (Skach, 2009; Shao and Hegde, 2011). Since this process is co-translational, the cellular machinery must recognize each transmembrane segment (TMS) in the nascent chain, insert the TMS into the hydrophobic core of the phospholipid (PL) bilayer in the correct orientation, and direct the hydrophilic PMP loops that flank a TMS to opposite sides of the membrane. These complex operations are accomplished by the ribosome and translocon, acting as a coupled functional unit, along with other proteins in and on both sides of the membrane (Johnson, 2009; Lin et al., 2011a; Lin et al., 2011b). Yet many mechanisms involved in PMP integration are largely unknown or controversial.

One unresolved aspect of PMP integration is the nature of translocon protein involvement. Translocon proteins were linked to integration early on when they were photocrosslinked by nascent membrane proteins (Thrift et al., 1991; High et al., 1991). Probes positioned in the middle of a TMS were then found to photocrosslink phospholipids upon entering the translocon, a result that suggested the membrane interior is exposed directly to the aqueous translocon pore (Fig. 1A; Martoglio et al., 1995). Since the lipid core of the bilayer is nonpolar and TMSs are comprised primarily of nonpolar amino acids, TMSs were presumed to partition directly into the bulk lipid of the membrane from the aqueous translocon pore, driven by a thermodynamically favorable hydrophobic association. This model has been widely accepted because: (i) some TMS photocrosslinking to translocon proteins was shortlived (Mothes et al., 1997; Heinrich et al., 2000; Heinrich and Rapoport, 2003); (ii) the archaeal crystal SecYEβ structure suggested that the two clamshell domains of SecY separate to form a lateral opening for TMS passage into membrane lipid (van den Berg et al., 2004); and (iii) the efficiency of TMS integration into the bilayer correlated with total TMS hydrophobicity (Hessa et al., 2005).

Fig. 1. When does a TMS enter the bulk lipid?

Fig. 1

Open in a new tab

(A) The translocon (yellow) is viewed from above the plane of the membrane. A nascent chain TMS (black) is depicted as having entered the aqueous pore (white), from which the TMS moves directly (red arrow) into the bulk lipid of the ER membrane (gray)(Martoglio et al., 1995; Heinrich et al., 2000). (B) A TMS helix remains adjacent to translocon proteins even though more than 400 Å of nascent chain have emerged from the ribosome (blue)(Do et al., 1996) and are exposed to the cytosol (Liao et al., 1997). Ion movement through the pore is inhibited by the combined actions of, among others, BiP (green) and a J-domain-containing ER membrane protein (magenta)(Haigh and Johnson, 2002; Alder et al., 2005). (C) Acceptor-labeled PLs (BOP-PE; red) are distributed by diffusion within the bulk lipid. After release from the translocon, donor-acceptor proximity is maximal because the TMS labeled with a donor dye (green) is completely surrounded by BOP-PE. (D) A TMS donor adjacent to or inside the translocon has a reduced proximity to acceptors because BOP-PE is excluded from area occupied by the translocon and associated proteins. Distances, sizes, and concentrations are not to scale.

Yet there is considerable crosslinking evidence that various nascent chain TMSs remain adjacent to the Sec61α, Sec61β, and/or TRAM proteins of the mammalian translocon for extended periods of time (e.g., Thrift et al., 1991; Do et al., 1996; Meacock et al., 2002: Sadlish et al., 2005; Sauri et al., 2005; Kida et al., 2007). High-resolution crosslinking experiments with probes projecting from 3 or 4 different sides of a TMS helix further suggested that helices in nascent chains were not freely and randomly rotating in the lipid phase, but instead were bound at the translocon in a specific orientation relative to TRAM and/or Sec61α (Meacock et al., 2002; McCormick et al., 2003; Saksena et al., 2004; Sadlish et al., 2005; Pitonzo et al., 2009; Cross and High, 2009). Indeed, the translocon may contain multiple TMS binding sites since, for example, 4 of 6 aquaporin TMSs photocrosslinked to Sec61α at one length of nascent chain (Sadlish et al., 2005) and 3 opsin TMSs were retained at the translocon until termination (Ismail et al., 2008). Thus, a second model (Fig. 1B) suggests that TMSs do not move rapidly into the lipid phase upon reaching the translocon, but instead are retained at specific translocon locations until being released into the bulk lipid.

The two opposing models differ primarily in the extent of TMS interaction with translocon proteins and the timing of TMS immersion into bulk lipid. Since the detection of TMStranslocon protein proximity by crosslinking depends upon the location and accessibility of TMS (photo)crosslinking probes as well as TMS identity (McCormick et al., 2003; Saksena et al., 2004; Sadlish et al., 2005; Cross and High, 2009), the absence of TMS-translocon protein crosslinking does not necessarily mean the TMS has been released into the lipid phase. Especially since a fully assembled mammalian translocon includes many proteins besides the core proteins that interact with nascent proteins. These translocon-associated proteins may also bind and retain proteins, even after translation has terminated (Wilson et al., 2005). In reality, the timing of TMS entry into bulk lipid has never been determined experimentally, despite the fact that this timing dictates the initiation and kinetics of nascent membrane protein TMS assembly and loop folding during integration.

We have here used a variation of the fluorescence resonance energy transfer (FRET) technique to detect and quantify the proximity of TMSs to PLs in bulk lipid during integration and thereby determine directly when a TMS enters the lipid phase of the membrane. Among other things, this approach has revealed that individual PMP TMSs and signal-anchor TMSs do not diffuse rapidly into lipid. Instead, TMSs are retained at the translocon until their release into the lipid phase is triggered by either nascent chain cleavage from the tRNA (termination) or the arrival of another TMS at the translocon.

Results

Approach

After leaving the translocon, a TMS is immersed in the membrane's lipid phase. Since such TMSs are completely surrounded by PL molecules (Fig. 1C), the TMS-PL proximity is maximal. In contrast, the proximity between PLs and a TMS positioned in or adjacent to a translocon is substantially reduced because proteins in or associated with the translocon occupy a substantial fraction of the surface area close to the TMSs (Fig. 1D). The time (= length of nascent chain) at which a TMS is released into the bulk lipid of the membrane can therefore be determined by identifying the nascent chain length at which FRET is maximized.

When a fluorescent dye (the donor) absorbs a photon, its excitation energy can be transferred to a second dye or chromophore (the acceptor) without the emission of a photon if the acceptor has the appropriate spectral properties. The efficiency of transfer, E, depends on several variables, but most importantly on the distance between donor and acceptor. Most FRET experiments measure and/or monitor the separation between a single donor dye and a single acceptor dye at specific sites in the same molecule (e.g., Woolhead et al., 2004; Lin et al., 2011b) or molecular complex (e.g., Johnson et al., 1982). However, protein interactions and topography at the two-dimensional membrane surface are sometimes characterized using a variation of the FRET technique that introduces acceptor-labeled PL molecules into liposomes. The acceptor-PLs diffuse freely in liposomes and hence distribute randomly and uniformly in the plane of the membrane. A donor can therefore transfer energy to any acceptor that is sufficiently close. When proteins with a donor dye at a specific site bind to liposomes containing acceptor-labeled PL, the observed E can determine, among other things, the height of an enzyme active site or protein domain above the surface (e.g., Husten et al., 1987; Yegneswaran et al., 1999; Ramachandran et al., 2005), or the kinetics and thermodynamics of protein binding to the membrane (e.g., Duffy et al., 1992). This technique cannot quantify distances in natural membranes because the unknown PL surface area and distribution of membrane proteins prevent an accurate determination of acceptor density. However, changes in E with natural membranes do reflect changes in donor-acceptor separation.

To monitor TMS-PL proximity during co-translational integration, a donor dye must be positioned at a TMS in the nascent chain. Since nascent proteins are only 0.1% of the total protein in a typical sample, selective labeling can only be achieved by incorporating the donor dye into a nascent protein as it is being synthesized by the ribosome. This approach requires a modified aminoacyl-tRNA that recognizes a specific codon, but incorporates, instead of its natural amino acid, an amino acid analog with the donor dye (here 7-nitrobenz-2-oxa-1,3-diazole = NBD) covalently attached to its side chain (Johnson et al., 1976; Johnson, 2005). In this study, in vitro translations contained signal recognition particle (SRP), ER microsomes, full-length or truncated mRNAs with a single in-frame Lys or Cys codon, and either εNBD-LystRNALys (Crowley et al., 1993) or NBD-S-Cys-tRNACys (Alder et al., 2008a), respectively. Since truncated mRNAs have no stop codon, no termination occurs at the end of the mRNA, and nascent chains, each of the same length, remain bound to the ribosome as peptidyltRNAs.

Acceptor dyes were positioned at the cytosol-membrane interface by covalently attaching BODIPY 576/589 (BOP; Invitrogen) to the headgroup of dioleylphosphatidylethanolamine to yield BOP-PE (Supp. Info.; Fig. S1A). Purified BOP-PE (Fig. S1B) fused with salt-washed rough ER microsomes (KRMs), and the resulting BOPPE• KRMs were purified away from free BOP-PE micelles by sedimentation through sucrose (Fig. S1C). Since the average donor-acceptor separation is less if the surface density of acceptor dyes (σ) is high, E is higher at higher BOP-PE concentrations. Yet both targeting and signal sequence cleavage were reduced by the presence of BOP-PE (Fig. S1D). Thus, the BOP-PE•KRMs used in this study were prepared at a constant 39 pmol BOP-PE/A280 unit KRM to obtain an acceptable balance between the magnitudes of E and donor intensity (the latter reflects the number of KRM-bound proteins). This concentration amounts to 1 BOP-PE per 328 PL (0.3 mole%; Supp. Info.). We have not attempted to estimate the BOP-PE σ in KRMs because the amount and disposition of surface area occupied by proteins is unknown, as is the extent and distribution of BOP-PE association with ER membrane proteins. But the R0 for NBD-BOP FRET is 47 Å (Lin et al., 2011b) and the translocon diameter is ~100 Å in native ER membranes (Hanein et al., 1996), so E between an NBD and a BOP on opposite sides of a translocon is essentially 0.

FRET-detected TMS movement into bulk lipid

Full-length 111p, a single-spanning membrane protein (SSMP; Fig. 2A), was translated with an NBD dye positioned in the middle of its vesicular stomatitis virus G protein (VSVG) TMS and released into the bulk lipid to establish E for maximal TMS-BOP-PE proximity. E averaged 46% for fully integrated TMSs in KRMs with 0.3 mole% BOP-PE (Fig. 2B). When full-length 111p mRNAs that lacked only the stop codon were translated, the resulting integration intermediates (ribosome-translocon complexes or RTCs) contained full-length 111p molecules that were still covalently attached to ribosome-bound tRNA (Fig. 1B) and had an average E about half of that seen with 111p released into the bulk lipid (Fig. 2B). Termination of 111p translation therefore effects a substantial change in NBD-TMS proximity to BOP-PEs.

Fig. 2. SSMP TMS proximity to BOP-PE.

Fig. 2

Open in a new tab

(A) Topogenic sequences in SSMP 111p: VSVG TMS (green); SS, preprolactin signal sequence (orange). NBD (red) is attached at residue 75. (B) Average NBD-to-BOP E values (± S.D.; n = 3) are shown for 111p232 that has been terminated normally (green), released from tRNA by puromycin (yellow), or is still attached to the tRNA (magenta). (C) Nascent, normally terminated (FL), or puromycin-reacted 111p232 with an εANB-Lys probe at 75 were photolyzed, immunoprecipitated with antibodies to Sec61α (α), Sec61β (β) or TRAM (T), and analyzed by SDS-PAGE. Photoadducts containing Sec61α (●), Sec61β (▲) and TRAM (♦) are indicated; arrow, unreacted 111p. See also Fig. S1.

In this type of FRET experiment, an excited donor dye can transfer its excitation energy to any of several nearby acceptor dyes, and the observed E is a summation of donor FRET to acceptors located different distances from the donor. An increase in E could therefore result from: (i) an increase in donor quantum yield, Q, that increases R0, the distance between a single donor and a single acceptor at which E = 50%, and thereby increases the number of acceptors within range; (ii) an increase in BOP-PE σ that increases in the number of nearby acceptors and hence decreases average donor-acceptor separation; and/or (iii) an increase in lipid phase surface area around the donor that increases the number of nearby acceptors without changing σ.

NBD fluorescence lifetime (τ) is directly proportional to Q if absorbance is constant, and the average lifetimes, <τ>, were very similar before and after termination for each of the nascent chains examined (Table S1), as were the acceptor absorbance and donor emission spectra and also the donor anisotropies (0.25 before and 0.23 after termination). Thus, R0 was not significantly altered by nascent chain release. Also, the diffusion of 0.3 mole% BOPPE acceptors at the membrane surface will not be detectably affected by either termination of translation more than 100 Å above the membrane surface or the release of a few TMSs into the lipid phase. Similarly, the distribution of any BOP-PE molecules that happen to associate with microsomal proteins is unlikely to be altered by termination or TMS integration. Thus, σ is probably unchanged by nascent chain release from the RTC.

Instead, termination must alter NBD-TMS proximity to lipid surface area and its 0.3 mole% BOP-PE. After termination and TMS integration into the lipid phase, the NBD-TMSs were completely surrounded by BOP-PE in the lipid phase and E was maximal (Fig. 1C). Prior to termination, the 111p TMSs were most likely located in or alongside translocons because they occupy substantial surface area and hence markedly reduce the lipid surface area and number of BOP-PE within range of the donor without altering σ (Fig. 1D). If a TMS in a tRNA-bound 111p had partitioned into the bulk lipid prior to termination, its proximity to BOP-PE would not be altered by severing the aminoacyl bond. We conclude that the 111p TMS is retained at the translocon until termination, even though the nascent chain TMS-tRNA tether is more than 500 Å long if fully extended (Fig. 1B).

TMS release is prevented by attachment to the tRNA

The only chemical difference between the terminated and non-terminated full-length 111p samples was the ester bond between nascent protein and tRNA. To determine whether this covalent bond affected E, non-terminated RTCs were treated with puromycin, an antibiotic that reacts with a peptidyl-tRNA to release the nascent chain from the tRNA. Since exposure to puromycin increased the E for the RTCs to the same maximal level observed with fully integrated and released 111p (Fig. 2B), nascent chain release from the tRNA allowed the TMS to diffuse away from the ribosome and translocon and move into the bulk lipid. This FRET result reveals that 111p must be covalently esterified to a tRNA to remain adjacent to the translocon; otherwise, the TMS and nascent protein diffuse into the bulk lipid.

TMSs are adjacent to translocons prior to release

To further characterize TMS-translocon proximity and proteins that may be involved in retaining the TMS, a single photoreactive probe was incorporated into 111p using Nε-(5-azido-2-nitrobenzoyl)-Lys-tRNALys (εANB-Lys-tRNALys). Parallel aliquots of non-terminated, terminated, and puromycin-terminated full-length 111p were photolyzed and then immunoprecipitated by antibodies specific for Sec61α, Sec61β, or TRAM. Non-terminated 111p TMS primarily photocrosslinked TRAM and Sec61α (Fig. 2C), as seen earlier (Do et al., 1996; McCormick et al., 2003), although the occasional reaction with Sec61β suggests that all three targets are dynamically accessible to the TMS probe. Yet no TMS-translocon crosslinking was detected after 111p was released from the tRNA. Thus, the disappearance of TMS-translocon photocrosslinking coincides with FRET-detected TMS movement into the bulk lipid.

The translocon retains two TMSs

The appearance of a second TMS (TMS2) in a nascent chain elicits major structural rearrangements in and on both sides of the ER membrane (Lin et al., 2011a). To determine what effect, if any, TMS2 entry into the translocon would have on TMS1 retention, the second TMS of opsin (OP2) was inserted into 111p (Fig. 3A). The resulting PMP was designated 2TML12K1n to represent an n-residue nascent protein with 2 TMSs separated by a loop of 12 residues (Fig. 3B) and a lysine codon-specified probe in TMS1 (K1). When the NBD-TMS1 to BOP-PE FRET efficiency was quantified for full-length 2TML12K1 that was terminated either normally or by puromycin, the observed FRET efficiencies were the same as determined for 111p (Figs. 2F, 3C). Thus, PMP TMS1 was surrounded by bulk lipid after release from the tRNA and RTC. But for non-terminated 2TML12K1281 RTCs, E was 2-fold less than for terminated nascent chains (Fig. 3C). Thus, TMS1 was not released into the bulk lipid even though TMS2 had long since entered the translocon (TMS2 is tethered to the tRNA by 175 residues, ~600 Å of fully extended nascent chain). Since E was the same for TMS1 in fulllength 2TML12K1 and 111p both before and after termination, the positioning of TMS1 in or at the translocon is largely unaffected by the presence of TMS2 and the major structural changes it initiates at the translocon (Lin et al., 2011a).

Fig. 3. TMS1 and TMS2 are retained at the translocon.

Fig. 3

Open in a new tab

(A) Topogenic sequences for PMPs with an opsin 2 TMS (OP2; yellow) are identified as in Fig. 2A. NBD (red) is positioned in either TMS1 (K1) or TMS2 (K2). (B) 2TM topology before (i) and after (ii) termination. (C) E values (± S.D.; n = 3–4) of each NBD-TMS to BOP-PE FRET before and after PMP release from tRNA. (D) Photocrosslinking of TMS1 and TMS2 to translocon components before and after termination. Symbols are defined in Fig. 2C. (E) A longer TMS1-TMS2 loop had little effect on E (± S.D.; n = 4–5). (F) Inverting TMS1 and TMS2 in the PMP did not alter protein topology (Lin et al., 2011a) and had little effect on E (± S.D.; n = 3). See also Table S1.

When the donor dye was incorporated into TMS2 instead of TMS1 and the resulting full-length 2TML12K2 protein was terminated either normally or by puromycin, the observed NBD-TMS2 to BOP-PE FRET efficiencies were nearly equivalent to those seen with NBD-TMS1 (Fig. 3C). Both TMSs of the PMP were therefore integrated into the bulk lipid after being released from the tRNA. However, E was 50% less before than after termination. Thus, both TMS1 and TMS2 were retained at the translocon until the nascent chain was released from the tRNA. The translocon therefore has at least two sites that retain TMSs.

PMP TMS1 and TMS2 are adjacent to translocon proteins

Close proximity of 2TML12 to translocon proteins was detected by placing a photoreactive probe into TMS1 or TMS2 in parallel reactions. As can be seen in Fig. 3D, tRNA-bound 2TML12K1/2281 nascent chains photocrosslinked to Sec61α, Sec61β, and TRAM, while no photoadducts were evident after TMSs were released into the bulk lipid. Thus, TMS1 and TMS2 were each in close proximity to translocon proteins prior to termination, but not after.

TMS release is unaffected by loop size

After increasing the inter-TMS loop in 2TML12 by 41 residues, TMS release for the resulting 2TML53 protein (Fig. 3A) was examined as above. E increased by about two-fold for both TMS1 in 2TML53K1 and TMS2 in 2TML53K2 when translation was allowed to terminate normally (Fig. 3E). Thus, both TMS1 and TMS2 were retained by the translocon and not allowed to diffuse into the bulk lipid, despite ~140 Å of fully extended polypeptide linking the two TMSs. TMS retention by the translocon therefore appears independent of the size of the loop separating the two TMSs and the time needed to synthesize it.

Reversing bilayer orientation does not alter TMS release

When the TMSs of 2TML53 were inverted to form a new PMP, 2INVL54 (Fig. 3A), the opsin 2 and VSVG TMSs were integrated into the ER membrane successfully (Lin et al., 2011a) despite the fact that each TMS was inserted into the bilayer in a direction opposite its natural orientation (Nlum for VSVG; Ncyt for OP2). Yet exchanging the very hydrophobic and uncharged TMS1 for the less hydrophobic, charged, and longer TMS2 (Lin et al., 2011a) did not alter their FRET-detected retention by the translocon (Fig. 3F): TMS release into the bulk lipid was termination-dependent for both TMS1 and TMS2.

TMS3 is retained at the translocon

To determine the effect of TMS3 on PMP TMS retention at the translocon, the third opsin TMS (OP3) was inserted behind OP2 in 2TML12 to yield 3TML12,18 (Fig. 4A). A single donor NBD dye was incorporated into each TMS in turn, and E was determined for each NBD-TMS in full-length 3TML12,18 both before and after its release from the tRNA (Fig. 4B). After normal termination, all three NBD-TMSs had E values ≥ 40% (Fig. 4C), consistent with their being completely surrounded by bulk lipid and BOP-PE. However, just prior to termination, E for NBD-TM3 was half its maximum (Fig. 4C). TMS3 was therefore retained at the translocon and did not partition into bulk lipid until the nascent chain was severed from the tRNA.

Fig. 4. TMS3 stays at translocon, but triggers TMS1 and TMS2 release into bulk lipid.

Fig. 4

Open in a new tab

(A) Topogenic sequences for PMPs with an opsin 3 TMS (OP3; magenta) are identified as in Fig. 3A. NBD (red) is positioned in TMS1 (K1), TMS2 (K2), or TMS3 (K3). (B) 3TM topology before (i) and after (ii) termination. (C) E values (± S.D.; n = 3) of each NBD-TMS to BOP-PE FRET before and after PMP release from tRNA. (D) A longer TMS2-TMS3 loop had little effect on E (± S.D.; n = 4). (E) Photocrosslinking of TMS1, TMS2, and TMS3 to translocon components before and after termination of PMP. (F) Nascent chain length dependence of TMS2 and TMS3 photocrosslinking to Sec61α. Symbols are defined in Fig. 2C.

The FRET-detected retention of opsin TMS3 in the translocon observed here differs from the crosslinking-detected early release of opsin TMS3 reported earlier (Ismail et al., 2006). This discrepancy may result from differing flanking sequences in the two studies and/or from a conformational change that rendered a Cys residue inaccessible for crosslinking.

TMS3 triggers TMS1 and TMS2 release

TMS1 was unaffected by TMS2 arrival at the translocon (Fig. 3C, E, F), but TMS3 entry into the translocon dramatically affected both TMS1 and TMS2. Their E values were maximal both before and after termination (Fig. 4C), thereby indicating had each moved into the lipid phase prior to termination. TMS2 was also released if the TMS2-TMS3 loop was 50 residues longer (Fig. 4D). Consistent with this interpretation, no TMS1 or TMS2 photoadducts were found prior to termination after exchanging each NBD for a photoreactive probe, but TMS3 photocrosslinked TRAM, Sec61α, and Sec61β (Fig. 4E).

A comparison of the nascent chain length dependence of TMS3 arrival at and TMS2 departure from the translocon revealed a strong correlation between these events. When the nascent chain was 173 residues long, TMS2 was adjacent to Sec61α and TMS3 was not. But at 184 residues, TMS2 no longer photocrosslinked to Sec61α, but TMS3 did (Fig. 4F). The release of TMS2 and TMS1 into bulk lipid prior to termination therefore appears to be triggered by the entry of TMS3 into the translocon.

A type I signal-anchor sequence does not enter the bulk lipid immediately

The above membrane proteins were targeted to the translocon by a cleavable signal sequence to uncouple ribosomal targeting to the translocon from nascent chain integration into the bilayer. To determine whether a type I (Nlum) signal-anchor (SA) TMS enters the bulk lipid more quickly than internal TMSs, FRET was used to determine when the SA left the translocon. An NBD was incorporated into the SA of Lep1, a variant of the wild-type E. coli leader peptidase that lacks its second TMS (Fig. 5A; Mothes et al., 1997; Heinrich et al., 2000; McCormick et al., 2003), at the naturally occurring Cys 33 using NBD-S-Cys-tRNACys. After normal termination, E was 53% for Lep1 (Fig. 5B), consistent with its full release into the bulk lipid (E was higher here because SA NBDs were located nearer the membrane surface and closer to the BOPs than other TMS-NBDs). In contrast, the E for non-terminated full-length Lep1 in RTCs was about half that of the released Lep1 (Fig. 5B). The SA was therefore in or at the translocon prior to termination.

Fig. 5. SA proximity to bulk lipid and translocon proteins.

Fig. 5

Open in a new tab

(A) Location of Lep SA sequence (blue) and NBD dye (red) in Lep1 derivative (McCormick et al., 2003) that lacked the second TMS. (B) Nascent chain length dependence of NBD-SA to BOP-PE FRET E (± S.D.; n = 3–4) before and after termination. (C) Sec61α proximity to four different helical surfaces of SA was assessed by photoadduct formation (●) before and after termination. (D) Nascent chain length dependence of Sec61α photocrosslinking from four different SA probe locations. Symbols are defined in Fig. 2C.

When four successive residues (25–28) within the SA were replaced, one at a time, by εANB-Lys in the SA helix in Lep1212, only tRNA-bound Lep1 photocrosslinked to translocon proteins (Fig. 5C). No photoadducts were seen wtih Lep1 after termination. Thus, the SA was adjacent to translocon proteins until termination occurred, as was true for internal TMSs.

Lep SA occupies different sites at the RTC

To further characterize SA location relative to bulk lipid, E was measured as a function of nascent chain length. Surprisingly, the NBD at the C-terminal (cytoplasmic) end of the SA was located much closer to the lipid phase when it first emerged from the ribosomal tunnel (E = 44%) than before termination. Then, as the nascent chain lengthened, the separation between the bulk lipid and the SA C-terminus increased until E reached a constant 30% (Fig. 5B). The increased NBD-BOP separation could be explained either by SA movement within the RTC that positions NBD in or perpendicular to the plane of the membrane further from BOP-PEs and/or by the association or rearrangement of ER membrane proteins within the RTC as the nascent chain grows that displaces bulk lipid and BOP-PEs surrounding the RTC.

SA proximity to translocon proteins was examined using photoprobes that projected in four different directions from the TMS helix. None of the probes in the 80-residue nascent chain photocrosslinked to Sec61α, which indicates that SA was not adjacent to Sec61α in this RTC even though SA had exited the tunnel (Fig. 5D). The 100-residue Lep1 probes at 25, 26, and especially 28 reacted covalently with Sec61α, but no Sec61α photoadducts were formed by the probe at 27 (Fig. 5D). Thus, the 27 side of the helix was either facing away and/or too far from Sec61α to crosslink it. In contrast, only probes at 26 and 27 crosslinked to Sec61α in RTCs with nascent chains of 120–212 residues (Fig. 5C, D). The asymmetry and reproducibility of the photoadduct patterns for each nascent chain ≥100 residues reveal that the SA helix is not oriented randomly, but is bound in a specific orientation within the translocon of each RTC. However, since the 100mer pattern differs from those of the other nascent chains, the SA apparently moved rotationally and/or translationally within the translocon relative to Sec61α as the nascent chain lengthened from 80 t0 100 and then from 100 to 120. Thus, both FRET and photocrosslinking suggest that the entry of the SA into the translocon involves 3 experimentally distinguishable locations relative to both the bulk lipid and translocon core proteins.

TMS retention is temperature-independent

To determine if TMS release from the translocon was temperature sensitive, a sample was purified and prepared at 4°C and spectral measurements were made at 4°C. The same sample was then incubated at 37°C for 30 min, and spectral measurements were repeated after the sample had equilibrated back to 4°C. The same sample was then incubated a second time at 37°C for 30 min before spectral measurements were repeated at 37°C. Both the non-terminated and terminated versions of five different proteins were examined (Table 1). The 4°C data agree with the Fig. 2–5 data: TMS1 in 111p, TMS2 in 2TM, TMS3 in 3TM, and SA in Lep1 are all bound to the translocon prior to termination because the terminated E (Eterm) values are twice as large as the tRNA-bound E (EtRNA) values, and TMS2 in 3TM was in the bulk lipid prior to termination since Eterm and EtRNA were ~equal.

Table 1. Temperature dependence of NBD-TMS to BOP-PE FRET.

The microsomal BOP-PE concentration was slightly higher for these experiments, so E was slightly higher.

Samples E (%)
4°C, 4°C		 E (%)
37°C, 4°C		 E (%)
37°C, 37°C
tRNA-bound 111p281 25 ± 3 28 ± 2 17 ± 2
terminated 111pFL 47 ± 1 43 ± 4 31 ± 3
Eterminated / EtRNA-bound 1.9 1.5 1.8
tRNA-bound 2TML12K2281 22 ± 2 24 ± 3 16 ± 3
terminated 2TML12K2FL 43 ± 2 40 ± 2 31 ± 1
Eterm / EtRNA 2.0 1.7 1.9
tRNA-bound 3TML12,18K2305 44 ± 4 41 ± 2 30 ± 1
terminated 3TML12,18K2FL 48 ± 2 42 ± 2 29 ± 3
Eterm / EtRNA 1.1 1.0 1.0
tRNA-bound 3TML12,18K3305 20 ± 2 15 ± 0
terminated 3TML12,18K3FL 40 ± 2 29 ± 3
Eterm / EtRNA 2.0 1.9
tRNA-bound Lep1212 30 ± 1 23 ± 2
terminated Lep1212 53 ± 2 42 ± 5
Eterm / EtRNA 1.8 1.8
Open in a new tab

The key observation is that Eterm/EtRNA was the same at 4°C and 37°C (Table 1) for all samples. The lower absolute values of E at 37°C than at 4°C result solely from the temperature dependence of fluorescence: NBD lifetime decreases as the temperature increases (Table S1), and hence Q, R0, and E are all less at 37°C. TMS retention and release therefore appear to be insensitive to temperature-dependent variations in bulk lipid fluidity and phase separation.

Translocons bind two TMSs simultaneously

The separation between any two points in a single nascent chain can be monitored by FRET (Woolhead et al., 2004; Lin et al., 2011b). To determine the proximity of TMS1 to TMS2 at different stages of integration, a 2TML12DA1-2 (Fig. 6A) mRNA with a single amber stop codon in TMS1 and a single Lys codon in TMS2 was translated in the presence of donorlabeled εNBD-Lys-tRNALys and acceptor-labeled εBOP-Lys-tRNAamb, an amber suppressor tRNA that translates the amber codon. Following normal termination and release into the bulk lipid, Eterm was 19% (Fig. 6B); the loop linking the two TMSs prevented them from separating completely. But prior to termination, when the nascent chain was long enough for both TMSs to have entered the translocon, EtRNA was 30% and remained so while the nascent chain lengthened by more than 100 residues (Fig. 6B). The decreased E upon termination reveals that the TMSs moved relative to each other translationally and/or rotationally such that donor-acceptor separation was increased. Since the dyes were closer together when translocon-bound than after termination, the RTC must force the TMSs into a conformation that has a smaller donor-acceptor separation than the conformation the free nascent chain adopts by itself in the bilayer. Moreover, since the RTC has to bind both TMSs to reduce the distance between them, the two TMSs are located simultaneously at the translocon (cf. Fig. 3C).

Fig. 6. TMS1-TMS2 separation.

Fig. 6

Open in a new tab

(A) Donor (green) and acceptor (red) dyes were incorporated into TMS2 and TMS1, respectively. (B) E (± S.D.; n = 3) was determined before and after termination.

Discussion

Two-dimensional FRET provides a unique approach for directly detecting the insertion of nascent chain TMSs into the lipid phase of natural ER membranes. Donor-labeled TMS movement from the translocon into membrane lipid containing acceptor-labeled PLs elicits an increase in FRET efficiency because the NBD-TMS moves from a largely protein environment lacking acceptors to being surrounded by bulk lipid and BOP-PEs, thereby increasing donoracceptor proximity. The most striking result is the clear distinction between TMSs "in" and "not in" the lipid phase of the membrane: each TMS experienced a two-fold increase in E upon entering the bulk lipid. Moreover, TMS movement into lipid was actively triggered and regulated by translation termination or TMS arrival at the RTC. Thus, RTC proteins play an active and prolonged role in membrane protein integration.

What holds a TMS in the translocon? Since the extent of TMS release was the same at 4°C and 37°C, it is unlikely that a lipid phase change significantly influences TMS release activity in natural membranes. Instead, these data indicate that TMS retention is mediated by nascent chain interactions with protein components of the RTC, not by lipid fluidity and phase effects. Further evidence that retention is protein-mediated was provided by FRET data showing that TMS1 and TMS2 were closer together before than after termination. To decrease TMS1-TMS2 separation and maintain a more compressed conformation, RTC proteins must simultaneously bind to each TMS and prevent them from relaxing into their preferred, more open conformation.

What RTC proteins mediate TMS retention? A TMS helix in bulk lipid would be free to rotate, and photoreactive probes projecting from different sides of such a randomly oriented helix would react equally well with a nearby translocon protein. Yet probes projecting from 3 or 4 different helical surfaces of each TMS examined here had very different and reproducible extents of photocrosslinking to TRAM and/or Sec61α prior to termination (Fig. 5C, D; McCormick et al., 2003). Thus, none of these TMSs rotated freely at the RTC. Instead, each TMS helix bound in a specific non-random orientation to translocon protein TMSs and/or other proteins at the bilayer surfaces (e.g., ribosomal). Since FRET-detected movement of TMSs into the lipid often correlated with the loss of TMS photocrosslinking to TRAM and Sec61α (Fig. 2–5), TRAM and Sec61α appear to be involved in TMS retention. Direct interactions between nascent chain TMSs and translocon protein TMSs probably occur since changing a single residue in a TMS significantly altered its photocrosslinking targets (Heinrich et al., 2000; McCormick et al., 2003). However, ribosomes, the oligosaccharylransferase, and/or other translocon-associated proteins may also contribute to the binding of a TMS and its flanking residues since TMSs also crosslink to PAT-10 (Meacock et al., 2002; Ismail et al., 2006; Ismail et al., 2008) and importin α-16 (Saksena et al., 2006) prior to, and ribophorin I after (Wilson et al., 2005), termination of translation.

Is TMS movement through the RTC a simple one-step partitioning? By trapping Lep1 integration at different stages, SA was found in 3 distinct environments after emerging from the ribosome tunnel and prior to release (Fig. 5D). SA integration is therefore a multi-step operation. The integration of a Type II SA into the ER membrane also involves multiple steps (Devaraneni et al., 2011). Although a detailed examination of internal TMSs has yet to be done, it seems likely that all TMS passage through and release from RTCs involves multiple steps, including TMS recognition, orientation, movement, and binding prior to release.

How does one reconcile the complex SA release reported here with the opposite conclusions of Mothes et al. (1997) and Heinrich et al. (2000)? Heinrich et al. happened to place a single probe at position 28, and when it photocrosslinked to Sec61α for only a short time, they concluded that the SA partitioned quickly into the lipid phase. We also found that a probe at 28 photocrosslinked to Sec61α efficiently only when the Lep1 was 100 residues, but not when it was 120 residues or longer (Fig. 5D). However, we also observed that probes at position 26 and 27 crosslinked to Sec61α until termination (Fig. 5D). These results emphasize the importance of using the high-resolution photocrosslinking approach with 4 different probe sites, as the more comprehensive data reveal that the Lep TMS rotates and/or changes position as it moves within the RTC during integration. Moreover, the FRET data confirm that the TMS remains in the RTC until translation is terminated, but changes positions within the RTC early in integration (Fig. 5B).

What triggers TMS release into bulk lipid? Since all TMSs examined here and those in previous studies save one (Pitonzo et al., 2009) were released when translation terminated normally, it is easy to imagine that termination factor-dependent conformational changes disrupt the translocon, ribosomal, and/or other protein interactions that bind each TMS in the RTC. Yet puromycin also triggered TMS release (Fig. 2B, 3C; e.g., Pitonzo et al., 2009; Cross and High, 2009). The RTC machinery therefore appears to recognize when the nascent chain is no longer attached to a tRNA, but it is not clear how this is done. To further complicate matters, in one case a TMS was retained at the translocon after translation terminated, and its subsequent release was ATP-dependent (Pitonzo et al., 2009).

Since TMS3 arrival at the RTC triggers TMS1 and TMS2 release into the lipid (Fig. 4C, D), TMS release is regulated by more than the loss of the covalent bond between tRNA and the nascent chain. This result is significant for several reasons. First, the TMS3-dependent release of TMS1 and TMS2 constitutes an internal control of the FRET technique that shows TMS release can be detected without termination. Second, the data (Fig. 4C, 4D, 6B) show that RTCs can bind at least two TMSs simultaneously. Similarly, certain pairs of translocon-bound aquaporin and opsin TMSs were released when a third TMS arrived (Sadlish et al., 2005; Ismail et al., 2006). Yet other studies suggest that the translocon's TMS binding capacity is greater than two (Borel and Simon, 1996; Sadlish et al., 2005; Kida et al., 2007; Ismail et al., 2008). Third, since the arrival of one TMS at the RTC triggers the release of two TMSs, the mechanism of TMS passage through the translocon is more complex than an obligatory one-by-one sequential displacement model. Fourth, in the absence of termination, TMS release requires the RTC to recognize both an incoming nascent chain TMS and also the occupancy level of the RTC to determine whether an already bound TMS(s) must be released. Thus, the state of the nascent chain and its structural elements control the timing of TMS release. Other mechanisms that might trigger TMS release include TMS interactions with another TMS in the nascent chain and/or the folding of a PMP loop or domain. However, the mechanisms that govern recognition, binding, and release of nascent chain TMSs at the RTC are still obscure.

Whatever those mechanisms, the criteria for nascent chain TMS recognition by the RTC are not overly stringent since the four TMSs examined here have very different sequences, hydrophobicities, lengths, and number of charged residues (Lin et al., 2011a; McCormick et al., 2003). Moreover, inverting the order of TMSs in the protein sequence and their orientation in the bilayer did not alter the timing of their release. Thus, in terms of TMS release, RTCs appear to recognize primarily stretches of nonpolar residues as TMSs rather than specific sequences or properties, as was true of TMS recognition within the ribosome tunnel (Lin et al., 2011a).

The ramifications of regulating TMS release into the lipid phase of the ER membrane are substantial and significant. PMP biogenesis requires the insertion of TMSs into the bulk lipid of the bilayer in the correct orientation, the arrangement of TMSs within the bilayer, the folding of loops on each side of the bilayer, and the interactions of loops and/or TMSs with each other and/or with other proteins and ligands. The chronological sequence of these occurrences will be protein dependent, but TMS movement away from the RTC likely initiates the TMS assembly, loop folding, and interaction events that ultimately generate a native state. The timing of TMS release therefore regulates when successive TMS-delineated segments of membrane proteins assemble and fold during co-translational integration. Furthermore, unlike metabolic pathways that are often regulated only at the first step, the controlled release of multiple PMP TMSs would provide repeated opportunities to initiate the folding of individual domains and thereby stage their assembly into the native protein.

Experimental Procedures

Plasmids, mRNAs and tRNAs

Lysine-free PMP, 111p, and pPL plasmids are described elsewhere (Lin et al., 2011a; Lin et al., 2011b), as is the Lep1 derivative (Heinrich et al., 2000). Site-specific sequence alterations were made using QuikChange (Agilent Technologies) to obtain a unique probe location: a single amber stop codon was introduced at Lep1 residue 25, 26, 27, or 28 for photocrosslinking and at 2TML12DA1-2 residue 75 for FRET, while Cys171 and Cys177 were converted to Ala to limit Lep1 labeling to Cys33. Primary sequences were confirmed by DNA sequencing. Full-length or truncated mRNAs were transcribed in vitro using SP6 polymerase and PCR-produced DNA fragments of the desired length. Aminoacyl-tRNA analogs, prepared and purified as before (Johnson et al., 1976; Crowley et al., 1993; Alder et al., 2008b), were obtained from tRNA Probes, College Station, TX, as were SRP, ER microsomes, and wheat germ translations.

FRET experiments

E was quantified by the acceptor-dependent decrease in donor emission intensity (Supp. Info.). Four matched samples were prepared in parallel that differed only in the presence or absence of donor and/or acceptor dyes, and they were designated D (donor-containing), DA (donor- and acceptor-containing), A (acceptor-containing) and B (a blank sample with no donor or acceptor dyes). Wheat germ translations of full-length or truncated mRNAs in the presence of SRP, KRMs or BOP-PE•KRMs, and various combinations of aminoacyl-tRNA analogs yielded RTCs that were purified for the two-dimensional FRET experiments. For the point-to-point FRET experiments, RTCs were prepared and purified as before using εBOP-Lys-tRNAamb to incorporate the acceptors into TMS1 (Supp. Info.; Lin et al., 2011b).

Photocrosslinking

Translations containing εANB-Lys-tRNALys/amb were prepared, illuminated, and analyzed as before (Supp. Info.; McCormick, 2003 #3245; Lin, 2011 #4179}.

Supplementary Material

01

Acknowledgments

We thank Yiwei Miao and Yuanlong Shao for their outstanding technical assistance, and NIH grant GM26494 and Robert A. Welch Foundation Chair grant BE-0017 for support.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Alder NN, Jensen RE, Johnson AE. Fluorescence mapping of mitochondrial TIM23 complex reveals a water-facing, substrate-interacting helix surface. Cell. 2008a;134:439–450. doi: 10.1016/j.cell.2008.06.007. [DOI] [PubMed] [Google Scholar]
  2. Alder NN, Shen Y, Brodsky JL, Hendershot LM, Johnson AE. The Molecular Mechanisms Underlying BiP-Mediated Gating of the Sec61 Translocon of the Endoplasmic Reticulum. J Cell Biol. 2005;168:389–399. doi: 10.1083/jcb.200409174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Alder NN, Sutherland J, Buhring AI, Jensen RE, Johnson AE. Quaternary structure of the mitochondrial TIM23 complex reveals dynamic association between Tim23p and other subunits. Mol Biol Cell. 2008b;19:159–170. doi: 10.1091/mbc.E07-07-0669. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Borel AC, Simon SM. Biogenesis of polytopic membrane proteins: membrane segments assemble within translocation channels prior to membrane integration. Cell. 1996;85:379–389. doi: 10.1016/s0092-8674(00)81116-2. [DOI] [PubMed] [Google Scholar]
  5. Cross BCS, High S. Dissecting the physiological role of selective transmembrane-segment retention at the ER translocon. J Cell Sci. 2009;122:1768–1777. doi: 10.1242/jcs.046094. [DOI] [PubMed] [Google Scholar]
  6. Crowley KS, Reinhart GD, Johnson AE. The signal sequence moves through a ribosomal tunnel into a noncytoplasmic aqueous environment at the ER membrane early in translocation. Cell. 1993;73:1101–1115. doi: 10.1016/0092-8674(93)90640-c. [DOI] [PubMed] [Google Scholar]
  7. Devaraneni P, Conti BJ, Matsumura Y, Yang Z, Johnson AE, Skach WR. Stepwise Insertion and Inversion of a Type II Signal Anchor Sequence in the Ribosome-Sec61 Complex. Cell. 2011;146:134–147. doi: 10.1016/j.cell.2011.06.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Do H, Falcone D, Lin J, Andrews DW, Johnson AE. The cotranslational integration of membrane proteins into the phospholipid bilayer is a multistep process. Cell. 1996;85:369–378. doi: 10.1016/s0092-8674(00)81115-0. [DOI] [PubMed] [Google Scholar]
  9. Duffy EJ, Parker ET, Mutucumarana VP, Johnson AE, Lollar P. Binding of factor VIIIa and factor VIII to factor IXa on phospholipid vesicles. J Biol Chem. 1992;267:17006–17011. [PubMed] [Google Scholar]
  10. Haigh NG, Johnson AE. A new role for BiP: Gating the aqueous ER translocon pore during membrane protein integration. J Cell Biol. 2002;156:261–270. doi: 10.1083/jcb.200110074. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Hanein D, Matlack KES, Jungnickel B, Plath K, Kalies K-U, Miller KR, Rapoport TA, Akey CW. Oligomeric rings of the Sec61p complex induced by ligands required for protein translocation. Cell. 1996;87:721–732. doi: 10.1016/s0092-8674(00)81391-4. [DOI] [PubMed] [Google Scholar]
  12. Heinrich SH, Mothes W, Brunner J, Rapoport TA. The Sec61p complex mediates the integration of a membrane protein by allowing lipid partitioning of the transmembrane domain. Cell. 2000;102:233–244. doi: 10.1016/s0092-8674(00)00028-3. [DOI] [PubMed] [Google Scholar]
  13. Heinrich SU, Rapoport TA. Cooperation of transmembrane segments during the integration of a double-spanning protein into the ER membrane. EMBO J. 2003;22:3654–3663. doi: 10.1093/emboj/cdg346. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hessa T, Kim H, Bihlmaier K, Lundin C, Boekel J, Andersson H, Nilsson I, White SH, von Heijne G. Recognition of transmembrane helices by the endoplasmic reticulum translocon. Nature. 2005;433:377–381. doi: 10.1038/nature03216. [DOI] [PubMed] [Google Scholar]
  15. High S, Flint N, Dobberstein B. Requirements for the membrane insertion of signal-anchor type proteins. J Cell Biol. 1991;113:25–34. doi: 10.1083/jcb.113.1.25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Husten EJ, Esmon CT, Johnson AE. The active site of blood coagulation factor Xa. Its distance from the phospholipid surface and its conformational sensitivity to components of the prothrombinase complex. J Biol Chem. 1987;262:12953–12961. [PubMed] [Google Scholar]
  17. Ismail N, Crawshaw SG, Cross BCS, Haagsma AC, High S. Specific transmembrane segments are selectively delayed at the ER translocon during opsin biogenesis. Biochem J. 2008;411:495–506. doi: 10.1042/BJ20071597. [DOI] [PubMed] [Google Scholar]
  18. Ismail N, Crawshaw SG, High S. Active and passive displacement of transmembrane domains both occur during opsin biogenesis at the Sec61 translocon. J Cell Sci. 2006;119:2826–2836. doi: 10.1242/jcs.03018. [DOI] [PubMed] [Google Scholar]
  19. Johnson AE. The Co-translational Folding and Interactions of Nascent Protein Chains: A New Approach Using Fluorescence Resonance Energy Transfer. FEBS Lett. 2005;579:916–920. doi: 10.1016/j.febslet.2004.11.046. [DOI] [PubMed] [Google Scholar]
  20. Johnson AE. The structural and functional coupling of two molecular machines, the ribosome and the translocon. J Cell Biol. 2009;185:765–767. doi: 10.1083/jcb.200902014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Johnson AE, Adkins HJ, Matthews EA, Cantor CR. Distance moved by transfer RNA during translocation from the A site to the P site on the ribosome. J Mol Biol. 1982;156:113–140. doi: 10.1016/0022-2836(82)90462-4. [DOI] [PubMed] [Google Scholar]
  22. Johnson AE, Woodward WR, Herbert E, Menninger JR. Nε-acetyllysine transfer ribonucleic acid: A biologically active analogue of aminoacyl transfer ribonucleic acids. Biochemistry. 1976;15:569–575. doi: 10.1021/bi00648a018. [DOI] [PubMed] [Google Scholar]
  23. Kida Y, Morimoto F, Sakaguchi M. Two translocating hydrophilic segments of a nascent chain span the ER membrane during multispanning protein topogenesis. J Cell Biol. 2007;179:1441–1452. doi: 10.1083/jcb.200707050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Liao S, Lin J, Do H, Johnson AE. Both lumenal and cytosolic gating of the aqueous ER translocon pore is regulated from inside the ribosome during membrane protein integration. Cell. 1997;90:31–41. doi: 10.1016/s0092-8674(00)80311-6. [DOI] [PubMed] [Google Scholar]
  25. Lin P-J, Jongsma CG, Liao S, Johnson AE. Transmembrane Segments of Nascent Polytopic Membrane Proteins Control Cytosol/ER Targeting during Membrane Integration. J Cell Biol. 2011a;195:41–54. doi: 10.1083/jcb.201103117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lin P-J, Jongsma CG, Pool MR, Johnson AE. Polytopic Membrane Protein Folding at L17 in the Ribosome Tunnel Initiates Cyclical Changes at the Translocon. J Cell Biol. 2011b;195:55–70. doi: 10.1083/jcb.201103118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Martoglio B, Hofmann MW, Brunner J, Dobberstein B. The protein-conducting channel in the membrane of the endoplasmic reticulum is open laterally toward the lipid bilayer. Cell. 1995;81:207–214. doi: 10.1016/0092-8674(95)90330-5. [DOI] [PubMed] [Google Scholar]
  28. McCormick PJ, Miao Y, Shao Y, Lin J, Johnson AE. Cotranslational protein integration into the ER membrane is mediated by the binding of nascent chains to translocon proteins. Mol Cell. 2003;12:329–341. doi: 10.1016/s1097-2765(03)00304-6. [DOI] [PubMed] [Google Scholar]
  29. Meacock SL, Lecomte FJL, Crawshaw SG, High S. Different Transmembrane Domains Associate with Distinct Endoplasmic Reticulum Components during Membrane Integration of a Polytopic Protein. Mol Biol Cell. 2002;13:4114–4129. doi: 10.1091/mbc.E02-04-0198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mothes W, Heinrich SU, Graf R, Nilsson I, von Heijne G, Brunner J, Rapoport TA. Molecular mechanism of membrane protein integration into the endoplasmic reticulum. Cell. 1997;89:523–533. doi: 10.1016/s0092-8674(00)80234-2. [DOI] [PubMed] [Google Scholar]
  31. Pitonzo D, Yang Z, Matsumura Y, Johnson AE, Skach WR. Sequence-specific retention and regulated integration of a nascent membrane protein by the endoplasmic reticulum Sec61 translocon. Mol Biol Cell. 2009;20:685–698. doi: 10.1091/mbc.E08-09-0902. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Ramachandran R, Tweten RK, Johnson AE. The Domains of a Cholesterol-Dependent Cytolysin Undergo a Major FRET-Detected Rearrangement during Pore Formation. Proc Natl Acad Sci, USA. 2005;102:7139–7144. doi: 10.1073/pnas.0500556102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Sadlish H, Pitonzo D, Johnson AE, Skach WR. Sequential triage of transmembrane segments by Sec61α during biogenesis of a native multispanning membrane position. Nature Struct Mol Biol. 2005;12:870–878. doi: 10.1038/nsmb994. [DOI] [PubMed] [Google Scholar]
  34. Saksena S, Shao Y, Braunagel S, Summers MD, Johnson AE. Cotranslational Integration and Initial Sorting at the Endoplasmic Reticulum Translocon of Proteins Destined for the Inner Nuclear Membrane. Proc Natl Acad Sci, USA. 2004;101:12537–12542. doi: 10.1073/pnas.0404934101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Saksena S, Summers MD, Burks JK, Johnson AE, Braunagel SC. Importin α-16: a translocon associated protein that may facilitate sorting of integral membrane proteins to the nuclear envelope. Nature Struct Mol Biol. 2006;13:500–508. doi: 10.1038/nsmb1098. [DOI] [PubMed] [Google Scholar]
  36. Sauri A, Saksena S, Salgado J, Johnson AE, Mingarro I. Double-Spanning Plant Viral Movement Protein Integration into the ER Membrane Is SRP-Dependent, Translocon-Mediated, and Concerted. J Biol Chem. 2005;280:25907–25912. doi: 10.1074/jbc.M412476200. [DOI] [PubMed] [Google Scholar]
  37. Shao S, Hegde RS. Membrane Protein Insertion at the Endoplasmic Reticulum. Annu. Rev. Cell Dev. Biol. 2011;27:25–56. doi: 10.1146/annurev-cellbio-092910-154125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Skach WR. Cellular Mechanisms of Membrane Protein Folding. Nature Struct Mol Biol. 2009;16:606–612. doi: 10.1038/nsmb.1600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Thrift RN, Andrews DW, Walter P, Johnson AE. A nascent membrane protein is located adjacent to ER membrane proteins throughout its integration and translation. J Cell Biol. 1991;112:809–821. doi: 10.1083/jcb.112.5.809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. van den Berg B, Clemons WMJ, Jr, Collinson I, Modis Y, Hartmann E, Harrison SC, Rapoport TA. X-ray structure of a protein-conducting channel. Nature. 2004;427:36–44. doi: 10.1038/nature02218. [DOI] [PubMed] [Google Scholar]
  41. Wilson CM, Kraft C, Duggan C, Ismail N, Crawshaw SG, High S. Ribophorin I Associates with a Subset of Membrane Proteins after Their Integration at the Sec61 Translocon. J Biol Chem. 2005;280:4195–4206. doi: 10.1074/jbc.M410329200. [DOI] [PubMed] [Google Scholar]
  42. Woolhead C, McCormick PJ, Johnson AE. Nascent membrane and secretory proteins differ in FRET-detected folding far inside the ribosome and in their exposure to ribosomal proteins. Cell. 2004;116:725–736. doi: 10.1016/s0092-8674(04)00169-2. [DOI] [PubMed] [Google Scholar]
  43. Yegneswaran S, Smirnov MD, Safa O, Esmon NL, Esmon CT, Johnson AE. Relocating the active site of activated protein C eliminates the need for its protein S cofactor. A fluorescence resonance energy transfer study. J Biol Chem. 1999;274:5462–5468. doi: 10.1074/jbc.274.9.5462. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS406731-supplement-01.doc (245KB, doc)

RESOURCES