1. Volume 146, Issue 1, Pages 134-147 (July 2011)
Stepwise Insertion and Inversion of a Type II Signal Anchor Sequence in the Ribosome- Sec61 Translocon Complex 
Prasanna K. Devaraneni, Brian Conti, Yoshihiro Matsumura, Zhongying Yang, Arthur E. Johnson, William R. Skach 
Cell 
Volume 146, Issue 1, Pages (July 2011)
DOI: /j.cell
Copyright © 2011 Elsevier Inc. Terms and Conditions

2. Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

3. Figure 1 Nascent Chain Targeting to Sec61α
(A) Autoradiogram of truncated in vitro-translated AQP4-TM1.P showing total products (T), supernatant (S), and membrane pellet (P) fractions analyzed by SDS-PAGE. Peptidyl-tRNA bands (asterisk) and prematurely released nascent chains lacking tRNA (double asterisk) are indicated.
(B) Fraction of nascent chains (as in A) that remained ER associated ± NaCl treatment as indicated.
(C) Translation products containing a photoactive crosslinker (ɛANB-Lys) at residue 28 (and WT constructs lacking UAG codon) were UV irradiated and pelleted. Translation products were quantified by phosphorimaging (see also Figure S1). Equal amounts were immunoprecipitated with Sec61α antisera and subjected to SDS-PAGE.
(D) Quantification of photocrosslinks (as in C) after correcting for WT signal.
(E and F) Autoradiogram showing photoadducts to residues 2, 28, 44, and 65 immunoprecipitated with Sec61α (E) or TRAM (F). Graphs show mean (n ≥ 3 ± SEM).
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

4. Figure 2 TM1 Inserts Head-First into the Translocon
(A) Schematic diagram of RNC and RTC showing relative location of residues 2 and 44.
(B and C) Stern Volmer plots obtained for RNCs and RTCs truncated at codon 71 and containing ɛNBD-Lys at residue 2 (B) or 44 (C) (see also Figure S2). Ksv was determined before and after microsome permeabilization with melittin. Results show mean (n ≥ 3 ± SEM).
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

5. Figure 3
Initiation of TM1 Inversion from a Type I to a Type II Topology
Ksv values were determined as in Figure 2 for probes located at residue 2 (A) or 44 (B) in RTCs before (gray bars) and after (black bars) membrane permeabilization. Data show average Ksv values obtained for indicated chain lengths (n ≥ 3 ± SEM). (C) Schematic of RTC showing probable location of fluorescent probes (circles).
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

6. Figure 4 Nascent Chain Shielding by the RTC
(A and B) Transcripts containing L44C were translated in the absence (A) or presence (B) of ER microsomes. Pelleted RNCs and RTCs were incubated with PEG-Mal ± SDS as indicated. Pegylated (P) and unpegylated (UP) peptidyl-tRNA bands are indicated (see also Figure S3).
(C) Fraction of pegylated nascent chains (as in A and B) were quantified and plotted against chain length.
(D and E) Fraction of pegylated RTCs containing Cys44 (D) or Cys34, Cys49, Cys65 (E) following permeabilization with digitonin or melittin.
(F) Pegylation of Cys34, Cys44, and Cys49 in RTCs solubilized with TX-100.
(G) Pegylation efficiency of residue Cys9 in intact and digitonin solubilized RTCs (see also Figure S4). Graphs show mean (n = 3 ± SEM).
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

7. Figure 5
Ribosome Shielding Is Salt Sensitive, Reversible, and Required for TM1 Inversion
(A) Schematic RTC showing potential effect of NaCl before and after TM1 inversion.
(B–E) RTCs containing a Cys residue at position 34 (B), 44 (C), 49 (D), or 65 (E) were pegylated ± addition of 0.5 M NaCl. Fraction of pegylated peptidyl tRNA bands was determined as in Figure 4.
(F) RTCs containing L44C (truncated at residue 98 or 110 aa) were incubated with (lanes 3 and 4) or without (lanes 1 and 2) 0.5 M NaCl and pegylated directly (lanes 1–4) or repelleted and pegylated in the presence (lanes 7 and 10) and absence (lanes 6 and 9) of 0.5 M NaCl.
(G) Quantification of experiments (as in F) showing no preincubation (white bars), preincubation without NaCl (gray bars), or preincubation with NaCl (black bars). Mean values (n = 3, ±SEM).
(H and I) Pegylation efficiency of Cys9 in RTCs digested with RNase and treated as indicated (see also Figures S4 and S5). Graphs show mean (n = 3, ± SEM).
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

8. Figure 6
Proteolytic Susceptibility of the Ribosome-Translocon Junction
(A) Autoradiogram showing truncated peptidyl-tRNA bands ± PK digestion (downward arrows show protected bands).
(B) PK digestion of nascent chains labeled with [14C]Lys at indicated UAG stop codons. Five to 7 kDa N-terminal fragments (bracket) contain residues 2 and 44 while C-terminal peptidyl-tRNA fragments (horizontal arrow) contain residue 65. For truncation 110, intensity of latter bands reflects partial removal by PK. Double asterisks indicate released nascent chains.
(C) RTCs were subject to PK digestion as in (A), but in the presence of 0.5 M NaCl and/or digitonin as indicated. PK protected Peptidyl tRNA bands in the presence of NaCl (downward arrowhead.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

9. Figure 7 Type I Signal Anchor Insertion
(A) Schematic showing engineered N-linked glycosylation site (N-glc) and mutations used to reverse TM1 topology.
(B) PK protection of (membrane targeted) type I and type II AQP4-TM1.P ± N-glc. Asterisk indicates full-length protein (see also Figure S6). Double asterisk shows a minor population cleaved at a cryptic signal peptidase site (described previously [Foster et al., 2000]). Cleavage is observed only for truncations >133 aa (not shown). Downward arrow indicates glycosylated band (see also Figure S6). Graph shows percent of chains with translocated C terminus (type II topology, mean of two experiments).
(C and D) Mean pegylation efficiency of Cys46 for type I construct (n = 3, ± SEM). Truncations are at same sites as type II constructs, but numbering reflects addition of Arg residues (E41R3).
(E) Protease protection of type I RTCs at indicated truncations. Graph shows mean C terminus translocation efficiency for type I and type II nascent chains (n = 3, ± SEM [type II] or average of two experiments [type I]).
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

10. Figure S1
In Vitro Expression and Photocrosslinking, Related to Figure 1
(A and B) AQP4-TM1.P cDNA was truncated at indicated residues, transcribed in vitro and translated in RRL supplemented with canine pancreatic microsomes. SDS-PAGE gel shows [35S]-Met-labeled, RNase treated translation products before (A) after (B) correcting for variation in translation efficiency and microsome recovery. This method was used to quantify Sec61α photoadducts in Figure 1C,D (as described in (Sadlish et al., 2005)). Truncations 66 & 71 reflect aberrant migration of peptides which was verified by site specific labeling (not shown).
(C) AQP4-TM1.P transcripts containing a unique UAG codon at residue Val2, Leu28, Leu44 or Gln65 were truncated at the indicated sites and translated in the presence of 1 μM ɛANB-Lys-tRNAamb under safe-light conditions. Crosslinking was performed by irradiation on ice for 10 min with collimated 300–350 nM UV light from a 500-W mercury arc lamp (Oriel). Translation products were analyzed before and after UV irradiation by SDS-PAGE, EN3HANCE (Perkin-Elmer) fluorography and autoradiography. Polypeptides that read-through the stop codon are present at the bottom of each lane (asterisk). UV-specific photoadducts to predicted translocon components are indicated by downward arrowheads. Additional photoadducts to residue 65 observed at truncations likely represents crosslinking to ribosomal proteins (Liao et al., 1997).
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

11. Figure S2
Collisional Quenching of Defined Secretory and Transmembrane Proteins in RRL, Related to Figure 2
Because collisional quenching has been used to date only on substrates expressed in wheat germ extract, we first validated its ability to discriminate topology of assembled integration intermediates with known topology that were generated in the RRL translation system.
(A) cDNA of the secretory substrate preprolactin (lacking the first 9 codons (Crowley et al., 1994)) and containing a unique UAG codon at residue 64 was truncated 8 residues from the C terminus at codon 221 by PCR (sites refer to WT pPL sequence). Transcript was translated in RRL in the presence of ER microsomes and supplemented with ɛNBD-[14C]Lys-tRNAamb or [14C]Lys-tRNAamb (spectral control). Microsomes were isolated and fluorescence emission intensity of NBD (Ex468nm/Em530nm) was measured at increasing concentrations of iodide ions before and after addition of melittin. Data is presented as a Stern-Volmer plot where the slope of the line [(F0/F) - 1] versus [I-] yields a quenching constant (Ksv) for intact RTCs of 2.1 ± 0.2. Following membrane permeabilization with melittin, Ksv increased to 4.5 ± 0.2. Thus, iodide accessibility the ɛNBD-Lys probe located within the lumenal domain of a nascent translocation intermediate is restricted by the intact microsome membrane. We noted that the baseline Ksv obtained here is slightly higher than published values for shorter translocation intermediates (Crowley et al., 1994). This may be due to the presence of a minor population of improperly targeted nascent chains, associated polysomes carrying short nascent chains, or minor differences in the microsomes used here which here have not been pretreated with high salt or EDTA. Despite this, the increase in Ksv following membrane permeabilization (∆Ksv) is in good agreement with previously published results (Crowley et al., 1994; Crowley et al., 1993; Liao et al., 1997).
(B) In full length AQP4, Leu44 resides in the first extracellular loop that is cotranslationally translocated into the ER lumen (yellow circle, (Foster et al., 2000; Shi et al., 1995)).
(C) Microsomes containing in vitro-synthesized full length AQP4 with an ɛNBD-Lys probe at residue 44 were prepared and analyzed as in panel A. Stern-Volmer plot shows quenching constants obtained before and after melittin addition and confirms the lumenal localization of the probe at this position.
(D) Microsomes were processed as in panel C but ɛNBD-Lys fluorescence was measured under identical conditions 0, 32, 64, 100 min after addition of KI. Results show no appreciable change in quenching during incubation period and a 2-fold increase upon melittin permeabilization, indicating that microsomes remain stably impermeable to iodide during the experiment. Error bars represent SEM.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

12. Figure S3
Validation of PEG-Mal as a Tool for Probing Nascent Protein Topology, Related to Figure 3
(A) The secretory control protein bovine preprolactin was translated in RRL in the presence of ER microsomes. Microsomes were isolated by pelleting and incubated on ice with PEG-Mal with or without Triton X-100 for 3 hr or 5 hr. Prolactin contains 6 cysteines located at residues 34, 41, 88, 204, 221 and 229. Unpegylated (UP) and pegylated (P) protein species are indicated at right. Data shows that cysteines were stably protected from pegylation by the membrane barrier in the absence of detergent.
(B) Preprolactin cDNA was truncated codon 86 aa and translated in the presence or absence of microsomes. Translation products were pelleted and incubated with PEG-Mal for 3 hr on ice in the presence or absence of TX-100. Pegylation was observed under both conditions prior to membrane targeting (RNCs), but only after TX-100 solubilization for RTCs. Thus translocated cysteine residues that reside in the ER lumen are shielded from PEG-Mal. Unpegylated (UP) and pegylated (P) bands designate peptidyl-tRNA species that are derived from bona fide nascent chains retained in intact RTCs or RNCs. Small tRNA-released bands are also evident at the bottom of the gel. Although these latter bands undergo pegylation, they are not included in our analysis because it is unknown whether the peptidyl-tRNA bond is hydrolyzed during translation or during processing and SDS-PAGE.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

13. Figure S4
Accessibility of N-Terminal Engineered Cys Residues to PEG-Mal, Related to Figure 4 and Figure 5
(A) Schematic of AQP4-TM1.P showing sequence and location of Cys residues introduced into the N-terminal flanking residues of TM1. A cysteine was also placed 2 residues N-terminal to Met1 position in an extension (−2Cys).
(B) Pelleted microsomes containing translation products shown in A (and truncated at residue 153) were solubilized in 0.2% SDS and incubated in PEG-Mal as in Figure S3. Pegylated (P) and unpegylated (UP) peptidyl-tRNA bands are indicated. Pegylation efficiency was quantified by phosphorimaging and expressed as fraction pegylated calculated by: Fraction Pegylated = P/(U+P). Graph shows mean, n = 2. For reasons that remain unclear, pegylation observed for N-terminal residues was substantially less than C-terminal residues (Cys44) with Cys9 showing the highest efficiency (∼15%).
(C) Cys9 pegylation in intact RTCs in the presence and absence of digitonin or additional 0.5 M NaCl as indicated. Pegylated (P) and unpegylated (UP) peptidyl-tRNA bands are indicated. Results indicate luminal location of the Cys residue.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

14. Figure S5 RNase Digestion of Assembled RTCs, Related to Figure 5
(A) AQP4-TM1.P containing Cys9 was truncated as indicated and translated in RRL containing ER microsomes. Samples were treated with RNase (10μg/ml) for 10 min at 24°C and analyzed by SDS-PAGE. Puromycin treatment (1 mM) is also shown as a control. Peptidyl-tRNA band (asterisk) and released nascent chains (double asterisk) are indicated. Migration of smaller fragments is distorted by endogenous RRL globin.
(B) Translation products as in panel A but containing Cys49 instead of Cys9 were analyzed prior to and following RNase digestion. Peptidyl-tRNA (asterisk) and released peptides (double asterisk) are indicated.
(C) Microsomes containing nascent chains shown in panel B were isolated, digested with RNase, and incubated in PEG-Mal before and after membrane solubilization with TX-100. Cys49 shows no increase in pegylation following detergent treatment for truncations 88 and 98, in contrast to truncation 133. Pegylated (P) and unpegylated (UP) bands are indicated.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

15. Figure S6
OST Accessibility to N Terminus in Type I and Type II SAs, Related to Figure 7
(A) Schematic showing sequence of N-terminus, location of the engineered NSS glycosylation consensus site (N-glc) and additional mutations (K5D, K14D, and E41R3) used to convert TM1 to a type I SA. Expected topology of WT and type I mutant is shown.
(B) To test whether the N-terminal glycosylation site might be recognized by luminal OST during transient head first insertion, full length type II (WT) and the type I (mutant) proteins were translated in the presence of ER microsomes. Glycosylated bands (downward arrowhead) were confirmed by addition of an inhibitory peptide (NYT) as in panel D (not shown). Glycosylation was observed only for the type I construct and glycosylation efficiency was unaffected by addition of cyclohexamide, which slows translation rate thereby increasing the available time that the N-terminus might remain exposed to the lumen prior to TM inversion. Asterisk indicates full length protein; double asterisk, shows minor population that is cleaved by signal peptidase at a cryptic site in the fusin protein as described previously (Foster et al., 2000; Shi et al., 1995).
(C and D) Expression of truncated nascent chains further showed glycosylation occurred only on the type I construct and only after the nascent chain reached a length of more than 110 aa. At this stage of synthesis, the type II SA has moved away from its optimal crosslinking site near Sec61 and inversion of the TM has already begun. This also corresponds to loss of ribosome shielding of the type I SA and movement of C-terminal residues into the cytosol. Thus accessibility of the consensus site to OST depends not only on exposure to the ER lumen, but also on the environment of the TM within the RTC and the specific stage of topogenesis.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions