Efficient detection of proteins retro-translocated from the ER to the cytosol by in vivo biotinylation

Abstract

Retro-translocation from the ER to the cytosol of proteins within the secretory pathway takes place on misfolded molecules that are targeted for degradation by the cytosolically located 26S proteasome complex. Retro-translocation occurs also for other proteins (such as calreticulin) that, despite being synthesized and transported to the ER, are in part dislocated to the cytosol. We have taken advantage of the E. coli derived biotin-ligase (BirA) expressed in the cytosol of mammalian cells to specifically biotin-label in vivo proteins within the secretory pathway that undergo retro-translocation. We validated the method using four different proteins that are known to undergo retro-translocation upon different conditions: the human trans-membrane protein MHC class-I α chain (MHC-Iα), the Null Hong Kong mutant of the secretory α1 anti-trypsin (NHK-α1AT), the immunoglobulin heavy chain (HC) and the ER chaperone calreticulin (Crt). We observed specific mono-biotinylation of cytosolically dislocated molecules, resulting in a novel, reliable way of determining the extent of retro-translocation.

Figure 1. Scheme of in vivo biotinylation of retro-translocated proteins and of genetic constructs.

(a), membrane or secretory proteins with the BAP-tag localized to the luminal side of the ER. Upon retro-translocation the cytosolic BirA covalently adds a biotin to the single acceptor lysine within BAP. (b), scheme of the principal constructs used. The BAP and SV5 tags are indicated. In each case the signal leader peptide (sec) is indicated.

Figure 2. Retro-translocation of MHC-Iα.

WB-ra of cellular extracts of HEK293 cells transfected with: (a) BAP-MHC-Iα and MHC-Iα-BAP, co-expressed with cyt-BirA, or BAP-MHC-Iα co-expressed with sec-BirA, as indicated; (b) BAP-MHC-Iα and cyt-BirA and US2 or the US2 mutants US2-ΔC and US2-C133S (left panel), or US11 or US11 mutant US11-Q192L (right panel), as indicated. (c) Cytofluorimetry of HEK293 cells co-transfected with BAP-MHC-Iα and either cyt-BirA (alone or with US2 and US11), or sec-BirA, and stained with anti-SV5 mAb (left panel) or Streptavidin-QuantumDot (right panel). (d) WB-ra of supernatants of HEK293 cells co-transfected with a secretory BAP-tagged scFv control protein and cyt-BirA, BAP-MHC-Iα and US2 or US11. (e) WB-ra of cellular extracts of HEK293 cells co-transfected with t-BAP-mIgE, cyt-BirA and US2 or US11, as indicated. All blots were developed with anti-SV5 mAb, and anti-tubulin where indicated.

Figure 3. Proteasome inhibition effect on biotinylation of MHC-Iα.

(a) WB-ra of cellular extracts of HEK293 cells co-transfected with BAP-MHC-Iα and cyt-BirA (control) and, where indicated, with US2 or US11 in the absence (−) or presence of MG132 (MG; 50 µM for 4 h) or Bortezomib (Bort.; 50 µM for 4 h). (b) Quantification of the relative levels of biotinylated MHC-Iα shown in (a) calculated as the ratio between biotinylated vs. non-biotinylated form in a given lane. (c) WB-ra of PNGaseF (PNG) treated, affinity-purified biotinylated BAP-MHC-Iα, derived from MG132-treated and untreated cells co-expressing US2. All blots were developed with anti-SV5 mAb and where indicated with anti-tubulin. Open arrowheads indicate de-glycosylated BAP-MHC-Iα.

Figure 4. Trypsin sensitivity of biotinylated MHC-Iα.

WB-ra of microsomes-containing cell lysates (microsomes) derived from cells expressing BAP-MHC-Iα, cyt-BirA and US2 or US11 and treated or not with trypsin, as indicated. As a control, an aliquot of cells was directly lysed in SDS sample buffer (total). Before lysis cells were treated with 10 µM MG132 for 16 h. Open arrowheads indicate de-glycosylated molecules, while arrows indicate MHC-Iα with the cytosolic C-terminal tail digested by trypsin.

Figure 5. Retro-translocation of secretory proteins.

WB-ra of supernatants and/or cellular extracts, as indicated, of HEK293 cells co-transfected with: (a), α1AT-BAP or the mutant NHK-α1AT-BAP and cyt-BirA; (b), NHK-α1AT-BAP and cyt-BirA in the presence of MG132 (10 µM for 16 h) and digested or not (mock) with PNGaseF. Open arrow and arrowheads indicate de-glycosylated full-length NHK-α1AT-BAP and NHK-α1AT-BAP fragments, respectively, while filled arrow and arrowheads indicate the corresponding biotinylated bands. (c) WB-ra of microsomes-containing cell lysates (microsomes) derived from cells expressing NHK-α1AT-BAP and cyt-BirA and treated with MG132 (10 µM for 16 h) and, where indicated, digested with trypsin. NP40 indicates the same microsomes-containing lysates treated with detergent to solubilise ER membranes, thus making also luminal proteins accessible to trypsin. (d) WB-ra of supernatants and cellular extracts from cells expressing HC-BAP (HC) and cyt-BirA with (+) or without (−) LC. All blots were developed with anti-SV5 mAb or anti-tubulin; the LC in (d), left panel, was visualized because of the secondary anti-mouse IgG antibody used.

Figure 6. Retro-translocation of calreticulin.

(a) WB-ra of cellular extracts (developed with anti-SV5 mAb) of HEK293 cells co-transfected with Crt or the mutant Crt-ΔC and cyt-BirA. Two representative different experiments are shown. (b) Quantification of the relative levels of biotinylation of Crt and Crt-ΔC (obtained from the WB-ra) when co-expressed with cyt-BirA or sec-BirA. Histograms show the results of three independent experiments; error bars indicate one standard deviation.

Figure 7. Determination of retro-translocation by ELISA.

(a) Scheme of the ELISA used to monitor biotinylation of the retro-translocated fraction. Anti-SV5 mAb coating ensures capture of both, biotinylated and not-biotinylated BAP-MHC-Iα, which are then revealed with HRP-conjugated StrAv (only biotinylated MHC-Iα) and with anti-roTag (total MHC-Iα). (b) Retro-translocation levels (developed with HRP-conjugated StrAv) of BAP-MHC-Iα in HEK293 cells transfected alone or with US2 or US11, in the presence or absence of MG132 (50 µM for 4 h), as indicated. (c) Proportion of retro-translocated BAP-MHC-Iα (developed in parallel with HRP-conjugated StrAv and anti-roTag) in HEK293 cells transfected alone or with US2 or US11, expressed as the fraction of biotinylated BAP-MHC-Iα relative to the total amount of BAP-MHC-Iα. Histograms show the results of three independent experiments; error bars indicate one standard deviation. (d) WB-ra of samples used in (c), developed with anti-SV5 mAb or anti-tubulin. The arrow indicates the position of the non-biotinylated MHC-Iα.