Different effects of Sec61α, Sec62 and Sec63 depletion on transport of polypeptides into the endoplasmic reticulum of mammalian cells

Abstract

Co-translational transport of polypeptides into the endoplasmic reticulum (ER) involves the Sec61 channel and additional components such as the ER lumenal Hsp70 BiP and its membrane-resident co-chaperone Sec63p in yeast. We investigated whether silencing the SEC61A1 gene in human cells affects co- and post-translational transport of presecretory proteins into the ER and post-translational membrane integration of tail-anchored proteins. Although silencing the SEC61A1 gene in HeLa cells inhibited co- and post-translational transport of signal-peptide-containing precursor proteins into the ER of semi-permeabilized cells, silencing the SEC61A1 gene did not affect transport of various types of tail-anchored protein. Furthermore, we demonstrated, with a similar knockdown approach, a precursor-specific involvement of mammalian Sec63 in the initial phase of co-translational protein transport into the ER. By contrast, silencing the SEC62 gene inhibited only post-translational transport of a signal-peptide-containing precursor protein.

Fig. 1.

Effect of SEC61A1 gene silencing on ribosome content of the human ER. Transfected HeLa cells were grown on poly-lysine-coated sapphire discs in 6 cm dishes for 96 hours and subjected to freeze substitution prior to the embedding. Ultrathin sections were cut parallel to the cell monolayer, collected on copper grids, stained with uranyl acetate and lead citrate, and analyzed on a Tecnai 12 Biotwin EM. Representative images are shown. The arrowheads point to putative ER cisternae. The ribosome content of the ER in control siRNA and the two SEC61A1 siRNAs was quantified as 16.9±1.2% (n=19 pictures; 82.5 μm membrane), 6.5±0.5% (n=11; 69.5 μm) and 6.5±0.7% (n=14, 55.7 μm) ribosomes/1 μm. Scale bars: 1 μm (upper row), 0.5 μm (middle row) and 0.2 μm (lower row).

Fig. 2.

Effect of SEC61A1 gene silencing on signal peptide- and tail-anchor-dependent protein transport into the human ER. Before treatment with digitonin for preparation of semi-permeabilized cells, HeLa cells were treated with the indicated SEC61A1 siRNA or control siRNA for 96 hours, where indicated, for the last 48 hours with SEC61A1 plasmid or control plasmid. A representative western blot for SEC61A1 silencing is shown in supplementary material Fig. S1A. The indicated precursor polypeptides were imported into the indicated ER fractions under co- (A,B) or post-translational (C,D) transport conditions, or incubated in the presence of buffer as a negative control. The membranes were subjected to sequestration analysis, followed by SDS-PAGE and phosphorimaging. Only the areas of interest for single gels are shown. Processing efficiency was calculated for the mature proteins in the gel shown (processed and/or glycosylated) as a percentage of the precursor plus mature protein. g, a glycosylated polypeptide; p, precursor; PK, proteinase K; Note that reticulocyte lysate contains residual amounts of ER membrane that can give rise to minute amounts of mature proteins in the absence of added membranes.

Fig. 3.

Effect of SEC62 gene silencing on signal peptide- and tail-anchor-dependent protein transport into the ER. Before treatment with digitonin for preparation of semi-permeabilized cells, HeLa cells were treated with the indicated SEC62 siRNA or control siRNA for 96 hours (A–F), where indicated; TX100, Triton X-100, for the last 48 hours with SEC62 plasmid or control plasmid (F). (A) A representative western blot documenting SEC62 silencing. The indicated precursor polypeptides were imported into the indicated ER fractions under co- (C,E) or post-translational (B,D,F) transport conditions, or incubated in the presence of buffer as a negative control. The membranes were subjected to sequestration analysis, followed by SDS-PAGE and phosphorimaging. Only the areas of interest for a single gel are shown. Processing efficiency was calculated for the gel shown as described in the legend to Fig. 2. g, glycosylated polypeptide; p, precursor; PK, proteinase K; TX100, Triton X-100.

Fig. 4.

The absence of Sec63 affects protein transport into the ER in a precursor-specific manner. (A–C) Before treatment with digitonin for preparation of semi-permeabilized cells, HeLa and NIH/3T3 cells were treated with the indicated SEC63/Sec63 siRNA or control siRNA for 96 hours and, where indicated for the last 48 hours with SEC63 plasmid or control plasmid (C). (A) A representative western blot for SEC63 silencing in HeLa cells is shown. The asterisk indicates an unrelated polypeptide in HeLa cells. (D) Murine kidney Sec63^+/+ and Sec63^−/− cells were treated with digitonin for preparation of semi-permeabilized cells. (B–D) The indicated precursor polypeptides were imported into the cellular ER under co- or post-translational (ppcecA, TA proteins) transport conditions or incubated in the presence of buffer as a negative control. Transport reactions were either followed by sequestration analysis and/or by SDS-PAGE and phosphorimaging. Only the areas of interest from single gels are shown. The efficiencies of modification by signal peptidase or oligosaccharyl transferase are indicated for the mature proteins in the gel shown as a percentage of the precursor plus mature protein. g, glycosylated polypeptide; p, precursor; PK, proteinase K.

Fig. 5.

The transport of prion protein is affected by the lack of Sec63. Murine kidney Sec63^+/+ and Sec63^−/− cells were treated with digitonin for preparation of semi-permeabilized cells. The indicated precursor polypeptides were imported into the indicated ER fractions under co-translational transport conditions. The membranes were subjected to sequestration analysis, followed by SDS-PAGE and phosphorimaging. Only the areas of interest for single gels are shown. Glycosylation efficiency was calculated for the shown gel and is given as mature protein as a percentage of precursor plus mature protein. g, glycosylated polypeptide; p, precursor; PK, proteinase K; RM, rough microsomes; TX100, Triton X-100; *PrP precursor was efficiently ubiquitylated; the characteristic ladder of mono-, di- or triple-ubiquitylated PrP was observed (Rane et al., 2008). RMs were less efficient at glycosylation than semi-permeabilized cells; therefore, processing by signal peptidase can be seen in addition to glycosylation.