JCS046094 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. P2X2 is delivered to the ER membrane by SRP. Diagram shows the presumptive location of the first TM of P2X2Q56C-108 relative to the peptidyl transferase centre of the ribosome. Numbers indicate residues in each sub-region of the fragment and Q56C shows the position of the cysteine probe from which crosslinking mediated. A truncated 108-residue HA-tagged Q56C P2X2 fragment was synthesised in vitro in a rabbit reticulocyte lysate system for 7 minutes in the presence of ³⁵Smethionine. Ribosome-nascent chain-SRP complexes were then isolated by ultracentrifugation. H2O (ΔRM; lanes 1 to 4) or canine pancreatic rough microsome and 1 mM GTP (RM + GTP; lanes 5 to 8) was then added and the samples incubated for a further 10 minutes to facilitate nascent chain transfer. Samples were then treated with either BMH (+) or a solvent control (-) before immunoprecipitation with anti-HA (lanes 1, 2, 5 and 6), anti-SRP54 (lanes 3 and 7) or anti-Sec61α (lanes 4 and 8) antisera. The P2X2 108 nascent chain is visible at ∼15 kDa in samples following immunoprecipitation with anti-HA antiserum and BMH dependent adducts with SRP54 (see ‘54’) were detected at ∼64 kDa following immunoprecipitation with anti-SRP54 serum. Two species of P2X2 adducts with Sec61α of ∼55 kDa (α) and ∼64 kDa (αβ; see also supplementary material Fig. S3) following immunoprecipitation with anti-Sec61α antiserum. An unidentified adduct at ∼33 kDa is labelled with an asterisk. The loss of crosslinking to SRP54 coincides with the gain of adducts with Sec61α and indicates that in the presence of ER membranes, the P2X2 nascent chain has been transferred from the targeting complex to the ER translocon.

• Supplemental Figure S2 -

  Fig. S2. Membrane integration and membrane-dependant glycosylation of P2X2. (A) An mRNA transcript encoding full-length Glu-Glu epitope tagged P2X2 was translated in vitro in either the absence (lanes 1 and 2) or presence (lanes 3 and 4) of ER microsomes. In the latter case, membranes and associated components were isolated by ultracentrifugation before immunoprecipitation (IP) with an anti-Glu-Glu antibody whilst P2X2 translated in the absence of ER membrane was isolated by IP directly from the translation reaction. EndoH treatment was performed on the protein A sepharose bound polypeptides before denaturation and analysis by SDS-PAGE. Glycosylated P2X2 was found only on translation in the presence of ER membrane (lane 3, labelled P2X2ψ). Non-glycosylated P2X2 is detected at ∼60 kDa (labelled P2X2). Cartoon shows the native topology of the full-length P2X2 protein, the approximate position of the glycans and the Glu-Glu tag (GG). (B) Full-length P2X2 Glu-Glu and a 108 residue HA-tagged N-terminal fragment of P2X2 were synthesised in vitro in the presence of ER microsomes. Membranes and membrane-integrated proteins were isolated by ultracentrifugation through a high-salt sucrose cushion. The truncated P2X2 encoded by an mRNA template lacking a stop-codon, was treated with puromycin/EDTA to simulate the termination of translation and release the polypeptide from the ribosome. The ‘total’ (T) sample, and the supernatant (S) and pellet (P) fractions resulting from extraction with alkaline sodium carbonate solution, were denatured and analysed directly. Full-length P2X2 can be seen at ∼66 kDa in the T and P samples (lanes 1 and 2, labelled P2X2 FL). The truncated species runs at ∼15 kDa and is also present only in the T and P fractions (lanes 4 and 5, labelled P2X2 108), indicating that it is refractive to extraction by the alkaline and is therefore fully integrated into the membrane. (C) Truncated P2X2 transcripts of varying lengths were translated in the presence of ER membrane and isolated by ultracentrifugation through a high-salt sucrose cushion before treatment with 500 µg/ml proteinase K (pK) either in the presence or absence of 1% Triton X-100 (TX100) and analysed by SDS-PAGE. Major protected fragments are shown by white triangles that represent stalled nascent chains, proteolytically cleaved at the N-terminus. A minor unidentified product labelled with an asterisk. In each case the trapped chains showed a single predominant species that was protected from protease digestion.

• Supplemental Figure S3 -

  Fig. S3. Identification of P2X2 integration intermediate cross-linked adducts. (A) Diagram shows the experimental set up for B and C and the relative position of the cysteine probe in P2X2Q56C. (B) P2X2Q56C-127 (lanes 1 to 12) or a cysteine-null P2X2-127 (lanes 13-18) were translated in the presence of ER microsomes and treated either with cycloheximide or puromycin as indicated before crosslinking with BMH and immunoprecipitation with anti-HA, anti-Sec61α (α), anti-Sec61β (β) or a non-related serum (NR). Cycloheximide-dependent adducts to Sec61α and a secondary crosslink to Sec61α × Sec61β with the P2X2 chain are shown by ‘α’ and ‘α × β’, respectively. (C) P2X2Q56C-108HA was translated in vitro in the presence ER microsomes and subjected to crosslinking. Samples were immunoprecipitated with anti-HA and analysed directly (lanes 1 and 2) or with anti-Sec61α before denaturing and re-immunoprecipitation with anti-Sec61α (α; lane 3), anti-Myc tag (lane 4) or anti-HA tag (lane 5) antiserum to formally identify crosslinked adduct to Sec61α.

• Supplemental Figure S4 -

  Fig. S4. Quantitative comparison of immunoprecipitated P2X2 crosslinked adducts. (A) P2X2Q56C-108 was translated and crosslinked as indicated before serial dilution to 25% (lanes 3 and 4) or 50% (lanes 5 and 6) relative to the total reaction (100%, lanes 7 and 8). These samples were then subjected to IP with either the anti-HA tag antiserum (lanes 1-3, 5 and 7) or anti-Sec61α antiserum (lanes 4, 6 and 8). ‘P2X2 108’ indicates the P2X2 nascent chain, whereas ‘α’ and ‘α × β’ indicate low and high molecular mass crosslinked adducts to Sec61α, respectively. (B) Values for band intensity were standardised against the uncrosslinked nascent chain to account for any variation in translation efficiency and then normalised against the 100% input samples to give relative intensities. IP with antisera recognising Sec61α results in a significant over-representation of the crosslinking efficiency, most likely because it may also recover adducts with products that are not recognised by the anti-HA antiserum and are therefore not at the precise stage of biosynthesis intended by the experiment. In addition, competing non-crosslinked Sec61 complexes may saturate the reaction. Quantification of products recovered the HA immunoprecipitation is therefore the most robust and accurate approach and allows for a direct comparison between adducts with different components of the ER translocon.

• Supplemental Figure S5 -

  Fig. S5. In vitro translations of 25 different 127-residue N-terminal fragments of P2X2 were performed in a rabbit reticulocyte lysate system supplemented with ER microsomes and ³⁵S-methionine. Cross-linking was induced by the addition of BMH, versus a solvent only control (DMSO) before denaturation and subsequent immunoprecipitation with anti-HA (HA; lanes 1 and 2 for each variant), anti-Sec61α (α; lane 3 for each variant), anti-Sec61β (β; lane 4 for each variant), anti-TRAM (T; lane 5 for each variant) or a non-related (NR) antisera. Adducts to Sec61β are indicated by ‘β’, adducts with Sec61α are indicated by ‘α’, adducts to TRAM are indicated by ‘T’ and adducts to Sec61α × Sec61β are indicated by ‘αβ’. In the case of IPs for Sec61α, a slightly faster-migrating adduct that does not contain the HA tag was also sometimes also recovered using this antiserum (see for example I50C, compare lanes 2 and 3), and was therefore excluded from the quantitative analysis (see also Ismail et al., 2006). Adducts to a presently unidentified high molecular weight component are shown by ‘χ’. This analysis was independently repeated and quantified to inform the selection of optimal probes sites for an extended analysis of P2X2 integration.