We extended thermal proteome profiling to detect transmembrane protein–small molecule interactions in cultured human cells. When we assessed the effects of detergents on ATP-binding profiles, we observed shifts in denaturation temperature for ATP-binding transmembrane proteins. We also observed cellular thermal shifts in pervanadate-induced T cell–receptor signaling, delineating the membrane target CD45 and components of the downstream pathway, and with drugs affecting the transmembrane transporters ATP1A1 and MDR1.
Subscribe to Journal
Get full journal access for 1 year
only $21.58 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Martinez Molina, D. et al. Science 341, 84–87 (2013).
Savitski, M.M. et al. Science 346, 1255784 (2014).
Huber, K.V. et al. Nature 508, 222–227 (2014).
Jafari, R. et al. Nat. Protoc. 9, 2100–2122 (2014).
Miettinen, T.P. & Bjorklund, M. Mol. Pharm. 11, 4395–4404 (2014).
Franken, H. et al. Nat. Protoc. 10, 1567–1593 (2015).
Overington, J.P., Al-Lazikani, B. & Hopkins, A.L. Nat. Rev. Drug Discov. 5, 993–996 (2006).
Huber, K.V. et al. Nat. Methods doi:10.1038/nmeth.3590 (21 September 2015).
Bantscheff, M. et al. Nat. Biotechnol. 25, 1035–1044 (2007).
Bantscheff, M. et al. Nat. Biotechnol. 29, 255–265 (2011).
O'Shea, J.J., McVicar, D.W., Bailey, T.L., Burns, C. & Smyth, M.J. Proc. Natl. Acad. Sci. USA 89, 10306–10310 (1992).
Matsumoto, M. et al. Proteomics 9, 3549–3563 (2009).
Krejsa, C.M. et al. J. Biol. Chem. 272, 11541–11549 (1997).
Dai, R., Wilson, D.J., Geders, T.W., Aldrich, C.C. & Finzel, B.C. ChemBioChem 15, 575–586 (2014).
Chen, G. et al. Hum. Mol. Genet. 21, 4530–4536 (2012).
Fantus, I.G., Kadota, S., Deragon, G., Foster, B. & Posner, B.I. Biochemistry 28, 8864–8871 (1989).
Garrod, D.R., Fisher, C., Smith, A. & Nie, Z. Cell Adhes. Migr. 2, 161–166 (2008).
Hsu, F. & Mao, Y. Biochim. Biophys. Acta 1851, 698–710 (2015).
Laursen, M., Yatime, L., Nissen, P. & Fedosova, N.U. Proc. Natl. Acad. Sci. USA 110, 10958–10963 (2013).
Wallstab, A., Koester, M., Bohme, M. & Keppler, D. Br. J. Cancer 79, 1053–1060 (1999).
Becher, I. et al. ACS Chem. Biol. 8, 599–607 (2013).
Stoevesandt, O. et al. Mol. Cell. Proteomics 6, 503–513 (2007).
Kruse, U. et al. Leukemia 25, 89–100 (2011).
Werner, T. et al. Anal. Chem. 86, 3594–3601 (2014).
Savitski, M.M. et al. Anal. Chem. 83, 8959–8967 (2011).
Savitski, M.M. et al. J. Am. Soc. Mass Spectrom. 21, 1668–1679 (2010).
Savitski, M.M. et al. J. Proteome Res. 12, 3586–3598 (2013).
Omasits, U., Ahrens, C.H., Muller, S. & Wollscheid, B. Bioinformatics 30, 884–886 (2014).
We thank J. Stuhlfauth for cell culture; M. Jundt, K. Kammerer, M. Klös-Hudak, M. Paulmann, I. Tögel and T. Rudi for expert technical assistance; and C.-W. Chung, J.W. Polli and M. Zamek-Gliszczynski for helpful suggestions.
F.B.M.R., D.E., T.W., H.F., D.C., C.D., M.F.S., M.B., M.M.S. and G.D. are employees and/or shareholders of Cellzome GmbH and GlaxoSmithKline, companies that funded the work.
Integrated supplementary information
Supplementary Figure 1 Mild detergents do not resolubilize heat-precipitated proteins and thus are in principle compatible with TPP.
K562 cells were incubated at 37 °C or at 70 °C for two minutes, followed by cell lysis in PBS, or PBS supplemented with a set of nonionic, ionic and zwitterionic detergents at the concentrations indicated. Each treatment was performed as full biological replicate. Soluble fractions were analyzed by SDS-PAGE followed by staining of proteins with colloidal Coomassie. The result indicates that cell lysis after heat treatment in the presence of mild detergents does not lead to resolubilization of aggregated proteins, unlike SDS extraction which does resolubilize heat-precipitated proteins.DDM = n-Dodecyl-β-D-Maltopyranoside; OTG = n-octyl-β-d-thioglucopyranoside; No det. = no detergent
Supplementary Figure 2 The mild detergent NP-40 did not substantially affect the thermal stabilization of ATP-binding proteins after the addition of ATP to cell extract.
(a) 293 proteins with the GO annotation as “ATP-binding” and 1,233 other proteins pass the requirements for melting curve fitting and reproducibility in the vehicle conditions (for details see Methods section) in both +NP-40 and -NP-40 experiments. (b) In the set of proteins shown in (a), 136 “ATP-binding” proteins and 70 other proteins are stabilized upon addition of ATP by more than 1°C in both replicates of the +NP-40 condition, (c) 123 “ATP-binding” and 52 other proteins are stabilized by more than 1 °C in both replicates of the –NP-40 condition and (d) 99 “ATP-binding” and 27 other proteins are stabilized by more than 1 °C in both replicates of both conditions.Note that because most ATP-binding proteins have an ATP affinity in the high µM to mM range, they are expected to exhibit only small Tm shifts upon addition of excess ATP. To assess the coverage of “ATP-binding” proteins in the +NP-40 and -NP-40 conditions, we considered shifts exceeding 1 °C in both biological replicates in either condition. The less strict requirement of 1 °C stabilization (compared to our normal requirement of a p-value below 0.05, see Methods) is expected to lead to false positives in the non-“ATP-binding” proteins, but the main fraction of stabilized proteins are still “ATP-binding” proteins. In particular, when proteins stabilized by 1 °C in both conditions are considered, the proportion of stabilized non-ATP binding proteins drops substantially as one would expect as a result of small Tm shifts becoming more robust with more experiments. The variation in the number of ATP binding proteins between the detergent and non-detergent condition as well as the overlap are in line with the typical reproducibility (i.e., around 80%) expected from biological replicates analyzed by shotgun proteomics.
Supplementary Figure 3 TCR pathway components that showed thermal shifts after pervanadate treatment.
Many components of the T-cell receptor pathway exhibited thermal shifts after pervanadate treatment. Of the 47 proteins mapped to this pathway, 34 were identified with 11 being significantly shifted (data analysis by Ingenuity Pathway Analysis software). The T-cell receptor pathway had the second highest P value (7.9×10−7) after the 3-phosphoinositide degradation pathway (1.7×10−7) (Supplementary Table 7).
Several components of the 3-phosphoinositide degradation pathway exhibited thermal shifts after pervanadate treatment. Of the 155 proteins mapped to this pathway, 66 were identified with 17 being shifted by more than 1.5 °C (data analysis by Ingenuity Pathway Analysis software - data in Supplementary Table 7). Notably all but one of the 17 pervanadate-affected proteins represent phosphatases:ACP1 phosphotyrosine protein phosphataseCDC25B M-phase inducer phosphatase 2KIAA1274 Prot-tyrosine_phosphatase-likeMTM1 Prot-tyrosine_phosphatase-likeMTMR1 Lipid phosphatase with CX5R motif (Prot-tyrosine_phosphatase-like)MTMR14 Lipid phosphatase with CX5R motif (Prot-tyrosine_phosphatase-like)MTMR2 Lipid phosphatase with CX5R motif (Prot-tyrosine_phosphatase-like)MTMR3 Lipid phosphatase with CX5R motif (Prot-tyrosine_phosphatase-like)MTMR4 Lipid phosphatase with CX5R motif (Prot-tyrosine_phosphatase-like)PPP3CA Protein phosphatase 3, catalytic subunit, alpha isozymePPP4R1 Serine/threonine-protein phosphatase 4PTPN2 Tyr-protein phosphatasePTPN6 Tyr-protein phosphatasePTPRC Tyr-protein phosphataseRNGTT Prot-tyrosine_phosphatase-likeRASA1 Ras GTPase activating protein - tyrosine phosporylatedSYNJ1 Lipid phosphatase with CX5R motif
Supplementary Figure 6 Treatment of K562 cells with ouabain changed the thermal stability of Na+ and K+ pump subunits.
(a) Predicted topology of the Na+/K+ pump α-subunit ATP1A1 in the plasma membrane (wlab.ethz.ch/protter/) and chemical structure of ouabain. (b) Ouabain treatment of K562 cells leads to altered thermal behavior of ATP1A1. Cells were treated with 1 µM ouabain for 1 h, subjected to heat-treatment across the indicated temperature range and lysed in PBS supplemented with 0.4% NP-40. After centrifugation soluble fractions were analyzed by western blot using an ATP1A1 antibody. The 37 °C fraction derived from vehicle treated cells was loaded on both gels which allowed the calculation of a correction factor to compensate for differences in protein transfer efficiency. Western blot results from three full biological replicates are shown. (c) Concentration-dependent stabilisation of the Na+/K+ pump subunits ATP1A1 and ATP1B3. Two independent isothermal dose-response experiments were performed at 63 °C with ouabain-treated K562 cells. Soluble fractions were analyzed by mass spectrometry.
Supplementary Figure 7 Treatment of Caco-2 cells with elacridar led to thermal destabilization of the cognate target transporter MDR1.
(a) Predicted topology of MDR1 in the plasma membrane (wlab.ethz.ch/protter/) and chemical structure of elacridar. (b) Elacridar treatment of Caco-2 cells leads to altered thermal behavior of MDR1. Cells were treated with 1 µM elacridar for 1 h, trypsinized, subjected to heat-treatment across the indicated temperature range and lysed in PBS supplemented with 0.8% NP-40. After centrifugation soluble fractions were analyzed by western blot using an anti-MDR1 antibody. The 37 °C and 41 °C fractions derived from vehicle treated cells were loaded on both gels to enable correction of differences in protein transfer efficiency. Data shown summarize a representative experiment out of three independent experiments done. (c) Concentration-dependent destabilisation of MDR1 – western blot analysis. Caco-2 cells were treated with a concentration range of elacridar, followed by heating to either 54 °C or 56 °C (four replicates per temperature) and lysis in PBS supplemented with 0.8% NP-40. After centrifugation soluble fractions were analyzed by western blot using an anti-MDR1 antibody. To all SDS-PAGE gels an aliquot of untreated HepG2 cell lysate was added as reference (labelled ”Ref.”). (d) Concentration-dependent destabilisation of MDR1- mass spectrometry analysis. Five conditions of replicate four, 54 °C (vehicle and compound treated samples at 500 nM, 100 nM, 20 nM and 0.8 nM elacridar) and the corresponding conditions of replicate four, 56 °C were analyzed by mass spectrometry.
Supplementary Figures 1–7 (PDF 1525 kb)
Thermal proteome profiling data showing effect of ATP on proteins in lysate extracted in absence of detergent. (XLSX 7182 kb)
Thermal proteome profiling data showing effect of ATP on proteins in lysate extracted in presence of detergent. (XLSX 8547 kb)
Comparison of effect of ATP in detergent versus non-detergent conditions (XLSX 1567 kb)
Thermal proteome profiling data obtained in intect cells showing effect of pervanadate on proteins and subsequent lysis in absence of detergent. (XLSX 8378 kb)
Thermal proteome profiling data obtained in intect cells showing effect of pervanadate on proteins and subsequent lysis in presence of detergent. (XLSX 9894 kb)
Comparison of effect of pervanadate on proteins in intact cells extracted post-heating in presence or absence of detergent. (XLSX 2334 kb)
Pathway analysis on proteins shifted by pervanadate treatment using Ingenuity Pathway Analysis software. (XLSX 29 kb)
Effect of a concentration range of oubain to the thermal stability proteins at 63 °C (XLSX 616 kb)
Effect of a concentration range of elacridar to the thermal stability proteins at 54 °C and 56 °C (XLSX 1382 kb)
About this article
Cite this article
Reinhard, F., Eberhard, D., Werner, T. et al. Thermal proteome profiling monitors ligand interactions with cellular membrane proteins. Nat Methods 12, 1129–1131 (2015). https://doi.org/10.1038/nmeth.3652
Analytical Chemistry (2021)
Cell surface thermal proteome profiling tracks perturbations and drug targets on the plasma membrane
Nature Methods (2021)
Journal of Proteome Research (2021)
Mutant thermal proteome profiling for characterization of missense protein variants and their associated phenotypes within the proteome
Journal of Biological Chemistry (2020)
Homogeneous label-free protein binding assay using small-molecule-labeled DNA nanomachine with DNAzyme-Based chemiluminescence detection