Fig 1.
Scheme of the primary and the secondary screening procedures.
In the primary screen (upper panel), construct CRF1R.GFP was used, a GFP-tagged GPCR possessing a cleavable signal peptide which uses the SRP-Sec61 targeting/translocation pathway. The secondary screen (lower panel) was performed with unfused, soluble GFP which does not use the SRP-Sec61 targeting/translocation pathway. See the text for details.
Fig 2.
Schematic depiction of the results of the 1052 hit compounds under selection and deselection conditions.
For each compound (dots), the relative GFP fluorescence of the screening target in the primary screen (CRF1R.GFP) was plotted against the relative GFP fluorescence of the target in the secondary screen (soluble, unfused GFP). Hit compounds are indicated in green, the 5 compounds used for further analysis are indicated in red (see below). Data represent mean values of three technical replicates (see S1 Fig for the replicates). 100% fluorescence = mean values of 16 induced plate control samples; 0% fluorescence = mean values of 16 non-induced plate control samples (plate control samples were without compound).
Fig 3.
Chemical structures and properties of the compounds FMP-503533, FMP-214534, FMP-401319, FMP-208236 and FMP-214219.
Left panels. Chemical structures of the compounds. Central panels. Rating of the compounds (red line) for their fulfilment of “Lipinski’s rule of five” for druggable compounds: H donors, H acceptors, hydrophobicity (ALogP), molecular mass (MW), and polar surface area (PSA). Calculation of the physicochemical properties was made using the SwissADME server (www.swissadme.ch). Right panels. Concentration response curves for the biosynthesis inhibition of the screening target CRF1R.GFP. The assay was carried out as in the primary screen using 384-well plates and different concentrations of the compounds. Expression of CRF1R.GFP was quantified fluorometrically via its GFP signals. The IC50 values are indicated. Data points are mean values of triplicates (±SD) and represent % of the maximal activity.
Fig 4.
Whole cell assay: biosynthesis of various target proteins in HEK 293 cells treated with compounds FMP-503533, FMP-214534, FMP-401319, FMP-208236 and FMP-214219.
To analyze activity and selectivity of the compounds, HEK 293 cells were transiently transfected with the original screening target CRF1R.GFP (red column) and various GFP-tagged integral membrane proteins. Cells were treated with the compounds (FMP-503533, 20 μM; FMP-214534, 25 μM; FMP-401319, 25 μM; FMP-208236, 20 μM; FMP-214219, 10 μM) or DMSO for 19 h and the total GFP fluorescence of the cells was analyzed using flow cytometry as a measure of biosynthesis. Note that compounds FMP-214534, FMP-401319, FMP-208236 and FMP-214219 affected the biosynthesis of all target proteins whereas compound FMP-503533 impaired only biosynthesis of the original screening target CRF1R.GFP. Columns represent mean values of biosynthesis (% of matched DMSO control, red line) calculated out of 3–10 independent experiments indicated in the individual panels ±SD. P values: p ≤ 0.001 (***), p ≤ 0.01 (**), p ≤ 0.05 (*), p ≥ 0.1 (not significant, ns).
Fig 5.
Cell free assay: activities of the compounds against the SRP-Sec61 targeting/translocation pathway.
(A) Representative digital autoradiograms of the in vitro transcribed/translated and translocated CRF1R-pPL chimaera in the absence or presence of the indicated compounds (300 μM). NCs of 78 residues without a stop codon were transcribed/translated in vitro in the cell free rabbit reticulocyte lysate system. Because of an intact peptidyl-tRNA bond, NCs stay attached to the ribosome and appear as NC-tRNA complexes (lane 1, NC-tRNA). Due to spontaneous hydrolysis of the peptidyl-tRNA bond, a fraction of the [35S]-labeled peptides is visible as prematurely released NCs (lane 1, NC). Loaded ribosomes are subsequently mixed with RM (lanes 2–5, RM) to allow for ribosome docking onto the Sec61α protein-conducting channels. Note that the presence of RM also releases extra NCs from tRNA in an unproductive way. Finally, NCs are released from the ribosomes by addition of PU (lanes 4–5, PU), resulting in signal peptide-cleaved NCs that have been translocated into the ER lumen (SPcNC, 55 residues, red arrow). In the diagram on the lower right panel, each black bar represents the relative amount of translocated protein versus total protein (translocated and precursor) after compound treatment (lane 5), in comparison to the corresponding DMSO control (= 100% translocation; lane 4). Bars are mean ± SD.; n ≥ 2. See also S1 Table for the data. P values: p < 0.0001 (***), p ≤ 0.05 (*). (B) Schematic depiction of the cell free assay following in vitro transcription/translation of construct CRF1R-pPL and addition of RM. The effects of PU treatment (right panel) or leaving the sample untreated (left panel) on the NC-tRNAs are shown. See also (A) above for details. SP = signal peptide; SPase = signal peptidase.
Fig 6.
Substructure-based back screening of compound FMP-401319.
Left panel. Structure of the compounds. Right panel. Commercially available derivatives of FMP-401319 (25 μM each) were tested for their inhibitory action in the whole cell biosynthesis assay as described in the legend of Fig 3. While some of the compounds showed a decreased activity (exemplary results: compounds FMP-401319-6, FMP-401319-17 and FMP-401319-43), one compound (FMP-401319-3) increased biosynthesis inhibition of some target proteins in comparison to FMP-401319. Columns represent mean values of biosynthesis (% of matched DMSO control, red line) calculated out of 5 independent experiments ±SD. P values: p ≤ 0.001 (***), p ≤ 0.01 (**), p ≤ 0.05 (*), p ≥ 0.1 (not significant, ns).
Fig 7.
Compound FMP-401319-3 inhibits cotranslational translocation of CRF1R-pPL concentration-dependently at the level of the Sec61 complex.
(A) Schematic depiction of the effects of PU, PK and a PU/PK combination on targeted and non-targeted ribosomes containing CRF1R-pPL NC-tRNAs in the cell free transcripition/translation/translocation assay. See text for details. The individual panels of Fig 7A refer to the autoradiogram shown in 7B. (B) Representative digital autoradiogram of the translated and translocated CRF1R-pPL chimaera in the absence or presence of different concentrations of compound FMP-401319-3 (3 μM, 30 μM, 300 μM), as described in the legend to Fig 5. All samples are with RM. Equal aliquots of the translated material were treated with PK (lanes 9–16, PK) or left untreated (lanes 1–8). SPcNCs that have been translocated into the ER lumen (red arrow) are protected inside the microsomal vesicles and thus PK-resistant. Note that at 300 μM of FMP-401319-3, most peptide chains could be rescued in the intact NC form after PK treatment (lane 14). (C) Interpretation of the results following FMP-401319-3 treatment. Without compound treatment (left panel), the NC is fully translocated, the signal peptide is cleaved-off and the SPcNC fragment appears. Treatment with FMP-401319-3 (right panel), however, interferes with a step before the growing peptide chain has been translocated and reached the luminal side of the ER. Some NC may be positioned between ribosome exit tunnel and the cytosolic side of the Sec61 complex in such a way that some parts are exposed to the cytosol and accessible in the experiment to PK. (D) Quantification of the data shown in Fig 7B and from 2 additional, independent experiments. Graph in upper panel shows the percentage of translocation for different concentrations of compound FMP-401319-3, similar as explained in the legend to Fig 5A. Bars are mean of three digital autoradiograms ± SD; n = 3. See also S1 Table for the data. Lower panel displays the results from equivalent experiments with the archetypical mixed type A/type B inhibitor cotransin (a representative digital autoradiogram is given in S3 Fig). It was previously shown that the signal peptide of the CRF1R is cotransin-sensitive [7]. Bars are mean of three digital autoradiograms ± SD; n = 3. See also S1 Table for the data.
Fig 8.
Compound FMP-401319-3 inhibits cotranslational translocation of CRF1R-pPL in a post targeting step.
(A) Schematic depiction of the experiment. CRF1R-pPL NCs of 78 residues were translated in the absence of microsomes before administration to the RM for targeting. NCs were left untreated or were treated with FMP-401319-3, either applied to RM for pretreatment (pre), or applied to the RNC/RM mixture 15 minutes after initiation of targeting but before PU release (post). (B) Representative autoradiogram of the translated and translocated CRF1R-pPL chimaera in the absence (-) or presence (+) of compound FMP-401319-3 (300 μM). Pre = drug treatment before targeting. Post = drug treatment initiated 15 minutes (i.e., 10 + 5) after targeting. The NC and the SPcNC fragments are indicated.