Targeted dephosphorylation of SMAD3 as an approach to impede TGF-β signaling

Summary TGF-β (transforming growth factor-β) signaling is involved in a myriad of cellular processes and its dysregulation has been implicated in many human diseases, including fibrosis and cancer. TGF-β transcriptional responses are controlled by tail phosphorylation of transcription factors SMAD2 and SMAD3 (mothers against decapentaplegic homolog 2/3). Therefore, targeted dephosphorylation of phospho-SMAD3 could provide an innovative mechanism to block some TGF-β-induced transcriptional responses, such as the transcription of SERPINE-1, which encodes plasminogen activator inhibitor 1 (PAI-1). Here, by developing and employing a bifunctional molecule, BDPIC (bromoTAG-dTAG proximity-inducing chimera), we redirected multiple phosphatases, tagged with bromoTAG, to dephosphorylate phospho-SMAD3, tagged with dTAG. Using CRISPR-Cas9 technology, we generated homozygous double knock-in A549 bromoTAG/bromoTAGPPM1H/dTAG/dTAGSMAD3 cells, in which the BDPIC-induced proximity between bromoTAG-PPM1H and dTAG-SMAD3 led to a robust dephosphorylation of dTAG-SMAD3 and a significant decrease in SERPINE-1 transcription. Our work demonstrates targeted dephosphorylation of phospho-proteins as an exciting modality for rewiring cell signaling.

(C) Cytotoxicity of HDPIC was measured using CellTox Green Assay (Promega) by treating wild-type (WT) U2OS or A549 cells with HDPIC at the indicated concentrations for 24 h.DMSO was included as a negative control and MG132 (20 μM, 24 h) or the lysis buffer provided in the assay kit used as a positive control.Fluorescence was then measured using a PHERAstar HEK293 WT or HEK293 dTAG/dTAG PPP2CA cells were treated for 24 h with DMSO, dTAG-13 PROTAC (100 nM), HDPIC (1 µM) or a combination of dTAG-13 and HDPIC.Cells were lysed and processed as in (D).
(F) U2OS cells stably expressing Halo-SMAD3 alone or in combination with dTAG-PPM1H were serum-starved (16 h) before 1 h stimulation with TGFβ (5 µg/L).0 h samples were lysed at this point.For other samples, TGFβ stimulation was removed and cells were placed in fresh serum-free medium without TGFβ with DMSO, HDPIC (1 µM) or no treatment.Samples were then lysed and processed as in (D).(E) Preliminary dose response of BDPIC-mediated targeted dephosphorylation of dTAG-SMAD3 in A549 bromoTAG/bromoTAG PPM1H/ dTAG/dTAG SMAD3 cells.Following serum-starvation (16 h), cells were stimulated with TGFβ (5 µg/L) for 1 h.0 h time points were lysed at this moment.
For 2 h time points, TGFβ stimulation was removed by washout and fresh serum-free medium without TGFβ was added to cells, along with DMSO or the indicated concentrations of BDPIC for 2 h.SB-505124 was employed as a control.Cells were then lysed and processed as in

Synthesis of BDPIC (bromoTAG-dTAG proximity-inducing chimera)
To ensure access of the full methodology for BDPIC synthesis, we include the identical detailed methodology presented here also in the manuscript by Zhao et al that was jointly submitted for publication in iScience.

General comments
All chemicals were purchased from commercial vendors and used without further purification.
Ortho-AP1867 were prepared as described in 4 .
Flash column chromatography and Prep HPLC were performed by using Buchi PrepChrom C-700, prepacked Buchi Sepacore Flash Cartridges and HPLC C18 column Gemini NY(RP)C18 110, 21.2×150 mm, 10 µn particle size.Details about the conditions for preparative HPLC are provided in the experimental procedures.
NMR spectra were recorded on a Bruker Ascend 500 MHz.Chemical shifts are reported in parts per million referenced to residual solvent peaks (CDCl3= 7.26 ppm).Only the chemical shifts of the major rotamer are reported.The following abbreviations were used in reporting spectra, s (singlet), d (doublet), t (triplet), q (quartet), m (multiplet), dd (doublet of doublets), bs (broad signal).Low resolution mass spectra and analytical HPLC traces were recorded on an Agilent Technologies 1200 series HPLC connected to an Agilent Technologies 6130 quadrupole LC/MS, connected to an Agilent diode array detector.The column used was a Waters XBridge column (50 mm × 2.1 mm, 3.5 μm particle size), with a gradient from 5 % to 95% of acetonitrile in water (with 0.1 % of formic acid) over 3 or 7 minutes.The flow rate was 0.7 mL/min.To a solution of 5 (0.85 g, 1.34 mmol) in THF (5 ml) was added 1M solution of TBAF in THF (1.47 ml, 1.47 mmol) at 0 °C and the mixture was stirred at RT for overnight.Water (5 ml) was added and extracted with EtOAc twice, organic layer washed with brine and concentrated.
The mixture was stirred for 12 h at room temperature, quenched with saturated NaHCO3 (2 mL) and extracted with DCM (10 ml).The organic layer washed with brine and concentrated.
plate reader (ex: 480 nm em: 530 nm).Data represent n=3.Values are shown as a mean fluorescence reading normalized to DMSO controls ± SD. (D) Engagement of Halo-POI by HDPIC was shown by band-shift of small (~35 kDa) artificial FLAG-NLS(nuclear localization signal)-Halo-HiBiT protein upon HDPIC treatment.U2OS cells stably expressing FLAG-NLS-Halo-HiBiT were treated for 2 h with the indicated concentrations of HDPIC or an equivalent volume of DMSO.Cells were lysed before extracts were resolved by SDS-PAGE, transferred to nitrocellulose membrane and immunoblotted with the indicated antibodies.Three independent experiments are shown together.(E) Engagement of dTAG-POI by HDPIC was shown by means of a competition-style assay.

(
G) Western blotting screening of multiple A549 Halo SMAD3 knock-in clones, with the clone of interest, Clone 19, highlighted in red.Cells were lysed and processed as in (D).(H) Confirmation of multiple knock-in clones, with clone 19 highlighted in red, by amplification of the target SMAD3 genomic region through polymerase chain reaction.DNA from parental WT A549 cells was included as a negative control.(I) Confirmation of insertion of knock-in at desired genomic locus by sequencing of the target locus in A549 Halo/WT SMAD3 clone 19 cells.(J) To show effect of HDPIC on Halo-SMAD3 phosphorylation, A549 Halo/WT SMAD3 knock-in cells were serum-starved (16 h) prior to 1 h co-treatment with TGFβ (5 µg/L) and SB-505124 (1 µM) or DMSO.0 h time points were lysed at this moment.For 4 h time points, TGFβ stimulation was removed by washout and fresh serum-free medium without TGFβ was added to cells, along with DMSO, HDPIC (200 nM) or PhosTAC7 (200 nM) for 4 h.Samples were then lysed and processed as in (D).(K) Structure of PhosTAC7.Data are representative of n=3 independent experiments, except (G-I).

Figure S4 : 2 .
Figure S4: Quantification of relative Western blot signal intensities from biological repeats of Figure S3F.In relation to Fig. S3 and Fig. 2. Quantification of n=3 independent experiments of Fig. S3F, showing mean ± SD phospho-dTAG-SMAD3/total dTAG-SMAD3 levels, relative to the 0 h sample for each cell line.

Figure S5 :
Figure S5: Characterization of HDPIC-Neg and identification of retained interaction with HaloTag at micromolar concentrations.In relation to Fig. S3 and Fig. 2. (A) Structure of HDPIC-Neg.(B) Cytotoxicity of HDPIC-Neg was measured using CellTox Green Assay (Promega) as in (Fig. S3C), in WT U2OS and A549 cells.Data represent n=3.Values are shown as a mean fluorescence reading normalized to DMSO controls ± SD. (C) Engagement of Halo-POI by HDPIC-Neg was probed by a band-shift assay in U2OS FLAG-NLS-Halo-HiBiT cells following 2 h treatment with the indicated concentrations of HDPIC-Neg, HDPIC (active) or an equivalent volume of DMSO.Cells were lysed before extracts were resolved by SDS-PAGE, transferred to nitrocellulose membrane and immunoblotted with the indicated antibodies.Data are representative of n=3 independent experiments.

Figure S7 : 4 .
Figure S7: Quantification of relative Western blot signal intensities from biological repeats of Figure 4C.In relation to Fig. 4. Quantification of n=3 independent experiments of Fig. 4C, showing mean ± SD phospho-dTAG-SMAD3/total dTAG-SMAD3 levels, relative to the first lane of DMSO treatment for each cell line.

Figure S8 :
Figure S8: Generation and validation of A549 bromoTAG/bromoTAG PPM1H/ dTAG/dTAG SMAD3 knock-in cells.In relation to Fig. 5. (A) Western blot of A549 wild-type (WT) and A549 bromoTAG/bromoTAG PPM1H/ dTAG/dTAG SMAD3 clone 51 cells.For Western blotting, cells were lysed before extracts were resolved by SDS-PAGE, transferred to nitrocellulose membrane and subjected to immunoblotting with the indicated antibodies.(B) Confirmation of knock-in by amplification of the target SMAD3 genomic region through polymerase chain reaction, with clone 51 highlighted in red.DNA from parental A549 bromoTAG/bromoTAG PPM1H cells was included as a negative control while DNA from knock-in A549 dTAG/dTAG SMAD3 clone 5 cells was included as a positive control.Samples were all run on the same gel and non-relevant samples have been omitted from between samples of interest.(C) Confirmation of insertion of knock-in at desired genomic locus by sequencing of the target locus in A549 bromoTAG/bromoTAG PPM1H/ dTAG/dTAG SMAD3 clone 51 cells.(D) Validation that endogenous dTAG-SMAD3 in A549 bromoTAG/bromoTAG PPM1H/ dTAG/dTAG SMAD3 clone 51 cells retain endogenous function and is responsive to TGFβ stimulation and capable of subsequently increasing PAI-1 protein levels.Here, cells were stimulated with TGFβ (5 µg/L) or control and treated with SB-505124 or DMSO for 6 h prior to being processed as in (A).