Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Jan 8;10(1):66.
doi: 10.1038/s41467-018-07923-2.

The Transcription Factor STAT5 Catalyzes Mannich Ligation Reactions Yielding Inhibitors of Leukemic Cell Proliferation

Affiliations
Free PMC article

The Transcription Factor STAT5 Catalyzes Mannich Ligation Reactions Yielding Inhibitors of Leukemic Cell Proliferation

Ee Lin Wong et al. Nat Commun. .
Free PMC article

Abstract

Protein-templated fragment ligations have been established as a powerful method for the assembly and detection of optimized protein ligands. Initially developed for reversible ligations, the method has been expanded to irreversible reactions enabling the formation of super-additive fragment combinations. Here, protein-induced Mannich ligations are discovered as a biocatalytic reaction furnishing inhibitors of the transcription factor STAT5. STAT5 protein catalyzes multicomponent reactions of a phosphate mimetic, formaldehyde, and 1H-tetrazoles yielding protein ligands with greatly increased binding affinity and ligand efficiency. Reactions are induced under physiological conditions selectively by native STAT5 but not by other proteins. Formation of ligation products and (auto-)inhibition of the reaction are quantified and the mechanism is investigated. Inhibitors assembled by STAT5 block specifically the phosphorylation of this protein in a cellular model of acute myeloid leukemia (AML), DNA-binding of STAT5 dimers, expression of downstream targets of the transcription factor, and the proliferation of cancer cells in mice.

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Discovery of phosphate-mimetic fragment 3. a Fluorescently labeled phosphotyrosine peptide 1 was used in an FP assay for the screening of a fragment library furnishing 4-amino-furazan-3-carboxylic acid 3 as a phosphate-mimetic. Phosphotyrosine-mimetic fragment 4-formyl-phenyl phosphate 2 was employed to investigate fragment hits for second site binding. b–c Molecular docking results of fragments 2 and 3 into homology model of human STAT5b-SH2 domain, generated from the published structure of STAT5a (PDB accession codes, 1Y1U [http://dx.doi.org/10.2210/pdb1Y1U/pdb]). Hydrogen bonds with key residues in the hydrophilic binding pocket of the STAT5-SH2 domain were illustrated as red dashed lines
Fig. 2
Fig. 2
Expansion of fragment 3 through protein-induced reactions. a Amidation of 3 yielded compounds 4 and 5, which were inactive in the FP assay. b Mannich ligation was investigated as an alternative fragment expansion method to obtain the active compounds 6–19 containing a linker with reduced steric hindrance and better structural flexibility
Fig. 3
Fig. 3
Assembly of STAT5 inhibitor 10 through protein-induced Mannich ligations. a FA was tolerated at up to 250 µm in the FP assay of MBP-STAT5b-SH2 (n = 3). FA did not affect the STAT5b protein stability in the TSA at ≤ 250 µm (data shown are one representative of n = 3). c Protein-induced formation of 10 in the FP assay with increasing FA concentrations (n = 3). d Protein-induced formation of 10 from fragments 3 and 1H-tetrazole 25 with increasing FA concentrations in the TSA (ΔTm = 7 °C) (data shown are one representative of n = 3). e Formation of 10 detected in the HPLC-QTOF-MS. 1: No formation of 10 from FA, fragments 3 and 25 (all 250 µm) at pH 7.4 in MOPS buffer without protein. 2: Protein-induced formation of compound 10 with protein MBP-STAT5b-SH2 (250 nm) at pH 7.4. 3–6: Inhibition of protein-induced formation of 10 by peptide 1 (3,4) or inhibitor 16 (5,6). 7,8: No formation of compound 10 in the presence of Maltose Binding Protein (1 µm) or GST-STAT3 (250 nm). 9–11: Compound 10 was not formed in the presence of phosphatases, SHP2 (250 nm) and PTP1B (250 nm) nor in the presence of mutant STAT5b-N642A at pH 7.4 in MOPS buffer. 12: Formation of compound 10 at pH 5.0 without protein. Data shown are one representative example of n = 3. f 3D-binding model of the STAT5b:10 complex. Hydrogen bonds are illustrated as red dashed lines. g, Mechanism of the protein-induced formation of 10 from 3 with FA and 25 (R = H). (i) Binding of 3 via Arg618, Ser622, and Asn642, activation of FA with Asn642. (ii) Activation of the forminium cation of 3, coordination of the incoming tetrazolium anion of 25 by Asn642 leads to formation of Mannich ligation product 10 (iii). Error bars denote mean ± S.D
Fig. 4
Fig. 4
Inhibitor 10 blocks STAT5 dimerization, DNA-binding, and is active in CETSA and photocrosslinking assays in cellular lysates. a 10 stabilizes MBP-STAT5b-SH2 shifting the melting temperature (ΔTm) by 9 °C in the TSA (data shown are one representative example of n = 3). b 10 and 16 inhibit formation of trimeric (STAT5b)2-DNA complexes in an ELISA (n = 3). c (i) 10 inhibits binding of STAT5 dimers, isolated from nuclear extracts of BaF3/FLT3-ITD cells, to its target DNA in the electro mobility shift assay (EMSA) and (ii) shows selectivity for disrupting STAT5a,b-DNA complexes in the TransAM® STAT family ELISA (n = 3). d–e 10 binds to STAT5a and STAT5b in complex cellular lysates as demonstrated by cellular TSA (CETSA), resulting in shifted melting curves at 50 µm compared with vehicle (DMSO). Relative STAT5a and STAT5b band intensities were plotted against corresponding incubation temperatures and fitted to Boltzmann sigmoidal curve. The blue triangle represents DMSO and red circle represents compound 10 (n = 3). f Binding of 10 to STAT5 in BaF3/FLT3-ITD cell lysate was determined by photo-crosslinking (n = 3). (i) The dual-labeled (carboxyfluorescein and biotin) peptide probe 27 binds to STAT5 with submicromolar affinity and photocrosslinked (ii) with target proteins by activating the 4-phosphoncarbonyl residue that acts as a photoactive phosphotyrosine-mimetic. (iii) Cross-linked proteins can be isolated by biotin pull-down using Neutravidine beads. (iv) Displacement of 27 (100 µm) by compound 10 (50 µm) in BaF3/FLT3-ITD cell lysates (1 mg ml−1) resulted in significantly reduced photo-crosslinking of 27 and STAT5 as demonstrated in the western blotting using STAT5 antibodies (right lane), whereas other biotinylated proteins were not reduced (middle lane). For an uncropped image of c, d, e, f see Supplementary Figure 10. Error bars denote mean ± S.D. and p values are considered as follows: *p value < 0.05; **p value < 0.01; and ***p value < 0.001. Statistical analyses were performed using one-way ANOVA or the two-tailed Student’s t tests where appropriate
Fig. 5
Fig. 5
Activity, target occupancy, and functional effects of STAT5 inhibitor 16 tested in STAT5-dependent cells. a 16 blocks tyrosine phosphorylation of STAT5 in a dose-dependent manner as shown by western blot analysis in BaF3/FLT3-ITD cells after 6 h treatment. Relative STAT5 phosphorylation levels were plotted as after quantification using Image J software. Immunoblotting for beta-actin was used as a control for uniform protein loading (n = 3). b Compound 16 inhibits transcriptional activity of STAT5 in BaF3/FLT3-ITD cells as measured by normalized Fluc/Rluc ratio in dual luciferase reporter assay (n = 3). c Expression of downstream targets of STAT5 Pim1,BcL-xL and Cis was reduced after 18 h of treatment with compound 16. Gene expression was quantified by quantitative PCR (n = 3). d Compound 16 inhibits the proliferation of BaF3/FLT3-ITD cells after 48 h as determined by the Alamar Blue assay (n = 3). e, i BaF3/FLT3-ITD cells were treated with compound 16 after which annexin-V/propidium iodide staining and flow cytometry were performed (gating strategy as in Supplementary Figure 11a, data shown are one representative of n = 3); (ii) Intracellular levels of phosphorylated STAT5 were evaluated by flow cytometry after 6 h exposure of cells to compound 16 for 50 µm (gating strategy as in Supplementary Figure 11b, data shown are one representative of n = 3). f–g, In-cell occupancy of STAT5a and STAT5b by compound 16 in BaF3/FLT3-ITD, determined using ITDRF. ITDRF of compound 16 on STAT5a and STAT5b denaturization at 60 °C for 3 min based on raw data from western blotting chemiluminescence readings (n = 3). CETSA, cellular thermal shift assay; ITDRF, isothermal dose–response fingerprint; OC50, the concentration at which 50% of the STAT5 in the cell was occupied by inhibitor. For an uncropped image of panels f and g see Supplementary Figure 10. Error bars denote mean ± S.D. and p values are considered as follows: **p value < 0.01; and ***p value < 0.001. Statistical analyses were performed using one-way ANOVA or the two-tailed Student’s t tests where appropriate
Fig. 6
Fig. 6
Synergy of inhibitor 16 and kinase inhibitor midostaurin (PKC412). Activity tested in a murine cancer model. a MV-411 cells were treated with PKC412 (10 nm) or compound 16 (10 µm) alone or in combination and incubated for 24 h followed by annexin-V/ propidium iodide staining and flow cytometry. Apoptosis was quantitated for three independent experiments (n = 3) and error bars denote mean ± S.D. b Cell viability assays were carried out by treating MV-411 cells with compound 16 (10 µm) and PKC412 (10 nm) alone or in combination. The number of viable cells was distinguished using an ATP-dependent bioluminescence assay (CellTiter-Glo, Promega) (n = 3, error bars denote mean ± S.D.). c Combination index (CI) plot showing the synergistic effect of compound 16 and PKC412 in MV-411 cells. CI values were generated using CalcuSyn software (Conservion, Ferguson, MO) and plotted as a function of fractional growth inhibition (n = 3) (Fa) where Fa = (A570 control−A570 treated)/A570control. CI values of < 1, = 1, and > 1 indicate synergism, additivity, and antagonism, respectively. Error bars denote mean ± S.D. d Relative STAT5 phosphorylation levels were plotted based on the raw data from western blotting chemiluminescence readings to study the synergistic effect of both compounds on STAT5 phosphorylation reduction. MV-411 cells were treated with compound 16 or PKC412 alone or in combination and incubated for 6 h (n = 3, error bars denote mean ± S.D.). e The corresponding body weight changes in non-xenografted mice during compound 16 treatments (n = 6, error bars denote mean ± S.D.). f Compound 16 significantly inhibits tumor growth in BaF3/FLT3-ITD xenograft tumor model. Time course of tumor growth suppressed by compound 16 (200 mg kg−1) in mice bearing BaF3/FLT3-ITD tumor (n = 6, error bars denote mean ± S.D.)

Similar articles

See all similar articles

Cited by 2 articles

References

    1. Jaegle M, et al. Protein-templated fragment ligations - from molecular recognition to drug discovery. Angew. Chem. Int. Ed. 2017;56:7358–7378. doi: 10.1002/anie.201610372. - DOI - PMC - PubMed
    1. Herrmann A. Dynamic combinatorial/covalent chemistry: a tool to read, generate and modulate the bioactivity of compounds and compound mixtures. Chem. Soc. Rev. 2014;43:1899–1933. doi: 10.1039/C3CS60336A. - DOI - PubMed
    1. Mondal M, Hirsch AKH. Dynamic combinatorial chemistry: a tool to facilitate the identification of inhibitors for protein targets. Chem. Soc. Rev. 2015;44:2455–2488. doi: 10.1039/C4CS00493K. - DOI - PubMed
    1. Ramström O, Lehn JM. Drug discovery by dynamic combinatorial libraries. Nat. Rev. Drug. Discov. 2002;1:26–36. doi: 10.1038/nrd704. - DOI - PubMed
    1. Schmidt MF, Groves MR, Rademann J. Dynamic substrate enhancement for the identification of specific, second-site-binding fragments targeting a set of protein tyrosine phosphatases. Chembiochem. 2011;12:2640–2646. doi: 10.1002/cbic.201100414. - DOI - PubMed

Publication types

MeSH terms

Feedback