Hirsutinolide Series Inhibit Stat3 Activity, Alter GCN1, MAP1B, Hsp105, G6PD, Vimentin, TrxR1, and Importin α-2 Expression, and Induce Antitumor Effects against Human Glioma

Abstract

We report that hirsutinolide series, 6, 7, 10, 11, 20, and 22, and the semisynthetic analogues, 30, 31, 33, and 36, inhibit constitutively active signal transducer and activator of transcription (Stat)3 and malignant glioma phenotype. A position 13 lipophilic ester group is required for activity. Molecular modeling and nuclear magnetic resonance structural analyses reveal direct hirsutinolide:Stat3 binding. One-hour treatment of cells with 6 and 22 also upregulated importin subunit α-2 levels and repressed translational activator GCN1, microtubule-associated protein (MAP)1B, thioredoxin reductase (TrxR)1 cytoplasmic isoform 3, glucose-6-phosphate 1-dehydrogenase isoform a, Hsp105, vimentin, and tumor necrosis factor α-induced protein (TNAP)2 expression. Active hirsutinolides inhibited anchorage-dependent and three-dimensional spheroid growth, survival, and migration of human glioma lines and glioma patients' tumor-derived xenograft cells harboring constitutively active Stat3. Oral gavage delivery of 6 or 22 inhibited human glioma tumor growth in subcutaneous mouse xenografts. The inhibition of Stat3 signaling represents part of the hirsutinolide-mediated mechanisms to induce antitumor effects.

Figure 1. Chemical structures of hirsutinolide natural products and semi-synthetic derivatives

(A and B) Hirsutinolide natural products; and (C) newly semi-synthesized analogs of 22.

Figure 2. Hirsutinolide natural products inhibit Stat3 activation

(A and B) Nuclear extracts containing activated (A) Stat3 from NIH3T3/v-Src fibroblasts or (B) Stat1, Stat3 and Stat5 from EGF-stimulated NIH3T3/hEGFR were pre-incubated with the designated hirsutinolides for 30 min at room temperature prior to incubating with the radiolabeled hSIE probe that binds Stat1 and Stat3 or MGFe probe that binds Stat1 and Stat5 and performing EMSA analysis; bands corresponding to STAT:DNA complexes in gel were quantified using ImageQuant and represented as percent of control (B (i), lower panel); (C and D) Stat3 DNA-binding activity with EMSA analysis using the hSIE probe that binds Stat3 of nuclear extracts prepared from U251MG or U373MG cells treated with the designated hirsutinolides at (C) 5 μM for 0–24 h or (D) 0–5 μM for 1.5 h; (E–G) Immunoblotting analysis of whole-cell lysates prepared from (E and F) U251MG cells treated with the designated hirsutinolides at (E) 5 μM for 0–24 h or (F) 0–5 μM for 2 h, or (G) human glioma patients-derived xenograft cells (i) G6-G102, untreated, or (ii) G22, treated with 5 μM for 0–24 h 6, 10, or 22 and probing for pStat3, Stat3 or GAPDH. Positions of proteins or DNA-bound STATs in gel are labeled; control lane (c, 0) represents whole-cell lysates or nuclear extracts prepared from 0.025% DMSO-treated cells or nuclear extracts pre-treated with 0.025% DMSO. Bands corresponding to Stat3:DNA complexes were scanned and quantified using ImageJ, plotted against concentration of agent from which IC₅₀ values were derived. Data are representative of 2–3 independent determinations. *Position of supershifted complex.

Figure 3. Hirsutinolides do not modulate Stat1, Stat5, JAK2, Akt, Erk1/2, SOCS3, or PTP induction

(A–C) Immunoblotting analysis of whole-cell lysates of equal total protein prepared from (A) NIH3T3/hEGFR fibroblasts pre-treated with or without 5 μM of 6, 10, 20 or 22 for 3 h prior to stimulation with 1 μg/ml EGF for 12 min, (B) U251MG cells treated with 5 μM 6, 10 or 22 for 0–24 h, or (C) U251MG cells treated with 5 μM 6, 10 or 22 for 3 h in the absence or presence of 10–300 μM Na₃VO₄ and probing for pStat3, Stat3, pStat1, Stat1, pStat5, Stat5, pErk1/2, Erk1/2, pS-Akt, Akt, pJAK2, JAK2, SOCS3, PTP1B, β-actin or GAPDH. Positions of proteins in gel are labeled; control lanes (0, −) represent whole-cell lysates prepared from 0.025% DMSO-treated cells. Data are representative of 2–4 independent determinations.

Figure 4. NMR analysis and molecular modeling of Stat3:compound interactions and the effect of over-expression of Stat3 DNA-binding domain on hirsutinolide activity

(A) Molecular modeling and docking of 33, 31, 29, 22, 6 and 7 into the Stat3 DNA-binding domain pocket showing the interactions with key amino acid residues; (B and C) Overlay of the ¹H-¹³C HMQC spectra of (B) wild-type Stat3, free (black) or bound to 100 μM (red) or 200 μM (green) 6 and 10, and (C) wild-type Stat3, free (red) or bound to 100 or 200 μM (blue) 7, 9, 11 or 22 and showing residues with significant changes in either resonance line-widths or NMR chemical shifts that are indicated by arrowheads; and (D) immunoblots of (i) FLAG or GAPDH, or (ii) pStat3 or Stat3 in whole-cell lysates from U251MG cells transiently-transfected with Stat3 FLAG-tagged N-terminal (NTD), Coiled coil (CCD), C-terminal (CTD) or DNA-binding (DBD) domain and treated with 5 μM 6 for 1.5 h. Positions of proteins in gel are labeled; control lane (−) represents whole-cell lysates prepared from cells treated with 0.025% DMSO. Data are representative of 2 independent determinations.

Figure 4. NMR analysis and molecular modeling of Stat3:compound interactions and the effect of over-expression of Stat3 DNA-binding domain on hirsutinolide activity

(A) Molecular modeling and docking of 33, 31, 29, 22, 6 and 7 into the Stat3 DNA-binding domain pocket showing the interactions with key amino acid residues; (B and C) Overlay of the ¹H-¹³C HMQC spectra of (B) wild-type Stat3, free (black) or bound to 100 μM (red) or 200 μM (green) 6 and 10, and (C) wild-type Stat3, free (red) or bound to 100 or 200 μM (blue) 7, 9, 11 or 22 and showing residues with significant changes in either resonance line-widths or NMR chemical shifts that are indicated by arrowheads; and (D) immunoblots of (i) FLAG or GAPDH, or (ii) pStat3 or Stat3 in whole-cell lysates from U251MG cells transiently-transfected with Stat3 FLAG-tagged N-terminal (NTD), Coiled coil (CCD), C-terminal (CTD) or DNA-binding (DBD) domain and treated with 5 μM 6 for 1.5 h. Positions of proteins in gel are labeled; control lane (−) represents whole-cell lysates prepared from cells treated with 0.025% DMSO. Data are representative of 2 independent determinations.

Figure 4. NMR analysis and molecular modeling of Stat3:compound interactions and the effect of over-expression of Stat3 DNA-binding domain on hirsutinolide activity

(A) Molecular modeling and docking of 33, 31, 29, 22, 6 and 7 into the Stat3 DNA-binding domain pocket showing the interactions with key amino acid residues; (B and C) Overlay of the ¹H-¹³C HMQC spectra of (B) wild-type Stat3, free (black) or bound to 100 μM (red) or 200 μM (green) 6 and 10, and (C) wild-type Stat3, free (red) or bound to 100 or 200 μM (blue) 7, 9, 11 or 22 and showing residues with significant changes in either resonance line-widths or NMR chemical shifts that are indicated by arrowheads; and (D) immunoblots of (i) FLAG or GAPDH, or (ii) pStat3 or Stat3 in whole-cell lysates from U251MG cells transiently-transfected with Stat3 FLAG-tagged N-terminal (NTD), Coiled coil (CCD), C-terminal (CTD) or DNA-binding (DBD) domain and treated with 5 μM 6 for 1.5 h. Positions of proteins in gel are labeled; control lane (−) represents whole-cell lysates prepared from cells treated with 0.025% DMSO. Data are representative of 2 independent determinations.

Figure 5. Hirsutinolides inhibit in vitro cell proliferation, colony survival, and migration, and the expression of Stat3-regulated genes

(A) Single-cell cultures of U251MG, SF295, MDA-MB-231, MCF7 and NIH3T3 cells treated once with 0–5 μM 6, 10 or 22 and allowed to grow until large colonies were visible, which were stained with crystal violet, counted and plotted. U251MG colony formation was completely inhibited by 1 μM 6 or 22; (B) 5-bromo-2′-deoxyuridine (BrdU) incorporation analysis for proliferating U251MG cells treated with 0 or 5 μM 6 or 22 for 3–24 h. Images were captured under a fluorescence microscope and analyzed with ImageJ software; (C) cultured U251MG cells were wounded, treated once with 0–5 μM 6, 10 or 22 and allowed to migrate to the denuded area over 19 h and imaged; (D) cell cycle distribution analysis of U251MG cells treated or untreated (DMSO) with 15 μM Cmpd1 for 24 or 72 h, processed by propidium iodide (PI) staining, and analyzed by flow cytometry for DNA content, which is plotted; and (D) immunoblots of pStat3, Stat3, c-Myc, Bcl-2, Bcl-xL, Mcl-1 or GAPDH from whole-cell lysates from U251MG cells treated with 5 μM 6 for 0–24 h. Positions of proteins in gel are shown; control lane (0) represents 0.025% DMSO-treated cells or whole-cell lysate preparation from 0.025% DMSO-treated cells. Values are the mean ± S.D., n=3–6. Data are representative of 2–3 independent determinations.

Figure 6. Antitumor effects against human glioma tumor xenografts in vivo

(A) Mice bearing U251MG subcutaneous tumor xenografts were administered 6 or 22 via oral gavage, 2 mg/kg or vehicle (1% DMSO) every other day for 33 days. Tumor sizes, measured every 3 days, were converted to tumor volumes and plotted against days of treatment; and (B) immunoblots of pStat3, Stat3 or GAPDH in tumor tissue lysates prepared from control and treated mice. Positions of proteins in gel are labeled; control lanes (c) represent tissue lysates prepared from mice treated with 1% DMSO. Values, mean ± S.D., n=6. * - <0.05.