A Critical Function for the Transcription Factors GLI1 and GLI2 in the Proliferation and Survival of Human Mast Cells

Abstract

Mast cell hyperactivity and accumulation in tissues are associated with allergy and other mast cell-related disorders. However, the molecular pathways regulating mast cell survival in homeostasis and disease are not completely understood. As glioma-associated oncogene (GLI) proteins are involved in both tissue homeostasis and in the hematopoietic system by regulating cell fate decisions, we sought to investigate the role for GLI proteins in the control of proliferation and survival of human mast cells. GLI1 transcripts were present in primary human mast cells and mast cell lines harboring or not activating mutations in the tyrosine kinase receptor KIT (HMC-1.1 and HMC-1.2, and LAD2 cells, respectively), while GLI2 transcripts were only present in HMC-1.1 and HMC-1.2 cells, suggesting a role for oncogenic KIT signaling in the regulation of GLI2. Reduction in GLI activity by small molecule inhibitors, or by shRNA-mediated knockdown of GLI1 or GLI2, led to increases in apoptotic cell death in both cultured human and murine mast cells, and reduced the number of peritoneal mast cells in mice. Although GLI proteins are typically activated via the hedgehog pathway, steady-state activation of GLI in mast cells occurred primarily via non-canonical pathways. Apoptosis induced by GLI silencing was associated with a downregulation in the expression of KIT and of genes that influence p53 stability and function including USP48, which promotes p53 degradation; and iASPP, which inhibits p53-induced transcription, thus leading to the induction of p53-regulated apoptotic genes. Furthermore, we found that GLI silencing inhibited the proliferation of neoplastic mast cell lines, an effect that was more pronounced in rapidly growing cells. Our findings support the conclusion that GLI1/2 transcription factors are critical regulators of mast cell survival and that their inhibition leads to a significant reduction in the number of mast cells in vitro and in vivo, even in cells with constitutively active KIT variants. This knowledge can potentially be applicable to reducing mast cell burden in mast cell-related diseases.

Trial registration: ClinicalTrials.gov NCT00001756.

Keywords: GLI; KIT; apoptosis; hedgehog signaling pathway; mast cell; proliferation.

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2022 Falduto, Pfeiffer, Zhang, Yin, Metcalfe and Olivera.

Figures

Figure 1

Modulation of GLI1/2 expression in human mast cells. (A) Simplified illustration of the canonical hedgehog signaling pathway and some of the small molecules used for its activation or inhibition. (B) Relative GLI1 mRNA levels (2-ΔΔCq) in human mast cell lines (HMC-1.1, HMC-1.2, LAD2) and primary human mast cells (hMC) after 20 h incubation with the SMO agonist SAG (200 nM) or the PTCH1 ligand SHH (500 ng/mL). (C) Relative GLI1 mRNA levels (2-ΔΔCq) in mast cells transduced with lentiviral particles containing two separate shRNA constructs to knockdown SUFU (see Supplementary Figure 2). (D) Relative GLI1 and GLI2 mRNA levels (2-ΔΔCq) after 5 h culture with the GLI1/2 inhibitor HPI-1 (20 or 40 µM) or with the SMO inhibitor vismodegib (40 µM). Results are expressed as Mean ± SD of three independent experiments. GAPDH and ACTB were used for normalization. Each individual experiment was done in triplicate. Two-way ANOVA followed by Dunnet multiple comparisons test was used for statistical analysis. **p < 0.01; ****p < 0.0001.

Figure 2

Ptch1 haploinsufficiency provides a proliferative advantage in mouse BMMC. (A) Relative mRNA levels (2-ΔCq) of Ptch1, Smo, and Gli1 in fresh bone marrow cells and after 7 weeks in culture, when >98% of cells are mast cells. Gapdh was used for normalization. (B, C) Percentage of mast cells (Kit+ FcεRI+ cells) (B) and total mast cell numbers (C) at the indicated times after initiation of the bone marrow culture (n=3 mice/group). Similar results were obtained in additional sets of cultures. The bracket in C indicates statistical significance between the two curves using 2-way ANOVA. The SD bars are not visible in some of the points due to the low variability between those cultures. (D) Proliferation assay using Cell Trace Violet (CTV) by FACS. BMMC were stained with the dye, at week 2 or 6 of culture. After washing, they were placed in culture media, and 6 days later, analyzed for CTV intensity and cell death staining. Representative dot-plots in FcϵRI+ Kit+ gated cells are shown on the left. On the right, the average of CTV median intensity in cultures is shown (n=3 mice/group). (E, F) CTV staining (E) and dead cell staining (F) of 6-week-old BMMC in the presence of the Smo agonist SAG (200 nM) or the Gli1/2 antagonist GANT61 (20 µM) for 6 days (n=3 mice/group). Results are shown as Mean ± SD. Unpaired t test or 2-way ANOVA followed by Dunnet multiple comparisons test were used for statistical analysis. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Figure 3

GLI1/2 inhibitors reduce human mast cell growth by decreasing viability and proliferation. (A) HMC-1.1 (left panel), HMC-1.2 (middle panel), and LAD2 (right panel) cells were cultured for 72 h (HMC-1.1 and HMC-1.2 cells) or 7 days (LAD2 cells) with increasing concentrations of the indicated inhibitors of GLI1/2 (HPI-1, GANT61 and ATO) or SMO (vismodegib). Cell growth was assessed by measuring relative units of fluorescence (RFU) that represent viable cells using a Cyquant assay; or counting viable cells with a cell counter (LAD2 cells). Results were normalized to vehicle control (represented as concentration 0 µM) and are expressed as Mean ± SD of three independent experiments. Two-way ANOVA followed by Dunnet multiple comparisons test (compared to vehicle) were used for statistical analysis. Approximated IC50 values, estimated using a non-linear fit between the inhibitor concentration and the normalized response (variable slope) with GraphPad Prism 9, were for HPI-1: 25 µM in HMC-1.1 and 16 µM in HMC1.2; and for GANT61: 15 µM in HMC-1.1 and 14 µM in HMC1.2. (B) Representative dot-plots of three independent experiments in which HMC-1.2 and LAD2 were cultured for 72 h or 7 days, respectively, in the presence of the indicated inhibitors (20 µM). Cells were stained with Cell Trace Violet (CTV) before the incubation and at the end of the experiment stained with green dead cell stain. The percentage of dead cells and median fluorescence intensity (MFI) for CTV is shown in each dot-plot. (C) Representative histograms of primary human mast cells (hMC; 1 healthy donor out of 2) cultured for 5 days with the indicated inhibitors (20 µM) and stained with green dead cell stain. Averages of CTV MFI and percentages of dead cells in three independent experiments are shown in Supplementary Figure 4. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Figure 4

ShRNA-mediated silencing of GLI1/2 reduces human mast cell growth by decreasing viability and proliferation. (A) HMC-1.2 and LAD2 cells were transduced with lentiviral particles containing GLI1 or GLI2 shRNA constructs (two independent constructs for each). After proper selection, relative GLI1 and GLI2 mRNA levels (2-ΔΔCq) were quantified by qPCR. GAPDH and ACTB were used for normalization. (B) After proper selection numbers of viable HMC-1.2 (left) and LAD2 (right) cells were assessed using the Cyquant assay (relative fluorescence units, RFU) or a cell counter, respectively. Cell numbers were assessed after 72 h in HMC-1.2 cells and after 7 days in LAD2 cells. Results are expressed as Mean ± SD of three independent experiments. One-way or 2-way ANOVA followed by Dunnet multiple comparisons test were used for statistical analysis. (C) After proper selection, cell proliferation and cell death were assessed as described in Figure 3. Representative dot-plots of two independent experiments are shown. The percentage of dead cells and median fluorescence intensity (MFI) of Cell Trace Violet (CTV) fluorescence within the live cells gate are shown in each dot-plot. Averages of CTV MFI and percentages of dead cells in two independent experiments are shown in Supplementary Figure 5. *p < 0.05; **p < 0.01; ****p < 0.0001.

Figure 5

GLI1/2 inhibition causes apoptosis in human mast cells. (A) HMC-1.1, HMC-1.2, LAD2, and primary human mast cells (hMC) were cultured in the presence of HPI-1, GANT61, or vismodegib (20 µM) for 48 h (HMC-1.1 and HMC-1.2 cells), 6 days (LAD2 cells) and 4 days (hMC). The percentages of annexin V positive and dead cells were quantified by flow cytometry. Representative dot-plots from one of three independent experiments are shown. The bar graph represents the average percentage of apoptotic cells (annexinV+/dead cell stain-) in three independent experiments using the indicated cells and treatments. (B) Caspase 3/7 activity was quantified by measuring relative units of fluorescence (RFU) of a caspase 3/7 substrate in HMC-1.1 and HMC 1.2 after 48h incubation with increasing concentrations of inhibitors, vehicle control is represented as concentration 0 µM. Results expressed as Mean ± SD of three independent experiments. Two-way ANOVA followed by Dunnet multiple comparisons test (compared to vehicle) were used for statistical analysis. *p < 0.05; ***p < 0.001; ****p < 0.0001.

Figure 6

GLI1 inhibition stabilizes p53 and alters the gene expression of pro-apoptotic proteins. (A) Gene expression profiling of differentially expressed genes (2-ΔΔCq) within the mitochondrial apoptotic pathway (p53-regulated genes are marked by arrows and in bold). RNA from HMC-1.2 cells was extracted after an overnight incubation with HPI-1 or GANT61 (40 µM) or after proper selection of GLI1 shRNA (construct 1)-transduced cells. Vehicle and non-target mRNA levels are represented by dotted lines. Z-scores heatmap of the top seven canonical pathways modulated in GLI1 ShRNA (construct 1)-transduced cells, determined by using an IPA analysis. (B) HMC-1.2 cells were incubated overnight in the presence of HPI-1 (40 µM), cell lysates were obtained and p53 was quantified by western blot. β-actin was used as a loading control. The western blot shows a representative experiment and the average quantification of p53 protein expression, normalized to β-actin, in 3 independent experiments, are shown in the bar graph (Mean ± SD). Full-length blots are shown in Supplementary Figure 6A. (C) RNA from human mast cell lines was obtained after incubation for 5 h with HPI-1 (20 or 40 µM) and relative mRNA levels (2-ΔΔCq) of iASSP and USP48 was determined by qPCR. GAPDH and ACTB were used for normalization. Results are expressed as Mean ± SD of three independent experiments. Illustration represents the regulation of p53 levels and function by iASSP and USP48. Unpaired t test or 2-way ANOVA followed by Dunnet multiple comparisons test were used for statistical analysis. **p < 0.01; ***p < 0.001; ****p < 0.0001.

Figure 7

GLI1 inhibition downregulates KIT expression and phosphorylation in human mast cells. (A) Relative KIT mRNA levels (2-ΔΔCq) in mast cell lines or primary human mast cells (hMC) after 5h incubation with HPI-1 (20 or 40 µM) or 48h after transduction with lentiviral particles containing GLI1 shRNA (construct 1), no puromycin selection was conducted. Results are expressed as Mean ± SD of three independent experiments. GAPDH and ACTB were used for normalization. (B) HMC-1.1, HMC-1.2 and LAD2 cells were cultured overnight with HPI-1 (20 or 40µM) and lysed. Proteins were resolved in SDS-PAGE. KIT and phosphorylated-KIT (pKIT) were quantified by western-blot. β-actin was used as a loading control. The blots show representative images, and the bar graphs are the average relative band intensities of KIT and phospho-KIT (normalized by β-actin) in 2 independent experiments (Mean ± SD). Unpaired t test or 2-way ANOVA followed by Dunnet multiple comparisons test were used for statistical analysis. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Full-length blots are shown in Supplementary Figure 6B.

Figure 8

Gli1/2 inhibitors reduce peritoneal mast cell numbers in vivo. (A) Scheme illustrating the protocol for treatment of mice with the indicated Gli1/2 inhibitors. (B) Peritoneal lavage was obtained 2 days after the last injection with vehicle control (PBS : EtOH; 95:5%), HPI-1 (2 mg/Kg), or GANT61 (2 mg/Kg). Cells were stained with anti-FcϵRI and anti-Kit antibodies and separated by FACS. Percentages of mast cell (FcϵRI+ Kit+ cells) among total live peritoneal cells for each treatment are shown (left). The number of viable cells in the peritoneal lavage after these treatments was measured using a cell counter (right panel). Results are expressed as Mean ± SD (n=4 mice/group). One-way ANOVA followed by Dunnet multiple comparisons test were used for statistical analysis. (C) Representative dot-plots from one mouse per group are shown. *p < 0.05; **p < 0.01.