SD‐36 promotes growth inhibition and induces apoptosis via suppression of Mcl‐1 in glioma

Abstract Glioma is one of the most commonly observed tumours, representing approximately 75% of brain tumours in the adult population. Generally, glioma therapy includes surgical resection followed by radiotherapy and chemotherapy. The transcription factor STAT3 (signal transducer and activator of transcription 3) is a promising target for the treatment of cancer and several other diseases. At nanomolar concentrations, SD‐36 induces rapid cellular degradation of STAT3 but cannot degrade other STAT proteins. The current study demonstrates the therapeutic efficacies of the STAT3 degraders SD‐36 against glioma, as well as understanding the elucidating mechanisms and identifying molecular markers that determine cell sensitivity to STAT3 degraders. Glioma cell lines possessed similar response patterns to SD‐36 but different responses to the STAT3 inhibitor Stattic. SD‐36 potently induced apoptosis in glioma cells along with a reduction in Mcl‐1 levels, which are critical for mediating the induction of apoptosis and enhancing TMZ‐induced apoptosis. Accordingly, SD‐36 sensitizes the antitumour effect of TMZ in patient‐derived xenograft. In addition, the downregulation of Mcl‐1 expression‐mediated antitumour effect of SD‐36 was analysed in cell‐derived xenograft. These observations need to be validated clinically to confirm the efficacy of STAT3 degraders in glioma.

in its state with no unphosphorylation. 8 Oncogenesis is promoted by macroenvironment inflammatory signals and the tumour microenvironment and sustained via the expression and release of factors that facilitate survival, including IL-6 (interleukin-6), 8,9 thereby creating an effective loop of endless feedback of autocrine and paracrine signalling to initiate and sustain reprogramming of metabolism and stalling apoptosis. Moreover, interactions of STAT3 and supercomplexes of transcription factors facilitate signalling pathways of oncogenes. 5,10 The PROTAC (PROteolysis TArgeting Chimera) technique was found during the identification of peptides or small chemical ligands that particularly bind with endogenous E3 ligases, including FBXW1A (F-box/WD repeat-containing protein 1A, also called b-TRCP1), MDM2 (mouse double minute 2 homolog), the von Hippel-Lindau tumour suppressor, and CRBN (cereblon). 11,12 PROTAC is structurally a small molecule consisting of two functional parts, a highly specific 'warhead' towards the POI (binding protein of interest) and a second part that acts as a ligand recognized by the E3 ligase and is connected via a linker. 13 PROTACs provide an alternate approach for targets that may be undruggable, including transcription factors. 14 SD-36, one of the STAT3 degraders, possibly induces in vitro and in vivo STAT3 protein degradation and exhibits high selectivity compared with other STAT members. 15 SD-36 inhibits the growth of acute myeloid leukaemia and anaplastic large-cell lymphoma cell lines, 16 although its effect on glioma is still elusive.
In the current study, we compared the effects of SD-36 and Stattic on the growth of glioma cells and demonstrated their therapeutic mechanisms.

| Cell culture
The cell lines U251 and U87 from humans were procured from the ATCC (Manassas, VA) and grown in RPMI 1640 medium along with foetal bovine serum (5%) in humid air with 5% CO 2 at 37℃.
Stable Mcl-1 overexpression was achieved by infecting a given cell line for 24 hours with lentiviruses carrying empty vector or Mcl-1 as described previously. 17,18 On receiving the cells, annual detection of mycoplasma was performed using the Mycoplasma Detection Kit Nozan MycoAlert (Sigma, USA) to make sure that they were free from any mycoplasma. SD-36 and Temozolomide (TMZ) were obtained from MedChemExpress. S63845 was purchased from Chemietek (Indianapolis, IN).

| Cell viability assay
Cell viability was analysed as previous study. 18 Seeding

| Matrigel invasion assay
Transwell chambers in 24-well plates were got from ThermoFisher (USA) and used to perform the invasion assay. Matrigel was used to coat the Transwell chamber (BD Bioscience, USA). Cells were trypsinized 24 hours following SD-36 treatment and were then incubated in the top chamber for 24 hours without serum. The medium supplemented with serum (10%) was placed in the bottom chamber.
Non-invaded cells were scraped off. Cells that invaded through the filter were fixed with 4% paraformaldehyde and stained with haematoxylin.

| Apoptosis
The apoptosis detection kit with annexin V/PI from BD Biosciences (San Jose, CA) was used to detect apoptosis. Further, the ELISA kit for Cell Death Detection from Roche Diagnostics (Indianapolis, IN) was used to determine DNA fragments associated with histones as per the provided instructions. Western blotting was carried out to determine apoptosis and protein cleavage.

| Western blotting
Western blotting was performed following previous studies. 19,20 In brief, glioma cells were lysed for 20 minutes using RIPA (radioimmunoprecipitation assay) buffer with protease inhibitors on ice and then centrifuged for 60 minutes at 20 000 g and 4℃. Then, the BCA assay using a Pierce BCA protein assay kit from Thermo Scientific, Inc, was carried out to determine the protein content in the separated supernatants. Fifty micrograms of protein were resolved on a 10% PAGE gel and electronically transferred onto nitrocellulose membranes for 60 minutes using a Bio-Rad Trans-Blot (both from Bio-Rad Laboratories). Blocking buffer (0.05% PBS and 1% skim milk) was used on the membrane for 60 minutes at room temperature, followed by the addition of primary antibodies for one full day at 4℃.
The membrane was washed using washing buffer (0.05% PBS and Tween-20), and a secondary antibody conjugated with HRP (horseradish peroxidase) was carried out for 60 minutes at room temperature. This step was followed by three washes using wash buffer, treatment with the western blotting detection reagent ECL Prime

| Real-time PCR
Real-time PCR was performed following previous studies. 21   week, ip), or a combination. Callipers were used to measure tumour volumes and determined using the formula V = ½ × L × W 2 , where V is tumour volume, W is tumour width, and L is tumour length. As the treatment ended, the weights of the mice were determined, and the mice were euthanized by CO 2 asphyxiation. The tumours were then removed, weighed, and frozen in liquid nitrogen for further analyses.

| Xenograft
Approval for experiments on animals was obtained from the Xingtai People's Hospital Institutional Animal Care and Use Committee (IACUC). For treatments, SD-36 (5 mg/kg/d; twice/week, ip) was used. Callipers were used to measure tumour volumes and determined using the formula V = 1/2 × L×W 2 . As the treatment ended, the weights of the mice were determined and euthanized by CO 2 asphyxiation. The tumours were then removed, weighed, and frozen in liquid nitrogen for further analyses.

| Immunohistochemical (IHC) staining
Antigen extraction was performed using 0.01 mol/L citrate buffer after pre-treatment of the sections with xylene and rehydration through a graded alcohol series. Background activity was removed with hydrogen peroxide. Goat Serum (Sigma-Aldrich, USA) was then used to treat the tissue sections for 30 minutes to block non-specific binding. Afterward, the indicated primary antibodies were added and the slides were incubated overnight at 4℃. After DAB staining, counterstained the slices with haematoxylin, and mounted in neutral gum to observe the paraffin sections. The tissues were then analysed by a bright-field microscope.

| Statistical analysis
Two-sided unpaired Student's t tests analysed the statistical significance of variations between two groups in terms of tumour sizes or weights in the case of equal variances. The same software was used to examine the data to confirm the assumptions for using the t tests.
One-way ANOVA assessed the differences between multiple treatments. At P < .05, outcomes were statistically significant.

| SD-36 suppresses the cell growth, invasion and induces apoptosis of human glioma cell lines
We first tested the responses of U87 and U251 glioma cell lines to the STAT3 degrader SD-36 and Stattic using a 3-day cell viability assay and found that the survival rate of these cell lines decreased more strongly with SD-36 in general than Stattic treatment ( Figure 1A).
Next, we determined the effects of SD-36 on invasion in glioma cells. As shown in Figure 1B, SD-36 treatment decreases cell invasion in U87 and U251 cells, which was detected by Transwell assay.
Furthermore, the effect of SD-36 on epithelial-mesenchymal transition (EMT) was detected by western blotting. Our findings revealed that SD-36 treatment resulted in increase in E-cadherin and decrease in N-cadherin, vimentin and snail in U87 and U251 cells ( Figure 1C).
To determine whether SD-36 suppresses the growth of glioma cells via apoptotic induction, we next compared the effects of  Figure 1E).
Moreover, SD-36-induced apoptosis was blocked by z-VAD-fmk pre-treatment ( Figure 1F). Together, these results indicate that the tested STAT3 degrader SD-36 possesses more potent activities than Stattic in inducing apoptosis of human glioma cells.

| SD-36 effectively decreases STAT3 levels and exerts different effects from Stattic on the modulation of Mcl-1 levels in human glioma cells
To

| Mcl-1 suppression is required for SD-36-induced apoptosis in glioma cells
To

| SD-36 synergizes with TMZ to enhance apoptosis
We

| SD-36 sensitizes TMZ-induced apoptosis in PDX mouse model
The effect of SD-36 and TMZ in the PDX model was evaluated. Each single agent was compared against the SD-36 and TMZ combination in the model ( Figure 5A-C). The combination was found to exhibit Cotreatment induced more apoptosis in tumours than single treatment ( Figure 5H). The above data indicate that SD-36 sensitizes TMZ in the PDX model.

| Effective inhibition of human glioma xenograft growth by SD-36
Next, we investigated the effect of SD-36 on the growth of pa-

| DISCUSS ION
A common cancer type of the central nervous system, glioma, is a primary reason for deaths due to cancer because of high-grade invasion and growth of glioma cells. 24 Treatment for glioma patients includes radiotherapy, surgery, chemotherapy, and a combination of these. 25 Furthermore, patients often have a poor prognosis with a 5-year survival rate of nearly 10%. 26 Clinically, a common drug for chemotherapeutic glioma treatment is TMZ. 27 Nevertheless, TMZ resistance is often observed during treatment. 3,28 Hence, it is imperative to discover new agents for better outcomes by examining the mode of occurrence and progression of glioma.
The current study has clearly shown that the novel STAT3 degrader SD-36 15 30 Tumour proliferation and progression are promoted by activated STAT3 via gene expression regulation participating not only in cancer cell invasion and survival but also in immune escape and angiogenesis in the microenvironment of the tumour. 31 Constitutive activation of STAT3 occurs in solid and haematologic tumours, but activation is transient in normal cells. 5 Therefore, STAT3 is a potentially attractive target for cancer treatment.
In our study, we describe the novel finding that SD-36 de- The molecules related to TMZ resistance in GBM include MGMT (O6-methylguanine-DNA methyltransferase), MPG (Nmethylpurine DNA glycosylase), and HIF-1α (hypoxia-inducible factor 1α). 3 This resistance against TMZ is mediated by enhancing MGMT levels. 32 In particular, the methylation of the MGMT promoter is a predictive biomarker associated with alkylating chemotherapy, especially TMZ. 33 Various other mechanisms of action recently associated with TMZ resistance have been observed. 34 For instance, downregulating HIF-1α expression increased the sensitivity of GBM cells to TMZ. 35 In GBM, hedgehog (Hh) signalling is deregulated and correlates with TMZ resistance. 36 Further, connexins, especially Cx43 or the gap junction protein connexin 43, participate in the malignant glioma microenvironment and confer resistance of GBM cells to chemotherapeutic agents, such as TMZ. 37 Recently, PI3K pathway activation was also shown to be related to TMZ resistance. 38  In this study, we observed that SD-36 possessed more potent activity than Stattic in inhibiting the growth of human glioma.
Downregulation of Mcl-1 expression is necessary for SD-36-induced apoptosis. Moreover, SD-36 enhances the antitumour activity of TMZ in glioma cells and PDX model. Therefore, further improvement in the compound or optimization of its dosages or administration routes to achieve better anticancer activity with acceptable safety profiles and future testing of its anticancer activity in the clinic are warranted.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.