Abstract

Background

Triple-negative breast cancer (TNBC) is an aggressive cancer type associated with poor prognosis and limited therapeutic options. Due to the absence of targetable receptors, conventional chemotherapy remains the primary treatment approach. ERCC1 is a critical component of the nucleotide excision repair system, responsible for repairing DNA damaged by platinum-based agents like carboplatin. Heat shock response (HSR) is a fundamental, stress-induced defense mechanism that supports cancer cell survival.

Aims

This study aims to investigate the effects of HSR inhibition by KNK437 on ERCC1 gene expression and carboplatin sensitivity in the TNBC cell line MDA-MB-231.

Methods

The IC₅₀ values of carboplatin and KNK437 were determined using WST-8 cytotoxicity assay. ERCC1 gene expression levels were quantified by real-time quantitative polymerase chain reaction. Apoptotic and necrotic cell death induced by carboplatin and KNK437 was assessed by flow cytometry using FITC-Annexin V assay.

Results

The IC₅₀ values of carboplatin and KNK437 were 247.5 µM and 89.74 µM, respectively. Carboplatin or KNK437 monotherapy significantly decreased ERCC1 expression by 42.8% and 49.5%, respectively, while their combined application caused a 54.9% reduction. Furthermore, co-treatment markedly increased total cell death by 34.1% compared to carboplatin alone. Interestingly, necrosis induced by carboplatin shifted toward apoptosis upon co-treatment with KNK437, as confirmed by light microscopy and flow cytometry.

Conclusion

HSR inhibition by KNK437 enhances carboplatin sensitivity in TNBC cells and downregulates ERCC1 gene expression. Given the aggressive nature of TNBC and its limited treatment options, our results suggest that KNK437 may offer therapeutic advantages when combined with carboplatin, particularly in contexts where carboplatin-induced necrosis contributes to inflammation-related complications.

Introduction

Triple-negative breast cancer (TNBC) is a distinct breast cancer subtype characterized by the lack of estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2 (HER2) proteins (Bou Zerdan et al., 2022). TNBC represents a heterogeneous group of clinically challenging tumors, distinguished by early recurrence, poor overall survival, and development of resistance to chemotherapy. Compared to other subtypes, the therapeutic options for TNBC are limited as it does not respond well to hormone-based therapies or HER2-targeting drugs. Additionally, only a small fraction of TNBC patients respond effectively to novel targeted treatment methods and immunotherapeutic approaches. Therefore, platinum-based chemotherapy, either individually or in combination with these novel agents, continues to serve as the mainstay of TNBC treatment (Lee, 2023). However, there is an urgent need to identify new prognostic and predictive biomarkers for TNBC that can help improve treatment outcomes and guide more effective and individualized therapeutic strategies.

Platinum-based agents are compounds that exert cell apoptosis or necrosis by inducing the formation of adducts and crosslinks within the DNA helix, ultimately disrupting replication and transcription (Brabec & Kasparkova, 2005; de Sousa, 2014). A meta-analysis incorporating 12 clinical studies covering 4,580 patients reported that platinum-based chemotherapy significantly improves pathological complete response rates and prognosis in TNBC patients, with greater efficacy compared to non-platinum regimens (Lin et al., 2022). Carboplatin is a second-generation platinum-based chemotherapeutic agent that exhibits a more favorable toxicity profile compared to its predecessor, cisplatin, with notably suppressed nephrotoxicity, neurotoxicity, and gastrointestinal side effects. Several meta-analyses have demonstrated that the inclusion of carboplatin in neoadjuvant/adjuvant chemotherapy markedly improved disease-free and overall survival in TNBC (Bian et al., 2021; Pathak et al., 2022).

The nucleotide excision repair (NER) pathway is a critical DNA repair mechanism that removes a wide range of helix-distorting lesions, such as intrastrand crosslinks and bulky adducts induced by ultraviolet (UV) radiation, chemical mutagens, and certain chemotherapeutic agents. Excision repair cross-complementation group 1 (ERCC1) is a key component protein of the NER pathway; it forms a heterodimeric complex with Xeroderma Pigmentosum group F (XPF), an endonuclease, which catalyzes an excision step close to the site of DNA damage, crucial for the restoration of genomic integrity (Marteijn et al., 2014). Therefore, ERCC1 expression levels may be useful in predicting the responses of TNBC patients to platinum-based chemotherapy, with high or low ERCC1 levels associated with a poor or favorable response, respectively (Hashmi et al., 2025; Ma & El Kashef, 2017; Sidoni et al., 2008).

Heat shock response (HSR) is a highly conserved cell defense mechanism induced by various forms of stress, functioning to maintain protein homeostasis (proteostasis) under unfavorable conditions. Heat shock factor1 (HSF1) is the principal transcriptional regulator of the HSR and the expression of heat shock proteins (HSPs)/chaperones. Beyond its classical role in proteostasis, HSF1 promotes oncogenic transformation by coordinating a broad transcriptional network that encompasses pathways involved in protein synthesis, cell proliferation, survival, adhesion, and energy metabolism (Dai et al., 2007; Mendillo et al., 2012; Yun et al., 2019). KNK437 is a synthetic benzylidene lactam that inhibits the HSR by suppressing HSF1 transcriptional activity, and thus the expression of various HSPs, thereby exerting antitumor effects against various cancer types (Powers & Workman, 2007; Yun et al., 2019).

A recent study investigating the stability dynamics of the ERCC1—XPF complex proposed that the complex maintains its stability and cellular levels with the help of DCAF7, the proper folding of which is mediated by the molecular chaperone TRiC (Kawara et al., 2019). Based on this finding as well as evidence from previous research, we hypothesized that KNK437 may alter ERCC1 expression in TNBC cells, thereby modulating the cytotoxicity induced by carboplatin. Therefore, the present study aims to elucidate the impacts of the HSR inhibitor KNK437 on ERCC1 gene expression and carboplatin sensitivity in TNBC cells.

Materials and Methods

Human TNBC Cancer Cell Culture

The human TNBC cell line MDA-MB-231, obtained from the American Type Culture Collection (VA, USA), was used as a study model. The cells were cultured in GlutaMAX™ Dulbecco’s Modified Eagle Medium (Thermo Fisher Scientific, MA, USA) supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 µg/mL streptomycin. Cells were cultured at 37 °C in a Heracell 240i™ incubator and humidified atmosphere containing 5% CO₂ (Thermo Fisher Scientific) and subcultured every 2—3 days upon reaching a confluency of approximately 85%—90%. For passaging, the cells were detached using 0.05% trypsin—EDTA at 37 °C. Viable cells were counted with a hemocytometer after Trypan blue staining and reseeded at appropriate densities into new plates. All cell culture reagents used were purchased from Gibco (Thermo Fisher Scientific).

Cytotoxicity Analysis and IC₅₀ Determination

To determine the IC₅₀ values of carboplatin (C2538; Sigma-Aldrich Chemicals, MO, USA) and KNK437 (HY-100110; MedChemExpress, NJ, USA), cells were seeded into 96-well plates at a density of 1 × 10⁴ cells/well, 24 h before drug treatments. A wide range of concentrations, each at 2-fold serial dilution, was employed to accurately determine the IC₅₀. The cells were treated with carboplatin or KNK437 at final concentrations of 2000, 1000, 500, 250, 125, 62.5, 31.25, 15.62, 7.81 µM; and 500, 250, 125, 62.5, 31.25, 15.62, 7.81 µM, respectively, for 48 h. Then, cell viability was assessed using the WST-8 assay (Abcam, Cambridge, UK), according to the manufacturer’s instructions. In brief, 10 μL of the WST-8 reagent was added to each well, and the plates were incubated at 37 °C for 2 h. The OD₄₆₀ was then measured using a Syn4 spectrophotometer (Biotek Instruments, Inc., VT, USA), in duplicate. The mean absorbance values were corrected by subtracting the absorbance of blank wells containing only culture medium. Dose-response curves were generated by setting the viability of untreated control cells to 100%, and IC₅₀ values were calculated using the software GraphPad Prism version 10.0 (https://www.graphpad.com/).

Total RNA Isolation

Cells were seeded into 6-well plates at a density of 2.5 × 10⁵ cells/well. The next day, they were treated with various concentrations of dimethyl sulfoxide (DMSO) (vehicle control for KNK437), carboplatin, KNK437, or carboplatin + KNK437 for 48 h. Following treatments, the culture medium was removed, and the cells were gently washed with 1X PBS. Then, TRIzol reagent (Invitrogen, Thermo Fisher Scientific) was added, the lysates were collected into a microcentrifuge tube by scraping, and incubated at room temperature for 10 min to allow complete homogenization. Chloroform was added to the lysates at a ratio of 1:5 (v/v), followed by vigorous shaking and incubation at room temperature for 5 min. For phase separation, solutions were cold-centrifuged using a 3K30 refrigerated benchtop centrifuge (Sigma Aldrich) at 11,500×g for 15 min. The clear upper aqueous phase containing the total RNA was carefully transferred to a fresh tube. An equal volume of 70% RNase-free ethanol was added slowly and gently mixed by pipetting. The mixture was then loaded onto a silica-based spin column provided with the FavorPrep™ RNA isolation kit (Favorgen Biotech Corp., Ping Tung, Taiwan), and then the manufacturer’s instructions were followed. Total RNA was finally eluted in 50 µL of RNase-free water and stored at -80 °C until further use. RNA concentration was measured, and purity was assessed with a Jenway Genova Nano spectrophotometer (Bibby Scientific, Staffordshire, UK).

cDNA Synthesis and Quantitative Real-Time Polymerase Chain Reaction (qPCR)

For cDNA synthesis, 500 ng of the total RNA from each sample was reverse transcribed in a 20 µL reaction volume containing random hexamer primers, oligo(dT), and MMLV reverse transcriptase, provided with the iScript™ cDNA synthesis kit (Bio-Rad Laboratories, CA, USA), following the manufacturer’s protocol. The thermal cycle profile used was as follows: 5 min at 25 °C (primer annealing), 30 min at 46 °C (cDNA synthesis), and 1 min at 95 °C (enzyme inactivation); GeneAmp 9700 PCR System (Applied Biosystems, Thermo Fisher Scientific). Freshly prepared cDNA was processed immediately for qPCR. The qPCR reaction mix with a total volume of 20 µL contained 1 µL of cDNA and 250 nM each of the forward and reverse primers specific to ERCC1 and the reference GAPDH. An SsoAdvanced™ Universal SYBR^® Green Supermix kit (Bio-Rad) was used, and the protocol provided was applied. The primer sequences were designed in-house using the PrimerBLAST tool (https://www.ncbi.nlm.nih.gov/tools/primer-blast/), set up with the universal primer design criteria. The 5’—3’ sequences of the forward (F) and reverse (R) primers were F, GAPDH: CCATCTTCCAGGAGCGAGATC; R, GAPDH: GGCAGAGATGATGACCCTTTTG; and F, ERCC1: TTTGGCGACGTAATTCCCGA; R, ERCC1: CCTGCTGGGGATCTTTCACA. The thermal cycle profile employed was 30 sec at 98 °C (initial denaturation), followed by 40 cycles each of 10 sec at 98 °C (denaturation) and 30 sec at 60 °C (annealing/extension), and a CFX Connect™ Real-Time PCR Detection System (Bio-Rad) was employed. Melting curves were analyzed at the end of each run to confirm amplicon specificity. All reactions were performed in duplicate. Relative fold changes in ERCC1 expression levels between the different drug treatment groups were calculated based on mean C_t values and utilizing the 2^‒ΔΔCt method, and analyzed with REST 2009 software (QIAGEN, Venlo, Netherlands).

Agarose Gel Electrophoresis

To verify the specificity and expected sizes of the qPCR amplicons (146 and 175 bp for GAPDH and ERCC1, respectively), agarose gel electrophoresis was performed. Briefly, 2% agarose (HS-8000, Prona, Biomax) gel was prepared in 1X TAE buffer and added with a SafeView™ Classic Nucleic Acid Gel stain (Applied Biological Materials, BC, Canada). After the qPCR run was completed, 10 µL of the amplicons from two randomly selected wells per target gene (GAPDH and ERCC1) was mixed with 2 µL of 6X DNA loading dye (TriTrack, Thermo Fisher Scientific), and loaded into the gel wells along with a DNA ladder (GeneRuler 50 bp, Thermo Fisher Scientific). Electrophoresis was carried out at 100 V for 45 min using a Sub-Cell GT Horizontal Electrophoresis System (Bio-Rad), and the gel was subsequently imaged under UV transillumination on a ChemiDoc Go Imaging System (Bio-Rad).

Apoptosis/Necrosis Analysis by Annexin V Assay

To examine the effects of KNK437 on carboplatin-induced cytotoxicity in MDA-MB-231 cells, apoptosis and necrosis were assessed using an FITC-Annexin V apoptosis detection kit (BD Biosciences, NJ, USA), following the manufacturer’s instructions. Cells were seeded into 6-well plates at a density of 2.5 × 10⁵ cells/well. The next day, cells were treated with various concentrations of DMSO (vehicle control for KNK437), carboplatin, or KNK437, or combinations of carboplatin and KNK437 for 48 h. Next, the culture medium of each well containing the floating, dead cells was taken into a fresh tube; later, the media were pooled along with the corresponding adherent cells collected by trypsinization. The cells were gently washed with cold 1X PBS, and approximately 1 × 10⁵ cells were resuspended in 100 µL of 1X Annexin V binding buffer. For staining, 2.5 µL of FITC-Annexin V and 2.5 µL of 7-AAD were added to each sample, gently mixed, and incubated at room temperature for 20 min in the dark. Finally, 400 µL of 1X Annexin V binding buffer was added to each tube, and cells were immediately analyzed using a BD FACS Calibur™ flow cytometer (BD Biosciences). Fluorescence signals in the FL1 (530/30) channel for FITC-Annexin V and the FL3 (670 LP) channel for 7-AAD were detected. A total of 20,000 gated events were acquired per sample, and quadrant lines were set employing the appropriate single- and double-stained controls. Data were analyzed using the BD CellQuest Pro™ software (BD Biosciences), and the proportions of necrotic (7-AAD⁺ only), late apoptotic/necrotic (7-AAD⁺ and FITC-Annexin V⁺), and total dead (the sum of these two populations) cells were determined (in %) based on quadrant statistics.

Statistical Analysis

Statistical analyses and IC₅₀ value calculations were performed using GraphPad Prism software version 10.0 (GraphPad Inc., CA, USA). Data are presented as the mean ± standard deviation (SD). Statistical evaluation of the relative fold changes in gene expression was conducted using the REST 2009 software (Qiagen). Student’s t-test was applied for pairwise comparisons in the Annexin cell death assay. The p-values were considered statistically significant at *p < 0.05, **p < 0.01, and ***p < 0.001.

Results

Determination of IC₅₀ Values for Carboplatin and KNK437 in MDA-MB-231 Cells

To determine the IC₅₀ values of carboplatin and KNK437, MDA-MB-231 cells were seeded in 96-well plates and treated with 2-fold serially increasing concentrations of carboplatin (7.81 µM to 2 mM) or KNK437 (7.81 µM to 500 µM) for 48 h. Cell viability was measured by the WST-8 assay, and dose-response curves were generated utilizing GraphPad Prism. The IC₅₀ values obtained were 247.5 µM for carboplatin (Figure 1A) and 89.74 µM for KNK437 (Figure 1B), comparable to those of previous reports (Barrio et al., 2013; Wang et al., 2010).

Effects of Carboplatin and KNK437 on ERCC1 Gene Expression as Single or Combined Treatments

To investigate the effects of carboplatin and KNK437, administered either alone or in combination, on ERCC1 expression in MDA-MB-231 cells, a series of real-time qPCR experiments was conducted. Before proceeding with the main investigations, the lengths of the qPCR amplicons were verified by agarose gel electrophoresis. As shown in Figure 2A, amplicon bands corresponding to the expected sizes of the target genes—146 bp for GAPDH and 175 bp for ERCC1—were detected. The sharpness and positions of the bands, without any non-specific products, confirmed both the accuracy and specificity of the primer pairs designed for the target genes.

To assess the effect of carboplatin on ERCC1 gene expression, the IC₅₀ of the compound was used as the median dose, based on which the treatment concentrations were determined in a series of 2-fold dilutions. Accordingly, cells were treated with increasing concentrations of carboplatin (dissolved in water) ranging from 62.5 µM to 1 mM for 48 h. qPCR analysis revealed that ERCC1 expression decreased significantly by 30%—40% across the concentrations tested compared to the untreated control (Figure 2B). Notably, carboplatin did not exhibit any dose-dependent suppression, as expression was reduced to similar levels, regardless of concentration.

To evaluate the effects of KNK437 on ERCC1 expression, the cells were treated with increasing concentrations of KNK437 (dissolved in DMSO) ranging from 31.25 µM to 500 µM for 48 h. This concentration range was estimated based on the relative fold difference between the IC₅₀ values of carboplatin and KNK437. To determine whether DMSO influenced ERCC1 expression, the cells were also treated with only 2.5% (v/v) DMSO, the same concentration to which the cells were exposed when treated with the highest dose of KNK437 at 500 µM. As shown in Figure 2C, qPCR analysis revealed that ERCC1 expression was remarkably reduced by 40%—60% across all concentrations tested, compared to the untreated control. However, treatment with DMSO alone also suppressed ERCC1 expression by 60%, a level that was similar to that observed with 500 µM KNK437, indicating that the downregulation observed at 500 µM may not be due to KNK437, but rather DMSO (Figure 2C). Therefore, we decided to perform an additional set of qPCR experiments using DMSO alone to determine a DMSO concentration that did not interfere significantly with ERCC1 expression.

The cells were exposed for 48 h to DMSO alone at 0.078%, 0.156%, 0.312%, and 0.625% (v/v), corresponding to the DMSO contents of the 15.62 µM, 31.25 µM, 62.5 µM, and 125 µM KNK437 treatments, respectively. Compared to the untreated control, ERCC1 expression levels did not show any statistically significant alterations at any concentration tested (p > 0.05) (Figure 2D). However, as a slight decline was observed at 0.625% DMSO, a onefold lower concentration—0.312%—was selected for the subsequent combination treatment experiments. A comparative analysis of Figures 2C and 2D clearly indicated that 62.5 µM KNK437 reduced ERCC1 expression by approximately 50%, an effect which was also independent of any interference by DMSO.

Considering the proportional difference between the IC₅₀ values of the two agents, 125 µM carboplatin was used in combination with 62.5 µM KNK437. To assess the synergistic effect of carboplatin and KNK437 on ERCC1 expression, the cells were treated with only 125 µM carboplatin, only 62.5 µM KNK437, or a combination of the two for 48 h. Compared to the untreated control, carboplatin or KNK437 alone suppressed gene expression by 42.8% and 49.5%, respectively, whereas their combined application induced a 54.9% reduction (Figure 2E). Such results with single-agent treatments were consistent with our findings obtained previously at the two respective doses (Figure 2B for 125 µM carboplatin and Figure 2C for 62.5 µM KNK437). When the reduction ratios with single and combined treatments were compared, it was observed that KNK437 exerted a more robust influence than carboplatin in suppressing ERCC1 gene expression (Figure 2E, carboplatin alone vs. combination: 21%, KNK437 alone vs. combination: 10%).

 Effects of Carboplatin and KNK437 on Cell Death in MDA-MB-231 Cells as Single or Combined Treatments  

To investigate the cytotoxic effects of carboplatin and KNK437, administered either alone or in combination, on MDA-MB-231 cells, a series of flow cytometry-based Annexin V assays was conducted. Based on previous experience from our studies investigating the impacts of carboplatin on the same cell line, a 1 mM dose was selected, as it consistently induced total cell death at ~40%. It was regarded as the baseline appropriate for the drug combination experiments, considering the potential of the second drug to either enhance or suppress this cytotoxicity ratio. Based on the proportional differences between the IC₅₀ values of the two agents, doses of 250 and 500 µM KNK437 were selected.

Cells were treated with 2.5% DMSO alone (v/v; corresponding to the DMSO content at 500 µM KNK437), 1 mM carboplatin, 250 µM or 500 µM KNK437, and varying combinations of both drugs, for 48 h. The cells were subsequently visualized under an inverted phase-contrast light microscope (Figure 3A). Compared to the untreated cells (control), DMSO-only-treated cells exhibited a phenotypic shift toward an elongated, spindle-like morphology. Similar morphological changes were observed in the cells treated with 250 and 500 µM KNK437, suggesting that these effects were largely induced by DMSO. In contrast, the cells treated with only 1 mM carboplatin displayed morphological features, such as membrane disruption, cell disintegration, and swelling, which are characteristic of necrosis (white arrows, Figure 3A). Remarkably, a combination of carboplatin and KNK437 dramatically shifted the morphology, to a one characterized by rounded cells with intact membranes and evident shrinkage, indicative of apoptosis (black arrows, Figure 3A). The cell morphologies observed were consistent with the criteria well-established in the literature for distinguishing apoptosis and necrosis (Balvan et al., 2015).

Following the morphological analyses, to evaluate the cytotoxic impacts of carboplatin and KNK437, FITC-Annexin V assays were performed, followed by flow cytometry. MDA-MB-231 cells were treated for 48 h with carboplatin and KNK437, either alone or in combination, at the concentrations mentioned before, and the proportions of necrotic (7-AAD⁺ only), late apoptotic/necrotic (7-AAD⁺ and FITC-Annexin V⁺), and total number of dead cells were quantified (%) by applying quadrant statistics (Figures 3B and 3C). Thus, our results indicate that KNK437 alone did not induce any remarkable cytotoxicity. At a lower dose of 250 µM, the total cell death rate remained comparable to that of the untreated and DMSO controls (~10%), while the 500 µM dose induced only a modest enhancement of 21%. In contrast, 1 mM carboplatin alone resulted in a total cell death of 42% (Figure 3C), with a majority of the dead cells located within the necrotic (lower right) quadrant (Figure 3B), consistent with their morphology, typical of necrotic cells, previously observed under light microscopy (Figure 3A). Remarkably, co-treatment with 250 µM KNK437 did not induce any substantial change compared to carboplatin alone (both 42%—44%, Figure 3C). However, a distinct shift from necrosis to late apoptosis was observed, reflected by a redistribution of the cell population from the lower right to the upper right quadrants (Figure 3B). Such an effect was more pronounced with a combination of carboplatin with 500 µM KNK437, at which the total cell death enhanced to 56.3%, representing a significant increase of 34.1% compared to carboplatin alone (Figure 3C). Such a trend was again largely due to the shift from necrosis (carboplatin alone) to apoptosis (combination with KNK437) (Figure 3B), consistent with their respective morphologies previously observed using light microscopy (Figure 3A). Collectively, our light microscopy and flow cytometry findings demonstrate that necrosis induced by carboplatin is replaced by apoptosis upon co-treatment with KNK437.

Discussion

Platinum-based chemotherapy, either as a monotherapy or combined with other cancer therapeutics, remains the cornerstone of current TNBC treatments. However, a critical need to discover reliable prognostic and predictive biomarkers and thereby enhance clinical outcomes, guide personalized treatment approaches, and elucidate the mechanisms underlying therapeutic resistance and recurrence still arises (Lee, 2023). The present study investigated the effects of single and combined applications of carboplatin and the HSR inhibitor KNK437 on apoptosis or necrosis-induction and ERCC1 gene expression in the TNBC cell line MDA-MB-231.

The NER pathway is a mechanism critical to the repair of DNA damage caused by platinum-based chemotherapeutic agents. Upregulation of the genes involved in the NER enables cancer cells to effectively repair such damage, ultimately contributing to the development of resistance (de Sousa, 2014).

Elevated ERCC1 mRNA and protein levels in cisplatin-resistant human ovarian cancer cells have been reported (Li et al., 2000). Similarly, ERCC1 mRNAs were markedly higher in tumor biopsies of cisplatin-resistant gastric cancer patients (Metzger et al., 1998). In contrast, XPA, XPF, and ERCC1 expression was low in cisplatin-sensitive testicular cancer cell lines, indicating their limited capacity to mitigate cisplatin-induced DNA damage (Welsh et al., 2004). Similarly, in our study, since MDA-MB-231 cells are not resistant to carboplatin, their primal response to carboplatin-induced DNA damage was to suppress ERCC1 expression (Figure 2B), and commit to cell death rather than to DNA repair and survive (Figure 3C). Moreover, a combination of carboplatin and KNK437 further decreased ERCC1 expression (Figure 2E), thereby enhancing the sensitivity of MDA-MB-231 cells to carboplatin and ultimately increasing cell death levels compared to that induced by carboplatin alone (Figure 3C). This finding of ours where HSF1 was chemically inhibited by KNK437, is consistent with a previous report in which the shRNA-mediated knockdown of HSF1 also rendered MDA-MB-231 cells more sensitive to carboplatin (Desai et al., 2013).

DMSO is a polar, amphipathic, aprotic organic molecule that is ideal for dissolving poorly soluble polar and non-polar compounds, due to which it is widely used in toxicology and pharmacology studies, and also for the cryopreservation of cells. Although conventionally DMSO is considered non-toxic at concentrations ≤ 10% (v/v), exposure of 3D cardiac and hepatic microtissues to DMSO at levels as low as 0.1% (v/v) markedly altered the functioning of cellular processes, including significant changes in transcription and translation as well as the epigenetic landscape, as revealed by comprehensive transcriptomic, proteomic, and DNA methylation profiling (Verheijen et al., 2019). To further elucidate the cytotoxic effects of DMSO at the concentration ranges (0.1%—1.5% [v/v]) typically used in cell-based studies, epithelial colon cancer cells were treated with DMSO; and it has been reported that DMSO alters both DNA and protein topologies, in addition to significantly suppressing the expression of certain proteins at concentrations ≥ 0.5% (v/v) (Tuncer et al., 2018). Similarly, we observed that DMSO, used as a solvent for KNK437, reduced ERCC1 expression at concentrations ≥ 0.625% (v/v) (Figures 2C and 2D). Therefore, additional qPCR assays using increasing concentrations of only DMSO were conducted to determine a concentration that did not impact ERCC1 transcription. Consequently, we identified 0.312% (v/v) as the maximum level that did not induce any detectable background suppression of ERCC1 expression, and thus, 62.5 μM was the highest KNK437 concentration that could be used free of any interference by DMSO (Figure 2D). The subsequent qPCR experiments involving combined drug treatments employed this level as the reference (Figure 2E).

The Annexin V cell death assays conducted later demonstrated that even 2.5% (v/v) DMSO, corresponding to 500 µM KNK437, did not remarkably enhance cell death compared to the untreated control (Figure 3C), but induced distinct changes in cell morphology, including elongation and the production of spindle-like extensions (Figure 3A). In line with the findings of the two studies mentioned earlier, our results highlight the need for considerable caution when using DMSO as a solvent, as its toxic effects can manifest at the level of gene expression without being reflected by cell death rate. Although DMSO is widely regarded as a “universal solvent” with a diverse range of biological applications, its potential adverse effects should not be overlooked, and the working concentration should be minimized to the extent possible.

The effects of DMSO on TNBC cell morphology observed in our study are consistent with several previous reports. For instance, polyclonal human ovarian cancer cells exposed to 1% DMSO for several days displayed a similar spindle-like morphology and a monolayer growth pattern characterized by contact inhibition (Grunt et al., 1991), as was also reported in several human lymphoblastoid and HeLa cell lines exposed to 2% DMSO suggesting that such a morphological response was independent of tissue origin (Aranda-Anzaldo et al., 2024). Our observations agreed with these studies, as exposure to 2.5% (v/v) DMSO caused the MDA-MB-231 cells to adopt a spindle-like morphology (Figure 3A). Similar morphological changes were also detected in cells treated with 250 and 500 µM KNK437 (Figure 3A), which were most likely due to the background effect of DMSO.

KNK437 inhibits the HSR by blocking the transcriptional activity of HSF1 (Powers & Workman, 2007). Our observation that ERCC1 gene expression decreased significantly even at low doses of KNK437 (31.25 µM, Figure 2C) suggested that ERCC1 might indeed be a genuine transcriptional target of HSF1. To explore this possibility, we comprehensively reviewed the literature and examined the supplementary data of high-throughput genome-wide transcriptomic and chromatin immunoprecipitation microarray studies conducted on HSF1. In two such reports, ERCC1 was listed among the genes that are transcriptionally upregulated by HSF1 (Kovács et al., 2019; Page et al., 2006). These microarray-based findings provide a possible explanation for the reduction in ERCC1 expression induced by KNK437, and further support the notion that ERCC1 might be a bona fide HSF1 target gene. Future studies employing chromatin immunoprecipitation to assess HSF1 binding at the ERCC1 promoter, combined with functional luciferase reporter assays, are warranted to provide direct evidence of such transcriptional regulation.

Carboplatin acts by binding to DNA and inducing the formation of adducts and crosslinks, which disrupt replication and transcription, thereby activating signaling pathways that ultimately lead to apoptosis or necrosis (Brabec & Kasparkova, 2005; de Sousa, 2014). In our study, morphological aberrations such as membrane disruption, disintegration, and swelling observed in cells treated with 1 mM carboplatin were indicative of necrosis (white arrows, Figure 3A). This finding was also supported by our flow cytometry-based apoptosis/necrosis analysis, which revealed necrosis as the predominant mode of death in MDA-MB-231 cells exposed to only carboplatin (Figure 3B). In conjunction, these results indicate that carboplatin primarily induces necrosis rather than apoptosis in MDA-MB-231 cells. Such an association between carboplatin and necrosis is also supported by the literature, as evidenced by reports of a scleroderma patient with lung cancer who developed digital necrosis following chemotherapy (Clowse & Wigley, 2003) and of a lymphocytic leukemia patient who died due to acute liver necrosis following carboplatin therapy (Hruban et al., 1991).

When cells were treated with a combination of carboplatin and KNK437, marked changes in cell morphology were observed compared to those treated with only carboplatin, such as rounded, shrunken cells with a largely preserved membrane integrity, which are characteristic features of apoptosis (black arrows, Figure 3A). These results were supported by our further flow cytometry-based apoptosis/necrosis analysis, where a redistribution of the cell population from the lower right to the upper right quadrants indicated apoptosis as the predominant mode of cell death in response to combined carboplatin and KNK437 exposure (Figure 3B). Notably, the cell morphologies observed were fully consistent with the well-established characteristics for apoptosis and necrosis reported in the literature (Balvan et al., 2015). In summary, these results indicate that in MDA-MB-231 cells, necrosis induced by carboplatin alone is replaced by apoptosis when combined with KNK437. The pro-apoptotic influence of KNK437 is corroborated by a previous study where the inhibition of HSR by KNK437 significantly enhanced the apoptosis induced by the proteasome inhibitor bortezomib in multiple myeloma cells and mouse embryonic fibroblasts, as evidenced by increased Annexin V staining and DNA fragmentation (Voorhees et al., 2007). Altogether, our study demonstrates that KNK437 co-treatment promotes apoptosis at low (250 µM) and high (500 µM) doses, and enhances  carboplatin-induced cytotoxicity by 34.1% at the higher dose (Figure 3C). These findings underline the promising potential of KNK437 to improve the efficacy of carboplatin chemotherapy against TNBC.

Study Limitations

There were several limitations to our study. First, the drug concentrations used to assess ERCC1 gene expression could not be directly employed in the Annexin V cell death assays, as these doses were too low to produce any differential cytotoxic response in this assay. However, as the higher KNK437 doses (250 and 500 µM) applied could have interfered with ERCC1 gene expression due to their high DMSO content, the effects observed in cell death might result from other biological actions of KNK437. Second, only a single TNBC cell line was used throughout the study, which may limit the generalizability of our findings to other TNBC models. Third, ERCC1 expression was assessed only at the mRNA level; therefore, potential post-transcriptional regulations or functional effects at the protein level could not be evaluated. Future studies involving additional TNBC cell lines and functional analyses of ERCC1 protein would help provide a more comprehensive understanding of the underlying mechanisms.

Conclusion

In conclusion, our findings reveal that the chemical inhibition of the HSR by KNK437 downregulated ERCC1 expression and increased carboplatin sensitivity in TNBC cells. Interestingly, necrosis triggered by carboplatin was replaced by apoptosis when it was administered in combination with KNK437. Excessive inflammation evoked by chemotherapy-induced necrosis at the tumor microenvironment is well known to contribute to adverse clinical outcomes such as therapy resistance, tumor progression, angiogenesis, and metastasis (Akkız et al., 2025; Vyas et al., 2014). In such cases, as suggested by the present study, redirecting tumor cells from necrosis to apoptosis and thereby evading inflammation through KNK437 co-administration may offer significant therapeutic advantages for the treatment of TNBC—an aspect that requires further investigation by more comprehensive studies in the future.