Dose- and Time-Dependent Halofuginone Cytotoxicity in HaCaT Keratinocytes: Implications for Cholesteatoma

Abstract

Background and Objectives

Cholesteatomas are chronically progressive keratinizing epithelial lesions that locally destroy tissue and erode bone. Surgery is the primary treatment for cholesteatomas; however, tumor recurrence and complications have prompted the search for more effective medical therapies. Halofuginone (HF) inhibits multiple pathways implicated in cholesteatoma pathogenesis. Therefore, the dose- and time-dependent cytotoxic effects of HF on the spontaneously transformed human adult skin keratinocyte (HaCaT) cell line were investigated, and the IC₅₀ values of HF were determined as a preliminary step in evaluating the potential of using HF as a pharmacological treatment for cholesteatoma.

Materials and Methods

An in vitro experiment was conducted using HaCaT cells, which is an immortalized human keratinocyte cell line. Cells were treated with HF at 0.1 μM to 100 μM for 24 and 48 hours. Cell viability was measured via an MTT assay, and the IC₅₀ values were calculated using nonlinear regression analysis.

Results

HF treatment substantially dose- and time-dependently reduced HaCaT cell viability. The IC₅₀ values at 24 and 48 hours were 2.74 μM and 0.24 μM, respectively, reflecting a more than ten-fold increase in cytotoxic potency over time. The differences between the treatment groups were significant at both time points (p=0.017 and p=0.001 at 24 and 48 hours, respectively).

Conclusions

HF exerts potent cytotoxic effects on keratinocytes, with efficacy increasing with exposure duration. These findings support the further study of HF as a pharmacological agent for modulating epithelial proliferation in cholesteatomas and related disorders.

Introduction

Cholesteatoma is a chronic, non-neoplastic lesion characterized by aberrant proliferation and migration of keratinizing epithelium, frequently associated with local inflammation and bone destruction. Although its pathogenesis remains incompletely understood, growing evidence suggests that several signaling pathways, including transforming growth factor-beta (TGF-β)/Smad signaling pathway [1,2], phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT) pathway [3,4], and interleukin-6 (IL-6)/Janus kinase (JAK)/signal transducer and activator of transcription 3 (STAT3) pathway [5,6], are dysregulated in cholesteatoma tissue. Halofuginone (HF), a febrifugine-derived alkaloid, has been shown to inhibit these pathways across various fibrotic and neoplastic models [7–11]. Given this overlap in molecular targets, HF emerges as a promising candidate for modulating cholesteatoma-related epithelial dysregulation. However, its effects on keratinocyte viability have not been extensively characterized in this context.

HaCaT cells, an immortalized human keratinocyte line, are widely used as an in vitro model to study epidermal proliferation, differentiation, and inflammation. Importantly, HaCaT cells share phenotypic and signaling characteristics with the keratinizing squamous epithelium seen in cholesteatoma, including responsiveness to TGF-β, PI3K/Akt, and STAT3 pathways [6,12,13]. These similarities make HaCaT cells a suitable model for evaluating the cytotoxic and modulatory effects of HF on keratinocyte biology relevant to cholesteatoma.

In this study, we assessed the cytotoxic effects of HF on HaCaT keratinocytes by examining cell viability across a range of concentrations at two time points (24 and 48 hours). Particular emphasis was placed on determining IC₅₀ values to quantify its potency over time. Given that HaCaT cells share key molecular features with cholesteatoma epithelium and are responsive to multiple dysregulated pathways, this model was employed to evaluate HF’s potential as a multi-pathway modulator in the context of cholesteatoma.

Materials and Methods

Ethical approval for this study was obtained from the Hacettepe University Non-Interventional Clinical Research Ethics Committee with the approval number GO 21/821 (Date: June 29, 2021; Decision No: 2021/13-36).

Study design

Cell culture and treatment with HF

HaCaT cells were cultured under standard conditions from cells that were previously passaged and stored in our laboratory. Cells were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM; Gibco), supplemented with 10% fetal bovine serum (FBS; Gibco), 1% penicillin-streptomycin (100 U/mL and 100 μg/mL, respectively), and incubated at 37°C in a humidified atmosphere containing 5% CO₂. Cells were passaged at 70%–80% confluence using trypsin-EDTA (0.25%) and used for experiments between passages 5–15 to ensure consistency.

To assess the cytotoxic and apoptotic effects of HF, cells were seeded in appropriate culture plates (96-well for MTT assay, 6-well for fluorescence staining) and allowed to adhere overnight. The next day, cells were treated with various concentrations of HF (e.g., 0.1, 0.3, 1, 3, 10, 20, and 100 μM), diluted in culture medium. Control groups received vehicle treatment (e.g., dimethyl sulfoxide [DMSO] at concentrations not exceeding 0.1%). The cells were incubated with the treatment solutions for 24 and 48 hours.

Cell viability assay (MTT) and apoptotic cell staining

Cell viability following HF treatment was assessed using the MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) assay. After 24 and 48 hours of treatment, 20 μL of MTT solution (5 mg/mL in PBS) was added to each well of the 96-well plate, and the cells were incubated for an additional 2–4 hours at 37°C. The resulting formazan crystals were dissolved by adding 100 μL of DMSO to each well. Absorbance was measured at 570 nm using a microplate reader (BioTek or equivalent). The cell viability percentage was calculated relative to the control group, and IC₅₀ values were determined using nonlinear regression analysis in GraphPad Prism software.

For apoptosis detection, dual fluorescent staining with acridine orange (AO) and propidium iodide (PI) was performed. HaCaT cells were seeded into 6-well plates and treated with HF as described above. At the end of the 24 and 48-hour incubation periods, cells were gently washed with PBS and stained with a mixture of AO (10 μg/mL) and PI (10 μg/mL) for 2 minutes at room temperature in the dark. The stained cells were observed immediately under an inverted microscope (Olympus IX70) equipped with appropriate filters (Fig. 1).

Fig. 1.

Inverted microscopic images of HaCaT cells stained with acridine orange and propidium iodide at 24 hours (A) and 48 hours (B).

Statistical analysis

Statistical analyses were performed using SPSS (version 26.0; IBM Corp.). Descriptive statistics were presented as frequencies and percentages for categorical variables, and as mean±standard deviation (SD) or median (minimum–maximum) for continuous variables, as appropriate. Since the data did not meet the assumptions of normality, non-parametric statistical methods were applied. Comparisons of cell viability between multiple groups were performed using the Kruskal–Wallis H test. When significant overall differences were identified, pairwise comparisons were conducted using the Mann–Whitney U test. To control for multiple testing, Holm–Bonferroni correction was applied. Sequential pairwise comparisons were predefined between adjacent concentrations (control–0.1 μM, 0.1–0.3 μM, 0.3–1 μM, 1–3 μM, 3–10 μM, 10–20 μM, 20–100 μM). A p-value <0.05 after Holm–Bonferroni correction was considered statistically significant. In addition, IC₅₀ values (half-maximal inhibitory concentration) were calculated for both 24-hour and 48-hour treatment groups using nonlinear regression analysis. A four-parameter logistic (4PL) model was fitted to dose-response data using the curve fitting function in Python (version 3.11) with the SciPy library (version 1.11), and IC₅₀ was determined as the concentration corresponding to 50% cell viability. Graphical visualizations of the fitted dose-response curves were also generated using log-transformed concentration scales.

Results

To assess the cytotoxic effects of HF on HaCaT cells, cell viability was evaluated at 24- and 48-hours following treatment with increasing concentrations of the compound (ranging from 0.1 μM to 100 μM). The Kruskal–Wallis H test revealed statistically significant differences in cell viability between groups at both 24 hours (p=0.017) and 48 hours (p=0.001) (Table 1). The box plots show the decrease in absorbance and cell viability with increasing HF concentration (Fig. 2).

Table 1.

HaCaT cell viability at 24 and 48 hours

	Cell viability (%)
	24 h			48 h
	Mean±SD	Median (min–max)	p-value	Mean±SD	Median (min–max)	p-value
Control	100.0±52.36	84.9 (56.7–183)	0.017*	100.0±21.26	106.65 (69.5–117.2)	0.001*
0.1 μM	77.23±38.53	72.95 (a30.4–125.9)		73.9±12.74	75.35 (57–87.9)
0.3 μM	93.8±38.18	113.9 (36.5–122.6)		42.26±1.86	42.3 (40.4–44.9)
1 μM	108.23±20.31	116.95 (75–126.1)		30.15±2.98	29.75 (27–35.6)
3 μM	69.53±18.56	68.85 (37.9–94.1)		16.58±1.88	17.3 (14.1–18.4)
10 μM	55.35±6.01	52.15 (50.4–63.4)		8.48±3.16	7.1 (5.5–13.1)
20 μM	54.28±11.45	52.15 (40–74.1)		6.42±2.66	5.5 (4.8–11.8)
100 μM	52.42±12.58	53.05 (31.4–68.8)		5.17±1.16	4.65 (4–7)

Kruskal–Wallis test.

^*p<0.05, statistically significant.

SD, standard deviation; min, minimum; max, maximum.

Fig. 2.

Dose- and time-dependent effects of halofuginone on HaCaT cell viability. A: 24-h results. B: 48-h results. Data are presented as boxplots (median, interquartile range, whiskers, outliers). Cytotoxicity was more pronounced at 48 hours.

At 24 hours, a dose-dependent decrease in cell viability was observed (Fig. 2A). While no statistically significant differences were found between the control and concentrations up to 100 μM individually, a noticeable decline in cell viability occurred between 1 μM and 3 μM, indicating the onset of cytotoxicity. At 24 hours, no statistically significant differences were detected between sequential concentrations after Holm–Bonferroni correction. Although raw p-values suggested a modest decrease in cell viability at higher doses, none of these comparisons remained significant, indicating that cytotoxicity did not reach a robust level within the first 24 hours (Fig. 3A). According to nonlinear regression, the half-maximal inhibitory concentration (IC₅₀) for 24-hour exposure was calculated to be approximately 2.74 μM (Fig. 4A).

Fig. 3.

Sequential pairwise comparisons of halofuginone-induced cytotoxicity in HaCaT cells at 24 hours (A) and 48 hours (B). Bar graphs represent mean±standard deviation of cell viability (%) at each concentration. Pairwise comparisons were performed between sequential concentrations (control–0.1 μM, 0.1–0.3 μM, 0.3–1 μM, 1–3 μM, 3–10 μM, 10–20 μM, 20–100 μM) using the Mann–Whitney U test with Holm–Bonferroni correction. *Significant differences (p<0.05 after correction).

Fig. 4.

Dose-response curve of halofuginone on HaCaT cells at 24-h (A) and 48-h (B) time points.

The cytotoxic effect of HF became markedly more pronounced at 48 hours (Fig. 2B). Statistically significant differences in viability were observed between the control group and all concentrations ≥0.3 μM (p<0.05). At 48 hours, overall group differences were significant, and sequential pairwise analyses revealed clear dose-dependent effects. Holm–Bonferroni–adjusted pairwise comparisons confirmed significant decreases in cell viability between 0.3 μM and 1 μM (p=0.022), 1 μM and 3 μM (p=0.015), and 3 μM and 10 μM (p=0.015). Other sequential comparisons (Control–0.1 μM, 0.1–0.3 μM, 10–20 μM, 20–100 μM) did not reach statistical significance. These findings demonstrate that the cytotoxic effect of HF becomes more pronounced after 48 hours, with a clear stepwise reduction in viability across increasing concentrations (Fig. 3B). Nonlinear regression yielded an IC₅₀ value of approximately 0.24 μM (Fig. 4B), further highlighting the time-dependent increase in HF potency. Compared to 24-hour results, this represents a more than ten-fold decrease in IC₅₀, confirming enhanced cytotoxicity with prolonged exposure.

Discussion

Agents investigated in the treatment of cholesteatoma

There are many studies in the literature on the pathogenesis of cholesteatoma and potential target pathways [14,15]. A few treatments have been investigated, mostly aimed at hyperproliferative epithelium via fluorouracil [16–18], viral oncolytic therapy [19–21], and photodynamic therapy [22]; however, none are widely utilized in clinical practice.

Topical 5-fluorouracil (5-FU) has demonstrated clinical efficacy in both animal models and patient cohorts, particularly in attic and external auditory canal cholesteatomas, by reducing epithelial proliferation and keratin debris accumulation [16–18]. More recently, engineered oncolytic herpes simplex viruses have emerged as a promising modality, selectively eradicating cholesteatoma-derived keratinocytes both in vitro and in vivo, with up to 77% volume reduction observed in gerbil models [19–21]. Another novel approach involves immunotargeted photodynamic therapy using anti-EGFR (epidermal growth factor receptor)–coated nanocapsules, which achieved over 70% selective keratinocyte death while sparing middle ear mucosa in vitro [22]. Despite these advances, many of these agents require relatively high concentrations or complex delivery systems.

In their recent study investigating the effects of immunosuppressive agents on human epidermal keratinocytes isolated from adult skin and cholesteatoma-derived keratinocyte cultures, Uzun, et al. [23] demonstrated the cytotoxic potential of immunosuppressive agents. Compared to these repurposed agents, HF showed equal or superior cytotoxic efficacy at significantly lower concentrations. Remarkably, the IC₅₀ value of HF at 48 hours (0.24 μM) was substantially lower than the effective doses required for imiquimod (50 μg/mL) and tacrolimus (800 ng/mL), underscoring its high potency.

Our results reveal that HF exerts a potent, dose- and time-dependent cytotoxic effect on HaCaT keratinocytes, as evidenced by a striking reduction in IC₅₀ values from 2.74 μM at 24 hours to just 0.24 μM at 48 hours. This pronounced enhancement in efficacy over time indicates that HF may retain substantial biological activity even at low concentrations, particularly with extended exposure. By suppressing keratinocyte proliferation and viability, HF emerges as a promising candidate for pharmacological intervention in cholesteatoma. Nevertheless, further studies using cholesteatoma-derived keratinocytes are essential to clarify their therapeutic index and ensure safe clinical applicability.

Molecular targets and mechanisms of action of HF

In our study, the hypothesis of the inhibitory effect of HF on keratinocytes was based on the experiments carried out with HF on the pathways previously determined to be involved in the pathogenesis of cholesteatoma. In this section, we review the pathways involved in the pathogenesis of cholesteatoma, the studies carried out on HaCaT, and the studies in which the inhibitory effect of HF on these pathways was determined.

The TGF-β/Smad pathway has been extensively implicated in cholesteatoma pathophysiology. Increased expression of TGF-β1 and phosphorylated Smad2, coupled with reduced levels of Smad7, has been observed in cholesteatoma tissue, promoting extracellular matrix accumulation and fibrotic remodeling, similar to chronic wound healing responses [1]. TGF-β1 also suppresses hepatocyte growth factor in cholesteatoma-derived epithelium, reinforcing its profibrotic role [2]. HaCaT cells respond robustly to TGF-β1, with activation of ALK5-mediated signaling leading to migration and epithelial-mesenchymal transition [12]. Previous reports demonstrate that HF inhibits TGF-β1–induced Smad2/3 phosphorylation while upregulating Smad7, suggesting that HF may suppress proliferation and matrix deposition via blockade of TGF-β signaling [7,8].

The PI3K/AKT pathway is another critical mediator of cell survival and proliferation, often upregulated in cholesteatoma. Overexpression of miR-508-3p in cholesteatoma tissues leads to suppression of PTEN, resulting in PI3K/AKT activation and enhanced keratinocyte survival [3]. Restoration of PTEN expression via miR-34a-loaded nanoparticles reverses these effects and induces apoptosis in cholesteatoma cells [4]. In HaCaT keratinocytes, inflammatory stimuli such as TNF-α activate the PI3K/AKT/mTOR cascade, promoting cell proliferation and suppressing autophagy [13]. HF has been shown to inhibit PI3K/AKT signaling in multiple cancer models, including cisplatin-resistant lung cancer organoids, by suppressing gene expression involved in cell cycle and survival [9].

The IL-6/JAK/STAT3 axis is highly relevant in cholesteatoma, where IL-6 and phosphorylated STAT3 are significantly upregulated, promoting epithelial hyperproliferation [5]. Downregulation of miR-125b in cholesteatoma leads to overactivation of STAT3 and upregulation of its downstream targets, such as cyclin D1, survivin, and VEGF, facilitating cell cycle progression and resistance to apoptosis [6]. In HaCaT cells, STAT3 similarly governs proliferative behaviour; silencing STAT3 can reverse the pro-proliferative effects of miR-125b inhibition, confirming its central role [6]. HF has been shown to suppress STAT3 activation in Th17 cells by triggering amino acid starvation responses that lead to selective downregulation of STAT3 protein levels [10].

Limitations

However, several limitations must be acknowledged. First, our study utilized a single immortalized keratinocyte line (HaCaT), which, despite its widespread use as a keratinocyte model, does not fully reflect the heterogeneity and microenvironment of cholesteatoma tissue. Therefore, future studies incorporating primary cholesteatoma-derived keratinocytes are essential to validate the translational applicability of our results. Additionally, our assessment was limited to cell viability; mechanistic insights into apoptosis, proliferation, and signaling pathway activation were not directly examined. Therefore, while the observed cytotoxic effects are encouraging, further molecular analyses are necessary to elucidate the precise mechanisms of action. Another limitation of our study is that comparative potency analyses with other immunosuppressive agents could not be directly performed under the same experimental conditions. To our knowledge, only one previous study [23] has reported effective inhibitory concentrations of agents such as 5-FU, tacrolimus, imiquimod, and cyclosporine, but IC₅₀ values were not determined. Therefore, while we compared the IC₅₀ value of HF with the reported effective concentrations of these agents, this comparison may not be fully accurate. Future studies should perform direct IC₅₀ analyses under identical conditions to allow a more reliable assessment of comparative potency. Finally, neither HF nor other immunosuppressive agents have been evaluated for their local ototoxicity and safety profile in experimental models; such studies, particularly in animal cochlea, are necessary before reliable conclusions about safe dosage ranges can be drawn.

Future research directions

Future studies should aim to validate these findings using primary cholesteatoma-derived epithelial cells, which would provide more physiologically relevant data. Moreover, incorporating assays to evaluate apoptotic markers, cell cycle arrest, and pathway-specific protein expression would help clarify the biological impact of HF at the molecular level. In vivo investigations using animal models of cholesteatoma, or ex vivo cultures of human cholesteatoma tissue, could also offer valuable translational insights. Future research should also evaluate the local toxicity of HF on the middle ear mucosa in order to establish its safety profile. In vivo animal studies are particularly needed to assess both the therapeutic efficacy and potential adverse effects of HF before progressing to clinical applications.

Conclusion

The present study provides preliminary evidence that HF may serve as a promising molecular modulator for cholesteatoma through its multi-targeted effects on keratinocyte viability. Further research is warranted to explore its full therapeutic potential and to determine its safety and efficacy in clinically relevant systems.