Pharmacological Modulation of Autophagy Can Sensitize Acute Lymphoblastic Leukemia Cell Lines to Dexamethasone

Simple Summary

Dexamethasone (DEX) is a potent synthetic glucocorticoid (GC) widely used in treatment of acute lymphoblastic leukemia (ALL). Autophagy, a cellular recycling process, may play a dual role in GC resistance: in resistant cells, it often acts as a survival mechanism, whereas in sensitive cells, GC can induce autophagy before cell death. The present study examined how pharmacological modulation of autophagy, both its induction and inhibition, affects GC sensitivity in five ALL cell lines. Induction of autophagy with tamoxifen (TAM) was shown to successfully enhance GC sensitivity in most cell lines. These findings suggest that each ALL cell line may have an optimal basal level of autophagy, and that its dysregulation can effectively increase GC sensitivity.

Abstract

Background: The potent synthetic glucocorticoid (GC), dexamethasone (DEX), is a highly effective component of conventional chemotherapy for acute lymphoblastic leukemia (ALL). However, cases of GC resistance require elucidation of the underlying mechanisms and the development of new strategies to overcome them. GC-induced autophagy can play a dual role in GC resistance: it often acts as a salvage mechanism in resistant cells, while in sensitive cells, it is a mechanism leading to cell death. Methods: In the present study, cell death and autophagy, as well as their dependence on glucocorticoid receptors (GRs), were simultaneously monitored in DEX-treated ALL cell lines, both sensitive and resistant to GCs. Results: In GC-resistant cell lines, no changes in autophagy levels were observed after DEX treatment, whereas in GC-sensitive cell lines, autophagy elevation was associated with cell death. Blockade of GC receptors completely abolished DEX cytotoxicity in CCRF–CEM cells but not in RS4;11 cells, suggesting the participation of distinct, cell line-specific mechanisms. Furthermore, we investigated how pharmacological modulation of autophagy, both induction and inhibition, affects GC sensitivity. Autophagy induction with tamoxifen (TAM) successfully sensitized most cell lines to DEX. In CCRF–CEM cells, the sensitization effect was shown to correlate with increased apoptosis. In other cell lines, no increase in cell death was observed, suggesting decreased cell proliferation. Conclusions: These results suggest that each ALL cell line may have an optimal basal level of autophagy, and targeted dysregulation of this level may be an effective strategy for enhancing GC sensitivity.

1. Introduction

Acute lymphoblastic leukemia (ALL) is the most common type of cancer in pediatric patients [1]. Although the 5-year survival rate has increased considerably in high-income countries and conventional chemotherapy has proven to be well-tolerated in a large number of patients, survival remains lower in patients from low- and middle-income countries [1,2].

Glucocorticoids (GCs) are a key component of conventional chemotherapy used in all stages of treatment [1,3,4,5]. Among them, dexamethasone (DEX) is considered more effective, as patients receiving DEX show improved event-free survival [1,4]. However, most of the cytotoxicity and side effects are caused by the prolonged use of GCs, occurring more frequently in patients treated with DEX compared to prednisone [3]. Therefore, strategies have been proposed to reduce the dose and limit these complications. These include alternating its use weekly [5] and using co-treatments with clinically available drugs to reduce the required dose [6].

In patients with ALL, an initial poor response to GCs is often associated with an unfavorable prognosis, highlighting the importance of understanding the mechanisms that may mediate sensitivity or resistance to GCs (reviewed in [6]). In addition to GC resistance, caused by reduced expression or loss of GRs, several other mechanisms were discovered, driven by alterations in multiple signaling pathways and causing metabolic reprogramming [6]. One such mechanism is autophagy, a highly conserved recycling process that degrades a cell’s own components, as damaged organelles or protein aggregates. These components are sequestered in autophagosomes, and their contents are degraded through the lysosomal machinery. A constitutive form of autophagy, basal autophagy, occurs under normal physiological conditions and serves to maintain cellular homeostasis. In contrast, induced autophagy is a mechanism that is upregulated by cellular stress, like starvation and drug toxicity. It allows cells to survive by breaking down non-essential and damaged components and providing nutrients and structural elements for continued growth. The level of both the basal and induced autophagy was reported to be increased in many cancers [7,8].

A current debate exists regarding whether autophagy plays a pro-death or pro-survival role in ALL [9,10,11]. Given this duality, two different approaches can be proposed to overcome GC resistance in ALL. The first approach is to sensitize malignant cells by blocking autophagy as their primary survival mechanism. Most studies suggest that inhibiting autophagy is a promising strategy to eliminate leukemic blasts or counteract their resistance to various chemotherapeutic drugs [12,13,14]. Numerous clinical trials are evaluating combinations of potent autophagy inhibitors such as chloroquine (CQ) and hydroxychloroquine (HCQ) with various therapies to improve outcomes [15,16]. Nevertheless, it has also been observed that inhibiting autophagy in cancer increases the Warburg effect, an essential mechanism supporting cancer growth, which is characterized by an increased glucose consumption and lactate production despite the availability of oxygen [17]. Recently, an alternative approach has been proposed which involves inducing excessive autophagy, aiming to force death-resistant cancer cells to undergo autophagic cell death [18].

We previously hypothesized that an “optimal” level of autophagy is required for resistance to drug-induced cytotoxicity, and that both inhibition and stimulation of autophagy may help overcome drug resistance [19]. To test this hypothesis, in the present study we examined how autophagy modulators, both inhibitors and stimulants, affected DEX sensitivity in ALL cells. Our findings indicate that the response varies across cell lines, suggesting the existence of an optimal level of autophagy for each.

2. Materials and Methods

2.1. Cell Lines and Culture Conditions

All human cell lines used in this study were from the American Type Culture Collection (ATCC; Manassas, VA, USA). T-ALL: Jurkat (Clone E6-1, male, 14-year-old, ATCC^®TIB™), CCRF-CEM (female, 4-year-old, ATCC^®CCL-119™), MOLT-3 (male, 19-year-old, ATCC^®CRL-1552™); B-ALL: REH (female, 15-year-old ATCC^®CRL-8286™) and RS4;11 (female, 3-year-old, ATCC^®CRL-1873™). Chronic myeloid leukemia: K562 (female, 53-year-old, ATCC^®CCL-243™). Cells were maintained in suspension using Advanced RPMI 1640 medium containing 5% heat-inactivated fetal bovine serum (FBS), 10 mM HEPES, 1% GlutaMAX™, 100 U/mL penicillin, and 100 μg/mL streptomycin (Gibco, Thermo Fisher Scientific, Waltham, MA, USA). Cultures were kept in a humidified incubator at 37 °C with 5% CO₂. All experiments were conducted using cells in the logarithmic growth phase within 20 passages.

2.2. Reagents

All reagents were purchased from Sigma–Aldrich (San Luis, MO, USA): DEX (Cat. #D9184), chloroquine diphosphate salt (CQ, Cat. #C6628), rapamycin (RAP, Cat. #R0395), spautin-1 (SP-1, Cat. #SML0440), and TAM (Cat. #T5648). For stock solutions, DEX and SP-1 were dissolved in DMSO at 20 mM and 50 mM, respectively; RAP was dissolved in chloroform at 5 mM and TAM at 20 mM in ethanol. Solvent concentrations were maintained below 0.5% (v/v) at the maximum tested drug dosages to ensure no interference with cell viability.

2.3. Cell Viability Assay

To evaluate cell viability, a trypan blue exclusion test was performed. Cells were seeded at a density of 2.5 × 10⁵ cells/mL in 48-well plates and incubated for 24–72 h. After the incubation period, each sample was resuspended, and a 10 μL aliquot was taken for cell counting, using a Neubauer chamber and trypan blue dye solution (Sigma–Aldrich, Cat. #15250061). Results were normalized to the untreated control and expressed as the average of three or more independent experiments.

2.4. Autophagy and Cell Death by Flow Cytometry

Cells were cultured in the same conditions as cell viability assays. After the treatment and incubation period, 2.5 × 10⁵ cells were stained to assess autophagy, apoptosis, and necrosis, as mentioned below. To detect changes in autophagy induction, cell lines were stained with monodansylcadaverine 60 μM (MDC, Ex/Em max = 335/525 nm; Sigma-Aldrich, 30432). MDC is a lysosomotropic autofluorescent dye that accumulates in autophagolysosomes. Its fluorescence and retention increases under conditions that increase autophagy. For apoptosis detection, Annexin V conjugated to Alexa Fluor 488 (Annexin V-AF488, Ex/Em max = 495/519 nm; Thermo Fisher Scientific, A13201) was used; Annexin V binds specifically to externalized phosphatidylserine (PS) after apoptotic induction by inactivation of flippases activity. For necrosis assay, the nucleophilic marker propidium iodide (PI, Ex/Em max = 538/617 nm; Sigma–Aldrich, P4864) was used; PI was employed to identify cells with compromised plasma membranes. Cells were co-stained with MDC, Annexin V-AF488, and PI in Annexin V-binding buffer (Thermo Fisher Scientific, Cat. #V13246) for 30 min at room temperature in the dark. Data were acquired using a FACSCanto II flow cytometer (BD Biosciences, San Jose, CA, USA), where 10,000 events were filtered through a single-cell gate (debris and doublets excluded). Results were processed using FlowJo 10.5.3 software. The compensation procedure (MDC, Annexin V-AF488, and PI) was performed before data acquisition.

To assess basal autophagy, cells were stained with MDC (60 μM, 30 min, room temperature) and immediately analyzed by flow cytometry.

2.5. ROS Production

Intracellular reactive oxygen species (ROS) production was quantified using the non-fluorescent dye 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA; Sigma–Aldrich, Cat. #D6883). This dye is oxidized by intracellular ROS, resulting in the formation of the highly fluorescent compound DCF (Ex/Em max ≅ 492–495/517–527 nm). Briefly, cells treated with DEX 1 μM for 24 h, along with untreated controls, were collected and washed with PBS. Staining was performed using 20 μM DCFH-DA for 30 min. After the incubation period, the cells were rinsed again to remove any unincorporated dye and then resuspended in PBS. Fluorescence measurements of DCF were conducted using a GloMax plate reader (Promega, Madison, WI, USA) equipped with a 475 nm excitation and a 500–550 nm emission filters. Data from at least three independent experiments were averaged and normalized to the untreated control

2.6. Data Analysis and Statistics

Statistical analyses were performed using GraphPad Prism software (v.8.3). Data are presented as mean ± standard error (SEM) from at least three independent biological replicates (n ≥ 3) in an independent form. For multiple comparisons, one-way or two-way analysis of variance (ANOVA) was employed as appropriate. Specifically, Dunnett’s post hoc test was used to compare DEX-treated groups against the control; Tukey’s test was applied for comparisons between multiple groups treated with DEX and other pharmacological agents; and Sidak’s test was used for comparisons between two specific groups.

3. Results

3.1. DEX Sensitivity Varies Among All Cell Lines

In the present work, ALL cell lines with previously reported sensitivity to DEX [11,12,13,14] were used: DEX-resistant Jurkat, MOLT-3 (both T-ALL) and REH (B-ALL), DEX-sensitive RS4;11 (B-ALL), and CCRF–CEM (T-ALL) partially sensitive to DEX. Among resistant cell lines, REH is GRs-negative, whereas Jurkat and MOLT-3 show low basal levels of GRs expression [12,20,21].

Initially, sensitivity/resistance to DEX was confirmed in these cell lines. For this purpose, cells were treated with DEX for 24–72 h. Every 24 h, live cells were counted (Figure 1A) and cell death (apoptosis/necrosis) was evaluated (Figure 1B–D). The resistant cell lines maintained their viability over 75% at all concentrations and time points tested (Figure 1A,C and Figure S1). Conversely, in the DEX-sensitive CCRF–CEM and RS4;11 cell lines, cell death increased over time (Figure 1A,D and Figure S1).

Since increased levels of basal autophagy have been proposed as a potential contributor to drug resistance, we assessed this parameter in the studied lines (Figure 1E). Unexpectedly, we did not find a correlation between DEX resistance and basal autophagy levels, as resistant phenotypes displayed heterogeneous autophagic profiles. Although basal autophagy in the resistant cell line MOLT-3 was indeed higher than in other cell lines, this parameter in the other resistant line REH was significantly lower, and in the resistant line, Jurkat was practically the same as in the sensitive CCRF–CEM and RS4;11.

Our previous work demonstrated that excessive ROS production in DEX-sensitive T-ALL cell line CCRF–CEM is GRs-dependent, and provokes autophagy and cell death [19]. We then addressed the question of whether the level of ROS production in response to DEX treatment is different in DEX-sensitive and DEX-resistant cells. To assess the effects of DEX on ROS production, we treated both T-ALL and B-ALL cell lines with DEX (1 µM, 24 h). We observed no relationship between ROS levels and sensitivity to DEX. A significant increase in ROS after DEX treatment was observed in both resistant Jurkat and sensitive CCRF–CEM T-ALL cell lines. In contrast, ROS production in resistant MOLT-3 and REH, and in the DEX-sensitive B-ALL cell line RS4;11 was not changed after DEX treatment (Figure 1F).

Thus, the mechanisms of death of sensitive cells and the mechanisms of resistance differ in different cell lines, which apparently reflects the clinical situation and is a consequence of the high plasticity of leukemic cells.

3.2. GRs Blockade Does Not Completely Abolish DEX Cytotoxicity in RS4;11

While GRs activation is essential for the overall response to DEX in many scenarios, GCs and GRs seem to be involved in multifactorial cellular processes that can operate independently or involve crosstalk with other signaling pathways. Therefore, the aim of our subsequent experiments was to reveal whether binding to plasma GRs is a critical factor in mediating changes in cell viability in DEX-sensitive models. In these experiments, cells were pre-incubated with competitive GRs inhibitor Ru486 (1 µM, 20 min) before DEX was added. Of note, binding studies have shown that Ru486 and DEX share the same binding site [22], although Ru486 has greater [22,23] or equal [24] affinity for GRs, depending on the model evaluated.

Initially, DEX toxicity was evaluated at different times of treatment by live cell count (Figure 2A). We found that Ru486 completely protected CCRF–CEM cells from DEX cytotoxicity. However, in RS4;11 cells, the antagonistic activity of Ru486, although significant, was not complete, indicating involvement of other mechanisms.

Given that autophagy may represent either a rescue or death-driven mechanism, cell death and autophagy, and their dependence on GRs, were monitored simultaneously in DEX-treated cells (Figure 2B–E). For this, after incubation, DEX cells were triple-stained with MDC, PI, and Annexin V. To reveal if the autophagy level was increased in the surviving population in sensitive cell lines, it was assessed in live cells population (PI-Annexin V-, see experimental design at Figure 2B). We observed different scenarios in the two sensitive cell lines, which are described below.

In CCRF–CEM cells, partially sensitive to DEX, no significant changes in viability and autophagy level were observed at 24 h of treatment with DEX. At 72 h, a significant part of cells died (Figure 2D, first panel), and the autophagy level in the population of survived cells increased (Figure 2C, left panel). GR blockade prevented both the increase of autophagy and cell death induced by DEX (Figure 2C,D).

RS4;11 cells demonstrated high sensitivity to DEX: we observed significant decrease in the live cell population already at 24 h of treatment (Figure 2E, first panel), with slightly increased level of autophagy in it (Figure 2C, right panel). Both processes were prevented by GRs blockade. At 72 h, practically all cells in DEX-treated population died (Figure 2E). In a very small population of surviving cells, autophagy levels were significantly reduced compared to control levels, likely due to the poor ability of dying cells to retain the fluorochrome (Figure 2C). Pre-incubation with Ru486 only partially prevented cell death at 72 h (Figure 2E, first panel), but autophagy in surviving population was restored to the control level (Figure 2C, right panel).

Simultaneous monitoring of autophagic and dead populations is shown in the Figure 2D,E, (second and third panels). These experiments revealed that the population of living cells that retains MDC (Annexin V-MDC+) decreased after DEX treatment but restored by Ru486 pre-incubation (completely in CCRF-CEM and partially in RS4;11). The population of apoptotic cells that were still able to retain MDC (MDC+Annexin V+) and were likely in the early apoptotic phase increased after DEX treatment in the RS4;11 line. Cell populations that failed to retain MDC, most likely due to progressive plasma membrane damage, apparently belong to cells in the late stages of apoptosis (Annexin V+MDC-) and necrosis (Annexin V-MDC-). Annexin V+MDC- was increased in both cell lines, and Annexin V-MDC- in RS4;11. The appearance of these damaged populations was prevented by blocking GRs.

Based on the results presented here, it can be concluded that the toxicity of DEX in CCRF–CEM is completely dependent on binding to GRs, whereas in RS4;11 it is only partially dependent. Autophagy induced by DEX treatment in sensitive ALL cell lines is GR dependent and occurs prior to cell death.

3.3. Autophagy Is Manifested Upstream of Cell Death in DEX-Sensitive Cell Lines

In ALL, autophagy has been described as a double-edged sword, as it has the potential to promote both a chemoresistance and suppression of tumor growth [9]. To determine whether autophagy levels are altered in the DEX-resistant populations in different ALL cell lines, the viable population negative for apoptosis and necrosis after DEX treatment was selected for autophagy measurement (Figure 3A and Figure S2A). In resistant cell lines, no changes in the level of autophagy were observed after DEX treatment compared to controls at any of the concentrations and times evaluated, indicating that basal levels of autophagy in these cell lines are sufficient to recycle damaged organelles and macromolecules and counteract DEX toxicity (Figure 3A). In sensitive CCRF–CEM and RS4;11 cells, the population that remains resistant after DEX treatment presents a higher level of autophagy compared to the control (Figure 3A and Figure S2B). However, although autophagy levels in RS4;11 cells increased significantly at 24 and 48 h (Figure 3A), most of the cell population was dead after 72 h of treatment (Figure 1), indicating that increased autophagy represents an early event that did not protect cells against DEX toxicity. At this time point, a substantial fraction of cells exhibiting high autophagy levels had undergone cell death, whereas the remaining viable cells displayed lower autophagy levels, suggesting the presence of subpopulations with distinct sensitivities, as previously documented in other models of ALL.

To determine whether autophagy serves as a mechanism promoting survival or death in the overall cell population, subsequent experiments assessed autophagy in surviving and dying cell populations over time. Figure 3B shows representative flow cytometry dot plot, where viable cells with basal autophagy are Annexin V-MDC+ (Q5), early apoptotic cells are positive for Annexin V and still can retain MDC (Q6), whereas late apoptotic and necrotic cells with a damaged cell membrane could not retain the MDC dye (Q7 and Q8, respectively). As in the previous experiments, DEX neither induce death nor an increase of autophagy level in resistant cells Jurkat and REH (Figure 3C,E and Figure S2C,E). In contrast, in sensitive CCRF–CEM and RS4;11 cells treated with DEX, the population of viable cells (Annexin V-MDC+) was significantly reduced over time, while the population of dead cells was concurrently increasing, in particular those that were double positive (MDC+Annexin V+) likely corresponding to early apoptotic cells (Figure 3D,F and Figure S2D,F).

Collectively, these results indicate that in the CCRF–CEM and RS4;11 cell lines, the initial DEX-resistant population exhibits higher levels of autophagy compared to the sensitive population. Over time, cell death markers appear in this population, indicating that autophagy is significantly upregulated before the commitment to widespread cell death. In contrast, no increase in autophagy was observed in the fully resistant cell lines.

A concentration of 1 µM DEX was selected for subsequent experiments, as supplementary dose-response data indicated that higher concentrations (up to 100 µM) did not further enhance sensitivity, suggesting a saturation of the glucocorticoid-mediated response in these cell lines.

3.4. Autophagy Modulation Affects ALL Cell Lines Differently

To investigate whether modulation of autophagy affects the sensitivity of different leukemia cell lines to DEX, cells were treated with DEX (1 μM) in combination with one of the modulators. Particularly, we used the autophagy inducers RAP and TAM, and the autophagy inhibitors SP-1 and CQ.

RAP is known to induce autophagy by inhibiting the mammalian target of rapamycin (mTOR), a process that mimics cellular starvation [25]. In contrast, SP-1 suppresses the activity of ubiquitin-specific peptidases 10 and 13 (USP10 and USP13). This inhibition leads to the degradation of VSP34 and Beclin-1, two essential proteins for the early stages of autophagy induction [25]. CQ functions as a lysosomotropic agent. It enters the lysosome, where it becomes protonated, raising the lysosomal pH and thereby inhibiting the autophagosome–lysosome fusion. This mechanism makes CQ an autophagy blocker, as it prevents breakdown of autophagy cargo without affecting the initiation of autophagy [25,26]. The precise mechanism by which TAM induces autophagy remains unknown; however, it has been documented as an effective autophagy inducer in breast cancer [27], glioblastoma [28], and leukemias [29].

Autophagy modulation in leukemic cells by the aforementioned drugs was confirmed via MDC staining followed by flow cytometry, and by monitoring autophagic flux in Jurkat cells stably transfected with the LC3-mCherry-GFP construct using confocal microscopy (see methods in [29]). Dose-response experiments were performed for RAP and SP-1 to identify the minimum dose required for effective autophagy modulation (Figure S3A,B,D), whereas previously defined concentrations for TAM and CQ were employed [19,29] (Figure S3C).

Notably, RAP itself effectively inhibited cell growth in all tested lines, with the exception of RS4;11, as evidenced by a decrease in cell numbers compared to control samples. In the CCRF–CEM cell line, partially sensitive to DEX, the combination of DEX + RAP was significantly more effective than either of the two compounds where RAP thus increased sensitivity to DEX (Figure 4A).

Autophagy inhibition with SP-1 affected the viability of all cell lines except MOLT-3. Importantly, we found that SP-1 increases DEX sensitivity in Jurkat and CCRF-CEM cell lines (Figure 4B).

TAM either induced or enhanced DEX sensitivity in all leukemic cell lines, except REH (Figure 4C).

Blocking autophagic flux with CQ does not affect cell viability in any of the models evaluated, and sensitizes CCRF–CEM to DEX (Figure 4D). Although decreased viability in cells treated with combination of DEX and CQ in comparison to DEX treated cells was observed in resistant Jurkat and MOLT-3 at 24 and 48 h of treatment, the viability was restored at 72 h likely due to increased growth of surviving cells.

3.5. The Sensitizing Effect of TAM on DEX-Induced Cytotoxicity Varies Among Distinct ALL Cell Lines

TAM has been found to be a very effective inducer of autophagy in leukemia, as demonstrated in our earlier experiments on Jurkat cells [29] and confirmed in the present study on various T- and B-cell ALL cell lines (Figure 5). Given that TAM also increased DEX sensitivity in multiple cell lines (Figure 4), we sought to determine whether the observed reduction in cell count following treatment with DEX in combination with TAM was due to increased cytotoxicity (cell death) or solely a cytostatic effect (decreased proliferation).

In DEX resistant Jurkat, the combination of TAM and DEX did not enhance autophagy beyond the level observed with TAM alone (Figure 5A). Only a slight decrease in number of living cells was observed after TAM treatment, but the effect of TAM alone and in combined treatment with DEX and TAM was similar (Figure 5A), indicating that the significant decrease in Jurkat cell count after 72 h of combined treatment was largely due to reduced cell proliferation.

In CCRF–CEM cell line, where combination treatment with DEX and TAM efficiently decreased cell numbers at 48 and 72 h, the number of living cells also decreased when compared to DEX treatment. We observed an increase in the double-positive population MDC+Annexin V+, when cells were treated with DEX in combination with TAM, compared to either DEX or TAM alone. This population likely represents cells in the early stage of apoptosis that are still capable of retaining MDC. Notably, in this cell line, autophagy induced by the combination of DEX and TAM was significantly greater than that observed with either DEX or TAM alone (Figure 5B). Both of these results indicate that in this cell line, combination treatment induces an increase in autophagy that is not protective, reduces cell growth, and is associated with cell death.

Remarkably, REH cells were highly resistant not only to DEX but also to the combined treatment of DEX and TAM, although TAM induced a significant increase in autophagy levels in them (Figure 5C). It should be noted, however, that with combination treatment, the autophagy level was the same as with TAM alone.

In RS4;11 cells, which are highly sensitive to DEX and demonstrated significant increase of autophagy, the DEX + TAM combination did not induce a higher level of autophagy compared to either DEX or TAM monotherapy (Figure 5D). Although combination primarily caused an increase in the autophagy/apoptosis double-positive population after 72 h of treatment (Figure 5D,E), it did not increase general cytotoxicity (cell death).

In conclusion, the pronounced reduction in cell number observed in Jurkat and RS4;11 cells following combination therapy did not correlate with an increase in cell death, contrasting with DEX monotherapy; while in CCRF-CEM cell line, the observed decline in cell number is accompanied by a concomitant increase in cell death.

4. Discussion

Glucocorticoids are core chemotherapy drugs in ALL treatment, therefore GC resistance represents a major hurdle to effective therapy, as it predicts poor outcomes, relapse, and death. Resistance develops through genetic/epigenetic changes (like mutations, receptor loss, or pathway activation) that prevent leukemia cells from undergoing apoptosis. Understanding these mechanisms is vital to develop drugs that overcome resistance, making GCs effective again and improving patient survival.

GR expression varies among different ALL cell lines and ALL patients’ samples. Reduced GR expression or function is believed to be the primary mechanism by which leukemia cells acquire glucocorticoid resistance. A notable correlation exists between low or absent GR levels and relapsed T–ALL [30]. This observation is mirrored in DEX-resistant cell lines like Jurkat and CEM–C1, which exhibit reduced basal GR expression compared to their DEX-sensitive counterparts [31]. A similar decrease in receptor expression has been observed in DEX-resistant B–ALL cell lines relative to their parental lines [32]. For example, REH does not express GR [12] whereas RS4;11 shows high expression [20], and the expression level influences their sensitivity to DEX (Figure 1A–D).

Previous research has established that DEX uptake is GR-dependent, where CCRF–CEM cells incorporated DEX more quickly and efficiently than Jurkat cells [19]. The increase in ROS production is also dependent on the GRs. However, no direct correlations between DEX sensitivity and increased ROS levels in response to DEX treatment were reported in T–ALL [19] and B–ALL [33,34] cell lines. These observations were also confirmed in the present study (Figure 1F).

While the majority of effects of GCs are mediated by the classical genomic pathway involving the GR, complementary pathways were reported to contribute to its toxic effects [35]. In this regard, in the present study, blocking GR with Ru486 completely prevented the cytotoxic effects of DEX in the CCRF–CEM cell line, while in another highly sensitive cell line, RS4;11, even after prolonged exposure times, a remnant population remains whose viability is still affected by DEX (Figure 2), which may indicate the existence of alternative mechanisms. It has been documented that in environments with high GR density, such as RS4;11 cells, Ru486 may exhibit partial agonist activity [36,37]. Furthermore, to determine whether GR-independent effects are present, it is necessary to perform GR knockdown in addition to employing higher concentrations of Ru486 to ensure complete receptor saturation.

Although functional GRs are the primary target for GCs, the ability of leukemic cells to modulate autophagy is a critical adaptive mechanism that contributes to the response to GC treatment. In this study, we used simultaneous flow cytometry monitoring of DEX-induced autophagy and cell death to determine the relationship between these processes (Figure 3). We demonstrated that in DEX-sensitive cell lines, the surviving population exhibits a high number of autophagosomes, as detected by MDC staining. However, induction of autophagy did not have a protective effect, since this population subsequently expresses a classic marker of apoptosis, namely the externalization of PS, which was detected using Annexin V. In contrast, in DEX-resistant cells, these processes did not occur (Figure 3).

Although the main cause of GC resistance in REH cells is the loss of GRs, this cell line was reported to possess high basal autophagy levels [38], which may indicate autophagy as a general chemoresistance mechanism in this cell line. Similar to our observations, Laane and colleagues [39] reported increased autophagy in DEX-sensitive RS4;11 and SUP-B15 cells, while no induction of autophagy was shown in REH cells after DEX treatment. In our previous work, CCRF–CEM cells showed higher autophagy level than Jurkat cells after treatment with DEX [19]. In contrast, the DEX-resistant Raji and U-937 cell lines showed autophagy induction when treated with DEX, while CCRF–CEM cells did not [40].

The above-described findings indicate that each cell line or malignant clone in leukemia patients have an optimal autophagy flux range required to maintain proteostatic equilibrium and mitochondrial integrity under steady-state conditions, suggesting that both its positive and negative modulation can enhance sensitivity to DEX. Establishing a universal ‘optimal’ threshold remains challenging due to inter-patient heterogeneity. However, determining whether pharmacological modulation (induction or inhibition) triggers a transition from pro-survival autophagy to autophagic cell death or metabolic collapse is pivotal for designing effective co-treatments for leukemic patients.

Based on this assumption, in this study, we decided to test how pharmacological modulation of autophagy, namely its induction by RAP or TAM, or inhibition by SP-1, or blockade with CQ, affects sensitivity of leukemic cell lines to DEX.

We observed that although RAP and SP-1 do not affect normal cell morphology, they were able to reduce cell proliferation as measured by cell count. This aligns with previous reports indicating that these modulators decrease proliferation on both lymphoid and myeloid leukemia cell lines without inducing cell death [41,42,43,44,45].

Autophagy modulation with RAP, SP-1, and TAM impacted the viability of most leukemic cell lines. Specifically, RAP enhanced DEX sensitivity in CCRF–CEM cells (Figure 4A), consistent with findings in other ALL and lymphoma-derived cell lines [46]. SP-1 successfully sensitized both Jurkat and CCRF–CEM cells (Figure 4B), an effect previously documented only in myeloid leukemias treated with SP-1 in combination with imatinib [47] or 5-azacytidine [43]. Although existing literature suggests that CQ enhances DEX sensitivity [19,48], it was the least cytotoxic modulator in our experiments. We observed its sensitizing effect to DEX in all T-ALL cell lines tested, but this effect was not maintained over time in Jurkat and MOLT-3 cells, which may be due to increased proliferation of the surviving cell population (Figure 4D).

Importantly, our previous research identified TAM as an effective sensitizer of Jurkat cells to DEX [29]. In the present study, this effect was reproduced in most leukemic cell lines tested (Figure 4C). Due to this broad and consistent efficacy, TAM was selected for subsequent, more detailed investigation as autophagy modulator.

Recent guidelines have identified distinct mechanisms by which autophagy is integrated into regulated cell death pathways. Firstly, autophagy-dependent cell death (ADCD) is defined as cellular demise strictly attributable to the autophagic process, occurring independently of both apoptosis and necrosis. Secondly, autophagy-mediated cell death (AMCD) involves autophagy preceding cell death; however, it has not been conclusively demonstrated that the autophagic process actively switches the cell towards death. Finally, autophagy-associated cell death (AACD) describes instances where autophagy simply coincides with other established cell death pathways, such as apoptosis, acting as an accompanying event without exerting an active role in the ultimate cellular outcome. To distinguish ADCD from AMCD or AACD, inhibition of the pathways responsible for autophagy is required [26,49,50].

The analysis of the combined treatment with TAM and DEX in various cell lines reveals distinct, cell-specific outcomes. However, in Jurkat and RS4;11 cell models, the decreased cell viability is potentially attributable to a proliferation arrest, as TAM reduces viability without inducing significant cell death. Nevertheless, further studies characterizing the cell cycle or proliferation markers are required to definitively establish a cytostatic mechanism. While in CCRF–CEM cells, the decrease in viability was also accompanied by an increase in cell death. (Figure 5).

In Jurkat cells, the increase in the double-positive population (autophagy–apoptosis) induced by the TAM and DEX co-treatment was no greater than that elicited by TAM alone. Similarly, in RS4;11 cells, the combination of DEX and TAM at 24 h fails to induce a higher level of autophagy than either single agent alone (Figure 5D). Crucially, the subsequent increase in the double-positive population in RS4;11 is a delayed event, manifesting only after 72 h of treatment (Figure 5E). The same delayed effect was observed in CCRF–CEM cells, where the combined drug treatment ultimately induced both a higher autophagy level and a greater double-positive population relative to single-agent treatments, suggesting its involvement in the death process (Figure 5B). The data suggests that an optimal autophagy threshold may influence cell viability; however, additional studies are required to define the precise molecular boundaries of this phenomenon.

In light of evidence that increased autophagy is an early event prior to cell death (as detailed in Section 3.3), we hypothesize that the increase in autophagy induced by TAM drives the cells towards either AMCD or AACD. In this proposed scenario, the initial induction of autophagy is interpreted as a cellular survival strategy that becomes ineffective or fails over time, ultimately resulting in cell death. However, whether the observed cell death is strictly AMCD or merely AACD remains an open mechanistic question. Our findings provide a basis for future experiments aimed at dissecting these distinct processes.

This behavior was not reported in other cellular models at higher drug concentrations. In the TAM-resistant breast cancer cell line, the combined administration of DEX (100 µM) and TAM (33 µM) demonstrated significantly greater efficacy in inducing apoptosis compared to the use of each drug as monotherapy [51].

The cytostatic effect is frequently linked to disturbance in cellular metabolism [52]. Similarly, modulation of autophagy can alter diverse metabolic pathways in cancer cells, potentially creating survival or escape mechanisms. This highlights the critical need for combined treatments that simultaneously target both cell metabolism and autophagy modulation [17]. Furthermore, a lack of cell death induction may allow a resistant or metabolically adapted population to resume proliferation once chemotherapeutic agents are withdrawn.

In summary, while GR expression and functionality remain the primary determinants of DEX sensitivity in ALL, this study identifies autophagy modulation as a critical target for overcoming chemoresistance. Our data show that pharmacological induction of autophagy, particularly with TAM, effectively sensitizes leukemic cells to DEX treatment. Nevertheless, the observed cell-specific variability and the potential for metabolic adaptation suggest that a universal therapeutic threshold remains elusive. The obtained results suggest the feasibility of incorporating pharmacological modulators of autophagy and cellular metabolism into future complex combination treatment protocols. These drugs may be useful for stimulating active cell death and reducing the risk of relapse in resistant ALL clones.

5. Conclusions

In this study, we demonstrated that GC-resistant ALL cell lines do not exhibit significant alterations in cell viability or autophagy levels after treatment with a wide range of DEX concentrations (up to 100 µM) for up to 72 h of observation. In contrast, in DEX-sensitive cell lines, a time-dependent decrease in viability was observed, associated with a marked induction of autophagy. It was confirmed that DEX sensitivity in ALL cell lines is GR-dependent.

TAM, which is capable to induce autophagy in ALL cells, effectively enhanced DEX sensitivity in most cell lines evaluated.

Our data suggest that disrupting the homeostatic balance of autophagy, primarily through its enhancement, can increase GC sensitivity in previously non-responsive cells. Therefore, further investigation of this phenomenon, along with the effect of autophagy inhibition on DEX sensitivity, represents an important area of research.