The MAO‐B Inhibitor Selegiline Reduces the Viability of Different Prostate Cancer Cell Lines and Enhances the Effects of Anti‐Androgen and Cytostatic Agents

ABSTRACT

The current treatments for advanced prostate adenocarcinoma (PAC) include the androgen receptor antagonist enzalutamide and docetaxel‐based chemotherapy. Elevated monoamine oxidase‐A (MAO‐A) mRNA expression and activity in tumorous prostate positively correlate with disease progression and therapy resistance. While MAO‐B mRNA expression was also demonstrated in PAC cell lines, its role remained unclear. Therefore, this study evaluates the effects of the irreversible MAO‐B inhibitor selegiline and rasagiline and their combinations with conventional therapies on androgen‐insensitive (PC‐3, DU145) and androgen‐sensitive (22Rv1, LNCaP, VCaP) PAC cell lines. MAO activity was determined by the MAO‐Glo luminescence assay, viability by the ATP‐based chemiluminescence method, proliferation by the Luna‐II automated cell counter, and mRNA expressions by RT‐qPCR. MAO‐B mRNA was stably expressed by all PAC cell lines, with the highest expression in 22Rv1 and LNCaP cells. Selegiline reduced MAO‐B activity by 75%–80% and decreased cell counts by 40%–50% at 100 μM in PC‐3 and 22Rv1 cells. Selegiline concentration‐dependently inhibited cell proliferation (100 μM–10 mM) and reduced viability (1–10 mM) similar to rasagiline in all cell lines. Combination with enzalutamide in 22Rv1, and with docetaxel in PC‐3 demonstrated potentiating and additive effects, respectively. Selegiline reduced FOXA1 and GLUT1 mRNA expressions related to cancer progression and metabolism in both cell lines, increased the apoptosis‐related BAX in PC‐3, and decreased AR, EGFR, and SNAI2 in 22Rv1 linked to proliferation and metastasis. These findings suggest potential for selegiline repurposing in both androgen‐sensitive and ‐insensitive PAC therapy by promoting apoptosis and inhibiting cancer growth and survival signals, respectively.

Distinct mechanisms of viability decreasing actions of selegiline in androgen‐insensitive PC3 and sensitive 22Rv1 PAC cell lines. MAO‐B inhibition in the mitochondrial membrane of PC3 cells leads to the upregulation of the BAX and downregulation of FOX and GLUT1 which potentially results in reduced glucose uptake and apoptosis induction. Meanwhile, besides inhibiting GLUT1‐dependent metabolic pathways, the major mechanism of 22Rv1 cells is the downregulation of EGFR and AR consequently reducing hormone signaling and tumor progression. MAO‐B, monoamine oxidase‐B; EGFR, epidermal growth factor receptor; FOXA1, forkhead box protein A1; HIF1α, hypoxia‐inducible factor 1A; GLUT1, glucose transporter 1; VEGF‐A, vascular endothelial growth factor A; KI67, Kiel 67; TWIST1, twist‐related protein 1; SNAI2, snail family transcriptional repressor 2; BAX, BCL2 associated X; AR, androgen receptor.

Abbreviations

AR

androgen receptor

BAX

BCL2 associated X

BMI1

B‐lymphoma Mo‐MLV insertion region 1

DHX9

DExH‐Box helicase 9

EGF

epidermal growth factor

EGFR

epidermal growth factor receptor

FOXA1

forkhead box protein A1

GLUT1

glucose transporter 1

HIF1A

hypoxia‐inducible factor 1A

HPRT1

hypoxanthine phosphoribosyltransferase 1

IPO8

importin 8

MAO

monoamine‐oxidase

MAP2K1

mitogen‐activated protein kinase kinase 1

MAPK1

mitogen‐activated protein kinase 1

MAPK14

mitogen‐activated protein kinase 14

PAC

prostate adenocarcinoma

PDK1

phosphoinositide‐dependent kinase 1

ROS

reactive oxygen species

SNAI2

snail family transcriptional repressor 2

STAT3

signal transducer and activator of transcription 3

TWIST1

twist‐related protein 1

VEGFA

vascular endothelial growth factor A—‐KI67—‐kiel 67

1. Introduction

Prostate adenocarcinoma (PAC) is the second most common cancer and the second leading cause of cancer‐related death among men in the United States and Europe [1].

The majority of PAC cases are androgen‐sensitive; therefore, androgen deprivation therapy is typically used as the first‐line treatment. However, PAC cells can eventually become androgen‐insensitive, leading to disease progression. In the treatment of metastatic castration‐resistant PAC, docetaxel is primarily used as a chemotherapeutic agent. Still, the median progression‐free survival remains limited to approximately 18.5 months [2]. The development of novel, original drugs is a time‐consuming, costly, and high‐risk process. Consequently, drug repurposing (also known as repositioning)—the application of an already approved drug or its combination for a new therapeutic application—represents a potentially more efficient and economical alternative.

Monoamine‐oxidase (MAO) is a mitochondrial enzyme that catalyzes the oxidative deamination of monoamine neurotransmitters and exogenous amines. It exists in two isoforms, MAO‐A, and MAO‐B, which are expressed in most cell types, most abundantly in the brain, but are also present with different distribution patterns in peripheral tissues [3].

The role of MAO‐A in PAC is well documented. It is expressed at both mRNA and protein levels in the basal epithelial cells of normal and malignant prostate tissue, where it counteracts differentiation and promotes the development of aggressive, dedifferentiated cancer subtypes. Elevated MAO‐A protein expression has been shown to correlate with malignancy in PAC [4]. Therefore, blocking of MAO‐A with an irreversible MAO‐A selective inhibitor clorgyline suppresses cell proliferation in various PAC cell lines. Furthermore, it is also interesting that MAO‐A transcript levels were increased in these cell lines after docetaxel treatment [5]. In a mouse 22Rv1 xenograft model, clorgyline, and phenelzine suppressed tumor growth and inhibited the development of resistance to enzalutamide [6]. Preliminary positive results from an ongoing clinical trial suggest that phenelzine may be effective in treating patients with biochemically recurrent, castration‐sensitive prostate cancer [7] (ClinicalTrials.gov Identifier: NCT02217709). However, the long‐term use of phenelzine in PAC can be significantly limited due to the development of adverse effects and unfavorable metabolic interactions with other drugs [8].

The role of MAO‐B has been implicated in colorectal carcinoma, where higher mRNA and protein expression was observed in tumor tissue than in non‐tumoral tissue. It also showed a positive correlation with epithelial‐to‐mesenchymal transition [9]. The largely unexplored and potentially underestimated role of MAO‐B in PAC, along with the hypothesis that MAO‐B inhibition may offer therapeutic benefit, prompted us to investigate selected MAO‐B inhibitors alone and in combinations in various PAC cell lines.

The potential beneficial effect of MAO‐B inhibitors is also based on some of their known properties. First, MAO‐B is detectable in PAC and positively correlates with poor clinical outcomes [10, 11]. Second, PAC progression is associated with oxidative stress [12], which can initiate a cascade of unfavorable cellular reactions. The protective effect of the irreversible MAO‐B inhibitor selegiline against oxidative stress has been demonstrated experimentally [13]. Third, selegiline, originally approved for the treatment of Parkinson's disease, exerts a long‐lasting therapeutic effect, explained by the neosynthesis of the MAO‐B enzyme, and has demonstrated safety in long‐term use. Moreover, its metabolic profile suggests no interactions with drugs routinely used in PAC therapy. Fourth, MAO‐B inhibitors are also capable of inhibiting MAO‐A activity [14]. A recent publication reporting a positive correlation between elevated stromal MAO‐B expression at both mRNA and protein levels correlating with poor clinical outcomes in PAC further supports our hypothesis [10, 11].

Although multiple factors are likely to contribute to poor outcomes, reduced survival, and therapy resistance, tumor hypoxia is recognized as one of the most significant negative prognostic indicators, for example, in head and neck cancer and bladder cancer [15].

Tumor hypoxia, characterized by elevated expression of the key regulator hypoxia‐inducible factor 1A (HIF1A), arises due to the high proliferation rate of cancer cells [16, 17]. HIF1A mRNA and protein levels are increased in various cancers, where HIF1A promotes tumor progression by regulating genes involved in metabolism, growth, survival, angiogenesis, and metastasis [18]. In PAC, disease progression is driven in part by VEGFA‐induced angiogenesis [19]. Epidermal growth factor receptor (EGFR) a key regulator of differentiation and proliferation, might further contribute to metastasis and poor clinical outcomes [20]. Kiel 67 (KI67), a widely used proliferation marker [21], is strongly associated with tumor cell growth [22] and shows a correlation with the expression of apoptosis‐promoting BCL2‐associated X (BAX) protein [23] and glucose transporter 1 (GLUT1) [24]. Elevated GLUT1 protein expression has also been linked to metastasis, facilitated by invasion and epithelial‐mesenchymal transition, which is regulated by transcription factors such as signal transducer and activator of transcription 3 (STAT3), twist‐related protein 1 (TWIST1) and snail family transcriptional repressor 2 (SNAI2) [25, 26, 27, 28]. Forkhead box protein A1 (FOXA1) protein is highly expressed in the prostate epithelium, where it promotes morphogenesis and maturation [29], and its physical interaction with the androgen receptor (AR) has been confirmed [30].

We investigated the effects of the irreversible, selective MAO‐B inhibitors selegiline and rasagiline alone as well as in combinations with the chemotherapeutic agent docetaxel and AR antagonist enzalutamide. For comparison, the MAO‐A inhibitors phenelzine, clorgyline, and moclobemide were also included in our experiments. To model the heterogeneity of PAC phenotypes, we assessed the effects of MAO inhibitors across five distinct cell lines: the androgen‐sensitive, slow‐growing LNCaP and VCaP, the fast‐proliferating 22Rv1, and the androgen‐insensitive, rapidly proliferating PC‐3 and DU145 lines. In addition, we investigated potential mechanisms of action by quantifying the mRNA expression of key genes involved in signaling pathways associated with cell proliferation, tumor progression, survival, metabolism, and apoptosis.

2. Materials and Methods

2.1. Test Compounds, Cell Lines, and Treatment Protocols

Selegiline hydrochloride (cat. nr.: M003) and enzalutamide (MDV3100) were purchased from Merck KGaA (Darmstadt, Germany), phenelzine sulphate (Cat. nr.: P6777) from Sigma‐Aldrich (St. Louis, MO, USA) and docetaxel from Avidin Ltd. (Cat. nr: S1148, Szeged, Hungary). Clorgyline hydrochloride [31], moclobemide hydrochloride [32], and rasagiline mesylate [33] were synthesized as previously described.

All cell lines were purchased from ATCC (Virginia, USA). The following cell lines were used for this research: androgen‐insensitive DU145 (HTB‐81, RRID: CVCL_0105) and PC‐3 (CRL‐1435, RRID: CVCL_0035), androgen‐sensitive LNCaP (CRL‐1740, RRID: CVCL_1379), 22Rv1 (CRL‐2505, RRID: CVCL_1045) and VCaP (CRL‐2876, RRID: CVCL_2235). Cell cultures were routinely tested for mycoplasma by PCR and regularly monitored for signs of bacterial, fungal, or other microbial contamination. All cultures were confirmed to be free of contamination throughout the experimental period. The androgen receptor (AR)‐positive 22Rv1 cells were used for the selegiline and enzalutamide combination treatment based on their sensitivity and doubling time. For the selegiline and docetaxel combination assay, AR‐negative, selegiline‐sensitive PC‐3 cells were used.

The cells were cultured in RPMI 1640 medium (Lonza, Basel, Switzerland) supplemented with 10% heat‐inactivated fetal bovine serum (Sigma‐Aldrich, Burlington, MA, USA) and penicillin–streptomycin (Thermo Fisher Scientific, Waltham, MA, USA) in a humidified atmosphere of 5% CO₂ at 37°C. All cells were plated at a density of 5000 cells/well in 96‐well cell culture plates.

For MAO inhibitors, the following concentrations were used: 1 μM, 10 μM, 100 μM, 1 mM, 10 mM, except for clorgyline where the highest concentration was 1 mM (n = 6) and for selegiline, where additional concentrations of 250 μM, 500 μM, 750 μM, 2 mM, and 3 mM were used for PC‐3 and 22Rv1 cells. The MAO inhibitors were dissolved in the cell culture medium prior to each experiment. For combinational assays, selegiline was used at concentrations of 250 μM, 500 μM, 750 μM and 1000 μM; enzalutamide at 10 μM and 50 μM and docetaxel at 1 μM (n = 6–10). DMSO was used as a solvent control for enzalutamide and docetaxel.

2.2. Assessment of Proliferation by Cell Counting

PC‐3 and 22Rv1 cells were seeded onto 24‐well plates and cultured for 48 h following treatment with 10 μM, 100 μM, 1 mM, 5 mM, and 10 mM of selegiline. The cells were passaged using a trypsin–EDTA solution (Lonza, Basel, Switzerland) and cell counting was performed using trypan blue (Sigma‐Aldrich, Burlington, MA, USA) with the help of a Luna‐II automated cell counter (Logos Biosystems, Anyang‐si, Gyeonggi‐do, South Korea).

2.3. MAO Activity Assay

Total MAO and MAO‐B enzyme activities were measured using the bioluminescent MAO‐Glo assay systems (cat. no.: V1401, Promega Corp, Madison, WI). Briefly, control cells and 100 μM selegiline‐treated cells were harvested and lysed in 1X Reporter Lysis Buffer (Promega Corp, Madison, WI), then frozen at −80°C to achieve complete cell lysis. Twenty‐five microliters of cell lysate was incubated with 25 μL of 2X MAO or 2X MAO‐B Substrate Solution at RT for 3 h. Fifty microliters of Luciferin Detection Reagent was added, and the mixture was incubated at RT for 20 min. A luminometric measurement was performed then on the EnSpire Multimode Plate Reader (Perkin Elmer, Waltham, MA, USA). The total MAO and MAO‐B activities in each sample were normalized to their respective protein concentrations, which were determined using the BioRad DC Protein Assay (Bio‐Rad Laboratories, Hercules, CA).

2.4. Viability Assay

The cells were seeded in 96‐well tissue culture plates and cultured for 48 h following treatment. Cell viability was determined by measuring their intracellular ATP of the cells using the Promega CellTiter‐Glo Luminescent Cell Viability Assay (cat. no: G7571, Promega, USA), following the manufacturer's instructions. The intensity of the emitted light is proportional to the intracellular ATP content of the cells being measured [34]. Luminometric measurement was performed using an EnSpire Multimode Plate Reader.

2.5. Ingenuity Pathway Analysis (IPA) Analysis

Interpretative phenomenological analysis was used to predict intracellular target molecules for the gene expression assay of selegiline‐treated cells. IPA analysis was performed using IPA (QIAGEN) software version 122103623 to test our hypothesis‐driven list of genes and molecules against known associations described in the literature and databases. We used the My Pathways, Connect, and Path Designer tools in the software to search the IPA knowledgebase and create a customized pathway visualization. We identified the following processes: activation, chemical‐protein interactions, expression, inhibition, localization, modification, molecular cleavage, phosphorylation, protein‐DNA interactions, protein–protein interactions, reaction, regulation of binding, transcription, translocation, and ubiquitination. We used the default settings considering all direct and indirect interactions.

2.6. RNA Isolation, TaqMan Assay and PCR Gel Electrophoresis

The untreated PC‐3, DU145, 22Rv1, and LNCaP cell lines, as well as the PC‐3 and 22Rv1 cells treated with 100 μM selegiline, were homogenized in 1 mL of TRI‐reagent. The total RNA was isolated using Direct‐zol RNA MicroPrep (Zymo Research, Irvine, CA, USA), according to the manufacturer's instructions. The samples were treated with 1 U of DNase I to eliminate any genomic DNA contamination. The amount and purity of the RNA were determined using a Jenway Genova Nano Micro‐Volume Spectrophotometer (Fisher Scientific, Loughborough, UK). cDNA synthesis was performed using a High‐Capacity Reverse Transcription Kit (Applied Biosystems Thermo Fisher Scientific, Waltham, MA, USA) with 500 ng of RNA.

TaqMan assays were conducted in a QuantStudio 5 system (Life Technologies Magyarország Ltd., Budapest, Hungary) in a 96‐well block, using the geometric means of hypoxanthine phosphoribosyltransferase 1 (HPRT1) and importin 8 (IPO8) as reference genes [35]. The reaction volume contained 1× SensiFAST Probe Lo‐ROX mix (Meridian Bioscience, Memphis, TN, USA), 400 nM of probe primer mix (forward and reverse) and 20 ng cDNA. The FAM‐conjugated TaqMan Gene Expression Assays (Thermo Fischer Scientific, Waltham, MA, USA) used to amplify the target loci are listed in Table 1. Gene expression ratios were calculated using the Ct method.

TABLE 1.

List of TaqMan probes used for gene expression analysis in this study.

Gene symbol	Official full name	Assay ID
HPRT1	Hypoxanthine phosphoribosyltransferase 1	Hs02800695_m1
IPO8	Importin 8	Hs00914057_m1
MAO‐A	Monoamine oxidase A	Hs00165140_m1
MAO‐B	Monoamine oxidase B	Hs01106246_m1
AR	Androgen receptor	Hs00171172_m1
GLUT1	Glucose transporter 1	Hs00892681_m1
VEGF‐A	Vascular endothelial growth factor A	Hs00900055_m1
SNAI2	Snail family transcriptional repressor 2	Hs00161904_m1
TWIST1	Twist‐related protein 1	Hs00361186_m1
HIF1A	Hypoxia‐inducible factor 1A	Hs00153153_m1
KI67	Kiel 67	Hs01032443_m1
EGFR	Epidermal growth factor receptor	Hs01076078_m1
BAX	BCL2 associated X	Hs00180269_m1
FOXA1	Forkhead box protein A1	Hs04187555_m1
STAT3	Signal transducer and activator of transcription 3	Hs00374280_m1

The PCR products were then electrophoresed on a 2% agarose gel containing Eco‐safe (Pacific Image Electronics, New Taipei, Taiwan) at 70 V for 40 min, before being visualized using a BioRad Gel Doc XR+ (Bio‐Rad laboratories, Hercules, CA) transilluminator.

2.7. Statistical Analysis

One‐way ANOVA followed by the Holm–Sidak's multiple comparisons test was used to assess the statistical significance of MAO activity, monotherapy treatments, and cell proliferation. Concentration–response curves were fitted using nonlinear regression with 95% confidence intervals. Multiple pairwise F‐tests were conducted to compare the EC₅₀ values among all experimental groups. mRNA expression data from untreated cell lines were analyzed using one‐way ANOVA followed by Dunn's post hoc test. For the selegiline‐treated groups, an unpaired t‐test was used. One‐way ANOVA with Tukey's multiple comparisons test was used for combined treatments with selegiline, docetaxel, and enzalutamide. In all cases, p < 0.05 was considered statistically significant.

3. Results

3.1. MAO‐A and MAO‐B Are Expressed at mRNA Levels and Functionally Active in PAC Cell Lines

Androgen‐sensitive (22Rv1, LNCaP) and insensitive (PC‐3, DU145) PAC cell lines show basal MAO‐A and MAO‐B mRNA expression, with higher MAO‐A levels observed in 22Rv1 and LNCaP and higher MAO‐B levels in DU145 cells (Figure 1A).

FIGURE 1.

MAO mRNA expression and activity in prostate adenocarcinoma (PAC) cell lines. (A) Expression of monoamine‐oxidase‐A (MAO‐A), monoamine‐oxidase‐B (MAO‐B) and androgen receptor (AR) mRNA in PC‐3, 22Rv1, DU145, and LNCaP PAC cell lines. Columns represent the mean ± SEM of mRNA fold change of MAO‐A, MAO‐B, and AR (*p < 0.05, ***p < 0.001, one‐way ANOVA with Dunn's post hoc test). (B) Total MAO and MAO‐B activity in PC‐3 and 22Rv1 cell lines after 48 h treatment with 100 μM selegiline. Columns represent the mean ± SEM of MAO activity values as a percentage (n = 12) obtained from two independent series of experiments. **p < 0.01, ***p < 0.001 versus respective control group, one‐way ANOVA followed by Holm–Sidak's multiple comparison test. (C) Cell numbers and concentration‐response curves of PC‐3 and 22Rv1 cells after 48 h selegiline treatment. Columns and curves represent the mean ± SEM of cell viability values expressed as a percentage (n = 12–30) obtained from three independent experiments. *Indicates a significance between treatment groups and respective controls *p < 0.05, ***p < 0.001, one‐way ANOVA followed by Holm–Sidak's multiple comparison test.

The transcript levels of AR mRNA are highly expressed in 22Rv1 and LNCaP, but are undetectable in PC‐3 and DU145 cells (Figure 1A). In comparison with 22Rv1, PC‐3 exhibited significantly lower total MAO and MAO‐B activities, measuring 60% ± 6.38% and 72% ± 9.7%, respectively. Treatment with 100 μM selegiline effectively reduced MAO‐B activity to the same levels in both cell lines (Figure 1B). Selegiline treatment induced a concentration‐dependent decrease in cell number in both cell lines, with statistically significant reductions observed at 100 μM (34.2% in PC‐3 and 43% in 22Rv1). At 10 mM, selegiline caused an 89% reduction in PC‐3 cell number and induced a viability reduction approaching baseline background levels in 22Rv1 (Figure 1C).

3.2. MAO‐B Inhibitors Significantly Reduce the Viability of PAC Cell Lines Similarly to MAO‐A Blockers

At a concentration of 1 mM, selegiline and rasagiline caused a significant decrease in cell viability of approximately 30%–50% across all tested cell lines; except for DU145, for which a significant effect was only observed at 10 mM (Figure 2). In PC‐3 and 22Rv1—selected as representative cell lines for further analysis—selegiline treatment resulted in a significant reduction in cell viability in PC‐3 at concentrations of 750 μM in 22Rv1 at 100 μM.

FIGURE 2.

Effects of different MAO inhibitors on the viability of prostate adenocarcinoma (PAC) cell lines. Columns represent the mean ± SEM of cell viability percentages from n = 6 to 16 experiments, obtained from two to four independent experimental series. (*indicates a significant difference between the treatment group and the respective control; *p < 0.05, **p < 0.01, ***p < 0.001, one‐way ANOVA followed by Holm–Sidak's multiple comparison test).

A significant decrease in cell viability was observed following the treatment with 100 μM of the non‐selective MAOI phenelzine and the irreversible MAO‐A inhibitor clorgyline, except in LNCaP cells, where phenelzine caused a significant effect already at 10 μM. Moclobemide, a reversible MAO‐A inhibitor, significantly reduced viability in PC‐3 and 22Rv1 cells at 1 mM concentration, while DU145 cells showed a significant viability decrease only at a higher concentration of 10 mM. In contrast, LNCaP cells showed a significant viability reduction at a lower concentration of 100 μM. Interestingly, ATP‐based viability increased in DU145 cells upon treatment with 1 mM moclobemide, and in PC‐3 cells following 1 μM and 10 μM phenelzine treatment (Figures 2 and 3).

FIGURE 3.

Concentration‐response curves of MAO inhibitors on the viability of prostate adenocarcinoma (PAC) cell lines. Curves represent the mean ± SEM of cell viability percentages from n = 6 to 16 experiments, obtained from two to four independent experimental series.

In PC‐3, 22Rv1, and LNCaP cells, selegiline showed a significantly higher EC₅₀ than phenelzine and clorgyline. Moclobemide treatment was significantly less effective than selegiline in the 22Rv1 and LNCaP cells. The corresponding EC₅₀ values together with confidence intervals and the result of the statistical analysis are provided in Table 2 and Tables S1–S10. EC₅₀ values for DU145 cells were not determined due to the lack of a concentration‐dependent response, except for phenelzine (Table 2).

TABLE 2.

EC₅₀ values of MAO inhibitors on the viabilities of prostate adenocarcinoma (PAC) cell lines.

Compound	Cell line	EC50 (μM)	95% CI bottom	95% CI top
Selegiline	PC‐3	2863	−68.5 to −27.7	99.4–109.7
	DU145	N/A	Very wide	Very wide
	22Rv1	2028	−58.1 to −21	96.8–108.6
	LNCaP	1538	−30.2 to −3.4	95–103.7
Rasagiline	PC‐3	3662	−53.4 to −20.6	95.7–100.6
	DU145	N/A	Very wide	Very wide
	22Rv1	1537	−23 to −8.6	98.2–102.9
	LNCaP	3212	−60.4 to −12.2	90.6–98.4
Phenelzine	PC‐3	88.27	−12.5 to 1.5	105.0–116.1
	DU145	100.6	−6.6 to 0.1	97.7–102.9
	22Rv1	331.4	−15.9 to −2.9	99.4–106.9
	LNCaP	23.9	−11.2 to 12.2	87.7–112.5
Clorgyline	PC‐3	259.9	−35.4 to −17.6	97.3–102.0
	DU145	N/A	Very wide	Very wide
	22Rv1	193.8	−27.9 to −13.6	101.6–106.4
	LNCaP	193.4	−29.8 to −10.1	96.4–103.0
Moclobemide	PC‐3	2967	−39.4 to −19.8	97.6–101.5
	DU145	N/A	Very wide	Very wide
	22Rv1	7767	−125.3 to −53.4	98.1–101.4
	LNCaP	9553	−1246 to −38.7	90.6–97.3

Note: N/A: Not applicable due to the absence of a concentration‐response relationship.

3.3. Addition of Selegiline Enhances the Inhibitory Effects of the Chemotherapeutic Agent Docetaxel and the Androgen Receptor Antagonist Enzalutamide on PAC Cell Viability

The selegiline‐docetaxel combinational assay was conducted using the androgen‐insensitive PC‐3 cell line. Selegiline exerted a concentration‐dependent effect on PC‐3 cell viability, reaching the statistically significant 30% reduction at 750 μM. Docetaxel (1 μM) alone caused a significant 15% decrease in viability, while its combination with 750 and 1000 μM selegiline resulted in significantly greater inhibitory effects of 40% and 36%, respectively. Selegiline, when administered as monotherapy, exerted no significant effect on cell viability at lower concentrations. However, in combination with docetaxel, it led to a reduction in ATP‐based viability (Figure 4A,C).

FIGURE 4.

Effect of selegiline in combination with docetaxel and enzalutamide on the viability of androgen‐insensitive PC‐3 and androgen‐sensitive 22Rv1 cell lines. (A) Effects of selegiline (250, 500, 750, 1000 μM) docetaxel (1 μM), and their combinations on PC‐3 cell viability. (B) Effects of selegiline (250, 500, 750, 1000 μM), enzalutamide (10, 50 μM), and their combinations on 22Rv1 cell viability. (C) Concentration‐response curves of selegiline in combination with docetaxel and enzalutamide in PC‐3 and 22Rv1 cell lines, respectively. Columns and curves represent the mean ± SEM of viability percentages from n = 6 to 10, obtained from 3 independent experimental series. Columns and curves represent the mean ± SEM of viability percentages (n = 6–10), obtained from three independent experimental series. *Indicates a significance between the treatment group versus respective vehicle controls; *p < 0.01, **p < 0.001; $ indicates the significant difference between the selegiline‐treated group and the corresponding combinational treatment groups; and # indicates a significant difference compared to the respective treatments (see indicating lines). One‐way ANOVA followed by Tukey's multiple comparisons test.

In order to evaluate the potentiating effect of the combination of selegiline and enzalutamide, the AR‐positive 22Rv1 cell line was used. In 22Rv1, selegiline treatment also led to a concentration‐dependent decrease in viability, with significant reductions of 14%, 22%, and 26% observed at 500, 750, and 1000 μM, respectively. In contrast, the administration of enzalutamide alone at a concentration of 10 μM did not result in any observable inhibitory effect. However, at a concentration of 50 μM, enzalutamide induced a significant 32% reduction in viability. Ten microliter enzalutamide significantly potentiated the effect of selegiline when combined. The combined effect was significantly greater than that of 750 or 1000 μM selegiline administered alone.

Furthermore, the inhibitory effects of combining 50 μM enzalutamide with 500–1000 μM selegiline were significantly greater than the effects of either treatment alone (Figure 4B,C).

3.4. Selegiline and MAO‐B Are Connected to Proliferation, Progression, Angiogenesis, Metabolism, Metastatis, and Apoptosis‐Related Pathways in PAC

To explore the potential tumor‐associated signaling pathways linked to MAO‐A and MAO‐B, as well as associated molecular mechanisms in PAC, IPA was used for further analysis. In the pathway diagram, gray arrows represent general tumor‐specific interactions, while blue arrows indicate interactions specific to PAC. Solid lines represent direct interactions, while dotted lines indicate indirect interactions. As illustrated in Figure 5 selegiline exerts a direct inhibitory effect on MAO‐B and an indirect inhibitory effect on MAO‐A. Both isoforms generate H₂O₂ and reactive oxygen species (ROS) as byproducts.

FIGURE 5.

Graphical summary generated in Ingenuity Pathway Analysis core analysis representing network of major protein markers of angiogenesis, cell proliferation, glucose metabolism and metastasis of prostate adenocarcinoma (PAC). Graphical summary was generated to overview the major biological themes of PAC. Solid lines represent direct, while dotted lines indicate indirect interactions. Gray lines represent tumor‐specific, while blue ones prostate cancer‐specific connections. The members selected for further qPCR analysis are indicated red. MAO‐A, monoamine oxidase‐A; MAO‐B, monoamine oxidase‐B; H₂O₂, hydrogen peroxide; ROS, reactive oxygen species; EGF, epidermal growth factor; EGFR, epidermal growth factor receptor; MAPK1, mitogen‐activated protein kinase 1; MAP2K1, mitogen‐activated protein kinase kinase 1; MAPK14, mitogen‐activated protein kinase 14; PDK1, phosphoinositide‐dependent kinase 1; FOXA1, forkhead box protein A1; BMI1, B‐lymphoma Mo‐MLV insertion region 1; DHX9, DExH‐Box helicase 9; AR, androgen receptor; HIF1A, hypoxia‐inducible factor 1A; GLUT1, glucose transporter 1; VEGF, vascular endothelial growth factor; KI67, kiel 67; TWIST1, twist‐related protein 1; SNAI2, snail family transcriptional repressor 2; BAX, BCL2 associated X; STAT3, signal transducer and activator of transcription 3.

MAO‐B has indirect relationships through various interactions with AR (via MAPK1 and SNAI2), VEGF, EGFR (via MAPK1 and MAPK14), GLUT1 (via H₂O₂ production), TWIST1 (via MAP2K1), SNAI2 (via MAP2K1 and H₂O₂), KI67 (via VEGF), BAX (via AR) and FOXA1 (via AR).

3.5. mRNA Expression Changes Related to Tumor Growth and Apoptosis in Response to Selegiline Treatment

Based on these molecular predictions, we examined the mRNA expression of the corresponding genes in untreated, as well as selegiline‐treated PC‐3 and 22Rv1 cells.

Basal GLUT1 and BAX mRNA levels were similar in both androgen‐insensitive PC‐3 and androgen‐sensitive 22Rv1 cells. 22Rv1 expressed higher basal levels of VEGFA, EGFR, TWIST1, KI67, and FOXA1 mRNA, and lower transcript levels of HIF1A and SNAI2 compared to PC‐3 (Figure 6A).

FIGURE 6.

Expression of angiogenesis‐, proliferation‐, apoptosis‐, glucose metabolism‐, and metastasis‐related mRNA in PC‐3 and 22Rv1 cell lines. (A) Expression levels of GLUT1, BAX, VEGFA, EGFR, TWIST1, KI67, FOXA1, HIF1A, and SNAI2 mRNA in untreated 22Rv1 cells relative to untreated PC‐3. (*indicates a significant difference in mRNA expression of PC‐3 and 22Rv1, *p < 0.05, ***p < 0.001, unpaired t‐test). (B) mRNA expression in PC‐3 and 22Rv1 cell lines following 48 h treatment with 100 μM selegiline. Genes related to tumor progression, angiogenesis, proliferation, glucose metabolism, and metastasis were analyzed. (*indicates a significant difference between the treatment group and the respective control; *p < 0.05, **p < 0.01, ***p < 0.001, unpaired t‐test).

Treatment with 100 μM selegiline led to a significant reduction in GLUT1 and FOXA1 transcript levels in both cell lines, without affecting HIF1A and KI67 mRNA expression. Selegiline induced a significant increase in the expression of VEGFA, TWIST1, and BAX mRNA in PC‐3 cells, while it caused a decrease in the AR, EGFR, SNAI2, and BAX mRNA in 22Rv1 cells. STAT3 mRNA expression was not detected in PC‐3 cells and remained unchanged in 22Rv1 cells following selegiline treatment. (Figure 6B).

4. Discussion

This is the first study to demonstrate the effects of the selective, irreversible MAO‐B inhibitor selegiline in both androgen‐sensitive and ‐insensitive PAC cell lines. Our findings show that selegiline significantly reduces total MAO and MAO‐B enzyme activity, as well as decreases cell viability. These effects may be mediated, at least in part, through the modulation of hypoxia, angiogenic processes, and glucose metabolic pathways. Furthermore, selegiline enhances or potentiates the cytotoxic effects of the chemotherapeutic agent docetaxel and the androgen receptor antagonist enzalutamide.

Although MAO‐A has been considered a promising therapeutic target due to its elevated protein expression and its correlation with poor prognosis in metastatic PAC, its clinical use is limited by an unfavorable side‐effect profile. To date, only the non‐selective MAOI phenelzine has entered phase II clinical investigation for PAC, where it demonstrated a reduction in PSA levels (ClinicalTrials.gov Identifier: NCT02217709) [7].

In contrast, the oncological potential of MAO‐B blockers remains largely unexplored, representing a novel therapeutic avenue [36]. Nevertheless, recent studies have indicated a role for MAO‐B in the pathogenesis of several cancers, including colorectal cancer, breast cancer, and oral squamous cell carcinoma [37]. Notably, following the submission of the patent application and the initiation of a clinical trial, elevated stromal MAO‐B protein levels were reported in castration‐resistant PAC patients compared to hormone‐sensitive patients. The elevated MAO‐B protein level is associated with the neoplasm marker chromogranin A in adjacent tumor tissue of castration‐resistant patients and is linked to poor clinical outcomes [10]. Importantly, silencing of MAO‐B in normal human prostate stromal cells did not affect viability or cell cycle, while in both androgen‐sensitive and ‐insensitive cancer, MAO‐B silencing significantly reduced the proliferation rate, cell migration, and invasion [11].

The MAO‐B mRNA expression levels differ in the investigated PAC cell lines supporting their distinct MAO activities. The MAO‐B‐mediated cytotoxic effect of selegiline is suggested by a significant reduction in cell number and viability in response to lower concentrations, along with similar effects of rasagiline and the IPA analysis. The significantly higher level of MAO‐B activity in 22Rv1 cells compared to PC‐3 cells might explain the higher potency of selegiline in this cell line, as demonstrated by its lower EC₅₀ value.

Interestingly, treatment with 1–10 μM phenelzine in PC‐3 and 1 mM moclobemide in DU145 cells resulted in elevated intracellular ATP levels. Physiological MAO‐A activity exerts a significant influence on cellular redox homeostasis and mitochondrial function. The knockdown of MAO‐A in human neuroblastoma and gastric cancer cells has been shown to reduce ROS, enhance mitochondrial complex I activity, and increase ATP production while simultaneously suppressing cell migration and invasion [38, 39]. Similarly, selegiline treatment significantly reduced viable cell count at a 100 μM concentration, while ATP‐based viability was reduced only at 1 mM. Recent literature has demonstrated that harmol, a modulator of MAO‐B and GABA‐A, has the capacity to enhance mitochondrial markers and ATP levels in differentiated C2C12 myotubules, highlighting the mitochondrial impact of MAO inhibition [40].

The role of MAO enzymes, along with MAPK and EGF signaling pathways and ROS production, has been implicated in the development, maintenance, and therapeutic resistance of cancer [37, 41, 42, 43, 44]. Although MAO‐A inhibitors have shown potential to prevent taxane resistance, data remain inconsistent due to tumor heterogeneity and molecular variability [45].

We observed that selegiline significantly downregulated GLUT1 transcript levels in both 22Rv1 and PC‐3 cells. Since GLUT1 inhibition impairs glycolysis and cell growth, this may contribute to the observed reduction in cell viability. Previous studies have shown that GLUT1 knockdown reduces tumor volume in xenograft models using both enzalutamide‐sensitive and ‐resistant cell lines [46]. In our experiments, selegiline treatment caused decreased FOXA1 mRNA levels in both cell lines and reduced the AR mRNA in 22Rv1 cells. These results are supported by earlier studies demonstrating that FOXA1 silencing in LNCaP cells significantly reduced AR activity and cell proliferation [47]. These results suggest that selegiline exerts its effect on PAC cells, at least in part, by disrupting AR signaling and its downstream transcriptional network. This finding is consistent with previous research conducted with clorgyline [48].

We detected higher baseline KI67 transcript levels in 22Rv1 than in PC‐3 cells. While elevated KI67 is often associated with aggressive tumors, it may also enhance immunogenicity, potentially providing therapeutic benefit [49].

Elevated HIF1A transcript level was reported in normoxic conditions in cancers and is linked to antiandrogen therapy resistance and PAC progression [50]. HIF1A regulates a wide range of genes, including VEGF, a known marker of angiogenesis, poor prognosis, and metastasis [19, 51, 52]. Anti‐angiogenic drugs such as the anti‐VEGF monoclonal antibody bevacizumab, VEGF‐receptor‐linked tyrosine kinase inhibitor sorafenib, and thalidomide influencing the angiogenic signaling pathways have long been suggested in PAC therapy [53]. Therefore, the combination of selegiline with anti‐angiogenic drugs may overcome the potential pro‐angiogenic side effects of selegiline.

In response to selegiline treatment, a decrease in EGFR mRNA levels was observed in 22Rv1 cells. EGFR protein is known to be elevated and associated with hormone resistance and cancer progression in PAC patients. The observed reduction following selegiline administration suggests a potential therapeutic benefit for androgen‐sensitive PAC [54].

Basal SNAI2 mRNA levels were higher in PC‐3, compared to 22Rv1 cells, that can be explained by its metastatic origin, and supported by previous studies also showing elevated levels of SNAI2 mRNA in distant metastases, compared to adjacent lymph node metastases and primary tumor samples [26, 55, 56].

While a previous study reported decreased TWIST1 transcript level following MAO‐B knockdown in both androgen‐sensitive and ‐insensitive cell lines, our findings showed different mRNA levels between PC‐3 and 22Rv1 cell lines [11], suggesting stromal‐tumor crosstalk after selegiline treatment. Furthermore, STAT3 [28] mRNA was undetectable in PC‐3 and unaltered by selegiline in 22Rv1.

The expression of pro‐apoptotic BAX mRNA was also different in androgen‐sensitive and ‐insensitive PAC cell lines in response to selegiline treatment, with higher levels measured in PC‐3 cells. This elevation suggests that selegiline may promote apoptosis via mitochondrial membrane permeabilization in androgen‐insensitive cells [57].

In summary, we demonstrate that micromolar concentrations of selegiline reduce PAC cell viability presumably via MAO‐B inhibition. Selegiline reduces the viability of both PC‐3 and 22Rv1 cells primarily through impairing metabolic and glycolytic processes. Furthermore, it also induces apoptosis in androgen‐insensitive PC‐3 cells, while it suppresses AR signaling and EGFR mRNA expression in androgen‐sensitive 22Rv1 cells.

It is concluded that incorporating selegiline into conventional PAC treatment regimens could improve clinical outcomes by improving prognosis, overcoming therapy resistance, and prolonging the survival of PAC patients. Based on these and other compelling preclinical findings, we recently filed a patent application describing the therapeutic benefits of MAO‐B inhibitors in PAC (WO2021024005) and initiated a Phase IIa clinical trial to evaluate selegiline in combination with docetaxel in PAC patients (ClinicalTrials.gov Identifier: NCT04586543).

Author Contributions

Anita Steib: data curation, methodology, visualization, writing – original draft. Krisztina Pohóczky: formal analysis, supervision, visualization, writing – original draft, writing – review and editing. Norbert Tóth: methodology, visualization. Viktória Kormos: methodology, writing – review and editing. József Kun: methodology, visualization. Tamás Kálai: methodology, writing – review and editing. László Mangel: conceptualization, writing – review and editing. Péter Mátyus: conceptualization, funding acquisition, writing – review and editing. Zsuzsanna Helyes: conceptualization, formal analysis, funding acquisition, writing – original draft, writing – review and editing.

Consent

The authors have nothing to report.

Conflicts of Interest

Author Péter Mátyus is also employed by E‐Group ICT Software Plc. The authors declare no conflicts of interest. The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of this manuscript, or the initiation to submit it for publication; nonetheless, the funders gave their mutual consent to the publication.

Supporting information

Figure S1: Effects of different MAO inhibitors on the viability of VCaP cell line. Columns represent the means ± SEM of the percentage of viability of n = 6 experiments obtained from two independent series. (*indicates the significance between the treatment group vs. respective controls ***p < 0.001, one‐way ANOVA followed by Holm–Sidak's multiple comparisons).

Figure S2: Concentration‐response curves of MAO inhibitors on the viability of VCaP cell line. Curves represent the means ± SEM of the percentage of viability of n = 6 experiments obtained from two independent series.

Table S1: EC₅₀ values of MAO inhibitors on the viability of VCaP cell line.

Table S2: Statistical comparisons of EC₅₀ values of MAO inhibitors on the viability of PC‐3 prostate adenocarcinoma (PAC) cell line. Multiple pairwise F‐tests were conducted to compare the EC₅₀ values between all experimental groups. The table shows the statistics performed using n = 6–16 experiments, obtained from four independent experimental series.

Table S3: Statistical comparisons of EC₅₀ values of MAO inhibitors on the viability of 22Rv1 prostate adenocarcinoma (PAC) cell line. Multiple pairwise F‐tests were conducted to compare the EC₅₀ values between all experimental groups. The table shows the statistics performed using n = 6–15 experiments, obtained from four independent experimental series.

Table S4: Statistical comparisons of EC₅₀ values of MAO inhibitors on the viability of LNCaP prostate adenocarcinoma (PAC) cell line. Multiple pairwise F‐tests were conducted to compare the EC₅₀ values between all experimental groups. The table shows the statistics performed using n = 6 experiments, obtained from two independent experimental series.

Table S5: Statistical comparisons of EC₅₀ values of MAO inhibitors on the viability of VCaP prostate adenocarcinoma (PAC) cell line. Multiple pairwise F‐tests were conducted to compare the EC₅₀ values between all experimental groups. The table shows the statistics performed using n = 6 experiments, obtained from two independent experimental series.

Table S6: Comparison of EC₅₀ values for selegiline‐induced viability changes in the analyzed prostate adenocarcinoma (PAC) cell lines. Multiple pairwise F‐tests were conducted to compare the EC₅₀ values between all experimental groups. The table shows the statistics performed using n = 6–16 experiments, obtained from two to four independent experimental series.

Table S7: Comparison of EC₅₀ values for rasagiline‐induced viability changes in the analyzed prostate adenocarcinoma (PAC) cell lines. Multiple pairwise F‐tests were conducted to compare the EC₅₀ values between all experimental groups. The table shows the statistics performed using n = 6 experiments, obtained from two independent experimental series.

Table S8: Comparison of EC₅₀ values for phenelzine‐induced viability changes in the analyzed prostate adenocarcinoma (PAC) cell lines. Multiple pairwise F‐tests were conducted to compare the EC₅₀ values between all experimental groups. The table shows the statistics performed using n = 6–9 experiments, obtained from two to three independent experimental series.

Table S9: Comparison of EC₅₀ values for clorgiline‐induced viability changes in the analyzed prostate adenocarcinoma (PAC) cell lines. Multiple pairwise F‐tests were conducted to compare the EC₅₀ values between all experimental groups. The table shows the statistics performed using n = 6 experiments, obtained from two independent experimental series.

Table S10: Comparison of EC₅₀ values for moclobemide‐induced viability changes in the analyzed prostate adenocarcinoma (PAC) cell lines. Multiple pairwise F‐tests were conducted to compare the EC₅₀ values between all experimental groups. The table shows the statistics performed using n = 6 experiments, obtained from two independent experimental series.

Steib A., Pohóczky K., Tóth N., et al., “The MAO‐B Inhibitor Selegiline Reduces the Viability of Different Prostate Cancer Cell Lines and Enhances the Effects of Anti‐Androgen and Cytostatic Agents,” Pharmacology Research & Perspectives 13, no. 5 (2025): e70173, 10.1002/prp2.70173.

Data Availability Statement

The authors have nothing to report.