Molecular crossroads: identifying MAPK proteins bridging SMAD and dopamine pathways in breast cancer

ABSTRACT

The transforming growth factor-beta (TGF-β)/SMAD signaling pathway, mitogen-activated protein kinase (MAPK) signaling cascade, and dopamine receptor activity are all implicated in tumor progression. This study investigates molecular interactions among these pathways, identifying MAPK proteins that bridge SMAD and dopamine signaling in the context of breast cancer pathogenesis. A cohort of 405 breast cancer patients was categorized into molecular subtypes: luminal A (n = 130), luminal B HER2-negative (n = 100), luminal B HER2-positive (n = 96), non-luminal HER2-positive (n = 36), and triple-negative breast cancer (TNBC; n = 43). Transcriptomic profiling using microarrays and bioinformatics-based network analysis revealed differentially expressed genes across subtypes. Quantitative reverse transcription polymerase chain reaction (qRT-PCR) was used to validate 11 key transcripts, and protein expression was quantified by enzyme-linked immunosorbent assay (ELISA). MicroRNA (miRNA) regulatory interactions were analyzed to assess post-transcriptional modulation. Among 167 differentially expressed genes, 14 were consistently altered across all subtypes. These included cell division cycle 42 (CDC42), KRAS proto-oncogene, GTPase (KRAS), and transforming growth factor beta 1 (TGFB1), which were upregulated, as well as fibroblast growth factor 2 (FGF2), fibroblast growth factor 7 (FGF7), and insulin-like growth factor 1 (IGF1), which were downregulated. miRNA analysis revealed miR-221, miR-222, and miR-16-5p as regulators of these pathways. ELISA confirmed reduced KIT, IGF1, and FGF family proteins in tumor tissues, with KRAS significantly upregulated. Protein interaction analysis highlighted key hubs linking MAPK, SMAD, and dopamine signaling. This study elucidates crucial molecular intersections between MAPK, SMAD, and dopamine pathways, identifying potential biomarkers and therapeutic targets for breast cancer.

Introduction

Breast cancer remains one of the most prevalent and lethal malignancies Worldwide [1], driven by intricate molecular mechanisms that govern tumor initiation, progression, and resistance to therapy [2]. The tumor microenvironment and intracellular signaling networks play a crucial role in shaping disease outcomes, yet the complex interplay between key oncogenic pathways remains incompletely understood [3–7]. Among these pathways, transforming growth factor-beta (TGF-β) signaling is particularly significant due to its paradoxical role in cancer progression [8]. In early-stage breast cancer, TGF-β acts as a tumor suppressor by inhibiting cell proliferation and inducing apoptosis. However, in later stages, it promotes tumor invasiveness, metastasis, and immune evasion by enhancing epithelial-to-mesenchymal transition (EMT) [9] and remodeling the extracellular matrix [10–12]. The intracellular effectors of TGF-β signaling, known as Suppressor of Mothers Against Decapentaplegic (SMAD) proteins, serve as transcriptional regulators that mediate these tumorigenic effects [13,14].

In parallel, emerging evidence suggests that the dopaminergic system, traditionally associated with neurotransmission, plays a role in tumor biology, including in breast cancer [15,16]. Dopamine, acting through its dopamine receptors (DRs), has been shown to influence tumor growth, angiogenesis, and immune cell infiltration [17,18]. Some studies suggest that dopamine receptor activation can suppress tumor progression by reducing inflammatory signaling and inducing apoptosis, while others highlight its role in enhancing proliferation in certain tumor subtypes [19]. However, the precise mechanisms by which dopamine signaling interacts with other oncogenic pathways in breast cancer remain largely elusive [20–22].

A key regulatory network that could mediate crosstalk between the TGF-β/SMAD pathway and dopamine signaling is the Mitogen-Activated Protein Kinase (MAPK) signaling cascade [23,24]. The MAPK pathway serves as a critical hub for transducing extracellular signals into intracellular responses, influencing cell proliferation, survival, and differentiation. In breast cancer, the MAPK pathway – comprising key subfamilies such as extracellular signal-regulated kinases (ERK1/2), c-Jun N-terminal kinases (JNK), and p38 MAPKs – has been implicated in resistance to therapy, metastasis, and tumor immune evasion [25,26]. The interplay between these pathways is particularly intriguing, as TGF-β signaling, primarily mediated by SMAD proteins, can also activate non-canonical pathways, including the MAPK cascade, influencing tumor progression and drug resistance [27,28]. Additionally, dopamine receptors modulate MAPK activity through G-protein-coupled receptor (GPCR) signaling, impacting cell survival and proliferation. SMAD proteins have been shown to interact with MAPK effectors, altering transcriptional responses to TGF-β [29,30], and emerging evidence suggests that dopamine may regulate TGF-β activity via MAPK-dependent mechanisms, further influencing EMT and immune responses. However, the precise molecular intermediates bridging these pathways remain unidentified [25,26,30].

Despite significant advancements in breast cancer research, the functional interconnection between the TGF-β/SMAD pathway, dopamine signaling, and the MAPK cascade remains poorly understood. Most studies have examined these pathways in isolation, failing to explore their potential molecular convergence [31]. Given the critical role of MAPK proteins in transducing oncogenic signals, identifying the specific MAPK components that bridge TGF-β/SMAD and dopamine receptor signaling could provide novel insights into breast cancer pathophysiology and therapeutic resistance [32].

Furthermore, recent studies have highlighted the pivotal role of microRNAs (miRNAs) – small non-coding RNAs that regulate gene expression at the post-transcriptional level – in modulating these signaling pathways [33,34]. miRNAs influence breast cancer progression by targeting key regulators within the TGF-β, dopamine, and MAPK signaling networks, thereby affecting cellular proliferation, apoptosis, and metastasis. Understanding the regulatory role of miRNAs in these pathways may provide new opportunities for therapeutic interventions [35–37].

This study aims to bridge this knowledge gap by identifying the MAPK proteins that mediate crosstalk between the TGF-β/SMAD and dopamine pathways in breast cancer. By integrating bioinformatics analysis with experimental validation, we seek to unravel the molecular interactions that drive tumor progression and therapeutic response.

Material and methods

Patient selection and sample collection

The study cohort included patients diagnosed with different breast cancer subtypes, with luminal A accounting for 130 cases, luminal B comprising 196 cases (subdivided into HER2-negative (100 cases) and HER2-positive (96 cases)), non-luminal HER2-positive totaling 36 cases, and triple-negative breast cancer (TNBC) identified in 43 patients. Tumor samples, along with matched non-cancerous control tissues, were surgically removed. Histopathological evaluation was conducted intraoperatively to confirm negative surgical margins. If tumor-free margins were not ensured, additional tissue resection was performed. Tumor classification followed the Tumor, Nodes, and Metastases (TNM) system, with all patients meeting the T1N0M0 criteria.

Among luminal A patients, 18% were classified as G1, 37% as G2, and 45% as G3. Age-wise, 33% were younger than 50 years, while 67% were older, with an average BMI of 30.78 ± 2.76 kg/m². Hypertension was present in 12% of cases, diabetes mellitus type 2 in 8%, and a combination of both conditions in 7%, whereas 72% had no reported comorbidities.

In the luminal B HER2-negative group, 31% of patients were in G1, 57% in G2, and 12% in G3. A total of 32% were under 50 years of age, while 68% were older, with an average BMI of 30.18 ± 4.56 kg/m². Hypertension affected 12%, diabetes mellitus type 2 was observed in 7%, and 6% had both conditions, whereas 75% reported no comorbidities.

For luminal B HER2-positive cases, G1 classification was found in 24%, G2 in 59%, and G3 in 17%. A majority (80%) of patients were over 50 years old, while 20% were younger, with an average BMI of 32.09 ± 6.19 kg/m². Hypertension and diabetes were noted in 12% and 9% of cases, respectively, while 6% had both conditions, and 73% had no reported comorbidities.

Among non-luminal HER2-positive patients, 25% were classified as G1, 33% as G2, and 42% as G3. Approximately 25% were younger than 50 years, whereas 75% were older, with an average BMI of 33.18 ± 5.67 kg/m². Hypertension was present in 11%, diabetes mellitus type 2 in 7%, and both conditions were observed in 7%, while 74% of patients had no associated comorbidities.

For TNBC patients, G1 tumors were observed in 32%, G2 in 49%, and G3 in 19%. In terms of age, 23% of cases were younger than 50, while 77% were older, with an average BMI of 34.67 ± 2.98 kg/m². Hypertension was found in 12%, diabetes mellitus type 2 in 6%, and a combination of both in 9%, while 73% of patients had no reported comorbidities.

Control samples

Matched non-cancerous breast tissue was obtained from the same patients who provided tumor specimens. Control tissue was collected during surgical resection from sites located at least 2 cm from the tumor margin to minimize the risk of malignant cell contamination. Intraoperative histopathological examination was performed to confirm the absence of neoplastic or preneoplastic changes. If tumor-free margins were not ensured, additional resection was undertaken to obtain adequate control material.

Immediately after excision, control samples were rinsed in sterile phosphate-buffered saline (PBS) to remove residual blood, snap-frozen in liquid nitrogen, and stored at − 80°C until further processing. total ribonucleic acid (RNA) and protein isolation procedures for control tissues were performed in parallel with tumor samples, using identical reagents, workflows, and quality control measures. For RNA analysis, TRIzol reagent (Invitrogen Life Technologies, California, USA) extraction was followed by purification with the RNeasy Mini Kit (QIAGEN, Valencia, CA, USA) and DNase I (Fermentas International Inc., Burlington, Canada) treatment. RNA integrity and concentration were verified by agarose gel electrophoresis and spectrophotometric measurement.

In all molecular assays, control samples served as the baseline reference for expression analysis. For quantitative reverse transcription polymerase chain reaction (qRT-PCR), relative expression levels were normalized to β-actin (ACTB) and expressed as fold change compared to the mean value of the control group using the 2^−ΔΔCt method. For protein studies (Enzyme-Linked Immunosorbent Assay, ELISA, Western blot), lysates were prepared from control tissues in the same manner as for tumor specimens, ensuring identical handling conditions.For ELISA and Western blot analyses, control group values were used as the reference for statistical comparisons with the respective breast cancer subtypes.

RNA extraction and purification

RNA was isolated from tissue samples using TRIzol reagent (Invitrogen Life Technologies, California, USA) in accordance with the manufacturer’s recommended protocol. Further purification was performed using the RNeasy Mini Kit (QIAGEN, Valencia, CA, USA) to remove potential contaminants. To eliminate residual genomic DNA, samples were treated with DNase I (Fermentas International Inc., Burlington, Canada). The integrity of RNA was assessed via 1% agarose gel electrophoresis stained with ethidium bromide, and RNA concentration was quantified using spectrophotometric analysis at 260 nm to determine purity and yield.

Gene expression analysis of the MAPK pathway in the context of SMAD and Dopamine signaling

Genes associated with the MAPK, SMAD, and dopamine signaling pathways were identified using data from the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://pathcards.genecards.org/, accessed on March 1, 2025). To explore interactions between these pathways, a network analysis was conducted using STRING Database (v11.0) [38], focusing on the hsa04010 MAPK signaling pathway. Initially, MAPK pathway genes were cross-referenced with differentially expressed genes from previously published studies on SMAD-related signaling [39] to identify overlapping elements. A similar analysis was performed to establish connections between MAPK pathway genes and differentially expressed genes from studies on dopamine signaling [40]. The intersection of these datasets allowed the identification of MAPK proteins that interact with both SMAD and dopamine pathways, designating them as convergent molecular nodes within these signaling networks.

To analyze differential gene expression between tumor and control tissues, HG-U133_A2 microarrays (Affymetrix, Cat. No. 902416) were utilized. Of the 22,277 probes present on the microarray, specific targets corresponding to MAPK pathway genes were identified. The methodological workflow included complementary DNA (cDNA) synthesis, RNA amplification, and hybridization, followed by fluorescence-based detection using the Affymetrix Gene Array Scanner 3000 7 G.

miRNA-Mediated regulatory interactions linking MAPK, SMAD, and dopamine pathways

To explore miRNAs regulation of MAPK, SMAD, and dopamine pathways-associated genes, a microarray-based miRNA expression analysis was conducted using the GeneChip miRNA 2.0 Array (Affymetrix, Santa Clara, CA, USA). Differentially expressed miRNAs between tumor and control samples were identified using TargetScan [41] and miRanda [42] databases. Predictions of miRNA-mRNA interactions were classified based on confidence scores: interactions scoring above 90 were considered highly reliable, while those below 60 required additional validation [13,42].

qRT-PCR validation

To confirm microarray findings, qRT-PCR)was conducted using the SensiFast SYBR No-ROX One-Step Kit (Bioline, London, UK) according to the manufacturer’s guidelines. Gene expression levels were calculated using the 2^−ΔΔCt method, with values normalized against ACTB as an internal control. Expression changes were interpreted as follows: a fold change of 1 represented baseline control levels, values exceeding 1 indicated upregulation, and values below 1 signified gene silencing. Primer sequences used in this study are detailed in Table 1. All primers were procured from Sigma-Aldrich (St. Louis, MO, USA).

Table 1.

Nucleotide sequence of primers.

mRNA	Nucleotide sequence
CDC42	Forward	5’- TAGGCCTGTGAGTTGGGAAG −3’
	Reverse	5’- AGTATCCCTGACCGTTTGC −3’
EGF	Forward	5’- CAGGGAAGATGACCACCACT-3’
	Reverse	5’- TCTCGGTACTGACATCGCTC-3’
EGFR	Forward	5’- AAAACCGGACTGAAGGAGCT-3’
	Reverse	5’- TGGATCACACTTTTGGCAGC −3’
ERBB3	Forward	5’- TCCTCCCAATTCCTGTGCAT-3’
	Reverse	5’- CCACCACCACTTCCTGAGAT-3’
FGF2	Forward	5’- AGTCTTCGCCAGGTCATTGA3’
	Reverse	5’- CCTGAGTATTCGGCAACAGC-3’
FGF7	Forward	5’- TTGTGGCAATCAAAGGGGTG-3’
	Reverse	5’- CATTTCCCCTCCGTTGTGTG-3’
IGF1	Forward	5’- TCTGAATCTTGGCTGCTGGA-3’
	Reverse	5’- TGTGCTTCTTGACGACTTGC-3’
KIT	Forward	5’- ACTGTGGCCGTTATCTGGAA-3’
	Reverse	5’- GAAGTGCCCCTGAAGTACCT-3’
KRAS	Forward	5’- GCCCAATCCATTAGCGACAG-3 ’
	Reverse	5’- AGCGAGACTCTGACACCAAA-3 ’
RASA1	Forward	5’- CACCACCAGAGCCAGTAGAA-3’
	Reverse	5’- CGGCCCACCTCTTCTACTAG-3’
TGFB1	Forward	5’- TGAACCGGCCTTTCCTGCTTCTCATG −3’
	Reverse	5’- CGGAAGTCAATGTACAGCTGCCGC-3’
ACTB	Forward	5’-TCACCCACACTGTGCC CATCTACGA-3’
	Reverse	5’-CAGCGGAACCGCTCATTGCCAATGG-3’

CDC42, Cell Division Cycle 42; EGF, Epidermal Growth Factor; EGFR, Epidermal Growth Factor Receptor; ERBB3, Erb-B2 Receptor Tyrosine Kinase 3; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase; RASA1, RAS p21 Protein Activator 1; TGFB1, Transforming Growth Factor Beta 1; ACTB, beta actin.

Protein expression analysis via ELISA

The protein expression profile of KIT, IGF1, FGF2, FGF7, and KRAS was evaluated using ELISA, employing commercially available kits from MyBioSource, Inc., San Diego, CA, USA. The assay was conducted on tissue samples, ensuring high specificity and sensitivity in detecting protein concentrations across different breast cancer subtypes. Standard curves and controls were included in each experiment to validate the accuracy and reproducibility of the measurements. The detection range and sensitivity of each assay were optimized according to the manufacturer’s protocols to achieve reliable quantification.

ACTB (ab8227; Abcam Limited, Cambridge, UK) was used as the endogenous control protein. Primary antibodies against KIT (ab32363), KRAS (ab275876), FGF2 (ab208687), FGF7 (ab131162), IGF1 (ab134140) were purchased from Abcam. HRP-conjugated goat anti-rabbit IgG (ab97051; Abcam) was used as the secondary antibody. The optical density of each blot lane was measured using Kodak MI 4.5SE software (Kodak, Rochester, NY, USA).

Protein expression analysis by Western blot

Protein expression levels of KIT, KRAS, FGF2, FGF7, and IGF1 were further assessed by Western blot analysis. Equal amounts of total protein (30 µg per lane) were resolved by SDS – polyacrylamide gel electrophoresis (SDS – PAGE), Proteins were subsequently transferred to polyvinylidene difluoride (PVDF, 0.45 µm pore size, Thermo Fisher, Waltham, MA, USA; catalog numb 88,518).

Membranes were blocked for 1 h at room temperature in 1× Tris-buffered saline (TBS) containing 0.1% Tween-20 (Sigma-Aldrich, St. Louis, MO, USA; cat. no. 91414) supplemented with 5% nonfat dry milk powder (Merck/Sigma-Aldrich, St. Louis, MO, USA; cat. no. NIST1549A) to reduce nonspecific binding. Primary antibodies against KIT (ab32363), KRAS (ab275876), FGF2 (ab208687), FGF7 (ab131162), IGF1 (ab134140), and the loading control β-actin (ACTB; ab8227) were purchased from Abcam (Cambridge, UK) and used at the manufacturer-recommended dilutions. Membranes were incubated with primary antibodies overnight at 4°C under gentle agitation.

Following three washes in TBST (10 min each), membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary goat anti-rabbit IgG antibody (ab97051; Abcam) for 1 h at room temperature. Immunoreactive bands were visualized using an enhanced chemiluminescence detection system (ECL Prime, GE Healthcare, Little Chalfont, UK) and imaged with a ChemiDoc MP Imaging System (Bio-Rad, Hercules, CA, USA).

Band intensity was quantified using Kodak MI 4.5SE software (KODAK, Rochester, NY, USA).and normalized to the corresponding ACTB signal. All experiments were performed in triplicate, and mean normalized values were used for statistical comparisons between tumor subtypes and control tissues.

Statistical analysis

All statistical computations were performed using Statistica 13.0 PL and the Affymetrix Transcriptome Analysis Console. Data normality was tested with the Shapiro-Wilk test, and statistical significance was set at p < 0.05. One-way analysis of variance (ANOVA), corrected for multiple comparisons using the Benjamini-Hochberg procedure, was employed alongside Tukey’s post hoc test or Student’s t-test for mean comparisons.

To evaluate survival outcomes, Kaplan – Meier survival curves (http://kmplot.com/) [43,44] were plotted, and statistical differences between high and low expression groups were analyzed using the log-rank test. Additionally, hazard ratios (HRs) with 95% confidence intervals (CIs) were calculated to determine the prognostic significance of gene expression on overall survival (OS).

Gene interaction networks were analyzed using the STRING Database 11.0 (accessed March 1, 2025) [38]. Functional enrichment was quantified using log10 (observed/expected) values, with significance determined by the false discovery rate (FDR), adjusted via the Benjamini-Hochberg method. Gene clustering was performed using the Markov Cluster (MCL) algorithm, optimizing the inflation parameter to ensure biological relevance while maintaining pathway-level resolution.

To determine the appropriate sample size, calculations were conducted assuming a 95% confidence interval. Based on national breast cancer incidence data from Poland (2019), a minimum sample size of 377 patients was deemed statistically adequate. The subtype distribution in our cohort was compared with literature-reported prevalence: luminal A (32.10% vs. 23.7%), luminal B HER2-negative (38.8% in both), luminal B HER2-positive (23.7% vs. 14%), HER2-positive (8.89% vs. 11.2%), and TNBC (10.62% vs. 12.3%) [45,46].

Results

Microarray and qRT-PCR profile of common genes for MAPK, SMAD, and dopamine signaling pathways samples in comparison with control tissue

Among the 167 mRNAs common to all three analyzed signaling pathways (−2.0 < FC > 2.0; p < 0.05), the number of genes differentiating each breast cancer type from the control group was as follows: 33 mRNAs for luminal A (LumA), 34 mRNAs for luminal B HER2-positive (LumB+), 32 mRNAs for luminal B HER2-negative (LumB-), 45 mRNAs for HER2-positive (HER2+), and 79 mRNAs for triple-negative breast cancer (TNBC). Additionally, 14 of these mRNAs distinguished tumor-altered samples from controls, regardless of the breast cancer subtype.

A Venn diagram illustrated the genes characteristic of specific breast cancer subtypes or those shared among multiple groups (Figure 1). The transcriptional activity changes of these 11 mRNAs across the five breast cancer subtypes are presented in Table 2.

Figure 1.

Venn diagram of genes differentiating breast cancer subtypes from control tissue.

Analysis was based on transcriptomic data from biological replicates as follows: luminal A (n = 130), luminal B HER2− (n = 100), luminal B HER2+ (n = 96), non-luminal HER2+ (n = 36), TNBC (n = 43), and matched controls. One technical replicate was performed for each biological replicate (one patient). Statistical significance was assessed using one-way ANOVA with Benjamini-Hochberg correction (p < 0.05).

Table 2.

Genes differentiating tumor-altered samples from controls, Independent of breast cancer subtype (−2.0 < FC > 2.0; p < 0.05).

ID	mRNA	Luminal A vs. control	Luminal B HER2− vs. control	Luminal B HER2+ vs. control	Non-luminal HER2+ vs. control	TNBC vs. control
208727_s_at	CDC42	2.82 ± 0.31	3.23 ± 0.44	2.74 ± 0.91	6.45 ± 0.54	12.12 ± 2.13
226400_at		4.79 ± 0.45	4.36 ± 0.43	4.06 ± 0.34	5.78 ± 0.55	5.29 ± 0.14
206254_at	EGF	−2.79 ± 0.78	−4.79 ± 0.51	−3.07 ± 0.42	−3.17 ± 0.91	−2.99 ± 0.18
224999_at	EGFR	−2.62 ± 0.21	−4.98 ± 0.34	−4.30 ± 0.54	−2.35 ± 0.11	−8.65 ± 1.14
201983_s_at		−3.19 ± 0.31	−7.13 ± 0.40	−5.08 ± 0.91	−2.08 ± 0.17	2.30 ± 0.45
202454_s_at	ERBB3	3.37 ± 0.54	3.50 ± 0.43	3.21 ± 0.17	2.11 ± 0.16	−6.97 ± 0.81
204421_s_at	FGF2	−2.48 ± 0.19	−2.91 ± 0.56	−2.74 ± 0.46	−3.22 ± 0.44	−3.87 ± 0.21
204422_s_at		−3.04 ± 0.36	−3.31 ± 0.67	−3.55 ± 0.43	−3.49 ± 0.81	−2.33 ± 0.45
205782_at	FGF7	−2.97 ± 0.61	−3.88 ± 0.71	−4.40 ± 0.81	−4.57 ± 0.51	−7.67 ± 1.12
209540_at	IGF1	−2.86 ± 0.35	−7.09 ± 0.51	−9.40 ± 1.34	−15.26 ± 1.91	−15.12 ± 2.19
205051_s_at	KIT	−3.94 ± 0.41	−6.03 ± 0.48	−11.04 ± 1.98	−8.05 ± 1.11	−12.84 ± 1.71
214352_s_at	KRAS	2.54 ± 0.45	3.05 ± 0.41	3.22 ± 0.12	3.40 ± 0.32	−2.09 ± 0.55
210621_s_at	RASA1	4.95 ± 0.54	4.73 ± 0.48	3.38 ± 0.13	4.40 ± 0.15	2.16 ± 0.19
203085_s_at	TGFB1	4.40 ± 0.56	3.88 ± 0.78	3.69 ± 0.54	2.57 ± 0/57	4.64 ± 0.76

LumA, luminal A; LumB, luminal B; HER2, human epidermal growth factor receptor 2; TNBC, triple-negative breast cancer; C, control; CDC42, Cell Division Cycle 42; EGF, Epidermal Growth Factor; EGFR, Epidermal Growth Factor Receptor; ERBB3, Erb-B2 Receptor Tyrosine Kinase 3; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase; RASA1, RAS p21 Protein Activator 1; TGFB1, Transforming Growth Factor Beta 1. ID, number of the probe; HER2, human epidermal growth factor receptor 2; TNBC, triple-negative breast cancer; C, control; CDC42, Cell Division Cycle 42; EGF, Epidermal Growth Factor; EGFR, Epidermal Growth Factor Receptor; ERBB3, Erb-B2 Receptor Tyrosine Kinase 3; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase; RASA1, RAS p21 Protein Activator 1; TGFB1, Transforming Growth Factor Beta 1.

In the next stage of the study, we conducted qRT-PCR for 11 transcripts that differentiated tumor samples from controls, regardless of the breast cancer subtype, to validate the microarray experiment results. The findings are presented in Figure 2. The table presents the fold change in gene expression levels for 11 genes across five breast cancer subtypes compared to control samples, which served as the basis for the figure. cell division cycle 42 (CDC42) exhibits the most significant upregulation, particularly in TNBC and non-luminal HER2+ subtypes. KRAS proto-oncogene, GTPase (KRAS), RASA1, and transforming growth factor beta 1 (TGFB1) also show consistent overexpression, with RASA1 reaching a fold change of 5.12 in luminal A and 4.88 in luminal B HER2−, while TGFB1 is notably elevated in luminal A and TNBC. ERBB3 is moderately upregulated across most subtypes, particularly in luminal B HER2− (3.5) and luminal A (3.14), whereas KRAS is highest in luminal B HER2+ (3.41) and non-luminal HER2+ (3.4). Conversely, EGF, EGFR, FGF2, FGF7, IGF1, and KIT show lower expression variations, with some values below 1, indicating downregulation. EGF and EGFR remain relatively unchanged or slightly reduced in most subtypes, except for EGFR in TNBC (2.13). FGF2, FGF7, and IGF1 exhibit minimal differences, with FGF7 showing the lowest fold change in TNBC (0.11). Overall, the qRT-PCR data corroborates the microarray findings.

Figure 2.

Expression profile of selected genes obtained by qRT-PCR across breast cancer subtypes and controls.

Data represent mean ± SD from luminal A (n = 130), luminal B HER2− (n = 100), luminal B HER2+ (n = 96), non-luminal HER2+ (n = 36), TNBC (n = 43), and matched controls. One technical replicate was performed per biological replicate (one patient). Expression was normalized to ACTB. One-way ANOVA with Tukey’s post hoc test was used for statistical comparisons (p < 0.05). HER2, human epidermal growth factor receptor 2; TNBC, triple-negative breast cancer; CDC42, Cell Division Cycle 42; EGF, Epidermal Growth Factor; EGFR, Epidermal Growth Factor Receptor; ERBB3, Erb-B2 Receptor Tyrosine Kinase 3; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase; RASA1, RAS p21 Protein Activator 1; TGFB1, Transforming Growth Factor Beta 1.

Prediction of gene expression regulation by miRnas in MAPK, SMAD, and dopamine signaling pathways

Table 3 presents the expression levels of miRNAs potentially involved in the regulation of selected genes, with significant changes (p < 0.05; FC > 2 or < −2) across different breast cancer subtypes compared to controls.

Table 3.

Expression of miRnas potentially involved in the regulation of selected genes (p < 0.05; FC > 2 or < −2).

mRNA	miRNA	Target score	Fold change
			Luminal A vs. control	Luminal B HER2− vs. control	Luminal B HER2+ vs. control	Non-luminal HER2+ vs. control	TNBC vs. control
KIT	miR-221	95	2.34 ± 0.32	2.56 ± 0.32	3.18 ± 0.45	3.45 ± 0.18	4.01 ± 0.18
	miR-222	95	1.87 ± 0.43	2.12 ± 0.54	2.76 ± 0.15	2.98 ± 0.71	3.01 ± 0.19
KRAS	miR-300	90	−2.34 ± 0.17	−2.71 ± 0.51	−3.12 ± 0.43	−2.98 ± 0.65	−3.98 ± 0.51
FGF2 FGF7	miR-16-5p	100 97	4.56 ± 0.18	3.98 ± 0.22	4.37 ± 0.14	4.56 ± 0.98	4.18 ± 0.13
IGF1 KIT	mir-5011-5p	100 97	−2.17 ± 0.15	−2.91 ± 0.32	−2.99 ± 0.51	−3.45 ± 0.65	−4.91 ± 0.18

HER2, human epidermal growth factor receptor 2; TNBC, triple-negative breast cancer; C, control; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase.

KIT is regulated by miR-221 and miR-222, both showing a progressive upregulation, with the highest expression observed in TNBC (4.01 ± 0.18 and 3.01 ± 0.19, respectively). Similarly, FGF2 and FGF7 are targeted by miR-16-5p, which is consistently upregulated across all subtypes, reaching its highest expression in luminal A (4.56 ± 0.18). In contrast, KRAS is regulated by miR-300, which is downregulated across all subtypes, with the most significant suppression in TNBC (−3.98 ± 0.51). Additionally, KIT is also influenced by miR-5011-5p, exhibiting strong downregulation, especially in TNBC (−4.91 ± 0.18).

Concentration of KIT, KRAS, FGF2, FGF7, and IGF1 in breast cancer tissues and control at the protein level

Table 4 presents the concentration levels of selected proteins across different breast cancer subtypes and the control group (p < 0.05), determined using ELISA, highlighting significant alterations in expression patterns.

Table 4.

Concentration of selected proteins in breast cancer subtypes and control group, determined using ELISA (p < 0.05).

Protein	Control	LumA	HER2-negative LumB	HER2-positive LumB	HER2-positive	TNBC
KIT [pg/mL]	1875.23 ± 31.11	below detection threshold*	below detection threshold*	below detection threshold*	below detection threshold*	below detection threshold*
KRAS [ng/mL]	165 ± 13.76	590 ± 28.19*	765.18 ± 67.19*	981.10 ± 90.87*	1323.98 ± 87.14*	432.91 ± 19.91*
FGF2 [pg/mL]	378.19 ± 18.91	165.10 ± 34.123*	160.19 ± 12.34*	67.19 ± 11.01*	54.19 ± 9.18*	50.19 ± 7.65*
FGF7 [ng/mL]	3.23 ± 0.12	2.19 ± 0.18*	1.76 ± 0.19*	1.17 ± 0.11*	below detection threshold*	1.34 ± 0.19*
IGF1[ng/mL]	3.76 ± 0.65	2.19 ± 0.19*	1.21 ± 0.18*	0.98.15*	0.76 ± 0.13*	0.71 ± 0.12*

LumA, luminal A; LumB, luminal B; HER2, human epidermal growth factor receptor 2; TNBC, triple-negative breast cancer; C, control; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase; *p < 0.05 vs. control.

KIT was detected in the control group (1875.23 ± 31.11 pg/mL) but remained below the detection threshold in all breast cancer subtypes, indicating its possible downregulation in tumor tissues. KRAS showed a marked increase in expression across all cancer subtypes, rising from 165 ± 13.76 ng/mL in controls to the highest level in HER2-positive tumors (1323.98 ± 87.14 ng/mL), suggesting its potential role in tumor progression.

FGF2 exhibited a progressive decline, with concentrations dropping from 378.19 ± 18.91 pg/mL in controls to 50.19 ± 7.65 pg/mL in TNBC, indicating significant downregulation in aggressive tumor subtypes. Similarly, FGF7 levels were highest in controls (3.23 ± 0.12 ng/mL) but declined across all cancer subtypes, with the lowest detectable concentration in HER2-positive tumors (1.17 ± 0.11 ng/mL), while being undetectable in HER2-positive breast cancer.

IGF1 levels also decreased across all cancer subtypes, dropping from 3.76 ± 0.65 ng/mL in controls to 0.71 ± 0.12 ng/mL in TNBC, with the sharpest decline observed in HER2-positive (0.76 ± 0.13 ng/mL).

The protein expression patterns detected by Western blot were consistent with the quantitative data obtained via ELISA. Figure 3 presents representative Western blot images demonstrating the specificity of the antibody reaction and sample integrity, as confirmed by the ACTB loading control (molecular weight 42 kDa). Normalized relative to ACTB, the band optical density for KIT (molecular weight 145 kDa) in the control samples was 0.95 ± 0.08, whereas in the tumor samples the signal was either absent or present at levels near the detection limit, with very low OD values, consistent with ELISA results showing KIT levels below the detection threshold. The band optical density for KRAS (molecular weight 21 kDa) in the control samples was 0.13 ± 0.02, while in the tumor samples it was significantly increased: 0.60 ± 0.05 in luminal A, 0.77 ± 0.07 in HER2-negative luminal B, 0.95 ± 0.09 in HER2-positive luminal B, 1.33 ± 0.11 in HER2-positive, and 0.45 ± 0.03 in TNBC The band optical density for FGF2 (molecular weight 18 kDa) in the control samples was 1.10 ± 0.07, while it was significantly reduced in the tumor subtypes: 0.45 ± 0.05 in luminal A, 0.42 ± 0.04 in HER2-negative luminal B, 0.21 ± 0.03 in HER2-positive luminal B, 0.18 ± 0.02 in HER2-positive, and 0.15 ± 0.02 in TNBC The band optical density for FGF7 (molecular weight 28 kDa) in control samples was 1.05 ± 0.06. It progressively decreased in tumor subtypes: 0.72 ± 0.05 in luminal A, 0.61 ± 0.06 in HER2-negative luminal B, 0.41 ± 0.04 in HER2-positive luminal B, and 0.47 ± 0.05 in TNBC. In HER2-positive samples, FGF7 was below the detection threshold The band optical density for IGF1 (molecular weight 10 kDa) in control samples was 1.00 ± 0.08. In tumor subtypes, IGF1 levels were progressively reduced: 0.58 ± 0.06 in luminal A, 0.34 ± 0.05 in HER2-negative luminal B, 0.26 ± 0.04 in HER2-positive luminal B, 0.20 ± 0.03 in HER2-positive, and 0.18 ± 0.02 in TNBC (p < 0.05).

Figure 3.

Representative Western blots of selected proteins across breast cancer subtypes and controls. blots were derived from tissue samples of luminal a (n = 130), luminal B HER2− (n = 100), luminal B HER2+ (n = 96), non-luminal HER2+ (n = 36), TNBC (n = 43), and controls. One technical replicate was performed per biological replicate (one patient). Band intensity was normalized to ACTB. Statistical comparisons used one-way ANOVA and Tukey’s post hoc test (p < 0.05).

LumA, luminal A; LumB, luminal B; HER2, human epidermal growth factor receptor 2; TNBC, triple-negative breast cancer; C, control; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase.

Protein interaction network for common proteins across the three analyzed signaling pathways

The presented protein-protein interaction (PPI) network illustrates the relationships between 11 proteins common to the three analyzed signaling pathways (Figure 4). The network consists of 11 nodes (proteins) and 49 edges (interactions), significantly exceeding the expected number of edges (16). The average node degree is 8.91, indicating that each protein interacts, on average, with nearly nine other proteins in the network. The average local clustering coefficient of 0.915 suggests a highly interconnected network with a strong tendency for proteins to form clusters. The PPI enrichment p-value (9.9e − 12) confirms that the observed interactions are highly significant and unlikely to occur randomly, reinforcing the functional relevance of these proteins within the MAPK, SMAD, and dopamine signaling pathways. Key hub proteins with the highest interaction degrees include EGF (10), EGFR (10), KIT (10), KRAS (10), and CDC42 (9), suggesting their crucial roles in mediating signal transduction across the three pathways. The strong connectivity and clustering within this network point to cooperative interactions among these proteins, which may be vital in cancer-related signaling and potential therapeutic targeting.

Figure 4.

Relationship network for the selected PI3K/AKT/mTOR pathway differentiation genes generated in the STRING database.

CDC42, Cell Division Cycle 42; EGF, Epidermal Growth Factor; EGFR, Epidermal Growth Factor Receptor; ERBB3, Erb-B2 Receptor Tyrosine Kinase 3; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase; RASA1, RAS p21 Protein Activator 1; TGFB1, Transforming Growth Factor Beta 1.

Overall survive analysis

Overall survival analysis was performed for the studied mRNAs (Figures 5–9).

Figure 5.

Overall survival analysis in luminal a breast cancer. (n = 130).

CDC42, Cell Division Cycle 42; EGF, Epidermal Growth Factor; EGFR, Epidermal Growth Factor Receptor; ERBB3, Erb-B2 Receptor Tyrosine Kinase 3; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase; RASA1, RAS p21 Protein Activator 1; TGFB1, Transforming Growth Factor Beta 1.

Figure 6.

Overall survival analysis in luminal B HER2-negative breast cancer (n = 100).

CDC42, Cell Division Cycle 42; EGF, Epidermal Growth Factor; EGFR, Epidermal Growth Factor Receptor; ERBB3, Erb-B2 Receptor Tyrosine Kinase 3; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase; RASA1, RAS p21 Protein Activator 1; TGFB1, Transforming Growth Factor Beta 1.

Figure 7.

Overall survival analysis in luminal B HER2-positive breast cancer (n = 96).

CDC42, Cell Division Cycle 42; EGF, Epidermal Growth Factor; EGFR, Epidermal Growth Factor Receptor; ERBB3, Erb-B2 Receptor Tyrosine Kinase 3; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase; RASA1, RAS p21 Protein Activator 1; TGFB1, Transforming Growth Factor Beta 1.

Figure 8.

Overall survival analysis in non-luminal HER2-positive breast cancer (n = 36).

CDC42, Cell Division Cycle 42; EGF, Epidermal Growth Factor; EGFR, Epidermal Growth Factor Receptor; ERBB3, Erb-B2 Receptor Tyrosine Kinase 3; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase; RASA1, RAS p21 Protein Activator 1; TGFB1, Transforming Growth Factor Beta 1.

Figure 9.

Overall survival analysis in triple-negative breast cancer. (n = 43).

CDC42, Cell Division Cycle 42; EGF, Epidermal Growth Factor; EGFR, Epidermal Growth Factor Receptor; ERBB3, Erb-B2 Receptor Tyrosine Kinase 3; FGF2, Fibroblast Growth Factor 2; FGF7, Fibroblast Growth Factor 7; IGF1, Insulin-Like Growth Factor 1; KIT, KIT Proto-Oncogene, Receptor Tyrosine Kinase; KRAS, KRAS Proto-Oncogene, GTPase; RASA1, RAS p21 Protein Activator 1; TGFB1, Transforming Growth Factor Beta 1.

In luminal A cancer, low levels of KRAS, RASA1 and TGFB1 were associated with worse survival (Figure 5).

On the other hand, poorer survival in luminal B HER2-negative cases was linked with high levels of CDC42, ERBB3, FGF7, TGFB1 and low IGF1 level (Figure 6).

In the case of luminal B HER2-positive cancer, low levels of CDC42, EGF, ERBB3, KIT and high levels of EGFR were associated with worse overall survival (Figure 7).

The analysis also showed that high levels of EGFR, FGF7 and low levels of FGF2, IGF1, RASA1 are associated with worse survival in non-luminal HER2-positive cases (Figure 8).

In addition, analysis revealed a relationship between low levels of FGF2, KRAS and poorer survival in TNBC patients (Figure 9).

Discussion

Breast cancer is a highly heterogeneous disease classified into several molecular subtypes, each exhibiting distinct biological behaviors and clinical outcomes. The primary subtypes include luminal A, luminal B (HER2-negative and HER2-positive), HER2-enriched, and triple-negative breast cancer (TNBC) [47–51]. Luminal A tumors are generally associated with a favorable prognosis, while TNBC and HER2-positive tumors often exhibit more aggressive progression and higher resistance to conventional therapies [3–7].

Despite extensive research into the molecular drivers of breast cancer, a critical gap remains in understanding the intricate interplay between key oncogenic pathways that contribute to tumor heterogeneity, progression, and therapeutic resistance [52–54]. While studies have focused on individual pathways such as the TGF-β/SMAD pathway, MAPK signaling, and dopamine receptor activity, few have investigated their interconnected regulatory networks [25,26,30,55]. This study addresses this knowledge gap by identifying MAPK proteins as crucial molecular bridges between SMAD and dopamine signaling, thereby uncovering novel mechanisms that may underlie breast cancer pathogenesis.

Our findings underscore the central role of the MAPK pathway as a molecular hub integrating signals from multiple oncogenic pathways [56]. The MAPK cascade plays a pivotal role in transducing extracellular signals into intracellular responses that regulate cellular proliferation, differentiation, and survival [56]. Notably, our study reveals that MAPK-related proteins, particularly KRAS, serve as critical mediators of crosstalk between SMAD-driven TGF-β signaling and dopamine receptor activation [57]. The upregulation of KRAS across all breast cancer subtypes, particularly in HER2-positive tumors, reinforces its role in aggressive tumor behavior [58,59]. Moreover, interactions between KRAS, CDC42, TGFB1, and ERBB3 suggest a cooperative mechanism through which MAPK signaling contributes to tumor progression, therapeutic resistance, and potential immune evasion [60]. Interestingly, the overall survival analysis indicated an association of poorer prognosis with low KRAS in patients with luminal A cancer and TNBC. We also noted an association of poorer survival with low CDC42 and ERBB3 levels in luminal B HER2-positive cancer. In the case of luminal B HER2-negative, it was the opposite. These findings highlight the necessity of analyzing multiple interconnected pathways rather than investigating signaling cascades in isolation [61–63].

The significance of integrating multiple pathway analyses is further exemplified by our exploration of the TGF-β/SMAD axis, a pathway with paradoxical roles in cancer [8]. Our findings indicate that key SMAD-associated genes, particularly TGFB1, are upregulated in luminal A and TNBC subtypes, reinforcing their involvement in tumor progression [64,65] Notably, the activation of SMAD proteins appears to converge with MAPK signaling, as evidenced by shared regulatory nodes such as RASA1 and CDC42. These interactions suggest that TGF-β signaling does not operate in isolation but rather integrates with other oncogenic pathways to shape tumor behavior [66]. Understanding these interdependencies is crucial for developing targeted therapeutic strategies that address the multifaceted nature of breast cancer signaling [66].

In addition to protein-coding genes, our study highlights the critical role of miRNAs in post-transcriptional gene regulation within these signaling networks [67]. In breast cancer, miRNAs act as both oncogenic drivers and tumor suppressors, influencing key biological processes such as proliferation, apoptosis, angiogenesis, and metastasis [68,69]. Our findings identify miR-221 and miR-222 as significant regulators of KIT mRNA expression, with their upregulation in TNBC suggesting a potential role in tumor aggressiveness [70,71]. Similarly, miR-16-5p, which targets FGF2 and FGF7, is upregulated across all subtypes, indicating a broader regulatory impact on fibroblast growth factor signaling [72–74]. Conversely, miR-300 is downregulated, leading to increased KRAS expression and subsequent activation of oncogenic MAPK signaling [75]. These findings underscore the intricate regulatory landscape orchestrated by miRNAs, reinforcing their potential as biomarkers and therapeutic targets [76].

Additionally, our study also provides compelling evidence that dopamine receptor signaling, traditionally associated with neurotransmission, plays a functional role in breast cancer pathogenesis [40,77]. The observed downregulation of KIT, IGF1, and FGF family proteins in tumor tissues suggests that dopamine signaling may exert tumor-suppressive effects by modulating inflammatory and growth factor pathways [78–80]. However, the interplay between dopamine receptor activity and MAPK signaling remains complex, as certain dopamine receptor subtypes have been implicated in both pro- and anti-tumorigenic roles depending on the cellular context [81,82]. In addition, our overall survival analysis indicated that low KIR levels were associated with poorer survival in luminal B HER2-positive cases. Similar observations were noted for low IGF1 levels and luminal B HER2-negative cancer. In contrast, high FGF7 levels were associated with poorer survival in HER2-negative cancer and non-luminal HER2-positive cancer, while low FGF2 levels were associated with poorer survival in patients with non-luminal HER2-positive cancer and TNBC. Our findings suggest that dopamine receptor activation may influence breast cancer progression through MAPK-dependent mechanisms, further emphasizing the need to investigate pathway interconnectivity in cancer research.

From a clinical perspective, identifying MAPK proteins as molecular intersections between SMAD and dopamine pathways has profound implications for precision oncology [83–85]. Given the central role of KRAS in mediating oncogenic crosstalk, targeted therapies against KRAS or its downstream effectors may be particularly effective for HER2-positive and TNBC subtypes. Additionally, the observed dysregulation of miRNAs highlights the potential for miRNA-based therapies that modulate key signaling regulators, thereby offering novel avenues for therapeutic intervention [86]. Furthermore, the differential expression of growth factors such as IGF1, FGF2, and FGF7 across breast cancer subtypes suggests that personalized therapeutic approaches targeting specific pathway alterations may enhance treatment efficacy [87].

The differential expression of MAPK-, SMAD-, and dopamine pathway – associated proteins suggests distinct functional consequences in breast cancer pathogenesis [88]. KRAS upregulation, observed across all tumor subtypes, is expected to sustain constitutive activation of MAPK signaling [89], thereby promoting proliferation, survival, and EMT while reducing apoptotic sensitivity [90]. Elevated CDC42 levels likely enhance cytoskeletal reorganization, cell motility, and invasive potential [91]. Increased TGFB1 expression, particularly in luminal A and TNBC, may facilitate extracellular matrix remodeling, immune suppression, and metastasis [92,93]. Conversely, the downregulation of KIT, IGF1, FGF2, and FGF7 could reflect a shift away from normal growth factor – dependent signaling toward alternative oncogenic drivers, potentially altering angiogenesis patterns and tumor – stromal interactions [94]. Collectively, these expression changes indicate a molecular reprogramming that favors tumor progression, metastatic dissemination, and resistance to cell death [95,96].

While dopamine receptor signaling canonically modulates MAPK activity through GPCR – mediated cascades [29,30], our findings suggest that KRAS represents an additional convergence node linking dopamine and SMAD signaling via MAPK [97]. These mechanisms are not mutually exclusive: GPCR activation can initiate upstream signaling that feeds into KRAS-dependent MAPK activation, while KRAS itself can integrate signals from multiple receptor systems, including GPCRs and receptor tyrosine kinases (RTKs) [98,99]. The pronounced upregulation of KRAS thus refines and further substantiates, rather than contradicts, the proposed dopamine – MAPK connection [100].

Despite providing valuable insights into the interplay between MAPK, SMAD, and dopamine signaling pathways in breast cancer, this study has several limitations. The relatively small sample sizes for certain subtypes, such as non-luminal HER2-positive and TNBC, may limit statistical power and the generalizability of our findings. Additionally, while our transcriptomic and bioinformatics analyses identified key molecular interactions, the absence of in vitro and in vivo functional validation restricts our ability to confirm causative relationships. Furthermore, the complex nature of breast cancer signaling suggests that additional pathways and regulatory mechanisms beyond those analyzed in this study may contribute to tumor progression and therapeutic resistance. Future research should incorporate larger patient cohorts, experimental validation, and mechanistic studies to further elucidate the functional roles of these signaling interactions.

This study provides novel insights into the convergence of MAPK, SMAD, and dopamine signaling pathways in breast cancer, supported by multi-layered analyses including transcriptomic profiling, protein quantification, and miRNA expression. The use of matched controls and multiple molecular techniques strengthens the internal validity of the results.

However, several limitations must be acknowledged. First, the observational and cross-sectional design precludes any inference of causality. While we observed significant associations between molecular signatures and clinical subtypes, the study cannot determine whether these alterations are drivers or consequences of tumor progression. Second, the absence of functional assays, such as in vitro knockdown or in vivo validation, limits the mechanistic interpretation of miRNA – mRNA and protein interaction networks. This introduces potential bias regarding the directionality of the observed interactions.

Third, the study cohort was drawn from a single center, which may introduce selection bias and limit the generalizability of the findings. Additionally, some subgroups (e.g. TNBC and non-luminal HER2-positive cases) had relatively small sample sizes, reducing statistical power and increasing the risk of type I or II errors. The magnitude of this limitation is most notable in survival analyses, where underpowered subgroup comparisons may lead to unstable estimates.

Fourth, although ELISA and Western blot analyses were carefully conducted, technical limitations may affect the precision of protein quantification. For example, KIT protein was undetectable in tumors, which could reflect true biological downregulation or assay insensitivity. This uncertainty introduces potential measurement bias. Similarly, predicted miRNA – mRNA interactions were based on in silico tools without experimental validation (e.g. luciferase assays), and confidence scores alone may not reliably distinguish functional regulation from coincidental expression patterns.

Finally, relevant clinical variables – such as treatment regimens, menopausal status, and immune microenvironment – were not fully controlled for, representing possible confounders that may have influenced gene expression or survival outcomes. The direction of such confounding could either amplify or obscure the true effects of the pathways analyzed.

Future studies should include larger, multi-institutional cohorts with more diverse patient profiles, and incorporate experimental validation to confirm the biological relevance of proposed regulatory mechanisms. Addressing these limitations will be critical to translating these molecular findings into clinically actionable biomarkers or therapeutic targets.

Funding Statement

This research received no external funding.

Author contributions

Conceptualization, P.B. and B.O.G; methodology, T.S. and D.B.; formal analysis, T.S.; M.C. and A.S.; resources, P.O.; P.W. and K.B.; data curation, N.Z.; and B.O.G.; writing – original draft preparation, P.B.; T.S.; and R.N.; writing – review and editing, P.B.; T.S.; and B.O.G, supervision, B.O.G; visualization, K.B. and N.D.; project administration, B.O.G. All authors have read and agreed to the published version of the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Institutional review board statement

This study adhered to the ethical principles outlined in the 2013 Declaration of Helsinki regarding research involving human participants. The study protocol received approval from the Bioethical Committee of the Regional Medical Chamber in Krakow (Approval No. 81/KBL/OIL/2023, issued on March 10, 2023).

Informed consent statement

Written informed consent was obtained from all subjects involved in the study.

Data availability statement

All data was included in the paper.

Supplementary Information

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15384101.2025.2549097