Expression pattern and prognostic significance of CDKs in breast cancer: An integrated bioinformatic study

Abstract

BACKGROUND:

Globally, breast cancer (BC) has become one of the most prevalent malignancies and the leading cause of tumor-related deaths among women. Dysregulation of the cell cycle is a well-known hallmark of cancer development and metastasis. CDKs are essential components of the cell-cycle regulatory system with aberrant expression in a variety of cancers, including BC. In the development of targeted cancer treatment, reestablishing the regulation of the cell cycle by modulation of CDKs has emerged as a promising approach.

METHODS:

Herein, we used a bioinformatic approach to assess the expression pattern, prognostic and diagnostic importance, and clinical relevance of CDKs in BC. Additionally, we conducted a functional enrichment analysis of deregulated CDKs using the STRING and KEGG databases to delineate the role of CDKs in breast tumorigenesis.

RESULTS:

Gene expression analysis revealed substantial deregulation of CDKs in BC, with CDK1, CDK11A, and CDK18 showing a fold change of $>\pm$ 1.5. Also, metastatic tumors showed high expression of CDK1 in the single cell RNA sequencing analysis of primary and metastatic breast tumors. Additionally, it was found that dysregulated CDK expression affects overall survival (OS) and relapse-free survival (RFS) of BC patients.

CONCLUSION:

The study’s multimodal analytical methodologies imply that modulating CDKs for BC treatment is a promising approach.

Keywords

Breast cancer CDKs gene expression prognostic factor bioinformatics cell cycle

1. Introduction

Breast cancer (BC) is one of the most common malignancies as well as the leading cause of tumor-related mortality among women globally. In the United States alone, 1700 people will die from cancer each day, with breast, lung, and colorectal cancers accounting for the majority of fatalities in women [1]. BC incidence rates are rising in transitional nations such as South America, Africa, and high-income Asian countries (Japan and the Republic of Korea) due to substantial changes in lifestyle, social, and constructed environments brought about by developing economies [2]. In addition, there has been a rapid rise in BC mortalities in Sub-Saharan African countries, largely due to the underdeveloped health infrastructure and circumstances that result in poor survival [2]. BC is a highly heterogeneous solid tumor with more than 20 distinct subgroups differing in appearance, genetics, and clinical behavior. Based on gene expression, BC has been categorized into luminal A, luminal B, HER2-enriched, and triple-negative breast cancer (TNBC) [3]. TNBC accounts for approximately 20% of all BC cases and is defined by the lack of estrogen (ER) and progesterone (PR) receptors, as well as the lack of amplification or overexpression of human epidermal growth factor receptor 2 (HER2) [3, 4]. Despite advancements in diagnostics and treatment modalities, BC continues to be a problem due to low treatment response and the development of recurring and metastatic cancers [5].

Several novel drugs targeting crucial pathways in cellular development and regulation are being developed and tested in combination treatment to improve the effectiveness of therapeutic regimens and extend the survival of cancer patients [6]. The cell cycle and subsequent cellular development are controlled by a highly coordinated cell cycle control system involving cyclin-dependent kinases (CDKs) and cyclins [7, 8]. Interactions amongst cyclin-dependent kinases (CDKs) and cyclins are critical in maintaining the regulation of cell cycle progression [8]. Loss of control over the cell cycle, leads to unrestricted growth, a classic hallmark of cancer. CDK modulation to reestablish cell cycle control has thus been a promising alternative in the development of targeted cancer therapy [9]. In the present study, we evaluated the expression of CDKs in clinical samples of BC patients by utilizing TCGA BRCA datasets available on the UCSC XENA and Gepia2 Web portals. We report here that the expression of CDKs is significantly deregulated in BC. In addition, deregulated CDK expression was found to affect both overall survival (OS) and relapse free survival (RFS) in patients with BC. Enrichment studies showed that CDKs are essential components of neoplastic pathways and that modulating CDKs in combination with conventional therapies will be a promising approach to treat BC patients.

2. Materials and methods

2.1 Identification of differentially expressed CDKs in breast cancer

The UCSC XENA online database was utilized to analyze the expression pattern of CDK in breast cancer and a heatmap was generated [10]. The Gepia2 database (http://gepia2.cancer-pku.cn/), an online data mining platform, was further used to analyze the fold change in expression of CDKs in BC [11]. The database has 1085 breast tumor samples and 291 GTEx samples. We also utilized the UALCAN database, a comprehensive web resource for analyzing cancer OMICS data, to evaluate the expression of highly deregulated CDKs in different subclasses, age groups, and ethnicities of breast cancer patients [12].

2.2 Kaplan-Meier plotter

The Kaplan-Meier plotter (https://kmplot.com/), is a web resource with gene expression data and information regarding the survival of BC patients [13]. The mRNA gene chip BC dataset of the KM Plotter has RFS data for 4929 and OS data for 1879 patients. The patients were grouped into two cohorts based on the median expression of the gene. The effect of the deregulated genes on OS and RFS of the two defined groups was analyzed by KM-survival curves and the hazard ratio intervals and log-rank $P$ -value were calculated using the KM plotter.

2.3 ROC plotter analysis

To evaluate the correlation between expression levels of the highly deregulated CDKs, viz., CDK1, CDK11A, and CDK18 with sensitivity to endocrine therapy and HER2 directed therapies, the ROC Plotter portal was accessed [14]. The ROC Plotter (http://www.rocplot.org/) is an online transcriptome-level validation tool, capable of linking gene expression with response to therapy using transcriptome-level data of 3,104 breast cancer patients for predictive biomarkers.

2.4 bc-GenEXminer

Breast cancer Gene-Expression Miner V4.5 (bC-GenEXminer v4.5) (http://bcgenex.centregauducheau.fr/BC-GEM) is a web-based bioinformatics tool for annotated breast cancer transcriptomic data [15, 16]. bc-GenEXMiner was used to examine the association of CDK1, CDK11A, and CDK18 expression levels with different clinical parameters of breast tumor patients viz, hormonal status, Her2 enrichment, and SBR grade and p53 status.

Table 1
Log2 fold change of CDKs in breast cancer

Gene symbol	Gene ID	Median (tumor)	Median (normal)	Log2 (fold change)	Adjp
CDK1	ENSG00000170312.15	27.889	3.03	2.842	2.82E-184
CDK2	ENSG00000123374.10	31.111	24.16	0.352	4.31E-15
CDK3	ENSG00000250506.6	2.77	8.43	$-$ 1.323	5.18E-47
CDK4	ENSG00000135446.16	113.881	72.962	0.635	1.49E-52
CDK5	ENSG00000164885.12	19.35	6.32	1.475	2.42E-164
CDK6	ENSG00000105810.9	2.29	3.41	$-$ 0.423	3.04E-09
CDK7	ENSG00000134058.10	32.271	17.12	0.877	5.30E-61
CDK8	ENSG00000132964.11	7.82	6.78	0.181	1.16E-06
CDK9	ENSG00000136807.11	39.341	50.112	$-$ 0.341	3.11E-12
CDK10	ENSG00000185324.21	26.91	54.719	$-$ 0.997	5.51E-36
CDK11A	ENSG00000008128.22	14.93	45.711	$-$ 1.552	1.04E-38
CDK15	ENSG00000138395.14	0.44	0.85	$-$ 0.361	3.08E-16
CDK16	ENSG00000102225.15	76.341	54.651	0.475	1.50E-36
CDK17	ENSG00000059758.7	15.34	20.77	$-$ 0.414	1.03E-05
CDK18	ENSG00000117266.15	2.77	11.92	$-$ 1.777	3.10E-14
CDK19	ENSG00000155111.14	3.93	2.925	0.329	1.29E-03
CDK20	ENSG00000156345.17	6.39	5.33	0.223	5.81E-05

2.5 Gene-gene interactions and Protein-protein interaction analysis

For gene-gene interaction analysis, we utilized the online portal GENEMANIA (https://genemania.org/) to predict the function and association of selected CDKs [17]. The Search Tool for the Retrieval of Interacting Genes (STRING database ver.11.0b) (http://stringdb.org), a biological database designed to construct and study functional interactions between proteins, was utilized to construct a PPI network of deregulated CDKs with a confidence score $\geqslant$ of 0.7 [18]. The PPI network was further analyzed and visualized using Cytoscape software (version 3.8.2) [19]. The plug-in of Molecular Complex Detection (MCODE) in Cytoscape software was put in to get the notable modules in the PPI network [20]. The HUB nodes with the top 5 degrees in the PPI network were obtained using the cytohubba plugin [21].

2.6 Functional enrichment of highly deregulated CDKs

The Mayaan lab online database (https://maayanlab.cloud/Enrichr/) was used for the gene ontology function and KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway enrichment analyses of CDK1, CDK11A, and CDK18 [22]. The GO and KEGG terms with FDR $<$ 0.05 were regarded as significant functions and pathways. We further performed functional analysis of highly deregulated CDKs in BC using available literature.

2.7 CDKs exploration in single cell sequencing database (TME)

The Tumor Immune Single-cell Hub (TISCH) was first processed to visualize significantly deregulated CDK expression across several datasets in BC tissues at the single-cell level [23]. We investigated expression of CDK1, CDK11A, and CDK18 in primary and metastatic breast cancer datasets to determine the relationship between CDK expression patterns and tumor progression.

3. Results

3.1 Expression of CDKs is deregulated in BC

The expression pattern of CDKs in BC was examined using UCSC XENA. CDK expression analysis in the TCGA BrCa dataset revealed significant CDK deregulation in BC. Figure 1. Further analysis using the Gepia2 database revealed the log2 fold change of different CDKs in BC Table 1. CDK1 was found to be highly upregulated, with a log2 fold change of 2.84 and a p-value of 2.82E-184, whereas CDK11A and CDK18 were found to be downregulated, with log2 fold changes of $-$ 1.552 and $-$ 1.77, respectively. Figure 2. The expression pattern of highly deregulated CDKs among different BC subclasses, age groups, and ethnicities was analyzed using UALCAN, and it was found that TNBC patients express high levels of CDK1, followed by Her2 enriched and luminal subtypes. Also, CDK1 was found highly upregulated in women under the age group of 20–40 yrs and African-American ethnicity Fig. 3. CDK18 was found to be highly downregulated in HER2 enriched breast tumors and Asian women. CDK11A, however, didn’t show significant results in expression patterns among different BC subclasses, age groups, and ethnicities.

Figure 1.

mRNA expression pattern of CDKs in breast cancer patients. Heat map displaying the expression patterns of CDKs. XENA was used to extract CDK expression. Comparing expression of CDKs amongst GTEx (normal breast, $n=$ 291), Solid Tissue Normal (tissue surrounding the tumor ( $n=$ 114), and primary tumor (cancer, $n=$ 1097).

Figure 2.

Expression pattern of CDKs in Breast cancer. A. Box Plots of the expression pattern of CDKs analyzed using TCGA BRCA and GTEx datasets using UCSC XENA. B. Expression profile of highly deregulated CDK1, C, CDK11A, and D, CDK18 in BC using Gepia2 database $p$ value $>$ 0.05.

Figure 3.

Expression pattern of highly deregulated CDKs in BC subtypes, ethnicity, and age groups. A. CDK1 showed high expression in TNBC patients with high statistical difference, while as CDK18 showed significant deregulated expression between in luminal and TNBC groups but lacked difference in HER2 enriched BC patients. CDK11A lacked statistical significance in expression with BC subtypes. B. CDK1 and CDK18 showed low median expression in Caucasian women, with CDK18 highly expressed in African-American women ( $p$ -value $>$ 0.0001). Low statistical significance was found in expression of CDK11A within races. C. For age, CDK1 showed high median expression in women under the age group of 21–40 y ( $p$ -value $>$ 0.0001), while CDK18 showed low median expression in women under the age group of 61–80 y. CDK11A didn’t show a significant difference in mRNA levels with age ( $p$ -value $>$ 0.08).

Additionally, we explored the expression pattern of deregulated CDKs in TISCH database, a single sequence database focusing on tumor microenvironment. We performed expression pattern analysis in two BrCa dataset with primary and metastatic tumors. The study revealed that CDK1 is upregulated in metastatic tumors compared to primary tumors. Moreover, CDK1 expression showed heterogeneity in expression pattern across the metastatic tumor cell population. CDK11A and CDK18 showed low expression, however stromal cells in the TME of primary tumor showed high expression of CDK11A compared to CDK18 Fig. 4.

Figure 4.

Expression pattern of deregulated CDKs in primary and metastatic tumors using single cell RNA-seq database TISCH.

3.2 Deregulation of CDKs is associated with a worse prognosis in breast cancer patients

The KM Plotter, which contains gene expression information as well as the survival probability of breast cancer patients, was used to investigate the survival probability of highly deregulated CDKs such as CDK1, CDK11A, and CDK18. BC patients were divided into two cohorts based on median expression, with the two groups termed as low and high expression groups. The relapse-free survival (RFS) analysis was performed on 4929 BC patients for 300 months, and it was found that BC patients with low CDK1 expression have better RFS than patients with higher CDK1 mRNA levels, with a hazard ratio of 1.68 and a log-rank p-value of 1e-16 (Fig. 5a). Furthermore, BC patients with high CDK11A and CDK18 expression had better RFS survival, with HR values of 0.69 and 0.8, respectively. The correlation between overall survival (OS) and CDK expression was investigated in 1879 BC patients, and it was discovered that BC patients with high mRNA levels of the CDK1 gene have poor overall survival compared to patients with low survival, with a hazard ratio (HR) of 1.44 and a log-rank p-value of 0.00016 (Fig. 5b). Low expression of CDK11A and CDK18, on the other hand, is associated with a worse OS in BC patients, with HRs of 0.76 and 0.84, respectively.

Figure 5.

Expression of CDKs correlates with survival outcomes. A. Correlation between expression of CDKs with relapse-free survival (RFS) in BC patients. B. Correlation between expression of CDKs with overall survival in BC patients.

Figure 6.

Diagnostic analysis of deregulated CDKs in BC receiving endocrine therapy. A. ROC plot for the Pathological complete response of breast cancer patients receiving endocrine therapy. B. ROC plot for RFS of breast cancer patients receiving endocrine therapy.

3.3 CDKs’ potential diagnostic significance in BC

Receiver operating characteristic (ROC) curve analysis revealed several highly predictive values for deregulated CDKs. CDK1 was shown to be an extremely sensitive biomarker for individuals receiving endocrine therapy. On the other hand, CDK11A and CDK18 exhibited limited specificity and sensitivity as biomarkers in patients taking endocrine therapies Fig. 6. On the other hand, patients with high CDK18 expression reacted better to hormonal therapy than patients with low expression. Additionally, CDKs revealed increased sensitivity and specificity as prognostic biomarkers in patients with breast cancer following HER2-directed treatments Fig. 7. Moreover, BC patients receiving HER2-directed therapies with low CDK1 expression had a high pathological complete response with a specificity of 62% and sensitivity of 65 %. CDK11A was found to be less sensitive as a diagnostic marker. CDK18, on the other hand, predicted both pathological complete response and relapse-free survival in BC patients receiving anti-HER2 therapy with high sensitivity and specificity. Besides that, BC patients with high CDK18 expression responded better to anti-HER2 therapy.

Figure 7.

Diagnostic analysis of deregulated CDKs in BC receiving anti-HER2 therapy. A ROC plot for the pathological complete response of breast cancer patients receiving anti-HER2 therapy. B. ROC plot for RFS of breast cancer patients receiving anti-HER2 therapy.

3.4 Correlation of deregulated CDKs with clinical-physiological characteristics in BC

The bc-GenEXMiner database was used to examine the association between severely deregulated CDKs, especially CDK1, CDK11A, and CDK18, and common clinical-pathological characteristics in breast cancer patients. CDK1 expression was shown to be significantly higher in breast cancers lacking hormone receptors (ER and PR) ( $p$ value-0.0001). CDK1 expression, on the other hand, was enhanced in HER2-enriched breast cancers ( $p$ value-0.0001). In terms of SBR grade, it was discovered that increased CDK1 expression was connected with SBR3. Additionally, individuals with mutant p53 had significantly greater CDK1 mRNA levels than patients with wild-type p53 (Fig. 8).

Figure 8.

Clinicopathological parameters and CDKs in breast cancer. bC-GenEXminer analysis of deregulated CDKs showed a high association with clinicopathological parameters of BC. A. CDK1, B. CDK11A, C. CDK18.

CDK11A expression was considerably lower in patients with ER-negative breast cancer and much elevated in tumors enriched in HER2. CDK11A expression, however, did not vary substantially from PR, SBR grade, or p53 status. CDK18, on the other hand, was shown to be significantly downregulated in patients with breast cancer who expressed hormone receptors (ER, & PR), demonstrating a strong relationship with hormone receptor negativity. CDK18 mRNA levels were discovered to be low in BC patients with HER2 amplification and were observed to correlate with SBR3 grade. In BC patients with wild type 53, CDK18 expression was shown to be substantially linked with p53 status.

3.5 PPI and GGI networks of CDKs

The PPI network was constructed using the STRING database and the confidence score of the interactions was set as $>$ 0.7. The constructed network had 47 nodes and 461 edges, with an average node degree of 28 and a PPI enrichment $p$ -value $<$ 1.0e-16. The Cytoscape plugin, MCODE, was further utilized to obtain the significant module, and MCODE analysis revealed 31 nodes and 394 edges in the significant module of the constructed network Fig. 9A.

Figure 9.

Protein-protein interaction and Gene-gene interaction analysis of deregulated CDKs. A. Protein-protein interaction (PPI) network of CDK1 and CDK11A gene constructed with STRING tool had. B. MCODE plugin analysis created a significant module in the constructed network with 31 nodes and 394 edges. C. The top 5 hub genes were identified using the cytoHubba tool in Cytoscape software. D. Gene-gene interaction network of deregulated CDKs.

CDK11A formed a significant PPI network with 14 nodes and 30 edges. CDK11A showed direct interactions with LEO1, a component of the PAF1 complex (PAF1C), which has multiple functions during transcription by RNA polymerase II and is implicated in the regulation of development and maintenance of embryonic stem cell pluripotency. CDK11A, also showed linkage with CCNL1 (Cyclin-L1), involved in pre-mRNA splicing and a prime candidate proto-oncogene in head and neck squamous cell carcinomas. CDK11A however didn’t show any interactions with CDK1 or CDK18, indicating that CDK11A acts independently of these two CDKs Fig. 9B.

Furthermore, using the cytohubba plugin, the top five genes regarded as hub genes in the constructed PPI network based on degree were found out to be Aurora kinase A (AURKA), Cyclin A2 (CCNA2), CDK2, CCNB1, and CDK1. The advanced options set as degree cut-off $=$ 2, K-Core $=$ 2, and Node Score Cut-off $=$ 0.2.

We further analyzed gene-gene interactions of CDK1, CDK11A, and CDK18, and it was found that deregulated CDKS show high interactions with wee1, CDC25C, CCNB1, and other genes involved in the regulation of the cell cycle. Gene-gene interaction analysis further revealed that deregulated CDKs are central to serine/threonine protein kinase complex, mitotic cell cycle checkpoint, DNA damage checkpoint and signal transduction by p53 class mediator.

3.6 Gene ontology enrichment and pathways analysis

Using the Enrichr database, we identified signaling pathways and gene ontology features related to CDK1. GO analysis revealed that CDK1 was highly enriched in biological processes among the three GO terms. The CDK1 gene was found to be enriched in golgi disassembly, positive regulation of the G2/M transition of the mitotic cell cycle, regulation of glial cells, regulation of embryonic development, and other biological processes Fig. 10A. In molecular function (MF) terms, CDK1 showed enrichment in histone kinase activity, protein kinase activity, phosphotransferase activity, RNA Poly II carboxy-terminal domain kinase activity, etc. Fig. 10B. In the cellular compartment group, the suggested GO was primarily enriched in spindle microtubule, mitotic spindle, nuclear chromosome part, centrosome, etc. Fig. 10C.

Figure 10.

Gene enrichment and pathway analysis. (A)–(C) Gene Ontology analysis of the BP, CC, and MF and, (D) KEGG pathway enrichment analysis of CDK1, CDK11, and CDK18 genes in breast cancer.

For pathway studies, the KEGG database was evaluated. According to KEGG Human 2019, CDK1 is associated with the cell cycle, P53 signalling pathway, progesterone mediated oocyte maturation, cellular senescence, and cell death pathways Fig. 10D.

Figure 11.

Literature-based functional analysis of CDK1 in BC.

Additionally, a literature-based functional analysis of the deregulated CDKs demonstrated that CDKs regulate key aspects of breast cancer growth and metastasis Fig. 11. CDK1 was found to be central to tumor cell proliferation. Cellular knockdown of RBM7 (RNA binding motif protein 7) reduced CDK1, resulting in decreased cell proliferation and cell cycle arrest in the G1 phase. Besides, circular RNAs have been reported to regulate CDK1 activity in BC. circ-Ccnb1 was found interacting with CCNB1 and CDK1 complex and inducing dissociation of the CCNB1-CDK1 complex, resulting in reduced cell migration and invasion potential of tumor cells. RO3306, a CDK1 inhibitor, significantly reduced CDK1, PDK1, and $\beta$ -catenin expression and inhibited the epithelial to mesenchymal transition in tumor cells. Additionally, treatment with RO3306 reduced sorafenib resistance in resistant models. A literature-based survey further revealed that TP53 and TP73 regulate the expression of the negative regulator of CDK1, p21 ${}^{\text{CIP1}}$ . Loss of p21 ${}^{\text{CIP1}}$ resulted in enhanced CDK1 activity and induction of whole chromosome instability (W-CIN) mediated by CDK1. Furthermore, mi-RNAs were also found to be central to CDK1 and CDK18 regulation in BC.

4. Discussion

Breast cancer is the leading cause of tumor-related deaths in women and one of the most common incident malignancies [2]. Though advances in early detection and advances in therapies have improved the OS and RFS of breast tumor patients, the development of metastatic and resistant tumors requires urgent focus [24, 25]. Also, tumors are complex and don’t merely comprise the highly proliferating tumor cells, but also the surrounding cellular and non-cellular components of the tumor defined as the tumor microenvironment (TME) [26, 27]. Tumor cells and the surrounding stroma influence each other through secreted factors and physical interaction and enhance tumor progression, metastasis, and therapeutic resistance [27]. The stromal cells have been found to induce epigenetic changes in the tumor cells and influence gene expression, aiding in tumor progression and affecting prognosis [28].

The cell cycle is an evolutionarily conserved process, a requisite and regulated phenomenon in mammalian cell growth and development. Abnormalities in the cell-cycle control system are one of the hallmarks of human cancers, including BC [29]. Several studies have analyzed the role of CDKs in controlling the growth, development, and survival outcomes of tumor patients. However, a deep understanding of the role of CDKs in tumor biology is needed to develop rationale therapies for the betterment of cancer patients [30]. In the present study, we employed a bioinformatic approach to unravel the expression pattern, prognostic and diagnostic significance, and functional role of the deregulated CDKs in breast tumor patients. Our analysis revealed that CDKs are highly deregulated in BC, with CDK1, CDK11A, and CDK18 deregulated with a log2 FC of $\pm$ 1.5.

The CDK1 gene was found to be substantially expressed in individuals with BC, and prognostic analysis indicated that increased CDK1 expression was associated with a worse OS and RFS. These data suggest that CDK1 plays a critical role in tumor development promotion. Interestingly, patients with reduced CDK1 expression had a favorable response to endocrine and anti-HER2 treatments, indicating that CDK1 has a high diagnostic potential. Further investigation established a link between CDK1 and the HER2 receptor, since patients with elevated levels of HER2 expressed a high level of CDK1. Additionally, HER2 $+$ patients with high CDK1 expression responded poorly to anti-HER2 therapy, suggesting a function for CDK1 in protecting tumor cells against HER2 therapies. CDK1 overexpression was also associated with a high SBR3 grade and a p53 mutation. CDK1 was also identified with gene-gene interactions and PPI networks, demonstrating that CDK1 plays a critical role in the control of cell proliferation. CDKs, notably CDK1, have been reported to be dysregulated in a range of malignancies, including breast cancer [31]. CDK1 is a catalytic subunit of a protein kinase complex, responsible for controlling the transition from G1 to S phase and from G2 to M phase of the cell cycle [31]. As various physiological processes like apoptosis, differentiation, and ageing are regulated by the mediators of the cell cycle control system, targeting the CDKs is an effective means for the treatment of breast cancer. Thus, inhibition of CDK1 can potentially be used to inhibit tumor cell growth in combination with other anticancer approaches [32, 33]. In epithelial ovarian cancer, CDK1 was found to regulate the cell cycle and proliferation, and its high expression correlated with a worse prognosis. Furthermore, inhibiting CDK1 expression improved tumor cell response to chemotherapeutics [34]. Also, a recent study revealed that dinaciclib-mediated inhibition of CDK1 expression induced cell cycle arrest in lung cancer [35]. Ding and colleagues concluded that the five hub genes: CDK1, CCNB2, ACACB, PPARG, and MAD2L1 are highly involved in the regulation and thus can be used as diagnostic markers in ductal carcinoma in situ (DCIS) [36]; thus, suggesting that there is an association of CDK1 with breast cancer. Interestingly, inhibition of CDK1 has been regarded as a worthy anticipating biomarker to evaluate the efficacy of anti-cancer therapy, as is evident from ROC plot curves (Figs 5 and 6) for BC patients receiving endocrine and anti-HER2 therapies [31].

In contrary to CDK1, CDK11A and CDK18 were considerably downregulated in patients with breast cancer, with both being highly downregulated in HER2 $+$ women. Additionally, increased CDK11A and CDK18 expression was associated with improved OS and RFS, indicating that CDK11A and CDK18 inhibit tumor development and progression and prolong the survival of BC patients. Likewise, ROC plots suggested that BC patients with high CDK11A and CDK18 expression react more favorably to hormone therapy and achieve a higher pCR. The findings for CDK11A, on the other hand, were less significant ( $p>$ 0.05).

Furthermore, a literature-based survey demonstrated a key role for CDK1 in breast tumorigenicity. For example, miR-424 downregulated in BC was found to be central to CDK1 regulation. Upregulation of miR-424 induces cell cycle arrest and regulates CDK1 expression via ERK1/2 signaling. The expression of CDK1 was found downregulated in cells overexpressing miR-424 with reduced cell survival, demonstrating that targeting CDK1 is a promising candidate in breast tumorigenicity [37]. Also, inhibiting CDK1 with R03306 enhanced sensitivity to sorafenib in resistant models with reduced EMT potential [38].

Additionally, TP53 and TP73 act as tumor suppressors and induce expression of the CDK1 inhibitor, p21 ${}^{\text{CIP1}}$ . Loss of p21 ${}^{\text{CIP1}}$ due to loss of TP53 and TP73 results in enhanced CDK1 activity and induction of whole chromosome instability (W-CIN) mediated by CDK1 [39]. Also, it was observed that cells with induced CDK1 expression showed abnormal microtubule assembly rates and W-CIN. Another study demonstrated that the knockdown of RBM7 (RNA binding motif protein 7) reduced CDK1, resulting in decreased cell proliferation and cell cycle arrest in the G1 phase, indicating RBM7 as a key target in regulating CDK1 expression [40]. Besides, circular RNAs have been reported to regulate CDK1 activity in BC. circ-CCNB1 was found interacting with CCNB1 and the CDK1 complex, and inducing dissociation of the CCNB1-CDK1 complex, resulting in reduced cell migration and invasion potential of tumor cells [41]. Besides, delivery of circ-CCNB1 inhibited the growth of tumor in a mice model and prolonged survival, demonstrating that targeting the CDK1-CCNB1 axis via circ-CCNB1 may be a promising therapeutic approach in addition to conventional treatments.

In terms of clinical characteristics, CDK11A showed association with ER and HER2 receptor with ER $+$ patients and HER2 $+$ patients expressing high mRNA levels, correlating with low-grade tumors. A recent study found that CDK11 is an anti-metastatic gene in ER-positive breast cancer and that CDK11’s control of integrin $\beta$ 3 via ER signaling suppression may be part of a signaling cascade driving breast cancer invasion [42].

BC patients with high expression of CDK18 receiving anti-HER2 therapy showed a high response and had better pCR and RFS and the results were highly significant with $p<$ 0.0001. Moreover, CDK18 showed high sensitivity to anti-HER2 therapy response, indicating that the expression of CDK18 can be used as a diagnostic biomarker in BC patients undergoing HER2-directed therapies. Also, the bc-GenEXMiner analysis further revealed that ER – ve BC patients express high levels of CDK18, which may be responsible for the high sensitivity of ER – ve tumors to chemotherapy. These results are as per the findings of a recent study that demonstrated that decreased CDK18 expression was linked to worse patient survival. A higher level of CDK18 protein expression may suggest a better response to chemotherapy. These findings imply that CDK18 mRNA and protein levels are controlled differentially inside cancer cells, with possibly divergent implications for patient outcomes [43].

Previously, CDK18 was found to interact with RAD9, a part of the 9-1-1 replication stress signaling complex [3]. This is a critical axis for establishing an appropriate cellular response to replication stress and, by extension, to chemotherapy [44]. CDK18 expression may impact tumor cell biological responses to chemotherapy. The authors used CRISPR activation to enhance endogenous CDK18 to better understand the function of CDK18 amplification in breast cancer. Cells were more prone to accumulating DNA damage after CDK18 amplification, as seen by staining with a biomarker of double-strand breaks, -H2AX. Importantly, -H2AX activation was pan-nuclear, indicating a widespread problem with DNA replication. Furthermore, the response to replication stress caused by restricting nucleotides was reduced in cells expressing high amounts of CDK18. The scientists discovered a defect in ATR signaling activation, as well as an increase in susceptibility to DNA damaging compounds that impede nucleotide synthesis, such as methotrexate or 5-FU [43, 45]. A literature-based survey further revealed that CDK18 is regulated by miR-4732-5p with enhanced expression at metastatic sites such as lymph nodes, indicating that CDK18 at metastatic sites may regulate the migratory potential of breast tumor cells [46].

Together, these findings provide a picture of CDKs in breast cancer, but elucidating the molecular mechanisms involved in CDK regulation will aid in establishing the rationale for CDK modulation in breast cancer research.

5. Conclusion

In conclusion, our study found that CDK1, CDK11A and CDK18 may serve as possible predictors of OS and RFS in patients with breast cancer following hormonal or anti-HER2 therapy. Additionally, although modulating deregulated CDKs may be a potential strategy for treating BC patients, well-designed prospective pre-clinical and clinical studies are required to elucidate the molecular mechanism behind CDK differential expression and its function in BC.

Author contributions

Conception: MAM.

Interpretation or analysis of data: UM and SS.

Preparation of the manuscript: UM.

Revision for important intellectual content: MAM and BA.

Supervision: MAM.

Funding

This work was funded by the JK Science Technology & Innovation council DST, Govt. of J&K, India with grant No. JKST&IC/SRE/885-87 sanctioned to Dr. Manzoor Ahmad Mir.

Footnotes

Acknowledgments

Mr. Umar Mehraj is a recipient of a Senior research fellowship (SRF) from UGC-CSIR, Govt. of India.

Conflict of interest

The authors declare no conflicts of interest.

References

Siegel

R.L.

Miller

K.D.

Fuchs

H.E.

and Jemal

, Cancer statistics, 2021, CA: A Cancer Journal for Clinicians 71 (2021), 7–33.

Sung

Ferlay

Siegel

R.L.

Laversanne

Soerjomataram

Jemal

and Bray

, Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries, CA: A Cancer Journal for Clinicians 71 (2021), 209–249.

Koboldt

Fulton

McLellan

Schmidt

Kalicki-Veizer

McMichael

Fulton

Dooling

Ding

and Mardis

, Comprehensive molecular portraits of human breast tumours, Nature 490 (2012), 61–70.

Yin

Duan

J.-J.

Bian

X.-W.

and Yu

S.-c.

, Triple-negative breast cancer molecular subtyping and treatment progress, Breast Cancer Research 22 (2020), 1–13.

Waks

A.G.

and Winer

E.P.

, Breast cancer treatment: A review, Jama 321 (2019), 288–300.

Rozeboom

Dey

and De

, ER+ metastatic breast cancer: Past, present, and a prescription for an apoptosis-targeted future, American Journal of Cancer Research 9 (2019), 2821.

Spring

L.M.

Wander

S.A.

Zangardi

and Bardia

, CDK 4/6 inhibitors in breast cancer: Current controversies and future directions, Current Oncology Reports 21 (2019), 1–9.

Malumbres

and Barbacid

, Cell cycle, CDKs and cancer: A changing paradigm, Nature Reviews Cancer 9 (2009), 153–166.

Mayer

E.L.

, Targeting breast cancer with CDK inhibitors, Current Oncology Reports 17 (2015), 20.

10.

Goldman

M.J.

Craft

Hastie

Repečka

McDade

Kamath

Banerjee

Luo

Rogers

Brooks

A.N.

Zhu

and Haussler

, Visualizing and interpreting cancer genomics data via the Xena platform, Nat Biotechnol 38 (2020), 675–678.

11.

Tang

Kang

Gao

and Zhang

, GEPIA: A web server for cancer and normal gene expression profiling and interactive analyses, Nucleic Acids Res 45 (2017), W98–w102.

12.

Chandrashekar

D.S.

Bashel

Balasubramanya

S.A.H.

Creighton

C.J.

Ponce-Rodriguez

Chakravarthi

and Varambally

, UALCAN: A portal for facilitating tumor subgroup gene expression and survival analyses, Neoplasia 19 (2017), 649–658.

13.

Györffy

Lanczky

Eklund

A.C.

Denkert

Budczies

and Szallasi

, An online survival analysis tool to rapidly assess the effect of 22,277 genes on breast cancer prognosis using microarray data of 1,809 patients, Breast Cancer Res Treat 123 (2010), 725–731.

14.

Fekete

J.T.

and Györffy

, ROCplot. org: Validating predictive biomarkers of chemotherapy/hormonal therapy/anti-HER2 therapy using transcriptomic data of 3,104 breast cancer patients, International Journal of Cancer 145 (2019), 3140–3151.

15.

Jézéquel

Campone

Gouraud

Guérin-Charbonnel

Leux

Ricolleau

and Campion

, bc-GenExMiner: An easy-to-use online platform for gene prognostic analyses in breast cancer, Breast Cancer Research and Treatment 131 (2012), 765–775.

16.

Jézéquel

Frénel

J.-S.

Campion

Guérin-Charbonnel

Gouraud

Ricolleau

and Campone

, bc-GenExMiner 3.0: New mining module computes breast cancer gene expression correlation analyses, Database 2013 (2013).

17.

Warde-Farley

Donaldson

S.L.

Comes

Zuberi

Badrawi

Chao

Franz

Grouios

Kazi

and Lopes

C.T.

, The GeneMANIA prediction server: Biological network integration for gene prioritization and predicting gene function, Nucleic Acids Research 38 (2010), W214–W220.

18.

Szklarczyk

Franceschini

Wyder

Forslund

Heller

Huerta-Cepas

Simonovic

Roth

Santos

and Tsafou

K.P.

, STRING v10: Protein-protein interaction networks, integrated over the tree of life, Nucleic Acids Research 43 (2015), D447–D452.

19.

Smoot

M.E.

Ono

Ruscheinski

Wang

P.-L.

and Ideker

, Cytoscape 2.8: New features for data integration and network visualization, Bioinformatics 27 (2011), 431–432.

20.

Bader

G.D.

and Hogue

C.W.V.

, An automated method for finding molecular complexes in large protein interaction networks, BMC Bioinformatics 4 (2003), 1–27.

21.

Chin

C.-H.

Chen

S.-H.

H.-H.

C.-W.

M.-T.

and Lin

C.-Y.

, cytoHubba: Identifying hub objects and sub-networks from complex interactome, BMC Systems Biology 8 (2014), 1–7.

22.

Kuleshov

M.V.

Jones

M.R.

Rouillard

A.D.

Fernandez

N.F.

Duan

Wang

Koplev

Jenkins

S.L.

Jagodnik

K.M.

and Lachmann

, Enrichr: A comprehensive gene set enrichment analysis web server 2016 update, Nucleic Acids Research 44 (2016), W90–W97.

23.

Sun

Wang

Han

Dong

Zheng

Shi

Wang

and Ren

, TISCH: A comprehensive web resource enabling interactive single-cell transcriptome visualization of tumor microenvironment, Nucleic Acids Research 49 (2021), D1420–D1430.

24.

Anastasiadi

Lianos

G.D.

Ignatiadou

Harissis

H.V.

and Mitsis

, Breast cancer in young women: An overview, Updates in Surgery 69 (2017), 313–317.

25.

Benson

J.R.

and Jatoi

, The global breast cancer burden, Future Oncology 8 (2012), 697–702.

26.

Mehraj

Dar

A.H.

Wani

N.A.

and Mir

M.A.

, Tumor microenvironment promotes breast cancer chemoresistance, Cancer Chemotherapy and Pharmacology (2021), 1–12.

27.

Mehraj

Ganai

R.A.

Macha

M.A.

Hamid

Zargar

M.A.

Bhat

A.A.

Nasser

M.W.

Haris

Batra

S.K.

Alshehri

Al-Baradie

R.S.

Mir

M.A.

and Wani

N.A.

, The tumor microenvironment as driver of stemness and therapeutic resistance in breast cancer: New challenges and therapeutic opportunities, Cellular Oncology (2021).

28.

Mehraj

Qayoom

and Mir

M.A.

, Prognostic significance and targeting tumor-associated macrophages in cancer: New insights and future perspectives, Breast Cancer (2021), 1–17.

29.

Çoban

E.A.

Tecimel

Şahin

and Deniz

A.A.H.

, Targeting cancer metabolism and cell cycle by plant-derived compounds, Cell Biology and Translational Medicine 8 (2019), 125–134.

30.

Chou

Quigley

D.A.

Robinson

T.M.

Feng

F.Y.

and Ashworth

, Transcription-associated cyclin-dependent kinases as targets and biomarkers for cancer therapy, Cancer discovery 10 (2020), 351–370.

31.

Izadi

Nikkhoo

Hojjat-Farsangi

Namdar

Azizi

Mohammadi

Yousefi

and Jadidi-Niaragh

, CDK1 in breast cancer: Implications for theranostic potential, Anti-Cancer Agents in Medicinal Chemistry (Formerly Current Medicinal Chemistry-Anti-Cancer Agents) 20 (2020), 758–767.

32.

Asplund

Edqvist

P.H.D.

Schwenk

J.M.

and Pontén

, Antibodies for profiling the human proteome – The H uman P rotein A tlas as a resource for cancer research, Proteomics 12 (2012), 2067–2077.

33.

Ahmed

, Integrated network analysis reveals FOXM1 and MYBL2 as key regulators of cell proliferation in non-small cell lung cancer, Frontiers in Oncology 9 (2019), 1011.

34.

Huang

Wang

Zhong

Liu

Jiang

Wang

Fang

and Yang

, The expression of CDK1 is associated with proliferation and can be a prognostic factor in epithelial ovarian cancer, Tumor Biology 36 (2015), 4939–4948.

35.

Danilov

A.V.

Orr

Godek

Mustachio

L.M.

Sekula

Liu

Kawakami

Johnson

F.M.

and Compton

D.A.

, Dinaciclib induces anaphase catastrophe in lung cancer cells via inhibition of cyclin-dependent kinases 1 and 2, Molecular Cancer Therapeutics 15 (2016), 2758–2766.

36.

Ding

Z.-H.

Shang

A.-Q.

Zhang

Y.-J.

Wei

L.-Q.

Wang

W.-W.

and Yang

, Docking of CDK1 with antibiotic drugs revealed novel therapeutic value in breast ductal cancer in situ, Oncotarget 8 (2017), 61998.

37.

Xie

Song

Hua

Zhao

and Ji

, MicroRNA-424 serves an anti-oncogenic role by targeting cyclin-dependent kinase 1 in breast cancer cells, Oncology Reports 40 (2018), 3416–3426.

38.

C.X.

Wang

X.Q.

Chok

S.H.

Man

Tsang

S.H.Y.

Chan

A.C.Y.

K.W.

Xia

and Cheung

T.T.

, Blocking CDK1/PDK1/β-Catenin signaling by CDK1 inhibitor RO3306 increased the efficacy of sorafenib treatment by targeting cancer stem cells in a preclinical model of hepatocellular carcinoma, Theranostics 8 (2018), 3737.

39.

Schmidt

A.-K.

Pudelko

Boekenkamp

J.-E.

Berger

Kschischo

and Bastians

, The p53/p73-p21 CIP1 tumor suppressor axis guards against chromosomal instability by restraining CDK1 in human cancer cells, Oncogene 40 (2021), 436–451.

40.

P.-W.

Zhang

Zhu

Dai

X.-Y.

Cheng

Shi

Wei

J.-F.

and Ding

, Oncogenic action of the exosome cofactor RBM7 by stabilization of CDK1 mRNA in breast cancer, NPJ Breast Cancer 6 (2020), 1–10.

41.

Fang

W.W.

Awan

F.M.

Dong

and Yang

B.B.

, The circular RNA circ-Ccnb1 dissociates Ccnb1/Cdk1 complex suppressing cell invasion and tumorigenesis, Cancer Letters 459 (2019), 216–226.

42.

Chi

Huang

Wang

Zhou

Wang

Xiao

Cai

Zhou

and Wu

, CDK11 p58 inhibits ERα-positive breast cancer invasion by targeting integrin β3 via the repression of ERα signaling, Bmc Cancer 14 (2014), 1–10.

43.

Barone

Arora

Ganesh

Abdel-Fatah

Moseley

Ali

Chan

S.Y.T.

Savva

Schiavone

and Carmell

, The relationship of CDK18 expression in breast cancer to clinicopathological parameters and therapeutic response, Oncotarget 9 (2018), 29508.

44.

Barone

Staples

C.J.

Ganesh

Patterson

K.W.

Bryne

D.P.

Myers

K.N.

Patil

A.A.

Eyers

C.E.

Maslen

and Skehel

J.M.

, Human CDK18 promotes replication stress signaling and genome stability, Nucleic Acids Research 44 (2016), 8772–8785.

45.

Braams

and D’Angiolella

, Keeping CDK18 in balance to prevent DNA replication stress in breast cancer, Oncotarget 9 (2018), 37610.

46.

Wang

Y.W.

Zhao

Yuan

X.Y.

Liu

Zhang

Wang

Zhu

and Ma

, miR-4732-5p promotes breast cancer progression by targeting TSPAN 13, Journal of Cellular and Molecular Medicine 23 (2019), 2549–2557.

Expression pattern and prognostic significance of CDKs in breast cancer: An integrated bioinformatic study

Abstract

BACKGROUND:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Materials and methods

2.1 Identification of differentially expressed CDKs in breast cancer

2.2 Kaplan-Meier plotter

2.3 ROC plotter analysis

2.4 bc-GenEXminer

Table 1 Log2 fold change of CDKs in breast cancer

2.6 Functional enrichment of highly deregulated CDKs

2.7 CDKs exploration in single cell sequencing database (TME)

3. Results

3.1 Expression of CDKs is deregulated in BC

5. Conclusion

Author contributions

Funding

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Log2 fold change of CDKs in breast cancer