Human BRCA cell lines MDA-MB-231, MDA-MB-468, MCF-7, and BT-549 as well as normal breast epithelial cells MCF-10A were purchased (22). BRCA cells were grown in complete high glucose DMEM (Wisent, USA), enriched with 10% FBS, 100 μg/ml pen/strep (Hyclone, USA). MCF10A were cultured in mammary epithelial cell basal medium (MEBM, Lonza, USA) supplemented with 100 ng/ml cholera toxin. All cells were grown at 37°C, 5% CO₂ in a humidified incubator.

DCTPP1 knockdown (shDCTPP1) or DCTPP1 overexpression (DCTPP1), and a scrambled sequence (SCR) or a negative control (NC) sequence, respectively, were used according to the manufacturer's instructions. Plasmid sequences were validated via sequencing (GenePharma, China). We cultured the cells were in 6-well plates at 30% confluence, followed by inoculation with the retroviruses. Polybrene (5 μg/ml) was used to enhance infection efficiency. Stably transfected cells (puromycin-resistant) were selected by treating with puromycin (2 μg/ml) for 2 weeks. DCTPP1-overexpressing MCF7 and MDA-MB-468 cells, and control (NC) cells were inoculated into 6-well plates and cultured overnight. Next, siRNA (GenePharma, China) and non-targeting control siRNA were transfected using lipofectamine® 3000 (Invitrogen, USA) using manufacturer instructions. The sequences of the siRNAs were: Sense 5′-gatccgcccttcaagaggagcttattcaagagataagctcctcttgaagggcttttttacgcgt g-3′ and antisense 5′-aattcacgcgtaaaaaagcccttcagaggagcttatctcttga agctcctcttgaagggcg-3′. miRNA sequences were as follows: miR-378a-3p mimic: cuggacuuggagucagaagg, mimic-NC: agugcauguuaugccuacg, miR-378a-3p inhibitor: aguucagguucugacuccu, inhibitor-NC: ugguccguguaggccuacua.

Isolation of total RNA was carried out with the Trizol reagent (TaKaRa, USA). cDNA was generated from 1 μg of RNA via reverse-transcription with the Primescript RT Reagent (TaKaRa, USA). The FastStart Universal SYBR Green Master (Roche, USA) was employed to perform RT-qPCR on a real-time PCR instrument (Applied Biosystems, USA). GAPDH and U6 were used as reference genes. The primers are indicated in Supplementary Table 1.

Cells were lysed using RIPA buffer (Thermo Fisher, USA) enriched with 0.1% protease inhibitor, 1% phosphatase inhibitor, as well as 1% PMSF. Fractionation of the proteins was done on SDS-PAGE gel and then the proteins transfer-embedded onto NC membranes (Millipore, USA). Afterward, membranes were blocked with 5% skimmed milk in PBS for 2 h, followed by overnight incubation with anti-β-actin (Cell Signaling Technology, USA) and anti-DCTPP1 (Abgent Inc., USA) at 4°C. They were then washed thrice, 10 min each with PBST, and incubated for 1 h with indicated secondary antibodies.

Cell proliferation was examined using a CCK-8 kit (Dojindo, Japan) as per the manufacturer's protocol. 2 × 10³ cells/well, in 200 μl of cell culture media, were cultured onto 96-well plates and cultured at 37°C for 4 h. The cell culture medium was then replaced with media enriched with 10% CCK8, then incubation of the cells was performed at 37°C for 2 h. Thereafter, a microplate reader was employed to determine the absorbance at 450 nm.

Cells were seeded in 6-well plates before transfection with DCTPP1-siRNA, mimic of miR-378a-3p, or their NC control for 48 h. After that, fixation with 75% absolute ethanol was performed for 60 s, followed by rinsing twice with PBS, then staining by Giesma (Sigma, USA) for 15 min, and dried at room temperature. The colonies with ≥50 cells/well were then counted.

Breast cancer cells (MCF-7 and MDAMB-468) were inoculated in 6-well plates before transfection with DCTPP1-siNRA or control group for 48 h. TUNEL assay kit (Roche, Germany) was used to test MCF-7 and MDA-MB-468 cell apoptosis. After that, cells fixation with 4% PFA (paraformaldehyde) for 15 min. Then, MCF7 and MDA-MB-468 were blocked in 0.1% Triton X-100 for 1 h. Cells were then treated with a TUNEL reaction mixture at room temperature for 1 h. Cell nuclei were stained by DAPI for 5 min.

Where specified, cells were inoculated and grown in triplicates for 24 h, followed by co-transfection with DCTPP1-3′-UTR clones or mutant clones with pRL-TK Renilla plasmid and miR-378a-3p mimic. After the elapse of 48 h, the luciferase enzyme activity of transfected cells was assessed using a dual luciferase assay kit (Promega, USA) following the manufacturer's instructions.

All animal experiments adhered to guidelines by the Institutional Animal Care and Use Committee of the Harbin Medical University. We randomly split 28, 4-week old female BALB/c nude mice, weighing 18–22 g into four groups. Stable DCTPP1-siRNA, miR-378a-3p mimic, NC MCF-7 cells, or control cells (1 × 10⁶ cells in 100 μl of PBS) were subcutaneously administered into mammary fat pads of the mice, and tumor volume measured weekly using calipers. Tumor volume was given by the formula: (tumor length × width × height)/2. After 6 weeks, we sacrificed the mice and measured the final tumor weight.

We transfected the 8 × 10² MCF-7, and MDA-MB-468 cells with DCTPP1 siRNA, control siRNA, miR-378a-3p mimic, or mimic of miR-NC were planted into 6 cm dishes and cultured for 10 d. Staining of colonies by 0.1% crystal violet in 20% methanol was done for 15 min. Five-hundred cells as the standard for a Clonogenic. They were then imaged, and visible colonies determined.

MCF-7 and MDA-MB-468 cells were seeded onto glass coverslips and grown for 24 h to 50–60% density. They were then treated with DCTPP1 siRNA, or Control-si, for 48 h and fixed. They were then stained using anti-γH2A antibody (CST, USA) for 1 h and then incubation with secondary antibody performed for 20 min at RT. They were then counterstained with DAPI and mounted on prolong® diamond antifade (Applied biosystems, USA).

To determine DCTPP1 transcription levels in breast cancer, we analyzed gene expression cohorts in the ONCOMINE (https://www.oncomine.org), UALCAN (http://ualcan.path.uab.edu/analysis.html), Gene Expression Profiling Interactive Analysis (GEPIA) (http://gepia.cancer-pku.cn/index.html). Metascape (http://metascape.org) databases to explore the interaction and function of DCTPP1 co-expressed genes. Enrichment of GO terms including biological process, cellular component, and molecular function, and KEGG pathways was carried on using the Metascape online tools. Functional terms with p-value ≤ 0.01 and minimum count ≥3 were considered statistically significant. The most significant term within each cluster was chosen as representative of this cluster. Then, the association between the significant terms was established as a network, in which terms with similarities >0.3 were connected. Protein–protein interaction enrichment assessment was done on BioGrid, InWeb_IM, as well as OmniPath. The Molecular Complex Detection (MCODE) algorithm was used to determine the densely connected network components.

All experiments were done three times unless otherwise indicated. Data were analyzed using the SPSS 20.0 (IBM). For continuous variables, the Students t-test was employed to establish statistically remarkable differences between groups. P < 0.05 signified statistical significance.

We initially evaluated DCTPP1 transcriptional expression in multiple BRCA datasets on TCGA and Gene Expression Omnibus (GEO). Figure 1A shows the DCTPP1 expression profile in 33 cancers (GEPIA). Analysis of DCTPP1 expression in three Oncomine datasets relative to normal tissues revealed DCTPP1 overexpression in BRCA tissue relative to normal breast tissue (Figures 1B–D, p ≤ 0.01). Fold differences were all >1.5. Analysis of DCTPP1 expression in BRCA tumors vs. normal tissues using UALCAN revealed that regardless of age, gender, disease stage, nodal metastasis, major subclasses, or menopause status, DCTPP1 transcription levels were remarkably elevated in BRCA patients than in healthy controls (Figure 2). Thus, DCTPP1 has diagnostic potential in BRCA.

We then investigated if DCTPP1 expression correlates with BRCA prognosis. The impact of DCTPP1 expression on survival was evaluated using GEPIA. Notably, DCTPP1 expression significantly impacts BRCA prognosis [OS HR = 1.9, Logrank p = 0.00018, p(HR) = 0.00021; Figure 3A]. Analysis of the effects of DCTPP1 expression on the prognosis of BRCA subtypes demonstrated that high DCTPP1 expression levels were significantly linked to the prognosis of Luminal A and B subtypes [Luminal A: OS HR = 2, Logrank p = 0.011, p(HR) = 0.013, Luminal B OS HR = 2, Logrank p = 0.048, p(HR) = 0.052; Figures 3D,E]. However, DCTPP1 levels did not correlate with prognosis of basal-like and HER2+, non-luminal subtypes [basal-like/triple-negative OS HR = 2.2, Logrank p = 0.12, p(HR) = 0.13, HER2+, non-luminal OS HR = 2.8, Logrank p = 0.076, p(HR) = 0.089; Figures 3B,C].

To explore the role of DCTPP1 in BRCA cancer, we first test the expression of DCTPP1 in different BC cell lines include the TNBC cell line. We found that DCTPP1 is up-regulated in MCF-7 and MDA-MB-468 (Supplementary Figure 1). To assess DCTPP1 effects on proliferation, it was overexpressed or silenced in MCF-7 and MDA-MB-468. Western blot and RT-qPCR analysis verified transfection efficiency (Supplementary Figures 2A–D). CCK-8 analysis of DCTPP1 effects on cell growth revealed that DCTPP1 silencing suppressed MCF-7 and MDA-MB-468 proliferation (Figures 4A,B). The clonogenic analysis revealed that DCTPP1 knockdown suppressed BRCA tumorigenic potential (Figure 4C). Additionally, KI67 staining was used to assess the effect of DCTPP1 on BRCA proliferation and revealed that DCTPP1 silencing suppresses BRCA cell proliferation (Figure 4D). TUNEL assay was used to test the cell apoptosis after treatment with DCTPP1 inhibition. Our data showed that silencing DCTPP1 can induce MCF-7 and MDA-MB-468 cell apoptosis (Figure 4E). Together, these data indicate that DCTPP1 silencing suppresses BRCA cell proliferation in vitro.

To explore the impact of DCTPP1 in BRCA in vivo, DCTPP1-deficient MCF-7 and controls were transplanted into the mammary fat pads of mice. Our analysis revealed low DCTPP1 levels in DCTPP1-siRNA tumor tissue relative to control tissue (Figure 4F). Moreover, DCTPP1-siRNA tumor volume growth was slower relative to control tumors (Figure 4G). At 6 weeks, DCTPP1-siRNA tumor weights were significantly lighter than control tumors (Figure 4H). Moreover, DCTPP1-siRNA tumors expressed lower Ki67 levels relative to controls (Figure 4I). Together, these data show that DCTPP1-knockdown suppresses tumorigenesis in vivo.

To investigate the biological role of DCTPP1 in BRCA, the top 100 genes co-expressed with DCTPP1 TCGA BRCA datasets were obtained. Metascape analysis was used to identify the pathways and processes enriched in all co-expression genes, including GO biological processes as well as reactome gene sets. Functional terms with p ≤ 0.01 and a minimum count of three were selected. Co-expressed genes, were mainly enriched in GO BP terms like DNA repair, protein localization to chromosome, and regulation of intracellular estrogen receptor signaling (Figure 5A and Table 1).

γH2A is an established marker of DNA damage (23). Relative to mock-silenced cells, DCTPP1-deficient induced BRCA cells had significantly higher γH2A levels (Figure 5B). These data indicate that DCTPP1-deficient accelerated DNA damage repair.

Regarding the functional analysis of these core modules (Figure 5A and Table 2), they are enriched in the DNA repair signaling pathway. DNA repair signaling influences cancer progression (24). Western blot analysis of the DNA repair signaling pathway factors, XRCC1, PARP1, and RAD51, revealed their elevation upon DCTPP1 upregulation in BRCA cells (Figure 5C).

Given that miRNAs are implicated in BRCA progression (9, 25), we screened the publicly available databases, TargetScan and PITA, for miRNAs that may modulate DCTPP1 and found that miR-378a-3p has an optional seed unit for DCTPP1 (Figures 6A,B). Relative to negative controls, RT-qPCR and western blot analyses revealed DCTPP1 suppression in all the cell lines under miR-378a-3p overexpression (Figures 6C,D). Moreover, luciferase enzyme activity was repressed by miR-378a-3p in BRCA cells transfected with wild-type DCTPP1 3′-UTR (Figure 6E).

Evaluation of the expression of miR-378a-3p in BRCA patients' TCGA datasets revealed that it is significantly downregulated (p ≤ 0.05; Figure 7A). Moreover, we established that the expression of miR-378a-3p was dramatically lower in BRCA cell lines relative to MCF-10A (Figure 7B). To assess the function of miR-378a-3p in BRCA progression, we transfected the MCF-7 cells and MDA-MB-468 cells, which have high miR-378a-3p measures with the mimic of miR-378a-3p, and those with the lowest expression of miR-378a-3p with miR-378a-3p repressor (Supplementary Figure 2E). CCK-8 analysis revealed that the silencing of miR-378a-3p enhanced cell growth, while its overexpression markedly suppressed the growth of MCF-7 cells and MDA-MD-468 cells (Figures 7C,D). Consistent with these data, colony formation assay and ki67 staining indicated suppressed BRCA cell growth upon miR-378a-3p overexpression (Figures 7E,F). A mouse tumor xenograft revealed higher miR-378a-3p levels in miR-378a-3p mimic bearing tumors relative to the NC group (Figure 7G). The average number of harvested nodules (Figure 7H) and tumor weight (Figure 7I) were significantly lower in miR-378a-3p mimic bearing mice relative to the NC group. IHC analysis of Ki67 expression revealed lower ki67 levels in miR-378a-3p mimic bearing tumors relative to the NC group (Figure 7J). These data demonstrate that miR-378a-3p negatively modulates DCTPP1 in BRCA and that miR-378a-3p suppresses BRCA development in vitro and in vivo.