Abstract

   UNC-5 Homolog B (UNC5B) is a member of the dependence receptor family
   that regulates cell survival and apoptosis in a ligand-dependent
   manner. UNC5B plays an important role in the development of multiple
   cancers, including colorectal, bladder, and thyroid cancer. However,
   the exact expression pattern and mechanism of UNC5B in breast cancer
   have not been well elucidated. Here, we showed that UNC5B expression
   was significantly upregulated in breast cancer using bioinformatics
   analysis and experimental validation. High UNC5B expression was
   correlated with poor overall survival in breast cancer patients. UNC5B
   knockdown inhibited breast cancer cell proliferation and metastasis and
   compromised PI3K/Akt signaling activation. In summary, UNC5B is a
   promising diagnostic and prognostic biomarker and targeting UNC5B is a
   potential strategy for individualized breast cancer treatment.

   Keywords: UNC5B, breast cancer, prognosis, metastasis, functional
   networks

INTRODUCTION

   Worldwide, breast cancer is the most commonly diagnosed cancer and the
   leading cause of cancer death in women [[38]1]. In the last few
   decades, the treatment efficacy of breast cancer has improved
   tremendously, and various prognostic biomarkers such as estrogen
   receptor (ER), progesterone receptor (PR), and human epidermal growth
   factor receptor 2 (HER2) have been identified [[39]2, [40]3]. With the
   help of these molecular biomarkers, endocrine therapy such as tamoxifen
   and targeted therapy such as trastuzumab have been used as first-line
   options for ER-positive and HER2-positive breast cancer patients,
   respectively [[41]4, [42]5]. Despite the great success of early
   screening and biomarker-based therapies in breast cancer treatment,
   some patients already have distant metastases when diagnosed with
   breast cancer and some patients acquire resistance to therapies such as
   tamoxifen [[43]6], which severely impairs their prognosis [[44]7].
   Therefore, effective diagnostic biomarkers and new therapeutic targets
   for breast cancer are urgently needed.

   UNC-5 Homolog B (UNC5B), first described as an axon-guiding
   transmembrane receptor, is a member of the human UNC5 homologs (UNC5H)
   comprising UNC5A, UNC5B, UNC5C, and UNC5D [[45]8]. UNC5B plays crucial
   roles in multiple biological processes, including regulating vascular
   morphogenesis and acting as a repellent in axonal migration [[46]9,
   [47]10]. In addition, UNC5B belongs to the dependence receptor family,
   which is defined by common functional characteristics and initiates two
   distinct signaling pathways depending on the presence or absence of
   ligand [[48]11]. UNC5B can trigger apoptosis by activating
   death-associated protein kinase (DAPK) in the absence of ligand
   [[49]12], whereas its proapoptotic effect is blocked in the presence of
   netrin-1 [[50]13], a survival factor in tumorigenesis [[51]14–[52]17].
   Because of its well-known effects on apoptosis and cell survival, UNC5B
   is implicated in tumorigenesis. UNC5B expression is dysregulated in
   many cancers, including colorectal, bladder, and thyroid cancer
   [[53]18–[54]20] and UNC5B knockdown suppresses the proliferation,
   migration, and invasion of thyroid cancer cells [[55]20]. Nevertheless,
   the expression level of UNC5B and its clinical relevance in breast
   cancer have not been fully elucidated. Moreover, the molecular function
   and underlying mechanism of UNC5B in breast cancer are unclear.

   In this study, we showed that UNC5B was frequently overexpressed in
   breast cancer. High UNC5B expression was correlated with poor prognosis
   in breast cancer patients. We also noted that UNC5B inhibition
   attenuated the proliferation and metastasis of breast cancer cells and
   demonstrated that UNC5B depletion inhibited PI3K/Akt signaling. Our
   findings highlight the significance of UNC5B in breast cancer,
   including its promising diagnostic and prognostic value and potential
   as a therapeutic target.

RESULTS

UNC5B is predominant in breast cancer

   Based on the ONCOMINE database ([56]http://www.oncomine.org), we first
   analyzed the mRNA expression of UNC5H receptors in multiple cancers and
   normal tissues ([57]Figure 1A). Compared with normal breast tissues,
   UNC5B mRNA expression was significantly upregulated in breast cancer
   tissues in 14 datasets (p < 0.05) ([58]Table 1). Analysis of the breast
   cancer dataset available through The Cancer Genome Atlas (TCGA) and
   GTEx database confirmed a marked increase in UNC5B expression in breast
   cancer compared to normal breast tissues (p < 0.05), whereas there were
   no significant differences in UNC5A, UNC5C, and UNC5D expression
   between breast cancer and normal tissues ([59]Figure 1B). Thus, further
   analyses focused on UNC5B. We further analyzed the UNC5B expression in
   32 cancer types utilizing the TCGA database. UNC5B mRNA expression was
   consistently higher in breast cancer tumors than in adjacent normal
   tissues (p < 0.001) ([60]Supplementary Figure 1A). Meanwhile, UNC5B
   expression was significantly higher in cholangiocarcinoma, kidney renal
   clear cell carcinoma, and thyroid carcinoma compared with adjacent
   normal tissues (p < 0.001) ([61]Supplementary Figure 1A). We
   subsequently analyzed UNC5B expression in three major histological
   subtypes of breast cancer, including invasive ductal breast carcinoma,
   lobular breast carcinoma, and mixed lobular and ductal breast
   carcinoma. Consistently, UNC5B expression was higher in tumors than in
   adjacent normal tissues among these three breast cancer subtypes
   ([62]Supplementary Figure 1B–[63]1I).

Figure 1.

   [64]Figure 1
   [65]Open in a new tab

   UNC5B is overexpressed in breast cancer (ONCOMINE and GEPIA). (A)
   Different mRNA expression of UNC5H receptors between cancer tissues and
   normal tissues. (B) mRNA expression levels of UNC5H receptors in breast
   cancer tissues and normal breast tissues. *, p < 0.05.

Table 1. Significant changes in UNC5B expression at the transcriptional level
between different histological subtypes of breast cancer and normal breast
tissues (ONCOMINE).

   Types of breast cancer VS. normal breast tissue Fold change p value
   t-test Overexpression gene rank Reference
   Invasive Ductal Breast Carcinoma 4.337 3.53E-6 5.772 top 1% Turashvili
   Breast [[66]43]
   Invasive Lobular Breast Carcinoma 2.382 0.046 1.938 top 9% Turashvili
   Breast [[67]43]
   Ductal Breast Carcinoma in Situ Stroma 4.726 1.18E-8 8.285 top 1% Ma
   Breast 4 [[68]44]
   Invasive Ductal Breast Carcinoma Stroma 4.731 4.45E-6 6.394 top 1% Ma
   Breast 4 [[69]44]
   Invasive Breast Carcinoma 3.318 8.58E-25 12.692 top 2% TCGA Breast^a
   Invasive Ductal Breast Carcinoma 3.512 8.39E-38 19.232 top 2% TCGA
   Breast^a
   Invasive Lobular Breast Carcinoma 4.147 2.97E-19 12.105 top 1% TCGA
   Breast^a
   Mixed Lobular and Ductal Breast Carcinoma 5.947 1.45E-4 6.706 top 5%
   TCGA Breast^a
   Male Breast Carcinoma 6.661 1.32E-10 25.621 top 1% TCGA Breast^a
   Intraductal Cribriform Breast Adenocarcinoma 7.027 2.41E-6 21.314 top
   2% TCGA Breast^a
   Invasive Ductal Breast Carcinoma 3.705 8.00E-7 10.125 top 5% Zhao
   Breast [[70]45]
   Lobular Breast Carcinoma 2.827 9.18E-6 8.016 top 4% Zhao Breast
   [[71]45]
   Invasive Ductal Breast Carcinoma Stroma 3.741 0.001 4.155 top 4%
   Karnoub Breast [[72]46]
   Invasive Breast Carcinoma 3.456 0.002 6.470 top 9% Gluck Breast
   [[73]47]
   [74]Open in a new tab

   ^aTCGA: The Cancer Genome Atlas.

   We then evaluated correlations between UNC5B expression and
   clinicopathological features in breast cancer patients. A total of
   1,097 primary breast tumor samples and 114 normal breast tissues were
   analyzed in UALCAN ([75]http://ualcan.path.uab.edu). Across gender,
   race, menopause status, molecular subtypes, and cancer stages, UNC5B
   mRNA expression was always higher in breast cancer patients than in
   healthy individuals ([76]Figure 2). Note that white breast cancer
   patients tended to have higher UNC5B expression than African-American
   patients (p < 0.001) and Asian patients (p < 0.05), implying ethnic
   diversity of UNC5B expression ([77]Figure 2C). In addition, the luminal
   subtype expressed UNC5B at a higher level than the HER2-positive (p <
   0.001) and triple-negative subtypes (p < 0.001) ([78]Figure 2E).

Figure 2.

   [79]Figure 2
   [80]Open in a new tab

   The mRNA expression of UNC5B in breast cancer patients with distinct
   clinicopathological features (UALCAN). (A) UNC5B mRNA expression in
   normal breast tissues and breast cancer tissues. (B) UNC5B mRNA
   expression in normal breast tissues and breast cancer tissues
   classified by gender. (C) UNC5B mRNA expression in normal breast
   tissues and breast cancer tissues classified by ethnicity. (D) UNC5B
   mRNA expression in normal breast tissues and breast cancer tissues
   classified by menopause status. (E) UNC5B mRNA expression in normal
   breast tissues and breast cancer tissues classified by molecular
   subtype. (F) UNC5B mRNA expression in normal breast tissues and breast
   cancer tissues classified by cancer stage. *, p < 0.05; **, p < 0.01;
   ***, p < 0.001.

UNC5B protein expression in breast cancer

   We further evaluated UNC5B protein expression in diverse breast cancer
   cell lines and MCF 10A cells, a nontumorigenic breast epithelial cell
   line, by performing Western blot. As expected, UNC5B expression was
   upregulated in eight breast cancer cell lines compared to MCF 10A cells
   ([81]Figure 3A and [82]3B).

Figure 3.

   [83]Figure 3
   [84]Open in a new tab

   UNC5B protein expression in breast cancer (Western blot and IHC). (A)
   UNC5B expression in MCF 10A cells and diverse breast cancer cell lines.
   (B) The relative UNC5B protein expression in MCF 10A cells and diverse
   breast cancer cell lines. (C) Representative IHC image of UNC5B
   expression in breast cancer tissues and normal breast tissues.

   In addition, we analyzed UNC5B expression in breast cancer tissues and
   normal breast tissues using immunohistochemistry (IHC). UNC5B was
   expressed at a higher level in breast cancer tissues than in normal
   tissues ([85]Figure 3C). Considering the close association of UNC5B
   with breast cancer, UNC5B could act as a potential diagnostic marker of
   breast cancer.

High UNC5B expression associated with poor prognosis in breast cancer
patients

   We then investigated the prognostic value of UNC5B in breast cancer.
   Interestingly, survival analysis using GEPIA
   ([86]http://gepia.cancer-pku.cn/index.html) indicated that increased
   UNC5B mRNA expression was significantly associated with poor overall
   survival (OS) in breast cancer patients (p = 0.026) ([87]Figure 4A). By
   using a breast cancer tissue microarray (TMA), we further evaluated the
   prognostic value of UNC5B at the protein level. The results of the IHC
   staining revealed that high UNC5B protein expression was correlated
   with poor OS in breast cancer patients (p = 0.0125) ([88]Figure 4B and
   [89]4C). These results suggest that UNC5B might be a promising
   prognostic biomarker in breast cancer treatment.

Figure 4.

   [90]Figure 4
   [91]Open in a new tab

   The prognostic value of UNC5B expression in breast cancer (GEPIA and
   IHC). (A) The prognostic value of UNC5B mRNA expression in breast
   cancer patients, analyzed by GEPIA. (B) The prognostic value of UNC5B
   protein expression in breast cancer patients. (C) Representative IHC
   images of high expression and low expression of UNC5B in breast cancer
   tissues.

Genomic alterations of UNC5B and their biological functions in breast cancer

   We next evaluated genomic alterations of UNC5B in breast cancer using
   cBioPortal ([92]http://cbioportal.org). UNC5B was altered in 24 of 180
   (13%) breast cancer patients ([93]Figure 5A), and amplification was the
   most frequent type of UNC5B alteration in breast cancer (10.13%)
   ([94]Figure 5B). The Network model of cBioPortal was used to generate
   an interaction network of the 20 most frequently altered
   UNC5B-neighboring genes. The results showed that cancer-related genes,
   including PIK3CA, PIK3CD, PIK3R1, DAPK1, DAPK3, RAC1, RAC2, and GRB2,
   were associated with UNC5B alterations ([95]Figure 5C). Detailed
   characteristics of UNC5B-neighboring gene alterations are shown in
   [96]Supplementary Table 1.

Figure 5.

   [97]Figure 5
   [98]Open in a new tab

   Genomic alterations of UNC5B and the biological interaction network in
   breast cancer (cBioPortal). (A, B) Types and frequencies of UNC5B
   alterations in breast cancer. (C) The biological interaction network of
   UNC5B and its neighboring altered genes in breast cancer. UNC5B is
   represented with a thick border, and the interaction types are shown in
   the legend.

   To further investigate the functions of UNC5B and the 20 most
   frequently altered UNC5B-neighboring genes, we performed Gene Ontology
   (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway
   enrichment analyses using the “clusterProfiler” package in R [[99]21].
   These genes were primarily involved in the composition of the
   phosphatidylinositol 3-kinase (PI3K) complex, membrane structures, and
   cell-cell junctions ([100]Figure 6A). Similarly, these gene alterations
   primarily affected the regulation of PI3K signaling and protein kinase
   B signaling ([101]Figure 6B). Moreover, these gene alterations
   significantly affected molecular functions, such as PI3K activity,
   protein binding, and signaling adaptor activity ([102]Figure 6C). In
   the KEGG pathway analysis, the results showed obvious enrichments in
   cancer-associated signaling pathways, including the Ras signaling
   pathway ([103]Figure 6D and [104]6E).

Figure 6.

   [105]Figure 6
   [106]Open in a new tab

   Enrichment analysis of UNC5B and its neighboring altered genes in
   breast cancer. The functions of UNC5B and the 20 most frequently
   altered UNC5B-neighboring genes were predicted in Gene Ontology (GO)
   and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment
   analyses using the “clusterProfiler” package in R. (A) Cellular
   components. (B) Biological processes. (C) Molecular functions. (D) KEGG
   pathway enrichment analysis. (E) Detailed annotation of the Ras
   signaling pathway regulated by UNC5B-associated gene alterations. Nodes
   marked in red represent altered genes.

Knockdown of UNC5B in breast cancer cells inhibited their proliferation and
metastasis

   To elucidate the specific role of UNC5B in breast cancer, we performed
   shRNA-mediated UNC5B knockdown in T-47D and MDA-MB-231 breast cancer
   cells. Quantitative RT-PCR revealed that UNC5B mRNA expression was
   significantly inhibited by UNC5B-shRNA (UNC5B-SH1 and UNC5B-SH2) (p <
   0.0001) ([107]Figure 7A). Western blot analysis confirmed a marked
   decrease in UNC5B expression in UNC5B-shRNA treated breast cancer cells
   compared to the negative control (NC)-shRNA treated cells ([108]Figure
   7B). As expected, UNC5B depletion significantly attenuated the
   proliferation of breast cancer cells (p < 0.01) ([109]Figure 7C).
   Consistently, the silencing of UNC5B compromised the colony formation
   ability of breast cancer cells ([110]Figure 7D and [111]7E). We further
   investigated the effect of UNC5B on breast cancer metastasis. As shown
   in [112]Figure 7F–[113]7I, the silencing of UNC5B significantly
   suppressed the migration and invasion capacities of breast cancer
   cells. Collectively, these results emphasized the essential role of
   UNC5B in breast cancer cell proliferation and metastasis.

Figure 7.

   [114]Figure 7
   [115]Open in a new tab

   UNC5B knockdown inhibited the proliferation and metastatic capacities
   of breast cancer cells. (A) Quantitative PCR analysis of UNC5B
   expression in T-47D and MDA-MB-231 cells after UNC5B-shRNA
   introduction. (B) Western blot analysis of UNC5B expression in T-47D
   and MDA-MB-231 cells after UNC5B-shRNA introduction. (C) The
   proliferation of T-47D and MDA-MB-231 cells after UNC5B knockdown. (D,
   E) Representative images and quantified results of colonies formation
   in T-47D and MDA-MB-231 cells after UNC5B knockdown. (F, G)
   Representative images and quantified results of transwell migration
   assay in T-47D and MDA-MB-231 cells after UNC5B knockdown. (H, I)
   Representative images and quantified results of matrigel-treated
   transwell invasion assay in T-47D and MDA-MB-231 cells after UNC5B
   knockdown. Data represent three independent experiments. Data are
   expressed as mean ± s.e.m. *p < 0.05, **p < 0.01, ***p < 0.001, ****p <
   0.0001.

Knockdown of UNC5B in breast cancer cells inhibited PI3K/Akt pathway
activation

   To further explore UNC5B co-expressed genes in breast cancer, we used
   the LinkedOmics database ([116]http://linkedomics.org/login.php) to
   analyze mRNA sequencing data from 1,093 breast cancer patients in TCGA.
   As shown in the volcano plot ([117]Supplementary Figure 2A), 6,672
   genes were significantly positively correlated with UNC5B and 6,161
   genes were significantly negatively correlated (false discovery rate
   (FDR) < 0.05). COL5A1 was the most significantly positively correlated
   with UNC5B (Pearson correlation = 0.68, p = 1.557e-147)
   ([118]Supplementary Figure 2B). The top 50 genes positively and
   negatively correlated with UNC5B in breast cancer are shown in heat
   maps ([119]Supplementary Figure 2C and [120]2D).

   The functions of these UNC5B co-expressed genes were explored using GO
   and KEGG pathway enrichment analyses. Proteins from these genes were
   mainly located in the extracellular matrix, adherens junctions, and
   membrane regions ([121]Figure 8A). These genes were involved in
   multiple biological processes, including small GTPase-mediated
   signaling transduction and extracellular structure organization
   ([122]Figure 8B). Similarly, these genes remarkably affected the
   molecular functions of cell adhesion molecule binding and GTPase
   binding ([123]Figure 8C). Notably, KEGG enrichment analysis indicated
   widespread changes in biological pathways, among which the PI3K/Akt
   signaling pathway was prominently enriched ([124]Figure 8D). To
   demonstrate whether UNC5B regulates the PI3K/Akt signaling pathway in
   breast cancer, we analyzed PI3K/Akt pathway activation using Western
   blot analysis. As shown in [125]Figure 8E–[126]8F, the phosphorylation
   of PI3K and Akt were significantly blunted in UNC5B-knockdown breast
   cancer cells compared with breast cancer cells treated with NC-shRNA,
   suggesting the essential role of UNC5B in PI3K/Akt pathway activation.

Figure 8.

   [127]Figure 8
   [128]Open in a new tab

   UNC5B knockdown compromised PI3K/Akt signaling activation in breast
   cancer cells. (A–D) The functions of UNC5B and genes significantly
   co-expressed with UNC5B were predicted in GO and KEGG pathway
   enrichment analyses using the “clusterProfiler” package in R. (A)
   Cellular components. (B) Biological processes. (C) Molecular functions.
   (D) KEGG pathway enrichment analysis. (E, F) Western blot analysis of
   PI3K and Akt phosphorylation in breast cancer cells after UNC5B
   knockdown. Data represent three independent experiments. Data are
   expressed as mean ± s.e.m. **p < 0.01, ***p < 0.001, ****p < 0.0001.

DISCUSSION

   Metastasis and drug resistance are critical challenges in the clinical
   management of breast cancer. Despite advances in noninvasive breast
   surveillance, including breast ultrasound, mammography, and MRI,
   approximately 4-6% of breast cancer patients have metastases when
   diagnosed, compromising the effectiveness of subsequent therapies
   [[129]22]. In addition, drug resistance, such as chemoresistance in
   triple-negative breast cancer patients, usually leads to tumor
   progression and poor prognosis [[130]23]. Hence, identifying specific
   diagnostic biomarkers and novel therapeutic targets would be incredibly
   valuable in the treatment of breast cancer.

   In the present study, we demonstrated for the first time that UNC5B
   mRNA expression was significantly higher in breast cancer tissues than
   in normal breast tissues using transcriptional data from more than
   1,000 breast cancer samples in diverse databases. Further subgroup
   analyses confirmed that UNC5B expression was upregulated in breast
   cancer tissues consistently, regardless of histological subtypes and
   other clinicopathological features, including gender, race, menopause
   status, molecular subtype, and cancer stage. Similarly, we verified
   that UNC5B expression was upregulated in breast cancer cell lines and
   breast cancer tissues using Western blot and IHC, respectively. These
   findings suggest UNC5B as a promising diagnostic biomarker in breast
   cancer. Interestingly, in 2003, Thiebault et al. noted that UNC5B
   expression was downregulated in breast cancer compared with normal
   breast tissues, which was contrary to our findings [[131]24]. This
   conclusion came from the analysis of a dot blot array using a complete
   rat UNC5H2 cDNA probe that could hybridize UNC5A, UNC5B, and UNC5C cDNA
   rather than specifically binding to UNC5B. This lack of specificity may
   be the key difference between this conclusion and ours.

   To understand how UNC5B influences breast cancer prognosis, we analyzed
   the expression of UNC5B and its prognostic value in breast cancer using
   the TCGA data in GEPIA. The results showed that increased UNC5B
   expression was significantly correlated with poor OS in breast cancer
   patients. Protein level validation in a breast cancer TMA confirmed
   that high UNC5B expression could serve as a biomarker for poor
   prognosis in breast cancer. Intriguingly, the expression of UNC5H
   receptors was downregulated in colorectal cancer, and UNC5C
   loss-of-function was associated with intestinal tumor progression in
   mice [[132]24, [133]25]. In this regard, UNC5H receptors were proposed
   to be tumor suppressors. Nevertheless, netrin-1 expression in the mouse
   gut reduced intestinal cell apoptosis and facilitated intestinal tumor
   development [[134]26]. Hence, UNC5H receptors could serve as
   conditional tumor suppressors depending on the presence of netrin-1,
   theoretically providing two ways to impact tumorigenesis [[135]27].
   First, UNC5H expression could be reduced to block the proapoptotic
   effect of these receptors in the absence of netrin-1, as occurs in
   colorectal cancer cells. Second, the antiapoptotic effect of these
   receptors could be enhanced in the presence of netrin-1. Along this
   line, UNC5B overexpression in breast cancer as observed in our study
   could be a selective advantage for tumorigenesis, as increased
   expression of netrin-1 was seen in the majority of breast tumors with
   metastatic tendency [[136]14]. In addition, our results showed that
   UNC5B knockdown significantly inhibited the proliferation and
   metastasis of breast cancer cells, underscoring the essential role of
   UNC5B in breast cancer development.

   Recent studies have emphasized the importance of genomic alterations in
   cancer development [[137]28]. The identification of several mutational
   cancer drivers, including ER, HER-2, PIK3CA, and CDK4/6, promoted the
   development of stratified medicine in breast cancer [[138]29]. Here, we
   discovered that UNC5B was altered in 24 of 180 sequenced breast cancer
   patient samples (13%) and that the major type of UNC5B alteration was
   amplification. Interestingly, further analysis of the interaction
   network of UNC5B and its neighboring altered genes indicated that
   genomic alterations to these genes mainly influenced the activities of
   key kinases in signaling transduction pathways, including the Ras
   signaling pathway. Indeed, the Ras signaling pathway plays a vital role
   in regulating cell survival and abnormal Ras signal transduction
   promotes cancer development [[139]30]. Thus, UNC5B is a promising
   target for individualized breast cancer management.

   Various studies have demonstrated the importance of the PI3K/Akt
   signaling pathway in breast cancer [[140]31]. Tumor cells aberrantly
   activate the PI3K/Akt pathway to enhance cell proliferation, survival,
   and drug resistance [[141]32]. However, the use of PI3K inhibition in
   clinical trials showed only a modest effect in breast cancer patients
   [[142]33, [143]34], mainly because of drug-related toxicities and
   compensatory mechanisms [[144]35]. In our study, UNC5B co-expressed
   genes were significantly enriched in the PI3K/Akt signaling pathway and
   UNC5B depletion in breast cancer cells markedly inhibited the PI3K/Akt
   pathway. This was consistent with the previous finding that UNC5B
   regulated cell survival through PI3K-mediated activation of Akt in
   response to stresses [[145]36]. Consequently, targeting UNC5B may
   enhance the clinical application of PI3K inhibitors and improve the
   prognosis of breast cancer patients.

   In summary, our study provides a comprehensive understanding of the
   role of UNC5B in breast tumorigenesis and better defines the value of
   UNC5B in breast cancer management. Our data revealed that upregulated
   UNC5B expression in breast cancer could be a promising diagnostic
   biomarker. High UNC5B expression and poor OS are correlated, further
   suggesting UNC5B as a potential prognostic biomarker in breast cancer.
   Moreover, UNC5B knockdown mitigated the aggressiveness of breast cancer
   cells and compromised PI3K/Akt pathway activation, suggesting UNC5B as
   a new therapeutic target for breast cancer. One limitation of our study
   was that the UNC5B genomic alteration data in breast cancer were
   obtained from online databases and were not validated in our datasets.
   In addition, the molecular functions of UNC5B in breast cancer were
   evaluated in vitro. Further work is required to validate the spectrum
   of UNC5B genomic alterations, and in vivo experimental investigations
   are needed to validate the essential role of UNC5B in breast cancer,
   which could contribute to the development of precision medicine.

MATERIALS AND METHODS

Ethics statement

   This study was approved by the Medical Ethics Committee of the Second
   Affiliated Hospital of Zhejiang University School of Medicine. All
   datasets were obtained from public databases and all data were
   collected with written informed consent.

ONCOMINE database

   ONCOMINE database ([146]http://www.oncomine.org) is a comprehensive
   cancer microarray database and visualized data-mining platform
   [[147]37]. In our study, the mRNA expression of UNC5H receptors between
   different types of cancers and adjacent normal tissues were analyzed by
   the ONCOMINE database. The thresholds used were p value of 0.05, fold
   change of 2, and gene rank of top 10%. UNC5B transcriptional expression
   in different histological subtypes of breast cancer was also assessed
   by ONCOMINE, and Student’s t-test was used to generate a p value.

TIMER database

   TIMER ([148]https://cistrome.shinyapps.io/timer/) is an integrated
   resource for dynamic analysis of gene expression profiles and
   tumor-immune interactions across 32 cancer types [[149]38]. We used the
   Dfii ExP module to determine UNC5B expression of tumor tissues and
   adjacent normal tissues in 32 cancer types from TCGA.

UALCAN

   UALCAN ([150]http://ualcan.path.uab.edu) is an interactive web
   interface for in-depth analyses of level 3 RNA-seq and clinical data of
   31 cancer types from the TCGA database [[151]39]. We used the
   Expression module to explore the relative UNC5B mRNA expression of
   invasive breast carcinoma and adjacent normal breast tissues based on
   different clinicopathologic parameters, including gender, ethnicity,
   menopause status, molecular subtypes, and cancer stages. Student’s
   t-test was used to generate a p value with p < 0.05 considered
   statistically significant.

Cell culture

   MCF 10A cells and the human breast cancer cell line T-47D were
   purchased from ATCC (Gaithersburg, Maryland), and other cell lines,
   including MCF-7, BT-474, SK-BR-3, MDA-MB-231, MDA-MB-468, BT-549, and
   Hs 578T, were obtained from the Cell Bank of the Chinese Academy of
   Sciences (Shanghai, China). All cell lines used were analyzed by short
   tandem repeat profiling, and the culture conditions of these cell lines
   followed ATCC protocols.

UNC5B knockdown in breast cancer cells

   For stable knockdown of UNC5B, lentivirus stably expressing UNC5B-shRNA
   or control-shRNA (OBiO, Shanghai, China) was introduced into breast
   cancer cells. Knockdown efficiency was verified by quantitative RT-PCR
   and Western blot. The sequences of UNC5B-shRNA and NC-shRNA were as
   follows. UNC5B-SH1:
   5’-CCGGCCACACAGATCTACTTCAATTCAAGAGATTGAAGTAGATCTGTGTGGTTTTTTG-3’;
   UNC5B-SH2:
   5’-CCGGCCAACTTCCTGCTCACCATTTCAAGAGAATGGTGAGCAGGAAGTTGGTTTTTTG-3’; NC:
   5’-CCGGTTCTCCGAACGTGTCACGTTTCAAGAGAACGTGACACGTTCGGAGAATTTTTTG-3’.

Quantitative real-time PCR

   Total RNA was extracted using TRIzol reagent (Invitrogen, Boston, MA)
   and cDNA was synthesized with the Hifair III cDNA Synthesis Kit
   (Yeasen, 11141ES10, Shanghai, China). Real-time PCR was carried out
   with the Hieff UNICON qPCR SYBR Green Master Mix Kit (Yeasen,
   11199ES03, Shanghai, China) on an Applied Biosystems 7500 Fast
   Real-Time PCR System. Relative mRNA expression was determined using the
   2^-ΔΔCt method. Primers used are as follows: UNC5B (Human)-forward:
   CACGGGCGAGTCCTATTC; UNC5B (Human)-reverse: CGGCTCCTCCACCAAGTA; GAPDH
   (Human)-forward: ACAACTTTGGTATCGTGGAAGG; GAPDH (Human)-reverse:
   GCCATCACGCCACAGTTTC.

Western blot

   Total protein was extracted using RIPA buffer (Boster, Pleasanton, CA)
   supplemented with a protease and phosphatase inhibitor cocktail (Thermo
   Fisher Scientific, Boston, MA) and quantified using the BCA Protein
   Assay (Thermo Fisher Scientific, Boston, MA) according to
   manufacturer’s instructions. Protein lysates were separated using 10%
   SDS-PAGE and transferred to PVDF membranes. Subsequently, membranes
   were blocked with 5% non-fat milk and incubated with UNC5B (Bioss,
   11492R, 1:1000, Woburn, MA), phospho-PI3K p85α (phospho Y607) (Abcam,
   ab182651, 1:1000, Cambridge, MA), PI3K p85α (Abcam, ab191606, 1:1000,
   Cambridge, MA), phospho-Akt (Ser473) (Cell Signaling Technology, 4060S,
   1:2000, Danvers, MA), Akt (Cell Signaling Technology, 2920S, 1:2000,
   Danvers, MA), and GAPDH (Boster, BM1985, 1:2000, Pleasanton, CA) at 4°
   C overnight, followed by a one-hour incubation with secondary
   antibodies. The signal was detected using an enhanced chemiluminescence
   detection kit (Boster, Pleasanton, CA). Relative protein expression was
   measured with ImageJ and normalized over GAPDH.

Tissue microarray and immunohistochemistry

   A tissue microarray (TMA) containing 160 breast cancer tissues and 10
   normal breast tissues was purchased from Outdo Biotech (Shanghai,
   China). The slide was deparaffinized and rehydrated before antigen
   retrieval. After endogenous peroxidase activity was blocked, the slide
   was blocked with 10% normal goat serum and incubated with a UNC5B
   polyclonal antibody (Thermo Fisher Scientific, PA5-67631, 1:25, Boston,
   MA) at 4 °C overnight. The slide was then incubated with secondary
   antibody, stained with diaminobenzidine, and counterstained with
   hematoxylin.

   The UNC5B expression was determined by two pathologists according to
   the immunoreactivity score (IRS). The immunostaining intensity
   (negative, 0; weak, 1+; moderate, 2+; strong, 3+) and the percentage of
   positive cells (0-5%, 0; 6%-25%, 1+; 26%-50%, 2+; 51%-75%, 3+;
   76%-100%, 4+) were recorded, and the IRS was calculated by multiplying
   the two scores. Because four tumor cores were lost during IHC staining,
   156 tumor cores were analyzed, and tumor samples were classified into
   two groups, high expression and low expression, by their median
   expression. p values were generated by the log-rank test with p < 0.05
   considered statistically significant.

GEPIA analysis

   The GEPIA ([152]http://gepia.cancer-pku.cn/index.html) is a web-based
   tool for diverse gene expression analysis based on 9,736 tumors and
   8,587 normal samples from the TCGA and the GTEx databases [[153]40].
   Here, we used the Expression module to analyze the mRNA expression of
   UNC5H receptors in breast cancer and normal breast tissues. We also
   used the Survival module to evaluate the prognostic value of UNC5B in
   breast cancer. Breast cancer samples were separated into two groups,
   high expression and low expression, by their median expression, and p
   values were determined by the log-rank test.

cBioPortal analysis

   The cBioPortal for Cancer Genomics ([154]http://cbioportal.org) is a
   user-friendly web resource that provides access to multiple cancer
   genomic data sets [[155]41]. We explored the metastatic breast cancer
   project (Provisional, October 2018) dataset and analyzed genomic
   alterations of UNC5B, which contained mutations and putative
   copy-number alterations from GISTIC. The OncoPrint module provided an
   overview of the genomic profile in UNC5B, and the Network module was
   used to visualize biological interactions between UNC5B and the 20 most
   frequently altered neighboring genes in breast cancer. GO and KEGG
   pathway enrichment analyses of altered neighboring genes were conducted
   with the “clusterProfiler” package in R.

Cell proliferation assay

   Breast cancer cell proliferation was determined using a CCK-8 kit
   (Dojindo, CK04, Tokyo, Japan). Briefly, 2,500 cells/well were seeded
   into a 96-well plate and incubated at 37 °C. After culturing for 24-96
   h, cells were incubated with CCK-8 reagent according to manufacturer’s
   instructions and absorbance at 450 nm was measured.

Cell colony formation assay

   A total of 1,000 cells/well were seeded into a 6-well plate and
   incubated at 37 °C. After culturing for 2 weeks, colonies were fixed,
   stained with crystal violet, and counted. Experiments were repeated at
   least three times independently.

Cell migration and invasion assay

   Cell migration and invasion assays were performed using Transwell
   chambers (Corning, 3422, NYC), without Matrigel (Corning, 356230, NYC)
   for migration assays and with Matrigel for invasion assays. Briefly,
   T-47D cells (10 × 10^4 cells/chamber) or MDA-MB-231 cells (5 × 10^4
   cells/chamber) in 200 μL of serum-free DMEM or L15 medium were seeded
   in the upper chambers and 500 μL of corresponding medium containing 10%
   FBS were placed in the lower chambers. After 24 h, cells on the lower
   surface of the membrane were fixed and stained with crystal violet.
   Cells were counted in five random fields under a microscope.

LinkedOmics database

   The LinkedOmics database ([156]http://linkedomics.org/login.php)
   encompasses multi-omics data for 32 cancer types from the TCGA project
   and provides comprehensive analyses of more than a billion data points
   [[157]42]. We used the LinkFinder module to analyze and visualize
   UNC5B-correlated genes that were differentially expressed in 1,093
   breast cancer patients. Pearson’s correlation test was used to measure
   the strength of the correlations. Significantly co-expressed genes
   correlated with UNC5B were further extracted using the criterion of
   FDR<0.05, and GO and KEGG enrichment analyses were performed with the
   “clusterProfiler” package in R.

Statistical analysis

   Data were analyzed in GraphPad Prism 7 (GraphPad Software) and were
   presented as the mean ± s.e.m. Unless otherwise stated, statistical
   significance was analyzed by either two-tailed, Student’s t-test or
   one-way ANOVA. p values < 0.05 were considered statistically
   significant.

Supplementary Material

   Supplementary Figures
   [158]aging-12-103639-s002..pdf^ (655.8KB, pdf)
   Supplementary Table 1
   [159]aging-12-103639-s001..pdf^ (341.3KB, pdf)

Footnotes

   AUTHOR CONTRIBUTIONS: Y.D.C. and S.J.W. designed the study; S.J.W.,
   X.Y.G., H.H.C., and Y.X.L. conducted the bioinformatics analyses;
   S.J.W., X.Z., and K.Z. performed the experimental validation; S.J.W.,
   X.Y.G., and J.J.Z. wrote the manuscript; J.J.Z. and Y.D.C. reviewed the
   paper.

   CONFLICTS OF INTEREST: The authors declare no potential conflicts of
   interest.

   FUNDING: This work was supported by the Key Program of the Natural
   Science Foundation of Zhejiang Province (LZ16H160002), the Zhejiang
   Provincial Program for the Cultivation of High-level Innovative Health
   Talents, the National Natural Science Foundation of China (81702866),
   and the Preclinical and Multi-center Basket Clinical Trial of the
   Multi-kinase Inhibitor TT-00420 (2019ZX09301158).

REFERENCES