Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Nov 1.
Published in final edited form as: Steroids. 2019 Jul 22;151:108463. doi: 10.1016/j.steroids.2019.108463

Paxillin Regulated Genomic Networks in Prostate Cancer

Xiaoting Ma 1, Anindita Biswas 2, Stephen R Hammes 1,*
PMCID: PMC6802295  NIHMSID: NIHMS1536518  PMID: 31344408

Abstract

Paxillin is extensively involved in focal adhesion signaling and kinase signaling throughout the plasma membrane and cytoplasm. However, recent studies in prostate cancer suggest that paxillin also plays a critical role in regulating gene expression within the nucleus, serving as a liaison between cytoplasmic and nuclear MAPK and Androgen Receptor (AR) signaling. Here we used RNA-seq to examine the transcriptome in several human prostate cancer cell lines. First, we examined paxillin effects on androgen-mediated transcription in control or paxillin-depleted AR-positive LNCaP and C4–2 human prostate cancer cells. In androgen-dependent LNCaP cells, we found over 1000 paxillin-dependent androgen-responsive genes, some of which are involved in endocrine therapy resistance. Most paxillin-dependent AR-mediated genes in LNCaP cells were no longer paxillin-dependent in androgen-sensitive, castration-resistant C4–2 cells, suggesting that castration-resistance may markedly alter paxillin effects on genomic AR signaling. To examine the paxillin-regulated transcriptome in the absence of androgen signaling, we performed RNA-seq in AR-negative PC3 human prostate cancer cells. Paxillin enhanced several pro-proliferative pathways, including the CyclinD/Rb/E2F and DNA replication/repair pathways. Additionally, paxillin suppressed pro-apoptotic genes, including CASP1 and TNFSF10. Quantitative PCR confirmed that these pathways are similarly regulated by paxillin in LNCaP and C4–2 cells. Functional studies showed that, while paxillin stimulated cell proliferation, it had minimum effect on apoptosis. Thus, paxillin appears to be an important transcriptional regulator in prostate cancer, and analysis of its transcriptome might lead to novel approaches toward the diagnosis and treatment of this important disease.

Keywords: Paxillin, Prostate Cancer, Androgen Receptor, Transcription, Proliferation

Introduction

Prostate cancer is the most frequently diagnosed non-skin cancer and the second leading cause of cancer death in United States. With the incidence of prostate cancer rising every year, the burden of this disease is steadily increasing globally [1]. Fortunately, the majority of prostate cancer patients have localized low-grade disease that is readily treated and cured with surgery or radiation; however, the subset of patients with more advanced disease have a poor prognosis, despite the development of novel treatment modalities. Specifically, these patients are first treated with androgen deprivation, which is usually very effective at controlling prostate cancer burden for 12–18 months. Unfortunately, at this point many patients develop a more aggressive, castration resistant prostate cancer (CRPC) [2]. Current treatments of CRPC, including abiraterone and enzalutamide, can extend life further; however, approximately 15%−25% CRPC patients are still non-responders, and even those who do respond still have limited long term survival [3–5]. Thus, there is a need to identify new therapeutic targets and develop new treatment methods for advanced prostate cancer, as well as to uncover novel biomarker for advanced disease that can be used to detect aggressive cancers earlier.

Comprehensive genomic studies in prostate cancer have identified multiple recurrent genomic networks, including androgen receptor signaling, kinase signaling, and DNA repair pathways, that are altered during cancer initiation and progression [6]. With the advent of next-generation sequencing technologies, recent genomic studies uncovered recurrent mutations in several genes, including SPOP, FOXA1, MED12, IDH [7]. In addition to the well-known alterations in genomic regions involving 8p, 8q, 10q23, these studies have also revealed novel complex large-scale genomic alterations, common ETS translocations, and androgen receptor amplifications [8]. These altered genomic networks can exhibit profound influence on reprogramming prostate cancer cells to become more proliferative, invasive, and androgen independent. With more in-depth understanding of the prostate cancer genome, it has become increasingly important to identify key regulators and connectors between these networks to develop better cancer therapies. Furthermore, identifying the various pathologic gene signatures of prostate cancers has profoundly impacted the treatment and prognosis of a variety of other cancers, including but not limited to breast, colon, and pancreatic cancers [9–11].

Our group had previously shown that the scaffold molecule paxillin may play a critical role in prostate cancer progression [12]. Paxillin is well known as a cytoplasmic adapter protein. The major functions of paxillin are regulating membrane and cytoplasmic structures at focal adhesions, as well as mediating kinase signaling throughout the plasma membrane and cytoplasm [13]. Paxillin belongs to the LIM domain protein family, which, among other transcriptional modulators, includes the androgen receptor (AR) coregulator Hic-5 [14]. In addition to its role outside of the nucleus, recent studies indicate that paxillin is also localized and has an important role in the nucleus [15–18]. Our group has reported that paxillin played a role in both extranuclear and nuclear signaling in prostate cancer cells. Outside of the nucleus, paxillin serves a regulator of cytoplasmic ERK signaling in response to both androgen and growth factors. Inside the nucleus, paxillin may serve as a mediator of AR- and ERK-mediate transcription [19–21]. In fact, we found that paxillin served as a critical liaison between extanuclear and intranuclar AR and ERK signaling. Finally, we showed that paxillin was overexpressed in human prostate cancer tumor microarrays, suggesting that it may serve as an important biomarker for prostate cancer.

Here we take a closer look at the genomic actions of paxillin in prostate cancer. The goal of this study is to create an atlas of paxillin-regulated genes and networks in several different prostate cancer cell lines that can be used by researchers to uncover novel pathways that might serve as potential diagnostic markers therapeutic targets. Our findings demonstrate that paxillin regulates a network of androgen responsive genes in androgen dependent cell lines that may be related to hormone resistance. In addition, we find that development of castration resistance significantly alters the network of androgen responsive genes, as well as the role of paxillin in regulating these genes. Finally, paxillin regulates pro-proliferative and anti-apoptotic genes in both androgen responsive and castration resistant prostate cancer cells, which may contribute accelerated cell proliferation and tumor progression. We conclude that paxillin is a broad regulator of prostate cancer genomic programming and may play a critical role in regulating tumor progression in response to androgens and other growth factors.

Materials and Methods

Cell lines

PC3(ATCC), LNCaP (ATCC), C4–2((from Ganesh Raj, University of Texas Southwestern, Dallas, TX) cells are cultured in RPMI media (Gibco, Gaithersburg, MD) supplied with 10% Fetal Bovine Serum (Seradigm, Salt Lake City, UT) and 1% penicillin -streptavidin (Gibco, Gaithersburg, MD). RWPE-1 cells (ATCC, Manassas, VA) were cultured in keratinocyte media supplied with prequalified human recombinant Epidermal Growth Factor 1–53 (EGF 1–53) 5 ng/mL and Bovine Pituitary Extract (BPE) 0.05 mg/mL (Invitrogen, Carlsbad, CA). Cells were maintained at 37 °C, 95% air, and 5% CO2. Experiments were performed with cells below passage 25. PC3, LNCaP, and C4–2 cells were verified by ATCC using STR profiling.

siRNA transfection

SMARTpool siRNA against paxillin (ON-TARGETplus human PXN siRNA, L-005163–00-005) and a non-targeted siRNA pool (ON-TARGETplus Control siRNA, D-001810–01-20) were used in knockdown experiments (Dharmacon, Lafayette, CO). DharmaFECT2/3 was used as the transfection reagent and transfection were performed as described in the instructions. LNCaP and C4–2 cells were transfected with siRNA for 48 hours, followed by overnight serum starvation and 24hours of 2.5nM DHT treatment, PC3 cells were transfected with siRNA for 72 hours, and RNA was isolated for the subsequent experiments. Paxillin and GFP plasmids in the vector of pcDNA3.1 were used in paxillin overexpression experiments and RWPE-1 cells were transfected with lipofectamine 3000 (Invitrogen, Carlsbad, CA) for 72 hours.

Western Blot

Cells were lysed in Mammalian Cell lysis buffer (Abcam, ab179835, Cambridge, MA) for protein isolation and proteins were separated by 12.5% SDS-PAGE gel (BioRad, Hercules, CA). Western blots were performed as described previously [12, 22, 23]. Primary antibodies used were as follows: anti-mouse-Paxillin (1:1000)(Becton Dickinson, Franklin Lakes, NJ), anti-rabbit-PARP-1(1:1000)(Cell Signaling, Danvers, MA). Secondary antibodies included Goat-anti-mouse IgG (Biorad, #170–6516, Hercules, CA) and Goat-anti-Rabbit IgG (Biorad, #170–6515)(1:4000).

Quantitative PCR (qPCR)

RNA was isolated with E.Z.N.A total RNA Kit I (OMEGA, #R6834–02, N orcross, GA) according to manufacturer’s instructions. Taqman RNA-to-Ct 1 step kit (Applied Biosystems, #4392938, Beverly, MA) was used for quantitative-PCR. Taqman primer of PXN (Hs01104424), CCND1 (Hs99999004), RB1 (Hs01078066), E2F1 (Hs00153451), TYMS (Hs00426586), CASP1 (Hs00354836), TNFSF10 (Hs00921974), KLK3 (Hs02576345), NKX3.1 (Hs00171834) were used for experiments and the human GAPDH primer (Hs03929097) was used as an internal control (Applied Biosystems, Beverly, MA). Delta delta Ct was calculated and relative gene expression was assessed by comparing to the control group as indicated.

Bromodeoxyurinidine (BrdU) Proliferation assay

After 72 hours of transfection, BrdU (Cell signaling, #6813, Danvers, MA) was added to the same wells where transfections were conducted, and cells were incubated for 2 hours at 37 °C. Without washing, cells were then fixed with 4% paraformaldehyde for 30 minutes. After rinsing with PBS, 2N HCI was added to the fixed cells for 30 minutes to denature the DNA. Nonspecific epitopes were blocked by treating cells with 5% normal horse serum prepared in PBS with 0.2% Triton X-100 for 2 hours. Samples were then stained with anti-BrdU antibody overnight, followed by incubation with either HRP-conjugated secondary antibody for absorbance reading as described in manufacture’s protocol, or with fluorescent secondary antibody (1:200) (Texas Red-goat-anti-mouse, Invitrogen, #1848474, Carlsbad, CA) for fluorescent imaging. Hoechst was used for nuclear staining. Photos were merged and processed using ImageJ (NIH, Bethesda, MD).

Cell Apoptosis assay

PC3 and LNCaP cells were stained with Annexin V and PI to identify apoptotic cells according to the manufacturer’s protocol. Annexin V positive (marker for apoptosis) and propidium iodide (PI) negative (marker for positive viability) cells were considered to be the apoptotic population and the percentage of apoptotic cells was calculated using FlowJo software.

Cell cycle analysis

PC3, LNCaP and C4–2 cells were transfected with siRNA for 72 hours and fixed with cold 70% ethanol for 30 minutes at 4 °C, followed by washing with PBS and ribonuclease A treatment (500ng/mL) for 30 minutes to eliminate RNA. Cells were then stained with 100 ul of PI for 15 minutes in the dark. An LSRII flow cytometer from BD was used for the experiments and FlowJo software was used for data analysis. Cell populations in G0G1, S, G2 phases were identified by PI signal and quantified as percentage of all cell populations.

RNAseq analysis

Total RNA was isolated using the RNeasy Plus Kit (Qiagen, Gaitherburg, MD). RNAseq was performed by the UR Genomics Research Center. Illumina compatible library construction was performed using the TruSeq Stranded mRNA Sample Preparation Kit (Illumina, San Diego, CA, USA). Amplified libraries were hybridized to the Illumina single-end flow cell and amplified using the cBot (Illumina) at a concentration of 8 pM per lane. Single-end reads of 100 nt were generated per sample and aligned to the human genome. For the RNAseq experiment of PC3 cells, differential expression gene list was generated by Cufflinks-2.0.2 (cuffdiff), and a q value of 0.05 was used as the cutoff threshold [24] [25]. For the RNAseq experiments of LNCaP and C4–2 cells, differential expression gene lists were generated by R-DESeq2, and a p adjusted value of 0.05 was used as the cutoff threshold [26]. Secondary analysis was conducted using online software, Ingenuity Pathway Analysis (Qiagen, Gaitherburg, MD) and GeneSet Enrichment Analysis (GSEA) [27]. The entire RNA-seq analysis is available on GEO (GSE131731).

Statistics

Group differences were analyzed using standard two-tailed Student’s t-tests. Experiments were done three times with triplicate samples in each experiment. Values are expressed as mean +/− standard error of the mean (s.e.m.). P < 0.05 was considered statistically significant.

Results

Paxillin regulates many androgen-responsive genes, including some related to endocrine therapy resistance, in an androgen-dependent prostate cancer cell line

Since we previously described that paxillin mediates extranuclear AR signaling pathway, we first examined paxillin’s role in regulating intranuclear AR actions. We therefore performed RNA-seq analysis on control and paxillin-depleted LNCaP cells treated with 2.5nM DHT. Cells were transfected with control or anti-paxillin siRNA pools for 48hrs and then treated with either vehicle or 2.5nM DHT for 24 hours. The paxillin knockdown efficiency was validated by quantitative PCR (Figure 1B). The entire RNA-seq analysis is available on GEO (GSE131731). Using a cutoff adjusted p value less than 0.05, we identified 1128 genes that were regulated by DHT in LNCaP cells treated with nonspecific siRNA. Surprisingly, we found only 137 genes that were regulated by DHT in LNCaP cells treated with siRNA directed against paxillin (Figure 1C). This result therefore suggests that the majority of androgen-regulated genes in LNCaP cells depend on the presence of paxillin. To further investigate the composition of these paxillin- and DHT-dependent genes, the 1017 paxillin- and androgen-dependent genes were analyzed by the GeneSet Enrichment Analysis (GSEA) online platform. Interestingly, a large subset of the these paxillin-dependent genes are related to endocrine therapy resistance in prostate or breast cancer (Figure 1D). A detailed list of those endocrine therapy resistant gene groups is shown in Supplemental Figure 1. Importantly, many of the “signature” androgen-responsive genes, including KLK3(PSA) and NKX3.1, are present in the overlapping group that includes 111 genes, suggesting that the androgen regulated transcription of these genes is not dependent on paxillin. The full list of the 111 DHT regulated paxillin-independent genes is shown in Supplemental Figure 2. To further confirm that KLK3 and NKX3.1 expression is not dependent on paxillin, we performed qPCR and calculated relative gene expression in LNCaP cell treated with DHT with or without knocking down paxillin (Figure 2). Although the overall KLK3 and NKX3.1 expression level is reduced in the paxillin knockdown cells, the magnitude of DHT induction of these two genes remains the same in control versus paxillin-knockdown LNCaP cells (approximately 5 - fold), confirming that while paxillin appears to enhance their overall expression, it is not involved in the regulating effects of AR genomic signaling on their expression.

Figure 1. Paxillin regulates the expression of androgen responsive genes in LNCaP cells.

Figure 1.

Open in a new tab

A. Experimental design of RNA-seq studies in LNCaP and C4–2 cells. LNCaP or C4–2 cells were treated with either paxillin-target siRNA (PXN siRNA) or non-specific target siRNA (NSP siRNA) for 48 hours followed by overnight serum starvation and 2.5nM DHT treatment for 24 hours. Each group includes three pooled replicates, and each experiment was performed three times. B. Paxillin knockdown efficiency in LNCaP cells was validated by qPCR before RNA-seq experiments (p<0.01 by t-test). Protein expression was also reduced by over 80% (not shown). C-D. Bioinformatic analysis of paxillin-dependent androgen-responsive genes in LNCaP cells. DHT regulated genes in NSP siRNA transfected LNCaP cells (yellow) were compared with PXN siRNA transfected cells (blue) in the Venn diagram. The 1,017 paxillin-dependent androgen-responsive genes were further analyzed by Gene Set Enrichment Analysis (GSEA) database and the overlapping gene groups are shown in the pie chart (D).

Figure 2. qPCR validation of paxillin independence of signature androgen-responsive genes.

Figure 2.

Open in a new tab

LNCaP cells were transfected with either NSP siRNA or PXN siRNA and reduced paxillin expression level are confirmed in the left panels. KLK3 and NKX3.1 relative expression levels were measured after DHT treatment for 24 hours. (compared to untreated, *P<0.05, N=3 by t-test). Panel A represents paxillin (PXN) knockdown for panel B, and panel C represents PXN knockdown for panel D. All values were normalized to GAPDH mRNA expression. ANOVA followed by comparison tests for B and D confirmed that, while DHT promoted significant increases in mRNA expression relative to vehicle under all conditions (p<0.05), no significant difference between the magnitude of these inductions were noted between NSP and PXN siRNA conditions.

Paxillin-regulated gene transcription is significantly different in androgen-sensitive versus androgen-responsive but resistant prostate cancer cell lines

To more carefully investigate paxillin-dependent androgen-responsive genes, we conducted a similar experiment as described in Figure 1 in LNCaP-derived castration-resistant C4–2 cells. Notably, The C4–2 cell line is a subline derived from a LNCaP xenograft co-culture with fibroblast cells in castrated nude mice, and it is therefore considered to be androgen sensitive (AR positive) but castration resistant (can grow in the absence of androgen). In contrast to LNCaP cells, we found considerably more androgen-regulated genes using the same analysis (6302 in C4–2 versus 1,128 in LNCaP). With paxillin knockdown, we uncovered 1,842 paxillin-dependent and androgen-responsive genes, 4,460 paxillin-independent but androgen-responsive genes, and 1,374 genes only regulated by androgen in the absence of paxillin (Figure 3A). These results suggest that, although C4–2 cells grown in the absence of androgens (and are therefore “androgen-independent”), they are in fact equally or more responsive to androgens compared to LNCaP cells. By comparing the 1,842 paxillin-dependent and androgen-responsive genes in C4–2 cells to the 1,017 paxillin-dependent and androgen-responsive genes in LNCaP cells, we discovered that only 85 genes overlapped (Figure 3C), meaning that there is a dramatic shift in paxillin’s role in regulating androgen responses in C4–2 cells. In contrast, comparing the 111 androgen-responsive but paxillin-independent genes from LNCaP with the 4,460 androgen responsive paxillin-independent genes from C4–2 uncovered the finding that most of androgen responsive paxillin-independent genes (91 of 111) in LNCaP are still paxillin-independent in C4–2 cells (Figure 3B), including the “signature” androgen-responsive genes, such as KLK3(PSA) and NKX3.1. When specifically comparing the aforementioned endocrine therapy-resistant genes expressed in LNCaP cells, we found that only 10 of the 131 genes were still both paxillin-dependent and androgen-responsive in C4–2 cells. In fact, 43/131 of the endocrine therapy resistant gene observed in LNCaP cells were no longer androgen responsive in C4–2 cells. Together, these data suggest that there is a dramatic switch in C4–2 cells whereby expression of endocrine therapy resistant genes become less dependent on androgens and paxillin.

Figure 3. Venn diagrams of paxillin-dependent and paxillin-independent androgen-responsive genes in LNCaP and C4–2.

Figure 3.

Open in a new tab

A. Comparison of DHT-regulated genes in NSP siRNA transfected C4–2 cells (yellow) with PXN siRNA transfected C4–2 cells (blue). Knockdown of paxillin is confirmed by qPCR on the right. B. Comparison of DHT-regulated paxillin-independent genes between LNCaP (yellow) and C4–2 (blue) cells. C. Comparison of DHT-regulated paxillin-dependent genes between LNCaP (yellow) and C4–2 (blue) cells.

Paxillin regulates pro-proliferative, pro-apoptotic gene networks in castration resistant cells

Previous work from our laboratory indicated that paxillin might regulate proliferation in prostate cancer cells independent of androgen signaling. To further investigate the paxillin-regulated whole genomic network without the interference of androgen receptor signaling, we performed RNA-seq analysis in paxillin siRNA transfected AR negative PC3 cells compared to paxillin intact PC3 cells. Depletion of paxillin dramatically modified mRNA expression in PC3 cells as shown in the heat map of selected genes in Figure 4A. Figure 4B shows a list of genes that were most upregulated by paxillin (expression decreased with paxillin knockdown), while Figure 4C shows a list of genes mostly down-regulated by paxillin (expression increased with paxillin knockdown). Data analysis of paxillin-dependent genes by IPA suggested that genomic networks in cancer, cell cycle, cellular growth and proliferation, and cellular development were most upregulated by paxillin expression. In contrast, networks related to cell death and apoptosis were most suppressed by paxillin (Figure 5A).

Figure 4. RNA-seq analysis of paxillin-dependent, differentially expressed genes in PC3 cells.

Figure 4.

Open in a new tab

A. Heat map of differentially expressed genes between NSP siRNA and PXN siRNA transfected PC3 cells. Red=upregulated and Green=downregulated by paxillin. B. List of top 20 upregulated genes by paxillin (expression went down with paxillin knockdown) with log2 fold change (log2FC) and q value (False Discovery Rate, FDR). C. List of top 20 downregulated genes by paxillin (expression went up with paxillin knockdown) with log2 fold change and q value. q value less than 1E-20 are displayed as 0 by the software.

Figure 5. Ingenuity pathway analysis of paxillin-dependent genes in PC3.

Figure 5.

Figure 5.

Open in a new tab

A. Using cutoffs described in the methods section, disease and function analysis of paxillin-dependent genes are shown with highlighted major gene functions (red rectangles). Orange color represents genes upregulated by paxillin and blue color represents genes downregulated by paxillin. Box size reflects gene numbers. B-C. Major genomic network analysis of paxillin dependent genes. Red represents genes markedly upregulated by paxillin, pink represents genes moderately upregulated by paxillin, and green represents genes downregulated by paxillin. Representative DNA replication genes include TYMS (B), representative apoptotic genes include CASP1 (C). D. The cyclinD1 pathway is predicted to be activated by paxillin. Orange colored genes such as CCND1, E2F1, and MYC promote this pathway and are upregulated by paxillin, while blue colored genes such as RB1 and TP53 suppress this pathway and are downregulated by paxillin. Genes shown in Figure 6 include CCND1, E2F1, Rb, TYMS, CASP1 and are labeled with red stars.

A closer look at the cell cycle and DNA replication network (Figure 5B) revealed the upregulation of several cyclins, cyclin-dependent kinases, and E2F family members. Genes upregulated by paxillin are noted in pink. Cell death and survival gene pathways (Figure 5C) were generally downregulated by paxillin (green). Notably, multiple genes in the CyclinD-Rb-E2F pathway, which is involved in cell cycle progression, were regulated by paxillin, with predicted activation of cell cycle promoters cyclinD1(CCND1), E2F1, E2F3, and MYC (orange), and inhibition of tumor suppressors retinoblastoma-1(RB1), retinoblastoma-like-1, p53(TP42) and cyclin-dependent kinase inhibitor 1(CDKN1A) (blue), as shown in Figure 5D.

To validate the RNA-seq data, we chose several of the genes that were most dramatically regulated by paxillin (some are marked by red stars in Figure 5B–D) and performed quantitative PCR in PC3 cells. As expected, paxillin knockdown reduced the expression of cell cycle promoter genes CCND1, E2F1, and TYMS while activating pro-apoptotic genes CASP1 and TNFSF10 (Figure 6A). Quantitative PCR in LNCaP and C4–2 cells demonstrated similar paxillin-mediated regulation of these genes (Figures 6B&C), confirming that our findings in PC3 cells extend to other human prostate cancer cell lines, and that paxillin’s positive effect on cell cycle genes and suppressive effects on pro-apoptotic genes might represent generalized genomic effects of paxillin.

Figure 6. qPCR validation of paxillin-dependent proliferative and pro-apoptotic genes.

Figure 6.

Open in a new tab

A-C. PC3, LNCaP, or C4–2 cells were transfected and either NSP siRNA (black bar) or PXN siRNA (white bar), and Ct values of PXN, CCND1, RB1, E2F1, TYMS, CASP1, and TNFS10 were examined by qPCR and relative gene expression were calculated. *p<0.05 relative to NSP siRNA (n=3). All values were normalized to GAPDH mRNA expression.

Notably, using the same cutoffs as in Fig 4A (PC3 cells, log2-fold change>1.5), heat maps of the RNASeq data comparing control versus paxillin knock-down in unstimulated LNCaP and C4–2 cells revealed hundreds of PXN-regulated mRNAs in LNCaP cells and far fewer in C4–2 cells (supplemental figure 3). Some of the pro-proliferative and pro-apoptotic genes demonstrated by qPCR in figure 6 were present in the LNCaP heat maps (e.g., TYMS and E2F1) but not in the C4–2 maps. When the selection fold-change was loosened, most genes from figure 6 and 7 appeared with both cell lines. These data again emphasize that paxillin may be functioning quite differently in castration sensitive (LNCaP) versus resistant (C4–2) cells. Furthermore, the differences in RNASeq versus qPCR data remind us of the limitations of RNASeq and the need for confirmation by other quantitative analyses.

Figure 7. Paxillin promotes prostate cancer cell proliferation.

Figure 7.

Open in a new tab

A-C. BrdU assay was used to evaluate cell proliferation rate in PC3, LNCaP and C4–2 cells. Absorbance at 450nm in NSP/PXN siRNA transfected cells were measured and calculated. D. RWPE-1 cells were transfected with WT-PXN or control GFP plasmid and BrdU incorporation was measured and calculated. *p<0.05 relative to NSP siRNA (A-C) of control GFP transfected (D) (n=3). Paxillin knockdown and rescue was confirmed by Western blot (inset for A-D). E. PC3, LNCaP or C4–2 cells were transfected with NSP versus PXN siRNA and stained with BrdU antibody (red) for immunofluorescence imaging. Hoechst (blue) was used to stain cell nuclei. This experiment was repeated 3 times with similar results.

Paxillin promotes cell proliferation and cell cycle progression

To determine whether the RNASeq data reflects physiology, we next sought to examine paxillin’s effect on prostate cancer cell proliferation. While paxillin’s role in cellular movement has been shown in various models [28, 29], and our previous work suggested that paxillin improves prostate cancer cell survival (using MTT assays) and tumor progression [12], no studies to date have directly measured paxillin effects on proliferation and cell cycle. We used siRNAs directed against paxillin mRNA to knock down paxillin expression in PC3, LNCaP and C4–2 cells, using non-specific siRNA as a control, and measured the effects of paxillin knockdown on proliferation using BrdU assays. In all three cell lines, we measured significant reduction of cell proliferation in paxillin knockdown cells compared to control cells (Figure 7A–C). In contrast, when we transiently overexpressed paxillin in the benign prostate epithelial cell line, RWPE-1 (which normally expresses less paxillin than the tumor cell lines), we noted an increase in cell proliferation relative to cells transfected with a plasmid expressing GFP (Figure 7D). To confirm these results visually, we used immunofluorescence of BrdU staining on the three prostate cancer cell lines, and again saw marked reduction in BrdU incorporation upon paxillin knockdown (Figure 7E).

Since paxillin significantly regulates genes involved in the cell cycle, we next investigated whether cell cycle progression was affected by paxillin depletion. Using flow cytometry, PXN siRNA or NSP siRNA treated PC3 cells were stained with PI and each phase of cell cycle was monitored and recorded by flow cytometry (Figure 8A). In this experiment, paxillin knockdown resulted in a greater number of cells resting in the G0G1 phase, which was quantitatively validated (Figure 8B). Cell with reduced paxillin expression were significantly more likely to be in G0G1 phase, with a concomitant reduction in the number of cells in S phase. Together these data confirm that paxillin might promote prostate cancer cell proliferation by augmenting the progression from G0G1 to S.

Figure 8. Depletion of paxillin induces cell cycle arrest in PC3 cells.

Figure 8.

Open in a new tab

A. Cell cycle was measured by flow cytometry with PI staining in PC3 cells transfected with either NSP siRNA or PXN siRNA. Each cell cycle phase was measured and recorded by FACS. B. Percentage of cells in each cell cycle was calculated and compared. *p<0.05 relative to NSP siRNA (n=3).

Paxillin manifests minimal effect on cell apoptosis

After using paxillin knockdown to demonstrate the physiologic relevance of the RNAseq data regarding proliferation, we next used flow cytometry to determine whether paxillin-knockdown would promote apoptosis, as suggested by the RNASeq data. We again used siRNA to suppress paxillin expression in PC3 and LNCaP prostate cancer cells. Annexin V and PI double staining were then used to distinguish apoptotic cells from live (non-apoptotic) and dead cells. In Figure 9, the lower right quadrant with Annexin V positive and PI negative staining represents the apoptotic cell population. In both PC3 and LNCaP cells, there was no significant increase in apoptotic cells after paxillin knockdown. Thus, while our RNA-seq data suggest that paxillin regulates some apoptotic genes and apoptotic pathways in prostate cancer cell lines, paxillin expression in fact seems to have minimal effect on cell apoptosis using physiologic assays, suggesting that paxillin’s effects on proliferation and cell cycle might be more significant in-vivo.

Figure 9. Paxillin has minimal effect on cell apoptosis.

Figure 9.

Open in a new tab

PC3 and LNCaP cells were transfected with NSP/PXN siRNA and double stained with Annexin V and PI. Percentage of apoptotic cells (Annexin V positive and PI negative) were measured and analyzed by FACS. Minimal difference of apoptotic cell population was seen between NSP siRNA and PXN siRNA.

Discussion

With all the advancements in sequencing technologies, the importance of global gene expression studies, especially in disease models such as cancer, has attracted much attention in recent years. Unlike breast or colon cancer, there is no clearly dominant gene signature that can be used to explain the mechanism of tumor progression or predict the prognosis of prostate cancer. However, genome wide analysis indicates that both genetic and epigenetic alterations are important contributing factors in prostate carcinogenesis. One androgen responsive gene, NKX3.1, which is a critical regulator of epithelial differentiation and stem cell function, has been reported to be down regulated in up to 85% of high-grade PIN lesions and adenocarcinomas [30]. In contrast, the MYC oncogene is overexpressed in many PIN lesions and most carcinomas [31]. Other molecular pathways, such as PTEN, Akt/mTOR, and MAPK signaling, also have been implicated in the initiation and/or development of prostate cancer [32]. Together, these data and others suggest that investigation of the whole genomic alterations, rather than examination of single genes or even individual signaling pathways, may be a more useful approach in understanding the complex nature of prostate cancer development and progression. In this paper, we described the genomic network of one molecule, paxillin, which we have found to be up-regulated in prostate cancer and to potentially play an important role in prostate cancer progression.

Paxillin was originally purified as a cytoskeletal protein that binds vinculin and Focal Adhesion Kinase (FAK). The FAK mediated paxillin activation has been implicated as a critical step in the formation of nascent focal complex and turnover of mature focal adhesions [33]. Paxillin contains five amino terminal LD domains and four C-terminal LIM domains that interact with a variety of proteins. It also contains multiple phosphorylation sites within or outside of the abovementioned domains that can be modified by various kinases. Therefore, paxillin serves as a scaffolding protein at the membrane and in the cytoplasm that modulates the spatial and temporary signal transductions in cells. Paxillin dysregulation or mutation has been reported to play a critical role in a variety of cancers. For example, paxillin mutation of A127T is associated with lung cancer progression and metastasis [34]. In addition, upregulation of paxillin had been reported in brain and CNS tumors, breast cancer, cervical cancer, and ovarian cancer [35–38]. Paxillin not only functions directly in tumor cells, but may also contribute to cancer progression by modulating the tumor microenvironment. Evidence suggests that, in colon cancer, paxillin is up-regulated in tumor associated macrophages, resulting in enhanced M2 macrophage polarization, ultimately leading to increased tumor cell proliferation and invasion [39]. Although paxillin plays important roles in cancer via extranuclear actions, so far, there are few studies showing a nuclear function of paxillin. One example comes from Hozak and colleagues, who showed that paxillin directly regulates IGF2 and H19 gene clusters during fetal development [16].

Our group previously reported that the expression of certain DHT-induced AR target gene expression were reduced in the absence of paxillin [12]. However, our studies here demonstrate that, while depletion of paxillin indeed markedly suppresses baseline PSA and Nkx3.1 gene expression, this reduction occurs without androgen stimulation. In fact, we observe that the relative DHT-induced gene expression remains the same in both paxillin knockdown cells and paxillin intact cells (Figure 2), suggesting that expression of these well-established genes in response to androgen is in fact not paxillin dependent.

In contrast to these “signature” androgen-dependent, paxillin-independent genes, our RNA-seq analysis of paxillin- and androgen- regulated genes in LNCaP cell and C4–2 cells revealed profoundly different transcriptomes between these two cell lines. The LNCaP cell line was established from a lymph node metastatic lesion of human prostatic adenocarcinoma [40], and represents androgen responsive phenotype. The C4–2 cell line is a subline derived from LNCaP cells after undergoing co-culture with fibroblast cells in castrated nude mice, which resulted in a castration resistant prostate cancer cell model. Although C4–2 is derived from LNCaP, they display disparate morphology and growth capacity. It has been reported that exome sequencing detected 2,188 and 3,840 mutations in LNCaP and C4–2B (another variant of the C4–2 cell line) cells respectively. The transcriptome analysis also revealed that 457 genes are upregulated and 246 genes are downregulated in C4–2B compared to LNCaP cells, and the most important changes were in the focal adhesion and ECM-receptor interaction pathways [41]. Therefore, since paxillin functions in both of these pathways, it makes sense that it may possess different characteristics and regulate a divergent transcriptome in C4–2 cells compared to LNCaP, which in turn could explain why so many of the paxillin-dependent genes in LNCaP cells were no longer paxillin-dependent in C4–2 cells (Figure 3C). Interestingly, not only are most of the endocrine therapy resistant genes no longer paxillin dependent in the C4–2 cell lines (121 out of 131), but also one third (43/131) of the endocrine therapy resistant genes seen in LNCaP cells were no longer androgen responsive at all in C4–2 cells, suggesting that evolution from androgen sensitive to the castration resistant status dramatically changed the whole genomic programming in androgen signaling.

While direct comparison of androgen-sensitive genes in LNCaP versus C4–2 revealed many differences, one common theme in all three cells lines tested (PC3, LNCaP, and C4–2) is that paxillin promotes expression of proliferative genes while it suppresses expression of some pro-apoptotic genes. Accordingly, by analyzing the RNA-seq data of the androgen-independent PC3 cell, we identified several paxillin regulated pathways and signature genes associated with cell proliferation and apoptosis. Importantly, physiologic assays confirm paxillin’s importance in proliferation (Figure 8, 9) but not apoptosis (Figure 9), indicating that paxillin’s proliferative effects are likely dominant in-vivo, and illustrating the importance of following up the pathway analysis with functional studies.

One gene that was markedly upregulated by paxillin is TYMS, which encodes thymidylate synthase, an essential enzyme involved in the DNA repair pathway. TYMS has been shown to be associated with 5 fluorouracil chemoresistance in various cancers. Colorectal cancer patients who harbor genetic variants of TYMS are susceptible to an earlier relapse after 5 fluorouracil-based chemotherapy [42]. Interestingly, a recent study of prostate tumor progression demonstrated that expression of AR and TYMS were frequently observed in circulating tumor cells (CTCs) of patients with advanced metastatic disease [43]. Our qPCR data confirmed that paxillin upregulated TYMS gene expression in prostate cancer cells, which might further indicate that overexpression of paxillin was associated with poor prognosis in prostate cancer.

While TYMS represents an individual gene seemingly enhanced by paxillin, the CyclinD1-Rb-E2F1 pathway appears to be the major cell proliferation pathway upregulated by paxillin. CyclinD1 belongs to the cyclin family, which is a protein family mainly involved in cell cycle regulation. Unlike the other two D-cyclins, cyclinD1 overexpression has been observed in a variety of cancers, including non-small cell lung cancer, head and neck cancer, colorectal cancer and breast cancer [44–47]. Although studies demonstrated that cyclinD1 overexpression might not be the single driver of oncogenesis, emerging evidence suggests that nuclear retention of cyclinD1 may play a critical role in its oncogenicity [48]. Interestingly, a splice variant of cyclinD1, cyclinD1b is induced during disease progression in various cancer types, and it cooperates with AR to modulate a large transcriptional network that is associated with a metastasis phenotype in prostate cancer [49, 50]. Furthermore, cyclinD1 is regulated by another paxillin family member, Hic-5, which competes with cyclinD1 for CRM1 during nuclear export and results in cyclinD1 nuclear retention and promotes anchorage independent cell growth and survival [51]. In addition, our group previously showed that, in clinical samples, both paxillin and cyclinD1 are overexpressed in prostate cancer tissue compared to normal prostate tissue [12]. Together, our data demonstrate that cyclinD1 may be a key functional molecule modulated by paxillin genomic signaling that markedly affects disease progression. Finally, paxillin also has been reported to interact with CRM1 during nuclear transportation, thus it is possible that it may utilize a similar method of regulating cyclinD1 associated gene transcription and cell proliferation as Hic-5 [15].

In this study, we use RNASeq analysis to reveal the complex transcriptional landscape affected by paxillin in cell line models. By comparison of three large scale RNA-seq analysis, we provide extensive resources in the exploration of the paxillin transcriptome as well as androgen-regulated genomic networks. We validate some genes from our analysis via quantitative PCR in multiple prostate cancer cell lines, as well as paxillin’s effects on proliferation in the same cells. While we acknowledge the limitations of using cell lines versus human prostate cancer samples, we also recognize that the nuclear actions of paxillin are just beginning to be appreciated, and that the first step toward understanding the transcriptomic importance of paxillin in prostate or any cancer requires discovery and analysis in uniform cell line samples. Our hope is that this valuable resource may offer further opportunities to identify important gene clusters or gene signatures in prostate cancer that can be used to identify novel diagnostic or therapeutic targets to help treat human diseases.

Supplementary Material

1

Acknowledgements:

We would like to thank Jason Myers, M.S., Lead Bioinformatics Analyst in the University of Rochester Genomics Research Center for his help analyzing the RNA-Seq data. Funding for this work was supported by the National Institutes of Health R01GM101709 and R01CA193583.

Footnotes

Declaration of Interest: There is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • [1].Pishgar F, Ebrahimi H, Saeedi Moghaddam S, Fitzmaurice C, Amini E, Global, Regional and National Burden of Prostate Cancer, 1990 to 2015: Results from the Global Burden of Disease Study 2015, The Journal of urology 199(5) (2018) 1224–1232. [DOI] [PubMed] [Google Scholar]
  • [2].Harris WP, Mostaghel EA, Nelson PS, Montgomery B, Androgen deprivation therapy: progress in understanding mechanisms of resistance and optimizing androgen depletion, Nature clinical practice. Urology 6(2) (2009) 76–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [3].Crona DJ, Milowsky MI, Whang YE, Androgen receptor targeting drugs in castration-resistant prostate cancer and mechanisms of resistance, Clinical pharmacology and therapeutics 98(6) (2015) 582–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [4].Scott LJ, Enzalutamide: A Review in Castration-Resistant Prostate Cancer, Drugs 78(18) (2018) 1913–1924. [DOI] [PubMed] [Google Scholar]
  • [5].Gartrell BA, Saad F, Abiraterone in the management of castration-resistant prostate cancer prior to chemotherapy, Ther Adv Urol 7(4) (2015) 194–202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [6].Shen MM, Abate-Shen C, Molecular genetics of prostate cancer: new prospects for old challenges, Genes & development 24(18) (2010) 1967–2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [7].Barbieri CE, Baca SC, Lawrence MS, Demichelis F, Blattner M, Theurillat JP, White TA, Stojanov P, Van Allen E, Stransky N, Nickerson E, Chae SS, Boysen G, Auclair D, Onofrio RC, Park K, Kitabayashi N, MacDonald TY, Sheikh K, Vuong T, Guiducci C, Cibulskis K, Sivachenko A, Carter SL, Saksena G, Voet D, Hussain WM, Ramos AH, Winckler W, Redman MC, Ardlie K, Tewari AK, Mosquera JM, Rupp N, Wild PJ, Moch H, Morrissey C, Nelson PS, Kantoff PW, Gabriel SB, Golub TR, Meyerson M, Lander ES, Getz G, Rubin MA, Garraway LA, Exome sequencing identifies recurrent SPOP, FOXA1 and MED12 mutations in prostate cancer, Nature genetics 44(6) (2012) 685–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [8].Rubin MA, Demichelis F, The Genomics of Prostate Cancer: emerging understanding with technologic advances, Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc 31(S1) (2018) S1–11. [DOI] [PubMed] [Google Scholar]
  • [9].Glinsky GV, Krones-Herzig A, Glinskii AB, Malignancy-associated regions of transcriptional activation: gene expression profiling identifies common chromosomal regions of a recurrent transcriptional activation in human prostate, breast, ovarian, and colon cancers, Neoplasia (New York, N.Y.) 5(3) (2003) 218–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [10].Song K, Su W, Liu Y, Zhang J, Liang Q, Li N, Guan Q, He J, Bai X, Zhao W, Guo Z, Identification of genes with universally upregulated or downregulated expressions in colorectal cancer, Journal of gastroenterology and hepatology (2018). [DOI] [PubMed] [Google Scholar]
  • [11].Newhook TE, Blais EM, Lindberg JM, Adair SJ, Xin W, Lee JK, Papin JA, Parsons JT, Bauer TW, A thirteen-gene expression signature predicts survival of patients with pancreatic cancer and identifies new genes of interest, (1932–6203 (Electronic)). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [12].Sen A, De Castro I, Defranco DB, Deng FM, Melamed J, Kapur P, Raj GV, Rossi R, Hammes SR, Paxillin mediates extranuclear and intranuclear signaling in prostate cancer proliferation, The Journal of clinical investigation 122(7) (2012) 2469–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Schaller MD, Paxillin: a focal adhesion-associated adaptor protein, Oncogene 20(44) (2001) 6459–72. [DOI] [PubMed] [Google Scholar]
  • [14].Deakin NO, Pignatelli J, Turner CE, Diverse roles for the paxillin family of proteins in cancer, Genes & cancer 3(5–6) (2012) 362–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Fung HYJ, Fu S-C, Chook YM, Nuclear export receptor CRM1 recognizes diverse conformations in nuclear export signals, eLife 6 (2017) e23961. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [16].Marasek P, Dzijak R, Studenyak I, Fiserova J, Ulicna L, Novak P, Hozak P, Paxillin-dependent regulation of IGF2 and H19 gene cluster expression, Journal of cell science 128(16) (2015) 3106–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Woods AJ, Roberts MS, Choudhary J, Barry ST, Mazaki Y, Sabe H, Morley SJ, Critchley DR, Norman JC, Paxillin associates with poly(A)-binding protein 1 at the dense endoplasmic reticulum and the leading edge of migrating cells, The Journal of biological chemistry 277(8) (2002) 6428–37. [DOI] [PubMed] [Google Scholar]
  • [18].Sathe AR, Shivashankar GV, Sheetz MP, Nuclear transport of paxillin depends on focal adhesion dynamics and FAT domains, Journal of cell science 129(10) (2016) 1981–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [19].Rasar M, DeFranco DB, Hammes SR, Paxillin regulates steroid-triggered meiotic resumption in oocytes by enhancing an all-or-none positive feedback kinase loop, The Journal of biological chemistry 281(51) (2006) 39455–64. [DOI] [PubMed] [Google Scholar]
  • [20].Sen A, Prizant H, Light A, Biswas A, Hayes E, Lee HJ, Barad D, Gleicher N, Hammes SR, Androgens regulate ovarian follicular development by increasing follicle stimulating hormone receptor and microRNA-125b expression, Proceedings of the National Academy of Sciences of the United States of America 111(8) (2014) 3008–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Miedlich SU, Taya M, Young MR, Hammes SR, Paxillin and embryonic PolyAdenylation Binding Protein (ePABP) engage to regulate androgen-dependent Xenopus laevis oocyte maturation - A model of kinase-dependent regulation of protein expression, Molecular and cellular endocrinology 448 (2017) 87–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Ma X, Hayes E, Prizant H, Srivastava RK, Hammes SR, Sen A, Leptin-Induced CART (Cocaine- and Amphetamine-Regulated Transcript) Is a Novel Intraovarian Mediator of Obesity-Related Infertility in Females, Endocrinology 157(3) (2016) 1248–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].Evaul K, Hammes SR, Cross-talk between G protein-coupled and epidermal growth factor receptors regulates gonadotropin-mediated steroidogenesis in Leydig cells, The Journal of biological chemistry 283(41) (2008) 27525–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, van Baren MJ, Salzberg SL, Wold BJ, Pachter L, Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation, Nature biotechnology 28(5) (2010) 511–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Trapnell C, Hendrickson DG, Sauvageau M, Goff L, Rinn JL, Pachter L, Differential analysis of gene regulation at transcript resolution with RNA-seq, Nature biotechnology 31(1) (2013) 46–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [26].Love MI, Huber W, Anders S, Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2, Genome Biol 15(12) (2014) 550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [27].Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, Mesirov JP, Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles, Proceedings of the National Academy of Sciences 102(43) (2005) 15545–15550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [28].Panetti TS, Tyrosine phosphorylation of paxillin, FAK, and p130CAS: effects on cell spreading and migration, Frontiers in bioscience : a journal and virtual library 7 (2002) d143–50. [DOI] [PubMed] [Google Scholar]
  • [29].Schaller MD, FAK and paxillin: regulators of N-cadherin adhesion and inhibitors of cell migration?, The Journal of cell biology 166(2) (2004) 157–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [30].Ornstein DK, Cinquanta M, Weiler S, Duray PH, Emmert-Buck MR, Vocke CD, Linehan WM, Ferretti JA, Expression studies and mutational analysis of the androgen regulated homeobox gene NKX3.1 in benign and malignant prostate epithelium, The Journal of urology 165(4) (2001) 1329–34. [PubMed] [Google Scholar]
  • [31].Gurel B, Iwata T, Koh CM, Jenkins RB, Lan F, Van Dang C, Hicks JL, Morgan J, Cornish TC, Sutcliffe S, Isaacs WB, Luo J, De Marzo AM, Nuclear MYC protein overexpression is an early alteration in human prostate carcinogenesis, Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc 21(9) (2008) 1156–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [32].Mitchell T, Neal DE, The genomic evolution of human prostate cancer, British journal of cancer 113(2) (2015) 193–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [33].Zaidel-Bar R, Milo R, Kam Z, Geiger B, A paxillin tyrosine phosphorylation switch regulates the assembly and form of cell-matrix adhesions, Journal of cell science 120(Pt 1) (2007) 137–48. [DOI] [PubMed] [Google Scholar]
  • [34].Mackinnon AC, Tretiakova M, Henderson L, Mehta RG, Yan BC, Joseph L, Krausz T, Husain AN, Reid ME, Salgia R, Paxillin expression and amplification in early lung lesions of high-risk patients, lung adenocarcinoma and metastatic disease, Journal of clinical pathology 64(1) (2011) 16–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [35].Sun LH, Yang FQ, Zhang CB, Wu YP, Liang JS, Jin S, Wang Z, Wang HJ, Bao ZS, Yang ZX, Jiang T, Overexpression of Paxillin Correlates with Tumor Progression and Predicts Poor Survival in Glioblastoma, CNS neuroscience & therapeutics 23(1) (2017) 69–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [36].Vadlamudi R, Adam L, Tseng B, Costa L, Kumar R, Transcriptional up-regulation of paxillin expression by heregulin in human breast cancer cells, Cancer Res 59(12) (1999) 2843–6. [PubMed] [Google Scholar]
  • [37].Liu Q, Wang J, Tang M, Chen L, Qi X, Li J, Yu J, Qiu H, Wang Y, The overexpression of PXN promotes tumor progression and leads to radioresistance in cervical cancer, Future oncology (London, England) 14(3) (2018) 241–253. [DOI] [PubMed] [Google Scholar]
  • [38].Deng B, Zhang S, Miao Y, Han Z, Zhang X, Wen F, Zhang Y, Adrenomedullin expression in epithelial ovarian cancers and promotes H08910 cell migration associated with upregulating integrin alpha5beta1 and phosphorylating FAK and paxillin, Journal of experimental & clinical cancer research : CR 31 (2012) 19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [39].Zhang LL, Zhang LF, Shi YB, Down-regulated paxillin suppresses cell proliferation and invasion by inhibiting M2 macrophage polarization in colon cancer, Biological chemistry 399(11) (2018) 1285–1295. [DOI] [PubMed] [Google Scholar]
  • [40].Horoszewicz JS, Leong SS, Kawinski E, Karr JP, Rosenthal H, Chu TM, Mirand EA, Murphy GP, LNCaP model of human prostatic carcinoma, Cancer Res 43(4) (1983) 1809–18. [PubMed] [Google Scholar]
  • [41].Spans L, Helsen C, Clinckemalie L, Van den Broeck T, Prekovic S, Joniau S, Lerut E, Claessens F, Comparative genomic and transcriptomic analyses of LNCaP and C4–2B prostate cancer cell lines, PloS one 9(2) (2014) e90002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [42].Zawiah M, Yousef AM, Kadi T, Yousef M, Majdalawi K, Al-Yacoub S, Al-Hiary R, Tantawi D, Mukred R, Ajaj AR, Early disease relapse in a patient with colorectal cancer who harbors genetic variants of DPYD, TYMS, MTHFR and DHFR after treatment with 5-fluorouracil-based chemotherapy, Drug metabolism and personalized therapy 33(4) (2018) 201–205. [DOI] [PubMed] [Google Scholar]
  • [43].Russo GI, Bier S, Hennenlotter J, Beger G, Pavlenco L, van de Flierdt J, Hauch S, Maas M, Walz S, Rausch S, Bedke J, Morgia G, Stenzl A, Todenhofer T, Expression of tumour progression-associated genes in circulating tumour cells of patients at different stages of prostate cancer, BJU international 122(1) (2018) 152–159. [DOI] [PubMed] [Google Scholar]
  • [44].Jin M, Inoue S, Umemura T, Moriya J, Arakawa M, Nagashima K, Kato H, Cyclin D1, p16 and retinoblastoma gene product expression as a predictor for prognosis in non-small cell lung cancer at stages I and II, Lung cancer (Amsterdam, Netherlands) 34(2) (2001) 207–18. [DOI] [PubMed] [Google Scholar]
  • [45].Bartkova J, Lukas J, Muller H, Strauss M, Gusterson B, Bartek J, Abnormal patterns of D-type cyclin expression and G1 regulation in human head and neck cancer, Cancer Res 55(4) (1995) 949–56. [PubMed] [Google Scholar]
  • [46].Qin A, Yu Q, Gao Y, Tan J, Huang H, Qiao Z, Qian W, Inhibition of STAT3/cyclinD1 pathway promotes chemotherapeutic sensitivity of colorectal caner, Biochemical and biophysical research communications 457(4) (2015) 681–7. [DOI] [PubMed] [Google Scholar]
  • [47].Bartkova J, Lukas J, Muller H, Lutzhoft D, Strauss M, Bartek J, Cyclin D1 protein expression and function in human breast cancer, Int J Cancer 57(3) (1994) 353–61. [DOI] [PubMed] [Google Scholar]
  • [48].Kim JK, Diehl JA, Nuclear cyclin D1: an oncogenic driver in human cancer, Journal of cellular physiology 220(2) (2009) 292–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [49].Lu F, Gladden AB, Diehl JA, An alternatively spliced cyclin D1 isoform, cyclin D1b, is a nuclear oncogene, Cancer Res 63(21) (2003) 7056–61. [PubMed] [Google Scholar]
  • [50].Burd CJ, Petre CE, Morey LM, Wang Y, Revelo MP, Haiman CA, Lu S, Fenoglio-Preiser CM, Li J, Knudsen ES, Wong J, Knudsen KE, Cyclin D1b variant influences prostate cancer growth through aberrant androgen receptor regulation, Proceedings of the National Academy of Sciences of the United States of America 103(7) (2006) 2190–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [51].Mori K, Hirao E, Toya Y, Oshima Y, Ishikawa F, Nose K, Shibanuma M, Competitive nuclear export of cyclin D1 and Hic-5 regulates anchorage dependence of cell growth and survival, Molecular biology of the cell 20(1) (2009) 218–32. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1536518-supplement-1.pptx (2MB, pptx)

RESOURCES