PHD-finger domain protein 5A functions as a novel oncoprotein in lung adenocarcinoma

Abstract

Background: PHD-finger domain protein 5A (PHF5A) is a highly conserved small transcriptional regulator also involved in pre-mRNA splicing; however, its biological functions and molecular mechanisms in non-small cell lung cancer (NSCLC) have not yet been investigated. The purpose of this study was to determine the functional relevance and therapeutic potential of PHF5A in lung adenocarcinoma (LAC).

Methods: The expression of PHF5A in LAC tissues and adjacent non-tumor (ANT) tissues was investigated using immunohistochemistry of a tissue microarray, qRT-PCR, western blot and bioinformatics. The function of PHF5A was determined using several in vitro assays and also in vivo assay by lentiviral vector-mediated PHF5A depletion in LAC cell lines.

Results: PHF5A was highly upregulated in LAC tissues compared with the ANT counterparts, and closely associated with tumor progression and poor patient prognosis. These results were further confirmed by findings of the TCGA database. Moreover, functional studies demonstrated that PHF5A knockdown not only resulted in reduced cell proliferation, increased cell apoptosis, and cell cycle arrest, but also suppressed migration and invasion in LAC cells. PHF5A silencing was also found to inhibit LAC tumor growth in nude mice. Microarray and bioinformatics analyses revealed that PHF5A depletion led to dysregulation of multiple tumor signaling pathways; selected factors in key signaling pathways were verified in vitro.

Conclusions: The data suggest for the first time that PHF5A is an oncoprotein that contributes to LAC progression by regulating multiple signaling pathways, and may constitute a prognostic factor and potential new therapeutic target in NSCLC.

Keywords: Lung adenocarcinoma; PHF5A; Prognostic biomarker; Proliferation; Tumor invasion.

Fig. 1

PHF5A overexpression is associated with lung adenocarcinoma (LAC) progression and poor prognosis. a PHF5A expression in LAC tissue microarrays. Representative images of LAC tissue samples with strong nuclear staining (brown), with adjacent noncancerous normal lung tissues showing weakly positive PHF5A staining. b The staining index (SI) of PHF5A expression in LAC tissues was significantly higher than that of normal paired samples. c-e PHF5A expression was confirmed in three fresh paired primary LAC tissues and matched adjacent non-tumor tissues from the same patient, by IHC (c), qRT-PCR (d) and Western blot (e). f Kaplan-Meier survival curves for LAC patients with high and low PHF5A expression levels, respectively. LAC patients with high PHF5A expression showed poorer survival compared with the low PHF5A group (n = 57; P = 0.043, ^# compared by Gehan-Breslow-Wilcoxon test). g-h TCGA data indicated that Phf5a mRNA levels in LAC tissues were associated with T stage (g) and N stage (h). i Kaplan-Meier survival curves comparing LAC patients with low and high Phf5a expression levels (n = 149; P = 0.010, ^★compared by the log-rank test; TCGA). High and low expression levels were based on the median value of Phf5a mRNA. ^**P < 0.01

Fig. 2

Effect of PHF5A knockdown on cell proliferation and colony formation in H1299 cells. a Phf5a mRNA levels from four common LAC cell lines were assessed by qRT-PCR. b-d Lentivirus-mediated shPHF5A knockdown in H1299 cells was performed via lentiviral infection (b), and qRT-PCR (c) and Western blot (d) were used to assess silencing efficacy. e Cellomics assay images and fluorescence quantification over five days in shCtrl and shPHF5A-transfected H1299 cells. f Colony formation assay was used to evaluate H1299 cell growth after PHF5A knockdown. ^**P < 0.01; ^**compared with the shCtrl group (e)

Fig. 3

Effect of PHF5A knockdown on cell cycle progression and apoptosis in H1299 cells. a Cell cycle was assessed in H1299 cells by flow cytometry five days after transfection with the indicated shRNAs. Representative flow-cytograms are shown, as well as diagrams quantifying cell fractions in the G₀/G₁, S, and G₂/M phases. b Apoptosis was evaluated by flow cytometry in PHF5A knockdown and control H1299 cells. Representative flow-cytograms are shown, and apoptotic rates were derived as the percentages of Annexin V-APC positive cells. ^**P < 0.01

Fig. 4

Effect of PHF5A knockdown on cell proliferation, cell cycle, and apoptosis in H1975 cells. a-c shRNA against Phf5a was conducted by lentivirus infection, and Phf5a silencing efficacy in H1975 cells was determined at both mRNA (qRT-PCR) and protein (Western blot) levels. d During five days of continuous cell counting by fluorescence microscopy, the quantity of shPHF5A-transfected cells decreased gradually compared with control values. e The clonogenic ability of H1975 cells was declined after PHF5A knockdown as evidenced by colony formation assay. f Cell cycle analysis using FACS in shCtrl and shPHF5A-transfected H1975 cell lines. g Apoptosis was analyzed in PHF5A knockdown and control H1975 cells by FACS after Annexin V staining. ^*P < 0.05; ^**P < 0.01

Fig. 5

PHF5A silencing inhibits tumor growth in vivo. a Representative images of tumor-bearing mice. b Xenografts from mice in each group. c Tumor volumes were measured at the indicated times. d Quantitative IVIS intensity was obtained at 30 days. e Mean weights of tumors obtained from mice at 30 days. f-g IHC staining demonstrated that PHF5A suppression was established in xenografts, and cell proliferation ability was inhibited in vivo, as indicated by Ki67 levels. ^**P < 0.01; ^**compared with the shCtrl group (c)

Fig. 6

Effect of PHF5A knockdown on migration and invasion in LAC cells. a Wound healing assay was performed to assess the migratory potential of control or PHF5A knockdown cells at indicated time points (original magnification, × 100). b Quantitation of migration area from control and PHF5A-silenced cells. c Transwell invasion assay was carried out to measure the cell invasive ability of control or PHF5A knockdown cells (original magnification, × 100). d Quantitation of cell invasion from control and PHF5A-silenced cells. **P < 0.01

Fig. 7

Microarray analysis of H1299 cells after PHF5A knockdown. a Hierarchical cluster analysis of shCtrl and shPHF5A-transfected H1299 cells. A total of 1112 genes were differentially expressed. Heat-map colors represent mean-centered fold change expression in log-scale. b Disease and function enrichment analysis based on gene numbers, expression levels, and significance probability suggested that among the retrieved categories, genes related to “Cancer” ranked first with the greatest changes. c Pathway enrichment analysis for related genes, based on significance probability, suggested that among the analyzed gene sets, genes related to the IGF-1 signaling pathway ranked second in showing the greatest changes in expression levels. d Fold changes of three putative IGF-1 pathway genes, including IGFBP3, PIK3CB, and AKT2, as well as cell cycle/apoptosis related genes, including DDIT3, Skp2, and P53, based on microarray analysis. e-f Validation of microarray data by qRT-PCR (e) and Western blot (f). IGFBP3, DDIT3, P53 were significantly up-regulated, while PIK3CB, AKT2, and Skp2 were significantly down-regulated upon PHF5A knockdown. g Knowledge-based interactive network for the selected targets in the IGF-1 pathway as well as cell cycle/apoptosis related molecules was constructed with the Reactome database. Red and green denote up-regulated and down-regulated genes, respectively. NMJ, neuromuscular junction (c). ^*P < 0.05; ^**P < 0.01, compared with the shCtrl group. A higher quality of Fig. 7 is available as Additional file 6