PFKFB4 promotes lung adenocarcinoma progression via phosphorylating and activating transcriptional coactivator SRC-2

The expression profile of PFKFB4 and SRC family at mRNA level in lung adenocarcinoma tissues was analyzed from UALCAN Platform (http://​ualcan.​path.​uab.​edu/​index.​html) which is a comprehensive, user-friendly, and interactive web resource for analyzing cancer OMICS data.

A549 cells which were acquired from Shanghai Chinese Academy of Sciences Cell Bank (Shanghai, China) were transfected with SRC-2 #1 siRNA, SRC-2 #2 siRNA or control siRNA, respectively. Cells were harvested after transfection 72 h. Total RNA was extracted from cells with Trizol reagent (Invitrogen, Carlsbad, CA, USA). RNA purity was analyzed using NanoDrop (Quawell UV–visible spectrophotometer) and Agilent 2100 Bioanalyzer (Santa Clara, CA, USA). Afterwards, transcriptome sequencing was performed on the Illumina HiSeq™ 2500 (NxGenBio Life Sciences, New Delhi) platform. Clean reads were got by filtering for low-quality then assembled. The contigs were organized based on the overlap regions to get transcript sequences. Fold change with an absolute value of log₂ ratio ≥ 5 was set as the threshold to obtain the transcripts of which expression was significant difference.

NCI-H1975, A549 and 293 T cell lines were acquired from Shanghai Chinese Academy of Sciences Cell Bank (Shanghai, China). The lung adenocarcinoma cell line NCI-H1975 was cultured in RPMI1640 medium (Invitrogen, USA) and A549 was cultured in F12K (Invitrogen, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, Grand Island, NY, USA) and 100 μg/ml streptomycin and 100 IU/ml penicillin (Invitrogen, USA). 293 T cells were cultured in DMEM medium (Invitrogen, USA) supplemented with 10% FBS and 1% penicillin/streptomycin. All cells were cultured at 37 °C with 5% CO₂. The transfection experiments were performed using Lipofectamine 2000 Reagent (Invitrogen, USA). The wildtype plasmids were purchased from OriGene (Rockville, MD, USA). SRC-2 S487A was constructed by site-directed mutation. The sequences of shRNA against PFKFB4 and siRNAs against SRC-2 were shown in Table 1.

293 T cells were co-transfected with pLKO shRNA constructs, PAX2, MD2 helper plasmids using Lipofectamine 2000 Reagent (Invitrogen, USA). Following transfection, the lentivirus supernatants were collected. A549 and NCI-H1975 were infected for 3 days and selected in the presence of puromycin (1 μg/ml) for at least one week.

Cell proliferation was measured using cell counting kit-8 (CCK-8; Dojindo Molecular Technologies, Rockville, MD, USA). Cells were transfected as indicated. Cells were seeded into 96-well plates at a density of 3 × 10³ cells/well. CCK-8 reagent was added into the wells at indicated time points and incubated at 37 °C for 2 h. The absorbance was measured at 450 nm in a microplate reader (Promega, Madison, WI, USA).

For migration analysis, cells transfected after 48 h were seeded into the top compartment of the Transwell chamber (8-μm pore size; BD Biosciences, San Jose, CA, USA) in a 24-well plate at a proper density. DMEM medium containing 10% FBS was added to the bottom chamber. When cultured for 24 h, cells on the upper surface of the membrane were scraped. Cells which went through the membrane were fixed with 4% paraformaldehyde and stained with DAPI. The images were captured using the microscope. For invasion analysis, the upper Transwell chamber was coated with matrigel. The following steps were conducted as the migration assay. Cells going through the matrigel were collected and counted. Each experiment was performed in triplicate.

Cells were harvested with lysis buffer with protease and phosphatase inhibitor cocktail (Millipore). Lysates were precleared with Protein A/G Agarose beads (Pierce) at 4 °C for 1 h, after washed then incubated with antibodies against Myc (Cell signaling Technology, #2276), or HA (Abcam, ab18181), or PFKFB4 (GeneX Health) respectively at 4 °C overnight. Preblocked agarose beads added to the lysates mixture were incubated at 4 °C for another 1 h. The immunocomplexes were eluted for western blot analysis.

Proteins were extracted using lysis buffer and separated by 10% SDS-PAGE (Invitrogen) and blotted onto Polyvinylidene Fluoride (PVDF) membranes. Membranes were incubated with primary antibodies at 4 °C overnight. Then membranes were incubated with the secondary antibodies at room temperature for one hour. Proteins were detected using the enhanced chemiluminescence (ECL) kit (Millipore, Burlington, MA, USA). The antibodies against SRC-2 phospho Ser469, phospho Ser487, phospho Ser493, phospho Ser499 and phospho Ser736 were prepared from GeneX Health Co,. Ltd.

Total RNA was extracted from cells using Trizol reagent (Invitrogen). RNA was reversely transcribed into cDNA using PrimeScript RT Master Mix (TaKaRa, Dalian, China). The cDNA was subjected on Real-time polymerase chain reaction (PCR) analysis using the SYBR qPCR-detection-Kit (TaKaRA, Dalian, China) on the Real-time PCR System (ABI7500, ABI, Oyster Bay, NY, USA). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the internal control. The sequences of the primers used for RT-PCR were listed in Table 2.

All data were analyzed using GraphPad Prism5 software (GraphPad Software, San Diego, CA). Results were presented as the mean ± SD. Statistical significance was determined by the Student’s t-test.

It has been reported that metabolic enzyme PFKFB4 could promote breast cancer progression by activating the oncogenic steroid receptor coactivator-3 (SRC-3) [2]. Since we found that both PFKFB4 and SRC protein had a higher expression at mRNA level in 515 LUAD samples compared to 59 normal tissues (Fig. 1a, b) from UALCAN Platform (http://​ualcan.​path.​uab.​edu/​index.​html), we wondered whether PFKFB4 also had an interaction with SRC family proteins in LUAD. We next performed co-immunoprecipitation using Myc-tag in 293 T cells transfected with myc-PFKFB4 together with SRC-1, SRC-2 or SRC-3 respectively. We observed that PFKFB4 only interacted with SRC-2 but not SRC-1 or SRC-3 (Fig. 1c). Next, we validated the direct interaction between SRC-2 and PFKFB4 by co-precipitating SRC-2 with HA-tag in PFKFB4-expressing 293 T cells (Fig. 1d). Furthermore, endogenous co-immunoprecipitation results also confirmed that PFKFB4 interacted with SRC-2 in lung adenocarcinoma cells A549 and NCI-H1975 (Fig. 1e, f).

To investigate the interactive relationship between PFKFB4 and SRC-2, firstly we selected two kinds of validated PFKFB4 shRNAs. The PFKFB4-knockdown stable cell lines were infected by lentivirus. The silencing efficiency was detected as shown in Fig. 2a, b. Both shPFKFB4s significantly reduced PFKFB4 expression compared to negative control (sh-NC) in A549 and NCI-H1975 cell lines. The shPFKFB4 #1 group had a better knockdown efficiency in comparison to the shPFKFB4 #2 group and would be used in subsequent functional experiments. We observed that PFKFB4-knockdown did not change SRC-2 expression at protein level (Fig. 2c, d). What’s more, the phosphorylated levels of SRC-2 at Ser487, of which the site was associated with transcription activity, was remarkably diminished when PFKFB4 expression was reduced (Fig. 2e, f). At the same time, the other phosphorylation sites of SRC-2 at Ser469, Ser493, Ser499, Ser736 had no change, indicating that PFKFB4 interacted with SRC-2 and phosphorylated SRC-2 at Ser487. Furthermore, we observed that the phosphorylated SRC-2 at S487 could also be detected in precipitant from SRC-2 immunoprecipitation of lung adenocarcinoma cell A549 and NCI-H1975 (Fig. 2g). Consistently, compared to those with overexpression of SRC-2 WT, co-IP of PFKFB4 would precipitate lower level of SRC-2 in cells with overexpression of the phosphor-deficient SRC-2 mutant (SRC-2 S487A) (Fig. 2h). Taken together, these results indicated that PFKFB4 could interact and phosphorylate SRC-2 at Ser487.

To identify the potential mechanism of PFKFB4-SRC-2 in lung adenocarcinoma, we performed a transcriptome sequencing using A549 cells transfected with si-SRC-2 #1 and si-NC. CARM1 expression was the most downregulated gene among these genes (Fig. 3a). The screening result was further validated by qRT-PCR. CARM1 expression at mRNA level was the most decreased among the downregulated genes in si-SRC-2 #1-transfected cells compared to control (Fig. 3b). Western blot showed that CARM1 expression at protein level was also reduced when SRC-2 was knocked down (Fig. 3c, d). The CARM1 protein level was decreased much more obviously in si-SRC-2 #1-transfected A549 and NCI-H1975 cells as si-SRC-2 #1 had a higher silencing efficiency compared to si-SRC-2 #2.

To further explore the functional effects of the phosphorylation of SRC-2 at Ser487 in lung adenocarcinoma progression, we constructed a phosphorylation-deficient SRC-2 mutant plasmid (SRC-2 S487A). We transfected SRC-2 wildtype (SRC-2 WT) or SRC-2 S487A in 5MPN-treated (PFKFB4 inhibitor) A549 and NCI-H1975 cells. Western blot analysis showed that SRC-2 expression at protein level was not changed in 5MPN-treated A549 and NCI-H1975 cells while highly expressed with overexpression of SRC-2 WT or SRC-2 S487A. What’s more, CARM1 expression was decreased with 5MPN treatment and could only be rescued by SRC-2 WT overexpression but not SRC-2 S487A overexpression (Fig. 4a, b). The results suggested that PFKFB4 phosphorylated SRC-2 at Ser487 and activated its transcriptional activity so that CARM1 was up-regulated by phosphorylated SRC-2 at Ser487. Cell Counting Kit-8 (CCK-8) analysis showed that the decreased cell proliferation induced by 5MPN treatment in A549 and NCI-H1975 cells could be rescued by co-overexpression of SRC-2 WT but not SRC-2 S487A (Fig. 4c, d). Furthermore, overexpression of SRC-2 WT but not SRC-S487A in LUAD cells with low expression of PFKFB4 also restored the suppressed migration and invasion ability induced by PFKFB4 downregulation (Fig. 4e, f). In accord with the above results, we observed the similar functional effects by knocking down PFKFB4 with shRNA (Additional file 1: Figure S1). Above all, the results indicated that PFKFB4 promoted LUAD cells proliferation, migration and invasion dependent on the phosphorylation of SRC-2 at Ser487.

To understand the precise molecular mechanism of PFKFB4-SRC-2-drived lung adenocarcinoma progression, we overexpressed CARM1 in si-SRC-2 #1-transfected LUAD cells, of which SRC-2 had been mostly knockdown. We observed that the decreased CARM1 expression at mRNA and protein levels in si-SRC-2#1-transfected A549 and NCI-H1975 cells could be rescued by co-overexpression of CARM1 (Fig. 5a, b). CCK-8 assays showed that overexpression of CARM1 in SRC-2-knockdown LUAD cells could blocked the suppressed A549 and NCI-H1975 cells proliferation resulted from si-SRC-2 #1 transfection (Fig. 5c). In addition, the blockade of migration and invasion abilities induced by SRC-2 knockdown could be restored by CARM1 co-overexpression (Fig. 5d, e). These data revealed that CARM1 was transcriptional up-regulated by phosphorylated SRC-2 at Ser487, enhancing lung adenocarcinoma cells proliferation, migration and invasion.

The above results indicated that suppressed effects resulted from SRC-2 knockdown could be mediated by CARM1 overexpression, but it was unclear that whether the inhibitory effects induced by PFKFB4 knockdown could be rescue by CARM1. Thus, we overexpressed CARM1 in PFKFB4-knockdown cells and detected the functional effects. The results showed that the decreased CARM1 expression at mRNA and protein levels led to by PFKFB4 knockdown could be rescued by CARM1 overexpression in A549 and NCI-H1975 cells (Fig. 6a, b). Moreover, the suppressed proliferation ability of PFKFB4 knockdown in lung adenocarcinoma cells was reversed upon overexpression of CARM1 in PFKFB4-knockdown A549 and NCI-H1975 cells (Fig. 6c). Furthermore, we also observed that CARM1 overexpression rescued the inhibitory effect of PFKFB4 knockdown on cell migration and invasion of A549 and NCI-H1975 cells (Fig. 6d, e). Taken together, these data showed that overexpression of CARM1 could abolish the suppressed effects induced by PFKFB4 knockdown, which suggested that PFKFB4 promoted lung adenocarcinoma cells proliferation through SCR-2/CARM1 axis.