JMJD5 inhibits lung cancer progression by regulating glucose metabolism through the p53/TIGAR pathway

Metabolic reprogramming is considered one of the main driving forces for tumor progression, providing energy and substrates of biosynthesis to support rapid neoplastic proliferation. Particularly, the tumor suppressor protein p53 was shown to revert the Warburg effect and play complex roles in regulating glucose metabolism. Jumonji C domain-containing protein 5 (JMJD5) has previously been reported as a negative regulator of p53. However, the role of JMJD5 in p53-mediated metabolic reprogramming remains elusive. Here, we discovered that knockdown of JMJD5 significantly enhances TIGAR expression in p53 wild-type non-small cell lung cancer (NSCLC) cells, which could further suppress glycolysis and promote the pentose phosphate pathway. Besides, JMJD5 knockdown promotes the NSCLC cell proliferation in vitro and xenograft tumor growth in vivo, while silencing TIGAR can abolish this effect. Low JMJD5 expression levels are associated with elevated TIGAR levels and correlates with poor prognosis in lung cancer patients. Taken together, our findings suggest that JMJD5 is a key regulator of tumor glucose metabolism by targeting the p53/TIGAR metabolic pathway.


Introduction
As a hallmark of cancer, metabolic reprogramming is critical for tumorigenesis and tumor progression by providing energy and substrates of biosynthesis to support cancer cell proliferation [1]. It is best characterized by the "Warburg effect"-tumor cells prefer to utilize glucose through glycolysis rather than oxidative phosphorylation (OXPHOS) to generate lactate and ATP even in the normal aerobic environment [2]. Glycolysis also confers cancer cells to other advantages, especially by providing diverse metabolic intermediates for various biosynthetic pathways [3]. In addition, the pentose phosphate pathway (PPP), which branches from glycolysis at the first committed step, serves a crucial role in supporting cancer cell survival and growth by generating pentose phosphate for nucleic acid synthesis and nicotinamide adenine dinucleotide phosphate (NADPH) for fatty acid synthesis and combating oxidative stress [4,5].
The p53 tumor suppressor acts as a sensor of stress and is a regulator of various cellular processes, including cell cycle arrest, DNA repair, apoptosis, senescence, and metabolism [6]. Although apoptosis, cell cycle arrest, and senescence were widely believed to mediate the main function of p53 in tumor suppression, the disruption of these functions is not sufficient to induce cancer [7]. Instead, multiple studies in mouse models (e.g. p53 3KR/3KR knock-in mouse model) have highlighted its metabolic roles in suppressing cancer progression, mainly through the inhibition of glycolysis and promotion of OXPHOS. In this context, p53 exerts this metabolic regulation by regulating the expression of metabolic genes, such as p53-induced glycolysis and apoptosis regulator (TIGAR), glucose transporters, parkin, synthesis of cytochrome c oxidase 2 (SCO2), glutaminase 2 (GLS2), and malic enzyme; or modulating metabolic enzyme activities, such as glucose-6-phosphate dehydrogenase (G6PD) and hexokinase 2 (HK2) [8,9].
Here, we report that the p53 key-response gene TIGAR was significantly upregulated as JMJD5 knockdown in nonsmall cell lung cancer (NSCLC) cells. We show that JMJD5 deletion promotes lung cancer cell proliferation by regulating glucose flux between glycolysis and PPP through the p53/TIGAR pathway. Furthermore, JMJD5 is negatively related to the expression of TIGAR and correlates with better prognosis in lung cancer patients.

Cell culture and transfection
Human NSCLC cell lines A549, H1650, H1299, and Calu-6 were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) and cultured in RPMI 1640 medium (Invitrogen, USA) supplemented with 10% fetal calf serum (FBS). All cells were cultured at 37 °C in a humidified incubator with 5% CO 2 .
Cells transfected with plasmids were using X-treme GENE HP DNA Transfection Reagent (Roche, Switzerland) and cells transfected with siRNAs were using Lipofectamine RNAiMAX (Invitrogen, USA) according to the manufacturer's instructions.
Stable knockdown of JMJD5 was performed with the lentiviral expression system using shRNA. HEK293FT cells were co-transfected with pLVX shJMJD5 and two packaging plasmids (psPAX2 and pMD2.G) to produce lentiviruses. The shRNA sequences targeting JMJD5 were as follows:sense:GAT CCG CCA CTG AGC TCT TCT ACG ACT CGA GTC GTA GAA GAG CTC AGT GGT TTT TG,antisense:AAT TCA AAA ACC ACT GAG CTC TTC TAC GAC TCG AGT CGT AGA AGA GCT CAG TGG CG.

Western blotting
Total protein was extracted from cells using RIPA lysis buffer (Millipore, USA). Proteins were separated by SDS-PAGE, transferred to a nitrocellulose membrane, and incubated with specific primary antibodies. Then, the membrane was incubated with Alex 680-or IR 800-conjugated secondary antibody for Odyssey CLx analysis (Li-COR, USA).

Quantitative real-time PCR (RT-qPCR)
Total RNA was isolated from cells using RNAiso Plus (Takara, Japan) and was reverse-transcribed into cDNA using the PrimeScript RT reagent kit (Takara, Japan) according to the manufacturer's instructions. qRT-PCR was performed with SYBR Premix Ex Taq (Takara, Japan) on a CFX96 Touch system (Bio-Rad, USA). The primer sequences were as follows:

Dual-luciferase reporter assay
The promoter sequence of TIGAR that contains the p53 binding site (p53 BS: CGG CAG GTC TTA GAT AGC TT) was synthesized by PCR and subcloned into a pGL3-Basic vector (Promega, USA). JMJD5 siRNA, Myc-JMJD5, p53 siRNA, or Flag-p53 were transfected or co-transfected into cells firstly, after 4 ~ 6 h transfection, cells with the fresh medium were co-transfected with pRL-TK and pGL3-TIGAR promoter plasmids. After 24 h transfection, cells were lysed and relative luciferase activity was measured using Dual-Luciferase Reporter Assay System (Promega, USA).

Measurement of glucose uptake, lactate production, G6PDH activity, and NADPH/NADP + Ratio
Glucose levels in the culture medium were measured using the Glucose Colorimetric Assay Kit (BioVision, USA). Intracellular lactate levels were measured using a lactate acid assay kit (Solarbio, China). G6PDH activity was measured using G6PDH Activity Assay Kit (Solarbio, China). NADP + /NADPH ratios were measured with a NADP + / NADPH quantitation colorimetric kit (BioVision, USA). All these experiments were performed with kits according to the manufacturer's protocols.
For the colony formation assay, 500 ~ 1000 cells were seeded into a 6-well plate and incubated for about 10 days. Then cells were fixed with 4% paraformaldehyde for 15 min and stained with crystal violet for 15 min. Finally, cells were washed with deionized water and air-dried. Images were taken and colony numbers were counted under the microscope (Olympus, Japan).

Immunohistochemistry (IHC)
A cohort of 25 lung adenocarcinoma patient specimens was obtained from the Second Affiliated Hospital of Zhejiang University School of Medicine. The study was approved by the ethics committee of Zhejiang University School of Medicine (2017026). The IHC assay was performed using an Envision Detection System (DAKO, CA) according to the manufacturer's instruction with anti-JMJD5 (sc-377440) and anti-TIGAR (sc-166290) antibodies. The staining results were assessed and confirmed by two independent investigators blinded to the clinical data.

Nude mice xenograft
4-week-old BALB/c nude mice were purchased from SLAC Laboratory animal corporation (Shanghai, China). 5 × 10 6 of A549 or H1299 cells were subcutaneously injected into the nude mice (n = 5 per group). Xenograft tumor sizes were measured by a vernier caliper with two perpendicular diameters and tumor volumes were calculated according to the formula: 0.5 × length × width 2 . After 4 weeks the mice were sacrificed and the xenograft tumors were harvested, weighed, and photographed. All animal experiments were approved by the Laboratory Animals Welfare Ethics Review Committee of Zhejiang University.

Statistical analysis
All experiments in this study were repeated independently at least three times and data were presented as mean ± SD. GraphPad Prism 8.0 was used for statistical analysis. Comparison between the two groups was analyzed by student's t-test or log-rank t-test. P-value < 0.05 was considered to be statistically significant.

JMJD5 negatively regulates the expression of TIGAR in a p53-dependent manner
To explore the role of JMJD5 in p53-regulated glucose metabolism in lung cancer, we analyzed the expression profiles of p53 downstream targets involved in the Warburg effect. Knockdown of JMJD5 significantly increased the expression of TIGAR in p53 wild type (p53-wt) cell lines (A549, H1650) but not in p53-null cell lines (H1299, Calu-6), whereas overexpression of JMJD5 decreased the level of TIGAR only in p53-wt cells (Fig. 1a-c). Other metabolism genes (PKM2, HK2, Glut1, G6PD), however, remained unchanged or altered discordantly at protein and mRNA levels.
TIGAR is one of the key p53-response genes in glucose metabolism [27]. To determine whether JMJD5-mediated changes in TIGAR expression depend on p53, we constructed a pGL3-TIGAR promoter reporter vector containing a p53 binding site and performed the dual-luciferase reporter assay in A549 and H1299 cells (Fig. 1d). Overexpression of JMJD5 reduced the transcriptional activity of TIGAR promoter and silencing of JMJD5 remarkably increased it in A549 cells but not in H1299 cells (Fig. 1e,  f). Moreover, knockdown of p53 abolished the enhanced promoter activity induced by JMJD5 loss in A549 cells (Fig. 1g). Overexpression of JMJD5 notably reduced the activity of TIGAR promoter in p53-overexpressed H1299 cells (Fig. 1h). All these results indicate that Dual-luciferase reporter assay was performed to measure the transcriptional activity of the TIGAR promoter after JMJD5 overexpression (e) or JMJD5 silencing (f) with or without p53 alteration (g, h). NC: negative control; Ctr: control; p53 BS: p53 binding site; EV: empty vector. The results were expressed as mean ± SD. ns: not significant, * /# P < 0.05, ** P < 0.01 JMJD5 negatively regulates the expression of TIGAR in a p53-dependent manner.

JMJD5 regulates glucose metabolism mainly through the p53/TIGAR axis
As a fructose-2,6-bisphosphatase (F2,6bPase), TIGAR functions as an important regulator in metabolic pathways by inhibiting glycolysis and increasing PPP flow [28,29]. Therefore, we examined the effect of JMJD5 on glycolysis by measuring glucose uptake and lactate production in lung cancer cells. The depletion of JMJD5 resulted in decreased glucose uptake and lactate production, which was more pronounced in p53-wt cells (Fig. 2a, b). The PPP activity was also assessed by detecting the G6PDH activity and NADPH/ NADP + ratio. As shown in Fig. 2c, d, silencing of JMJD5 significantly increased the G6PDH activity and NADPH/ NADP + ratio in p53-wt cells but there was no detectable increase in p53-null cells. Then we established stable JMJD5 knockdown cell lines by shRNA and further inhibited the expression of TIGAR (Fig. 2e). TIGAR silencing partially restored the reduced lactate production and suppressed the elevated G6PDH activity induced by JMJD5 knockdown in p53-wt cells (Fig. 2f, g). In p53-deficient H1299 cells, however, there was no apparent change after JMJD5 depletion (Fig. 2f, g). These data suggest that JMJD5 influences glucose metabolism mainly through the p53/TIGAR axis.

JMJD5 knockdown promotes cancer cell proliferation via TIGAR
Next, we investigated the role of JMJD5 in cancer cell proliferation which is largely supported by metabolic reprogramming. MTT and colony formation assay in A549 cells revealed that the stable knockdown of JMJD5 promoted cell proliferation. This effect was markedly reduced by further and lactate production (f) and G6PD activity (g) were detected in lung cancer cells with JMJD5 stably knockdown and TIGAR silencing. NC: negative control; Ctr: control. The results were expressed as mean ± SD. * /# P < 0.05, ** /## P < 0.01 silencing TIGAR (Fig. 3a, c and d). Whereas in H1299 cells, JMJD5 knockdown has no obvious effects on cell proliferation ( Fig. 3b-d). In JMJD5 overexpressed cells, we found a declined colony-forming ability in A549 and H1299 cells (Fig. 3e, f). Our results suggest that loss of JMJD5 could promote NSCLC cell proliferation mainly dependent on TIGAR regulation.

JMJD5 knockdown promotes xenograft tumor growth
To verify the role of JMJD5 in vivo, A549 and H1299 cells were injected subcutaneously into nude mice to develop xenograft tumors. In comparison with the control group, loss of JMJD5 exhibited larger volumes (Fig. 4a, b) and higher weights (Fig. 4c) in A549 xenograft tumors. However, the growth of H1299 xenograft tumors had no significant difference between JMJD5-shRNA and the control group ( Fig. 4d-f). We then further inhibited the expression of TIGAR in tumors and found a notable reverse on such a facilitative effect (Fig. 4a-c). Altogether, these results suggest that JMJD5 knockdown displays a tumor growth-promoting effect in p53-wt cells mainly through the regulation of TIGAR.

JMJD5 is negatively related to TIGAR expression and correlates with better prognosis in lung cancer patients
To address the clinical significance of our findings, we analyzed The Cancer Genome Atlas (TCGA) database and identified a significant downregulation of JMJD5 and an upregulation of TIGAR in lung adenocarcinoma (LUAD) tissues as compared with matched normal tissues (n = 51) (Fig. 5a). The genomic data of clinical LUAD samples were acquired from cBioPortal database, and we discovered that the mRNA expression of JMJD5 and TIGAR were negatively correlated in p53-wt samples (n = 120) (Fig. 5b). Consistently, a strong negative correlation was also identified between the mRNA levels of these two genes in p53-wt NSCLC cell lines from the Cancer Cell Line Encyclopedia (CCLE) database (Fig. 5c). We further performed IHC analysis to examine the protein expression of JMJD5 and TIGAR in LUAD samples, confirming that JMJD5 was downregulated and TIGAR was upregulated (n = 25) in tumor tissues compared to adjacent and H1299 cells after JMJD5 knockdown and TIGAR silencing (c, d), as well as JMJD5 stably overexpression (e, f). Ctr: control; EV: empty vector. The results were expressed as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001 normal tissues (Fig. 5d, e). Survival analysis of the TCGA data from two GSE cohorts (GSE30219 and GSE31210) and the OncoLnc database showed that high levels of JMJD5 were associated with significantly longer overall survival (OS) of the patients (Fig. 5f), whereas high levels of TIGAR were correlated with a shorter OS (Fig. 5g). Moreover, a combination of high JMJD5 and low TIGAR level exhibited a better prognosis of the patients (Fig. 5h). Taken together, these results demonstrate that JMJD5 is negatively related to TIGAR expression and correlates with better prognosis in p53-wt lung cancer patients.

Discussion
Cancer cells reprogram cellular metabolism to meet the energy and substrate needs of tumor development. Although it is well established that p53 plays a crucial role in regulating glucose homeostasis and tumor metabolism, how its upstream regulator and downstream metabolic network coordinate their effects is not completely understood [7][8][9]. In the current study, we show that the JmjC protein JMJD5 acts as a negative regulator of p53-mediated glycolytic regulation.
JMJD5 knockdown specifically upregulates the expression of p53 downstream gene TIGAR and promotes lung cancer cell proliferation by switching glucose flux from glycolysis to PPP. JMJD5 was initially identified as a histone demethylase, and its functions were extended to hydroxylation and proteolysis of non-histone proteins later [24,[30][31][32]. By interacting with metabolic enzymes or mediators, JMJD5 performs versatile functions in metabolic regulation largely independent of its enzymatic function. It has been reported that JMJD5 facilitates the nuclear translocation of PKM2, a critical enzyme involved in tumor glucose metabolism, and promotes aerobic glycolysis and tumor progression in breast, prostate, and glioblastoma cancer [17,18,21]. Whereas in pancreatic cancer, JMJD5 inhibits glycolysis and cell proliferation by negatively regulating c-Myc and its downstream glycolytic gene expression [25]. Intriguingly, we previously showed that JMJD5 is downregulated in lung cancer cells and negatively regulates p53 transcriptional activity by associating with its DNA binding domain, thereby facilitating cell cycle progression in response to DNA damage [23], [26]. In this study, we further revealed that JMJD5 suppresses the expression of glycolytic regulator TIGAR by On the other hand, it has been reported that there exist dual roles of p53 in the regulation of tumor metabolism. As enhanced aerobic glycolysis is a cancer-associated trait, the ability of p53 to suppress glycolysis and promote OXPHOS represents its tumor-suppressive activity [9,33]. However, p53-mediated pro-survival effects may also be beneficial for cancer cells to survive various stresses, such as protecting cancer cells from toxic levels of reactive oxygen species (ROS) and inducing cell cycle arrest to overcome nutrient deprivation [34,35]. In this case, TIGAR is one of the critical targets of p53. TIGAR inhibits phosphofructose kinase 1 (PFK1) activity by degrading its allosteric activator fructose-2,6-bisphosphate (F2,6BP), thereby inhibiting glycolysis and promoting PPP [27,29,36]. TIGAR could also increase the expression of G6PD, the PPP rate-limiting enzyme that catalyzes glucose-6-phosphate(G-6-P) into NADPH and 5-ribose phosphate (R-5-P), which is beneficial for the nucleotide synthesis and antioxidation effect [37,38]. Moreover, under hypoxic conditions, hypoxia-inducible factor 1 (HIF1α) promotes TIGAR to translocate to the mitochondria and binds to HK2, resulting in an increase in HK2 activity and a reduction of ROS-induced cell death [35]. TIGAR also inhibits autophagy and apoptosis by reducing oxidative stress or activating the mTOR signaling pathway [39]. Therefore, our findings that JMJD5 knockdown significantly upregulates TIGAR in p53-wt cells could contribute to the pro-survival effect of p53 in lowering ROS oxidative damage and enhancing DNA synthesis and repair, thus promoting cancer cell survival and proliferation. Lung cancer remains the most commonly diagnosed cancer and the leading cause of cancer-related mortality worldwide [40]. Although the p53 gene (TP53) is somatically mutated in over 50% of all types of lung cancer, its role in lung cancer prognosis is still unclear. A recent highthroughput analysis reported that the use of p53 target gene expression signatures was more prognostically predictive than p53 itself [41]. In the present study, we discovered that TIGAR is upregulated whereas JMJD5 is downregulated in lung cancer tissues, and both of them are correlated with the overall survival of the patients. Notably, a combination of high JMJD5 and low TIGAR levels exhibited a better prognosis for the disease. Our results, therefore, highlight the potential of using JMJD5 and TIGAR as prognostic markers for lung cancer, especially in p53-wt patients.
In summary, our study illustrates that JMJD5 plays a critical role in tumor metabolism by regulating TIGAR-mediated switching from glycolysis to PPP which is p53-dependent. Knockdown of JMJD5 could promote NSCLC cell proliferation in vitro and xenograft tumor growth in vivo by upregulating TIGAR expression, while silencing TIGAR can abrogate this effect. We also discovered that JMJD5 has a lower expression whereas TIGAR has a higher expression in lung cancer tissues, with an inverse correlation with disease prognosis. Altogether, our results suggest that JMJD5 is a negative regulator of the p53/TIGAR pathway, which may provide new ideas and strategies for diagnosis and targeted therapy in lung cancer.
Author contributions All authors have contributed to this study. SJ, QH, LG: experimental design, material preparation, data collection and analysis; SJ, LG: writing the first draft of the manuscript; SJ: funding acquisition. All authors commented on previous versions, and read and approved the final manuscript.
Funding This work was supported by the Zhejiang Provincial Natural Science Foundation of China (LY20H160040 and LY18H160024) and the National Natural Science Foundation of China (81772919).
Data availability All the datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare no conflict of interest. Fig. 5 JMJD5 is negatively related to TIGAR expression and correlates with better prognosis in p53 wild-type LUAD tissues. (a) The expression of JMJD5 and TIGAR in tumor and matched normal tissues of LUAD samples (n = 51) from the TCGA database. The correlation between JMJD5 and TIGAR expression in p53-wt LUAD tissue samples from cBioPortal database (b) and p53-wt NSCLC cell lines from CCLE database (c). IHC analysis (d) and representative images (e) of JMJD5 and TIGAR in LUAD and adjacent normal tissues (n = 25). The scale bar represents 100 μm. Kaplan-Meier survival curves by log-rank tests on LUAD patients stratified by the expression levels of JMJD5 (f), TIGAR (g), or both of the two proteins (h) for overall survival from GSE datasets and OncoLnc database. WT: wild type. The results were expressed as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001 ◂