progression
Sheng-You Liao1●I-Ying Kuo2●Yu-Ting Chen2●Pao-Chi Liao3●Ya-Fen Liu2●Hsin-Yi Wu4●Wu-Wei Lai5● Yi-Ching Wang 1,2
Received: 22 November 2018 / Revised: 1 May 2019 / Accepted: 29 May 2019 / Published online: 9 August 2019
© The Author(s), under exclusive licence to Springer Nature Limited 2019
Abstract
ZNF322A is an oncogenic zinc-finger transcription factor. Our published results show that ZNF322A positively regulates transcription of alpha-adducin (ADD1) and cyclin D1 (CCND1) to promote tumorgenicity of lung cancer. However, the upstream regulatory mechanisms of ZNF322A protein function remain elusive. Here, we demonstrate that AKT could phosphorylate ZNF322A by in vitro kinase assay and cell-based mass spectrometry analysis. Overexpression of AKT promoted ZNF322A protein stability and transcriptional activity, whereas these effects were inhibited by knockdown of AKT or treating with AKT inhibitor. We studied AKT-mediated phosphorylation sites, viz. Thr-150, Ser-224, Thr-234, and Thr-262. ZNF322A phosphorylation at Thr-262 by AKT promoted ZNF322A protein stability thus increased ADD1 promoter activity. Interestingly, phosphorylation at Thr-150, Ser-224, and Thr-234 enhanced transcription activity without affecting protein stability of ZNF322A. Chromatin immunoprecipitation and DNA affinity precipitation assays showed that ZNF322A phosphorylation defective mutants Thr-150A, Ser-224A, and Thr-234A attenuated chromatin binding and DNA binding affinity to ADD1 and CCND1 promoters compared with wild-type ZNF322A. Furthermore, AKT-mediated Thr-150, Ser-224, Thr-234, and Thr-262 phosphorylation promoted lung cancer cell growth and metastasis in vitro and in vivo.
Clinically, expression of phosphorylated ZNF322A (p-ZNF) correlated with actively phosphorylated AKT (p-AKT) in tumor specimens from 150 lung cancer patients. Multivariate Cox regression analysis indicated that combined p-AKT and p-ZNF expression profile was an independent factor to predict the clinical outcome in lung cancer patients. Our results reveal a new mechanism of AKT signaling in promoting ZNF322A protein stability and transcriptional activity in lung cancer cell, xenograft, and clinical models.
Introduction
ZNF322A, also known as ZNF388 or ZNF489, is a zinc-finger transcription factor that belongs to the Cys2His2type krüppel-like zinc-finger protein family. ZNF322A consists of 11 tandem repeats of zincfinger motif. Li and associates found that ZNF322A is mostly expressed in the nucleus and ZNF322A activates SRE and AP-1 transcriptional activity [1]. Our previous study showed that ZNF322A over-expression correlates with poor prognosis in both Asian and Caucasian lung cancer patients [2]. In addition, ZNF322A promotes lung tumor growth, metastasis, and stemness properties partially through activating promoter activity of alpha-adducin (ADD1) and cyclin D1 (CCND1), while suppressing promoter activity of p53 and c-Myc [3,4]. In addition, we found that deregulation of CK1 δ-GSK3β-FBXW7α signaling axis results in prolonged
* Yi-Ching Wang
1 Institute of Basic Medical Sciences, College of Medicine, National Cheng Kung University, Tainan 70101, Taiwan
2 Department of Pharmacology, College of Medicine, National Cheng Kung University, Tainan 70101, Taiwan
3 Department of Environmental and Occupational Health, College of Medicine, National Cheng Kung University, Tainan 704, Taiwan
4 Instrumentation Center, National Taiwan University, Tainan 10617, Taiwan
5 Department of Surgery, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan 70101, Taiwan
Supplementary informationThe online version of this article (https://
doi.org/10.1038/s41388-019-0928-x) contains supplementary material, which is available to authorized users.
1234567890();,: 1234567890();,:
ZNF322A stability to promote cancer progression [5].
Taken together, ZNF322A is an oncogenic transcriptional factor that contributes to lung tumorigenesis.
It has been shown that kinases and activation signaling axes regulate the protein stability and activity of transcrip-tion factors [6–13]. For example, AKT phosphorylates Sox2 at threonine (Thr, T) 118, which blocks lysine 119 methylation-mediated degradation of Sox2 [14]. Moreover, HSF-1 is phosphorylated by AKT at serine (Ser, S) 326 to enhance Slug promoter activity thereby inducing breast cancer cell metastasis upon HER2 overexpression [15].
AKT-mediated Sp1 phosphorylation at S42, T679, and S698 promotes CCR7 expression upon COX-2 stimulation [16]. Phosphorylation of RUNX2 by AKT enhances DNA binding affinity and transcriptional activity of RUNX2 to promote osteoblast and chondrocyte differentiation [17].
These studies indicate that AKT-mediated phosphorylation of transcription factors through regulating their DNA binding affinity, transcription complex formation, and pro-tein stability is associated with cancer progression.
Here we demonstrate that ZNF322A is phosphorylated by AKT at multiple sites including 150, Ser-224, Thr-234, and Thr-262. Phosphorylation at Thr-150, Ser-224, and Thr-234 activates the transcriptional activity of ZNF322A by increasing the binding affinity to the targeted promoters, while Thr-262 phosphorylation prolongs half-life of ZNF322A protein. Such a coordinated phosphor-ylation mediated by AKT enhances ZNF322A protein activity. Clinically, expression pattern of both phosphory-lated AKT AKT) and phosphoryphosphory-lated ZNF322A (p-ZNF) in tumor specimens of lung cancer patients can be an independent factor for predicting survival outcome.
Results
EGF/AKT signal enhances ZNF322A protein stability
It is well known that AKT is the downstream effector of epidermal growth factor (EGF) signaling pathway [18–20].
Activation of AKT positively correlates with the EGF receptor level in non-small cell lung cancer patients [21, 22]. Therefore, we first examined whether EGF/AKT signaling regulated ZNF322A protein expression. The immunoblotting data showed that EGF stimulation activated AKT activity accompanied by an increase of ZNF322A protein expression, whereas AKT inhibitor MK2206 treat-ment, but not ERK inhibitor U0126, significantly attenuated ZNF322A protein expression (Fig. 1a). These results demonstrated that EGF/AKT signaling enhanced ZNF322A protein expression.
AKT-mediated phosphorylation regulates functions of transcription factors including their transcriptional activity
and protein stability. Therefore, we investigated whether AKT regulates protein expression level of ZNF322A. The immunoblotting results showed that overexpression of AKT increased expression of endogenous ZNF322A protein in both H1299 and H460 cells (Fig. 1b, c). Conversely, ecto-pically expressed HA-ZNF322A protein level was decreased in cells treated with or without AKT inhibitor MK2206 (Fig. 1d, e). ZNF322A protein expression level was also decreased in siRNA-mediated AKT knockdown (si-AKT#1 and #2) cells (Fig. S1a, b). Importantly, change in AKT expression or activity did not affect ZNF322A mRNA level (Figs. 1f, g, and S1c), indicating that AKT regulates ZNF322A protein expression via post-translational modification.
Next, we examined whether AKT promoted ZNF322A protein stability. Cycloheximide (CHX) chase assay showed that overexpression of AKT prolonged ZNF322A protein half-life (Fig. 1h). Conversely, inhibition of AKT activity (Fig. 1i) or knockdown of AKT (Fig. S1d, e) diminished ZNF322A protein level. These data suggest that AKT increases ZNF322A protein expression via enhancing pro-tein stability of ZNF322A.
AKT promotes ZNF322A transcriptional activity
Furthermore, we addressed the question whether AKT regulates the transcriptional activity of ZNF322A. We previously reported that ZNF322A binds to AP-1 element in promoters of cancer-related genes such as alpha-adducin (ADD1) [3]. Our luciferase promoter activity assay showed that ADD1 promoter activity was increased by EGF sti-mulation, whereas the activity was suppressed upon AKT inhibitor treatment (Fig. 1j). Notably, overexpression of AKT enhanced ADD1 promoter activity whereas the activity was abolished by knockdown of ZNF322A (si-ZNF) (Fig. 1k). Conversely, overexpression of ZNF322A significantly increased ADD1 promoter activity, whereas treatment with AKT inhibitor decreased the ZNF322A transcriptional activity (Fig. 1l). Similar results on down-regulation of ZNF322A transcriptional activity in ADD1 promoter were observed in cells transfected with si-AKT oligos (Fig. S1f, g). These data suggest that AKT promotes ZNF322A transcriptional activity.
ZNF322A is a protein substrate of AKT kinase
To determine whether AKT phosphorylates ZNF322A, we performed radioactive in vitro kinase assay using recom-binant AKT and GST-tagged ZNF322A purified from E.
coli. We observed that ZNF322A was phosphorylated by AKT in a dose-dependent manner (Fig.2a). In addition, p-ZNF was detected by anti-phospho-AKT substrate (p-AKT substrate) antibody using nonradioactive in vitro kinase
6724 S.-Y. Liao et al.
assay (Fig. 2b). Moreover, we performed LC-MS/MS analysis to identify ZNF322A phosphorylation sites of AKT using p-ZNF recombinant protein. Indeed, we iden-tified several putative ZNF322A phosphorylation sites that were potential targets of AKT including Thr-150, Ser-224, and Thr-234 (Fig. S2). In addition, we also observed ZNF322A phosphorylation sites Thr-234 and Thr-262 from PhosphoSitePlus online database (https://www.
phosphosite.org/proteinAction.action?id=18154&show AllSites=true).
Next, we generated phospho-defective ZNF322A mutant recombinant protein including T150A, S224A, T234A, and T262A and performed in vitro kinase assay to determine whether these phosphorylation sites are regulated by AKT.
Our data showed reduced AKT-mediated phosphorylation signal in ZNF322A phosphorylation-defective mutant of T150A, S224A, T234A, and T262A compared with wild-type (WT)-ZNF322A (Fig. 2c). The immunoprecipitation (IP) and reverse IP results showed that AKT interacted with ZNF322A in lung cancer cells (Fig. 2d, e). These data Fig. 1 AKT upregulated ZNF322A protein expression, protein
stabi-lity, and transcription activity. a Immunoblot analysis of cell lysates derived from H1299 cells expressing HA-ZNF322A. Cells were starved in serum free medium for 12 h and then treated with or without EGF alone or together with AKT inhibitor MK2206 (2.5μM) or ERK inhibitor U0126 (10μM) for 3 h. Immunoblot analysis of endogenous ZNF322A in H1299 (b) or H460 (c) cells overexpressing AKT.
Immunoblot analysis of ectopically expressed HA-ZNF322A protein level in H1299 (d) or H460 (e) cells treated with or without AKT inhibitor MK2206. RT-qPCR analysis of mRNA level of ZNF322A in H1299 cells overexpressed with AKT (f) or treated with or without AKT inhibitor (g). Immunoblot analysis of cell lysates from H1299 cells expressing HA-ZNF322A cotransfected with or without AKT for
24 h (h) or treated with or without AKT inhibitor for 3 h (i) prior to cycloheximide (CHX, 20μg/ml) treatment at indicated times. Nor-malized fold changes are labeled below the blots. Quantification of ZNF322A band intensities was normalized to GAPDH, and then normalized to the time-point: 0 min. j–l Dual luciferase activity assays were performed using ADD1 promoter in H1299 cells. Cells expres-sing WT-ZNF322A were treated with or without EGF alone or toge-ther with AKT inhibitor MK2206 (j). Dual luciferase activity assay in AKT overexpressed H1299 cells transfected with si-Ctrl or ZNF322A siRNAs (k) or in cells treated with or without AKT inhibitor MK2206 (l). Data are presented as mean ± SEM. P-values determined using two-tailed Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001 AKT-mediated phosphorylation enhances protein stability and transcription activity of ZNF322A to. . . 6725
suggest that ZNF322A is a novel protein substrate of AKT and can be phosphorylated at Thr-150, Ser-224, Thr-234, and Thr-262.
ZNF322A Thr-262 phosphorylation by AKT promotes ZNF322A protein stability thus increases ZNF322A transcriptional activity
Since AKT could promote ZNF322A protein stability and transcriptional activity, we further examined that ZNF322A phosphorylation sites were responsible for ZNF322A protein stability. We first generated cells expressing HA-tagged ZNF322A phosphorylation-defective and mimetic mutant T150A/E, S224A/E, T234A/E, and T262A/E proteins.
Immunoblotting assay showed that the protein expression levels of T150A/E, S224A/E, and T234A/E were similar to that of WT-ZNF322A in both H1299 and H460 cells (Fig.
S3a–c). Notably, we observed that the protein expression of
phosphorylation-defective T262A-ZNF322A protein was lower than that of WT-ZNF322A (Fig. 3a). In addition, the protein expression of T262A-ZNF322A was restored to WT-ZNF322A expression level when cells were treated with proteasome inhibitor MG132 (Fig. 3b). Furthermore, we performed CHX assay to analyze the protein half-life of these ZNF322A mutant proteins. We observed that the protein half-life of phosphorylation-mimetic T262E-ZNF322A mutant was extended as compared with WT-ZNF322A, whereas the protein half-life of phosphorylation-defective T262A-ZNF322A mutant was shorter than that of WT-ZNF322A (Fig. 3c, d). As expected, the protein half-life of T150A/E, S224A/E, and T234A/E mutant proteins was similar to that of WT-ZNF322A (Fig. S3d–f). These data suggest that ZNF322A Thr-262 phosphorylation by AKT promotes ZNF322A protein stability.
Next, we determined whether Thr-262 phosphorylation affects transcriptional activity of ZNF322A in cells Fig. 2 ZNF322A was a protein substrate of AKT kinase. a The
radioactive in vitro kinase assay demonstrated that AKT phosphory-lated ZNF322A in a dose-dependent. b The nonradioactive in vitro kinase assay confirmed that ZNF322A was phosphorylated by AKT.
cThe nonradioactive in vitro kinase assay revealed that ZNF322A phosphorylation-defective mutants T150A-, S224A-, T234A, and T262A-ZNF322A attenuated ZNF322A phosphorylation by AKT compared with WT-ZNF322A. d, e ZNF322A and AKT interaction
were examined using immunoprecipitation (IP)-Western blot analysis.
ZNF322A and AKT were immunoprecipitated from total cell lysate of H1299 cells overexpressing GFP-tagged ZNF322A (GFP-ZNF322A) and/or AKT (d) or HA-ZNF322A and/or AKT (e). The pulled down proteins were detected by anti-GFP (d) or anti-AKT (e). An immu-noblot of longer exposure time to detect AKT input is shown to justify the interaction of GFP-ZNF322A with endogenous AKT (d)
6726 S.-Y. Liao et al.
transfected with EV, WT-, T262A-, or T262E-ZNF322A vector. Luciferase promoter activity data showed that ADD1 promoter activity was decreased in cells expressing phosphorylation-defective T262A-ZNF322A compared with cells expressing WT-ZNF322A (Left, Fig. 3e), while the decreased of ADD1 promoter activity could be blocked by proteasome inhibitor MG132 treatment (Right, Fig.3e).
We further examined the mRNA expression of ADD1 and cyclin D1 (CCND1), another transcription target of ZNF322A [3], in H1299 and H460 cells transfected with EV, WT-, T262A-, or T262E-ZNF322A vector. As expec-ted, the mRNA expression level of ADD1 and CCND1 was increased upon WT-ZNF322A overexpression, whereas cells expressing T262A-ZNF322A mutants could not
upregulate ADD1 and CCND1 mRNA (Fig. 3f–h). These results suggest that AKT-mediated ZNF322A phosphor-ylation at Thr-262 promotes ZNF322A protein stability thus enhances ZNF322A transcriptional activity.
AKT phosphorylates ZNF322A at Thr-150, Ser-224, and Thr-234 sites to enhance ZNF322A
transcriptional activity
Our data so far suggest that phosphorylation of ZNF322A at Thr-150, Ser-224 and Thr-234 did not affect ZNF322A protein stability (Fig. S3). Therefore, we examined the effect of these phosphorylation sites on the transcriptional activity of ZNF322A. Promoter activity assay showed that Fig. 3 ZNF322A Thr-262 phosphorylation promoted ZNF322A
pro-tein stability thus increased ZNF322A transcriptional activity.
a Immunoblot analysis of cell lysates from H460 cells ectopically expressing empty vector (EV), WT-, T262A-, or T262E-HA-ZNF322A protein. b Immunoblot analysis of cell lysates from H1299 cells expressing EV, WT-, T262A-, or T262E-HA-ZNF322A treated with DMSO (Left) or MG132 (10μM) (Right) for 6 h.
Immunoblot analysis of cell lysates from H1299 cells expressing WT-, T262E (c), or T262A-ZNF322A (d) mutants for 24 h prior to CHX
treatment. Normalized fold changes are labeled below the blots. e Dual luciferase activity assay was performed using ADD1 promoter in H1299 cells expressing WT- or T262A/E-ZNF322A treated with DMSO (Left) or MG132 (Right). RT-qPCR analysis of mRNA level of ZNF322A (f), ADD1 (g), or CCND1 (h) in H1299 cells expressing EV, WT-, T262A-, or T262E-ZNF322A. Data are presented as mean
± SEM. P-values determined using two-tailed Student’s t-test.
*P < 0.05; **P < 0.01; ***P < 0.001
AKT-mediated phosphorylation enhances protein stability and transcription activity of ZNF322A to. . . 6727
T150A-, S224A- and T234A-ZNF322A phosphorylation-defective mutants attenuated ADD1 and CCND1 promoter activities compared with WT-ZNF322A (bars gray vs. sla-shed gray, Fig. 4a–c). Moreover, overexpression of AKT increased WT-ZNF322A-induced promoter activity (bars gray vs. black, Fig.4a–c), whereas AKT failed to enhance promoter activity of ADD1 or CCND1 in cells expressing T150A-, S224A-, or T234A-ZNF322A mutants (Bars black vs. slashed dark gray, Fig. 4a–c). In addition, phosphorylation-mimetic mutant T150E, S224E, or T234E exerted transcription activity comparable to WT-ZNF322A (Fig. S4a–c). Together, these data suggest that AKT phos-phorylates ZNF322A at Thr-150, Ser-224, and Thr-234 sites to enhance ZNF322A transcriptional activity.
We further examined the mRNA expression of ADD1 and CCND1 in H1299 and H460 cells expressing T150A/E-, S224A/E-, or T234A/E-ZNF322A phospho-mutants. As expected, the mRNA expression level of ADD1 and CCND1 was increased upon WT-ZNF322A overexpression, whereas cells expressing phosphorylation-defective T150A-, S224A-, or T234A-ZNF322A mutants could not upregulate ADD1 and CCND1 mRNA (Fig. 4d). In addition, the increased ADD1 mRNA by WT-ZNF322A was attenuated by treat-ment with AKT inhibitor MK2206, whereas expression of ADD1 mRNA was not affected in phosphorylation-mimetic mutants T150E-, S224E-, or T234E-ZNF322A treated with AKT inhibitor (Fig. S4d). These data indicate that AKT phosphorylates ZNF322A at Thr-150, Ser-224, and Thr-234 to promote the transcriptional activation of ZNF322A, lead-ing to an increase in mRNA level of the target genes such as ADD1 and CCND1.
Thr-150, Ser-224, and Thr-234 phosphorylation increases DNA binding affinity of ZNF322A and recruitment of ZNF322A transcriptional complex to the promoter region ofADD1 and CCND1
Given that ZNF322A Thr-150, Ser-224, and Thr-234 phosphorylation by AKT increased ADD1 and CCND1 promoter activities to up-regulate their mRNA expression, we thus investigated whether 150, Ser-224, and Thr-234 phosphorylation sites affect the recruitment of ZNF322A transcriptional complex to the ADD1 or CCND1 promoter region. Our chromatin immunoprecipitation (ChIP) assay data showed that WT-ZNF322A was indeed bound to ADD1 or CCND1 promoter (bars gray vs. white, Fig.4e–g), whereas the binding ability to ADD1 or CCND1 promoter of T150A-, S224A- or T234A-ZNF322A mutant was abolished compared with that of WT-ZNF322A (bars gray vs. slashed gray, Fig. 4e–g). Furthermore, we found that AKT enhanced ADD1 promoter binding ability of WT-ZNF322A (bars gray vs. black, Fig. 4e–g), whereas over-expression of AKT did not markedly induce promoter
recruitment ability of T150A-, S224A- or T234A-ZNF322A mutant (bars black vs. slashed dark gray, Fig. 4e–g). In addition, phosphorylation-mimetic mutant T150E, S224E, or T234E possessed promoter recruitment ability compar-able to WT-ZNF322A (Fig. S4e–g).
Transcription factor activity regulated by post-translational modification has been reported to influence DNA binding affinity [23,24]. This prompted us to further investigate whether phosphorylation at T150, S224, or T234 at ZNF322A enhance their DNA binding affinity by in vitro DNA affinity immunoprecipitation assay (DAPA). Notably, we observed that the impairment of T150A-, S224A-, or T234A-ZNF322A binding to ADD1 promoter probe as compared with WT-ZNF322A (Fig. 4h). Together, these results indicate that AKT-mediated T150, S224, and T234 phosphorylation enhances the DNA binding affinity and recruitment of ZNF322A transcriptional complex to the promoter region of ADD1 and CCND1.
AKT-mediated Thr-150, Ser-224, or Thr-234
phosphorylation enhances lung cancer cell growth in vitro and in vivo
Results from our laboratory previously demonstrated that ZNF322A activates CCND1 expression to enhance lung cancer cell growth [3]. Therefore, we investigated whether phosphorylation ZNF322A at T150, S224, T234, or T262 enhance lung cancer cell proliferation. The proliferation assay results showed that WT-ZNF322A enhanced lung cancer cell proliferation compared with EV group. Of note, lung cancer cell proliferation was inhibited by T262A-ZNF322A compared with WT-T262A-ZNF322A, while T262E-ZNF322A promoted lung cancer cell proliferation to a level similar to WT-ZNF322A (Fig.5a). Next, we investigated the effect of AKT on cancer cell growth through phosphoryla-tion of ZNF322A at T150, S224, or T234 and we found that WT-ZNF322A-mediated induction of lung cancer cell pro-liferation could be induced by AKT overexpression in cells.
However, such proliferation enhancement was attenuated in cells expressing T150A, S224A, or T234A-ZNF322A even in the presence of AKT overexpression (lines blue vs. green;
red vs. purple, Fig.5b–d).
We further examined the effects of AKT-mediated acti-vation of ZNF322A on tumor growth in vivo using sub-cutaneous injection model including five animal groups.
Mice were injected with control cells (EV/EV), or cells expressing WT-ZNF322A (EV/WT), coexpressing AKT and WT-ZNF322A (AKT/WT), expressing a phosphorylation-deficient mutant (EV/S224A) or coexpressing AKT plus a phosphorylation-deficient mutant (AKT/S224A). The Ser224 was chosen based on the molecule modeling data that phosphorylation on Ser224 residue could increase the inter-action between ZNF322A and DNA (Fig. S5a). Tumor
6728 S.-Y. Liao et al.
volume and tumor weight were increased in the group injected with EV/WT cells and further enhanced in group with AKT/WT compared with EV/EV control group at 3 weeks post injection. Conversely, tumor growth promoted by AKT-mediated ZNF322A phosphorylation was greatly reduced in EV/S224A phosphorylation-deficient group and AKT/S224A coexpression group (Fig.5e–h). These in vivo results collaborated with in vitro data that AKT-mediated ZNF322A phosphorylation promotes lung tumor growth.
AKT-mediated Thr-150, Ser-224, or Thr-234 phosphorylation enhances lung cancer cell migration in vitro and tumor metastasis in vivo
Previous study showed that ZNF322A activated ADD1 expression to induce lung cancer cell migration [3].
Therefore, we evaluated whether ZNF322A phosphoryla-tion sites regulate lung cancer cell migraphosphoryla-tion (Fig.6a). We
Therefore, we evaluated whether ZNF322A phosphoryla-tion sites regulate lung cancer cell migraphosphoryla-tion (Fig.6a). We