Repression of Akt3 gene transcription by the tumor suppressor RIZ1

RIZ1 has been studied as a tumor suppressor and may play a role in metabolic diseases related to the Western style diet, such as cancer and obesity. The Akt pathway is known to play a role in both cancer and obesity, and a link between Akt and RIZ1 has also been found. To better understand the role of RIZ1 in obesity and cancer, we investigated how RIZ1 regulates the expression of Akt3. We found that overexpression of RIZ1 in HEK293 cells reduced the expression of Akt3 protein. Luciferase reporter activity of Akt3 gene promoter was significantly reduced in cells co-transfected with RIZ1. Recombinant proteins of RIZ1 was able to bind the Akt3 promoter in vitro, and chromatin immunoprecipitation assay also demonstrated the ability of RIZ1 binding to the Akt3 promoter in vivo. Overexpression of RIZ1 increased H3K9 methylation on the Akt3 promoter. These results identify Akt3 as a target of RIZ1 regulation and expand our understanding of the Akt pathway in cancer and obesity.


Results
RIZ1 on Akt3 expression in tissue culture. Previous studies by our group have shown that RIZ1 could regulate the activity of Akt and influence the mRNA expression of Akt3 in mouse 27 . Here, we further assessed whether the expression of Akt3 could be regulated by RIZ1. As shown in Fig. 1, overexpression of RIZ1 was accompanied by a significant decline in Akt3 protein level.
RIZ1 has been shown previously by reporter assays to regulate transcription 17 , so we next performed luciferase reporter assay to test RIZ1 effect on Akt3 promoter. We identified and cloned a candidate promoter containing a 1785 bp long GC rich fragment (including CpG island #146 as indexed by Ensemble database) located at the 5′-region of the mouse Akt3 gene ( Fig. 2A). And we also cloned a 718 bp long fragment from Akt3 gene 3′-region as a negative control. The results showed that the 5′-fragment but not the 3′-fragment was able to support the expression of the reporter gene in pGL3-Basic vector (Fig. 2B). Co-transfection of RIZ1-Flag plasmid and the Akt3 promoter reporter plasmid pGL3-Basic/CpG146 led to decreased luciferase activity (Fig. 2B). RIZ1 binding to the predicted sites of Akt3 gene promoter. Since the above results indicated RIZ1 regulation of Akt3 promoter, we next determined whether RIZ1 could directly bind to Akt3 promoter. Based on mouse and human genomic sequences of Akt3 gene available from the Ensembl database, we identified two CpG rich sites as candidate RIZ1 binding sites, including CpG:166 in the human gene and CpG:146 in the mouse gene (numbers mark the CpG counts as indexed by Ensemble). We designed PCR primers to cover these sites and performed chromatin immunoprecipitation (ChIP) assays to detect RIZ1 binding to these sites. RIZ1 binding was found in RIZ1-Flag plasmid transfected cells ( Fig. 3A and B) and in wild type (WT) mouse liver ( Fig. 3C and D).
To further verify RIZ1 binding to the Akt3 promoter, we performed electrophoresis mobility shift assay (EMSA) using recombinant GST-fusion proteins containing RIZ1 zinc finger motifs 1 to 3 (Zf1-3) and 8 (Zf8) (Fig. 4). Binding to the mouse promoter fragment CpG:146 was found for the RIZ1 Zf1-3 and the Zf8 proteins, which was abolished by an excess of unlabeled probe (Fig. 4B). However, these short truncated forms of RIZ1 could not affect Akt3 promoter reporter in co-transfection assays (data not shown).  Regulation of H3K9 methylation of Akt3 promoter by RIZ1. RIZ1 was a member of the SET/PR domain histone lysine methyltransferase family. We next examined RIZ1 effect on H3K9 methylation of Akt3 promoter. We performed ChIP-qPCR analysis with antibodies against transcriptionally repressive histone marks including H3K9me1, H3K9me2, and H3K9me3 (Fig. 5). In RIZ1-Flag transfected cells, we found increased levels of H3K9 mono-, di-, and tri-methylation in the Akt3 promoter.

Discussion
Estrogen receptor-negative breast cancer and androgen-independent prostate cancer lines show high levels of Akt3 mRNA 28 . In Akt3-knockdown cells, mitochondrial oxygen consumption rate was significantly reduced, indicating a critical role of Akt3 in mitochondrial respiration in human cancer cells 29 . And importantly, enhanced PI3K/Akt3 pathway activity is one of the main contributors in the genesis of melanoma 30 . Deregulated Akt activity has been shown to be responsible for 35-70% of advanced metastatic melanomas, and the majority of melanomas have elevated Akt3 expression and activity 31 . Akt3 may play an important role in the development of the disease and drug resistance 32,33 . Targeted inhibition of Akt3 in melanoma cell lines could decrease the pAkt level significantly 30,34 . However, the mechanism leading to deregulation of Akt3 remains to be better understood 35 . We in this study identified RIZ1 as a candidate repressor of Akt3 transcription. Our findings were consistent with known roles of RIZ1 and Akt3 in cancer pathogenesis. Future studies will be needed to determine whether RIZ1 (but not RIZ2) deregulation and Akt3 mis-expression may be coupled.
Insulin and Akt signaling modulates adipose tissue growth and adipogenesis 36 . Mouse embryonic fibroblasts (MEFs) lacking Akt display an inability to differentiate into adipocytes 37 , an RNAi-mediated decrease in Akt was found to block the differentiation of 3T3-L1 cells 38 . Since RIZ1 KO mouse exhibit diet-induced obesity and higher body weight, the link between RIZ1 and Akt3 found here may also play a role in obesity related pathways.
We here found that RIZ1 could bind to the promoter of Akt3 and suppress its expression. Like its related PR domain genes, RIZ1 also displays properties of transcription factors with a potential role in cell growth and tumorigenesis. RIZ1 gene products encode DNA-binding as well as transcription factor-binding activities 1,16,39 .
Recent studies provided evidence that malignant transformation results from a complex interplay of both genetic and epigenetic alterations affecting cell cycle regulation, cell proliferation, apoptosis, invasion, and angiogenesis, which finally lead to malignant phenotype development [40][41][42] . Histone methylation has an important role in transcriptional regulation 43 . Among them, H3K9 can be mono-, di-, or trimethylated and then induces distinct effects on chromatin compaction and gene expression 44 . The results here showing RIZ1 regulation of histone methylation on Akt3 promoter may help elucidate the link between histone methylation and cancer/obesity.

Materials and Methods
Cell culture.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting.
Cell lysates were prepared by incubating for 30 min in a buffer containing 25 mM Tris (pH 7.5), 75 mM NaCl, 5% glycerol, 2% SDS and protease/phosphatase inhibitors (Sigma) followed by centrifuging at 10,000 g for 10 min at room temperature. The protein concentration was measured using the BCA protein assay kit (Thermo Scientific). Proteins were separated by SDS-PAGE and transferred electrophoretically onto polyvinylidene fluoride (PVDF) membranes (Millipore), which were incubated overnight with the primary antibodies including mouse monoclonal anti-Akt3 (Santa Cruz Biotechnology), mouse monoclonal anti-Flag (Sigma), rabbit polyclonal anti-RIZ1 (Abcam), and rabbit polyclonal anti-β-actin (Sangon Biotech (Shanghai) Co., Ltd.). The bands were quantified by densitometry using ImageJ software.
Recombinant plasmid construction and luciferase assay. The mouse Akt3 gene promoter region containing CpG island 146 (CpG146) was generated by PCR using mouse genomic DNA. The 5′ primer, bearing a SacI site, was 5′-GCCTCGCCCAGGTGAATGT-3′, and the 3′ primer, bearing a NheI site, was 5′-CGCCAGCAGCGACAGCATCA-3′. The Akt3 promoter fragment was inserted into the pGL3-Basic vector, which includes luciferase as a reporter gene. For the luciferase assay, cells were washed with phosphate buffered saline and harvested with luciferase cell culture lysis reagent. Akt3 promoter activity in the cells was measured with the Luciferase Reporter Assay System using a Sirius luminometer (Titertek-Berthold). Luciferase activity was calculated in relative light units and normalized to the pCMV-RL vector containing the Renilla luciferase as control reporter.
ChIP assay. We performed ChIP assay using the Agarose ChIP Kit (Thermo Fisher Scientific PierceTM).

Electrophoretic Mobility Shift Assay (EMSA).
The human and mouse Akt3 promoter probe was synthetic by Sangon Biotech (Shanghai) Co., Ltd. as follows: 5′-TCCGCCCGCAATCTGTGTACCG TGTGTGGACGAATGCTTGCA-3′ (human), 5′-GGGGAGGGGCCGTCTGCGCATGTGGCGGGA GTGCGGGCGA-3′ (mouse). The DNA was included in the products amplified by human and mouse F1 and R1 primers. We used the LightShift ® Chemiluminescent EMSA Kit (Thermo Scientific) for the DNA binding analysis. The fragment was end-labeled using biotin 3′ end DNA labeling kit from Beyotime Biotechnology. Then the single strand DNA was annealed with the complementary chain to form a double strand DNA. Biotin end-labeled DNA was incubated with purified recombinant GST fusion proteins, followed by gel electrophoresis on a native polyacrylamide gel and transferred to a nylon membrane (Beyotime Biotechnology). The biotin end-labeled DNA was detected using the Streptavidin-Horseradish Peroxidase Conjugate (Beyotime Biotechnology) and the Chemiluminescent Substrate (Thermo Scientific).
Transfection experiments. pcDNA3.1/RIZ1-Flag was constructed by our team previously 6  Statistical analysis. Data were expressed as mean ± standard deviation of the mean (SD). Statistical analyses were performed using SPSS software version 19 (IBM Corporation). Significant differences among the treatment groups were assessed by one-way analysis of variance (ANOVA) and Tukey's multiple comparison tests. The level p < 0.05 were considered to indicate statistically significant differences.