EGR1 interacts with DNMT3L to inhibit the transcription of miR‐195 and plays an anti‐apoptotic role in the development of gastric cancer

Abstract EGR1 regulates the expression of its downstream target genes and may exert different biological effects in different tumours. We found that the expression of EGR1 was increased in gastric cancer (GC), and silencing the expression of EGR1 promoted the apoptosis of GC cells. Moreover, overexpression of EGR1 repressed the apoptosis of GC cells. Bioinformatics analysis showed that EGR1 had binding sites at the upstream promoter region of miR‐195; ChIP assays were applied to determine EGR1 occupancy of the miR‐195 promoter. The RT‐PCR results showed that EGR1 suppressed the expression of miR‐195. The mechanism by which EGR1 acts as a transcriptional repressor is still unclear. Bioinformatics analysis showed that EGR1 may interact with DNMT3L. We confirmed that EGR1 and DNMT3L formed a complex, and EGR1 was an important player in the transcriptional control of miR‐195. Overexpression of miR‐195 inhibited proliferation and promoted apoptosis in GC cells. We found a well‐matched miR‐195 binding site at the AKT3 3′‐UTR. Double luciferase reporter assays showed that AKT3 was a target of miR‐195, and silencing AKT3 repressed cell proliferation and promoted apoptosis. Our results indicated EGR1 may interact with DNMT3L to inhibit the miR‐195‐AKT3 axis and regulate the GC cell apoptosis.


| INTRODUC TI ON
a key role in cell growth, proliferation, apoptosis, migration, invasion and other physiological processes. EGR1 may exert different biological effects in different tumours. For example, in breast cancer, 6 bladder cancer 7 and lung cancer, 8 it acts as tumour suppressor gene, while in gastric cancer it promotes migration and invasion of gastric cancer cells, 9 suggesting that EGR1 may play the role of an oncogene in gastric cancer. EGR1 may act as both an activator and inhibitor of downstream target genes. The downstream molecules activated by EGR1 such as miR-152, 10 PTEN 11 and Siva-1, 12 and the downstream molecules inhibited such as Stathmin, 13 MEF2. 14 EGR1 interacts with Snail, 15 NF-kappaB, 16 Ref-1 17 to affect transcription. Bioinformatics analysis was used to analyse proteins bound to EGR1, and we identified DNA methyltransferase 3-like protein (DNMT3L) as a candidate. DNMT3L plays a key role in gene silencing by functioning as a repressor by directly binding to the HDAC1 protein. 18

| MTT assay
SGC-7901 and BGC-823 cells were seeded in 96-well plates. After 24 hours, they were transfected with the pre-miR-195 overexpression vector, miR-195 inhibitor, EGR1 vector, si-EGR1, si-AKT3 or their respective controls. The cells were then incubated for 24, 48 and 72 hours. MTT solution (20 μL) was added to each well and cells were incubated for 4 hours at 37°C. After that, the supernatants were discarded, and formazan crystals were dissolved in 150 μL dimethylsulphoxide (DMSO) and the absorbance was measured.

| Chromatin immunoprecipitation assay
The binding of EGR1 to the miR-195 promoter was detected by chromatin immunoprecipitation (ChIP). Protein/DNA complexes were obtained from BGC-823 cells, which were cross-linked with 1% formaldehyde for 15 minutes at room temperature, and quenched by adding glycine (0.125 mol/L) for 30 minutes. Cells were collected after rinsing two times with 5 mL of PBS and the nuclei were resuspended in Mg-NI, Mg-NIXP40, Ca-NI (with 0.5 mol/L EGTA) and lysis buffer (with protease inhibitors). The samples were sonicated by a cell cracker and the chromatin was sheared into ~200 bp fragments.
The sample was centrifuged to remove the insoluble material. 100 μl was used as input the lysates were divided into two parts and incubated with antibodies against EGR1 or IgG overnight at 4°C and then bound to protein G Sepharose (Invitrogen) for 2 hours at 4°C. The immunoprecipitates were consecutively washed twice byChIP lysis buffer, and finally TE buffer. The bound proteins were eluted from the beads by a solution containing 1% SDS and 0.1 mol/L sodium bicarbonate. Both the input and the samples were reverse cross-linked with proteinase K for 8 hours at 65°C. DNA was isolated using phenol/chloroform (Invitrogen). Promoter binding was detected by PCR using primers spanning the upstream regions of the miR-195 start sites. The primer sequences used are listed here:

| Western blot analysis
Protein was extracted with the radioimmunoprecipitation assay cell lysis buffer (Wolsen). Equal amounts of proteins in each sample were separated by 10% SDS polyacrylamide gels and electrophoretically transferred to an activated polyvinylidene difluoride membrane.

| Co-immunoprecipitation
Cellular proteins and protein complexes were extracted from SGC-7901 and BGC-823 cells by RIPA lysates. Then, the supernatant was added to the primary antibody. The antigen-antibody complex was placed on a slowly rotating shaker at 4°C for overnight. The DynabeadsTM Protein G (Invitrogen) were washed two times with PBS, and a 50% protein G beads working solution was prepared. This 50% protein G beads solution was then added into the sample solution, and the sample was placed on a slowly rotating shaker at 4°C for 4 hours or overnight. The sample was then centrifuged, the supernatant was discarded, and the protein G beads were collected. The beads were thereupon washed, and the supernatant was collected for Western blot.

| Statistical analysis
All data are presented as mean ± SEM. The Student's t test was used to evaluate differences between two groups. Data were considered to be statistically significant when P < .05.

| The miR-195 could inhibit proliferation and induce apoptosis in GC cells
To explore the function of miR-195 in gastric cancer, qRT-PCR was performed to detect the expression of miR-195 in GC and normal tissues. The results showed that miR-195 was downregulated in GC tissues compared to normal tissues ( Figure 1A). In addition, com-

| EGR1 affects miR-195 promoter activity in GC cells
To explore the mechanism of downregulation of miR-195 in GC, bioinformatic software was used to identify putative binding sequences at the miR-195 promoter region. We found that EGR1 binding sites were located upstream of the miR-195 gene The qRT-PCR was performed to detect the expression of EGR1, and the results showed that the expression of EGR1 mRNA levels was higher in GC tissues compared with their respective non-tumour tissue ( Figure 3G). These findings revealed that EGR1 and DNMT3L formed a complex that may act to repress transcription.

| EGR1 affects the apoptosis of gastric cancer cells
The qRT-PCR results demonstrated that si-EGR1 significantly silenced the expression of EGR1 ( Figure 4A

| AKT3 is a direct target of miR-195 and silencing the expression of AKT3 could inhibit proliferation and induce apoptosis in GC cells
The miRNA target prediction program was used to search for miR-195 target genes. We found a well-matched miR-195 binding site at the AKT3 3′-UTR. The sequence of miR-195 was highly conserved among species ( Figure 5A). The expression of AKT3 mRNA levels was higher in GC tissues compared with their respective non-tumour tissue ( Figure 5B). To determine whether AKT3 was a direct target of miR-195, the wt/mut AKT3 3′-UTR was transfected along with pre-miR-195 into HEK293 cells, and GLO was transfected with pre-miR-195 in HEK293 cells for control. After 24 hours, luciferase activity was markedly reduced in cells transfected with wt-AKT3 3′-UTR in comparison with control. The luciferase activity has no significant  Figure 5G). To verify the effect of silencing AKT3 on the protein level, we adopted Western blot analysis to detect the protein expression of AKT3, Bcl-2 and Bax. The results verified that after si-AKT3 transfection, the proteins AKT3 and BCL-2 were decreased compared with transfected si-control, while the protein expression of Bax was significantly upregulated in BGC-823 cells ( Figure 5H).

| EGR1 mediates miR-195 affect the GC progression by targeting AKT3
EGR1 bound to the promoter sequence of miR-195, repressing its expression. Moreover, EGR1 formed a complex with DNMT3L, which might be the reason for EGR1 acting as a transcriptional inhibitor.
Repressed expression of miR-195 promoted the GC cell progression, and inhibits the cell apoptosis by targeting AKT3 (Figure 6). These results indicated that EGR1 inhibited the transcription of miR-195. However, the mechanism by which EGR1 acts as a transcriptional repressor is still unclear. BioGrid software was used to forecast which protein could form a complex with EGR1. DNMT3L was identified as one of the candidates. Thus, we conducted co-IP assays to prove that EGR1 and DNMT3L indeed do form a complex and that DNMT3L acted primarily by regulation of the DNA methyltransferase.

| D ISCUSS I ON
The DNMT family has five members: DNMT1, DNMT2, DNMT3B and DNMT3L are highly expressed compared to non-tumour tissues, 24 and DNMT3L in testicular tumours is specifically expressed. 25 DNMT3L is a key factor affecting cell differentiation and tumour formation, and early studies have confirmed its transcriptional repression. 18 A previous study found that transcription factors such as FOS, MAFK, E2F3 and EGR1 could be interacted with DNMT3L through transcription factor array experiments, further suggesting that EGR1 may form a complex with DNMT3L. 26 Serine/threonine-specific protein kinase 3, AKT3, is a member of AKT family. There are three different AKT isoforms, AKT1, AKT2 and AKT3. 27 AKT is an important signalling pathway in regulating several cellular functions including nutrient metabolism, cell growth, apoptosis and survival. 28 AKT3 has been reported to be involved in cancer progression and plays important role in the regulations of various human cancers associated with miRNAs to modulate human cancers. [29][30][31] Combining experimental results with bioinformatics analysis, we speculate that miR-195 targets AKT3 to induce apoptosis in gastric cancer cells.
In conclusion, EGR1and DNMT3L formed a complex, which may be the reason why EGR1 was capable of inhibiting transcription of miR-195. Moreover, overexpression of miR-195 promoted apoptosis in gastric cancer cells by targeting AKT3, suggesting miR-195 acted as a tumour suppressor gene in the development of gastric cancer.

CO N FLI C T O F I NTE R E S T
The authors declare there is no conflict of interest.

AUTH O R ' S CO NTR I B UTI O N
YY and CH performed the experiments and wrote the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are available from the corresponding author upon request.