NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells

BBAGRM-00662; No. of pages: 11; 4C: 3, 4, 7, 8, 9 Biochimica et Biophysica Acta xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbagrm

3Q1

Fei Zhou a, Wei Wang a, Yujun Xing a,b, Tingting Wang a, Xinhui Xu a, Jinke Wang a,⁎ a b

The State Key Laboratory of Bioelectronics, Southeast University, Nanjing 210096, China Institute of Food Safety and Detection, Jiangsu Academy of Agricultural Sciences, Nanjing 210014, China

a r t i c l e

i n f o

R O

6

a b s t r a c t

Article history: Received 10 May 2013 Received in revised form 2 January 2014 Accepted 6 January 2014 Available online xxxx

D

P

As a transcription factor, NF-κB was demonstrated to regulate the expressions of miRNAs. However, only a few miRNAs have been identified as its targets so far. In this study, by using ChIP-Seq, Genechip and miRNA-Seq techniques, we identified 14 NF-κB target miRNAs in TNFα-stimulated HeLa Cells, including miR-1276, miR1286, miR-125b-1-3p, miR-219-1-3p, miR-2467-5p, miR-3200-3p, miR-449c-5p, miR-502-5p, miR-548d-5p, miR-30b-3p, miR-3620-5p, miR-340-3p, miR-4454 and miR-4485. Of these miRNAs, 8 detected miRNAs were also NF-κB target misRNAs in TNFα-stimulated HepG2 cells. We also identified 16 target genes of 6 miRNAs including miR-125b-1-3p, miR-1286, miR-502-5p, miR-1276, miR-219-1-3p and miR-30b-3p, in TNFα-stimulated HeLa cells. Target genes of miR-125b-1-3p and miR-1276 were validated in HeLa and HepG2 cells by reducing their expression of plasmids and mimics. Bioinformatic analysis revealed that two potential target genes of miR-1276, BMP2 and CASP9, were enriched in a disease phenotype. The former is enriched in osteoarthritis, and the latter is enriched in Type 2 diabetes and lung cancer, respectively. These findings suggested that this little known miRNA might play roles in these diseases via its two target genes of BMP2 and CASP9. The expression of miR125b-1 regulated by NF-κB has been reported in diverse cell types under various stimuli, this study found that its expression was also significantly regulated by NF-κB in TNFα-stimulated HeLa and HepG2 cells. Therefore, this miRNA was proposed as a central mediator of NF-κB pathway. These findings provide new insights into the functions of NF-κB in its target miRNA-related biological processes and the mechanisms underlying the regulation of these miRNAs. © 2014 Published by Elsevier B.V.

T

E

Keywords: NF-κB miRNA Target gene HeLa cell TNFα

R

E

C

7 8 9 10 11 12 14 13 15 16 17 18 19 20 21

O

4 5

F

2

NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells

1

44

R

43

1. Introduction

46

Nuclear factor-κB (NF-κB) was firstly identified in lymphocytes and subsequently found to function in various cells by regulating the expressions of a wide variety of genes [1]. The NF-κB family consisted of five members, p50, p52, p65/RelA, c-Rel, and RelB, which were classified into two groups: a group of p50 and p52, and the other group of RelA, RelB and c-Rel [2,3]. The members of the NF-κB family can form various heterodimers or homodimers to bind a DNA sequence motif known as κB sites in genome and thus regulates the transcriptions of many target genes through recruiting coactivators and corepressors [2]. The canonical NF-κB dimer is often sequestered in cytoplasm by binding its inhibitor, IκB. When the cells are exposed to stimuli, such as tumor necrosis factor α (TNFα), lipopolysaccharide (LPS), virus and ultraviolet (UV), IκB is degradated and the released NF-κB dimer transports into the nucleus where it binds to κB sites in genomes and regulates the transcription of its target genes [2,4]. Inducible regulation of gene expression allows

49 50 51 52 53 54 55 56 57 58 59 60

U

47 48

N C O

45

⁎ Corresponding author. Tel./fax: +86 25 83793620. E-mail addresses: [email protected] (F. Zhou), [email protected] (W. Wang), [email protected] (Y. Xing), [email protected] (T. Wang), [email protected] (X. Xu), [email protected] (J. Wang).

23 24 25 26 27 28 29 30 31 32 33 34 22 Q2 35 36 37 38 39 40 42 41

organisms to adapt to the changes of environmental, mechanical, chemical and microbiological stresses, which is essential for the survival, normal development and physiology of multicellular organisms [5]. Therefore, NF-κB plays a key role in the processes of emergency response to these changes [2,4,5]. It has been found that NF-κB controlled broad biological functions, such as cell survival, proliferation, differentiation, inflammation, immunity and tumorigenesis via its target genes [2,5]. In recent years, microRNA (miRNA) has been demonstrated to be another important class of molecules participating in gene expression regulation. MiRNA is a class of short (approximately 22 nt) endogenous noncoding RNA that can regulate gene expression at the posttranscriptional level and play a key role in many biological processes such as development, cell death, proliferation and immunity [3,6]. In the canonical biogenesis, miRNA genes are transcribed by either RNA polymerase II or RNA polymerase III to a long primary miRNA transcripts (pri-miRNA) with the size from several hundred nucleotides (nt) to several kilobases [7–9]. Pri-miRNA is recognized and cleaved in the nucleus by the microprocessor complex (consists of Drosha and DGCR8), resulting in an approximately 70-nt precursor hairpin RNA (pre-miRNA) [7–9]. Subsequently, pre-miRNA is processed to about 22-nt mature miRNA by Dicer in cytoplasm [9–11]. Finally, the mature miRNA is asymmetrically incorporated into the miRNA-induced silencing complex (RISC)

1874-9399/$ – see front matter © 2014 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.bbagrm.2014.01.006

Please cite this article as: F. Zhou, et al., NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbagrm.2014.01.006

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82

116 117

134

HeLa and HepG2 cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum, 100 units/ml penicillin and streptomycin at 37 °C in 5% CO2. For TNFα induction, cells were stimulated with 30 ng/ml TNFα (Sigma) in serumfree DMEM at 37 °C for 1 h. For siRNA transfection, the cells were firstly transfected with NF-κB p65 siRNA (Cell signaling Technology) at a concentration of 30 nM by Lipofectamine2000 (Invitrogen) for 48 h, then stimulated with 30 ng/ml TNFα in serum-free DMEM at 37 °C for 1 h. To regulate the expression of miRNA target genes, on one hand, HeLa cells were transfected for 48 h with GV268 vectors containing mir-1276 or mir-125b-1 gene fragments (Genechem, Shanghai) using Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocols. The blank GV268 vector was also transfected into HeLa cells as control. On the other hand, HeLa and HepG2 cells were transfected with doublestranded mimics of miR-1276 or miR125b-1-3p (GenePharma, Shanghai) using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. The double-stranded negative-control microRNA mimics (Genepharma, Shanghai) were also introduced into HeLa and HepG2 cells as negative control.

135

2.2. MiRNA-Seq and data analysis

136

Small RNA libraries were constructed for small RNA samples from HeLa cells, TNFα-treated HeLa cells and siRNA and TNFα dually treated HeLa cells by using the Illumina TrueSeq Small RNA Sample Preparation Kit (Illumina). Briefly, the total RNA (5 μg) was resolved on a denatured 8% polyacrylamide gel electrophoresis (PAGE), and a fraction of 18 nt to 30 nt was collected. The isolated small RNAs were sequentially ligated to a 3′ adapter using T4 RNA ligase 2 (New England Biolabs) and a 5′ adapter using T4 RNA ligase (New England Biolabs). The ligation

108 109 110 111

118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133

137 138 139 140 141 142 143

C

106 107

E

104 105

R

102 103

R

100 101

O

98 99

C

96 97

N

94 95

U

92 93

2.3. Real-time PCR

154

To detect the expression of mature miRNA, total RNA was isolated from HeLa and HepG2 cells with Trizol reagent (Invitrogen). An amount of 500 ng total RNA was reversely transcribed by using the BU-Script RT Kit (Biouniquer, Nanjing) in a 20 μl system. The reverse transcription primer was provided in Table S3. One microliter of reversely transcribed product was used to perform real-time PCR on a ABI StepOne Plus realtime PCR system (Applied Biosystems) with SYBR Green Real-time PCR Master Mix (Roche). The primers used to detect the expression of miRNAs were provided in Table S4. The expression of snRNA U6 was measured for normalizing the expressions of miRNAs. To detect expressions of protein genes, total RNA was firstly extracted from HeLa and HepG2 cells and reversely transcribed into cDNA by using PrimeScript RT Master Mix (Takara). The cDNAs were then detected with real-time PCR on ABI StepOne Plus real-time PCR systems (Applied Biosystems) with SYBR Green Real-time PCR Master Mix (Roche). The primers used to detect the expression of protein genes were also provided in Table S4. The expression of GAPDH was detected for normalizing the expressions of protein genes. All real-time PCR detections were performed in at least three technical replicates for each of three biological replicates. After each PCR reaction, the specificity of amplification was determined by performing melting curve analysis. Relative expression was calculated employing the comparative Ct method and the data were shown as mean values ± standard errors.

155 156

2.4. ChIP-Seq, data analysis and ChIP-real-time PCR

179

NF-κB ChIP-Seq was performed exactly as previously described [4]. In brief, HeLa cells (0.6–0.8 × 106/ml)were exposed to 30 ng/ml TNFα (Sigma) for 1 h, and then treated with 1% formaldehyde for 10 min at room temperature followed by quenching with glycine at the final concentration of 125 mM. HeLa cells were washed with PBS and swelled for 10 min in hypotonic lysis buffer. The cells were then subjected to 1 × RIPA buffer and the nuclear pellet was collected by centrifugation. The nuclear pellet was sonicated using a cell cracker for shearing chromatin into fragments at the length of 100 to 1000 bp. After sonication, the chromatin was centrifuged for collecting the clarified lysate. 10% of the diluted lysate was set aside for input, and then the clarified chromatin was divided in half and treated overnight with anti-NF-κB p65 (Abcam 7970) rabbit polyclonal antibody and normal rabbit IgG (Santa Cruz Sc-2027), respectively. The chromatin was then incubated with Dynabeads Protein A (Invitrogen) at 4 °C for 2 h for capturing NF-κB p65 binding DNA. Subsequently, the immunoprecipitates were successively washed three times with low wash buffer, high wash buffer and TE buffer. The protein–DNA complexes were then eluted from the beads with elution buffer. The elutes were added RNase A (0.5 mg/ml), Proteinase K (0.2 mg/ml) and NaCl (300 mM) and incubated overnight at 65 °C for reversing cross-linking. Finally, DNA was extracted from elutes with QIA-quick PCR purification kit (QIAGEN) for Illumina Solexa sequencing and PCR detecting. The raw data of ChIP-Seq reads was firstly cleared according to its unique genomic mapping and pairing. The ChIP-Seq reads were mapped onto unique positions on the human genome (hg19) by using

180

F

2.1. Cell culture and treatment

90 91

O

115

89

R O

2. Materials and methods

87 88

144 Q3 145

P

114

85 86

products were reversely transcribed using SuperScriptΠReverse Transcriptase (Life Technologies) and amplified with PCR (12 cycles). The PCR products were purified with electrophoresis of 6% polyacrylamide gel. The purified PCR products were then used to cluster and sequence with the Illumina HiSeq2000 (Illumina). The sequencing data was firstly qualified with FastQC software, and removed the sequence of primers, rRNA, tRNA, snRNA and snoRNA. Subsequently, the sequences at the length of 17 nt to 35 nt were obtained. The program of mirDeep was used to identify these miRNAs in genome. Finally, the miRNAs with differential expression were identified by two programs, edgeR and DESeq.

T

112 113

to silence its target genes at either the transcriptional or the translational level [12,13]. The miRNA gene, classified as a Class II gene along with all protein genes, can be elaborately controlled through various regulatory mechanisms including transactivation and transrepression by nuclear transcription factors [7,14]. MiRNAs exhibit tissue- or stage-specific expressions in development and respond to extracellular stimuli, also indicating that their expressions were tightly controlled [6,15]. The similar functions shared by miRNAs and NF-κB in cell proliferation, cell death, inflammation and development of human cancer have promoted researchers to explore the relationship of these two classes of regulators. Recently, some studies have reported that NF-κB could elaborately control the expressions of miRNA genes, such as miR-125b-1, miR-146, and miRNA-30b [3,14,16–18]. By these target miRNAs, NF-κB was found to be involved in some new biological processes, which shed new insights into the biological functions of this important transcription factor. However, up to date, only a few miRNAs have been identified to be regulated by NF-κB. TNFα is a well-known NF-κB activator and plays important roles in NF-κB-related inflammation deterioration. To find the possible miRNAs activated in this process and their potential functions, we performed this study for identifying the potential miRNAs regulated by NF-κB and their target genes. As a result, we have identified 16 NF-κB target miRNAs and several target genes of these miRNAs in TNFα-stimulated HeLa cells. We identified two target genes of miR-1276, BMP2 and CASP9, in which BMP2 was enriched in osteoarthritis and CASP9 was enriched in Type 2 diabetes and lung cancer, respectively. This finding suggested that this little known miRNA may play a role in these diseases via these target genes. We also found that a known target miRNA of NF-κB, miR125b-1, was also significantly regulated by NF-κB in TNF-α-stimulated HeLa cells, indicating that this miRNA was a central mediator of NF-κB pathway.

D

83 84

F. Zhou et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx

E

2

Please cite this article as: F. Zhou, et al., NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbagrm.2014.01.006

146 147 148 149 150 151 152 153

157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178

181 182 183 184 185 186 Q4 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205

3

O

F

F. Zhou et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx

U

N C O

R

R

E

C

T

E

D

P

R O

Fig. 1. NF-κB-regulating miRNAs in TNFα-stimulated HeLa cells. A. Numbers of miRNAs responding to TNFα stimulation and p65 siRNA interference with changed expression. Data obtained from both deseq and edgeR analysis. B. Expressions of mature miRNAs in TNFα-stimulated HeLa cells. The expressions of 18 miRNAs were regulated by NF-κB (p b 0.05 or 0.05 ≤ p b 0.2). The diagram was made according to deseq data. The suffix ‘5p’ and ‘3p’ represent the mature miRNA excised from 5′ arm or 3′ arm of its corresponding precursor, respectively.

Fig. 2. Detection of NF-κB-regulating miRNAs in TNFα-stimulated cells. A. MiRNA expression in TNFα-stimulated HeLa cells. B. MiRNA expression in TNFα-stimulated HepG2 cells. Relative quantitation (RQ) was calculated with the comparative Ct method and normalized against U6. *p ≤ 0.05; **p ≤ 0.01. C, control cells; T, TNFα-stimulated cells; S, siRNA-interfered cells.

Please cite this article as: F. Zhou, et al., NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbagrm.2014.01.006

229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250

F

O

R O

227 228

Firstly, the target genes of miRNAs were predicted by using three programs, miRanda [30], miRwalk [31] and DIANA-microT-CDS [32], respectively. The strict score threshold was set to 0.75 in the process of DIANA-microT-CDS prediction. In miRwalk prediction, the following parameters were adopted to predict miRNA binding sites on 3′ UTR and 2000 bp upstream flanked sequence of mRNA. The term of transcript was set the longest transcript of genes. Minimum seed length was 7 nt; p value was 0.05. The prediction results of miRanda and miRwalk were compared using the comparative analysis platform provided by the miRwalk web. Finally, the genes that were commonly predicted by three programs were regarded as the high-confidence miRNA targets.

P

225 226

262

D

223 224

2.5. MiRNA target gene prediction

2.6. Gene expression profiling and data analysis Gene expression profiling was performed exactly as previously described [4]. Briefly, total RNA was prepared from HeLa cells, TNFαtreated HeLa cells and siRNA and TNFα dually treated HeLa cells with a Trizol reagent (Invitrogen) according to the manufacturer's instructions. To profile the global gene expression, the qualified RNA samples were analyzed with Affymetrix Human Genome U133 Plus 2.0 Array according to standard Affymetrix protocols (Affymetrix China, Shanghai). The GeneChip® Operating Software was applied to initial array image analysis and quantification. Following normalization and background filtration, the signal intensity data of treated cells were compared with that of control cells, and the gene with intensity ratio over or less two folds was considered as upregulated or downregulated genes [33,34]. Genetic-association-DB-Disease analysis was performed by uploading the genes to databases of DAVID 6.7 system (http://david.abcc.ncifcrf. gov/).

T

221 222

C

219 220

E

217 218

R

215 216

R

213 214

O

212

251 252

C

210 211

data were analyzed based on the previous reports [27–29] as follows: the Ct input value was first adjusted to 100% of input by subtracting 3.322 cycles (log2 of 10) from the Ct value of 10% diluted input. The Ct values of NF-κB p65 antibody ChIPed sample and the negative sample were normalized to the Ct value of input sample as: ΔCtp65 = Ctp65 − (Ctinput − 3.322) and ΔCt IgG = Ct IgG − (Ctinput − 3.322). Next, a ΔΔCt value was calculated as: ΔΔCt = ΔCtp65 − ΔCt IgG. Then, the fold enrichments between NF-κB p65 antibody ChIPed sample and the negative sample were determined as the formula of fold enrichment = 2−ΔΔCt. The data were shown as mean values ± standard errors from three independent experiments.

N

208 209

ELAND [19] using the default setting allowing up to 2 mismatches. Peaks in the mapped ChIP-Seq reads were called with ChIP-Peak (http://ccg. vital-it.ch/chipseq) in parameters as follows: Window Width: 200; Vicinity Range: 200; Peak Threshold: 1; Count Cut-off: 10; select Repeat Masker. The peaks could be visualized with a genome browser of the University of California, Santa Cruz (UCSC). The target genes bound by NF-κB referred to the nearest RefSeq, Known gene, Genbank mRNA or EST. In this study, the miRNAs with high-confidence ChIP-Seq peaks (fold enrichment ≥ 10) in the genomic region from −100 kb upstream of pre-miRNA start site to + 100 kb after the transcription end site (TES) was determined as NF-κB-binding miRNAs according to previously reported long-range distance of NF-κB binding sites to its target genes [4,20–22]. To find κB sites in ChIP-Seq peaks, the sequences of ChIP-Seq peaks were searched with ten κB motifs reported by other previous studies [4], including GGGRNNTYCC (R: A, G; Y: T, C; N: A, G, C, T), GGRRNNYY CC, NGGRNTTYCC, RGGRNNHHYY (B: C, G, T; H: A, T. C), DGGGGGTTTY (D: A, G, T), GGKRRWKBHB (K: T, G; W: A, T), GKRVTTYCCV (V: A, C, G), GKVNWTYCCV, RGGGGRWKTW, NGGGGRWDDY. The motif GGGR NNTYCC came from TRANSFAC [23], and motif GGRRNNYYCC was derived from in vitro Protein binding microarray study [24]. NGGRN TTYCC was derived from ChIP-PET data in LPS-induced THP-1 cells [25]. These three motifs were regarded as canonical κB motifs. The remaining 7 κB motifs were obtained by a recent in vitro DNA-binding profiling technique, SELEX-Seq, with RELA-containing dimmers [26], which were regarded as non-canonical κB motifs. The motifs DGGGGGTTTY and GGKRRWKBHB, GKRVTTYCCV and GKVNWTYCCV, and RGGGGRWKTW and NGGGGRWDDY were generated from the top 50 and 1000 DNA binders to a RELA-p50 heterodimer, RELA–RELA homodimer, and RELA-p52 heterodimer, respectively. RGGRNNHHYY was regarded as a new expanded κB motif. To detect NF-κB binding to ChIP-Seq peaks with real-time PCR, 1.5 μl the NF-κB p65 antibody ChIPed DNA and an equal amount of 10% input material were subjected to real-time PCR on an ABI StepOne Plus realtime PCR system (Applied Biosystems) with SYBR Green Real-time PCR Master Mix (Roche) in a 10 μl reaction system. Besides, 1.5 μl Rabbit IgG ChIPed material was used as negative control, and 1.5 μl no DNA template (water) was used to monitor PCR process. STAT5A, a known NF-κB target gene, was used as positive gene control to evaluate the NF-κB p65 antibody ChIPed samples. The primers for ChIP-real-time PCR were designed according to the DNA sequences of a higher ChIPSeq peak at the 5′ proximal region of miRNA gene. The sequences of primers were shown in Table S5. To determine the fold differences (Fold enrichment) between the NF-κB p65 antibody ChIPed sample and the negative control (IgG ChIPed sample), the real-time PCR

U

206 207

F. Zhou et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx

E

4

Fig. 3. NF-κB target miRNAs. A. Number of NF-κB target miRNAs. Six miRNAs were up-regulated and 8 miRNAs were down-regulated by NF-κB. B. Number of ChIP-Seq peaks with fold enrichment ≥ 10 linked with miRNA genes.

Please cite this article as: F. Zhou, et al., NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbagrm.2014.01.006

253 254 255 256 257 258 259 260 261

263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290

F. Zhou et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx t1:1 t1:2

Table. 1 κB sites in ChIP-Seq peaks in the gene region of miRNA genes.

5

3.2. Identification of NF-κB-binding miRNAs

336 337 338

291

3. Results

Expression of most NF-κB-regulating genes required NF-κB binding to the κB sites in their gene regions [2,4]. In addition, several NF-κBregulating miRNAs have been reported to be bound by NF-κB in previous studies, such as mir-125b-1 and mir-30b [14,18]. Therefore, to find whether the above described 18 NF-κB-regulating miRNAs were bound by NF-κB in TNFα-stimulated HeLa cells, we performed a ChIP-Seq assay to identify NF-κB-binding miRNAs. As a result, we identified a total of 139,679 ChIP-Seq peaks with various fold enrichments (FEs), in which 41,194 peaks had enrichment folds ≥10. Next, we analyzed the correlation of these peaks with all pre-miRNAs in the human genome. The results showed that all peaks were related to 1439 miRNA genes and each miRNA gene was averagely linked with 97 peaks. Even to the peaks with FE ≥ 10, they linked with 1418 miRNA genes and each miRNA gene was averagely linked with 29 peaks. In order to find the high-confidence miRNAs bound by NF-κB, only miRNAs linked with peaks with FE ≥ 10 were regarded as NF-κB-binding miRNAs. Under this condition, we identified 1418 NF-κB-binding miRNAs in TNFαstimulated HeLa cells.

292

3.1. Identification of NF-κB-regulating miRNAs

3.3. Identification of NF-κB target miRNAs

355

293 294

In order to identify the miRNAs regulated by NF-κB in TNFαstimulated HeLa cells, we manipulated the endogenous NF-κB activity by TNFα stimulation and NF-κB p65 siRNA interference (Fig. S1A). Subsequently, we performed miRNA-Seq for identifying the miRNAs regulated by NF-κB. We identified the differentially expressed miRNAs by using two programs, edgeR and DESeq, in order to find highconfidence miRNAs that were commonly identified as differentially expressed miRNAs by two algorithms. The results showed that the expressions of 18 mature miRNAs were significantly changed by the activation (TNFα stimulation) and inhibition (p65 siRNA interference) of NF-κB activity in HeLa cells (Fig. 1A). Of these miRNAs, miR-125b1-3p, miR-2467-5p, miR-27a-5p, miR-4454 and miR-1286 were significantly upregulated (p b 0.05) by TNFα. However, under the p65 siRNA interference, miR-125b-1-3p was significantly downregulated (p b 0.05), and other miRNAs (such as miR-1286, miR-2467-5p, miR27a-5p, miR-4454) showed the decreased expression (0.05 ≤ p b 0.2). On the contrary, miR-340-3p, miR-3620-5p, miR548ay-5p, miR-548d-5p and miR-1276 were significantly downregulated by TNFα (p b 0.05), but their expression were recovered by p65 siRNA (0.05 bp b 0.2) (Fig. 1B). Therefore, we regarded these miRNAs as NF-κB-regulating miRNAs in TNFα-stimulated HeLa cells. To confirm the expression of miRNA, we detected the expressions of eight mature miRNAs in TNFα-stimulated HeLa cells by real-time PCR. The results showed that the expressions of miR-125b-1-3p, miR-5025p and miR-1286 were significantly upregulated by TNFα stimulation and markedly repressed by p65 siRNA interference, but miR-30b-3p, miR-219a-1-3p, miR-340-3p, miR-1276 and miR-3620-5p were significantly downregulated by TNFα stimulation and significantly recovered by p65 siRNA interference in HeLa cells (Fig. 2A). These results were in agreement with the results of miRNA-Seq. To further confirm these NF-κB-regulating miRNAs, we also manipulated the NF-κB activity in HepG2 cells by TNFα and NF-κB p65 siRNA treatments (Fig. S1B), and detected the expressions of the same miRNAs with real-time PCR. The results revealed that these miRNAs showed a similar expression profile in the two cell lines under TNFα and p65 siRNA treatments (Fig. 2B). Based on these results, we identified 18 NF-κB-regulating miRNAs in TNFα-stimulated HeLa cells, and some of these miRNAs were also regulated by NF-κB in TNFα-stimulated HepG cells. Among these NF-κBregulating miRNAs, miR-125b-1 and miR-30b have been previously reported to be regulated by NF-κB in H69 cells [14,16–18]. Therefore, we identified 16 new NF-κB-regulating miRNAs in TNFα-stimulated HeLa cells and eight miRNAs were also regulated by this transcription factor in TNFα-stimulated HepG2 cells.

In NF-κB-regulating miRNAs, some may be indirectly regulated and others may be directly regulated by this transcription factor. The latter has been attracted more attention than the former in the current researches of transcription factor and miRNA. We are also interested in and focused on the latter in this study. To find miRNAs that are directly regulated by NF-κB (namely NF-κB target miRNA), we compared the lists of NF-κB-regulating and NF-κB-binding miRNAs described above. The results showed that 14 NF-κB-regulating miRNAs were also bound by NF-κB (Fig. 3A), therefore, these miRNAs were considered as NF-κB target miRNAs. In these miRNAs, six were NF-κB-activated miRNA and 8 were NF-κB-repressed miRNA (Fig. 1). NF-κB positive and negative regulation to its target miRNAs is similar to its regulation of its target protein genes [4]. To confirm these NF-κB target miRNAs, we checked NF-κB binding to these miRNAs by searching ten κB motifs identified by previous studies [4] in all ChIP-Seq peaks linked with the miRNAs (Fig. 3B). The results showed that there were many κB sites in these miRNA-linked NF-κBbinding peaks and at least 2.75 κB sites in each peak (Table 1). The peaks of mir-3620 contained the most κB sites. This miRNA was linked with 40 peaks of FE ≥ 10 and there were a total of 746 κB sites in these peaks, averaged 18.65 κB sites per peak. The presence of κB sites in ChIP-Seq peaks demonstrated that NF-κB bound to its target miRNAs. (See Table 2.) To further confirm these NF-κB target miRNAs, we detected NF-κB binding to five of NF-κB target miRNAs with ChIP-real-time PCR. To each miRNAs, we firstly checked their ChIP-Seq peaks and found the peaks with the highest FE and nearest to miRNA genes. We also searched κB sits in these peaks. The results were shown in Fig. 4A–E. ChIP-real-time PCR was then performed to detect the NF-κB binding to the peak nearer to miRNA genes and determinate fold enrichment between a NF-κB p65 ChIPed sample and the negative control (IgG ChIPed sample). The fold enrichments determined by real-time PCR were over at least 17 (mir-219-1) indicated that NF-κB intensively bound to these peaks in TNFα-stimulated HeLa cells (Fig. 4F). We further detected and found that NF-κB also strongly bound to these peaks in TNFα-stimulated HepG2 cells and the fold enrichments were over at least 8-fold (mir-1286) (Fig. 4G). Based on these results, we identified 16 NF-κB target miRNAs in TNFα-stimulated HeLa cells and some of them were also found to be NF-κB target miRNAs in TNFα-stimulated HepG2 cells. Among these miRNAs, two miRNAs, miR-125b-1 and miR-30b, have been previously identified NF-κB target miRNAs in H69 cells [14,18]. Others were newly identified NF-κB target miRNAs. Of these new NF-κB target

356

miRNA genes

Number of peak (FE ≥ 10)

Total κB sites in all peaks

Averaged κB sites per peak

t1:4 t1:5 t1:6 t1:7 t1:8 t1:9 t1:10 t1:11 t1:12 t1:13 t1:14 t1:15 t1:16 t1:17

mir-1276 mir-1286 mir-125b-1 mir-219-1 mir-2467 mir-3200 mir-449c mir-502 mir-548d-2 mir-30b mir-3620 mir-340 mir-4454 mir-4485

28 44 4 38 33 45 23 4 21 23 40 53 11 39

370 658 11 464 293 335 148 28 102 158 746 373 24 286

13.21 14.95 2.75 12.21 8.88 7.44 6.43 7 4.86 6.87 18.65 7.04 2.18 7.33

307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335

O

R O

P

D

E

T

C

305 306

E

303 304

R

301 302

R

299 300

N C O

297 298

U

295 296

F

t1:3

Please cite this article as: F. Zhou, et al., NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbagrm.2014.01.006

339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354

357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 Q5 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398

6 t2:1 t2:2

F. Zhou et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx

Table 2 Results of genetic association DB disease categorization using DAVID 6.7. Term

Count

P Value

NF-κB-regulating genes

Fold enrichment

t2:4

Diabetes, type 2

27

0.0099

1.64

t2:5

Colorectal cancer

24

0.0007

t2:6

Lung cancer

20

0.0022

t2:7

Stomach cancer

13

0.0069

t2:8

Atherosclerosis, coronary

20

0.0009

t2:9 t2:10 t2:11 t2:12 t2:13 t2:14 t2:15

Sclerosis, systemic Polycysticovary syndrome Juvenile arthritis Osteoarthritis Pancreatitis, chronic Endometriosis Sepsis

8 7 7 9 7 9 7

0.0039 0.0033 0.0048 0.0088 0.0008 0.0098 0.0057

NAMPT, CCL2, PTGS2, LDLR, BDKRB1, SLC19A2, CDKN2B, SOS1, SERPINE1, INSR, KLF5, KLF6, ICAM1, IL6, IRS2, KLF12, CFB, ADM, GFPT1, MGEA5, CCL5, CASP9,ANXA1, SOD2, F3, FOXC2, TNF ICAM1, IL6, IRS2, CCL2, PTGS2, NFKBIA, NFKB1, KRAS, CDKN2B, SERPINE1, MTR, GDF15, INSR, APC, TNF, GSTA4, CYP1B1, IL8, RRM2B, MLH3, FAM46A, SOD2, IL1B, PLAU KLF6, ICAM1, IL6, PTGS2, AKAP9, KRAS, MTR, PDCD5, APC, SAT1, NBN, GSTA4, CYP1B1, IL8, RRM2B, ATR, LIG4, SOD2, CASP9, IL1B IL6, PTGS2, KRAS, APC, NBN, TNF, IL8, SOD2, GSK3B, MTR,IL1B, HSPA13, IL1A ICAM1, CCL2, LDLR, PTGS2, EDN1, BDKRB1, CXCL16, MTR, SERPINE1, CFH, TNFAIP3, INSR, PTAFR, TNF, OLR1, LMAN1, CCL5, SOD2, IL1B, IL1A IL6, CCL2, EDN1, TNF, IL8, IL1B, CCL5, IL1A IL6, FST, SERPINE1, INSR, STS, TNF, GSK3B IL6, CCL2, IRF1, TNF, IL1B, SLC26A2, IL1A PTGS2, RHOB, TNFAIP6, BMP2, TNF, IL8, IL1B, SLC26A2,IL1A ICAM1, IL6, CCL2, TNF, IL8, IL1B, SOD2 ICAM1, IL6, KRAS, SERPINE1, NRIP1, TNF, CYP1B1, IL1B, CCL5 IL6, CXCL2, SERPINE1, TNF, IL1B, IL1A, SOD2

O

R O

407

To identify the target genes of NF-κB target miRNAs, we firstly detected the global expression profiles of HeLa cells, TNFα-treated HeLa cells and p65 siRNA/TNFα dually treated HeLa cells by using DNA microarrays. As a result, 432 upregulated genes (Fold change ≥ 2.0) and 582 downregulated genes (Fold change ≤ −2.0) were found (Fig. 5 and Table S1). It was found that some typical known target genes of NF-κB were included in these NF-κB-regulating genes, such as CD83, IL6, CCL20, NFKBIA and TNFAIP3 (Fig. 5 and Table S1). The functions of these NF-κB-regulating genes were preliminary investigated by using DAVID 6.7 online servers under the Genetic Association DB Disease class with the thresholds of at least six genes in each term (p ≤ 0.01). The results revealed that 66 genes were significantly enriched in various disease phenotypes, such as colorectal cancer, lung cancer, and Type 2 diabetes, terms of these diseases enriched 24, 20, 27 genes, respectively (Table 1). Next, we predicted the putative target genes of NF-κB target miRNAs by using miRanda [30], miRwalk [31] and DIANA-microT-CDS [32] online. The genes that were commonly predicted by three programs were considered as putative target genes of miRNAs. As a result, a total of 576 genes were predicted as the target genes of miRNAs (Table S2). We compared these predicted target genes of miRNAs with NF-κB-regulating genes (Table S1), 16 predicted miRNA target genes were found to be regulated by NF-κB (Fig. 6A). Moreover, under the TNFα stimulation and p65 siRNA interference, the expressions of these genes were reversely correlated with the expressions of miRNAs in HeLa cells. For example, PPP1R12A was one of the predicted target genes of miR-125b-1-3p; the expression of miR-125b-1-3p was enhanced by TNFα stimulation and repressed by p65 siRNA interference, but the expression of PPP1R12A was repressed by TNFα-stimulation and enhanced by p65 siRNA interference. Thereby, this gene was regarded as the target gene of miR-125b-1-3p (Fig. 6B). Likewise, the gene MAP3K2 was regarded as the target gene of miR-502-5p, and the

418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438

C

E

R

R

416 417

O

414 415

C

412 413

N

410 411

U

408 409

2.27

3.81 4.56 4.25 2.99 5.87 2.93 4.11

P

3.4. Identification of target genes of miRNAs

2.38

genes SIAE, BMP2, CASP9 and NR4A3 were regarded as target genes of miR-1276. The gene NR4A3 was also targeted by miR-219-1-3p. In a recent study, we identified 584 NF-κB direct target protein genes and 480 NF-κB indirect target protein genes in HeLa cells by TNFα stimulation and p65 siRNA interference [4]. In comparison with these genes, we found that some of them were also target genes of NF-κB target miRNAs, such as PPP1R15B, GNAI1 and THBS1 (Fig. 6B), suggesting that these genes may be dually regulated by NF-κB and NF-κB target miRNAs. Anyway, we found that some of target genes of NF-κB target miRNAs were NF-κB indirect target genes, such as SIAE, NR4A3, CASP9, BMP2, PPP1R12A and MAP3K2 (Fig. 6B), suggesting that NF-κB may regulate these genes via its target miRNAs. To validate these target genes of NF-κB target miRNAs, we detected the expressions of two miRNAs, miR-125b-3p and miR-1276, in HeLa cells that were transfected with plasmids containing mir-1276 or mir-125b-1 gene fragments. We found that when the expressions of two miRNAs were significantly upregulated in the transfected cells (Fig. S2A), the expressions of PPP1R12A, BMP2, CASP9, NR4A3, PPP1R15B and SIAE were significantly down regulated (Fig. 7A). These results indicate that PPP1R12A is a target gene of miR-125b-3p, and BMP2, CASP9, NR4A3, PPP1R15B and SIAE were target genes of miR-1276. To further confirm these results, we also transfected HeLa cells with mimics of miR-125b-3p and miR-1276 and found that along with overexpressions of two miRNAs (Fig. S2B), the expressions of these genes were repressed or significantly repressed in HeLa cells (Fig. 7B). In addition, we also found that when the expressions of two miRNAs were upregulated by transfecting their mimics in HepG2 cells (Fig. S2C), the expressions of their corresponding target genes were similarly inhibited or significantly inhibited in this cell line (Fig. 7C).

439

3.5. The central role of miR-125b-1 in NF-κB signal pathway

468

The miRNA miR-125b-1 has been reported as an important cancerrelated miRNA [39–41]. In this study, we identified this miRNA as a target of NF-κB and identified PPP1R12A as its target gene in TNFαstimulated HeLa and HepG2 cells. The protein encoded by PPP1R12A, MYPT1, could modulate cell survival [42,43]. In addition, the expression of miR-125b-1 was also up-regulated by NF-κB in biliary epithelial cells following Cryptosporidium parvum infection or LPS stimulation [14,18], in primary human brain cells stimulated by amyloid beta 42 (Aβ42) peptides and interleukin-1beta (IL-1β) [44], and in human keratinocyte and embryonic kidney cells radiated with ultraviolet [45]. The activation

469

D

406

402 403

2.11

T

404 405

miRNAs, several have been reported to be involved in cancer progression. For instance, miR-340 has been reported to inhibit colorectal cancer cell growth [35] and breast cancer cell migration and invasion [36]. Overexpression of miR-548d inhibited pancreatic cell proliferation and sensitization to gemcitabine [37]. miR-548d-5p and miR-1286 were found to be downregulated by two most commonly used clinical agents, cisplatin and 5-fluorouracil (5-FU), in esophageal cancer cells [38].

400 401

2.10

E

399

F

t2:3

Please cite this article as: F. Zhou, et al., NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbagrm.2014.01.006

440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467

470 471 472 473 474 475 476 477 478 Q6

7

N C O

R

R

E

C

T

E

D

P

R O

O

F

F. Zhou et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx

Fig. 4. ChIP-Seq peaks and their ChIP-real-time PCR detection in TNFα-stimulated cells. A–E. Presentation of ChIP-Seq peaks (FE ≥ 10) linked with 5 miRNA genes. The highest peaks (red arrow labeled) are enlarged and their start points are labeled with numbers. Stars refer to the ChIP-real-time PCR detected peaks which were also enlarged and pointed out their start points. The numbers of κB sites found in peaks are shown in brackets. F–G. Fold enrichment between NF-κB p65 antibody ChIPed material and rabbit IgG ChIPed material (negative control) in TNFα-stimulated HeLa cells (F) and HepG2 cells(G) analyzed by ChIP-real-time PCR, respectively. Fold enrichment determination was described in materials and methods.

480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495

of NF-κB might result in the upregulation of miR-125b in human breast cancer cells exposed to Taxol, an effective chemotherapeutic agent for treatment of cancer patients [46]. However, miR-125b was also reported to be downregulated by NF-κB in mouse macrophages [16]. Owing to the limitation of test techniques at that time, the researchers were not capable of distinguishing the mature forms of miR-125b-1 and miR-125b-2 by using probe assay [16]. These data demonstrate that the NF-κB regulates this miRNA in a cell and stimuli independent format, suggesting the importance of this regulation in mammalian cells. On the other hand, it has been reported that miR-125a and miR125b could constitutively activate NF-κB by repressing TNFα-induced protein 3 (TNFAIP3) [47]. In this study, we found that there were putative miR-125b-1-3p binding sites on the 3′ UTR region of TNFAIP3 by using two predictive algorithms, miRanda and miRwalk. In our recent study, TNFAIP3 was identified as a target gene of NF-κB, its expression was upregulated 13-fold by TNFα and downregulated 199-fold by p65

U

479

siRNA in HeLa cells [4]. Other previous studies also identified TNFAIP3 as a target gene of NF-κB [48]. Therefore, we propose that miR-125b-1 plays a central role in the NF-κB signal pathway. Briefly, the cellular stimuli such as TNFα and LPS activated the NF-κB pathway; miR-1251 is thus upregulated by NF-κB and then represses the expression of PPP1R12A. In this way, a positive feedback loop is established in the NF-κB signal pathway via miR-125b-1-3p repression of TNFAIP3, a key player in the negative feedback regulation of NF-κB signaling in response to multiple stimuli [49] (Fig. 8). Because TNFAIP3 is a critical inhibitor of NF-κB signaling [49], its repression by miR-125b-1-3p may contribute to NF-κB-related pathogenesis [50].

496

4. Discussion

507

497 498 499 500 Q7 501 502 503 504 505 506

NF-κB was firstly identified in lymphocyte [1]. As an inducible 508 transcription factor [2,4], it plays a central role in regulating cell survival, 509 differentiation, inflammation, immune, and tumorigenesis via its target 510

Please cite this article as: F. Zhou, et al., NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbagrm.2014.01.006

F. Zhou et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx

O

F

8

526 527

P

D

E

T

C

524 525

E

522 523

R

520 521

R

518 519

O

516 517

bound by NF-κB but their expressions were not regulated by this factor was similar to the fact that many protein genes were also bound by NFκB but their expressions were not regulated by this factor [4]. In our recent study, we found that only the expression of 4.3% NF-κB-binding protein genes was regulated by this factor [4]. In line with our findings, the expression of many NF-κB-binding genes on chromosome 22 was also not regulated by the TNFα-activated NF-κB [22], and only a small fraction of RelA-binding genes (24%) were associated with gene expression variation in LPS-stimulated THP-1 cells [20]. These data suggest that the binding of NF-κB to the miRNA genes just provides an opportunity for its target miRNA selectivity; the activation of NF-κB-binding miRNAs may require other cell-specific partners. In this study, we identified 14 NF-κB target miRNAs in HeLa cells (Fig. 1 and 3). Some of these miRNAs were found to be NF-κB target miRNAs in HepG2 cells (Fig. 2 and 4). Of these miRNAs, some have been reported to play roles in cellular processes. For instance, miR-340 inhibited colorectal cancer cell growth [35] and breast cancer cell

C

514 515

genes [2,5]. Until now, it has been known that NF-κB provides a critical mechanistic link between inflammation and cancer [51]. In this study, we found that NF-κB may play roles in the development of colorectal cancer, lung cancer, and diabetes by performing the Genetic Association DB Disease class analysis to the genes regulated by this transcription factor (Table 1). So the complex biological functions of NF-κB are a great challenge for its application as a therapeutic target. Therefore, identification of all NF-κB target genes including miRNAs is crucial to uncover its whole functions in cells and disclose the mechanisms underlying its controls of cellular processes and disease development. It has been reported that NF-κB could bind to miRNA genes and regulate their expressions, such as miR-125b-1 and miR-30b [14,18]. In this study, we globally identified the binding profile of NF-κB to miRNA genes in the human genome. As a result, we found that NF-κB bound 1418 miRNA genes. However, only the expressions of 14 of these miRNAs (0.99%) were regulated by NF-κB in the TNFα-stimulated HeLa cells (Fig. 3). This phenomenon that many miRNA genes were

N

512 513

U

511

R O

Fig. 5. NF-κB-regulating protein genes in TNFα-stimulated HeLa cells. Scatter plots showed the fold changes of gene expression under the TNFα stimulation and siRNA interference of 1014 NF-κB-regulating protein genes. The TNFα-upregulated and -downregulated genes were shown in red and green, respectively. Several known NF-κB target genes such as TNFAIP3 were labeled out. Arrow shows a new NF-κB-regulating gene that was identified as a common target gene of two NF-κB target miRNAs, miR-1276 and miR-219-1-3p.

Fig. 6. Target genes of NF-κB target miRNAs in TNFα-stimulated HeLa cells. A. Comparative analysis of the predicted target genes of NF-κB target miRNAs and NF-κB-regulating genes. B. The target genes of NF-κB target miRNAs. The gene NR4A3 was targeted by two different miRNAs. The arrows represented the genes clustered in disease phenotypes (Table 1). The dot showed NF-κB target genes.

Please cite this article as: F. Zhou, et al., NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbagrm.2014.01.006

528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544

9

P

R O

O

F

F. Zhou et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx

553

T

C

E

R

551 552

R

549 550

N C O

547 548

migration and invasion [36]. Overexpression of miR-548d resulted in pancreatic cell proliferation suppression via increased apoptosis and cell cycle arrest, and also led to a sensitization to gemcitabine [37]. MiR-548d-5p and miR-1286 were deregulated in esophageal cancer cells treated with cisplatin and 5-fluorouracil (5-FU) [38]. NF-κB should be involved in these pathological processes via these target miRNAs. These data provide new clues for studying NF-κB functions and the mechanism underlying the regulation of above described miRNAs in these biological processes.

U

545 546

E

D

Fig. 7. Expressions of the target genes of two miRNAs in miRNA overexpressed cells. A. HeLa cells transfected with mir-125b-1 and mir-1276 plasmids. B. HeLa cells transfected with mir125b-1-3p and mir-1276 mimics. C. HepG2 cells transfected with mir-125b-1 and mir-1276 mimics. Relative quantitation (RQ) was calculated with the comparative Ct method and normalized against GAPDH. *p ≤ 0.05; **p ≤ 0.01. C, control cells; 125, cells transfected with miR-125b-1-3p plasmid or mimic; 1276, cells transfected withmiR-1276 plasmid or mimic.

Fig. 8. Schematic illustration of the central role of mir-125-1 in NF-κB signal pathway. TNFAIP3 is a target gene of NF-κB; its expression can be upregulated by the activation of NF-κB. TNFAIP3 is an important negative regulator of NF-κB activity. Mir-125-1 formed a positive feedback loop for regulating NF-κB activity via its repression of its target gene TNFAIP3. Mir-125-1 also repressed the expression of PPP1R12A.

In this study, we also identified 16 target genes of NF-κB target miRNAs in TNFα-stimulated HeLa cells (Fig. 6). Of these target genes, the target genes of miR-125b-1-3p andmiR-1276 were confirmed in HeLa and HepG2 cells (Fig. 8). These data provide new insights into the functions of these miRNAs. For example, BMP2 and CASP9 were two target genes of miR-1276, and BMP2 was enriched in the related disease phenotype of osteoarthritis [52] (Table 1), and CASP9 was enriched in both the disease phenotype of Type 2 diabetes and lung cancer [53,54] (Table 1). However, to our knowledge, miR-1276 has been little studied up to date. Therefore, these findings suggest that this little known miRNA may function in the pathological process of osteoarthritis, diabetes and lung cancer via its target genes, BMP2 and CASP9. MEKK2 is a Ser/Thr protein kinase belonging to the MEKK/ STE11 subgroup of the MAP/ERK kinase family [55]; its mRNA was identified as a potential target of miR-502-5p in present study. This finding suggests that miR-502-5p may be involved in the MAP/ERK signal pathway. The NR4A3/EWS fusion protein was discovered in approximately 75% of extraskeletal myxoid chondrosarcoma (EMC) tumors, but very little is known about the mechanisms underlying its generation of tumors [56]. We found that miR-1276 and miR-219-1-3p commonly targeted NR4A3; this finding provides a new clue for understanding the tumor progression. MiRNA miR-125b-1 is transcribed from a locus on chromosomes 11q23 and belongs to the miR-125 family that consists of three homologs, miR-125a, miR-125b-1 and miR-125b-2. Members of this family play crucial roles in many different cellular processes such as cell differentiation, proliferation and apoptosis via their target genes [57]. In the present study, we identified PPP1R12A as a target gene of miR-125b-13p. PPP1R12A encodes protein MYPT1, its downexpression rescued mitotic arrest caused by PLK1 depletion in HeLa cells [42] and its overexpression induced apoptosis of NIH3T3 cells [43]. This study revealed that miR-125b-1-3p was an NF-κB-activated miRNA in TNFα-stimulated HeLa and HepG2 cells. In agreement with our results, the NF-κB-dependent expression of miR-125b-1 was also found in other cells, such as biliary epithelial cells (H69) [14,18], primary

Please cite this article as: F. Zhou, et al., NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbagrm.2014.01.006

554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588

623

Appendix A. Supplementary data

624 625

Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.bbagrm.2014.01.006.

626

References

627 628 629 630 631 632 633 634 Q9 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654

[1] R. Sen, D. Baltimore, Multiple nuclear factors interact with the immunoglobulin enhancer sequences, Cell 46 (1986) 705–716. [2] M.S. Hayden, S. Ghosh, Shared principles in NF-kappaB signaling, Cell 132 (2008) 344–362. [3] X. Ma, L.E. Becker Buscaglia, J.R. Barker, Y. Li, MicroRNAs in NF-kappaB signaling, J. Mol. Cell Biol. 3 (2011) 159–166. [4] Y. Xing, F. Zhou, J. Wang, Subset of genes targeted by transcription factor NF-kappaB in TNFalpha-stimulated human HeLa cells, Funct. Integr. Genomics (2012). [5] M.S. Hayden, S. Ghosh, NF-kappaB, the first quarter-century: remarkable progress and outstanding questions, Genes Dev. 26 (2012) 203–234. [6] V. Ambros, The functions of animal microRNAs, Nature 431 (2004) 350–355. [7] Y. Lee, M. Kim, J. Han, K.H. Yeom, S. Lee, S.H. Baek, V.N. Kim, MicroRNA genes are transcribed by RNA polymerase II, EMBO J. 23 (2004) 4051–4060. [8] G.M. Borchert, W. Lanier, B.L. Davidson, RNA polymerase III transcribes human microRNAs, Nat. Struct. Mol. Biol. 13 (2006) 1097–1101. [9] A.M. Denli, B.B. Tops, R.H. Plasterk, R.F. Ketting, G.J. Hannon, Processing of primary microRNAs by the Microprocessor complex, Nature 432 (2004) 231–235. [10] G. Hutvagner, J. McLachlan, A.E. Pasquinelli, E. Balint, T. Tuschl, P.D. Zamore, A cellular function for the RNA-interference enzyme Dicer in the maturation of the let-7 small temporal RNA, Science 293 (2001) 834–838. [11] A. Grishok, A.E. Pasquinelli, D. Conte, N. Li, S. Parrish, I. Ha, D.L. Baillie, A. Fire, G. Ruvkun, C.C. Mello, Genes and mechanisms related to RNA interference regulate expression of the small temporal RNAs that control C. elegans developmental timing, Cell 106 (2001) 23–34. [12] D.S. Schwarz, G. Hutvagner, T. Du, Z. Xu, N. Aronin, P.D. Zamore, Asymmetry in the assembly of the RNAi enzyme complex, Cell 115 (2003) 199–208. [13] H. Guo, N.T. Ingolia, J.S. Weissman, D.P. Bartel, Mammalian microRNAs predominantly act to decrease target mRNA levels, Nature 466 (2010) 835–840.

614 615

620 621

C

612 613

E

610 611

R

608 609

R

606 607

O

604 605

C

602 603

N

600 601

U

598 599

F

622

This work was supported by the National Natural Science Foundation of China (61171030); and Technology Support Program of Jiangsu (BE2012741). We gratefully acknowledge the assistance of Dr. Qingyu Ge in the miRNA expression detections. The authors declare no conflict of interest.

596 597

O

618 619

595

R O

Acknowledgments

593 594

P

617

591 592

[14] R. Zhou, G. Hu, A.Y. Gong, X.M. Chen, Binding of NF-kappaB p65 subunit to the promoter elements is involved in LPS-induced transactivation of miRNA genes in human biliary epithelial cells, Nucleic Acids Res. 38 (2010) 3222–3232. [15] D.P. Bartel, MicroRNAs: genomics, biogenesis, mechanism, and function, Cell 116 (2004) 281–297. [16] E. Tili, J.J. Michaille, A. Cimino, S. Costinean, C.D. Dumitru, B. Adair, M. Fabbri, H. Alder, C.G. Liu, G.A. Calin, C.M. Croce, Modulation of miR-155 and miR-125b levels following lipopolysaccharide/TNF-alpha stimulation and their possible roles in regulating the response to endotoxin shock, J. Immunol. 179 (2007) 5082–5089. [17] K.D. Taganov, M.P. Boldin, K.J. Chang, D. Baltimore, NF-kappaB-dependent induction of microRNA miR-146, an inhibitor targeted to signaling proteins of innate immune responses, Proc. Natl. Acad. Sci. U. S. A. 103 (2006) 12481–12486. [18] R. Zhou, G. Hu, J. Liu, A.Y. Gong, K.M. Drescher, X.M. Chen, NF-kappaB p65-dependent transactivation of miRNA genes following Cryptosporidium parvum infection stimulates epithelial cell immune responses, PloS Pathog. 5 (2009) e1000681. [19] Z. Ning, A.J. Cox, J.C. Mullikin, SSAHA: a fast search method for large DNA databases, Genome Res. 11 (2001) 1725–1729. [20] C.A. Lim, F. Yao, J.J. Wong, J. George, H. Xu, K.P. Chiu, W.K. Sung, L. Lipovich, V.B. Vega, J. Chen, A. Shahab, X.D. Zhao, M. Hibberd, C.L. Wei, B. Lim, H.H. Ng, Y. Ruan, K.C. Chin, Genome-wide mapping of RELA(p65) binding identifies E2F1 as a transcriptional activator recruited by NF-kappaB upon TLR4 activation, Mol. Cell 27 (2007) 622–635. [21] A. Antonaki, C. Demetriades, A. Polyzos, A. Banos, G. Vatsellas, M.D. Lavigne, E. Apostolou, E. Mantouvalou, D. Papadopoulou, G. Mosialos, D. Thanos, Genomic analysis reveals a novel nuclear factor-kappaB (NF-kappaB)-binding site in Alu-repetitive elements, J. Biol. Chem. 286 (2011) 38768–38782. [22] R. Martone, G. Euskirchen, P. Bertone, S. Hartman, T.E. Royce, N.M. Luscombe, J.L. Rinn, F.K. Nelson, P. Miller, M. Gerstein, S. Weissman, M. Snyder, Distribution of NF-kappaB-binding sites across human chromosome 22, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 12247–12252. [23] V. Matys, E. Fricke, R. Geffers, E. Gossling, M. Haubrock, R. Hehl, K. Hornischer, D. Karas, A.E. Kel, O.V. Kel-Margoulis, D.U. Kloos, S. Land, B. Lewicki-Potapov, H. Michael, R. Munch, I. Reuter, S. Rotert, H. Saxel, M. Scheer, S. Thiele, E. Wingender, TRANSFAC: transcriptional regulation, from patterns to profiles, Nucleic Acids Res. 31 (2003) 374–378. [24] A. Nijnik, R. Mott, D.P. Kwiatkowski, I.A. Udalova, Comparing the fine specificity of DNA binding by NF-kappaB p50 and p52 using principal coordinates analysis, Nucleic Acids Res. 31 (2003) 1497–1501. [25] C.A. Lim, F. Yao, J.J. Wong, J. George, H. Xu, K.P. Chiu, W.K. Sung, L. Lipovich, V.B. Vega, J. Chen, A. Shahab, X.D. Zhao, M. Hibberd, C.L. Wei, B. Lim, H.H. Ng, Y. Ruan, K.C. Chin, Genome-wide mapping of RELA(p65) binding identifies E2F1 as a transcriptional activator recruited by NF-kappaB upon TLR4 activation, Mol. Cell 27 (2007) 622–635. [26] D. Wong, A. Teixeira, S. Oikonomopoulos, P. Humburg, I.N. Lone, D. Saliba, T. Siggers, M. Bulyk, D. Angelov, S. Dimitrov, I.A. Udalova, J. Ragoussis, Extensive characterization of NF-kappaB binding uncovers non-canonical motifs and advances the interpretation of genetic functional traits, Genome Biol. 12 (2011) R70. [27] C. Settembre, R. De Cegli, G. Mansueto, P.K. Saha, F. Vetrini, O. Visvikis, T. Huynh, A. Carissimo, D. Palmer, T.J. Klisch, A.C. Wollenberg, D. Di Bernardo, L. Chan, J.E. Irazoqui, A. Ballabio, TFEB controls cellular lipid metabolism through a starvationinduced autoregulatory loop, Nat. Cell Biol. 15 (2013) 647–658. [28] A. Mukhopadhyay, B. Deplancke, A.J. Walhout, H.A. Tissenbaum, Chromatin immunoprecipitation (ChIP) coupled to detection by quantitative real-time PCR to study transcription factor binding to DNA in Caenorhabditis elegans, Nat. Protoc. 3 (2008) 698–709. [29] J. Nishida, Y. Li, Y. Zhuang, Z. Huang, H. Huang, IFN-gamma suppresses permissive chromatin remodeling in the regulatory region of the Il4 gene, Cytokine 62 (2013) 91–95. [30] D. Betel, M. Wilson, A. Gabow, D.S. Marks, C. Sander, The microRNA.org resource: targets and expression, Nucleic Acids Res. 36 (2008) D149–D153. [31] H. Dweep, C. Sticht, P. Pandey, N. Gretz, miRWalk-database: prediction of possible miRNA binding sites by “walking” the genes of three genomes, J. Biomed. Inform. 44 (2011) 839–847. [32] M. Reczko, M. Maragkakis, P. Alexiou, I. Grosse, A.G. Hatzigeorgiou, Functional microRNA targets in protein coding sequences, Bioinformatics 28 (2012) 771–776. [33] B.M. Bolstad, R.A. Irizarry, M. Astrand, T.P. Speed, A comparison of normalization methods for high density oligonucleotide array data based on variance and bias, Bioinformatics 19 (2003) 185–193. [34] M.K. Kerr, M. Martin, G.A. Churchill, Analysis of variance for gene expression microarray data, J. Comput. Biol.s 7 (2000) 819–837. [35] Y. Sun, X. Zhao, Y. Zhou, Y. Hu, miR-124, miR-137 and miR-340 regulate colorectal cancer growth via inhibition of the Warburg effect, Oncol. Rep. 28 (2012) 1346–1352. [36] Z.S. Wu, Q. Wu, C.Q. Wang, X.N. Wang, J. Huang, J.J. Zhao, S.S. Mao, G.H. Zhang, X.C. Xu, N. Zhang, miR-340 inhibition of breast cancer cell migration and invasion through targeting of oncoprotein c-Met, Cancer 117 (2011) 2842–2852. [37] H. Heyn, S. Schreek, R. Buurman, T. Focken, B. Schlegelberger, C. Beger, MicroRNA miR-548d is a superior regulator in pancreatic cancer, Pancreas 41 (2012) 218–221. [38] R. Hummel, T. Wang, D.I. Watson, M.Z. Michael, M. Van der Hoek, J. Haier, D.J. Hussey, Chemotherapy-induced modification of microRNA expression in esophageal cancer, Oncol. Rep. 26 (2011) 1011–1017. [39] Y. Zhang, L.X. Yan, Q.N. Wu, Z.M. Du, J. Chen, D.Z. Liao, M.Y. Huang, J.H. Hou, Q.L. Wu, M.S. Zeng, W.L. Huang, Y.X. Zeng, J.Y. Shao, miR-125b is methylated and functions as a tumor suppressor by regulating the ETS1 proto-oncogene in human invasive breast cancer, Cancer Res. 71 (2011) 3552–3562.

T

616

human neuronal-glial (HNG) cells [44], human keratinocyte (HacaT) cells [45], and human embryonic kidney (HEK293) cells [45]. In a recent our study, we found that NF-κB target protein genes identified in different cell lines in response to various stimuli was significantly different [4]. Therefore, the NF-κB-dependent expressions of miR-125b-1 in various cells suggest that this miRNA may play a central role in the NF-κB signal pathway, and modulate cell survival through its target genes such as PPP1R12A (Fig. 8). In conclusion, we identified 14 NF-κB target miRNAs in TNFαstimulated HeLa cells and 16 target genes of 6 miRNAs, including miR125b-1-3p, miR-1286, miR-502-5p, miR-1276, miR-219-1-3p and miR-30b-3p. Two potential target genes of miR-1276, BMP2 and CASP9, were overrepresented in the GO terms of osteoarthritis, and Type 2 diabetes and lung cancer, respectively, suggesting that this little known miRNA may play a role in these diseases. One potential target gene of miR125b-1-3p, PPP1R12A, has the function of regulating cell survival, suggesting that this miRNA plays a role in the regulation of cell survival. In addition, the expression of a known NF-κB target miRNA, miR125b-1, was found significantly regulated by NF-κB in TNFα-stimulated HeLa and HepG cells, in combination with its NF-κB-dependent expression fashion reported in other cell types under different stimuli, we firstly proposed this miRNA as a central mediator in NF-κB signal pathway. This miRNA formed a positive feedback loop of NF-κB activity regulation via its target gene TNFAIP3 and modulate cells survivals via another target gene PPP1R12A. These findings provide new insights into the functions of NF-κB in its target miRNA-related biological processes and the mechanisms underlying the regulation of these miRNAs.

D

589 590

F. Zhou et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx

E

10

Please cite this article as: F. Zhou, et al., NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbagrm.2014.01.006

655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740

F. Zhou et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx

[49] L. Vereecke, R. Beyaert, G. van Loo, The ubiquitin-editing enzyme A20 (TNFAIP3) is a central regulator of immunopathology, Trends Immunol. 30 (2009) 383–391. [50] K. Honma, S. Tsuzuki, M. Nakagawa, S. Karnan, Y. Aizawa, W.S. Kim, Y.D. Kim, Y.H. Ko, M. Seto, TNFAIP3 is the target gene of chromosome band 6q23.3-q24.1 loss in ocular adnexal marginal zone B cell lymphoma, Genes Chromosom. Cancer 47 (2008) 1–7. [51] J.A. DiDonato, F. Mercurio, M. Karin, NF-kappaB and the link between inflammation and cancer, Immunol. Rev. 246 (2012) 379–400. [52] I. Papathanasiou, K.N. Malizos, A. Tsezou, Bone morphogenetic protein-2-induced Wnt/beta-catenin signaling pathway activation through enhanced low-densitylipoprotein receptor-related protein 5 catabolic activity contributes to hypertrophy in osteoarthritic chondrocytes, Arthritis Res. Ther. 14 (2012) R82. [53] E.J. Anderson, E. Rodriguez, C.A. Anderson, K. Thayne, W.R. Chitwood, A.P. Kypson, Increased propensity for cell death in diabetic human heart is mediated by mitochondrial-dependent pathways, Am. J. Physiol. Heart Circ. Physiol. 300 (2011) H118–H124. [54] J. Feng, Y. Liu, N. Dobrilovic, L.M. Chu, C. Bianchi, A.K. Singh, F.W. Sellke, Altered apoptosis-related signaling after cardioplegic arrest in patients with uncontrolled type 2 diabetes mellitus, Circulation 128 (2013) S144–S151. [55] J.L. Blank, P. Gerwins, E.M. Elliott, S. Sather, G.L. Johnson, Molecular cloning of mitogen-activated protein/ERK kinase kinases (MEKK) 2 and 3. Regulation of sequential phosphorylation pathways involving mitogen-activated protein kinase and c-Jun kinase, J. Biol. Chem. 271 (1996) 5361–5368. [56] C. Filion, Y. Labelle, Identification of genes regulated by the EWS/NR4A3 fusion protein in extraskeletal myxoid chondrosarcoma, Tumour Biol. 33 (2012) 1599–1605. [57] Y.M. Sun, K.Y. Lin, Y.Q. Chen, Diverse functions of miR-125 family in different cell contexts, J. Hematol. Oncol. 6 (2013) 6.

O

F

[40] H. Nakanishi, C. Taccioli, J. Palatini, C. Fernandez-Cymering, R. Cui, T. Kim, S. Volinia, C.M. Croce, Loss of miR-125b-1 contributes to head and neck cancer development by dysregulating TACSTD2 and MAPK pathway, Oncogene (2013). [41] Y. Enomoto, J. Kitaura, M. Shimanuki, N. Kato, K. Nishimura, M. Takahashi, H. Nakakuma, T. Kitamura, T. Sonoki, MicroRNA-125b-1 accelerates a C-terminal mutant of C/EBPalpha (C/EBPalpha-C(m))-induced myeloid leukemia, Int. J. Hematol. 96 (2012) 334–341. [42] S. Yamashiro, Y. Yamakita, G. Totsukawa, H. Goto, K. Kaibuchi, M. Ito, D.J. Hartshorne, F. Matsumura, Myosin phosphatase-targeting subunit 1 regulates mitosis by antagonizing polo-like kinase 1, Dev. Cell 14 (2008) 787–797. [43] Y. Wu, A. Muranyi, F. Erdodi, D.J. Hartshorne, Localization of myosin phosphatase target subunit and its mutants, J. Muscle Res. Cell Motil. 26 (2005) 123–134. [44] W.J. Lukiw, NF-small ka, CyrillicB-regulated micro RNAs (miRNAs) in primary human brain cells, Exp. Neurol. 235 (2012) 484–490. [45] G. Tan, J. Niu, Y. Shi, H. Ouyang, Z.H. Wu, NF-kappaB-dependent microRNA-125b up-regulation promotes cell survival by targeting p38alpha upon ultraviolet radiation, Int. J. Biol. Chem. 287 (2012) 33036–33047. [46] M. Zhou, Z. Liu, Y. Zhao, Y. Ding, H. Liu, Y. Xi, W. Xiong, G. Li, J. Lu, O. Fodstad, A.I. Riker, M. Tan, MicroRNA-125b confers the resistance of breast cancer cells to paclitaxel through suppression of pro-apoptotic Bcl-2 antagonist killer 1 (Bak1) expression, Int. J. Biol. Chem. 285 (2010) 21496–21507. [47] S.W. Kim, K. Ramasamy, H. Bouamar, A.P. Lin, D. Jiang, R.C. Aguiar, MicroRNAs miR-125a and miR-125b constitutively activate the NF-kappaB pathway by targeting the tumor necrosis factor alpha-induced protein 3 (TNFAIP3, A20), Proc. Natl. Acad. Sci. U. S. A. 109 (2012) 7865–7870. [48] A. Krikos, C.D. Laherty, V.M. Dixit, Transcriptional activation of the tumor necrosis factor alpha-inducible zinc finger protein, A20, is mediated by kappa B elements, J. Biol. Chem. 267 (1992) 17971–17976.

R O

741 742 743 Q10 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768

11

U

N C O

R

R

E

C

T

E

D

P

798

Please cite this article as: F. Zhou, et al., NF-κB target microRNAs and their target genes in TNFα-stimulated HeLa Cells, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbagrm.2014.01.006

769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797