- Email: [email protected]
The tumor suppressor p33ING1b upregulates p16INK4a expression and induces cellular senescence
FEBS Letters 585 (2011) 3106–3112
journal homepage: www.FEBSLetters.org
The tumor suppressor p33ING1b upregulates p16INK4a expression and induces cellular senescence Na Li, Qian Li, Xiaoxiao Cao, Ganye Zhao, Lixiang Xue ⇑, Tanjun Tong ⇑ Research Center on Aging, Department of Biochemistry and Molecular Biology, Peking University Health Science Center, Beijing 100191, China
a r t i c l e
i n f o
Article history: Received 16 May 2011 Revised 29 August 2011 Accepted 29 August 2011 Available online 2 September 2011 Edited by Angel Nebreda Keywords: p33ING1b p16INK4a HAT Cellular senescence
a b s t r a c t ING1 protein is a tumor suppressor which plays significant roles in multiple cellular activities. p47ING1a and p33ING1b are major splice isoforms of ING1 and their roles in senescence need further investigations. Here we studied the functions of ING1 isoforms in cellular senescence and gene regulation, with focus on p16INK4a. We observe that p33ING1b protein is the major ING1 isoform expressed in 2BS human diploid fibroblasts. Overexpression of p33ING1b induces cellular senescence and upregulates p16INK4a expression in 2BS fibroblasts. p33ING1b upregulates p16INK4a transcription. p33ING1b and p300 bind to the p16INK4a promoter. p300/CBP-specific inhibitor curcumin can reverse the induction of p16INK4a by p33ING1b. These results help to better understand the function of ING1. Ó 2011 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
1. Introduction ING1 (inhibitor of growth1) protein functions as a type II tumor suppressor which plays a significant role in multiple cellular activities such as growth regulation, DNA repair, apoptosis and senescence [1]. ING1 gene was cloned in 1996 [2], and several years later, its isoforms p47ING1a, p33ING1b and p24ING1c were characterized [3]. Current evidence suggests many functions of ING1 rely on p53 [4,5]. However, some other results obtained from different cell lines implicated that ING1 could function in a p53-independent manner [6,7]. Given the facts that ING1 proteins are involved in chromatin remodeling through association with HAT (histone acetyl-transferase) or HDAC (histone deacetylase) complexes, it was suggested that ING1 proteins might regulate target gene expression through localized chromatin modification [8–10]. The first analysis of ING1 in cell senescence identified ING1 protein had pro-senescence effect, but at that time it was unable to tell which isoform of ING1 exert such effect [11]. Later, it was reported p33ING1b overexpression induced features of cellular senescence in human diploid fibroblasts cells [12,13]. But another study showed Abbreviations: ING, inhibitor of growth; HAT, histone acetyl-transferase; HDAC, histone deacetylase; SAHF, senescence-associated heterochromatic foci; PD, population doubling; qPCR, quantitative polymerase chain reaction ⇑ Corresponding authors at: Peking University Research Center on Aging, Department of Biochemistry and Molecular Biology, No.38, Xueyuan Road, Haidian District, Beijing 100191, China. Fax: +86 10 82802931. E-mail address: [email protected] (T. Tong).
that p47ING1a but not p33ING1b had the potential to promote senescent cell morphology [14]. Here we tried to dissect the expression level and function of ING1 isoforms in replicative cellular senescence. And it is well known that p16INK4a plays important role in replicative senescence. So we investigated whether ING1 isoforms could influence cellular senescence with focus on p16INK4a expression. And we attempted to dissect the molecular mechanism. Our results indicate that p33ING1b is the major isoform of ING1 expressed in 2BS fibroblasts. p33ING1b promotes cellular senescence by upregulation of p16INK4a expression. And p33ING1b interacts in vivo with the p16INK4a promoter and requires HAT activity to increase p16INK4a expression. These observations help to further understand the biological functions of ING1 isoforms in cellular senescence. 2. Materials and methods 2.1. Cell culture and reagent Human embryonic lung fibroblast cell line (2BS cells) was obtained from the National Institute of Biological Products (Beijing, China) and cultured as described previously [15]. 2BS cells are considered to be young at PD30 (population doubling30) or below and to be fully senescent at PD55 or above. Senescent cells are characterized by an irreversible growth arrest. Human embryonic kidney cell line (HEK293 cells) was cultured in DMEM medium, containing 10% fetal bovine serum. Curcumin,
0014-5793/$36.00 Ó 2011 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2011.08.044
N. Li et al. / FEBS Letters 585 (2011) 3106–3112
3107
with purity P95%, was purchased from Sigma (St. Louis, MO, USA) and used at indicated concentration.
(Leica). Representative photographs from three independent experiments are shown.
2.2. Plasmids construction
2.8. Luciferase activity assay
Expression plasmids of p47ING1a and p33ING1b cDNA in the pCI vector were gifts of Professor Karl Riabowol. Plasmids were digested by EcoRI and XbaI, and cloned into pcDNA3.1 (Invitrogen) or pLVX-IRES-ZsGreen1 (Clontech) linearized with the same enzymes. 50 -fragment of human p16INK4a promoter was digested with XhoI and HindIII or exemplified by PCR from pSIR-EGFP vector [16], and inserted into pGL3-Basic Luciferase Report Vector linearized with the same enzymes (Promega).The p16 shRNA sequence was as previously described [17]. The template oligonucleotides were chemical synthesized and inserted into the pMSCV-puromiR30 vector.
Twenty-four or forty-eight hours after transfection, the activities of luciferase were quantified in a luminometer (Centro LB 960; Berthold Technologies) using the Dual Luciferase Reporter Assay System (Promega). All experiments were performed in triplicate.
2.3. Transfection All plasmids were purified with Qiagen Plasmid Midi Kits. Cells were transfected with plasmids coated by lipofectamine 2000 (Invitrogen) following the manufacturer’s indications. To over-express p33ING1b or knock down p16INK4a in 2BS cells, recombinant viruses were produced according to manufacturer’s indications (Clontech). 2.4. Preparation of total RNA and qPCR RNA preparation and qPCR (quantitative polymerase chain reaction) were performed according to manufacturer’s instructions (invitrogen). The primer sequences were as follows: p47ING1a 50 -CGGGCCTAGCGCAATAAC-30 and 50 -CCCTCGGGTCA GCAAAGAT-30 ; p33ING1b 50 -TCCCTGCCTTTCGACTTG-30 and 50 -GCTTGCTGTTG GGCTTGT-30 ; p16INK4a 50 -CAACGCACCGAATAGTTACGG-30 and 50 -GCGCAGTT GGGCTCCG-30 ; p53 50 -AATGCCAGAGGCTGCTCCC-30 and 50 -GGTGTGGAATCA ACCCAC-30 ; b-actin 50 -AGCGAGCATCCCCCAAAGTT-30 and 50 -GGGCACGAAGGCT CATCATT-30 . 2.5. Western blot Western bolts were performed as described [15]. The antibodies used are anti-p16INK4a (Clone 16P04; NeoMarkers), anti-p53 (sc-126; Santa Cruze), anti-ING1 (Santa Cruze; Millipore or Abcam, the antibody from Abcam only recognizes p33 ING1b ), anti-H3K9me3 (ab8898; Abcam), anti-p300 (05–257; Millipore), anti-b-actin (Santa Cruze) and anti-tubulin (Millipore). 2.6. SA-b-gal assays The 2BS cells were washed twice in PBS, fixed in 3% formaldehyde, and washed again in PBS. The cells were incubated overnight at 37 °C (without CO2) with freshly prepared SA-b-gal staining solution as described [18]. 2.7. SAHF and Immunofluorescence SAHF (senescence-associated heterochromatic foci) (DAPI foci) were detected by staining with DAPI at room temperature. Image collection was by a confocal microscope [18]. Immunofluorescence were performed as previous described. Stained cells were examined with a confocal laser microscope
2.9. Chromatin immunoprecipitation Chromatin immunoprecipitations (ChIPs) were performed according to manufacturer’s instructions (Upstate). Re-ChIP was performed as previously described [18].The sequences of the primers for ChIP assays were as follows: Primer-1 1.7 kb upstream 50 -CACAGGAAAAATTAGAAC-30 50 -CCGAGATCGCGCCAT-30 ; Primer-2 1.0 kb upstream 50 -TTCCCAAAGTGCTGGGAT-30 50 -GGGGTTGTTGTGAGTTTA-30 ; Primer-3 200 bp upstream 50 -CCCCG ATTCA ATTTG GCAGT-30 50 -CAGCG TTGGC AAGGA AGGAG-30 ; Primer-4 End of INK4A exon 1a 50 -AGAGGGTCTGCAGCGG-30 50 -TCGAAGCGCTACCTGATTCC-30 ; GAPDH promoter 50 -TACTAGCGGTTTTACGGGCG-30 50 -TCGAACAGGAGGAGCAGAGAGCGA-30 . 2.10. Statistics All values are expressed as means ± S.D. in the figures. Student0 s t-test was used to perform statistical analyses. Differences with ⁄ P < 0.05 were considered significant.
3. Results 3.1. Expression levels of p47ING1a and p33ING1b in replicative senescence process of 2BS cells Reportedly, p47ING1a and p33ING1 are two major isoforms of ING1 in human cells. They have conserved C-terminus, while have distinct N-terminus with different affinities for HATs and HDACs (Fig. 1A). We tested the total protein and mRNA levels of ING1 isoforms in 2BS cells. p16INK4a expression was also detected at the same time to indicate the senescent state of 2BS cells. The results showed that the RNA levels of p47ING1a were higher in middle-aged and senescent 2BS cells than young 2BS cells; and p33ING1b mRNA expression was upregulated in middle-aged 2BS cells and decreased in young and senescent cells (Fig. 1B). The total protein levels of p47ING1a were too low to detect in 2BS cells; while p33ING1b protein levels were consistent with its mRNA levels. As ING1 protein participates in gene transcriptional control, we detected ING1 expression in chromatin fractions of 2BS fibroblasts. The results showed that p33ING1b expression increased in chromatin fractions of middle-aged and senescent 2BS cells. Notably, protein and mRNA levels of p16INK4a increased in the middle-aged and senescent 2BS cells in which chromatin fraction levels of p33ING1b increased (Fig. 1C). 3.2. p33ING1b induces senescent markers and p16INK4a expression in 2BS cells To test whether p33ING1b plays roles in senescence progress of 2BS cells, we detected the cellular senescence markers in p33ING1b
3108
N. Li et al. / FEBS Letters 585 (2011) 3106–3112
A
188
1
p47ING1a
422 NCR
NLS
PHD
PBR
HDACs
1
p33ING1b
PB
HDACs
NCR
HATs
NLS
257 PHD
279 PBR
HDACs
C
B
212
126
45 PIP
Y
M
S
Relative mRNA level (fold)
p47ING1a 20 18 16 14 12 10 8 6 4 2 0
Y M S
p33ING1b p16 β-actin
Y p47
p33
p16
M
S p33ING1b Histones
Fig. 1. The expression levels of p47ING1a and p33ING1b in young, middle-aged and senescent 2BS cells. (A) Schematic representation of INGI isoforms p47ING1a and p33ING1b. PIP = PCNA-interacting protein motif; PB = partial bromodomain; NCR = novel conserved region; NLS = nuclear localization sequence; PHD = plant homeodomain; PBR, polybasic region. (B) The mRNA expression of ING1 transcripts by qPCR analysis. (C) Total protein levels (upper panel) and chromatin levels (lower panel) of ING1 isoforms in young (Y; PD17), middle-aged (M; PD43) and senescent (S; PD65) 2BS cells. The histones were detected by coomassie blue staining. A representative result of three separate experiments is shown.
-transfected 2BS cells. The results showed that 2BS cells overexpressing of p33ING1b efficiently induced the expression of SA-bgal activity, flattened senescent cell morphology, and SAHF DNA staining pattern which is indicated by co-localization with H3K9Me3 (Fig. 2A and B). To gain insights into the possible relationship of ING1 isoforms with senescence-related genes, we investigated the effects of ING1 isoforms on the expressions of p16INK4a and p53 in 2BS cells. qPCR and Western blot results showed that introduction of p33ING1b upregulated p16INK4a mRNA and protein levels (Fig. 2C and D). Moreover, p16INK4a promoter activity was upregulated by p33ING1b (Fig. 2E). Reportedly, ING1 proteins regulate p53 protein by posttranscriptional mechanisms [4,5]. Consistent with that, our results showed that p53 protein levels were slightly upregulated by p33ING1b, while the mRNA level of p53 did not increase (Fig. 2C and D). These results suggested that p33ING1b may play roles in cellular senescence partly through upregulation of p16 INK4a transcription. To further verify that p33ING1b could regulate senescence via p16INK4a, we evaluated the role of p16 in mediating the senescence effects by p33ING1b in 2BS cells. 2BS cells infected with both p33ING1b and p16-shRNA or with p33ING1b or p16-shRNA alone were monitored. The results showed that cells treated with p33ING1b alone revealed strong SA-b-gal staining and p16 increasing, whereas cells doubly treated with p33ING1b and p16-shRNA reversed the senescence phenotypes. As expected, p16-shRNA-infected cells displayed young phenotypes (Fig. 2F). To confirm the effects of p33ING1b on p16INK4a expression, we overexpressed p33ING1b in HEK293 cells, and analyzed the protein expressions, mRNA levels and promoter activities of p16INK4a. p53 levels were also detected as controls. The results showed that p33ING1b markedly increased p16INK4a protein level, mRNA level
and promoter activity in HEK293 cells (Fig. 2G–I). p53 mRNA expression did not increase, which is in agreement with 2BS fibroblasts. These results showed a well correlation of promoter activity, mRNA levels and protein levels of p16INK4a upon ectopic expression of p33ING1b both in 2BS fibroblasts and HEK293 cells; suggesting p16INK4a expression was regulated by p33ING1b at the transcriptional level. 3.3. p33ING1b activates and binds to p16INK4a promoter region in vivo To investigate the mechanism of regulation of p16INK4a transcription by p33ING1b, we examined the effects of p33ING1b on deletion constructs of p16INK4a transcription regulatory regions in HEK293 cells. We used the constructs from 178 bp downstream (+178 bp) to 1765 bp upstream (1765 bp) of the transcriptional start site of p16INK4a. The luciferase activity assays showed that p33ING1b activated p16INK4a promoter regions, and the main regions contributing to over-expression of p16INK4a by p33ING1b were located in the region from +178 to 140 bp of p16INK4a transcriptional start site (Fig. 3A). Then we examined the in vivo DNA-binding activity of p33ING1b to p16INK4a promoter. The antibody used for ChIP assay only recognized p33ING1b isoform. In HEK293 cells, both RT–PCR and qPCR for ChIP analysis showed that p33ING1b specifically bound 200 bp upstream of the p16INK4a transcription start site (primer set 3). This region is also the p16INK4a promoter region significantly activated by p33ING1b (Fig. 3A and B). We also found that ectopic expressed p33ING1b bound the p16INK4a promoter in HEK293 cells, which is the same region of p16 promoter as the endogenous p33ING1b bound (Fig. 3C, upper panel). In 2BS cells, endogenous p33ING1b specifically bound this region of the p16INK4a promoter (primer set 3); moreover,
3109
N. Li et al. / FEBS Letters 585 (2011) 3106–3112
A
B
p33
Ctrl
p33
Ctrl
H3K9Me3
DAPI
D Ctrl
Relative mRNA level (fold)
C p33 p16 p53 p33 actin
ctrl p33
5 4
*
3 2 1
E
7
Luciferase activity (fold)
p33
6
ctrl p33
*
5 4 3 2 1
0
0 p53
control
p16
p1 3+
Ctrl
p3
3
6-s
p3
p1
Ctr
l
hR
6-s
NA
F
hR
NA
p16
p16-shRNA
p33
p33+ p16-shRNA
p16 p33
H Ctrl
p33 p16 p53 p33 actin
6 5
I
ctrl p33 *
4 3 2 1 0
Luciferase activity (fold)
G
Relative mRNA level (fold)
actin
6 5
ctrl p33
*
4 3 2 1 0
p16
p53
control
p16
Fig. 2. Overexpression of p33ING1b induce cellular senescence and p16INK4a level. (A) SA-b-gal activity (blue) and (B) SAHF formation (left panel) in vector or p33ING1b-infected stable cells. Immunostaining of H3K9me3 in p33ING1b-infected cells (right panel). White arrows indicated co-localization of H3K9me3 at SAHF. (C) Western blots and (D) qPCR analysis of p16INK4a and p53 expressions in vector and p33ING1b-infected stable 2BS cells. (E) Stable infected 2BS cells were transfected with control promoter or p16 promoter. Luciferase activities were measured 24 h later. Values have been equalized relative to the value obtained with control empty vector (set at 1) and are the average of three independent determinations. ⁄P < 0.05. (F) Western blots and SA-b-gal analysis of the effects of knocking down p16INK4a on p33ING1b-induced senescence in 2BS cells. (G) Western blots and (H) qPCR analysis of p16INK4a and p53 expressions in vector or p33ING1b-transfected HEK293 cells. (I) Vector or p33ING1b-transfected HEK293 cells were transfected with control promoter or p16 promoter. Luciferase activities were measured as described. ⁄P < 0.05.
the binding of p33ING1b increased in middle-aged and senescent 2BS cells (Fig. 3C, lower panel). Therefore these results indicate that p33ING1b activates and specifically binds to the p16INK4a promoter region and imply again that p33ING1b plays a role in p16INK4a transcriptional regulation. 3.4. p33ING1b and p300 interact with the p16INK4a promoter to regulate its transcription Reportedly, p33ING1b interacts with proteins associated with HAT activity, such as p300/CBP [8]. To examine whether the effect of p33ING1b on p16INK4a transcription was dependent on p300/CBP HAT activity, Re-ChIP was performed using an anti-p33ING1b antibody in HEK293 cells, followed by reimmunoprecipitation using
an anti-p300 antibody, and then detection of PCR-amplified signal in the p16INK4a promoter region. The results showed that both p33ING1b and p300 associate with the regulatory region of the p16INK4a promoter (Fig. 4A). Curcumin has been reported to be a p300/CBP-specific cell permeable HAT inhibitor [19,20]. Here curcumin was used to confirm the effects of p300/CBP HAT on p33ING1b-induced p16INK4a expression. After introduction of control vector or p33ING1b, 2BS and HEK293 cells were incubated with 1–20 lM curcumin. The results showed that p16INK4a protein expressions were increased in p33ING1b-overexpressed 2BS and HEK293 cells without curcumin treatment. While following the addition of curcumin, p33ING1b -induced increase of p16INK4a protein expressions reduced more than vector control cells. Treatment of cells with 20 lM curcumin
3110
N. Li et al. / FEBS Letters 585 (2011) 3106–3112
A
7
p33
6
Luciferase activity (fold)
B
control
p16
*
-1700
5
-200
-1000
+461
* *
4
* 3
2
1
2
3
4
1
+178
-140
-315
-735
IgG
-1765bp
1
C
p33
Input
0
2
3
4
1
2
4 1
3
2
3
4 p16
Input
HEK293
IgG p33
GAPDH
p16 GAPDH
Input Y
M
p33 S
Y
M
2BS
IgG S
Y
M
Bound/Input (fold)
5
S p16 GAPDH
4
IgG p33
3 2 1 0 1
2
3
4
Fig. 3. p33ING1 activates and specifically binds to p16INK4a promoter region. (A) Vector and p33ING1b-transfected HEK293 cells were transfected with constructs of p16 regulatory regions from 178 bp downstream to 1765 bp upstream p16 transcription start site. Luciferase activities were measured as described. ⁄P < 0.05. (B) Schematic diagram of the p16 gene promoter and the end of INK4a exon 1a (upper panel). ChIP analysis was performed using primers indicated by red bars in HEK293 cells. The precipitated DNA was amplified by RT–PCR (middle panel) and qPCR (lower panel). Values are expressed relative to the controls, which were set as 1. (C) ChIP assays of p33ING1b transfected-HEK293 cells (upper panel); and normal young (Y), middle-aged (M) and senescent (S) 2BS cells at p16 promoter using the primer set 3. The antibody used for ChIP recognized p33ING1b but not p47ING1a.
reversed completely the effects of p33ING1b on p16INK4a protein levels both in 2BS and HEK293 cells (Fig. 4B and C). In addition, the luciferase assays for p16INK4a promoter activities showed a similar tendency (Fig. 4D). Taken together, these results indicated that p33ING1b bound to p16INK4a promoter region in vivo, and used at least partly HAT-dependent mechanism for transcriptional induction of p16INK4a. 4. Discussion Although recent studies began to attract more attention to ING1 isoforms, their biological functions have not been fully elucidated up to date [1,21]. In the present study, we have investigated the expressions of the two major ING1 isoforms in young, middle-aged and senescent 2BS fibroblasts and detected their functions on gene expression and cell senescence. We observed the mRNA levels of p47INGla increased in middle-aged and senescent human diploid 2BS fibroblasts, however, its protein levels are very low in 2BS cells; suggesting some posttranscriptional mechanisms may regulate p47INGla protein levels in 2BS cells. p33ING1b is the major ING1 isoform expressed in 2BS fibroblasts, its mRNA and total protein expression elevated in middle-aged 2BS cells when p16INK4a levels began increasing, but its expression declined when cells entered senescent state. Reportedly, ING1 proteins regulate gene
expression through regulating chromatin organization [1,8,9,13,14]. By extracting the chromatin proteins, we found that the expression of p33ING1b increases in chromatin fractions of both middle-aged and senescent 2BS cells, with a more markedly increase in middle-aged cells (Fig. 1). A recent study also published chromatin expression increase of p33ING1b in Ras-induced senescence in IMR-90 fibroblasts [13]. Furthermore, we found that the binding between p33ING1b and p16INK4a promoter in middle-aged and senescent 2BS cells also increased (Fig. 3C). These results indicate that p33ING1b might play a role in increasing p16INK4a gene transcription in middle-aged and senescent 2BS fibroblasts. p16INK4a is an important gene in provoking and maintaining cellular senescence, there are several proteins regulate its transcription in senescent cells, such as Ets1, Id1, PPARc, B-MYB and Lsh [18,22– 24]. It is clear that the p16INK4a promoter is subject to multiple levels of control, and therefore p16INK4a regulation cannot be explained by a single isolated pathway. Here our observations suggest that p33ING1b might be one of the important factors in regulating p16INK4a expression in cellular senescence. Although p47ING1a was reported to induce senescent phenotype, it is controversial regarding p33ING1b in cellular senescence [12–14]. As p33ING1b protein is the major isoform of ING1 expressed in 2BS cells, we examined its role in cellular senescence. The results showed that overexpression of p33ING1b induced some
3111
N. Li et al. / FEBS Letters 585 (2011) 3106–3112
A
1st IP Input
IgG p33 p33 p300 IgG p300
2nd IP
p16 GAPDH
B
2BS Ctrl
C HEK293
1
0
Ctrl
5
1
10
5 10
20 Curcumin
p33
0
1
5 10
20 Curcumin
p16
p16
p33
p33
tubulin
tubulin
20 Curcumin
p33
0
1
5
10
20 Curcumin
p16
p16
p33
p33
tubulin
tubulin
2.5 Luciferase activity (fold)
D
0
*
2.0 *#
*
*# *#
1.5 *#
1.0 0.5 0.0
Ctrl p33 p33 p33 p33 p33 p33 Curcumin (μM) ING1b
0
0
1
INK4a
2.5
5
10
20
INK4a
Fig. 4. p33 and p300 bind to the promoter of p16 to induce p16 transcription. (A) Re-ChIP analysis showing the colocalization of p33ING1b with p300 at the p16 promoter using the primer set 3 in HEK293 cells. (B) 2BS cells and (C) HEK293 cells were infected or transfected with vector or p33ING1b as indicated. Cells were then treated with DMSO vehicle or various concentration of curcumin (1, 5, 10, 20 lM). Cells were collected for total protein 12 h after treatment. Western blots were used to detect indicated proteins. (D) HEK293 cells were transfected with control vector or p33ING1b as indicated, together with pGL3-p16 promoter. 24 h after transfection, cells were treated with DMSO vehicle or various concentration of curcumin (1, 2.5, 5, 10, 20 lM). 12 h after treatment, luciferase activities were measured.⁄P < 0.05 compared with vector control; #P < 0.05 compared with p33ING1b.
aspects of a senescent phenotype in 2BS fibroblasts, suggesting the important roles of p33ING1b in implementation of the replicative senescent state. It appears that senescence is not triggered by a single, linear series of events, but instead is regulated by a complex signaling network. p53 and p16INK4a play important roles in both tumor suppression and cellular senescence. Reportedly, ING1 proteins are functionally linked to the p53 pathway via mechanisms including control of p53 protein stability or posttranslational modifications [4,5]. Consistent with that, we observed that p53 mRNA levels did not increase upon p33ING1b overexpression. However, we found that p33ING1b increases p16INK4a gene mRNA, promoter activity and protein expression in both 2BS and HEK293 cells. And knocking down p16INK4a expression rescued the senescent effect by p33ING1b (Fig. 2). These results suggest that p33ING1b might increase p16INK4a expression at transcriptional level in cellular senescence, and p16INK4a is required for p33ING1b-induced premature senescence. Recent studies have shown that ING1 proteins may regulate gene transcription by epigenetic control [13,14]. p33ING1b binds with the transcription co-activator CBP/p300, which is associated with gene transcription activation [8]. So we examined whether p33ING1b regulated p16INK4a expression through this mechanism. We demonstrated that p33ING1b binds to and activates p16INK4a
promoter to increase its transcription. Furthermore, we found that p33ING1b and p300 co-localized in the promoter region of p16INK4a. After using p300/CBP-specific cell permeable HAT inhibitor curcumin [19,20], p33ING1b-mediated induction of p16INK4a was reduced in both 2BS and HEK293 cells (Fig. 4). These results suggest that p33ING1b might exert its transcriptional activation of p16INK4a through HAT-dependent mechanism. In addition to p33ING1b, we also examined the functions of overexpression of p47ING1a in 2BS fibroblasts and HEK293 cells. Unlike p33ING1b, overexpression of p47ING1a showed discrepant effects on p16INK4a expression in these two cell lines (data not shown). More investigations are needed to illustrate this discrepancy. As p33ING1b protein is the major isoform expressed in 2BS fibroblasts, in this study we focused on the function of p33ING1b in cellular senescence and gene regulation. In summary, we observed that p33ING1b protein is the major ING1 isoform expressed in 2BS fibroblasts. Overexpression of p33ING1b induces some markers of senescence in 2BS cells. Moreover, we established for the first time that p33ING1b increases p16INK4a expression in 2BS and HEK293 cells, p33ING1b specifically binds and activates the p16INK4a promoter and p300/CBP HAT specific inhibitor curcumin can reverse the effect of p33ING1b on p16INK4a. As the effect and mechanism of p33ING1b protein in
3112
N. Li et al. / FEBS Letters 585 (2011) 3106–3112
cellular senescence have not been fully demonstrated, our results may help better understand the function of p33ING1b in cellular senescence. Acknowledgements This work was supported by Grant No. 2007CB507400 from the National Basic Research Programs of China. We are sincerely grateful to Professor Karl Riabowol, who generously provided the pCI-ING1 plasmids that were used in this study. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.febslet.2011.08.044. References [1] Russell, M., Berardi, P., Gong, W. and Riabowol, K. (2006) Grow-ING, Age-ING and Die-ING: ING proteins link cancer, senescence and apoptosis. Exp. Cell Res. 312, 951–961. [2] Garkavtsev, I., Kazarov, A., Gudkov, A. and Riabowol, K. (1996) Suppression of the novel growth inhibitor p33(ING1) promotes neoplastic transformation. Nat. Genet. 14, 415–420. [3] Saito, A., Furukawa, T., Fukushige, S., Koyama, S., Hoshi, M., Hayashi, Y. and Horii, A. (2000) P24/ING1-ALT1 and p47/ING1-ALT2, distinct alternative transcripts of p33/ING1. J. Hum. Genet. 45, 177–181. [4] Kataoka, H. et al. (2003) ING1 represses transcription by direct DNA binding and through effects on p53. Cancer Res. 63, 5785–5792. [5] Leung, K.M., Po, L.S., Tsang, F.C., Sin, W.Y., Lau, A., Ho, H. and Poon, R. (2002) The candidate tumor suppressor ING1b can stabilize p53 by disrupting the regulation of p53 by MDM2. Cancer Res. 62, 4890–4893. [6] Tsang, F.C., Po, L.S., Leung, K.M., Lau, A., Siu, W.Y. and Poon, R. (2003) ING1b decreases cell proliferation through p53-dependent and -independent mechanisms. FEBS Lett. 553, 277–285. [7] Coles, A.H., Liang, H.L., Zhu, Z.Q., Marfella, C.G., Kang, J., Imbalzano, A.N. and Jones, S.N. (2007) Deletion of p37(Ing1) in mice reveals a p53-independent role for Ing1 in the suppression of cell proliferation, apoptosis, and tumorigenesis. Cancer Res. 67, 2054–2061. [8] Vieyra, D. et al. (2002) Human ING1 proteins differentially regulate histone acetylation. J. Biol. Chem. 277, 29832–29839. [9] Skowyra, D. et al. (2001) Differential association of products of alternative transcripts of the candidate tumor suppressor ING1 with the mSin3/HDAC1 transcriptional corepressor complex. J. Biol. Chem. 276, 8734–8739.
[10] Kuzmichev, A., Zhang, Y., Erdjument-Bromage, H., Tempst, P. and Reinberg, D. (2002) Role of the Sin3-histone deacetylase complex in growth regulation by the candidate tumor suppressor p33(ING1). Mol. Cell Biol. 22, 835–848. [11] Garkavtsev, I. and Riabowol, K. (1997) Extension of the replicative life span of human diploid fibroblasts by inhibition of the p33(ING1) candidate tumor suppressor. Mol. Cell Biol. 17, 2014–2019. [12] Goeman, F., Thormeyer, D., Abad, M., Serrano, M., Schmidt, O., Palmero, I. and Baniahmad, A. (2005) Growth inhibition by the tumor suppressor p33ING1 in immortalized and primary cells: Involvement of two silencing domains and effect of Ras. Mol. Cell Biol. 25, 422–431. [13] Abad, M., Moreno, A., Palacios, A., Narita, M., Blanco, F., Moreno-Bueno, G., Narita, M. and Palmero, I. (2011) The tumor suppressor ING1 contributes to epigenetic control of cellular senescence. Aging Cell 10, 158–171. [14] Soliman, M.A., Berardi, P., Pastyryeva, S., Bonnefin, P., Feng, X., Colina, A., Young, D. and Riabowol, K. (2008) ING1a expression increases during replicative senescence and induces a senescent phenotype. Aging Cell 7, 783–794. [15] Xue, L.X., Wu, J.F., Zheng, W.J., Wang, P.C., Li, J., Zhang, Z.Y. and Tong, T.J. (2004) Sp1 is involved in the transcriptional activation of p16(INK4) by p21(Waf1) in HeLa cells. FEBS Lett. 564, 199–204. [16] Wang, W., Wu, J.F., Zhang, Z.Y. and Tong, T.J. (2001) Characterization of regulatory elements on the promoter region of p16(INK4a) that contribute to overexpression of p16 in senescent fibroblasts. J. Biol. Chem. 276, 48655– 48661. [17] Narita, M., Narita, M., Krizhanovsky, V., Nuñez, S., Chicas, A., Hearn, S.A., Myers, M.P. and Lowe, S.W. (2006) A novel role for high-mobility group A proteins in cellular senescence and heterochromatin formation. Cell 126, 503–514. [18] Zhou, R., Han, L.M., Li, G.D. and Tong, T.J. (2009) Senescence delay and repression of p16(INK4a) by Lsh via recruitment of histone deacetylases in human diploid fibroblasts. Nucleic Acids Res. 37, 5183–5196. [19] Balasubramanyam, K., Varier, R.A., Altaf, M., Swaminathan, V., Siddappa, N.B., Ranga, U. and Kundu, T.K. (2004) Curcumin, a novel p300/CREB-binding protein-specific inhibitor of acetyltransferase, represses the acetylation of histone/nonhistone proteins and histone acetyltransferase-dependent chromatin transcription. J. Biol. Chem. 279, 51163–51171. [20] Marcu, M.G., Jung, Y.J., Lee, S., Chung, E.J., Lee, M.J., Trepel, J. and Neckers, L. (2006) Curcumin is an inhibitor of p300 histone acetylatransferase. Med. Chem. 2, 169–174. [21] Gong, W., Suzuki, K., Russell, M. and Riabowol, K. (2005) Function of the ING family of PHD proteins in cancer. Int. J. Biochem. Cell B. 37, 1054–1065. [22] Huang, Y., Wu, J.F., Li, R.Z., Wang, P.C., Han, L.M., Zhang, Z.Y. and Tong, T.J. (2011) B-MYB delays cell aging by repressing p16INK4a transcription. Cell. Mol. Life Sci. 68, 893–901. [23] Zheng, W.J., Wang, H.Y., Xue, L.X., Zhang, Z.Y. and Tong, T.J. (2004) Regulation of cellular senescence and p16INK4a expression by Id1 and E47 proteins in human diploid fibroblast. J. Biol. Chem. 279, 31524–31532. [24] Gan, Q.N., Huang, J., Zhou, R., Niu, J., Zhu, X.J., Wang, J., Zhang, Z.Y. and Tong, T.J. (2008) PPARc accelerates cellular senescence by inducing p16INK4a expression in human diploid fibroblasts. J. Cell Sci. 121, 2235–2245.
journal homepage: www.FEBSLetters.org
The tumor suppressor p33ING1b upregulates p16INK4a expression and induces cellular senescence Na Li, Qian Li, Xiaoxiao Cao, Ganye Zhao, Lixiang Xue ⇑, Tanjun Tong ⇑ Research Center on Aging, Department of Biochemistry and Molecular Biology, Peking University Health Science Center, Beijing 100191, China
a r t i c l e
i n f o
Article history: Received 16 May 2011 Revised 29 August 2011 Accepted 29 August 2011 Available online 2 September 2011 Edited by Angel Nebreda Keywords: p33ING1b p16INK4a HAT Cellular senescence
a b s t r a c t ING1 protein is a tumor suppressor which plays significant roles in multiple cellular activities. p47ING1a and p33ING1b are major splice isoforms of ING1 and their roles in senescence need further investigations. Here we studied the functions of ING1 isoforms in cellular senescence and gene regulation, with focus on p16INK4a. We observe that p33ING1b protein is the major ING1 isoform expressed in 2BS human diploid fibroblasts. Overexpression of p33ING1b induces cellular senescence and upregulates p16INK4a expression in 2BS fibroblasts. p33ING1b upregulates p16INK4a transcription. p33ING1b and p300 bind to the p16INK4a promoter. p300/CBP-specific inhibitor curcumin can reverse the induction of p16INK4a by p33ING1b. These results help to better understand the function of ING1. Ó 2011 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
1. Introduction ING1 (inhibitor of growth1) protein functions as a type II tumor suppressor which plays a significant role in multiple cellular activities such as growth regulation, DNA repair, apoptosis and senescence [1]. ING1 gene was cloned in 1996 [2], and several years later, its isoforms p47ING1a, p33ING1b and p24ING1c were characterized [3]. Current evidence suggests many functions of ING1 rely on p53 [4,5]. However, some other results obtained from different cell lines implicated that ING1 could function in a p53-independent manner [6,7]. Given the facts that ING1 proteins are involved in chromatin remodeling through association with HAT (histone acetyl-transferase) or HDAC (histone deacetylase) complexes, it was suggested that ING1 proteins might regulate target gene expression through localized chromatin modification [8–10]. The first analysis of ING1 in cell senescence identified ING1 protein had pro-senescence effect, but at that time it was unable to tell which isoform of ING1 exert such effect [11]. Later, it was reported p33ING1b overexpression induced features of cellular senescence in human diploid fibroblasts cells [12,13]. But another study showed Abbreviations: ING, inhibitor of growth; HAT, histone acetyl-transferase; HDAC, histone deacetylase; SAHF, senescence-associated heterochromatic foci; PD, population doubling; qPCR, quantitative polymerase chain reaction ⇑ Corresponding authors at: Peking University Research Center on Aging, Department of Biochemistry and Molecular Biology, No.38, Xueyuan Road, Haidian District, Beijing 100191, China. Fax: +86 10 82802931. E-mail address: [email protected] (T. Tong).
that p47ING1a but not p33ING1b had the potential to promote senescent cell morphology [14]. Here we tried to dissect the expression level and function of ING1 isoforms in replicative cellular senescence. And it is well known that p16INK4a plays important role in replicative senescence. So we investigated whether ING1 isoforms could influence cellular senescence with focus on p16INK4a expression. And we attempted to dissect the molecular mechanism. Our results indicate that p33ING1b is the major isoform of ING1 expressed in 2BS fibroblasts. p33ING1b promotes cellular senescence by upregulation of p16INK4a expression. And p33ING1b interacts in vivo with the p16INK4a promoter and requires HAT activity to increase p16INK4a expression. These observations help to further understand the biological functions of ING1 isoforms in cellular senescence. 2. Materials and methods 2.1. Cell culture and reagent Human embryonic lung fibroblast cell line (2BS cells) was obtained from the National Institute of Biological Products (Beijing, China) and cultured as described previously [15]. 2BS cells are considered to be young at PD30 (population doubling30) or below and to be fully senescent at PD55 or above. Senescent cells are characterized by an irreversible growth arrest. Human embryonic kidney cell line (HEK293 cells) was cultured in DMEM medium, containing 10% fetal bovine serum. Curcumin,
0014-5793/$36.00 Ó 2011 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2011.08.044
N. Li et al. / FEBS Letters 585 (2011) 3106–3112
3107
with purity P95%, was purchased from Sigma (St. Louis, MO, USA) and used at indicated concentration.
(Leica). Representative photographs from three independent experiments are shown.
2.2. Plasmids construction
2.8. Luciferase activity assay
Expression plasmids of p47ING1a and p33ING1b cDNA in the pCI vector were gifts of Professor Karl Riabowol. Plasmids were digested by EcoRI and XbaI, and cloned into pcDNA3.1 (Invitrogen) or pLVX-IRES-ZsGreen1 (Clontech) linearized with the same enzymes. 50 -fragment of human p16INK4a promoter was digested with XhoI and HindIII or exemplified by PCR from pSIR-EGFP vector [16], and inserted into pGL3-Basic Luciferase Report Vector linearized with the same enzymes (Promega).The p16 shRNA sequence was as previously described [17]. The template oligonucleotides were chemical synthesized and inserted into the pMSCV-puromiR30 vector.
Twenty-four or forty-eight hours after transfection, the activities of luciferase were quantified in a luminometer (Centro LB 960; Berthold Technologies) using the Dual Luciferase Reporter Assay System (Promega). All experiments were performed in triplicate.
2.3. Transfection All plasmids were purified with Qiagen Plasmid Midi Kits. Cells were transfected with plasmids coated by lipofectamine 2000 (Invitrogen) following the manufacturer’s indications. To over-express p33ING1b or knock down p16INK4a in 2BS cells, recombinant viruses were produced according to manufacturer’s indications (Clontech). 2.4. Preparation of total RNA and qPCR RNA preparation and qPCR (quantitative polymerase chain reaction) were performed according to manufacturer’s instructions (invitrogen). The primer sequences were as follows: p47ING1a 50 -CGGGCCTAGCGCAATAAC-30 and 50 -CCCTCGGGTCA GCAAAGAT-30 ; p33ING1b 50 -TCCCTGCCTTTCGACTTG-30 and 50 -GCTTGCTGTTG GGCTTGT-30 ; p16INK4a 50 -CAACGCACCGAATAGTTACGG-30 and 50 -GCGCAGTT GGGCTCCG-30 ; p53 50 -AATGCCAGAGGCTGCTCCC-30 and 50 -GGTGTGGAATCA ACCCAC-30 ; b-actin 50 -AGCGAGCATCCCCCAAAGTT-30 and 50 -GGGCACGAAGGCT CATCATT-30 . 2.5. Western blot Western bolts were performed as described [15]. The antibodies used are anti-p16INK4a (Clone 16P04; NeoMarkers), anti-p53 (sc-126; Santa Cruze), anti-ING1 (Santa Cruze; Millipore or Abcam, the antibody from Abcam only recognizes p33 ING1b ), anti-H3K9me3 (ab8898; Abcam), anti-p300 (05–257; Millipore), anti-b-actin (Santa Cruze) and anti-tubulin (Millipore). 2.6. SA-b-gal assays The 2BS cells were washed twice in PBS, fixed in 3% formaldehyde, and washed again in PBS. The cells were incubated overnight at 37 °C (without CO2) with freshly prepared SA-b-gal staining solution as described [18]. 2.7. SAHF and Immunofluorescence SAHF (senescence-associated heterochromatic foci) (DAPI foci) were detected by staining with DAPI at room temperature. Image collection was by a confocal microscope [18]. Immunofluorescence were performed as previous described. Stained cells were examined with a confocal laser microscope
2.9. Chromatin immunoprecipitation Chromatin immunoprecipitations (ChIPs) were performed according to manufacturer’s instructions (Upstate). Re-ChIP was performed as previously described [18].The sequences of the primers for ChIP assays were as follows: Primer-1 1.7 kb upstream 50 -CACAGGAAAAATTAGAAC-30 50 -CCGAGATCGCGCCAT-30 ; Primer-2 1.0 kb upstream 50 -TTCCCAAAGTGCTGGGAT-30 50 -GGGGTTGTTGTGAGTTTA-30 ; Primer-3 200 bp upstream 50 -CCCCG ATTCA ATTTG GCAGT-30 50 -CAGCG TTGGC AAGGA AGGAG-30 ; Primer-4 End of INK4A exon 1a 50 -AGAGGGTCTGCAGCGG-30 50 -TCGAAGCGCTACCTGATTCC-30 ; GAPDH promoter 50 -TACTAGCGGTTTTACGGGCG-30 50 -TCGAACAGGAGGAGCAGAGAGCGA-30 . 2.10. Statistics All values are expressed as means ± S.D. in the figures. Student0 s t-test was used to perform statistical analyses. Differences with ⁄ P < 0.05 were considered significant.
3. Results 3.1. Expression levels of p47ING1a and p33ING1b in replicative senescence process of 2BS cells Reportedly, p47ING1a and p33ING1 are two major isoforms of ING1 in human cells. They have conserved C-terminus, while have distinct N-terminus with different affinities for HATs and HDACs (Fig. 1A). We tested the total protein and mRNA levels of ING1 isoforms in 2BS cells. p16INK4a expression was also detected at the same time to indicate the senescent state of 2BS cells. The results showed that the RNA levels of p47ING1a were higher in middle-aged and senescent 2BS cells than young 2BS cells; and p33ING1b mRNA expression was upregulated in middle-aged 2BS cells and decreased in young and senescent cells (Fig. 1B). The total protein levels of p47ING1a were too low to detect in 2BS cells; while p33ING1b protein levels were consistent with its mRNA levels. As ING1 protein participates in gene transcriptional control, we detected ING1 expression in chromatin fractions of 2BS fibroblasts. The results showed that p33ING1b expression increased in chromatin fractions of middle-aged and senescent 2BS cells. Notably, protein and mRNA levels of p16INK4a increased in the middle-aged and senescent 2BS cells in which chromatin fraction levels of p33ING1b increased (Fig. 1C). 3.2. p33ING1b induces senescent markers and p16INK4a expression in 2BS cells To test whether p33ING1b plays roles in senescence progress of 2BS cells, we detected the cellular senescence markers in p33ING1b
3108
N. Li et al. / FEBS Letters 585 (2011) 3106–3112
A
188
1
p47ING1a
422 NCR
NLS
PHD
PBR
HDACs
1
p33ING1b
PB
HDACs
NCR
HATs
NLS
257 PHD
279 PBR
HDACs
C
B
212
126
45 PIP
Y
M
S
Relative mRNA level (fold)
p47ING1a 20 18 16 14 12 10 8 6 4 2 0
Y M S
p33ING1b p16 β-actin
Y p47
p33
p16
M
S p33ING1b Histones
Fig. 1. The expression levels of p47ING1a and p33ING1b in young, middle-aged and senescent 2BS cells. (A) Schematic representation of INGI isoforms p47ING1a and p33ING1b. PIP = PCNA-interacting protein motif; PB = partial bromodomain; NCR = novel conserved region; NLS = nuclear localization sequence; PHD = plant homeodomain; PBR, polybasic region. (B) The mRNA expression of ING1 transcripts by qPCR analysis. (C) Total protein levels (upper panel) and chromatin levels (lower panel) of ING1 isoforms in young (Y; PD17), middle-aged (M; PD43) and senescent (S; PD65) 2BS cells. The histones were detected by coomassie blue staining. A representative result of three separate experiments is shown.
-transfected 2BS cells. The results showed that 2BS cells overexpressing of p33ING1b efficiently induced the expression of SA-bgal activity, flattened senescent cell morphology, and SAHF DNA staining pattern which is indicated by co-localization with H3K9Me3 (Fig. 2A and B). To gain insights into the possible relationship of ING1 isoforms with senescence-related genes, we investigated the effects of ING1 isoforms on the expressions of p16INK4a and p53 in 2BS cells. qPCR and Western blot results showed that introduction of p33ING1b upregulated p16INK4a mRNA and protein levels (Fig. 2C and D). Moreover, p16INK4a promoter activity was upregulated by p33ING1b (Fig. 2E). Reportedly, ING1 proteins regulate p53 protein by posttranscriptional mechanisms [4,5]. Consistent with that, our results showed that p53 protein levels were slightly upregulated by p33ING1b, while the mRNA level of p53 did not increase (Fig. 2C and D). These results suggested that p33ING1b may play roles in cellular senescence partly through upregulation of p16 INK4a transcription. To further verify that p33ING1b could regulate senescence via p16INK4a, we evaluated the role of p16 in mediating the senescence effects by p33ING1b in 2BS cells. 2BS cells infected with both p33ING1b and p16-shRNA or with p33ING1b or p16-shRNA alone were monitored. The results showed that cells treated with p33ING1b alone revealed strong SA-b-gal staining and p16 increasing, whereas cells doubly treated with p33ING1b and p16-shRNA reversed the senescence phenotypes. As expected, p16-shRNA-infected cells displayed young phenotypes (Fig. 2F). To confirm the effects of p33ING1b on p16INK4a expression, we overexpressed p33ING1b in HEK293 cells, and analyzed the protein expressions, mRNA levels and promoter activities of p16INK4a. p53 levels were also detected as controls. The results showed that p33ING1b markedly increased p16INK4a protein level, mRNA level
and promoter activity in HEK293 cells (Fig. 2G–I). p53 mRNA expression did not increase, which is in agreement with 2BS fibroblasts. These results showed a well correlation of promoter activity, mRNA levels and protein levels of p16INK4a upon ectopic expression of p33ING1b both in 2BS fibroblasts and HEK293 cells; suggesting p16INK4a expression was regulated by p33ING1b at the transcriptional level. 3.3. p33ING1b activates and binds to p16INK4a promoter region in vivo To investigate the mechanism of regulation of p16INK4a transcription by p33ING1b, we examined the effects of p33ING1b on deletion constructs of p16INK4a transcription regulatory regions in HEK293 cells. We used the constructs from 178 bp downstream (+178 bp) to 1765 bp upstream (1765 bp) of the transcriptional start site of p16INK4a. The luciferase activity assays showed that p33ING1b activated p16INK4a promoter regions, and the main regions contributing to over-expression of p16INK4a by p33ING1b were located in the region from +178 to 140 bp of p16INK4a transcriptional start site (Fig. 3A). Then we examined the in vivo DNA-binding activity of p33ING1b to p16INK4a promoter. The antibody used for ChIP assay only recognized p33ING1b isoform. In HEK293 cells, both RT–PCR and qPCR for ChIP analysis showed that p33ING1b specifically bound 200 bp upstream of the p16INK4a transcription start site (primer set 3). This region is also the p16INK4a promoter region significantly activated by p33ING1b (Fig. 3A and B). We also found that ectopic expressed p33ING1b bound the p16INK4a promoter in HEK293 cells, which is the same region of p16 promoter as the endogenous p33ING1b bound (Fig. 3C, upper panel). In 2BS cells, endogenous p33ING1b specifically bound this region of the p16INK4a promoter (primer set 3); moreover,
3109
N. Li et al. / FEBS Letters 585 (2011) 3106–3112
A
B
p33
Ctrl
p33
Ctrl
H3K9Me3
DAPI
D Ctrl
Relative mRNA level (fold)
C p33 p16 p53 p33 actin
ctrl p33
5 4
*
3 2 1
E
7
Luciferase activity (fold)
p33
6
ctrl p33
*
5 4 3 2 1
0
0 p53
control
p16
p1 3+
Ctrl
p3
3
6-s
p3
p1
Ctr
l
hR
6-s
NA
F
hR
NA
p16
p16-shRNA
p33
p33+ p16-shRNA
p16 p33
H Ctrl
p33 p16 p53 p33 actin
6 5
I
ctrl p33 *
4 3 2 1 0
Luciferase activity (fold)
G
Relative mRNA level (fold)
actin
6 5
ctrl p33
*
4 3 2 1 0
p16
p53
control
p16
Fig. 2. Overexpression of p33ING1b induce cellular senescence and p16INK4a level. (A) SA-b-gal activity (blue) and (B) SAHF formation (left panel) in vector or p33ING1b-infected stable cells. Immunostaining of H3K9me3 in p33ING1b-infected cells (right panel). White arrows indicated co-localization of H3K9me3 at SAHF. (C) Western blots and (D) qPCR analysis of p16INK4a and p53 expressions in vector and p33ING1b-infected stable 2BS cells. (E) Stable infected 2BS cells were transfected with control promoter or p16 promoter. Luciferase activities were measured 24 h later. Values have been equalized relative to the value obtained with control empty vector (set at 1) and are the average of three independent determinations. ⁄P < 0.05. (F) Western blots and SA-b-gal analysis of the effects of knocking down p16INK4a on p33ING1b-induced senescence in 2BS cells. (G) Western blots and (H) qPCR analysis of p16INK4a and p53 expressions in vector or p33ING1b-transfected HEK293 cells. (I) Vector or p33ING1b-transfected HEK293 cells were transfected with control promoter or p16 promoter. Luciferase activities were measured as described. ⁄P < 0.05.
the binding of p33ING1b increased in middle-aged and senescent 2BS cells (Fig. 3C, lower panel). Therefore these results indicate that p33ING1b activates and specifically binds to the p16INK4a promoter region and imply again that p33ING1b plays a role in p16INK4a transcriptional regulation. 3.4. p33ING1b and p300 interact with the p16INK4a promoter to regulate its transcription Reportedly, p33ING1b interacts with proteins associated with HAT activity, such as p300/CBP [8]. To examine whether the effect of p33ING1b on p16INK4a transcription was dependent on p300/CBP HAT activity, Re-ChIP was performed using an anti-p33ING1b antibody in HEK293 cells, followed by reimmunoprecipitation using
an anti-p300 antibody, and then detection of PCR-amplified signal in the p16INK4a promoter region. The results showed that both p33ING1b and p300 associate with the regulatory region of the p16INK4a promoter (Fig. 4A). Curcumin has been reported to be a p300/CBP-specific cell permeable HAT inhibitor [19,20]. Here curcumin was used to confirm the effects of p300/CBP HAT on p33ING1b-induced p16INK4a expression. After introduction of control vector or p33ING1b, 2BS and HEK293 cells were incubated with 1–20 lM curcumin. The results showed that p16INK4a protein expressions were increased in p33ING1b-overexpressed 2BS and HEK293 cells without curcumin treatment. While following the addition of curcumin, p33ING1b -induced increase of p16INK4a protein expressions reduced more than vector control cells. Treatment of cells with 20 lM curcumin
3110
N. Li et al. / FEBS Letters 585 (2011) 3106–3112
A
7
p33
6
Luciferase activity (fold)
B
control
p16
*
-1700
5
-200
-1000
+461
* *
4
* 3
2
1
2
3
4
1
+178
-140
-315
-735
IgG
-1765bp
1
C
p33
Input
0
2
3
4
1
2
4 1
3
2
3
4 p16
Input
HEK293
IgG p33
GAPDH
p16 GAPDH
Input Y
M
p33 S
Y
M
2BS
IgG S
Y
M
Bound/Input (fold)
5
S p16 GAPDH
4
IgG p33
3 2 1 0 1
2
3
4
Fig. 3. p33ING1 activates and specifically binds to p16INK4a promoter region. (A) Vector and p33ING1b-transfected HEK293 cells were transfected with constructs of p16 regulatory regions from 178 bp downstream to 1765 bp upstream p16 transcription start site. Luciferase activities were measured as described. ⁄P < 0.05. (B) Schematic diagram of the p16 gene promoter and the end of INK4a exon 1a (upper panel). ChIP analysis was performed using primers indicated by red bars in HEK293 cells. The precipitated DNA was amplified by RT–PCR (middle panel) and qPCR (lower panel). Values are expressed relative to the controls, which were set as 1. (C) ChIP assays of p33ING1b transfected-HEK293 cells (upper panel); and normal young (Y), middle-aged (M) and senescent (S) 2BS cells at p16 promoter using the primer set 3. The antibody used for ChIP recognized p33ING1b but not p47ING1a.
reversed completely the effects of p33ING1b on p16INK4a protein levels both in 2BS and HEK293 cells (Fig. 4B and C). In addition, the luciferase assays for p16INK4a promoter activities showed a similar tendency (Fig. 4D). Taken together, these results indicated that p33ING1b bound to p16INK4a promoter region in vivo, and used at least partly HAT-dependent mechanism for transcriptional induction of p16INK4a. 4. Discussion Although recent studies began to attract more attention to ING1 isoforms, their biological functions have not been fully elucidated up to date [1,21]. In the present study, we have investigated the expressions of the two major ING1 isoforms in young, middle-aged and senescent 2BS fibroblasts and detected their functions on gene expression and cell senescence. We observed the mRNA levels of p47INGla increased in middle-aged and senescent human diploid 2BS fibroblasts, however, its protein levels are very low in 2BS cells; suggesting some posttranscriptional mechanisms may regulate p47INGla protein levels in 2BS cells. p33ING1b is the major ING1 isoform expressed in 2BS fibroblasts, its mRNA and total protein expression elevated in middle-aged 2BS cells when p16INK4a levels began increasing, but its expression declined when cells entered senescent state. Reportedly, ING1 proteins regulate gene
expression through regulating chromatin organization [1,8,9,13,14]. By extracting the chromatin proteins, we found that the expression of p33ING1b increases in chromatin fractions of both middle-aged and senescent 2BS cells, with a more markedly increase in middle-aged cells (Fig. 1). A recent study also published chromatin expression increase of p33ING1b in Ras-induced senescence in IMR-90 fibroblasts [13]. Furthermore, we found that the binding between p33ING1b and p16INK4a promoter in middle-aged and senescent 2BS cells also increased (Fig. 3C). These results indicate that p33ING1b might play a role in increasing p16INK4a gene transcription in middle-aged and senescent 2BS fibroblasts. p16INK4a is an important gene in provoking and maintaining cellular senescence, there are several proteins regulate its transcription in senescent cells, such as Ets1, Id1, PPARc, B-MYB and Lsh [18,22– 24]. It is clear that the p16INK4a promoter is subject to multiple levels of control, and therefore p16INK4a regulation cannot be explained by a single isolated pathway. Here our observations suggest that p33ING1b might be one of the important factors in regulating p16INK4a expression in cellular senescence. Although p47ING1a was reported to induce senescent phenotype, it is controversial regarding p33ING1b in cellular senescence [12–14]. As p33ING1b protein is the major isoform of ING1 expressed in 2BS cells, we examined its role in cellular senescence. The results showed that overexpression of p33ING1b induced some
3111
N. Li et al. / FEBS Letters 585 (2011) 3106–3112
A
1st IP Input
IgG p33 p33 p300 IgG p300
2nd IP
p16 GAPDH
B
2BS Ctrl
C HEK293
1
0
Ctrl
5
1
10
5 10
20 Curcumin
p33
0
1
5 10
20 Curcumin
p16
p16
p33
p33
tubulin
tubulin
20 Curcumin
p33
0
1
5
10
20 Curcumin
p16
p16
p33
p33
tubulin
tubulin
2.5 Luciferase activity (fold)
D
0
*
2.0 *#
*
*# *#
1.5 *#
1.0 0.5 0.0
Ctrl p33 p33 p33 p33 p33 p33 Curcumin (μM) ING1b
0
0
1
INK4a
2.5
5
10
20
INK4a
Fig. 4. p33 and p300 bind to the promoter of p16 to induce p16 transcription. (A) Re-ChIP analysis showing the colocalization of p33ING1b with p300 at the p16 promoter using the primer set 3 in HEK293 cells. (B) 2BS cells and (C) HEK293 cells were infected or transfected with vector or p33ING1b as indicated. Cells were then treated with DMSO vehicle or various concentration of curcumin (1, 5, 10, 20 lM). Cells were collected for total protein 12 h after treatment. Western blots were used to detect indicated proteins. (D) HEK293 cells were transfected with control vector or p33ING1b as indicated, together with pGL3-p16 promoter. 24 h after transfection, cells were treated with DMSO vehicle or various concentration of curcumin (1, 2.5, 5, 10, 20 lM). 12 h after treatment, luciferase activities were measured.⁄P < 0.05 compared with vector control; #P < 0.05 compared with p33ING1b.
aspects of a senescent phenotype in 2BS fibroblasts, suggesting the important roles of p33ING1b in implementation of the replicative senescent state. It appears that senescence is not triggered by a single, linear series of events, but instead is regulated by a complex signaling network. p53 and p16INK4a play important roles in both tumor suppression and cellular senescence. Reportedly, ING1 proteins are functionally linked to the p53 pathway via mechanisms including control of p53 protein stability or posttranslational modifications [4,5]. Consistent with that, we observed that p53 mRNA levels did not increase upon p33ING1b overexpression. However, we found that p33ING1b increases p16INK4a gene mRNA, promoter activity and protein expression in both 2BS and HEK293 cells. And knocking down p16INK4a expression rescued the senescent effect by p33ING1b (Fig. 2). These results suggest that p33ING1b might increase p16INK4a expression at transcriptional level in cellular senescence, and p16INK4a is required for p33ING1b-induced premature senescence. Recent studies have shown that ING1 proteins may regulate gene transcription by epigenetic control [13,14]. p33ING1b binds with the transcription co-activator CBP/p300, which is associated with gene transcription activation [8]. So we examined whether p33ING1b regulated p16INK4a expression through this mechanism. We demonstrated that p33ING1b binds to and activates p16INK4a
promoter to increase its transcription. Furthermore, we found that p33ING1b and p300 co-localized in the promoter region of p16INK4a. After using p300/CBP-specific cell permeable HAT inhibitor curcumin [19,20], p33ING1b-mediated induction of p16INK4a was reduced in both 2BS and HEK293 cells (Fig. 4). These results suggest that p33ING1b might exert its transcriptional activation of p16INK4a through HAT-dependent mechanism. In addition to p33ING1b, we also examined the functions of overexpression of p47ING1a in 2BS fibroblasts and HEK293 cells. Unlike p33ING1b, overexpression of p47ING1a showed discrepant effects on p16INK4a expression in these two cell lines (data not shown). More investigations are needed to illustrate this discrepancy. As p33ING1b protein is the major isoform expressed in 2BS fibroblasts, in this study we focused on the function of p33ING1b in cellular senescence and gene regulation. In summary, we observed that p33ING1b protein is the major ING1 isoform expressed in 2BS fibroblasts. Overexpression of p33ING1b induces some markers of senescence in 2BS cells. Moreover, we established for the first time that p33ING1b increases p16INK4a expression in 2BS and HEK293 cells, p33ING1b specifically binds and activates the p16INK4a promoter and p300/CBP HAT specific inhibitor curcumin can reverse the effect of p33ING1b on p16INK4a. As the effect and mechanism of p33ING1b protein in
3112
N. Li et al. / FEBS Letters 585 (2011) 3106–3112
cellular senescence have not been fully demonstrated, our results may help better understand the function of p33ING1b in cellular senescence. Acknowledgements This work was supported by Grant No. 2007CB507400 from the National Basic Research Programs of China. We are sincerely grateful to Professor Karl Riabowol, who generously provided the pCI-ING1 plasmids that were used in this study. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.febslet.2011.08.044. References [1] Russell, M., Berardi, P., Gong, W. and Riabowol, K. (2006) Grow-ING, Age-ING and Die-ING: ING proteins link cancer, senescence and apoptosis. Exp. Cell Res. 312, 951–961. [2] Garkavtsev, I., Kazarov, A., Gudkov, A. and Riabowol, K. (1996) Suppression of the novel growth inhibitor p33(ING1) promotes neoplastic transformation. Nat. Genet. 14, 415–420. [3] Saito, A., Furukawa, T., Fukushige, S., Koyama, S., Hoshi, M., Hayashi, Y. and Horii, A. (2000) P24/ING1-ALT1 and p47/ING1-ALT2, distinct alternative transcripts of p33/ING1. J. Hum. Genet. 45, 177–181. [4] Kataoka, H. et al. (2003) ING1 represses transcription by direct DNA binding and through effects on p53. Cancer Res. 63, 5785–5792. [5] Leung, K.M., Po, L.S., Tsang, F.C., Sin, W.Y., Lau, A., Ho, H. and Poon, R. (2002) The candidate tumor suppressor ING1b can stabilize p53 by disrupting the regulation of p53 by MDM2. Cancer Res. 62, 4890–4893. [6] Tsang, F.C., Po, L.S., Leung, K.M., Lau, A., Siu, W.Y. and Poon, R. (2003) ING1b decreases cell proliferation through p53-dependent and -independent mechanisms. FEBS Lett. 553, 277–285. [7] Coles, A.H., Liang, H.L., Zhu, Z.Q., Marfella, C.G., Kang, J., Imbalzano, A.N. and Jones, S.N. (2007) Deletion of p37(Ing1) in mice reveals a p53-independent role for Ing1 in the suppression of cell proliferation, apoptosis, and tumorigenesis. Cancer Res. 67, 2054–2061. [8] Vieyra, D. et al. (2002) Human ING1 proteins differentially regulate histone acetylation. J. Biol. Chem. 277, 29832–29839. [9] Skowyra, D. et al. (2001) Differential association of products of alternative transcripts of the candidate tumor suppressor ING1 with the mSin3/HDAC1 transcriptional corepressor complex. J. Biol. Chem. 276, 8734–8739.
[10] Kuzmichev, A., Zhang, Y., Erdjument-Bromage, H., Tempst, P. and Reinberg, D. (2002) Role of the Sin3-histone deacetylase complex in growth regulation by the candidate tumor suppressor p33(ING1). Mol. Cell Biol. 22, 835–848. [11] Garkavtsev, I. and Riabowol, K. (1997) Extension of the replicative life span of human diploid fibroblasts by inhibition of the p33(ING1) candidate tumor suppressor. Mol. Cell Biol. 17, 2014–2019. [12] Goeman, F., Thormeyer, D., Abad, M., Serrano, M., Schmidt, O., Palmero, I. and Baniahmad, A. (2005) Growth inhibition by the tumor suppressor p33ING1 in immortalized and primary cells: Involvement of two silencing domains and effect of Ras. Mol. Cell Biol. 25, 422–431. [13] Abad, M., Moreno, A., Palacios, A., Narita, M., Blanco, F., Moreno-Bueno, G., Narita, M. and Palmero, I. (2011) The tumor suppressor ING1 contributes to epigenetic control of cellular senescence. Aging Cell 10, 158–171. [14] Soliman, M.A., Berardi, P., Pastyryeva, S., Bonnefin, P., Feng, X., Colina, A., Young, D. and Riabowol, K. (2008) ING1a expression increases during replicative senescence and induces a senescent phenotype. Aging Cell 7, 783–794. [15] Xue, L.X., Wu, J.F., Zheng, W.J., Wang, P.C., Li, J., Zhang, Z.Y. and Tong, T.J. (2004) Sp1 is involved in the transcriptional activation of p16(INK4) by p21(Waf1) in HeLa cells. FEBS Lett. 564, 199–204. [16] Wang, W., Wu, J.F., Zhang, Z.Y. and Tong, T.J. (2001) Characterization of regulatory elements on the promoter region of p16(INK4a) that contribute to overexpression of p16 in senescent fibroblasts. J. Biol. Chem. 276, 48655– 48661. [17] Narita, M., Narita, M., Krizhanovsky, V., Nuñez, S., Chicas, A., Hearn, S.A., Myers, M.P. and Lowe, S.W. (2006) A novel role for high-mobility group A proteins in cellular senescence and heterochromatin formation. Cell 126, 503–514. [18] Zhou, R., Han, L.M., Li, G.D. and Tong, T.J. (2009) Senescence delay and repression of p16(INK4a) by Lsh via recruitment of histone deacetylases in human diploid fibroblasts. Nucleic Acids Res. 37, 5183–5196. [19] Balasubramanyam, K., Varier, R.A., Altaf, M., Swaminathan, V., Siddappa, N.B., Ranga, U. and Kundu, T.K. (2004) Curcumin, a novel p300/CREB-binding protein-specific inhibitor of acetyltransferase, represses the acetylation of histone/nonhistone proteins and histone acetyltransferase-dependent chromatin transcription. J. Biol. Chem. 279, 51163–51171. [20] Marcu, M.G., Jung, Y.J., Lee, S., Chung, E.J., Lee, M.J., Trepel, J. and Neckers, L. (2006) Curcumin is an inhibitor of p300 histone acetylatransferase. Med. Chem. 2, 169–174. [21] Gong, W., Suzuki, K., Russell, M. and Riabowol, K. (2005) Function of the ING family of PHD proteins in cancer. Int. J. Biochem. Cell B. 37, 1054–1065. [22] Huang, Y., Wu, J.F., Li, R.Z., Wang, P.C., Han, L.M., Zhang, Z.Y. and Tong, T.J. (2011) B-MYB delays cell aging by repressing p16INK4a transcription. Cell. Mol. Life Sci. 68, 893–901. [23] Zheng, W.J., Wang, H.Y., Xue, L.X., Zhang, Z.Y. and Tong, T.J. (2004) Regulation of cellular senescence and p16INK4a expression by Id1 and E47 proteins in human diploid fibroblast. J. Biol. Chem. 279, 31524–31532. [24] Gan, Q.N., Huang, J., Zhou, R., Niu, J., Zhu, X.J., Wang, J., Zhang, Z.Y. and Tong, T.J. (2008) PPARc accelerates cellular senescence by inducing p16INK4a expression in human diploid fibroblasts. J. Cell Sci. 121, 2235–2245.