Báo cáo khoa học: Transcription factor specificity protein 1 (SP1) and activating protein 2a (AP-2a) regulate expression of human KCTD10 gene by binding to proximal region of promoter pot

Transcription factor specificity protein 1 (SP1) and
activating protein 2a (AP-2a) regulate expression of human
KCTD10 gene by binding to proximal region of promoter
Rushi Liu1,*, Aidong Zhou1,*, Daolong Ren1,*, Ailan He1, Xiang Hu1, Wenfeng Zhang1,
Liping Yang1, Mingjun Liu1, Hong Li1, Jianlin Zhou1, Shuanglin Xiang1 and Jian Zhang1,2
1 Key Laboratory of Protein Biochemistry and Development Biology of State Education Ministry of China, Hunan Normal University, China
2 Model Organisms Division, Shanghai Second Medical University, China

Keywords
AP-2a; KCTD10; promoter; regulatory
element; SP1
Correspondence
J. Zhang, Key Laboratory of Protein
Biochemistry and Development Biology of
State Education Ministry of China, College
of Life Sciences, Hunan Normal University,
Changsha, Hunan 410081, China
Fax: +86 731 887 2792
Tel: +86 731 887 2792
E-mail:
*These authors contributed equally to this
work
(Received 25 October 2008, revised 7
December 2008, accepted 11 December
2008)
doi:10.1111/j.1742-4658.2008.06855.x

Potassium channel tetramerization domain-containing 10 gene (KCTD10)
belongs to the polymerase delta-interacting protein 1 (PDIP1) gene family.
Mouse KCTD10 was thought to interact with proliferating cell nuclear
antigen and the small subunit of polymerase d, and to have roles in DNA

repair, DNA replication and cell-cycle control. To better understand the
regulatory mechanism of KCTD10 expression, we characterized the promoter of human KCTD10 containing a 639 bp fragment of the 5¢-flanking
region ()609 ⁄ +30). A primer extension assay identified the transcription
start site as an A, 63 bp upstream of the start codon. The promoter region
contains neither a TATA box nor a CCAAT box, but a CpG island was
found near to the transcription start site. Deletion mutagenesis showed that
the region from )108 to +30 was indispensable to the promoter activity,
and site-directed mutation analysis in this proximal promoter region of
KCTD10 revealed two important transcription regulatory elements of specificity protein 1 (SP1) and activating protein-2 (AP-2). An in vivo chromatin
immunoprecipitation assay further demonstrated that SP1 and AP-2a could
bind to this proximal promoter region. Moreover, using a luciferase reporter assay, quantitative real-time RT-PCR and western blot analysis, both
overexpression and RNA interference of SP1 and AP-2a indicated that the
binding of SP1 to the proximal promoter region stimulated the promoter
activity and endogenous KCTD10 expression, whereas binding of AP-2a to
this region showed opposite effects.

Rat potassium channel tetramerization domain-containing 10 (KCTD10) gene was recently cloned and
identified as a new member of the polymerase deltainteracting protein 1 (PDIP1) gene family. The amino
acid sequence of rat KCTD10 shares high levels of
identity with other members in this family; for example, 65.1% identity with PDIP1 and 66.7% identity
with tumor necrosis factor alpha-induced protein 1

(TNFAIP1), respectively. Like PDIP1 and TNFAIP1,
KCTD10 protein also contains a BTB ⁄ POZ domain
and a potassium channel tetramerization (K-tetra)
domain (a relative of BTB ⁄ POZ domain) at its
N-terminus, and a proliferating cell nuclear antigen
(PCNA)-binding motif at its C-terminus [1].
PCNA, a multifunctional protein, plays critical
roles in a variety of eukaryotic cellular processes,

Abbreviations
AP-2, activating protein-2; ChIP, chromatin immunoprecipitation; DPE, downstream promoter element; KCTD10, potassium channel
tetramerization domain-containing 10; PCNA, proliferating cell nuclear antigen; PDIP1, polymerase delta-interacting protein 1; Sp1, specificity
protein 1; TNFAIP1, tumor necrosis factor alpha induced protein 1; TNF-a, tumor necrosis factor-alpha; TSS, transcription start site.

1114

FEBS Journal 276 (2009) 1114–1124 ª 2009 Hunan Normal University. Journal compilation ª 2009 FEBS


R. Liu et al.

including DNA replication, DNA repair and cellcycle progression by interacting with proteins
involved in these processes. The best-understood role
of PCNA was as a slide clamp to tether DNA polymerase d to its template for high processive DNA
synthesis [2,3]. Many proteins have been shown to
bind PCNA; they all contain a consensus PCNAbinding motif, which was initially identified in p21
[4]. Recent studies have shown that PDIP1 and
TNFAIP1 could interact with PCNA and stimulate
PCNA-dependent
polymerase d
activity
[5,6].
KCTD10 can also interact with the small subunit of
DNA polymerase d (p50) and PCNA [1]. These findings suggested that PCNA might function as a regulatory target of the PDIP1 gene family including
KCTD10, coordinating DNA replication, DNA repair
and cell-cycle progression.
Tumor necrosis factor-alpha (TNF-a) is a multifunctional cytokine involved in a variety of biological
activities, such as apoptosis, proliferation, B-cell activation and some inflammatory responses [7]. It is

known that TNF-a induces cell proliferation in liver
regeneration after hepatocyte loss caused by surgical
resection or chemical injury [8]. Like other members of
the PDIP1 gene family, KCTD10 is also supposed to
be induced by TNF-a [1]. KCTD10 is an interacting
partner of PCNA and the small subunit of pol d, so it
probably plays some roles in cell proliferation by linking TNF-a signaling to DNA synthesis. Recently, a
transgenic Caenorhabditis elegans model of Alzheimer’s
disease was achieved by expressing human b-amyloid
peptide [9]. In the brain of transgenic C. elegans,
TNFAIP1 expression was found to be markedly
increased; and the brain region with minimum pathological symptoms showed the highest expression of
TNFAIP1, suggesting that TNFAIP1 may have a
protective function during Alzheimer’s disease progression [9]. These studies indicated that the PDIP1 gene
family, including KCTD10, might function in the proliferation and development of cells, or in disease
progression.
Although the function of KCTD10 is very important, regulation of its expression remains unclear. We
characterized a 639 bp genomic fragment in the
5¢-flanking region of human KCTD10; and identified
the binding sites of two transcription factors, specificity protein 1 (SP1) and activating protein-2a (AP-2a),
in the promoter region of KCTD10. We found that the
proximal promoter region from )108 to +30 was
indispensable for basal promoter activity of KCTD10;
and SP1 and AP-2a can regulate the promoter activity
and endogenous KCTD10 expression oppositely
through binding this region.

SP1 and AP-2a regulate KCTD10

Results

Analysis of genomic structure and identification
of transcription start site of human KCTD10
To determine the genomic structure of human
KCTD10, a full-length cDNA sequence of human
KCTD10 was used to search the human genome database. The results showed that human KCTD10 was
mapped to chromosome 12q24.11 region, spanned
 21.3 kb and contained five exons. Mouse and
human KCTD10 share identical organization and
exons with similar length (Fig. 1A). In the human
genome, KCTD10 was arranged in the reverse direction to its neighbor gene – ubiquitin protein ligase
E3B (UBE3B, NM_183415). The 5¢-flanking region
was very short, only  380 bp between the start the
codon of human KCTD10 and 5¢ cDNA end of
UBE3B (Fig. 1B).
The transcription start site (TSS) of KCTD10 was
determined by primer extension assay. An antisense
primer on the 3¢-end of the first exon was labeled with
[32P]ATP[cP], and extended with AMV reverse transcriptase using total RNAs from HeLa cells as
template; simultaneously, sequencing reaction was performed using the same primer. Nucleotide A, 63 bp
upstream of the start codon, was identified as the TSS
by sequence reading (Fig. 2).
Identification of the cis-regulatory region and
trans-regulatory factors responsible for the
promoter activity of human KCTD10
Computer-aided analysis was performed to predict
the potential cis-elements regulating KCTD10 expression in the 5¢-flanking region, a 639 bp genomic
sequence upstream of the start codon of KCTD10.
We did not find any TATA box or CCAAT box
around the TSS, but checked out a CpG island with
64.7% GC content throughout the whole 639 bp

sequence. The 5¢-flanking region of mouse KCTD10
displayed very similar genomic organization to that
of human KCTD10.
To characterize potential sequences involved in the
regulation of KCTD10 expression, the 5¢-flanking
region was amplified, and inserted into the upstream
of pTAL-Luc – a promoter-free luciferase reporter
vector. A series of 5¢ deletion constructs with a common 3¢-end except for construct P ()609 ⁄ )241) were
generated using different primers (Table 1). Luciferase
activities of these constructs were assayed in transfected HeLa cells. As shown in Fig. 3, construct P
()609 ⁄ +30) containing the whole 639 bp sequence

FEBS Journal 276 (2009) 1114–1124 ª 2009 Hunan Normal University. Journal compilation ª 2009 FEBS

1115


SP1 and AP-2a regulate KCTD10

R. Liu et al.

A

B

Fig. 1. Genomic structure of human and
mouse KCTD10 genes, and the 5¢-flanking
region sequence of human KCTD10.
(A) Genomic organization of human and
mouse KCTD10 genes. Solid boxes indicate

coding region, and open boxes indicate
5¢UTR and 3¢UTR. (B) 5¢-Flanking region of
human KCTD10 from )609 to +30. KCTD10
was arranged in the reverse direction to its
neighboring gene ubiquitin protein ligase
E3B (UBE3B, NM_183415); TSS of human
KCTD10 is denoted +1, and the start codon
(ATG) is indicated by solid box. The consensus binding sites of the transcription factors
with high regulatory activities are boxed,
other sites with no or weak regulatory
activities are underlined.

produced  17.5-fold higher luciferase activity compared with the control pLuc vector. Taking this as a
100% base, deletion from )609 to )343 showed a
7.4% increase; construct P ()609 ⁄ )241), in which
271 bp of the 3¢-end was deleted, presented almost
no promoter activity; by contrast, construct P
()241 ⁄ +30) containing this 271 bp of the 3¢-end
showed only a 8.7% increase. These results suggested
that the functional promoter region was located in
the 271 bp region of the 3¢-end (from )241 to +30).
Further deletion extended to )203 showed an
 28.1% decrease compared with P ()241 ⁄ +30),
whereas deletion to )108 showed an  67.4%
increase compared with P ()203 ⁄ +30). This indicated
that the potential positively-regulating region might
exist from )241 to )203, and the potential negativeregulating region might exist from )203 to )108.
Deletion to )13 sharply decreased luciferase activity
to only 1.68% of P ()609 ⁄ +30). The 5¢-flanking
region contained the functional promoter; within it,

the region from )108 to +30 is essential for the promoter activity.
Using tfsearch and matinspector programs,
multiple potential binding sites of transcription factors were found in the promoter region from )108
to +30 (Fig. 1B), including two binding sites for
Sp1, AP-2a and c-myb, and one CACCC box.
1116

Mutational analysis revealed that mutation of the
upstream Sp1 site (Sp1.1) decreased the promoter
activity by 46.9%, whereas mutation of the downstream AP-2a site (AP-2.2) strongly increased the
promoter activity by 88.1% (Fig. 4A). However,
although other sites were mutated individually, either
no effects (c-myb.2, AP-2.1 and Sp1.2, data not
shown) or very modest effects (mutation of the
c-myb.1 site decreased the promoter activity by
28.6%, and mutation of CACCC box increased the
promoter activity by 27.4%) were observed (Fig. 4A).
Taken together, Sp1.1 and AP-2.2 were two important elements related to the regulation of human
KCTD10 expression.
By aligning the 5¢-flanking region of both human
and mouse KCTD10 promoters (human from )144 to
)45, mouse from )143 to )44 relative to the start
codon), we found that two sequences shared great
identities, and Sp1.1 sites were highly conserved
(Fig. 4B). Although one nucleotide varied from G to T
in the AP-2.2 site (from )77 to )69) of mouse
KCTD10 promoter, the sequence (GCCCCCTGC)
might also be able to be recognized by AP-2a,
and play similar role in the regulation of KCTD10
expression. Thus, we proposed that the regulatory

mechanisms of KCTD10 might be conserved between
human and mouse.

FEBS Journal 276 (2009) 1114–1124 ª 2009 Hunan Normal University. Journal compilation ª 2009 FEBS


R. Liu et al.

SP1 and AP-2a regulate KCTD10

Roles of Sp1 and AP-2a in the regulation of the
promoter activity and the expression of human
KCTD10

Fig. 2. Transcription start site of human KCTD10. Antisense primer
from +21 to +40 was labeled with [32P]ATP[cP] using T4 polynucleotide kinase, then annealed to total RNAs template from HeLa
cells, and extended with AMV reverse transcriptase. DNA
sequence ladder was obtained using the same primer and separated on the same gel. The extended product is indicated by the
arrow (PE means primer extension).

Binding of Sp1 and AP-2a to the corresponding
sites in the promoter of human KCTD10 in vivo
To confirm that Sp1 and AP-2a bind to the authentic endogenous cis-elements in the KCTD10 promoter, we generated a pool of DNA fragments from
HeLa cell lysates using chromatin immunoprecipitation (ChIP) with antibodies against Sp1 or AP-2a.
The ChIP-DNA was used as a template for PCR
amplification of the genomic regions containing Sp1
or AP-2a binding elements in the KCTD10 promoter, and representative ChIP-PCR results were
shown in Fig. 5. The chromatin fragment containing
Sp1- or AP-2a-binding sites in the KCTD10 promoter was precipitated by antibodies against Sp1 or
AP-2a but not by control IgG, indicating that

endogenous Sp1 or AP-2a proteins specifically bound
the Sp1 or AP-2a sites, and worked in the basal
state to regulate KCTD10 transcription, respectively.

To test the effects of Sp1 and AP-2a binding on
KCTD10 promoter activity, we performed a luciferase
assay in HeLa and HepG2 cells by overexpression of
Sp1 or AP-2a. In HeLa cells, co-transfected pCMVSp1 increased the luciferase activity of construct P
()241 ⁄ +30) by 58.3%; but only weakly increased the
luciferase activity of the Sp1.1 site mutated construct
Mt-Sp1.1 (Fig. 6A). In AP-2a-deficient HepG2 cells,
several reporter constructs containing 5¢-flanking
region of KCTD10 such as P ()609 ⁄ +30), P
()108 ⁄ +30) and P ()13 ⁄ +30) presented higher luciferase activities than in AP-2a-normal HeLa cells
(Fig. 6B); and co-transfected pCMV–AP-2a deceased
the luciferase activity of construct P ()241 ⁄ +30) in a
dose-dependent manner (Fig. 6C). These data suggested that Sp1 upregulated, but AP-2a downregulated, the promoter activity of human KCTD10.
Finally, to determine if the expression changes of
AP-2a or Sp1 alter the KCTD10 level, we checked
endogenous KCTD10 expression in HeLa cells at the
mRNA and protein levels using quantitative real-time
RT-PCR and western blot through overexpressing or
suppressing AP-2a or Sp1. As shown in Fig. 7A,
mRNA levels in HeLa cells were measured using quantitative real-time RT-PCR, and presented as base 10
logarithms. For AP-2a, overexpression of AP-2a (A1,
purple bar versus yellow bar) led to a 5.6-fold decrease
of KCTD10 at its transcription level (A3, purple bar
versus yellow bar, base 10 logarithms); whereas AP-2a
knockdown (A1, red bar versus yellow bar) caused a
2.34-fold induction of KCTD10 mRNA level (A3, red

bar versus yellow bar, base 10 logarithms). For Sp1,
its overexpression (A1, green bar versus yellow bar)
resulted in a 12.0-fold increase (A3, green bar versus
yellow bar, base 10 logarithms); whereas its knockdown (A1, blue bar versus yellow bar) resulted in a
1.9-fold decrease in KCTD10 mRNA level (A3, blue
bar versus yellow bar, base 10 logarithms). Western
blot verified these results from quantitative real-time
RT-PCR again. As shown in Fig. 7B–D, whereas AP-2a
suppression by RNA interference (RNAi) increased
KCTD10 (lane 3 in Fig. 7B,D), AP-2a overexpression
decreased KCTD10 (lane 2 in Fig. 7B,D). However,
overexpression of Sp1 increased KCTD 10 (Fig. 7C
lane 2, Fig. 7D lane 4), but silencing of Sp1 decreased
KCTD10 (Fig. 7C lane 3, Fig. 7D lane 5). So we
concluded that the expression of KCTD10 could be
regulated by Sp1 positively, but by AP-2a negatively.

FEBS Journal 276 (2009) 1114–1124 ª 2009 Hunan Normal University. Journal compilation ª 2009 FEBS

1117


SP1 and AP-2a regulate KCTD10

R. Liu et al.

Table 1. Oligonucleotides used in this study.
Name

Sequence(5¢- to -3¢)

Purpose

PLuc-KCTD10-R
P ()609 ⁄ +30)-F1
P ()343 ⁄ +30)-F2
P ()241 ⁄ +30)-F3
P ()203 ⁄ +30)-F4
P ()108 ⁄ +30)-F5
P ()13 ⁄ +30)-F6
P ()609 ⁄ )241)-R
AP-2a ⁄ chipF
AP-2a ⁄ chipR
Sp1 ⁄ chipF
Sp1 ⁄ chipR
AP-2a ⁄ si
Sp1 ⁄ si
KCTD10FP
KCTD10RP
Sp1FP
Sp1RP
AP-2aFP
AP-2aRP
b-actinFP
b-actinRP

CCAAGCTTCGGACTGAGAGAGGCAGGAA
GGGGTACCTGGAGCACACACGCCAGATC
GGGGTACCCGACGCACTACCGCCATCGT
GGGGTACCCTTCTTCGCCCGGGAAGGAA

GGGGTACCGCAGCTGAAATAGCGGAGGT
GGGGTACCAACGGACGGCTTAAGACGTT
GGGGTACCCCGGCTGGCGTGAGCTGGGT
CCAAGCTTTTCCTTCCCGGGCGAAGAAG
GGGGCGGAAGTGGGGTG
GAAAAGTCGGAGGACG
GGACTACCTGGAGTGATGCCTAA
CCCATCAACGGTCTGGAACT
GCUCCACCUCGAAGUACAATT
NNAGCGCUUCAUGAGGAGUGA
TTGCGGTTTAGGTACATCCA
TGTGGCTCTGTGGAACATTT
GGACTACCTGGAGTGATGCCTAA
CCCATCAACGGTCTGGAACT
CAACGTTACCCTGCTCACATCA
CAGGTCGGTGAACTCTTTGCA
GCGCGGCTACAGCTTCA
CTTAATGTCACGCACGATTTCC

Mutagenesis

ChIP
ChIP
ChIP
ChIP
RNAi
RNAi

Real-time PCR

Fig. 3. Function analysis of 5¢-flanking region of human KCTD10 promoter. Constructs containing sequentially deleted fragments of human
KCTD10 5¢-flanking region (from )609 to +30) were transfected into HeLa cells, luciferase activities were then measured 36 h after transfection. The region from )108 to +30 is essential for the promoter activity. Data (means ± SD) were presented as the percentage of the
control, P ()609 ⁄ +30). Vacant vector pTAL-luc only containing a basal TATA-like promoter was used as experimental control.

Discussion
Our data showed that KCTD10 contains neither a
canonical TATA box nor a CCAAT motif in the promoter region, but a cluster of CpG dinucleotides near
to the TSS. Although only a single TSS of human
KCTD10 was detected 63 bp upstream of the start
codon, and deletion mutagenesis demonstrated that the
region from )108 to +30 was indispensable for basal
promoter activity, we still could not exclude the exis1118

tence of other potential weak initial sites or other weak
promoter region which were often found in CpG
islands containing and TATA-free promoters [10–12].
Typically, GC-rich promoter regions lack TATA or
DPE core elements, and are frequently found to be
bound with transcription factor Sp1. Sp1 helps to
maintain the hypomethylation of CpG islands [13,14],
and interacts with some components of the basal transcription complex [15,16]; therefore it plays a critical
role in the assembly of the transcription start complex

FEBS Journal 276 (2009) 1114–1124 ª 2009 Hunan Normal University. Journal compilation ª 2009 FEBS


R. Liu et al.

SP1 and AP-2a regulate KCTD10

A

Fig. 4. Identification of potential binding
sites of transcription factors in the proximal
promoter region. (A) Site-directed mutation
analysis of c-myb.1, CACCC-box, Sp1.1 and
AP-2.2 sites in the proximal promoter region
of human KCTD10. Data (means ± SD)
were presented as the percentage versus
the wild-type control P ()203 ⁄ +30). The
cross (·) indicates the mutated sites.
SREBP-1, sterol regulatory element binding
protein-1. (B) Alignment of the 5¢-flanking
sequences upstream of the start codon of
human and mouse KCTD10.

B

Fig. 5. In vivo ChIP assays of Sp1.1 and AP-2.2 binding sites in the
proximal promoter region of human KCTD10. Mouse monoclonal
anti-Sp1 (lane 2), or rabbit polyclonal anti-(AP-2a) (lane 5), or nonimmune mouse and rabbit IgG (lanes 3, 6) were used in ChIP. The
DNA fragments derived from AP-2a-specific or Sp1-specific immunoprecipitations were amplified using primers flanking the binding
elements of Sp1 or AP-2a. A portion of the total input was used as
positive control (lanes 1, 4).

to selectively activate transcription [13,14,17,18].
In vivo ChIP assays confirmed that Sp1 could specifically recognize the GC-box element in the proximal
promoter region of KCTD10. Luciferase assays in
HeLa cells indicated that mutation of the Sp1-binding
site decreased the promoter activity by 46.9%,

whereas overexpression of Sp1 increased the promoter
activity by 58.3%. The expression of endogenous
KCTD10 was downregulated 1.9-fold by Sp1 silencing, and upregulated 12.0-fold by Sp1 overexpression.
Western blot results of KCTD10 were identical to
quantitative real-time RT-PCR. Sp1 is an essential

and positive regulator for the transcription of human
KCTD10.
The binding affinity and transcriptional specificity of
Sp1 can be altered by interaction with other cofactors
in the binding sites near to Sp1 recognition motif, such
as C ⁄ EBP [19], NF-jB [20], AP-1 [21] or AP-2 [22,23].
In particular, AP-2 binding sites have been found
in most Sp1-dependent promoters. AP-2 was a tissuespecific transcription factor, playing critical roles in the
regulation of gene expression during mammalian development, differentiation and carcinogenesis [24–27]. It
was reported that AP-2 could stimulate AP-2-dependent promoter activities in a tissue-specific manner
[28], but suppress Sp1-dependent housekeeping promoter activities [22,23,29]. In this study, we determined
the functional role of AP-2a in the regulation of
KCTD10. Mutation of the AP-2-binding site increased
KCTD10 promoter activity by 88.1%, whereas overexpression of AP-2a in HepG2 cells inhibited promoter
activity in a dose-dependent manner. AP-2a silencing
increased the expression of endogenous KCTD10 by
2.3-fold, whereas AP-2a overexpression decreased the
expression of endogenous KCTD10 by 5.6-fold. AP-2a
is a negative regulator for KCTD10.
Transcriptional interference, in which overexpression
of one transcription factor resulted in inhibition of
another, was often found in the regulation of eukaryotic gene expression. The repression of AP-2 on Sp1dependent transcriptional activity has been explained
in three models: (a) steric hindrance model – AP-2

FEBS Journal 276 (2009) 1114–1124 ª 2009 Hunan Normal University. Journal compilation ª 2009 FEBS

1119


SP1 and AP-2a regulate KCTD10

R. Liu et al.

Fig. 6. Effects of Sp1 and AP-2a expression level on the promoter
activity of human KCTD10. (A) The relative luciferase activities of P
()241 ⁄ +30) and Mt-Sp1.1 mutant were examined in pCMV–Sp1transfected HeLa cells. pCMV–myc was used to equilibrate the
plasmid quantity, and pCMV–LacZ was used as a transfection efficiency control. Data (means ± SD) were presented as the percentages of wild-type promoter control. (B) Constructs P ()609 ⁄ +30), P
()108 ⁄ +30), P ()13 ⁄ +30) and pTAL-luc control were transfected
into HeLa or HepG2 cells, respectively. Luciferase activities were
presented as the percentages of pTAL-luc control. (C) The relative
luciferase activities of P ()241 ⁄ +30) were inhibited by AP-2a overexpression in a dose-dependent manner in pCMV-AP-2a transfected HepG2 cells. pCMV–myc was used to equilibrate the plasmid
quantity, and pCMV–LacZ was used as a transfection efficiency
control.

A

B

C

represses Sp1-dependent promoter activity by interfering with the activation of transcription initiation by
Sp1 via specific steric interference machinery [29]; (b)
1120

interaction model – the interaction between Sp1 and
AP-2 affects formation of the Sp1–DNA complex [23];
(c) competing model – the AP-2 and Sp1 binding sites
overlap, and the AP-2 binding competes with the SP1
binding [22]. Our results reveal that the repressive
activity of AP-2a on KCTD10 transcription works
through inhibition of Sp1 activity. It has been proposed that the ratio of Sp1 to AP-2a might be responsible for the transcriptional state of Sp1-dependent
promoters; for example, Chen et al. have shown that
K3 keratin gene transcription was regulated by the
ratio of Sp-1 to AP-2 in differentiating rabbit corneal
epithelial cells [29]. In our case, even though there were
similar levels of Sp1 in both HeLa and HepG2 cells,
KCTD10 promoter activity is significantly lower in
HeLa cells than in HepG2 cells; this is probably due
to the AP-2 deficiency. When AP-2a was overexpressed
in HepG2 cells, KCTD10 promoter activity decreased
in a dose-dependent manner. However, AP-2a silencing increased KCTD10 transcription, whereas AP-2a
overexpression repressed KCTD10 transcription in
HeLa cells. It is possible that the relative ratio of Sp1
to AP-2a in a particular cell environment determined
the activation or repression of human KCTD10.
Our previous study demonstrated that mouse PDIP1
could also be regulated by Sp1 and AP-2, and the regulating sites were conserved in human, mouse and rat
[30], which implied that the PDIP1 gene family might
have similar regulating mechanisms in different species.
It has been reported that Sp1 binding to the promoters
were critical for the activation of TNF-a- stimulated
ADAM17 [31] and MCP-1 genes [32]; however, we did
not detect significant changes in KCTD10 expression
in HeLa cells after TNF-a induction (data not shown).

This suggested that the isolated 5¢-flanking region of
KCTD10 might not contain a TNF-a responsive
element. Although our results confirmed the roles of
two transcription factors, Sp1 and AP-2a, in the

FEBS Journal 276 (2009) 1114–1124 ª 2009 Hunan Normal University. Journal compilation ª 2009 FEBS


R. Liu et al.

A

SP1 and AP-2a regulate KCTD10

Experimental procedures
Mapping TSS by primer extension

B

C

D

Fig. 7. Endogenous expression change in human KCTD10 after
overexpressing, or suppressing Sp1 and AP-2a. (A) mRNA levels in
HeLa cells were measured using quantitative real-time RT-PCR,
and presented as base 10 logarithms. A1, mRNA levels of AP-2a
after overexpressing Ap-2a, or suppressing AP-2a by RNA interference; A2, mRNA levels of SP1 after overexpressing Sp1, or suppressing Sp1 by RNA interference; A3, mRNA levels of human
KCTD10 after overexpressing Ap-2a or Sp1, or suppressing AP-2a
or Sp1 by RNA interference. mRNA levels in HeLa cells transfected

with Myc vector were used as negative controls. In A3, folds of
human KCTD10 mRNA levels versus negative control were calculated. All experiments were run three times. (B–D) The protein
expressions of AP-2a, Sp1 and KCTD10 after overexpressing Ap-2a
or Sp1, or suppressing AP-2a or Sp1 by RNA interference in HeLa
cells were examined using western blotting. b-Actin was used as
internal control; Si-NC represents scramble siRNA oligos which can
not knockdown AP-2a, SP-1 and KCTD10.

regulation of KCTD10 expression, other regulatory
proteins involved in the transcription control of
KCTD10 still need to be further investigated.

The TSS of human KCTD10 was determined using primer
extension kit (Promega, Madison, WI, USA) according to
the manufacturer’s instruction. Briefly, an anti-sense primer
(5¢-TCTCCAAACCCGGACTGAGA-3¢) corresponding to
the position from +21 to +40 was labeled with [32P]ATP[cP]
using T4 polynucleotide kinase, and purified by precipitation
in 75% ethanol. The labeled primer was incubated with
25 lg of total RNA from HeLa cells at 62 °C for 30 min,
and then extended with AMV reverse transcriptase at 42 °C
for 1 h; sequencing reaction was further performed using
CycleReaderÔ DNA Sequencing kit (Invitrogen, San Diego,
CA, USA) with the same labeled primer. The products of
primer extension and sequencing reaction were resolved on
6% denaturing polyacrylamide sequencing gel.

Computer analysis of KCTD10 promoter
structures, cloning of the promoter fragment,
construction of deletion and point-mutated vectors

Promoter searching was performed using proscan (http://
www-bimas.cit.nih.gov) and promoterinspector (http://
www.genomatix.de) programs. Transcription factor binding
sites in the 5¢-flanking region of human KCTD10 were predicted using tfsearch ( />TFSEARCH.html). CpG island in this region was analyzed
() using the criteria of CG percentages over 50%, calculated or expected CpG ratio over 0.6
and a minimal length of 200 bases.
The human genomic chromosome 12 clone (NT_
009775.15) containing KCTD10 was identified by searching
the human genome using full-length KCTD10 cDNA
sequence. A 639 bp genome fragment upstream of the start
codon (ATG) and other five 5¢-deletion mutants were
obtained by PCR from human genomic DNA using the
3¢-end primer pLuc-KCTD10-R and the upstream 5¢-end
primers F1 to F6 (Table 1), and subcloned into the KpnI ⁄
HindIII sites of pTAL-Luc vector (Clontech, Mountain
View, CA, USA). A 3¢-deletion mutant was constructed
using the primers P ()241 ⁄ +30)-F3 and P ()609 ⁄ )241)-R.
Site-directed mutants derived from the P ()203 ⁄ +30)
construct were generated by overlapping extension PCR as
described previously [10]. In brief, in the first round, two
PCR were performed in parallel using construct P
()203 ⁄ +30) as a template: one with a wild-type forward
primer P ()203 ⁄ +30)-F4 and a reverse mutated primer, the
other with a forward mutated primer (complementary to
the reverse mutated primer) and a wild-type reverse primer
PLuc–KCTD10-R; in the second round, equal molar
mixture of the two PCR products was used as template,
P ()203 ⁄ +30)-F4 and pLuc–KCTD10-R were used as
primers. The final PCR products were similarly subcloned

FEBS Journal 276 (2009) 1114–1124 ª 2009 Hunan Normal University. Journal compilation ª 2009 FEBS

1121


SP1 and AP-2a regulate KCTD10

R. Liu et al.

into pTAL–luc vector. All constructs were confirmed by
DNA sequencing.

Luciferase reporter assay
HeLa and HepG2 cells were cultured in DMEM (Invitrogen) supplemented with 10% fetal bovine serum, 2 mm
l-glutamine, 100 mL)1 penicillin and 100 lgỈmL)1 streptomycin at 37 °C in a 5% CO2 incubator. When cells reach
90% confluence in 24-well plate, the culture medium was
replaced with serum- and antibiotic-free medium 3 h before
transfection; then 0.5 lg of luciferase reporter plasmids
were co-transfected with 0.3 lg of b-galactosidase expression vector pCMV-LacZ using LipofectamineÔ 2000
Reagent (Invitrogen); 6 h after transfection, cells were
washed, and fresh medium supplemented with serum and
antibiotics was put back again; cell lysates were collected
36 h after transfection and assayed for luciciferase activity
using the Luciferase Assay System (Promega).

Mammalian expression vectors of human Sp1
and AP-2a
The human Sp1 coding sequence was amplified from HeLa
first-strand cDNA using forward primer with EcoRI site
(underlined) and reverse primer with SalI site in 5¢-end (forward primer: 5¢-AGAATTCCTGCCACCATGAGCGAC

CA-3¢, reverse primer: 5¢-CGTCGACTG ATCTCAGAA
GCCATTTTGC-3¢). The PCR product was subcloned into
pCMV–HA (Clontech) at EcoRI ⁄ SalI sites. The human
AP-2a coding sequence was amplified from a human brain
cDNA library using the primer shown in Table 1, and subcloned into pCMV–HA at EcoRI ⁄ XhoI sites. Plasmids
pCMV–Sp1 and pCMV–AP-2a were confirmed by sequencing, and used to overexpress Sp1 and AP-2a in mammalian
cells. All expressed proteins were confirmed by SDS–PAGE
and western blotting.

ChIP assay
Sequence-specific DNA binding activities of Sp1 and AP-2a
to KCTD10 promoter were confirmed by in vivo ChIP
assays. EZ ChIPÔ Kit (Upstate, Temecula, CA, USA) was
used following the manufacturer’s instruction. Briefly, cells
were cross-linked for 10 min in 1.6% formaldehyde solution; for every 2 mg of protein extracts, 10 lg of mouse IgG
bound to protein A–Sepharose was used for 1 h pre-clear
with shaking at 4 °C; protein extracts were then immunoprecipitated with 2 lg of rabbit polyclonal anti-(AP-2a) or
2 lg of mouse monoclonal anti-Sp1 (Santa Cruz Biotech,
Santa Cruz, CA, USA) by rotating overnight at 4 °C. To
generate the pool of DNA fragments, protein extracts from
10 cm plate culture of HeLa cells were immunoprecipitated;
the immunocomplexes were then washed and eluted; after

1122

reversion of cross-link, the DNA fragments were purified
using spin columns. The PCR primers for ChIP assay
(shown in Table 1) were designed to flank the Sp1 and
AP-2a sites in KCTD10 promoter.

Real-time quantitative RT-PCR and western blot
Specific small interfering RNA (siRNA) against AP-2a
(5¢-GCUCCACCUCGAAGUACAATT-3¢) and Sp1 (5¢-NN
AGCGCUUCAUGAGGAGUGA-3¢) [12] were synthesized
from Invitrogen to suppress the endogenous expression of
Sp1 and AP-2a; pCMV–Sp1 and pCMV–AP-2a were used
to overexpress Sp1 and AP-2a. The cultured cells were
transfected using LipofectamineÔ 2000 according to the
manufacturer’s instruction; 24 h later after transfection,
total RNAs were extracted from 107 cells with TRIzol
reagent (Invitrogen); RT-PCR was carried out using a twostep strategy following the manufacturer’s manual (Promega). The primers for quantitative real-time RT-PCR of
KCTD10 were shown in Table 1, and b-actin was used for
internal normalization. In parallel, nuclear extracts were
prepared from the transfected cells; after 5 min heating and
separation on 12.5% denaturing polyacrylamide gel, proteins were transferred onto nitrocellulose membranes, and
then analyzed by western blot using rabbit polyclonal
serum against AP-2a or Sp1 (Santa Cruz Biotech).

Acknowledgements
This work was supported in part by the National Natural Science Foundation of China (No. 30771082,
30571005), the 973 project of Ministry of Science
and Technology of China (No. 2005CB522505,
2006CB943506), the Hunan Provincial Natural Science
Foundation of China (No. 08JJ3082), Project of Hunan
Science and Technology Commission: run foundation of
large-scale precision instrument and management of
experiment animals (2060599), Project of Changsha
Science and Technology Department (K0803107-21)
Changsha, the Cultivation Fund of the Key Scientific
and Technical Innovation Project, Ministry of Education of China (No. 705041), the Provincial Science &

Technology Department of Hunan (05FJ4016), and provincial Education Department of Hunan (No. 05C391,
07C563).

References
1 Zhou J, Ren K, Liu X, Xiong X, Hu X & Zhang J
(2005) A novel PDIP1-related protein, KCTD10, that
interacts with proliferating cell nuclear antigen and
DNA polymerase delta. Biochim Biophys Acta 1729,
200–203.

FEBS Journal 276 (2009) 1114–1124 ª 2009 Hunan Normal University. Journal compilation ª 2009 FEBS


R. Liu et al.

2 Tan CK, Castillo C, So AG & Downey KM (1986) An
auxiliary protein for DNA polymerase-delta from fetal
calf thymus. J Biol Chem 261, 12310–12316.
3 Prelich G, Tan CK, Kostura M, Mathews MB, So AG,
Downey KM & Stillman B (1987) Functional identity
of proliferating cell nuclear antigen and a DNA polymerase-delta auxiliary protein. Nature 326, 517–520.
4 Warbrick E (1998) PCNA binding through a conserved
motif. Bioessays 20, 195–199.
5 He H, Tan CK, Downey KM & So AG (2001) A tumor
necrosis factor alpha- and interleukin 6-inducible protein that interacts with the small subunit of DNA
polymerase delta and proliferating cell nuclear antigen.
Proc Natl Acad Sci USA 98, 11979–11984.
6 Zhou J, Hu X, Xiong X, Liu X, Liu Y, Ren K, Jiang T,
Hu X & Zhang J (2005) Cloning of two rat PDIP1-related
genes and their interactions with proliferating cell nuclear

antigen. J Exp Zool A Comp Exp Biol 303, 227–240.
7 Fiers W, Beyaert R, Boone E, Cornelis S, Declercq W,
Decoster E, Denecker G, Depuydt B, De Valck D,
De Wilde G et al. (1995) TNF-induced intracellular signaling leading to gene induction or to cytotoxicity by
necrosis or by apoptosis. J Inflamm 47, 67–75.
8 Fausto N (2000) Liver regeneration. J Hepatol 32, 19–31.
9 Link CD, Taft A, Kapulkin V, Duke K, Kim S, Fei Q,
Wood DE & Sahagan BG (2003) Gene expression analysis in a transgenic Caenorhabditis elegans Alzheimer’s
disease model. Neurobiol Aging 24, 397–413.
10 Arman M, Calvo J, Trojanowska ME, Cockerill PN,
Santana M, Lopez-Cabrera M, Vives J & Lozano F
(2004) Transcriptional regulation of human CD5:
important role of Ets transcription factors in CD5
expression in T cells. J Immunol 172, 7519–7529.
11 Lavie L, Maldener E, Brouha B, Meese EU & Mayer J
(2004) The human L1 promoter: variable transcription
initiation sites and a major impact of upstream flanking
sequence on promoter activity. Genome Res 14, 2253–
2260.
12 Gum JR Jr, Hicks JW, Crawley SC, Dahl CM, Yang
SC, Roberton AM & Kim YS (2003) Initiation of transcription of the MUC3A human intestinal mucin from
a TATA-less promoter and comparison with the
MUC3B amino terminus. J Biol Chem 278, 49600–
49609.
13 Brandeis M, Frank D, Keshet I, Siegfried Z, Mendelsohn M, Nemes A, Temper V, Razin A & Cedar H
(1994) Sp1 elements protect a CpG island from de novo
methylation. Nature 371, 435–438.
14 Macleod D, Charlton J, Mullins J & Bird AP (1994)
Sp1 sites in the mouse aprt gene promoter are required
to prevent methylation of the CpG island. Genes Dev 8,

2282–2292.
15 Gill G, Pascal E, Tseng ZH & Tjian R (1994) A glutamine-rich hydrophobic patch in transcription factor Sp1
contacts the dTAFII110 component of the Drosophila

SP1 and AP-2a regulate KCTD10

16

17

18

19

20

21

22

23

24

25

26

27

28

TFIID complex and mediates transcriptional activation.
Proc Natl Acad Sci USA 91, 192–196.
Chiang CM & Roeder RG (1995) Cloning of an intrinsic human TFIID subunit that interacts with multiple
transcriptional activators. Science 267, 531–536.
Baker DL, Dave V, Reed T & Periasamy M (1996)
Multiple Sp1 binding sites in the cardiac ⁄ slow twitch
muscle sarcoplasmic reticulum Ca2+-ATPase gene promoter are required for expression in Sol8 muscle cells.
J Biol Chem 271, 5921–5928.
Cohen HT, Bossone SA, Zhu G, McDonald GA &
Sukhatme VP (1997) Sp1 is a critical regulator of the
Wilms’ tumor-1 gene. J Biol Chem 272, 2901–2913.
Lopez-Rodriguez C, Botella L & Corbi AL (1997)
CCAAT-enhancer-binding proteins (C ⁄ EBP) regulate
the tissue specific activity of the CD11c integrin gene
promoter through functional interactions with Sp1
proteins. J Biol Chem 272, 29120–29126.
Perkins ND, Edwards NL, Duckett CS, Agranoff AB,
Schmid RM & Nabel GJ (1993) A cooperative interaction between NF-kappa B and Sp1 is required for
HIV-1 enhancer activation. EMBO J 12, 3551–3558.
Lee W, Haslinger A, Karin M & Tjian R (1987) Activation of transcription by two factors that bind promoter
and enhancer sequences of the human metallothionein
gene and SV40. Nature 325, 368–372.
Getman DK, Mutero A, Inoue K & Taylor P (1995)
Transcription factor repression and activation of the
human acetylcholinesterase gene. J Biol Chem 270,
23511–23519.
Xu Y, Porntadavity S & St Clair DK (2002) Transcriptional regulation of the human manganese superoxide
dismutase gene: the role of specificity protein 1 (Sp1) and

activating protein-2 (AP-2). Biochem J 362, 401–412.
Zhang J, Hagopian-Donaldson S, Serbedzija G, Elsemore J, Plehn-Dujowich D, McMahon AP, Flavell RA
& Williams T (1996) Neural tube, skeletal and body
wall defects in mice lacking transcription factor AP-2.
Nature 381, 238–241.
Eckert D, Buhl S, Weber S, Jager R & Schorle H
(2005) The AP-2 family of transcription factors. Genome
Biol 6, 246.
Turner BC, Zhang J, Gumbs AA, Maher MG, Kaplan
L, Carter D, Glazer PM, Hurst HC, Haffty BG & Williams T (1998) Expression of AP-2 transcription factors
in human breast cancer correlates with the regulation of
multiple growth factor signalling pathways. Cancer Res
58, 5466–5472.
Pellikainen JM & Kosma VM (2007) Activator protein2 in carcinogenesis with a special reference to breast
cancer – a mini review. Int J Cancer 120, 2061–2067.
Benson LQ, Coon MR, Krueger LM, Han GC, Sarnaik
AA & Wechsler DS (1999) Expression of MXI1, a Myc
antagonist, is regulated by Sp1 and AP2. J Biol Chem
274, 28794–28802.

FEBS Journal 276 (2009) 1114–1124 ª 2009 Hunan Normal University. Journal compilation ª 2009 FEBS

1123


SP1 and AP-2a regulate KCTD10

R. Liu et al.

29 Chen TT, Wu RL, Castro-Munozledo F & Sun TT

(1997) Regulation of K3 keratin gene transcription by
Sp1 and AP-2 in differentiating rabbit corneal epithelial
cells. Mol Cell Biol 17, 3056–3064.
30 Zhou J, Fan C, Zhong Y, Liu Y, Liu M, Zhou A, Ren
K & Zhang J (2005) Genomic organization, promoter
characterization and roles of Sp1 and AP-2 in the basal
transcription of mouse PDIP1 gene. FEBS Lett 579,
1715–1722.

1124

31 Bzowska M, Jura N, Lassak A, Black RA & Bereta J
(2004) Tumour necrosis factor-alpha stimulates expression of TNF-alpha converting enzyme in endothelial
cells. Eur J Biochem 271, 2808–2820.
32 Ping D, Boekhoudt G, Zhang F, Morris A, Philipsen
S, Warren ST & Boss J (2000) Sp1 binding is critical
for promoter assembly and activation of the MCP-1
gene by tumor necrosis factor. J Biol Chem 275,
1708–1714.

FEBS Journal 276 (2009) 1114–1124 ª 2009 Hunan Normal University. Journal compilation ª 2009 FEBS