Báo cáo khoa học: Identification and characterization of four novel peptide motifs that recognize distinct regions of the transcription factor CP2 doc

Identification and characterization of four novel peptide
motifs that recognize distinct regions of the transcription
factor CP2
Ho Chul Kang
1
, Bo Mee Chung
1
, Ji Hyung Chae
1
, Sung-Il Yang
2
, Chan Gil Kim
3
and
Chul Geun Kim
1
1 Department of Life Science, Hanyang University, Korea
2 Department of Pharmacology, Konkuk University, Korea
3 Department of Biotechnology, Konkuk University, Korea
CP2 was discovered initially in mouse as a transcrip-
tion factor that binds to and stimulates transcription
from the a-globin promoter [1]. Initially, CP2 was
known to be a homologue of Drosophila Grainyhead
(Grh, also known as NTF-1 or Elf-1) [2–9]. However,
recent identification of the mammalian and Drosophila
homologue genes of CP2 and Grh revealed that CP2
and Grh belong to separate phylogenetic groups [10–
13]. CP2 has also been found in other organisms as
diverse as Drosophila, chicken and human, comprising
a highly conserved family. CP2 is named differently
depending upon species: for instance, dCP2 [10,12] in

Drosophila, cCP2 in chicken [14], and LBP-1c ⁄ LSF in
human [15]. Six isoforms (LBP-1a, -1b, -1c, -1d, -9 and
-32) of human CP2 proteins have been identified [15–
17], whereas four CP2 isoforms (CP2a ⁄ NF2d9, CP2b,
CP2c ⁄ CP2 and CRTR-1) have been reported in mice
[10,18,19].
Keywords
CP2; gene regulation; peptide motif; phage
display; protein–protein interaction;
transcription factor
Correspondence
C. G. Kim, Department of Life Science,
Hanyang University, Haengdangdong 17,
Sungdong-gu, Seoul 133-791, Korea
Fax: +82 2 2296 5996
Tel: +82 2 2290 0957
E-mail:
(Received 17 December 2004, revised 6
January 2005, accepted 10 January 2005)
doi:10.1111/j.1742-4658.2005.04564.x
Although ubiquitously expressed, the transcriptional factor CP2 also exhib-
its some tissue- or stage-specific activation toward certain genes such as
globin in red blood cells and interleukin-4 in T helper cells. Because this
specificity may be achieved by interaction with other proteins, we screened
a peptide display library and identified four consensus motifs in numerous
CP2-binding peptides: HXPR, PHL, ASR and PXHXH. Protein-database
searching revealed that RE-1 silencing factor (REST), Yin-Yang1 (YY1)
and five other proteins have one or two of these CP2-binding motifs.
Glutathione S-transferase pull-down and coimmunoprecipitation assays
showed that two HXPR motif-containing proteins REST and YY1 indeed

were able to bind CP2. Importantly, this binding to CP2 was almost abol-
ished when a double amino acid substitution was made on the HXPR
sequence of REST and YY1 proteins. The suppressing effect of YY1 on
CP2’s transcriptional activity was lost by this point mutation on the HXPR
sequence of YY1 and reduced by an HXPR-containing peptide, further
supporting the interaction between CP2 and YY1 via the HXPR sequence.
Mapping the sites on CP2 for interaction with the four distinct CP2-bind-
ing motifs revealed at least three different regions on CP2. This suggests
that CP2 recognizes several distinct binding motifs by virtue of employing
different regions, thus being able to interact with and regulate many cellu-
lar partners.
Abbreviations
APP, b-amyloid precursor protein; BSA, bovine serum albumin; CKII-a, casein kinase II-a; EMSA, electrophoretic mobility shift assay; GST,
glutathione S-transferase; HDAC1, histone deacetylase 1; HRP, horseradish peroxidase; LTR, long-terminal repeat; NLS, nuclear localization
signal; PEG, polyethylene glycol; PKC-d, protein kinase C-d; REST, RE-1 silencing factor; PIAS1, protein inhibitor of activated STAT1; TFIIE-a,
a subunit of transcription factor IIE; TS, thymidylate synthase; YY1, Yin-Yang1.
FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS 1265
Numerous data point to the transcriptional activa-
tion of a range of genes by CP2, leading to functional
diversity of CP2 in various species. In chicken, cCP2
regulates the lens-specific expression of the aA-crystal-
line gene [14]. In mammals, CP2 contributes to the
preferential recruitment of the b-globin locus control
region to the c-globin promoter during fetal erythro-
poiesis [20–25]. CP2 participates in cell-cycle regulation
by binding to and thus modulating transcription of the
thymidylate synthase (TS) promoter in growth-stimula-
ted human cells of late G
1
phase [26–28]. CP2 is also

implicated in genetic diseases such as Alzheimer’s dis-
ease [29,30]. In addition, CP2 has been shown to be
able to modulate transcription from some viral pro-
moters; it stimulates transcription from the viral SV40
major late promoter [31], but represses transcription
from HIV-1 long-terminal repeat (LTR) [16].
It has become apparent that these various functions
of CP2 are mediated by its interaction with different
tissue- or species-specific molecules. Lens-specific tran-
scription of aA-crystalline appears to involve cooper-
ation of cCP2 with a putative lens-specific factor
[14]. Binding of CP2 to the stage selector element of
the proximal c-globin gene promoter requires the
formation of a stage selector protein [24] that is a
heteromeric complex between CP2 and NF-E4, a fetal
erythroid-specific partner protein [25]. Because Fe65 is
a ligand of the Alzheimer’s b-amyloid precursor pro-
tein (APP) [32,33] and the aberrations of CP2 and ⁄ or
APP in neuronal cells promote neuronal apoptosis in
the Alzheimer’s disease brain [34,35], CP2 appears to
be linked with Fe65-mediated Alzheimer’s disease.
Repression of HIV-1 LTR transcription by CP2 is
accomplished via the binding of CP2 to another tran-
scription factor, Yin-Yang1 (YY1), and the resultant
recruitment of histone deacetylase 1 (HDAC1) [36,37].
Despite the extensive literature, the discovery of the
molecules that bind to CP2 and account for its diverse
functions is far from being complete. As a strategy to
reveal these molecules, we screened a phage display
library that provides a pool of up to 10

9
random
peptides. We were able to isolate a number of CP2-
interacting peptides and discovered that they contained
some new motifs that play an important role in the
interaction between CP2 and its binding proteins.
Results
Isolation of CP2-binding phage clones through
a phage display library screening
In an attempt to identify targets with which CP2 could
potentially interact, we screened a highly complex M13
phage library that contains  1.9 · 10
9
different
recombinant phage clones. Each of these phage dis-
plays a random 12-mer peptide on its coat, by virtue
of its capsid pVIII protein being tethered to the pep-
tide by a (Gly
4
Ser)
3
peptide linker. Each peptide is
composed of 12 amino acids randomly encoded by the
NNK codon (N ¼ A, C, G, T; K ¼ G, T). To enhance
identification of CP2-interacting peptides, a full-length
CP2 protein was used as bait for this screening. Sev-
eral rounds of amplified selection were undertaken on
phage that exhibited the ability to bind to GST–CP2.
Following each round of selection, the titer of
GST–CP2-bound phage (measured as pfu) increased

consecutively (data not shown), indicating successful
biopanning. A number of phage clones was obtained
from this screening. Because GST–CP2 fusion protein
was used in this screening (Fig. 1A), we eliminated the
phage that may have been bound to GST rather than
CP2. To this end, the abilities of each phage to bind
to GST–CP2 and GST were compared in ELISA
assays as described in Experimental procedures. As
shown in Fig. 1B,C, some phage (e.g. clones 01, 04,
09, 11 and 12) bound to GST–CP2 no better than to
GST. However, many phage clones, such as 05, 08, 13,
21 and 31, exhibited stronger binding to GST–CP2,
indicating that they bound specifically to CP2.
Identification of CP2-binding motifs
Approximately 100 clones showed highly specific bind-
ing to CP2 and were subjected to DNA sequencing for
their expressed peptides. Only two clones were found
three times; the remaining clones were diverse in terms
of their sequences. Searching the PIR-PSD data-
base ( />html) revealed that none of these sequences exactly
matched those of any known proteins. Interestingly,
however, 15 sequences could be categorized into four
classes with some common motifs using the clustal w
multiple sequence alignment program (http://clustalw.
genome.jp); the remainder did not show any common
motif. The deduced amino acid sequences of those 15
peptides are shown in Table 1. Class I mostly contains
His-X-Pro-Arg (HXPR) amino acids with minor con-
served variations between aromatic and imidazole
groups His (H), Tyr (Y) and Trp (W). X, indicating

any amino acid, was frequently found to be Pro or
His. The second motif, found in several other phage
clones, is Pro-His-Leu (PHL), usually flanked by Leu,
Ala, or Ser. The third and fourth classes exhibit an
Ala-Ser-Arg (ASR) and Pro-X-His-X-His (PXHXH)
motif, respectively. These data suggest that CP2 can
interact with a variety of peptides, each of which
Identification of CP2-binding peptide motifs H. C. Kang et al.
1266 FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS
possesses at least one of the four distinct CP2-binding
motifs.
Despite this carefully controlled screening proce-
dure we remained concerned about the potential
existence of false positives in the selected set of pep-
tides, which might originate from the highly complex
phage sequences. Therefore, as a second approach to
confirm the bona fide association of CP2 with the
peptides presented in Table 1, we performed in vitro
binding assays using purified peptides instead of
phage. Six peptides were chosen for this experiment:
five representing one of the four putative CP2-bind-
ing motifs and one internal control peptide which
had a high binding affinity to GST rather than CP2.
To facilitate expression and purification in Escheri-
chia coli, peptides were fused to the C-terminus of
GST (Fig. 2A). Whole-cell extracts were prepared
from cells expressing HA-tagged CP2, and incubated
with each GST-fused peptide. Complex formation
between the GST–peptide and HA–CP2 was deter-
mined by GST pull-down and immunoblot analyses.

Although control peptide was not able to bind to
CP2, all the other peptides containing the putative
CP2-binding motifs consistently showed a strong
interaction with CP2 (Fig. 2B). Taken together, these
results indicate that the phage clones presented in
Table 1 genuinely bind to CP2 through their coat-
displayed exogenous peptides with some common
motifs.
AB C
Fig. 1. Isolation of phage expressing CP2-binding peptides from a phage display library. (A) Coomassie Brilliant Blue-stained gel of GST–CP2
used as bait in the screening. GST–CP2 was purified by glutathione–agarose affinity chromatography as described in Experimental procedures.
(B) Abilities of some phage clones isolated from a random phage display library to bind to GST-CP2. Each phage is designated by a number lis-
ted at the bottom of the graph. Considerable numbers of phage clones, including 03, 05, 08, 10, 13, 17, 21, 31 and 54, show specific binding
to GST–CP2, whereas others such as 01, 04, 09, 11 and 12 bind equally to GST–CP2 and GST. In each experiment 1.23 · 10
12
phage were
used and all assays were performed in triplicate. Specific binding to CP2 was obtained by subtracting the absorbance value of background
GST binding from that of GST–CP2 binding. The results are presented as mean ± standard error. (C) CP2-binding activity is proportionate to
the number of phage. The extent of binding to GST–CP2 or GST was determined by measuring absorbance at 405 nm from ELISA.
Table 1. Deduced amino acid sequences of the 12-mer peptides in
phage clones that showed highly specific binding to CP2. Standard
single letter amino acid abbreviations are used. Conserved amino
acids are indicated in bold.
Classes
Selected
clone number Amino acid sequences
IA 13 HKSHLHFH P PRP
26 PPHKH H PRQPAM
32 H V PRMHKHNLQM
43 H N PRAESSSHLR

50 H P PRGHLVPTTP
53 HHFH V PRSNOSP
54 HQRHH H PRKSPE
61 HKPHMH G PRPWA
74 ILGAH H PRPIKP
81 HTKEFH P PRGIW
IB 05 HERRESNY P QRP
69 AHRSRW S PRPSY
II 21 HKFHQHRLPHLA
24 HKFHQHLAPHLA
29 HSPHLSHRHLLR
35 HFKHHKSLPHLG
64 S PFLSHRHYVPH
III 03 HITYSYVASRPS
08 HNMHKHSASRIH
49 HTPSQHHASRVT
18 YDLWWNSPFSAS
IV 20 W P H H H H TRLSTV
31 VWKHP H H H H YKR
33 W P Y H W H PAMKKN
99 GHTHKP Q H F H RS
H. C. Kang et al. Identification of CP2-binding peptide motifs
FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS 1267
Mapping the sites of interaction between CP2
and each CP2-binding motif
Because four CP2-binding motifs were identified as
above, we were interested in whether these distinct
motifs recognize the same region in CP2. CP2 is com-
posed of 502 amino acids, whose structure can be
divided into six regions: N-terminal transcriptional

activation (amino acids 1–63), N-terminal Elf-1
(amino acids 63–244), the basic (amino acids 244–
250), SPXX (amino acids 250–403), Q ⁄ P (amino acids
403–413) and C-terminal acidic (amino acids 413–502)
regions [17,38]. To date, the functional role of each
region has not been fully studied, although the carb-
oxyl-flanking region of SPXX has recently been
shown to be involved in interactions with YY1 [36].
We mapped the binding site of each CP2-binding
peptide on CP2. Several N- and ⁄ or C-terminally dele-
ted CP2 mutants were generated (Fig. 3A) and their
abilities to bind either with GST-tagged YY1
(Fig. 3B) or with representative phage clones each dis-
playing one of the CP2-binding motifs (Fig. 3C) were
examined. GST pull-down assays showed that YY1
bound only to those CP2 proteins that contained a
region corresponding to amino acids 306–396. This
result was anticipated, as this region of CP2 has been
shown to be involved in the interaction with YY1
[36]. This putative CP2-binding region was then fur-
ther confirmed in ELISA experiments in which phage
clone 13 (containing an HXPR motif) and other CP2
mutant constructs were employed (Fig. 3C). This
phage was able to bind only to those CP2 mutant
proteins with amino acids 306–396. Interestingly, this
region of CP2 was found to be involved in interac-
tions with not only phage clone 05, which has
YXQR, a subset of HXPR motif, but also clone 08,
which possesses an ASR motif. These results indicate
that two CP2-binding motifs, HXPR (found in phage

clones 13 and 05) and ASR (found in phage clone
08) recognize a region of amino acids 306–396 that
corresponds to the carboxyl-flanking region of the
SPXX region. However, phage clones 21 and 31
were found to bind to regions other than SPXX.
Clone 21, which represents a PHL motif, bound to
the C-terminal part of the Elf-1 domain (amino acids
134–243). The Elf-1 domain was also found to pro-
vide an interacting site for clone 31, a representative
of the PXHXH motif. For interaction with clone 31,
however, the N-terminal part of Elf-1 (amino acids
67–134) was required.
A
B
Fig. 2. Purified peptides possessing the proposed CP2-binding motifs bind to CP2 in vitro. (A) Amino acid sequences of five representative
GST-tagged peptides that contain CP2 (05, 08, 13, 21 and 31) and one GST-binding peptide (04). Residues underlined in black correspond to
the 12-mer peptide sequences. (B) In vitro binding between HA–CP2 and the purified peptides. Whole-cell extracts from 293T cells expres-
sing HA–CP2 were mixed with each of the purified GST-tagged peptides and the complex formation was analyzed by GST pull-down and
immunoblotting against anti-HA serum. Coomassie Brilliant Blue staining and anti-HA blotting of the input materials ascertain the equivalent
amounts of extracts and peptides between each pull-down samples being employed. Arrow indicates the position of HA–CP2.
Identification of CP2-binding peptide motifs H. C. Kang et al.
1268 FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS
Recognition of proteins containing CP2-binding
motifs and verification of the CP2-binding ability
of REST and YY1 that contain the HXPR motif
As stated above, none of the 12-mer peptide sequences
exactly matched those of any known proteins. Having
discovered common 3–5 amino acid motifs on these
peptides, we decided to see if there were any proteins
in the database that had these motifs. We searched

iProClass (PIR ± Swiss-Prot ⁄ TrEMBL) and the PIR-
PSD database using the PIR pattern ⁄ peptide match
search tool. As a result, several proteins were recog-
nized, including REST, YY1, protein inhibitor of acti-
vated STAT1 (PIAS1), protein kinase C-d (PKC-d),
A
B
C
EIf-1 BD
SPXX
Q/P
Net Acidic
Fig. 3. Identification of the domains on CP2, with which CP2-binding peptides interact. (A) Schematic representation of full-length CP2 com-
posed of six regions and various N- and ⁄ or C-terminally deleted CP2 mutants. Regions recognized by CP2-binding peptides were determined
by the experiments shown in (B) and (C). (B) GST pull-down assays showing the binding of YY1 to various CP2 deletion mutants. Lysates
(500 lg of protein) from cells over-expressing HA–CP2 deletion mutant proteins were incubated for 2 h with GST–YY1 immobilized on gluta-
thione–Sepharose beads. Bound proteins were eluted with SDS sample buffer, resolved by SDS ⁄ PAGE, and analyzed by immunoblotting
with anti-HA and anti-GST sera. None of the CP2 deletion mutant proteins showed detectable level of binding to GST itself (GST panel).
Western blot using an anti-HA serum shows that similar amounts of CP2 deletion mutants were used in each lane (INPUT panel). (C) Binding
of various CP2 deletion mutants to phage clones in which each displays a 12-mer peptide containing a representative CP2-binding motif as
in Fig. 1. Binding of each phage clone to the GST-tagged CP2 proteins was determined by ELISA as described in Experimental procedures.
Absorbance at 405 nm was used to determine extent of binding. Shown are the specific bindings to CP2, obtained after subtracting the non-
specific bindings to GST. Similar amounts of CP2 deletion mutants were used in each assay, as confirmed by Western blot using an anti-
GST serum (subset figure in a box).
H. C. Kang et al. Identification of CP2-binding peptide motifs
FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS 1269
casein kinase II-a (CKII-a), ETS-1 and the a subunit
of transcription factor IIE (TFIIE-a) (Table 2). REST
and YY1 have one HXPR motif and PIAS1 has a
PHL motif, whereas PKC-d, CKII-a and ETS-1 have

two motifs. The two protein kinases, PKC-d and
CKII-a, have both PHL and ASR motifs, whereas
ETS-1 has PHL and HXPR. TFIIE-a contains a
PXHXH motif. Among these proteins, YY1 is the only
one that has been shown to interact with CP2, whereas
the remaining proteins are novel in that their ability to
interact with CP2 has not been reported, although
their physiological functions and roles are well charac-
terized.
Of the proteins that have CP2-binding motifs, we
further investigated the CP2 binding of the two
HXPR-motif-containing proteins, REST and YYI.
Both REST and YY1 function as a transcription
modulator by recruiting and utilizing HDAC family
proteins [36,37,39]. REST acts largely as a repressor
for expression of neuronal genes in non-neuronal tis-
sues [51,52]. YY1 regulates the transcription of a vari-
ety of genes [53,54], but its activity could be activating
or inhibiting, depending upon the promoter context
and cellular environment [55,56]. It is known that
HDAC1 recruitment requires prior interaction of YY1
with CP2 [36,37], and the first zinc finger domain is
crucial for this interaction [36,57,58]. Importantly, we
found that the HXPR motif is located within this
region. In REST, a HXPR motif is found to reside in
the DNA-binding domain clustered with eight zinc
fingers [40]. We investigated whether REST as well as
YY1 can associate with CP2. To this end, 293T cells
were cotransfected with EGFP–CP2 and HA–REST
(Fig. 4A) or CP2 and HA–YY1 (Fig. 4B) and the cell

lysates were immunoprecipitated with appropriate anti-
bodies. A significant amount of REST was shown to
be present in anti-EGFP immunoprecipitates and con-
versely, CP2 was found in anti-HA immunoprecipitates
(Fig. 4A). Likewise, YY1 and CP2 could be coimmu-
noprecipitated with the counterpart anti-HA or anti-
CP2 sera (Fig. 4B). These results indicate that both
REST and YY1 bind to CP2 in vivo.
The HXPR motif is important in the interaction
of YY1 and REST with CP2
To verify that YY1 and REST interact with CP2 via
the HXPR motif, we generated mutants for each pro-
tein that had a double (PRfiAA) or single (HfiA)
amino acid substitution on HXPR and determined
their binding to CP2. We performed GST pull-down
experiments in which GST-tagged YY1 proteins were
incubated with 293T whole-cell extracts expressing
HA–CP2 (Fig. 5A). HA–CP2 was retained by GST–
YY1 but not by GST itself (GST-pull down panel),
indicating that YY1 binds to CP2 in vitro. Under con-
ditions in which equivalent amounts of wild-type YY1
and the two YY1 mutants were expressed (Input
panel), the single point mutant of YY1 possessed
Table 2. Proteins containing the putative CP2-binding motifs.
Putative CP2-binding proteins Conserved motifs Functions
REST HXPR Transcriptional repressor of neuronal genes in non-neuronal tissues [39,40]
Yin-Yang1 HXPR Transcriptional repressor and activator [36,41]
PIAS1 PHL Transcriptional repressor and activator, SUMO-E3 ligase [42–44]
PKC-d PHL and ASR Commitment to terminal erythroid differentiation [45,46]
CKII-a PHL and ASR Commitment to terminal erythroid differentiation [47,48]

ETS-1 PHL and HXPR Erythroid-specific gene activation [49]
TFIIE, a subunit PXHXH Transcription factor IIE, a subunit (TFIIE a) [50]
A
B
Fig. 4. REST and YY1 strongly interact with CP2 in vivo. (A) Coim-
munoprecipitation assays. 293T cells were cotransfected with
EGFP–CP2 and HA–REST or HA–REST(P272A ⁄ R273A) plasmids
and total cell lysates were immunoprecipitated against either anti-
HA or anti-EGFP sera. Bound proteins were eluted with SDS sam-
ple buffer, resolved by electrophoresis on SDS)8% polyacrylamide
gels. The presence of coimmunoprecipitated proteins was detected
by probing with the appropriate antibodies. (B) Coimmunoprecipita-
tion of CP2 and HA-YY1 was similarly assayed by using anti-HA
and -CP2 antibodies. Input, one-tenth of the cell extracts used for
IP; IP, immunoprecipitation.
Identification of CP2-binding peptide motifs H. C. Kang et al.
1270 FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS
CP2-binding activity that was approximately one third
that of wild-type YY1. The double-point mutant of
YY1 did not show any detectable level of CP2 binding.
We also demonstrated that the REST double-point
mutant (PRfiAA) exhibited much less ability to bind
to CP2 in a coimmunoprecipitation assay (Fig. 4A,
right panel). These results clearly indicate that the
binding of YY1 and REST to CP2 are mediated by
the HXPR motif.
That HXPR is a motif for binding of YY1 to CP2
was further confirmed by competition with an HXPR
motif-containing peptide (i.e. peptide 13) and by YY1
point mutants. The extent of the CP2–YY1 interaction

was indirectly measured by assaying the transcriptional
activity of CP2, because YY1 decreases the transcrip-
tional activity of CP2 by forming a complex with it
[36]. A luciferase assay was employed using a synthetic
promoter that contains four copies of a CP2-binding
element in the mouse a-globin promoter [21]. Cells
were transfected with CP2, YY1, YY1 double- or
single-point mutant, and peptide 13 plasmids in var-
ious combinations and the CP2 promoter-driven luci-
ferase activity was measured (Fig. 5B). CP2 stimulated
luciferase activity quite well, whereas YY1 and peptide
A
B
Fig. 5. A HXPR motif mediates the interatcion of YY1 with CP2. (A) GST pull-down assays. Whole-cell extracts (500 lg of protein) containing
HA–CP2 protein were incubated with glutathione–Sepharose bead-immobilized GST, GST–YY1, GST–YY1(P322A ⁄ R323A) or GST–
YY1(H320A) proteins for 2 h. Bound proteins were eluted with SDS sample buffer, resolved by electrophoresis on SDS) 8% polyacrylamide
gels, and analyzed by immunoblotting with anti-HA or anti-GST sera. (B) Effect of YY1 point mutants and a competitor peptide, on
CP2-driven luciferase activity. A pGL3-based luciferase plasmid was engineered to contain a region of the a-globin promoter with CP2-bind-
ing elements. 293T cells were cotransfected with this reporter and one, some or all of the following plasmids as indicated at the bottom of
the plot: HA–CP2, HA–YY1, HA–YY1(P322A ⁄ R323A), HA–YY1(H320A) and FLAG–peptide 13 plasmids. To eliminate DNA concentration-
dependent variation of transfection efficiency and reporter expression between transfections, total amount of plasmid DNA was adjusted to
2 lg by supplementing the pcDNA3–FLAG vector to test vectors. All other conditions are described in detail in Experimental procedures.
Experiments were performed three times and the results were analyzed statistically using the Student’s t-test. P values were obtained by
comparison of bar 5 with bar 2 (*P £ 0.001), bar 5 with bar 6 (**P £ 0.01) and bar 14 with bar 16 (**P £ 0.01).
H. C. Kang et al. Identification of CP2-binding peptide motifs
FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS 1271
13 alone did not (Fig. 5B; lane 1 vs. lanes 2, 9 and 10),
showing that specific transcription is driven by CP2.
YY1 was able to suppress this CP2-driven luciferase
activity in a dose-dependent manner (Fig. 5B; lane 2

vs. lanes 3–5). When a plasmid encoding peptide 13
was cotransfected, however, the suppressing effect of
YY1 on CP2’s transcriptional activity was reduced in
proportion to the amount of peptide 13 (Fig. 5B; lane
5 vs. lane 6 and lane 4 vs. lanes 7–8). Furthermore, we
found that the YY1 double-point mutation on HXPR
could not suppress CP2’s transcriptional activity
(Fig. 5B; lane 5 vs. lane 13) but the YY1 single-point
mutation showed significant suppression of CP2 activ-
ity although it was not completely inhibited (Fig. 5B;
lane 5 vs. lane 16). Taken together, these results indi-
cate that the HXPR motif can confer on YY1 or
REST the ability to bind to CP2 and the physiological
role of CP2 is mediated by the HXPR motif.
Discussion
Since CP2 was first isolated almost 15 years ago, much
effort has been put into finding its functional roles. As
a result, it is now known that CP2 is not only a ubi-
quitous transcription factor expressed in most tissues,
but is also involved in tissue- or stage-specific tran-
scription activation of some genes. It remains, how-
ever, unclear how CP2’s activity is regulated and how
CP2 exerts tissue- and stage-specific functions. This
regulation and specificity can certainly be achieved by
various mechanisms, one of which would be inter-
action with some other proteins, as found for a
number of cellular examples including NF-jB with
IjB. An M13 phage display library that provides up to
10
9

peptides composed of 12 random amino acids was
screened in this study. Approximately 100 phage clones
that express CP2-binding peptides on their coats were
isolated, and specific binding of some of these clones
to CP2 was verified using purified 12-mer peptides.
Sequence analysis showed that some clones could be
grouped into four classes each having a unique consen-
sus sequence of 3–5 amino acids, whereas the remain-
der did not display a consensus sequence (Table 1).
The consensus motifs are as follows: HXPR for Class
I clones, PHL for Class II, ASR for Class III and
PXHXH for Class IV. Class I clones can have some
substitutions at positions 1 and 3; Y or W is found at
position 1 in two clones and Q at position 3 in one
clone. For Class IV clones, positions 2 and 4 show a
preference for H, although Y, Q W, and F are also tol-
erated.
We became aware of the presence of these CP2-
binding motifs in REST, YY1, PIAS1, PKC-d and
three other proteins (Table 2). GST pull-down and co-
immunoprecipitation assays indicate that REST and
YY1 indeed bind to CP2 in vitro and in vivo (Fig. 4).
Although YY1 is known to interact with CP2 [36],
REST binding to CP2 is a novel finding. We also pro-
vide the first evidence that the HXPR motif mediates
the binding of these two proteins to CP2. YY1 and
REST mutants having substitutions of H to A or PR
to AA on the HXPR motif had significantly reduced
or virtually no CP2-binding capability (Figs 4A and
5A). In addition, an HXPR motif-containing peptide

was able to repress the interaction of YY1 and CP2
(Fig. 5B). It is noteworthy that HXPR encloses a
newly discovered PR motif [59] with which the WW-A
domain has been shown to interact. Because Fe65,
which binds to CP2 [33] and is thought to be patho-
genically associated with Alzheimer’s disease [60], has
a WW-A domain [59] it might be of interest to explore
the possible association of Fe65 with YY1 and REST.
Using various deletion mutants of CP2, we have
shown that the carboxyl-flanking region (more specif-
ically, amino acids 306–396) of the SPXX region on
CP2 is involved in the binding of the ASR motif as
well as HXPR (Fig. 3). However, we do not know
whether these two motifs recognize the same
sequences in CP2, and thus further study is required
to dissect this region. If the two motifs bind to over-
lapping, or the same, residues, the question might be
raised as to whether the two motifs can compete
with each other for binding to CP2. Because several
proteins, including PKC-d and CKII-a, possess the
ASR motif (Table 2), it will be interesting to know
whether PKC-d and CKII-a can also alter the effect
of YY1 on CP2. We also demonstrated that, unlike
the HXPR and ASR motifs, PHL and PXHXH
motifs bind to the N- and C-terminal parts of the
Elf-1 domain on CP2, respectively (Fig. 3). Because
the Elf-1 domain has been thought to primarily
mediate DNA binding by CP2 [38], this is the first
time that an additional, novel function has been
assigned to the Elf-1 domain of CP2.

Our list of polypeptides that bind to CP2 shows a
preponderance of histidines, implying the importance
of this amino acid residue in the CP2-binding motif
of proteins that interact with CP2. Indeed, two or
three histidine residues are found around the HXPR
motif of both YY1 and REST. We can exclude
possible nonspecific interactions between polypeptides
and the plastic material of the plate well during the
phage screening, which may result in enrichment of
histidine-rich peptides, as it has been reported that
plastics prefer tyrosine and tryptophan but not histi-
dine [61].
Identification of CP2-binding peptide motifs H. C. Kang et al.
1272 FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS
Taken together, our findings suggest that CP2 is
able to receive a variety of single and ⁄ or combined
regulatory inputs by utilizing its different regions to
interact with various CP2-binding motifs. Thus, this
might be one of the mechanisms by which CP2 exerts
multifunctional roles in various cellular processes
including apoptosis, proliferation and differentiation
via differential transcriptional regulations of target
genes.
Experimental procedures
Cell culture and luciferase assay
293T cells were maintained in Dulbecco’s modified Eagle’s
medium (Invitrogen-Gibco ⁄ BRL, Carlsbad, CA, USA) sup-
plemented with 10% fetal bovine serum (Hyclone, Logan,
UT, USA). For luciferase assays, cells (5 · 10
4

) were plated
on 24-well plates and transfected with various combinations
of reporter and expression vectors (total 2 lg DNA) using
a calcium phosphate method [62]. At 48 h after transfec-
tion, cells were harvested in NaCl ⁄ P
i
and resuspended in
passive lysis buffer (Promega, Madison, WI, USA) and
dual (firefly and renilla) luciferase activities were measured
using a luminometer Lumat LB9501 (Berthold, Gaithers-
burg, MD, USA). Firefly luciferase expression was normal-
ized against renilla luciferase. Each experiment was
performed in duplicate and repeated at least three times.
Plasmid constructs
CP2 cDNA fragments were cloned in-frame with the GST
coding sequence in prokaryotic pGEX-4T1 vectors (Amer-
sham-Pharmacia, Piscataway, NJ, USA) and also cloned
in-frame with the HA tag sequence in the eukaryotic
pCMV-HA vector (Clontech, Palo Alto, CA, USA).
EGFP–CP2 was generated by PCR cloning into pEGFP-N1
(Clontech) using following primers; 5¢-G
AAGCTTAT
GGCCTGGGCTCTGAAG-3¢ and 5¢-C
GGTACCGCCTT
GAGAATGACATGATAG-3¢, which contain HindIII and
KpnI sites, respectively (underlined). GST-tagged CP2-
deletion constructs were generated by PCR from pGEX-
4T1–CP2 or pCMV-HA–CP2. GST- or FLAG-tagged 12-mer
peptides were prepared as follows: phage DNA was amplified
by PCR using primers 5¢-CTTTAGT

GGTACCTTTC
TATTC-3¢ and 5¢-GTATGGGATTTTG
CTCGAGAACTT
TC-3¢, which introduced KpnI and XhoI sites, respectively
(underlined). The PCR products were then inserted into the
pGEM-Teasy vector (Promega). The EcoRI ⁄ XhoI frag-
ments from these pGEM-Teasy–peptide constructs were
cloned into pGEX-4T1 (Amersham-Pharmacia) or pcDNA-
FLAG vectors. Two oligonucleotides containing the nuclear
localization signal (NLS) and EcoRI and KpnI restriction
sequences (underlined) were inserted into the respective sites
of pcDNA-FLAG)12-mer peptide vector: 5¢-CG
GAATTC
CCCCCAAAAAAGAAGAGAAAGAT-3¢ and 5¢-GG
GG
TACCCCGTCTTCTATCTTTCTCTTCTTT-3¢. The REST
cDNA fragment was generated by PCR from the pcDNA-
FLAG–REST vector with specific primers (see below) and
then cloned in frame into the pCMV–HA vector. The
pcDNA–FLAG–REST vector was kindly provided by G
Thiel (University of Saarland Medical Center, Germany),
and HA–YY1 and GST–YY1 vectors were from Y Shi
(Harvard Medical School, Cambridge, MA, USA) and E
Seto (University of South Florida, Tampa, FL, USA),
respectively.
Site-directed mutagenesis
A PCR-based site-directed mutagenesis approach was used
to generate two YY1 point mutants, YY1(H320A) and
YY1(P322A ⁄ R323A) that have amino acid substitution of
the His residue at position 320 or both Pro and Arg resi-

dues at positions 322 and 323 to Ala. For cDNA construc-
tion of each mutant, two products were generated covering
bases 1–975 and 949–1245 of the respective mutant YY1
cDNA. pCMV–HA–YY1 was employed as a template and
the following primers (bases changed for point mutations
are underlined) were used: forward primer (5¢-GGAATT
CTCATGGCCTCGGGCGACACC-3¢) corresponding to
bases 1–18 of YY1 cDNA; reverse oligomer (5¢-GCTCGA
GTCACTGGTTGTTTTTGGCC-3¢) for bases 1227–1245;
reverse (5¢-GTGGAC
TGCGGCTCCGTGGGTGTG-3¢)
and forward (5¢-CAC ACCCAC GGA
GCCGCAGTC CAC- 3¢)
oligonucleotides for Pro ⁄ Arg to Ala ⁄ Ala mutations in bases
952–975; reverse (5¢-GTGGACTCTGGGACC
GGCTGTG
TGCAG-3¢) and forward (5¢-CTGCACACA
GCCGGT
CCCAGAGTCCAC-3¢) oligonucleotides for His to Ala
mutation in bases 949–975. The two overlapping PCR frag-
ments were mixed and added as a template in the second
PCR, which used 1–18 and 1227–1245 base oligomers as a
forward and a reverse primer, respectively. This resulted in
the generation of 1245 bp products containing the point
mutations PRfiAA or HfiA of YY1. These products
were directly cloned into pGEM-T Easy vector (Promega).
Following digestion with EcoRI and XhoI, fragments
were cloned into pGEX-4T2 or pCMV–HA vectors and
sequencing analysis was performed to confirm the point
mutations. A plasmid containing point mutation in

REST, pCMV–HA–REST(P272A ⁄ R273A), was generated
by the same method. pcDNA–FLAG–REST was used as a
template and the following primers (bases changed for
point mutations are underlined) were used: forward primer
(5¢-GGAATTCTCATGGCCACCCAGGTAATGG-3¢) cor-
responding to bases 1–19 of REST cDNA and reverse
primer (5¢-CTCGAGTTACTCCTGCCCTTGAGCTGC-3¢)
for bases 3274–3294; reverse primer (5¢-GTGTATACTT
TCGCTGCAAAATGGTTTC-3¢) and forward primer
H. C. Kang et al. Identification of CP2-binding peptide motifs
FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS 1273
(5¢-GAAACCATTTTGCAGCGAAAGTATACAC-3¢) for
Pro ⁄ Arg to Ala ⁄ Ala mutations in bases 803–830. The two
overlapping PCR fragments were mixed and added as a
template in the second PCR that used 1–19 and 3274–3294
base oligomers as a forward and a reverse primer, respect-
ively. This resulted in the generation of a 3294 bp product
containing the PRfiAA mutation of REST. This product
was directly cloned into pGEM-T Easy vector (Promega).
Following digestion with EcoRI and XhoI, the fragment
was cloned into the pCMV–HA vector.
Purification of GST fusion proteins
GST-fused proteins were expressed in E. coli BL21 (pLys).
Cell extracts were incubated with glutathione–agarose beads
(Sigma-Aldrich, St. Louis, MO, USA), and the beads were
washed extensively with NaCl ⁄ P
i
. Proteins were eluted with
20 mm reduced glutathione. The protein concentration was
determined using the Bradford-based protein assay (Bio-

Rad, Hercules, CA, USA) and the purity was checked by
SDS ⁄ PAGE.
Peptide phage library screening
GST–CP2 or GST itself was immobilized onto microtiter
plates by directly coating the wells with 10 lLof
100 lgÆmL
)1
proteins in TBS (50 mm Tris ⁄ HCl, pH 7.5,
150 mm NaCl) for 8 h at 4 °C in a humidified container.
The wells were rapidly washed six times with TBST [TBS,
0.1% (v ⁄ v) Tween-20]. To remove any potential GST-bind-
ing phage, the library was precleared three times using
GST. Approximately 1.2 · 10
12
phage (New England Bio-
labs, Beverly, MA, USA) in 100 lL TBS were added
sequentially to two GST-coated wells and incubated for
10 min at room temperature in each round of preclearing.
Precleared phage were then added to GST–CP2-coated
wells in 200 lL TBS ⁄ 0.1% (w ⁄ v) bovine serum albumin
(BSA) ⁄ 0.05% (w ⁄ v) GST and incubated for 2 h at 4 °C.
Unbound phage were removed by washing the wells with
cold TBS ⁄ 0.05% (v ⁄ v) Tween 20. Bound phage were eluted
with 100 lL solution of glycine ⁄ HCl (0.2 m, pH 2.2) and
0.1% (w ⁄ v) BSA. After immediate neutralization with
10 mm Tris ⁄ HCl (pH 9.1), the eluted phage were amplified
by infecting log-phase E. coli (strain ER2738). The superna-
tants of the bacterial culture were treated with polyethylene
glycol (PEG) and NaCl for 8 h on ice, and the phage
precipitates were recovered by centrifugation. To kill any

residual bacteria, the phage resuspension in TBS was heat-
treated at 70 °C for 15 min. The collected phage were then
subjected to five more rounds of amplification as above.
The phage titer of every round was determined by infecting
200 lL of a mid-log phase culture of E. coli with 10 lLof
serially diluted phage and plating on LB ⁄ IPTG ⁄ Xgal
plates. The number of resulting colonies was used to deter-
mine plaque-forming units per milliliter. The ability of the
isolated phage to bind CP2 was confirmed by an ELISA
assay where GST–CP2-bound phage in the microtiter wells
were colorimetrically detected using a horseradish
peroxidase (HRP)-conjugated anti-M13 serum (Amersham-
Pharmacia) and 2,2¢-azino-bis (3-ethylbenzothiazoline-6-
sulfonic acid) buffer (Sigma-Aldrich). To eliminate the
false-positive signals, control experiments with GST (10 lg
protein) coated wells were performed concurrently.
Deduced peptide sequence determination
To amplify individual phage clones for DNA sequencing
analysis, single colonies were isolated from the titering
plates and used to infect 1 mL of a 100-fold diluted over-
night culture of ER2738 E. coli. The cultures were incuba-
ted for 5 h with shaking at 37 °C and the phage-containing
supernatant was cleared by centrifugation (13 000 g,30sat
4 °C). For DNA sequencing, 500 lL of the phage stock
was precipitated with 200 lL of PEG ⁄ NaCl and the phage-
containing pellet was resuspended in 100 lL of iodide buf-
fer (10 mm Tris ⁄ HCl, pH 8.0, 1 mm EDTA, 4 m NaI). The
preferential precipitation of single-stranded phage DNA
was accomplished by ethanol precipitation. The DNA pellet
was then resuspended in 30 lL of Tris ⁄ EDTA buffer

(pH 8.0). DNA sequencing of recombinant phage inserts
was carried out using Dideoxy-Terminator Sequencing
Chemistry from CEQ 2000 XL DNA Analysis System
(Beckman, Urbana, IL, USA) with the primer 5¢-CCCTCA
TAGTTAGCGTAACG-3¢.
Coimmunoprecipitation and GST pull-down
assays
For coimmunoprecipitation assays, 293T cells were trans-
fected with either CP2 alone or CP2 together with HA–
REST, HA–REST(P272A ⁄ R273A) or HA–YY1 expression
vectors. Forty-eight hours after transfection, cells were sol-
ubilized with precooled lysis buffer A (30 mm Hepes,
pH 7.4, 100 mm NaCl, 1 mm EGTA, 1% (v ⁄ v) Triton X-
100, 0.1% (w ⁄ v) BSA, protease inhibitor cocktail [Sigma-
Aldrich] and 1 mm phenylmentanesulfonyl fluoride]. To
these extracts (200 lg of protein), appropriate antibodies
and 10 lL of protein A–Sepharose beads (Sigma-Aldrich)
were added. After incubation for 1 h at room temperature,
the beads were washed three times with NaCl ⁄ P
i
⁄ 0.6%
(v ⁄ v) Tween 20 (PBST), and proteins were eluted by boiling
in SDS sample buffer [50 mm Tris ⁄ HCl, pH 6.8, 100 mm
dithiothreitol, 2% (w ⁄ v) SDS, 0.1% (w ⁄ v) bromophenol
blue, and 10% (v ⁄ v) glycerol]. Proteins were fractionated
by SDS ⁄ PAGE and transferred to a nitrocellulose mem-
brane in transfer buffer [25 mm Tris, 40 mm glycine, 0.05%
(w ⁄ v) SDS, 20% (v ⁄ v) methanol]. The membrane (Bio-
Rad) was blocked with PBST ⁄ nonfat dried milk for 1 h,
followed by 1 h incubation with the appropriate antibodies

and finally treated with a secondary HRP-conjugated goat
Identification of CP2-binding peptide motifs H. C. Kang et al.
1274 FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS
anti-mouse IgG (Pierce, Rockford, IL, USA) at 1 : 10 000
in PBST for 1 h. All of these immunoblotting procedures
were performed at room tempertaure, and membranes were
washed several times with PBST for 15 min between steps.
Membrane were then treated with ECL reagents (Amer-
sham-Pharmacia) and exposed to X-ray film for 5–20 s.
For GST pull-down assays, glutathione–Sepharose beads
(Amersham-Pharmacia) bound with GST–YY1, GST–
YY1(P322A ⁄ R323A), GST–YY1(H320A), GST)12-mer
peptide or GST were incubated with the extracts prepared
from 293T cells transiently expressing HA–CP2 in buffer B
[50 mm potassium phosphate, pH 7.5, 100 mm KCl, 10%
(v ⁄ v) glycerol, 0.1% (v ⁄ v) Triton X-100] for 2 h at 4 °C.
The beads were washed four times with buffer B omitting
glycerol and Triton X-100 [53]. Beads were then boiled for
5 min in SDS sample buffer, analyzed by 10% SDS ⁄ PAGE
and immunoblotted with specific antibodies.
Acknowledgements
We thank Drs Seto, Mu
¨
ller, Shi, Shank and Thiel for
their donations of the plasmid constructs and technical
inputs. We also thank Brian Watson for his critical
reading and comment on the manuscript. This work
was supported by a Korea Research Foundation
Grant (KRF-2002-015-CP0283). HCK and JHC were
supported by the Brain Korea 21 Project from the

Ministry of Education and Human Resources of
Korea through the Research Group on Stem Cells and
Early Development.
References
1 Barnhart KM, Kim CG, Banerji SS & Sheffery M (1988)
Identification and characterization of multiple erythroid
cell proteins that interact with the promoter of the murine
alpha-globin gene. Mol Cell Biol 8, 3215–3226.
2 Bray SJ, Burke B, Brown NH & Hirsh J (1989)
Embryonic expression pattern of a family of Drosophila
proteins that interact with a central nervous system reg-
ulatory element. Genes Dev 3, 1677–1688.
3 Dynlacht BD, Attardi LD, Admon A, Freeman M &
Tjian R (1989) Functional analysis of NTF-1, a devel-
opmentally regulated Drosophila transcription factor
that binds neuronal cis elements. Genes Dev 3, 1677–
1688.
4 Biggin MD & Tjian R (1988) Transcription factors that
activate the Ultrabithorax promoter in developmentally
staged extracts. Cell 53, 699–711.
5 Soeller WC, Poole SJ & Kornberg T (1988) In vitro
transcription of the Drosophila engrailed gene. Genes
Dev 2, 68–81.
6 Huang JD, Dubnicoff T, Liaw GJ, Bai Y, Valentine
SA, Shirokawa JM, Lengyel JA & Courey AJ (1995)
Binding sites for transcription factor NTF-1 ⁄ Elf-1 con-
tribute to the ventral repression of decapentaplegic.
Genes Dev 9, 3177–3189.
7 Liaw GJ, Rudolph KM, Huang JD, Dubnicoff T, Cou-
rey AJ & Lengyel JA (1995) The torso response element

binds GAGA and NTF-1 ⁄ Elf-1, and regulates tailless
by relief of repression. Genes Dev 9, 3163–3176.
8 Attardi LD & Tjian R (1993) Ectopic expression of
wild-type or a dominant-negative mutant of transcrip-
tion factor NTF-1 disrupts normal Drosophila develop-
ment. Genes Dev 7, 1341–1353.
9 Uv AE, Thompson CR & Bray SJ (1994) The Droso-
phila tissue-specific factor Grainyhead contains novel
DNA-binding and dimerization domains which are con-
served in the human protein CP2. Mol Cell Biol 14,
4020–4031.
10 Wilanowski T, Tuckfield A, Cerruti L, O’Connell S,
Saint R, Parekh V, Tao J, Cunningham JM & Jane SM
(2002) A highly conserved family of mammalian devel-
opmental transcription factors related to Drosophila
grainyhead. Mech Dev 114, 37–50.
11 Tuckfield A, Clouston DR, Wilanowski TM, Zhao L-L,
Cunningham JM & Jane SM (2002) Binding of the
RING polycomb proteins to specific target genes in
complex with the grainyhead-like family of developmen-
tal transcription factors. Mol Cell Biol 22, 1936–1946.
12 Ting SB, Wilanowski T, Cerruti L, Zhao LL, Cunning-
ham JM & Jane SM (2003) The identification and char-
acterization of human Sister-of-Mammalian Grainyhead
(SOM) expands the grainyhead-like family of develop-
mental transcription factors. Biochem J 370, 953–962.
13 Kudryavtseva EI, Sugihara TM, Wang N, Lasso RJ,
Gudnason JF, Lipkin SM & Andersen B (2003) Identifi-
cation and characterization of Grainyhead-like epithelial
transactivator (GET-1), a novel mammalian Grainy-

head-like factor. Dev Dyn 226, 604–617.
14 Murata T, Nitta M & Yasuda K (1998) Transcription
factor CP2 is essential for lens-specific expression of the
chicken alphaA-crystallin gene. Gene Cell 3, 443–457.
15 Shirra MK, Zhu Q, Huang HC, Pallas D & Hansen U
(1994) One exon of the human LSF gene includes con-
served regions involved in novel DNA-binding and
dimerization motifs. Mol Cell Biol 14, 5076–5087.
16 Yoon JB, Li G & Roeder RG (1994) Characterization of
a family of related cellular transcription factors which
can modulate human immunodeficiency virus type 1 tran-
scription in vitro. Mol Cell Biol 14, 1776–1785.
17 Lim LC, Swendeman SL & Sheffery M (1992) Mole-
cular clonging of the alpha-globin transcription factor
CP2. Mol Cell Biol 12, 828–835.
18 Sueyoshi T, Kobayashi R, Nishio K, Aida K, Moore R,
Wada T, Handa H & Negishi M (1995) A nuclear fac-
tor (NF2d9) that binds to the male-specific P450 (Cyp
2d-9) gene in mouse liver. Mol Cell Biol 15, 4158–4166.
H. C. Kang et al. Identification of CP2-binding peptide motifs
FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS 1275
19 Rodda S, Sharma S, Scherer M, Chapman G & Rathjen
P (2001) CRTR-1, a developmentally regulated tran-
scriptional repressor related to the CP2 family of tran-
scription factors. J Biol Chem 276, 3324–3332.
20 Chae JH, Lee YH & Kim CG (1999) Transcription fac-
tor CP2 is crucial in hemoglobin synthesis during ery-
throid terminal differentiation in vitro. Biochem Biophys
Res Commun 263 , 580–583.
21 Chae JH & Kim CG (2003) CP2 binding to the promo-

ter is essential for the enhanced transcription of globin
genes in erythroid cells. Mol Cells 15, 40–47.
22 Amrolia PJ, Gabbard W, Cunningham JM & Jane SM
(1998) Maximal activity of an erythroid-specific enhan-
cer requires the presence of specific protein binding sites
in linked promoters. J Biol Chem 273, 13593–13598.
23 Jane SM, Ney PA, Vanin EF, Gumucio DL & Nienhuis
AW (1992) Identification of a stage selector element in
the human gamma-globin gene promoter that fosters
preferential interaction with the 5¢ HS2 enhancer when
in competition with the beta-promoter. EMBO J 11,
2961–2969.
24 Zhou W, Clouston DR, Wang X, Cerruti L, Cunning-
ham JM & Jane SM (2000) Induction of human fetal
globin gene expression by a novel erythroid factor,
NF-E4. Mol Cell Biol 20 , 7662–7672.
25 Jane SM, Nienhuis AW & Cunningham JM (1995)
Hemoglobin switching in man and chicken is mediated
by a heteromeric complex between the ubiquitous tran-
scription factor CP2 and a developmentally specific pro-
tein. EMBO J 14, 97–105.
26 Bruni P, Minopoli G, Brancaccio T, Napolitano M,
Faraonio R, Zambrano N, Hansen U & Russo T (2002)
Fe65, a ligand of the Alzheimer’s beta-amyloid precur-
sor protein, blocks cell cycle progression by down-
regulating thymidylate synthase expression. J Biol Chem
277, 35481–35488.
27 Powell CM, Rudge TL, Zhu Q, Johnson LF & Hansen
U (2000) Inhibition of the mammalian transcription fac-
tor LSF induces S-phase-dependent apoptosis by down

regulating thymidylate synthase expression. EMBO J
19, 4665–4675.
28 Solis C, Aizencang GI, Astrin KH, Bishop DF & Des-
nick RJ (2001) Uroporphyrinogen III synthase erythroid
promoter mutations in adjacent GATA1 and CP2 ele-
ments cause congenital erythropoietic porphyria. J Clin
Invest 107, 753–762.
29 Lambert JC, Goumidi L, Vrieze FW, Frigard B, Harris
JM, Cummings A, Coates J, Pasquier F, Cottel D,
Gaillac M, et al. (2000) The transcriptional factor
LBP-1c ⁄ CP2 ⁄ LSF gene on chromosome 12 is a genetic
determinant of Alzheimer’s disease. Hum Mol Genet 9,
2275–2280.
30 Taylor AE, Yip A, Brayne C, Easton D, Evans JG,
Xuereb J, Cairns N, Esiri MM & Rubinsztein DC
(2001) Genetic association of an LBP-1c ⁄ CP2 ⁄ LSF gene
polymorphism with late onset Alzheimer’s disease.
J Med Genet 38, 232–233.
31 Kim CH, Heath C, Bertuch A & Hansen U (1987) Spe-
cific stimulation of simian virus 40 late transcription in
vitro by a cellular factor binding the simian virus 40 21-
base-pair repeat promoter element. Proc Natl Acad Sci
USA 84, 6025–6029.
32 Russo T, Faraonio R, Minopoli G, De Candia P, De
Renzis S & Zambrano N (1998) Fe65 and the protein
network centered around the cytosolic domain of the
Alzheimer’s beta-amyloid precursor protein. FEBS Lett
434, 1–7.
33 Zambrano N, Minopoli G, de Candia P & Russo T
(1998) The Fe65 adaptor protein interacts through its

PID1 domain with the transcription factor
CP2 ⁄ LSF ⁄ LBP1. J Biol Chem 273, 20128–20133.
34 Kim HS, Kim EM, Lee JP, Park CH, Kim S, Seo JH,
Chang KA, Yu E, Jeong SJ, Chong YH & Suh YH
(2003) C-Terminal fragments of amyloid precursor pro-
tein exert neurotoxicity by inducing glycogen synthase
kinase-3beta expression. FASEB J 17, 1951–1953.
35 Kashour T, Burton T, Dibrov A & Amara FM (2003)
Late simian virus 40 transcription factor is a target of
the phosphoinositide 3-kinase ⁄ Akt pathway in anti-
apoptotic Alzheimer’s amyloid precursor protein signal-
ling. Biochem J 370, 1063–1075.
36 Coull JJ, Romerio F, Sun JM, Volker JL, Galvin KM,
Davie JR, Shi Y, Hansen U & Margolis DM (2000)
The human factors YY1 and LSF repress the human
immunodeficiency virus type 1 long terminal repeat via
recruitment of histone deacetylase 1. J Virol 74, 6790–
6799.
37 Romerio F, Gabriel MN & Margolis DM (1997)
Repression of human immunodeficiency virus type 1
through the novel cooperation of human factors YY1
and LSF. J Virol 71, 9375–9382.
38 Zhong F, Swendeman SL, Popik W, Pitha PM & Sheff-
ery M (1994) Evidence that levels of the dimeric cellular
transcription factor CP2 play little role in the activation
of the HIV-1 long terminal repeat in vivo or following
superinfection with herpes simplex virus type 1. J Biol
Chem 269, 21269–21276.
39 Ballas N, Battaglioli E, Atouf F, Andres ME, Chenow-
eth J, Anderson ME, Burger C, Moniwa M, Davie JR,

Bowers WJ et al. (2001) Regulation of neuronal traits
by a novel transcriptional complex. Neuron 31, 353–365.
40 Lietz M, Hohl M & Thiel G (2003) RE-1 silencing tran-
scription factor (REST) regulates human synaptophysin
gene transcription through an intronic sequence-specific
DNA-binding site. Eur J Biochem 270, 2–9.
41 Berndt JA, Kim JG, Tosic M, Kim C & Hudson LD
(2001) The transcriptional regulator Yin Yang 1 acti-
vates the myelin PLP gene. J Neurochem 77, 935–942.
42 Liu B, Liao J, Rao X, Kushner SA, Chung CD, Chang
DD & Shuai K (1998) Inhibition of Stat1-mediated gene
Identification of CP2-binding peptide motifs H. C. Kang et al.
1276 FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS
activation by PIAS1. Proc Natl Acad Sci USA 95,
10626–10631.
43 Tan J, Hall SH, Hamil KG, Grossman G, Petrusz P,
Liao J, Shuai K & French FS (2000) Protein inhibitor
of activated STAT-1 (signal transducer and activator of
transcription-1) is a nuclear receptor coregulator
expressed in human testis. Mol Endocrinol 14, 14–26.
44 Kotaja N, Karvonen U, Janne OA & Palvimo JJ (2002)
PIAS proteins modulate transcription factors by func-
tioning as SUMO-1 ligases. Mol Cell Biol 22, 5222–
5234.
45 Leng L, Yu F, Dong L, Busquets X, Osada S, Richon
VM, Marks PA & Rifkind RA (1993) Differential modu-
lation of protein kinase C isoforms in erythroleukemia
during induced differentiation. Cancer Res 53, 5554–
5558.
46 Sparatore B, Pessino A, Patrone M, Passalacqua M,

Melloni E & Pontremoli S (1995) Changes in calcium
influx affect the differentiation of murine erythroleukae-
mia cells. Biochem J 305, 285–290.
47 Ouyang L, Chen X & Bieker JJ (1998) Regulation of
erythroid Kruppel-like factor (EKLF) transcriptional
activity by phosphorylation of a protein kinase casein
kinase II site within its interaction domain. J Biol Chem
273, 23019–23025.
48 Biekerm JJ, Ouyang L & Chen X (1998) Transcriptional
factors for specific globin genes. Ann NY Acad Sci 850,
64–69.
49 Marziali G, Perrotti E, Ilari R, Lulli V, Coccia EM,
Moret R, Kuhn LC, Testa U & Battistini A (2002) Role
of Ets-1 in transcriptional regulation of transferrin
receptor and erythroid differentiation. Oncogene 21,
7933–7944.
50 Ohkuma Y (1997) Multiple functions of general tran-
scription factors TFIIE and TFIIH in transcription:
possible points of regulation by trans-acting factors.
J Biochem 122, 481–489.
51 Chong JA, Tapia-Ramirez J, Kim S, Toledo-Aral JJ,
Zheng Y, Boutros MC, Altshuller YM, Frohman MA,
Kraner SD & Mandel G (1995) REST: a mammalian
silencer protein that restricts sodium channel gene
expression to neurons. Cell 80, 949–957.
52 Schoenherr CJ & Anderson DJ (1995) The neuron-restric-
tive silencer factor (NRSF): a coordinate repressor of
multiple neuron-specific genes. Science 267, 1360–1363.
53 Galvin KM & Shi Y (1997) Multiple mechanisms of
transcriptional repression by YY1. Mol Cell Biol 17,

3723–3732.
54 Thomas MJ & Seto E (1999) Unlocking the mechanisms
of transcription factor YY1: are chromatin modifying
enzymes the key? Gene 236, 197–208.
55 Shi Y, Seto E, Chang LS & Shenk T (1991) Transcrip-
tional repression by YY1, a human GLI-Kruppel-
related protein, and relief of repression by adenovirus
E1A protein. Cell 67, 377–388.
56 Ficzycz A & Ovsenek N (2002) The Yin Yang 1 tran-
scription factor associates with ribonucleoprotein
(mRNP) complexes in the cytoplasm of Xenopus
oocytes. J Biol Chem 277, 8382–8387.
57 Lee JS, Galvin KM & Shi Y (1993) Evidence for physi-
cal interaction between the zinc-finger transcription
factors YY1 and Sp1. Proc Natl Acad Sci USA 90,
6145–6149.
58 Yao YL, Yang WM & Seto E (2001) Regulation of
transcription factor YY1 by acetylation and deacetyla-
tion. Mol Cell Biol 21, 5979–5991.
59 Bedford MT, Sarbassova D, Xu J, Leder P & Yaffe
MB (2000) A novel pro-Arg motif recognized by WW
domains. J Biol Chem 275, 10359–10369.
60 Zambrano N, Buxbaum JD, Minopoli G, Fiore F, De
Candia P, De Renzis S, Faraonio R, Sabo S, Cheetham
J, Sudol M & Russo T (1997) Interaction of the phos-
photyrosine interaction ⁄ phosphotyrosine binding-related
domains of Fe65 with wild-type and mutant Alzheimer’s
beta-amyloid precursor proteins. J Biol Chem 272,
6399–6405.
61 Adey NB, Mataragnon AH, Rider JE, Carter JM &

Kay BK (1995) Characterization of phage that bind
plastic from phage-displayed random peptide libraries.
Gene 156, 27–31.
62 Jordan M, Schallhorn A & Wurm FM (1996) Transfect-
ing mammalian cells: optimization of critical parameters
affecting calcium-phosphate precipitate formation.
Nucleic Acids Res 24, 596–601.
63 Ahn SJ, Han SJ, Mo HJ, Chung JK, Hong SH, Park
TK & Kim CG (1998) Interaction of phospholipase C
gamma 1 via its COOH-terminal SRC homology 2
domain with synaptojanin. Biochem Biophys Res
Commun 244, 62–67.
H. C. Kang et al. Identification of CP2-binding peptide motifs
FEBS Journal 272 (2005) 1265–1277 ª 2005 FEBS 1277