Báo cáo khoa học: Nucling interacts with nuclear factor-jB, regulating its cellular distribution potx

Nucling interacts with nuclear factor-jB, regulating its
cellular distribution
Li Liu
1,
*,, Takashi Sakai
1,
*, Nam Hoang Tran
1
, Rika Mukai-Sakai
1,2
, Ryuji Kaji
2
and Kiyoshi Fukui
1
1 Division of Enzyme Pathophysiology, Institute for Enzyme Research, University of Tokushima, Japan
2 Institute of Health Biosciences, University of Tokushima Graduate School, Japan
Nucling was originally isolated from murine embryo-
nal carcinoma cells as a novel protein, the expression
of which was upregulated during cardiac muscle diff-
erentiation [1]. We have reported previously that
Nucling, as a proapoptotic factor, is associated with
the apoptosome pathway. It has been shown that
Nucling recruits the Apaf1/procaspase-9 complex for
the induction of stress-induced apoptosis [2]. In
addition, Nucling acts as an important factor during
1-methyl- 4-phe nyl-1, 2,3,6-tetrahydropyridine-induced
neurotoxicity through activation of the apoptosome
pathway [3]. We have also reported that Nucling
mediates apoptosis by inhibiting the expression of
galectin-3, an antiapoptotic factor, through interfer-
ence with nuclear factor-jB (NF-jB) signaling [4]. In

Nucling-deficient (Nucling
)/)
) mice, a high occurrence
of inflammatory lesions of preputial glands was
observed in males, and hepatocellular carcinoma aris-
ing from cholestatic hepatitis was frequently detected
in aged mice. Furthermore, we have confirmed that the
activation of NF-jB is involved in these pathological
changes (T. Sakai, L. Liu, X. Teng, N. Ishimaru,
R. Mukai-Sakai, N. H. Tran, N. Sano, Y. Hayashi,
R. Kaji & K. Fukui, unpublished results). Although
Keywords
IjBa; nuclear factor-jB; nuclear
translocation; Nucling; tumor necrosis
factor-a
Correspondence
K. Fukui, Institute for Enzyme Research,
University of Tokushima, 3-18-15 Kuramoto-
cho, Tokushima 770-8503, Japan
Fax: +81 88 633 7431
Tel: +81 88 633 7430
E-mail:
yPresent address
Department of Physics, Graduate School of
Science, Kyoto University, Japan
*These authors contributed equally to this
work
(Received 11 April 2008, revised 30
December 2008, accepted 5 January 2009)
doi:10.1111/j.1742-4658.2009.06888.x

Nucling is an Apaf1-binding proapoptotic protein involved in apoptosome-
mediated apoptosis. Luciferase assays have revealed that the activation of
nuclear factor-jB induced by tumor necrosis factor-a, interleukin-1b and
lipopolysaccharide is downregulated by the overexpression of Nucling in
HEK293 cells. Moreover, the expression of endogenous cyclooxygenase 2,
tumor necrosis factor-a and galectin-3, the end-point molecules in the path-
way for the activation of nuclear factor-jB, as well as nuclear factor-jB
(p65) itself, is upregulated in Nucling gene-deficient mouse embryonic
fibroblasts, suggesting that nuclear factor-jB is a target of Nucling.
Subsequent study has revealed that Nucling physically interacts with
nuclear factor-jB (p65 and p50) and that the binding domain of Nucling is
its amino-terminal region (amino acids 1–466) containing ankyrin repeats.
Overexpression of Nucling prevents the translocation of nuclear factor-jB
into the nucleus. In addition, the cytoplasmic retention of endogenous
nuclear factor-jB in resting cells is not observed in Nucling gene-deficient
mouse embryonic fibroblasts. These results reveal a novel function of
Nucling as a suppressor of nuclear factor-jB, mediated by its cytoplasmic
retention through physical interaction.
Abbreviations
COX2, cyclooxygenase 2; EMSA, electrophoretic mobility shift assay; FITC, fluorescein isothiocyanate; IL-1b, interleukin-1b; LPS,
lipopolysaccharide; MEF, mouse embryonic fibroblast; NF-jB, nuclear factor-jB; TK, thymidine kinase; TNFa, tumor necrosis factor-a.
FEBS Journal 276 (2009) 1459–1470 ª 2009 The Authors Journal compilation ª 2009 FEBS 1459
these findings strongly indicate that Nucling constitutes
a missing link between inflammation and cancer via
the NF-jB pathway, the mechanism underlying the
regulation of this pathway by Nucling remains to be
elucidated.
The classic form of NF- jB is the heterodimer of the
p50 and p65 subunits, which is sequestered in the cyto-
plasm as an inactive complex bound to its inhibitor

protein IjB [5,6]. NF-jB is activated by various stim-
uli, including tumor necrosis factor (TNF), interleukin-1
(IL-1) and lipopolysaccharide (LPS) [7,8]. It is then
transported to the nucleus, where it binds to specific
sequences in the promoter or enhancer region of target
genes. Activated NF-jB can mediate NF-jB-regulated
gene transcription, the products of which are involved
in antiapoptosis, cell proliferation and invasion [9,10].
In this study, we propose a new regulatory mechanism
for the activation of NF-jB mediated by Nucling.
Nucling, as a regulatory factor for apoptosis, links the
NF-jB pathway and mediates the nuclear translocation
and activation of NF-jB. Nucling consists of a polypep-
tide of 1411 amino acids, containing an ankyrin repeat,
a leucine zipper motif and t-SNARE coiled-coil
domains. These domains may play some role in the for-
mation of complexes. Recently, we have reported that
Nucling interacts and downregulates the expression of
galectin-3, an antiapoptotic factor [4]. In addition,
Nucling has been found to upregulate Apaf1 expression
and recruit the Apaf1/procaspase-9 complex for the
induction of apoptosis following proapoptotic stress [2].
Moreover, we have identified several positive clones by
yeast two-hybrid screening for the protein interacting
with Nucling using NuclingDN as ‘bait’ [4]. Therefore, it
is conceivable that Nucling is involved in several signal-
ing pathways through interactions with possible regula-
tory factors. We have observed previously that a
deficiency of Nucling expression achieved using gene
targeting increases the basal level of activated NF-jBin

mouse embryonic fibroblasts (MEFs) [4]. However, it is
unclear whether or not Nucling itself is a regulator of
the activation of NF-jB. In this study, experiments were
carried out to investigate the molecular mechanism
underlying this regulatory role of Nucling in the NF-jB
pathway.
Results
Ectopic expression of Nucling downregulates the
activation of NF-jB following cytokine treatment
To test whether Nucling is involved in the activation
of NF-jB, we first examined the NF-jB-mediated acti-
vation of a luciferase reporter gene in the presence or
absence of transfected Nucling. We introduced
pFLAG-Nucling or pFLAG empty vector (mock)
together with a reporter plasmid comprising three
repeats of the NF-jB site upstream of a minimal thy-
midine kinase (TK) promoter and a luciferase gene
into HEK293 cells. The activation of NF- jB was eval-
uated with the dual-luciferase reporter assay system
(Fig. 1). In the mock transfection, the activation of
NF-jB was upregulated during stimulation with TNFa
(lane 2), IL-1b (lane 3) and LPS (lane 4) from the rest-
ing level (lane 1). However, the transfection of Nucling
clearly reduced TNFa-, IL-1b- and LPS-induced
NF-jB activity (lanes 5–8).
Nucling is involved in the regulation of NF-jB
signaling
We have reported that Nucling inhibits the expression
of galectin-3, an NF-jB-regulated gene, by suppressing
the activation of NF-jB [4]. To elucidate the general

role of Nucling in the NF-jB signaling pathway, we
TNFα
IL-1β
LPS
–
–
–
0
5
10
15
20
25
30
35
Nuclin
g
Mock
–
+
–
+
–
–
–
–
+
–
–
–

–
+
–
+
–
–
–
–
+
Relative activation
12 3 4 5 67 8
Fig. 1. Ectopic Nucling inhibits the transcriptional activity of NF-jB
following TNFa treatment. HEK293 cells were cotransfected with
the NF-jB reporter gene and Flag-Nucling or Flag empty vector
(mock). At 16 h post-transfection, cells were stimulated by TNFa,
IL-1b or LPS. Then, the activation of NF-jB was assessed by mea-
surement of luciferase activity. Each sample was analyzed at least
three times.
Nucling regulates cellular distribution of NF-jB L. Liu et al.
1460 FEBS Journal 276 (2009) 1459–1470 ª 2009 The Authors Journal compilation ª 2009 FEBS
attempted to investigate whether Nucling also affects
other NF-jB-regulated genes. Therefore, we compared
the protein and RNA levels of TNFa and cyclooxy-
genase 2 (COX2), as NF-jB-regulated genes, in MEFs
isolated from wild-type (WT) and Nucling
)/)
mice.
Four independently prepared WT and Nucling
)/)
MEFs were investigated. The results showed that

TNFa was upregulated in its expression in Nucling
)/)
MEFs in every case (Fig. 2A, panel 1; Fig. 2B,
panel 1; Fig. S1). COX2 was also upregulated in
Nucling
)/)
MEFs in two of four samples (Fig. 2A,
panel 2; Fig. 2B, panel 2). No significant difference
for COX2 between WT and Nucling
)/)
MEFs was
observed in two remaining samples (Fig. S1). COX2 is
not usually expressed in most tissues [11]. In the pres-
ent study, a weak band of COX2 protein was observed
in WT MEFs. We considered that the expression of
COX2 was induced by the presence of nuclear NF-jB-
p65 in WT MEFs in half of the four samples (Fig. 2A,
panel 6, lane 1). The upregulated expression of gale-
ctin-3 and the absence of Nucling were found in
Nucling
)/)
MEFs as a positive control (Fig. 2A, panels
3, 4). In addition, the nuclear translocation of NF-jB-
p65 was also confirmed in the nuclear fraction of
Nucling
)/)
MEFs (Fig. 2A, panel 6). Moreover, semi-
quantitative and real-time RT-PCR analyses revealed
that the transcription of COX2 was not upregulated in
Nucling

)/)
MEFs by TNFa (Fig. 2C; Fig. S1). This
result indicates that the NF-jB signaling pathway is
impaired in Nucling
)/)
MEFs. Taken together with the
results of the reporter assays in this study, we
concluded that Nucling is important for the regulation
of the NF-jB signaling pathway.
Nucling interacts with NF-jB
Considering that Nucling downregulates the activation
of NF-jB, we investigated whether the two physically
interact. COS7 cells were transfected with the expres-
sion vector for NF-jB-p50, NF-jB-p105 or NF-j B-
p65 with pFLAG-Nucling or a pFLAG vector. At
24 h post-transfection, cell lysates were immunoprecip-
itated using a mouse IgG1 against FLAG peptide
(anti-FLAG M5). Figure 3B,C illustrates that all three
NF-jB subunits interact with the full-length Nucling
and NuclingDC (amino acids 1–466, containing the
ankyrin repeat region; a deletion mutant lacking the
C-terminal region; Fig. 3A) directly in COS7 cells.
Concerning NuclingDN (amino acids 814–1411; a dele-
tion mutant lacking the N-terminal region; Fig. 3A), a
faint band was observed for p65, but no band was
detected for p50 or p105. The faint band was consid-
ered to be nonspecific because it was very weak
compared with the level of NuclingDN. However, no
interaction occurred between IjBa and Nucling
(Fig. 3D). Thus, we concluded that Nucling interacts

A
B
C
Fig. 2. Nucling is involved in the regulation of the NF-jB signaling
pathway. (A) Whole MEFs of Nucling
+/+
and Nucling
)/)
mice were
used to determine the levels of TNFa, COX2, galectin-3 and Nucling
(panels 1–4). Nuclear proteins from Nucling
+/+
and Nucling
)/)
MEFs
were used to determine the level of NF-jB-p65 (panel 6). The equal
loading of protein in each lane was confirmed using mouse IgG1
against b-actin (panels 5 and 7). (B) RT-PCR was carried out to
detect the transcripts of the TNFa and COX2 genes. Bar graphs
show the compiled means ± standard deviation for results of densi-
tometric scanning from three experiments for TNFa and COX2,
quantified by
NIH-IMAGE software. Data were normalized to the
density of Nucling
+/+
MEFs for the TNFa/b-actin and COX2/b-actin
ratios, respectively. The values for Nucling
)/)
MEFs were statis-
tically compared with those for Nucling

+/+
MEFs. *P < 0.05. (C)
RT-PCR was carried out to detect the transcripts of COX2 genes
following TNFa stimulation.
L. Liu et al. Nucling regulates cellular distribution of NF-jB
FEBS Journal 276 (2009) 1459–1470 ª 2009 The Authors Journal compilation ª 2009 FEBS 1461
A
B
C
D
Fig. 3. Nucling interacts with NF-jB-p50 and NF-jB-p65 but not with IjBa. (A) Primary structure of full-length Nucling and the deletion mutants
(NuclingDC, amino acids 1–466; NuclingDN, amino acids 814–1411) used for coimmunoprecipitation analysis. Ank, SN and LZ represent the
ankyrin repeat region, t-SNARE coiled-coil domain and leucine zipper motif, respectively. (B) COS7 cells were transiently cotransfected with
Flag-Nucling/Flag-NuclingDC/Flag-NuclingDN/Flag-vector and HA-NF-jB-p50/HA-NF-j B-p105. Lysates were subjected to a coimmunoprecipita-
tion assay with anti-Flag M5. Then, an immunoblot analysis was performed with rat IgG against HA. The presence of Nucling/
NuclingDC/NuclingDN and NF-jB-p50/NF-jB-p105 in the same lysates was verified by immunoblotting with anti-Flag M5 and antibody against
HA, respectively. (C) COS7 cells were cotransfected with Flag-Nucling/Flag-NuclingDC/Flag-NuclingDN/Flag-vector and NF-jB-p65, respectively.
The same method as in (B) was used for coimmunoprecipitation assay. Immunoprecipitates were subjected to western blotting using anti-
NF-jB-p65. The expression levels of NF-jB-p65 and Flag-Nucling/Flag-NuclingDC/Flag-NuclingDN in cellular lysates were confirmed by western
blotting. (D) COS7 cells were transfected with Flag-Nucling, Flag-NuclingDC or Flag-vector. Lysates were prepared and incubated with the anti-
Flag M5 for coimmunoprecipitation assay. The expression levels of endogenous IjBa in cellular lysates were confirmed by western blotting.
The expression of Flag-Nucling and Flag-NuclingDC was confirmed by western blotting with anti-Flag M5.
Nucling regulates cellular distribution of NF-jB L. Liu et al.
1462 FEBS Journal 276 (2009) 1459–1470 ª 2009 The Authors Journal compilation ª 2009 FEBS
with NF-jB-p50, NF-jB-p105 and NF-jB-p65, possi-
bly through its N-terminal domain containing ankyrin
repeats.
To confirm the presence of a protein complex
containing Nucling and NF-jB in primary cells, the
cytosolic fraction of WT or Nucling

)/)
MEFs was sep-
arated under nondenaturing conditions. Immunoblots
of these native gels revealed Nucling or NF-jB-p50
immunoreactivity within several distinct complexes in
fractions from TNFa-treated WT MEFs (Fig. 4A). We
identified five bands, which were reactive to both
antibodies (Fig. 4Aa–e). Although bands a and b were
specifically detected in WT MEFs (arrowhead), bands
c–e were detected in both WT and Nucling
)/)
MEFs.
Thus bands c–e were considered to be nonspecific. As
a next step, cytosolic proteins from TNFa-treated WT
or Nucling
)/)
MEFs were fractionated into native
complexes under nondenaturing conditions (first
dimension), and subsequently separated into individual
components (second dimension) by placing a native gel
slice at a horizontal position as a stack above an SDS-
PAGE denaturing gel (Fig. 4B). Second-dimension gels
were also transferred and immunoblotted, confirming
that Nucling (molecular mass, 160 kDa) was present in
A
BC
j
i
h
g

f
j’
Fig. 4. Nucling forms a complex with NF-jB-p50 in primary MEF cells. (A) Nucling-containing or NF-j B-p50-containing complexes were
observed in TNFa-treated WT MEFs, but not in Nucling
)/)
MEFs on native electrophoresis. Fifty micrograms of the cytosolic fraction were
subjected to nondenaturing electrophoresis (first dimension). Proteins were transferred and immunoblotted with a rabbit serum against
Nucling (anti-mNucl.mid). The arrows beside the photographs indicate the positions of the bands common to mNucl.mid and anti-NF-kB1
immunoreactive membranes. Common complexes in lanes 1, 3, 5 and 7 are marked with arrowheads. (B) Cytosolic fractions prepared from
WT MEFs (+/+) or Nucling
)/)
MEFs ()/)) were analyzed as the second dimension. Immunoblotting of the second-dimension gel of cytosolic
proteins from the TNFa-treated WT MEFs (left panel) or Nucling
)/)
MEFs (right panel) revealed the presence of Nucling (160 kDa) and p50
(50 kDa, h and i) in the complexes indicated as a, b and c in the native first-dimension gel of WT MEFs (A). Arrows a–e indicate the positions
of bands a–e in (A). Some other high molecular mass dots were detected in the membrane of WT MEFs for p50 (f and g). Arrows in the
lower panel for WT MEFs (left panel) indicate Nucling, which was not detected in Nucling
)/)
MEFs (right panel). p65 was detected in WT
and Nucling
)/)
MEFs (arrows j and j¢). IjBa was also detected in both MEFs (arrows in IjBa panels).
L. Liu et al. Nucling regulates cellular distribution of NF-jB
FEBS Journal 276 (2009) 1459–1470 ª 2009 The Authors Journal compilation ª 2009 FEBS 1463
complexes a, b and c in Fig. 4A (arrows in the bottom
panel in Fig. 4B). We also confirmed the presence of
p50 in complexes a and c (arrows h and i in Fig. 4B).
We observed two spots immunoreactive to the p50
antibody at the position of band c (arrows f and g in

Fig. 4B). In addition, there was a spot for p65 in the
lane for band c (arrow j in Fig. 4B); however, it was
observed in the same position in the membrane of
Nucling
)/)
MEFs (arrow j¢). We could not detect IjBa
in the lanes for bands a–c (Fig. 4B). Thus, we con-
cluded that Nucling interacts with NF-jB-p50 and
possibly with NF-jB-p65 in primary cells.
Nucling influences the localization of NF-jB
To verify the interaction between Nucling and NF-jB
proteins in vivo, we next examined whether the ectopic
expression of Nucling affects the distribution of p65 or
p50, which are overexpressed in COS7 cells. A double
immunostaining assay was performed after the coex-
pression of Nucling/p50 (Fig. 5A–D) or Nucling/p65
(Fig. 5E–H). The transfection of NF-jB alone (p50 or
p65) resulted in a nuclear localization (Fig. 5A,E).
However, cytoplasmic staining of NF-jB (p50 or p65)
was observed in approximately 80–85% of cells when
Nucling or NuclingDC was overexpressed by cotrans-
fection (Fig. 5B,C,F,G). However, when NuclingDN
was coexpressed with NF-jB (p50 or p65), over 65%
of cells showed nuclear staining of NF-jB (p50 or
p65) (Fig. 5D,H). The cytosolic or nuclear localization
of NF-jB (p50 and p65) was quantified and analyzed
statistically (Fig. 5I,J). The results clearly showed that
the ectopic expression of Nucling inhibits the nuclear
localization of NF-jB in vivo. In addition, the distribu-
tion of ectopic NF-jB apparently overlapped with that

of Nucling (Fig. 5B,F) or NuclingDC (Fig. 5C,G), but
not with that of NuclingDN (Fig. 5D,H), suggesting
that Nucling interacts with NF-jB through its N-ter-
minal region, and the interaction is sufficiently strong
to regulate the cellular localization of NF-jB.
To assess the potential of endogenous Nucling
to regulate the nuclear translocation of endogenous
NF-jB, we investigated the endogenous distribution of
NF-jB (p65 and p50) in WT and Nucling
)/)
MEFs
(Fig. 5K–N,K¢–N¢). In WT MEFs, both p50 and p65
were found exclusively in the cytoplasm (Fig. 5K,
K¢,M,M¢). In contrast, in Nucling
)/)
MEFs, p50 and
p65 were distributed throughout the cell, including the
nucleus and cytoplasm (Fig. 5L,L¢,N,N¢).
Based on these observations, we concluded that
Nucling physically interacts with and can regulate the
intracellular localization of NF-jB.
Nucling is important for the strict regulation of
the classical NF-jB signaling pathway
To verify the physiological function of Nucling in the
classical NF-jB signaling pathway, we investigated the
changes in the expression of p65, p50 and IjBa in cyto-
solic and nuclear lysates of MEFs following TNFa
treatment. A western blot analysis and electrophoretic
mobility shift assay (EMSA) indicated that the NF-jB
signaling pathway in WT MEFs was functioning nor-

mally. After TNFa treatment, a decrease in IjBa (lane 3
in Fig. 6A, bar for IjBa at 15 min in Fig. 6B) followed
by the expression of IjBa (lane 5 in Fig. 6A, bar for
IjBa at 30 min in Fig. 6B), increases in nuclear p50 and
p65 (lane 3 in Fig. 6A, bar for p50 and p65 at 15 min in
Fig. 6B) and an increase in NF-jB complexes (p65/p50
and p50/p50) (lane 4 in Fig. 6C) were evident in WT
MEFs. In Nucling
)/)
MEFs, a decrease in IjBa (lane 4
in Fig. 6A) was observed following TNFa treatment,
but it was not as marked as that in WT MEFs (lane 4 in
Fig. 6A). In addition, the nuclear translocation of p65
and p50 after TNFa treatment was impaired in
Nucling
)/)
MEFs (lanes 4 and 6 in Fig. 6A, bar for p50
and p65 at 15 min in Fig. 6B). Although the relative
levels of p65 and p50 were lower in the nuclear lysate
from Nucling
)/)
MEFs than in that from WT MEFs
(Fig. 6B), intense signals for p65 in the cytosol were
Fig. 5. Nucling mediates the localization of NF-jB. (A–D) COS7 cells were cotransfected with Flag-Vector/Flag-Nucling/Flag-NuclingDC/
Flag-NuclingDN and pHA-p50, respectively. Flag-Nucling and the mutant of Flag-Nucling were detected with an FITC-conjugated monoclonal
antibody against the Flag epitope (green). HA-p50 was detected with antibody against HA and Texas Red-conjugated secondary antibody
(red). Although the transfected cells were cultured in appropriate media with the pan-caspase inhibitor (zVADfmk), some cells transfected
with Flag-Nucling showed an apoptotic morphology (e.g. A). (E–H) COS7 cells were cotransfected with Flag-Vector/Flag-Nucling/Flag-
NuclingDC/Flag-NuclingDN and the p65 expression vector, respectively. Flag-Nucling and the mutant of Flag-Nucling were detected with an
FITC-conjugated monoclonal antibody against the Flag epitope (green). NF-jB-65 was detected with p65 antibody and rhodamine-conjugated

secondary antibody (red). Representative, double-stained images are shown. Scale bars represent 20 lm (A–H, K–N) or 10 lm(K¢–N¢). (I, J)
Bar graphs show the cytosolic (filled bar) or nuclear (open bar) localization of NF-jB-p50 (I) and NF-jB-p65 (J). The data are the
means ± standard deviation for three independent experiments. (K–N) Nucling
+/+
and Nucling
)/)
MEFs were cultured and analyzed by immu-
nofluorescent staining. Endogenous NF-jB-p65 and NF-jB-p50 were detected with a primary antibody against p65 and p50, respectively,
and an FITC-conjugated secondary antibody.
Nucling regulates cellular distribution of NF-jB L. Liu et al.
1464 FEBS Journal 276 (2009) 1459–1470 ª 2009 The Authors Journal compilation ª 2009 FEBS
observed in Nucling
)/)
MEFs (lanes 4 and 6 in Fig. 6A).
There was little or no upregulation of NF-jB (p65/p50
and p50/p50) activity on TNFa treatment (Fig. 6C).
These results strongly suggest that Nucling is not just a
suppressor of NF-jB in resting cells, but a regulator for
the activation of NF-jB in stimulated cells.
A
B
C D
E
I
J
F
G
H
K
L

M
N
K'
L'
M'
N'
L. Liu et al. Nucling regulates cellular distribution of NF-jB
FEBS Journal 276 (2009) 1459–1470 ª 2009 The Authors Journal compilation ª 2009 FEBS 1465
p65/β-actin
*
**
0
5-
0
5-
10-
0
1-
2-
*
IκBα/β-actin
IκBα/β-actin
0
1-
p50/β-actin
p50/β-actin
p65/β-actin
0
5-
0

4-
2-
Nucling:
TNFα:
Cytosol
Nucleus
Cold
Nucling+/+ MEF
Nucling-/- MEF
TNFα: 0 min
15 min 30 min
p65 in nucleus /total (%)
**
**
50
100
p50 in nucleus/total (%)
IκBα in cytosol/total (%)
TNFα: 0 min
15 min 30 min
0
0
50
90
**
TNFα: 0 min
15 min 30 min
0
50
90

A
C
B
Nucling regulates cellular distribution of NF-jB L. Liu et al.
1466 FEBS Journal 276 (2009) 1459–1470 ª 2009 The Authors Journal compilation ª 2009 FEBS
In resting Nucling
)/)
cells (Nucling
)/)
MEFs), the
activation of NF-jB (p65/p50) was upregulated
(Fig. 6B). Then, we investigated the expression of IjBa
in Nucling
)/)
and WT MEFs. However, the level in
Nucling
)/)
MEFs was almost the same as that in WT
MEFs in the cytosol or in total, with a slight but sig-
nificant difference in the nucleus (Fig. 6A, lanes 1 and
2, Fig. 6B). A decrease in IjBa was observed in
Nucling
)/)
MEFs treated with TNFa (Fig. 6A, lane 4),
indicating the degradation of IjBa. However, the
decrease was less marked in Nucling
)/)
MEFs than in
WT MEFs. As IjBa itself is regulated by NF-jB,
IjBa expression may be induced following the degra-

dation caused by TNFa, as we observed the expression
of IjBa at 30 min in Fig. 6A (lane 5 in cytosol). The
activation of NF-jB in resting cells may suppress the
dramatic decrease in the level of IjBa in Nucling
)/)
MEFs. In addition, there was no difference between
WT and Nucling
)/)
MEFs in the cellular expression of
IjBa by immunocytochemistry (data not shown). A
considerable amount of IjBa was detected in the
nucleus of resting cells (Fig. 5B, lanes 1 and 2, and
data not shown for immunocytochemistry).
Discussion
In this study, we have demonstrated that Nucling neg-
atively regulates the activation of NF-jB. The inhibi-
tory function proposed for Nucling in Fig. 1 was
based on several publications [12–14]. We hypothesized
that the inhibitory effect might be caused by the cyto-
plasmic retention of NF-jB via interaction with
Nucling. In order to check the importance of the inter-
action between Nucling and NF-jB, we performed a
two-dimensional native SDS-PAGE analysis in addi-
tion to immunoprecipitation. In this experiment, we
confirmed that Nucling and NF-jB-p50 form a com-
plex in primary cells (MEFs).
The activation of NF-jB increases the expression of
genes that encode proinflammatory mediators, called
cytokines, and activates genes that regulate the balance
between cell proliferation and cell death. COX2, as an

NF-jB-regulated proinflammatory gene, is expressed
essentially in inflammatory lesions and tumor tissues
[11,15]. TNFa is regarded as a critical mediator, bridg-
ing inflammation and tumorigenesis. The inhibition of
tumor and/or stromal TNFa may provide a novel ther-
apeutic strategy for cancer [8,16]. In this study, we
demonstrated that the expression of not only galectin-
3, but also COX2 and TNFa, was mediated by
Nucling via the inhibition of the activation of NF-jB.
Moreover, the upregulated expression of galectin-3,
COX2 and TNFa may play an important role in the
pathological changes in Nucling
)/)
mice. We have
observed several inflammatory conditions in some
Nucling
)/)
mice, including preputial gland abscesses
[4], liver dysfunction and hepatic diseases (data not
shown). Further detailed studies are necessary to reveal
precisely how these NF-jB-regulated mediators are
involved in the inflammation and tumorigenesis
induced by a deficiency of Nucling.
In this study, we confirmed the upregulation of TNFa
in three independent clones of resting Nucling
)/)
MEFs.
However, the upregulation of COX2 in Nucling
)/)
MEFs was clone dependent. Some redundant pathways

for the regulation of COX2 may occasionally work in
resting Nucling
)/)
MEFs. However, impairment of the
upregulation of TNFa and COX2 in Nucling
)/)
MEFs
following treatment with TNFa was consistently
observed in every case, indicating the importance of Nu-
cling for the activation of the NF-jB pathway.
Nucling was originally characterized as an apopto-
sis-inducing factor [2]. We observed that its overex-
pression led to apoptosis in several cell lines, including
COS7, NIH3T3 and HEK293 in the absence of a cas-
pase inhibitor ([2] and data not shown). We revealed
that Nucling reduces the expression of galectin-3, an
antiapoptotic molecule, and increases that of Apaf1,
an apoptosis-activating factor [2]. Here, we have
shown that Nucling is also a key regulator for NF-jB
Fig. 6. Nucling is important for the regulation of the NF-jB signaling pathway. (A) WT MEFs (+/+) or Nucling
)/)
MEFs ()/)) were treated
with TNFa for 0, 15 or 30 min, and cytosolic and nuclear cell lysates were prepared. Western blotting was performed to check the expres-
sion levels of NF-jB-p65, NF-jB-p50, IjBa or b-actin in the lysates. The levels of b-actin revealed that the amounts of applied cell lysates
were nearly the same. Relative protein levels are shown in a bar graph below each of the representative images. Three individual trials were
performed and analyzed. Data are presented as the mean ± standard error of the mean. *P < 0.05; **P < 0.005. (B) The bar charts show
the relative protein levels of p65, p50 and IjBa in the nucleus/total (for p65 and p50) or in the cytosol/total (for IjBa) (total = cyto-
sol + nucleus). Values relative to the level of b-actin were used for the analysis. (C) NF-jB (p65/p50 and p50/p50) DNA-binding activity in
nuclear extracts from WT MEFs (+/+) or Nucling
)/)

MEFs ()/)) with or without TNFa treatment. Where indicated, 4 lg of specific antibody
(anti-p65 in lanes 10 and 11, and anti-p50 in lanes 12 and 13) were added to the reaction mixture to demonstrate the specific binding of the
NF-jB-p65 and NF-jB-p50 proteins. For competition experiments, 100 ng of double-stranded cold competitor (cold probe) was added to the
reaction (lanes 8 and 9). Results from one of two independent experiments are shown.
L. Liu et al. Nucling regulates cellular distribution of NF-jB
FEBS Journal 276 (2009) 1459–1470 ª 2009 The Authors Journal compilation ª 2009 FEBS 1467
during TNFa treatment. NF-jB (p65) was consistently
activated in Nucling
)/)
cells. However, the activation
of p65 by TNFa was partly impaired in Nucling
)/)
cells (Fig. 6). These observations indicate that Nucling
is critical for the cellular distribution of NF-jB during
its activation. Contrary to the result for p65, the
upregulation of p50 and p105 expression was not so
dramatic in the nuclear fraction of resting Nucling
)/)
MEFs compared with WT MEFs (Fig. 6A,B, and data
not shown). However, we also observed the nuclear
distribution of p50/p105 in Nucling
)/)
MEFs
(Fig. 5L). From the results of immunoprecipitation
(Fig. 3) and two-dimensional nondenaturing gel elec-
trophoretic analysis (Fig. 4), we consider the interac-
tion between p50 and Nucling to be much stronger
than that between p65 and Nucling. In addition, Nu-
cling usually distributes in the perinucleus, not the
nucleus, and is harvested in the nuclear fraction. On

the basis of these observations, we speculate that a
considerable amount of cytoplasmic p50 interacting
with Nucling can be recovered in the nuclear fraction.
This may explain why there was no dramatic difference
between Nucling
)/)
and WT MEFs with regard to the
expression of p50 in the nuclear fraction.
Previously, we have also observed that the expres-
sion of Nucling is upregulated by several cellular stres-
sors, including H
2
O
2
, UV irradiation and anoikis ([2]
and data not shown). Nucling is particularly sensitive
to TNFa. As observed here, TNFa induces the activa-
tion of NF-jB. The upregulation of Nucling expres-
sion may be important for the regulation of the
activation of NF-jB following treatment with TNFa.
According to the standard model of classical NF-jB
signaling, the activation of NF-jB is promoted by the
degradation of the inhibitors of NF-jB, the IjBs. This
family of mainly cytoplasmic inhibitors currently has
eight members: IjBa,IjBb,IjBc,IjBe,IjBf, Bcl-3,
as well as p105 and p100. In unstimulated cells, NF-jB
forms a complex with the IjBs and is thereby locked
into the cytoplasm. IjBa is the best-characterized IjB,
and is also a target of NF-jB itself. Thus, we focused
on IjBa for further investigation. We observed a

decrease in IjBa following TNFa treatment in
Nucling
)/)
MEFs, indicating that the degradation of
IjBa is independent of Nucling. We actually observed
a change in the expression of IjBa in Nucling
)/)
MEFs during TNFa treatment, which can be explained
by an impairment of the regulation of NF-jB
signaling.
In conclusion, we propose that Nucling is important
for the regulation of the distinct intracellular
distribution of NF-jB in its resting and activated
states.
Materials and methods
Cell lines and culture conditions
HEK293, MEF and COS7 cells were used. The cells were
maintained in DMEM with 100 unitsÆmL
)1
penicillin,
100 lgÆmL
)1
streptomycin, 5 lm mercaptoethanol and 10%
(v/v) fetal bovine serum. They were cultured in plastic tissue
culture plates at 37 °C in an atmosphere of 5% CO
2
/95% air
in a humidified incubator. For propagation, cultures were
split, and the growth medium was replenished every 3–4 days.
NF-jB reporter assay

HEK293 cells cultured on 24-well plates (3 · 10
5
cells) were
transfected with a reporter plasmid comprising three repeats
of the NF-jB site upstream of a minimal TK promoter and
a firefly luciferase gene in the pGL-2 vector, and the vector
phRL-TK comprising a TK promoter upstream of a renilla
luciferase gene [17], together with the Nucling plasmid or a
negative control vector. PhRL-TK was used to exclude
nonspecific global effects of the transfections on transcrip-
tion. After 16 h, the transfected cells were treated with or
without TNFa (10 ngÆmL
)1
), IL-1b (5 ngÆmL
)1
) or LPS
(100 ngÆmL
)1
) for 1 h, and then harvested in a luciferase
lysis buffer. Luciferase assays were performed with the
Dual-LuciferaseÒ Reporter Assay System (Promega,
Tokyo, Japan) using a luminometer (Lumat LB 9507;
Berthold Technologies, Wildbad, Germany).
Subcellular fractionation and western blotting
Subcellular fractionation was performed as described previ-
ously [1]. In brief, semiconfluent cells were washed twice
with 5 mL of cold NaCl/P
i
, harvested by scraping, divided
into two tubes and centrifuged (100 g for 5 min). Half of

the cell pellets were harvested as whole-cell extracts. The
remaining cell pellets were resuspended in 1 mL of cold
hypotonic buffer (42 mm KCl, 10 mm Hepes and 5 mm
MgCl
2
) supplemented with protease inhibitor cocktail
(0.1 mm phenylmethanesulfonyl fluoride, 2 lgÆmL
)1
leupep-
tin and aprotinin) and incubated for 30 min on ice. Nuclei
were removed by centrifugation at 600 g for 10 min (nuclear
fraction). Cytosolic fractions were prepared as supernatants
separated by additional centrifugation at 100 000 g for
90 min. Whole-cell and nuclear pellets were resuspended in
cold extraction buffer (1% NP-40, 0.5% sodium deoxycho-
late and 0.1% SDS in 2 · NaCl/P
i
) supplemented with pro-
tease inhibitor cocktail, and the pellets were broken by
passing the suspension through a 26G needle. Following
incubation on ice for 30 min, cell extracts were obtained as
supernatants by centrifugation at 100 000 g for 5 min. The
protein concentrations of the extracts were estimated using
the BCA kit (Pierce, Rockford, IL, USA). Ten micrograms
of protein were loaded on to an SDS-PAGE gel, transferred
Nucling regulates cellular distribution of NF-jB L. Liu et al.
1468 FEBS Journal 276 (2009) 1459–1470 ª 2009 The Authors Journal compilation ª 2009 FEBS
onto an ImmobilonÔ Transfer Membrane (Millipore, Bed-
ford, MA, USA), and western blotting was performed. The
antibodies used were an NF-jB p65 mouse monoclonal

antibody (F-6) (Santa Cruz Biotechnology, Santa Cruz, CA,
USA), a rabbit polyclonal antibody against IjBa (C-21)
(Santa Cruz Biotechnology), a rabbit polyclonal antibody
against COX2 (Cayman Chemical, Ann Arbor, MI, USA)
and an antibody against TNFa (R&D Systems, Minnea-
polis, MN, USA). Western blot analysis was carried out
according to standard procedures using an ECL detection
kit (Amersham-Pharmacia, Uppsala, Sweden).
Coimmunoprecipitation
COS7 cells were transiently cotransfected with plasmids using
the NucleofectorÔ system (Amaxa, Ko
¨
ln, Germany). At 16 h
post-transfection, cells were harvested and lysed in lysis buffer
[50 mm Tris/HCl, pH 7.4, 150 mm NaCl, 0.25% deoxycholic
acid, 1% NP-40, 1 mm EDTA and CompleteÔ protease
inhibitor cocktail (Roche Diagnostics, Tokyo, Japan)]. Cell
extracts were clarified by centrifugation at 12 000 g for
20 min at 4 °C, and the supernatant was immunoprecipitated
with anti-Flag M2 affinity gel (Sigma-Aldrich, Tokyo, Japan)
by rotating overnight at 4 °C. The beads were washed 10
times with lysis buffer and suspended in SDS sample loading
buffer. Subsequently, the lysates and coimmunoprecipitated
samples were subjected to an immunoblot assay with anti-
Flag M2 horseradish peroxidase conjugate (Sigma) and
anti-HA horseradish peroxidase conjugate (3F10) (Roche).
RT-PCR and real-time RT-PCR
Total RNA from MEFs was extracted from Nucling
+/+
and Nucling

)/)
mice using TRIZOL
Ò
Reagent (Invitrogen
Life Technologies, Carlsbad, CA, USA). cDNA synthesis
was performed using a Superscript Preamplification System
for First Strand cDNA Synthesis Kit (Invitrogen). The fol-
lowing primers were used: for TNFa,5¢-ATGAGCACAG
AAAGCATGATCC-3¢ and 5¢-CCAAAGTAGACCTGCC
CGGACTC-3¢; for COX2, 5¢-AAAACCGTGGGGAATGT
ATGAGC-3¢ and 5¢-GATGGGTGAAGTGCTGGGCAA
AG-3¢; for b-actin, 5¢-TTGTAACCAACTGGGACGATAT
GG-3¢ and 5¢-GATCTTGATCTTCATG GTGCTAGG-3¢.
PCR was performed as follows: for TNFa, 30 cycles
(94 °C, 1 min; 60 °C, 45 s; 72 °C, 2 min); for COX2, 25
cycles (94 °C, 20 s; 60 °C, 40 s; 72 °C, 40 s); for b-actin, 25
cycles (95 °C, 1 min; 57 °C, 2 min; 72 °C, 2 min). PCR was
performed on a DNA thermal cycler (GeneAmp PCR system
9700; Applied Biosystems, Foster City, CA, USA). For real-
time RT-PCR, transcript levels of TNFa, COX2 and b-actin
were examined using a PTC-200 DNA Engine Cycler (MJ
Research Incorporated, Waltham, MA, USA) with SYBR
Premix Ex Taq (Takara, Kyoto, Japan). Results were calcu-
lated with the software of the DNA Engine Opicon System
(Roche Molecular System, Inc., Alameda, CA, USA).
Confocal immunofluorescence microscopic
analysis
COS7 cells were plated on to Lab-Tek Chamber Slides
(Nalge Nunc International, Rochester, NY, USA), tran-
siently cotransfected with Flag-Nucling/HA-NF-jB-p50 or

Flag-Nucling/NF-jB-p65, Flag-NuclingDC/HA-NF-jB-p50
or Flag-NuclingDC/NF-jB-p65, Flag-NuclingDN/HA-NF-
jB-p50 or Flag-NuclingDN/NF-jB-p65, or Flag-vector/HA-
NF-jB-p50 or Flag-vector/NF-jB-p65 (Fig. 1A), and then
cultured overnight. First, the cells were fixed with 4% para-
formaldehyde in NaCl/P
i
for 15 min at room temperature
and immersed in NaCl/P
i
containing 0.1% Triton X-100 for
3 min on ice. For blocking, the cells were incubated with 3%
bovine serum albumin in NaCl/P
i
for 1 h. Then, they were
incubated with the anti-HA (3F10) or NF-jB-p65 antibody
(F-6) overnight at room temperature, and the slides were
incubated with Texas Red-conjugated anti-goat IgG (ICN
Biomedicals, Inc., Aliso Viejo, CA, USA) or rhodamine-con-
jugated anti-mouse IgG (Chemicon International, Inc.,
Temecula, CA, USA) for 1 h. Next, the cells were incubated
with fluorescein isothiocyanate (FITC)-conjugated anti-
FLAG M2 (Sigma) overnight at 4 °C. Finally, cells on the
slides were mounted with anti-fade solution, sealed and
examined using a confocal laser scanning microscope and
software (Leica TCS NT, Heidelberg, Germany). Between
each step, the slides were washed three times with NaCl/P
i
buffer.
EMSA

The NF-j B oligonucleotide (5¢-AGTTGAGGGGACT
TTCCCAGGC-3¢, Promega) was labeled with [c-
32
P]ATP
(NEN Life Science Products, Inc., Boston, MA, USA) using
T4 polynucleotide kinase. MEF nuclear extracts (1 lgÆlL
)1
)
were added to 5 lL of gel shift binding buffer [20% (v/v)
glycerol, 5 mm MgCl
2
, 2.5 mm EDTA, 2.5 mm dithiothrei-
tol, 250 mm NaCl, 50 mm Tris/HCl, pH 7.5, and
0.25 mgÆmL
)1
poly(dI-dC)]. For competition or gel shift
experiments, binding was performed in the presence of 1 lL
(1.75 pmolÆlL
)1
) of unlabelled NF-jB oligonucleotide or
5 lg of NF-jB-p65 antibody (F-6) (Santa Cruz Biotechnol-
ogy) and NF-jB-p50 antibody (E-10) (Santa Cruz Biotech-
nology), respectively. The mixture was incubated for 20 min
on ice before the addition of 1 lLof
32
P-labelled NF-jB
oligonucleotide as probe. The resulting complexes were
resolved by electrophoresis on a 4% polyacrylamide gel.
Finally, the gel was dried and analyzed by autoradiography.
Two-dimensional native/denaturing PAGE of

protein complexes
WT or Nucling
)/)
MEF cell lysates were fractionated into
a cytosolic fraction as described previously [1]. The lysate
was resolved onto 2–15% precast gels (Daiichi Pure Chemi-
L. Liu et al. Nucling regulates cellular distribution of NF-jB
FEBS Journal 276 (2009) 1459–1470 ª 2009 The Authors Journal compilation ª 2009 FEBS 1469
cals Co. Ltd., Tokyo, Japan). For western blotting, the
native gel was soaked in blot buffer (20 mm Tris-base,
150 mm glycine and 0.08% SDS) for 10 min, and denatured
proteins were transferred to polyvinylidene difluoride mem-
branes (Millipore) in the same buffer using the semidry
blotting technique. Immunodetection was carried out
according to standard procedures and was visualized by
the ELC method (Amersham-Pharmacia). For the two-
dimensional analysis, individual lanes were cut out
from the first-dimension native gel and layered on top of a
15–25% gradient resolving gel, and a 7% stacking gel was
poured over and around the native gel slice.
Acknowledgements
We thank T. Fujita for the plasmid containing NF-jB-
p65 and NF-jB-p50 cDNA, and T. Takemori and
M. Matsumoto for the NF-jB-dependent luciferase
pGL-2 vector. This work was supported by a Grant-
in-Aid for Scientific Research and a Grant for the 21st
Century COE Program from the Ministry of Educa-
tion, Science, Sports and Culture of Japan for the
Promotion of Science, and a Research Grant from the
Ministry of Health, Labor and Welfare of Japan.

References
1 Sakai T, Liu L, Shishido Y & Fukui K (2003) Identifica-
tion of a novel, embryonal carcinoma cell-associated
molecule, nucling, that is up-regulated during cardiac
muscle differentiation. J Biochem (Tokyo) 133, 429–436.
2 Sakai T, Liu L, Teng X, Mukai-Sakai R, Shimada H,
Kaji R, Mitani T, Matsumoto M, Toida K, Ishimura K
et al. (2004) Nucling recruits Apaf-1/pro-caspase-9 com-
plex for the induction of stress-induced apoptosis. J Biol
Chem 279, 41131–41140.
3 Teng X, Sakai T, Liu L, Sakai R, Kaji R & Fukui K
(2006) Attenuation of MPTP-induced neurotoxicity and
locomotor dysfunction in Nucling-deficient mice via
suppression of the apoptosome pathway. J Neurochem
97, 1126–1135.
4 Liu L, Sakai T, Sano N & Fukui K (2004) Nucling
mediates apoptosis by inhibiting expression of
galectin-3 through interference with nuclear factor
kappaB signalling. Biochem J 380, 31–41.
5 Huxford T, Huang DB, Malek S & Ghosh G (1998)
The crystal structure of the IkappaBalpha/NF-kappaB
complex reveals mechanisms of NF-kappaB inactiva-
tion. Cell 95 , 759–770.
6 Jacobs MD & Harrison SC (1998) Structure of an
IkappaBalpha/NF-kappaB complex. Cell 95, 749–758.
7 Baldwin AS Jr (1996) The NF-kappa B and I kappa B
proteins: new discoveries and insights. Annu Rev
Immunol 14, 649–683.
8 Kaltschmidt B, Kaltschmidt C, Hofmann TG, Hehner
SP, Droge W & Schmitz ML (2000) The pro- or anti-

apoptotic function of NF-kappaB is determined by the
nature of the apoptotic stimulus. Eur J Biochem 267,
3828–3835.
9 Kucharczak J, Simmons MJ, Fan Y & Gelinas C (2003)
To be, or not to be: NF-kappaB is the answer – role of
Rel/NF-kappaB in the regulation of apoptosis. Onco-
gene 22, 8961–8982.
10 Pahl HL (1999) Activators and target genes of Rel/NF-
kappaB transcription factors. Oncogene 18, 6853–6866.
11 Williams CS, Mann M & DuBois RN (1999) The role
of cyclooxygenases in inflammation, cancer, and devel-
opment. Oncogene 18, 7908–7916.
12 Park MY, Jang HD, Lee SY, Lee KJ & Kim E (2004)
Fas-associated factor-1 inhibits nuclear factor-kappaB
(NF-kappaB) activity by interfering with nuclear trans-
location of the RelA (p65) subunit of NF-kappaB.
J Biol Chem 279, 2544–2549.
13 van Heel DA, Hunt KA, Ghosh S, Herve M & Play-
ford RJ (2006) Normal responses to specific NOD1-
activating peptidoglycan agonists in the presence of the
NOD2 frameshift and other mutations in Crohn’s dis-
ease. Eur J Immunol 36, 1629–1635.
14 Van Huffel S, Delaei F, Heyninck K, De Valck D &
Beyaert R (2001) Identification of a novel A20-binding
inhibitor of nuclear factor-kappa B activation termed
ABIN-2. J Biol Chem 276, 30216–30223.
15 Bauer MK, Lieb K, Schulze-Osthoff K, Berger M,
Gebicke-Haerter PJ, Bauer J & Fiebich BL (1997)
Expression and regulation of cyclooxygenase-2 in rat
microglia. Eur J Biochem 243, 726–731.

16 Szlosarek P, Charles KA & Balkwill FR (2006) Tumour
necrosis factor-alpha as a tumour promoter. Eur J
Cancer 42, 745–750.
17 Kashiwada M, Shirakata Y, Inoue JI, Nakano H, Oka-
zaki K, Okumura K, Yamamoto T, Nagaoka H & Take-
mori T (1998) Tumor necrosis factor receptor-associated
factor 6 (TRAF6) stimulates extracellular signal-regu-
lated kinase (ERK) activity in CD40 signaling along a
ras-independent pathway. J Exp Med 187, 237–244.
Supporting information
The following supplementary material is available:
Fig. S1. Nucling is involved in the regulation of the
NF-jB signaling pathway.
This supplementary material can be found in the
online version of this article.
Please note: Wiley-Blackwell is not responsible for
the content or functionality of any supplementary
materials supplied by the authors. Any queries (other
than missing material) should be directed to the cor-
responding author for the article.
Nucling regulates cellular distribution of NF-jB L. Liu et al.
1470 FEBS Journal 276 (2009) 1459–1470 ª 2009 The Authors Journal compilation ª 2009 FEBS