The chromosomal protein HMGN2 mediates
lipopolysaccharide-induced expression of b -defensins in
A549 cells
Lu-Xia Deng
1,2
, Gui-Xia Wu
1
, Yue Cao
1
, Bo Fan
1
, Xiang Gao
1
, Lin Luo
3
and Ning Huang
1,4
1 Research Unit of Infection and Immunity, West China School of Preclinical and Forensic Medicine, Sichuan University, Chengdu, China
2 Chongqing Lummy Pharmaceutical Co. Ltd, Chongqing, China
3 Department of Anesthesiology, West China College of Stomatology, Sichuan University, Chengdu, China
4 State Key Laboratory of Biotherapy, West China Hospital, Sichuan, Chengdu, China
Introduction
Antimicrobial peptides are produced by many different
tissues of the body when the innate immune system is
activated by lipopolysaccharide (LPS) or other inflam-
matory cytokines and chemokines. Defensins are small
cationic antimicrobial peptides which have usually been
categorized into three major families based on the
molecular structure: a-, b-, and h-defensins [1–3]. They
regulate both the innate and adaptive immune
responses, exhibiting broad-spectrum antimicrobial

member of the high mobility group superfamily that affects chromatin
function, modulates the expression of HBD-2 in A549 cells treated by lipo-
polysaccharide. Mechanistically, HMGN2 prolongs the retention time and
enhances the accumulation of nuclear factor jB p65 in the nucleus, and
promotes the acetylation of p65 through increasing histone acetyltransfer-
ase activity and enhancing p65-Ser536 phosphorylation. Additionally, chro-
matin immunoprecipitation reveals that HMGN2 and p65 synergistically
promote their specific binding to HBD-2 promoter, thereby affecting the
downstream transcription. Taken together, these results suggest that
HMGN2 acts as a positive modulator of nuclear factor jB signalling to
promote lipopolysaccharide-induced b-defensin expression.
Abbreviations
AA, anacardic acid; ChIP, chromatin immunoprecipitation; Co-IP, co-immunoprecipitation; Glut2, type 2 glucose transporters; HAT, histone
acetyltransferase; HBD, human b-defensin; HDAC, histone deacetylase; HMG, high mobility group; HMGA, HMG-AT-hook family; HMGB,
HMG-box family; HMGN2, high mobility group protein N2; IjBa, NF-jB inhibitor a; LPS, lipopolysaccharide; MAPK, mitogen-activated protein
kinase; NF-jB, nuclear factor jB; PE-H, pEGFPN1-HMGN2; Psi-H, pSilencer-HMGN2-2; siRNA2, small interfering RNA2; TNF-a,
tumour necrosis factor a; TSA, trichostatin.
2152 FEBS Journal 278 (2011) 2152–2166 ª 2011 The Authors Journal compilation ª 2011 FEBS
cells has been shown to be regulated through several
signalling pathways, such as the nuclear factor jB
(NF-jB) pathway, the p38 mitogen-activated protein
kinase (MAPK) pathway, the c-Jun N-terminal kinase
pathway and the phosphatidylinositol-3-kinase ⁄ AKT
pathway [7–11].
The chromatin-associated high mobility group
(HMG) superfamily includes the HMG-AT-hook
(HMGA) family, HMG-box (HMGB) family and
HMG nucleosome binding (HMGN) family and mod-
ulates a wide range of DNA-dependent activities such
as chromatin structure, post-translational modifica-

between HATs and HDACs thereby affecting p65 acet-
ylation; and (b) HMGN2 itself or through interaction
with p65 binds to the chromatin in the promoter
region of b-defensin genes to enhance HBD-2 expres-
sion. This research aimed to confirm the initial hypoth-
esis through transient transfection and luciferase
experiments, the activity of HAT and HDAC blocking
experiments, chromatin immunoprecipitation (ChIP)
assay and co-immunoprecipitation (Co-IP) etc.
Results
Gene expression profiles after HMGN2
knockdown or ⁄ and LPS stimulation in A549 cells
As the first step to characterizing the function of
HMGN2 in transcriptional regulation in response to
LPS, we prepared knockdown HMGN2 in A549 cells
and found that HMGN2 expression at both mRNA
and protein levels was significantly downregulated in
A549 cells treated by HMGN2-specific small interfer-
ing RNA2 (SiRNA-HMGN2-2) (Figs 1A,B and S1).
Next we employed a cDNA microarray to examine the
effect of reduced endogenous HMGN2 level and ⁄ or
LPS treatment on gene expression profiles in A549
cells (Figs S2 and S3). The results showed that
HMGN2 downregulation altered the expression of over
4% of 31 000 genes by twofold or more (Table 1).
B N M 1 2 3
HMGN2
Histone 3
B N 1 2 3
-actin

blank
Psi
Psi-H/PE-H
Psi-H/PE
PE-H
PE
Psi-H
blank
Maker
Psi
Psi-H/PE-H
Psi-H/PE
Fig. 1. Efficient knockdown of HMGN2 in
A549 cells. Representative RT-PCR (A) or
western blot (B) results showing that
HMGN2 mRNA and protein levels were
reduced by over 80% after siRNA2 treat-
ment. b-actin and histone H3 served as the
loading control for RT-PCR and western blot,
respectively. B, blank group; N, normal
group (without LPS treatment); M, marker.
1–3, different HMGN2 siRNA. (C) Represen-
tative RT-PCR and western blot results
showing the expression of HMGN2 mRNA
and protein in the different established sta-
ble A549 cells. (D) Values are presented as
mean ± SD for at least five independent
experiments performed in triplicate.
*P < 0.01 versus blank group.
L X. Deng et al. HMGN2 mediates expression of b-defensins

autosomal sex-reversal)
NM_000591.1 CD14 5.3 )1.0 4.1 CD14 antigen
NM_002228.3 JUN 4.8 )2.9 2.2 v-jun sarcoma virus 17 oncogene homolog (avian)
NM_001569.2 IRAK1 4.0 )1.4 2.3 Interleukin-1 receptor-associated kinase 1
NM_002751.5 TRAF6 )3.2 2.9 )1.3 TNF receptor-associated factor 6
NM_014261.1 TRIF 2.8 )1.4 1.9 TIR domain containing adaptor inducing interferon-beta
NM_003266.2 TLR4 4.9 )2.2 3.6 Toll-like receptor 4
NM_014294.3 TRAM1 1.1 1.4 2.6 Translocation associated membrane protein 1
NM_005252.2 FOS 3.2 1.9 5.7 v-fos FBJ murine osteosarcoma viral oncogene homolog
NM_003225.2 TFF1 1.2 3.2 4.8 Trefoil factor 1 (breast cancer, estrogen-inducible sequence
expressed in)
NM_005080.2 XBP1 3.0 1.3 4.9 X-box binding protein 1
BU739862 MED19 )0.9 )0.5 )1.1 Mediator of RNA polymerase II transcription, subunit 19
homolog (yeast)
NM_003187.3 TAF9 )0.5 )1.1 )1.2 TAF9 RNA polymerase II, TATA box binding protein (TBP)
associated factor, 32kDa
BU739862 LCMR1 0.6 1.6 2.8 Mediator of RNA polymerase II transcription, subunit 19
homolog (yeast)
BX537584 PC4 3.8 3.9 7.0 Activated RNA polymerase II transcription cofactor 4
NM_000125.1 ESR1 0.5 )3.1 )2.7 Estrogen receptor 1
NM_001904.2 CTNNB1 4.8 3.4 7.2 Catenin (cadherin-associated protein), beta 1, 88kDa
NM_001792.2 CDH2 4.2 2.6 7.7 Cadherin 2, type 1, N-cadherin (neuronal)
NM_003508.1 FZD9 2.8 1.7 5.1 Frizzled homolog 9 (Drosophila)
NM_012193.2 FZD4 3.14 2.1 5.1 Frizzled homolog 4 (Drosophila)
NM_004625.3 WNT7A 2.5 1.3 5.5 Wingless-type MMTV integration site family, member 7A
NM_003392.3 WNT5A 3.3 1.9 4.7 Wingless-type MMTV integration site family, member 5A
NM_002335.1 LRP5 1.4 1.9 4.2 Low density lipoprotein receptor-related protein 5
NM_002336.1 LRP6 1.4 2.3 3.4 Low density lipoprotein receptor-related protein 6
NM_153426.1 PITX2 0.4 4.1 4.5 Paired-like homeodomain transcription factor 2
NM_022454.2 SOX17 3.5 )2.4 1.8 SRY (sex determining region Y) box 17

in A549 cells
Since microarray analysis showed that the expression
of HBD-2 was significantly changed after LPS induc-
tion and HMGN2 knockdown, next we aimed to con-
firm these results by RT-PCR and western blot. We
employed a variety of A549 cells with stable transfec-
tions of pSilencer-HMGN2-2 (Psi-H), pEGFPN1-
HMGN2 (PE-H), control siRNA (Psi, PE), wild-type
A549 cells (blank), and reintroduction of HMGN2
expression vector or control vector to HMGN2 knock-
down A549 cells (Psi-H ⁄ PE-H or Psi-H ⁄ PE, respec-
tively; Figs 1C,D, S4 and S5). These cells were treated
with 0, 20, 40, 60, 80 or 100 lgÆmL
)1
LPS for 24 h
and then the RNA and protein were isolated for RT-
PCR and western blot analysis. The results demon-
strated that LPS induced HBD-2 expression in a dose-
dependent manner which can be abolished by
HMGN2 knockdown. However, reintroduction of
HMGN2 into HMGN2 knockdown cells recovered
LPS-induced expression of HBD-2 (Fig. 2A–D). Taken
together, these data suggest that HMGN2 is crucial
for LPS-induced HBD-2 expression.
To verify that the decreased levels of HBD-2 protein
and transcripts are indeed linked to the expression
levels of HMGN2, the plasmid expressing double-point
mutant HMGN2-S24, 28E was prepared (this protein
enters the nucleus but does not bind to chromatin);
then we examined the levels of HBD-2 expression level

somal dominant)
NM_005406.1 ROCK1 0.6 )2.4 )2.5 Rho-associated, coiled-coil containing protein kinase 1
NM_018890.2 RAC1 3.1 1.6 3.7 Ras-related C3 botulinum toxin substrate 1 (rho family, small
GTP binding protein Rac1)
BQ051103 HMGN2 )0.3 )8.5 )7.9 High mobility group nucleosomal binding domain 2
BM907805 H3F3A 3.1 )2.1 1.8 H3 histone, family 3A
BQ940876 HIST1H1C )0.3 )2.1 )2.4 Histone 1, H1c
L X. Deng et al. HMGN2 mediates expression of b-defensins
FEBS Journal 278 (2011) 2152–2166 ª 2011 The Authors Journal compilation ª 2011 FEBS 2155
cantly (Fig. 2E). Overall, the results support an impor-
tant role of HMGN2 with chromatin regulation in
inducible HBD-2 expression.
HMGN2 regulates NF-jB activity in A549 cells
To reveal potential mechanisms by which HMGN2
regulates LPS-induced HBD-2 expression, we first
examined whether HMGN2 could modulate NF-jB
activity in A549 cells because the promoter region of
HBD-2 contains four NF-jB binding sites [4]. We
observed that LPS led to increased NF-jB levels in the
nucleus and decreased NF-jB levels in the cytoplasm.
However, the LPS-induced change of NF-jB distribu-
tion was markedly attenuated by HMGN2 knockdown,
indicating that loss of HMGN2 inhibits LPS-induced
NF-jB accumulation in the nucleus. On the other
hand, gain of HMGN2 due to overexpression repro-
A
B
D
C
FE

*
*
*
*
*
*
*
*
20 40 60 80 100
HBD-2
/
β-actin ratio
0.2
0.8
0.6
0
0.4
1.0
HBD-2 Western blotting
*
*
*
*
*
*
*
*
*
*
Psi-H/PE group

HBD2
β-actin
HBD2
β-actin
HBD2
β-actin
HBD2
Psi-H group
B group
Psi-H/PE group
β-actin
HBD2
PE-H group
β-actin
HBD2
PE group
Psi group
0 20 40 60 80 100
Cell
Transfected
HMGN2
HBD-2
β-actin
HMGN2
–
A549
(PEGFPN1-HM
GN2)
HMGN2
–

blank group. (C) Representative western
blot results showing the expression of
HMGN2 protein in the different established
stable A549 cells. b-actin served as the load-
ing control. (D) Photodensitometric analysis
of western blot is presented as mean ± SD
for at least five independent experiments
performed in triplicate. *P < 0.01 versus
blank group. (E) Western blot analysis of
stably transfected HMGN2
)
A549 cells
expressing either HMGN2 or the HMGN2-
S24, 28E. A549 HMGN2
)
denotes control,
non-transformed HMGN2
)
cells. (F) Values
of western blots are expressed as relative
expression compared with b-actin and pre-
sented as mean ± SD for at least five inde-
pendent experiments performed in triplicate.
HMGN2 mediates expression of b-defensins L X. Deng et al.
2156 FEBS Journal 278 (2011) 2152–2166 ª 2011 The Authors Journal compilation ª 2011 FEBS
ducibly prompted the accumulation of NF-jB in the
nucleus (Fig. 3A–D).
To further confirm that HMGN2 modulates LPS-
induced NF-jB activation in A549 cells, A549 cell lines
harbouring a transient NF-jB-dependent luciferase

*
*
*
*
*
*
0
20 40 60 80 100
Psi-H/PE group
PE-H group
PE group
B group
Psi-H group
Psi group
Psi-H/PE-H group
A
B
Psi-H/
PE-H
PE
Psi
Psi-H/
PE
Psi-H
PE-H
Cytoplasm Nucleus
LPS stimulated concentrations (µg·mL
–1
)
2

PE
Psi-H
PE-H
0
0
5
5
1
5
1
5
3
0
3
0
6
0
6
0
1
2
0
1
2
0
1
8
0
1
8

B group
Psi-H group
Psi group
Psi-H/PE-H group
E
A fold induction
(compared to mock infection)
*
*
*
Psi-H/PE group
PE-H group
PE group
Bgroup
Psi-H group
Psi group
Psi-H/PE-H group
P
E
-
H
P
E
B
P
s
i
-
H
P

p65 in the different established stable A549 cells. (D) Photodensitometric analysis of western blot is presented as mean ± SD for at least
five independent experiments performed in triplicate. *P < 0.01 versus blank group. (E) HMGN2 promotes the transcription activity of NF-jB.
The cells were transfected with NF-jB luciferase reporter and treated by LPS. The luciferase activity is presented as mean ± SD for at least
five independent experiments performed in triplicate. *P < 0.01 versus blank group. The luciferase activity in the blank group untreated by
LPS (mock) was utilized as the control value.
L X. Deng et al. HMGN2 mediates expression of b-defensins
FEBS Journal 278 (2011) 2152–2166 ª 2011 The Authors Journal compilation ª 2011 FEBS 2157
HMGN2 modulates the acetylation of p65
The transcription factor NF-jB activity is known to
be regulated by reversible acetylation through HATs
and HDACs. Anacardic acid (AA) inhibits HAT activ-
ity while trichostatin (TSA) inhibits HDAC activity.
Therefore, we used these reagents to treat A549 cells.
Western blot analysis for p65-Lys310 acetylation in
AA-treated or TSA-treated cells demonstrated clearly
(Table 2) that the acetylation of p65-Lys310 was
decreased in HMGN2 knockdown cells with
100 lgÆmL
)1
LPS for 5, 15, 30, 60 or 120 min and
increased in HMGN2 overexpressing cells with
100 lgÆmL
)1
LPS for 60 and 120 min but the later dif-
ference was not statistically significant (Fig. 4A,B),
suggesting that HMGN2 promotes the acetylation of
p65-Lys310 mainly by enhancing HAT activity. Since
p65-Lys310 acetylation depends on p65-Ser536 phos-
phorylation, we also examined the effect of HMGN2
on p65-Ser536 phosphorylation and found that

whole cell extract
0.4
0.6
0.2
0
0.8
5 15 30 60 120
*
*
1.0
*
*
*
*
The acetylation level of p65-Lys
310
DC
Nucleus
Psi-H group
Psi-H/TSA group
B group
PE-H group
PE-H/AA group
LPS stimulated time (min)
LPS stimulated time (min)
*
*
*
Psi-H/TSA
PE-H group

*
*
60
Fig. 4. HMGN2 increases the HAT activity
to promote NF-jB activation. The A549 cells
were pre-incubated with 25 l
M AA for 4 h
or 100 n
M TSA for 18 h and treated with
100 lgÆmL
)1
LPS for 5, 15, 30, 60 or
120 min. (A) Representative western blot
results showing the acetylation of p65-
Lys310 in the indicated A549 cells. (B)
Photodensitometric analysis of western blot
is presented as mean ± SD for at least five
independent experiments performed in
triplicate. *P < 0.01 versus blank group.
P < 0.01 PE-H ⁄ AA versus PE-H group. (C)
Representative western blot results show-
ing the phosphorylation of p65-Ser536 in the
indicated A549 cells. (D) Photodensitometric
analysis of western blot is presented as
mean ± SD for at least five independent
experiments performed in triplicate.
*P < 0.01 versus blank group.
HMGN2 mediates expression of b-defensins L X. Deng et al.
2158 FEBS Journal 278 (2011) 2152–2166 ª 2011 The Authors Journal compilation ª 2011 FEBS
HMGN2 binds to HBD-2 promoter upon LPS

Thus we hypothesized that HMGN2 and p65 may pre-
assemble into a regulatory complex on HBD-2 pro-
moter. However, Co-IP experiments demonstrated that
p65 and HMGN2 did not assemble into a complex in
the nuclear compartment because HMGN2 antibody
efficiently precipitated HMGN2 from nuclear extracts
of A549 cells but the precipitates did not contain p65
(Fig. 6B). Likewise, anti-p65 immunoprecipitated p65
efficiently and the precipitates did not contain
HMGN2 (Fig. 6B). Next, we examined whether
HMGN2 and p65 share the same binding sites on
HBD-2 promoter. ChIP analysis indicated that the
antibodies to HMGN2 and p65 efficiently immunopre-
cipitated chromatin containing the respective proteins
(Fig. 6C), but the chromatin with HMGN2 enrichment
does not contain p65 (Fig. 6C, upper left panel) and
the chromatin with p65 enrichment does not contain
HMGN2 (Fig. 6C, lower right panel). Based on these
results we conclude that HMGN2 could not interact
with p65 in nuclear extracts and they did not share the
same chromatin binding site.
In contrast, results of reciprocal sequential ChIP
analysis revealed that chromatin that was sequentially
immunoprecipitated with HMGN2 and p65 antibody
was enriched in the DNA of HBD-2 promoter (P1)
while the HBD-2 promoter was enriched in chromatin
that was sequentially immunoprecipitated with anti-
p65 and anti-HMGN2 (Fig. 6D,E), suggesting that
HMGN2 and p65 mutually promote their binding to
the promoter of HBD-2.

P14
P13
P12
P11
P10
P9
P8
P7
P6
P5
P4
P3
P2
P1
P0
G1
G2
G3
G4
G5
G6
G7
G8
Fig. 5. HMGN2 is enriched in HBD-2 pro-
moter chromatin in A549 cells. (A) Enrich-
ment of each DNA sequence in the HMGN2
or HMGN1 immunoprecipitate relative to
the input DNA is normalized and plotted as
the position of the PCR primer pair within
the HBD-2 gene locus. Each point is an

consistent with those achieved through RT-PCR and
western blotting, including the change of HBD-2
expression. Compared with A549 cells stimulated by
LPS, HBD-2 expression was 50% less in HMGN2
knockdown cells and was over 30% higher in HMGN2
overexpressing cells. In addition, reintroduction of
HMGN2 re-expression led to the recovery of HBD-2
expression by over 70% in HMGN2 knockdown cells.
Overall, these findings prove that HMGN2 plays an
essential role in LPS-induced HBD-2 expression in
A549 cells.
Next we aimed to elucidate the molecular mecha-
nism by which HMGN2 regulates HBD-2 expression.
HBD-2 promoter contains several binding sites for
transcription factors including NF-jB, NF-IL-6 and
p65
HMGN2
IP
:
Fold enrichment at P1
A
B
C
Control IgG
Control IgG
α-p65
α-p65
Control lgG
Control lgG
No lgG

HBD-2 P1
Input
control IgG
α-p65
+–
α-HMGN2
–+
+–α-HMGN2
–+α-p65
1
st
: α-p65
2
nd
: α-HMGN2
1
st
: α-HMGN2
2
nd
: α-p65
1
st
ChIP
2
nd
ChIP
Normal
HMGN2
–

trol. (E) ChIP assay showing that HMGN2 or
p65 bound to HBD-2 chromatin at P1.
HMGN2 mediates expression of b-defensins L X. Deng et al.
2160 FEBS Journal 278 (2011) 2152–2166 ª 2011 The Authors Journal compilation ª 2011 FEBS
AP-1 [36–38]. Previous studies found that the upregu-
lation of HBD-2 promoter activity was mainly depen-
dent on NF-jB in A549 cells, and LPS is known to
induce the activation of NF-jB. Therefore, we pro-
posed that HMGN may mediate LPS-induced HBD-2
expression through the NF-jB signalling pathway. To
examine this possibility we first performed western blot
analysis to show that the accumulation of p65 protein
in the nucleus, indicative of NF-jB activation, was
increased in HMGN2 overexpresssing A549 cells but
decreased in HMGN2 knockdown cells. Next, we
employed NF-jB luciferase reporter assay to quantify
NF-jB activation [39]. The NF-jB-induced luciferase
activity was significantly diminished in HMGN2
knockdown cells and increased in HMGN2 overex-
pressing cells. Based on these data we could conclude
that HMGN2 is crucial for LPS-induced NF-jB acti-
vation.
NF-jB activation is known to be reciprocally regu-
lated by RelA ⁄ p65 acetylation and deacetylation medi-
ated by HATs and HDACs. HDACs and HATs are
enzymes that influence transcription by selectively de-
acetylating or acetylating the e-amino groups of lysine
located near the amino termini of core histone pro-
teins. Acetylation of p65 at lysines 218, 221 and 310
by HATs including the general transcriptional coacti-

LL-37, a member of the antimicrobial peptide that
protects the urinary tract against invasive bacterial
infection [43].
In the current study, AA (HAT inhibitor) was
selected to test whether blocking HATs diminishes the
acetylation level of the p65-Lys310 subunit in the PE-H
transfected A549 cells. If HMGN2 promotes the acety-
lation of p65 through increasing HAT activity, AA pre-
treatment would reduce the acetylation level of p65-
Lys310 in the nucleus in the PE-H ⁄ AA group compared
with the PE-H group. On the other hand, TSA (HDAC
inhibitor) was used to examine the acetylation level of
p65-Lys310 in the Psi-H transfected A549 cells. If
HMGN2 promotes the acetylation of p65 through
inhibiting HDAC activity, the TSA pretreatment would
augment the acetylation level of p65-Lys310 in the
nucleus in the Psi-H ⁄ TSA group compared with the
Psi-H group. Because p65-Lys310 acetylation was only
allowed when p65-Ser536 was phosphorylated, the p65
global phosphorylation status on Ser536 was also tested
by western blotting in five groups.
The results demonstrated that the amount of p65-
Lys310 acetylation in the nucleus of A549 cells in the
PE-H group was low compared with that in the pres-
ence of AA (Fig. 4A,B). Adding AA to the HMGN2
overexpressing cells helped to reduce the acetylation
level of p65-Lys310 in the nuclei to the normal level,
while adding TSA to the HMGN2 knockdown cells
did not bring the acetylation level of p65-Lys310 in the
nucleus to normal; the difference was not statistically

Moreover, a previous study had reported that it is likely
that homeostatic mechanisms, perhaps involving
HMGN2, compensate for the loss of the HMGN1 pro-
tein based on structural similarity among them. Indeed,
the ChIP analyses in that study indicated an increase of
HMGN2 in the Sox9 chromatin obtained from
HMGN1
) ⁄ )
mice, suggesting functional redundancy
among these proteins [31]. Moreover, the downregula-
tion of HMGN2 decreased the binding of p65 to HBD-
2 promoter significantly. To address the potential rela-
tionship between HMGN2 and p65 in the regulation of
the HBD-2 promoter region, we did reciprocal sequen-
tial precipitation experiments and observed that
HMGN2 and p65 specifically target the HBD-2 pro-
moter. Co-IP and ChIP experiments demonstrate that
HMGN2 does not interact with NF-jB directly and the
two proteins do not form a complex in the nucleus and
or share the same binding sites throughout the entire
chromatin. These findings are consistent with previous
studies reporting the relationship between HMGB1 and
glucocorticoid receptor in glucocorticoid receptor bind-
ing sites and HMGN3 and PDX1 (a Glut2 transcription
factor). In particular, HMGN3 and PDX1 mutually
reinforce their specific binding to the Glut2 chromatin
without interaction with each other or sharing the same
binding sites in Glut2 chromatin [28,44]. Based on these
data, we propose that HMGN2 and p65 mutually rein-
force their specific binding to the chromatin in the pro-

Cruz, CA, USA), p65 (for ChIP; Abcum, Cambridge, UK),
acetyl-p65 (Lys310) (Cell Signaling, Danvers, MA, USA),
b-actin (Santa Cruz), histone H3 (Abcum) and HBD-2
(Santa Cruz). The affinity pure HMGN2 antibodies were
prepared in our laboratory. Escherichia coli LPS was from
Sigma (St Louis, MO, USA).
Cell culture
Human lung adenocarcinoma A549 cells were obtained
from the School of Pharmaceutical Science, Sun Yat-Sen
University (Guangzhou, China). These cells were cultured
in Dulbecco’s modified Eagle’s medium (DMEM) supple-
mented with 10% fetal bovine serum (FBS) at 37 °Cina
humidified incubator containing 5% CO
2
.
siRNA transfection
Three HMGN2-specific siRNAs (1, 2 and 3) (SiRNA-HMG
N2-1: sense, 5¢-CUA AUA GAA UGU CUC CAA ATT-3¢;
antisense, 5¢-UUU GGA GAC AUU CUA UUA GTG-3¢;
SiRNA-HMGN2-2: sense, 5¢-AGU CAG GGU CGG CUU
GUG ATT-3¢; antisense, 5¢-UCA CAA GCC GAC CCU
GAC UTT-3¢; SiRNA-HMGN2-3: sense, 5¢-UAA UAG
AAU GUC UCC AAA GTT-3¢; antisense, 5¢-CUU UGG
AGA CAU UCU AUU AGT-3¢) and two non-specific con-
trol siRNAs (positive siRNA sense, 5¢-GAC TTC ATAA
GGCG CATGC-3 ¢; antisense, 5 ¢- GCATGCGCCTTAT
GAAGTC-3¢; negative siRNA sense, 5¢-AUU GUA UGC
GAU CGC AGA CTT-3¢; antisense, 5¢-GUC UGC GAU
CGC AUA CAA UGA-3¢) were designed and synthesized by
RiboBio (Guangzhou, China). The positive control siRNA

hybridization buffer. The samples were hybridized competi-
tively under coverslips to the microarray slides at 50 °C for
16 h in a dark humidity chamber and washed as follows:
2 · NaCl ⁄ Cit for 5 min, 0.5 · NaCl ⁄ Cit for 5 min, and
30% ethanol ⁄ water for 30 s at 42 °C. Hybridization experi-
ments were performed in duplicate using cDNA derived
from the four different groups.
Images were scanned and analysed by the Agilent
GeneChip Scanner. The original signals of the images
were normalized using the reference spots on the slide
and the fluorescence signals were balanced and amended
using the genespring software. Three biological replicates
(independent RNA isolations), each with a technical repli-
cate (dye swap), were performed for each condition.
Genes with a Cy5 ⁄ Cy3 signal ratio ‡ 2.0 were considered
upregulated, whereas those with a ratio £ 0.5 were down-
regulated.
Plasmid constructs
Short hairpin (sh) RNA constructs were prepared using pSi-
lencer 4.0-CMV (Ambion, Austin, TX, USA) as the vector.
The HMGN2-shRNA plasmids were named pSilencer-
HMGN2-1 and pSilencer-HMGN2-2, respectively (HM
GN2-shRNA-1: sense, 5¢-GATCCTCTGCGAGGTTGTC
TGCTATTCAAGAGATAGCAGACAACCTCGCAGAT
CA-3¢; antisense, 5¢-AGCTTGATCTGCGAGGTTGTCTG
CTATCTCTTGA ATAGCAGACAACCTCGCAGAG-3¢;
HMGN2-shRNA-2: sense, 5¢-GATCCA AATGGAGATGC
CAAAACATTCAAGAGATGTTTTGGCATCTCCATTT
TCA-3¢;antisense, 5¢-AGCTTGAAAATGGAGATGCCAA
AACATCTCTTGAATGTTTTGGCATCTCCATTTG-3¢).

based on the killing curve patterns
for A549 cells. Single clones were picked and tested for
Table 4. A549 cell groups in this study.
Groups Description
PE-H group A549 cells stably transfected with
PEGFPN1-HMGN2
PE group A549 cells stably transfected with
PEGFPN1
B group A549 cells treated by LPS
Psi-H group A549 cells stably transfected with
pSilencer-HMGN2-2
Psi group A549 cells stably transfected with
pSilencer
Psi-H ⁄ PE-H group A549 cells stably transfected with
pSilencer-HMGN2-2 and transfect-
ed with PEGFPN1-HMGN2
Psi-H ⁄ PE group A549 cells stably transfected with
pSilencer-HMGN2-2 and transfect-
ed with PEGFPN1
Table 3. A549 cell groups in microarray analysis.
Groups Description
Blank group (group A) A549 cells untreated
HMGN2 knockdown
group (group B)
A549 cells transfected with
HMGN2-specific siRNA2
LPS group (group C) A549 cell treated by 100 lgÆmL
)1
LPS
HMGN2 knockdown

Western blot
Whole cell lysate was extracted from A549 cells using Pro-
tein Extraction Reagent (Pierce, Rockford, IL, USA).
Alternatively, the cytoplasmic and nuclear extracts were
extracted from A549 cells using NE-PER nuclear and cyto-
plasmic extraction reagents (Pierce) following the manu-
facturer’s instructions. The protein concentration of the
extract was determined using the BCA protein assay
(Pierce). The proteins were separated with SDS ⁄ PAGE and
transferred to poly(vinylidene difluoride) membranes. The
membranes were incubated with HBD-2 (1 : 1000, Santa
Cruz), HMGN2 (1 : 500, Santa Cruz), NF-jB p65
(1 : 1000, Santa Cruz), histone H3 (1 : 1000, Santa Cruz)
or b-actin (1 : 1000, Santa Cruz) antibody followed by
incubation with the horseradish peroxidase conjugated anti-
goat IgG serum (Santa Cruz). Finally the membranes were
developed using ECL Plus reagent (Pierce).
Luciferase assay
Stable A549 cells were placed in 24-well plates at a density
of 2.5 · 10
5
cells per well. After 24 h the cells were
co-transfected with 200 ng reporter plasmid pGL3-3 · jB-
Luc and pRL-TK (Promega) using Lipfectomine 2000 (In-
vitrogen). Then the cells were untreated or treated with
LPS (100 lgÆmL
)1
) for 4 h before the cell lysate was col-
lected by passive lysis for luciferase activity assay. The
experiments were done in triplicate for each group.

(1999) Defensins: key players or bystanders in infection,
injury, and repair in the lung? J Allergy Clin Immunol
104, 1131–1138.
2 Schutte BC & McCray PB Jr (2002) b-defensins in lung
host defense. Annu Rev Physiol 64, 709–748.
3 Oppenheim JJ, Biragyn A, Kwak LW & Yang D (2003)
Roles of antimicrobial peptides such as defensins in
innate and adaptive immunity. Ann Rheum Dis 62,
ii17–ii21.
4 Tsutsumi-Ishii Y & Nagaoka I (2003) Modulation of
human beta defensin-2 transcription in pulmonary
epithelial cells by lipopolysaccharide-stimulated
Table 5. Primers for HMGN2 and HBD-2.
Primers Sequences
HMGN2 sense 5¢-CGATG CCC AAG AGA AAG G-3¢
HMGN2 antisense 5¢-GCAA CTT GGC ATC TCC AGC A-3¢
HBD-2 sense 5¢-CCAGCCATCAGCCATGAGGGT-3¢
HBD-2 antisense 5¢-GGAGCCCTTTCTGAATCCGCA-3¢
HMGN2 mediates expression of b-defensins L X. Deng et al.
2164 FEBS Journal 278 (2011) 2152–2166 ª 2011 The Authors Journal compilation ª 2011 FEBS
mononuclear phagocytes via proinflammatory cytokine
production. J Immunol 170, 4226–4236.
5 Goldman MJ, Anderson GM, Stolzenberg ED, Kari
UP, Zasloff M & Wilson JM (1997) Human b-defensin-
1 is a salt-sensitive antibiotic in lung that is inactivated
in cystic fibrosis. Cell Press 88, 553–560.
6 Liu L, Roberts AA & Ganz T (2003) By IL-1 signaling,
monocyte-derived cells dramatically enhance the epider-
mal antimicrobial response to lipopolysaccharide.
J Immunol 170, 575–580.

15 Bustin M (1999) Regulation of DNA-dependent activi-
ties by the functional motifs of the high-mobility-group
chromosomal proteins. Mol Cell Biol 19, 5237–5246.
16 Cleynen I, Brants JR, Peeters K, Deckers R, Debiec-
Rychter M, Sciot R, Van De Ven WJ & Petit MM
(2007) HMGA2 regulates transcription of the Imp2
gene via an intronic regulatory element in cooperation
with nuclear factor-jB. Mol Cancer Res 5, 363–372.
17 Mantovani F, Covaceuszach S, Rustighi A, Sgarra R,
Heath C, Goodwin GH & Manfioletti G (1998) NF-jB
mediated transcriptional activation is enhanced by the
architectural factor HMGI-C. Nucleic Acids Res
26,
1433–1439.
18 Zaalouk TK, Elliott MB, George JT & McDonald V
(2004) Differential regulation of b-defensin gene expres-
sion during Cryptosporidium parvum infection. Infect
Immun 72, 2772–2779.
19 Ashburner BP, Westerheide SD & Baldwin AS Jr
(2001) The p65 (RelA) subunit of NF-kB interacts with
the histone deacetylase (HDAC) corepressors HDAC1
and HDAC2 to negatively regulate gene expression.
Mol Cell Biol 21, 7065–7077.
20 Chen L, Fischle W, Verdin E & Greene WC (2001)
Duration of nuclear NF-jB action regulated by revers-
ible acetylation. Science 293, 1653–1657.
21 Chen LF, Mu Y & Greene WC (2002) Acetylation of
RelA at discrete sites regulates distinct nuclear functions
of NF-jB. EMBO J 21, 6539–6548.
22 Gao Z, Chiao P, Zhang X, Zhang X, Lazar MA, Seto

Bustin M (2009) The nucleosome binding protein
HMGN3 modulates the transcription profile of pancre-
atic beta cells and affects insulin secretion. Mol Cell
Biol 29, 5264–5276.
29 Korner U, Bustin M, Scheer U & Hock R (2003)
Developmental role of HMGN proteins in Xenopus
laevis. Mech Dev 120, 1177–1192.
30 Rubinstein YR, Furusawa T, Lim JH, Postnikov YV,
West KL, Birger Y, Lee S, Nguyen P, Trepel JB & Bu-
stin M (2005) Chromosomal protein HMGN1 modulates
the expression of N-cadherin. FEBS J 272, 5853–5863.
L X. Deng et al. HMGN2 mediates expression of b-defensins
FEBS Journal 278 (2011) 2152–2166 ª 2011 The Authors Journal compilation ª 2011 FEBS 2165
31 Furusawa T, Lim JH, Catez F, Birger Y, Mackem S &
Bustin M (2006) Down-regulation of nucleosomal bind-
ing protein HMGN1 expression during embryogenesis
modulates Sox9 expression in chondrocytes. Mol Cell
Biol 26, 592–604.
32 Amen M, Espinoza HM, Cox C, Liang X, Wang J,
Link TM, Brennan RG, Martin JF & Amendt BA
(2007) Chromatin-associated HMG-17 is a major regu-
lator of homeodomain transcription factor activity
modulated by Wnt ⁄ b-catenin signaling. Nucleic Acids
Res 36, 462–476.
33 Belova GI, Postnikov YV, Furusawa T, Birger Y &
Bustin M (2008) Chromosomal protein HMGN1
enhances the heat shock-induced remodeling of Hsp70
chromatin. J Biol Chem 283, 8080–8088.
34 Feng Y, Huang N, Wu Q & Wang B (2005) HMGN2:
a novel antimicrobial effector molecule of human

induced NF-jB activation. EMBO J 26, 1150–1162.
42 Schwab M, Reynders V, Loitsch S, Steinhilber D,
Schroder O & Stein J (2008) The dietary histone
deacetylase inhibitor sulforaphane induces human
b-defensin-2 in intestinal epithelial cells. Immunology
125, 241–251.
43 Schauber J, Iffland K, Frisch S, Kudlich T, Schmausser
B, Eck M, Menzel T, Gostner A, Lu
¨
hrs H & Scheppach
W (2004) Histone-deacetylase inhibitors induce the
cathelicidin LL-37 in gastrointestinal cells. Mol Immunol
41, 847–854.
44 Agresti A, Scaffidi P, Riva A, Caiolfa VR & Bianchi
ME (2005) GR and HMGB1 interact only within chro-
matin and influence each other’s residence time. Mol
Cell
18, 109–121.
45 Mohamed OA, Bustin M & Clarke HJ (2001) High-
mobility group proteins 14 and 17 maintain the timing
of early embryonic development in the mouse. Dev Biol
229, 237–249.
46 Livak KJ & Schmittgen TD (2001) Analysis of relative
gene expression data using real-time quantitative PCR
and the 2
)DD
Ct Method. Methods 25, 402–408.
Supporting information
The following supplementary material is available:
Fig. S1. HMGN2 expression is not affected by LPS

copy-edited or typeset. Technical support issues arising
from supporting information (other than missing files)
should be addressed to the authors.
HMGN2 mediates expression of b-defensins L X. Deng et al.
2166 FEBS Journal 278 (2011) 2152–2166 ª 2011 The Authors Journal compilation ª 2011 FEBS