Content uploaded by Takao Hamakubo
Author content
All content in this area was uploaded by Takao Hamakubo on Mar 22, 2016
Content may be subject to copyright.
(343), ra88. [doi: 10.1126/scisignal.2005522]7Science Signaling
Hamakubo (September 16, 2014)
Shuying Jiang, Makoto Naito, Kouhei Tsumoto and Takao
Matsubara, Takashi Minami, Naotaka Yamaguchi, Kenji Inoue,
Kenji Daigo, Makoto Nakakido, Riuko Ohashi, Rie Fukuda, Koichi
histone-mediated endothelial cell cytotoxicity in sepsis
Protective effect of the long pentraxin PTX3 against
This information is current as of September 16, 2014.
The following resources related to this article are available online at http://stke.sciencemag.org.
Article Tools
http://stke.sciencemag.org/content/7/343/ra88
article tools:
Visit the online version of this article to access the personalization and
Materials
Supplemental http://stke.sciencemag.org/content/suppl/2014/09/12/7.343.ra88.DC1.html
"Supplementary Materials"
Related Content
http://stke.sciencemag.org/content/sigtrans/4/199/ra75.full.html
http://stke.sciencemag.org/content/sigtrans/6/258/eg2.full.html
's sites:ScienceThe editors suggest related resources on
References http://stke.sciencemag.org/content/7/343/ra88#BIBL
This article cites 49 articles, 13 of which you can access for free at:
Glossary http://stke.sciencemag.org/cgi/glossarylookup
Look up definitions for abbreviations and terms found in this article:
Permissions http://www.sciencemag.org/about/permissions.dtl
Obtain information about reproducing this article:
reserved.
DC 20005. Copyright 2014 by the American Association for the Advancement of Science; all rights
American Association for the Advancement of Science, 1200 New York Avenue, NW, Washington,
(ISSN 1937-9145) is published weekly, except the last December, by theScience Signaling
on September 16, 2014http://stke.sciencemag.org/Downloaded from on September 16, 2014http://stke.sciencemag.org/Downloaded from
SEPSIS
Protective effect of the long pentraxin PTX3
against histone-mediated endothelial cell
cytotoxicity in sepsis
Kenji Daigo,
1
* Makoto Nakakido,
2†
Riuko Ohashi,
3,4
Rie Fukuda,
1
Koichi Matsubara,
3
Takashi Minami,
5
Naotaka Yamaguchi,
6
Kenji Inoue,
7
Shuying Jiang,
3,8,9
Makoto Naito,
3
Kouhei Tsumoto,
2
Takao Hamakubo
1‡
Pentraxin 3 (PTX3), a member of the long pentraxin subfamily within the family of pentraxins, is a soluble
pattern recognition molecule that functions in the innate immune system. Innate immunity affords the
infected host protection against sepsis, a potentially life-threatening inflammatory response to infection.
Extracellular histones are considered to be the main cause of septic death because of their cytotoxic effect
on endothelial cells, which makes them a potential therapeutic target. We found that PTX3 interacted with
histones to form coaggregates, which depended on polyvalent interactions and disorder in the secondary
structure of PTX3. PTX3 exerted a protective effect, both in vitro and in vivo, against histone-mediated
cytotoxicity toward endothelial cells. Additionally, the intraperitoneal administration of PTX3 reduced
mortality in mouse models of sepsis. The amino-terminal domain of PTX3, which was required for coag-
gregation with histones, was sufficient to protect against cytotoxicity. Our results suggest that the host-
protective effects of PTX3 in sepsis are a result of its coaggregation with histones rather than its ability
to mediate pattern recognition. This long pentraxin–specific effect provides a potential basis for the
treatment of sepsis directed at protecting cells from the toxic effects of extracellular histones.
INTRODUCTION
Pentraxin 3 (PTX3) is one of a number of soluble pattern recognition mol-
ecules (PRMs), which are key molecules in the innate immune system,
and it belongs to the long pentraxin subfamily within the family of pen-
traxins (1,2). PTX3 plays several roles in the first-line activity of the host-
protective response, such as the recognition of specific pathogens and the
subsequent opsonization that occurs, as well as regulation of complement
and inflammation (3–5). PTX3 also plays a role in angiogenesis and fe-
male fertility (1,6,7). This multifunctionality of PTX3 is due to its ability
to bind to many different ligands (5,8). PTX3 has two domains: a unique
N-terminal domain and a conserved C-terminal pentraxin domain, and
most of its ligands bind to PTX3 in a domain-specific manner (2,9,10).
PTX3 forms a multimeric structure through intermolecular disulfide
bonds, and it has an N-glycosylation site (5,8).
The innate immune system plays crucial roles in the pathophysiology
of sepsis, which is a major cause of death in developed countries (11). At
the onset of sepsis, infectious events initiate the innate immune response
through PRMs by sensing pathogen-associated molecular patterns
(PAMPs) (12,13). Subsequently, certain innate immune cells, such as
macrophages, secrete proinflammatory mediators in response to PRMs
(13,14). An excessive immune reaction and exposure to proinflammatory
cytokines result in a systemic inflammatory reaction (13). In response to
the innate immune reaction, neutrophils are attracted to, and activated at,
the site of infection or damaged tissue. An overaccumulation of neutro-
phils leads to the multiorgan failure that is the direct cause of septic death,
by exerting an excessive inflammatory response and concomitant cytotox-
icity (14). In sepsis, the circulating concentration of PTX3 increases sub-
stantially (15–17). In addition to PAMPs, host-derived molecules, called
damage-associated molecular patterns (DAMPs; also known as alarmins),
also stimulate the PRM-mediated inflammatory and innate immune re-
sponses (18,19). In the case of sterile inflammation, such as that induced
by trauma, burn, ischemia, or hemorrhage, which exhibits almost the same
pathophysiology as that of sepsis, DAMPs are the major activators of
PRMs (12). DAMPs participate not only in sterile inflammation but also
in sepsis pathophysiology (12,13,20). Thus, DAMPs are regarded as an
additional target for the treatment of sepsis.
Extracellular histones are DAMPs (21,22), and they are a major cause
of septic death (23). They are present in the plasma of septic patients, and
they exert cytotoxic activity toward endothelial cells. Mice injected with
histones display damage to the endothelium and ultimately develop intra-
alveolar hemorrhage. PTX3 binds to certain histones (24,25), and we pre-
viously observed that extracellular histones are bound to PTX3 in plasma
from septic patients (26). Although it is clear that the recognition of certain
PAMPs and the activation of the subsequent innate immune response are
some of the host-protective activities of PTX3 during sepsis, from the ev-
idence discussed earlier, we hypothesized the possibility of an additional
role for PTX3 in sepsis in the form of host protection against extracellular
histones. Here, we investigated the detailed mechanism by which PTX3
interacted with histones and how it exerted its cytoprotective effect
against histone damage. These results suggest that PTX3 plays a major
1
Department of Quantitative Biology and Medicine, Research Center for Ad-
vanced Science and Technology, The University of Tokyo, Tokyo 153-8904, Ja-
pan.
2
Laboratory of Medical Proteomics, Institute of Medical Science, The
University of Tokyo, Tokyo 108-8639, Japan.
3
Division of Cellular and Molecular
Pathology, Department of Cellular Function, Niigata University Graduate School
of Medical and Dental Sciences, Niigata 951-8510, Japan.
4
Department of Pa-
thology, Niigata University Medical and Dental Hospital, Niigata 951-8520, Ja-
pan.
5
Laboratory for Vascular Biology, Research Center for Advanced Science
and Technology, The University of Tokyo, Tokyo 153-8904, Japan.
6
Department of
Emergency and Critical Care Medicine, Juntendo University Nerima Hospit al, To-
kyo 177-8521, Japan.
7
Department of Cardiology, Juntendo University Nerima
Hospital, Tokyo 177-8521, Japan.
8
Niigata College of Medical Technology, Niigata
950-2076, Japan.
9
Perseus Proteomics Inc., Tokyo 153-0041, Japan.
*Present address: Humanitas Clinical and Research Center, Rozzano, Milan
20089, Italy.
†Present address: Department of Medicine, University of Chicago, Chicago,
IL 60637, USA.
‡Corresponding author. E-mail: hamakubo@qbm.rcast.u-tokyo.ac.jp
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 16 September 2014 Vol 7 Issue 343 ra88 1
on September 16, 2014http://stke.sciencemag.org/Downloaded from
host-protective role in the response to sepsis, and thus, it has potential value
for sepsis treatment.
RESULTS
Interaction pattern between histone and PTX3
The binding between immobilized histone H1 and PTX3 (24), as well as
inhibition of the binding of immobilized PTX3 to fibroblast growth factor
2(FGF2)byhistones(25) were previously reported; however, a compre-
hensive analysis of the ability of PTX3 to interact with each histone has
not been reported. Thus, we sought to characterize the interactions be-
tween linker and core histones and PTX3. First, we used an enzyme-linked
immunosorbent assay (ELISA) to assess the interactions between immobil-
ized histones and PTX3. We observed increased signals when full-length
PTX3 was incubated with histones H1, H3, and H4, whereas we observed
relatively decreased signals in the cases of histones H2A and H2B (Fig. 1A).
All signals were increased in the presence of calcium ions (Fig. 1A). We
further checked the interaction preference of the PTX3 domains with his-
tones in experiments with PTX3 domain fragment proteins. The extent of
interaction of the N-terminal domain of PTX3 with all the histones was
greater than that of the C-terminal domain of PTX3 (Fig. 1B and fig.
S1A). The oligomerization-defective PTX3 N-terminal domain, in which
all of the cysteine residues were changed to serine residues, abolishes the
ability of PTX3 to bind to ligands (26,27). Compared to the signals gen-
erated by the interaction of wild-type PTX3 with all of the histones, the
signals generated by the oligomerization-defective PTX3 N-terminal do-
main mutant were decreased (fig. S1B). These results indicate that all of
the histones directly bound to PTX3, preferably through its N-terminal
oligomerization domain.
We next measured apparent interaction affinities by surface plasmon
resonance (SPR). We found that PTX3 had a higher affinity for histones
H1, H3, and H4 than it had for histones H2A and H2B (Fig. 1C and f ig.
S2). There was also an increase in its affinity in the presence of calcium
ions (fig. S2). Both full-length PTX3 and the PTX3 N-terminal domain
showed high affinity for histones (fig. S2, A and C). In comparison, the
oligomerization-deficient PTX3 N-terminal domain showed lower affinity
for histones (fig. S2, B and C). Note that the affinity of the PTX3 N-terminal
domain for all histones was similar to that of the full-length protein (fig.
S2C). We then investigated the region(s) in the histones that were respon-
sible for the interaction with PTX3 in experiments with synthesized pep-
tides (20 residues in length) that covered all of the regions of histones H3
and H4 (fig. S3A and table S1). We focused on histones H3 and H4 be-
cause these histones are more cytotoxic toward endothelial cells than are
the other histones (23). We analyzed the interactions between the immo-
bilized histone fragment peptides and PTX3 by ELISA. Several peptides
exhibited binding to full-length PTX3, and most of these preferentially
bound to the N-terminal domain of PTX3 (fig. S3B). These results suggest
that there are multiple PTX3-interacting regions in histones H3 and H4.
Coaggregation of PTX3 with histones
To analyze the histone-PTX3 complex from a physicochemical view, we
undertook the preparation of an in-solution complex by the addition of
PTX3 to a solution of histone H4. By measuring the UV (ultraviolet)–visible
spectrum of the histone H4 and PTX3 mix-
ture, we observed a dose-dependent increase
in the absorbance spectrum (fig. S4). This
result suggested that the spectrum pattern
originated from the light scattering of micro-
particles. In addition to the electron micro-
scopic observation of amorphous aggregates
in the histone H4–PTX3 mixture (Fig. 2A),
it appeared that the in-solution histone-PTX3
interaction induced coaggregates. Note that
the addition of histones to serum or plasma
causes a precipitate to form (28), and fibri-
nogen forms an aggregate with histones (29).
Thus, we analyzed the formation of the
histone-PTX3 coaggregate. Using the ab-
sorbance at 310 nm as a measurement of
the turbidity of solutions, we analyzed the
extent of coaggregation among histones and
PTX3 domains. All of the histones formed
coaggregates with PTX3, and higher turbid-
ity was observed in coaggregates contain-
ing the PTX3 N-terminal domain than in
aggregates containing the PTX3 C-terminal
domain (Fig. 2, B and C).
Induction of histone-PTX3
aggregate formation by
polyvalent interactions
and unfolding
To investigate the mechanisms by which
histone-PTX3 aggregates formed, we set
out to determine the stoichiometry of this
aggregate. Various molar ratio mixtures
H1
A
100
10
1
0
A450
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Ca
2+
EDTA/EGTA
H2A
100
10
1
0
H2B
rhPTX3 (ng/ml)
100
10
1
0
H3
100
10
1
0
H4
100
10
1
0
B
A450
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
ng/ml
500
50
5
0.5
0.05
0
H4 Ca
2+
ng/ml
500
50
5
0.5
0.05
0
H4 EDTA/EGTA
PTX3_Full
PTX3_N
PTX3_C
C
0
100
200
300
400
500
Response [RU]
H4 Ca
2+
0200 600 1000
Time [s]
0 200 600 1000
Time [s]
H4 EDTA/EGTA
PTX3_Full
PTX3_N
PTX3_C
KD (M)
H4 Ca
2+
H4 EDTA/EGTA
PTX3_Full 1.07 ± 1.28 x
10
–11
3.44 ± 0.29 x
10
–11
PTX3_N 3.35 ± 0.53 x
10
–10
3.46 ± 0.14 x
10
–10
PTX3_C 1.04 ± 1.69 x
10
–9
3.22 ± 5.45 x
10
–7
Fig. 1. Binding pattern and affinity of PTX3 and histones. (A) The extent of binding of recombinant, non-
tagged human PTX3 (rhPTX3) to the indicated histone proteins in the presence (Ca
2+
) and absence
(EDTA/EGTA) of calcium was measured by ELISA as described in Materials and Methods. The horse-
radish peroxidase (HRP)–conjugated anti-PTX3 monoclonal antibody PPZ-1228 was used for detection.
Data are from three independent experiments performed in duplicate. (B) The extent of binding of the
indicated recombinant tagged human PTX3 fragments to histone H4 in the presence (left) and absence
(right) of calcium was measured by ELISA. An HRP-conjugated anti-myc antibody was used for detection.
Data are from three independent experiments performed in duplicate. (C) Binding sensorgrams and
calculated affinities of the interaction between histone H4 and the indicated recombinant tagged human
PTX3 fragments in the presence (Ca
2+
) and absence (EDTA/EGTA) of calcium as determined by SPR
measurements. SPR sensorgrams are representative of two or three independent experiments. The
calculated affinities shown below the graphs are from two or three independent experiments. The extents
of binding and affinities of the PTX3 fragments and other histones can be found in figs. S1 and S2.
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 16 September 2014 Vol 7 Issue 343 ra88 2
on September 16, 2014http://stke.sciencemag.org/Downloaded from
of histone H4 and PTX3 were prepared, and, after removal of any aggregates
by centrifugation, the concentrations of both proteins in the supernatant were
measured (Fig. 3A). The concentration of PTX3 in the supernatant was
decreased in the presence of increasing concentrations of histone H4,
and was undetectable at a PTX3 monomer/histone H4 molar ratio of be-
tween 1:1 and 1:2 (Fig. 3B). These data suggest that full-length PTX3
formed aggregates with histone H4 at ratios between 1:1 and 1:2, whereas
the ratios were 1:2 for the N-terminal domain and 1:1 for the C-terminal
domain of PTX3. Furthermore, PTX3 was recovered from the supernatant
in the case where there was an excess molar ratio of histone H4 (Fig. 3B).
This suggests that, similar to the case of polyclonal antibody–antigen com-
plex formation, histone-PTX3 aggregation is the result of a polyvalent in-
teraction. We further investigated aggregate formation in experiments with
an oligomerization-defective PTX3 N-terminal domain. Although it was
less potent than wild-type PTX3, the oligomerization-defective PTX3
N-terminal domain formed aggregates with histones (fig. S5A), and the
stoichiometry of its aggregation with histone H4 was similar to that of
wild-typePTX3(fig.S5B).
To investigate other mechanisms of aggregate formation, we examined
whether histone-PTX3 aggregates could be assessed by staining with thio-
flavin T (ThT) and 8-anilino-1-naphthalenesulfonic acid ammonium salt
(ANS), which are fluorescent dyes commonly used for the detection of
aggregate formation by bstrand stacking or exposure of hydrophobic
amino acid residues (30–32), but we did not observe any signals from
histone-PTX3 aggregates (fig. S5, C and D). Thus, we next analyzed the
mixture of the PTX3 N-terminal domain and histone H4 by circular di-
chroic (CD) spectroscopy to detect any change in the secondary structure
in the histone-PTX3 complex. From the spectrum pattern of the original
state of the PTX3 N-terminal domain, the presence of an ahelix was sug-
gested (Fig. 3C, upper panel), as is predicted from the amino acid sequence
(6). In the histone-PTX3 mixture, the spectra were changed from the orig-
inal state (Fig. 3C, lower panel). Because these spectra did not exhibit any
typical secondary structures, the result suggests that PTX3 lost its sec-
ondary structure when complexes were formed. Together, these data suggest
PTX3 (
µ
g/ml)
010 30 50
Turbidity (310 nm)
0.00
0.01
0.02
0.03
0.04
0.06
0.08
0.05
0.07
0.00
0.01
0.02
0.03
0.04
0.06
0.08
0.05
0.07
0.00
0.01
0.02
0.03
0.04
0.06
0.08
0.05
0.07
Histones
Control
010 30 50 0103050
PTX3_Full
PTX3_N PTX3_C
0.00
0.01
0.02
0.03
0.04
0.05
– H1 H2A H2B H3 H4
Turbidity (310 nm)
–
PTX3_C
PTX3_N
PTX3_Full
AB
C
Fig. 2 . Histone-PTX3 binding causes aggre gate formation. (A) Electronmicro-
scopic image of the aggregates formed in a mixture of histone H4 (0.5 mg/ml)
and the N-terminal domain of wild-type PTX3 (0.17 mg/ml). Scale bar, 0.2 mm.
Image is representative of two independent experiments. (B) The indicated
recombinant histone proteins (50 mg/ml) were mixed with each of the indi-
cated PTX3 fragments (30 mg/ml), and the turbidity of the resulting mixtures
was measured as described in Materials and Methods. Data are means ± SD
from two independent experiments. (C) Calf thymus histones (Histones,
50 mg/ml) were mixed with a range of concentrations of the indicated PTX3
fragments, and the turbidity of the resulting mixtures was measured as de-
scribed earlier. Data are means ± SD from two independent experiments.
PTX3-histone
mixture
Centrifuge
Supernatant
SYPRO Ruby
Precipitant
nm
200 210 220 230 240 250 260
CD [m deg]
–6
–4
–2
0
2
4
6N-terminal wild type+
Histone H4
N-terminal mutant+
Histone H4
CD [m deg]
–6
–4
–2
0
2
4
6N-terminal wild type
nm
200 210 220 230 240 250 260
N-terminal mutant
Histone H4
A
PTX3_Full/histone H4 molar ratio
PTX3_N/histone H4 molar ratio
PTX3_N
Histone H4
1:0
1:0.25 1:0.5
1:1 1:2 1:4 1:8
PTX3_C/histone H4 molar ratio
PTX3_C
Histone H4
1:0
1:0.25 1:0.5
1:1 1:2 1:4 1:8
PTX3_Full
Histone H4
1:0
1:0.25 1:0.5
1:1 1:2 1:4 1:8
B
C
Fig. 3 . The molecular mechanisms of histone-
PTX3 aggregate formation. (A) Method
used to determine the stoichiometry of ag-
gregate formation. (B) Determination of the
stoichiometry of histone-PTX3 aggregates.
The indicated PTX3 fragments (50 mg/ml)
were mixed with the indicated molar ratios
of histone H4 protein, aggregates were re-
moved by centrifugation, and the residual
PTX3 and histone H4 proteins in the super-
natant were resolved by SDS–polyacrylamide
gel electrophoresis. The gels were stained
with SYPRO Ruby. Images are representative
of two independent experiments. (C)Top:CD
spectra of PTX3 N-terminal wild type (25 mg/ml),
PTX3 N-terminal mutant (25 mg/ml), and histone
H4 (5 mg/ml). Bottom: CD spectra of mixtures
of the indicated PTX3 proteins and histone
H4. To ensure adequate measurements,
concentrations of PTX3 and histone H4 that
resulted in no aggregation were used (fig.
S5E). Before all experiments, samples were
dialyzed with tris-buffered saline (TBS)
containing 4 mM CaCl
2
. Spectra are repre-
sentative of two independent experiments.
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 16 September 2014 Vol 7 Issue 343 ra88 3
on September 16, 2014http://stke.sciencemag.org/Downloaded from
that histone-PTX3 aggregate formation results from both the formation of a
large molecular mass complex by a polyvalent interaction and a disordered
aggregation caused by unfolding of the ahelix of PTX3.
Protective effect of PTX3 against histone-mediated
cytotoxicity toward endothelial cells
The specific histone-PTX3 interaction we observed prompted us to further
examine its role in modulating histone-mediated cytotoxicity toward endo-
thelial cells (23). We found that histone-dependent cytotoxicity toward en-
dothelial cells was attenuated by PTX3 in a dose-dependent manner
(Fig. 4A). The PTX3-mediated cytoprotective effect was exhibited in
the case of all of the histones (fig. S6A). PTX3 also blocked histone-
dependent calcium fluxes in endothelial cells (Fig. 4B and fig. S6B).
Furthermore, PTX3 substantially suppressed further calcium fluxes
when it was added to culture medium after the histones had been added
(Fig. 4B, right panel).
The association of histones with plasma
membranes is crucial for their toxicity to-
ward endothelial cells (33). Thus, we next
checked whether PTX3 blocked the associ-
ation of fluorescently labeled histones with
endothelial cells. From confocal microscop-
ic imaging, we found that, in addition to
associating with the plasma membrane,
histones were also internalized by the cells
(fig. S6, C and D). We observed that his-
tone aggregates formed in response to the
addition of PTX3, and these precipitates
increased in number in a time-dependent
manner (Fig. 4C). We used flow cytometric
analysis to evaluate the effect of PTX3 in
blocking histone accumulation in endothe-
lial cells. We found that histone accumulation
was decreased by PTX3 in a dose-dependent
manner (Fig. 4D). These results suggest
that the histone-PTX3 aggregates suppress
the cytotoxic effects of histones by prevent-
ing the association of histones with endo-
thelial cells. Finally, we observed that the
N-terminal domain of PTX3 exhibited a
cytoprotective effect that was similar to that
of full-length PTX3 (Fig. 4E).
In vivo cytoprotective effects
of PTX3 against
extracellular histones
We further examined the protective effects
of PTX3 against histone-mediated cytotox-
icity in a mouse model of histone infusion.
The N-terminal domain of PTX3 was ad-
ministered to the mice because it was suf-
ficient for aggregate formation (fig. S5A)
and it had a cytoprotective effect against his-
tones (fig. S7A). Administration of PTX3
substantially reduced mortality in mice in-
fusedwithahighdoseofhistone(60mg/kg)
(Fig. 5A), which typically induces mouse
lethality within 1 hour (23,33,34).
To investigate the host-protective ef-
fects elicited by PTX3, we performed his-
tological examinations of mice infused with
a sublethal dose of histones (50 mg/kg).
Histone-infused mice exhibited hemorrhage
in the connective tissues of the bronchiovas-
cular bundle in the lung (Fig. 5B) and in the
hilum of the lung (Fig. 5C). The extent of
these hemorrhages was reduced by PTX3
(Fig. 5, B and C). Histone infusion inflicted
PTX3_Full (µg/ml)
0 1020304050
MFI
0
50
100
150
200
250
300
350
Histones
0 µg/ml
50 µg/ml
100 µg/ml
90 min 20 min 10 min
Histones
Histones+
PTX3_Full
MFI
PTX3_Full (
µ
g/ml)
0 1020304050
0
5
10
15
200
400
600
800
1000
100
µ
g/ml
50
µ
g/ml
100
µ
g/ml
50
µ
g/ml
–
Histones
BSA
Histones + PTX3_N
101
100102
PI
103104
Histones +PTX3_C
Histones
0
25
50
Count
75
Control
Histones + PTX3_Full
s
0
60
120
180
240
300
360
420
480
540
600
Number of cells with
increasing fluorescence
0
2
4
6
8
10
12
14
16
Control
PTX3_Full
*
**
***
***
***
***
***
***
***
***
***
Histones + PTX3
s
0
60
120
180
240
300
360
420
480
540
600
Control
PTX3_Full
Histones
PTX3
*
**
*
*
*
*
*
*
*
*
AB
CDE
Fig. 4. PTX3-mediated protection of endothelial cells from cytotoxicity caused by extracellular histones. (A)
The ability of PTX3 to protect human umbilical vein endothelial cells (HUVECs) from the cytotoxic effects of
the indicated concentrations of calf thymus histones (Histones) was assessed by staining the cells with
propidium iodide (PI) and determining the mean fluorescence intensity (MFI) of PI staining. Data are means ±
SD from three independent experiments. (B) The ability of PTX3 to suppress histone-mediated calcium
flux in EA.hy926 cells was determined by measuring a calcium indicator dye in loaded cells. The fluores-
cence images were taken at 5-s intervals with a microscope, and the numbers of cells that exhibited
increased fluorescence intensity within groups of 25 cells in each set of six frames were counted. Left:
Histone (Control) or a histone-PTX3 mixture (PTX3_Full) was added. Right: Histone was added at 0 s,
and then either medium (Control) or PTX3 (PTX3_Full) was subsequently added at 180 s. Data are means ±
SD from three independent experiments. *P<0.05,**P< 0.01, and ***P< 0.001 compared to control. (C)
EA.hy926 endothelial cells stained with CellTracker Orange were incubated for the indicated times with
Alexa Fluor 488–labeled histones (40 mg/ml) alone or in the presence of full-length PTX3 protein (20 mg/ml).
The localization of the histones in relation to the endothelial cells was observed by confocal microscopy.
Scale bar, 100 mm. Images are representative of three independent experiments. (D) EA.hy926 cells were
incubated with Alexa Fluor 488–conjugated histones or bovine serum albumin (BSA) in the presence of the
indicated concentrations of full-length PTX3 protein. Cells were then analyzed by flow cytometry to mea-
sure the MFI of bound fluorescently labeled protein. Data are means ± SD from three independent
experiments. (E) HUVECs were treated with calf thymus histones (Histones, 100 mg/ml) in the presence
or absence of the indicated human PTX3 fragments (40 mg/ml). Cells were then incubated with PI, and the
extent of PI staining was measured by flow cytometric analysis as described earlier. Data are representa-
tive of three independent experiments.
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 16 September 2014 Vol 7 Issue 343 ra88 4
on September 16, 2014http://stke.sciencemag.org/Downloaded from
damage on the smooth muscle fibers in the pulmonary veins (Fig. 5D),
which suggested the source of the massive bleeding. Such damage was
not observed in the pulmonary artery (fig. S7B). Infused histones also in-
duced breakage of the argyrophilic fibers in the alveoli and alveolar hem-
orrhage (Fig. 5E). This damage was also suppressed by PTX3 (Fig. 5, D
and E). Electron microscopic analysis revealed endothelial cell damage,
such as cytoplasmic vacuolation, and the detachment of endothelial cells from
the basal laminae in the peripheral pulmonary artery, effects that were sup-
pressed by PTX3 (Fig. 5F). Histone infusion not only causes cytotoxicity-
induced hemorrhage in vivo but also induces thrombocytopenia (34,35);
however, we found that PTX3 did not ameliorate histone-mediated throm-
bocytopenia (fig. S7C). Together, the results from our in vivo experiments
suggest that the reduced mortality was mainly a result of the protective
effect of PTX3 against histone-mediated cytotoxicity toward endothelial
cells and that this suppressed hemorrhaging.
Host-protective effects of PTX3 in mouse
models of sepsis
The protective effect of the N-terminal domain of PTX3 against histone-
mediated cytotoxicity in vivo implied that it had therapeutic potential in sep-
sis. Thus, we investigated its effects further in mouse models of sepsis. We
found that PTX3 substantially reduced mortality in mice injected with lipo-
polysaccharide (LPS) (Fig. 6A). Histone H3 was detectable in the circulation
24 hours after the mice were injected with LPS, as determined by Western
blotting analysis, and this increase in histone H3 concentration was suppressed
by PTX3 (Fig. 6B and fig. S8). The plasma concentrations of interleukin-6
(IL-6) and vascular endothelial growth factor (VEGF) were increased in
mice injected with LPS, an effect that was suppressed by PTX3 (Fig. 6C).
Histological examination of the LPS-injected mice showed the recruitment
of macrophages to the lung and the liver (Fig. 6D), as well as neutrophil in-
filtration into microvessels at alveolar walls (Fig. 6E). These effects were all
substantially suppressed by PTX3 (Fig. 6F). We also observed that PTX3
resulted in a substantial reduction in mortality in mice subjected to cecal li-
gation and puncture (CLP), which is a widely used murine sepsis model (36),
even when it was administered after the CLP operation (Fig. 6G).
DISCUSSION
Here, we report that PTX3 protects endothelial cells from extracellular
histone–mediated cytotoxicity in vitro and in vivo. The administration of
the N-terminal domain of PTX3 reduced mortality in mouse models of
histone infusion and sepsis, which suggests that this domain is suff icient
for the host-protective effect of PTX3 against septic lethality. PTX3 caused
coaggregation with histones (Fig. 2). The molecular mechanisms under-
lying the coaggregation involved the formation of a supermolecule by
polyvalent interactions, as well as the unfolding of the ahelix of PTX3
(Fig. 3) (37). That the N-terminal domain of PTX3, which is independent
of the C-terminal pentraxin domain, participated in coaggregation with
histones (Fig. 2, B and C) implies a correlation between the extent of
coaggregation and the effect on histone-mediated cytotoxicity (Fig. 4).
A previous study of PTX3-transgenic mice demonstrated that an artificial-
ly increased amount of PTX3 afforded protection against septic lethality
020 40 60 80 100
0
20
40
60
80
100 N-terminal
wild type
Control
P = 0.0018
Time after challenge (min)
Percent survival
N-terminal wild typeControl
H&E staining
Elastica van
Gieson staining
N-terminal wild typeControl
Reticulin staining
N-terminal wild typeControl N-terminal wild typeControl
N-terminal wild type
Control
AB C
DE F
Fig. 5. Analysis of the protective effects of the N-terminal domain of PTX3
against histone infusion in mice. Histone-infused mice were treated with the
PTX3 N-terminal domain (N-terminal wild type) or control buffer (Control). (A)
Survival rates of mice intravenously injected with histones (60 mg/kg) over
time. Data are from five mice for each treatment. (Bto F) PTX3-mediated
suppression of hemorrhage and other damage in the lungs. Lungs removed
from mice 120 min after they were infused with sublethal doses of histones
(50 mg/kg) were stained with (B and C) hematoxylin and eosin (H&E), (D)
(top) H&E, (bottom) Elastica van Gieson, and (E) reticulin, or (F) were ana-
lyzed by electron microscopy. Histone-induced hemorrhages were ob-
served in (B) the bronchovascular bundle (arrows) and (C) hilum of the
lung (arrows). (D) Histone-induced disruption of smooth muscle fibers
and desquamation of endothelium and elastic fiber were observed in pul-
monary veins (arrows). (E) Histone-induced rupture of the alveolar walls was
observed (arrows). (F) Histone-induced cytotoxicity of endothelial cells was
observed by electron microscopy. The areas indicated by boxes are en-
larged and are shown below. Hemorrhaging and other damage in the mice
were suppressed by the administration of the N-terminal domain of PTX3.
Scale bars, 500 mm(BandC),50mm (D and E), and 25 mm (F). Data in (B)
to (F) are representative of three mice for each treatment.
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 16 September 2014 Vol 7 Issue 343 ra88 5
on September 16, 2014http://stke.sciencemag.org/Downloaded from
(38). Our results suggest the possibility that, in addition to pathogen rec-
ognition in the innate immune system (3–5), PTX3 exerts a host-protective
effect in sepsis by forming a tight interaction with histones such that the
histone-PTX3 complexes become disordered aggregates that are prevented
from interacting with the endothelial cell surface.
Compared to C-reactive protein, PTX3 is a local PRM, and its concen-
tration in the circulation is less (39). Thus, we believe that the main physiolog-
ical role of PTX3 is not protection against histone-dependent cytotoxicity,
but rather the elimination of invading microbes through opsonization, reg-
ulation of the complement cascade, and neutralization by neutrophil extra-
cellular traps (2,40). In turn, in the case of severe sepsis, the abundance of
PTX3 is increased, and it might exert an effect as a systemic PRM. Under
these conditions, PTX3 gains an additional function, namely, affording
protection against histone-dependent cytotoxicity. When the amount of ex-
tracellular histone protein is greater than the amount of physiological
PTX3, multiorgan failure occurs (20); thus, administration of exogenous
PTX3 might be a feasible treatment approach under such conditions.
The high-affinity interaction between PTX3 and various histone proteins,
which was determined by ELISA and SPR analysis, appears to be a result of
both multivalent interactions (figs. S2 and S3) and aggregation (Fig. 2). The
unexpected observation of histone-PTX3 aggregation was mainly observed
with the PTX3 N-terminal domain; however, although it had weak interac-
tions, the C-terminal domain did form aggregates. This result raises the pos-
sibility that the short pentraxins, C-reactive protein and serum amyloid P
components, which form oligomeric structures (1) and bind to histones (41,42),
may also form aggregates with histones. Because we could not detect either b
strand stacking or a pattern of exposure of hydrophobic residues (fig. S5, C and
D), this suggests that histone-PTX3 aggregate formation is a result of poly-
valent interactions and the unfolding of the ahelix of PTX3 (Fig. 3). Although
the coaggregation of unstable or disordered proteins has been reported previ-
ously (43,44), to our knowledge, this is the f irst report of aggregate formation
induced by unfolding as a result of interaction with a heterogeneous protein.
Extracellular histones are the major mediators of death in sepsis (23).
Although the detailed mechanisms of cytotoxicity are not fully known, the
0 2 4 6 8 10
0
20
40
60
80
100
N-terminal
wild type (n = 16)
Control (n = 16)
P = 0.0028
Time after challenge (days)
Percent survival
LPS
IL-6 (ng/ml)
0
10
20
30
40
50
60
70
*
*
Time after challenge
(hours)
Time after challenge
(hours)
0624
Control
N-terminal
wild type
0624
VEGF (ng/ml)
0.0
0.5
1.0
1.5
2.0
2.5
*
Histone H3
Control
N-terminal
wild type
IgG L
chain
H3
Lung neutrophil
Control N-terminal wild type
0123456
0
20
40
60
80
100
N-terminal
wild type (n = 11)
Control (n = 10)
P = 0.0099
Time after challenge (days)
CLP
Percent survival
Control N-terminal wild type
Lung
macrophage
Liver
macrophage
AB
FG
DE
C
**
0
2
4
6
8
Control
N-terminal
wild type
Count
/6.25 mm
2
Control
N-terminal
wild type
0
20
40
60
80
100
**
Count
/6.25 mm
2
Control
N-terminal
wild type
0
20
40
60
80
120
100
*
Count
/0.05
mm
2
Lung
macrophage
Liver
macrophage
Lung
neutrophil
Fig. 6. Analysis of the host-
protective effects of the N-
terminal domain of PTX3
against sepsis in mouse
models. (Ato F)Micewere
treated with LPS and then
were treated with either the
N-terminal domain of PTX3
(N-terminal wild type) or
control buffer (Control). (A)
Survival rates of mice intra-
venously injected with LPS
over time. Data are from
16 mice for each treatment. (B) The pres-
ence of plasma histone H3 protein was
detected 24 hours after injection with
LPS. Data are from five mice for each
treatment. Analysis of the time course
is shown in fig. S8. (C) The plasma con-
centrations of IL-6 and VEGF in the LPS-
injected mice at the indicated times
were measured by ELISA. Data are
means ± SEM from five mice for each
treatment. *P< 0.05. (D) Macrophage
recruitment into the lung (top) and liver
(bottom) 24 hours after LPS injection was determined by immunostaining with an anti-F4/80 antibody. Scale bar, 50 mm. Images are representative of
four mice for each treatment. (E) Neutrophil recruitment into the lung 24 hours after LPS injection was determined by esterase staining. Scale bar, 50 mm.
Images are representative of five mice for each treatment. (F) Numbers of macrophages in the lung and the liver, and of neutrophils in the lung. *P< 0.05,
**P< 0.005. Data are means ± SD from four or five mice for each treatment. (G) Survival rates of mice subjected to CLP. Mice were treated with gentamicin
alone (Control) or in the presence of the N-terminal domain of PTX3 (N-terminal wild type). Data are from the indicated numbers of mice for each treatment.
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 16 September 2014 Vol 7 Issue 343 ra88 6
on September 16, 2014http://stke.sciencemag.org/Downloaded from
endosome-associated Toll-like receptor 9 (TLR9) (21) and the plasma
membrane–associated TLR2 and TLR4 (22) are implicated in histone cy-
totoxicity in a mouse model of fatal liver injury. Although previous studies
indicated that extracellular histones localized only at the plasma mem-
brane (33,34), we observed their partial internalization (fig. S6C), which
suggests the possible participation of endosome-associated TLRs. We hy-
pothesize that aggregation induced by PTX3 binding and disruption of the
secondary structure of histones could prevent the internalization of extra-
cellular histones. That PTX3 had protective effects against extracellular
histone–dependent cytotoxicity was further supported by the observation
that histone-dependent calcium flux in endothelial cells was suppressed by
PTX3 even when it was added several minutes after treatment with his-
tones (Fig. 4B). We speculate that histone-mediated cytotoxicity is a rela-
tively slow process and that it can be interrupted by the potent interaction
between PTX3 and histones.
In addition to our cell-based analysis, we demonstrated that PTX3 had
a protective effect against extracellular histones in vivo. In a mouse model
of histone challenge, administration of PTX3 blocked massive lung hem-
orrhage and vascular pathological lesion without affecting platelet function
(Fig. 5). This result highlights a distinct feature of PTX3 compared with
other reported protective molecules, such as anti–histone H4 antibody
(23), heparin (34), C-reactive protein (33), and recombinant thrombomo-
dulin (35). Although these molecules also bind to histones, the distinctive
feature of PTX3 is its capacity to cause coaggregation, which enables
PTX3 to bind rapidly and irreversibly to histones.
In our efforts to confirm the effects of administration of the N-terminal
domain of PTX3 in mouse models of sepsis, we observed that PTX3
markedly rescued lethality even after the CLP operation. These findings
suggest that the PTX3 N-terminal domain, which is sufficient for cytopro-
tection against histones, is a potential treatment for sepsis. PTX3 suppressed
both proinflammatory signals (Fig. 6C) and the plasma concentration of
histone in mice 24 hours after injection with LPS (Fig. 6B). Li et al. de-
tected histone H3 in the plasma of mice 3 hours after injection with LPS
(45). The discrepancy between that study and our current study may be due
to the difference in the dose of LPS that was injected. Suppression of the
proinflammatory responses to histones may be a result of the other effects
of PTX3 on the innate immune system and inflammation (2,3). The
aggregation-induced attenuation of histone cytotoxicity that we showed here
may enable the rapid and permanent removal of toxic molecules and thus
may provide a new strategy for the treatment of sepsis.
MATERIALS AND METHODS
Reagents
Recombinant histones were purchased from New England BioLabs Inc.
Calf thymus histones was purchased from Roche. The anti-PTX3 mouse
monoclonal antibodies PPZ-1228 [immunoglobulin G2b (IgG2b)] and
HRP-conjugated PPZ-1228 were generated as previously described (26).
The HRP-conjugated anti-6xHis antibody (Anti-His-tag HRP-DirecT) was
purchased from Medical & Biological Laboratories. The anti-myc antibody
(9E10) was purchased from Santa Cruz Biotechnology. The anti–histone
H3 antibody (ab1791) was purchased from Abcam. The synthetic human
histone H3 and H4 peptide fragments were obtained from Sigma. The
peptide sequences are provided in table S1.
Expression and purification of PTX3 proteins
Recombinant, nontagged human PTX3 (rhPTX3) protein was expressed
and purified as previously described (26). Recombinant, tagged human
PTX3 fragments (PTX3_Full, PTX3_N, and PTX3_C) were expressed
and purified as previously described (26). The recombinant tagged N-terminal
domain of human PTX3 (N-terminal wild type) and its oligomerization-
defective mutant, in which all of the cysteine residues were changed to
serine residues (N-terminal mutant), were expressed and purified as de-
scribed previously (26). Removal of endotoxin from bacterially expressed
PTX3 proteins for cell-based and animal experiments was performed with
Detoxi-Gel Endotoxin Removing Columns (Thermo Fisher Scientific
Inc.) according to the manufacturer’s instructions.
Binding assays for PTX3 proteins and histone proteins
Binding assays were performed as previously described (26). Briefly, his-
tone or a histone peptide fragment was immobilized on a polystyrene 96-
well plate. The plate was blocked with buffer containing 1% BSA. As the
primary reaction, each recombinant PTX3 protein diluted in 1% BSA
buffer was added. To detect 6xHis-tagged PTX3 protein, an HRP-conjugated
anti-6xHis antibody was used for the secondary reaction. To detect myc-
tagged PTX3 protein, an anti-myc antibody was used for the secondary
reaction, whereas a peroxidase-conjugated F(ab′)
2
fragment goat anti-
mouse IgG (H + L) was used for the third reaction. The plate was devel-
oped with soluble 3,3′,5,5′-tetramethylbenzidine (TMB) reagent, and the
reaction was stopped with TMB stop buffer. The absorbance was then read
at 450 and 630 nm. To investigate the calcium dependency of the interac-
tion between PTX3 and histone proteins, 4 mM CaCl
2
or4mMeachof
EDTA and EGTAwere added to all of the buffers.
SPR measurement
All of the SPR experiments were performed with a Biacore T200. Im-
mobilization of each recombinant human histone protein on the sensor chip
CM5 was performed with an Amine Coupling Kit using a standard coupling
protocol. The reaction conditions were set at 5 ml/min at 25°C, with HBS-P
[10 mM Hepes (pH 7.4), 150 mM NaCl, 0.05% Tween 20] as the running
buffer. All of the flow cell surfaces were activated by a 1:1 mixture of 1-ethyl-
3-[3-dimethylaminopropyl]carbodiimide hydrochloride/N-hydroxysuccinimide
(EDC/NHS) for 7 min. Next, each recombinant histone protein, diluted with
10 mM sodium acetate (pH 5.0), was injected for 7 min. Deactivation of ex-
cess reactive groups was achieved with a 7-min injection of ethanolamine.
One flow cell, which was left uncaptured in the samples, was used as a con-
trol. The kinetic analysis was performed with a single-cycle kinetics method.
With the reaction conditions set at 30 ml/min and 25°C, five twofold dilutions
of the analytes were injected sequentially for 2 min of association, followed by
2 min of dissociation. The highest concentration of each analyte was as fol-
lows: 1 nM for PTX3_Full; 5 nM for PTX3_N; 50 nM for PTX_C; 5 nM for
the N-terminal wild type; and 50 nM for the N-terminal mutant. Regenera-
tion of the surface was achieved by an injection of 0.1% SDS for 7 min. To
investigate the calcium dependency of the interaction, 3 mM CaCl
2
or 3 mM
each of EDTA and EGTA was added to the buffer during the kinetic anal-
ysis. The kinetics were calculated with Biacore T200 evaluation software.
Turbidity measurement
Turbidity measurement was performed with a NanoDrop spectro-
photometer (Thermo Scientific). Before the turbidity measurements,
samples were dialyzed with TBS containing 4 mM CaCl
2
.
Transmission electron microscopy
Transmission electron microscopy analysis was performed as previously
described (46). Briefly, the histone-PTX3 mixtures were mixed and then
fixedina1%OsO
4
solution for 1 hour. A drop of the mixture was placed
on a grid provided with a supporting collodion film. Photographs were
taken under transmission electron microscopy (S700, Hitachi).
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 16 September 2014 Vol 7 Issue 343 ra88 7
on September 16, 2014http://stke.sciencemag.org/Downloaded from
Fluorescent probe–based aggregation assay
All of the fluorescence measurements were performed with Fusion Uni-
versal Microplate Reader (PerkinElmer). To detect aggregation caused by
bsheet stacking, 5 mM (final concentration) ThT (Sigma) was added to
the samples. To detect hydrophobicity-mediated aggregation, 1 mM(final
concentration) ANS (Sigma) was added to the samples. The excitation and
emission maximums for each measurement were set to 425 and 485 nm
for ThT and 330 and 485 nm for ANS, respectively.
CD analysis
CD spectra were recorded on a J-820 spectropolarimeter (JASCO). Measure-
ments were performed in TBS containing 4 mM CaCl
2
at 25°C with a quartz
cuvette with a path length of 0.1 cm (GL Sciences Inc.). All of the measurements
were performed with the indicated concentrations of TBS containing 4 mM
CaCl
2
. Samples were scanned five times at 20 nm/min with a bandwidth of
0.1 nm and a response time of 1 s over the wavelength range 190 to 260 nm.
The data were averaged, and the spectrum of the buffer sample was subtracted.
Cytotoxicity assays
HUVECs were cultured in EGM2 medium (Clonetics) and were used for
experiments within the first six passages. For the cytotoxicity assay,
cultured HUVECs were washed with phosphate-buffered saline (PBS)
and incubated with various histones mixed with or without PTX3 in
Opti-MEM medium at 37°C for 1 hour, and then were stained by adding
PI (10 mg/ml, Sigma) for 5 min at 37°C. The cells were washed with PBS;
detached with PBS, 0.2% Pluronic F-68, and 1 mM EDTA; and then
subjected to flow cytometric analysis.
Ca
2+
signal imaging
The human umbilical vein cell line EA.hy926 cells (American Type
Culture Collection) were cultured on glass-based dishes (Asahi Glass
Co. Ltd.) in Dulbecco’s modified Eagle’s medium (DMEM) supplemented
with 10% fetal bovine serum. Cells were preloaded with Fluo 4-AM (Do-
jindo) according to the manufacturer’s instructions. After washing with
PBS, cells were incubated with Hanks’balanced salt solution (HBSS)
containing 3 mM CaCl
2
. Cells were treated with histones in the absence
or presence of PTX3 in HBSS containing 3 mM CaCl
2
, and then fluores-
cence images were acquired at 5-s intervals by confocal laser scanning
microscopy (FV1000, Olympus). The acquired images were analyzed with
MetaMorph imaging software (v6.2, Molecular Devices).
Analysis of histone localization with endothelial cells
Calf thymus histones or BSA was labeled with an Alexa Fluor 488 protein-
labeling kit (Invitrogen) according to the manufacturer’s instructions.
EA.hy926 cells were cultured on glass-based dishes. Cells were preloaded
with CellTracker Orange Fluorescent Probe (Takara Bio) according to the
manufacturer’s instructions. After washing with PBS, cells were incubated
with HBSS containing 3 mM CaCl
2
. Cells were treated with Alexa Fluor–
conjugated histones with or without PTX3 in HBSS containing 3 mM
CaCl
2
. Cells were washed with PBS, fixed with 4% paraformaldehyde
in PBS for 10 min at room temperature, washed with PBS, and incubated
with HBSS containing 3 mM CaCl
2
. Fluorescence images were captured
by confocal laser scanning microscopy (FV1000, Olympus).
Animal experiments
We used 6- to 12-week-old male C57BL/6 mice for the animal experiments.
All animal studies were approved by the Institutional Animal Care and
Use Committees at the University of Tokyo. For the histone infusion mod-
el, mice were intravenously injected with calf thymus histones at 60 or
50 mg/kg body weight. Mice were treated with or without PTX3 (N-terminal
wild type, 12 mg/kg body weight) by intraperitoneal injection just before the
histones were infused. Lung samples were fixed in formalin, embedded,
sectioned, stained with H&E, and subjected to Elastica van Gieson staining
and reticulin staining. Paraformaldehyde fixation was used for electron mi-
croscopic observation. For the LPS-induced endotoxemia model, mice were
injected with LPS (Lot. 090M4030, Sigma, 16 mg/kg body weight). Mice
were treated with or without PTX3 (N-terminal wild type, 5 mg/kg body
weight) by intraperitoneal injection 2 hours before they were challenged
with LPS. Measurement of the concentrations of IL-6 and VEGF in the plas-
ma of LPS-injected mice was performed as previously described (47). The
livers and lungs were removed 24 hours after LPS injection, fixed in 10% buf-
fered formalin, and embedded in paraffin. To identify neutrophils, histological
slides were deparaffinized and rehydrated. T he slides were stained for naphthol
AS-D chloroacetate esterase according to the manufacturer’s instructions
(Naphthol AS-D Choloroacetate Esterase Kit, Sigma-Aldrich) and were coun-
terstained with hematoxylin. The neutrophils within these areas, identified by
red cytoplasmic staining and typical polymorphonuclear morphology, were
then counted. With an imaging software (NIS- Elements D, Nikon Inc.), at least
0.25 mm
2
or more areas were selected randomly and averaged. The results
from each count were expressed as neutrophil number per 0.05 mm
2
for the
section. For immunohistochemistry, the specimens were stained with an anti-
mouse F4/80 rat monoclonal antibody (BMA Biomedicals) as previously de-
scribed (48). Stained cells were counted in 10 randomly chosen 6.25-mm
2
squares with an eyepiece (Olympus WHK 10×/20L). For CLP-induced sep-
sis model, the CLP procedure was performed as described previously (49).
Mice were treated with gentamicin (5 mg/kg body weight) and PTX3 (N-
terminal wild type, 5 mg/kg body weight) or gentamicin alone by intra-
peritoneal injection 4 hours after the CLP procedure was performed.
Statistics
Results are means ± SD unless otherwise stated. We analyzed calcium im-
aging data with Bonferroni post-tests. We analyzed mouse survival with the
log-rank test. P< 0.05 was considered statistically significant. Other statis-
tical analyses were performed with Pvalues calculated by the Student’sttest.
SUPPLEMENTARY MATERIALS
www.sciencesignaling.org/cgi/content/full/7/343/ra88/DC1
Fig. S1. Analysis of histone-PTX3 interactions by ELISA.
Fig. S2. Analysis of histone-PTX3 interactions by SPR.
Fig. S3. Analysis of the interaction between histone peptides and PTX3.
Fig. S4. UV-visible spectrum of a histone-PTX3 mixture.
Fig. S5. Assessments of histone-PTX3 aggregation.
Fig. S6. PTX3-mediated protection against cytotoxicity that is caused by the association of
histones with the endothelium.
Fig. S7. Effects of PTX3 in histone-infused mice.
Fig. S8. Time-course analysis of plasma histone H3 concentrations in LPS-treated mice.
Table S1. List of the histone H3 and H4 fragment peptides.
REFERENCES AND NOTES
1. C. Garlanda, B. Bottazzi, A. Bastone, A. Mantovani, Pentraxins at the crossroads be-
tween innate immunity, inflammation, matrix deposition, and female fertility. Annu. Rev.
Immunol. 23, 337–366 (2005).
2. B. Bottazzi, A. Doni, C. Garlanda, A. Mantovani, An integrated view of humoral innate
immunity: Pentraxins as a paradigm. Annu. Rev. Immunol. 28, 157–183 (2010).
3. A. Mantovani, C. Garlanda, A. Doni, B. Bottazzi, Pentraxins in innate immunity: From
C-reactive protein to the long pentraxin PTX3. J. Clin. Immunol. 28,1–13 (2008).
4. A. Doni, C. Garlanda, B. Bottazzi, S. Meri, P. Garred, A. Mantovani, Interactions of the
humoral pattern recognition molecule PTX3 with the complement system. Immunobiology
217, 1122–1128 (2012).
5. A. Mantovani, S. Valen tino, S. Gentile, A. Inforzato, B. Bottazzi, C. Garlanda, The
long pentraxin PTX3: A para digm for humoral pattern recogn ition molecules. Ann.
N. Y. Acad. Sci. 1285,1–14 (2013).
6. M. Presta, M. Camozzi, G. Salvatori, M. Rusnati, Role of the soluble pattern recognition
receptor PTX3 in vascular biology. J. Cell. Mol. Med. 11, 723–738 (2007).
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 16 September 2014 Vol 7 Issue 343 ra88 8
on September 16, 2014http://stke.sciencemag.org/Downloaded from
7. A. Inforza to, S. Jaillon, F. Moa lli, E. Barbati, E. B onavita, B. Bottaz zi, A. Mantovani,
C. Garlanda, The lo ng pentraxin PTX3 at the crossroads bet ween innate immuni ty
and tissue remodelling. Tissue Antigens 77,271–282 (2011).
8. A. Inforzato, P. C. Reading, E. Barbati, B. Bottazzi, C. Garlanda, A. Mantovani, The
“sweet”side of a long pentraxin: How glycosylation affects PTX3 functions in innate
immunity and inflammation. Front. Immunol. 3, 407 (2013).
9. L. Deban, B. Bottazzi, C. Garlanda, Y. M. de la Torre, A. Mantovani, Pentraxins: Mul-
tifunctional proteins at the interface of innate immunity and inflammation. BioFactors
35, 138–145 (2009).
10. P. Cieślik, A. Hrycek, Long pentraxin 3 (PTX3) in the light of its structure, mechanism
of action and clinical implications. Autoimmunity 45, 119–128 (2012).
11. G. S. Martin, D. M. Mannino, S. Eaton, M. Moss, The epidemiology of sepsis in the
United States from 1979 through 2000. N. Engl. J. Med. 348, 1546–1554 (2003).
12. A. Castellheim, O. L. Brekke, T. Espevik, M. Harboe, T. E. Mollnes, Innate immune
responses to danger signals in systemic inflammatory response syndrome and sepsis.
Scand. J. Immunol. 69, 479–491 (2009).
13. M. Jedynak, A . Siemiątkowski, K. Rygasiewicz, Molecular basics of sepsis development.
Anaesthesiol. Intensive Ther. 44, 221–225 (2012).
14. M. Aziz, A. Jacob, W. L. Yang, A. Matsuda, P. Wang, Current trends in inflammatory and
immunomodulatory mediators in sepsis. J. Leukoc. Biol. 93, 329–342 (2013).
15. B. Muller, G. Peri, A. Doni, V. Torri, R. Landmann, B. Bottazzi, A. Mantovani, C irculating
levels of the long pentraxin PTX3 correlate with severity of infection in critically ill patients.
Crit. Care Med. 29, 1404–1407 (2001).
16. T. Mauri, G. Bellani, N. Patroniti, A. Coppadoro, G. Peri, I. Cuccovillo, M. Cugno, G. Iapichino,
L. Gattinoni, A. Pesenti, A. Mantovani, Persisting high levels of plasma pentraxin 3 over the
first days after severe sepsis and septic shock onset are associated with mortality. Intensive
Care Med. 36,621–629 (2010).
17. R. Uusitalo-Seppälä, R. Huttunen, J. Aittoniemi, P. Koskinen, A. Leino, T. Vahlberg,
E. M. Rintala, Pentraxin 3 (PTX3) is associated with severe sepsis and fatal disease
in emergency room patients wi th suspected infection: A prospective cohort study.
PLOS One 8, e53661 (2013).
18. M. E. Bianchi, DAMPs, PAMPs and alarmins: All we need to know about danger. J.
Leukoc. Biol. 81,1–5 (2007).
19. K. Newton, V. M. Dixit, Signaling in innate immunity and inflammation. Cold Spring
Harb. Perspect. Biol. 4, a006049 (2012).
20. S. D enk, M. Perl, M. Huber-Lang, Damage- and pathogen-associated molecular
patterns and alarmins: Keys to sepsis? Eur. Surg. Res. 48, 171–179 (2012).
21. H. Huang, J. E vankovich, W. Yan, G. Nace, L. Zhang, M. Ross, X. Liao, T. Billiar, J. Xu,
C. T. Esmon, A. Tsung, Endogenous histones function as alarmins in sterile inflammatory
liver injury through Toll-like receptor 9 in mice. Hepatology 54, 999–1008 (2011).
22. J. Xu, X. Zhang, M. Monestier, N. L. Esmon, C. T. Esmon, Extracellular histones are
mediators of death through TLR2 and TLR4 in mouse fatal liver injury. J. Immunol.
187, 2626–2631 (2011).
23. J. Xu, X. Zhang, R. Pelayo, M. Monestier, C. T. Ammollo, F. Semeraro, F. B. Taylor,
N. L. Esmon, F. Lupu, C. T. Esmon, Extracellular histones are major mediators of
death in sepsis. Nat. Med. 15, 1318–1321 (2009).
24. B. Bottazzi, V. Vouret-Craviari, A. Bastone, L. De Gioia, C. Matteucci, G. Peri, F. Spreafico,
M. Pausa, C. D’Ettorre, E. Gianazza, A. Tagliabue, M. Salmona, F. Tedesco, M. Introna,
A. Mantovani, Multimer formation and ligand recognition by the long pentraxin PTX3.
Similarities and differences with the short pentraxins C-reactive protein and serum
amyloid P component. J. Biol. Chem. 272, 32817–32823 (1997).
25. M.Rusnati,M.Camozzi,E.Moroni,B.Bottazzi,G.Peri,S.Indraccolo,A.Amadori,A.Mantovani,
M. Presta, Selective recognition of fibroblast growth factor-2 by the long pentraxin PTX3
inhibits angiogenesis. Blood 104,92–99 (2004).
26. K. Daigo, N. Yamaguchi, T. Kawamura, K. Matsubara, S. Jiang, R. Ohashi, Y. Sudou, T. Kodama,
M. Naito, K. Inoue, T. Hamakubo, The proteomic profile of circulating pentraxin 3 (PTX3) complex
in sepsis demonstrates the interaction with azurocidin 1 and other components of neutro-
phil extracellular traps. Mol. Cell. Proteomics 11, M111.015073 (2012).
27. A. Inforzato, C. Baldock, T. A. Jowitt, D. F. Holmes, R. Lindstedt, M. Marcellini, V. Rivieccio,
D. C. Briggs, K. E. Kadler, A. Verdoliva, B. Bottazzi, A. Mantovani, G. Salvatori, A. J. Day,
The angiogenic inhibitor long pentraxin PTX3 forms an asymmetric octamer with two
binding sites for FGF2. J. Biol. Chem. 285, 17681–17692 (2010).
28. A. D. Pemberton, J. K. Brown, N. F. Inglis, Proteomic identif ication of interactions
between histones and plasma proteins: Implications for cy toprotection. Proteomics
10, 1484–1493 (2010).
29. S. L. Gonias, J. J. Pasqua, C. Greenberg, S. V. Pizzo, Precipitation of fibrinogen, fibrinogen
degradation products and fibrin monomer by histone H3. Thromb. Res. 39,97–116 (1985).
30. M. Biancalana, S . Koide, Molecular mechanism of Thioflavin-T binding to amyloid fibrils.
Biochim. Biophys. Acta 1804, 1405–1412 (2010).
31. J. R. Daban, M. D. Guasch, Exposed hydrophobic regions in histone oligomers studied
by fluorescence. Biochim. Biophys. Acta 625, 237–247 (1980).
32. R. Khurana, C. Coleman, C. Ionescu-Zanetti, S. A. Carter, V. Krishna, R. K. Grover,
R. Roy, S. Singh, Mechanism of thioflavin T binding to amyloid fibrils. J. Struct. Biol.
151, 229–238 (2005).
33. S. T. Abrams, N. Zhang, C. Dart, S. S. Wang, J. Thachil, Y. Guan, G. Wang, C . H. Toh, Human
CRP defends against the toxicity of circulating histones. J. Immunol. 191, 2495–2502 (2013).
34. T. A. Fuchs, A. A. Bhandari, D. D. Wagner, Histones induce rapid and profound
thrombocytopenia in mice. Blood 118, 3708–3714 (2011).
35. M. Nakahara, T. Ito , K. Kawahara, M. Yamamoto, T. Nagasato, B. Shrestha, S. Yamada,
T. Miyauchi, K. Higuchi, T. Takenaka, T. Yasuda, A. Matsunaga, Y. Kakihana, T. Hashiguchi,
Y. Kanmura, I. Maruyama, Recombinant thrombomodulin protects mice against histone-
induced lethal thromboembolism. PLOS One 8, e75961 (2013).
36. L. Dejager, I. Pinheiro, E. Dejonckheere, C. Libert, Cecal ligation and puncture: The
gold standard model for polymicrobial sepsis? Trends Microbiol. 19, 198–208 (2011).
37. C. M. Dobson, Protein folding and misfolding. Nature 426, 884–890 (2003).
38. A. A. Dias, A. R. Goodman, J. L. Dos Santos, R. N. Gomes, A. Altmeyer, P. T. Bozza,
M. F. Horta, J. Vilcek, L. F. Reis, TSG-14 transgenic mice have improved survival to
endotoxemia and to CLP-induced sepsis. J. Leukoc. Biol. 69, 928–936 (2001).
39. K. Daigo, A. Mantovani, B. Bottazzi, The yin-yang of long pentraxin PTX3 in inflam-
mation and immunity. Immunol. Lett. 161,38–43 (2014).
40. K. Daigo, T. Hamakubo, Host-protective effect of circulating pentraxin 3 (PTX3) and
complex formation with neutrophil extracellular traps. Front. Immunol. 3, 378 (2012).
41. T. W. Du Clos, L. T. Zlock, R. L. Rubin, Analysis of the binding of C-reactive protein to
histones and chromatin. J. Immunol. 141, 4266–4270 (1988).
42. P. S. Hicks, L. Saunero-Nava, T. W. Du Clos, C. Mold, Serum amyloid P component binds to
histones and activates the classical complement pathway. J. Immunol. 149, 3689–3694 (1992).
43. H. Olzscha, S. M. Schermann, A. C. Woerner, S. Pinkert, M. H. Hecht, G. G. Tartaglia,
M. Vendruscolo, M. Hayer-Hartl, F. U. Hartl, R. M. Vabulas, Amyloid-like aggregates
sequester numerous metastable proteins with essential cellular functions. Cell 144,
67–78 (2011).
44. J.Xu,J.Reumers,J.R.Couceiro,F.DeSmet,R.Gallardo,S.Rudyak,A.Cornelis,J.Rozenski,
A. Zwolinska, J. C. Marine, D. Lambrechts, Y. A. Suh, F. Rousseau, J. Schymkowitz, Gain of
function of mutant p53 by coaggregation with multiple tumor suppressors. Nat. Chem. Biol. 7,
285–295 (2011).
45. Y. Li, B. Liu, E. Y. Fukudome, J. Lu, W. Chong, G. Jin, Z. Liu, G. C. Velmahos, M. Demoya,
D. R. King, H. B. Alam, Identification of citrullinated histone H3 as a potential serum protein
biomarker in a lethal model of lipopolysaccharide-induced shock. Surgery 150, 442–451 (2011).
46. M. Naito, E. Wisse, Filtration effect of endothelial fenestrations on chylomicron
transport in neonatal rat liver sinusoids. Cell Tissue Res. 190, 371–382 (1978).
47. T. Minami, K. Yano, M. Miura, M. Kobayashi, J. Suehiro, P. C. Reid, T. Hamakubo, S. Ryeom,
W. C. Aird, T. Kodama, The Down syndrome critical region gene 1 short variant promoters
direct vascular bed–specific gene expression during inflammation in mice. J. Clin. Invest. 119,
2257–2270 (2009).
48. T. Yamamoto, C . Kaizu, T. Kawasaki, G. Hasegawa, H. Umezu, R. Ohashi, J. Sakurada,
S. Jiang, L. Shultz, M. Naito, Macrophage colony-stimulating factor is indispensable for repop-
ulation and differentiation of Kupffer cells but not for splenic red pulp macrophages in osteo -
petrotic (op/op) mice after macrophage depletion. Cell Tissue Res. 332, 245–256 (2008).
49. D. Rittirsch, M. S. Huber-Lang, M. A. Flierl, P. A. Ward, Immunodesign of experimental
sepsis by cecal ligation and puncture. Nat. Protoc. 4,31–36 (2009).
Acknowledgments: We thank C. Saruta and K. Suga for their excellent technical assistance.
We thank A. Mantovani [Istituto Clinico Humanitas IRCCS (Istituti di Ricovero e Cura a Carat-
tere Scientifico) and University of Milan] for critical reading and valuable suggestions about the
manuscript. We also thank K. Boru of Pacific Edit for the review of the manuscript. Funding:
This work was supported by Japan Grants-in-Aid for Scientific Research 20221010, 21590397,
and 25220205 from the Ministry of Education, Culture, Sports, Science and Technology, col-
laborative research of the University of Tokyo and JSR Corp., and Young Scientists Develop-
ment Program, Research Center for Advanced Science and Technology at University of Tokyo
(funded by Fujifilm Corp.). Author contributions: K.D. and T.H. conceived the study idea and
designed the experiments; K.D. and R.F. expressed and purified recombinant PTX3 proteins;
N.Y. and K.I. provided hypotheses and ideas on cytotoxic assay; K.D. performed ELISA and
cytotoxic assays; K.D., R.F., and T.M. designed and conducted mouse experiments; R.O.,
K.M., S.J., and M. Naito designed, conducted, and analyzed data from histological examina-
tions and electron microscopic analysis; K.D. conducted SPR analysis; K.T. provided ideas on
the aggregation assay; K.D., M. Nakakido, and K.T. designed and conducted aggregation as-
says and CD spectra measurements; K.D., K.T., and T.H. analyzed the data; and K.D. and T.H.
wrote the manuscript. Competing interests: K.D., K.I., K.T., and T.H. are co-inventors of a
patent on “Agent for trea ting or preventing sys temic inflammatory re sponse syndrome”
(WO2013191280 A1). The other authors declare that they have no competing interests.
Submitted 27 May 2014
Accepted 29 August 2014
Final Publication 16 September 2014
10.1126/scisignal.2005522
Citation: K. Daigo, M. Nakakido, R. Ohashi, R. Fukuda, K. Matsubara, T. Minami,
N. Yamaguchi, K. Inoue, S. Jiang, M. Naito, K. Tsumoto, T. Hamakubo, Protective
effect of the long pentraxin PTX3 against histone-mediated endothelial cell cytotoxicity
in sepsis. Sci. Signal. 7, ra88 (2014).
RESEARCH ARTICLE
www.SCIENCESIGNALING.org 16 September 2014 Vol 7 Issue 343 ra88 9
on September 16, 2014http://stke.sciencemag.org/Downloaded from
