Acute administration of recombinant Angiopoietin-1 ameliorates
multiple-organ dysfunction syndrome and improves survival in murine sepsis
Sascha Davida,b,1, Joon-Keun Parka,1, Matijs van Meursc,d, Jan G. Zijlstrac, Christian Koeneckee,
Claudia Schrimpfa, Nelli Shushakovaa, Faikah Guelera, Hermann Hallera, Philipp Kümpersa,f,⇑
aDepartment of Nephrology & Hypertension, Hannover Medical School, Carl-NeubergStrasse 1, D- 30625, Hannover, Germany
bCenter of Vascular Biology Research, Beth Israel Deaconess Medical Center, Harvard Medical School, 99 Brookline Ave, Boston, MA, USA
cDepartment of Critical Care, University Medical Center Groningen, Hanzeplein 1, 9713 GZ, Groningen, The Netherlands
dDepartment of Pathology and Medical Biology, University Medical Center Groningen, Hanzeplein 1, 9713 GZ Groningen, The Netherlands
eInstitute of Immunology, Hannover Medical School, Carl-NeubergStrasse 1, D-30625, Hannover, Germany
fDepartment of Medicine D, Division of General Internal Medicine, Nephrology, and Rheumatology, University Hospital Münster, Albert-Schweitzer-Strasse 33,
48149 Münster, Germany
a r t i c l ei n f o
Received 4 January 2011
Received in revised form 22 March 2011
Accepted 5 April 2011
Available online 30 April 2011
Multiple-organ dysfunction syndrome
a b s t r a c t
Introduction: Endothelial activation leading to vascular barrier breakdown plays an essential role in the
pathophysiology of multiple-organ dysfunction syndrome (MODS) in sepsis. Increasing evidence suggests
that the function of the vessel-protective factor Angiopoietin-1 (Ang-1), a ligand of the endothelial-spe-
cific Tie2 receptor, is inhibited by its antagonist Angiopoietin-2 (Ang-2) during sepsis. In order to reverse
the effects of the sepsis-induced suppression of Ang-1 and elevation of Ang-2 we aimed to investigate
whether an intravenous injection of recombinant human (rh) Ang-1 protects against MODS in murine
Methods: Polymicrobiological abdominal sepsis was induced by cecal ligation and puncture (CLP). Mice
were treated with either 1 lg of intravenous rhAng-1 or control buffer immediately after CLP induction
and every 8 h thereafter. Sham-operated animals served as time-matched controls.
Results: Compared to buffer-treated controls, rhAng-1 treated septic mice showed significant improve-
ments in several hematologic and biochemical indicators of MODS. Moreover, rhAng-1 stabilized endo-
thelial barrier function, as evidenced by inhibition of protein leakage from lung capillaries into the
alveolar compartment. Histological analysis revealed that rhAng-1 treatment attenuated leukocyte infil-
tration in lungs and kidneys of septic mice, probably due to reduced endothelial adhesion molecule
expression in rhAng-1 treated mice. Finally, the protective effects of rhAng-1 treatment were reflected
by an improved survival time in a lethal CLP model.
Conclusions: In a clinically relevant murine sepsis model, intravenous rhAng-1 treatment alone is suffi-
cient to significantly improve a variety of sepsis-associated organ dysfunctions and survival time, most
likely by preserving endothelial barrier function. Further studies are needed to pave the road for clinical
application of this therapy concept.
? 2011 Elsevier Ltd. All rights reserved.
Septic multiple-organ dysfunction syndrome (MODS) is caused
by widespread endothelial-cell activation and subsequent vascular
barrier breakdown due to exaggerated systemic inflammation in
response to a pathogen. It is characterized by an increased
expression of pro-inflammatory cytokines, systemic capillary leak-
age with tissue edema, recruitment and transmigration of leuko-
cytes, and vasodilation refractory to vasopressors [1,2]. MODS is
the leading cause of death in non-coronary intensive care units
and mortality estimates for sepsis accompanied by shock and
MODS range as high as 85% [2,3]. Despite advances in understand-
1043-4666/$ - see front matter ? 2011 Elsevier Ltd. All rights reserved.
Abbreviations: MODS, multiple-organ dysfunction syndrome; Ang-1, Angiopoie-
tin-1; Ang-2, Angiopoietin-2; rhAng-1, recombinant human Ang-1; CLP, cecal
ligation and puncture; ICAM-1, Inter-Cellular Adhesion Molecule-1; VCAM-1,
Vascular Cell Adhesion Molecule-1; ALI, acute lung injury; ARDS, acute respiratory
distress syndrome; AKI, acute kidney injury; PBS, phosphate-buffered saline; PFA,
paraformaldehyde; BAL, broncho-alveolar lavage; IL-6, Interleukin-6; MCP-1,
macrophage chemoattractant protein-1.
⇑Corresponding author at: Department of Medicine D, Division of General
Internal Medicine, Nephrology, and Rheumatology, University Hospital Münster,
Albert-Schweitzer-Strasse 33, 48149 Münster, Germany. Tel.: +49 251 8347516;
fax: +49 251 8346979.
E-mail addresses: firstname.lastname@example.org, email@example.com (P.
1Contributed equally to the manuscript and are both considered first authors.
Cytokine 55 (2011) 251–259
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/issn/10434666
ing of the pathophysiological alterations the mainstay of treatment
remains nonspecific supportive care . In particular, specific anti-
mediator strategies proved essentially ineffective, probably due to
the redundancy, heterogeneity and complexity of the hosts ´ (dis-
proportionate) immune response . Thus, an important goal in
pre-clinical research is to identify and validate novel therapeutic
approaches to prevent sepsis-associated syndromes such as acute
lung injury, acute respiratory distress syndrome (ALI/ARDS) and
acute kidney injury (AKI). Recently, the endothelial-specific Angio-
poietin-Tie2 ligand–receptor system has been recognized as a ma-
jor signalling pathway that controls vascular inflammation and
permeability, key features in the pathogenesis of ALI/ARDS and
AKI, in a non-redundant manner.
Angiopoietins are angiogenic factors essential for vascular
development, maturation, and inflammation [5–7]. As circulating
or matrix-bound molecules, Angiopoietin-1 (Ang-1) and its con-
text-dependent antagonist Angiopoietin-2 (Ang-2) bind to the
extracellular domain of the tyrosine kinase receptor Tie2 that is
predominantly expressed on endothelial cells [8,9]. Produced by
vascular smooth-muscle cells and precursor pericytes, constitutive
Ang-1 expression and low-level Tie2 phosphorylation probably
represent a control pathway that maintains vessel integrity, sup-
presses inflammatory gene expression, and prevents transmigra-
tion of leukocytes [10,11]. The importance and non-redundancy
of Ang-1/Tie2 signalling is illustrated by either Ang-1?/?and
Tie2?/?knockout mice which die in utero owing to severe vascular
remodelling defects causing perturbed vascular integrity [6,5].
Ang-2 is expressed in endothelial cells, where it is stored in gran-
ules, the so-called Weibel–Palade bodies . The release of Ang-2
upon activation of the endothelium with for instance thrombin,
histamine, or hypoxia disrupts the constitutive Ang-1/Tie2 signal-
ling by preventing Ang-1 from binding to the receptor . Conse-
quently,lossof Tie2 signalling
inflammation, ultimately leading to capillary leakage and endothe-
lial barrier breakdown [13,14].
In human endotoxemia and sepsis, circulating Ang-1 levels re-
main unchanged, or even decrease, whereas Ang-2 is rapidly re-
leased by the activated endothelium [15–17]. Hence, the balance
between circulating Ang-1 and Ang-2 is shifted in favor of the more
dynamic player Ang-2. Elevated Ang-2 levels in plasma from criti-
cally ill patients seem to correlate with the extent of pulmonary
vascular leak in ALI/ARDS , the severity of AKI , and inde-
pendently predict outcome [13,16,18–22]. In mice, injection of re-
combinant Ang-2 alone is sufficient to provoke pulmonary vascular
leak and congestion .
Consistent with the opposing roles of Ang-1 and Ang-2 on the
non-redundant Tie2 receptor pathway, mice receiving adenoviral
constructs encoding either native or engineered Ang-1 (AdAng-1)
are largely protected from vascular barrier breakdown and show
improved survival during endotoxemia [23–26]. Thus, the applica-
tion of Ang-1 may have potential as an endothelium-targeted ther-
apeutic agent in patients with septic shock and MODS [27–29].
However, as human gene therapy (with AdAng-1) is not yet feasi-
ble, we particularly wanted to test if intravenously administered
recombinant human Ang-1 (rhAng-1) is protective in a murine ce-
cal ligation and puncture (CLP) model of sepsis.
Eight- to 10-wk-old 129 SV mice (20–25 g), which we found
susceptible to AKI after lethal CLP (data not shown), were obtained
from Charles River (The Charles River Laboratories; Sulzfeld, Ger-
many). Mice were maintained on mouse chow and tap water ad
libitum in a temperature- controlled chamber at 24 ?C with a
12:12-h light–dark cycle. All procedures were approved by the lo-
cal committee for care and use of laboratory animals and were per-
2.1.1. Experimental sepsis model
Polymicrobial sepsis in mice was induced by CLP. In brief, mice
were anesthetized using isofluorane (induction 3%, maintenance
1.5%, oxygen flow 3 L/min); after median laparotomy, the cecum
was exposed, and a 5–0 silk ligature was placed 10 mm from the
cecal tip. The cecum was punctured twice with an 18-gauge needle
and gently squeezed to express a 1-mm column of fecal material.
The length of 10 mm and the needle size of 18-gauge have been
chosen to create a 50% lethality to approximately 24 h (lethal model
with MODS, i.e. 48-h mortality 100%). In order to additionally inves-
tigate a milder course of sepsis, CLP was performed using a 24-
gauge needle (sublethal model); the ligation was placed as de-
scribed above. After repositioning of the bowel, the abdomen was
closed in layers using 4–0 surgical sutures. Mice were fluid resus-
citated with pre-warmed normal saline (1 mL) intraperitoneally
(i.p.) immediately after the procedure. Sham animals underwent
the same procedure except for ligation and puncture of the cecum.
All experiments have been carried out at the same time of the day.
2.1.2. Ang-1 treatment
A fixed-dose of 1 lg rhAng-1 (R&D Systems, Minneapolis, MN)
(solved in 100 lL phosphate-buffered saline per injection) or the
same volume of control buffer (100 lL phosphate-buffered saline)
were administered intravenously immediately after CLP procedure,
as well as every 8 h thereafter until the end of the experiments.
This arbitrary dosing schedule was based on data from the litera-
ture  and on theoretical considerations.
2.1.3. Animal groups and in vivo experiments
In a first subgroup, mice were anesthetized with isofluorane for
blood sampling and subsequently sacrificed for tissue sampling at
6 and 16 h after CLP or sham surgery (n = 6 per group in the sub-
lethal CLP model, and n = 8 per group in the lethal CLP model). Kid-
neys were removed, cut in thirds and either fixed for 20 h in 3.75%
Paraformaldehyde in Soerensen’s phosphate buffer and embedded
in paraffin for histologic examination, snap frozen in isopentane
(?40 ?C) for cryostat sectioning or frozen in liquid nitrogen and
stored at ?80 ?C for qPCR analysis.
In a second subgroup, mice were anesthetized with isofluorane
for blood sampling and subsequently sacrificed for broncho-alveo-
lar lavage (BAL) at 6 and 16 h after CLP or sham surgery (n = 6 per
group, sub-lethal CLP model). Mice were intubated with a 21 G
cannula via tracheostomy and BAL fluid was obtained by injecting
and slowly withdrawing three aliquots of 1 mL of PBS each (n = 6
per group in the sub-lethal CLP model). BAL fluid was separated
from cellular components by centrifugation at 800g for five min-
utes at 4 ?C and stored at ?80 ?C. Total protein concentration in
the BAL was determined by an automated method using an Olym-
pus AU 400 analyzer (Beckman Coulter Inc., Krefeld, Germany). In
order to visualize manifest tissue changes best, all histological
analysis were performed at the late time-point only. Therefore,
lungs were harvested at 16 h after CLP or sham procedure, fixed
for 20 h in 3.75% PFA in Soerensen’s phosphate buffer, and embed-
ded in paraffin for histologic examination.
In a third subgroup, survival time was recorded hourly up to
36 h after lethal CLP (n = 10 rhAng-1 treatment, n = 15 PBS treat-
ment) or sham surgery (n = 5), respectively.
S. David et al./Cytokine 55 (2011) 251–259
2.2. Blood and serum measurements
Blood samples were obtained from the cavernous sinus using a
capillary under isofluorane anesthesia. Whole blood cell counts
(EDTA) were performed by an automated method using the Veter-
inary Hematological Analyzer (VET ABC, Scil animal care company,
Gurnee, IL). Serum creatinine and urea concentrations were mea-
sured by an automated method using an Olympus AU 400 analyzer
(Beckman Coulter Inc., Krefeld, Germany). Serum levels of the pro-
inflammatory cytokines Interleukin-6 (IL-6) and macrophage che-
moattractant protein-1 (MCP-1) were quantified by bead-based
flow cytometry assay (CBA Kit, Becton Dickinson, Franklin Lakes,
New Jersey, USA) according to the manufacturer’s instructions.
2.3. Tissue analysis
We expected changes in tissue adhesion molecules to be more
likely to observe in manifest sepsis stages. Therefore, immunofluo-
rescence (IF) was performed exclusively 16 h after CLP/sham on
formalin fixed paraffin sections (2 lm) using the following primary
antibodies: rat anti-mouse Inter-Cellular Adhesion Molecule-1
(ICAM-1 or CD54, clone KAT-1; AbD Serotec, Germany), rat anti-
mouse Vascular Cell Adhesion Molecule-1 (VCAM-1 or CD106,
clone MVCAM.A, AbD Serotec), rabbit anti-mouse E-selectin
(CD62E, Santa Cruz Biotech, USA) and rat anti-mouse Gr-1 positive
neutrophils (clone 7/4; AbDSerotec). ICAM-1 and VCAM-1 staining
was performed on ice cold acetone-fixed cryosections (6 lm), E-
selectin and Gr-1 staining on 4% PFA-fixed paraffin sections
(2 lm). Paraffin-embedded sections were dewaxed, dehydrated
and antigen-demasked with 0.05% trypsin. Nonspecific binding
sites were blocked with 10% normal donkey serum (Jackson Immu-
noResearch Laboratories, PA, USA) for 30 min. Thereafter sections
were incubated with the primary antibody for 1 h. All incubations
were performed in a humid chamber at room temperature. For
fluorescent visualization of bound primary antibodies, sections
were further incubated with Cy3-conjugated secondary antibodies
(Jackson ImmunoResearch Laboratories, PA, USA) for 1 h in the
dark. For negative controls the staining procedure was performed
as described without the primary antibodies. Specimens were ana-
lyzed using a Zeiss Axioplan-2 imaging microscope with the digital
image-processing program AxioVision 4.3 (Zeiss, Jena, Germany).
Analysis of infiltrating neutrophils was done by enumerating Gr-
1-positive cells in lung- and kidney-tissue sections (n = 5 per
group). Data are expressed as mean number of 10 (lung) or 20 (kid-
ney) randomly chosen, non-overlapping fields per section. Analysis
of adhesion molecule expression in specific vascular sections (i.e.
glomeruli, arteries, and capillaries) in the renal and pulmonary vas-
culature was done by semi-quantitative scoring as follows: score
0 = no, score 1 = very weak, score 2 = weak, score 3 = moderate,
score 4 = strong, and score 5 = very strong expression (n = 5 per
group). Data are expressed as mean score of 40 glomeruli 20 ves-
sels, or 20 randomly chosen, non-overlapping interstitial fields
(peri-tubular capillaries) per section. The investigator had no
knowledge of the treatment group assignment.
2.3.2. Gene expression analysis by quantitative RT-PCR
RNA was extracted from 20 ? 5 lm cryosections from mouse
kidney and isolated using the RNeasy Mini Plus Kit (Qiagen, West-
burg, Leusden, The Netherlands) according to the manufacturer’s
instructions. Integrity of RNA was determined by gel electrophore-
sis. RNA yield (OD 260) and purity (OD 260/OD 280) were mea-
sured by an ND-1000 UV–Vis spectrophotometer (NanoDrop
Technologies, Rockland, DE). One microgram of RNA was reverse-
transcribed using SuperScript III reverse transcriptase (Invitrogen,
Breda, The Netherlands) and random hexamer primers (Promega,
Leiden, The Netherlands). The Assay-on-Demand primers (ABI Sys-
tems, Foster City, CA) used in the PCR included the housekeeping
Mm00443242_m1), ICAM-1 (assay ID Mm00516023_m1), VCAM-
1 (assay ID Mm00449197_m1). Duplicate real-time RT-PCR
analyses were executed for each sample, and the obtained thresh-
old cycle values (CT) were averaged. According to the comparative
CT method described in the ABI manual, gene expression was
normalized to the expression of the housekeeping gene, yielding
2.3.3. Quantification of Tie2 protein levels by ELISA
To quantify the amount of Tie2 protein in the renal tissues of
mice, 15 ? 10 lm kidney slices were homogenized in 50 mM Tris
HCl buffer (pH 7.5) containing 150 mM NaCl and protein inhibitor
cocktail (Sigma–Aldrich, Germany) and centrifuged at 13,000g for
15 min. Total protein was determined by DC Protein Assay (Bio-
Rad Laboratories, Hercules, CA), before quantification of Tie2 by
ELISA (mouse Tie2 MTE200, R&D Systems, Minneapolis, MN)
according to the manufacturer’s instructions.
2.4. Statistical analysis
Data are presented as means and standard error of mean (SEM).
Multiple comparisons were analyzed for significant differences by
using the One-way analysis of variance (ANOVA) with the Tukey as
a post hoc test for multiple comparisons. Kaplan–Meier plots were
used to illustrate survival between treatment groups and statistical
assessment was performed by the log-rank test. Animals still alive
at 36 h after CLP were censored at 36 h. All tests were two-sided
and significance was accepted at P < 0.05. GraphPad Prism Version
5.02 (GraphPad Prism Software Inc., San Diego, California, USA)
was used for data analysis and figure preparation.
3.1. Ang-1 prevents pulmonary vascular leakage and attenuates
neutrophil accumulation in lung tissue
To test whether exogenous application of rhAng-1 can protect
from sepsis-associated vascular leakage we analyzed the protein
content in BAL fluid at 6 and 16 h after sub-lethal CLP. Total protein
increased roughly 2-fold at 16 h after sub-lethal CLP in the buffer-
treated mice, indicating vascular leakage and ALI due to sepsis. In
contrast, rhAng-1 was sufficient to protect the integrity of the
microvascular barrier as shown by a significantly lower protein
concentration in the BAL fluid at 16 h after sub-lethal CLP
(Fig. 1A). Next, we used immunofluorescence staining to localize
and quantify the protein expression of E-selectin within the pul-
monary vessels. We found that sub-lethal CLP significantly induced
up-regulation of E-selectin, which was significantly suppressed in
rhAng-1 injected mice (Fig. 1B). In addition we quantified Gr-1
stained lung-tissue section to determine if rhAng-1 treatment re-
duced neutrophil influx in lung tissue as well. Consistent with
BAL protein levels and semiquantitative E-selectin data, rhAng-1
treatment led to a significant reduction of neutrophil accumulation
compared with the placebo treated CLP mice (Fig. 1C).
3.2. Ang-1 treatment ameliorates leucopoenia and platelet
consumption without affecting cytokine levels
We observed a ?50% reduction in systemic leukocyte counts
at 16 h after induction of sub-lethal CLP. Interestingly, rhAng-1
treated mice were protected from sepsis-associated leucopoenia
S. David et al./Cytokine 55 (2011) 251–259
compared to buffer-treated controls (Fig. 2A). Likewise, circulating
rhAng-1 treatment prevented the ?40% decline in platelet counts
at 16 h after sub-lethal CLP, indicating less platelet consumption
(Fig. 2B). As expected, red blood cell counts remained unchanged
throughout the study period (data not shown). We also investi-
gated whether rhAng-1 treatment led to a reduction in serum lev-
els of pro-inflammatory cytokines. Interestingly, serum IL-6
(Fig. 2C) and MCP-1 levels (Fig. 2D) increased significantly after
sub-lethal CLP, but were not different between the rhAng-1 and
control-buffer treated mice.
3.3. Ang-1 prevents acute kidney injury (AKI)
Mice that underwent sub-lethal CLP exhibited only mild in-
creases in serum urea levels, whereas serum creatinine levels did
not change significantly (Supplementary data file 1). Therefore,
we investigated the effect of rhAng-1 treatment on AKI in a more
severe (so-called ‘lethal’) CLP model (48-h mortality 100%, see
material and methods). This lethal dose of CLP resulted in a signif-
icant increase in serum creatinine and urea levels in buffer-treated
controls (1.5 and 3.5-fold at 16 h, respectively). Treatment with
rhAng-1 was sufficient to protect from septic-AKI, as evidenced
by significantly lower serum creatinine and urea levels in rhAng-
1 treated mice compared to controls (Fig. 3A). As there was no con-
sistent damage detectable on light microscopy examination (H&E
staining, data not shown), we quantified Gr-1 stained kidney-tis-
sue sections to determine whether rhAng-1 treatment reduced re-
nal neutrophil influx (Fig. 3B) similar to what we found in lungs. As
shown by semi-quantitative analysis, rhAng-1 indeed led to a sig-
nificant reduction of neutrophil accumulation in the intravascular
and interstitial compartments compared with the placebo treated
CLP mice (Fig. 3C).
3.4. Ang-1 reduces up-regulation of renal vascular adhesion molecules
in a section-specific manner
Previous animal studies implied that Ang-1 gene transfer re-
duced the expression of endothelial adhesion molecules, such as
Inter-Cellular Adhesion Molecule-1 (ICAM-1) and Vascular Cell
Adhesion Molecule-1 (VCAM-1) in experimental sepsis. First, we
Fig. 1. Effects of recombinant human (rh) Angiopoietin-1 (Ang-1) on pulmonary vascular leakage, adhesion molecule expression and neutrophil accumulation in lung tissue
after sub-lethal cecal ligation and puncture (CLP) or sham operation. Recombinant Ang-1 or control buffer was injected into the cavernous sinus at CLP induction and at 8 h
thereafter. (A) Six and 16 h after CLP and 16 h after sham operation, broncho-alveolar lavage (BAL) was performed, and the protein content in BAL fluid was measured (n = 5 in
each group). Immunofluorescence study and quantification of (B) E-selectin expression (red, [400?]) and C) Gr-1 stained neutrophils (red, [200?], mean number per high
power field [No./HPF]) in lung-tissue harvested at 16 h after CLP or sham treatment. Scoring of E-selectin expression was done as follows: Score 0 = no, score 1 = very weak,
score 2 = weak, score 3 = moderate, score 4 = strong, and score 5 = very strong expression. Data are expressed as means ± SEM (n = 5 mice/group).⁄P < 0.05;⁄⁄P < 0.01 vs.
sham.?P < 0.05;?P < 0.01 vs. CLP at the same time point. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this
S. David et al./Cytokine 55 (2011) 251–259
Fig. 2. Effects of rhAng-1 on blood measurements after sub-lethal CLP or sham operation. Recombinant Ang-1 or control buffer was injected into the cavernous sinus at CLP
induction and at 8 h thereafter. Six and 16 h after CLP, and 16 h after sham treatment, blood was drawn from the cavernous sinus, and (A) white blood counts (WBC) and (B)
platelet counts were performed by an automated hematological analyzer. Serum levels of (C) Interleukin-6 (IL-6) and (D) macrophage chemoattractant protein-1 (MCP-1)
were measured by bead-based flow cytometry. Data are expressed as means ± SEM (n = 5 mice/group).⁄P < 0.05;⁄⁄P < 0.01 vs. sham.
Fig. 3. Effects of rhAng-1 on renal function and neutrophil influx after lethal CLP or sham operation. Recombinant Ang-1 or control buffer was injected into the cavernous
sinus at CLP induction and at 8 h thereafter. Six and 16 h after CLP, and 16 h after sham treatment, blood was drawn from the cavernous sinus, and (A) serum creatinine and
urea levels determined by an automated analyzer (n = 8 in each group). (B) Immunofluorescence study and quantification (mean number per high power field [No./HPF]) of
infiltrating neutrophils (red) in Gr-1 stained kidney-tissue sections at 16 h after lethal CLP or sham treatment (200?). Data are expressed as means ± SEM (n = 5 mice/group).
⁄P < 0.05;⁄⁄P < 0.01 vs. sham.?P < 0.05;?P < 0.01 vs. CLP at the same time point. (For interpretation of the references to color in this figure legend, the reader is referred to the
web version of this article.)
S. David et al./Cytokine 55 (2011) 251–259
investigated the mRNA expression levels of ICAM-1 and VCAM-1 in
whole kidney homogenates of sham operated and either rhAng-1
or control-buffer treated mice at 16 h after lethal CLP. As expected,
ICAM-1 and VCAM-1 transcript levels were significantly higher in
lethal CLP compared to sham treatment. Surprisingly, ICAM-1
and VCAM-1 transcript levels were statistically not different be-
tween the rhAng-1 and control-buffer treated mice, respectively
(Supplementary data file 2). Next, we used immunofluorescence
staining to localize and quantify the protein expression of ICAM-
1 and VCAM-1 among specific vascular sections within the renal
vasculature. We found that CLP significantly induced up-regulation
of ICAM-1 in glomeruli and peri-tubular capillaries, which was
ameliorated in rhAng-1 injected mice (Fig. 4A).
CLP-induced up-regulation of VCAM-1 expression was seen in
arteries and, to a lesser extent, in peri-tubular capillaries, but
was not observed in the glomerular endothelium (not shown). In
contrast to control buffer, rhAng-1 treatment significantly reduced
VCAM-1 expression in arteries at 16 h after CLP. However, rhAng-1
did not ameliorate up-regulation of VCAM-1 in peri-tubular capil-
laries after CLP (Fig. 4B).
3.5. Tie2 reduction during sepsis is dose-dependent and not rescued by
We have recently reported a transient downregulation of Tie2
mRNA and protein in the kidneys during LPS injection in mice
[31,32]. Thus, we investigated the expression levels of Tie2 tran-
scripts in kidney homogenates of rhAng-1 or control-buffer treated
mice at 16 h after sub-lethal and lethal CLP, respectively (Fig. 5A).
Consistent with our previous work, Tie2 transcript levels were sig-
nificantly reduced at 16 h after CLP. Interestingly, Tie2 mRNA
downregulation occurred in a dose-dependent fashion with lowest
levels detected in the lethal CLP model. Tie2 mRNA expression was
not different between rhAng-1 or control-buffer treatment in the
sub-lethal or lethal CLP model, respectively. Similarly, Tie2 mRNA
reduction was paralleled by a modest but significant decrease in
Tie2 protein levels as determined by tissue ELISA (Fig. 5B).
3.6. Ang-1 improves survival in lethal CLP
We hypothesized that the beneficial effects of rhAng-1 on sev-
eral features of MODS should be reflected by a better outcome. In-
Fig. 4. Effects of rhAng-1 on endothelial adhesion molecule expression in the renal vasculature after lethal CLP or sham operation. Recombinant Ang-1 or control buffer was
injected into the cavernous sinus at CLP induction and at 8 h thereafter. Kidney-tissues were harvested at 16 h after lethal CLP or sham treatment. Immunofluorescence study
and quantification of (A) ICAM-1 and (B) VCAM-1 protein expression among specific sections within the renal vasculature (i.e. glomeruli [400?], arteries, and peritubular
capillaries [200?]). Scoring was done as follows: Score 0 = no, score 1 = very weak, score 2 = weak, score 3 = moderate, score 4 = strong, and score 5 = very strong expression.
Data are expressed as means ± SEM (n = 5 mice/group).⁄P < 0.05;⁄⁄P < 0.01 vs. sham.?P < 0.05;?P < 0.01 vs. CLP at the same time point.
S. David et al./Cytokine 55 (2011) 251–259
deed, rhAng1 treatment could improve survival time in mice suf-
fering from a lethal dose of CLP, even without treating sepsis per
se (log-rank test: P < 0.001). Median survival after lethal CLP was
17 h in the buffer-treated control group but 30 h in the rhAng-1
treated group. The improved survival time of rhAng-1 treated mice
is visualized by Kaplan–Meier curves in Fig. 6.
In 2005 Witzenbichler et al.  were the first to report that an
Ang-1 encoding adenovirus (AdAng-1) protects mice from LPS-in-
duced endotoxic shock, as evidenced by improved cardio-pulmon-
ary function and survival. Since then, several investigators have
confirmed the beneficial effects of gene transfer of native or engi-
neered AdAng-1 variants in LPS-induced systemic capillary leakage
, AKI  and ALI/ARDS  in detail. However, from a clinical
perspective, all of the aforementioned studies share two important
limitations: First, in LPS-induced endotoxemia a well-defined bac-
terial strain or endotoxin is used to bring about the onset of sepsis.
In human sepsis, the pathogenic bacteria are often not known, and
mixed infections that involve both Gram-negative and Gram-posi-
tive bacteria are common . Second, although gene therapy con-
genetically determined and idiopathic diseases, systemic delivery
of Ang-1 using viral vectors may have limited clinical applicability
in critically-ill patients, particularly as it takes some time to reach
the peak plasma concentration of Ang-1 (i.e. 48 h).
According to the translational scope of our research, the design
of the current feasibility study is different from the above men-
tioned studies regarding (A) the choice of a clinically more relevant
sepsis model (i.e. polymicrobial CLP instead of LPS), (B) thorough
assessment of MODS and mortality, the benchmark of any success-
ful treatment regimen, and (C) the therapeutic approach of Ang-1
delivery (i.e. intravenous injection of rhAng-1 instead of AdAng-1
48 h pre-treatment). Collectively, our data show that intravenous
rhAng-1 treatment alone was sufficient to significantly improve a
variety of sepsis-associated organ dysfunctions and survival time,
most likely by preserving endothelial barrier function. This finding
supports the potential utility of rhAng-1 in the treatment of septic
The improvement in survival time and the protection from sev-
eral features of MODS is likely due to a direct anti-permeability ef-
fect of rhAng-1 on capillaries, protecting from systemic leakage
and subsequent distributive shock. Detailed in vitro studies have
provided mechanistic evidence, that Ang-1 prevents vascular per-
meability by regulating the endothelial cytoskeleton through coor-
dinated and opposite effects on the Rho GTPases Rac1 and RhoA
[13,34]. In fact, protein leakage from lung capillaries into the alve-
olar compartment was completely abolished in the lungs of rhAng-
1 treated septic mice, and leukocyte infiltration into lung tissue
was markedly reduced. Our observations are in agreement with
previously published data demonstrating that Ang-1 can inhibit
LPS-induced pulmonary hyperpermeability and leukocyte infiltra-
tion in vivo [24,25,34]. However, as AdAng-1 protects from hypo-
tension and improves cardiac output , it remains to be seen
whether the vascular protective effect of Ang-1 is only due to inhi-
bition of capillary leakage, but also by exerting effects on vascular
tone that change hydrostatic pressure to favor extravasation and
overt hypotension. The latter seems possible as Ang-1 increased
the vasoconstriction to phenylephrine in mesenteric arterioles ex-
posed to endotoxin .
Leukocyte trafficking across the vascular endothelium is criti-
cally dependent on endothelial adhesion molecules, such as
ICAM-1, VCAM-1 and E-Selectin [1,2]. In vitro, Ang-1 suppresses
adhesion molecule expression and reduces neutrophil transmigra-
tion in response to various pro-inflammatory stimuli [36–38]. Con-
sistently, leukocyte infiltration was reduced in the lung and kidney
from rhAng-1 treated mice. This was also reflected by less leuko-
cyte extravasation from the blood following rhAng-1 injection. In
fact, rhAng-1 treated mice were essentially protected from septic
Fig. 5. Effects of rhAng-1 on Tie2 receptor expression in the kidneys after CLP or
sham operation. Recombinant Ang-1 or control buffer was injected into the
cavernous sinus at CLP induction and at 8 h thereafter. Kidney-tissues were
harvested at 16 h after sub-lethal/lethal CLP or sham treatment, respectively. (A)
Tie2 mRNA expression levels in kidney homogenates are shown as relative to
GAPDH and determined by quantitative RT-PCR. (B) Tie2 protein expression in
kidney homogenates as determined by ELISA methodology. Data are expressed as
means ± SEM (n = 5 mice/group).⁄P < 0.05;⁄⁄P < 0.01 vs. sham.
Fig. 6. Effects of rhAng-1 on survival after lethal CLP or sham operation. Kaplan–
Meier curves showing the effect of rhAng-1 on survival after lethal CLP or Sham
operation. Recombinant Ang-1 (n = 15) or control buffer (n = 10) was injected into
the cavernous sinus at CLP induction and at 8 h thereafter. The difference between
the curves (CLP vs. CLP+Ang-1) was determined by using a log-rank test. Sham-
operated animals served as additional control group (n = 5).
S. David et al./Cytokine 55 (2011) 251–259
AKI as evidenced by preserved renal function parameters, although
we were forced to use a much higher CLP severity model (so called
‘lethal’) to induce AKI in the first place. However, in contrast to the
robust suppressive effect of rhAng-1 on E-selectin expression in
pulmonary vessels, the suppression of ICAM-1 and VCAM-1
expression in the kidneys from septic mice was relatively modest
and compartment-specific. Admittedly this work does not satisfac-
tory answer the question how rhAng-1 improves kidney function
but gives space for further hypothesizing. Our renal compart-
ment-specific findings regarding adhesion molecule expression
might mechanistically be involved but not exclusively responsible
for the better organ function. In fact the only modest suppression
of ICAM-1 and VCAM-1 suggests that an improvement of other fac-
tors such as systemic and intra-renal hemodynamics might be
accountable for the preserved renal function in rhAng-1 treated
In accordance with this notion, Kim and co-workers  re-
ported recently that pre-treatment with an engineered variant of
AdAng1 prevented the decrease in renal blood flow and mean arte-
rial pressure while improving glomerular filtration rate in LPS-
Furthermore, we have recently reported a transient downregu-
lation of Tie2 mRNA and protein in the kidneys during LPS injection
in mice . In the current study we show that this is also the case
in CLP and that this Tie2 downregulation is proportional to the
severity of the CLP model used. The regulation of Tie2 in sepsis
has so far not been explored mechanistically. In HUVECs, ligation
of Tie2 by Ang-1, leads to internalization of the receptor and deg-
radation of Tie2 without affecting its synthesis . The lack of
Tie2 modulation via rhAng-1 treatment might be explained by
the much higher Ang-1 concentration used in vitro to induce the
receptor internalization. Our findings might reflect enhanced deg-
radation of Tie2 induced directly or indirectly by so far undetected
It is important to recognize certain limitations of this study.
First of all, hemodynamics and cardiac function were not analyzed
in the present study. Thus, we can only speculate on the presum-
ably beneficial effects of rhAng-1 on cardiac output, arterial blood
pressure and renal blood flow that have been reported previously
[25,26]. Second, the rhAng1 dosage (1 lg per injection) and appli-
cation schedule (immediately after CLP and every 8 h thereafter)
was mainly based on previous reports and theoretical consider-
ations . Using the same rhAng-1 dosage, Hegeman et al. re-
ported that intravenously administered rhAng-1 reduced the
inflammation and granulocyte infiltration in a mouse model of
ventilator-induced lung injury . Optimizing the dosage and
schedule of administration of rhAng1 could further improve out-
come in septic MODS. However, the administration of rhAng-1 pro-
tein at high doses is costly and putative side-effects, such as
[10,32,40]. Third, administration of rhAng-1 was not able to im-
prove relative survival, but mean survival time by 75%. In a clinical
scenario, a supportive rhAng-1 regimen might provide ICU physi-
cians with extra time to localize septic foci, specify antibiotic strat-
egies, and accomplish sepsis treatment bundles. Fourth, despite
repeated attempts, we were unable to stain for E-selectin in kidney
tissue and for ICAM-1 or VCAM-1 in lung tissue. Lastly, we have to
mention that we were forced to use a higher dose of CLP to inves-
tigate rhAng-1’s effect on renal function (because the sub-lethal
dose did not induce relevant AKI). We did not investigate the ef-
fects of this lethal CLP dose in other organs.
The current feasibility study provides first evidence that intra-
venous rhAng-1 treatment can indeed reproduce the findings
pursued with AdAng-1 in a clinical relevant sepsis model. In detail,
rhAng-1 significantly protected against a variety of sepsis-associ-
ated organ dysfunctions and improved survival time, most likely
by preserving endothelial barrier function. In contrast to former
anti-mediator trials in critical care, the absence of redundant sys-
tems to take over the function of Ang/Tie2 has the advantage that
the effect of therapeutic interventions cannot easily be bypassed.
Although results from the current study are promising, further
experimental and translational research is urgently needed before
Ang/Tie2 modulation can be introduced into the clinic.
We thank Petra Berkefeld (Hannover, Germany) and Rianne M.
Jongman (UMCG, Groningen) for extraordinary technical assis-
tance. This work was supported by an internal research grant (to
P.K.) from the Medical School Hannover (HiLF). S.D. is a scholar
of the German Research Foundation DA1209/1-1.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.cyto.2011.04.005.
 Vincent JL, De Backer D. Microvascular dysfunction as a cause of organ
dysfunction in severe sepsis. Crit Care 2005;9(Suppl. 4):S9–S12.
 Aird WC. The role of the endothelium in severe sepsis and multiple organ
dysfunction syndrome. Blood 2003;101:3765–77.
 Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR.
Epidemiology of severe sepsis in the United States: analysis of incidence,
outcome, and associated costs of care. Crit Care Med 2001;29:1303–10.
 Dellinger RP, Levy MM, Carlet JM, Bion J, Parker MM, Jaeschke R, et al. Surviving
Sepsis Campaign: international guidelines for management of severe sepsis
and septic shock. Crit Care Med 2008;36:296–327.
 Davis S, Aldrich TH, Jones PF, Acheson A, Compton DL, Jain V, et al. Isolation of
angiopoietin-1, a ligand for the TIE2 receptor, by secretion-trap expression
cloning. Cell 1996;87:1161–9.
 Suri C, Jones PF, Patan S, Bartunkova S, Maisonpierre PC, Davis S, et al. Requisite
role of angiopoietin-1, a ligand for the TIE2 receptor, during embryonic
angiogenesis. Cell 1996;87:1171–80.
 Thurston G, Suri C, Smith K, McClain J, Sato TN, Yancopoulos GD, et al. Leakage-
resistant blood vessels in mice transgenically overexpressing angiopoietin-1.
 Wong AL, Haroon ZA, Werner S, Dewhirst MW, Greenberg CS, Peters KG. Tie2
expression and phosphorylation in angiogenic and quiescent adult tissues. Circ
 Fiedler U, Krissl T, Koidl S, Weiss C, Koblizek T, Deutsch U, et al. Angiopoietin-1
and angiopoietin-2 share the same binding domains in the Tie-2 receptor
involving the first Ig-like loop and the epidermal growth factor-like repeats. J
Biol Chem 2003;278:1721–7.
 Brindle NP, Saharinen P, Alitalo K. Signaling and functions of angiopoietin-1 in
vascular protection. Circ Res 2006;98:1014–23.
 Augustin HG, Koh GY, Thurston G, Alitalo K. Control of vascular morphogenesis
and homeostasis through the angiopoietin-Tie system. Nat Rev Mol Cell Biol
 Fiedler U, Scharpfenecker M, Koidl S, Hegen A, Grunow V, Schmidt JM, et al.
The Tie-2 ligand angiopoietin-2 is stored in and rapidly released upon stimu-
lation from endothelial cell Weibel-Palade bodies. Blood 2004;103:4150–6.
 Parikh SM, Mammoto T, Schultz A, Yuan HT, Christiani D, Karumanchi SA, et al.
Excess circulating angiopoietin-2 may contribute to pulmonary vascular leak
in sepsis in humans. PLoS Med 2006;3:e46.
 Roviezzo F, Tsigkos S, Kotanidou A, Bucci M, Brancaleone V, Cirino G, et al.
Angiopoietin-2 causes inflammation in vivo by promoting vascular leakage. J
Pharmacol Exp Ther 2005;314:738–44.
 David S, van Meurs M, Kumpers P. Does low angiopoietin-1 predict adverse
outcome in sepsis? Crit Care 2010;14:180.
 Kumpers P, van Meurs M, David S, Molema G, Bijzet J, Lukasz A, et al. Time
course of angiopoietin-2 release during experimental human endotoxemia and
sepsis. Crit Care 2009;13:R64.
 Mankhambo LA, Banda DL, Study Group, Jeffers G, White SA, Balmer P, et al.
The role of angiogenic factors in predicting clinical outcome in severe bacterial
infection in Malawian children. Crit Care 2010;14:R91.
 Van der Heijden M, Nieuw Amerongen GP, Koolwijk P, Van Hinsbergh VW,
Groeneveld AB. Angiopoietin-2 permeability oedema, occurrence and severity
of ALI/ARDSin septicand non-septic
S. David et al./Cytokine 55 (2011) 251–259
 Kumpers P, Hafer C, David S, Hecker H, Lukasz A, Fliser D, et al. Angiopoietin-2 Download full-text
in patients requiring renal replacement therapy in the ICU: relation to acute
kidney injury, multiple organ dysfunction syndrome and outcome. Intensive
Care Med 2010;36:462–70.
 Kumpers P, Lukasz A, David S, Horn R, Hafer C, Faulhaber-Walter R, et al.
Excess circulating angiopoietin-2 is a strong predictor of mortality in critically
ill medical patients. Crit Care 2008;12:R147.
 Giamarellos-Bourboulis EJ, Kanellakopoulou K, Pelekanou A, Tsaganos T,
Kotzampassi K. Kinetics of angiopoietin-2 in serum of multi-trauma
patients: correlation with patient severity. Cytokine 2008;44:310–3.
 Lukasz A, Hellpap J, Horn R, Kielstein JT, David S, Haller H, et al. Circulating
angiopoietin-1 and angiopoietin-2 in critically ill patients: development and
clinical application of two new immunoassays. Crit Care 2008;12:R94.
 Huang YQ, Sauthoff H, Herscovici P, Pipiya T, Cheng J, Heitner S, et al.
Angiopoietin-1 increases survival and reduces the development of lung edema
induced by endotoxin administration in a murine model of acute lung injury.
Crit Care Med 2008;36:262–7.
 Hwang JA, Lee EH, Lee SD, Park JB, Jeon BH, Cho CH. COMP-Ang1 ameliorates
endotoxemia. Biochem Biophys Res Commun 2009;381:592–6.
 Kim DH, Jung YJ, Lee AS, Lee S, Kang KP, Lee TH, et al. COMP-angiopoietin-1
 Witzenbichler B, Westermann D, Knueppel S, Schultheiss HP, Tschope C.
Protective roleof angiopoietin-1
 van der Heijden M, van Nieuw Amerongen GP, Chedamni S, vanHinsbergh VW,
Johan Groeneveld AB. The angiopoietin-Tie2 system as a therapeutic target
in sepsis and acute lung injury. Expert Opin Ther Targets 2009;13:39–53.
 van Meurs M, Kumpers P, Ligtenberg JJ, Meertens JH, Molema G, Zijlstra JG.
Bench-to-bedside review: angiopoietin signalling in critical illness – a future
target? Crit Care 2009;13:207.
 van Nieuw Amerongen GP, Groeneveld AB. A plethora of angiopoietin-2 effects
during clinical sepsis. Crit Care 2010;14:166.
endothelialtight junctions during
 Childs EW, Tharakan B, Byrge N, Tinsley JH, Hunter FA, Smythe WR.
hyperpermeability following hemorrhagic shock. Am J Physiol Heart Circ
 Van Meurs M, Kurniati NF, Wulfert F, Asgeirsdottir SA, de Graaf IA, Satchell
SC, et al. Shock-induced stress induces loss of microvascular endothelial
Tie2 in the kidney which is not associated with reduced glomerular barrier
function. Am J Physiol Renal Physiol 2009;297(2):F272–81. Epub 2009
 Hegeman MA, Hennus MP, van MM, Cobelens PM, Kavelaars A, Jansen NJ, et al.
Angiopoietin-1 treatment reduces inflammation but does not prevent
ventilator-induced lung injury. PLoS One 2010;5:e15653.
 Rittirsch D, Hoesel LM, Ward PA. The disconnect between animal models of
sepsis and human sepsis. J Leukoc Biol 2007;81:137–43.
 Mammoto T, Parikh SM, Mammoto A, Gallagher D, Chan B, Mostoslavsky G,
et al. Angiopoietin-1 requires p190 RhoGAP to protect against vascular leakage
in vivo. J Biol Chem 2007;282:23910–8.
 Hall E, Brookes ZL. Angiopoietin-1 increases arteriolar vasoconstriction to
phenylephrine during sepsis. Regul Pept 2005;131:34–7.
 Kim I, Moon SO, Park SK, Chae SW, Koh GY. Angiopoietin-1 reduces VEGF-
stimulated leukocyte adhesion to endothelial cells by reducing ICAM-1,
VCAM-1, and E-selectin expression. Circ Res 2001;89:477–9.
 Gamble JR, Drew J, Trezise L, Underwood A, Parsons M, Kasminkas L, et al.
Angiopoietin-1 is an antipermeability and anti-inflammatory agent in vitro
and targets cell junctions. Circ Res 2000;87:603–7.
 Pizurki L, Zhou Z, Glynos K, Roussos C, Papapetropoulos A. Angiopoietin-1
inhibits endothelial permeability, neutrophil adherence and IL-8 production.
Brit J Pharmacol 2003;139:329–36.
 Bogdanovic E, Nguyen VP, Dumont DJ. Activation of Tie2 by angiopoietin-1 and
angiopoietin-2 results in their release and receptor internalization. J Cell Sci
 Kumpers P, Nickel N, Lukasz A, Golpon H, Westerkamp V, Olsson KM, et al.
Circulating angiopoietins in idiopathic pulmonary arterial hypertension. Euro
Heart J 2010;31:2291–300.
S. David et al./Cytokine 55 (2011) 251–259