Bone marrow derived mesenchymal stem cells inhibit inflammation and preserve vascular endothelial integrity in the lungs after hemorrhagic shock.

Abstract

Hemorrhagic shock (HS) and trauma is currently the leading cause of death in young adults worldwide. Morbidity and mortality after HS and trauma is often the result of multi-organ failure such as acute lung injury (ALI) and acute respiratory distress syndrome (ARDS), conditions with few therapeutic options. Bone marrow derived mesenchymal stem cells (MSCs) are a multipotent stem cell population that has shown therapeutic promise in numerous pre-clinical and clinical models of disease. In this paper, in vitro studies with pulmonary endothelial cells (PECs) reveal that conditioned media (CM) from MSCs and MSC-PEC co-cultures inhibits PEC permeability by preserving adherens junctions (VE-cadherin and β-catenin). Leukocyte adhesion and adhesion molecule expression (VCAM-1 and ICAM-1) are inhibited in PECs treated with CM from MSC-PEC co-cultures. Further support for the modulatory effects of MSCs on pulmonary endothelial function and inflammation is demonstrated in our in vivo studies on HS in the rat. In a rat "fixed volume" model of mild HS, we show that MSCs administered IV potently inhibit systemic levels of inflammatory cytokines and chemokines in the serum of treated animals. In vivo MSCs also inhibit pulmonary endothelial permeability and lung edema with concurrent preservation of the vascular endothelial barrier proteins: VE-cadherin, Claudin-1, and Occludin-1. Leukocyte infiltrates (CD68 and MPO positive cells) are also decreased in lungs with MSC treatment. Taken together, these data suggest that MSCs, acting directly and through soluble factors, are potent stabilizers of the vascular endothelium and inflammation. These data are the first to demonstrate the therapeutic potential of MSCs in HS and have implications for the potential use of MSCs as a cellular therapy in HS-induced lung injury.

Full text

Page 1

Bone Marrow Derived Mesenchymal Stem Cells Inhibit
Inflammation and Preserve Vascular Endothelial Integrity
in the Lungs after Hemorrhagic Shock
Shibani Pati1*, Michael H. Gerber1, Tyler D. Menge1, Kathryn A. Wataha1, Yuhai Zhao1, John Adam
Baumgartner1, Jing Zhao3, Phillip A. Letourneau1, Maria P. Huby1, Lisa A. Baer1, John R. Salsbury1,
Rosemary A. Kozar1, Charles E. Wade1, Peter A. Walker3, Pramod K. Dash2, Charles S. Cox Jr.3, Marie-
Francoise Doursout4, John B. Holcomb1
1Department of Surgery and Center for Translational Injury Research, University of Texas Health Science Center at Houston, Houston, Texas, United States of America,
2Department of Neurobiology and Anatomy, University of Texas Health Science Center at Houston, Houston, Texas, United States of America, 3Department of Pediatric
Surgery, University of Texas Health Science Center at Houston, Houston, Texas, United States of America, 4Department of Anesthesiology, University of Texas Health
Science Center at Houston, Houston, Texas, United States of America
Abstract
Hemorrhagic shock (HS) and trauma is currently the leading cause of death in young adults worldwide. Morbidity and
mortality after HS and trauma is often the result of multi-organ failure such as acute lung injury (ALI) and acute respiratory
distress syndrome (ARDS), conditions with few therapeutic options. Bone marrow derived mesenchymal stem cells (MSCs)
are a multipotent stem cell population that has shown therapeutic promise in numerous pre-clinical and clinical models of
disease. In this paper, in vitro studies with pulmonary endothelial cells (PECs) reveal that conditioned media (CM) from MSCs
and MSC-PEC co-cultures inhibits PEC permeability by preserving adherens junctions (VE-cadherin and b-catenin). Leukocyte
adhesion and adhesion molecule expression (VCAM-1 and ICAM-1) are inhibited in PECs treated with CM from MSC-PEC co-
cultures. Further support for the modulatory effects of MSCs on pulmonary endothelial function and inflammation is
demonstrated in our in vivo studies on HS in the rat. In a rat ‘‘fixed volume’’ model of mild HS, we show that MSCs
administered IV potently inhibit systemic levels of inflammatory cytokines and chemokines in the serum of treated animals.
In vivo MSCs also inhibit pulmonary endothelial permeability and lung edema with concurrent preservation of the vascular
endothelial barrier proteins: VE-cadherin, Claudin-1, and Occludin-1. Leukocyte infiltrates (CD68 and MPO positive cells) are
also decreased in lungs with MSC treatment. Taken together, these data suggest that MSCs, acting directly and through
soluble factors, are potent stabilizers of the vascular endothelium and inflammation. These data are the first to demonstrate
the therapeutic potential of MSCs in HS and have implications for the potential use of MSCs as a cellular therapy in HS-
induced lung injury.
Citation: Pati S, Gerber MH, Menge TD, Wataha KA, Zhao Y, et al. (2011) Bone Marrow Derived Mesenchymal Stem Cells Inhibit Inflammation and Preserve
Vascular Endothelial Integrity in the Lungs after Hemorrhagic Shock. PLoS ONE 6(9): e25171. doi:10.1371/journal.pone.0025171
Editor: Christian Schulz, Heart Center Munich, Germany
Received June 22, 2011; Accepted August 26, 2011; Published September 28, 2011
Copyright: ? 2011 Pati et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by funding from Mission Connect: The Institute for Rehabilitation and Research and the National Institutes of Health
T32GM008792-11. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: shibani.pati@uth.tmc.edu
Introduction
Traumatic injury is currently one of the leading causes of death
worldwide. One of the hallmarks of hemorrhagic shock (HS), a
condition resulting from rapid blood loss after traumatic injury, is
the onset of a systemic response that results in endothelial injury,
inflammation, aberrant coagulation, tissue edema and end organ
injury [1]. HS is associated with a high incidence of acute lung
injury (ALI) and acute respiratory distress syndrome (ARDS) (14–
20% of ICU patients) that results in significant morbidity and
mortality [2–7]. There are currently few treatments beyond
supportive therapy to treat lung edema and injury. Over the past
ten years, the therapeutic potential of the multipotent bone
marrow derived stem cell population known as mesenchymal stem
cells (MSCs) have been investigated in multiple disease states
including sepsis [8], acute renal failure [9], graft vs. host disease
[10], ALI [11], and myocardial infarction [12]. MSCs have
multiple characteristics that make them attractive candidates for
clinical translation: 1) They are relatively easy to isolate, 2) They
grow rapidly in culture, and 3) They are able to home to sites of
active tissue injury where they are thought to modulate
inflammation and vascular function [13]. There are currently
154 clinical trials registered with Clinicaltrials.gov investigating the
use of adult MSCs in a number of conditions; however, there are
no clinical trials investigating the use of MSCs in HS, ALI or
ARDS despite the significant pre-clinical data to support their use
and therapeutic benefits. In our past work, we have shown that
intravenously (IV) administered MSCs have potent stabilizing
effects on the vascular endothelium in injury [14] and are capable
of inhibiting blood brain barrier (BBB) permeability after
traumatic brain injury (TBI) via modulation of the adherens
junctions (AJs) proteins: VE-cadherin and b-catenin. In endothe-
PLoS ONE | www.plosone.org1 September 2011 | Volume 6 | Issue 9 | e25171

Page 2

lial cells (ECs) these AJ proteins form molecular ‘‘zippers’’
between neighboring ECs, hence regulating paracellular perme-
ability and tissue edema [15]. In addition, our past work has
demonstrated that the influence of MSCs on endothelial per-
meability and barrier function is likely mediated by a secreted
factor(s) that is released as a result of MSC-endothelial cell
interactions. Based upon our past work and that of others
showing that MSCs have potent anti-inflammatory effects in
multiple disease states [13] we hypothesized that MSCs would
have similar stabilizing effects in the lungs exposed to HS. To
address our hypothesis we chose to study pulmonary endothelial
cell-MSC interactions in vitro and the effects of MSCs in vivo in a
rat model of HS and resuscitation.
Methods
Ethics Statement
All experiments involving the use of animals were in compliance
with the National Institutes of Health Guide for the Care and Use of
Laboratory Animals and were approved by the University of Texas
Health Science Center at Houston’s Institutional Animal Care and
Use Committee, #HSC-AWC-10-073.
Primary cells and cell lines
First passage human MSCs and PECs were purchased from
Lonza (Walkersville, MD). MSCs were cultured in MSC growth
media (MSC-GM, Lonza); PECs were maintained in EGM-2MV
media. MSCs were used at passage 3–7 for all experiments. U937s
were obtained from ATCC (Bethesda, MD) and passaged in
RPMI with 10% fetal bovine serum. All cell lines were maintained
at 37uC and 5% CO2.
Flow cytometry
MSCs were characterized by flow cytometry as described
previously [16]. PECs were treated with CM from MSCs,
PEC+MSCs, and PECs alone. Cells were stimulated with TNFa,
50 ng/ml for four hours. After treatment, cells were harvested and
stained with antibodies to VCAM-1 (Becton Dickinson, Franklin
Lakes, NJ), ICAM-1 (Becton Dickinson), E-Selectin (Becton
Dickinson), and P-Selectin (Becton Dickinson). Cells were stained
with secondary antibodies and ran on a flow cytometer (BD
FacsCalibur, BD Biosciences, San Jose, CA).
Flow Cytometry of Adhesion Marker Expression on PECs
Conditioned media was made in three groups: (1) PECs alone,
(2) MSCs alone, (3) PECs/MSCs co-culture. EGM-2 (Lonza) was
added to wells as culture medium. After 48 hours of culture, the
media was removed and stored at 280uC. PECs grown to
confluence on 6 well plates and growth media was replaced with
the conditioned media in the above groups. The experiment was
performed in replicates of four. After 12 hour incubation, TNFa
(50 ng/mL) was added to the wells and incubated for 4 hours. The
cells were then collected and stained with fluorophore conjugated
antibodies to I-CAM, V-CAM, E-selectin, and P-selectin (BD
Biosciences). The cells were then analyzed on the BD LSR II (BD
Biosciences) flow cytometer.
Conditioned media (CM)
PECs were co-cultured in contact with MSCs. 56105cells were
seeded on a 4.67 cm2surface area. The ratio of MSCs to PECs
was 0.2 determined by previously reported experiments [14,16].
CM was harvested from co-cultures after 24 hours. See schematic
in Figure 1.
Endothelial cell permeability
Collagen coated 0.4 mm pore size inserts were obtained from
BD Biosciences. PECs were seeded at 40,000 cells per insert well in
a total volume of 500 ml of media and cultured for 8 hours to allow
for cell attachment and adhesion. PEC monolayer permeability
was tested 8 hours later by adding 10 ml of 10 mg/ml 40 kD
FITC-Dextran (Sigma-Aldrich, St. Louis, MO) to the upper
chamber of each well. Monolayers were treated with 10 ng/ml
VEGF-A165 from R&D (Minneapolis, MN), one hour prior to
addition of FITC-Dextran-40 kD. Measurements were deter-
mined with a fluorimeter using excitation and emission wave-
lengths of 485 nm and 530 nm, respectively, from samples
obtained 30 minutes after addition of FITC-Dextran. Data
represent the mean 6 standard error of the mean (SEM) from
three independent experiments. For experiments in which the
effects of CM from MSCs or MSC/PEC co-cultures were studied,
the media from PEC monolayer co-cultures that had been plated
on collagen coated inserts and allowed to adhere was replaced with
undiluted CM and maintained in culture for four hours prior to
addition of FITC-Dextran. Data represent the mean 6 SEM from
three independent experiments.
In vitro VE-cadherin/b-catenin staining (PECs)
Passage 3 PECs were seeded into 8-well glass chambered slides
and grown for 48 hours at 37uC. The cells were then treated with
basal growth media or CM from PEC-MSC co-cultures, PECs
alone, or MSCs alone for 60 minutes at 37uC. To induce
permeability, recombinant human VEGF-A (R&D Systems #293-
VE-010, Minneapolis, MN) was added at 50 ng/mL and
incubated for 30 minutes at 37uC. The cells were then fixed in
2% formaldehyde and blocked in TBST+2.5% normal goat serum
for 60 minutes at room temperature. A rabbit a-VE-cadherin
antibody (1:400 dilution, Cell Signaling #2500 Beverly, MA) and
a mouse a-b-catenin antibody (1:200 dilution, Cell Signaling
#2677) were applied overnight at 4uC and detected using an
Alexa 488 a-rabbit antibody (1:500 dilution, Molecular Probes
#A-11034, Invitrogen, Carlsbad, CA) and an Alexa 568 a-mouse
antibody (1:500 dilution, Invitrogen Molecular Probes #A-11031,
Invitrogen). The next day the slides were mounted using ProLong
Gold anti-fade reagent with DAPI (Molecular Probes P-36931,
Invitrogen) to obtain nuclear staining. Images of the cells were
taken at 406 on a Nikon A1R confocal microscope (Nikon
Instruments, Inc, Melville, NY).
Leukocyte binding assay
PECs were grown to confluence on 96 well plates. 1610‘4 cells/
well were seeded and incubated at 37uC for 2 days or until
confluent. Adhesion molecule expression was stimulated by the
addition of TNFa (50 ng/ml). Cells were treated with CM for
1.5 hours. CM was generated from co-cultured cells and cells
grown alone in PEC growth media. U937 cells were labeled with
Calcein-AM (Invitrogen), added to wells and allowed to adhere for
one hour. Non-adherent cells were gently washed away in
phosphate buffered saline (PBS) and labeled cells that remained
were quantified by fluorescent reading on the Synergy 2
Mircoplate Reader (Biotek, Winooski, VT) at 490 nm wavelength
excitation and 520 nm emission.
Statistical analysis
Data in all graphs is shown as mean 6 SEM. One-way
ANOVA was used for statistical analysis of dose response
permeability studies, adhesion markers and in vivo quantitation of
cytokines and inflammatory infiltrates.
Pulmonary Effects of MSCs in Hemorrhagic Shock
PLoS ONE | www.plosone.org2 September 2011 | Volume 6 | Issue 9 | e25171

Page 3

Figure 1. Conditioned media (CM) inhibits permeability and restores adherens junctions (AJs). Figure 1A, CM from MSCs and MSC-PEC
co-cultures inhibits PEC permeability in vitro. A schematic depicting how CM is prepared is shown. MSCs and PECs are cultured separately and
together in co-culture. CM is collected 24 hours later and used in a transwell assay of PEC permeability to 40 kD FITC-Dextran. Figure 1A shows that
CM from both MSCs and MSC-PEC co-cultures inhibits permeability induced by VEGF-A (10 ng/ml). (*) signs indicate significance p,0.05 between
MSC and VEGF and MSC-PEC and VEGF groups. Figures 1B–E, CM from MSCs and MSC-PEC co-cultures restore AJs. PECs were cultured overnight.
After 24 hours, breakdown of AJs (b-catenin and VE-Cadherin) is induced by treatment with VEGF-A (10 ng/ml). Cells were subsequently treated with
CM. Figures 1B–E show that MSC and MSC-PEC CM restored AJ breakdown as determined by confocal microscopy imaging of cultures stained with
antibodies to VE-Cadherin (red) and b-Catenin (green).
doi:10.1371/journal.pone.0025171.g001
Pulmonary Effects of MSCs in Hemorrhagic Shock
PLoS ONE | www.plosone.org3September 2011 | Volume 6 | Issue 9 | e25171

Page 4

Animals and HS model
Surgical Preparation: Under isoflurane anesthesia, male Sprague-
Dawley rats (Harlan Laboratories, Indianapolis, IN) were
instrumented by placement of 23 g sterile polyurethane tubing
(Access Technologies, Skokie, IL.) into the femoral artery and
vein. The catheters were subcutaneously tunneled and exited the
dorsal neck. After a 3-day recovery period, the arterial catheter
was connected to a physiological pressure transducer (MLT844,
distributed by AD Instruments, Colorado Springs, CO). Pressures
and heart rates were recorded by a PowerLab 4/30 (AD
Instruments). Rats were randomly assigned to one of four groups.
Group I: Sham. Animals received no injury or treatment. Group
II: HS. Animals received hemorrhage alone. Group III:
HS+resuscitation with Lactated Ringer’s Solution (LR). Animals
received hemorrhage and were then treated with LR (36 shed
blood volume). Group IV: LR+MSC. Animals experienced
hemorrhage, were resuscitated with LR as Group III and also
26106MSCs were administered with LR at 1 and 24 hours, re-
suspended in 100 ml of PBS. Blood samples were taken at 2 hours
post hemorrhage and pressures recorded at baseline, end of
hemorrhage, 1, 2, and 96 hours after hemorrhage. Hemorrhage
was achieved by withdrawing 2 ml/100 g body weight of blood
from the venous or arterial catheter over 10 minutes. This yielded
a 30% blood loss. This procedure was the same for all animals
receiving hemorrhage. Resuscitation was administered at one hour
post hemorrhage. LR was administered at a volume of 3 times the
shed blood volume over 30 minutes. After 96 hours, the rats were
euthanized under isoflurane anesthesia.
Tissue harvesting and sectioning protocol for lung tissue
The lungs were harvested, and used for measurements of wet-to-
dry ratios, molecular biology studies (left side, one lobe) histopa-
thology, and immunohistochemistry (right side, four lobes). Ninety-
six hours after the protocol was started, animals were euthanized
under intubated anesthesia (5% isoflurane through a 16620G
Terumo Surflo I.V catheter). The chest was opened and a
hemostatic clamp was placed on the left main bronchus and
2.5 ml of a solution of 0.5 M sucrose and OCT (Tissue Tek
embedding medium for frozen tissue, Sakura Fineteck USA Inc.,
Torrance, CA) were mixed in equal volume and injected through
theendotrachealtube,tofilltherightlobe.Therightmainbronchus
was clamped, cut over the clamp to secure the embedding media
inside the lobes and tied with 4-O silk. The tissue was immediately
dropped in HistoPrep disposable base molds (Fisher Scientific,
Pittsburg, PA) filled with OCT, wrapped in labeled foil and snap
frozen in liquid nitrogen. This side of the lung was stored at 280uC
and is cryosectioned later for immunohistochemistry staining.
Another piece of lung was fixed in formalin for paraffin fixation
and histopathological analysis. The tissue was moved from the
280uC freezer to the cryostat (Shandon Cryotome SME, Thermo
Fisher Scientific, Waltham, MA). Sections (6 mm) were cut at a
working temperature of 217uC, and mounted to microscope slides
(VWR Vista VisionTMHistoBondH slides, VWR International,
Radnor, PA). The slides were kept at 280uC until stained.
Immunohistochemistry
Lung sections (6 mm) were cut on a cryostat, mounted on slides,
post-fixed for 20 minutes with 220uC methanol, and rehydrated
in PBS. Sections were incubated in primary antibodies (1.0 mg/ml
in PBS containing 2.5% goat serum and 0.25% Triton X-100) for
48 hours at 4uC, washed in PBS, and then incubated with Alexa
Fluor-conjugated species-specific secondary antibodies for 3 hours
at room temperature. Sections were cover slipped with Fluor-
mount-G (Fisher Scientific), and imaged using an upright
Olympus BX-51 microscope (Bio-Rad Laboratories, Hercules,
CA) attached to a MagnaFire digital camera (Optronics, Goleta,
CA). Bisbenzamide H was added in the final TBS wash sequence
to obtain nuclear staining. Antibodies used: monoclonal to a-
myeloperoxidase antibody (MPO) (Abcam, Cambridge, MA),
CD68 antibody (Abcam), VE-cadherin (Cell Signaling), Occludin-
1 (Invitrogen), Claudin-1 (Invitrogen) and Human mitochondrial
antibody (Millipore, Concord, MA).
Lung wet-to-dry ratios
The clamp was removed from the left bronchus, and the left
lung was harvested. One piece of this lobe was taken, moved to a
small microcentrifuge tube, snap frozen in liquid nitrogen and
stored at 280uC for molecular biology studies (measurement of
cytokines in tissue). The other piece was placed on a labeled plastic
dish. The dish containing the lung was weighed immediately after
harvesting (day 0 weight), and then stored in a 50uC oven to dry
for 72 hours and reweighed (day 3 weight) to calculate the wet-to-
dry ratio. The ratios between the day 0 and day 3 weights are the
wet-dry/dry ratios and are considered a measurement of lung
edema.
Cytokine and chemokines by Bioplex (tissue and serum)
Cytokines were measured in the tissue at four days and in the
serum at two hours. Two hours was picked as the main time point
to determine inflammatory changes since previous studies with this
model had shown that the maximum changes in inflammatory
cytokines occur between 1–4 hours after HS [17–19]. The Bio-
Plex cytokine assay system (Bio-Rad Laboratories) and concen-
trations of IL-1a, IL-1b, IL-6, IL-10, MIP-1a, MCP-1 and TNFa
were simultaneously evaluated using a commercially available
multiplex bead-based immunoassay (Rat 9-Plex, Bio-Rad Labo-
ratories). The assay was performed per the manufacturer’s
instructions and the details have been previously published by
our group and others [20]. High standard curves (low RP1 target
value) for each soluble cytokine were used, ranging from 2 to
32,000 pg/ml. A minimum of 100 beads per cytokine region were
evaluated and recorded. Values with a coefficient of variation
beyond 10% were not included in the final data analysis. All
samples were run in duplicate.
Results
Conditioned media from MSCs inhibits pulmonary
endothelial cell permeability in vitro
In our past work with human umbilical vein endothelial cells
(HUVECS), we have shown HUVECs in contact with MSCs
produce a soluble factor(s) that inhibits paracellular permeability
and may be responsible for the enhanced barrier integrity in
MSC/HUVEC co-cultures [14]. To gain further insight into the
mechanism by which MSCs could potentially modulate PEC
permeability, we investigated whether MSCs could inhibit
paracellular permeability in PECs. We tested the CM from cells
grown in two different formats (MSCs alone and PEC+MSCs) for
their ability to reduce paracellular permeability induced by
VEGF-A. In each format, the total number of cells seeded in
the culture was held constant at 26105cells per well. For the co-
cultured conditions, MSCs were added to give a final MSC to
PEC ratio of 0.2; we have previously used this ratio to demonstrate
effects on endothelial function [14,16]. Conditioned media was
harvested 24 hours after seeding and tested for their effect on
permeability (Figure 1A). Permeability to 40 kD FITC-Dextran
for each culture condition was assessed and normalized to the
permeability values obtained using CM from PECs alone.
Pulmonary Effects of MSCs in Hemorrhagic Shock
PLoS ONE | www.plosone.org4September 2011 | Volume 6 | Issue 9 | e25171

Page 5

Figure 1B demonstrates that the permeability of cultured PECs is
induced by VEGF-A and is inhibited by CM from MSC as well as
MSC-PEC. This finding differs from our previous results with
HUVECs, where the effects on permeability were only noted in
the CM from HUVEC-MSC co-cultures, indicating that the
effects may be dependent on the endothelial cell type since the
endothelium is highly heterogeneous and varies greatly between
the organ of origin [21].
MSCs increase VE-cadherin/b-catenin interaction at the
cell surface of PECs
As mentioned previously, AJs (VE-cadherin and b-catenin) are
critical regulators of vascular integrity and permeability [22,23].
To determine if b-catenin is involved in the decrease in barrier
permeability we observed (Figure 1B), we looked at the effects of
the MSCs on b-catenin and VE-cadherin localization in PECs
treated with CM from MSCs and MSCs+PECs as described
above. Figures 1C–E reveal that CM from MSCs and MSC-PEC
co-cultures restore AJs (b-catenin and VE-cadherin) staining at the
surface of PECs that have been treated with VEGF-A. PECs
treated with CM from MSCs and MSC-PEC co-cultures
demonstrate enhanced VE-cadherin and b-catenin immunoreac-
tivity at the membrane and an enhanced propensity to grow in
clusters. Furthermore our findings with CM suggest, as we have
found in past work, that a secreted factor(s) from MSCs is
responsible for improving the barrier integrity of PECs. Interest-
ingly, in these studies with PECs, we find that CM from MSCs
alone has potent effects on AJs and permeability.
MSCs inhibit leukocyte adhesion to PECs
Multiple factors are responsible for the compromise of the
endothelial barrier in HS-induced lung injury. In addition to
factors such as hypoxia and thrombin, inflammatory changes
caused by cytokine and chemokine release, leukocyte adhesion,
diapedesis and infiltration into the lungs have all been shown to
contribute to the clinical severity and outcome in ALI/ARDS [4].
We sought to determine if MSCs modulate leukocyte binding to
PECs stimulated with the inflammatory cytokine TNFa. In these
studies, PECs were treated with CM from MSCs and MSC-PEC
co-cultures. To stimulate inflammatory cell (U937) binding, PECs
were treated with TNFa (50 ng/ml) which increased U937 cell
binding as depicted in Figure 2A. U937 cells are a monocytoid line
that we have previously used to study leukocyte-endothelial
adhesion [24]. Binding of calcein-labeled cells to treated PECs
was quantified by relative fluorimetric. Binding studies reveal that
pre-treatment of the PECs with CM media from MSC or MSC-
PEC co-cultures inhibit U937 binding to the PECs (Figure 2B).
Flow cytometric analysis of the treated PECs reveals that CM from
MSC+PECs co-culture, but not MSC CM, inhibit TNFa
mediated increases in ICAM-1, and VCAM-1, suggesting a
contact-mediated effect. E-selectin and P-selectin were unchanged
by either treatment, suggesting that the effects of the MSC-PEC
CM is on firm adhesion of leukocytes and not tethering and rolling
processes typically regulated by the selectins (Figures 2C–2F).
Taken together, our findings suggest that the therapeutic effects of
MSCs are not solely due to enhancement of barrier integrity, but
do have direct effects on inflammatory cell binding to the
endothelium. It is of interest to note that we find effects of the
MSC CM media alone on EC permeability, but in the case of
leukocyte binding, the effects are limited to co-cultured MSC-PEC
CM. These findings suggest that the soluble factors mediating
barrier integrity may be different from those mediating inflam-
mation and leukocyte adhesion, indicating a decoupling of these
effects. To determine if our in vitro findings translate into similar
effects in a rat model of HS, we administered MSCs IV after
hemorrhage with specific focus upon the effects on endothelial
integrity and inflammation in the lungs.
IV Administration of MSCs in rats subjected to HS does
not affect mean arterial pressure (MAP) in treated
animals
Using a well characterized rat model of mild, non lethal HS
[14,25], we sought to determine if administration of IV MSCs
after hemorrhage affected MAP. Rats were randomly assigned to
four main groups with anticipated 100% survival in all groups.
Animals underwent hemorrhage with removal of 2 ml/100 g of
blood over 10 minutes. MAP fell to approximately 30 mmHg in
all groups. After a 60 minute period of shock, animals were
resuscitated. Groups were as follows (Figure 3A): Group I: Sham-
animals received no injury or treatment. Group II: HS- animals
received hemorrhage alone. Group III: HS+LR. Animals
underwent hemorrhage and were then resuscitated with LR.
Group IV: LR+MSC- rats were resuscitated with LR as in Group
III plus 26106MSCs were administered with LR at 1 hour post
hemorrhage and a second dose in 100 ml of LR at 24 hours after
hemorrhage. Dosing of MSCs was determined by our past work
and that of others in disease models of vascular stability [16,26].
The logic behind our selection of LR was to mimic the early stages
of resuscitation in trauma patients when MSCs could potentially
be used as an adjunct to resuscitation. The goal would be to
administer MSCs at an early time point to promote vascular
stability and prevent the development of ALI/ARDS or multi-
organ failure (MOF). Human MSCs were obtained commercially
(see methods) from four young, healthy donors for use in all
experiments. Our choice to use human MSCs in the rat was
prompted by our past work showing no acute evidence of rejection
and previous findings that human MSCs may act differently then
rat MSCs [27,28]. Prior to using these cells for in vitro and in vivo
experiments, fluorescence activated cell sorting (FACS) analysis
was performed to examine the expression of characteristic MSC
cell surface markers. Cells were confirmed to display characteristic
MSC cell surface markers [16], including high levels of CD44 and
CD105, and an absence of CD31 expression (data not shown).
Figures 3B and 3C show that there are no significant changes in
hemorrhage volume and weight respectively between animals in
all three groups subjected to hemorrhage. Figure 3D demonstrates
no differences in MAP in all three groups. The MAP dropped soon
after hemorrhage and was essentially restored in all groups by one
hour post hemorrhage.
IV administration of MSCs in rats subjected to HS inhibits
edema and inflammatory cell infiltrates in the lungs of
treated animals
To determine if lung edema and injury were generated by the
hemorrhagic insult, lungs were taken for wet-to-dry analysis. Wet-
to-dry-analysis (Figure 4A) revealed that mild edema is present in
Group II (HS alone). Resuscitation with LR (Group III) did not
significantly increase the amount of lung edema above HS (Group
II), but MSCs (Group IV) did indeed significantly decrease the
amount of edema compared to LR (Group III) and HS (Group II),
thus this data suggests that IV MSCs may have modulator effects
upon vascular stability in the lungs after HS. Histopathological
(H&E) analysis (Figure 4B) of tissue sections from rat lungs indicate
that animals treated with MSCs show attenuated injury. Further
histopathological analysis reveals that MSC treated animals have
decreased numbers of CD68 positive cells, a general surface
marker of leukocytes and lymphocytes, in the lungs compared to
Pulmonary Effects of MSCs in Hemorrhagic Shock
PLoS ONE | www.plosone.org5 September 2011 | Volume 6 | Issue 9 | e25171