Acute lung injury (ALI) and end-stage acute respiratory distress syndrome (ARDS) are among the most common causes of death in intensive care units. Activation and damage of pulmonary endothelium is the hallmark of ALI/ARDS. Recent studies have demonstrated the importance of circulating endothelial progenitor cells (EPCs) in maintaining normal endothelial function as well as endothelial repairing after vascular injury. Here, the authors present the first study demonstrating the therapeutic potential of EPCs in a rabbit model of ALI/ARDS.
Circulating EPCs were obtained from rabbits using Ficoll centrifugation. One week after culturing, ALI was induced in rabbits by oleic acid (75 mg/kg, intravenous), and autologous EPCs were transplanted intravenously. Vasomotor function of isolated pulmonary artery and degrees of lung injury were assessed 2 days later.
Endothelial dysfunction in the pulmonary artery was significantly attenuated in rabbits treated with EPCs, whereas the endothelium-independent relaxation responses were not different. Expression of inducible nitric oxide synthase was suppressed in the pulmonary artery of EPC-treated animals. Infiltration of leukocytes in the lung parenchyma was significantly reduced after EPC transplantation. EPCs also decreased water content, hyaline membrane formation, and hemorrhage in lungs.
The authors demonstrated that autologous transplantation of EPCs preserves pulmonary endothelial function and maintains the integrity of pulmonary alveolar-capillary barrier. Transplantation of EPCs can be a novel cell-based, endothelium-targeted therapeutic strategy for prevention and treatment of ALI/ARDS.
During the past decade, evidence continues to accumulate on the capability of circulating endothelial progenitor cells (EPCs) in neovascularization and vasculogenesis after their first discovery by Asahara et al. 6in 1997. These circulating endothelial cell progenitors share certain common embryonic cell markers with hematopoietic stem cells, including Flk-1, Tie-2, and CD34.7Circulating EPCs are mobilized from bone marrow to peripheral circulation by cytokines, growth factors, and ischemic conditions during endothelial injury and by pharmaceutics such as statins.8The endothelial regenerative potential of EPCs has been under intensive investigation in a variety of animal models of vascular injury, including hind limb ischemia,9myocardial ischemia,10carotid artery injury,11and vascular graft survival.12Several clinical studies, including randomized controlled trials, also reported some encouraging outcomes in patients with critical limb ischemia and myocardial infarction after administration of EPCs.13–16Overall, these studies in diseased animals or patients implicate that circulating EPCs exert important therapeutic function as an endogenous repair mechanism in maintaining integrity of the endothelial monolayer by replacing denuded parts of the injured artery and/or forming new vessels by direct incorporation and paracrine-mediated effects. However, there is currently no published report demonstrating the therapeutic effects of EPCs in animal models or human subjects with ALI/ARDS. Recently, Burnham et al. 17reported an important observational study measuring the circulating EPCs in patients with ALI. Peripheral blood mononuclear cells were collected from patients with ALI, and colony-forming units of EPCs were counted 7 days after incubation. Their results showed that EPC colony numbers were twofold higher in patients with ALI compared with healthy volunteers. In the 45 patients with ALI, improved survival was associated with a higher EPC colony count, and mortality was significantly increased in ALI patients with EPC colony-forming units below 35. Yamada et al. 18also demonstrated that bone marrow–derived progenitor cells accumulated in the inflammatory site of lung after lipopolysaccharide-induced ALI and differentiated into pulmonary endothelial cells. Currently, regeneration of pulmonary endothelium has been suggested as one of the potential therapeutic targets for ALI/ARDS.19Therefore, the current study tested the effects of EPCs on pulmonary endothelial regeneration, as a novel cell-based, endothelium-targeted therapeutic strategy in a rabbit model of ALI.
Materials and Methods
Isolation and Culturing of EPCs
Peripheral blood was obtained from healthy human volunteers and New Zealand White rabbits via the peripheral vein (20 ml) and ear artery (10 ml/kg), respectively. Peripheral blood mononuclear cells were isolated by density gradient centrifugation with Ficoll-Plaque Plus (Amersham Biosciences, Buckinghamshire, United Kingdom).20Mononuclear cells were then washed and plated on six-well plates coated with human fibronectin (Sigma-Aldrich Corp., St. Louis, MO), supplemented with endothelial growth medium 2 (Cambrex Corp., Charles City, IA). At day 7 of culture, the adherent cells (known as early EPCs) were harvested by trypsinization for analysis or transplantation.11All procedures were performed in accordance with the guidelines of the Institutional Animal Care and Use Committee and were approved by the Institutional Review Board (The National Cheng Kung University, Tainan, Taiwan).
Characterization of EPCs
The early EPCs isolated from rabbits were characterized as previously described.11,20Briefly, cells were incubated subsequently with acetyl-LDL (10 μg/ml; Molecular Probes, Carlsbad, CA) and isolectin (5 μg/ml; Molecular Probes). The staining of acetyl-LDL and isolectin in cultured EPCs was detected under fluorescence confocal microscopy at absorption wavelengths of 555 and 495 nm, respectively. In a separated experiment, isolated rabbit mononuclear cells were cultured on four-well glass slides coated with human fibronectin. On day 7, the attached EPCs were fixed with 4% paraformaldehyde and permeabilized with 0.1% Triton X-100 (Amresco, Inc., Solon, OH) in phosphate buffer solution. The cells were incubated with primary antibodies against Flk-1 (a specific endothelial cell marker; Santa Cruz Biotechnology, Santa Cruz, CA) and processed for indirect immunofluorescence staining. The expression of Flk-1 was analyzed by confocal microscopy.
Induction of HO-1 and MnSOD in Cultured Human EPCs
Because mature rabbit endothelial cells are not available in our laboratory, mature endothelial cells, namely human umbilical vein endothelial cells, were used to compare with the day-7 EPCs. The protein expression of heme oxygenase 1 (HO-1) and manganese superoxide dismutase (MnSOD) in human EPCs was compared with that in human umbilical vein endothelial cells. Human umbilical vein endothelial cells and human EPCs were incubated with different concentrations of hemin (0–50 μm; Sigma-Aldrich Corp.) in endothelial basic medium 2 for 2 h at 37°C. Solutions were washed three times with phosphate buffer solution. Cells were supplemented with endothelial basic medium and cultured for another 12 h. Expression of HO-1 and MnSOD was determined by Western blotting.
Rabbit Model of ALI and Transplantation of EPCs
Rabbits (2.5–3 kg) were anesthetized with intramuscular injection of ketamine (35 mg/kg), and oleic acid (75 mg/kg; Sigma-Aldrich Corp.) was infused over 20 min via the right ear vein. Pulmonary endothelial damage occurs within hours after oleic acid infusion; interstitial/intraalveolar edema and lung hemorrhage with vascular congestion develop by 24 h.21Autologous EPCs (approximately 105cells in a volume of 200 μl saline) or placebo (200 μl saline) were transplanted via the ear vein of each rabbit 30 min after infusion of oleic acid. Immediately before transplantation, a fluorescent cell tracker, CM-DiI (Molecular Probes, Carlsbad, CA), was used to label EPCs to track their homing in the pulmonary vasculature.11Sham-operated animals received either placebo or autologous EPC transplantation without injection of oleic acid. Rabbits were allowed to recover spontaneously from anesthesia after the procedures. Forty-eight hours later, rabbits were killed by intravenous administration of pentobarbital (250 mg/kg). Right and left main pulmonary arteries were isolated, and lung tissues were excised.
Pulmonary artery rings (approximately 2 mm long) were mounted in organ chambers containing 25 ml Krebs solution (118.6 mm NaCl, 4.7 mm KCl, 2.5 mm CaCl2, 1.2 mm MgSO4, 1.2 mm KH2PO4, 25.1 mm NaHCO3, 10.1 mm glucose, 0.026 mm EDTA) at 37°C (94% O2–6% CO2). Changes in force were recorded continuously using an isometric force-displacement transducer (Grass FT03; Grass Instrument, West Warwick, RI). Each ring was gradually stretched to 2.5 g. After a 45-min equilibration period, the rings were contracted by addition of KCl (40 mm) or increasing concentrations of phenylephrine (10−9to 10−5m; Sigma-Aldrich Corp.). Concentration–response curves were obtained by cumulative addition of acetylcholine (10−9to 10−5m; Sigma-Aldrich Corp.), and a nitric oxide donor (DEA-NONOate, 10−9to 10−5m; Sigma-Aldrich Corp.) during contraction to a median effective concentration (EC50) of phenylephrine. Papaverine (3 × 10−4m; Sigma-Aldrich Corp.) was used to induce complete relaxations of the vessels. All experiments were performed in intact vessels without mechanical removal of pulmonary endothelium.
Western Blot Analysis
Soluble protein extracts (30–50 μg) isolated from pulmonary artery were loaded into polyacrylamide gels (9–12%) and transferred onto nitrocellulose membranes. Anti–inducible nitric oxide synthase (iNOS; BD Biosciences, San Jose, CA), anti–HO-1 (Santa Cruz Biotechnology), and anti-MnSOD (BD Biosciences) antibodies were used. After washing, the membranes were incubated with horseradish peroxidase–linked secondary antibodies, and bands were visualized using enhanced chemiluminescence (Amersham Pharmacia, Buckinghamshire, United Kingdom). Protein levels were quantified by scanning densitometry (Scion Image, Scion Corp., Frederick, MD).
Measurement of Lung Dry-to-wet Ratio
The intermediate lobe of right lung was excised and weighed immediately. Lung tissues were dried in an oven at 80°C for 12 h and reweighed. The lung wet-to-dry ratios were obtained by dividing the mass of the initial specimen by the mass of the dried specimen.
Histologic Examination of Pulmonary Artery and Lung Tissues
Frozen isolated pulmonary artery and lung tissues were sectioned and examined under fluorescence microscopy for the detection of DM-DiI–labeled cells. Other lung tissues were immersed in 10% buffered formal saline and fixed for at least 24 h. Biopsies were processed through increasing grades of alcohol and embedded in paraffin wax. Hematoxylin and eosin–stained lung sections were examined under a light microscope and photographed. An investigator blinded to the treatment was assigned to measure areas of hyaline membrane formation and hemorrhage in the lung sections using the TwinCAD software (TCAM Development, Taipei, Taiwan). Infiltration of polymorphonuclear cells (PMNs) in the lung tissue was also measured using the PMN ratio, as previously described.22Briefly, numbers of PMNs and non-PMNs were counted from 40 randomly selected fields in the five domains of each tissue slide under light microscopy (400×). Each observed area was defined with a graticule fitted in the eyepiece (an effective calibrated area of 2,830 μm2). A ratio of PMNs to non-PMNs was calculated for each tissue slide. According to our previous report, normal human lung tissues yield PMN ratios of 1.3–2.3%, whereas sections obtained from pneumonia or lung abscess yield higher PMN ratios of 8.3–21.8%.22
Results are presented as mean ± SEM. Data were compared by an unpaired t test or analysis of variance, as appropriate. Statistical significance was accepted at a level of P < 0.05.
Characterization of Phenotypes of EPCs
Cultured day-7 rabbit EPCs exhibited phenotyping of endothelial cells, including incorporation of acetyl-LDL and isolectin and expression of Flk-1 (figs. 1A–E). Formation of monolayer colonies with “cobblestone” appearance was also observed 2 weeks after culturing in endothelial basic medium 2 (fig. 1F).11
Expression of HO-1 and MnSOD in Isolated EPCs
To determine the endogenous antioxidant capacity in EPCs, we studied the expression of HO-1 and MnSOD in these premature endothelial cells isolated from human peripheral blood after incubation with hemin for 2 h. Compared with human umbilical vein endothelial cells, human EPCs expressed higher levels of MnSOD in unstimulated conditions, and expression of HO-1 was also more significantly up-regulated in a dose-dependent manner after exposure to hemin (fig. 2).
Detection of Transplanted EPCs in the Pulmonary Circulation
Fluorescence-labeled EPCs were detected on the lumens of intralobular pulmonary arterioles, indicating the homing of transplanted EPCs on the injured pulmonary endothelium (fig. 3).
Assessment of Vascular Function of Pulmonary Artery
Vascular function of pulmonary artery was assessed by isometric tension recordings. Contraction responses of isolated pulmonary artery to KCl and phenylephrine were significantly enhanced in rabbits treated with EPCs, but EC50s of phenylephrine were not different (table 1). Transplantation of EPCs attenuated the impaired endothelium-dependent relaxations of pulmonary artery to acetylcholine in rabbits with ALI (fig. 4A). High concentration of acetylcholine (10−5m) induced significantly greater relaxations in the EPC group compared with controls (81.9 ± 4.8% vs. 45.1 ± 10.2%, respectively; P < 0.001, n = 6–8). In fact, high concentrations of acetylcholine (10−7to 10−5m) induced contractile responses in the control group (fig. 4A). However, the median concentrations of acetylcholine for inducing 50% relaxations (EC50R) of precontracted pulmonary artery in the control and EPC-treated groups were similar (4.76 ± 1.55 × 10−8m vs. 5.23 ± 1.55 × 10−8m, respectively; P = 0.88, n = 6–8). On the other hand, endothelium-independent relaxations were similar between the two treatment groups (fig. 4B), suggesting that function of vascular smooth muscle was not affected by transplantation of EPCs. Because of superimposition of the concentration–response curves between sham-treated animals receiving placebo or EPC transplantation, results of the latter were not shown.
Expression of iNOS in Pulmonary Artery
Protein expression of iNOS was up-regulated in the pulmonary artery of rabbits with oleic acid–induced ALI (fig. 5). Compared with controls, autologous transplantation of EPCs significantly reduced expression of iNOS in pulmonary artery (fig. 5).
Determination of Degrees of Lung Injury
Disruption of alveolar–capillary barrier was assessed by measuring water and fluid content in the lungs. The lung wet-to-dry ratio was significantly reduced in rabbits received autologous transplantation of EPCs (fig. 6). Degrees of lung parenchymal damage were determined under light microscopy. Formation of hyaline membrane and hemorrhage were dramatically reduced in the lung tissue of rabbits treated with EPCs (fig. 7). Because infiltration of PMNs is the characteristic histologic changes during the acute exudative phase (first 24–48 h) of ALI/ARDS, we therefore examined the counts of PMNs in the lungs. Totals of 1,430 ± 68 and 1,410 ± 62 cells were counted from each lung section under light microscopy in the control and EPC-treated groups, respectively (P = 0.7, n = 3 animals in each group). A significantly higher PMN ratio was detected in lung tissue of the placebo-treated animals (fig. 8).
We present the first study demonstrating that transplantation of circulating EPCs attenuates lung injury and pulmonary artery endothelial dysfunction in rabbits with oleic acid–induced ALI. Transplantation of EPCs potentiates the relaxation response to acetylcholine in pulmonary arteries and reduces lung water content, suggesting that the therapeutic benefits of EPCs in ALI are most likely derived from their effect on reendothelization of the damaged pulmonary artery wall and alveolar–capillary membrane.
Acute lung injury and ARDS are characterized by diffuse pulmonary infiltration, increased pulmonary capillary permeability, and severe hypoxemia. The initial physiopathologic changes are damage of endothelial surfaces and disruption of the alveolar–capillary barrier function, and flooding alveolar spaces with fluid.23Hence, several pulmonary endothelium–targeted therapeutics, such as surfactants, pulmonary vasodilators, antioxidants, antiinflammatory agents, and statins, have been extensively tested in patients with ALI/ARDS, aiming to restore pulmonary endothelial function. However, none of these pharmacologic endothelium-targeted therapeutics reduce mortality of ALI/ARDS. In the current study, we used a novel cell-based, potent endothelial regenerating approach, namely transplantation of EPCs, to test our hypothesis that the high reendothelization capability of EPCs restores damaged pulmonary endothelium and thus provides therapeutic benefits in ALI. One of the foremost concerns for delivering cell-based therapeutics is the survival of target cells in unfavorable pathologic conditions, and homing of these cells in the injured tissues. First, we studied the endogenous antioxidant capacity of EPCs. In comparison with mature endothelial cells, early EPCs isolated from healthy human subjects expressed higher protein levels of MnSOD and HO-1. MnSOD is an abundant mitochondrial enzyme that catalyzes superoxide generated by respiratory chain activity into water and hydrogen peroxide,24and HO-1 is potent inducible antioxidant gene that modulates cell growth, inflammation, and apoptosis.25High expression of MnSOD in human EPCs was shown by two independent research groups,20,26but induction of HO-1 in these premature endothelial cells has not been previously demonstrated. Elevated intrinsic antioxidant enzymatic activity in EPCs may enable them to maintain endothelial regenerative function under conditions of severe oxidative stress, such as ALI/ARDS, and may also attribute to the paracrine cytoprotective effects. On the other hand, we located the transplanted EPCs on the lumen wall of pulmonary arterial system using fluorescence-conjugated cell tracers, indicating that EPCs homed in the pulmonary endothelium and may subsequently reendothelize the injured endothelium after intravenous administration. Nevertheless, we were not able to detect a large amount of DM-DiI–labeled cells from the serial lung sections of rabbits that received EPC transplantation and therefore were not able to provide quantitative data on this measurement. In our opinion, EPCs dispersed in the extremely large surface areas of pulmonary circulation after intravenous transplantation and thus reduced the ability of discovery from thin histologic sections. In addition, low homing efficacy of progenitors in the target organs has been reported after systemic or regional delivery, as a result of washout effect.27,28After 24 h, only 2–5% of the transplanted progenitor cells were detected in the bone marrow of recipient animals.29Muller-Ehmsen et al. 30showed that only 24% of the transplanted cells remained in the recipient heart 1 day after direct ventricular injection. Indeed, our results also support the general concept that apart from direct cell fusion, EPCs may mediate angiogenesis and reendothelization via their paracrine effects.
Vascular function of the isolated pulmonary artery was assessed by contraction and relaxation responses in organ bath experiments. Vasocontraction was induced by addition of high concentrations of extracellular potassium and phenylephrine. Contraction responses to 40 mm KCl and 10−5m phenylephrine were reduced in the pulmonary artery isolated from control animals. Because morphologic changes in media layer were unlikely, we therefore sought functional alteration in vascular smooth muscle. Consistent with previous reports,31,32Western analysis showed that expression of iNOS was significantly up-regulated in the pulmonary artery of animals with ALI, indicating the presence of oxidative stress in the injured pulmonary artery. However, transplantation of EPCs significantly suppressed the protein levels of iNOS in pulmonary artery. Induction of iNOS is commonly suggested to be responsible for diminished vasoconstriction of pulmonary artery due to overproduction of nitric oxide.33,34In addition, our result also speculates that transplantation of EPCs attenuates expression of iNOS. However, biologic mechanisms underlying regulation of iNOS by transplantation of EPCs require further investigation. We used different means to assess the function of pulmonary endothelium and the integrity of the alveolar–capillary barrier after placebo or EPC transplantation in rabbits with oleic acid–induced ALI. Endothelial dysfunction of pulmonary artery was confirmed by reduced relaxation response to acetylcholine in rabbits with ALI. Functional analysis of pulmonary artery isolated from rabbits that received EPC transplantation showed that contractile response to KCl and phenylephrine was restored, and endothelium-dependent relaxation response to acetylcholine was significantly potentiated. Instead of causing vasorelaxation, the pulmonary artery tended toward contraction in the control group during cumulative addition of higher concentrations of acetylcholine, suggesting the presence of endothelial injury. Acetylcholine induces vasodilation by releasing endothelium-derived relaxing factor in the presence of intact endothelium, whereas it causes vasoconstriction when the endothelium is removed or damaged.35,36Forty-eight hours after transplantation of EPCs, endothelium-dependent relaxation was partly restored (more than 80% of complete relaxation in comparison with 94.8% in sham-treated animals) with the absence of acetylcholine-induced vasocontraction. Together with the normal endothelial-independent relaxation response to nitric oxide donor, we conclude that the vascular regenerative effect of EPC in ALI is endothelium targeted.
The alveolar–capillary membrane is a blood-gas barrier formed by alveolar epithelium, pulmonary capillary endothelium, and lamina densa. This anatomical interface allows gas exchange; regulates solute and fluid flux between the alveolar surface, interstitium, and blood; and promotes active fluid clearance from the alveolar lumen to the interstitial space.37During the progression of ALI/ARDS, the alveolar–capillary membrane is disrupted by activation of pulmonary endothelium, expression of adhesion molecules, and massive infiltration of PMNs, hence followed by increased protein permeability across the barrier.5The characteristic pathologic changes after disruption of alveolar–capillary membrane are noncardiogenic pulmonary edema, hyaline membrane formation, and lung hemorrhage.38,39We examined the lung water content in the rabbits with ALI by measuring the lung wet-to-dry ratio. The ratio of wet-to-dry weight was significantly lower in the lungs of animals that received EPC transplantation. Histologic examination also revealed significantly reduced formation of hyaline membrane and area of hemorrhage in the lung sections of EPC-treated animals. Collectively, these data suggest that function of alveolar–capillary membrane was, at least partly, restored by administration of EPCs. In accord with isometric tension analysis of pulmonary artery, these findings reinforce the concept of endothelium-targeting therapeutic potential of EPC transplantation in ALI/ARDS.
Another biologic marker of ALI/ARDS is activation and transmigration of circulating neutrophils, or so-called PMNs, in the lung tissues.40In fact, elimination of lung PMNs by cyclophosphamide or antineutrophil antibodies diminished pulmonary edema in hemorrhage- or endotoxin-induced ALI in mice.41In the current study, we demonstrated that transplantation of EPCs reduced the numbers of PMN infiltrating the lungs of oleic acid–induced ALI. In addition to endothelial regeneration effect on the pulmonary artery and alveolar–capillary membrane, antiinflammatory effect of EPCs on the infiltration and function of PMNs in the lungs may provide supplementary therapeutic benefit in ALI. However, the actual mechanisms underlying suppression of PMNs after administration of EPCs require further investigation.
One of the major limitations to generalize these results to clinical practice is that development of ALI/ARDS is almost unpredictable; therefore, autologous transplantation of EPCs may not be clinically feasible. Furthermore, the pathophysiology of ALI/ARDS is a more complicated process than the simple animal model of oleic acid–induced ALI. Further investigation into human subjects is imperative.
In conclusion, we provide the first evidence demonstrating that transplantation of circulating EPCs with high intrinsic antioxidant activity improves endothelial function of the injured pulmonary artery and attenuates damage of alveolar–capillary barrier in ALI/ARDS. Administration of EPCs is a novel cell-based, pulmonary endothelium–targeted therapeutic strategy for ALI/ARDS.