Abstract
Human mesenchymal stromal cells diminish injury and enhance recovery and repair after ventilator-induced lung injury in animals
Current methods of isolating mesenchymal stromal cells result in a heterogeneous mix of cell types, which may be suboptimal
Pure subpopulations of bone marrow–derived human mesenchymal stromal cells were isolated on the basis of expression of the cell surface marker syndecan 2
Intravenous injection of these cells attenuated Escherichia coli–induced injury and enhanced resolution of ventilator-induced lung injury in rats, reducing lung inflammation and histologic injury and improving lung compliance and arterial oxygenation
Cells expressing syndecan 2 were more effective than those not expressing syndecan 2
Human mesenchymal stromal cells demonstrate promise for acute respiratory distress syndrome, but current studies use highly heterogenous cell populations. We hypothesized that a syndecan 2 (CD362)–expressing human mesenchymal stromal cell subpopulation would attenuate Escherichia coli–induced lung injury and enhance resolution after ventilator-induced lung injury.
In vitro studies determined whether CD362+ human mesenchymal stromal cells could modulate pulmonary epithelial inflammation, wound healing, and macrophage phagocytosis. Two in vivo rodent studies determined whether CD362+ human mesenchymal stromal cells attenuated Escherichia coli–induced lung injury (n = 10/group) and enhanced resolution of ventilation-induced injury (n = 10/group).
CD362+ human mesenchymal stromal cells attenuated cytokine-induced epithelial nuclear factor kappa B activation, increased epithelial wound closure, and increased macrophage phagocytosis in vitro. CD362+ human mesenchymal stromal cells attenuated Escherichia coli–induced injury in rodents, improving arterial oxygenation (mean ± SD, 83 ± 9 vs. 60 ± 8 mmHg, P < 0.05), improving lung compliance (mean ± SD: 0.66 ± 0.08 vs. 0.53 ± 0.09 ml · cm H2O−1, P < 0.05), reducing bacterial load (median [interquartile range], 1,895 [100–3,300] vs. 8,195 [4,260–8,690] colony-forming units, P < 0.05), and decreasing structural injury compared with vehicle. CD362+ human mesenchymal stromal cells were more effective than CD362− human mesenchymal stromal cells and comparable to heterogenous human mesenchymal stromal cells. CD362+ human mesenchymal stromal cells enhanced resolution after ventilator-induced lung injury in rodents, restoring arterial oxygenation (mean ± SD: 113 ± 11 vs. 89 ± 11 mmHg, P < 0.05) and lung static compliance (mean ± SD: 0.74 ± 0.07 vs. 0.45 ± 0.07 ml · cm H2O−1, P < 0.05), resolving lung inflammation, and restoring histologic structure compared with vehicle. CD362+ human mesenchymal stromal cells efficacy was at least comparable to heterogenous human mesenchymal stromal cells.
A CD362+ human mesenchymal stromal cell population decreased Escherichia coli–induced pneumonia severity and enhanced recovery after ventilator-induced lung injury.
HUMAN mesenchymal stromal cells are a promising therapeutic strategy for acute respiratory distress syndrome.1 Species-specific mesenchymal stromal cells demonstrate beneficial effects in diverse preclinical lung injury models, including pulmonary2–4 and abdominal sepsis,5–7 ventilator-induced lung injury,8 bleomycin-induced acute lung injury,9 and fibrosis.10,11 Importantly, mesenchymal stromal cells can restore lung function after established ventilator-induced lung injury12,13 ; this result underlines their reparative potential. More importantly from a translational perspective, human-derived mesenchymal stromal cells demonstrate efficacy in human lungs ex vivo14 and can reduce mortality in murine5,15 and ovine16 sepsis models. Our group and others have demonstrated that human mesenchymal stromal cells can reduce the severity of Escherichia coli pneumonia in rodent17 and murine18 models and in the ex vivo human lung,19 and mesenchymal stromal cells can enhance repair and recovery of function after ventilator-induced lung injury.20,21 Most recently, a phase 1, open-label, dose-escalation, multicenter clinical trial has demonstrated the safety of allogeneic bone marrow–derived human mesenchymal stromal cells in patients with moderate to severe acute respiratory distress syndrome.22
A potential concern exists in regard to the current approach to isolating human mesenchymal stromal cells. This relies largely on the isolation of the “plastic adherent” component of the bone marrow mononuclear cell population, and their subsequent in vitro characterization by standard surface markers and differentiation assays.23 This approach results in a heterogeneous mix of progenitors, lineage-restricted precursors, and fibroblasts,24 which raises two issues. First, this heterogeneous human mesenchymal stromal cell population may not be sufficiently pure to meet emerging regulatory requirements for Advanced Therapeutic Medicinal Products for clinical use. Second, using a specific subpopulation may result in a better-characterized therapeutic, with potentially less batch-to-batch variability, both of which would simplify clinical translation.
Syndecan 2 (also known as CD362) is expressed on the surface of a subpopulation of human mesenchymal stromal cells, enabling the selective isolation of this subpopulation by fluorescence-activated cell sorting, a significant advance over current isolation techniques. Our primary objective was to determine the effect of CD362+ human mesenchymal stromal cells in relevant in vitro and in vivo lung injury and repair models, and our secondary objective was to compare the degree of effect to that seen with heterogeneous human mesenchymal stromal cell populations. Pneumonia is the most common risk factor for acute respiratory distress syndrome,25 and ventilator-induced lung injury constitutes a major iatrogenic contributor to mortality in patients with acute respiratory distress syndrome.26 Given this, our overall hypothesis was that a CD362-expressing human mesenchymal stromal cell subpopulation would attenuate the development of bacterial-induced lung injury and enhance resolution and repair of lungs subjected to ventilator-induced lung injury. Our first specific hypothesis was that CD362+ human mesenchymal stromal cells would attenuate pulmonary epithelial inflammation, promote epithelial wound repair, and enhance macrophage phagocytosis in vitro. Our second specific hypothesis was that CD362+ human mesenchymal stromal cells would attenuate bacteria-induced lung injury, of importance given that infection is the most common27,28 and most severe cause of acute respiratory distress syndrome.29 Our third specific hypothesis was that CD362+ human mesenchymal stromal cells would enhance resolution of lung injury after ventilator-induced lung injury, of importance given the central importance of resolution and repair in acute respiratory distress syndrome.30 Our fourth specific hypothesis was that the degree of effect of CD362+ human mesenchymal stromal cells would be comparable to that seen with a standard heterogenous human mesenchymal stromal cell population.
Materials and Methods
All work was approved by the Animal Care Research Ethics Committee of the National University of Ireland (Galway, Ireland) and conducted under license from the Health Products Regulatory Authority (Dublin, Ireland). Specific-pathogen-free adult male Sprague Dawley rats (Charles River Laboratories, United Kingdom) weighing between 300 and 450 g were used in all experiments. A full description of the methods is available in Supplemental Digital Content 1, https://links.lww.com/ALN/B751.
Isolation and Characterization of CD362+/− Human Mesenchymal Stromal Cell Subpopulations
Informed consent was obtained for all bone marrow samples according to the Ethics Ref. C.A.02/08. Briefly, mononuclear cells were isolated by Ficoll density gradient centrifugation (GE Health Care Bio-Sciences, United Kingdom) and ACK Lysis Buffer (Life Technologies, USA) employed for erythrocyte lysis. Mononuclear cells were resuspended in fluorescence-activated cell sorting buffer and then analyzed for expression of CD235, CD45, CD271, and CD362 and viability with Sytox Blue dye (Supplemental Digital Content 1, https://links.lww.com/ALN/B751). With appropriate controls including “fluorescence minus ones,” sort gates were assigned and populations of interest were sorted by a BD Fluorescence-Activated Cell Sorting Aria (BD Biosciences, United Kingdom). Postsort purities were routinely 98% or more with viability of 90% or more. All human mesenchymal stromal cell populations were cultured under conditions of 37°C, 95% humidity, 5% CO2, and hypoxic conditions of 2% O2, until 70 to 80% confluent; the populations were then trypsinized and culture expanded to passage 3 to 4, whereupon they were used for experiments. Cell surface characterization of human mesenchymal stromal cell populations was based on positive expression of CD73, CD105, CD90, and MHCI and lack of expression of CD34, CD45, CD80, and CD86. A full list of the monoclonal antibodies and methods used for sorting and characterization can be found in Supplemental Digital Content 1, table e1, https://links.lww.com/ALN/B751. All flow cytometry data were analyzed with FlowJo software (Tree Star, Inc., USA).
Cell Lines and Preparation of Human Mesenchymal Stromal Cell–conditioned Medium
Primary human lung fibroblasts and U937 monocytic/macrophage cell line were obtained from American Type Culture Collection (ATCC, USA). Conditioned medium was generated from both heterogeneous and CD362+ human mesenchymal stromal cells (4 × 106 cells). After growth to 60% confluence, the cells were washed and re-fed with serum-free medium for 48 h; this medium was used for in vitro experiments.
In Vitro Determination of the Effects of CD362+ Human Mesenchymal Stromal Cells
Nuclear Factor κB Activation Assay.
A cell line derivative of type II alveolar A549 cells incorporating a stably transfected κb-luciferase reporter construct (Affymetrix, USA) was grown to confluence. Cell monolayers were randomized to interleukin 1β (10 ng/ml) or vehicle activation, treated with conditioned medium from heterogeneous, human mesenchymal stromal cell, CD362+ human mesenchymal stromal cells, or vehicle, harvested at 24 h, and assayed for luciferase content.31
Wound Healing Assay.
Single linear wounds were made in confluent A549 cell monolayers with a 200-μl pipette tip, as previously described.31 Monolayers were randomized to incubation in conditioned medium from heterogeneous human mesenchymal stromal cells, CD362+ human mesenchymal stromal cell cells, or vehicle, and the extent of epithelial restitution determined 24 h later.
Macrophage Phagocytosis Assay.
U937 cell-derived macrophages were seeded in six well plates with heterogeneous human mesenchymal stromal cell– or CD362+ human mesenchymal stromal cell–conditioned medium and exposed to fluorescein isothiocyanate–labeled Escherichia coli bacteria particles (Vybrant Phagocytosis Kit; Life Technologies, USA) for 4 h; phagocytic capacity was determined by quantification of cytosolic fluorescence.
In Vivo Experimental Protocols
Escherichia coli–induced Lung Injury.
Adult male Sprague Dawley rats were anesthetized by isoflurane inhalation and intraperitoneal ketamine, 40 mg/kg (Pfizer, United Kingdom). After laryngoscopy, the trachea was intubated with a 14-gauge catheter (BD Insyte; BD Biosciences), and 2 × 109 colony-forming units of Escherichia coli E5162 (serotype: O9 K30 H10) in a 300-µl phosphate-buffered saline suspension was instilled under direct vision; the animals were allowed to recover from anesthesia.17,32–35
Ventilator-induced Lung Injury.
Adult male Sprague Dawley rats were anesthetized with isoflurane, and intravenous access was obtained.12,13,36 Animals were intubated as above. Anesthesia was maintained with repeated boli of alfaxalone (Saffan; Schering Plough, United Kingdom) and paralysis with cisatracurium besylate, 0.5 mg · kg−1 (GlaxoSmithKline, Ireland). After baseline ventilation, static compliance was measured and ventilator-induced lung injury was induced with the following ventilator settings: pressure-controlled ventilation mode; Fio2, 0.3; inspiratory pressure, 35 cm H2O; respiratory rate, 18 min−1; inspiratory time to expiratory time ratio, 1:1; positive end-expiratory pressure, 0 cm H2O. After development of severe ventilator-induced lung injury, as evidenced by a 50% decrease in respiratory static compliance, injurious ventilation was discontinued, and the animals allowed to recover from anesthesia.36
Experimental Design
Escherichia coli Lung Injury.
Thirty minutes after intratracheal instillation of Escherichia coli bacteria, animals were randomized to receive (1) vehicle (phosphate-buffered saline, 300 μl); (2) CD362+ human mesenchymal stromal cells (1 × 107 cells/kg); (3) CD362− human mesenchymal stromal cells (1 × 107 cells/kg); or (4) heterogeneous human mesenchymal stromal cells (1 × 107 cells/kg); the degree of injury was assessed at 48 h.
Ventilator-induced Lung Injury.
Thirty minutes after discontinuation of injurious ventilation, animals were randomized to receive (1) vehicle; (2) heterogeneous human mesenchymal stromal cells (1 × 107 cells/kg); (3) CD362+ human mesenchymal stromal cells (1 × 107 cells/kg); or (4) fibroblasts (1 × 107 cells/kg); the effect on restoration of lung function and structure was assessed at 24 h. The cells doses chosen for both studies were those that demonstrated maximal effect for heterogenous human mesenchymal stromal cells in these models.17,20
Assessment of Lung Injury and Recovery
In Vivo Assessment.
Animals were anesthetized with intraperitoneal ketamine, 80 mg · kg−1 (Ketalar; Pfizer, Ireland), and xylazine, 8 mg · kg−1 (Xylapan; Vétoquinol, Ireland); intravenous and intraarterial access was secured, and a tracheostomy tube was inserted. Anesthesia was maintained with alfaxalone and paralysis with cisatracurium besylate, and mechanical ventilation commenced. Arterial blood pressure, airway pressure, lung static compliance, and arterial blood gas analyses were performed.37,38
Ex Vivo Analyses.
After exsanguination during anesthesia, bronchoalveolar lavage was performed; bronchoalveolar lavage fluid differential leukocyte counts and, in the pneumonia model, lung bacterial colony counts were completed. Bronchoalveolar lavage concentrations of tumor necrosis factor α, interleukin 1β, interleukin 6, cytokine-induced neutrophil chemoattractant 1, interleukin 10, and keratinocyte growth factor were determined by enzyme-linked immunosorbent assay (R&D Systems, United Kingdom), and total bronchoalveolar lavage protein was measured (Micro BCA; Pierce, USA). The left lung was isolated and fixed, and histologic lung damage determined with quantitative stereologic techniques.39 All ex vivo analyzes were performed by blinded investigators.
Statistical Analysis
A sample size of 10 animals per group was determined for a 4-group design based on previously published experimental data from these models by our group.12,13,17,33,36 Data are reported as means ± SD or as medians (interquartile range). Data were analyzed with Sigma Stat (SYSTAT software, USA). The distribution of all data was tested for normality with Kolmogorov–Smirnov tests. Data were analyzed by two-way or one-way ANOVA or ANOVA on Ranks (Kruskall–Wallis) as appropriate, with post hoc testing by Dunnett’s method, with the vehicle group as the single comparison group, or with Student–Newman–Keuls between-group comparisons, as appropriate. Underlying model assumptions were deemed appropriate on the basis of suitable residual plots. A two-tailed P value of less than 0.05 was considered significant.
Results
Identification and Isolation of CD362+/− Human Mesenchymal Stromal Cell Subpopulations
CD362+ human mesenchymal stromal cells were identified within the CD45−CD271high mononuclear cell population (fig. 1, A and B). The percentage populations found in a typical mononuclear cell preparation of human bone marrow are CD362+CD271high (CD362+ human mesenchymal stromal cells, 0.12% ± 0.06), CD362−CD271+ (CD362− human mesenchymal stromal cells, 1.70% ± 0.43), and CD362+CD271− (0.95% ± 0.26) (Supplemental Digital Content 2, https://links.lww.com/ALN/B750). With a fluorescence-activated cell sorting method of cell isolation, a pure CD362+ human mesenchymal stromal cell population demonstrates a significantly enhanced capacity to form colonies (13117 ± 2687 colonies / 1 × 105 cells) when compared to CD362− human mesenchymal stromal cells (123 ± 29.7 colonies / 1 × 105 cells) and heterogeneous human mesenchymal stromal cells, isolated by the standard plastic adherent method (8.5 ± 1.5 colonies / 1 × 105 cells) (fig. 1C). Once expanded in culture, all cell populations isolated, including the novel CD362+ human mesenchymal stromal cells (fig. 1D), met with current International Society for Cellular Therapy criteria for stromal cell protein expression.23
Effects of CD362+ Human Mesenchymal Stromal Cells In Vitro
Conditioned medium from CD362+ human mesenchymal stromal cell attenuated interleukin 1β–induced nuclear factor κβ activation in type II alveolar A549 cells to a comparable degree to that seen with heterogeneous human mesenchymal stromal cells (fig. 2A). CD362+ human mesenchymal stromal cell increased the rate of wound closure in A549 alveolar epithelial cultures 24 h after wound injury to a comparable degree to that seen with heterogeneous human mesenchymal stromal cells (fig. 2B). CD362+ human mesenchymal stromal cells were significantly more effective than heterogeneous human mesenchymal stromal cells in increasing the rate of phagocytosis in U937 monocytes-derived macrophages (fig. 2C).
CD362+ Human Mesenchymal Stromal Cells Decrease Escherichia coli–induced Lung Injury
Forty animals were entered into the experimental protocol, with 10 allocated to each of the groups. Two animals in the vehicle group died before injury assessment of Escherichia coli–induced lung injury; no other data were lost or excluded. There were no significant differences between the groups at baseline in terms of preinjury variables or the amount of instilled Escherichia coli bacteria.
CD362+, CD362−, and heterogeneous human mesenchymal stromal cells therapy each decreased the severity of Escherichia coli–induced lung injury compared to vehicle controls. CD362+, CD362−, and heterogeneous human mesenchymal stromal cells each significantly attenuated the decrease in arterial oxygenation compared to vehicle (mean ± SD: 77 ± 11, 72 ± 11, 83 ± 9 vs. 60 ± 8 mmHg, respectively, P < 0.001) (fig. 3A). CD362+, CD362−, and heterogeneous human mesenchymal stromal cells each significantly attenuated the decrease in lung static compliance compared to vehicle (mean ± SD: 0.70 ± 0.13, 0.66 ± 0.10, 0.66 ± 0.08 vs. 0.53 ± 0.09 ml · cm H2O−1, respectively, P < 0.001) (fig. 3B). Of importance, both CD362+ and heterogeneous human mesenchymal stromal cells, but not CD362− human mesenchymal stromal cells, attenuated the increase in lung microvascular permeability (fig. 3C) and reduced lung Escherichia coli counts (fig. 3D).
CD362+ human mesenchymal stromal cell therapy decreased overall alveolar inflammatory cell infiltration (fig. 3E), substantially decreasing the percentage of neutrophils in the alveolar fluid (P < 0.001) (fig. 3F). The absolute number of neutrophils was significantly reduced by CD362+ and heterogeneous human mesenchymal stromal cells, but not CD362− cells. CD362+ and heterogeneous human mesenchymal stromal cell, but not CD362− cells, attenuated the increase in bronchoalveolar lavage tumor necrosis factor α concentrations (fig. 4A), and all three cell types attenuated the increase in bronchoalveolar lavage interleukin 1β concentrations (fig. 4B). There was no significant effect of human mesenchymal stromal cell therapy on bronchoalveolar lavage interleukin 6 concentrations (fig. 4C). CD362+, CD362−, and heterogeneous human mesenchymal stromal cells all decreased bronchoalveolar lavage cytokine-induced neutrophil chemoattractant 1 (fig. 4D), increased bronchoalveolar lavage interleukin 10 concentrations (fig. 4E), and increased bronchoalveolar lavage keratinocyte growth factor (fig. 4F) compared to vehicle.
CD362+ human mesenchymal stromal cells were more effective in increasing recovery of airspace volume, decreasing alveolar thickening, and increasing recovery of airspace volume, as evidenced by increased alveolar airspace volume fraction and reduced alveolar tissue volume fraction, respectively, than heterogeneous or CD362− human mesenchymal stromal cells (fig. 5A). Representative histologic sections of lung demonstrate the greater degree of resolution of injury and alveolar infiltrates in each animal group (fig. 5, B–E).
CD362+ Human Mesenchymal Stromal Cells Enhance Injury Resolution after Ventilator-induced Lung Injury
Forty animals were entered into the experimental protocol, with 10 allocated to each of the groups. All animals survived the ventilator-induced lung injury protocol and the recovery period, and no data were lost or excluded. There were no baseline group differences and no significant difference in the duration of high-stretch ventilation required to induce ventilator-induced lung injury across the groups.
CD362+ human mesenchymal stromal cell therapy enhanced injury resolution after ventilator-induced lung injury compared to vehicle or fibroblast controls. Compared to vehicle or fibroblast therapy, CD362+ human mesenchymal stromal cells and heterogeneous mesenchymal stromal cells restored arterial oxygenation (mean ± SD: 89 ± 11, 91 ± 14, 113 ± 11, 119 ± 13 mmHg, respectively, P < 0.001) and lung static compliance (mean ± SD: 0.45 ± 0.07, 0.48 ± 0.17, 0.72 ± 0.15, 0.74 ± 0.09 ml · cm H2O−1, respectively, P < 0.001) (fig. 6, A and B) and decreased lung microvascular permeability (fig. 6, C and D). The effect of the CD362+ human mesenchymal stromal cells was comparable to that seen with heterogeneous human mesenchymal stromal cells. CD362+ human mesenchymal stromal cell therapy decreased overall alveolar inflammatory cell infiltration, substantially decreasing the percentage of neutrophils in the alveolar fluid (P < 0.001) (fig. 7, A and B). CD362+ human mesenchymal stromal cells decreased alveolar concentrations of interleukin 1β, interleukin 6, and cytokine-induced neutrophil chemoattractant 1, but they had no effect on interleukin 10 concentrations (fig. 7, C–F).
CD362+ human mesenchymal stromal cells decreased alveolar thickening and increased recovery of airspace volume, as evidenced by reduced alveolar tissue volume fraction and increased alveolar airspace volume fraction, respectively (fig. 8A). Representative histologic sections of lung demonstrate the greater degree of resolution of injury and alveolar infiltrates in the CD362+ human mesenchymal stromal cell–treated animals (fig. 8, B–E).
Discussion
In these studies we demonstrate that a novel homogenous subpopulation of human mesenchymal stromal cells, namely CD362+ human mesenchymal stromal cells, transplanted xenogenetically into the immune-competent rat, reduced lung bacterial counts and decreased both physiologic and histologic evidence of Escherichia coli–induced lung injury. Furthermore, CD362+ human mesenchymal stromal cells demonstrated some advantages over both heterogeneous and CD362− human mesenchymal stromal cells in reducing Escherichia coli–induced injury. We further demonstrate that CD362+ human mesenchymal stromal cells enhanced resolution and repair after ventilator-induced lung injury. We provide data characterizing this novel subpopulation and demonstrate potentially important mechanisms of action for CD362+ human mesenchymal stromal cells, including decreasing the pulmonary epithelial response to cytokine activation, enhancing epithelial wound healing, and increasing macrophage phagocytosis.
Advantages of CD362+ Human Mesenchymal Stromal Cell Subpopulation
The use of defined human mesenchymal stromal cell subpopulations, characterized by identification of specific cell surface antigens, has potential advantages with regard to clinical translation. Conventional human mesenchymal stromal cell isolation approaches rely heavily on the isolation of the “plastic adherent” component of the bone marrow mononuclear cell population and their subsequent characterization by standard surface markers and differentiation assays.23 This results in a heterogeneous cell population,24,40 which may not meet emerging regulatory requirements for Advanced Therapeutic Medicinal Products for clinical use. A marker that defines a pure human mesenchymal stromal cell population would remove the reliance on plastic adherence and be more likely to satisfy future licensing regulations. Second, using a specific subpopulation may result in a better-characterized therapeutic, with potentially less batch-to-batch variability.
While previous attempts have been made to isolate subpopulations of human mesenchymal stromal cells on the basis of on cell surface markers, previously described human mesenchymal stromal cell surface markers, including STRO-1, CD29, CD44, CD73, CD90, CD105, CD166, and MHC-1, are not unique to human mesenchymal stromal cells.41 The CD362 protein is expressed on the surface of human mesenchymal stromal cells, as evidenced by the fact that the anti-CD362 antibody binds to the surface of the CD45−CD271high mononuclear cell population, and can be used to denote a specific mesenchymal stromal cell subpopulation.
CD362+ Human Mesenchymal Stromal Cells Attenuated Escherichia coli–induced Acute Respiratory Distress Syndrome
CD362+ human mesenchymal stromal cells attenuated the severity of Escherichia coli–induced acute lung injury, decreased physiologic indices of lung dysfunction, and reduced structural lung injury. Importantly, CD362+ human mesenchymal stromal cells appeared to possess some advantages over both CD362− human mesenchymal stromal cells and heterogeneous human mesenchymal stromal cells and exhibited the greatest range of effect, decreasing bacterial load and reducing histologic evidence of injury (Supplemental Digital Content 1, table e2, https://links.lww.com/ALN/B751). The finding that CD362+ human mesenchymal stromal cells significantly reduced lung Escherichia coli colony counts in our pneumonia model but that CD362− human mesenchymal stromal cells did not have this effect is an interesting and potentially important finding. This may be explained in part by the greater capacity of CD362+ human mesenchymal stromal cells to enhance macrophage phagocytosis, as demonstrated in our in vitro studies. Mesenchymal stromal cell enhancement of macrophage function has been demonstrated by our group17,32 and others5,18 to be a potentially important mechanism by which human mesenchymal stromal cells enhance bacterial clearance in preclinical injury models.
CD362+ Human Mesenchymal Stromal Cells Enhanced Repair after Ventilator-induced Lung Injury
CD362+ human mesenchymal stromal cells enhanced resolution of lung injury after ventilator-induced lung injury, as evidenced by a reduced alveolar-arterial oxygen gradient, improvements in lung compliance and lung permeability, a decrease in lung wet:dry weight ratios, and a decrease in alveolar fluid protein concentrations. Human mesenchymal stromal cell therapy also facilitated restoration of lung structure after ventilator-induced lung injury. The finding that fibroblasts did not have any therapeutic effect suggests that the reparative effects are specific to human mesenchymal stromal cells, whether heterogenous or CD362+. The potential of CD362+ human mesenchymal stromal cells to repair the injured lung is further supported by our in vitro data demonstrating their potential to enhance epithelial wound repair.
Limitations and Implications
These studies provide insights into the effects of CD362+ human mesenchymal stromal cells and suggest that additional studies are warranted. Of importance, we did not observe any adverse effects of administration of CD362+ human mesenchymal stromal cells in these studies, which confirms and extends previous findings in toxicology studies (personal communication with S.E.). However, there are some limitations to these studies that should be considered. First, while these studies are conducted in relevant preclinical models of acute lung injury and repair, caution is required in extrapolating to the clinical condition of acute respiratory distress syndrome. Second, we studied a single dose and administration route of CD362+ human mesenchymal stromal cells. This was based on our previous studies demonstrating that this is the most effective dose of heterogenous human mesenchymal stromal cells in these models of Escherichia coli lung injury17 and injury resolution after ventilator-induced lung injury.20 We utilized the intravenous route of administration on the basis of previous studies demonstrating that this route is at least as effective as other, more invasive routes, such as intrapulmonary or intraperitoneal administration.13,20 Additional dose–response studies, and studies using differing routes of administration, would provide further insights regarding these cells. Third, there is some degree of heterogeneity in the profile of effects of the CD362+ human mesenchymal stromal cells in our Escherichia coli injury model compared to the ventilator-induced lung injury repair model. While this is likely to be due, at least in part, to differences in the injury process underlying these different models, and in the time course of injury versus repair studies in these models, caution is nevertheless warranted in interpreting these data. While we demonstrate potentially important effects of CD362+ human mesenchymal stromal cells on epithelial inflammation and repair and on macrophage phagocytosis, additional in vivo mechanistic studies are required. Lastly, while CD362+ human mesenchymal stromal cells do demonstrate some potential advantages over heterogenous human mesenchymal stromal cells, particularly in regard to enhancement of macrophage function, clear superiority over heterogenous human mesenchymal stromal cells was not demonstrated in vivo. However, the demonstration of retention of therapeutic effect in this pure human mesenchymal stromal cell subpopulation, isolated by flow cytometry rather than current standard methods, constitutes an important advance.
Conclusions
In these studies, we demonstrate that transplantation of xenogeneic CD362+ human mesenchymal stromal cells into the immune-competent rat reduced Escherichia coli–induced lung injury and enhanced resolution of ventilator-induced lung injury.
Research Support
Supported by funding from the European Research Council (Brussels, Belgium) under the Framework 7 Programme (ERC-2007-StG 207777, to Dr. Laffey), Science Foundation Ireland (16/FRL/3845, to Dr. Laffey), the Health Research Board Ireland (HRA-POR-2015-1099, to Dr. O’Toole), and from Orbsen Therapeutics Ltd. (Galway, Ireland). The cells used in these studies were provided free of charge by Orbsen Therapeutics Ltd.
Competing Interests
Dr. Elliman is the Chief Scientific Officer, Dr. O’Flynn is the Head of Process Development, and Dr. Deedigan is Head of Analytical Development, at Orbsen Therapeutics Ltd. (Galway, Ireland), a company that is developing the CD362+ mesenchymal stromal cells for therapeutic purposes. Prof. O’Brien and Prof. Barry are directors of and equity holders in Orbsen Therapeutics. The other authors declare no competing interests.