Lung transplantation is an established lifesaving therapy for an increasing number of patients suffering from irreversible end-stage pulmonary diseases. Although early outcomes (hospital mortality and one-year survival rates) have dramatically improved over recent years, long-term survival rates are still far from those achieved with other solid organs. In the last decade, the 5-year survival rate has remained 50%,1 which contrasts with survival rates above 70% for heart, kidney, liver or pancreas recipients.2

Chronic lung allograft dysfunction (CLAD) as a whole, and bronchiolitis obliterans syndrome (BOS) as a specific phenotype, is the main cause of death after the first year post-lung transplant1 and represents the Achilles’ heel of this therapeutic option. Half of our patients will develop BOS by 5 years and more than 70% by 10 years after transplantation.1 Its pathological expression is termed bronchiolitis obliterans (BO) and, essentially, consists of a fibrotic process affecting the small airway. In the early phases a lymphocytic infiltrate is observed in the submucosa and epithelium, which is called lymphocytic bronchiolitis.3 Later, necrotic changes and ulceration occur in the epithelium and trigger a chronic inflammatory reaction, which results in fibroblastic migration into the airway lumen. The subsequent fibroproliferative process leads to small airway partial or total obstruction.4 The etiology of this process is still unknown. Although likely multifactorial, there is increasing evidence that innate and acquired immunological factors play a crucial role in its course.5,6

Stem cells have shown a consistent ability to modulate the inflammatory response and enhance the healing process of many disorders in different target tissues.7–9 Besides, the lack of Major Histocompatibility Complex (MHC) class I and II expressions on these cells prevents immunological rejection. Furthermore, it has been described that mesenchymal stem cells applied both locally and systemically may slow the development of bronchiolitis obliterans in other animal models because of this immunomodulatory ability.10

In this paper, we aim to assess the impact of allogeneic adult bone marrow-derived mesenchymal stem cells (BMSC) in an experimental rat model of BO using different routes of administration.

This is an experimental randomized study. The experimentation committee of Hospital Universitario Puerta de Hierro-Majadahonda, Spain, approved all phases of the project.

Bone marrow-derived mesenchymal stem cell (BMSC) isolation and characterization

BMSC isolation from male rats and phenotypic characterization were performed as described in previous publications from the Neuroscience Research Unit at our Hospital.12 BMSC were retrieved from tibia and femur of male rats by washing both epiphysis and diaphysis with alfa-minimun essential medium (alfa-MEM). Bone marrow was collected and dissociated by mechanical methods. The obtained homogeneous sample was then filtered and incubated at 37°C in 5% CO2 for 5 weeks. For characterization of the BMSC phenotype, we used flow cytometry analysis (Cytometer Cytomics FC500 MPL – Beckman Coulter, Inc., Brea, CA, USA, and CFC 500MXP software, version 2.2 – Beckman Coulter Inc., Miami, FL, USA). The following cell-surface markers were characterized CD11b, CD29, CD31, CD45 and CD90, with a suitable fluorochrome-conjugated isotype control (all antibodies were from AbD Serotec – Oxford, United Kingdom, all specific to rats). Only cells under 5 weeks in culture and with phenotypic characteristics for MSC published by International Society of Cellular Therapy (i.e., expression of CD29 and CD90 [95% positive] and absence of expression of CD11b, CD45 or CD31 [<5% positive]) were used (Fig. 1).13

Fig. 1.

Characterization of BMSC, by flow cytometry. Rat BMSC were labeled with CD29, CD31, CD34, CD44, CD45 and CD90 antibodies and analyzed by flow cytometry. BMSC were positive for CD29, CD44 and CD90 and negative for CD31, CD34 and CD45.

(0.32MB).

BOS hampers the long-term survival of lung transplant recipients. Regardless of the immunosuppressive regimen, only one-third of our patients will remain free from BOS 10 years after transplantation. Therefore, it is of paramount importance to find therapeutic strategies that prevent or attenuate the incidence of this condition.

Histopathologically, end-stage BOS consists of an excessive fibroproliferative thickening of the bronchiolar walls in response to epithelial injury. In the initial stages, lymphocytic infiltrates invade the epithelium and submucosal layer and the epithelium loses its architectural consistency. As the disease progresses, the bronchiolar lumen is progressively replaced by granulation tissue containing fibroblasts and myofibroblasts, and finally forms a fibrotic scar.3 Although this process was initially thought to be a direct result of T-cell mediated injury of graft structures,17 there is now evidence to support that antibody-mediated rejection (circulating antibodies to donor HLA molecules)18,19 and autoimmune responses to airway components (collagen V, Kα-1 tubulin) are also associated.20,21

Heterotopic tracheal transplantation into the lung parenchyma has been successfully used to reproduce the histologic changes of BO lesion and investigate angiogenesis and fibroproliferative tissue remodeling.14 However, some limitations have been described with regard to this animal model, such as the lack of connection of the graft to the airway or vessels, as well as the fact that BO occurs in the small airways in the clinical setting. Despite these limitations, we were able to reproduce the typical histological changes of BO lesions.

Our results show that BMSC are able to impact the histologic course of BOS at any time point and through any route of administration, moving from a pattern of dense fibrous scar tissue to another kind of histologic presentation, with various degrees of inflammation, edema and neovascularization.

The most impressive reductions in fibrosis were seen when BMSC were applied at day 7. In particular, the rats receiving BMSC via the endotracheal route at day 7 showed no fibrosis and no inflammation at all, suggesting that early endobronchial therapy should be the preferred choice.

Concerning the dosage of BMSC, the literature is unclear on its importance.10,12 The most effective concentration has not yet been established and warrants further research. Therefore, we planned to use the maximum tolerated concentration depending on the route of administration.

It has been described, both in vitro and in vivo, that mesenchymal stem cells (MSC) have immunomodulatory properties that promote or prevent innate and acquired inflammatory response, depending on the local environment. They can inhibit the proliferation and function of T-Lymphocytes, natural killer (NK) cells and dendritic cells (DC). MSC are able to prompt T cell expansion toward a regulatory phenotype (Treg) by the production of prostaglandin E2 (PGE2) and transforming growth factor β-1 (TGFβ-1).22 Similarly, MSC can shift the macrophage population toward a predominantly immunosuppressive M2 phenotype by secretion of IL-10, IL-2 and IL-13.23,24 The effect of MSC in B cells is controversial; some studies have shown inhibited proliferation, differentiation and immunoglobulin production,25–27 whereas other groups have reported stimulatory effects.28,29

The differentiation of BMSCs is strongly influenced by T cell activation and the type and concentrations of cytokines in the local environment at the time of the exposure, which explains the different properties they can show in different scenarios.30

This special feature of MSC supports our observation that the administration of these cells at the appropriate moment is critical to get the desired effect. The most impressive results were seen at day 7, when histopathological changes were at early stages in our model, with some degree of inflammatory activity. Prior to this (T0), there was no damage to restore and, later on (T14), established fibrotic changes exceeded their ability to completely reverse the process.

When tissue is damaged, circulating MSC are attracted to the site, and, as they have multipotent differentiation potential, they can differentiate into mature functional cells replacing damaged ones. Nonetheless, this is arguably a minor event, as MSC mainly exert their reparative effects by stimulating in situ progenitor cells, endothelial cells and fibroblasts to repair the injury through immunoregulation and growth factor secretion (i.e., paracrine mechanism).31

One of the most interesting findings in our experiment is the BMSC-mediated response to injury, developing an alveolar-like tissue and mimicking what is supposed to be a normal tissue in that environment. Moreover, those cells involved in this peculiar tissue showed BMSC markers and produced surfactant protein B, a specific biomarker of alveolar epithelial cells. Although it is impossible to proof whether this alveolar-like tissue is functional or not, this finding not only shows BMSC inclination to repair and restore normal tissue, but also it might be proof of their transdifferentiation ability,31–33 leading to mature cells of different lineages.

Another example of this is the finding of BMSC (CrY+) taking part of the walls of neovessels, which are morphologically identical to endothelial cells (Fig. 5).

Previous heterotopic tracheal transplant models have shown the capacity of stem cells to attenuate histopathological changes of BO after either topical or systemic administration.34–36 In this study, we have found that BMSC are able to nest and survive, detect an injury scenario, modify the histopathological course of BO lesions, and repair the structure. Moreover, although intravenous and locally injected BMSC have an impact in the BO lesions, the most remarkable effects are seen after endotracheal instillation of BMSC. This finding might mean that the endoscopic administration of BMSC should be considered for further experimental or clinical projects, while potential systemic collateral effects in other organs are minimized.

Given the fact that lung injury and repair is a complex process in a model of BOS, it is difficult to attribute our experimental results to one or another mechanism led by BMSC. It is clear that both paracrine, as the main mechanism, and differentiation, as a minor one, have a role in the modification of the histological course of BOS.

Although we have been able to find statistically significant results with very small sample sizes, using a control group and masking techniques, we must be cautious and await the results of this type of therapy in other models before generalizing our findings.

In the clinical scenario, there have been two phase I studies so far. Chambers et al.37 infused 2×106 BMSC from an unrelated donor to 10 patients with BOS via a peripheral vein twice a week for two weeks, without any procedure related toxicity. Similarly, Keller et al.38 infused 1–4 million of allogeneic mesenchymal stem cells into nine patients with moderate BOS without any negative impact in measured clinical, functional or laboratory variables, opening the door for the use of these cells in the setting of established BOS.

Bone marrow-derived mesenchymal stem cells modify the histopathological development of BO in this heterotopic tracheal transplantation animal model of BO. Moreover, they are able to express alveolar cell markers, such us pulmonary surfactant protein B.

Endotracheal administration of BMSCs might be considered an effective way of preventing BO development in this animal model.

This project has been granted by Fundación Mutua Madrileña (identification number: PR87532011).

David Gomez-de-Antonio: design, animal surgeries, interpretation of results, manuscript.

Jose Luis Campo-Cañaveral de la Cruz: design, experimental surgeries, interpretation of results, manuscript.

Mercedes Zurita: FISH and IHQ techniques.

Martín Santos: experimental surgeries anesthesia.

Carmen González Lois: pathological analysis of the samples.

Andrés Varela: manuscript edition.

Jesús Vaquero: manuscript edition.

The authors declare no conflict of interests.