Human umbilical cord mesenchymal stem cells reduce fibrosis of bleomycin-induced lung injury

Abstract

Acute respiratory distress syndrome is characterized by loss of lung tissue as a result of inflammation and fibrosis. Augmenting tissue repair by the use of mesenchymal stem cells may be an important advance in treating this condition. We evaluated the role of term human umbilical cord cells derived from Wharton's jelly with a phenotype consistent with mesenchymal stem cells (uMSCs) in the treatment of a bleomycin-induced mouse model of lung injury. uMSCs were administered systemically, and lungs were harvested at 7, 14, and 28 days post-bleomycin. Injected uMSCs were located in the lung 2 weeks later only in areas of inflammation and fibrosis but not in healthy lung tissue. The administration of uMSCs reduced inflammation and inhibited the expression of transforming growth factor-beta, interferon-gamma, and the proinflammatory cytokines macrophage migratory inhibitory factor and tumor necrosis factor-alpha. Collagen concentration in the lung was significantly reduced by uMSC treatment, which may have been a consequence of the simultaneous reduction in Smad2 phosphorylation (transforming growth factor-beta activity). uMSCs also increased matrix metalloproteinase-2 levels and reduced their endogenous inhibitors, tissue inhibitors of matrix metalloproteinases, favoring a pro-degradative milieu following collagen deposition. Notably, injected human lung fibroblasts did not influence either collagen or matrix metalloproteinase levels in the lung. The results of this study suggest that uMSCs have antifibrotic properties and may augment lung repair if used to treat acute respiratory distress syndrome.

Figure 1

The expansion and characterization of umbilical cord mesenchymal stem cells. A: Wharton’s jelly was dissected from the umbilical cord and placed in culture wells. The cells derived from the Wharton’s jelly were trypsinized following growth from the explanted tissue and cell sorted for a CD31⁻CD34⁻CD45⁻CD73⁺ population that were further propagated in culture. Characterization of the uMSCs demonstrates a fibroblastic appearance. B: Flow cytometry confirms the uMSCs are a homogenous cell population of CD31⁻CD34⁻CD45⁻CD73⁺ cells. C: Following propagation to confluence, uMSCs were cultured in conditioned media that would encourage the growth of osteoid tissue. Following 14 days in culture, Von Kossa staining (black) demonstrates calcium deposition that is suggestive of osteoblastic activity and bone formation. D: uMSCs were injected into the tail vein 24 hours following bleomycin injury. Using a specific anti-human anti-mitochondrial antibody, we isolated the injected uMSCs to the fibrotic areas of injury (brown staining and arrows) and to alveolar areas of the lung. Magnifications: ×50, ×100, and ×200.

Figure 2

uMSCs inhibit lung fibrosis. A: Histology of lung obtained from a SCID mouse at 7, 14, and 28 days following bleomycin lung injury and treatment with uMSCs. The control demonstrates normal lung architecture. There are mild infiltrates involving 10 to 25% of the lung at 7 days post-bleomycin injury. These changes are attenuated by uMSC injection at 7 days. At 14 days there is a significant increase in lung infiltrates involving 25 to 50% of the lung with some distortion of the lung architecture and collagen deposition. These changes are also attenuated by uMSC injection. At 28 days post-bleomycin injury, infiltrates occupy more than 75% of the lung with significant distortion of lung architecture and the presence of fibrotic foci. uMSC injection reduced the number of infiltrates as well as the number of fibrotic foci. uMSCs also preserved the lung architecture at 28 days. Magnification (all sections), ×100 (n = 8). B: The Ashcroft score of fibrosis shows a progressive elevation as lung injury evolves from 7 to 28 days after bleomycin injury. The levels of fibrosis at each time point are significantly reduced by uMSC treatment. Furthermore, alveolar and interstitial infiltrates progressively increased from 7 to 28 days following bleomycin lung injury. At each time point the number of infiltrates was significantly reduced by uMSC treatment. *P < 0.05 comparing bleomycin plus uMSCs to bleomycin alone for each time point of analysis (n = 8).

Figure 3

uMSCs inhibit the expression of profibrotic and proinflammatory cytokines. A: Transcripts for cytokines extracted from whole lung suspensions and analyzed by quantitative PCR demonstrate a significant increase in all cytokines at 14 days following bleomycin lung injury with significant attenuation of TGF-β, IFN-γ, and MIF following uMSC injection. *P < 0.05 comparing bleomycin plus uMSCs versus bleomycin alone (n = 6). The results are expressed as fold expression over normal lung. B: The enzyme-linked immunosorbent assay for TGF-β1 demonstrated a significant increase in TGF-β1 following bleomycin at 14 days with a significant reduction in levels with uMSCs C: Densitometry of the Western blot analysis for p-SMAD-2 a transcription factor mediating the actions of TGF-β1, done on whole lung protein extracts (14 days) demonstrate an increase in TGF-β1 following bleomycin-induced lung injury and a fall with uMSC treatment (n = 8). *P < 0.05 compared with healthy controls. +P < 0.05 comparing bleomycin plus uMSCs to bleomycin alone.

Figure 4

The uMSCs reduce collagen deposition. A: The hydroxyproline assay is a marker of collagen deposition. There is a significant elevation of collagen deposition following bleomycin-induced lung injury at 14 and 28 days but not at 7 days with significant abrogation by uMSC injection at 14 and 28 days. Notably there is no influence on collagen levels in the lung with injection of uMSCs into healthy mice. In contrast to uMSCs, injection of primary human lung fibroblasts into bleomycin-injured mice does not reduce the levels of collagen deposition. *P < 0.05 compared with healthy controls. +P < 0.05 comparing bleomycin plus uMSCs to bleomycin alone (n = 8). B: Transcripts for mouse collagen type 1 α 1, a marker for collagen deposition, extracted from whole lung suspensions and analyzed by quantitative PCR demonstrate a significant increase in collagen type 1 α 1 following 14 days of bleomycin injury. The expression of collagen type 1 α 1 mRNA in the bleomycin plus uMSC group was reduced in comparison with the bleomycin alone treated group. *P < 0.05 compared with healthy controls. +P < 0.05 comparing bleomycin plus uMSCs to bleomycin alone (n = 8). C: Masson’s trichrome staining for collagen deposition (blue) at 28 days demonstrates minimal collagen within control lung. There is an increase in collagen deposition with bleomycin that is reduced by uMSC treatment.

Figure 5

uMSCs augment the expression of MMPs in the injured lung. A: Zymographic analysis of whole lung suspensions shows expression of latent and active forms of MMP-9 and MMP-2 by all samples including healthy controls, healthy mice plus uMSC injection, bleomycin plus fibroblasts, bleomycin alone, as well as bleomycin plus uMSCs. We suggest that the product at 41 kd is a nonspecific catalytic product. B: Analysis of the protein bands on zymography by densitometry demonstrates an increase in MMP-2 activity with bleomycin lung injury. The increase in MMP-2 activity following bleomycin treatment was unaffected by fibroblast injection but further elevated above bleomycin levels with uMSC injection. *P < 0.05 compared with healthy controls. +P < 0.05 comparing bleomycin plus uMSCs to bleomycin alone. †P < 0.05 comparing bleomycin plus fibroblasts to bleomycin alone. (n = 6). C: There is no difference in MMP-13 levels between bleomycin and bleomycin plus uMSC-treated mice as analyzed by Western blots.

Figure 6

A: There is inhibition of TIMP expression following uMSC treatment. Transcripts for TIMPs extracted from whole lung suspensions and analyzed by quantitative PCR demonstrate a significant increase in TIMPs 1–4 at 2 weeks following bleomycin lung injury with attenuation of TIMPs 1, 3, and 4 by uMSC injection. The results are expressed as fold expression over normal lung. *P < 0.05 comparing bleomycin plus uMSCs to bleomycin alone (n = 8). B and C: Densitometry of the lung TIMP-1 and TIMP-2 bands analyzed by reverse zymography demonstrated a significant up-regulation of TIMP-2 protein levels following bleomycin-induced lung injury but a significant down-regulation of TIMP-2 protein expression by uMSC injection to bleomycin-treated mice. There are no significant changes in TIMP-1 levels between the groups. *P < 0.05 compared with healthy controls. +P < 0.05 comparing bleomycin plus uMSCs to bleomycin alone (n = 8).