Ahead of their time: hyperoxia injury induces senescence in developing lung fibroblasts

With time, organisms change, and the changes of aging are manifested in multiple processes often described as a slowing down of cellular machinery. These processes of slowed replication and defective cellular recycling are often grouped under the term “senescence,” a word that has its origins in the Latin senescere (“to grow old”). Senescence was first described by Hayflick and Moorhead (6) in 1961 when they observed deterioration of the replicative capacity of fetal lung cells after multiple passages in culture. Since then, many of the features of senescent cells have been described, chiefly that senescent cells have exited the cell cycle and rarely undergo apoptosis (9). In addition, senescent cells have been demonstrated to have both paracrine and autocrine effects on their cellular microenvironment largely through the secretion of chemokines and cytokines that define the senescence-associated secretory phenotype (SASP) (1).

Over the past 50 years, there has been a growing understanding of the molecular mechanisms underlying cellular senescence and an appreciation that cellular senescence may play a role in chronic lung diseases across the lifespan, including bronchopulmonary dysplasia (BPD), asthma, chronic obstructive pulmonary disease (COPD), and idiopathic pulmonary fibrosis (2, 7, 9, 11). Recent work has suggested that targeting the senescent cell population through senolytic therapy may provide a therapeutic strategy that would prevent or reverse these chronic lung diseases (5). Understanding the precise mechanisms whereby senescent cells contribute to chronic lung disease and how to target these mechanisms is an active area of scientific inquiry.

In this issue of the American Journal of Physiology-Lung Cellular and Molecular Physiology, You and colleagues (12) report that moderate hyperoxia exposure of developing lung fibroblasts induces a senescent phenotype, with increased cellular size and decreased proliferative capacity. Further, hyperoxia-exposed cells had reduced levels of autophagy when compared with controls and demonstrated increased expression of genes associated with a SASP profile including TNF-a, IL-1, IL-8, and PAI-1. Conditioned media from hyperoxia-exposed fibroblasts applied to naïve fibroblasts engendered a fibrotic response with increased deposition of collagen III, and increased fibroblast proliferation. Interestingly, in this data the hyperoxia-exposed cells themselves had decreased proliferation typical of senescent cells, whereas the conditioned media increased proliferation in naïve fibroblasts. Together, these data suggest that the SASP profile may elicit a different paracrine effect on neighboring cells from the autocrine effect on the SASP-secreting cells. In addition, the senescent cells themselves did not exhibit a fibrotic phenotype; however, their conditioned media fostered a profibrotic phenotype in naïve fibroblasts.

Besides broadening our appreciation for the contribution of senescence to the hyperoxia response in the developing lung, this article makes a significant methodologic contribution to those studying hyperoxia lung injury. Hyperoxia is used extensively to model bronchopulmonary dysplasia in vitro and in vivo (8). With in vitro systems there is often significant variability in what oxygen exposure defines “normoxia” in tissue culture conditions. While laboratories have historically cultured cells at 21% O₂, there is a growing appreciation that cells in the body typically grow under less oxygen tension than that provided by media incubated at 21% O₂, and that fetal cells are accustomed to growing with even lower oxygen exposure than cells derived from adult subjects (3). For this reason, stem cells are often cultured and differentiated in vitro at 1–5% O₂, and prior reports in other tissues suggest that 5% O₂ may represent tissue culture normoxia, with partial pressures of oxygen more consistent with what cells encounter in the organism (3). In this article, cells cultured at 40% demonstrated increased senescence features when compared with cells cultured at 21%. Additional experiments showed that cells cultured at 21% had a significantly increased senescent phenotype when compared with 5%, suggesting that the contribution of oxygen to the senescent phenotype was dose dependent, and that cell culture at 21% was sufficient to elicit some senescent features.

While it is possible that the chronic exposure to hyperoxia accelerates the aging process in developing lung fibroblasts resulting in a senescent phenotype, it is important to note that senescence is also a normally programmed mechanism during development that contributes to developmental patterning (10). Increased senescence after lung injury may represent a compensatory phenotype whereby developmental pathways are reactivated as an attempt to repair the lung (4), and the exposure to hyperoxia injury may lead to senescence similar to the ways that other developmental pathways are perturbed during saccular stage injury. Future work in the field seeks to understand and identify the subpopulations of fibroblasts that contribute to the mechanisms of injury, repair, and disease. Full characterization of the fibroblasts with a senescent phenotype after hyperoxia injury will yield important mechanisms into the role of senescence in the pathogenesis of BPD, will broaden our understanding of the role of senescent pathways in normal lung development, and will illuminate how these pathways might be targeted for therapy in the future.

GRANTS

This work was supported by NIH National Institute of Child Health and Human Development Grant K12 HD087023 (J. M. S. Sucre), National Heart, Lung, and Blood Institute Grants K08HL143051(J. M. S. Sucre) and K08HL127102 (E. Plosa), and The Francis Family Foundation (J. M. S. Sucre).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.M.S.S. drafted manuscript; E.J.P. edited and revised manuscript; J.M.S.S. and E.J.P. approved final version of manuscript.