Loss of lung mesenchymal identities in culture and a screening approach to identify restorative cues

Lung fibroblasts generate and respond to mechanical, biochemical, and matrix cues present in their microenvironment. With the advent of next-generation sequencing technologies, multiple studies describe transcriptionally unique fibroblast subpopulations in the human lung. However, limited published data suggest a loss of fibroblast native phenotypes and functions after culture ex vivo. In this study, we characterized changes in transcriptional programs of human lung mesenchyme isolated from freshly procured tissue and maintained in traditional cell culture conditions. Our results demonstrate that fibroblasts isolated and cultured in this manner adopt transcriptional programs largely distinct from those observed in vivo. To recapitulate distinct native fibroblast states in vitro, we sought to develop a screening approach to identify cues promoting native fibroblast identities. From published single-cell data, we identified unique transcriptional markers of alveolar and adventitial fibroblast subpopulations and validated the sensitivity of ELISAs for detecting changes in secreted markers of these fibroblast subpopulations. We then stimulated primary human lung fibroblasts with soluble cues known to act on fibroblasts, quantifying changes in secreted and transcriptional markers by ELISA and qPCR. Although our small pilot screen did not identify single cues capable of fully recapitulating fibroblast in vivo states, it established a system that can be expanded to broadly screen additional cues and pointed toward factors likely to be critical in developing better culture models for studying human lung fibroblast function and plasticity.<b>NEW & NOTEWORTHY</b> Recent studies highlight transcriptionally distinct fibroblast subpopulations in human lungs. We observed the loss of these native transcriptional programs as fresh isolated cells are maintained in traditional culture conditions. Identifying the signals defining native fibroblast identities will be pivotal to creating culture models that preserve unique subpopulations. The screening system developed here will allow the investigation of a broad selection of cues, leading to better culture models for studying human lung fibroblast function and plasticity.