Cartilages are a diverse group of tissues that form an essential component of the skeletal system. Cartilage-forming cells, aka chondrocytes, form cartilage by secreting a diverse array of structural molecules into the extracellular matrix (ECM), including multiple collagens and proteoglycans, conferring the tissue with shock-absorbing and/or elastic properties. In addition to its impact-buffering role in the axial and appendicular skeleton, cartilage plays an important part in the development and proper function of the skull, and many key regulators of chondrocyte specification and development in the head have been found. Conversely, the biology of many adult secondary or accessory cartilages such as those of the external ear or nasal capsule is still understudied. In this work, we show that, in mice, multiple head and neck cartilages, including neural crest-derived ear, nasal, thyroid, and epiglottic cartilages are composed of lipid-laden cartilage cells called lipochondrocytes. Initially identified by famed anatomist Franz von Leydig in the 1850s, lipochondrocytes were the subject of a series of studies in the 1960s and 1970s only to fall back into obscurity, and are still undescribed in modern studies.

Morphologically, lipochondrocytes resemble white adipocytes, but have distinct lipidomic and transcriptomic profiles. Importantly, lipochondrocytes lack many key components of lipid uptake and mobilization pathways, including many fatty acid transporters and lipolytic enzymes. Unlike adipocytes, the lipid droplets in ear lipochondrocytes are insensitive to metabolic stressors, namely obesity and caloric restriction. In ear cartilage, lipid droplet synthesis primarily depends on de novo lipogenesis, and its pharmacological or genetic inhibition disrupts proper cartilage formation, leading to mice having smaller and misshapen ears.

Further, we observed lipochondrocytes in several facial cartilages of multiple mammalian species, including in bats, primates, carnivores and even marsupials suggesting that a lipochondrocyte-like cell could have existed in the common ancestor of all extant therians. In this thesis, I bring the lipochondrocyte back into view using modern methods in molecular biology, and open new research avenues in the fields of cellular, developmental, and evolutionary biology.

Regeneration of hair follicles (HFs) and dermal adipocytes (DAs) occurs in mouse skin wounds upon large excisional wounding. Although HF regeneration is observed in African spiny mice of the genus Acomys and northern elephant seals after apex predator-inflicted wounding, laboratory rats do not display such regenerative phenotype. Such regeneration defect was observed in large excisional wound healing models in several rat strains, which undergo otherwise normal wound re-epithelialization. Inter-species transcriptome analyses between laboratory mouse and rat wound tissues attributed such lack of HF regeneration to differences in expression of inflammation markers, epigenetic remodelers and pleiotropic signaling molecules, including Satb1, Setd1b, Setdb1, and Whsc1l1. In mice, the origin of de novo HF regeneration has been partially elucidated, whereas the origin of DAs, a complex tissue that proceeds HF regeneration, remained elusive. Functional lineage tracing revealed the origin of DAs to be myofibroblastic. Bulk RNA-sequencing of genetically-labeled, FACS-purified myofibroblasts across a wound healing time course identified Zfp423 to be markedly up-regulated at a time-point coincident with initiation of DA regeneration. Pharmacological and genetic ablation/down-modulation of BMP signaling resulted in a significant DA regeneration defect. Because the origin of myofibroblasts appears to be tissue- and injury context-specific, the origin of myofibroblasts that contribute to DA regeneration in skin wounds was interrogated. Droplet-enabled single cell transcriptome analyses on unsorted, viable cells from wound dermal tissues collected prior the onset of HF regeneration was performed. Dimensionality reduction analyses revealed a large degree of cellular heterogeneity in the dermal compartment of early stage wounds. Furthermore, sub-clustering of wound fibroblasts further revealed a large degree of fibroblast heterogeneity. Pseudotime analyses revealed a putative fibroblast-myofibroblast differentiation trajectory and identified genes, including transcription factors, that may be important in myofibroblast differentiation in skin wounds in vivo. A subset of myofibroblasts expressed hematopoetic markers, most notably Lyz2, suggesting a common monocytic-origin. Full-length single cell RNA-sequencing and immunoblotting analyses of genetically labeled myofibroblasts confirmed these in silico observations. Bone marrow transplantation and functional lineage tracing using pan-hematopoetic Cre drivers demonstrated labeling of DA in regenerated skin wounds, suggesting that a population of hematopoetic-derived myofibroblasts contributes to regeneration of mouse skin wounds.

Background

Results

Conclusion