ABSTRACT: Limb development has long served as a model system for studying how a pool of progenitor cells undergoes coordinated spatial patterning and differentiation to produce complex and well-organized structures. Here, we identify a population of naïve-state limb progenitors and show that they differentiate progressively to form the skeleton in a complex nonconsecutive three-dimensional pattern. First, by performing single-cell RNA sequencing of the developing mouse forelimb we created an atlas of cellular compositions and transcriptional states. This atlas revealed three progenitor states: naïve, proximal and autopodial, as well as Msx1 as a marker for the naïve progenitors. In vivo lineage tracing analysis firmly established this role of Msx1 and localized the naïve progenitors to the outer margin of the developing limb, along the anterior-posterior axis. Sequential pulse-chase experiments showed that Msx1 naïve progenitors undergo progressive and simultaneous transition into both proximal and autopodial progenitors and then to Sox9+ chondroprogenitors. Interestingly, the progressive differentiation to chondroprogenitors occurred along all the skeletal segments concurrently, suggesting that skeleton form in a nonconsecutive way. Indeed, by tracking the spatiotemporal sequence of naïve progenitor differentiation, we found that the skeleton forms in a complex nonconsecutive three-dimensional pattern. Overall, these findings propose a new model for limb skeleton development. Limb development has long served as a model system for studying how a pool of progenitor cells undergoes coordinated spatial patterning and differentiation to produce complex and well-organized structures. Here, we identify a population of naïve-state limb progenitors and show that they differentiate progressively to form the skeleton in a complex nonconsecutive three-dimensional pattern. First, by performing single-cell RNA sequencing of the developing mouse forelimb we created an atlas of cellular compositions and transcriptional states. This atlas revealed three progenitor states: naïve, proximal and autopodial, as well as Msx1 as a marker for the naïve progenitors. In vivo lineage tracing analysis firmly established this role of Msx1 and localized the naïve progenitors to the outer margin of the developing limb, along the anterior-posterior axis. Sequential pulse-chase experiments showed that Msx1 naïve progenitors undergo progressive and simultaneous transition into both proximal and autopodial progenitors and then to Sox9+ chondroprogenitors. Interestingly, the progressive differentiation to chondroprogenitors occurred along all the skeletal segments concurrently, suggesting that skeleton form in a nonconsecutive way. Indeed, by tracking the spatiotemporal sequence of naïve progenitor differentiation, we found that the skeleton forms in a complex nonconsecutive three-dimensional pattern. Overall, these findings propose a new model for limb skeleton development. Limb development has long served as a model system for studying how a pool of progenitor cells undergoes coordinated spatial patterning and differentiation to produce complex and well-organized structures. Here, we identify a population of naïve-state limb progenitors and show that they differentiate progressively to form the skeleton in a complex nonconsecutive three-dimensional pattern. First, by performing single-cell RNA sequencing of the developing mouse forelimb we created an atlas of cellular compositions and transcriptional states. This atlas revealed three progenitor states: naïve, proximal and autopodial, as well as Msx1 as a marker for the naïve progenitors. In vivo lineage tracing analysis firmly established this role of Msx1 and localized the naïve progenitors to the outer margin of the developing limb, along the anterior-posterior axis. Sequential pulse-chase experiments showed that Msx1 naïve progenitors undergo progressive and simultaneous transition into both proximal and autopodial progenitors and then to Sox9+ chondroprogenitors. Interestingly, the progressive differentiation to chondroprogenitors occurred along all the skeletal segments concurrently, suggesting that skeleton form in a nonconsecutive way. Indeed, by tracking the spatiotemporal sequence of naïve progenitor differentiation, we found that the skeleton forms in a complex nonconsecutive three-dimensional pattern. Overall, these findings propose a new model for limb skeleton development.
