At resting, an average person inhales/exhales 5-8 litters of air. This air may vary in oxygen content, carry allergen, pollutants or pathogens. Each of these signals would trigger a specific set of responses. For example, allergen causes asthma while intermittent hypoxia causes pulmonary hypertension. How the specificity is established is poorly understood.

We recently found that pulmonary neuroendocrine cells, which represents less than 1% of cells in the lung, function as sensors on the airway wall (Branchfield et al., Science, 2016). These cells are the only innervated epithelial cells in lung, and act through secreting potent neuropeptides, neurotransmitters and amines. We found that overactivation of these cells leads to heightened baseline immune responses. We will test the hypothesis these cells serve as key nodes that mediate lung, immune and nervous system interactions.

Study of the pulmonary neuroendocrine cells also led us to investigate the molecular nature of the neural circuit that controls lung function. We are teaming up with neurobiologists, genomicists and physiologists to map the neural circuit that originates from and returns to the lung. Our goal is to understand which aspects of lung function is controlled by this neural circuit, and what underlie the specificity between inputs and outputs.

Mockup Image


Alveologenesis is the final step of lung maturation, which subdivides the alveolar region of the lung into smaller units called alveoli. Each of the nascent dividers serves as a new gas-exchange surface. Disruption of the final steps of lung development including alveologenesis results in alveoli simplification, as is seen in premature infants diagnosed with bronchopulmonary dysplasia (BPD). BPD is often associated with lifelong breathing deficiencies, pulmonary hypertension and accelerated decline in respiratory capacity.

To date, a majority of studies of alveologenesis rely on two-dimensional (2D) analysis of tissue sections. Given that an overarching theme of alveologenesis is thinning and extension of the epithelium and mesenchyme to facilitate gas exchange, often only a small portion of a cell or a cellular structure is represented in a single 2D plane. We have used a three-dimensional (3D) approach to examine the structural architecture and cellular composition of myofibroblasts,lipofibroblasts, alveolar type 2 cells and elastin extracellular matrix in normal as well as BPD-like mouse lungs (Branchfield et al., Dev Biol, 2016). The insights revealed by 3D reconstruction of the septae set the foundation for future investigations of the mechanisms driving alveologenesis, as well as causes of alveolar simplification in BPD. We have also used a genetic approach to investigate the role of signaling pathways such as FGF (Herriges et al, Developmental Cell, 2015).

Ongoing work in this direction interrogates how different elements of prematurity impacts long-term lung function. We are interested in why some BPD patients develop pulmonary hypertension while others do not, and why BPD patients are more susceptible to viral infections.

Mockup Image


In the trachea and main bronchi, airway smooth muscle and cartilage are localized to complementary domains surrounding the airway epithelium. Proper juxtaposition of these tissues ensures a balance of elasticity and rigidity that is critical for effective air passage. It was unknown how this tissue complementation is established during development. We carried out a study to dissect the developmental relationship between these tissues by genetically disrupting either smooth muscle formation or cartilage formation and assessing the impact on the remaining lineage (Hines et al., PNAS, 2013). We found that in both the trachea and main bronchi, loss of either smooth muscle or cartilage resulted in an increase in cell number of the remaining lineage, namely the cartilage or smooth muscle, respectively. However, only in the main bronchi, but not in the trachea, did the loss of either smooth muscle or cartilage lead to a circumferential expansion of the remaining lineage. In addition to changes in the smooth muscle, cartilage deficient tracheas displayed epithelial phenotypes, including decreased basal cell number, precocious club cell differentiation, and increased secretoglobin expression. Smooth muscle deficient tracheas also show cartilage segmentation defects. These findings suggest extensive crosstalk among the cartilage, smooth muscle and epithelium to shape the airway.

Ongoing work in this direction extends to include the vasculature and the neurons, and consider the airway microenvironment as a whole. We aim to test the possibility that in patients with malformation of either the airway cartilage or smooth muscle at birth, such as in tracheomalasia and cystic fibrosis patients, these developmental structural defects may contribute to the complex airway dysfunction that manifest at a later time.

Mockup Image


As the acquisition of high-resolution patient genomic data becomes routine, we seek to capture this opportunity and use advanced technology such as CRISPR/Cas9 genome editing to interrogate causal relationships between patient-specific genetic variants and phenotypes and disease mechanisms. Towards this goal, we started with congenital disorders as they are caused by single or small numbers of mutations with large effects.

In the past few years, we have studied a number of lung-related congenital disorders, including tracheo-esophageal fistula (Domyan et al., Development 2011), tracheobronchomalacia (Hines et al., PNAS 2013), and congenital diaphragmatic hernia (CDH) (Domyan et al., Dev Cell, 2013; Branchfield et al., Science, 2016, McCulley et al., in prep). Using CDH as an example, we have been collaborating with human geneticists to test the functional significance of the genomic variants found in their trio (patient and parents) whole exome or genome sequencing efforts. We have established a pipeline to first determine candidate gene expression in the developing lung (Herriges et al., Dev Dynamics, 2012), and then use CRISPR/Cas9 tools to recapitulate the mutations in mice, and determine which variants may be causal to the phenotypes.

Thus far, we have uncovered different mechanisms from distinct genetic models of CDH, suggesting that the identification of the genetic lesions in human CDH may inform personalized treatments. Furthermore, the studies of disease models also led to exciting basic biology discoveries (Domyan et al., Dev Cell, 2013; Branchfield et al., Science, 2016, McCulley et al., in prep). We look forward to expanding this line of investigation to more genes and diseases.

Mockup Image