About

Tushar Desai is a Professor of Medicine specializing in Pulmonary, Allergy, and Critical Care Medicine at Stanford University. He has practiced pulmonary medicine since 2002, focusing clinically on general pulmonary and Interstitial Lung Diseases such as Idiopathic Pulmonary Fibrosis (IPF). His research centers on understanding the causes of lung diseases and developing molecular and cell-based treatments, with particular attention to lung stem cells and the molecular signals that regulate their activity to repair and regenerate lung tissue after injury. His lab investigates key signaling pathways like Wnt and the role of TERT in lung stem cell biology, aiming to manipulate these signals for therapeutic purposes. Desai's work includes studying gene correction techniques in lung stem cells, such as CRISPR-mediated correction of CFTR mutations for cystic fibrosis, and developing protocols for autologous stem cell transplantation to treat lung diseases. He holds multiple academic appointments and is involved in translational research that bridges basic science and clinical applications, contributing to advancements in regenerative medicine and lung disease treatment.

Research topics

• Biology
• Medicine
• Immunology
• Pathology
• Virology
• Internal medicine
• Genetics
• Cell biology
• Computational biology
• Molecular biology
• Neuroscience
• Surgery
• Pediatrics
• Anatomy

Selected publications

• Abstract 1957: DNA methylation differentiates early-stage lung adenocarcinoma cells-of-origin with distinct smoking history and grade.

  Cancer Research · 2026-04-03

  article

  Abstract Lung cancer is a major cause of cancer-related mortality worldwide. Approximately 10%-20% of lung cancers occur in patients with no smoking history and are mostly lung adenocarcinoma (LUAD). Studies in mouse models showed that LUAD cases arising from alveolar epithelial type I (AT1) cells transitioning to alveolar type II (AT2) are less aggressive than those that originate directly from type 2 (AT2), but the origin of human LUAD remains unclear. Leveraging the bulk methylation array data from the EAGLE and Sherlock-Lung studies, two large multi-omics studies of lung cancer in people with and without smoking history, respectively, we infer the cells-of-origin for early-stage tumor samples and examine their associations with smoking history and grade. We impute cells-of-origin from bulk methylation array data of tumor and normal samples collected from 388 stage I LUAD cases without smoking history and 238 with smoking history. For the methylation reference profiles of normal lung cell types, we combined a published normal human cell methylation data set (Loyfer et al. 2023) from whole-genome bisulfite sequencing (WGBS) as well as purified AT1 and AT2 enzymatically converted methylation profiles sequenced in-house. We then perform principled feature engineering to select methylation sites that best distinguish cell types and represent each cell type in a balanced manner. To estimate the proportion of each lung cell type in the samples, we employ a combinatorial optimization strategy by formulating and solving the deconvolution as a quadratic program. The inferred compositions of tumor origins are then evaluated in relation to smoking history and tumor grade based on the International Association for the Study of Lung Cancer (IASLC) grading system. The tumor samples of LUAD cases without smoking history (n=238) have 11% higher AT1 presence compared to those with smoking history (n=388, p<0.05). Among cases without smoking history and with coherent grades, the tumor samples with well or moderate grade (n=96) have on average 13% higher AT1 presence than those with poor grade (n=102, p<0.05). Notably, tumor samples of stage I LUAD cases with no smoking history have the highest overall AT1 presence among all groups (on average 26%). All normal samples, regardless of smoking history and grade, have no significant difference in AT1 presence (on average 10%). Our study establishes a principled deconvolution framework to infer tumor origins in LUAD using DNA methylation data. Mirroring findings in mouse models, our results indicate that AT1 cells may generate LUAD in humans, and that this appears to occur more frequently in patients without smoking history and with better tumor grades. Our work adds a new dimension to the understanding of early-stage LUAD and sheds light on patient stratification, prognostic evaluation, and therapeutic targeting of tumor origin-specific vulnerabilities. Citation Format: Xuan Cindy Li, Diego Almanza, Phuc Huu Hoang, Thi-Van-Trinh Tranh, Nicholas H. Juul, Maximilian Diehn, Tushar J. Desai, Maria Teresa Landi. DNA methylation differentiates early-stage lung adenocarcinoma cells-of-origin with distinct smoking history and grade [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 1957.

  PublisherDOI

• Histological signatures map anti-fibrotic factors in mouse and human lungs

  Nature · 2025-03-19 · 22 citations

  article
  PublisherDOI

• The Discovery of a Novel Alveolar Type 2 Subpopulation That Express Tert and Function as Unipotent Stem Cells

  American Journal of Respiratory and Critical Care Medicine · 2025-05-01

  articleSenior author

  Abstract RATIONALE. Alveolar epithelial turnover is slow under homeostatic conditions, but the lung retains a remarkable regenerative capacity following injury. Disruptions in this renewal process contribute to pulmonary diseases such as idiopathic pulmonary fibrosis. While alveolar type II (AT2) cells serve as progenitors for the alveolar epithelium, their heterogeneity and functional specialization remain incompletely understood. A subset of AT2 cells expressing Axin2 (Axin2+ AT2) has been identified as a progenitor population with enhanced renewal potential. Given the known role of telomerase (TERT) in stem cell biology and pulmonary fibrosis, we sought to determinre whether TERT expression marks a distinct progenitor subset within the AT2 population.METHODS. To identify and trace TERT-expressing cells, we utilized a novel TERT-Cre-ERT2 > Rosa26-tdTomato mouse model. Immunohistochemistry (IHC) was performed to characterize TERT+ cells in the alveolar epithelium. Long-term lineage tracing (12 months) was performed to determine the expansion and differentiation potential of TERT+ AT2 cells. Injury-induced proliferation was evaluated using a butylated hydroxytoluene (BHT) injury model, which selectively damages AT1 cells. To compare TERT+ AT2 and Axin2+ AT2 populations, we crossed TERT-tdTomato and Axin2-rtTA-GFP mice, enabling dual labeling. Finally, bulk RNA sequencing was conducted on sorted AT2 subpopulations (Axin2+ AT2, TERT+ AT2, and Bulk AT2) to assess transcriptional differences.RESULTS. IHC analysis identified LAMP3+/Tomato+ AT2 cells (TERT+ AT2) as a rare AT2 subpopulation scattered throughout the alveoli. Longitudinal lineage tracing revealed a fivefold expansion of TERT+ AT2 cells over time, suggesting enhanced progenitor activity. However, unlike Axin2+ AT2 cells, TERT+ AT2 cells differentiated only into AT2 cells and not AT1 cells, indicating a distinct unipotent progenitor role.In response to BHT-induced AT1 injury, TERT+ AT2 cells exhibited increased proliferative capacity compared to bulk AT2 cells, yet they failed to differentiate into AT1 cells, reinforcing their lineage restriction. Pulse labeling in TERT-tdTomato;Axin2-rtTA-GFP mice confirmed TERT+ AT2 and Axin2+ AT2 as independent populations. RNA sequencing analysis showed TERT+ AT2 cells were transcriptionally distinct, with downregulation of Wnt target genes, E2F targets, and pathways regulating cell attachment and migration, suggesting a role in alveolar homeostasis rather than plasticity.CONCLUSION. Our findings establish TERT+ AT2 cells as a unique, unipotent progenitor population that contributes to AT2 maintenance during aging and injury, without transitioning to AT1 cells. This functionally distinguishes them from Axin2+ AT2 cells, highlighting cellular heterogeneity within the AT2 compartment and providing new insights into alveolar epithelial regeneration.

  PublisherDOI

• Neuroendocrine cells orchestrate regeneration through Desert hedgehog signaling

  Cell · 2025-06-09 · 4 citations

  articleOpen access
  PublisherOA PDFDOI

• Durable reconstitution of sinonasal epithelium by transplant of CFTR gene corrected airway stem cells

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-01-26

  preprintOpen access

  Abstract Modulator agents that restore cystic fibrosis transmembrane conductance regulator (CFTR) function have revolutionized outcomes in cystic fibrosis, an incurable multisystem disease. Barriers exist to modulator use, making local CFTR gene and cell therapies attractive, especially in the respiratory tract. We used CRISPR to gene-correct CFTR in upper airway basal stem cells (UABCs) and show durable local engraftment into recipient murine respiratory epithelium. Interestingly, the human cells recapitulate the in vivo organization and differentiation of human sinus epithelium, with little expansion or contraction of the engrafted population over time, while retaining expression of the CFTR transgene. Our results indicate that human airway stem cell transplantation with locoregional restoration of CFTR function is a feasible approach for treating CF and potentially other diseases of the respiratory tract.

  PublisherDOI

• An embryonic artery-forming niche reactivates in pulmonary arterial hypertension

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-05-07

  preprintOpen access

  Abstract Developmental mechanisms that precisely orchestrate cell fate and tissue architecture in organogenesis can be aberrantly reactivated to cause disease. Here we identify a previously uncharacterized population of endothelial niche cells, defined by the pioneer factor early B cell factor 1 (EBF1), that promotes both the development and pathological remodeling of pulmonary arteries (PAs). We show in the embryonic lung that the PA arises from an endothelial niche harbored within the vascular plexus, regulating Aplnr + progenitors that differentiate into arterial endothelial cells (ECs) and Ebf1 + ECs. Instead of directly incorporating into the PA endothelium, these Ebf1 + ECs secrete vasculotrophic signals that organize the proper expansion and arterialization of plexus progenitors with the recruitment of the mesenchymal cells that muscularize and ensheathe the maturing PA. Although essential in development, most Ebf1 + ECs disappear by completion of PA morphogenesis. In adult PAH, the embryonic artery-forming niche is reactivated. The normally quiescent Aplnr + general capillary stem cells re-enter the cell cycle and regenerate arterial ECs and Ebf1 + ECs. By expressing key vasculotrophic signals including Apelin, Cxcl12/Cxcr4, Notch, and Tgf-β, these Ebf1 + ECs promote angiogenesis, neo- arterialization, and neo-muscularization in a maladaptive process echoing their role in development. Our findings define a novel developmental mechanism featuring unusually transient PA organizing cells. We suggest their reemergence in adulthood drives vascular pathology and that targeting these Ebf1 + ECs could halt or reverse PAH.

  PublisherOA PDFDOI

• Efficacy of <i>Curcuma longa</i> extract in an<i> in vivo </i>model of <i>Rhizopus oryzae </i>infection in neutropenic mice

  International Journal of Pharmaceutical Sciences and Drug Research · 2025-03-30

  articleOpen access

  Mucormycosis, a life-threatening fungal infection caused by Rhizopus oryzae, disproportionately affects immunosuppressed individuals. Current treatments face significant limitations, necessitating alternative approaches. This study investigates the therapeutic potential of ethanolic extracts of Curcuma longa (C. longa), known for its antifungal and immunomodulatory properties, in a neutropenic mouse model of mucormycosis. Mice were infected with R. oryzae and treated with C. longa extracts at 100, 200 and 400 mg/kg doses. Key parameters assessed included survival rates, body weight, hematological profiles (RBC, HGB, PLT), organ weights, fungal burden in kidney and brain tissues, and immunological responses. The 400 mg/kg dose significantly improved survival rates, hematological recovery, and normalized organ weights, with reductions in fungal burden comparable to standard treatment. Mice also exhibited dose-dependent recovery in systemic health. However, incomplete recovery and some mortalities were noted, highlighting the need for further optimization. These findings suggest C. longa extracts possess antifungal, immunomodulatory, and systemic protective effects, offering a promising natural alternative for mucormycosis management in immunosuppressed hosts. Future research should focus on refining dosing strategies and establishing clinical safety and efficacy profiles.

  PublisherDOI

• Deep Learning for Predictive Forecasts and Brain Imaging in Memory Loss Diseases

  2024-09-20 · 5 citations

  article

  Because deep learning is so adept at finding patterns in complex, high-dimensional data, it has revolutionized computer vision, beating traditional machine learning models. The development of brain imaging technology led to extensive multim…ORIZATIONS: Deep learning algorithms, Alzheimer's illness, and classification This chapter presents a review of recent neuroimaging studies that used deep learning techniques to enhance AD classification and evaluation. Articles Published between January 2013 and July 2018 (searched on Pubmed and Google Scholar) Alzheimer's Disease Deep Learning Following a review of the articles, the findings were rated based on the classification technique and the kind of neuroimaging data utilized in each research. Twelve of the sixteen included research exclusively used deep learning models (CNN for eleven [5] – [14] and Bidirectional Long Short-Term Memory networks—Bi-LSTM for one study); the other four studies mixed classical machine learning with deep learning techniques [1] - [4]. The best accurate diagnosis result was obtained when these multimodal techniques were combined with biomarkers found in cerebrospinal fluid (used for further investigations). Deep learning algorithms have a lot of promise for the diagnostic criteria used to classify AD since they are getting more efficient. Understanding AD is still in its early stages, but deep learning and explainable technologies like these may help shed more light on the symptoms of the illness.

  PublisherDOI

• Physiological Modeling of the Vascularized Human Lung Organoid

  American Journal of Respiratory Cell and Molecular Biology · 2024-11-08 · 10 citations

  articleOpen access

  Human lung organoids (hLOs) derived from induced pluripotent stem cells (iPSCs) are of great interest, as they inform lung development, such as differentiation of lung epithelial subtypes in the distal alveolar unit. An unaddressed question is whether introducing endothelial cells (ECs) and vascularization provides a better representation of hLOs. Here we describe a method in which vessels become integrated with hLOs. hLOs were generated by combining human iPSC-derived lung progenitor cells (LPs) with ECs at varying LP:EC ratios. At the optimal combination of both cells, we observed vessel infiltration of hLOs compared to without ECs. Red blood cells were seen in hLOs implanted into kidney capsules of NOD/SCID mice. Both human and mouse ECs conjoined to form chimeric vessels in hLOs. The vascularized hLOs showed alveolar type II epithelial (ATII) cells and ATI cells, although there was no difference in 1:1 ATII/ATI ratio. We observed primitive airway sacs with alveolar epithelial cells lining the lumen of vascularized hLOs. Electron microscopy revealed surfactant production in ATII cells of vascularized hLOs in contrast to absence of vessels. The vascularized hLOs also mounted a robust inflammatory response characterized by influx of mouse neutrophils after challenging mice with LPS. Thus, interactions of ECs with LPs generated vascularized hLOs that induced ATII and ATI differentiation, although not reaching to the ratio of 1:9 seen in mature human lungs. hLOs also showed the LPS induced inflammatory response upon transplantation into recipient mice. Our results show the potential of vascularized hLOs for studying human lung development and inflammatory lung injury.

  PublisherOA PDFDOI

• Mapping Intra-pulmonary Tissue Neighborhoods Reveals Localized Fibrotic and Regenerative Motifs in Murine Pulmonary Fibrosis

  2024-04-30

  article
  PublisherDOI

Recent grants

• Next-Generation Genomic Imaging Technology

  NIH · $380k · 2018–2020

• Next-Generation Genomic Imaging Technology

  NIH · $380k · 2018–2020

• Identifying niche factors regulating distinct properties of AT2 stem cells

  NIH · $2.2M · 2018–2023

• NIH Grant R56HL127471

  NIH · $403k · 2017

• NIH Grant K08HL084095

  NIH · $660k · 2012

Frequent coauthors

• Sukki Cho

  Seoul National University Bundang Hospital

  37 shared

• Nam‐Hyuk Cho

  Seoul National University

  37 shared

• Jin-Haeng Chung

  Seoul National University Bundang Hospital

  37 shared

• Na‐Young Ha

  Seoul National University

  37 shared

• Mark A. Krasnow

  Howard Hughes Medical Institute

  21 shared

• Laure‐Emmanuelle Zaragosi

  Institut de Pharmacologie Moléculaire et Cellulaire

  18 shared

• Martijn C. Nawijn

  University Medical Center Groningen

  16 shared

• Rokhyun Kim

  Seoul National University

  13 shared

Labs

• Desai labPI

Awards & honors

• Elected Member, American Society for Clinical Investigation…
• Lung Force Gala Honoree, American Lung Association (2018)
• Woods Family Endowed Faculty Scholar in Pediatric Translatio…
• Stanford Medical Student Teaching Recognition, Stanford Univ…
• Robert Dawson Evans Fellow Excellence in Teaching Award, Bos…