Member (Student), Cardiovascular Institute
Education & Certifications
Bachelor of Arts, Harvard University, Biomedical Engineering (2016)
Statins improve endothelial function via suppression of epigenetic-driven EndMT.
Nature cardiovascular research
2023; 2 (5): 467-485
The pleiotropic benefits of statins in cardiovascular diseases that are independent of their lipid-lowering effects have been well documented, but the underlying mechanisms remain elusive. Here we show that simvastatin significantly improves human induced pluripotent stem cell-derived endothelial cell functions in both baseline and diabetic conditions by reducing chromatin accessibility at transcriptional enhanced associate domain elements and ultimately at endothelial-to-mesenchymal transition (EndMT)-regulating genes in a yes-associated protein (YAP)-dependent manner. Inhibition of geranylgeranyltransferase (GGTase) I, a mevalonate pathway intermediate, repressed YAP nuclear translocation and YAP activity via RhoA signaling antagonism. We further identified a previously undescribed SOX9 enhancer downstream of statin-YAP signaling that promotes the EndMT process. Thus, inhibition of any component of the GGTase-RhoA-YAP-SRY box transcription factor 9 (SOX9) signaling axis was shown to rescue EndMT-associated endothelial dysfunction both in vitro and in vivo, especially under diabetic conditions. Overall, our study reveals an epigenetic modulatory role for simvastatin in repressing EndMT to confer protection against endothelial dysfunction.
View details for DOI 10.1038/s44161-023-00267-1
View details for PubMedID 37693816
View details for PubMedCentralID PMC10489108
Leveraging Physiology and Artificial Intelligence to Deliver Advancements in Healthcare.
Artificial Intelligence (AI) in healthcare has generated remarkable innovation and progress in the last decade. Significant advancements can be attributed to the utilization of AI to transform physiology data to advance healthcare. In this review, we will explore how past work has shaped the field and defined future challenges and directions. In particular, we focus on three areas of development. First, we give an overview of AI, with special attention to the most relevant AI models. We then detail how physiology data has been harnessed by AI to advance the main areas of healthcare such as automating existing healthcare tasks, increasing access to care, and augmenting healthcare capabilities. Finally, we discuss emerging concerns surrounding the use of individual physiology data and detail an increasingly important consideration for the field, namely the challenges of deploying AI models to achieve meaningful clinical impact.
View details for DOI 10.1152/physrev.00033.2022
View details for PubMedID 37104717
From patterns to patients: Advances in clinical machine learning for cancer diagnosis, prognosis, and treatment.
Machine learning (ML) is increasingly used in clinical oncology to diagnose cancers, predict patient outcomes, and inform treatment planning. Here, we review recent applications of ML across the clinical oncology workflow. We review how these techniques are applied to medical imaging and to molecular data obtained from liquid and solid tumor biopsies for cancer diagnosis, prognosis, and treatment design. We discuss key considerations in developing ML for the distinct challenges posed by imaging and molecular data. Finally, we examine ML models approved for cancer-related patient usage by regulatory agencies and discuss approaches to improve the clinical usefulness of ML.
View details for DOI 10.1016/j.cell.2023.01.035
View details for PubMedID 36905928
SGLT2 inhibitor ameliorates endothelial dysfunction associated with the common ALDH2 alcohol flushing variant.
Science translational medicine
2023; 15 (680): eabp9952
The common aldehyde dehydrogenase 2 (ALDH2) alcohol flushing variant known as ALDH2*2 affects ∼8% of the world's population. Even in heterozygous carriers, this missense variant leads to a severe loss of ALDH2 enzymatic activity and has been linked to an increased risk of coronary artery disease (CAD). Endothelial cell (EC) dysfunction plays a determining role in all stages of CAD pathogenesis, including early-onset CAD. However, the contribution of ALDH2*2 to EC dysfunction and its relation to CAD are not fully understood. In a large genome-wide association study (GWAS) from Biobank Japan, ALDH2*2 was found to be one of the strongest single-nucleotide polymorphisms associated with CAD. Clinical assessment of endothelial function showed that human participants carrying ALDH2*2 exhibited impaired vasodilation after light alcohol drinking. Using human induced pluripotent stem cell-derived ECs (iPSC-ECs) and CRISPR-Cas9-corrected ALDH2*2 iPSC-ECs, we modeled ALDH2*2-induced EC dysfunction in vitro, demonstrating an increase in oxidative stress and inflammatory markers and a decrease in nitric oxide (NO) production and tube formation capacity, which was further exacerbated by ethanol exposure. We subsequently found that sodium-glucose cotransporter 2 inhibitors (SGLT2i) such as empagliflozin mitigated ALDH2*2-associated EC dysfunction. Studies in ALDH2*2 knock-in mice further demonstrated that empagliflozin attenuated ALDH2*2-mediated vascular dysfunction in vivo. Mechanistically, empagliflozin inhibited Na+/H+-exchanger 1 (NHE-1) and activated AKT kinase and endothelial NO synthase (eNOS) pathways to ameliorate ALDH2*2-induced EC dysfunction. Together, our results suggest that ALDH2*2 induces EC dysfunction and that SGLT2i may potentially be used as a preventative measure against CAD for ALDH2*2 carriers.
View details for DOI 10.1126/scitranslmed.abp9952
View details for PubMedID 36696485
Current and future perspectives of single-cell multi-omics technologies in cardiovascular research
Nat Cardiovasc Res
View details for DOI 10.1038/s44161-022-00205-7
Shifting machine learning for healthcare from development to deployment and from models to data.
Nature biomedical engineering
In the past decade, the application of machine learning (ML) to healthcare has helped drive the automation of physician tasks as well as enhancements in clinical capabilities and access to care. This progress has emphasized that, from model development to model deployment, data play central roles. In this Review, we provide a data-centric view of the innovations and challenges that are defining ML for healthcare. We discuss deep generative models and federated learning as strategies to augment datasets for improved model performance, as well as the use of the more recent transformer models for handling larger datasets and enhancing the modelling of clinical text. We also discuss data-focused problems in the deployment of ML, emphasizing the need to efficiently deliver data to ML models for timely clinical predictions and to account for natural data shifts that can deteriorate model performance.
View details for DOI 10.1038/s41551-022-00898-y
View details for PubMedID 35788685
Cannabinoid receptor 1 antagonist genistein attenuates marijuana-induced vascular inflammation.
Epidemiological studies reveal that marijuana increases the risk of cardiovascular disease (CVD); however, little is known about the mechanism. Δ9-tetrahydrocannabinol (Δ9-THC), the psychoactive component of marijuana, binds to cannabinoid receptor 1 (CB1/CNR1) in the vasculature and is implicated in CVD. A UK Biobank analysis found that cannabis was an risk factor for CVD. We found that marijuana smoking activated inflammatory cytokines implicated in CVD. In silico virtual screening identified genistein, a soybean isoflavone, as a putative CB1 antagonist. Human-induced pluripotent stem cell-derived endothelial cells were used to model Δ9-THC-induced inflammation and oxidative stress via NF-κB signaling. Knockdown of the CB1 receptor with siRNA, CRISPR interference, and genistein attenuated the effects of Δ9-THC. In mice, genistein blocked Δ9-THC-induced endothelial dysfunction in wire myograph, reduced atherosclerotic plaque, and had minimal penetration of the central nervous system. Genistein is a CB1 antagonist that attenuates Δ9-THC-induced atherosclerosis.
View details for DOI 10.1016/j.cell.2022.04.005
View details for PubMedID 35489334
Treating Duchenne Muscular Dystrophy: The Promise of Stem Cells, Artificial Intelligence, and Multi-Omics.
Frontiers in cardiovascular medicine
2022; 9: 851491
Muscular dystrophies are chronic and debilitating disorders caused by progressive muscle wasting. Duchenne muscular dystrophy (DMD) is the most common type. DMD is a well-characterized genetic disorder caused by the absence of dystrophin. Although some therapies exist to treat the symptoms and there are ongoing efforts to correct the underlying molecular defect, patients with muscular dystrophies would greatly benefit from new therapies that target the specific pathways contributing directly to the muscle disorders. Three new advances are poised to change the landscape of therapies for muscular dystrophies such as DMD. First, the advent of human induced pluripotent stem cells (iPSCs) allows researchers to design effective treatment strategies that make up for the gaps missed by conventional "one size fits all" strategies. By characterizing tissue alterations with single-cell resolution and having molecular profiles for therapeutic treatments for a variety of cell types, clinical researchers can design multi-pronged interventions to not just delay degenerative processes, but regenerate healthy tissues. Second, artificial intelligence (AI) will play a significant role in developing future therapies by allowing the aggregation and synthesis of large and disparate datasets to help reveal underlying molecular mechanisms. Third, disease models using a high volume of multi-omics data gathered from diverse sources carry valuable information about converging and diverging pathways. Using these new tools, the results of previous and emerging studies will catalyze precision medicine-based drug development that can tackle devastating disorders such as DMD.
View details for DOI 10.3389/fcvm.2022.851491
View details for PubMedID 35360042
Antitumor effects of iPSC-based cancer vaccine in pancreatic cancer.
Stem cell reports
Induced pluripotent stem cells (iPSCs) and cancer cells share cellular similarities and transcriptomic profiles. Here, we show that an iPSC-based cancer vaccine, comprised of autologous iPSCs and CpG, stimulated cytotoxic antitumor CD8+ Tcell effector and memory responses, induced cancer-specific humoral immune responses, reduced immunosuppressive CD4+ T regulatory cells, and prevented tumor formation in 75% of pancreatic ductal adenocarcinoma (PDAC) mice. We demonstrate that shared gene expression profiles of "iPSC-cancer signature genes" and others are overexpressed in mouse and human iPSC lines, PDAC cells, and multiple human solid tumor types compared with normal tissues. These results support further studies of iPSC vaccination in PDAC in preclinical and clinical models and in other cancer types that have low mutational burdens.
View details for DOI 10.1016/j.stemcr.2021.04.004
View details for PubMedID 33961792
Race and Genetics in Congenital Heart Disease: Application of iPSCs, Omics, and Machine Learning Technologies.
Frontiers in cardiovascular medicine
2021; 8: 635280
Congenital heart disease (CHD) is a multifaceted cardiovascular anomaly that occurs when there are structural abnormalities in the heart before birth. Although various risk factors are known to influence the development of this disease, a full comprehension of the etiology and treatment for different patient populations remains elusive. For instance, racial minorities are disproportionally affected by this disease and typically have worse prognosis, possibly due to environmental and genetic disparities. Although research into CHD has highlighted a wide range of causal factors, the reasons for these differences seen in different patient populations are not fully known. Cardiovascular disease modeling using induced pluripotent stem cells (iPSCs) is a novel approach for investigating possible genetic variants in CHD that may be race specific, making it a valuable tool to help solve the mystery of higher incidence and mortality rates among minorities. Herein, we first review the prevalence, risk factors, and genetics of CHD and then discuss the use of iPSCs, omics, and machine learning technologies to investigate the etiology of CHD and its connection to racial disparities. We also explore the translational potential of iPSC-based disease modeling combined with genome editing and high throughput drug screening platforms.
View details for DOI 10.3389/fcvm.2021.635280
View details for PubMedID 33681306
View details for PubMedCentralID PMC7925393
Universal intracellular biomolecule delivery with precise dosage control
2018; 4 (10): eaat8131
Intracellular delivery of mRNA, DNA, and other large macromolecules into cells plays an essential role in an array of biological research and clinical therapies. However, current methods yield a wide variation in the amount of material delivered, as well as limitations on the cell types and cargoes possible. Here, we demonstrate quantitatively controlled delivery into a range of primary cells and cell lines with a tight dosage distribution using a nanostraw-electroporation system (NES). In NES, cells are cultured onto track-etched membranes with protruding nanostraws that connect to the fluidic environment beneath the membrane. The tight cell-nanostraw interface focuses applied electric fields to the cell membrane, enabling low-voltage and nondamaging local poration of the cell membrane. Concurrently, the field electrophoretically injects biomolecular cargoes through the nanostraws and into the cell at the same location. We show that the amount of material delivered is precisely controlled by the applied voltage, delivery duration, and reagent concentration. NES is highly effective even for primary cell types or different cell densities, is largely cargo agnostic, and can simultaneously deliver specific ratios of different molecules. Using a simple cell culture well format, the NES delivers into >100,000 cells within 20 s with >95% cell viability, enabling facile, dosage-controlled intracellular delivery for a wide variety of biological applications.
View details for PubMedID 30402539
Harnessing cell pluripotency for cardiovascular regenerative medicine.
Nature biomedical engineering
2018; 2 (6): 392-398
Human pluripotent stem cells (hPSCs), in particular embryonic stem cells and induced pluripotent stem cells, have received enormous attention in cardiovascular regenerative medicine owing to their ability to expand and differentiate into functional cardiomyocytes and other cardiovascular cell types. Despite the potential applications of hPSCs for tissue regeneration in patients suffering from cardiovascular disease, whether hPSC-based therapies can be safe and efficacious remains inconclusive, with strong evidence from clinical trials lacking. Critical factors limiting therapeutic efficacy are the degree of maturity and purity of the hPSC-derived differentiated progeny, and the tumorigenic risk associated with residual undifferentiated cells. In this Review, we discuss recent advances in cardiac-cell differentiation from hPSCs and in the direct reprogramming of non-myocyte cells for cardiovascular regenerative applications. We also discuss approaches for the delivery of cells to diseased tissue, and how such advances are contributing to progress in cardiac tissue engineering for tackling heart disease.
View details for DOI 10.1038/s41551-018-0244-8
View details for PubMedID 31011193
A Rapid, High-Quality, Cost-Effective, Comprehensive and Expandable Targeted Next-Generation Sequencing Assay for Inherited Heart Diseases.
2015; 117 (7): 603-611
Thousands of mutations across >50 genes have been implicated in inherited cardiomyopathies. However, options for sequencing this rapidly evolving gene set are limited because many sequencing services and off-the-shelf kits suffer from slow turnaround, inefficient capture of genomic DNA, and high cost. Furthermore, customization of these assays to cover emerging targets that suit individual needs is often expensive and time consuming.We sought to develop a custom high throughput, clinical-grade next-generation sequencing assay for detecting cardiac disease gene mutations with improved accuracy, flexibility, turnaround, and cost.We used double-stranded probes (complementary long padlock probes), an inexpensive and customizable capture technology, to efficiently capture and amplify the entire coding region and flanking intronic and regulatory sequences of 88 genes and 40 microRNAs associated with inherited cardiomyopathies, congenital heart disease, and cardiac development. Multiplexing 11 samples per sequencing run resulted in a mean base pair coverage of 420, of which 97% had >20× coverage and >99% were concordant with known heterozygous single nucleotide polymorphisms. The assay correctly detected germline variants in 24 individuals and revealed several polymorphic regions in miR-499. Total run time was 3 days at an approximate cost of $100 per sample.Accurate, high-throughput detection of mutations across numerous cardiac genes is achievable with complementary long padlock probe technology. Moreover, this format allows facile insertion of additional probes as more cardiomyopathy and congenital heart disease genes are discovered, giving researchers a powerful new tool for DNA mutation detection and discovery.
View details for DOI 10.1161/CIRCRESAHA.115.306723
View details for PubMedID 26265630
DWI for Renal Mass Characterization: Systematic Review and Meta-Analysis of Diagnostic Test Performance
AMERICAN JOURNAL OF ROENTGENOLOGY
2015; 205 (2): 317-324
The objective of our study was to perform a systematic review and meta-analysis of the test performance of DWI in the characterization of renal masses.We performed searches of three electronic databases for studies on renal mass characterization using DWI. Methodologic quality was assessed for each study. We quantitatively analyzed test performance for three clinical problems: first, benign versus malignant lesions; second, clear cell renal cell carcinoma (RCC) versus other malignancies; and, third, high-versus low-grade clear cell RCCs. We summarized performance as a single pair of sensitivity and specificity values or a summary ROC curve.The studies in the literature were limited in both quantity and quality. For classification of benign versus malignant lesions, four studies with 279 lesions yielded a single summary estimate of 86% sensitivity and 78% specificity. For differentiation of clear cell RCC from other malignancies, five studies showed marked heterogeneity not conducive to meta-analysis. For differentiation of high-from low-grade clear cell RCCs, three studies with 110 lesions showed a threshold effect appropriate for summary ROC construction: The AUC was 0.83.Evidence suggests moderate accuracy of DWI for the prediction of malignancy and high-grade clear cell cancers, whereas DWI performance for ascertaining clear cell histologic grade remains unclear. To develop DWI as a noninvasive approach for the evaluation of solid renal masses, prospective studies with standardized test parameters are needed to better establish DWI performance and its impact on patient outcomes.
View details for DOI 10.2214/AJR.14.13930
View details for Web of Science ID 000358436000030
View details for PubMedID 26204281
Functional MnO nanoclusters for efficient siRNA delivery
2011; 47 (44): 12152-12154
A non-viral gene delivery nanovehicle based on Alkyl-PEI2k capped MnO nanoclusters was synthesized via a simple, facile method and used for efficient siRNA delivery and magnetic resonance imaging.
View details for DOI 10.1039/c1cc15408g
View details for Web of Science ID 000296342800035
View details for PubMedID 21991584
View details for PubMedCentralID PMC4620662