Drug development for the treatment of diseases in the central nervous system (CNS) is a long and costly process with a disproportionately lower chance of yielding successful drugs. Many drugs fail to reach the target in the brain in appreciable concentrations due to the blood brain barrier (BBB), a physiological barrier to efficient drug delivery due to its restriction of paracellular and transcellular molecular transport. Although mechanisms have been proposed to facilitate drug transport across the BBB without permanently compromising BBB integrity, there are a few existing therapeutics capable of delivering sufficient doses of drugs to the brain parenchyma. One major challenge is the inability to conduct a mechanistic study on the interaction of drug candidates with the BBB at molecular and cellular levels in the intact human brain or using animal models that cannot accurately recapitulate human diseases. This challenge highlights the importance of the development of in vitro models that mimic pathophysiological conditions of the human BBB.

To address this challenge, we develop a microengineered 'human neurovascular unit on a chip that makes it possible to achieve the structural and functional complexity of the neurovascular unit. We leverage advanced microengineering technologies to present cultured cells with controlled mechanical and biochemical cues with physiological relevance, leading to the development of microengineered biomimetic systems called ‘organ-on-a-chip technology’ that may simulate complex organ-level physiology. The innovative feature of our human NVU-on-a-chip 'hNVUoC’ system is the ability to incorporate critical elements including direct cell-cell interactions, shear flow, and bioinspired ECM that have been overlooked in existing in vitro models. We employ high throughput qPCR techniques to simultaneously analyze the expression of 81 BBB relevant endothelial specializations such as junctional proteins (ZO-1, Claudins, JAMs, etc.), specialized transporters (GLUT-1, CAT1, TfR, etc.) and drug resistant proteins (Pgp, ABCC1, ABCC4, etc.). Our findings indicate that our hNVUoC system provides a marked increase in physiological relevance relative to transwell culture systems.  

The overarching goal of our research is to create a representative example of innovative multidisciplinary engineering approaches to the development of good predictive models of the human neurovascular unit. Our strategies hold great potential to address the critical need recently identified by government regulatory agencies and virtually all major pharmaceutical companies for in vitro models of complex human diseases that can reliably predict drug efficacy and safety in humans. Given the increased awareness on the importance of physiological accurate model of the neurovascular unit and its applications to develop new therapeutics for complex CNS diseases, our approaches will contribute to high-throughput screening and analysis of novel drug candidates, having a broad impact on nanomedicine area for rapid clinical development. Our ultimate vision is to leverage these technological advances to achieve transformative breakthroughs that enable cost-effective development of therapeutic compounds for the prevention and treatment of CNS diseases.

We are currently evaluating therapeutic and diagnostic nanoparticles (nanomedicines) engineered for the treatment of CNS diseases including pediatric brain tumors and neurodegenerative disorders.

In collaboration with Prof. Tobey MacDonald at Emory University, we are developing theranostic nanoparticles that can cross the BBB and selectively bind to target tumor sites in medulloblastoma to examine targeted delivery of a sonic hedgehog inhibitor for the study of medulloblastoma therapeutics. Radiation and chemotherapies of brain tumors lead to adverse effects, getting associated with enormous health care costs. The proposed work will provide the unique approach for the development of a new drug delivery platform and its screening in a more physiologically relevant in vitro model of the human neurovascular unit for the detection and treatment of medulloblastoma, the most common malignant brain tumor in children. Advanced screening and treatment and potential derivatives will potentially improve human health and reduce overall costs associated with medulloblastoma and other brain tumors.

In collaboration with Dr. Allan Levey, Dr. Srikant Rangaraju, we are engineering biomimetic nanomaterials for targeted delivery of ion channel inhibitors and for enhanced clearance of amyloid beta. According to the Alzheimer’s Association, over 5.4 million Americans suffer from Alzheimer’s disease, and the cost of caring for AD is estimated to total $236 billion in 2016. Abnormal accumulation of amyloid beta, a peptide produced in neuronal cells, plays a key role in both early and late forms of AD, leading to the neuronal injury and synaptic dysfunction. Despite the debate about the role of amyloid beta in AD pathogenesis, substantial research supports the hypothesis that an imbalance between production and clearance of amyloid beta causes excess amyloid beta accumulation in the CNS, even as a very early, often initiating factor in AD. Recent work also showed that while increased amyloid beta production seems to be a central mechanism in rare, familial cases of early-onset AD due to mutations in amyloid precursor protein (APP) and presenilin (PSEN), reduced amyloid beta clearance appears to be the primary mechanism involved the commonly occurring form of late-onset sporadic AD. With no current treatment for AD, there is a large unmet need for the development of new therapeutics including enhanced amyloid beta clearance that improves the prognosis for AD patients.