Current Research Description
Drug discovery for Alzheimer's disease with JNK3-selective, brain penetrating small molecule(s) that are orally bioavailable: Our data suggest a model wherein JNK3 is cyclically activated as the primary summing node of multiple positive feedback paths in AÎ² peptide signaling. Initially, JNK3 phosphorylates APP to induce rapid endocytosis and processing to produce pathological AÎ²42. AÎ²42 then induces translational block by activating AMP-activated protein kinase (AMPK), which leads to ER stress, which further activates JNK3, thereby establishing a positive feedback loop. These data suggest that blocking JNK3 activation at any point can potentially halt or slow the subsequent progression of the disease. We are currently developing an orally bioavailable JNK3-selective small molecule inhibitor that can cross the blood brain barrier efficiently.
Understanding the role of metabolic dysfunction in Alzheimer's disease pathology: Another important finding in Yoon et al. (3) was that pathologic amyloid beta disrupts metabolic homeostasis in healthy neurons in culture. In further support, we found that new protein synthesis is significantly reduced in 5XFAD mice compared to controls, and the extent of reduction becomes greater upon treating mice with a high-fat diet. The reduction in new protein synthesis is the greatest in the hippocampus, suggesting that metabolic dysfunction may precede or correlate with cognitive decline.
(A) Identification of metabolic biomarkers for Alzheimer's disease and development of lab-on-a-chip for their detection: Our preliminary longitudinal metabolomics study using serum and urine from 5XFAD mice identified a potential dysfunction in the glycolytic pathway at the disease onset. In support, we found aberrant activation of pyruvate kinase M2 (PKM2) in microglia in these mice. Our goal is to identify aging- or AD-specific metabolites, validate them using human serum or CSF samples, and develop a lab-on-a-chip to detect the metabolite. This project is in collaboration with a chemist in metabolomics study and a biomedical engineer.
(B) Role of PKM2 activation in Alzheimer's disease: We found that PKM2 is mainly activated among microglia in 5XFAD mice. We are examining the role of PKM2 activation in microglia in AD mice using conditional knockout strategies.
(C) Understanding the connection between AD and another metabolic disease, type 2 diabetes (T2D): It is known that patients with type 2 diabetes (T2D) are twice as likely to develop Alzheimer's disease when old, suggesting there may be a common mechanism that contributes to Alzheimer's disease pathology. Although it is possible that T2D-associated pathology merely exacerbates a mechanism unique to Alzheimer's disease, the possibility is enticing that one or more common mechanisms exist, while none is known to date.
T2D is characterized as exhibiting resistance to both insulin and leptin with hyperglycemia. We found that 5XFAD mice develop leptin but not insulin resistance by 6 months of age even on normal chow. They do develop insulin resistance at this age if fed a high-fat diet. These results suggest that a mechanism that underlies leptin resistance may be shared between Alzheimer's disease and T2D at least in mouse models. A possible candidate is the mechanism by which AMPK is activated, while leptin inhibits it in the hypothalamus.
Along these lines, we discovered that JNK3 might be involved in regulating AMPK activation in the hypothalamus. Deleting JNK3 from 5XFAD mice restored leptin sensitivity at 6 month and AMPK activity is significantly reduced in JNK3 null mice: ghrelin failed to activate hypothalamic AMPK and AMPK-dependent phosphorylation of Raptor, an AMPK substrate, was significantly reduced in JNK3 null mice. In further support, we found that JNK3 phosphorylates the Î± subunit of AMPK at S356 in vitro. We have generated S356A knock-in mice both in AMPKÎ± and Î±2, and are in the process of analyzing their phenotype.
Role of proNGF and p75 signaling in loss of bladder control after spinal cord injury: Loss of bladder control is one of the most challenging outcomes facing spinal cord injured patients, with no drug treatments available at the present time. NGF has long been implicated in the development of bladder dysfunction under uropathological conditions and after spinal cord injury, being detected in urines. We discovered that it is in fact proNGF that is detected in urine after spinal cord injury in mammals as well as in humans. The urinary proNGF appears to disrupt the lumen lining cells by activating p75 on their surface, which contributes positively to overall micturition. When we blocked proNGF-p75 signaling after spinal cord injury with oral delivery of a small molecule, LM11A-31, that crosses the blood-brain/spinal cord barrier efficiently, the bladder function improved dramatically: The drug treated mice acquired spontaneous micturition weeks earlier than the control with attenuated hyperreflexia and normal bladder pressure. Since proNGF is the main NGF released in the CNS after spinal cord injury, and p75 is expressed in the main bladder circuits that are responsible for reflex voiding, we seek to decipher the mechanisms by which proNGF-p75 signaling contributes to micturition circuit after spinal cord injury.
Novel role of p75 in nociception: Loss of p75 neurotrophin receptor in global knockout mice (p75KO) results in reduced heat sensitivity. This phenotype has been mainly attributed to a ~30% loss of sensory neurons from the dorsal root ganglia (DRG) during development, which was supported by a 2-3 fold decrease in NGF sensitivity of p75KO sensory neurons, when they are cultured in isolation without Schwann cells. In addition, the p75KO mice exhibit hypomyelination in the periphery, having thinner myelin sheaths around the axons. This phenotype is presumed to be the result of losing the receptor in Schwann cells. P75 is, however, expressed by both sensory neurons and Schwann cells; to better understand the role of p75 in these two cell types, we generated conditional knockout mice lacking p75 in Schwann cells or in neurons.
Upon deleting p75 selectively in sensory neurons (Thy1-cre:p75fl/fl) or Schwann cells (Dhh-cre: p75fl/fl), we found no significant difference in myelin thickness, suggesting that the hypomyelination is likely the result of losing p75 in both cell types. Surprisingly, we also discovered that p75 in Schwann cells contributes to nociceptive behavior: Dhh-cre: p75fl/fl mice exhibited decreased heat sensitivity to the same extent as Thy1-cre:p75fl/fl mice. These unexpected results are reminiscent of a report that demonstrated nociceptive defects due to a loss of sensory neurons, upon blocking ErbB receptor signaling or FGF receptor signaling in non-myelinating Schwann cells. Indeed, our preliminary data illustrate that BDNF addition leads to activation of ErbB2 in Schwann cells, primarily due to a novel interaction between p75 and ErbB2. Furthermore, there was a 30% loss of sensory neurons in adult Dhh-cre: p75fl/fl mice. These results reveal a novel role for p75 in regulating sensory neuron survival through its function in Schwann cells.