Current Research Description
PROJECT 1: Determine the mechanisms by which mitochondrial homeostasis is deregulated under neuropathological conditions including Parkinson’s disease
Although it may not be the direct cause, mitochondrial dysfunction is a widespread phenomenon observed in almost all the neurodegenerative diseases. One major project in the lab is investigating the role of JNK3 and Pin1 in regulation of mitochondrial homeostasis.
Recently, we have discovered that JNK3 and Pin1 regulate cytochrome C release, a hallmark of apoptosis following spinal cord injury . We have used spinal cord injury as an initial model of neuropathology, since it allows a clearly defined temporal point when an apoptotic insult is delivered, unlike neurodegenerative diseases that develop progressively over an extended period to time. Following spinal cord injury, we found that JNK3 was the predominant JNK isoform whose activity increased dramatically with the injury, and its proapoptotic action involved facilitating the rapid degradation of a Bcl2 family member, Mcl-1. Mcl-1 proteins normally remain elevated by Pin1, which protects it from being ubiquitinated by direct binding. It is our hypothesis that Pin1 targets other proteins besides Mcl-1, thereby maintaining mitochondrial homeostasis under neurodegenerative conditions. At the onset of neurodegeneration, JNK3 or p38 is activated, translating various cellular and extracellular insults to cell death machinery inside the cell. Indeed, we have discovered that Pin1 binds HtrA2, a mitochondrial protease, which plays a critical role in Parkinson’s disease based on genetic studies [2-4]. HtrA2 itself is hyperphosphorylated in Parkinson’s patients, suggesting phosphorylation of HtrA2 plays a role in regulating its activity. We are currently employing MPTP model of Parkinson’s disease in Pin1 knockout mice to understand the role of Pin1 in Parkinson’s disease.
PROJECT 2: Determine the role of JNK3 in Alzheimer’s disease
One of the surprising findings in our spinal cord injury study was that while oligodendrocytes readily succumbed to JNK3-mediated apoptosis within a couple of days after spinal cord injury, apoptosis of motor neurons was unaffected in JNK3 knockout mice, even though JNK3 was activated among them . We hypothesize that in neurons, JNK3 induces autophagy rather than apoptosis by perturbing axonal transport.
We have discovered in JNK3-/- mice, a greater amount of APP is transported anterogradely in the sciatic nerve when measured at different times after ligation lesioning. We chose sciatic nerve lesioning as an initial paradigm, since JNK3 is activated after ligation injury (our unpublished data) and APP trafficking is easier to detect in the absence of neuronal cell bodies as it was reported initially . The observed increase in APP trafficking was accompanied by a decrease in Thr668 phosphorylation in JNK3-/- mice, suggesting that although JNK3 does not phosphorylate Thr668 in APP directly, it regulates its phosphorylation. Thr668 phosphorylation is believed to regulate BACE1-dependent cleavage . We therefore hypothesize that JNK3 plays a role in APP processing by regulating the normal trafficking of APP in the axons. We are currently testing whether deletion of JNK3 affects plaque formation and A?40/42 generation by crossing JNK3-/- mice with 5X-FAD mice, which express the mutant APP (Swe/F1/Lon) and mutant PS1 (M146L/L286V) under Thy1 promoter . We will also characterize the effect of JNK3 in the trafficking of APP-YFP, BACE1-GFP, and PS1-RFP by introducing them into hippocampal neurons from the JNK3+/+ and JNK3-/- mice. The transport rate of these fluorescent proteins will be analyzed using microfluidic chambers that separate the neuronal cell bodies from their processes.
PROJECT 3: Epigenetic regulation of cell death
The Pin1 binding site, p(S/T)-P motif, is the core consensus sequence for the proline-directed kinases, such as the MAP, GSK and CDK kinase families. Having discovered that an intricate interplay between JNK3 and Pin1 regulates cytochrome C release in vivo, which clearly applies to our spinal cord injury paradigm, we believe it likely that the observed dichotomous interaction between JNK3 and Pin1 occurs more widely. Our study established that JNK3 is the principle kinase that is activated in response to an insult to the CNS, and other reports that relied on knockout analyses similarly concluded that JNK3 is involved in initiating the apoptotic program under a wide range of pathological conditions, such as ischemia [8, 9], axotomy, [10-12], seizure , and in animal models of neurodegenerative diseases [14, 15]. We believe that identifying the in vivo targets of JNK3 that are oppositely regulated by Pin1 will be critical in understanding the fundamental mechanisms underlying neurodegeneration.
After injury, we detected JNK3 activity not only in the cytosol but also in the nucleus. Given the fact that many Pin1 binding partners are nuclear, we seek to identify new nuclear targets that are oppositely regulated by JNK3 and Pin1. The rationale is that although genetic components are clearly involved, they typically account for small fractions of affected individuals in many neurodegenerative diseases, suggesting that epigenetic mechanisms are likely to be involved. In support, a related MAP kinase, p38, was shown to phosphorylate BAF60, thereby influencing chromatin modification during muscle development , and we ourselves found that phosphorylation of H2A.x and H2B is altered in JNK3-/- mice.
We have performed chromatin-immunoprecipitation (ChIP) based screening ,, using Pin1 antibody in ChIP assays using the chromatins taken from the injured spinal cord of the JNK3 wild type and knockout mice. The resulting DNA fragments from ChIP assays were used to screen Affimetrix GeneChip Tiling Array sets. We found many genes that are known to be critical in neurodegenerative diseases, such as PS1, Bace, APP, and Parkin, in addition to well known Pin1 targets, such as p53. We have begun the yeast-1-hybrid screen to identify the proteins that bind the regulatory regions of these genes and characterize them biochemically as well as functionally. We are also repeating the ChIP on chip screen using JNK(3) antibody to discover the targets that are regulated by JNK3.
We believe the connection that we have discovered between JNK3 and Pin1 in mitochondrial dysfunction is novel, which is supported by the genes that we identified in the first Pin1 ChIP on chip screen. We would like to expand this study to the nucleus and identify new targets for JNK3 and Pin1. Since the JNK pathway is clearly involved in many neurodegenerative diseases, our hope is that the approach will lead us to unveiling novel targets that bear wide implications in these diseases as well as cancer. We believe that the more we delineate the mechanisms that are fundamental to this group of disease processes, the wider will be the scope of potential routes for therapeutic intervention.
PROJECT 4: Determine the mechanisms by which JNK3 is activated
Activation of the JNK pathway has been demonstrated in many neurodegenerative models as well as in actual cases of neurodegenerative diseases. It is our hypothesis that JNK3 activation is one of the necessary events that must occur prior to the onset of neurodegeneration. We discovered that cytochrome C is constitutively released in the spinal cord of Pin1 knockout mice without any injury . The level of cytochrome C that is constitutively released is similar to that released in the wild type mice that have undergone spinal cord injury, but neuronal apoptosis was not detected until mice were subjected to injury. These results suggest that cytochrome C release alone is not sufficient, and additional stimuli have to be incurred for cells to die. Our preliminary data suggest that one of the critical additional events is JNK3 activation, since in double knockout of JNK3 and Pin1, cytochrome C levels in the cytosol returned to levels close to those found in JNK3-/- after injury, although the basal level of cytochrome C release was not altered. Our goal is therefore to determine the mechanisms by which JNK3 is activated in vivo using spinal cord injury as an initial model but expand the findings to mouse models of Alzheimer’s, Parkinson’s, ALS, and Huntington’s diseases.
We have recently published that ApoER2 is necessary for axotomy-induced cell death in the adult cortex . Among the signaling pathways that ApoER2 activates, its ability to recruit JIP and JNK3 and not Dab, appears to be important for its ability to induce cell death . In the cortical axotomy model, deletion of p75 and JNK3 inhibits neuronal death ,, suggesting a functional interaction among ApoER2, p75 and JNK3. In line with this notion, we found that JIP1 mediates interaction between the p75 and ApoER2 in vitro. As JIP1 binds JNK3 , we believe that interaction between these two receptors is responsible for activating JNK3 in axotomy models of the cortex and the spinal cord.
We are breeding p75-/- mice with a knockin mouse line of ApoER2, in which the JIP binding domain is deleted . Using axotomy models of the double knockout mice, we will address whether the formation of a complex between p75 and ApoER2 is responsible for activating JNK3. In addition, we will determine whether the interaction is dependent on proNGF and/or ApoE4. This project is conducted in collaboration with Dr. Joachim Herz at UTSW.
PROJECT 5: A translational approach using a small molecule that disrupts proNGF-p75 interaction
For comprehensive functional recovery after spinal cord injury, one needs to minimize the death of affected neurons and myelinating oligodendrocytes. Using our biochemical data that suggest that proNGF binding to p75 is necessary for apoptosis of oligodendrocytes, we sought to minimize the loss of oligodendrocytes after injury by blocking p75 activation from the outset, where we disrupt the ability of proNGF to initiate p75-mediated death. Toward this end, Dr. Frank Longo at Stanford and I have developed and characterized a small monomeric compound, C31, that not only blocks proNGF-mediated oligodendrocyte apoptosis in vitro, but also crosses the blood brain barrier following intraperitoneal (IP) administration. C31 is especially well suited for such CNS therapeutics, since the ratio of the compound detected in the brain vs plasma is greater than 5.0. We thus plan to determine whether the post-injury delivery of the compound via IP injection will improve functional recovery. Our pilot study already suggests that even at a lower dose than that are required for inhibiting cytC release, C31 shows evidence of promoting functional recovery without any noticeable toxicity. We therefore propose to test the efficacy of C31 in a full-scale trial as a way to promote functional recovery of injured mice as a U01 in the future.