K Ulrich Ulli Bayer - US grants

DESCRIPTION (provided by applicant): The Ca +/calmodulin-dependent protein kinase II (CaMKII) is an important mediator of cellular Ca + signals. CaMKII is required for forms of learning and memory and is involved in regulation of dendritic morphology, synapse number and synaptic strength. CaMKII has also been implicated in cell cycle control, insulin secretion, and in ischemia-induced neuronal cell death. Many studies of CaMKII have relied on the pharmacological inhibitor KN93. However, KN93 also inhibits CaMKFV and, more importantly, voltage-dependent Ca2+ channels. Peptide inhibitors of CaMKII related to AIP are widely believed to be more specific, however, they also inhibit other CaM kinases as well as PKA. Thus, a better inhibitor would be a valuable and widely applicable tool. Such an inhibitor should be (i) CaMKII-specific, (ii) cellpermeable to allow easy use in cellular assays, (iii) readily available at reasonable cost or effort, and (iv) should be well characterized to allow educated interpretation of obtained data. As a starting point for such an inhibitor, a naturally occurring CaMKII inhibitory protein termed CaM-KIIN is a very promising source, as it does not affect activity of CaMKIV, PKC or PKA. In this proposal, we will first use peptides derived from the CaM-KIIN sequence to determine the minimally required region necessary for efficient and specific inhibition of CaMKII. We will then make cell permeable peptides based on such minimal inhibitory region, using the permeabilizing ant and tat sequences. The efficiency and specificity of these peptides will first be determined in biochemical assays. We will then determine their effect on substrate phosphorylation within neurons and a non-neuronal expression system; a suitable substrate to be tested is the AMPA-type glutamate receptor subunit GluR1. Then, the inhibitor will be tested for its action on the different modes of CaMKII activity (calcium/calmodulin-stimulated, autonomous by autophosphorylation, and autonomous by NR2B binding). The inhibitor developed and characterized in this proposal will allow easy testing of specific CaMKII functions in various physiological and pathological situations, and may thus lead to new therapeutic avenues, including for stroke.

The Ca2+/calmodulin-dependent protein kinase II (CaMKII) is required for forms of learning and memory. In brain, CaMKII is expressed at levels reminiscent of structural proteins (1-2% of total protein), and its 12meric holoenzyme structure allows multiple simultaneous protein interactions. However, structural roles of CaMKII or functions of its interactions with other protein are understood poorly at best. The major excitatory neuro- transmitter in mammalian brain, glutamate, induces two types of CaMKII translocation in hippocampal neurons: To postsynaptic sites upon brief stimulation (0.5-2 min) and to extra-synaptic clusters after extended exposure (3-5 min). Both types of translocation require the multivalent CaMKII holoenzyme structure (unpublished observations). Brief glutamate treatment can induce forms of synaptic plasticity thought to underlie learning and memory, while extended exposure leads to excitotoxicity, a cellular model for ischemic cell death. The goal of this proposal is to test our hypotheses that synaptic translocation requires binding to the NMDA-type glutamate receptor subunit NR2B and is involved in synaptic plasticity, while formation of extrasynaptic clusters is mediated by CaMKII self-association and is involved in ischemic cell death. Elucidating the mechanisms and functions of CaMKII targeting will aid our understanding of synaptic principles underlying higher brain functions and behavior, and provide new therapeutic avenues for ischemic damage. We will accomplish our goal in the following aims (using rat hippocampal and cortical cultures as main model systems for studies in neurons): (1) Determine protein interactions required for subcellular CaMKII localization in neurons. We will identify CaMKII mutations and inhibitors that prevent NR2B-binding or self-association in vitro and determine their effect on GFP-CaMKII localization in neurons. (2) Determine neuroprotective versus neurotoxic effects of CaMKII activity, synaptic targeting, and extrasynaptic clustering. We will make use of an inhibitor of CaMKII activity and clustering, knockdown of expression by RNAi, and overexpressionof CaMKII wildtype and mutants with specific impairments in activity, regulation and targeting. (3) Determine functions of CaMKII in regulation of synapse number and strength. We will make use of the same tools as in (2) and determine effects on synapse number and strength in established assays.

DESCRIPTION (provided by applicant): The Ca2+/calmodulin-dependent protein kinase II (CaMKII) and the NMDA-type glutamate receptor (NMDAR) subunit GluN2B are two central mediators of long-term potentiation (LTP), a form of synaptic plasticity thought to underlie both physiological and maladaptive addiction-related learning and memory.. The proposal will test the hypotheses that transient disruption of the CaMKII/GluN2B complex (a) persistently reverses amphetamine- and cocaine-induced addiction- related behavior and (b) acutely interferes with memory consolidation but does not reverse established memory. These hypotheses are based on recent preliminary data on addiction behavior and on recent results with normal LTP. Notably, the results of this study will be equally significant even if they lead to rejection of our hypotheses (for instance in case we instead find that inhibition of CaMKII activity is sufficient for reversal of addiction behavior, without requirement for disruption of th CaMKII/GluN2B complex). (Aim 1) We will here first establish a pharmacological treatment that disrupts the CaMKII/GluN2B complex in vivo. (Aim 2) Then, we will determine the effect of such treatment on reversal of addiction behavior. This will directly determine feasibility of a novel therapeutic approach to addiction. (Aim 3) Finally, we will determine the effect on normal memory, which is also clinically important. Acute but reversible interference with memory acquisition and even erasure of recent not yet consolidated memory would be clinically acceptable side-effects. However, while an unexpected erasure also of already consolidated memory would be of high scientific impact, it would need to be overcome in therapy development (for instance by targeting a specific brain region, i.e. the NAc, which could be done but would require more intensive care). We will here utilize a highly rigorous pharmaco-genetic approach. While the pharmacological treatment enables determining the therapeutic relevance (by enabling the temporal distinction between prevention and reversal of addiction behavior), two specific mouse mutant will determine target-specificity: a CaMKII knock-out mouse will test for CaMKII-specificity, while a mouse with mutant GluN2B incapable of CaMKII binding will test for the CaMKII/GluN2B complex as the specific target. Importantly, this approach also overcomes potential compensatory effects that are frequently seen in mutant mice: If the mutant mice still develop addiction behavior (either to normal or to somewhat reduced levels) due to such compensatory effects, the pharmacological treatment should be no longer effective in these mice, if their mutation is indeed the relevant target of the treatment.

DESCRIPTION (provided by applicant): Each year in the United States ~600,000 people suffer from cardiac arrest (CA) and receive cardiopulmonary resuscitation (CPR), resulting in hypoxia-ischemia (HI) of the brain and consequently in severe neurological deficits in most of the survivors. No pharmacological treatment is available to improve survival or long-term neurological outcome. Ischemic damage in the brain following cerebral ischemia induced by CA is in large part due to glutamate excitotoxicity triggered by a pathological flood calcium through NMDA-type glutamate receptors (NMDAr), ultimately resulting in neuronal cell death. While the pathology and cascade of events leading to injury following cerebral ischemia is complex, overstimulation of NMDAr is considered the major triggering spark for excitotoxicity and ischemic neuronal damage. Excitotoxic stimulation of NMDAr triggers a series of Ca2+-dependent signaling pathways, including activation of Ca2+/calmodulin-dependent protein kinase II (CaMKII). Our recent findings, and preliminary data, demonstrate that inhibition of CaMKII is a novel approach to minimizing downstream effects of excessive glutamatergic stimulation (excitotoxicity). CaMKII is well established as a major mediator of physiological glutamate signaling involved in synaptic plasticity, particularly synaptic potentiation in CA1 neurons, likel contributing to hippocampal learning. Two major forms of CaMKII regulation have been described, stimulated and autonomous CaMKII activity. Stimulated activity is induced by Ca2+/CaM to CaMKII, and prolonged autonomous (Ca2+-independent) activity is induced by autophosphorylation of residue T286 and/or direct binding to NMDAr subunit NR2B at the synapse (which also mediates CaMKII accumulation at the synapse). Additionally, our preliminary studies reveal a novel link between CaMKII autonomy and nitric oxide (NO), which is produced during excitotoxicity following cerebral ischemia and contributes to oxidative stress and neuronal damage. Our preliminary data indicate that NO-induced nitrosylation/oxidation of CaMKII promotes autonomous CaMKII activity, directly and by protecting T286 from de-phosphorylation. We recently demonstrated that our new CaMKII inhibitor tatCN21 (which blocks both stimulated and autonomous activity) provides robust neuroprotection both in vitro and following experimental stroke. In contrast, traditional CaMKII inhibitors (which block only stimulated activity) do not provide post-insult neuroprotection. This indicates that inhibition of autonomous, but not stimulated CaMKII activity, is a relevant drug target for post-insult neuroprotection. [Importantly, our preliminary results demonstrate that administration of tatCN21 after cardiac arrest results in significant neuroprotection.] The current proposal will utilize our novel mouse CA/CPR model to take advantage of several mutant mouse strains deficient in each form of autonomous CaMKII activity to unravel the complex interactions between these forms of CaMKII activity and their relative contribution to ischemic neuronal cell death. We hypothesize that (i) each autonomy mechanism contributes to neuronal cell death, and that (ii) T286-autophosphorylation mediated CaMKII autonomy is of most direct importance, that (iii) the novel nitrosylation mechanisms contributes to ischemic damage by prolonging T286 phosphorylation, and that (iv) NR2B-binding enables efficient nitrosylation via localizing CaMKII near nNOS.

Project Summary/Abstract Past translational disappointments with stroke have tainted the perception of feasibility also for global cerebral ischemia (GCI). However, stroke and GCI are clearly distinct conditions. The translational potential in GCI is arguably much higher; yet, compared to stroke, much fewer efforts have been made to date (likely due to the smaller market share). GCI is caused by near- drowning or ?suffocation, as well as during heart bypass surgeries. However, most cases result from cardiac arrest (CA) followed by cardio-vascular resuscitation (CPR). Each year, approximately 600,000 Americans suffer from CA and receive CPR. Approximately 60% of the survivors have long-lasting moderate to severe cognitive impairments. Compared to stroke patients, CA/CPR patients are much more homogenous (duration of ischemia; time elapsed between ischemia and hospitalization; always full reperfusion at a well-defined time point). This is expected to provide a significant advantage in future clinical trials (as decreased variability significantly increases statistical power). Our preliminary studies have identified inhibitors that showed high therapeutic potential in a mouse model of CA/CPR that closely mimics the human condition. The inhibitors where injected i.v. at a clinically relevant timepoint after CA/CPR, and showed dramatic improvement. Such improvement was seen also over therapeutic hypothermia alone (the current standard of care); when both treatments were combined, only residual and barely detectable neuronal cell death was seen. In patients, cell death after 2-3 days (measured by serum levels of neuron- specific enolase) is strongly correlated with severity of the long-term outcome. Endpoints tested here additionally included functional effects on synaptic plasticity in the surviving neurons and behavioral outcome in a learning/memory task. Target validation has been done in mutant mice. Mode of target engagement has been characterized biochemically and structurally. The optimized lead compound is highly potent tight binder, with an in vitro IC50<0.4 nM and full therapeutic efficacy in vivo even at 0.01 mg/kg i.v.. The compound is highly selective, shows no hERG channel binding at the therapeutic concentration, and no apparent adverse health effects in mice even at 100fold of the therapeutic dose. Target-specific potential risk-factors have been evaluated in mice, and did not pose problems. Shelf-stability of the compound is very good. This project will collect the remaining data that are required for moving forward to IND- enabling safety studies, in the following four specific aims. Aim 1: Lead optimization through initial safety studies. We will initiate safety studies with a focus on the highest risk factors. If needed, this will be followed by optimization of the lead compound and/or the delivery route. Aim 2: Efficacy in a non-rodent model, specifically determined in a pig CA/CPR model that is established by local collaborators. Two initial doses will be used and efficacy will be compared to mouse based on PK comparison. Additional regimens will be determined based on the results of this and of the subsequent aims. Aim 3: PK and PK/PD relationship, will be determined in both test species after a single bolus and correlated to efficacy. Based on the outcome, booster treatment will be designed and tested. Functional target engagement will be verified by biochemical activity assays. Additionally, we will compare plasma protein binding between the test species and human. Aim 4: Optimization of the treatment regimen in mouse, by testing additional time points and if subsequent booster injections can further improve efficacy. Additionally, long-term functional outcomes after these treatments will be tested. The nature of the optimal treatment regimen will inform the required safety studies during the next development phase.