Katherine Conant - US grants

DESCRIPTION: An abundance of data suggests that products of activated monocyte-derived cells may contribute to the development of HIV dementia. Prominent monocytic infiltration of the central nervous system (CNS) is observed in association with this condition. Moreover, the degree of macrophage activation and the CNS levels of macrophage derived products, including TNF-alpha and prostaglandin E2, correlate with dementia severity. One mechanism by which the products of activated monocytes may contribute to disease is through direct effects, including TNF-alpha and/or eicosanoid mediate neurotoxicity. Another mechanism by which these products may contribute to disease, however, is through their ability to stimulate the production of matrix metalloproteinases (MMPs) from CNS- resident cells. MMPs can affect the ability of inflammatory cells to cross the blood brain barrier (BBB) and to migrate within the CNS parenchyma itself. In addition, through their effects on BBB permeability, MMPs may allow systemic cytokines and other toxins to more easily enter the CNS. It is also possible, however, that through their effects on neuronal lands and receptors, MMPs could more directly affect neuronal function. Therefore, in order to improve our understanding of those mechanisms that might contribute to the development of HIV dementia, and to potentially gain information with respect to specific therapeutics, the applicants aim to test the following hypotheses: 1) that specific cytokines, which are elevated in association with HIV dementia, increase the production of select MMPs from CNS derived cells, 2) that these MMPs may be neurotoxic, and 3) that the production, release and/or activity of these MMPs can be reduced by specific antagonists. Experiments will include analysis of supernatants from cytokine stimulated cells by ELISA and Western blot, as well as by collagen, laminin an gelatin substrate zymography. In addition, cultured neurons will be stimulated with select MMPs, as well as with supernatants from cytokine stimulated CNS derived cells, both in the presence and in the absence of specific antibodies/antagonists. Also, compounds including the hydroxamic acid maramastat as well as select anti- oxidants and anti-inflammatory drugs, will be tested for their ability to inhibit the production, release and/or activity of specific MMPs.

HIV infection continues to rise in drug users, putting them at greater risk of developing dementia. The pathobiology of HIV dementia is only partly understood, however, it is clear that disruption of the blood brain barrier facilitates the entry of peripheral blood monocytes into the brain. Activated and HIV-1 infected monocytes within the central nervous system (CNS) can in turn damage neighboring cells through the release of infectious virus and/or potent neurotoxins. It has recently been shown that drugs of abuse such as cocaine can disrupt the blood brain barrier (BBB), though the associated mechanisms are not well understood. Critical components of this barrier include extracellular matrix proteins such as collagen IV and laminin. Extracellular matrix proteins are regulated by an endogenous family of proteins called matrix metalloproteases (MMPs). These proteases are released not only by activated T cells and monocytes as they migrate through the BBB, but also by resident cells of the brain parenchyma. It has been shown that injection of MMPs into the parenchyma of the brain causes a disruption of the BBB which is followed by leukocytic infiltration of the central nervous system (CNS). Moreover, MMPs produced within the brain parenchyma may have more direct effects on CNS structure and function in that extracellular matrix proteins are known to affect synaptic structure as well as neuronal survival. In the present study, we will therefore address the possibility that both HIV-1 proteins and drugs of abuse will increase MMP production by brain-derived cells. We will also examine the mechanisms by which these substances may affect MMP production, and identify clinically tolerable compounds which may inhibit such production.

DESCRIPTION: The pathobiology of HIV dementia (HIVD) is only partly understood, however, evidence suggests that disruption of the blood brain barrier (BBB) may play a role. Structural and functional alterations in the BBB have been detected in association with HIVD, and these alterations would be expected to facilitate the ingress of both inflammatory cells and serum-derived toxins. Extracellular matrix proteins, including those which are critical to the integrity of the BBB, are regulated by an endogenous family of proteins called matrix metalloproteases (MMPs). These proteases are released not only by activated T cells and monocytes as they migrate through the BBB, but also by resident cells of the brain parenchyma. It has been shown that injection of MMPs into the parenchyma of the brain causes a disruption of the BBB, which is followed by leukocytic inflitration of the central nervous system (CNS). Moreover, MMPs produced within the brain parenchyma may have more direct effects on CNS structure and function in that extra cellular matrix proteins are known to affect synaptic structure as well as neuronal survival. In the present study, we will address the possibility that activated and HIV-1 infected monocytes release substances which can in turn increase the release of MMPs from resident cells of the CNS. We will also address the possibility that supernatants from activated and HIV-infected monocytes, as well as the MMPs that they induce, will affect the structure and function of an in vitro BBB model. These studies are novel in that the effects of supernatants from activated and HIV infected monocytes on MMP production by resident cells of the CNS are unknown, and in general, little is known about MMP regulation in the CNS. Also, the proposed studies will use both primary brain-derived HIV isolates, which differ from laboratory adapted strains, and normal human cells, which also differ significantly from those of rodent or tumor origin.

DESCRIPTION (provided by applicant): Pathological evidence suggests that levels of both thrombin and its principal receptor, proteinase activated receptor-1 (PAR-l), are increased in associated with HIV encephalitis. Thrombin can enter the CNS through a damaged blood brain barrier, and it is also produced by immune activated cells of the brain parenchyma. Potential effects of thrombin which may be important to the pathogenesis of HIVD include its ability to degrade matrix proteins that support cell adhesion and survival, as well as its ability to signal through PARs. Activation of endothelial cell PARs has been linked to changes in cell shape, increased chemokine release, and increased adhesion molecule expression. Activation of neuronal PARs has also been demonstrated and is thought to underlie some of the previously reported effects of thrombin on these cells, including neurotoxicity and neurite retraction. Of interest, HIV dementia (HIVD) has been associated with evidence of neurite retraction including dendritic simplification and reduced synaptic density. The responsible mechanisms, however, are not well understood. In the present R21 application, we propose to test the hypothesis that thrombin acts on specific neuronal PARs to stimulate neurite retraction through the activation of glycogen synthase kinase 3b (GSK3b), a critical mediator of cell shape and survival. GSK3b has been shown to phosphorylate microtubule-associated proteins including tau, and increased phosphorylation of tau has been linked to neurite retraction in a number of experimental paradigms. We will also identify the mechanisms that underlie thrombin's effects on GSK3b. While previous studies have suggested that G protein subtypes typically associated with alterations in Rho GTPase activity can affect the activity of GSK3b, none have shown that those linked to decreased adenylate cyclase activity can do the same. We will therefore test the hypothesis that thrombin stimulation of neurons affects both a decrease in PKA-dependent serine phosphorylation and an increase in RhoA dependent tyrosine phosphorylation of GSK3b. The results from these experiments should prompt future studies to determine whether other HIVD related stimuli that cause neurite retraction target similar pathways. Ultimately, we hope to gain a better understanding of mechanisms that contribute to altered neuronal morphology in the setting of HIVD, and to determine whether it may be useful to target GSK3b and/or specific G protein coupled receptors.

[unreadable] DESCRIPTION (provided by applicant): Evidence suggests that integrins may be critical to the establishment and maintenance of neuronal networks. Integrins are involved in the directed migration of both neurons and their processes, and a number of studies suggest they are also involved in the stabilization of synapses. The ability of integrins to play a role in neural networking is understandably influenced by the nature and availability of integrin binding ligands. These ligands include cell surface receptors, soluble molecules such as the semaphorins, and select proteins of the extracellular matrix (ECM). Recent studies have shown that both neural networking and the availability of integrin binding ligands may be altered in the setting of central nervous system (CNS) inflammation. For example, in simian immunodeficiency virus encephalitis (SIVE), synaptic damage and loss of perineuronal matrix has been demonstrated. Similarly, in human immunodeficiency dementia, a loss of dendritic arbor has been observed. And with respect to changes that may occur in development, in utero infection with cytomegalovirus has been linked to disrupted cortical organization. While multiple mechanisms are likely to contribute to disordered neural networking in the context of inflammation, matrix metalloproteinase (MMP) mediated changes in integrin function are likely to play an important role. MMPs degrade constituents of the ECM that are known to interact with integrins, and select MMPs have also been shown to directly associate with cell surface integrins. In the present proposal, we outline plans to test the hypothesis that MMPs will interfere with neuronal migration and with the formation or maintenance of synapses. We will also test the hypothesis that such effects are dependent on neuronal integrins. The results of our studies should improve our understanding of those CNS inflammatory diseases that are characterized by elevated levels of MMPs. [unreadable] [unreadable] [unreadable]

[unreadable] DESCRIPTION (provided by applicant): Synaptic damage has been described with both HIV dementia (HIVD) and METH use. For example, dendritic injury has been well described in HIV encephalitis (HIVE), and METH using patients with HIV encephalitis (HIVE) show greater loss of synaptic protein immunoreactivity than do HIVE non-METH users. Matrix metalloproteinases (MMPs) are zinc dependent enzymes whose levels may be substantially elevated in association with HIVE. Further both HIV proteins and methamphetamine have been shown to increase MMP release from cultured neurons and glia. Though well studied for their effects outside of the brain, studies of MMPs in the CNS have generally been limited to their effects on proteins of the blood brain barrier. Of particular interest, however, is the potential for MMPs to target proteins critical to synaptic structure and function. In vitro studies, often using non-neural systems, have shown that MMPs target molecules including syndecans, cadherins, and SIRP-1a. Of interest, each of these molecules is known to play a role in synaptic adhesion. Moreover, as will be shown to follow, we have preliminary data which suggests that select MMPs target molecules that play a predominant role in synaptic function. One such molecule is the NR1 subunit of the NMDA type glutamate receptor. The present application will therefore outline plans to test the hypothesis that MMPs are critical mediators of synaptic injury occurring with HIV proteins and METH. Dissociated and slice cultures will be treated with HIV proteins and METH, in the presence or absence of specific MMP inhibitors, and then evaluated for evidence of MMP mediated synaptic injury. MMP levels, select synaptic adhesion molecules, and NMDA receptor subunits will be examined, as will be neuronal death. Synaptic structure will also be evaluated by confocal and electron microscopy. In parallel studies, METH treated mice that over express HIV-1 Tat or gp120 will be treated with a broad spectrum MMP inhibitor, or vehicle control, and CNS tissues will be similarly evaluated. The results of the studies proposed in this R21 application should determine whether MMPs are elevated in HIV protein/METH treated cultures and animals. More importantly, the proposed studies should determine whether MMPs likely play a role in synaptic injury occurring with METH use and HIV infection, and whether general or selective MMP antagonists should be considered for the treatment of specific patients. [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): A variety of inflammatory conditions induce tissue damage through excess proteolysis. However, the extent to which proteolysis of neuronal substrates can contribute to brain pathology is less well understood. A particularly important family of proteolytic enzymes is the matrix metalloproteinases (MMPs), zinc- dependent, secreted endopeptidases that can cleave extracellular matrix proteins as well as cytokines and cell surface receptors. MMPs are produced within the brain and their production may be increased by pro-inflammatory cytokines and 2-amyloid. Moreover, elevated expression of MMPs has been reported in Alzheimer's disease, vascular dementia, and HIV-associated dementia. While some studies have examined the possibility that MMPs are neurotoxic, a full understanding of how elevated levels of these enzymes may influence neuronal physiology is lacking. Of particular importance is the potential for MMPs to influence neuronal networks via an effect on the synapse. Preliminary data shows that at least one MMP, MMP-7 inhibits synaptic vesicle release in vitro. Such data also shows that immunoreactivity for select SNARE protein, known to play a role in synaptic vesicle release, is reduced by exogenous MMP-7. The central hypothesis of the present proposal is that MMP-7 alters synaptic vesicle release via direct intrasynaptic cleavage of SNARE protein(s). The possibility that MMP-7 enters neurons via endocytosis and then mediates intrasynaptic cleavage of essential SNARE protein(s) will be examined. The rationale for the research is that if MMP-7 has direct effects on synaptic structure and function, it may contribute to synaptic dysfunction in the setting of CNS inflammation, and excess proteolysis may be considered as a target for neuroprotective therapeutics. The purpose of the application is to understand how MMP-7, a protease whose levels are increased in the brain inflammations, influences neuron to neuron communication. The relevance of this research to Public Health is that it may explain how inflammation contributes to neurological dysfunction. Excess levels of select proteases may be considered as novel targets for neuro-protective drugs.

DESCRIPTION (provided by applicant): Classical microglial activation may contribute to neuronal injury occurring with HIV associated neurological disorders (HAND). Emerging evidence suggests that matrix metalloproteinases (MMPs) could play a critical role in stimulating such activation. For example, inhibition of MMP activity blocks microglial activation in response to stimuli including lipopolysaccharide. In addition, minocycline, a potent inhibitor of MMP expression and activity, abrogates microglial activation occurring with osmotic demyelination as well as with simian immunodeficiency viral (SIV) infection. The mechanisms by which MMPs might activate microglia are not completely understood. Of interest, however, is the potential for MMPs to generate ligands for integrins that are highly expressed on microglia. Integrin dependent signaling can in turn stimulate changes associated with an activated phenotype. That this mechanism could be important is underlined by data showing that inhibition of microglial integrin expression, or function, blocks microglial phagocytosis and migration. In addition, recent studies have shown that select integrin antagonists can substantially reduce microglial activation and associated neurotoxicity in more than one disease model. In previous publications we have shown that HIV proteins can increase MMP release from brain derived cells, and that MMP levels are increased in spinal fluid samples from patients with HAND. In the present application, we hypothesize that these MMPs generate specific cell adhesion molecule (CAM) fragments that will in turn engage microglial integrins. Our focus on CAM fragments is based on several considerations. CAMs are easily accessible by virtue of their proximity to the cell surface, an area where MMP activity may be concentrated. We and others have shown that MMPs stimulate ectodomain shedding of these molecules, and elevated levels of soluble forms can be detected in spinal fluid samples from patients with brain inflammation. Moreover, we have recently published data showing that the shed domain of at least one CAM can interact with a microglial integrin that has been well linked to an activated phenotype. In the present R21 proposal we plan to identify microglial integrin-binding ligands that may be increased in association with HIV, to investigate the hypothesis that these ligands stimulate classical, pro-inflammatory microglial activation, and to determine whether this activation is of sufficient magnitude to be inimical to vulnerable neurons. We will focus on CAMs that are widely expressed in the CNS, including the immunoglobulin (Ig) domain containing intercellular cell adhesion molecules (ICAMs) and synaptic CAMs (synCAMs), as well as the cadherins. We will also focus on integrins that mediate microglial activation in other disease models, such as LFA-1 and Mac- 1. This dual PI grant relies on expertise related to MMPs (KC), as well as microglia and dopaminergic neurons (KMZ). The possibility that MMPs contribute to microglial activation in the setting of HIV is yet untested and clinically relevant, in that it would allow MMP inhibitor to be used in an attempt to reduce such activation. Further study of the receptors that underlie MMP dependent effects is also clinically relevant. At least one soluble CAM can interact with LFA-1, a microglial integrin for which a clinically tolerable antagonist has been developed.

DESCRIPTION (provided by applicant): Protease activated receptor-1 (PAR-1) is a G protein coupled receptor (GPCR) that is highly expressed on neurons and microglia. Levels of PAR-1 activators, including plasmin and matrix metalloproteinases (MMPs), are substantially increased in HIV associated neurological disorders (HAND) and one study has shown that levels of PAR-1 are increased as well. Though not yet studied in the context of HAND, PAR-1 antagonists can prevent microglial activation and neurotoxicity in animal models of Parkinson's disease and cerebral ischemia. Recent studies have shown that select GPCRs can associate with ?-arrestins to activate the kinase glycogen synthase kinase-3??(GSK-3?) through a novel, non-canonical signaling pathway. Importantly, GSK- 3?? inhibitors have been shown to reduce neuronal injury in response to HAND relevant stimuli. In addition, increased GSK-3?? activity has been implicated in HAND relevant pathology including microglial activation, neurotoxicity, and long term depression of synaptic transmission. In the present dual PI R01 proposal, we hypothesize that excess activation of PAR-1 stimulates non- canonical GPCR dependent signaling pathways to measurably contribute to HAND relevant microglial activation and neuronal injury. Our plan will be to test underlying mechanisms with in vitro studies (Aims 1 and 2) and the relative in vivo importance of this pathway to phenotypic changes observed in mouse models (Aim 3). Preliminary data in support of our hypothesis will include evidence for GSK-3?? activation and cognitive impairment in mice that overexpress a potent PAR-1 agonist. Preliminary data will also show evidence for PAR-1 dependent, non-canonical signaling, in CNS derived cells isolated from these animals. Innovation comes from the study of a relatively unexplored receptor as related to HAND, the study of a novel signaling pathway for this receptor, and the use of a unique mouse model. Innovation also comes from techniques that include recordings of neuronal activity via multielectrode arrays, and small animal magnetic resonance spectroscopy. The overall goal of our proposal is to identify targets for adjunct therapeutics. If PAR-1 activation can stimulate HAND relevant pathology through increased GSK-3?? activity, novel drugs being developed to more specifically inhibit the activity of this kinase could be considered for treatment of this condition. Moreover, newly developed orally available PAR-1 antagonists that are now in clinical trials for coronary artery disease might be considered for treatment of the same.

Due to strong excitatory input, reliable GABA release and fast firing, parvalbumin expressing (PV) neurons are thought to represent critical pacemakers for synchronous network events. PV neurons also represent the predominant GABAergic neuronal population that is enveloped by the perineuronal net (PNN), a lattice like extracellular matrix that is thought to localize glutamatergic input. Disruption of the PNN has been linked to reductions in PV excitability. Importantly, deficits in PV excitability influence synchronous network events critical to both attention and long-term memory consolidation. In support of this, recent studies have demonstrated that reduced glutamatergic input to hippocampal PV cells, through knockout of PV selective glutamate receptors or a reduction in presynaptic glutamatergic innervation, is linked to increases in sharp wave ripple (SWR) density and deficits in long term memory consolidation. PNN processing occurs through the actions of specific proteases. While metalloproteinases of the ?a disintegrin and metalloproteinase with thrombospondin motifs? (ADAMTS) and secreted matrix metalloproteinase (MMP) family members can cleave specific PNN components, the latter may be particularly important in the background of human immunodeficiency virus (HIV) infection. Soluble MMPs are expressed by neurons and microglia and known to digest PNN components including aggrecan and brevican. In addition, while ADAMTS protein expression is not detected in astrocytes in a simian immunodeficiency virus (SIV) model, PNN degrading MMPs are highly expressed by astrocytes, the most numerous cell type in the brain. Moreover, in murine models of brain injury, selective MMP inhibition reduces PNN remodeling. It has previously been demonstrated that human HIV encoded Tat protein can increase the expression and/or cellular release of MMP-9, a potent modulator of PNN processing. Tat protein is detectable in the cerebrospinal fluid of individuals receiving combination anti-retroviral treatment (cART). In new preliminary data included herein, we show that Tat significantly increases release of MMP-13 from astrocytes. Moreover, we see active forms of MMP-13 in brain tissue lysates from virologically suppressed HIV-infected individuals. In preliminary studies, MMP-13 can efficiently cleaves PNN components. Published work has linked MMP-13 expression to HIV infection, and also shown reduced PNN integrity in the background HIV associated cognitive dysfunction (HAND). Importantly, however, causes and potentially critical neurophysiological consequences of PNN disruption in the setting of HAND have not been well examined. In the present application, we plan to test the hypothesis that HIV relevant stimuli including Tat can stimulate MMP-dependent PNN processing in vitro and in vivo, with consequent effects on hippocampal PV activity, neuronal population dynamics and memory consolidation. !

Drugs that target monoaminergic transmission represent a first line treatment for major depressive disorder, a debilitating condition that affects approximately 12-17 % of the U.S. population at some point during the individual?s lifetime [1]. Though a full understanding of the mechanisms that underlie antidepressant efficacy in responsive individuals is not yet fully appreciated, emerging evidence supports a role for enhanced excitatory transmission as a mediator of symptomatic remission. Increased excitatory transmission can occur through at least two non-mutually exclusive mechanisms. The first involves increased function of excitatory neurons through relatively direct mechanisms and, in the setting of antidepressant treatment, this likely involves enhanced dendritic arborization and spine formation. A second non- mutually exclusive potential mechanism is that reduced inhibition of excitatory neurons could contribute to antidepressant efficacy. Consistent with this possibility, GABAergic interneuron-mediated cortical inhibition has been linked to reduced ? activity [2], a rhythm that is also diminished in models of depression [3]. Remission of depressive symptoms induced by chronic restraint stress correlates with restoration of ? activity [3]. Due to strong excitatory input, reliable GABA release and fast firing, PV neurons are thought to represent critical pacemakers for synchronous network events [7, 8]. PV neurons also represent the predominant GABAergic neuronal population enveloped by the perineuronal net (PNN), a structure that is thought to localize glutamatergic input [10]. Disruption of the PNN reduces PV excitability [11], and has been linked to enhanced lateral diffusion of glutamate receptor subunits [10]. PNN disruption can also enhance cortical ? activity [13]. The presence of a robust PNN is thus thought to facilitate PV activation, while a relatively weak PNN can instead reduce the same. A limited number of published studies suggest that monoamine modulating antidepressants can reduce integrity of the perineuronal net to potentially facilitate neuroplasticity [14, 15]. Recent work has shown increased PNN deposition with social defeat stress-induced depressive behavior in rats [16]. This is associated with memory impairment that can be rescued with imipramine. Importantly, the mechanisms by which monoamine reuptake inhibitors reduce PNN integrity, and the mechanisms linking increased PNN deposition with depressive behavior, remain largely unexplored. We plan to test the hypothesis that specific monoamine reuptake inhibitors will increase the expression of plasticity-relevant matrix metalloproteinases (MMPs) in brain regions important to depression and that these MMPs will stimulate a) dendritic arborization and spine formation in pyramidal cells as well as b) processing of PNNs that surround PV expressing interneurons. We further propose that select monoamine reuptake inhibitors will act via MMP-dependent mechanisms to a) influence excitatory/inhibitory balance to increase network activity critical to attention and memory (? and sharp wave ripples) and b) ameliorate depression related behaviors.