Cristina Backman - US grants

CREB (cAMP response element binding protein) is a protein member of the bZIP subfamily of transcription factors. Activation of CREB occurs via phosphorilation of a single serine residue. Once phosphorylated, CREB dimmers, bind to specific CRE (cAMP response element) sites on target genes and regulates gene expression. CREB is of particular interest in drug addiction because its activation is downstream of the cAMP-signaling pathway, whose upregulation has been extensively characterized as an adaptation to chronic exposure to drugs of abuse (Chao and Nestler 2004). Upregulation of the cAMP pathway and activation of CREB seems to be crucial for the effects of drugs on the brain reward and motivational systems, such as the nucleus accumbens, ventral tegmental area, amygdala and frontal cortex, to name a few. In several of these regions, activation of CREB occurs in response to acute and chronic administration of different drugs of abuse, such as opiates, stimulants and alcohol. The regulation of CREB phosphorylation and the function of CREB in addiction vary with respect to multiple parameters, including the identity of the substance (opiates versus cocaine), the nature of the exposure (acute versus chronic), and the CNS region or neuronal pathways involved (Nestler 2004). A leading hypothesis is that drug-induced activation of CREB in these motivation centers of the brain underlies some of the common core features of drug addiction seen clinically, by inducing gene expression that leads to permanent molecular changes or drug induced plasticity. Kandel and collaborators were able to demonstrate that inactivation of CREB in hippocampal CA1 neurons impairs learning in the Morris water maze by interfering with some forms of long-term memory formation (Pittenger et al., 2002). To perturb CREB function, a transgenic mouse that expresses KCREB (a mutant of human CREB that is a potent dominant-negative inhibitor) through a forebrain specific promoter was generated. In accordance, over expression of CREB in CA1 neurons facilitated the establishment of long-lasting LTP in hippocampal slices (Barco et al., 2002). Interestingly, recent studies by Bonci and collaborators have shown that a single exposure to cocaine induces long-term potentiation in dopamine neurons (Ungless et al., 2001), a mechanism that, as in memory formation, may contribute to synaptic plasticity and permanent neuronal changes, which may progressively lead to drug addiction. Whether CREB affects the formation of LTP in dopaminergic cells remains to be investigated. Also, Nestler and collaborators have shown that over expression of a dominant negative mutant CREB in the area of the nucleus accumbens (through the neuron specific-enolase promoter) increases the rewarding effects of cocaine, possibly by regulating dynorphin expression. As CREB is a molecular component of ubiquitous nature in the mesolimbic dopamine system, it would be advantageous if we could dissociate its effects by conditionally inactivating its function in separate sets of mesolimbic neurons, one set at the time. This approach would allow us to better understand the specific roles that CREB plays in addiction and other conditions affecting the mesolimbic system.[unreadable] [unreadable] In our laboratory, we are studying genetically modified mice containing a tetracycline inducible dominant inhibitor CREB gene (KCREB) under the control of specific endogenous neuronal promoters. By using a system for temporal, spatial and cell-type specific control of gene expression, KCREB will be induced only in neurons expressing either the dopamine transporter (DAT), the dopamine receptor 1 (Drd1), or 2 (Drd2) and only after treatment with doxycycline (a tetracycline analog). These different knock-in mice will help us determine the functions that CREB mediates during addiction in specific neuronal types of the mesolimbic dopamine system. These conditional animals are being developed with the use of a single DNA construct that has been tested in vitro (see Bdckman et al., 2004). We will first determine if the genetic modifications introduced in this new animal strain produce the expected phenotype. Upon verification of a biologically functional mutation, these animals will be used for further in depth studies of CREB function during drug addiction. Exhausting behavioral, morphological and functional studies of these animals will generate data that correlates CREB inactivation in D1-, D2- or DAT-positive neurons to addiction.[unreadable] [unreadable] We have generated all the strains described above and are currently under study.

The resulting mutant mice showed neuronal hypertrophy, and an increased number of dopaminergic neurons and fibers in the ventral mesencephalon. Interestingly, quantitative microdialysis studies in Pten KO mice revealed no alterations in basal DA extracellular levels or evoked DA release in the dorsal striatum, despite a significant increase in total DA tissue levels. Striatal dopamine receptor D1 (DRD1) and prodynorphin (PDyn) mRNA levels were significantly elevated in KO animals, suggesting an enhancement in neuronal activity associated with the striatonigral projection pathway, while dopamine receptor D2 (DRD2) and preproenkephalin (PPE) mRNA levels remained unchanged. In addition, PTEN inactivation protected DA neurons and significantly enhanced DA-dependent behavioral functions in KO mice after a progressive 6OHDA lesion. These results provide further evidence about the role of PTEN in the brain and suggest that manipulation of the PTEN/Akt signaling pathway during development may alter the basal state of dopaminergic neurotransmission and could provide a therapeutic strategy for the treatment of Parkinsons disease, and other neurodegenerative disorders. PTEN deletion or Akt/PKB activation in dopamine neurons of the ventral midbrain results in remarkable hypertrophy of the substantia nigra and VTA. Our initial characterization of a DAT-PTEN KO strain has provided a clear definition of some of the neuroadaptations in the mesolimbic and nigrostriatal systems, and clearly show dopamine neurotransmission is permanently altered in PTEN KO mice. However, while DAT-PTEN KO animals are viable and appear behaviorally competent, an in depth study of behavioral parameters will clarify if the lack of PTEN interferes with essential functions related to the dopamine system in young, adult and aged animals. Studies performed over the past few years have clearly shown that phenotypes caused by specific genetic modifications are strongly influenced by genes unlinked to the target locus. This problem is exacerbated through the use of Cre-LoxP models as two strains, often containing their own (obscure) genetic backgrounds, are crossed through very specific breeding schedules to generate control and experimental animals. Clearly, it becomes important to avoid the use of a mixed genetic background so complex as to preclude any reasonable use of controls and prevent replication by other investigators. To perform complex behavioral studies, we will use a c57bl/6 congenic DAT-PTEN KO mouse line, generated in our lab. A new mechanism in the brain of rats that may mediate the rewarding and reinforcing properties of drugs of abuse has recently been discovered. This mechanism involves the physical interaction between two proteins in midbrain dopamine neurons, the tumor suppressor PTEN and the brain specific receptor for serotonin, the 5-HT2c receptor (5-HT2cR). Blocking the interaction of PTEN with 5-HT2cR prevents the development of conditioned place preference to nicotine and marijuana. In addition, PTEN has been shown to physically interact with the NR1 subunit of NMDAR in hippocampus, and PTEN downregulation decreases NMDAR surface expression. These studies suggest PTEN in dopamine neurons may directly modulate functions intimately linked to the development of addiction, and dopamine mediated cognition, such as responses to reward and motivation. We will use a congenic PTEN-KO line to analyze in detail the behavioral profile of KO animals in relationship to drug abuse, overall locomotor performance, as well as other dopamine related cognitive functions. We have shown Pten deletion in differentiated DA neurons causes a significant increase in the number and size of surviving neurons in both the mesolimbic and nigrostriatal projecting pathways (see above). Because at the time of Pten deletion DA neurons have already completed mitosis and phenotypic determination, it is unlikely the reported increase in DA neurons is due to an increase in newly formed neurons. It is thus likely PTEN ablation preserves DA neurons that normally would undergo apoptosis due to the lack of target support, by repressing the initiation of apoptotic pathways. We are now intrigued about several aspects induced by PTEN deletion in dopamine cells: Do all dopamine neurons preserved in DAT-PTEN KO animals project to target areas and form functional connections? Can exposure to an enriched environment enhance dopaminergic function in young and aged KO animals? Do the mesolimbic and nigrostriatal dopamine projections remain functional into aging? Is the PTENless aging dopaminergic system more or less resistant to neurotoxic insults applied during different stages of the mouse life span? Are PTENless dopamine neurons prone to form tumors? Obviously, the tremendous adaptations observed in the PTENless dopaminergic system during development, may pose problems for interpretation of results;however, as previous studies have shown Akt/PKB activation in the adult dopaminergic system can also result in remarkable hypertrophy and plasticity of the nigra and VTA, the results obtained in this study may provide valuable insights into how PTEN ablation changes dopamine function at the molecular and behavioral levels, and the long-term consequences of such adaptations. We believe these studies are important, as manipulations of the PTEN pathway are being considered as a possible venue for therapeutic strategies involving the brain.

DAT-NR1 KO mice were implanted with a bilateral cannula in the VTA to deliver specific NMDAR antagonists during a cocaine sensitization regimen, and our results have shown that sensitization can be equally blocked in DAT-NR1 KO and WT animals by simultaneous injection of an NMDAR antagonist in the VTA along with peripheral cocaine. Our experiments are now aimed at ruling out or including the mesocortical DA cell population as the mediator of VTA NMDA dependent sensitization. We are currently performing patch-clamp recordings from retrograde labeled neurons projecting to the prefrontal cortex in DAT-NR1 KO to determine if these dopamine neurons, due to low DAT levels and therefore lack of Cre, may express NMDAR currents. In addition, TH-NR1 KO animals are being generated, and if viable, will be used to determine the response to cocaine sensitization, as in these KO animals NMDAR currents will be eliminated in all dopamine cells. Possible pitfalls when using TH-NR1 animals is the unspecificity of the NR1 deletion, as NMDAR will be deleted in all cathecholaminergic brain areas. However, if TH-NR1 KO animals do not develop cocaine induced sensitization it will suggest the likelihood of an active role of the DA mesocortical projection for the initiation of sensitization.

To investigate if the use of genetically modified iPS cells can get around some of the logistical problems imposed by the use of fetal cell transplants, we propose to study if it is possible to manipulate the mechanisms of dopamine-neuronal differentiation and survival in our favor, to generate transformed ES and iPS cells that enhance their therapeutic value when transplanted into the brain of hemiparkinsonian mice. Genetically altered mouse ES and iPS cell lines will be generated by using mouse colonies with specific mutations known to affect the function of dopamine cells. For example, in a previous study we have shown ablation of the gene PTEN (phosphatase and tensin homolog) in dopamine neurons, leads to dopamine hypertrophic effects and a significant increase in neuronal protection. We will generate mouse ES and iPS cells from an established transgenic mouse colony genetically manipulated to ablate PTEN activity in dopamine neurons. Differentiated dopamine cells lacking PTEN activity will be transplanted in the brain of hemiparkinsonian mice to study the benefits of such strategies for dopamine enrichment and survival from transplanted cells and replacement of dopaminergic function.

We have developed a mouse model in which dopamine neurons lack the ability to synthesize BDNF (BDNF knockout mice), by expressing Cre recombinase through the dopamine transporter locus in a BDNF loxP transgenic mouse. This mouse model provides a tool to determine how BDNF synthesized exclusively in DA neurons contributes to the shaping and function of the dopaminergic system. BDNF is highly expressed in midbrain DA neurons, and as a secreted molecule it could act on DA neurons, as a paracrine factor, or have an effect on surrounding neurons and in target areas by anterograde transport. We intend to utilize the BDNF knockout mouse model to determine how the ablation of BDNF in DA neurons affects the state of midbrain dopaminergic neurotransmission during development and in adult animals. In addition, as expression of BDNF in DA neurons has been shown to shape the neuro-plasticity observed during learning and reward, we intend to define how the lack of BDNF exclusively in DA neurons may affect behavioral outputs related to the reward system, and to define what cellular mechanisms and pathways intrinsic to DA neurons may be directly affected by BDNF expression.

In our laboratory, we recently developed a Cre-loxP transgenic model to specifically inactivate β-catenin in DA neurons of the ventral midbrain (β-catenin KO mice). β-catenin KO mice are born at expected Mendelian radios. Adult mice present lower body weight compared with age-matched controls. Stereological cell counts do not reveal differences in the number of tyrosine hydroxylase positive neurons in the substantia nigra and ventral tegmental area of β-catenin KO mice. However, behavioral studies showed that β-catenin deficient mice are hyperactive in a novel environment, although no changes were found in the circadian locomotor activity pattern. To assess the dynamics of DA release in the striatum, fast-scan cyclic voltammetric recordings were performed in brain slices obtained from β-catenin KO mice and WT controls. Preliminary studies showed an increase of DA release in the striatum of KO animals when compared to controls. Striatal prodynorphin (PDyn) mRNA levels were significantly elevated in KO animals, suggesting an enhancement in neuronal activity associated with the striatonigral projection pathway, while preproenkephalin (PPE) mRNA levels remained unchanged. These results suggest β-catenin plays an essential role in the development and maintenance of dopaminergic neurotransmission. Our lab is currently focused on elucidating what molecular mechanisms, mediated by β-catenin ablation, result in the observed changes in behavior and dopaminergic neurotransmission.