University of Florida Thesis Or Dissertation Formatting

NEUROBIOLOGICAL SUBSTRATES OF POST-COCAINE COGNITIVE FUNCTION AND DRUG SEEKING: ROLE OF METABOTROPIC GLUTAMATE RECEPTOR 5

CHRISTINA M. GOBIN

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2019

To my parents, Kadar and Deborah Gobin

ACKNOWLEDGMENTS

I thank my parents for taking an interest in my studies and supporting my pursuit of a doctorate. I thank my best friends, Gabriella Powell and Steffanie Jacobson for the experiences that we shared which sparked my initial interest in neuroscience research and psychopharmacology. I thank my master’s advisor, Dr. Jamie Tartar for encouraging my pursuit of both master’s and doctoral degrees. I thank my mentor, Dr.

Schwendt and committee member, Dr. Knackstedt for guiding me throughout the dissertation process, expanding my intellectual interests, and making me a better research scientist. I thank my other committee members, Drs. Bizon and Dallery for their valuable input on my projects. I thank Dr. Muehlmann for teaching me cytochrome oxidase histochemistry. I thank my undergraduate research assistants who helped run my behavioral experiments. I thank John Shallcross as my lab mate for teaching me fluorescent insitu hybridization and microscopy. I also thank John Shallcross as a partner for our invaluable experiences connecting over art, music, neuroscience, intellectually stimulating conversations, conferences, crafting, plants, saltwater fish tanks, cooking, snacks, and Stark. Last, I thank my cat, Stark (Dr. Kinase) for being such a nice and perceptive friend, staying up late with me while I work and offering his love when needed.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 9

LIST OF FIGURES ...... 10

ABSTRACT ...... 12

CHAPTER

1 INTRODUCTION ...... 14

1.1 Cocaine Addiction ...... 14 1.1.1 Cocaine and Cognitive Dysfunction ...... 14 1.1.2 Cocaine and PFC Dysfunction ...... 15 1.1.3 Cognitive Deficits and Relapse ...... 18 1.1.4 Section Summary ...... 19 1.2 Working Memory ...... 20 1.2.1 Delayed Match-to-Sample Tasks Can be Used to Investigate Working Memory ...... 20 1.2.2 The Dorsolateral PFC in Humans and Non-Human Primates is Implicated in Working Memory ...... 21 1.2.3 The Prelimbic Cortex in Rodents is Homologous to the Human dlPFC to Investigate Working Memory ...... 22 1.3 The Role of the dlPFC and PrL in Cocaine Addiction ...... 24 1.3.1 The Role of the Human dlPFC in Cocaine Seeking ...... 24 1.3.2 The Role of the Rodent PrL and Cocaine Seeking ...... 25 1.4 The Role of Glutamate Receptors ...... 28 1.4.1 The Role of Glutamate Receptors in Working Memory ...... 31 1.4.2 The Role of Glutamate Receptors in Cocaine Seeking ...... 34 1.5 Conclusions ...... 36 1.5.1 General Hypotheses ...... 37 1.5.2 Dissertation Objectives: ...... 37

2 NEURAL FEATURES OF POST-COCAINE COGNITIVE DEFICITS AND DRUG SEEKING ...... 38

2.1 Materials and Methods ...... 42 2.1.1 Animals and Experimental Design ...... 42 2.1.2 Surgical Procedures ...... 42 2.1.3 Extended Access Cocaine Self-Administration and Abstinence ...... 43 2.1.4 Operant Delayed Match-to-Sample/Non-Match-to-Sample Task ...... 44 2.1.5 Context + Cue Relapse Test ...... 46 2.1.6 Tissue Collection ...... 46

2.1.7 Cytochrome Oxidase Histochemistry...... 47 2.1.8 Western Blotting ...... 48 2.1.9 Fluorescent in situ Hybridization ...... 49 2.1.10 Statistics ...... 50 2.2 Results ...... 50 2.2.1 Rats Escalated Cocaine Intake during Extended Access Cocaine Self- Administration ...... 50 2.2.2 Rats with a History of Extended Access Cocaine Self-Administration Showed Working Memory Impairments Related to Prior Cocaine Intake and Later Drug Seeking...... 51 2.2.3 Reversal Learning was Not Impaired in Rats with a History of Extended Access Cocaine Self-Administration. NMS Task Behavior Predicted Subsequent Drug Seeking...... 53 2.2.4 Relapse to Cocaine-Seeking on Day 90 of Post-Cocaine Abstinence was Accompanied by Increased Arc mRNA within mGlu5-Positive Cells in the PrL...... 54 2.2.5 Metabolic Activity in the PrL Positively Correlated with Prior Working Memory Performance in Cocaine Rats Only ...... 55 2.2.6 Greater Expression of the mGlu5 Monomer and Homer 1b/c in the PrL was Exhibited in Cocaine Rats. Expression of the mGlu5 Monomer Negatively Correlated with Prior Working Memory Performance ...... 56 2.3 Discussion ...... 57 2.4 Limitations ...... 67 2.5 Future Directions...... 68 2.6 Conclusions ...... 69

3 INVESTIGATING PRELIMBIC MARKERS DURING PROLONGED ABSTINENCE ...... 79

3.1 Materials and Methods ...... 80 3.1.1 Animals ...... 80 3.1.2 Surgical Procedures ...... 80 3.1.3 Extended Access Cocaine Self-Administration and Abstinence ...... 80 3.1.4 Tissue Collection ...... 81 3.1.5 Cytochrome Oxidase Histochemistry...... 81 3.1.6 Western Blotting ...... 81 3.1.7 Statistics ...... 81 3.2 Results ...... 82 3.2.1 Rats Escalated Cocaine Intake during Extended Access Cocaine Self- Administration ...... 82 3.2.2 PrL Metabolic Activity did Not Differ between Rats with a History of Cocaine or Saline during Home-Cage Abstinence ...... 82 3.2.3 Expression of mGlu5 and mGlu5-Associated Proteins in the PrL did Not Differ between Rats with a History of Cocaine or Saline during Home- Cage Abstinence ...... 82 3.3 Discussion ...... 83 3.4 Limitations ...... 84

3.5 Conclusions and Future Directions ...... 85

4 THE ROLE OF THE PRELIMBIC CORTEX IN WORKING MEMORY...... 91

4.1 Materials and Methods ...... 91 4.1.1 Animals ...... 91 4.1.2 Operant Delayed Match-to-Sample Task ...... 92 4.1.3 Tissue Collection ...... 92 4.1.4 Fluorescent in situ Hybridization ...... 92 4.1.5 Statistical Analysis ...... 93 4.2 Results ...... 93 4.2.1 Naïve Rats Exhibited Demand Dependent Performance during Baseline Testing. Naïve Rats Performed Worse in the High-Load DMS Task Compared to the Low-Load DMS Task ...... 93 4.2.2 Naïve Rats Demonstrated Greater mGlu5-Dependent Activation during the High Load Compared to the Low Load DMS Task. Performance in the High Load Condition Negatively Correlated with mGlu5-Dependent Activation ...... 93 4.3 Discussion ...... 94 4.4 Conclusions and Future Directions ...... 95

5 MODULATION OF MGLU5 RECEPTORS ON POST-COCAINE WORKING MEMORY AND DRUG SEEKING ...... 99

5.1 Materials and Methods ...... 102 5.1.1 Animals ...... 102 5.1.2 Drugs ...... 102 5.1.3 Surgery and Cocaine Self-Administration ...... 103 5.1.4 Operant Delayed Match-to-Sample Task ...... 104 5.1.5 Relapse Tests ...... 104 5.1.6 Locomotor Testing ...... 105 5.1.7 Statistical Analysis ...... 105 5.2 Results ...... 106 5.2.1 Rats Escalated Cocaine Intake during Extended Access Cocaine Self- Administration ...... 106 5.2.2 Rats did Not Differ in DMS Task Acquisition or Baseline Performance Prior to Treatment ...... 107 5.2.3 MTEP Impaired while CDPPB had Delayed Pro-Cognitive Effects on Working Memory Performance ...... 107 5.2.4 Both MTEP and CDPPB Decreased Drug-Seeking in a Context + Cue Relapse Test ...... 108 5.2.5 Drug Treatments did Not Alter Spontaneous Novelty-Induced Locomotion ...... 109 5.3 Discussion ...... 109 5.4 Limitations ...... 115 5.5 Future Directions...... 115 5.6 Conclusions ...... 115

6 GENERAL DISCUSSION ...... 121

6.1 Future Directions...... 122 6.2 Conclusions ...... 124

LIST OF REFERENCES ...... 126

BIOGRAPHICAL SKETCH ...... 160

LIST OF TABLES

Table page

1-1 Protein levels of mGlu5-associated proteins in the PrL on day 90...... 78

3-1 Protein levels of mGlu5-associated proteins in the PrL on day 45...... 90

LIST OF FIGURES

Figure page

2-1 Experimental timeline ...... 70

2-2 Extended access cocaine self-administration ...... 70

2-3 Delayed match-to-sample task performance...... 71

2-4 Nonmatch-to-sample task performance ...... 72

2-5 Context + cue relapse...... 72

2-6 Fluorescent in situ hybridization ...... 73

2-7 Cytochrome oxidase histochemistry ...... 74

2-8 Western blotting ...... 75

2-9 Delayed nonmatch-to-sample task performance ...... 76

2-10 Target ROI in the PrL selected for quantifying Arc mRNA expression...... 76

2-11 Representative immunoreactive bands of selected proteins on day 90...... 77

3-1 Experimental timeline ...... 86

3-2 Extended access cocaine self-administration ...... 86

3-3 Cytochrome oxidase histochemistry ...... 87

3-4 Western blotting ...... 88

3-5 Representative immunoreactive bands of selected proteins on day 45 ...... 89

4-1 Experimental timeline ...... 96

4-2 Delayed match-to-sample task performance ...... 96

4-3 Single delay DMS task performance ...... 96

4-4 Fluorescent in situ hybridization ...... 97

5-1 Experimental timeline ...... 117

5-2 Extended access cocaine self-administration ...... 117

5-3 Delayed match-to-sample task performance with drug treatments ...... 118

5-4 Context + cue relapse with drug treatments ...... 119

5-5 DMS Retesting of CDPPB, Vehicle and MTEP treated rats ...... 120

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

NEUROBIOLOGICAL SUBSTRATES OF POST-COCAINE COGNITIVE FUNCTION AND DRUG SEEKING: ROLE OF METABOTROPIC GLUTAMATE RECEPTOR 5

Christina M. Gobin

August 2019

Chair: Marek Schwendt Major: Psychology

Cocaine use disorder is associated with prefrontal cortex dysfunction and cognitive deficits that contribute to relapse susceptibility. Here, I used a rodent model to characterize cognitive deficits in a delayed match to sample task (DMS) after extended access cocaine self-administration. I also investigated immediate and long-term functional changes within the prelimbic cortex (PrL) in relation to cognitive performance and drug seeking. I found that cocaine rats displayed working memory impairments during a prolonged abstinence window, and cognitive performance correlated with later drug seeking. The cocaine group also demonstrated inefficient PrL activation and metabotropic glutamate receptor 5 (mGlu5) signaling in relation to cognitive performance and drug seeking as indicated by (1) greater metabolic activity related to working memory performance (2) mGlu5 dependent hyperactivity in response to cocaine-associated cues and (3) greater protein expression of the mGlu5 monomer and

Homer 1b/c. A separate group of cocaine rats without a history of cognitive testing did not demonstrate changes in mGlu5 and Homer 1b/c protein expression compared to saline rats, suggesting that post-cocaine cognitive dysfunction exhibited during a cognitive training period upregulated these proteins. Further, a group of naïve rats who

underwent only DMS testing exhibited mGlu5 dependent activation within the PrL that supported high versus low working memory demand. Finally, treatment with mGlu5 negative or positive allosteric modulators (NAM and PAM) was assessed on both working memory and drug seeking. I found that the mGlu5 NAM MTEP and the PAM

CDPPB both acutely decreased drug seeking, and repeated treatment with these compounds impaired or mildly enhanced working memory respectively. Taken together, these findings suggest post-cocaine PrL dysfunction that confers mGlu5 dependent hyperactivity to drug cues as well as compensatory metabolic activity and mGlu5 signaling to support working memory performance. As mGlu5 dependent activation in the PrL supported both drug seeking and working memory performance, treatment with

MTEP decreased cocaine seeking at the cost of cognitive function, suggesting that repeated inhibition of mGlu5 signaling may not be an ideal target. Rather, the mechanisms underlying the anti-relapse effects and mild pro-cognitive effects of

CDPPB should be further investigated.

CHAPTER 1 INTRODUCTION

1.1 Cocaine Addiction

Cocaine use disorder (CUD) is a chronic relapsing condition precipitated by maladaptive incentive salience, maintained by persistent craving and compounded by neuroplastic changes in brain reward and executive function that exacerbate resistance to abstinence (Koob & Volkow, 2010, 2016). These behavioral dysfunctions have been attributed to dysregulation within the mesocorticolimbic system. Limbic structures including the ventral tegmental area (VTA), nucleus accumbens (NAc), amygdala and hippocampus are involved in incentive salience attributed to drug reward and craving, whereas the prefrontal cortex (PFC) plays a role in modulating these limbic regions and maintaining higher order cognitive control (Connolly et al., 2012; Goldstein & Volkow,

2011). As such, PFC dysfunction may underlie persistent drug seeking (Goldstein &

Volkow, 2002, 2011). This chapter will specifically focus on the role of PFC dysfunction and accompanying cognitive impairments on hindering abstinence. A special focus on how glutamate is involved in cognitive function and drug seeking is explored.

1.1.1 Cocaine and Cognitive Dysfunction

Using neuropsychological test measures, cognitive deficits across multiple domains (attention, impulsivity, executive function, working memory, decision-making, problem solving, set-shifting, and cognitive flexibility) have been identified in both recreational and dependent users, demonstrated to be more pronounced with years of cocaine use and earlier age of onset, and shown to vary with length of abstinence

(Albein-Urios et al., 2012; Cunha et al., 2013; Ersche et al., 2011; Goldstein et al., 2004;

Hoff et al., 1996; Jovanovski, Erb, & Zakzanis, 2005; Kelley et al., 2005; Kübler,

Murphy, & Garavan, 2005; Pace-Schott et al., 2008; Potvin et al., 2014; Vonmoos et al.,

2013, 2014). For example, current cocaine use has been suggested to acutely normalize cocaine-induced cognitive impairments that resurface further into abstinence

(Pace-Schott et al., 2008), persist up to six months, and recover one year post cocaine

(Vonmoos et al., 2014). Additionally, metanalyses of studies assessing post-cocaine cognitive function reveal greatest impairments in attention, working memory, impulsivity, and verbal memory during intermediate abstinence (≤12 weeks), that presents as less pronounced in current users or users in prolonged abstinence (≥20 weeks) (Potvin et al., 2014). Nevertheless, regardless of abstinence lengths, deficits in executive function, learning, attention, and memory have been established in both recreational or dependent cocaine users with the most robust impairments exhibited in measures of attention and working memory (Jovanovski et al., 2005; Potvin et al., 2014; Vonmoos et al., 2013).

1.1.2 Cocaine and PFC Dysfunction

Supplementing the findings of cocaine-associated cognitive dysfunction, imaging studies of abstinent cocaine users suggest PFC hypofunction may contribute to cognitive impairments thwarting successful recovery (Goldstein et al., 2004; Goldstein &

Volkow, 2002). Most studies, with few exceptions (Garza-Villarreal et al., 2017;

Narayana et al., 2010), have identified hypofunction of several PFC regions as abnormalities in glucose metabolism and cerebral blood flow as well as reduced density, volume or grey matter, that has demonstrated to be more pronounced with years of cocaine use and persistent up to 4 months of abstinence (Connolly et al., 2013;

Ersche et al., 2011; Fein, Di Sclafani, & Meyerhoff, 2002; Goldstein & Volkow, 2002; Liu et al., 1998; Mackey & Paulus, 2013; Matochik et al., 2003; O’ Neill, Cardenas, &

Eyerhoff, 2001; Tanabe et al., 2009; Tumeh et al., 1990; Volkow et al., 1988; Volkow et al., 1992; Weber et al., 1993).

Additionally, structural changes have been observed within cocaine dependent samples in relation to length of abstinence (Bell et al., 2011). Decreased grey and white matter frontal cortical densities observed in cocaine users compared to controls, are more pronounced for current users compared to abstinent users (1-16 weeks). These differences have been hypothesized to reflect a recovery of cortical deficits or reassertion of cognitive control to sustain abstinence (Connolly et al., 2013; Hanlon et al., 2011). Similarly, studies assessing regional cerebral blood flow as a measure of metabolic activity report analogous recovery of reduced metabolic activity with increased abstinence (Kosten et al., 2004). However, opposite directionality has been reported with greater white matter integrity exhibited in current users compared to abstinent users. This observation has been suggested to represent an over utilization of white matter tracts to support craving or compensate for loss of function in other affected areas, of which abstinence may normalize this white matter (Bell et al., 2011).

In support of this hypothesis, current users exhibit greater white matter integrity within a region demonstrated to be hyperactive upon exposure to cocaine (Garavan et al., 2000;

Romero et al., 2010). Duration of cocaine use is also an important factor when interpreting these volumetric findings as more years of cocaine use is associated with decreased white and gray matter volumes (Connolly et al., 2013; Lim et al., 2008; Xu et al., 2010).

The functional significance of these neuroimaging findings is highlighted when correlated with neuropsychological or behavioral measures. Accordingly, several

studies reveal a positive relationship between PFC hypofunction and measures of impulsivity, compulsivity and attentional impairments (Camchong et al., 2011; Ersche et al., 2011; Romero et al., 2010). Further, investigation of neural activation during cognitive testing elucidates how cortical abnormalities may translate into cognitive dysfunction. In cocaine dependent individuals, hypoactivation of the cingulate cortex has been demonstrated during inhibitory control tasks (Bolla et al., 2004; Hester &

Garavan, 2004; Kaufman et al., 2003) and attention switching tasks (Kübler et al.,

2005). Additionally, hyperactivation among PFC regions related to task performance has been identified as a possible compensatory mechanism for individuals with CUD

(Bolla et al., 2003; Bolla et al., 2004; Connolly et al., 2012). Moreover, differential compensatory mechanisms may support different stages of abstinence. It has been interpreted that PFC hyperactivity compensates inhibitory control in short-term (1-5 weeks) abstinent users and behavioral monitoring in long term abstinent (40-102 weeks) users (Connolly et al., 2012).

The relationship between cortical activity and cognitive measures can be further elucidated by utilizing tasks that distinguish cognitive performance during easy versus hard phases of the same measure, such as working memory tasks wherein cognitive load can be manipulated. For example, during the n-back working memory task, activation in occipital, dorsolateral PFC (dlPFC) and parietal cortices have been identified (Tomasi & Caparelli, 2007) with increased activation of these regions supporting increasing load demands for healthy subjects (Tomasi et al., 2007a).

However, different activation patterns emerge for cocaine dependent individuals, who demonstrate significant impairments in this task, wherein they exhibit greater cortical

activation during overall task performance but less cortical activation during increasing working memory load demands compared to controls (Moeller et al., 2010; Tomasi et al., 2007b). This has been interpreted as neural inefficiency for cocaine dependent individuals in which they recruit larger network resources for working memory in low load versus high load conditions and exhibit hyperactivation during overall task performance as a compensatory attention mechanism for inefficient executive function.

Further, this aberrant working memory task activation was shown to be more pronounced in current versus non-current users (positive versus negative urine tests) for whom working memory impairments were more severe. This may suggest that neural inefficiency related to compromised working memory contributes to delayed abstinence.

1.1.3 Cognitive Deficits and Relapse

PFC dysfunction coinciding with cognitive impairments and varying with length of abstinence partially support the hypothesis that compromised higher order cognitive control may exacerbate resistance to abstinence. Further evidence can be obtained from studies assessing these measures in relation to treatment outcomes. Indeed, cognitive impairments in measures of attention, memory, inhibitory control and decision making at the time of treatment onset have been shown to predict poor treatment retention in cognitive behavioral therapy (or CBT) programs (Aharonovich et al., 2006;

Streeter et al., 2008) as well as relapse rates three months later (Verdejo-Garcia et al.,

2014). Additionally, greater white matter integrity within the frontal lobes of cocaine users at treatment onset predicted longer durations of abstinence during community based or CBT treatment programs, suggesting that greater white matter integrity supports cognitive control needed to maintain abstinence (Xu et al., 2010).This is further

evidenced by poor frontal lobe white matter integrity correlating with impulsivity and cognitive control impairments, indicative of unsuccessful treatment outcomes (Moeller et al., 2001; Moeller et al., 2005).

Activation during working memory tasks or tasks of inhibitory control has also been reported to predict treatment outcomes. Decreased thalamic activation during working memory task performance has been demonstrated to correlate with poor treatment outcomes (Moeller et al., 2010). Additionally, activation of cortico-striatal regions during Stroop task performance has been shown to correlate with greater duration of abstinence and treatment retention, emphasizing the relationship between the ability to exert cognitive control (mediated by recruitment of relevant neural resources) and successful abstinence (Brewer et al., 2008).

1.1.4 Section Summary

PFC dysfunction is expressed as cognitive deficits in cocaine dependent individuals and shown to be exacerbated by years of cocaine use, persistent up to several months, but remedied with prolonged abstinence. The importance of investigating PFC functional changes and accompanying cognitive deficits is highlighted by their relationship with poor treatment outcomes. Of interest is working memory since impairments in this domain are most robust in cocaine dependent individuals regardless of patterns of drug use and length of abstinence with a more difficult recovery of function in this domain for individuals with an early onset of cocaine use (Vonmoos et al., 2014). Additionally, tasks used to assess this domain enable the interrogation of

PFC processing in relation to easy versus difficult components of this measure that predicts poor treatment outcomes. Thus, working memory is an important domain to investigate in relation to cocaine relapse susceptibility.

1.2 Working Memory

Working memory requires the ability to temporally maintain and manipulate information in the absence of sensory input (relevant to the stored representation) that can be used to guide behavior (Cowan, 2008). Increasing working memory load demands either by increasing the delay length of the retention period or increasing the number of items to be remembered over time requires enhanced sustained attention and decreased responsivity to distracting stimuli (Simon et al., 2016). As such, intact attentional processes have been considered the basis for working memory (Cowan,

1997; Eriksson et al., 2015). Because working memory requires the interplay of selective attention and temporal storage in order to execute appropriate responses, it has been regarded as the foundation for most high order cognitive mechanisms such as reasoning, problem solving and language comprehension (Baddeley, 1986; Baddeley &

Hitch, 1974; Just & Carpenter, 1992). Indeed, working memory impairments covary with deficits in decision making and impulsivity (Bechara & Martin, 2004; Whitney, Jameson,

& Hinson, 2004).

1.2.1 Delayed Match-to-Sample Tasks Can be Used to Investigate Working Memory

One of the most commonly studied behavioral tasks to assess working memory in both humans and animals is the delayed match-to-sample task (DMS), and it has been regarded as a robust and sensitive task to identify differences in neurocognitive function within and between both normal and abnormal populations (Daniel, Katz, &

Robinson, 2016). Operant DMS tasks involve presentation of a stimulus followed by a delay period and a subsequent choice phase, wherein the selection of a matching stimulus is required for a reward. The delay period can be adjusted to control the

difficulty of the task with increasing delays requiring greater working memory capacity.

An advantage of the DMS task is that it can be similarly employed in non-human animals to assess working memory. Particularly, operant DMS tasks can be utilized to probe an animal’s working memory capacity of arbitrary events separate from aspects of cognitive function underlying performance in tasks with biological significance such as foraging for food in maze tasks or memory for a conspecific (Lind, Enquist, &

Ghirlanda, 2015).

1.2.2 The Dorsolateral PFC in Humans and Non-Human Primates is Implicated in Working Memory

Initial studies on monkeys have identified the involvement of the PFC in working memory. PFC lesions were shown to impair working memory (Jacobsen & Nissen,

1937), and recordings within the PFC demonstrated neuronal activity when information was being held over a delay period (Fuster, 1973). Moreover, the maintenance of working memory has been suggested to be mediated by a distributed network of frontal- striatal and thalamic regions (Ashby et al., 2005). Particularly, the dorsolateral PFC

(dlPFC) has been implicated in interconnecting these regions to support working memory (Curtis & D’Esposito, 2003; Daniel et al., 2016; Goldman-Rakic, 1995; Owen et al., 2005; Wager & Smith, 2003).

Humans with dlPFC lesions display working memory deficits (Barbey, Koenigs, &

Grafman, 2013; Manes et al., 2002; Tsuchida & Fellows, 2009). Evidence from neurophysiological unit recordings as well as fMRI studies have demonstrated persistent and sustained neuronal firing within this region during the retention phase of working memory tasks (Chafee & Goldman-Rakic, 1998; Jha & McCarthy, 2000; Kubota

& Niki, 1971; Leung, Gore, & Goldman-Rakic, 2002; Miller, Erickson, & Desimone,

1996; Sakai, Rowe, & Passingham, 2002; Zarahn, Aguirre, & D’Esposito, 1999).

Similarly, stimulation of the dlPFC using anodal transcranial direct current stimulation during working memory tasks has been shown to improve performance (Fregni et al.,

2005; Mulquiney et al., 2011; Ohn et al., 2008). Correspondingly, decreased activity within the dlPFC has been shown to coincide with impaired working memory performance (Pessoa et al., 2002; Sakai et al., 2002).

Understanding how dlPFC activation amidst the retention phase may support working memory has been further dissected by investigating how manipulation of working memory load affects this region’s maintenance of information. In studies varying delay lengths during the retention phase, activity within the dlPFC has been shown to be sustained throughout the entire delay length, suggesting that this region actively maintains representations long enough to guide behavior (Jha & McCarthy,

2000; Leung et al., 2002; Sakai et al., 2002; Zarahn et al., 1999). Additionally, studies lesioning this area, demonstrate delay-dependent impairments across multiple tasks used to measure working memory (Bauer & Fuster, 1976; Funahashi, Bruce, &

Goldman-Rakic, 1993). Overall, the general consensus within the literature is that the dlPFC supports online maintenance or manipulation of relevant internal and external representations to guide behaviors (Cabeza & Nyberg, 2000; Eriksson et al., 2015;

Fletcher & Henson, 2001; MacDonald et al., 2000; Petrides, 2000).

1.2.3 The Prelimbic Cortex in Rodents is Homologous to the Human dlPFC to Investigate Working Memory

The use of rodents to model behavioral and neural pathologies in addiction are of high clinical importance to investigate treatments that can restore dysfunction. As the dlPFC is a target of interest to assess higher order cognitive function, targeting the

analogous brain region in rodents is necessary to further understand neuroadaptations within translatable behavioral models. Indeed, a subregion of the dorsomedial PFC

(dmPFC), the prelimbic cortex (PrL), has been characterized in rodents as an anatomical and functional homologue to the human dlPFC (Vertes, 2004, 2006). Similar to the dlPFC, the PrL receives afferents from the medial dorsal thalamus (MT), hippocampus, and amygdala, and sends efferents to the NAc (Ongur & Price, 2000;

Uylings, Groenewegen, & Kolb, 2003). However, because the rodent cortex is agranular, it has been debated whether rodent models can provide useful information about cogntively complex human dlPFC function (Preuss, 1995). Nevertheless, the design of elegant behavioral tasks combined with lesion and electrophysiological studies provide a general consensus of the involvement of this region in higher order cognitive functions in rodents (Farovik et al., 2008; Granon & Poucet, 2000; Ongur &

Price, 2000; Perry et al., 2011; Seamans, Lapish, & Durstewitz, 2008; Uylings et al.,

2003).

Intact PrL function has been identified as necessary for optimal cognitive control and working memory performance (Vertes, 2004). As seen in the homologous human dlPFC, increased firing synchronization within PrL neurons during the delay period in a working memory task has been implicated in the encoding and representation of working memory (Jung et al., 2000; Jung et al., 1998; Yang et al., 2014). Additionally,

PrL neurons displaying dendritic atrophy as well as thin spine loss coincide with working memory deficits (Hains et al., 2009; Radley et al., 2015). As such, lesioning the mPFC/PrL or inactivating the PrL with muscimol has been shown to impair working memory performance in delayed alternation tasks or operant DMS tasks (Sloan, Good,

& Dunnett, 2006; van Haaren et al., 1988; Yang et al., 2014). Furthermore, the PrL’s involvement in working memory may be mediated by the MT as inactivation of either of these regions has been shown to impair working memory (Romanides, Duffy, & Kalivas,

1999).

1.3 The Role of the dlPFC and PrL in Cocaine Addiction

1.3.1 The Role of the Human dlPFC in Cocaine Seeking

In addition to its role in working memory, activation of the dlPFC has been associated with reactivity to cocaine-associated cues and craving in humans (Grant et al., 1996; Jasinska et al., 2014; Kober et al., 2016; Contia et al., 2016; Maas et al.,

1998). Interestingly, drug cue induced activation of this region may depend on whether the cocaine users are currently seeking treatment (Prisciandaro et al., 2014; Wilson,

Sayette, & Fiez, 2004). A review of 18 studies assessing cue-elicited craving using fMRI concluded that treatment seeking participants do not exhibit cue-elicited activation of the dlPFC (Wilson et al., 2004). Treatment-seeking participants may be exerting inhibitory control mechanisms when exposed to cues in order to maintain abstinence, whereas non-treatment seekers may be demonstrating greater cue-elicited activation because of anticipated use of the drug after the study. Thus, increased activation of the dlPFC may be involved in maintaining information about drug expectancy, planning and goals to obtain the drug, whereas these neural representations would not be relevant for treatment seekers who were not planning to later seek the drug. Indeed, a study assessing cue-elicited activation specifically in treatment seekers versus non-treatment seekers showed that untreated participants showed more reactivity in the dlPFC to cocaine-associated cues, and lower cue-induced dlPFC activation was correlated with increased motivation to positively change drug use (Prisciandaro et al., 2014). Thus, the

dlPFC may be an important region regulating reactivity to cocaine-associated cues to support successful abstinence. As such, modulation of this region with transcranial direct current stimulation or transcranial magnetic stimulation have been shown to reduce cocaine craving (Batista et al., 2015; Klauss et al., 2018) as well as cocaine use

(Terraneo et al., 2016).

1.3.2 The Role of the Rodent PrL and Cocaine Seeking

The PrL cortex receives dopaminergic projections from the VTA and has high expression of the dopamine transporter (DAT), suggesting the involvement of this region in response to cocaine (Jasinska et al., 2015). Additionally, the PrL primarily sends glutamatergic projections to the NAc core (Brog et al., 1993; Pennartz,

Groenewegen, & Lopes da Silva, 1994; Sesack & Pickel, 1992). Inactivation or activation of the PrL-NAc core circuit has been shown to respectively suppress or promote cocaine, context or cue-elicited cocaine seeking (Di Pietro, Black, & Kantak,

2006; Fuchs et al., 2005; Gipson et al., 2013; Knackstedt, Trantham-Davidson, &

Schwendt, 2014; McFarland & Kalivas, 2001; McFarland, Lapish, & Kalivas, 2003;

McLaughlin & See, 2003; Rebec & Sun, 2005; Stefanik et al., 2013; Wang et al., 2013;

West et al., 2014; Zavala et al., 2008). Furthermore, it has been demonstrated that neuronal activity in the PrL encodes information about the availability of the drug in response to a cue, whereas neuronal activity within the NAc encodes cocaine reward

(Rebec & Sun, 2005). Additionally, enhanced recruitments of PrL neurons has been exhibited during drug seeking upon reexposure to cocaine cues as well as during the resumption of cocaine self-administration (West et al., 2014). This increased PrL activity was only apparent for rats who had learned the cue association with the cocaine, suggesting a specific stimulus evoked responsiveness of this region to drug-associated

cues. Importantly, this pattern of PrL cell firing was demonstrated to be more robust further into abstinence, coinciding with increased drug seeking in response to cocaine cues. This phenomenon has been characterized as an incubation of cocaine craving effect wherein drug seeking in response to cocaine-associated cues linearly increases up to 3 months into abstinence (Freeman et al., 2008; Grimm et al., 2001), and the cue- elicited recruitment of PrL neural signaling has been implicated in this process (Pickens et al., 2011).

However, it should be noted that this role of the PrL is suggested when using traditional self-administration and cue-evoked reinstatement or relapse tests (Moorman et al., 2015). Studies using different methods have found PrL activity supports suppression of cocaine seeking when restraint is the contextually-appropriate behavior such as when cocaine seeking is paired with a stop signal (Mihindou et al., 2013;

Navailles et al., 2015), cocaine-paired foot shocks are used to inhibit responding

(Limpens et al., 2014), high-frequency self-administering rats are assessed (Martín-

García et al., 2014), or when a model of punishment-restraint cocaine seeking is used

(Chen et al., 2013).

For example, Martin-Garcia and colleagues (2014) found that high frequency compared to low frequency cocaine taking rats, showed greater neuronal activation in the PrL and NAc core as well as a robust positive correlation in activation between these regions after cocaine-induced reinstatement. This implicates the involvement of this circuit under conditions where relapse susceptibility is high. As such, optogenetic inactivation of the PrL reduced cocaine seeking in these rats after reinstatement.

However, the opposite effect of PrL optogenetic inactivation occurred in which a

potentiation of cocaine seeking during cocaine self-administration was observed. Thus, highly differential involvement of this region in cocaine taking compared to reinstated cocaine seeking may be mediated by frequency of drug use. Of note is that PrL inactivation was administered after a few cocaine self-infusions, and the potentiation of cocaine taking was observed during timeout periods, suggesting the role of the PrL in learning and controlling non-adaptive operant responses. As such, inactivation of this region following a few pairings might inhibit recruitment of this region to mediate these adaptive responses and result in a loss of control.

In another study employing a fear conditioning scenario wherein rats intermittently received a foot shock during cocaine self-administration, a similar role of the PrL was suggested (Chen et al., 2013). Using this paradigm, the emergence of shock-resistant and shock-aversive phenotypes was used to characterize compulsive seeking in rats (seeking despite a negative consequence). In shock- resistant/compulsive cocaine taking rats, reduced excitability in the PrL was exhibited after cocaine self-administration paired with foot shocks. Within these shock-resistant rats, optogenetic stimulation or inactivation of the PrL was shown to reduce and enhance compulsive cocaine seeking respectively. In traditional cocaine self- administration paradigms, drug cues paired with cocaine-induced dopamine release in the PrL and NAc are posited to enhance learning of incentive salience (for a review, see

(Shackman et al., 2011)). However, in the fear-conditioning paradigm, neuronal activation in the PrL may be necessary to learn to inhibit responding when variable punishment is expected.

Overall, these studies implicate the involvement of the PrL to control drug seeking. The PrL mediates learning response contingencies as well as positive and negative reinforcement of cocaine associated cues or instrumental responding which can underlie the promotion or suppression of cocaine seeking.

1.4 The Role of Glutamate Receptors

Glutamate mostly exerts excitatory neurotransmission in the brain and plays a role in synaptic plasticity, the strengthening or weakening of synapses in response to activity patterns. These actions are mediated by ionotropic and metabotropic glutamate

(mGlu) receptors (Ozawa, Kamiya, & Tsuzuki, 1998).

Ionotropic glutamate receptors consist of N-Methyl-D-aspartate (NMDA), α- amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA), and kainite receptors and their respective subtypes. The ionotropic receptors form an ion channel pore and upon activation, they increase the flux of sodium (NA+), potassium (K+) and for some subtypes, calcium (Ca2+), promoting cell excitability. In particular, NMDA receptors play a vital role in experience dependent changes underlying the efficacy of synaptic transmission (Lüscher & Malenka, 2012). NMDARs are unique in that they contain a magnesium (Mg2+) block within the pore channel. Depolarization of the postsynaptic neuron to remove that block combined with simultaneous glutamate binding to the receptor are necessary for NMDARs to conduct currents. Long term potentiation (LTP) or the strengthening of synapses occurs when more glutamate is released from the presynaptic cells and strongly activates NMDARs, enabling robust increases in intracellular Ca2+, (Malenka, 1994). High concentration of Ca2+, particularly on the spines, initiates signaling cascades including the activation of Ca2+/calmodulin- dependent protein kinase II (CaMKII) which leads to the insertion of AMPARs on the

membrane (Ehlers, 2000). On the other hand, a smaller magnitude of glutamate release modestly activates NMDARs, inducing a lower amount of Ca2+ influx, initiating different signaling cascades including activation of protein phosphatases which contribute to the internalization of AMPARs and result in the weakening of synapses or long-term depression (LTD). Additionally, LTP and LTD induction can depend on sensitivity of the

NMDAR on the postsynaptic cell in response to the same amount of glutamate release.

mGlu receptors are coupled via G proteins to ion channels, and upon activation, several intracellular signaling cascades can ensue, modulating cell excitability (Schoepp

& Conn, 1993).There are eight subtypes of mGlu receptors that have been classified into three groups based on sequence homology, ligand selectivity, and G protein coupling (Wang & Zhuo, 2012). Group I mGlu receptors (mGlu1 and mGlu5) are mostly located postsynaptically and activate via Gαq/G11 to stimulate phospholipase Cβ

(PLCβ). Group II (mGlu2 and mGlu3) and Group III (mGlu4, mGlu6, mGlu7, mGlu8) mGlu receptors are predominantly presynaptic (with the exception of mGlu6) and mostly act via Gαi to inhibit adenylyl cyclase, activate K+ channels and inhibit Ca2+ channels

(Niswender & Conn, 2010). Group I mGlus mediate gene transcription and translation

(Gerber, Gee, & Benquet, 2007; Gladding, Fitzjohn, & Molnár, 2009) as well as protein synthesis (Waung et al., 2008), implicating these receptors in synaptic plasticity. In this dissertation, my focus is on mGlu5 as mGlu5 signaling has been investigated to elucidate its involvement in synaptic plasticity within the context of cognitive function as well as cocaine addiction.

The activation of the mGlu5 receptor is involved in cell depolarization and enhanced neuronal excitability. This receptor is coupled via Gαq/11 to stimulate PLCβ

which results in the hydrolysis of phosphoinositides and generation of inositol 1,4,5- triphosphate (IP3) and diacyl glycerol (DAG). IP3 binds to its respective receptors on the endoplasmic reticulum and Ca2+ is subsequently released intracellularly from its stores.

Both Ca2+ and DAG then activate protein kinase C (PKC). A range of downstream effectors and protein kinase pathways can be activated by mGlu5 depending on the cell type and neuronal population. Moreover, these receptors can modulate additional signaling pathways downstream Gαq as well as cascades resulting from Gαi/o or Gαs signaling mechanisms (Hermans & Challiss, 2001; Niswender & Conn, 2010).

Additionally, binding and phosphorylation at the intracellular C-terminal domain of mGlu5 by extracellular signal regulated protein kinase (ERK), CaMKII, protein kinase A

(PKA) and PKC can mediate synaptic plasticity and regulate mGlu5 surface expression

(Liu et al., 2006; Mao et al., 2008; Marks et al., 2018; Uematsu et al., 2015).

Furthermore, mGlu5 has been shown to potentiate NMDARs though Ca2+ regulated interactions with CAMKII, PKC dependent signaling and PKC activated src (Benquet,

Gee, & Gerber, 2002; Chen, Liao, & Chan, 2011; Huang et al., 2001; Jin et al., 2013)

Additionally, mGlu5 mediated synaptic plasticity can be altered via coupling with scaffolding and other signaling proteins. Namely, the Homer family of scaffolding proteins include the constitutively expressed long forms: Homer 1b/c, Homer 2a/b and

Homer 3 as well as the inducible immediate early gene (IEG) short form: Homer 1a.

Both forms of Homer proteins contain Enabled/vasodilator-stimulated phosphoprotein homology 1 (EVH1) domains, but only the long forms of Homer contain a structurally unique coiled-coil domain comprised of EVH1 domains on each end that bind to the C- terminus of mGlu5 receptors, Shank proteins, and IP3 receptors, enabling the formation

of multimeric complexes with dimers of Shank protein and coupling of mGlu5 receptors with intracellular Ca2+ stores. On the other hand, the EVH1 domains on Homer 1a compete with those on the long forms of Homer to bind the C-terminus on those proteins and receptors and thus disrupt this physical coupling and alter mGlu5 dependent Ca2+ signaling (kinetics and peak amplitude) (for review: see (Fagni, Worley,

& Ango, 2002). Additionally, Homer 1b and Homer 1c indirectly couple mGlu5 to

NMDARs via Shank glutamate receptor complexes at the post synaptic density 95

(PSD95). Homer 1a competes for and disrupts this binding, and neuronal activation- induced expression of Homer 1a can directly couple mGlu5 to NMDA, wherein mGlu5 can inhibit rather than synergize with NMDA (Moutin et al., 2012). The competition between long and short Homer forms is specific to within dendritic spines, highlighting the role of Homer isoforms on mGlu5 receptors to control synaptic transmissions and remodeling of dendritic spines. Furthermore, Homer 1b and Homer 1c are involved in localizing mGlu5 receptors from the soma to the postsynaptic site, maintaining intracellular retention of this complex. However, as Homer 1a competes for the C terminus domain of mGlu5, upon neuronal activation, induction of Homer 1a triggers membrane expression of mGlu5.

1.4.1 The Role of Glutamate Receptors in Working Memory

As previously mentioned, maintained neuronal firing of pyramidal cells within the dlPFC for humans and primates or within the PrL for rodents supports working memory performance by holding relevant information online long enough to guide behaviors.

Investigating neurotransmitters modulating the afferents and efferents of these regions as well as receptor subtypes has provided information about the signaling mechanisms that may support this type of learning and memory.

As the PrL cortex receives glutamatergic input from the hippocampus and MT as well as dopaminergic input from the VTA, dopamine and glutamate signaling have been investigated and demonstrated to be involved in the recruitment of the dlPFC for working memory. For example, glutamate-induced metabolic activation as well as increases in extracellular dopamine within the dlPFC have been observed during working memory performance (Phillips, Ahn, & Floresco, 2004; Watanabe, Kodama, &

Hikosaka, 1997; Woodcock et al., 2018). Within the PrL, dopamine modulates neuronal activity via D1 receptors (Williams & Goldman-Rakic, 1995) and these D1 receptors can enhance neuronal excitability within the PrL via potentiation of NMDA receptors

(Wang & O’Donnell, 2001). Alternatively, D1 receptors can inhibit PrL neuronal excitability through interacting with Group I mGlus to induce LTD (Otani et al., 1999).

As such, a balance of dopamine and glutamate tone within the PrL is necessary to support working memory performance (Homayoun et al., 2004; Romanides et al.,

1999).

It is important to note that unlike enduring changes in synaptic strength important for LTP and LTD, the synaptic plasticity subserving working memory in the PFC has been demonstrated to be rapid and transient (Wang et al., 2013). The PFC pyramidal cells interconnect with dendritic spines wherein signaling cascades can rapidly induce the strengthening or weakening of network connectivity with neighboring synapses of

PFC neurons. This contrasts the involvement of dendritic spines in regulating long-term changes in synaptic strength seen for long-term hippocampal learning (Bourne 2007).

Particularly, the rapid strengthening of synapses generates the requisite sustained network connectivity to retain relevant representations (Monaco, Gulchina, & Gao,

2015), whereas the rapid weakening of network connections to the hippocampus thereafter fine tunes efficient control over temporary rather than long-term memory storage of the information (Arnsten et al., 2010; Eriksson et al., 2015; Laroche, Davis, &

Jay, 2000).

NMDARs have been shown to modulate this necessary strengthening and weakening of synapses (Wang et al., 2013). Accordingly, blockade of NMDARs has been shown to impair working memory and reduce connectivity of the PFC network

(Driesen et al., 2013; Romanides et al., 1999; Wang et al., 2013), whereas activating

NMDARs has been shown to improve working memory (McQuail et al., 2016). However, targeting ionotropic receptors is not ideal because it can produce adverse side effects and be potentially fatal due to their widespread distribution, fast kinetics and direct involvement in excitatory synaptic transmission (Cleva & Olive, 2012; Homayoun &

Moghaddam, 2006).

Alternatively, mGlu5 receptors have been investigated (Hovelsø et al., 2012) and shown to be involved in NMDAR dependent synaptic plasticity involved in learning and memory (Riedel, Platt, & Micheau, 2003; Simonyi, Schachtman, & Christoffersen, 2005;

Simonyi, Schachtman, & Christoffersen, 2010). Additionally, mGlu5 receptors are mostly located perisynaptically on dendritic spines where they are functionally coupled to NMDAR through Homer scaffolding (Romano et al., 1995; Shigemoto et al., 1997,

1993). Through potentiation of NMDARs, mGlu5 can modulate LTP and enhance various types of learning (Chen et al., 2011; Cleva et al., 2010; Ganella et al., 2016;

Gass & Olive, 2009; Lu et al., 1997; Manahan-Vaughan & Braunewell, 2005). Though most of these studies have investigated mGlu5-mediated synaptic plasticity within the

hippocampus in relation to learning, activation of mGlu5 similarly enhances NMDAR responses in the PFC (Attucci et al., 2001) to support working memory (Homayoun et al., 2004). Through its interaction with NMDA, mGlu5 exerts modulatory effects on cognitive processing such as working memory based on changes in PFC neuronal activity and can fine-tune the temporal pattern of neuronal firing within the PFC

(Homayoun & Moghaddam, 2006). For example, inhibition of mGlu5 can reduce burst firing activity of mPFC neurons but also potentiate NMDAR blockade-effects on firing rate, whereas the opposite has been exhibited using a positive allosteric modulator

(PAM) of mGlu5 (Homayoun & Moghaddam, 2006; Lecourtier et al., 2007). Accordingly, positive allosteric modulation of mGluR5 has been shown to be effective in remedying cognitive dysfunction via potentiation of NMDARs through fine tuning of PFC neuronal output (Cleva & Olive, 2011; Homayoun et al., 2004; LaCrosse et al., 2014). Yet, a recent study used a biased mGlu5 PAM that selectively potentiates coupling to Gαq rather than NMDAR and demonstrated that mGlu5 can also enhance working memory and executive function independent of modulation of NMDAR signaling (Rook et al.,

2015), highlighting the complexity of this receptor in modulating cognitive processing.

1.4.2 The Role of Glutamate Receptors in Cocaine Seeking

As mentioned previously, the mesocortiolimbic circuit is integral for reward learning and heavily implicated in the emergence of cocaine addiction. Cocaine elevates dopamine transmission in the NAc and PFC from the VTA. Repeated dopamine release within these regions from repeated drug use underlies attributing salience to the drug reward and learning behaviors to obtain it. On the other hand, glutamatergic afferents from the PFC and amygdala to the NAc are involved in the expression of addictive behaviors (Kalivas & O’Brien, 2008; Kelley, 2004). For example,

enhanced glutamate release from the PFC onto the VTA contributes to transient LTP in the VTA dopamine cells (Ungless et al., 2001) relevant for the development of drug self- administration (Jones & Bonci, 2005). Additionally, enhanced glutamate release from the PFC to NAc contributes to glutamate related cellular alterations such as enhanced

AMPA sensitivity on postsynaptic NAc cells (Kalivas & Hu, 2006) contributing to drug seeking (Suto et al., 2004). Further, upregulation of surface expression of the AMPA receptor subtype GLUR1 on the NAc has been shown to coincide with an incubation of cocaine craving effect (Conrad et al., 2008) and a resting LTP-like state during withdrawal, indicated by increased AMPA/NMDA ratio (Kourrich et al., 2007) and increased number of spines in the NAc (Robinson & Kolb, 2004). Consequently, this may prime or subserve robust drug learning, craving, difficulty extinguishing from drug cues, and resistance to abstinence. Accordingly, following cocaine cessation of several weeks, reexposure to cocaine, cues or stress increases glutamate release from the PrL to the NAc core which reinstates cocaine seeking behavior (Capriles et al., 2003;

Cornish & Kalivas, 2000; Di Ciano & Everitt, 2001; McFarland et al., 2004; McFarland et al., 2003; McLaughlin & See, 2003).This increase in glutamate induces a rapid postsynaptic change altering glutamate signaling and resulting in the emergence of an

LTD-like state.

In addition to alterations in glutamate receptor function and signaling after cocaine, dysregulated glutamate homeostasis occurs wherein the cystine-glutamate antiporter, responsible for exchanging extracellular cystine for intracellular glutamate becomes downregulated and results in decreased basal levels of glutamate in the NAc during withdrawal (Baker et al., 2003). Decreased basal glutamate levels provide less

tonic activation of the presynaptic mglu2/3 autoreceptors. With less inhibitory feedback, the activation of glutamatergic afferents from the PrL or BLA to the NAc in response to cocaine or cues is enhanced.

Taken together, these findings highlight that glutamate is integral in the development of drug taking and the expression of cocaine seeking behaviors.

Particularly, inhibiting glutamate along the PFC glutamatergic afferent to the NAc has revealed success in attenuating cocaine seeking in animal models (Capriles et al.,

2003; Knackstedt et al., 2014; Wang et al., 2013). As modulation of glutamate transmission is not ideal with ionotropic receptors, a special focus on targeting mGlu5 receptors has been considered for developing anti-relapse therapies (Olive, 2009).

Indeed, systemic or intra accumbens inhibition of mGlu5 has been demonstrated to limit cocaine reward and reduce cocaine seeking in response to cocaine, cues or stress triggers, broadening the appeal of mGlu5 inhibitors as therapeutics for cocaine relapse

(Chiamulera et al., 2001; Keck et al., 2013, 2014; Kenny et al., 2005; Knackstedt et al.,

2014; Knackstedt & Schwendt, 2016; Kumaresan et al., 2009; Lee et al., 2005; Li et al.,

2018; Martin-Fardon et al., 2009; Wang et al., 2013).

1.5 Conclusions

Cocaine addiction is associated with dysregulated glutamatergic transmission and cognitive impairments such as working memory that thwart successful abstinence.

Working memory relies on efficient glutamatergic transmission within the dlPFC/PrL to support increased neuronal firing, and positive allosteric modulation of mGlu5 exerts procognitive effects. On the other hand, robust glutamatergic transmission from the dlPFC/PrL elicits increased reactivity to drug cues that drive drug seeking, and rather inhibition of mGlu5 attenuates relapse. Thus, understanding the overlapping

neurobiological substrates of working memory and drug seeking within the PFC is important to investigate how repeated modulation of mGlu5 may be used to improve cognitive performance and inhibit drug seeking.

1.5.1 General Hypotheses

The general hypotheses of this dissertation are that deficits in working memory will be exhibited following chronic cocaine and will correlate with later drug seeking.

Alterations in PrL activity as well as mGlu5 signaling in relation to working memory deficits will be observed following chronic cocaine. Modulation of mGlu5 with a negative allosteric modulator (NAM) will decrease drug seeking but impair cognitive performance.

Alternatively, modulation of mGlu5 with a PAM will enhance cognitive performance without affecting relapse. To test these hypotheses, several aims were considered.

1.5.2 Dissertation Objectives:

1. Use of a rodent model to characterize working memory deficits in a DMS task after extended access cocaine self-administration.

2. Analyze markers of neuronal activity within the PrL in relation to working memory performance and drug seeking.

3. Assess mGlu5 protein signaling after chronic cocaine in relation to working memory performance and drug seeking.

4. Examine the effects of repeated modulation of mGlu5 with a NAM and PAM on working memory and drug seeking.

CHAPTER 2 NEURAL FEATURES OF POST-COCAINE COGNITIVE DEFICITS AND DRUG SEEKING

Cocaine use disorder is characterized by compulsive drug seeking, decreased ability to limit intake, and resistance to abstinence (American Psychiatric Association,

2013). A challenge to treat these behavioral deficits may be partially attributed to global cognitive impairment that has been reported in up to 30% of individuals with CUD

(Vonmoos et al., 2013). In particular, persistent dysfunction of the PFC and related deficits in higher order cognitive function such as attention, impulsivity, executive function, working memory, decision-making, problem solving, set-shifting, and cognitive flexibility have been identified in CUD (Albein-Urios et al., 2012; Cunha et al., 2013;

Ersche et al., 2011; Goldstein et al., 2004; Hoff et al., 1996; Jovanovski et al., 2005;

Kelley et al., 2005; Kübler et al., 2005; Pace-Schott et al., 2008; Potvin et al., 2014;

Vonmoos et al., 2013, 2014). Coinciding with these deficits, abnormalities in glucose metabolism (Goldstein & Volkow, 2002; Volkow et al., 1992), and cerebral blood flow

(Tumeh et al., 1990; Volkow et al., 1988; Weber et al., 1993) in the PFC, as well as lower cortical and gray matter volumes (Fein et al., 2002; Franklin et al., 2002; Liu et al.,

1998; Moreno-López et al., 2012) have been observed in abstinent cocaine users.

Critically, these cortical abnormalities may predict the probability of successful recovery.

For example, aberrant cortical activity (measured during cognitive testing) in abstinent human cocaine users has been shown to correlate with cognitive performance (Bolla et al., 2003; Camchong et al., 2011; Connolly et al., 2012; Moeller et al., 2010; Tomasi et al., 2007) and predict length of abstinence and treatment retention (Brewer et al., 2008).

And further, the severity of cognitive (decision making, attention, working memory, and executive function) impairments correlates with cocaine craving (Vonmoos et al., 2013),

mediates poor compliance to CBT programs (Aharonovich et al., 2006; Streeter et al.,

2008) and predicts relapse (Verdejo-Garcia et al., 2014). A subregion of the PFC in humans, the dlPFC, has been demonstrated to play a dual role in addiction, both attempting to maintain aspects of normal cognitive function as well as mediate craving and responses to drug cues in human drug users (Goldstein & Volkow, 2002, 2011;

Contia et al., 2016; Terraneo et al., 2016). As dlPFC dysfunction is believed to contribute to cognitive impairment and relapse susceptibility in CUD, development of translational animal models that explore neurobiological substrates of post-cocaine cognitive deficits and persistent cocaine-seeking (located within analogous brain circuits) is of high clinical importance. The goal of the current study was to test the hypothesis that chronic cocaine self-administration produces metabolic and glutamatergic changes within the rat PrL that coincide with decreased capacity of this brain region to support working memory function and/or to limit cocaine-seeking. To test this hypothesis, several critical considerations needed to be addressed: (1) Use of a validated rodent model of chronic cocaine use and relapse. Extended access cocaine self-administration in rodents is such a model, as it captures several key features of human CUD, such as escalation of cocaine intake (Allain & Samaha, 2018; Mandt et al.,

2015) and robust cocaine-seeking (Ferrario et al., 2005). Reactivity to cocaine- associated cues in this animal model persists (or even increases) during the initial 90- day period of abstinence and thus has been termed “incubation cocaine of craving”

(Freeman et al., 2008; Grimm et al., 2001; Lu et al., 2003) akin to the human condition of time-dependent increases in cue reactivity (Parvaz, Moeller, & Goldstein, 2016). (2)

Utilization of PFC-dependent cognitive tasks to evaluate covariation in post-cocaine

cognitive deficits and cocaine-seeking. Several studies to date reported impairments in working memory, reversal learning, or attention following extended access cocaine self- administration (Briand et al., 2008; Calu et al., 2007; Fijał et al., 2015; George et al.,

2008; Porter et al., 2011; Radley et al., 2015). However, none of these studies investigated whether the degree of post-cocaine deficits predicts the magnitude of cocaine seeking, as it has been observed in CUD.

Here, I utilized an operant version of the DMS task as the performance in this task depends on the functional integrity of the mPFC as demonstrated in rodents (Sloan et al., 2006). Importantly, this task can be modified to allow for the precise evaluation of working memory capacity (working memory load), as well as to assess cognitive flexibility (reversal learning) following a rule switch (Sloan et al., 2006; Yhnell, Dunnett,

& Brooks, 2016). Both of these cognitive parameters are impaired in cocaine users

(Albein-Urios et al., 2012; Cunha et al., 2013; Jovanovski et al., 2005; Kelley et al.,

2005; Kübler et al., 2005; Potvin et al., 2014; Vonmoos et al., 2013; 2014). (3)

Identification of neural markers that reflect immediate versus long-term functional changes in the PrL. In CUD, PFC dysfunction (Fein et al., 2002; Liu et al., 1998;

Matochik et al., 2003; O’ Neill et al., 2001; Tanabe et al., 2009) and corresponding behavioral deficits (Connolly et al., 2012; Moreno-López et al., 2012; Tanabe et al.,

2009; Tomasi et al., 2007b) have been detected after weeks or months of abstinence.

Therefore, neural markers to be used in animal studies investigating the consequences of chronic cocaine exposure (such as this study), need to capture changes in neural activity over this prolonged period. To that end, metabolic mapping using cytochrome oxidase (CO) histochemistry has been successfully utilized in animals to assess long-

term changes in brain activity (Bechard, Bliznyuk, & Lewis, 2017; Gonzalez-Lima &

Cada, 1994; Gonzalez-Lima & Jones, 1994; Riha, Rojas, & Gonzalez-Lima, 2011;

Spivey et al., 2011; Tanimura et al., 2011; Vélez-Hernández et al., 2014). Additionally, it is well established that intact glutamatergic function in the PFC maintains cognitive functioning in humans and rats (Dauvermann, Lee, & Dawson, 2017; Hernandez et al.,

2018; Homayoun & Moghaddam, 2010; Martinez et al., 2014; Woodcock et al., 2018;

Yuen et al., 2012). Dysregulated glutamatergic function (e.g. synaptic release or receptor expression) in the PFC has been shown to be disrupted in human cocaine users (Martinez et al., 2014) and involved in mediating responding to cocaine- associated cues in rats (Ben-Shahar et al., 2013; McFarland et al., 2003; Shin et al.,

2016). This study specifically focused on mGlu5 and its downstream signaling partners within the PrL because this receptor plays a key role in modulating glutamate release and enhancing pyramidal cell output in the mPFC (Kiritoshi et al., 2013) important for controlling behaviorally relevant synaptic plasticity in the cortex (Jew et al., 2013; Marek

& Zhang, 2008) that supports working memory performance (Hernandez et al., 2018;

Homayoun et al., 2004), cocaine seeking (Smith et al., 2016) and post-cocaine extinction learning (Ben-Shahar et al., 2013; Ghasemzadeh et al., 2011). In addition, abnormal numbers of this receptors have been reported in abstinent cocaine users

(Martinez et al., 2014). Finally, to assess mGlu5-dependent signaling during drug- seeking, I analyzed expression of activity-regulated cytoskeleton-associated gene (Arc) mRNA, specifically in mGlu5-positive cells in the PrL. Arc is an IEG activated in the frontal cortex during retrieval of cocaine-associated memories in rodents (Hearing et al.,

2008; Zavala et al., 2008; Ziółkowska et al., 2011), though its co-expression with mGlu5 after a relapse event has not been evaluated.

In summary, the goal of this study was to develop a comprehensive rodent model to investigate neurobiological substrates in the PrL that could mediate cognitive deficits and increased relapse vulnerability following chronic cocaine exposure.

2.1 Materials and Methods

2.1.1 Animals and Experimental Design

Adult male Sprague-Dawley rats (Charles River Laboratories; 275 g on arrival; N

= 28) were individually housed and maintained on a 12 hr reverse light/dark cycle (lights off at 0700), with ad libitum access to food and water, except as otherwise noted. After arrival, animals were first acclimated to the animal facility for at least one week prior to any manipulations. All animal procedures were approved by the Institutional Animal

Care and Use Committee of the University of Florida (IACUC) and were performed in accordance with the Guide for the Care and Use of Laboratory Animals. The overall experimental design is depicted in Figure 2-1 with individual behavioral procedures described below.

2.1.2 Surgical Procedures

All rats were implanted with jugular catheters and submitted to cocaine self- administration (or received saline) as previously described (Gobin & Schwendt, 2017) with some modifications. Briefly, rats were anesthetized with ketamine (87.5mg/kg, i.p.) and xylazine (5mg/kg, i.p.), and the jugular vein was implanted with Silastic tubing (Dow

Corning, Midland, MI). The tubing exited the vein via a subcutaneous incision made between the shoulder blades, where it connected to a stainless-steel cannula on the inside of a rat harness (Instech, Plymouth Meeting, PA). Rats were given a five-day

recovery period prior to self-administration during which they were treated with ketorolac

(3mg/kg) to provide analgesia for the first three days. Catheters were flushed daily with

0.1 ml of 100U/ml of heparinized saline. Following the recovery from surgery and for the reminder of the study, animals underwent a mild food restriction (15-20 g of food per day) to maintain animals at ~ 85% of their free-feeding weight (Carroll et al., 2002;

Gobin & Schwendt, 2017).

2.1.3 Extended Access Cocaine Self-Administration and Abstinence

Rats were randomly assigned to either cocaine or saline groups and underwent self-administration procedures using modified rat operant chambers (30 × 24 × 30 cm;

Med Associates, St. Albans, VT) equipped with two standard rat nose poke ports. The standard operant chambers were modified (mesh wire floor, striped wall pattern, vanilla scent) in order to provide a context that was distinct from the one used in the operant cognitive tasks as described in Section 2.1.4. Cocaine hydrochloride (NIDA Controlled

Substances Program; Research Triangle Institute, NC) was dissolved in 0.9% sterile physiological saline. Upon making a nose poke in the active port, rats assigned to the cocaine group received response-contingent 0.35 mg/100 ul infusions of cocaine on a fixed-ratio 1 (FR1) schedule of reinforcement. Each infusion length was 5s during which a light + tone cue was presented followed by a 20s ‘time-out’ period. During the time-out period, nose-poking in the active port was recorded but not reinforced. Nose-poking in the inactive port was recorded but never reinforced. Rats in the cocaine group underwent daily 1 hr (short access, ShA) sessions for 6 days, followed by daily 6 hr

(long access, LgA) sessions for 12 days. Rats assigned to the saline group, underwent

18 daily 2 hr sessions, during which they received 30 programmed infusions of saline paired with a 5s presentation of a light + tone cue. This was done in order to provide a

sensory and contextual experience like that of cocaine rats. After the conclusion of self- administration, animals experienced 90 days of abstinence during which they underwent daily behavioral testing as described below.

2.1.4 Operant Delayed Match-to-Sample/Non-Match-to-Sample Task

Rats were trained and tested in the DMS and delayed non-match-to-sample

(DNMS) tasks as previously described in (Beas, Setlow, & Bizon, 2013; Gobin &

Schwendt, 2017; Sloan et al., 2006) with few modifications. Daily training (or testing) sessions were conducted in standard rat operant chambers (30 × 24 × 30 cm; Med

Associates, St. Albans, VT) equipped with two levers. Rats received a single test session per day. Each session duration was 40 minutes and began with an illumination of the house light. This light remained on throughout the session, except for time-outs.

Within each session, individual trials were composed of three components - a sample component, a delay period and a choice component. In the sample component, a left or right lever was chosen at random by the computer such that each lever was presented approximately the same number of times in each session. Pressing the sample lever resulted in retraction of that lever, delivery of a sucrose pellet, and initiation of the delay interval with randomized delay durations. During the choice phase, both levers were extended, and a correct response of choosing the same lever presented in the sample phase resulted in delivery of a sucrose pellet. An incorrect response resulted in a 6s time-out period wherein no sucrose pellet was delivered, the house light was extinguished, and both levers were retracted. Rats were initially trained without any delays between the sample and choice component with a correction procedure to prevent the development of side biases prior to being trained at two delay sets: short delay set {0s, 1s, 2s, 3s, 4s, 5s, 6s} and intermediate delay set {0s, 2s, 4s, 8s, 12s,

16s}. Rats were required to reach a criterion of ≥ 80% correct for two consecutive days at each training set before progressing to the next training phase. Upon reaching criterion in the training delay sets, they entered the testing phase for two blocks of five days at the final delay set {0 s, 2s, 4s, 8s, 12s, 18s, 24s}. The series of variable delay intervals ranging from short to long in this task are reflective of low versus high working memory loads respectively and can be used to distinguish between delay dependent mnemonic deficits and non-specific impairments (Dunnett, 1993). High percent correct responding at short delays is indicative of task rule acquisition while compromised correct responding at increasing delay intervals (load demands) denote delay dependent deficits. On the other hand, impairments in performance regardless of delay length may designate impairments unspecific to the measure of working memory itself such as compromised motivation, attention, motor control or rule acquisition (Dunnett,

1993).

Next, rats were tested in the DNMS task involving a rule switch in which rats had to choose the lever opposite from the sample lever for a sucrose reward. Initially, six rats with a history of saline underwent testing in the DNMS that included randomized presentation of all delays {0s, 2s, 4s, 8s, 12s, 18s, 24s}. This version of the task was based on previously published procedures using male Lister-Hooded rats (Sloan et al.,

2006). However, this version of the DNMS task proved to be difficult for our cohort of male Sprague-Dawley rats, as they did not learn the rule switch amidst the assessment of working memory (below chance correct responding) for over one month (data for first ten days shown in Fig. 2-9). Therefore, these rats were eliminated from the DNMS experiment and subsequent testing. The DNMS task was modified to only include the 0s

delay in order to assess the effects of cocaine on reversal learning without reevaluating the capacity of working memory after the rule switch. This modified version is referred to as the non-match to sample task (NMS).

2.1.5 Context + Cue Relapse Test

On day 90 of abstinence, rats were reintroduced to the self-administration chambers to assess cocaine-seeking in response to the cocaine-associated context and cues during a 45-min long relapse test. Nose poking in the previously active ports generated contingent presentations of the light and tone cues previously paired with cocaine infusions, but cocaine infusions were no longer administered. The saline group received repeated non-contingent light and tone cue presentations every 4 minutes.

Cocaine-seeking was measured as the number of responses in the previously active port compared to the inactive port within the cocaine group as well as the number of responses in the active port compared between groups.

2.1.6 Tissue Collection

All rats were killed by rapid decapitation at the end of the relapse test. Brains were extracted, immediately frozen in 2-methylbutane and stored at -80°C. Brains were equilibrated at -20°C in a cryostat (Leica CM1950) and divided into hemispheres. The right hemisphere was dissected at approximately +3.72 and +2.76 mm relative to

Bregma (Paxinos & Watson, 2007) and PrL tissue was collected using a 2 mm micropunch (Harris Uni-Core, Ted Pella, Redding, CA), and frozen at -80°C for later western blotting. Alternating 40μm and 12μm coronal brain slices were collected from the left hemisphere at approximately +3.72 and +2.76 mm relative to Bregma (Paxinos

& Watson, 2005) for CO histochemistry and mRNA in situ hybridization. Brain slices

were mounted directly onto Superfrost Plus Gold slides (Fisher Scientific), dried at -

20°C for 15 min, and stored at -80°C.

2.1.7 Cytochrome Oxidase Histochemistry

CO histochemistry staining was conducted in accordance to previously published protocols (Gonzalez-Lima & Cada, 1994; Gonzalez-Lima & Jones, 1994). Briefly, standards were generated to ensure linearity of optical density measurements using homogenized brain tissue from non-subject male Sprague-Dawley rats, sectioned at increasing thicknesses (10, 20, 40, 60 and 80µm). Frozen slides were fixed for 5 minutes in a refrigerated 10% sucrose phosphate buffer solution containing 0.5% glutaraldehyde. Slides were then rinsed four times for 5 minutes each in 10% sucrose phosphate buffer to remove red blood cells. To enhance staining contrast, slides were pre-treated for ten minutes in a solution (pH 7.6) containing 0.05 M Tris buffer, 275 mg/l cobalt chloride, 10% sucrose and 0.5% dimethyl sulfoxide (DMSO). Following a brief rinse in a phosphate buffer, slides were then incubated for one hour at 37°C in an oxygen-saturated 0.1 M phosphate buffer (400mg Diaminobenzidine (DAB), 60 mg cytochrome c, 40g sucrose, 16 mg catalase and 2ml DMSO). DAB is oxidized to a visible indamine polymer to yield the staining reaction signal. The accumulation of this visible product is dependent on the continuous re-oxidation of cytochrome c, and increasing linearity of optical density measurements is indicative of greater CO reactivity. This suggests CO reactivity as a reliable indication of the neuron’s oxidative capacity, energy metabolism and resulting neuronal functional activity (Wong-Riley et al., 1998). To fix slides and stop the last reaction product, slides were incubated in buffered formalin at room temperature for 30 minutes. A final series of increasing ethanol (ETOH) baths (30% to 100% ETOH) were used to dehydrate slides. Slides were

cleared in CitraSolv and cover slipped with Permount (Fisher Scientific, Pittsburgh, PA,

USA).

Images were taken for each rat PFC slice using an Olympus BX51 microscope.

All imaging was done using the identical background lighting and exposure settings. For each rat section imaged, an adjacent blank section of the slide was captured. The same region of interest (ROI) was selected within the PrL cortex (dorsal frontal cortex layer

III/IV (Bregma 3.0 mm) for all rats. Total integrated density (I.D.) for each target ROI

(two sections per rat) and corresponding blank control regions was measured using

ImageJ software (Schneider, Rasband, & Eliceiri, 2012). The average difference in I.D. between the target ROIs and blank control regions were calculated for each rat and used as a measure of CO reactivity.

2.1.8 Western Blotting

Total protein was extracted from the tissue as previously described (Bilodeau &

Schwendt, 2016) and quantified using bicinchoninic assay according to the manufacturer’s instructions (BCA Protein Assay Kit, Thermo-Fisher, Waltham, MA).

Equal protein amounts (15 μg) were separated by SDS-PAGE (4–15% polyacrylamide) and transferred onto polyvinylidene difluoride (PVDF) membranes. Membranes were blocked for one hour in 5% milk/Tris-buffered saline with 0.1% Tween 20 (TBST), and probed overnight with the following primary antibodies: rabbit anti-mGlu5 (1:15000,

#AB5675), rabbit anti-mGlu1 (1:1000, Phosphosolutions #2031), rat anti-Homer 1b/c

(1:7500, #AB5877), rabbit anti-Homer 2a/b (1:5000, Synaptic Systems #160 203), mouse anti-Gαq (1:1500, New East Biosciences #26060), rabbit anti-PLCβ1 (1:5000,

#sc-9050), rabbit anti-CAMKII (1:5000, #sc-9035), rabbit anti-Homer 1a (1:2000,

Synaptic Systems #160 013), rabbit anti-Arc (1:25000, Synaptic Systems # 156 003),

rabbit anti-phospho-ERK (1:5000, Cell Signaling #9101S), rabbit anti-total ERK (1:5000,

Cell Signaling #9102), rabbit anti-phospho-PKC[s] (1:15000, Cell Signaling #2261), and mouse anti-total PKC (1:7500, EMD Millipore #05-983). Membranes were washed three times for ten minutes each in 5% milk/TBST, incubated with appropriate HRP- conjugated secondary antisera (Jackson Immuno Research, West Grove, PA).

Immunoreactive bands on the membranes were detected by ECL+ chemiluminescence using a high performance chemiluminescence film (GE Healthcare, Piscataway, NJ).

Equal loading and transfer of proteins were confirmed by blotting for the housekeeping protein calnexin (1:20-40,000, #ADI-SPA-860, Enzo Life Sciences, Farmingdale, NY,

USA). The integrated band density of each protein sample was measured using Gel-Pro

Analyzer software (Media Cybernetics, Rockville, MD, USA), normalized to its respective calnexin I.D. measure and expressed as percent value of saline controls.

2.1.9 Fluorescent in situ Hybridization

Fluorescent in situ hybridization (FISH) for Arc and GRM5 (mGluR5) mRNAs was performed with the RNAscope Multiplex Fluorescent Reagent Kit (ACDBio, Newark, CA) based on previously published procedures (Wang et al., 2012). Brain sections were fixed in 4% paraformaldehyde (PFA) for 15 min at 4℃, and then dehydrated in a series of graded ETOH concentrations for five min each (50%, 70%, 100%, 100%). Sections were dried at room temperature prior to undergoing a 25-min protease digestion using pretreatment #4 (ACDBio). RNAscope target probes for Arc (ACDBio: 317071-C1, lot

16078A) and GRM5 (ACDBio: 471241-C2, lot 17243B) were applied to each section, and slides were incubated 2 hr at 40°C for mRNA hybridization. Sections were then incubated with preamplifier and amplifier probes at 40°C (AMP 1, 30 min; AMP 2, 15 min; AMP 3, 30 min). AMP 4 Alt-C was selected so that Arc and GRM5 probes were

labeled with ATTO 550 and ATTO 647 fluorophores, respectively. Sections were counterstained with DAPI and coverslipped with ProLong Gold antifade mounting reagent (Thermofisher). Fluorescent images were obtained using a Nikon Ti-E inverted microscope fitted with a CFI Plan Apochromat Lambda 40x dry objective, a Nikon DS-

Fi2 camera, and NIS-elements software. Target regions were cropped from high magnification stitched images of each section. Quantification of Arc mRNA puncta was performed in MATLAB (Mathworks, Natick, MA) using TransQuant software (Halpern &

Itzkovitz, 2016). Cells expressing both Arc and GRM5 mRNA transcripts were identified manually with ImageJ (Schneider et al., 2012). Segmentation of all dual Arc/GRM5 positive cells was performed in TransQuant and within-cell Arc mRNA puncta were measured.

2.1.10 Statistics

IBM SPSS (Version 25) and GraphPad Prism (Version 5.03) software were used to analyze data. The alpha level was set at 0.05 for all statistical analyses used. Paired and unpaired samples t-tests (two-tailed) or one-way ANOVAs were used to assess group differences on single dependent measures. Repeated-measures ANOVAs were used to assess group differences across time points when appropriate. Bonferroni post- hoc tests were used to adjust for multiple comparisons. Two-tailed bivariate Pearson and partial correlations were conducted to compare continuous variables.

2.2 Results

2.2.1 Rats Escalated Cocaine Intake during Extended Access Cocaine Self- Administration

Rats in the cocaine group rapidly acquired self-administration and displayed clear discrimination between the active and inactive port (Fig. 2-2A). During both ShA and

LgA self-administration, rats demonstrated successful discrimination between active and inactive nose ports as indicated by a repeated measures ANOVA showing main effects of Port over the first six days (ShA: (F(1,110) = 16.03, p < 0.001) as well as over the next 12 days (LgA: F(1,242) = 55.91, p < 0.0001). Furthermore, rats displayed an escalation in cocaine intake (mg/kg) analyzed as previously described (Schwendt,

Reichel, & See, 2012; Schwendt et al., 2009) using a two-tailed paired-samples t-test to compare the mean cocaine intake from the first three days of LgA (M = 55.57, SD =

12.54) with the mean cocaine intake from the last three days of LgA (M = 68.7, SD =

13.10); (t(11) = 4.05, p < 0.01; Fig. 2-2B).

2.2.2 Rats with a History of Extended Access Cocaine Self-Administration Showed Working Memory Impairments Related to Prior Cocaine Intake and Later Drug Seeking.

After the completion of extended access cocaine self-administration, rats were trained in the DMS task. DMS task acquisition was calculated as the total number of days spent in training before meeting the criterion to enter the testing period. There were no group differences in the average number of days to reach criterion in the training sessions before DMS testing (t(26) = 0.93, n.s.; Fig. 2-3A). Percent correct was the primary dependent measure in the DMS task. Percent correct at each delay interval

{0 s, 2s, 4s, 8s, 12s, 18s, 24s} was averaged over each of the two five-day testing blocks. A repeated measures ANOVA was conducted with Delay and Block as the within-subjects factors and Group as the between-subjects factor. A repeated measures

ANOVA showed main effects of Delay (F(3.48, 90.48) = 297.34, p < 0.01) and Block

(F(1, 26) = 10.06, p < 0.01; Fig. 2-3B). Performance (% correct) for all rats was significantly more impaired as each delay interval increased (Fig. 2-3B), and overall percent correct significantly increased from Block one to Block two. Furthermore, there

was a main effect of Group (F(1, 26) = 16.34, p < 0.01) in which rats with a history of saline performed significantly better than rats with a history of cocaine (Fig. 2-3B).

There was no significant interaction.

Since rats were (a) trained until a criterion of 80% for the delays up to 12s and

(b) first introduced to the long delays (18s and 24s) during the ten day testing block (Fig.

2-3B), a repeated measures ANOVA was conducted to assess group differences in percent correct at only the two longest delays (18s and 24s) across the ten day testing block. Averaging percent correct at the 18s and 24s delays in this task for an overall measure of performance at long delays is based on previously published analyses

(Bañuelos et al., 2014). There was no main effect of Day and no interaction (Fig. 2-3C), but a main effect of Group (F(1, 26) = 10.4, p < 0.01) was found in which rats with a history of saline performed significantly better than rats with a history of cocaine (Fig. 2-

3C, 2-3D). Averaging performance for each rat across all long delay trials, rats with a history of cocaine compared to saline were significantly impaired (t(26) = 3.22, p < 0.05;

Fig. 2-3D). Pearson correlations were conducted to compare cocaine intake (mg/kg) with performance (% correct) at the long delays during each block of testing. A significant negative correlation was found between cocaine intake and performance at the 24s delay during block two (r = -.50, n = 12, p < 0.05) in which more cocaine intake correlated with worse performance at the highest delay during the second testing block.

Pearson correlations were also conducted to assess the relationship between DMS testing and later drug-seeking. A significant relationship was not found between percent correct at the long delays and number of active nose pokes (drug seeking) in the relapse test (r = 0.44, n = 11, n.s.). However, a significant negative correlation between

percent correct at the 0s delay and drug seeking during the relapse test was observed (r

= -0.52, n = 11, p < 0.05) in which rats who performed worse on this measure exhibited greater drug seeking in the relapse test.

2.2.3 Reversal Learning was Not Impaired in Rats with a History of Extended Access Cocaine Self-Administration. NMS Task Behavior Predicted Subsequent Drug Seeking.

After completing DMS testing, rats underwent assessment in the NMS task, where the delay was set to 0s (Fig. 2-4). The criterion for the NMS task was set at above chance responding (50% correct) for two consecutive days. Regardless of a history of cocaine or saline, rats did not differ in the average number of days to reach the criterion of above chance responding in the NMS task (t(19) = 0.05, n.s.; Fig. 2-4A).

One rat who failed to reach the criterion was excluded from additional testing. A repeated measures ANOVA was conducted between groups in the NMS task with Day as the within-subjects factor. The two groups did not differ in average percent correct across the fifteen-day testing block. There was a main effect of Day (F(3.34, 63.50) =

104.94, p < 0.01), but not a main effect of Group (F(1,19) = 0.19, n.s.) nor a significant interaction (Fig. 2-4B). Since both groups did not reach the criterion of above chance responding until the third block of testing, a one-way ANOVA was conducted to assess group differences in average percent correct during the third block of testing. Groups did not differ in their average percent correct over the third testing block (t(19) = 0.36, n.s.;

Fig. 2-4C). A Pearson correlation revealed no significant correlation between cocaine intake (mg/kg) and percent correct during the third block of testing (r = 0.40, n = 11, n.s.). A Pearson correlation did not reveal a significant relationship between reversal learning and drug seeking when comparing the average percent correct during the third block testing with active nose pokes in the relapse test (r = -0.38, n = 11, n.s.).

However, a significant negative correlation was found between percent correct on the last day of testing and drug seeking during the relapse test (r = -0.52, n = 11, p < 0.05) in which rats who demonstrated a greater impairment in switching the rule (lower percent correct responding) on the last day of testing exhibited greater drug seeking in the subsequent relapse test.

2.2.4 Relapse to Cocaine-Seeking on Day 90 of Post-Cocaine Abstinence was Accompanied by Increased Arc mRNA within mGlu5-Positive Cells in the PrL.

Ninety days following the end of extended access cocaine self-administration, rats were re-exposed to the cocaine-paired context with cues during a relapse test. A repeated measures ANOVA with Port (active versus inactive) as the within subjects factor and Group as the between subject factor revealed a main effect of Group F(1,19)

= 12.28, p < 0.01) and Port (F(1, 19) = 23.23, p < 0.0001) as well as a significant interaction between Port and Group (F(1, 19) = 14.80, p < 0.01; Fig. 2-5). Bonferroni post-tests revealed that rats with a history of cocaine versus saline exhibited significantly more responding in the active port than in the inactive port as well as significantly more responding in the active port overall (Fig. 2-5).

Arc mRNA was measured in PrL tissue collected from the cocaine and saline rats immediately after the relapse test as well as in age-matched rats killed from the home cage to assess PrL activation following re-exposure to cocaine-associated cues. PrL

Arc mRNA expression was quantified as the total Arc mRNA puncta per target ROI (Fig.

2-10) averaged from two PrL sections for each rat (Fig. 2-6). A one-way ANOVA revealed a significant difference between groups in PrL Arc mRNA expression (F(2, 21)

= 7.70, p < 0.01; Fig. 2-6B). Bonferroni’s Multiple Comparison Test indicated that rats with a history of cocaine expressed significantly more PrL Arc mRNA puncta compared

to both saline rats and home cage controls (Fig. 2-6A, 2-6B). Next, the average number of Arc mRNA puncta across all mGlu5 mRNA expressing cells within the target ROI was averaged across two brain sections per rat. Examination of Arc mRNA expression within

PrL mGlu5 positive cells using one-way ANOVA also indicated a significant difference between groups (F(2, 21) = 9.87, p < 0.01; Fig. 2-6C). Bonferroni’s Multiple

Comparison Test revealed that cocaine rats averaged higher expression of Arc mRNA within PrL mGlu5 expressing cells compared to both saline rats and home cage controls

(Fig. 2-6A, 2-6C).

2.2.5 Metabolic Activity in the PrL Positively Correlated with Prior Working Memory Performance in Cocaine Rats Only

PrL metabolic activity was expressed as CO reactivity, measured as I.D. within

PrL slices following chromogenic staining with CO histochemistry (Fig. 2-7A). There was no difference in CO reactivity between groups (t(19) = 0.67, n.s.). Since CO reactivity has been previously shown to lag behind behavior as a measure of activation within a particular region (Bechard et al., 2017; Tanimura et al., 2011), a partial correlation was conducted to control for the time point when rats were performing the cognitive measures. This is important because rats varied when they reached training criterion and ultimately began testing at different time points throughout the experiment. Partial correlations controlling for the number of days when rats were last tested in the cognitive tasks revealed a significant relationship between PrL CO reactivity and percent correct at the long delays (r = 0.47, n = 21, p < 0.05; Fig. 2-7B) and no significant relationship between CO reactivity and performance in the NMS task (r = -

0.41, n = 21, n.s.). The positive partial correlation between PrL CO reactivity and percent correct at the long delays was only significant for the cocaine group (r = 0.66, n

= 11, p < 0.05, Fig. 2-7C) and not the saline group (r = 0.20, n = 10, n.s.). There was no correlation between drug seeking and CO reactivity in the PrL (r = 0.51, n = 11, n.s.).

Partial correlation graphs (Fig. 2-7B, 2-7C) show the unstandardized residuals of each dependent measure as the x and y axis. After regressing each dependent measure (% correct at the long delays and CO reactivity) separately onto the variable being controlled for (time point of testing), unstandardized residual values are obtained from the actual values of each dependent measure subtracted from the predicted value of the regression equation. Partial correlation graphs showing unstandardized residuals of the dependent measures have been previously published (Moreno-López et al., 2015).

2.2.6 Greater Expression of the mGlu5 Monomer and Homer 1b/c in the PrL was Exhibited in Cocaine Rats. Expression of the mGlu5 Monomer Negatively Correlated with Prior Working Memory Performance

PrL protein expression was represented as the percentage of the I.D. values of the rats with a history of saline. One cocaine rat was excluded from western blotting analyses because tissue from the hemisphere used for western blotting was lost during collection. There were no significant differences between groups in the protein expression of: mGlu1, Homer 2a/b, Gαq, PLCβ, CaMKII, Homer 1a, Arc, phospho-ERK, total ERK (tERK), phospho-PKC substrate, and total PKC (tPKC) in the PrL (n.s.; Table

1 and Fig. 2-11). However, in rats with a history of extended access cocaine self- administration, significantly higher Homer 1b/c protein levels were detected (t(18) =

2.12, p < 0.05; Table 1 and Fig. 2-11). Groups also did not differ in expression of total mGlu5 (calculated as the sum of the normalized I.D. values of the monomer and dimer), t(18) = 2.00, n.s.; Fig. 2-8B). However, when examining I.D. of the mGlu5 monomer separately from the dimer, rats with a history of cocaine had significantly higher expression of the mGlu5 monomer (t(18) = 2.18, p < 0.05; Fig. 2-8D) while no group

difference was observed on the expression of the mGlu5 dimer (t(18) = 1.16, n.s.). To ensure these differences between groups were not confounded by relapse testing immediately prior to tissue collection, bivariate Pearson correlations were conducted between protein expression and the number of nose pokes in the active port. There was no significant correlation between active nose pokes and Homer 1b/c (r = 0.24, n = 10, n.s.), total mGlu5 (r = -0.041, n = 10, n.s.), mGlu5 monomer (r = -0.20, n = 10, n.s.) and mGlu5 dimer (r = 0.06, n = 10, n.s.). Bivariate Pearson correlations were conducted between mGlu5 expression and percent correct at the long delays in the DMS for all rats. This correlation using total mGlu5 revealed a significant negative correlation (r = -

0.50, n = 20, p < 0.05; Fig. 2-8C), driven by the monomer (r = -0.56, n = 20, p < 0.05;

Fig. 2-8E) and not significant for the dimer (r = -0.25, n = 20, n.s.).

2.3 Discussion

This study shows that rats with a history of extended access cocaine self- administration, display delay-dependent working memory impairments when compared to their saline counterparts, and cocaine intake predicts these impairments. Working memory capacity (as assessed in an mPFC-dependent operant DMS task) was selectively reduced, as neither acquisition of the DMS task, nor acquisition of the rule- switching (NMS task) was altered in cocaine rats. Importantly, decreased DMS performance represented a stable trait (detected 45-60 days post-cocaine) that occurred within a period of persistent cocaine-seeking (as detected on day 90). This study is also the first one to report on the relationship between the long-term metabolic activity of the PrL and past working memory performance with greater metabolic activity supporting working memory performance in cocaine rats. And finally, this study reveals dysregulation of PrL mGlu5 in cocaine rats as evidenced by increased expression of the

monomeric, inactive form of the mGlu5 receptor that negatively correlated with past working memory performance (for all rats) as well as upregulation of Arc mRNA expression detected in mGlu5-positive PrL cells immediately after reexposure to cocaine-associated cues.

Working memory deficits uncovered in this study are consistent with several previous pre-clinical reports (Fijał et al., 2015; George et al., 2008; Porter et al., 2011;

Radley et al., 2015). The validity of these (and this study’s) findings is supported by the fact that post-cocaine working memory deficits were detected across species (rats and rhesus monkeys), after varied types of access to cocaine self-administration (limited vs. extended) and using distinct working memory tasks (operant vs. T-maze). However, a critical consideration not investigated in these animal studies is the possible relationship between the degree of working memory impairment and drug seeking. Here, demand- dependent decreases in working memory performance occurred within a period of persistent relapse susceptibility (cocaine seeking). Thus, it was hypothesized that these two phenomena are interrelated, and both originate in the cocaine-induced dysfunction of the PrL. Indeed, several studies in human cocaine users found that cognitive impairments predict higher relapse rates and poorer treatment outcomes (Aharonovich et al., 2006; Brewer et al., 2008; Streeter et al., 2008; Verdejo-Garcia et al., 2014).

Here, rats with worse performance in the DMS task at the 0s delay later displayed greater drug seeking. On the other hand, the working memory performance at higher delays did not predict cocaine-seeking. There may be several ways to interpret these relationships between DMS performance and drug seeking. Performance in the task without any delays is less cognitively demanding as indicated by stable, above 90%

correct responding across treatment groups. Rather, incorrect responding without delays between the sample and choice phase indicates quick switching to the opposite lever and immediate responding when that lever is presented. The persistence of errors from this pattern of behavior despite demonstrating mastery of the task rules in this measure may indicate hyperactivity, decreased impulse control or attentional impairments, behavioral deficits seen to be predictive of treatment dropouts in CBT

(Aharonovich et al., 2006). On the other hand, performance in the DMS task at long delays was quite challenging but to a similar extent for all cocaine rats (within-group variability was less than 7%) despite considerable individual differences in the magnitude of drug-seeking. It is possible that slight variations in performance at the long delays were not sensitive enough to predict drug seeking. In addition, an interesting observation is that the correlation between percent correct at the long delays and drug seeking is positive. A possible explanation for this trend may be attributed to an overlap in reward related learning involved in both the DMS task as well as in retrieving memories for drug associated cues in a context where drug-incentivized learning previously occurred.

Next, the effects of chronic cocaine on reversal learning was evaluated. Here, reversal learning was assessed as the ability of rats to learn a new (non)match-to- sample rule in the NMS task. While some studies with human cocaine users reported reversal learning deficits (Cunha et al., 2013; Kelley et al., 2005), an effect of cocaine history on this measure using the NMS task was not observed. Overall, pre-clinical evidence on the effects of chronic cocaine on reversal learning is inconclusive (Bechard et al., 2018; Calu et al., 2007; Kantak et al., 2014) with possible discrepancies arising

from the sensitivity of the tasks or the strain of rat used. Here, the time point post cocaine as well as the history prior to reversal learning testing should be considered as the rats in the current study underwent extensive DMS training and testing until (on average) day 60 post-cocaine prior to testing in the NMS task. It is possible that a recovery of cognitive deficits with prolonged abstinence (Fijał et al., 2015; Vonmoos et al., 2014), or prior cognitive training and testing in a task that utilizes overlapping circuitry (Sloan et al., 2006) ‘erased’ differences between cocaine and saline groups.

Further, while no overall relationship between drug seeking and stable NMS performance was found, lower performance on the last day of NMS testing correlated with greater drug seeking, hinting a relationship between the two behavioral variables that need to be further explored in future studies. Finally, it should be noted that studies utilizing other self-administration regimens, like those that give rise to distinct ‘addictive’ vs. ‘non-addictive’ phenotypes (James et al., 2018; Kawa, Bentzley, & Robinson, 2016;

Zimmer, Oleson, & Roberts, 2012), might uncover more pronounced cognitive deficits in relation to drug seeking.

Previously, it has been demonstrated that the cognitive tasks employed here depend on intact mPFC function (Sloan et al., 2006). However, since the mPFC subregions, PrL and infralimbic (IL), play dissociable roles in controlling cognitive functions (Vertes, 2004) and drug-seeking (Moorman et al., 2015), the current study expands on the previous findings through investigation of neural substrates that may encode immediate and long-term changes in the PrL related to cognitive performance and drug seeking. Consistent with previous studies (Hearing et al., 2008; Zavala et al.,

2008; Ziółkowska et al., 2011), greater Arc mRNA expression was identified as a

measure of immediate PrL activation in response to cue-elicited cocaine seeking. The current study expands these earlier findings to show 1) that context + cue-elicited cocaine seeking after a prolonged abstinence period of 90 days (akin to ‘incubation of cocaine craving’ models) produces robust Arc mRNA expression in the PrL that 2) occurs in a subpopulation of mGlu5-positive cells. The significance of showing a cocaine-cue inducible PrL activation within mGlu5 expressing cells on day 90 highlights that targeting this receptor may have implications for attenuating drug responses even during prolonged abstinence where an incubation of craving phenomenon may be thwarting successful recovery. Although the current experimental design did not allow for further investigation of a possible overlap between cell populations and the circuitry supporting drug-seeking and working memory in cocaine rats, it highlights the role of mGlu5-expressing cells in the PrL during context+cue- elicited cocaine seeking. This corroborates previous findings on the role of PrL excitability (Benn et al., 2016;

McLaughlin & See, 2003), glutamatergic transmission (Luís et al., 2017) and mGlu5 activity (Timmer & Steketee, 2012) during cue-elicited cocaine seeking. With regards to a wider circuitry that supports cue-elicited cocaine-seeking, it is generally agreed that projections from the PrL to the NAc core (Gass & Chandler, 2013; Van den Oever et al.,

2010) and to the basolateral amygdala (Mashhoon, Wells, & Kantak, 2010; McLaughlin

& See, 2003) encode behavioral reactivity to cocaine-paired cues. After prolonged abstinence, or extensive self-administration experience, the corticostriatal circuit controlling drug-seeking progressively shifts from ventral to dorsomedial (dmSTR) and dorsolateral (dlSTR) striatum (Caprioli et al., 2017; Murray et al., 2015). Interestingly, recent studies dissecting the circuitry of operant working memory tasks in rodents and

primates highlighted the role of PrL projections to the dmSTR as well (Akhlaghpour et al., 2016; Liu et al., 2014; Ragozzino, 2007). Therefore, future studies need to address an intriguing possibility that dysfunction of overlapping neuronal populations within the

PrL is responsible both for cognitive (working memory) and motivational (drug-seeking) deficits and further probe the role of mGlu5 receptors in these circuits.

While immediate changes in PrL activity can inform us about the engagement of specific cells and circuits in studied behaviors, persistent behavioral pathologies require lasting functional neurobiological changes. To address this question, long-term changes in PrL metabolic activity was assessed using CO histochemistry. Even though CO reactivity in the PrL was analyzed immediately after a relapse test, it is unlikely that the test itself altered CO reactivity. This is based on observations that not acute (Vélez-

Hernández et al., 2014), but rather prolonged (days and weeks) periods of behavioral and neural activity alter oxidative capacity for energy metabolism quantified in this assay (Bechard et al., 2017; Gonzalez-Lima & Cada, 1994; Gonzalez-Lima & Jones,

1994; Riha et al., 2011; Spivey et al., 2011; Tanimura et al., 2011). Accordingly, a significant correlation was not found between drug seeking and CO reactivity in the PrL

(data not shown). Rather, since metabolic mapping using CO reactivity has been shown to lag behind behavior by up to 6-10 weeks (Bechard et al., 2017; Tanimura et al.,

2011), CO reactivity here was expected to correlate with prior cognitive measures evaluated around this time point.

First, group differences in overall PrL metabolic activity was not found like in a recent study that utilized non-contigent repeated cocaine administration (Vélez-

Hernández et al., 2014). However, it is noteworthy that these authors observed

decreased CO reactivity in several other cortical regions. This suggests that broader brain analysis needs to be conducted to determine 1) if extended cocaine self- administration alters CO reactivity across the frontal cortex and 2) whether this change persists over a prolonged drug-free period paralleling lower frontal metabolic activity found in abstinent human cocaine users (Goldstein & Volkow, 2002, 2011; Volkow et al., 1992). Unlike abstinent cocaine users, in our study, rats underwent extensive cognitive training and testing which likely produced long-term changes in brain metabolic activity itself and possibly diminished baseline differences in PrL activity between the groups. In support of this, greater metabolic activation within the PrL positively correlated with past working memory performance in the DMS task. This relationship was evident when controlling for the delay period between the time of DMS performance and the time of tissue analysis. Further, because the cocaine group performed worse in the working memory task, groups were separated in this analysis, and a significant positive correlation was found for the cocaine group only. It is important to note that rats underwent a cognitively distinct NMS task after the DMS, and performance in this subsequent task may have altered PrL CO reactivity and how it may relate to task performance. However, there was not a significant correlation between

PrL CO reactivity and NMS performance when controlling for the time point of testing.

Based from these findings, cocaine rats may be displaying an inefficient hypermetabolic activity to sustain working memory performance wherein they may need to work

‘cognitively harder’ during testing to perform better but still not as good as their saline counterparts. This interpretation is supported by the findings in human binge drinkers and rats binge-drinking alcohol, which show greater cortical activation during cognitive

tasks, suggesting greater recruitment of relevant neuronal resources to support cognitive performance, indicating neural inefficiency (Campanella et al., 2013; West,

Maynard, & Leasure, 2018). Likewise, compensatory attention and control processing has been implicated in human cocaine users as evidenced by hyperactivation in several cortical regions related to working memory performance (Tomasi et al., 2007b). These findings may also provide partial support for the hypothesis that an effect of cognitive testing may diminish any baseline group differences in PrL metabolic activation. As such, there is a need to separate the influence of cocaine vs. saline history from the influence of cognitive training itself, such as incorporating within-subject design to directly assess pre-/post-cocaine changes in metabolic activity, as well as assessing

CO reactivity after prolonged abstinence without any cognitive training.

These findings (induction of Arc in mGlu5-positive neurons in the PrL) and other recent reports (Hernandez et al., 2018) suggest that activation of mGlu5 receptor in the

PrL occurs during drug-seeking as well as while performing DMS working memory tasks. Therefore, basal protein expression of mGlu5 and its downstream signaling partners was investigated in this brain region. Proteins predicted to be unaffected by acute behavioral experiences (e.g. mGlu5 and long-form Homer proteins), as well as proteins downstream from mGlu5 receptors that rapidly respond to mGlu5 activation

(e.g. Homer1a, or pPKC substrates) were analyzed. Here, rats with a history of chronic cocaine exhibited greater expression of the mGlu5 monomer and Homer 1b/c scaffolding protein and further, monomeric mGlu5 levels negatively correlated with past

DMS working memory performance. It is important to note that the current experimental design is unable to dissociate effects of cognitive training or drug history on this greater

protein expression exhibited in the cocaine group. Given the time gap between the behavioral assessment and the protein analysis, one should approach the relationship between monomeric mGlu5 and working memory performance with caution. This result needs to be verified in follow-up studies that will place the time of behavioral and protein analysis closer together. One may interpret these findings (greater mGlu5 monomer expression for the cocaine group and an overall negative correlation with working memory) as (1) cognitive testing might have produced a compensatory upregulation of mGlu5 in the PrL, or (2) greater expression of the monomeric form of mGlu5 underlies poor performance. A previous study demonstrated that mGlu5 availability within the

PFC decreases during cocaine but not sucrose exposure, and drug withdrawal normalizes this effect (de Laat et al., 2018). Thus, an extended abstinence period in combination with extensive and particularly challenging cognitive testing employed in the current study may have upregulated this receptor to compensate for a possible early mGlu5 reduction from chronic cocaine exposure. A similar compensatory mechanism for mGlu5 dysfunction is hypothesized in schizophrenics where protein levels of mGlu5

(total, dimer, and monomer) and its trafficking molecules are increased within post- mortem hippocampus tissue and suggested to accompany their hippocampal- dependent cognitive deficits (Matosin et al., 2015).

It is also important to note that the negative correlation between mGlu5 expression and working memory is driven by expression of the monomer and not the dimer. It is unclear how to interpret this differential involvement of the monomeric versus the dimeric form of mGlu5 as it relates to working memory performance. However, differences in the functional significance between the dimer and monomer have been

characterized regarding the dimer as the ‘active’ functional form and the monomer as the ‘inactive’ form (El Moustaine et al., 2012). Greater monomeric ‘inactive’ expression of mGlu5 may support the idea that poorer working memory performance coincides with less active pool of mGlu5 receptors in the PrL. Further, a greater expression of this

‘inactive’ monomeric form as well as Homer 1b/c for the cocaine group may suggest group specific aberrant mGlu5 signaling related to cocaine history or as a compensatory mechanism to support post-cocaine cognitive dysfunction during cognitive training. It is important to consider that mGlu5 is expressed postsynaptically and intracellularly, and activation of different signaling cascades ensue depending on this receptor’s cellular location (Jong et al., 2014). However, because total homogenate preparations were used, it was not possible to differentiate if the mGlu5 monomer increase reflects alterations in the distribution of intracellular or cell-surface mGlu5. Yet, considering the role of Homer 1b/c in localizing mGlu5 to the post-synaptic density as well as scaffolding this receptor to signaling effectors (Ménard & Quirion, 2012; Shiraishi-

Yamaguchi & Furuichi, 2007), increased levels of Homer 1b/c alongside increased expression of the mGlu5 monomer of cocaine rats might suggest a dysregulation between Homer 1b/c and mGlu5, wherein selective scaffolding of the mGlu5 inactive monomeric form and/or the removal of mGlu5 active dimeric receptors from the membrane may be favored, further disabling the mGlu5 receptor function in this brain region. Interestingly, a significant negative correlation was found here between Homer

1b/c and mGlu5 dimer, but not monomer, expression specifically for the cocaine rats

(data not shown). Thus, increased Homer 1b/c protein may be involved in greater sequestering of the dimeric ‘active’ form away from the surface specifically for cocaine

rats as a result of cocaine history contributing to working memory impairments or during impaired learning within a cognitive training period. Alternatively, the increased protein expression may be located intracellularly, and as a compensatory mechanism to regulate post-cocaine cognitive function, Homer 1b/c may be involved in intracellular retention of the inactive monomeric form. Taken together, these findings suggest mGlu5 signaling dysfunction specific for the cocaine group that may be attributed to drug history or poor learning during a cognitive testing period that is expressed as cognitive deficits or a result of compensatory mechanisms to support post-cocaine cognitive dysfunction during a cognitive testing period.

2.4 Limitations

Whereas the current study revealed lasting differences in cognitive performance and dysregulation of neural markers in cocaine vs. saline rats, the chosen experimental design could not address all variables possibly influencing our findings. This includes the fact that baseline working memory performance prior to the self-administration regimen was not assessed. Therefore, it is possible that in addition to drug effects, preexisting individual differences in cognitive performance could have influenced our between-group comparison assessed after the end of the self-administration period.

Likewise, previous training on an unrelated operant task (such as self-administration) may influence future testing on other behavioral tasks. In the current study, saline rats do not acquire self-administration, and thus previous experience with operant responding (albeit a different context and distinct operandum) for rats in the cocaine group may differentially affect future cognitive performance and accompanying neural changes between groups. And finally, as mentioned above, while this study provides evidence that some neural markers of PrL activity correlate with the behavior, direct

causal relationships between the identified neuroadaptations and the behavior remain to be established.

2.5 Future Directions

The overall results suggest that alterations in PrL immediate and long-term neuronal activation as well as dysregulated mGlu5 signaling may accompany the working memory impairments and persistent drug seeking seen in rats with a history of chronic cocaine. It is imperative to consider the implication of these findings in the context of administering pharmacotherapies for future studies. The data showing activity dependent increases in PrL mGlu5 expressing cells in response to cocaine-conditioned cues complement the plethora of other data pointing to the important role of increased glutamatergic signaling from the PrL to the NAc in accompanying cue-elicited drug seeking. As such, targeting mGlu5 systemically or within the NAc with an mGlu5 NAM has been previously shown to attenuate responding to drug-associated cues

(Knackstedt & Schwendt, 2016; Knackstedt et al., 2014; Wang et al., 2013). However, the role that such a manipulation may have on cognitive function within this population should also be considered, especially since chronic cocaine exposure gives rise to cognitive deficits. These results point to the importance of neuronal activation within the

PrL and uncompromised mGlu5 function to accompany better working memory performance in cocaine rats. While treatment with an mGlu5 NAM reduces excessive glutamate release that may benefit the reduction of drug seeking, chronic treatment with this compound may adversely alter glutamate signaling within cortical areas implicated in cognitive function. Indeed, microinfusions of an mGlu5 NAM into the PrL has been previously shown to produce impairments in working memory performance using this same DMS task (Hernandez et al., 2018). As a result, the possible pharmacotherapy of

targeting mGlu5 with a NAM to decrease drug seeking may also impair cognitive performance. It is also important to consider the increase in PrL monomeric mGlu5 seen in rats within a paradigm of cognitive testing during prolonged abstinence after chronic exposure to cocaine compared to saline. This is important to consider because mGlu5 allosteric modulators can bind to the mGlu5 monomer and exert their effects in the absence of glutamate binding which may result in unwanted side effects. Thus, it is important to investigate how allosteric modulation of the mGlu5 receptor may affect cognitive performance and drug seeking within the context of a cognitive-based abstinence paradigm prior to relapse.

2.6 Conclusions

In conclusion, extended access cocaine self-administration produced persistent impairment of working memory (but not reversal learning) detected within a prolonged abstinence window of high relapse susceptibility, wherein components of these cognitive measures were predictive of drug-seeking. Next, analysis of two neurobiological variables that capture immediate and long-term changes in activity of the PrL uncovered hyperactivity in a subpopulation of mGlu5-positive cells in response to cocaine-associated cues as well as compensatory PrL metabolic activity accompanying working memory performance specific for cocaine rats. Additionally, greater expression of the mGlu5 monomer as well as Homer 1b/c was observed in cocaine rats and may suggest dysfunction in mGlu5 signaling as an underlying molecular feature related to post-cocaine working memory impairments. Considering the overlap between PrL-mediated cognitive function and drug seeking, as well as the role of mGlu5 in these processes, targeted circuitry-specific pharmacotherapies of this receptor should be further evaluated.

Figure 2-1. Experimental timeline. DMS: delayed match-to-sample task, NMS: nonmatch-to-sample. ShA: 1 hr short access to cocaine, LgA: 6 hr long access to cocaine.

Figure 2-2. Extended access cocaine self-administration. A) Rats discriminated between nose poking in the active versus inactive port during 6hr long access (LgA) to cocaine. There were main effects of Port and Day as well as a significant interaction. B) Rats showed an escalation of cocaine intake. Average cocaine intake from the last three days of LgA was significantly greater than the average cocaine intake from the first three days of LgA. Error bars represent SEM. *p < 0.05, n = 12.

Figure 2-3. Delayed match-to-sample task performance. A) There were no group differences in DMS task acquisition. Task acquisition was indicated as number of days to reach the criterion of 80% correct for two consecutive days at each training delay set. B) Rats with a history of cocaine versus saline were significantly impaired in overall working memory task performance. Task performance (% correct) was averaged for each delay over Block one (days 1-5) and Block two (days 6-10). There were main effects of Group, Delay, and Block. There was no significant interaction. C & D) Rats with a history of cocaine versus saline were significantly impaired in working memory task performance at the long delays. C) Task performance (% correct) was averaged at the long delays (18s and 24s) across the ten-day testing period. There was a main effect of Group but not Day. There was no significant interaction. D) Graph showing the main effect of Group from the data in Fig. C. Error bars represent SEM. *p < 0.05, n = 12-16/group.

Figure 2-4. Nonmatch-to-sample task performance. A) There were no group differences in NMS task acquisition. Task acquisition was indicated as number of days to reach the criterion of 50% correct for two consecutive days. B & C) There were no group differences in NMS task performance. B) Task performance (% correct) was averaged over the three blocks of testing. There was a main effect of Day but not Group. There was no significant interaction. The dashed line indicates task performance at chance (% correct at 50%). C) Graph showing task performance (% correct) during the final testing block when performance was above chance. Error bars represent SEM. n = 10-11/group.

Figure 2-5. Context + cue relapse. Rats with a history of cocaine demonstrated robust drug seeking. Drug seeking during a 45 min context + cue relapse test was measured as number of nose pokes in the previously active port. The cocaine group showed significantly greater number of nose pokes in the active port compared to the inactive port, as well as compared to nose pokes in the active port for rats with a history of saline. Error bars represent SEM. *#p < 0.05, n = 10-11/group.

Figure 2-6. Fluorescent in situ hybridization. A) Arc/mGlu5 mRNA labeling and quantification for cocaine and saline treated rats after a context + cue relapse test and age-matched rats killed from the home cage. Representative images of dual Arc/mGlu5 mRNA with DAPI nuclei counter staining in the PrL target ROI at 40x magnification (left). Insets are zoomed in on individual representative cells. Examples of images used for Arc mRNA puncta quantification with TransQuant software (right). Scale bars represent 25µm. B) Rats with a history of cocaine compared to saline and age-matched controls showed significantly more Arc mRNA expression within the PrL. Arc mRNA expression was measured as total number of Arc mRNA puncta within the PrL ROI. C) Rats with a history of cocaine compared to saline and age- matched controls expressed significantly more Arc mRNA in mGlu5 expressing cells within the PrL. An average was taken of the total number of Arc mRNA puncta in each mGlu5 positive cell within the PrL ROI. Error bars represent SEM. *#p < 0.05, n = 10-11/group.

Figure 2-7. Cytochrome oxidase histochemistry. A) (Left panel) Rat brain coronal section at the level of the PrL (+3.0 from Bregma (Paxinos & Watson, 2007) with highlighted location of the ROI selected for CO analysis. (Right panel) Representative images show CO reactivity (4x magnification) within the PrL in saline and cocaine rats. Dashed box represents ROI measuring 1x1mm. B) There was a significant positive correlation between DMS performance and CO reactivity across all rats. CO reactivity was measured as I.D. within the PrL ROI. The partial correlation graph shows the unstandardized residuals for % correct at the long delays (18s and 24s) averaged over the ten-day DMS testing block and CO reactivity in the PrL after regressing these variables upon the time point of DMS testing (the number of days between the last day of DMS testing and tissue collection). C) The significant positive correlation between DMS performance and CO reactivity was only present in the cocaine rats. The partial correlation graph is showing the data only from the cocaine group. *p < 0.05, n = 10-11/group.

Figure 2-8. Western blotting. A) PrL target ROI. Representative immunoreactive bands of mGlu5 protein as detected in rats with a history of cocaine versus saline at 90 days after the end of cocaine self-administration. Lanes display conditions in an alternating pattern (saline, cocaine, etc.) B) There were no group differences in total mGlu5 expression. Total mGlu5 expression within the PrL was measured as the sum of the I.D. values of the monomer and dimer normalized to calnexin and graphed as % of normalized I.D. in saline rats. C) Total mGlu5 expression (I.D.) negatively correlated with past DMS task performance (% correct averaged at the 18s and 24s delays over the ten-day testing block). D) Rats with a history of cocaine compared to saline expressed significantly more of the mGlu5 monomer. I.D. of the mGlu5 monomer was normalized to calnexin and graphed as % of normalized I.D. in saline rats. E) mGlu5 monomer expression negatively correlated with past DMS task performance. Error bars represent SEM. *p < 0.05, n = 10/group.

Figure 2-9. Delayed nonmatch-to-sample task performance. Rats displayed below chance performance in the DNMS task. Task performance (% correct) was averaged for each delay across each five-day testing block. The dashed line indicates task performance at chance (% correct at 50%). Data shown are from rats in the saline group excluded from later analyses. Error bars represent SEM. n = 6.

Figure 2-10. Target ROI in the PrL selected for quantifying Arc mRNA expression. Rat brain coronal section at the level of the PrL (+3.0 from Bregma; Paxinos and Watson, 2007) with the highlighted location of the ROI selected for Arc analysis. Highlighted box represents ROI measuring 350x350 µm.

Figure 2-11. Representative immunoreactive bands of selected proteins as detected in the PrL of saline and cocaine rats at 90 days after the end of self- administration. The following primary antibodies were used: mGlu1, Homer 1b/c, Homer 2a/b, Gαq, PLCβ1, CaMKII, Homer 1a, Arc, phospho-ERK, total ERK (tERK), phospho-PKCs, total PKC (tPKC), and the house-keeping control, calnexin (clnx).

Table 1-1. Protein levels of mGlu5-associated proteins in the PrL on day 90. Saline Cocaine mGlu1 100 ± 8.16 102.48 ± 8.03 Homer 1b/c 100 ± 8.34 126.21 ± 9.13* Homer 2a/b 100 ± 10.26 100.08 ± 10.43 Gαq 100 ± 8.95 114.40 ± 12.43 PLCβ 100 ± 11.59 103.57 ± 14.30 tPKC 100 ± 5.68 109.58 ± 4.89 CaMKII 100 ± 4.22 100.35 ± 4.61 pPKCs 100 ± 16.37 108.18 ± 8.50 p-ERK1/2 100 ± 9.49 75.26 ± 7.15 Homer 1a 100 ± 5.92 96.08 ± 5.59 Arc 100 ± 7.01 103.27 ± 7.86 Data represent mean immunoreactivity of each protein normalized to calnexin and expressed as percent of the Saline group ± SEM. *p < 0.05 vs. Saline, n = 10/group.

CHAPTER 3 INVESTIGATING PRELIMBIC MARKERS DURING PROLONGED ABSTINENCE

Chronic cocaine use manifests as PFC dysfunction and cognitive deficits that contribute to loss of control over drug seeking (Goldstein & Volkow, 2002, 2011). The neural substrates overlapping cognitive dysfunction and persistent drug seeking are important to investigate to identify druggable targets. In Chapter 2 (Gobin, Shallcross, &

Schwendt, 2019), I identified long-term functional changes in relation to working memory performance in rats with a history of cocaine vs saline. Greater protein expression of mGlu5 as well as Homer 1b/c was noted in the PrL of rats with a history of extended access cocaine self-administration compared to saline after prolonged abstinence. However, these rats underwent extensive cognitive training and testing during the abstinence period, and therefore, it is unknown if the reported differences are related to the observed deficits or due to a history of cocaine. As correlations between these long-term markers and cognitive performance emerged (positive correlation between PrL metabolic activity and working memory and a negative correlation between

PrL mGlu5 monomeric expression and working memory), I hypothesize that the differences between groups are attributed to cocaine-induced cognitive dysfunction during task performance. This chapter tested this hypothesis by quantifying (1) PrL metabolic activity and (2) PrL expression of mglu5 and some associated proteins including Homer 1b/c in rats with a history of cocaine versus saline following prolonged abstinence in the home cage without any cognitive assessment.

3.1 Materials and Methods

3.1.1 Animals

Adult male Sprague-Dawley rats (Charles River Laboratories; 275g on arrival; N

= 16 were first acclimated to the animal facility prior to any manipulation. They were housed individually, maintained on a 12 hr reverse light/dark cycle (lights off at 0700), given ad libitum access to water, and food restricted (15-20 g of food per day) to maintain ~ 85% of their free-feeding weight (except before surgery when given ad libitum food access) (Carroll et al., 2002; Gobin & Schwendt, 2017). All animal procedures were approved by the IACUC of the University of Florida and were performed in accordance with the Guide for the Care and Use of Laboratory Animals.

The overall experimental design is depicted in Figure 3-1 with individual behavioral procedures briefly described below.

3.1.2 Surgical Procedures

All rats were implanted with jugular catheters and submitted to cocaine self- administration (or received saline) as previously described in Chapter 2 and in published protocols (Gobin & Schwendt, 2017; Gobin et al., 2019). Catheters were flushed daily with 0.1 ml of 100U/ml of heparinized saline.

3.1.3 Extended Access Cocaine Self-Administration and Abstinence

Rats were randomly assigned to either cocaine or saline groups and underwent identical self-administration procedures to those used in Chapter 2. After the conclusion of self-administration, animals experienced 45 days of abstinence wherein they were left in their home cages with intermittent handling.

3.1.4 Tissue Collection

All rats were taken from the home cage and killed by rapid decapitation on day

45. Methods for brains extraction and tissue collection were identical to those used in

Chapter 2.

3.1.5 Cytochrome Oxidase Histochemistry

CO histochemistry staining, imaging and analysis were done as previously described in Chapter 2.

3.1.6 Western Blotting

Immunoblotting procedures and analysis were done as previously explained in

Chapter 2. PVDF membranes were probed with the following primary antibodies: rabbit anti-mGlu5 (1:15000, #AB5675), rabbit anti-mGlu1 (1:1000, Phosphosolutions #2031), rat anti-Homer 1b/c (1:7500, #AB5877), rabbit anti-PLCβ1 (1:5000, #sc-9050), rabbit anti-phospho-PKC[s] (1:15000, Cell Signaling #2261), and mouse anti-total PKC

(1:7500, EMD Millipore #05-983). Equal loading and transfer of proteins were confirmed by blotting for the housekeeping protein calnexin (1:20-40,000, #ADI-SPA-860, Enzo

Life Sciences, Farmingdale, NY, USA).

3.1.7 Statistics

IBM SPSS (Version 25) and GraphPad Prism (Version 5.03) software were used to analyze data. The alpha level was set at 0.05 for all statistical analyses used. Paired and unpaired samples t-tests (two-tailed) were used to assess group differences on single dependent measures. Repeated-measures ANOVAs were used to assess group differences across time points. Bonferroni post-hoc tests were used when appropriate.

3.2 Results

3.2.1 Rats Escalated Cocaine Intake during Extended Access Cocaine Self- Administration

All rats demonstrated successful discrimination between the active and inactive nose port throughout self-administration (Fig. 3-2A) as indicated by a repeated measures ANOVA showing main effects of Port over the first six days (ShA: F(1,70) =

4.67, p < 0.05) as well as over the next 12 days (LgA: F(1,154) = 186.6, p < 0.001).

Furthermore, rats displayed an escalation in cocaine intake (mg/kg) analyzed as previously described (Gobin et al., 2019; Schwendt, Reichel, & See, 2012; Schwendt et al., 2009) using a two-tailed paired-samples t-test to compare the mean cocaine intake from the first three days of LgA (M = 50.45, SD = 11.33) with the mean cocaine intake from the last three days of LgA (M = 78.84, SD = 15.31); (t(7) = 7.79, p < 0.001; Fig. 3-

2B).

3.2.2 PrL Metabolic Activity did Not Differ between Rats with a History of Cocaine or Saline during Home-Cage Abstinence

PrL metabolic activity was expressed as CO reactivity and measured as integrated density within PrL slices following chromogenic staining with CO histochemistry as described in Chapter 2 and published previously (Gobin et al., 2019)

(Fig. 3-3A). There was no difference in CO reactivity between groups (t(14) = 1.33, n.s.,

Fig. 3-3B).

3.2.3 Expression of mGlu5 and mGlu5-Associated Proteins in the PrL did Not Differ between Rats with a History of Cocaine or Saline during Home-Cage Abstinence

PrL protein expression was quantified as integrated density. The values for the cocaine rats were represented as a percentage of the values quantified for saline rats, and unpaired samples t-tests were conducted between values. There were no

significant differences between groups in the protein expression of total mGlu5

(calculated as sum of the normalized I.D. values of the monomer and dimer: t(14) =

0.55, n.s; Fig. 3-4B), mGlu5 monomer (t(14) = 0.73, n.s; Fig. 3-4C) and mGlu5 dimer

(t(14) = 0.34, n.s). Groups also did not differ in protein expression of mGlu1, Homer

1b/c, PLCβ, pPKCs, and tPKC (n.s., Fig. 3-5 and Table 3-1).

3.3 Discussion

Here, no differences were found in long-term markers within the PrL (metabolic activity and expression of mGlu5 and its associated proteins) between rats with a history of extended access cocaine self-administration or saline during prolonged abstinence. Similar to the findings in Chapter 2 and in another study (Vélez-Hernández et al., 2014), no differences in metabolic PrL activity were observed. In the study by

Velez-Hernandez and colleagues (2014), daily cocaine or saline injections resulted in differences in metabolic activity within areas of the frontal cortex, but the PrL was the exception. However, PrL metabolic activity during prolonged abstinence after extended access cocaine self-administration is unknown. While the study in Chapter 2 investigated PrL metabolic changes within that paradigm and found no differences after cocaine versus saline, the effect of cognitive training to normalize baseline metabolic changes could not be definitively excluded. In the current study, a similar unchanged

PrL metabolic activity is noted between drug groups without a history of cognitive training. Here, the measure of CO reactivity may be capturing a time point of early withdrawal when changes reflective of PFC hypofunction after cocaine may not be apparent yet. Additionally, CO reactivity may be capturing a time point of self- administration when cocaine but not saline rats are heavily involved in operant learning, and any differences in PrL metabolic activity may be normalized by this group specific

training experience. Still, the investigation of other frontal cortical areas observed to show differences in metabolic activity after noncontingent cocaine still warrant further exploration following extended access contingent cocaine paradigms.

Here, no differences were noted in protein expression of mGlu5 and its associated proteins. This was similar to another study that found no differences in PrL mGlu5 expression between rats with a history of extended access cocaine or saline during prolonged abstinence (30 days) (Ben-Shahar et al., 2013). While greater mGlu5 and Homer 1b/c protein expression was noted in Chapter 2, the discrepancy between the aforementioned and current findings may hint to an effect of cognitive training in the previous paradigm, reflected in the correlations that were previously observed. Indeed, effects of cognitive training on protein expression have been noted previously (Fijał et al., 2015). In the study by Fijal and colleagues (2015), saline and cocaine rats did not differ in protein expression of pCREB/CREB after prolonged home cage withdrawal, but alterations of this marker (upregulation in the PFC and dorsal striatum and downregulation within the hippocampus), were observed for cocaine rats compared to saline rats after T-maze training, a task where they exhibited delay dependent working memory impairments. Additionally, in regions where pERK/ERK expression did not differ in saline and cocaine rats after prolonged home cage withdrawal, alterations

(upregulation in hippocampus and dorsal striatum) were noted after T-maze training.

3.4 Limitations

It is important to note that while long-term markers were evaluated during a prolonged abstinence period akin to timepoints coinciding with incubation of cocaine craving as in the previous experiment in Chapter 2, the abstinence duration here was shorter (45 versus 90 days). Thus, the length of prolonged abstinence may differentially

alter these markers. However, differences in behavioral responses to drug cues and some neural alterations were not noted when previously examined across time points ranging from 30-100 days, but rather between early (< 10 days) versus prolonged abstinence (≥ 30 days) (Ben-Shahar et al., 2013; Freeman et al., 2008; Grimm et al.,

2001; Loweth, Tseng, & Wolf, 2014; Shin et al., 2016; Szumlinski et al., 2016). Of additional consideration is that rats in the previous study were killed immediately following a 45-min relapse test, whereas rats in this study were killed from the home cage. It is possible that differences noted in Chapter 2 may have been altered acutely by the relapse test, though no correlation between protein expression and behavioral responding in the relapse test was uncovered.

3.5 Conclusions and Future Directions

Taken together, these findings suggest that the changes in long-term functional markers of the PrL between rats with a history of chronic cocaine or saline noted in

Chapter 2 may be attributed to post-cocaine cognitive dysfunction exhibited during a cognitive training period. As long-term markers within the PrL may change with a history of cognitive performance and correlate with working memory performance, measures of immediate changes within this region that can be directly tracked to working memory performance are of interest. Particularly, activation patterns characterizing performance in low vs high demand working memory conditions are of interest. This will be investigated in Chapter 4.

Figure 3-1. Experimental timeline. ShA: 1 hr short access to cocaine, LgA: 6 hr long access to cocaine.

Figure 3-2. Extended access cocaine self-administration. A) Rats discriminated between nose poking in the active versus inactive port throughout self-administration. B) Rats showed an escalation of cocaine intake. Average cocaine intake from the last three days of LgA was significantly greater than the average cocaine intake from the first three days of LgA. Error bars represent SEM. *p < 0.05, n = 8

Figure 3-3. Cytochrome oxidase histochemistry. A) (Left panel) Rat brain coronal section at the level of the PrL (+3.0 from Bregma (Paxinos & Watson, 2007) with highlighted location of the ROI selected for CO analysis. (Right panel) Representative images show CO reactivity (4x magnification) within the PrL in saline and cocaine rats. Dashed box represents ROI measuring 1x1mm. B) Rats with a history of saline or cocaine did not differ in overall PrL metabolic activity. Error bars represent SEM. n = 8/group.

Figure 3-4. Western blotting. A) PrL target ROI. Representative immunoreactive bands of mGlu5 protein as detected in rats with a history of cocaine versus saline at 90 days after the end of self-administration. Lanes display conditions in an alternating pattern (saline, cocaine, etc.) B) There were no group differences in total mGlu5 expression. Total mGlu5 expression within the PrL was measured as the sum of the I.D. values of the monomer and dimer normalized to calnexin and graphed as % of normalized I.D. in saline rats. C) There were no group differences in expression of mGlu5 monomer. I.D. of the mGlu5 monomer was normalized to calnexin and graphed as % of normalized I.D. in saline rats. Error bars represent SEM. n = 8/group.

Figure 3-5. Representative immunoreactive bands of selected proteins as detected in the PrL of saline and cocaine rats at 45 days after the end of self- administration. The following primary antibodies were used: mGlu1, Homer 1b/c, PLCβ1, phospho-PKCs, total PKC (tPKC), and the house-keeping control, calnexin (clnx).

Table 3-1. Protein levels of mGlu5-associated proteins in the PrL on day 45. Saline Cocaine mGlu1 100 ± 7.78 97.21 ± 12.10 Homer 1b/c 100 ± 5.42 99.77 ± 5.63 PLCβ 100 ± 4.88 93.71 ± 11.61 tPKC 100 ± 4.46 92.97 ± 2.58 pPKCs 100 ± 16.23 84.64 ± 9.14

Data represent mean immunoreactivity of each protein normalized to calnexin and expressed as percent of the Saline group ± SEM, n = 10/group.

CHAPTER 4 THE ROLE OF THE PRELIMBIC CORTEX IN WORKING MEMORY

In Chapter 2, metabolic activity within the PrL positively correlated with working memory performance in a DMS task for rats with history of cocaine who exhibited impairments in this task. Additionally, mGlu5 expression of the monomeric form negatively correlated with past working memory performance. These findings suggest an important role of the PrL and specifically mGlu5 function within this region to support working memory performance. To elucidate these findings, I assessed neuronal activation within the PrL that can be directly related to working memory performance. Of interest is the role of mGlu5 dependent activation within this region during low versus high demand working memory conditions. I hypothesized that greater mGlu5 dependent activation within the PrL will support challenging demand measures of working memory.

To address this, I assessed PrL neuronal activation after working memory conditions with low or high demand using the IEG c-fos and quantified its mRNA expression within mGlu5 expressing cells.

4.1 Materials and Methods

4.1.1 Animals

Adult male Sprague-Dawley rats (Charles River Laboratories; 275g on arrival; N

= 20 were first acclimated to the animal facility prior to any manipulation. They were housed individually, maintained on a 12 hr reverse light/dark cycle (lights off at 0700), and given ad libitum access to water and, and food restricted (15-20 g of food per day) to maintain ~ 85% of their free-feeding weight as previously described (Carroll et al.,

2002; Gobin & Schwendt, 2017). All animal procedures were approved by the IACUC of the University of Florida and were performed in accordance with the Guide for the Care

and Use of Laboratory Animals. The overall experimental design is depicted in Figure 4-

1 with individual behavioral procedures briefly described below.

4.1.2 Operant Delayed Match-to-Sample Task

Rats were trained and tested in the DMS task as previously detailed in Chapter

2. The day following baseline DMS testing, rats were assigned to two different test conditions in a (30 min) modified version of the task wherein each trial either consisted of only a 0s or 24s delay to test low demand (easy) or high demand (hard) working memory respectively (n = 7-8/group). Five rats were reserved as home cage controls.

4.1.3 Tissue Collection

Rats were killed by rapid decapitation immediately following the low or high demand DMS tasks or after being taken from the home cage. Brains were extracted and tissues were collected and processed as previously described in Chapters 2 and 3.

4.1.4 Fluorescent in situ Hybridization

Fluorescent in situ hybridization (FISH) for c-Fos and GRM5 (mGluR5) mRNAs was performed with the RNAscope Multiplex Fluorescent Reagent Kit (ACDBio,

Newark, CA) based on published procedures (Wang et al., 2012) and as previously described in Chapter 2 (Gobin et al., 2019). RNAscope target probes for c-Fos

(ACDBio: 403591-C1, lot 18179C) and GRM5 (ACDBio: 471241-C2, lot 17243B) were applied to each section. The AMP 4 Alt-C was selected so that c-Fos and GRM5 probes were labeled with ATTO 550 and Alexa 647 fluorophores, respectively. Fluorescent images were obtained using an Olympus BX51 microscope with a UPLANApo 40x objective, CCD high resolution camera (Tucsen Photonics), and ISIcapture software.

Quantification of c-Fos mRNA puncta was performed in MATLAB (Mathworks, Natick,

MA) using TransQuant software (Halpern & Itzkovitz, 2016) as previously described in

Chapter 2.

4.1.5 Statistical Analysis

IBM SPSS (Version 25) and GraphPad Prism (Version 5.03) software were used to analyze data. The alpha level was set at 0.05 for all statistical analyses used. One- way ANOVAs were used to compare group differences on single dependent measures.

Two-tailed Pearson correlations were conducted to compare continuous variables.

4.2 Results

4.2.1 Naïve Rats Exhibited Demand Dependent Performance during Baseline Testing. Naïve Rats Performed Worse in the High-Load DMS Task Compared to the Low-Load DMS Task

A two way ANOVA with Block as the within subjects factor and Delay as the within subjects factor revealed a main effect of Delay (F(6, 216) = 260.5, p < .0001; Fig.

4-2). Rats in the high-load working memory test of 24s performed significantly worse compared to rats in the low-load condition of 0s F(1, 62) = 279.2, p < 0.0001; Fig. 4-3).

4.2.2 Naïve Rats Demonstrated Greater mGlu5-Dependent Activation during the High Load Compared to the Low Load DMS Task. Performance in the High Load Condition Negatively Correlated with mGlu5-Dependent Activation

c-Fos mRNA was measured in PrL tissue collected from the rats immediately after the DMS test as well as from age-matched rats killed from the home cage to assess PrL activation following low versus high load working memory conditions. PrL c-

Fos mRNA expression was quantified as the total c-Fos mRNA puncta per target ROI averaged from two PrL sections for each rat. A one-way ANOVA revealed a significant difference between groups in PrL c-FOS mRNA expression (F(2, 21) = 9.87, p < 0.01).

Bonferroni’s Multiple Comparison Test indicated that rats in the high load condition expressed significantly more PrL c-Fos mRNA puncta compared to rats in the low-load

condition and home-cage controls, and rats in the low-load condition expressed significantly more PrL c-Fos mRNA puncta compared to the home-cage controls (Fig. 4-

4B). Next, the average number of c-Fos mRNA puncta across all mGlu5 mRNA expressing cells within the target ROI was averaged across two brain sections per rat.

Examination of c-Fos mRNA expression within PrL mGlu5 positive cells using one-way

ANOVA also indicated a significant difference between groups (F(2, 21) = 9.87, p <

0.01). Bonferroni’s Multiple Comparison Test revealed that rats in the high-load condition averaged higher levels of c-Fos mRNA within PrL mGlu5 expressing cells compared to rats in the low-load condition and home cage controls, and rats in the low- load condition exhibited this same activation compared to the home cage controls (Fig.

4-4C). Bivariate Pearson correlations revealed significant negative correlations between total PrL c-Fos mRNA expression and performance during the high load condition (r = -

.72, n = 8, p < .05; Fig. 4-4D) as well as between PrL c-Fos mRNA expression within mGlu5 cells and performance during the high load condition (r = -.79, n = 8, p < .05; Fig.

4-4E).

4.3 Discussion

Here, naïve rats demonstrated greater mGlu5 dependent neuronal activation within the PrL during conditions of high vs low demand working memory, and this activation pattern negatively correlated with working memory performance. In Chapter

2, I reported that decreased mGlu5 monomeric expression and greater metabolic activation during the PrL supported better working memory performance at the high delay. However, this was determined several weeks after working memory performance

(with other behavioral measures tested in the interim). Further, I was unable to separate the influence from performance at other delays. The findings here supplement these

previous findings highlighting that greater activation with the PrL supports working memory performance at the high delay specifically for an impaired group with cocaine history. Since mGlu5 dependent activation is necessary during high demand working memory performance for naïve rats and is more pronounced when the task is more difficult (accuracy is lower), this supports my previous hypothesis that cocaine impaired rats exhibit an inefficient, compensatory PrL metabolic activation to perform better during challenging phases of the task during a prolonged cognitive training/testing period. As no such correlation was exhibited in saline rats during a prolonged training/testing period in Chapter 2, this suggests a normalized efficient neuronal activation to support high demand working memory performance in the less impaired saline rats.

4.4 Conclusions and Future Directions

The findings here emphasize the importance of PrL activation and optimal mGlu5 function to support working memory during high demand conditions. As increased activation from the PrL in an mGlu5 dependent way similarly supports reexposure to drug cues, the assessment of pharmacotherapies that target inhibition of mGlu5 to decrease relapse should concurrently evaluate such modulation of mGlu5 on working memory performance. The next chapter employs an experimental design to address this research aim.

Figure 4-1. Experimental timeline. ShA: 1 hr short access to cocaine, LgA: 6 hr long access to cocaine.

Figure 4-2. Delayed match-to-sample task performance. Working memory performance (% correct) decreased with increasing delay interval for both blocks, n = 19.

Figure 4-3. Single delay DMS task performance. Rats in the 24s delay condition were significantly impaired compared to rats in the 0s delay condition. Error bars represent SEM. *p < 0.0001, n = 7-8/group.

Figure 4-4. Fluorescent in situ hybridization. A) c-fos/mGlu5 mRNA labeling and quantification for rats after a 0s or 24s delay DMS task and age-matched rats killed from the home cage. Representative images of dual c-fos/mGlu5 mRNA with DAPI nuclei counter staining in the PrL target ROI at 40x magnification. Scale bars represent 25µm. B) Rats in the 24s condition compared to rats in the 0s condition and age-matched controls showed more c-fos mRNA expression in the PrL. Rats in the 0s condition showed significantly more c- fos mRNA expression in the PrL compared to age-matched controls. C-fos mRNA expression was measured as total number of c-fos mRNA puncta within the PrL ROI. C) Rats in the 24s condition compared to rats in the 0s condition and age-matched controls expressed significantly more c-fos mRNA in mGlu5 expressing cells within the PrL. Rats in the 0s condition expressed significantly more c-fos mRNA in mGlu5 expressing cells within the PrL compared to age-matched controls. An average was taken of the total number of c-fos mRNA puncta in each mGlu5 positive cell within the PrL ROI. D) Total c-fos mRNA expression in the PrL negatively correlated with working memory performance (% correct) for rats in the 24s condition. E) Average c- fos mRNA in mGlu5 expressing cells within the PrL negatively correlated with working memory performance for rats in the 24s condition. Error bars represent SEM. *#p < 0.05. n = 5-8/group.

CHAPTER 5 MODULATION OF MGLU5 RECEPTORS ON POST-COCAINE WORKING MEMORY AND DRUG SEEKING

Cocaine use disorder is characterized by an inability to limit drug intake despite negative consequences (Koob & Volkow, 2016). Currently, no available FDA-approved medications possess the safety profile or display clinical efficacy to reduce relapse rates in CUD. A challenge to treat CUD has been partially attributed to elevated cocaine craving (Gawin & Kleber, 1986) and increased reactivity to cocaine-associated cues

(Parvaz et al., 2016) that persists over extended periods of abstinence, a phenomenon also replicated in animals (Grimm et al., 2001). Several neuroadaptations that may be responsible for elevated cocaine-seeking after a drug-free period have been identified in animal studies. One of the best-characterized and replicated findings is that of glutamate release into the NAc during the reinstatement of drug-seeking in rats (for review see: Scofield et al. 2016). Though dysregulated glutamate homeostasis at the corticostriatal synapses is thought to be primarily responsible (McFarland et al., 2003), other glutamatergic inputs (such as those from the hippocampus, basolateral amygdala and mediodorsal thalamus) may also contribute to the elevated glutamate release during cocaine craving and relapse (Matzeu, Weiss, & Martin-Fardon, 2015; Pelloux et al., 2018; Rogers & See, 2007). Thus, it has been proposed that pharmacotherapies that limit excessive glutamate release, or reduce postsynaptic glutamate binding may promote abstinence (Uys & LaLumiere, 2008). Due to potentially severe adverse side effects of ionotropic glutamate receptor antagonists, pharmacological modulation of postsynaptic metabotropic glutamate receptor 5 (mGlu5) has been explored as a viable approach for the development of anti-relapse therapies (Olive, 2009). A large number of preclinical studies demonstrate that systemic and intra-accumbens administration of

mGlu5 antagonists or NAMs selectively limit cocaine reward and reduce cocaine- seeking, broadening the appeal of mGlu5 inhibitors as therapeutics for CUD

(Chiamulera et al., 2001; Keck et al., 2013, 2014; Kenny et al., 2005; Knackstedt &

Schwendt, 2016; Knackstedt et al., 2014; Kumaresan et al., 2009; Lee et al., 2005; Li et al., 2018; Martin-Fardon et al., 2009; Wang et al., 2013),

Despite this evidence advocating for the use of mGlu5 inhibitors in anti-relapse therapy, there are potential complications that may prevent clinical use of these compounds. As mGlu5 receptors play a central role in certain forms of synaptic plasticity (such as LTD), it is not surprising that brain-wide inhibition, or ablation of mGlu5 can produce learning and memory side-effects (for review see: Simonyi et al.

2005; Olive, 2010). When considering novel approaches to treatment of CUD, it is critical to evaluate and consider potential cognitive side-effects, since up to 30% of dependent cocaine users and 12% of recreational users already exhibit global neurocognitive impairment (Vonmoos et al., 2013). Specifically, deficits spanning multiple domains such as attention, working memory, and executive function have been documented in subjects diagnosed with CUD (Cunha et al., 2013; Goldstein et al., 2004;

Jovanovski et al., 2005; Kübler et al., 2005; Pace-Schott et al., 2008; Potvin et al., 2014;

Vonmoos et al., 2013, 2014). Moreover, cognitive deficits may compromise overall treatment outcomes in CUD, as their severity correlates with increased craving

(Vonmoos et al., 2013) and relapse susceptibility (Verdejo-Garcia et al., 2014), reduced impulse control (Albein-Urios et al., 2012), or decreased efficacy of CBT (Aharonovich et al., 2006; Streeter et al., 2008). In recent years, several known cognitive enhancers were evaluated as potential treatments for substance use disorders (for reviews, see:

100

Brady et al., 2011; Mahoney, 2018). However, these studies revealed no, or only modest benefit of cognitive enhancers on cocaine use or abstinence rates (Dackis et al.,

2005; Morgan et al., 2016; Nuijten et al., 2015), and did not concurrently investigate efficacy of these compounds on both cognitive function and cocaine relapse.

This study aimed to address this knowledge gap by evaluating the effects of mGlu5 NAMs and PAMs in an animal model of CUD that allows for the within-subject evaluation of cocaine-seeking and cognitive performance. To that end, cocaine-seeking in rats with a history of extended access cocaine self-administration was assessed after a period of prolonged abstinence that corresponded with a time point during which

“incubation of cocaine craving”, or elevated (‘incubated’) responses to cocaine- associated cues have been previously demonstrated in both humans (Parvaz et al.,

2016) and rodent models (Freeman et al., 2008; Grimm et al., 2001; Lu et al., 2003).

During this period, the effects of repeated treatment with mGlu5 allosteric modulators on working memory performance and cocaine-seeking were respectively evaluated in a

DMS task and upon re-exposure to cocaine paired context and cues (relapse test).

Using this experimental design in Chapter 2, extended-access cocaine self- administration produced not only robust cocaine-seeking, but also persistent working memory deficits that were associated with long-term changes in metabolic activity and mGlu5 expression within the PFC (Gobin et al., 2019). Here, the following hypotheses were tested: (1) repeated systemic administration of an mGlu5 NAM will impair working memory (2) repeated systemic treatment with an mGlu5 PAM will improve working memory and (3) acute administration of the mGlu5 NAM, but not PAM will attenuate cocaine-seeking. Here, the mGlu5 NAM 3-((2-Methyl-1,3-thiazol-4-yl)ethynyl)pyridine

101

hydrochloride (MTEP) and the mGlu5 PAM 3-Cyano-N-(1,3-diphenyl-1H-pyrazol-5- yl)benzamide (CDPPB) were chosen to be evaluated due to their low toxicity as well as the wealth of scientific data supporting anti-relapse properties of MTEP (Keck et al.,

2014; Knackstedt & Schwendt, 2016; Knackstedt et al., 2014; Kumaresan et al., 2009) and pro-cognitive properties of CDPPB (Fowler et al., 2013; LaCrosse et al., 2014;

Marszalek-Grabska et al., 2018; Reichel et al., 2011; Stefani & Moghaddam, 2010;

Uslaner et al., 2009).

5.1 Materials and Methods

5.1.1 Animals

Adult male Sprague-Dawley rats (Charles River Laboratories; 275g on arrival; N

= 38) were first acclimated to the animal facility prior to any manipulation. They were housed individually, maintained on a 12 hr reverse light/dark cycle (lights off at 0700), and given ad libitum access to food and water, except as noted below. All animal procedures were approved by the IACUC of the University of Florida and were performed in accordance with the Guide for the Care and Use of Laboratory Animals. A timeline of the overall experimental design is depicted in Figure 5-1 with individual behavioral procedures described below. Ten rats were excluded prior to assignment to receive drug or vehicle. Four of these rats died during the cocaine self-administration phase of the experiment, and six failed to reach the predetermined criterion during DMS training (N = 28). Additionally, one rat was considered an outlier during the relapse test according to a Grubb’s test for outliers and was thus excluded from analysis.

5.1.2 Drugs

Cocaine hydrochloride was acquired from the NIDA Controlled Substances

Program (Research Triangle Institute, NC) and was dissolved in 0.9% sterile

102

physiological saline. MTEP (Abcam Biochemicals, Cambridge, MA) and CDPPB (EAG

Laboratories, Maryland Heights, MO) were dissolved in 10% Tween 80 (Sigma-Aldrich) in 0.9% NaCl. Tween 80 (10%) in 0.9% NaCl was used as the vehicle solution. MTEP was neutralized to pH 7.2-7.4 with 1N NaOH. MTEP (3 mg/kg) was administered intraperitoneally (IP) ten minutes prior to behavioral testing. The dose, timing and route of MTEP administration was carefully selected based on previous reports. Specifically, this dose of MTEP resulted in a rapid and full receptor occupancy in adult Sprague-

Dawley rats that is maintained for at least one hour (Anderson et al. 2003); reduced cocaine seeking (Hao, Martin-Fardon, & Weiss, 2010; Martin-Fardon & Weiss, 2012;

Martin-Fardon et al., 2009), and was devoid of nonspecific effects on appetitive reinforcement or motor function (Gass et al., 2009; Martin-Fardon et al., 2009). CDPPB was administered subcutaneously at a dose of 30mg/kg, 20 minutes prior to behavioral testing based on previously published evidence showing pro-cognitive effects of this dose on different types of learning (Fowler et al., 2013; Olive, 2010; Reichel et al., 2011;

Stefani & Moghaddam, 2010) without locomotor side-effects (Gass & Olive, 2009).

5.1.3 Surgery and Cocaine Self-Administration

All rats underwent jugular catheter implantation, post-surgery recovery and cocaine self-administration as previously described in Chapters 2 and 3 and according to previously published protocols (Gobin & Schwendt, 2017; Gobin et al., 2019).

Beginning during cocaine self-administration and lasting throughout the duration of the experiment, rats were restricted to 15-20 g of food per day to maintain ~ 85% of free- feeding weight (Gobin & Schwendt, 2017; Gobin et al., 2019).

103

5.1.4 Operant Delayed Match-to-Sample Task

Rats were trained and tested in the DMS task as previously described in

Chapters 2, 3, and 4 as well as in published protocols (Beas et al., 2013; Gobin et al.,

2019; Sloan et al., 2006). Rats were initially trained without any delays between the sample and choice phase. A correction procedure was employed to prevent the development of side biases prior to being trained at two delay sets: short delay set {0s,

1s, 2s, 3s, 4s, 5s, 6s} and intermediate delay set {0s, 2s, 4s, 8s, 12s, 16s}. Rats were required to reach a criterion of ≥ 80% correct for two consecutive days at each training delay set. After reaching criteria, they entered the testing phase for two blocks of five days at the final delay set {0 s, 2s, 4s, 8s, 12s, 18s, 24s}. This phase is designated here as the ‘Baseline’ responding. Next, rats received systemic injections of drug (Vehicle,

CDPPB, MTEP) daily prior to DMS testing, for a block of five consecutive days. This was followed by another five-day block of DMS testing, during which no drugs were administered (‘Washout’). One group of rats (1xCDPPB group; n = 7) did not receive any drug treatments. Instead, after they completed the baseline DMS testing, they proceeded directly to relapse testing as described below. See the experimental timeline

(Fig. 5-1) for details.

5.1.5 Relapse Tests

On the day after the final day of DMS testing, all rats underwent a 45-min context+cue relapse test in the self-administration chamber, wherein the cues previously paired with cocaine were presented upon nose poking in the previously active port without cocaine delivery. Rats who received drug treatments prior to DMS testing were administered the same compounds prior to the relapse test. A second context+cue relapse test without prior drug treatments was conducted the following day

104

to assess washout drug effects. Additionally, to assess the effect of a single administration of CDPPB on ‘relapse’ without previous repeated treatment with the compound, one group of rats (n = 7) did not receive drug treatments during DMS testing and instead received a single injection of CDPPB 20 mins prior to the relapse test

(Fig.5-1). A group receiving a single injection of MTEP was not included since it has been previously demonstrated that a single systemic administration of MTEP reduces relapse (Knackstedt & Schwendt, 2016). Since rats reached criterion and finished testing in the DMS test at different time points, the range for the day of relapse testing was 45-80 days. It is unlikely this variable range significantly influenced our data, as responding to drug associated cues remains high and stable between 30 and 100 days of post-cocaine abstinence (Freeman et al., 2008).

5.1.6 Locomotor Testing

Locomotion was tested in a chamber containing sensors (San Diego Instruments;

40 cm length × 44 cm width x 37 cm) to detect beam breaks in 5-minute bins over the course of one hour. Before locomotor testing, rats received the same drug as previously administered prior to DMS testing and/or relapse testing.

5.1.7 Statistical Analysis

IBM SPSS (Version 25) and GraphPad Prism (Version 5.03) software were used to analyze data. The alpha level was set at 0.05 for all statistical analyses used. A repeated-measures ANOVA was used to analyze discrimination between the active and inactive nose poke ports across the 18 daily sessions of cocaine self-administration, with Port as the between-subjects factor and Day as the within-subjects factor. To assess cocaine escalation, a repeated-measures ANOVA was conducted to compare mean cocaine intake across the 12 daily LgA sessions of cocaine self-administration.

105

DMS task acquisition was calculated as the total number of days to meet criterion prior to entering the DMS testing phase. A one-way ANOVA was conducted to assess group differences in number of days to reach criterion prior to DMS testing. During the DMS testing phase, ‘percent correct’ was the primary dependent measure. Percent correct at each delay interval {0 s, 2s, 4s, 8s, 12s, 18s, 24s} was averaged over each five-day testing block. A repeated measures ANOVA was used to assess the effect of treatment on working memory performance for each condition with Block (Baseline, Drug,

Washout) as the between-subjects factor and Delay as the within-subject factors.

During the relapse tests, the measure of cocaine seeking was operationalized as the number of responses in the previously active port assessed between groups and also compared to within group responding in the inactive port. A two-way ANOVA was used to assess group differences in the number of nose pokes in the active and inactive ports during both context+cue relapse test days. A one-way ANOVA was conducted to assess group differences in locomotor performance. Bonferroni post-hoc tests were used when appropriate.

5.2 Results

5.2.1 Rats Escalated Cocaine Intake during Extended Access Cocaine Self- Administration

All rats demonstrated successful discrimination between the active and inactive nose poke port throughout self-administration (Fig. 5-2A). A repeated-measures

ANOVA revealed main effects of Port (F(1, 62) = 279.2, p < 0.0001) and Day (F(17,

1054) = 39.45, p < 0.0001) with a Port x Day interaction (F(17, 1054) = 45.32, p <

0.0001). A repeated measures ANOVA with Greenhouse-Geisser correction showed that rats displayed an escalation of cocaine intake (mg/kg) over the 12 days of LgA

106

sessions (F(5.20, 171.68) = 15.69, p < 0.001). Post-hoc tests using Bonferroni correction revealed that cocaine intake on LgA days 4-12 were significantly higher than on day 1 (p < 0.05; Fig. 5-2B). On average, total cocaine intake (± SEM) for the whole period of self-administration was 821.05 ± 32.86 mg/kg, with the long-access self- administration phase accounting for the majority of cocaine intake, 778.04 ± 29.75 mg/kg.

5.2.2 Rats did Not Differ in DMS Task Acquisition or Baseline Performance Prior to Treatment

A one-way ANOVA did not reveal a difference between groups on task acquisition prior to treatment (F(3, 25) = 1.35, n.s; Fig. 5-3A). Prior to behavioral pharmacology testing, rats were matched to treatment groups such that there were no group differences in total cocaine intake (F(3, 25) = 2.76, n.s.) and baseline DMS performance across all delays (no effect of Group F(3,25) = 1.58, n.s.; and no interaction F(18, 144) = 1.21, n.s).

5.2.3 MTEP Impaired while CDPPB had Delayed Pro-Cognitive Effects on Working Memory Performance

For the Vehicle treatment group, main effects of Delay (F(6, 98) = 29.27, p <

0.0001) and Block (F(2, 98) = 9.55, p < 0.001) were found without a significant Delay x

Block interaction (Fig. 5-3B). For the MTEP treatment group, main effects of Delay (F(6,

84) = 51.52, p < 0.0001) and Block (F(2, 84) = 26.99, p < 0.0001) as well as a Delay x

Block interaction (F(12, 84) = 3.05, p < 0.01) were found (Fig. 5-3C). Bonferroni post- tests revealed that MTEP treatment reduced the percent correct at the 12s, 18s, and

24s delays compared to baseline (ps < 0.01- 0.001). MTEP performance was also impaired at the 8s, 12s, 18s, and 24s compared to the washout block (ps < 0.001- 0.05).

(Fig. 5-3C). For the CDPPB treatment group, main effects of Delay (F(6, 70) = 50.43, p

107

< 0.0001) and Block (F(2, 70) = 9.06, p < 0.001) were found without a significant Delay x Block interaction (Fig. 5-3D). The effects of MTEP on working memory are not likely due to decreased appetitive behavior, as groups did not differ in total number of trials completed averaged over the five-day treatment block (F(2, 20) = 3.38, n.s). The mean

(± SEM) number of trials completed over the 5-day treatment period was: 102 ± 4.98

(Vehicle group), 101.7 ± 3.55 (MTEP group), and 116.5 ± 1.12 (CDPPB group).

To assess the possibility that CDPPB exerts gradual pro-cognitive effects that may emerge following additional drug-free testing, rats were retested following all other behavioral measures in this experiment (relapse and locomotion testing). This time- point was ~6 days after the washout block. All rats previously receiving CDPPB were able to be retested. However, only some previously treated Vehicle (n = 4) or MTEP (n

= 2) rats were able to be retested because most of them were already killed prior to deciding to conduct additional testing. A repeated measures ANOVA was conducted with Delays as the within-subjects factor and Block (previous washout block vs. follow- up testing block) as the between-subjects factor. A significant interaction was revealed only for rats previously treated with CDPPB (F(6, 35) = 6.22, p < 0.001; Fig. 5-5A) and not for rats previously treated with Vehicle (F(6, 21) = 1.61, n.s.; Fig. 5-5B) or MTEP

(F(6, 7) = 2.52, n.s.; Fig. 5-5C). Bonferroni post-tests revealed that an improvement was detected at the 12s (p < .01) and 24s (p < .001) delays for rats previously treated with

CDPPB.

5.2.4 Both MTEP and CDPPB Decreased Drug-Seeking in a Context + Cue Relapse Test

A two-way ANOVA revealed main effects of Port (F(1, 23) = 41.75, p < 0.0001),

Treatment (F(3, 23) = 3.76, p < 0.05) and a Port x Treatment interaction (F(3, 23) =

108

3.85, p < 0.05) during the first relapse test (Relapse test+drug; Fig. 5-4A). Bonferroni post-hoc tests showed that rats in the MTEP, repeated CDPPB, and 1xCDPPB groups exhibited less nose pokes in the previously active port compared to the Vehicle treatment group (ps < 0.01-0.001). Only rats in the Vehicle group significantly differed in the average number of nose pokes in the active port versus the inactive port (p <

0.001). On the following day (Relapse test – drug; Fig. 5-4B), a two-way ANOVA revealed main effects of Port (F(1, 23) = 58.77, p < 0.0001), but no effect of Treatment

(F(3, 23) = 1.36, n.s) and no significant Port x Treatment interaction (F(3, 23) = 2.17, n.s).

5.2.5 Drug Treatments did Not Alter Spontaneous Novelty-Induced Locomotion

Next, the doses of the drugs used in this study were evaluated on locomotor activity. A one-way ANOVA showed that treatment groups (Vehicle, MTEP, CDPPB) did not differ in the overall locomotion in a novel environment, measured as average number of beam breaks during a 1 hour session (F(2, 21) = 1.74, n.s.). Mean number of beam breaks (± SEM) for each group was as follows: Vehicle (5023 ± 385), MTEP

(4370 ± 229) and CDPPB (5529 ± 387).

5.3 Discussion

These data indicate that both allosteric inhibition and activation of mGlu5 reduces relapse to cocaine-seeking in rats, even after prolonged abstinence. At the same time, the dose of MTEP that decreased relapse also produced significant working memory deficits. On the other hand, allosteric activation of mGlu5 with CDPPB at a dose that reduced relapse, spared working memory, indicating that mGlu5 PAMs might offer a safer approach to reduce cocaine relapse.

109

Expanding on previous studies showing that systemic MTEP reduces cocaine- seeking (Keck et al., 2014; Knackstedt & Schwendt, 2016; Kumaresan et al., 2009), this is the first study to report that systemic MTEP administration also attenuates

(context+cue)-elicited cocaine-seeking following extended access cocaine self- administration and prolonged abstinence (45-80 days). In other words, MTEP attenuates cocaine-seeking in animals that typically show sustained elevation of drug- seeking and resistance to extinction, behavior previously described as ‘incubation of cocaine craving’ (Freeman et al., 2008; Grimm et al., 2001; Lu et al., 2003). Systemic

MTEP may be partially targeting the NAc to control drug seeking as previous studies reported that MTEP microinfused into the NAc was sufficient to reproduce the anti- relapse effects of systemic MTEP administration (Knackstedt et al., 2014; Li et al., 2018;

Wang et al., 2013). However, mGlu5 receptors are widely distributed in other brain regions (Rodrigues et al., 2002; Romano et al., 1995; Shigemoto et al., 1993) known to play a role in cue-elicited cocaine-seeking (Carelli, Williams, & Hollander, 2003; Pelloux et al., 2018). As previous studies have shown that intra-vmPFC or intra-dlSTR blockade of mGlu5 do not reduce relapse to cocaine-seeking (Ben-Shahar et al., 2013;

Knackstedt et al., 2014), anatomically separate populations of mGlu5 in the brain may play distinct roles in regulating drug-seeking. Thus, the possible contribution of mGlu5 receptors outside of the NAc to drug-seeking behavior warrants further, more comprehensive evaluation.

The MTEP-induced disruption of post-cocaine working memory performance was most apparent during longer delays (8-24s), suggesting a delay-dependent (or demand- dependent) impairment. These effects of MTEP were acute and did not persist past the

110

drug treatment period (denoted here as washout). In Chapter 2, the same regimen of cocaine self-administration used in the present study produced deficits in working memory performance that persisted over weeks of drug-free abstinence (Gobin et al.,

2019). As all rats used in the current study had a cocaine history, their working memory ability was likely already impaired prior to MTEP administration (though here we did not have a saline control to demonstrate this), indicating that treatment with this class of drugs may produce further cognitive impairment in recovering addicts. The deficits produced by MTEP here are of comparable magnitude to those resulting from mPFC lesions in the same task (Sloan et al., 2006). In addition to the mPFC, lesion and inactivation studies using match-to-sample and nonmatch-to-sample operant tasks have identified that circuitry supporting ‘normal’ working memory performance also includes the dmSTR (Akhlaghpour et al., 2016). Thus, systemic MTEP may be impairing working memory through inhibition of specific cortico-striatal circuits. The acute effects of MTEP here also suggest that mGlu5 activity is critical for ongoing working memory performance. Within the mPFC, mGlu5 is predominantly expressed postsynaptically on neurons projecting to various striatal regions, including the NAc and dmSTR (Romano et al., 1995). Indeed, administration of MTEP directly into the mPFC impairs working memory using the analogous DMS task (Hernandez et al., 2018).

As MTEP impaired working memory, an mGlu5 PAM was predicted to enhance working memory. Here, the effects of the mGlu5 PAM CDPPB were tested as this compound has been previously shown in other tasks to exert pro-cognitive effects

(LaCrosse et al., 2014; Reichel et al., 2011; Stefani & Moghaddam, 2010; Uslaner et al.,

2009). Surprisingly, pro-cognitive effects of CDPPB on post-cocaine working memory

111

performance were not noted during the treatment phase. Previously, the same dose of

CDPPB (30mg/kg) improved methamphetamine-induced episodic memory deficits

(Reichel et al., 2011), as well as MK-801-induced deficits in cognitive flexibility (Stefani

& Moghaddam, 2010) and spatial learning (Fowler et al., 2013). Several factors could account for the lack of effect found in this study. In these studies, CDPPB effects were observed in significantly impaired animals, with the discrimination or response accuracy near, or at chance (Reichel et al., 2011; Stefani & Moghaddam, 2010). And further, in these tasks facilitatory effects of CDPPB typically applied to new learning. In contrast, my previous experiments demonstrated that chronic cocaine produces sustained, but relatively mild cognitive impairment in rats assessed with the DMS task (Gobin &

Schwendt, 2017; Gobin et al., 2019) and this study). Also, in this study, CDPPB was evaluated on rats who had an extensive history of DMS training. And finally, while the difficulty of the task increased during testing (longer delays), new learning was not required to master the DMS task. To assess possible delayed effects of CDPPB on

DMS performance at the long delays, a continuation study was conducted (maintaining individual group designations) wherein rats’ working memory performance was re- evaluated after relapse and locomotor tests were completed (~6 days after the end of the washout period). Under these conditions, a working memory enhancement was observed in rats previously treated with CDPPB compared to their previous washout block, but not in rats with a history of Vehicle or MTEP treatment.

In contrast to the divergent effects MTEP and CDPPB on working memory, both drugs reduced cocaine-seeking triggered by cocaine-associated context and cues. The anti-relapse effects of CDPPB were not due to cumulative effects of prior repeated

112

treatment with this drug (5-day CDPPB treatment regimen during DMS testing), as single CDPPB administration (20 mins prior to the relapse test) also reduced cocaine- seeking. As intra-NAc administration of orthosteric mGlu5 agonist CHPG has been shown to promote cocaine-seeking or conditioned effects of cocaine (Benneyworth et al., 2019; Wang et al., 2013), potent anti-relapse effects of CDPPB may seem surprising. However, the relapse test as conducted in the current study is analogous to an extinction session on day 1 in studies that demonstrated extinction-enhancing effects of CDPPB (Cleva et al., 2011; Olive, 2010). Importantly, on extinction day 1, these studies show a similar CDPPB-induced decrease in overall responding to drug cues

(Cleva et al., 2011). Thus, the CDPPB-induced reduction in drug seeking may be attributed to acute enhancement of extinction learning occurring already within the first session.

Allosteric modulators typically regulate functional receptor activity specifically in brain areas where the endogenous agonist exerts its physiological effects (Stansley &

Conn, 2019). Thus, it is possible that MTEP and CDPPB (two mechanistically different compounds) may be acting at opposing brain regions to modulate mGlu5 function related to the expression of distinct behaviors (decreased motivation to seek the drug vs. enhancement of extinction learning) that would result in the same outcome

(decrease overall responding in the relapse test). MTEP may be decreasing the ability of drug-associated cues to drive PrL brain activation and behavior (Keck et al., 2014;

Kumaresan et al., 2009; Martin-Fardon et al., 2009) via an mGlu5-mediated inhibition of neuronal excitability within this region (Timmer & Steketee, 2012) or the NA core (Wang et al., 2013), but not infralimbic cortex (IL, (Ben-Shahar et al., 2013), whereas CDPPB

113

may be facilitating within-session extinction learning via an mGlu5-mediated activation within the IL (Ben-Shahar et al., 2013). Furthermore, the efficacy of these modulators in enhancing mGlu5 activity depends on the concentration of endogenous/orthostatic agonist (extracellular glutamate) and on mGlu5 agonist-modulator cooperativity factor

(Stansley & Conn, 2019). As such, region-specific dysregulation of mGlu5 signaling

(Gobin et al., 2019), mGlu5 function (Ben-Shahar et al., 2013) and glutamate homeostasis (Knackstedt & Kalivas, 2009) after chronic cocaine may further determine circuit-specific effects of systemic MTEP and CDPPB administration.

The current study also assessed cocaine-seeking one day after the initial relapse test (Relapse test +drug), to evaluate possible carryover effects of prior pharmacological manipulations on cocaine-seeking (Relapse test -drug). Here, prior treatment with

MTEP or CDPPB did not affect drug-seeking in this subsequent test. This finding is in agreement with previous observations that single intra-dSTR, but not systemic MTEP administration interferes with post-relapse extinction learning (Knackstedt & Schwendt,

2016; Knackstedt et al., 2014). It is possible that repeated MTEP administration is needed to decelerate extinction (Kim et al., 2014). In the case of CDPPB, a similar repeated daily administration may be necessary to produce extinction-enhancing effects across sessions (Olive, 2010). It should be noted that one study observed an effect of

CDPPB or MTEP administration on cocaine-seeking during a subsequent ‘undrugged’ extinction test (Perry et al., 2016). However, in this study, either compound was administered prior to a CS+ extinction session that was separate from operant extinction. Another factor not addressed in the aforementioned studies, is the resistance to extinction in rats with a history of extended access to cocaine (Freeman et al., 2008;

114

Grimm et al., 2001; Lu et al., 2003), and also observed in our Vehicle-treated rats. Thus, it is possible that longer extinction sessions (45mins vs. 2 hrs) are necessary to observe retarding or faciliatory effects of mGlu5 PAMs on extinction in these animals.

5.4 Limitations

It should be noted that multiple doses of MTEP or CDPPB were not tested in this paradigm, and as such, the cognitive side effects of MTEP and anti-relapse effects of

CDPPB noted here may not hold at other tested doses. As anti-relapse effects of MTEP are noted at lower doses of 0.5- 1mg/kg, these doses may not impair cognitive performance and should be further investigated. Additionally, other doses of CDPPB may provide more robust cognitive enhancing effects as well as anti-relapse effects.

5.5 Future Directions

While, I hypothesize region-specific effects of these compounds that should be further investigated, these types of evaluations will not fully elucidate their mechanism of action as distinct subpopulation of neurons within the cortico-striatal circuit have been identified to play dual roles in cocaine seeking or cocaine suppression (Warren et al.,

2016). As such, distinct labeling of neuronal ensembles in relation to how these compounds reduce cocaine seeking should be explored. Furthermore, the effect of mGlu5 NAMs and PAMs on cocaine induced glutamate release at corticostriatal synapses should also be investigated.

5.6 Conclusions

The current study is the first to investigate the effects of positive and negative mGlu5 allosteric modulation on both cognitive performance and persistent cocaine- seeking in the same subjects. I found that repeated systemic administration of the mGlu5 NAM MTEP impaired working memory performance, while repeated systemic

115

treatment with the mGlu5 PAM CDPPB had modest effects on working memory.

Interestingly, both compounds attenuated relapse to cocaine-seeking triggered by combined discrete and contextual cocaine cues, without affecting subsequent extinction learning. I hypothesize that systemic administration of MTEP reduces the ability of mGlu5 receptors within the cortico-striatal pathway to mediate cocaine-seeking, at the cost of inhibiting working memory processes. On the other hand, the ability of CDPPB to exert the same anti-relapse effect (though free of cognitive side-effects) makes mGlu5 allosteric modulation an attractive target for the development of future therapies of CUD.

These initial observations motivate further research into the brain circuitry and neurobiological mechanisms through which mGlu5 PAMs reduce drug seeking.

116

Figure 5-1. Experimental timeline. DMS: Delayed match-to-sample task. ShA: short, 1 hr access to cocaine self-administration. LgA: long, 6 hr access to cocaine self-administration.+drug: daily Vehicle, MTEP or CDPPB admininstration, - drug: no drug was administered.

Figure 5-2. Extended access cocaine self-administration. A) Rats discriminated between nose poking in the active versus inactive port throughout self-administration. There were main effects of Port and Day as well as a Port x Day interaction. B) Rats showed escalation of cocaine intake on days 4-12 of LgA cocaine self-administration. Error bars represent SEM. ***p < 0.001 vs. intake on Day 1 of LgA cocaine self-administration, n = 28.

117

Figure 5-3. Delayed match-to-sample task performance with drug treatments. A) Animals assigned to four treatment groups did not display pre-existing differences in task acquisition (number of days to reach criterion). B) Treatment with vehicle did not affect task performance (% correct) compared to baseline and washout blocks. C) Task performance significantly decreased during the MTEP (3 mg/kg, i.p.) treatment block compared to the baseline block at 12s, 18s, and 24s and compared to the washout block at 8s,12s, 18s, and 24s. D) Treatment with CDPPB (30 mg/kg, i.p.) did not alter task performance compared to baseline or washout blocks. Error bars represent SEM. ##p < 0.01, ###p < 0.001 MTEP vs. Baseline, *p < 0.05, ***p < 0.001 MTEP vs. Washout, n = 6-8/group.

118

Figure 5-4. Context + cue relapse with drug treatments. A) Rats with a history of vehicle prior to DMS testing and relapse exhibited greater # nose pokes in the active port compared to rats with a history of MTEP (3 mg/kg, i.p.) and CDPPB (30 mg/kg, i.p.) prior to DMS testing and relapse as well as compared to rats given only a single administration of CDPPB prior to relapse. Vehicle-treated rats also nose poked more in the active nose port compared to the inactive nose port. B) Groups did not differ on the # nose pokes in the previously active port on day 2. Error bars represent SEM. ***p < 0.001 vs. Vehicle active port nose pokes. ###p < 0.001 vs. inactive port nose pokes. n = 5- 8/group.

119

Figure 5-5. DMS Retesting of CDPPB, Vehicle and MTEP treated rats. Retesting started six days after the end of the initial washout period with no drugs administered. A) Rats previously treated with CDPPB showed an improvement in working memory performance at the 12s and 24s delays, as compared to the past Washout block. *p < 0.05 vs. Washout, n = 6. B) Rats previously treated with Vehicle showed no change in working memory performance compared to the previous Washout block, n = 4. C) Rats previously treated with MTEP showed no change in working memory perfomance compared with the previous Washout block, n = 2. Data represent an average of 5 daily DMS sessions. Error bars represent SEM.

120

CHAPTER 6 GENERAL DISCUSSION

In this dissertation, I employed a translational rodent model that evaluated the following behavioral measures (alone or in combination): extended access cocaine self- administration, working memory and cue-elicited cocaine seeking. I found that rats with a history of extended access cocaine self-administration displayed deficits in working memory but not reversal learning as assessed using an operant DMS/NMS task

(Chapter 2). Post-cocaine impairments were observed after a prolonged drug-free period but within a 90-day abstinence window when drug seeking in response to cocaine-associated cues was shown to be robust (Chapter 2). Some aspects of cognitive performance assessed with the operant DMS/NMS task that may reflect impairments in attention or inhibitory control predicted drug seeking (Chapter 2).

Metabolic activity within the PrL positively correlated with high demand working memory performance only for rats with a history of extended-access cocaine (Chapter 2).

Furthermore, rats with a history of both extended-access cocaine and DMS/NMS testing

(Chapter 2), but not solely a history of extended-access cocaine (Chapter 3), exhibited greater protein expression of the monomeric form of mGlu5 as well as Homer 1b/c, and the mGlu5 monomer negatively correlated with past high demand working memory performance. These results suggest a compensatory mechanism within the PrL or cortical inefficiency after cocaine to regulate post-cocaine cognitive dysfunction during a cognitive testing period. Furthermore, I found that mGlu5 dependent activation supports reexposure to cocaine-associated cues (Chapter 2) as well as high demand working memory performance (Chapter 4). Since these and other previously published findings suggest the involvement of mGlu5 in drug seeking and working memory performance,

121

the effects of mGlu5 allosteric modulation on both behaviors was assessed in Chapter

5. Consistent with other preclinical studies, acute systemic negative modulation of mGlu5 with MTEP decreased drug seeking (Keck et al., 2014; Knackstedt & Schwendt,

2016; Knackstedt et al., 2014; Kumaresan et al., 2009), but repeated systemic treatment dramatically impaired working memory performance. Additionally, acute systemic positive modulation of mGlu5 with CDPPB decreased drug seeking, while repeated treatment had neutral effects on working memory. Follow-up testing six days later revealed an enhancement in high demand working memory performance for the rats previously treated with CDPPB but not vehicle or MTEP, suggesting delayed pro- cognitive effects of CDPPB in this paradigm.

6.1 Future Directions

Future directions include (1) assessing group differences in baseline metabolic activity at different time points and within additional brain regions, (2) extending the investigation of other brain regions overlapping working memory dysfunction and drug seeking, and (3) exploring region-specific or cell-type specific effects of MTEP and

CDPPB on working memory and drug seeking.

As I found no differences in overall metabolic activation in the PrL between rats with a history of cocaine or saline following 45-day home-cage abstinence or 90-day cognitive testing/abstinence, additional time points should be examined. The metabolic activation assessed in rats killed on day 90 may have captured the time point of working memory testing six weeks prior wherein the compensatory increased metabolic activation to support working memory after cocaine might have erased baseline differences between groups. On the other hand, the metabolic activation assessed in rats killed on day 45 may have captured the time point of self-administration six weeks

122

prior when cocaine rats were engaged in operant drug learning. Again, this type of learning may have normalized any baseline differences. The assessment of metabolic activation within the PrL following 90-day home cage abstinence may instead provide more useful information about the effects of cocaine or saline on baseline metabolic activation without residual normalizing effects of operant learning. Additionally, other regions should be assessed as another study noted that non-contingent cocaine versus saline did not alter metabolic activation within the PrL but did for other PFC regions

(Vélez-Hernández et al., 2014).

Furthermore, while the focus of this dissertation was to investigate neurobiological substrates within the rodent PrL in relation to working memory deficits and drug seeking, the involvement of other regions overlapping these behavioral dysfunctions also warrant further investigation. Dissociable roles of the PrL and IL have been noted with the former being implicated in working memory and responding to drug cues and the latter involved in inhibiting drug responses and extinction learning (Ben-

Shahar et al., 2013; Vertes, 2004). Additionally, the dmSTR has been implicated in working memory performance as well as goal directed drug seeking (Akhlaghpour et al.,

2016; Murray et al., 2012). Accordingly, mGlu5 dependent activation within the IL and dmSTR should be assessed following a drug seeking event as well as after low and high working memory demand conditions.

Finally, region specific effects of MTEP and CDPPB should be explored to understand how systemic administration of these compounds can exert distinct effects on working memory but the same effects on reducing drug seeking. The effects of systemic administration of these compounds on mGlu5 dependent activation following a

123

working memory condition or drug seeking event should be assessed within multiple regions including the PrL, IL, NAc, dmSTR, and the mediodorsal thalamus. MTEP may be inhibiting mGlu5 dependent activation within the PrL, dmSTR, and or mediodorsal thalamus to impair working memory as well as within the PrL and NAc to decrease drug seeking. CDPPB may be enhancing mGlu5 dependent activation within the PrL and thalamus to improve working memory as well as within the IL to enhance extinction learning. Additionally, cue-induced glutamate release from the PrL that drives drug seeking, and respective increases and decreases in PrL and IL mGlu5 expression (Ben-

Shahar et al., 2013) may further drive region specific effects of these compounds to decrease drug seeking. MTEP may be decreasing drug seeking by decreasing cue- induced glutamate release within the PrL, while CDPPB may be enhancing extinction learning of drug cues by enhancing glutamate signaling within the IL.

6.2 Conclusions

In my dissertation, I showed that working memory impairments following cocaine overlap within a period of persistent responding to cocaine-associated cues. Greater metabolic activation and increased expression of the mGlu5 monomer in the PrL correlated with working memory performance and may reflect compensatory mechanisms of post-cocaine cognitive dysfunction. My findings highlight the role of the

PrL and more specifically, mGlu5 dependent activation to support both working memory demand as well as reexposure to cocaine-associated cues. Additionally, my results agree with several preclinical reports that mGlu5 is a viable target to decrease drug seeking (Keck et al., 2014; Knackstedt & Schwendt, 2016; Knackstedt et al., 2014;

Kumaresan et al., 2009). Further understanding how both positive and negative allosteric modulation of mGlu5 can decrease drug seeking may provide useful

124

information for how biased mGlu5 allosteric modulators may be developed to more robustly decrease drug seeking and enhance working memory.

125

LIST OF REFERENCES

Aharonovich, E., Hasin, D., Brooks, A., Liu, X., Bisaga, A., & Nunes, E. (2006). Cognitive deficits predict low treatment retention in cocaine dependent patients. Drug and Alcohol Dependence, 81(3), 313–322. https://doi.org/10.1016/j.drugalcdep.2005.08.003

Akhlaghpour, H., Wiskerke, J., Choi, J., Taliaferro, J., Au, J., & Witten, I. (2016). Dissociated sequential activity and stimulus encoding in the dorsomedial striatum during spatial working memory. eLife, 5(e19507). https://doi.org/10.7554/eLife.19507

Albein-Urios, N., Martinez-González, J., Lozano, O., Clark, L., & Verdejo-García, A. (2012). Comparison of impulsivity and working memory in cocaine addiction and pathological gambling: implications for cocaine-induced neurotoxicity. Drug and Alcohol Dependence, 126(1–2), 1–6. https://doi.org/10.1016/j.drugalcdep.2012.03.008

Allain, F., & Samaha, A. (2018). Revisiting long-access versus short-access cocaine self-administration in rats: Intermittent intake promotes addiction symptoms independent of session length. Addiction Biology. https://doi.org/10.1111/adb.12629

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: American Psychiatric Publishing.

Anderson, J., Bradbury, M., Giracello, D., Chapman, D., Holtz, G., Roppe, J., King, C., Cosford, N., & Varney, M. (2003). In vivo receptor occupancy of mGlu5 receptor antagonists using the novel radioligand [ 3 H]3-methoxy-5-(pyridin-2- ylethynyl)pyridine). European Journal of Pharmacology, 473(1), 35–40. https://doi.org/10.1016/S0014-2999(03)01935-6 Arnsten, A., Paspalas, C., Gamo, N., Yang, Y., & Wang, M. (2010). Dynamic network connectivity: a new form of neuroplasticity. Trends in Cognitive Sciences, 14(8), 365–375. https://doi.org/10.1016/j.tics.2010.05.003

Ashby, F., Ell, S., Valentin, V., & Casale, M. (2005). FROST: a distributed neurocomputational model of working memory maintenance. Journal of Cognitive Neuroscience, 17(11), 1728–1743. https://doi.org/10.1162/089892905774589271 Attucci, S., Carlà, V., Mannaioni, G., & Moroni, F. (2001). Activation of type 5 metabotropic glutamate receptors enhances NMDA responses in mice cortical wedges. British Journal of Pharmacology, 132(4), 799–806. https://doi.org/10.1038/sj.bjp.0703904

Baddeley, A. (1986). Working memory. (No. 11). New York, NY: Clarendon Press/Oxford Press. Retrieved from https://psycnet.apa.org/record/1986-98526-000

126

Baddeley, A., & Hitch, G. (1974). Working Memory. Psychology of Learning and Motivation, 8, 47–89. https://doi.org/10.1016/S0079-7421(08)60452-1

Baker, D., McFarland, K., Lake, R., Shen, H., Tang, X., Toda, S., & Kalivas, P. (2003). Neuroadaptations in cystine-glutamate exchange underlie cocaine relapse. Nature Neuroscience, 6(7), 743–749. https://doi.org/10.1038/nn1069

Bañuelos, C., Beas, B., McQuail, J., Gilbert, R., Frazier, C., Setlow, B., & Bizon, J. (2014). Prefrontal cortical GABAergic dysfunction contributes to age-related working memory impairment. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 34(10), 3457–3466. https://doi.org/10.1523/JNEUROSCI.5192-13.2014

Barbey, A., Koenigs, M., & Grafman, J. (2013). Dorsolateral prefrontal contributions to human working memory. Cortex, 49(5), 1195–1205. https://doi.org/10.1016/J.CORTEX.2012.05.022

Batista, E., Klauss, J., Fregni, F., Nitsche, M., & Nakamura-Palacios, E. (2015). A randomized placebo-controlled trial of targeted prefrontal cortex modulation with bilateral tDCS in patients with crack-cocaine dependence. International Journal of Neuropsychopharmacology, 18(12), pyv066. https://doi.org/10.1093/ijnp/pyv066

Bauer, R., & Fuster, J. (1976). Delayed-matching and delayed-response deficit from cooling dorsolateral prefrontal cortex in monkeys. Journal of Comparative and Physiological Psychology, 90(3), 293–302. https://doi.org/10.1037/h0087996

Beas, B., Setlow, B., & Bizon, J. (2013). Distinct manifestations of executive dysfunction in aged rats. Neurobiology of Aging, 34(9), 2164–2174. https://doi.org/10.1016/j.neurobiolaging.2013.03.019

Bechara, A., & Martin, E. (2004). Impaired decision making related to working memory deficits in individuals with substance addictions. Neuropsychology, 18(1), 152–162. https://doi.org/10.1037/0894-4105.18.1.152

Bechard, A., Bliznyuk, N., & Lewis, M. (2017). The development of repetitive motor behaviors in deer mice: effects of environmental enrichment, repeated testing, and differential mediation by indirect basal ganglia pathway activation. Developmental Psychobiology, 59, 390–399. https://doi.org/10.1002/dev.21503

Bechard, A., LaCrosse, A., Namba, M., Jackson, B., & Knackstedt, L. (2018). Impairments in reversal learning following short access to cocaine self- administration. Drug and Alcohol Dependence, 192, 239–244. https://doi.org/10.1016/J.DRUGALCDEP.2018.08.008

Bell, R., Foxe, J., Nierenberg, J., Hoptman, M., & Garavan, H. (2011). Assessing white matter integrity as a function of abstinence duration in former cocaine-dependent individuals. Drug and Alcohol Dependence, 114(2–3), 159–168. https://doi.org/10.1016/j.drugalcdep.2010.10.001

127

Ben-Shahar, O., Sacramento, A., Miller, B., Webb, S., Wroten, M., Silva, H., … Szumlinski, K. (2013). Deficits in ventromedial prefrontal cortex Group1 metabotropic glutamate receptor function mediate resistance to extinction during protracted withdrawal from an extensive history of cocaine self-administration. J Neurosci, 33(2), 495–506. https://doi.org/10.1523/JNEUROSCI.3710-12.2013

Benn, A., Barker, G., Stuart, S., Roloff, E., Teschemacher, A., Warburton, E., & Robinson, E. (2016). Optogenetic stimulation of prefrontal glutamatergic neurons enhances recognition memory. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 36(18), 4930–4939. https://doi.org/10.1523/JNEUROSCI.2933-15.2016

Benneyworth, M., Hearing, M., Asp, A., Madayag, A., Ingebretson, A., Schmidt, C. (2019). Synaptic depotentiation and mGluR5 activity in the nucleus accumbens drive cocaine-primed reinstatement of place preference. Journal of Neuroscience. https://doi.org/10.1523/JNEUROSCI.3020-17.2019

Benquet, P., Gee, C., & Gerber, U. (2002). Two distinct signaling pathways upregulate NMDA receptor responses via two distinct metabotropic glutamate receptor subtypes. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 22(22), 9679–9686. https://doi.org/10.1523/JNEUROSCI.22-22- 09679.2002

Bilodeau, J., & Schwendt, M. (2016). Post-cocaine changes in regulator of G-protein signaling (RGS) proteins in the dorsal striatum: relevance for cocaine-seeking and protein kinase C-mediated phosphorylation. Synapse, 70(10), 432–440. https://doi.org/10.1002/syn.21917

Bolla, K., Eldreth, D., London, E., Kiehl, K., Mouratidis, M., Contoreggi, C., … Ernst, M. (2003). Orbitofrontal cortex dysfunction in abstinent cocaine abusers performing a decision-making task. NeuroImage, 19(3), 1085–1094. https://doi.org/10.1016/S1053-8119(03)00113-7

Bolla, K., Ernst, M., Kiehl, K., Mouratidis, M., Eldreth, D., Contoreggi, C., … London, E. (2004). Prefrontal cortical dysfunction in abstinent cocaine abusers. Journal of Neuropsychiatry Clinical Neuroscience, 16(4), 456–464. https://doi.org/10.1176/appi.neuropsych.16.4.456

Bourne, J., & Harris, K. (2007). Do thin spines learn to be mushroom spines that remember? Current Opinion in Neurobiology, 17(3), 381–386. https://doi.org/10.1016/J.CONB.2007.04.009 Brady, K., Johnson, R., Gray, K., & Tolliver, B. (2011). Cognitive enhancers in the treatment of substance use disorders: clinical evidence. Pharmacology, Biochemistry, and Behavior, 99(2), 285–294. https://doi.org/10.1016/j.pbb.2011.04.017

128

Brewer, J., Worhunsky, P., Carroll, K., Rounsaville, B., & Potenza, M. (2008). Pre- treatment brain activation during stroop task is associated with outcomes in cocaine dependent patients. Biological Psychiatry, 64(11), 998–1004. https://doi.org/10.1016/j.biopsych.2008.05.024

Briand, L., Flagel, S., Garcia-Fuster, M., Watson, S., Akil, H., Sarter, M., & Robinson, T. (2008). Persistent alterations in cognitive function and prefrontal dopamine D2 receptors following extended, but not limited, access to self-administered cocaine. Neuropsychopharmacology, 33(12), 2969–2980. https://doi.org/10.1038/npp.2008.18 Brog, J., Salyapongse, A., Deutch, A., & Zahm, D. (1993). The patterns of afferent innervation of the core and shell in the accumbens part of the rat ventral striatum: Immunohistochemical detection of retrogradely transported fluoro-gold. The Journal of Comparative Neurology, 338(2), 255–278. https://doi.org/10.1002/cne.903380209

Cabeza, R., & Nyberg, L. (2000). Imaging cognition II: an empirical review of 275 PET and fMRI studies. Journal of Cognitive Neuroscience, 12(1), 1–47. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10769304

Calu, D., Stalnaker, T., Franz, T., Singh, T., Shaham, Y., & Schoenbaum, G. (2007). Withdrawal from cocaine self-administration produces long-lasting deficits in orbitofrontal-dependent reversal learning in rats. Learning & Memory, 14(5), 325– 328. https://doi.org/10.1101/lm.534807

Camchong, J., MacDonald, A., Nelson, B., Bell, C., Mueller, B., Specker, S., & Lim, K. (2011). Frontal hyperconnectivity related to discounting and reversal learning in cocaine subjects. Biological Psychiatry, 69(11), 1117–1123. https://doi.org/10.1016/j.biopsych.2011.01.008

Campanella, S., Peigneux, P., Petit, G., Lallemand, F., Saeremans, M., Noël, X., … Verbanck, P. (2013). Increased cortical activity in binge drinkers during working memory task: A preliminary assessment through a functional magnetic resonance imaging study. PLoS ONE, 8(4), e62260. https://doi.org/10.1371/journal.pone.0062260

Capriles, N., Rodaros, D., Sorge, R., & Stewart, J. (2003). A role for the prefrontal cortex in stress and cocaine-induced reinstatement of cocaine seeking in rats. Psychopharmacology, 168(1–2), 66–74. https://doi.org/10.1007/s00213-002-1283-z

Caprioli, D., Venniro, M., Zhang, M., Bossert, J., Warren, B., Hope, B., & Shaham, Y. (2017). Role of dorsomedial striatum neuronal ensembles in incubation of methamphetamine craving after voluntary abstinence. The Journal of Neuroscience, 37(4). https://doi.org/10.1523/JNEUROSCI.3091-16.2017

129

Carelli, R., Williams, J., & Hollander, J. (2003). Basolateral amygdala neurons encode cocaine self-administration and cocaine-associated cues. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 23(23), 8204– 8211. https://doi.org/10.1523/JNEUROSCI.23-23-08204.2003

Carroll, M., Morgan, A., Lynch, W., Campbell, U., & Dess, N. (2002). Intravenous cocaine and heroin self-administration in rats selectively bred for differential saccharin intake: phenotype and sex differences. Psychopharmacology, 161(3), 304–313. https://doi.org/10.1007/s00213-002-1030-5

Chafee, M., & Goldman-Rakic, P. (1998). Matching patterns of activity in primate prefrontal area 8a and parietal area 7ip neurons during a spatial working memory task. Journal of Neurophysiology, 79(6), 2919–2940. https://doi.org/10.1152/jn.1998.79.6.2919

Chen, B., Yau, H., Hatch, C., Kusumoto-Yoshida, I., Cho, S., Hopf, F., & Bonci, A. (2013). Rescuing cocaine-induced prefrontal cortex hypoactivity prevents compulsive cocaine seeking. Nature, 496(7445), 359–362. https://doi.org/10.1038/nature12024

Chen, H., Liao, P., & Chan, M. (2011). mGluR5 positive modulators both potentiate activation and restore inhibition in NMDA receptors by PKC dependent pathway. Journal of Biomedical Science, 18(1), 19. https://doi.org/10.1186/1423-0127-18-19

Chiamulera, C., Epping-Jordan, M., Zocchi, A., Marcon, C., Cottiny, C., Tacconi, S., … Conquet, F. (2001). Reinforcing and locomotor stimulant effects of cocaine are absent in mGluR5 null mutant mice. Nature Neuroscience, 4(9), 873–874. https://doi.org/10.1038/nn0901-873

Cleva, R., Gass, J., Widholm, J., & Olive, M. (2010). Glutamatergic targets for enhancing extinction learning in drug addiction. Current Neuropharmacology, 8(4), 394–408. https://doi.org/10.2174/157015910793358169

Cleva, R., Hicks, M., Gass, J., Wischerath, K., Plasters, E., Widholm, J., & Olive, M. (2011). mGluR5 positive allosteric modulation enhances extinction learning following cocaine self-administration. Behavioral Neuroscience, 125(1), 10–19. https://doi.org/10.1037/a0022339

Cleva, R., & Olive, M. (2011). Positive allosteric modulators of type 5 metabotropic glutamate receptors (mGluR5) and their therapeutic potential for the treatment of CNS disorders. Molecules (Basel, Switzerland), 16(3), 2097–2106. https://doi.org/10.3390/molecules16032097 Cleva, R., & Olive, M. (2012). mGlu receptors and drug addiction. Wiley Interdisciplinary Reviews. Membrane Transport and Signaling, 1(3), 281–295. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/22662312

130

Connolly, C., Bell, R., Foxe, J., & Garavan, H. (2013). Dissociated grey matter changes with prolonged addiction and extended abstinence in cocaine users. PloS One, 8(3), e59645. https://doi.org/10.1371/journal.pone.0059645

Connolly, C., Foxe, J., Nierenberg, J., Shpaner, M., & Garavan, H. (2012). The neurobiology of cognitive control in successful cocaine abstinence. Drug and Alcohol Dependence, 121(1–2), 45–53. https://doi.org/10.1016/J.DRUGALCDEP.2011.08.007

Conrad, K., Tseng, K., Uejima, J., Reimers, J., Heng, L., Shaham, Y., … Wolf, M. (2008). Formation of accumbens GluR2-lacking AMPA receptors mediates incubation of cocaine craving. Nature, 454(7200), 118–121. https://doi.org/10.1038/nature06995

Contia, C., Moscona, J., Miyuki, E., & Palaciosa, N. (2016). Dorsolateral prefrontal cortex activity and neuromodulation in crack-cocaine dependents during early abstinence. Journal of Neurology & Neurophysiology, 7(3). https://doi.org/10.4172/2155-9562.1000374

Cornish, J., & Kalivas, P. (2000). Glutamate transmission in the nucleus accumbens mediates relapse in cocaine addiction. The Journal of Neuroscience, 20.

Cowan, N. (1997). Attention and memory: an integrated framework (1st ed.). New York, NY: Oxford University Press. Retrieved from http://www.ovid.com/site/catalog/books/pdfs/3186.pdf

Cowan, N. (2008). What are the differences between long-term, short-term, and working memory? Progress in Brain Research, 169, 323–338. https://doi.org/10.1016/S0079-6123(07)00020-9

Cunha, P., Gonçalves, P., Ometto, M., dos Santos, B., Nicastri, S., Busatto, G., & de Andrade, A. (2013). Executive cognitive dysfunction and ADHD in cocaine dependence: searching for a common cognitive endophenotype for addictive disorders. Frontiers in Psychiatry, 4, 126. https://doi.org/10.3389/fpsyt.2013.00126

Curtis, C., & D’Esposito, M. (2003). Persistent activity in the prefrontal cortex during working memory. Trends in Cognitive Sciences, 7(9), 415–423. https://doi.org/10.1016/S1364-6613(03)00197-9

Dackis, C., Kampman, K., Lynch, K., Pettinati, H., & O’Brien, C. (2005). A double-blind, placebo-controlled trial of modafinil for cocaine dependence. Neuropsychopharmacology, 30(1), 205–211. https://doi.org/10.1038/sj.npp.1300600

Daniel, T., Katz, J., & Robinson, J. (2016). Delayed match-to-sample in working memory: A BrainMap meta-analysis. Biological Psychology, 120, 10–20. https://doi.org/10.1016/j.biopsycho.2016.07.015

131

Dauvermann, M., Lee, G., & Dawson, N. (2017). Glutamatergic regulation of cognition and functional brain connectivity: insights from pharmacological, genetic and translational schizophrenia research. British Journal of Pharmacology, 174, 3136– 3160. https://doi.org/10.1111/bph.v174.19/issuetoc de Laat, B., Weerasekera, A., Leurquin-Sterk, G., Bormans, G., Himmelreich, U., Casteels, C., & Van Laere, K. (2018). Glutamatergic biomarkers for cocaine addiction: a longitudinal study using MR spectroscopy and mGluR5 PET in self- administering rats. Journal of Nuclear Medicine, 59(6), 952–959. https://doi.org/10.2967/jnumed.117.202614

Di Ciano, P., & Everitt, B. (2001). Dissociable effects of antagonism of NMDA and AMPA/KA receptors in the nucleus accumbens core and shell on cocaine-seeking behavior. Neuropsychopharmacology, 25(3), 341–360. https://doi.org/10.1016/S0893-133X(01)00235-4

Di Pietro, N., Black, Y., & Kantak, K. (2006). Context-dependent prefrontal cortex regulation of cocaine self-administration and reinstatement behaviors in rats. European Journal of Neuroscience, 24(11), 3285–3298. https://doi.org/10.1111/j.1460-9568.2006.05193.x

Driesen, N., McCarthy, G., Bhagwagar, Z., Bloch, M., Calhoun, V., D’Souza, D., … Krystal, J. (2013). The impact of NMDA receptor blockade on human working memory-related prefrontal function and connectivity. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 38(13), 2613–2622. https://doi.org/10.1038/npp.2013.170

Dunnett, S. (1993). Operant delayed matching and non-matching to position in rats. Behavioural Neuroscience : A Practical Approach, 123–136. Retrieved from https://ci.nii.ac.jp/naid/20000724427/

Ehlers, M. (2000). Reinsertion or degradation of AMPA receptors determined by activity- dependent endocytic sorting. Neuron, 28(2), 511–525. https://doi.org/10.1016/S0896-6273(00)00129-X

El Moustaine, D., Granier, S., Doumazane, E., Scholler, P., Rahmeh, R., Bron, P., … Pin, J. (2012). Distinct roles of metabotropic glutamate receptor dimerization in agonist activation and G-protein coupling. Proceedings of the National Academy of Sciences of the United States of America, 109(40), 16342–16347. https://doi.org/10.1073/pnas.1205838109

Eriksson, J., Vogel, E., Lansner, A., Bergström, F., & Nyberg, L. (2015). Neurocognitive architecture of working memory. Neuron, 88(1), 33–46. https://doi.org/10.1016/j.neuron.2015.09.020

132

Ersche, K., Barnes, A., Jones, P., Morein-Zamir, S., Robbins, T., & Bullmore, E. (2011). Abnormal structure of frontostriatal brain systems is associated with aspects of impulsivity and compulsivity in cocaine dependence. Brain, 134(7), 2013–2024. https://doi.org/10.1093/brain/awr138

Fagni, L., Worley, P., & Ango, F. (2002). Homer as both a scaffold and transduction molecule. Sci. STKE, 2002(137), re8-re8. https://doi.org/10.1126/STKE.2002.137.RE8 Farovik, A., Dupont, L., Arce, M., & Eichenbaum, H. (2008). Medial prefrontal cortex supports recollection, but not familiarity, in the rat. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 28(50), 13428–13434. https://doi.org/10.1523/JNEUROSCI.3662-08.2008

Fein, G., Di Sclafani, V., & Meyerhoff, D. (2002). Prefrontal cortical volume reduction associated with frontal cortex function deficit in 6-week abstinent crack-cocaine dependent men. Drug Alcohol Depend, 68(1), 87–93. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2857690/pdf/nihms186759.pdf

Ferrario, C., Gorny, G., Crombag, H., Li, Y., Kolb, B., & Robinson, T. (2005). Neural and behavioral plasticity associated with the transition from controlled to escalated cocaine use. Biological Psychiatry, 58(9), 751–759. https://doi.org/10.1016/j.biopsych.2005.04.046

Fijał, K., Nowak, E., Leśkiewicz, M., Budziszewska, B., & Filip, M. (2015). Working memory deficits and alterations of ERK and CREB phosphorylation following withdrawal from cocaine self-administration. Pharmacological Reports, 67, 1–9. https://doi.org/10.1016/j.pharep.2015.01.013

Fletcher, P., & Henson, R. (2001). Frontal lobes and human memory: insights from functional neuroimaging. Brain : A Journal of Neurology, 124(Pt 5), 849–881. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/11335690

Fowler, S., Walker, J., Klakotskaia, D., Will, M., Serfozo, P., Simonyi, A., & Schachtman, T. (2013). Effects of a metabotropic glutamate receptor 5 positive allosteric modulator, CDPPB, on spatial learning task performance in rodents. Neurobiology of Learning and Memory, 99, 25–31. https://doi.org/10.1016/j.nlm.2012.10.010

Franklin, T., Acton, P., Maldjian, J., Gray, J., Croft, J., Dackis, C., … Childress, A. (2002). Decreased gray matter concentration in the insular, orbitofrontal, cingulate, and temporal cortices of cocaine patients. Biological Psychiatry, 51(2), 134–142. Retrieved from https://ac.els-cdn.com/S0006322301012690/1-s2.0- S0006322301012690-main.pdf?_tid=8500c51d-af83-4d09-a82b- 3e9574470811&acdnat=1538694973_0538a5a600cdc71afe91acfe90c9a7cb

133

Freeman, W., Patel, K., Brucklacher, R., Lull, M., Erwin, M., Morgan, D., … Vrana, K. (2008). Persistent alterations in mesolimbic gene expression with abstinence from cocaine self-administration. Neuropsychopharmacology, 33(8), 1807–1817. https://doi.org/10.1038/sj.npp.1301577

Fregni, F., Boggio, P., Nitsche, M., Bermpohl, F., Antal, A., Feredoes, E., … Pascual- Leone, A. (2005). Anodal transcranial direct current stimulation of prefrontal cortex enhances working memory. Experimental Brain Research, 166(1), 23–30. https://doi.org/10.1007/s00221-005-2334-6

Fuchs, R., Evans, K., Ledford, C., Parker, M., Case, J., Mehta, R., & See, R. (2005). The role of the dorsomedial prefrontal cortex, basolateral amygdala, and dorsal hippocampus in contextual reinstatement of cocaine seeking in rats. Neuropsychopharmacology, 30(2), 296–309. https://doi.org/10.1038/sj.npp.1300579

Funahashi, S., Bruce, C., & Goldman-Rakic, P. (1993). Dorsolateral prefrontal lesions and oculomotor delayed-response performance: evidence for mnemonic scotomas. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 13(4), 1479–1497. https://doi.org/10.1523/JNEUROSCI.13-04-01479.1993

Fuster, J. (1973). Unit activity in prefrontal cortex during delayed-response performance: neuronal correlates of transient memory. Journal of Neurophysiology, 36(1), 61–78. https://doi.org/10.1152/jn.1973.36.1.61 Ganella, D., Thangaraju, P., Lawrence, A., & Kim, J. (2016). Fear extinction in 17 day old rats is dependent on metabotropic glutamate receptor 5 signaling. Behavioural Brain Research, 298, 32–36. https://doi.org/10.1016/J.BBR.2014.12.010

Garavan, H., Pankiewicz, J., Bloom, A., Cho, J., Sperry, L., Ross, T., … Stein, E. (2000). Cue-induced cocaine craving: neuroanatomical specificity for drug users and drug stimuli. American Journal of Psychiatry, 157(11), 1789–1798. https://doi.org/10.1176/appi.ajp.157.11.1789

Garza-Villarreal, E., Chakravarty, M., Hansen, B., Eskildsen, S., Devenyi, G., Castillo- Padilla, D., … Gonzalez-Olvera, J. (2017). The effect of crack cocaine addiction and age on the microstructure and morphology of the human striatum and thalamus using shape analysis and fast diffusion kurtosis imaging. Translational Psychiatry, 7(5), e1122. https://doi.org/10.1038/tp.2017.92

Gass, J., & Chandler, L. (2013). The plasticity of extinction: Contribution of the prefrontal cortex in treating addiction through inhibitory learning. Frontiers in Psychiatry, 4, 46. https://doi.org/10.3389/fpsyt.2013.00046

Gass, J., & Olive, M. (2009). Positive allosteric modulation of mGluR5 receptors facilitates extinction of a cocaine contextual memory. Biological Psychiatry, 65(8), 717–720. https://doi.org/10.1016/j.biopsych.2008.11.001

134

Gass, J., Osborne, M., Watson, N., Brown, J., & Olive, M. (2009). mGluR5 antagonism attenuates methamphetamine reinforcement and prevents reinstatement of methamphetamine-seeking behavior in rats. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 34(4), 820– 833. https://doi.org/10.1038/npp.2008.140

Gawin, F., & Kleber, H. (1986). Abstinence symptomatology and psychiatric diagnosis in cocaine abusers. Archives of General Psychiatry, 43(2), 107–113. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/3947206

George, O., Mandyam, C., Wee, S., & Koob, G. (2008). Extended access to cocaine self-administration produces long-lasting prefrontal cortex-dependent working memory impairments. Neuropsychopharmacology, 33(10), 2474–2482. https://doi.org/10.1038/sj.npp.1301626

Gerber, U., Gee, C., & Benquet, P. (2007). Metabotropic glutamate receptors: intracellular signaling pathways. Current Opinion in Pharmacology, 7(1), 56–61. https://doi.org/10.1016/J.COPH.2006.08.008

Ghasemzadeh, M., Vasudevan, P., Giles, C., Purgianto, A., Seubert, C., & Mantsch, J. R. (2011). Glutamatergic plasticity in medial prefrontal cortex and ventral tegmental area following extended-access cocaine self-administration. Brain Research, 1413, 60–71. https://doi.org/10.1016/j.brainres.2011.06.041

Gipson, C., Kupchik, Y., Shen, H., Reissner, K., Thomas, C., & Kalivas, P. (2013). Relapse induced by cues predicting cocaine depends on rapid, transient synaptic potentiation. Neuron, 77(5), 867–872. https://doi.org/10.1016/j.neuron.2013.01.005

Gladding, C., Fitzjohn, S., & Molnár, E. (2009). Metabotropic glutamate receptor- mediated long-term depression: molecular mechanisms. Pharmacological Reviews, 61(4), 395–412. https://doi.org/10.1124/pr.109.001735

Gobin, C., & Schwendt, M. (2017). The effects of extended-access cocaine self- administration on working memory performance, reversal Learning and incubation of cocaine-seeking in adult male rats. Journal of Addiction & Prevention, 5(1), 1–6. https://doi.org/10.13188/2330-2178.1000035

Gobin, C., Shallcross, J., & Schwendt, M. (2019). Neurobiological substrates of persistent working memory deficits and cocaine-seeking in the prelimbic cortex of rats with a history of extended access to cocaine self-administration. Neurobiology of Learning and Memory. https://doi.org/10.1016/J.NLM.2019.03.007

Goldman-Rakic, P. (1995). Cellular basis of working memory. Neuron, 14(3), 477–485. https://doi.org/10.1016/0896-6273(95)90304-6

135

Goldstein, R., Leskovjan, A., Hoff, A., Hitzemann, R., Bashan, F., Khalsa, S., … Volkow, N. (2004). Severity of neuropsychological impairment in cocaine and alcohol addiction: association with metabolism in the prefrontal cortex. Neuropsychologia, 42(11), 1447–1458. https://doi.org/10.1016/J.NEUROPSYCHOLOGIA.2004.04.002

Goldstein, R., & Volkow, N. (2002). Drug addiction and its underlying neurobiological basis: Neuroimaging evidence for the involvement of the frontal cortex. The American Journal of Psychiatry, 159(10), 1642–1652. https://doi.org/10.1176/appi.ajp.159.10.1642

Goldstein, R., & Volkow, N. (2011). Dysfunction of the prefrontal cortex in addiction: neuroimaging findings and clinical implications. Nature Reviews Neuroscience, 12(11), 652–669. https://doi.org/10.1038/nrn3119

Gonzalez-Lima, F., & Cada, A. (1994). Cytochrome oxidase activity in the auditory system of the mouse: a qualitative and quantitative histochemical study. Neuroscience, 63(2), 559–578. https://doi.org/10.1016/0306-4522(94)90550-9

Gonzalez-Lima, F., & Jones, D. (1994). Quantitative mapping of cytochrome oxidase activity in the central auditory system of the gerbil: a study with calibrated activity standards and metal,intensified histochemistry. Brain Research, 660, 34–49. Retrieved from https://ac.els-cdn.com/0006899394908362/1-s2.0- 0006899394908362-main.pdf?_tid=41ec0ecc-7ed0-4c61-9bc3- 3d25f0ac8682&acdnat=1536972991_8cf9349a118a6c9b39ed0a5708ad06d6

Granon, S., & Poucet, B. (2000). Involvement of the rat prefrontal cortex in cognitive functions: a central role for the prelimbic area. Psychobiology, 28(2), 229–237. https://doi.org/10.3758/bf03331981

Grant, S., London, E., Newlin, D., Villemagne, V., Liu, X., Contoreggi, C., … Margolin, A. (1996). Activation of memory circuits during cue-elicited cocaine craving. Proceedings of the National Academy of Sciences of the United States of America, 93(21), 12040–12045. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/8876259

Grimm, J., Hope, B., Wise, R. & Shaham, Y. (2001). Incubation of cocaine craving after withdrawal. Nature, 412(6843), 141–142. https://doi.org/10.1038/35084134

Hains, A., Vu, M., Maciejewski, P., Van Dyck, C., Gottron, M., Arnsten, A. (2009). Inhibition of protein kinase C signaling protects prefrontal cortex dendritic spines and cognition from the effects of chronic stress. PNAS, 106(42), 17957–17962. Retrieved from http://www.pnas.org/content/106/42/17957.full.pdf

Halpern, B., & Itzkovitz, S. (2016). Single molecule approaches for quantifying transcription and degradation rates in intact mammalian tissues. Methods, 98, 134– 142. https://doi.org/10.1016/J.YMETH.2015.11.015

136

Hanlon, C., Dufault, D., Wesley, M., & Porrino, L. (2011). Elevated gray and white matter densities in cocaine abstainers compared to current users. Psychopharmacology, 218(4), 681–692. https://doi.org/10.1007/s00213-011-2360-y

Hao, Y., Martin-Fardon, R., & Weiss, F. (2010). Behavioral and functional evidence of metabotropic glutamate receptor 2/3 and metabotropic glutamate receptor 5 dysregulation in cocaine-escalated rats: factor in the transition to dependence. Biological Psychiatry, 68(3), 240–248. https://doi.org/10.1016/j.biopsych.2010.02.011

Hearing, M., Miller, S., See, R., & McGinty, J. (2008). Relapse to cocaine seeking increases activity-regulated gene expression differentially in the prefrontal cortex of abstinent rats. Psychopharmacology, 198(1), 77–91. https://doi.org/10.1007/s00213-008-1090-2

Hermans, E., & Challiss, R. (2001). Structural, signalling and regulatory properties of the group I metabotropic glutamate receptors: prototypic family C G-protein- coupled receptors. The Biochemical Journal, 359(Pt 3), 465–484. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/11672421

Hernandez, C., McQuail, J., Schwabe, M., Burke, S., Setlow, B., & Bizon, J. (2018). Age-related declines in prefrontal cortical expression of metabotropic glutamate receptors that support working memory. Eneuro, 5(3), ENEURO.0164-18.2018. https://doi.org/10.1523/ENEURO.0164-18.2018

Hester, R., & Garavan, H. (2004). Executive dysfunction in cocaine addiction: Evidence for discordant frontal, cingulate, and cerebellar activity. Journal of Neuroscience, 24(49), 11017-11022. https://doi.org/10.1523/JNEUROSCI.3321-04.2004

Hoff, A., Riordan, H., Morris, L., Cestaro, V., Wieneke, M., Alpert, R., … Volkow, N. (1996). Effects of crack cocaine on neurocognitive function. Psychiatry Research, 60(2–3), 167–176. https://doi.org/10.1016/0165-1781(96)02758-8

Homayoun, H., & Moghaddam, B. (2006). Bursting of prefrontal cortex neurons in awake rats is regulated by metabotropic glutamate 5 (mGlu5) receptors: rate- dependent influence and interaction with NMDA receptors. Cerebral Cortex, 16(1), 93–105. https://doi.org/10.1093/cercor/bhi087

Homayoun, H., & Moghaddam, B. (2010). Group 5 metabotropic glutamate receptors: role in modulating cortical activity and relevance to cognition. European Journal of Pharmacology, 639(1–3), 33–39. https://doi.org/10.1016/j.ejphar.2009.12.042

Homayoun, H., Stefani, M., Adams, B., Tamagan, G., & Moghaddam, B. (2004). Functional interaction between NMDA and mGlu5 receptors: effects on working memory, instrumental learning, motor behaviors, and dopamine release. Neuropsychopharmacology, 29(7), 1259–1269. https://doi.org/10.1038/sj.npp.1300417

137

Hovelsø, N., Sotty, F., Montezinho, L., Pinheiro, P., Herrik, K., & Mørk, A. (2012). Therapeutic potential of metabotropic glutamate receptor modulators. Current Neuropharmacology, 10(1), 12–48. https://doi.org/10.2174/157015912799362805

Huang, Y., Lu, W., Ali, D., Pelkey, K., Pitcher, G., Lu, Y., … MacDonald, J. (2001). CAKβ/Pyk2 kinase is a signaling link for induction of long-term potentiation in CA1 hippocampus. Neuron, 29(2), 485–496. https://doi.org/10.1016/S0896- 6273(01)00220-3

Jacobsen, C., & Nissen, H. (1937). Studies of cerebral function in primates. IV. The effects of frontal lobe lesions on the delayed alternation habit in monkeys. Journal of Comparative Psychology, 23(1), 101–112. https://doi.org/10.1037/h0056632 James, M., Mcglinchey, E., Vattikonda, A., Mahler, S., & Aston-Jones, G. (2018). Cued reinstatement of cocaine but not sucrose seeking is dependent on dopamine signaling in prelimbic cortex and is associated with recruitment of prelimbic neurons that project to contralateral nucleus accumbens core. International Journal of Neuropsychopharmacology, 21(1), 89–94. https://doi.org/10.1093/ijnp/pyx107

Jasinska, A., Chen, B., Bonci, A., & Stein, E. (2015). Dorsal medial prefrontal cortex (mPFC) circuitry in rodent models of cocaine use: implications for drug addiction therapies. Addiction Biology, 20(2), 215–226. https://doi.org/10.1111/adb.12132

Jasinska, A., Stein, E., Kaiser, J., Naumer, M., & Yalachkov, Y. (2014). Factors modulating neural reactivity to drug cues in addiction: a survey of human neuroimaging studies. Neuroscience and Biobehavioral Reviews, 38, 1–16. https://doi.org/10.1016/j.neubiorev.2013.10.013

Jew, C., Wu, C., Sun, H., Zhu, J., & Huang, J. (2013). mGluR5 ablation in cortical glutamatergic neurons increases novelty-induced locomotion. PLoS ONE, 8(8), 70415. https://doi.org/10.1371/journal.pone.0070415

Jha, A., & McCarthy, G. (2000). The influence of memory load upon delay-interval activity in a working memory task: an event-related functional MRI study. Journal of Cognitive Neuroscience, 12(2), 90–105. https://doi.org/10.1162/089892900564091

Jin, D., Guo, M., Xue, B., Mao, L., & Wang, J. (2013). Differential regulation of CaMKIIα interactions with mGluR5 and NMDA receptors by Ca(2+) in neurons. Journal of Neurochemistry, 127(5), 620–631. https://doi.org/10.1111/jnc.12434 Jones, S., & Bonci, A. (2005). Synaptic plasticity and drug addiction. Current Opinion in Pharmacology, 5(1), 20–25. https://doi.org/10.1016/J.COPH.2004.08.011

Jong, Y., Sergin, I., Purgert, C., & O’malley, K. (2014). Location-dependent signaling of the group 1 metabotropic glutamate receptor mGlu5. Mol Pharmacol, 86, 774–785. https://doi.org/10.1124/mol.114.094763

138

Jovanovski, D., Erb, S., & Zakzanis, K. (2005). Neurocognitive deficits in cocaine users: a quantitative review of the evidence. Journal of Clinical and Experimental Neuropsychology, 27(2), 189–204. https://doi.org/10.1080/13803390490515694

Jung, M., Qin, Y., Lee, D., & Mook-Jung, I. (2000). Relationship among discharges of neighboring neurons in the rat prefrontal cortex during spatial working memory tasks. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 20(16), 6166–6172. https://doi.org/10.1523/JNEUROSCI.20-16- 06166.2000

Jung, M., Qin, Y., McNaughton, B., & Barnes, C. (1998). Firing characteristics of deep layer neurons in prefrontal cortex in rats performing spatial working memory tasks. Cerebral Cortex, 8(5), 437–450. https://doi.org/10.1093/cercor/8.5.437

Just, M., & Carpenter, P. (1992). A capacity theory of comprehension: Individual differences in working memory. Psychological Review, 99(1), 122–149. https://doi.org/10.1037/0033-295X.99.1.122

Kalivas, P., & Hu, X. (2006). Exciting inhibition in psychostimulant addiction. Trends in Neurosciences, 29(11), 610–616. https://doi.org/10.1016/j.tins.2006.08.008

Kalivas, P., & O’Brien, C. (2008). Drug addiction as a pathology of staged neuroplasticity. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 33(1), 166–180. https://doi.org/10.1038/sj.npp.1301564

Kantak, K., Barlow, N., Tassin, D., Brisotti, M., & Jordan, C. (2014). Performance on a strategy set shifting task in rats following adult or adolescent cocaine exposure. Psychopharmacology, 231(23), 4489–4501. https://doi.org/10.1007/s00213-014- 3598-y

Kaufman, J., Ross, T., Stein, E., & Garavan, H. (2003). Cingulate hypoactivity in cocaine users during a GO-NOGO task as revealed by event-related functional magnetic resonance imaging. Journal of Neuroscience, 23(21), 7839–7843. https://doi.org/23/21/7839 [pii]

Kawa, A., Bentzley, B., & Robinson, T. (2016). Less is more: prolonged intermittent access cocaine self-administration produces incentive-sensitization and addiction- like behavior. Psychopharmacology, 233(19–20), 3587–3602. https://doi.org/10.1007/s00213-016-4393-8

Keck, T., Yang, H., Bi, G., Huang, Y., Zhang, H., Srivastava, R., … Xi, Z.. (2013). Fenobam sulfate inhibits cocaine-taking and cocaine-seeking behavior in rats: implications for addiction treatment in humans. Psychopharmacology, 229(2), 253– 265. https://doi.org/10.1007/s00213-013-3106-9

139

Keck, T., Zou, M., Bi, G., Zhang, H., Wang, X., Yang, H., … Newman, A. (2014). A novel mGluR5 antagonist, MFZ 10-7, inhibits cocaine-taking and cocaine-seeking behavior in rats. Addiction Biology, 19(2), 195–209. https://doi.org/10.1111/adb.12086

Kelley, A. (2004). Memory and addiction: shared neural circuitry and molecular mechanisms. Neuron, 44(1), 161–179. https://doi.org/10.1016/j.neuron.2004.09.016

Kelley, B., Yeager, K., Pepper, T., & Beversdorf, D. (2005). Cognitive impairment in acute cocaine withdrawal. Cognitive and Behavioral Neurology, 18(2), 108–112. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/15970730

Kenny, P., Boutrel, B., Gasparini, F., Koob, G., & Markou, A. (2005). Metabotropic glutamate 5 receptor blockade may attenuate cocaine self-administration by decreasing brain reward function in rats. Psychopharmacology, 179(1), 247–254. https://doi.org/10.1007/s00213-004-2069-2

Kim, J., Perry, C., Luikinga, S., Zbukvic, I., Brown, R., & Lawrence, A. (2014). Extinction of a cocaine-taking context that protects against drug-primed reinstatement is dependent on the metabotropic glutamate 5 receptor. Addiction Biology, 20(3), 482–489. https://doi.org/10.1111/adb.12142

Kiritoshi, T., Sun, H., Ren, W., Stauffer, S., Lindsley, C., Conn, P., & Neugebauer, V. (2013). Modulation of pyramidal cell output in the medial prefrontal cortex by mGluR5 interacting with CB1. Neuropharmacology, 66, 170–178. https://doi.org/10.1016/j.neuropharm.2012.03.024

Klauss, J., Anders, Q., Felippe, L., Ferreira, L., Cruz, M., Nitsche, M., & Nakamura- Palacios, E. (2018). Lack of effects of extended sessions of transcranial direct current stimulation (tDCS) over dorsolateral prefrontal cortex on craving and relapses in crack-cocaine users. Frontiers in Pharmacology, 9, 1198. https://doi.org/10.3389/fphar.2018.01198

Knackstedt, L., & Schwendt, M. (2016). mGlu5 receptors and relapse to cocaine- seeking: the role of receptor trafficking in postrelapse extinction learning deficits. Neural Plasticity, 2016, 9312508. https://doi.org/10.1155/2016/9312508

Knackstedt, L., Trantham-Davidson, H., & Schwendt, M. (2014). The role of ventral and dorsal striatum mGluR5 in relapse to cocaine-seeking and extinction learning. Addiction Biology, 19(1), 87–101. https://doi.org/10.1111/ADB.12061

Knackstedt, L., & Kalivas, P. (2009). Glutamate and reinstatement. Current Opinion in Pharmacology, 9(1), 59–64. https://doi.org/10.1016/J.COPH.2008.12.003

140

Kober, H., Lacadie, C., Wexler, B., Malison, R., Sinha, R., & Potenza, M. (2016). Brain activity during cocaine craving and gambling urges: An fMRI study. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 41(2), 628–637. https://doi.org/10.1038/npp.2015.193

Koob, G., & Volkow, N. (2010). Neurocircuitry of addiction. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 35(1), 217–238. https://doi.org/10.1038/npp.2009.110

Koob, G., & Volkow, N. (2016). Neurobiology of addiction: a neurocircuitry analysis. The Lancet. Psychiatry, 3(8), 760–773. https://doi.org/10.1016/S2215-0366(16)00104-8

Kosten, T., Tucker, K., Gottschalk, P., Rinder, C., & Rinder, H. (2004). Platelet abnormalities associated with cerebral perfusion defects in cocaine dependence. Biological Psychiatry, 55(1), 91–97. https://doi.org/10.1016/S0006-3223(03)00706- 6

Kourrich, S., Rothwell, P., Klug, J., & Thomas, M. (2007). Cocaine experience controls bidirectional synaptic plasticity in the nucleus accumbens. Journal of Neuroscience, 27(30), 7921–7928. https://doi.org/10.1523/JNEUROSCI.1859-07.2007

Kübler, A., Murphy, K., & Garavan, H. (2005). Cocaine dependence and attention switching within and between verbal and visuospatial working memory. European Journal of Neuroscience, 21(7), 1984–1992. https://doi.org/10.1111/j.1460- 9568.2005.04027.x

Kubota, K., & Niki, H. (1971). Prefrontal cortical unit activity and delayed alternation performance in monkeys. Journal of Neurophysiology, 34(3), 337–347. https://doi.org/10.1152/jn.1971.34.3.337

Kumaresan, V., Yuan, M., Yee, J., Famous, K., Anderson, S., Schmidt, H., & Pierce, R. (2009). Metabotropic glutamate receptor 5 (mGluR5) antagonists attenuate cocaine priming- and cue-induced reinstatement of cocaine seeking. Behavioural Brain Research, 202(2), 238–244. https://doi.org/10.1016/J.BBR.2009.03.039

LaCrosse, A., Burrows, B., Angulo, R., Conrad, P., Himes, S., Mathews, N., … Olive, M. (2014). mGluR5 positive allosteric modulation and its effects on MK-801 induced set-shifting impairments in a rat operant delayed matching/non-matching-to-sample task. Psychopharmacology, 232(1), 251–258. https://doi.org/10.1007/s00213-014- 3653-8

Laroche, S., Davis, S., & Jay, T. (2000). Plasticity at hippocampal to prefrontal cortex synapses: dual roles in working memory and consolidation. Hippocampus, 10(4), 438–446. https://doi.org/10.1002/1098-1063(2000)10:4<438::AID- HIPO10>3.0.CO;2-3

141

Lecourtier, L., Homayoun, H., Tamagnan, G., & Moghaddam, B. (2007). Positive allosteric modulation of metabotropic glutamate 5 (mGlu5) receptors reverses n- methyl-d-aspartate antagonist-induced alteration of neuronal firing in prefrontal cortex. Biological Psychiatry, 62(7), 739–746. https://doi.org/10.1016/j.biopsych.2006.12.003

Lee, B., Platt, D., Rowlett, J., Adewale, A., & Spealman, R. (2005). Attenuation of behavioral effects of cocaine by the metabotropic glutamate receptor 5 antagonist 2-methyl-6-(phenylethynyl)-pyridine in squirrel monkeys: comparison with dizocilpine. The Journal of Pharmacology and Experimental Therapeutics, 312(3), 1232–1240. https://doi.org/10.1124/jpet.104.078733

Leung, H., Gore, J., & Goldman-Rakic, P. (2002). Sustained mnemonic response in the human middle frontal gyrus during on-line storage of spatial memoranda. Journal of Cognitive Neuroscience, 14(4), 659–671. https://doi.org/10.1162/08989290260045882

Li, X., Peng, X., Jordan, C., Li, J., Bi, G., He, Y., … Xi, Z. (2018). mGluR5 antagonism inhibits cocaine reinforcement and relapse by elevation of extracellular glutamate in the nucleus accumbens via a CB1 receptor mechanism. Scientific Reports, 8(1), 3686. https://doi.org/10.1038/s41598-018-22087-1

Lim, K., Wozniak, J., Mueller, B., Franc, D., Specker, S., Rodriguez, C., … Rotrosen, J. (2008). Brain macrostructural and microstructural abnormalities in cocaine dependence. Drug and Alcohol Dependence, 92(1–3), 164–172. https://doi.org/10.1016/J.DRUGALCDEP.2007.07.019

Limpens, J., Schut, E., Voorn, P., & Vanderschuren, L. (2014). Using conditioned suppression to investigate compulsive drug seeking in rats. Drug and Alcohol Dependence, 142, 314–324. https://doi.org/10.1016/J.DRUGALCDEP.2014.06.037

Lind, J., Enquist, M., & Ghirlanda, S. (2015). Animal memory: a review of delayed matching-to-sample data. Behavioural Processes, 117, 52–58. https://doi.org/10.1016/j.beproc.2014.11.019

Liu, D., Gu, X., Zhu, J., Zhang, X., Han, Z., Yan, W., … Li, C. T. (2014). Medial prefrontal activity during delay period contributes to learning of a working memory task. Science, 346(6208), 458–463. https://doi.org/10.1126/science.1256573

Liu, F., Zhang, G., Hornby, G., Vasylyev, D., Bowlby, M., Park, K., … Andree, T. (2006). The effect of mGlu5 receptor positive allosteric modulators on signaling molecules in brain slices. European Journal of Pharmacology, 536(3), 262–268. https://doi.org/10.1016/J.EJPHAR.2006.02.003 Liu, X., Matochik, J., Cadet, J., & London, E. (1998). Smaller volume of prefrontal lobe in polysubstance abusers: a magnetic resonance imaging study. Neuropsychopharmacology, 18(4), 243–252. Retrieved from https://www.nature.com/articles/1395124.pdf

142

Loweth, J., Tseng, K., & Wolf, M. (2014). Adaptations in AMPA receptor transmission in the nucleus accumbens contributing to incubation of cocaine craving. Neuropharmacology, Vol. 76. https://doi.org/10.1016/j.neuropharm.2013.04.061 Lu, L., Grimm, J., Shaham, Y., & Hope, B. (2003). Molecular neuroadaptations in the accumbens and ventral tegmental area during the first 90 days of forced abstinence from cocaine self-administration in rats. Journal of Neurochemistry, 85(6), 1604– 1613. https://doi.org/10.1046/j.1471-4159.2003.01824.x

Lu, Y., Jia, Z., Janus, C., Henderson, J., Gerlai, R., Martin Wojtowicz, J., & Roder, J. (1997). Mice lacking metabotropic glutamate receptor 5 show impaired learning and reduced CA1 long-term potentiation (LTP) but normal CA3 LTP. The Journal of Neuroscience, 17(13), 5196-5205. Retrieved from http://www.jneurosci.org/content/jneuro/17/13/5196.full.pdf

Luís, C., Cannella, N., Spanagel, R., & Köhr, G. (2017). Persistent strengthening of the prefrontal cortex – nucleus accumbens pathway during incubation of cocaine- seeking behavior. Neurobiology of Learning and Memory, 138, 281–290. https://doi.org/10.1016/j.nlm.2016.10.003

Lüscher, C., & Malenka, R. (2012). NMDA receptor-dependent long-term potentiation and long-term depression (LTP/LTD). Cold Spring Harbor Perspectives in Biology, 4(6). https://doi.org/10.1101/cshperspect.a005710

Maas, L., Lukas, S., Kaufman, M., Weiss, R., Daniels, S., Rogers, V., … Renshaw, P. (1998). Functional magnetic resonance imaging of human brain activation during cue-induced cocaine craving. American Journal of Psychiatry, 155(1), 124–126. https://doi.org/10.1176/ajp.155.1.124

MacDonald, A., Cohen, J., Stenger, V., & Carter, C. (2000). Dissociating the role of the dorsolateral prefrontal and anterior cingulate cortex in cognitive control. Science, 288(5472), 1835–1838. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10846167

Mackey, S., & Paulus, M. (2013). Are there volumetric brain differences associated with the use of cocaine and amphetamine-type stimulants? Neuroscience and Biobehavioral Reviews, 37(3), 300–316. https://doi.org/10.1016/j.neubiorev.2012.12.003

Mahoney, J. (2018). Cognitive dysfunction in individuals with cocaine use disorder: Potential moderating factors and pharmacological treatments. Experimental and Clinical Psychopharmacology. https://doi.org/10.1037/pha0000245

Malenka, R. (1994). Synaptic plasticity in the hippocampus: LTP and LTD. Cell, 78(4), 535–538. https://doi.org/10.1016/0092-8674(94)90517-7

143

Manahan-Vaughan, D., & Braunewell, K. (2005). The metabotropic glutamate receptor, mGluR5, is a key determinant of good and bad spatial learning performance and hippocampal synaptic plasticity. Cerebral Cortex, 15, 1703–1713. https://doi.org/10.1093/cercor/bhi047

Mandt, B., Copenhagen, L., Zahniser, N., & Allen, R. (2015). Escalation of cocaine consumption in short and long access self-administration procedures. Drug and Alcohol Dependence, 149, 166–172. https://doi.org/10.1016/j.drugalcdep.2015.01.039

Manes, F., Sahakian, B., Clark, L., Rogers, R., Antoun, N., Aitken, M., & Robbins, T. (2002). Decision-making processes following damage to the prefrontal cortex. Brain : A Journal of Neurology, 125(Pt 3), 624–639. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/11872618

Mao, L., Liu, X., Zhang, G., Chu, X., Fibuch, E., Wang, L., … Wang, J. (2008). Phosphorylation of group I metabotropic glutamate receptors (mGluR1/5) in vitro and in vivo. Neuropharmacology, 55(4), 403–408. https://doi.org/10.1016/J.NEUROPHARM.2008.05.034 Marek, G., & Zhang, C. (2008). Activation of metabotropic glutamate5 (mGlu5) receptors induces spontaneous excitatory synaptic currents in layer V pyramidal cells of the rat prefrontal cortex. Neurosci Lett, 442(3), 239–243. https://doi.org/10.1016/j.neulet.2008.06.083

Marks, C., Shonesy, B., Wang, X., Stephenson, J., Niswender, C., & Colbran, R. (2018). Activated CaMKIIα binds to the mGlu5 metabotropic glutamate receptor and modulates calcium mobilization. Molecular Pharmacology, 94(6), 1352–1362. https://doi.org/10.1124/mol.118.113142 Marszalek-Grabska, M., Gibula-Bruzda, E., Bodzon-Kulakowska, A., Suder, P., Gawel, K., Talarek, S., … Kotlinska, J. (2018). ADX-47273, a mGlu5 receptor positive allosteric modulator, attenuates deficits in cognitive flexibility induced by withdrawal from “binge-like” ethanol exposure in rats. Behavioural Brain Research, 338, 9–16. https://doi.org/10.1016/J.BBR.2017.10.007

Martin-Fardon, R., Baptista, M., Dayas, C., & Weiss, F. (2009). Dissociation of the effects of MTEP [3-[(2-methyl-1,3-thiazol-4-yl)ethynyl]piperidine] on conditioned reinstatement and reinforcement: comparison between cocaine and a conventional reinforcer. The Journal of Pharmacology and Experimental Therapeutics, 329(3), 1084–1090. https://doi.org/10.1124/jpet.109.151357

Martin-Fardon, R., & Weiss, F. (2012). (-)-2-oxa-4-aminobicylco[3.1.0]hexane-4,6- dicarboxylic acid (LY379268) and 3-[(2-methyl-1,3-thiazol-4-yl)ethynyl]piperidine (MTEP) similarly attenuate stress-induced reinstatement of cocaine seeking. Addiction Biology, 17(3), 557–564. https://doi.org/10.1111/j.1369- 1600.2011.00345.x

144

Martín-García, E., Courtin, J., Renault, P., Fiancette, J.-F., Wurtz, H., Simonnet, A., … Deroche-Gamonet, V. (2014). Frequency of cocaine self-administration influences drug seeking in the rat: optogenetic evidence for a role of the prelimbic cortex. Neuropsychopharmacology, 39(10), 2317–2330. https://doi.org/10.1038/npp.2014.66

Martinez, D., Slifstein, M., Nabulsi, N., Grassetti, A., Urban, N., Perez, A., … Huang, Y. (2014). Imaging glutamate homeostasis in cocaine addiction with the metabotropic glutamate receptor 5 positron emission tomography radiotracer [11 C]ABP688 and magnetic resonance spectroscopy. Biological Psychiatry, 75, 165–171. https://doi.org/10.1016/j.biopsych.2013.06.026

Mashhoon, Y., Wells, A., & Kantak, K. (2010). Interaction of the rostral basolateral amygdala and prelimbic prefrontal cortex in regulating reinstatement of cocaine- seeking behavior. Pharmacology, Biochemistry, and Behavior, 96(3), 347–353. https://doi.org/10.1016/j.pbb.2010.06.005

Matochik, J., London, E., Eldreth, D., Cadet, J., & Bolla, K. (2003). Frontal cortical tissue composition in abstinent cocaine abusers: a magnetic resonance imaging study. NeuroImage, 19(3), 1095–1102. https://doi.org/10.1016/S1053-8119(03)00244-1

Matosin, N., Fernandez-Enright, F., Lum, J., Andrews, J., Engel, M., Huang, X., & Newell, K. (2015). Metabotropic glutamate receptor 5, and its trafficking molecules norbin and tamalin, are increased in the CA1 hippocampal region of subjects with schizophrenia. Schizophrenia Research, 166, 212–218. https://doi.org/10.1016/j.schres.2015.05.001

Matzeu, A., Weiss, F., & Martin-Fardon, R. (2015). Transient inactivation of the posterior paraventricular nucleus of the thalamus blocks cocaine-seeking behavior. Neuroscience Letters, 608, 34–39. https://doi.org/10.1016/j.neulet.2015.10.016

McFarland, K., Davidge, S., Lapish, C., & Kalivas, P. (2004). Limbic and motor circuitry underlying footshock-induced reinstatement of cocaine-seeking behavior. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 24(7), 1551–1560. https://doi.org/10.1523/JNEUROSCI.4177-03.2004

McFarland, K., & Kalivas, P. (2001). The circuitry mediating cocaine-induced reinstatement of drug-seeking behavior. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 21(21), 8655–8663. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/11606653

McFarland, K., Lapish, C., & Kalivas, P. W. (2003). Prefrontal glutamate release into the core of the nucleus accumbens mediates cocaine-induced reinstatement of drug- seeking behavior. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 23(8), 3531–3537. https://doi.org/10.1523/JNEUROSCI.23-08- 03531.2003

145

McLaughlin, J., & See, R. (2003). Selective inactivation of the dorsomedial prefrontal cortex and the basolateral amygdala attenuates conditioned-cued reinstatement of extinguished cocaine-seeking behavior in rats. Psychopharmacology, 168(1–2), 57–65. https://doi.org/10.1007/s00213-002-1196-x

McQuail, J., Beas, B., Kelly, K., Simpson, K., Frazier, C., Setlow, B., & Bizon, J. (2016). NR2A-containing NMDARs in the prefrontal cortex are required for working memory and associated with age-related cognitive decline. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 36(50), 12537–12548. https://doi.org/10.1523/JNEUROSCI.2332-16.2016

Ménard, C., & Mi Quirion, R. (2012). Successful cognitive aging in rats: a role for mGluR5 glutamate receptors, homer 1 proteins and downstream signaling pathways. PLoS ONE, 7(1). https://doi.org/10.1371/journal.pone.0028666

Mihindou, C., Guillem, K., Navailles, S., Vouillac, C., & Ahmed, S. (2013). Discriminative inhibitory control of cocaine seeking involves the prelimbic prefrontal cortex. Biological Psychiatry, 73(3), 271–279. https://doi.org/10.1016/J.BIOPSYCH.2012.08.011

Miller, E., Erickson, C., & Desimone, R. (1996). Neural mechanisms of visual working memory in prefrontal cortex of the macaque. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 16(16), 5154–5167. https://doi.org/10.1523/JNEUROSCI.16-16-05154.1996

Moeller, F., Dougherty, D., Barratt, E., Schmitz, J., Swann, A., & Grabowski, J. (2001). The impact of impulsivity on cocaine use and retention in treatment. Journal of Substance Abuse Treatment, 21(4), 193–198. https://doi.org/10.1016/S0740- 5472(01)00202-1

Moeller, F., Hasan, K., Steinberg, J., Kramer, L., Dougherty, D., Santos, R., … Narayana, P. (2005). Reduced anterior corpus callosum white matter integrity is related to increased impulsivity and reduced discriminability in cocaine-dependent subjects: diffusion tensor imaging. Neuropsychopharmacology, 30(3), 610–617. https://doi.org/10.1038/sj.npp.1300617

Moeller, F., Steinberg, J., Schmitz, J., Ma, L., Liu, S., Kjome, K., … Narayana, P. (2010). Working memory fMRI activation in cocaine-dependent subjects: association with treatment response. Psychiatry Research, 181(3), 174–182. https://doi.org/10.1016/j.pscychresns.2009.11.003

Monaco, S., Gulchina, Y., & Gao, W. (2015). NR2B subunit in the prefrontal cortex: A double-edged sword for working memory function and psychiatric disorders. Neuroscience & Biobehavioral Reviews, 56, 127–138. https://doi.org/10.1016/J.NEUBIOREV.2015.06.022

146

Moorman, D., James, M., McGlinchey, E., & Aston-Jones, G. (2015). Differential roles of medial prefrontal subregions in the regulation of drug seeking. Brain Research, 1628(Pt A), 130–146. https://doi.org/10.1016/j.brainres.2014.12.024

Moreno-López, L., Catena, A., Fernández-Serrano, M., Delgado-Rico, E., Stamatakis, E., Pérez-García, M., & Verdejo-García, A. (2012). Trait impulsivity and prefrontal gray matter reductions in cocaine dependent individuals. Drug and Alcohol Dependence, 125(3), 208–214. https://doi.org/10.1016/J.DRUGALCDEP.2012.02.012

Moreno-López, L., Perales, J., van Son, D., Albein-Urios, N., Soriano-Mas, C., Martinez- Gonzalez, J., … Verdejo-García, A. (2015). Cocaine use severity and cerebellar gray matter are associated with reversal learning deficits in cocaine-dependent individuals. Addiction Biology, 20(3), 546–556. https://doi.org/10.1111/adb.12143

Morgan, P., Angarita, G., Canavan, S., Pittman, B., Oberleitner, L., Malison, R., … Forselius, E. (2016). Modafinil and sleep architecture in an inpatient–outpatient treatment study of cocaine dependence. Drug and Alcohol Dependence, 160, 49– 56. https://doi.org/10.1016/J.DRUGALCDEP.2015.12.004

Moutin, E., Raynaud, F., Roger, J., Pellegrino, E., Homburger, V., Bertaso, F., … Perroy, J. (2012). Dynamic remodeling of scaffold interactions in dendritic spines controls synaptic excitability. The Journal of Cell Biology, 198(2), 251–263. https://doi.org/10.1083/jcb.201110101 Mulquiney, P., Hoy, K., Daskalakis, Z., & Fitzgerald, P. (2011). Improving working memory: exploring the effect of transcranial random noise stimulation and transcranial direct current stimulation on the dorsolateral prefrontal cortex. Clinical Neurophysiology, 122(12), 2384–2389. https://doi.org/10.1016/J.CLINPH.2011.05.009

Murray, J., Belin-Rauscent, A., Simon, M., Giuliano, C., Benoit-Marand, M., Everitt, B. J., & Belin, D. (2015). Basolateral and central amygdala differentially recruit and maintain dorsolateral striatum-dependent cocaine-seeking habits. Nature Communications, 6(1), 10088. https://doi.org/10.1038/ncomms10088

Narayana, P., Datta, S., Tao, G., Steinberg, J., & Moeller, F. (2010). Effect of cocaine on structural changes in brain: MRI volumetry using tensor-based morphometry. Drug and Alcohol Dependence, 111(3), 191–199. https://doi.org/10.1016/j.drugalcdep.2010.04.012

Navailles, S., Guillem, K., Vouillac-Mendoza, C., & Ahmed, S. (2015). Coordinated recruitment of cortical–subcortical circuits and ascending dopamine and serotonin neurons during inhibitory control of cocaine seeking in rats. Cerebral Cortex, 25(9), 3167–3181. https://doi.org/10.1093/cercor/bhu112

147

Niswender, C., & Conn, P. (2010). Metabotropic glutamate receptors: physiology, pharmacology, and disease. Annual Review of Pharmacology and Toxicology, 50, 295–322. https://doi.org/10.1146/annurev.pharmtox.011008.145533

Nuijten, M., Blanken, P., van den Brink, W., & Hendriks, V. (2015). Modafinil in the treatment of crack-cocaine dependence in the netherlands: results of an open-label randomised controlled feasibility trial. Journal of Psychopharmacology, 29(6), 678– 687. https://doi.org/10.1177/0269881115582151

O’ Neill, J., Cardenas, V., & Eyerhoff, D. (2001). Separate and interactive effects of cocaine and alcohol dependence on brain structures and metabolites: quantitative MRI and proton MR spectroscopic imaging. Addiction Biology, 6(4), 347–361. https://doi.org/10.1080/135562100

Ohn, S., Park, C., Yoo, W., Ko, M., Choi, K., Kim, G., … Kim, Y. (2008). Time- dependent effect of transcranial direct current stimulation on the enhancement of working memory. NeuroReport, 19(1), 43–47. https://doi.org/10.1097/WNR.0b013e3282f2adfd

Olive, M. (2009). Metabotropic glutamate receptor ligands as potential therapeutics for addiction. Current Drug Abuse Reviews, 2(1), 83–98. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/19630739

Olive, M. (2010). Cognitive effects of group I metabotropic glutamate receptor ligands in the context of drug addiction. European Journal of Pharmacology, 639(1–3), 47–58. https://doi.org/10.1016/j.ejphar.2010.01.029

Ongur, D., & Price, J. (2000). The organization of networks within the orbital and medial prefrontal cortex of rats, monkeys and humans. Cerebral Cortex, 10(3), 206–219. Retrieved from https://watermark.silverchair.com/100206.pdf?token=AQECAHi208BE49Ooan9kkh W_Ercy7Dm3ZL_9Cf3qfKAc485ysgAAAlgwggJUBgkqhkiG9w0BBwagggJFMIICQQ IBADCCAjoGCSqGSIb3DQEHATAeBglghkgBZQMEAS4wEQQMk5c5tyDme0a1Vh atAgEQgIICC0MVA6BDYrhihyYlTYrEjRRzfhFnuIjmNU2q814iYwzgX5CX

Otani, S., Auclair, N., Desce, J., Roisin, M., & Crépel, F. (1999). Dopamine receptors and groups I and II mGluRs cooperate for long-term depression induction in rat prefrontal cortex through converging postsynaptic activation of MAP kinases. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 19(22), 9788–9802. https://doi.org/10.1523/JNEUROSCI.19-22-09788.1999

Owen, A., McMillan, K., Laird, A., & Bullmore, E. (2005). N-back working memory paradigm: a meta-analysis of normative functional neuroimaging studies. Human Brain Mapping, 25(1), 46–59. https://doi.org/10.1002/hbm.20131

Ozawa, S., Kamiya, H., & Tsuzuki, K. (1998). Glutamate receptors in the mammalian central nervous system. Progress in Neurobiology, 54(5), 581–618. https://doi.org/10.1016/S0301-0082(97)00085-3

148

Pace-Schott, E., Morgan, P., Malison, R., Hart, C., Edgar, C., Walker, M., & Stickgold, R. (2008). Cocaine users differ from normals on cognitive tasks which show poorer performance during drug abstinence. The American Journal of Drug and Alcohol Abuse, 34(1), 109–121. https://doi.org/10.1080/00952990701764821

Parvaz, M., Moeller, S., & Goldstein, R. (2016). Incubation of cue-induced craving in adults addicted to cocaine measured by electroencephalography. JAMA Psychiatry, 73(11), 1127–1134. https://doi.org/10.1001/jamapsychiatry.2016.2181

Paxinos, G., & Watson, C. (2007). The rat brain in stereotaxic coordinates (6th ed.). New York, NY: Academic Press.

Pelloux, Y., Minier-Toribio, A., Hoots, J., Bossert, J., & Shaham, Y. (2018). Opposite effects of basolateral amygdala inactivation on context-induced relapse to cocaine seeking after extinction versus punishment. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 38(1), 51–59. https://doi.org/10.1523/JNEUROSCI.2521-17.2017

Pennartz, C., Groenewegen, H., & Lopes da Silva, F. (1994). The nucleus accumbens as a complex of functionally distinct neuronal ensembles: an integration of behavioural, electrophysiological and anatomical data. Progress in Neurobiology, 42(6), 719–761. https://doi.org/10.1016/0301-0082(94)90025-6

Perry, C., Reed, F., Zbukvic, I., Kim, J., & Lawrence, A. (2016). The metabotropic glutamate 5 receptor is necessary for extinction of cocaine-associated cues. British Journal of Pharmacology, 173(6), 1085–1094. https://doi.org/10.1111/bph.13437

Perry, J., Joseph, J., Jiang, Y., Zimmerman, R., Kelly, T., Darna, M., … Bardo, M. (2011). Prefrontal cortex and drug abuse vulnerability: translation to prevention and treatment interventions. Brain Research Reviews, 65(2), 124–149. https://doi.org/10.1016/J.BRAINRESREV.2010.09.001

Pessoa, L., Gutierrez, E., Bandettini, P., & Ungerleider, L. (2002). Neural correlates of visual working memory: fMRI amplitude predicts task performance. Neuron, 35(5), 975–987. https://doi.org/10.1016/S0896-6273(02)00817-6

Petrides, M. (2000). Dissociable roles of mid-dorsolateral prefrontal and anterior inferotemporal cortex in visual working memory. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 20(19), 7496–7503. https://doi.org/10.1523/JNEUROSCI.20-19-07496.2000

Phillips, A., Ahn, S., & Floresco, S. (2004). Magnitude of dopamine release in medial prefrontal cortex predicts accuracy of memory on a delayed response task. The Journal of Neuroscience, 24(2), 547-553. https://doi.org/10.1523/JNEUROSCI.4653-03.2004

149

Pickens, C., Airavaara, M., Theberge, F., Fanous, S., Hope, B., & Shaham, Y. (2011). Neurobiology of the incubation of drug craving. Trends in Neurosciences, 34(8), 411–420. https://doi.org/10.1016/j.tins.2011.06.001

Porter, J., Olsen, A., Gurnsey, K., Dugan, B., Jedema, H., & Bradberry, C. (2011). Chronic cocaine self-administration in rhesus monkeys: impact on associative learning, cognitive control, and working memory. J Neurosci, 31(13), 4926–4934. https://doi.org/31/13/4926 [pii]\r10.1523/JNEUROSCI.5426-10.2011

Potvin, S., Stavro, K., Rizkallah, É., & Pelletier, J. (2014). Cocaine and cognition. Journal of Addiction Medicine, 8(5), 368–376. https://doi.org/10.1097/ADM.0000000000000066

Preuss, T. (1995). Do rats have prefrontal cortex?: the rose-woolsey-akert program reconsidered. Journal of Cognitive Neuroscience, 7(1), 1–24. https://doi.org/10.1162/jocn.1995.7.1.1

Prisciandaro, J., McRae-Clark, A., Myrick, H., Henderson, S., & Brady, K. (2014). Brain activation to cocaine cues and motivation/treatment status. Addiction Biology, 19(2), 240–249. https://doi.org/10.1111/j.1369-1600.2012.00446.x

Radley, J., Anderson, R., Cosme, C, Glanz, R., Miller, M., Romig-Martin, S., & LaLumiere, R. (2015). The contingency of cocaine administration accounts for structural and functional medial prefrontal deficits and increased adrenocortical activation. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 35(34), 11897–11910. https://doi.org/10.1523/JNEUROSCI.4961- 14.2015

Ragozzino, M. (2007). The contribution of the medial prefrontal cortex, orbitofrontal cortex, and dorsomedial striatum to behavioral flexibility. Annals of the New York Academy of Sciences, 1121(1), 355–375. https://doi.org/10.1196/annals.1401.013

Rebec, G., & Sun, W. (2005). Neuronal substrates of relapse to cocaine seeking behavior: role of prefrontal cortex. Journal of the Experimental Analysis of Behavior, 84(3), 653–666. https://doi.org/10.1901/jeab.2005.105-04

Reichel, C., Schwendt, M., McGinty, J., Olive, M., & See, R. (2011). Loss of object recognition memory produced by extended access to methamphetamine self- administration is reversed by positive allosteric modulation of metabotropic glutamate receptor 5. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 36(4), 782–792. https://doi.org/10.1038/npp.2010.212

Riedel, G., Platt, B., & Micheau, J. (2003). Glutamate receptor function in learning and memory. Behavioural Brain Research, 140(1–2), 1–47. https://doi.org/10.1016/S0166-4328(02)00272-3

150

Riha, P., Rojas, J., & Gonzalez-Lima, F. (2011). Beneficial network effects of methylene blue in an amnestic model. NeuroImage, 54(4), 2623–2634. https://doi.org/10.1016/j.neuroimage.2010.11.023

Robinson, T., & Kolb, B. (2004). Structural plasticity associated with exposure to drugs of abuse. Neuropharmacology, 47, 33–46. https://doi.org/10.1016/J.NEUROPHARM.2004.06.025

Rodrigues, S., Bauer, E., Farb, C., Schafe, G., & LeDoux, J. (2002). The group I metabotropic glutamate receptor mGluR5 is required for fear memory formation and long-term potentiation in the lateral amygdala. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 22(12), 5219–5229. https://doi.org/10.1523/JNEUROSCI.22-12-05219.2002

Rogers, J., & See, R. (2007). Selective inactivation of the ventral hippocampus attenuates cue-induced and cocaine-primed reinstatement of drug-seeking in rats. Neurobiology of Learning and Memory, 87(4), 688–692. https://doi.org/10.1016/j.nlm.2007.01.003

Romanides, A., Duffy, P., & Kalivas, P. (1999). Glutamatergic and dopaminergic afferents to the prefrontal cortex regulate spatial working memory in rats. Neuroscience, 92(1), 97–106. https://doi.org/10.1016/S0306-4522(98)00747-7

Romano, C., Sesma, M., McDonald, C., O’malley, K., van den Pol, A., & Olney, J. (1995). Distribution of metabotropic glutamate receptor mGluR5 immunoreactivity in rat brain. The Journal of Comparative Neurology, 355(3), 455–469. https://doi.org/10.1002/cne.903550310

Romero, M., Asensio, S., Palau, C., Sanchez, A., & Romero, F. (2010). Cocaine addiction: diffusion tensor imaging study of the inferior frontal and anterior cingulate white matter. Psychiatry Research: Neuroimaging, 181(1), 57–63. https://doi.org/10.1016/J.PSCYCHRESNS.2009.07.004

Rook, J., Xiang, Z., Lv, X., Ghoshal, A., Dickerson, J., Bridges, T., … Conn, P. (2015). Biased mGlu5-positive allosteric modulators provide in vivo efficacy without potentiating mGlu5 modulation of NMDAR currents. Neuron, 86(4), 1029–1040. https://doi.org/10.1016/j.neuron.2015.03.063 Sakai, K., Rowe, J., & Passingham, R. (2002). Active maintenance in prefrontal area 46 creates distractor-resistant memory. Nature Neuroscience, 5(5), 479–484. https://doi.org/10.1038/nn846

Schneider, C., Rasband, W., & Eliceiri, K. (2012). NIH image to imageJ: 25 years of image analysis. Nat Methods, 9(7), 671–675. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5554542/pdf/nihms876304.pdf

151

Schoepp, D., & Conn, P. (1993). Metabotropic glutamate receptors in brain function and pathology. Trends in Pharmacological Sciences, 14(1), 13–20. https://doi.org/10.1016/0165-6147(93)90107-U

Schwendt, M., Reichel, C., & See, R. (2012). Extinction-dependent alterations in corticostriatal mGluR2/3 and mGluR7 receptors following chronic methamphetamine self-administration in rats. PLoS ONE, 7(3), e34299. https://doi.org/10.1371/journal.pone.0034299

Schwendt, M., Rocha, A., See, R., Pacchioni, A., McGinty, J., & Kalivas, P. (2009). Extended methamphetamine self-administration in rats results in a selective reduction of dopamine transporter levels in the prefrontal cortex and dorsal striatum not accompanied by marked monoaminergic depletion. The Journal of Pharmacology and Experimental Therapeutics, 331(2), 555–562. https://doi.org/10.1124/jpet.109.155770

Scofield, M., Heinsbroek, J., Gipson, C., Kupchik, Y., Spencer, S., Smith, A., … Kalivas, P. (2016). The nucleus accumbens: mechanisms of addiction across drug classes reflect the importance of glutamate homeostasis. Pharmacological Reviews, 68(3), 816–871. https://doi.org/10.1124/pr.116.012484

Seamans, J., Lapish, C., & Durstewitz, D. (2008). Comparing the prefrontal cortex of rats and primates: insights from electrophysiology. Neurotoxicity Research, 14(2– 3), 249–262. https://doi.org/10.1007/BF03033814

Sesack, S., & Pickel, V. (1992). Prefrontal cortical efferents in the rat synapse on unlabeled neuronal targets of catecholamine terminals in the nucleus accumbens septi and on dopamine neurons in the ventral tegmental area. The Journal of Comparative Neurology, 320(2), 145–160. https://doi.org/10.1002/cne.903200202

Shackman, A., Salomons, T., Slagter, H., Fox, A., Winter, J., & Davidson, R. (2011). The integration of negative affect, pain and cognitive control in the cingulate cortex. Nature Reviews Neuroscience, 12(3), 154–167. https://doi.org/10.1038/nrn2994

Shigemoto, R., Kinoshita, A., Wada, E., Nomura, S., Ohishi, H., Takada, M., … Mizuno, N. (1997). Differential presynaptic localization of metabotropic glutamate receptor subtypes in the rat hippocampus. Journal of Neuroscience, 17(19), 7503–7522. https://doi.org/10.1523/JNEUROSCI.17-19-07503.1997

Shigemoto, R., Nomura, S., Ohishi, H., Sugihara, H., Nakanishi, S., & Mizuno, N. (1993). Immunohistochemical localization of a metabotropic glutamate receptor, mGluR5, in the rat brain. Neuroscience Letters, 163(1), 53–57. https://doi.org/10.1016/0304-3940(93)90227-C

Shin, C., Serchia, M., Shahin, J., Ruppert-Majer, M., Kippin, T., & Szumlinski, K. (2016). Incubation of cocaine-craving relates to glutamate over-flow within ventromedial prefrontal cortex. Neuropharmacology, 102, 103–110. https://doi.org/10.1016/J.NEUROPHARM.2015.10.038

152

Shiraishi-Yamaguchi, Y., & Furuichi, T. (2007). The Homer family proteins. Genome Biology, 8(2), 206. https://doi.org/10.1186/gb-2007-8-2-206

Simon, S., Tusch, E., Holcomb, P., & Daffner, K. (2016). Increasing working memory load reduces processing of cross-modal task-irrelevant stimuli even after controlling for task difficulty and executive capacity. Frontiers in Human Neuroscience, 10. https://doi.org/10.3389/FNHUM.2016.00380

Simonyi, A., Schachtman, T., & Christoffersen, G. (2005). The role of metabotropic glutamate receptor 5 in learning and memory processes. Drug News & Perspectives, 18(6), 353. https://doi.org/10.1358/dnp.2005.18.6.927927

Simonyi, A., Schachtman, T., & Christoffersen, G. (2010). Metabotropic glutamate receptor subtype 5 antagonism in learning and memory. European Journal of Pharmacology, 639(1–3), 17–25. https://doi.org/10.1016/j.ejphar.2009.12.039

Sloan, H., Good, M., & Dunnett, S. (2006). Double dissociation between hippocampal and prefrontal lesions on an operant delayed matching task and a water maze reference memory task. Behavioural Brain Research, 171(1), 116–126. https://doi.org/10.1016/J.BBR.2006.03.030

Smith, A., Scofield, M., Heinsbroek, J., Gipson, C., Neuhofer, D., Roberts-Wolfe, D., … Kalivas, P. (2016). Accumbens nNOS interneurons regulate cocaine relapse. J Neurosci, 37(4), 742–756. https://doi.org/10.1523/JNEUROSCI.2673-16.2016

Spivey, J., Padilla, E., Shumake, J., & Gonzalez-Lima, F. (2011). Effects of maternal separation, early handling, and gonadal sex on regional metabolic capacity of the preweanling rat brain. Brain Research, 1367, 198–206. https://doi.org/10.1016/j.brainres.2010.10.038

Stansley, B., & Conn, P. (2019). Neuropharmacological insight from allosteric modulation of mGlu receptors. Trends in Pharmacological Sciences, 40(4), 240– 252. https://doi.org/10.1016/J.TIPS.2019.02.006

Stefani, M., & Moghaddam, B. (2010). Activation of type 5 metabotropic glutamate receptors attenuates deficits in cognitive flexibility induced by NMDA receptor blockade. European Journal of Pharmacology, 639(1–3), 26–32. https://doi.org/10.1016/j.ejphar.2010.01.028

Stefanik, M., Moussawi, K., Kupchik, Y., Smith, K., Miller, R., Huff, M., … LaLumiere, R. (2013). Optogenetic inhibition of cocaine seeking in rats. Addiction Biology, 18(1), 50–53. https://doi.org/10.1111/j.1369-1600.2012.00479.x

Streeter, C., Terhune, D., Whitfield, T., Gruber, S., Sarid-Segal, O., Silveri, M., … Yurgelun-Todd, D. (2008). Performance on the stroop predicts treatment compliance in cocaine-dependent individuals. Neuropsychopharmacology, 33(4), 827–836. https://doi.org/10.1038/sj.npp.1301465

153

Suto, N., Tanabe, L. M., Austin, J., Creekmore, E., Pham, C., & Vezina, P. (2004). Previous exposure to psychostimulants enhances the reinstatement of cocaine seeking by nucleus accumbens AMPA. Neuropsychopharmacology, 29(12), 2149– 2159. https://doi.org/10.1038/sj.npp.1300533

Szumlinski, K., Wroten, M., Miller, B., Sacramento, A., Cohen, M., Ben-Shahar, O., & Kippin, T. (2016). Cocaine self-administration elevates GluN2B within dmPFC mediating heightened cue-elicited operant responding. Journal of Drug Abuse, 2(2). Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/27478879 Tanabe, J., Tregellas, J., Dalwani, M., Thompson, L., Owens, E., Crowley, T., & Banich, M. (2009). Medial orbitofrontal cortex gray matter Is reduced in abstinent substance-dependent individuals. Biological Psychiatry, 65(2), 160–164. https://doi.org/10.1016/J.BIOPSYCH.2008.07.030

Tanimura, Y., King, M., Williams, D., & Lewis, M. (2011). Development of repetitive behavior in a mouse model: roles of indirect and striosomal basal ganglia pathways. International Journal of Developmental Neuroscience, 29, 461–467. https://doi.org/10.1016/j.ijdevneu.2011.02.004

Terraneo, A., Leggio, L., Saladini, M., Ermani, M., Bonci, A., & Gallimberti, L. (2016). Transcranial magnetic stimulation of dorsolateral prefrontal cortex reduces cocaine use: a pilot study. European Neuropsychopharmacology, 26(1). https://doi.org/10.1016/j.euroneuro.2015.11.011

Timmer, K., & Steketee, J. (2012). Examination of a role for metabotropic glutamate receptor 5 in the medial prefrontal cortex in cocaine sensitization in rats. Psychopharmacology, 221(1), 91–100. https://doi.org/10.1007/s00213-011-2548-1

Tomasi, D., & Caparelli, E. (2007). Macrovascular contribution in activation patterns of working memory. Journal of Cerebral Blood Flow and Metabolism : Official Journal of the International Society of Cerebral Blood Flow and Metabolism, 27(1), 33–42. https://doi.org/10.1038/sj.jcbfm.9600314

Tomasi, D., Chang, L., Caparelli, E., & Ernst, T. (2007). Different activation patterns for working memory load and visual attention load. Brain Research, 1132(1), 158–165. https://doi.org/10.1016/j.brainres.2006.11.030

Tomasi, D., Goldstein, R., Telang, F., Maloney, T., Alia-Klein, N., Caparelli, E., & Volkow, N. (2007). Widespread disruption in brain activation patterns to a working memory task during cocaine abstinence. Brain Research, 1171, 83–92. https://doi.org/10.1016/j.brainres.2007.06.102

Tsuchida, A., & Fellows, L. (2009). Lesion evidence that two distinct regions within prefrontal cortex are critical for n -back performance in humans. Journal of Cognitive Neuroscience, 21(12), 2263–2275. https://doi.org/10.1162/jocn.2008.21172

154

Tumeh, S., Nagel, J., English, R., Moore, M., & Holman, B. (1990). Cerebral abnormalities in cocaine abusers: demonstration by SPECT perfusion brain scintigraphy. Radiology, 176(3), 821–824. https://doi.org/10.1148/radiology.176.3.2389042

Uematsu, K., Heiman, M., Zelenina, M., Padovan, J., Chait, B., Aperia, A., … Greengard, P. (2015). Protein kinase A directly phosphorylates metabotropic glutamate receptor 5 to modulate its function. Journal of Neurochemistry, 132(6), 677–686. https://doi.org/10.1111/jnc.13038 Ungless, M., Whistler, J., Malenka, R., & Bonci, A. (2001). Single cocaine exposure in vivo induces long-term potentiation in dopamine neurons. Nature, 411(6837), 583– 587. https://doi.org/10.1038/35079077

Uslaner, J., Parmentier-Batteur, S., Flick, R., Surles, N., Lam, J., McNaughton, C., … Hutson, P. (2009). Dose-dependent effect of CDPPB, the mGluR5 positive allosteric modulator, on recognition memory is associated with GluR1 and CREB phosphorylation in the prefrontal cortex and hippocampus. Neuropharmacology, 57(5–6), 531–538. https://doi.org/10.1016/j.neuropharm.2009.07.022

Uylings, H., Groenewegen, H., & Kolb, B. (2003). Do rats have a prefrontal cortex? Behavioural Brain Research, 146(1–2), 3–17. https://doi.org/10.1016/J.BBR.2003.09.028

Uys, J., & LaLumiere, R. (2008). Glutamate: the new frontier in pharmacotherapy for cocaine addiction. CNS & Neurological Disorders - Drug Targets, 7(5), 482–491. https://doi.org/10.2174/187152708786927868

Van den Oever, M., Spijker, S., Smit, A., & De Vries, T. (2010). Prefrontal cortex plasticity mechanisms in drug seeking and relapse. Neuroscience & Biobehavioral Reviews, 35(2), 276–284. https://doi.org/10.1016/J.NEUBIOREV.2009.11.016 van Haaren, F., Van Zijderveld, G., Van Hest, A., & de Bruin, J. (1988). Acquisition of conditional associations and operant delayed spatial response alternation: effects of lesions in the medial prefrontal cortex. Behavioral Neuroscience, 102(4), 481– 488. https://doi.org/10.1037/0735-7044.102.4.481

Vélez-Hernández, M., Padilla, E., Gonzalez-Lima, F., & Jiménez-Rivera, C. (2014). Cocaine reduces cytochrome oxidase activity in the prefrontal cortex and modifies its functional connectivity with brainstem nuclei. Brain Research, 1542, 56–69. https://doi.org/10.1016/j.brainres.2013.10.017

Verdejo-Garcia, A., Albein-Urios, N., Martinez-Gonzalez, J., Civit, E., de la Torre, R., & Lozano, O. (2014). Decision-making impairment predicts 3-month hair-indexed cocaine relapse. Psychopharmacology, 231(21), 4179–4187. https://doi.org/10.1007/s00213-014-3563-9

155

Vertes, R. (2004). Differential projections of the infralimbic and prelimbic cortex in the rat. Synapse, 51(1), 32–58. https://doi.org/10.1002/syn.10279

Vertes, R. (2006). Interactions among the medial prefrontal cortex, hippocampus and midline thalamus in emotional and cognitive processing in the rat. Neuroscience, 142(1), 1–20. https://doi.org/10.1016/J.NEUROSCIENCE.2006.06.027

Volkow, N., Mullani, N., Gould, L., Adler, S., Krajewski, K. (1988). Cerebral blood flow in chronic cocaine users: a study with positron emission tomography. British Journal of Psychiatry, 152, 641–648. Retrieved from https://www.cambridge.org/core/services/aop-cambridge- core/content/view/201AD1B70106321917AFEA5481737A8D/S0007125000220321 a.pdf/cerebral_blood_flow_in_chronic_cocaine_users_a_study_with_positron_emis sion_tomography.pdf

Volkow, N., Hitzemann, R., Wang, G., Fowler, J., Wolf, A., Dewey, S., & Handlesman, L. (1992). Long-term frontal brain metabolic changes in cocaine abusers. Synapse, 11(3), 184–190. https://doi.org/10.1002/syn.890110303

Vonmoos, M., Hulka, L., Preller, K., Jenni, D., Baumgartner, M., Stohler, R., … Quednow, B. (2013). Cognitive dysfunctions in recreational and dependent cocaine users: role of attention-deficit hyperactivity disorder, craving and early age at onset. The British Journal of Psychiatry, 203, 35–43. https://doi.org/10.1192/bjp.bp.112.118091

Vonmoos, M., Hulka, L., Preller, K., Minder, F., Baumgartner, M., & Quednow, B. (2014). Cognitive impairment in cocaine users is drug-induced but partially reversible: evidence from a longitudinal study. Neuropsychopharmacology, 39(9), 2200–2210. https://doi.org/10.1038/npp.2014.71

Wager, T., & Smith, E. (2003). Neuroimaging studies of working memory: a meta- analysis. Cognitive, Affective & Behavioral Neuroscience, 3(4), 255–274. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/15040547

Wang, F., Flanagan, J., Su, N., Wang, L., Bui, S., Nielson, A., … Luo, Y. (2012). A novel in situ RNA analysis platform for formalin-fixed, paraffin-embedded tissues. JMDI, 14(1), 22–29. https://doi.org/10.1016/j.jmoldx.2011.08.002

Wang, H., & Zhuo, M. (2012). Group I metabotropic glutamate receptor-mediated gene transcription and implications for synaptic plasticity and diseases. Frontiers in Pharmacology, 3, 189. https://doi.org/10.3389/fphar.2012.00189

Wang, J., & O’Donnell, P. (2001). D(1) dopamine receptors potentiate nmda-mediated excitability increase in layer V prefrontal cortical pyramidal neurons. Cerebral Cortex (New York, N.Y. : 1991), 11(5), 452–462. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/11313297

156

Wang, M., Yang, Y., Wang, C., Gamo, N., Jin, L., Mazer, J., … Arnsten, A. (2013). NMDA receptors subserve persistent neuronal firing during working memory in dorsolateral prefrontal cortex. Neuron, 77(4), 736–749. https://doi.org/10.1016/j.neuron.2012.12.032

Wang, X., Moussawi, K., Knackstedt, L., Shen, H., & Kalivas, P. (2013). Role of mGluR5 neurotransmission in reinstated cocaine-seeking. Addiction Biology, 18(1), 40–49. https://doi.org/10.1111/j.1369-1600.2011.00432.x

Warren, B., Mendoza, M., Cruz, F., Leao, R., Caprioli, D., Rubio, F., … Hope, B. (2016). Distinct fos-expressing neuronal ensembles in the ventromedial prefrontal cortex mediate food reward and extinction memories. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 36(25), 6691–6703. https://doi.org/10.1523/JNEUROSCI.0140-16.2016

Watanabe, M., Kodama, T., & Hikosaka, K. (1997). Increase of extracellular dopamine in primate prefrontal cortex during a working memory task. Journal of Neurophysiology, 78(5), 2795–2798. https://doi.org/10.1152/jn.1997.78.5.2795

Waung, M., Pfeiffer, B., Nosyreva, E., Ronesi, J., & Huber, K. (2008). Rapid translation of arc/arg3.1 selectively mediates mGluR-dependent LTD through persistent increases in AMPAR endocytosis rate. Neuron, 59(1), 84–97. https://doi.org/10.1016/j.neuron.2008.05.014

Weber, D., Franceshi, D., Ivanovic, M., Atkins, H., Cabahug, C., Wong, C. (1993). SPECT and planar brain imaging in crack abuse: iodine-123-lodoamphetamine uptake and localization. Journal of Nuclear Medicine, 34(6), 899–907. Retrieved from http://jnm.snmjournals.org/content/34/6/899.full.pdf

West, E., Saddoris, M., Kerfoot, E., & Carelli, R. (2014). Prelimbic and infralimbic cortical regions differentially encode cocaine-associated stimuli and cocaine- seeking before and following abstinence. The European Journal of Neuroscience, 39(11), 1891–1902. https://doi.org/10.1111/ejn.12578

West, R., Maynard, M., & Leasure, J. (2018). Binge ethanol effects on prefrontal cortex neurons, spatial working memory and task-induced neuronal activation in male and female rats. Physiology & Behavior, 188, 79–85. https://doi.org/10.1016/J.PHYSBEH.2018.01.027

Whitney, P., Jameson, T., & Hinson, J. (2004). Impulsiveness and executive control of working memory. Personality and Individual Differences, 37(2), 417–428. https://doi.org/10.1016/J.PAID.2003.09.013

Williams, G., & Goldman-Rakic, P. (1995). Modulation of memory fields by dopamine Dl receptors in prefrontal cortex. Nature, 376(6541), 572–575. https://doi.org/10.1038/376572a0

157

Wilson, S., Sayette, M., & Fiez, J. (2004). Prefrontal responses to drug cues: a neurocognitive analysis. Nature Neuroscience, 7(3), 211–214. https://doi.org/10.1038/nn1200

Wong-Riley, M., Anderson, B., Liebl, W., & Huang, Z. (1998). Neurochemical organization of the macaque striate cortex: correlation of cytochrome oxidase with Na+K+ATPase, NADPH-diaphorase, nitric oxide synthase, and N-methyl-D- aspartate receptor subunit 1. Neuroscience, 83(4), 1025–1045. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/9502244

Woodcock, E., Anand, C., Khatib, D., Diwadkar, V., & Stanley, J. (2018). Working memory modulates glutamate levels in the dorsolateral prefrontal cortex during 1H fMRS. Frontiers in Psychiatry, 9, 66. https://doi.org/10.3389/fpsyt.2018.00066

Xu, J., DeVito, E., Worhunsky, P., Carroll, K., Rounsaville, B., & Potenza, M. (2010). White matter integrity is associated with treatment outcome measures in cocaine dependence. Neuropsychopharmacology, 35(7), 1541. https://doi.org/10.1038/NPP.2010.25

Yang, S., Shi, Y., Wang, Q., Peng, J., & Li, B. (2014). Neuronal representation of working memory in the medial prefrontal cortex of rats. Molecular Brain, 7(1), 61. https://doi.org/10.1186/s13041-014-0061-2

Yhnell, E., Dunnett, S., & Brooks, S. (2016). The utilisation of operant delayed matching and non-matching to position for probing cognitive flexibility and working memory in mouse models of Huntington’s disease. Journal of Neuroscience Methods, 265, 72–80. https://doi.org/10.1016/J.JNEUMETH.2015.08.022

Yuen, E., Wei, J., Liu, W., Zhong, P., Li, X., & Yan, Z. (2012). Repeated stress causes cognitive impairment by suppressing glutamate receptor expression and function in prefrontal cortex. Neuron, 73(5), 962–977. https://doi.org/10.1016/j.neuron.2011.12.033

Zarahn, E., Aguirre, G., & D’Esposito, M. (1999). Temporal isolation of the neural correlates of spatial mnemonic processing with fMRI. Cognitive Brain Research, 7(3), 255–268. https://doi.org/10.1016/S0926-6410(98)00029-9

Zavala, A., Osredkar, T., Joyce, J., & Neisewander, J. (2008). Upregulation of arc mRNA expression in the prefrontal cortex following cue-induced reinstatement of extinguished cocaine-seeking behavior. Synapse, 62(6), 421–431. https://doi.org/10.1002/syn.20502

Zimmer, B., Oleson, E., & Roberts, D. (2012). The motivation to self-administer is increased after a history of spiking brain levels of cocaine. Neuropsychopharmacology : Official Publication of the American College of Neuropsychopharmacology, 37(8), 1901–1910. https://doi.org/10.1038/npp.2012.37

158

Ziółkowska, B., Kiełbin´ski, M., Kiełbin´ski, K., Gieryk, A., Soria, G., Maldonado, R., & Przewłocki, R. (2011). Regulation of the immediate-early genes arc and zif268 in a mouse operant model of cocaine seeking reinstatement. Journal of Neural Transmission, 118(6), 877–887. https://doi.org/10.1007/s00702-011-0583-z

159

BIOGRAPHICAL SKETCH

Christina Gobin received her master’s degree in Experimental Psychology at

Nova Southeastern University where she investigated the effect of sleep quality on cognitive function and emotion processing. She received her PhD in Psychology-

Behavioral and Cognitive Neuroscience at the University of Florida in 2019 where she studied the neurobiology of cocaine addiction. Her research focused on identifying neurobiological substrates underlying post-cocaine cognitive function and drug seeking.

160