Opioid drugs, such as morphine, bind to opioid receptors in the brain and provide an analgesic, rewarding, and euphoric effect. Endogenous opioid peptides also bind to opioid receptors in response to natural rewards such as food and exercise. The hijacking of these natural reward circuits have been hypothesized to lead to addiction. Basal ganglia nuclei are rich in opioid neuropeptides and these opioid neuropeptides and their receptors have been implicated in supporting behavior reinforcement for rewards. The striatum, a major basal ganglia input nuclei, is a mu opioid receptor hotspot. The striatum is implicated in goal directed, habitual and decision-making behaviors, as well as drug addiction, but the role of striatal opioid signaling in the learning and successful completion of operant behaviors and drug reward have yet to be established. To understand how opioid neuropeptides within striatal microcircuits contribute to operant behaviors and opioid drug we used two main approaches. We genetically deleted mu opioid receptors (MORs) from striatal direct pathway neurons, or selectively removed MORs from different striatal regions using AAV-cre. In both cases we found that lever pressing for food rewards was unaffected. In parallel, to test the role of striatal MORs in opioid drug reward, using the same genetic and viral manipulations we ran mice through a morphine conditioned place preference (CPP) assay and found that these animals established a place preference for the morphine-paired chamber. These results indicate that MORs in striatal direct pathway neurons may not be involved in operant behaviors or opioid drug reward.

Cognitive pain modulation holds promise as the basis of pain therapies that do not rely on opioid analgesics. In placebo analgesia, prior experience and expectations lead to pain suppression in response to the administration of an inert substance. Placebo analgesia is associated with neural activity and endogenous opioid neuropeptide signaling in cortical and midbrain structures that control pain. Yet, how cortex exerts top-down control over pain modulatory circuits, and whether opioid analgesia relies on the same neural pathways remains unclear. Here we establish a placebo-like analgesia (PLAN) assay in mice that involves expectations about pain and show that both morphine- and placebo antinociception rely on antinociception neurons in the periaqueductal gray (PAG) that projects to the rostroventral medulla (RVM). In contrast, placebo-like antinociception, but not morphine analgesia, involves projections to the PAG from the medial prefrontal (mPFC) and anterior cingulate (ACC) cortices. Using a photoactivatable opioid antagonist to control the timing and location of drug action, we observed placebo analgesia to require endogenous opioid signaling in ACC, mPFC, and PAG. These results indicate that opioid drugs and cortically generated expectations about pain converge on the PAG to produce analgesia. By providing mechanistic explanations for longstanding observations in human subjects, these findings bridge clinical and preclinical studies into placebo analgesia

Opioids are among the most widely used medication classes globally and are frequently prescribed in the treatment of pain conditions. Paradoxically, some individuals taking opioids might develop a condition of increased pain sensitivity referred to as opioid-induced hyperalgesia (OIH). Understanding the underlying biological mechanisms of OIH will improve treatment outcomes. Current proposed mechanisms of OIH include neuroplastic changes to the descending pain modulatory pathway between the midbrain periaqueductal gray, brainstem rostral ventromedial medulla, and spinal cord Dorsal Horn. In order to extend current knowledge of activity-dependent changes in the descending pain modulatory pathway in OIH, we used immunohistology techniques to examine the activity of three well-studied activity markers– Fos, NPAS4, and pERK1/2– in the ventrolateral periaqueductal gray in a mouse model of OIH. Furthermore, we evaluated changes in these markers as a result of persistent Formalin-evoked pain during OIH. It was found that basal Fos expression was reduced in OIH mice compared to controls, but this difference was abolished when assessing formalin-evoked activity. Together, these data support a model of reduced descending antinociceptive control from the vlPAG to the RVM that may enhance concomitant RVM-mediated pain facilitation in OIH.

Endogenous opioids are neuropeptides, which are a type of specialized neurotransmitter that is poorly understand compared to classical neurotransmitters. It is unknown the time scale of action of neuropeptides, the spread of the peptide, or how one peptide might interact with multiple receptors. Mu and delta opioid receptors (MORs and DORs) are both present on hippocampal interneurons and both bind the opioid peptide enkephalin. It is debated whether or not these two functionally similar receptors interact when co-expressed in the same cells, which would have huge implications on drug design and development. Using electrophysiological assays with novel photoactivateable peptides, we found that DOR has faster onset kinetics and higher ligand potency to enkephalin than MOR, making it the dominant receptor. We found no evidence of functional interactions of MOR and DOR in assays for cross-desensitization and heteromer formation suggesting that the two receptors function independently even though they share signaling pathways. In a separate study, to ask about neuropeptide spread, we tested a novel genetically encoded fluorescent sensor, kLight, based on the kappa opioid receptor (KOR) to detect dynorphin, another opioid peptide. Using simultaneous dynorphin uncaging and imaging of kLight activation, we extracted the apparent diffusion coefficient of dynorphin in brain slices. This coefficient can be used to compare diffusion across multiple conditions and address the question of how far a neuropeptide spreads once it's released. These approaches leverage the latest tools to address the holes in our knowledge of opioid signaling and neuropeptide signaling more broadly.

Understanding the neural circuitry of the opioid system in the brain is a pressing clinical and scientific problem. Endogenous opioids are implicated in myriad behavioral functions including reward and analgesia. In a clinical setting, widely used opioid drugs can produce dangerous off-target effects including addiction and death. Pharmacology has long been used as a blunt tool of neuroscientists and clinicians for manipulating the central nervous system. The diverse effects of systemic opioids on the central nervous system (CNS) and the pitfalls of their clinical use highlight the need for advances in pharmacological tools. Photopharmacology, or the activation of drugs using light, is a potential innovation that will allow for new avenues of inquiry in the study of neural circuits. However, applying photopharmacology to study neural circuits in vivo has not been previously demonstrated as feasible. I show that we have developed a photoactivable opioid agonist, PhOX and antagonist, PhNX as a tool-kit for bidirectional manipulation of opioid receptors in vivo. First, I demonstrate that photoactivation of these compounds produces spatial and temporally specific effects on opioid receptor binding, behavior, and neural activity in vivo. Next, I apply photopharmacology to answer questions about opioid-dopamine interactions. Using same-site photoactivation and calcium monitoring in the Ventral Tegmental Area, I demonstrate, for the first time in vivo, suppression of inhibitory terminals by opioids. Finally, I use photopharmacology to study changes in mesolimbic circuitry under chronic morphine conditions.

Our perception of pain is an important survival mechanism to avoid potentially damaging noxious stimuli. However, chronic pain is debilitating. While opioids remain the best treatment for pain, chronic use can lead to dangerous side effects including addiction, tolerance, and respiratory depression that have contributed to the U.S. Opioid Crisis. A better understanding of the underlying circuitry that modulates pain perception and contributes to opioid analgesia will lead to alternative strategies to suppress pain in the absence of dangerous side effects. The supraspinal descending pain modulatory system (DPMS) shapes pain perception through terminals in spinal cord and is thought to be crucial for opioid analgesia. DPMS release of noradrenaline in the spinal dorsal horn inhibits pain. However, the circuit mechanisms by which the DPMS evokes this release remain unclear. Using anatomy, behavior, slice electrophysiology, and calcium activity measurements in transgenic mice, we establish a major role for the locus coeruleus (LC) in opioid-mediated antinociception. We also define the synaptic inputs by which canonical DPMS nodes, the ventrolateral periaqueductal gray (vlPAG) and rostroventral medulla (RVM), interface with LC. To this end, we discover that a significant portion of opioid antinociception is driven by excitatory output from vlPAG to LC and uncover a previously unstudied opioid-sensitive inhibitory output from RVM to LC, inhibition of which is antinociceptive. We also find that these circuit elements are responsive to noxious stimulation. Finally, we begin to define plasticity within the LC and excitatory inputs to LC in the context of persistent pain. These findings serve to revise circuit models of descending monoaminergic pain modulation and establish promising new targets for pain relief.

Alcohol use disorder (AUD) has major societal impact and few utilized treatments. New molecular targets with therapeutic potential and better understanding of the neurobiology of AUD are needed. Chronic ethanol exposure and consumption can lead to neurobiological changes in the brain’s reward and motivation systems that are viewed as that perpetuate harmful drinking. The development of compulsive drinking putatively involves a shift in striatal control from the NAc to the dorsomedial and then dorsolateral striatum as well as an imbalance between the dopamine receptor D1-direct and D2-indirect medium spiny neuron (MSN) pathways, whereby overactive dMSNs increase ethanol seeking and drinking behavior and underactive iMSNs putatively reduce inhibitory self-regulation. PDE10A is an enzyme of the phosphodiesterase enzyme family, a class of enzymes that regulate cellular responses by hydrolyzing cAMP/cGMP. PDE10A appears to encode the main PDE isoforms that control cAMP and cGMP levels of MSNs in the striatum, and the majority of PDE10A in the mouse striatum is the membrane-enriched PDE10A2 isoform. Recent studies suggest that inhibition of PDE10A family isoforms with different selective inhibitors can reduce alcohol selfadministration in models of excessive or dependent intake, which has raised interest in PDE10A inhibitors as potential treatments for AUD. AMG 579 is a small molecule potent, selective PDE10A inhibitor that recently showed promise for its drug-like physiochemical properties (e.g., lipophilic efficiency) and ability to dose-dependently reduce ethanol consumption in rats. However, it is unknown how AMG 579 alters activation of striatal D1-direct and D2-indirect MSN pathways at doses that are considered effective to reduce drinking as well as whether there is regional variation or sex differences in these effects. There also is uncertainty as to the specific role of the PDE10A2 isoform in the anti-drinking or molecular actions of PDE10A inhibitors like AMG 579. I decided to test the hypothesis that Pde10a2-knockout mice would show increased expression of Drd1, a direct pathway MSN marker, and the immediate-early genes (IEG) Egr1 and Fos as compared to wild-type (WT) mice reflecting decreased regulation in cAMP due to Pde10a2-knockout (KO). I also hypothesized that treatment with the PDE10A inhibitor AMG 579 would increase IEG expression, similar to what was observed previously with the PDE10A inhibitors MR1916 and TAK-063. Further, I hypothesized that treatment effects seen on IEG expression would be in wild-type mice only, which would indicate that the membrane-bound PDE10A2 isoform was involved in the ability of the PDE10A inhibitor to alter MSN activation, and thus absent in Pde10a2-knockout mice. These results were expected to be more pronounced in the dorsal striatum than the NAc, because PDE10A mRNA and protein are expressed at greatest levels in the caudate-putamen, and in male animals more their female counterparts, due to previously reported sex differences in synaptic striatal PDE10A levels and activation effects of the PDE10A inhibitor papaverine in vitro. MSN marker expression was measured using two methods: qPCR to measure overall Drd1, Drd2, Penk, Fos, and Egr1 expression and a method of fluorescent in-situ hybridization known as RNAscope to measure Drd1, Drd2 and Egr1 expression within Drd1- and Drd2- positive cell types. qPCR results show that both treatment with AMG 579 and Pde10a2- knockout decreased expression of Drd1, Drd2, Fos, and Egr1. Pde10a2-knockout seemed to reduce Penk expression as well. There were trends that suggested that there was a stronger reduction in Drd2 expression in AMG 579 KO mice in males than females but not in expression in Pde10a, Drd1, or either of the IEGs. RNAscope results showed that Egr1 expression within Drd1- and Drd2-expressing cells types as well as undefined cell types decreased as a result of treatment with AMG 579. The DMS and, to a lesser extent, the DLS, showed greater significance in Treatment effect than the NAc. Treatment with AMG 579 seemed to have a sexually dimorphic effect in the DMS that varied by gene target or cell type. Overall, the results show that AMG 579 acts differently than other PDE10A inhibitors in that it reduces, instead of increases, MSN marker and IEG expression; further study is needed to determine which method is better suited to modulate problematic drinking behavior

The Locus Coeruleus (LC) is the major source of noradrenaline in the brain and sends distinct projections to supraspinal brain structures and the spinal cord. While ascending projections from the dorsal LC are implicated in anxiogenic and aversive behavior, descending projections from the ventral LC to the dorsal horn of the spinal cord mediate noradrenaline-induced analgesia. In the LC, administration of opioid analgesics increases potassium channel conductance, leading to membrane hyperpolarization and decreased neuronal firing. Yet, why opioid analgesics paradoxically inhibit a major center for noradrenaline-mediated analgesia is unclear. The answer to this may lie in differential transcription of the mu opioid receptor (MOR) along the dorsal-ventral axis of the LC. We hypothesized that the dorsal portion of the LC, responsible for anxiety-generating behavior, may display higher quantities of MOR RNA transcripts compared to the analgesia-mediating ventral portion of the LC. To elucidate if MOR transcription in the LC follows a dorsal-ventral gradient, we performed RNA in situ hybridization on the LC. However, quantification of MOR transcripts in the LC showed no obvious dorsal-ventral gradient of MOR. This suggests that the answer to the paradoxical effect of opioids in the LC does not lie in inhomogeneous MOR transcription along its dorsal-ventral axis.

Nearly 20,000 people died from cocaine overdoses in 2020 (NIDA, 2020). Similarly, overdose deaths involving opioids reached 70,000 following the pandemic (NIDA, 2021). Despite the impact that these disorders have on society, there are no FDA approved drug treatments for CUD and only three for OUD. Therefore, more effective treatment options are needed.One possible target for new pharmacotherapies is the GABA-A system. Methylglyoxal, an endogenous product, is a GABA-A agonist (Distler et al. 2012). The enzyme GLO1, metabolizes MG; thus, GLO1 inhibitors increase levels of MG. Studies have shown the co-administration of a GABA-A agonist, Midazolam, with cocaine potentiates the locomotor activation of cocaine (Morris, 2008). Similarly, co-administration of fentanyl with the GLO1 inhibitor, pBBG, increased locomotor activation (Harp et al., 2022). These studies suggest a role for GABA-A in the response to cocaine and oxycodone. This thesis investigated the potential of GLO1 to alter behavioral responses to cocaine and oxycodone, including locomotor activation and reward seeking behavior, measured by the open field test (OFT) and the conditioned place preference (CPP) test. Results showed that co-administration of pBBG with cocaine, but not oxycodone, potentiates the locomotor activation typical with cocaine. In the conditioned place preference experiments mice showed a robust CPP following conditioning trials, but preference was not affected by pBBG. Although pBBG does not affect drug seeking behavior to cocaine or oxycodone in CPP, the interaction between cocaine and pBBG supports a relationship between GABAergic signaling and the effects of cocaine.