Aging and the Dynorphinergic System: Evaluation of Memory and Motor Systems in Prodynorphin Knockout Mice

University of Kentucky UKnowledge

University of Kentucky Doctoral Dissertations Graduate School

2007

AGING AND THE DYNORPHINERGIC SYSTEM: EVALUATION OF MEMORY AND MOTOR SYSTEMS IN PRODYNORPHIN KNOCKOUT MICE

Xuan V. Nguyen University of Kentucky, [email protected]

Right click to open a feedback form in a new tab to let us know how this document benefits ou.y

Recommended Citation Nguyen, Xuan V., "AGING AND THE DYNORPHINERGIC SYSTEM: EVALUATION OF MEMORY AND MOTOR SYSTEMS IN PRODYNORPHIN KNOCKOUT MICE" (2007). University of Kentucky Doctoral Dissertations. 569. https://uknowledge.uky.edu/gradschool_diss/569

This Dissertation is brought to you for free and open access by the Graduate School at UKnowledge. It has been accepted for inclusion in University of Kentucky Doctoral Dissertations by an authorized administrator of UKnowledge. For more information, please contact [email protected].

ABSTRACT OF DISSERTATION

Xuan V. Nguyen

The Graduate School

University of Kentucky

2007

AGING AND THE DYNORPHINERGIC SYSTEM: EVALUATION OF MEMORY AND MOTOR SYSTEMS IN PRODYNORPHIN KNOCKOUT MICE

ABSTRACT OF DISSERTATION

A dissertation submitted in partial fulfillment of the requirements for the degree in Doctor of Philosophy in the College of Medicine at the University of Kentucky

By Xuan V. Nguyen

Lexington, Kentucky

Director: Dr. Guoying Bing Associate Professor of Anatomy & Neurobiology

Lexington, KY

2007

ABSTRACT OF DISSERTATION

AGING AND THE DYNORPHINERGIC SYSTEM: EVALUATION OF MEMORY AND MOTOR SYSTEMS IN PRODYNORPHIN KNOCKOUT MICE

Dynorphins, endogenous peptide neurotransmitters expressed in the central nervous system, have been implicated in diverse pathophysiological processes, including excitotoxicity, chronic inflammation, traumatic injury, cognitive impairment, and motor dysfunction, with significant changes with aging or age-related disease processes. This has led to the hypothesis that the suppression of dynorphin expression would produce beneficial effects on learning and memory and motor function.

To assess the phenotypic manifestations of chronic suppression of endogenous dynorphin, knockout (KO) mice lacking the coding exons of the gene encoding the prodynorphin (Pdyn) precursor protein, were tested in a series of behavioral, biochemical, and molecular biological studies. Moderately aged Pdyn KO perform comparatively better than similarly aged wild-type (WT) mice in the water maze task, although no Pdyn effect was seen among young adult mice. In addition, young adult Pdyn KO mice show mildly improved performance on a passive avoidance task. Minimal baseline differences were noted in spontaneous locomotor activity in an open-field assay, but Pdyn deletion produced a relative sparing of motor dysfunction induced by the neurotoxin 1-methyl-4- phenyl-1,2,3,6-tetrahydropyridine (MPTP).

To investigate the relationship between aging and brain dynorphin expression in mice, we examined dynorphin peptide levels at varying ages in hippocampus, striatum, and frontal cortex of WT mice by quantitative radioimmunoassay. While aging produces progressive decline in Dyn B in striatum and frontal cortex, Dyn A shows an upward trend in frontal cortex without significant change in striatum. Systemic MPTP produces significant short-term elevations in dynorphin peptides that regress to below baseline by 7 days. HPLC analysis of striatal dopamine shows an age-dependent increase in basal dopamine levels in Pdyn KO mice, an effect that is abolished after MPTP. Western blotting experiments demonstrate that Pdyn deletion is associated with greater phosphorylation at the serine-40 site of tyrosine hydroxylase (TH) despite relatively less total TH immunoreactivity, suggesting a suppressive effect of dynorphins on dopamine synthesis. Microarray analysis of hippocampal tissue from young and aged WT and Pdyn KO mice reveals a number of functional groups of genes demonstrating altered expression. The results of this dissertation support a role of endogenous dynorphins in age-associated cognitive and motor dysfunction.

KEYWORDS: Dynorphin, Aging, Knockout mice, Learning and memory, Motor systems

Xuan V. Nguyen

October 17, 2007

AGING AND THE DYNORPHINERGIC SYSTEM: EVALUATION OF MEMORY AND MOTOR SYSTEMS IN PRODYNORPHIN KNOCKOUT MICE

Xuan V. Nguyen

Guoying Bing, M.D., Ph.D. Director of Dissertation

Jane Joseph, Ph.D. Director of Graduate Studies

October 17, 2007

RULES FOR THE USE OF DISSERTATIONS

Unpublished dissertations submitted for the Doctor's degree and deposited in the University of Kentucky Library are as a rule open for inspection, but are to be used only with due regard to the rights of the authors. Bibliographical references may be noted, but quotations or summaries of parts may be published only with the permission of the author, and with the usual scholarly acknowledgments.

Extensive copying or publication of the dissertation in whole or in part also requires the consent of the Dean of the Graduate School of the University of Kentucky.

A library that borrows this dissertation for use by its patrons is expected to secure the signature of each user.

Name Date

DISSERTATION

Xuan V. Nguyen

The Graduate School

University of Kentucky

2007

AGING AND THE DYNORPHINERGIC SYSTEM: EVALUATION OF MEMORY AND MOTOR SYSTEMS IN PRODYNORPHIN KNOCKOUT MICE

DISSERTATION

A dissertation submitted in partial fulfillment of the requirements for the degree in Doctor of Philosophy in the College of Medicine at the University of Kentucky

By Xuan V. Nguyen

Lexington, Kentucky

Director: Dr. Guoying Bing Associate Professor of Anatomy & Neurobiology

Lexington, KY

2007

DEDICATION

This dissertation is dedicated to my wife Erika, whose patience throughout the duration of my graduate studies enabled me to pursue the research contained herein and whose unconditional support and encouragement enabled me to complete it.

ACKNOWLEDGMENTS

The completion of the following dissertation would not have been possible without the expert scholarly direction of my Dissertation Chair, Dr. Guoying Bing, whose guidance and support proved indispensable. In addition, I would like to express my sincere thanks to my Outside Examiner Dr. Susan Barron and other members of my Dissertation Committee: Dr. Don Gash, Dr. Joe Springer, and Dr. Thomas Foster. In addition, Dr. Kurt Hauser, who had been on my committee until his departure from the University of Kentucky prior to the Final Exam, was a welcome source of information and feedback. The guidance and constructive input from these individuals not only helped to enhance the scientific quality of my dissertation project but also afforded me invaluable insights into the research process that I hope to use throughout my research career. Furthermore, I am indebted to the members of the Bing Lab, other members within the Department of Anatomy and Neurobiology, and multiple collaborators and instructors who have offered insightful comments, informal instruction, technical assistance, and supportive advice and encouragement during various stages of my research. Financial support for the research was derived from multiple sources, including the following external grants awarded to Dr. Guoying Bing: NS 044157 and DAMD17- 9919497. In addition, parts of this work were also supported by an individual training fellowship grant from the National Institute of Mental Health (MH65055-01) and a Howard Hughes Medical Institute predoctoral fellowship.

iii

TABLE OF CONTENTS

ACKNOWLEDGMENTS ...... iii LIST OF TABLES ...... vi LIST OF FIGURES ...... viii CHAPTER 1: Introduction ...... 1 Introduction to dynorphins ...... 1 Anatomical distribution of dynorphin peptides ...... 2 Regulation of dynorphin expression ...... 3 Pharmacology of dynorphins ...... 5 Review of physiological and pathological processes involving dynorphins ...... 6 Aging and the dynorphinergic system ...... 10 Overview of experimental aims ...... 12 CHAPTER 2: Prodynorphin knockout mice demonstrate diminished age- associated impairment in spatial water maze performance ...... 14 Introduction ...... 14 Materials and Methods ...... 15 Animal subjects ...... 15 Water maze spatial navigation testing ...... 16 Latency and path length analysis ...... 17 Probe trial analysis ...... 17 Swim speed measurement ...... 18 Dynorphin A extraction and radioimmunoassay (RIA) ...... 18 Statistical analysis ...... 18 Results ...... 19 Water maze ...... 19 Measurement of Dyn A in brain ...... 26 Discussion ...... 27 CHAPTER 3: Effects of aging and Pdyn deletion on nonspatial learning and memory and spontaneous locomotion in mice...... 31 Introduction ...... 31 Methods...... 32 Animal subjects ...... 32 Inhibitory avoidance ...... 33 Tail flick assay ...... 33 Open-field assay...... 34 Object recognition ...... 35 Results ...... 36 Inhibitory avoidance ...... 36 Open-field analysis ...... 38 Object recognition assay ...... 52 Discussion ...... 55

CHAPTER 4: Effects of prodynorphin deletion on striatal dopamine in mice during normal aging and in response to MPTP ...... 60 Introduction ...... 60 Methods...... 61 Animal subjects ...... 61 HPLC analysis ...... 62 Protein extraction and western blotting ...... 63 Statistical analysis ...... 64 Results ...... 64 Striatal dopamine is elevated in aged Pdyn KO mice ...... 64 Pdyn deletion enhances TH phosphorylation at Ser40 ...... 67 MPTP-induced dopamine loss is unaffected by Pdyn deletion ...... 70 Discussion ...... 73 CHAPTER 5: Radioimmunological quantitation of prodynorphin-derived peptides in mice in aging and an MPTP model ...... 79 Introduction ...... 79 Methods...... 80 Animal subjects ...... 80 Dynorphin extraction and radioimmunoassay (RIA) ...... 81 Statistical analysis ...... 82 Results ...... 83 Discussion ...... 87 CHAPTER 6: Microarray analysis of hippocampal gene expression changes with aging and prodynorphin deletion ...... 90 Introduction ...... 90 Methods...... 91 Animal subjects ...... 91 Microarray processing ...... 92 Probe set list selection ...... 92 Functional categorization of differentially regulated genes ...... 93 Results ...... 94 Effects of aging ...... 94 Effects of Pdyn deletion ...... 95 Discussion ...... 97 CHAPTER 7: Discussion and Conclusions ...... 115 Summary of effects of aging and Pdyn deletion ...... 115 Behavioral effects ...... 115 Molecular changes ...... 116 Role of dynorphins in MPTP neurotoxicity ...... 116 A working model for the role of dynorphins in aging ...... 117 Potential areas of further inquiry ...... 118 Extrapolation to human disease ...... 119 REFERENCES ...... 122 VITA ...... 138

LIST OF TABLES

Table 1.1. Amino acid composition of selected prodynorphin-derived peptides...... 2 Table 4.1. ANOVA results from analysis of effects of MPTP, age, and Pdyn genotype on dopamine and related monoamines...... 69 Table 6.1. Summary of probe set filtering and selection results...... 100 Table 6.2. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated with aging in Experiment A...... 101 Table 6.3. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated with aging in Experiment A...... 102 Table 6.4. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated in aged WT mice compared to young WT mice in Experiment A...... 102 Table 6.5. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated in aged WT mice compared to young WT mice in Experiment A...... 104 Table 6.6. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated in aged Pdyn KO mice compared to young Pdyn KO mice in Experiment A...... 105 Table 6.7. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated in aged Pdyn KO mice compared to young Pdyn KO mice in Experiment A...... 105 Table 6.8. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated with Pdyn deletion in Experiment A...... 105 Table 6.9. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated with Pdyn deletion in Experiment A...... 106 Table 6.10. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated in young Pdyn KO mice compared to young WT mice in Experiment A...... 106 Table 6.11. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated in young Pdyn KO mice compared to young WT mice in Experiment A...... 106 Table 6.12. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated in aged Pdyn KO mice compared to aged WT mice in Experiment A...... 108 Table 6.13. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated in aged Pdyn KO mice compared to aged WT mice in Experiment A...... 109 Table 6.14. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated in aged Pdyn KO mice compared to aged WT mice in Experiment B...... 109 Table 6.15. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated in aged Pdyn KO mice compared to aged WT mice in Experiment B...... 110

Table 6.16. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated with aging in Experiment A...... 110 Table 6.17. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated with aging in Experiment A...... 110 Table 6.18. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated in aged WT mice compared to young WT mice in Experiment A...... 111 Table 6.19. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated in aged WT mice compared to young WT mice in Experiment A...... 111 Table 6.20. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated in aged Pdyn KO mice compared to young Pdyn KO mice in Experiment A...... 111 Table 6.21. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated in aged Pdyn KO mice compared to young Pdyn KO mice in Experiment A...... 112 Table 6.22. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated with Pdyn deletion in Experiment A...... 112 Table 6.23. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated with Pdyn deletion in Experiment A...... 112 Table 6.24. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated in young Pdyn KO mice compared to young WT mice in Experiment A...... 112 Table 6.25. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated in young Pdyn KO mice compared to young WT mice in Experiment A...... 113 Table 6.26. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated in aged Pdyn KO mice compared to aged WT mice in Experiment A...... 113 Table 6.27. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated in aged Pdyn KO mice compared to aged WT mice in Experiment A...... 113 Table 6.28. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated in aged Pdyn KO mice compared to aged WT mice in Experiment B...... 114 Table 6.29. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated in aged Pdyn KO mice compared to aged WT mice in Experiment B...... 114

vii

LIST OF FIGURES

Figure 2.1. Cued water maze task performance and swim speed...... 20 Figure 2.2. Spatial water maze navigation performance...... 21 Figure 2.3. Quadrant analysis of probe trial performance in the water maze task...... 23 Figure 2.4. Platform crossings in water maze probe trials...... 24 Figure 2.5. Average proximity measurements during water maze probe trials...... 25 Figure 2.6. Dyn A quantitation by RIA in mouse brain...... 26 Figure 3.1. Inhibitory avoidance assay results in mice...... 36 Figure 3.2. Tail-flick latencies as a function of Pdyn genotype...... 37 Figure 3.3. Horizontal activity measurements during open-field analysis as a function of Pdyn genotype...... 38 Figure 3.4. Distance measurements during open-field analysis as a function of Pdyn genotype...... 39 Figure 3.5. Movement time during open-field analysis as a function of Pdyn genotype...... 40 Figure 3.6. Vertical activity during open-field analysis as a function of Pdyn genotype...... 41 Figure 3.7. Stereotypy during open-field analysis as a function of Pdyn genotype...... 41 Figure 3.8. Movement speed during open-field analysis as a function of Pdyn genotype...... 42 Figure 3.9. Horizontal activity measurements during open-field analysis as a function of aging...... 43 Figure 3.10. Distance traveled during open-field analysis as a function of aging...... 43 Figure 3.11. Movement time during open-field analysis as a function of aging...... 44 Figure 3.12. Vertical activity during open-field analysis as a function of aging...... 44 Figure 3.13. Stereotypy during open-field analysis as a function of aging...... 45 Figure 3.14. Movement speed during open-field analysis as a function of aging...... 45 Figure 3.15. Gender analysis of horizontal activity in open-field task...... 46 Figure 3.16. Gender analysis of distance traveled in open-field task...... 47 Figure 3.17. Gender analysis of movement time in open-field task...... 48 Figure 3.18. Gender analysis of vertical activity in open-field task...... 49 Figure 3.19. Gender analysis of stereotypy in open-field task...... 50 Figure 3.20. Gender analysis of movement speed in open-field task...... 50 Figure 3.21. Effects of MPTP and Pdyn deletion on locomotor activity...... 51 Figure 3.22. Effects of MPTP and Pdyn deletion on stereotyped behavior and motor speed...... 51 Figure 3.23. Effects of age and Pdyn deletion on object recognition using Protocol 1...... 52 Figure 3.24. Effects of age and Pdyn deletion on total exploration time during object recognition assay using Protocol 1...... 54 Figure 3.25. Effects of age and Pdyn deletion on object recognition using Protocol 2...... 55 Figure 4.1. Dopamine (DA) and its metabolites DOPAC and HVA as a function of age and Pdyn genotype...... 65

viii

Figure 4.2. Dopamine and serotonin turnover ratios as a function of age and Pdyn genotype...... 66 Figure 4.3. HPLC analysis of norepinephrine (NE), serotonin (5-HT), and the serotonin metabolite HIAA as a function of age and Pdyn genotype...... 67 Figure 4.4. Representative Western blot of mouse striatal homogenates...... 68 Figure 4.5. Densitometric quantitation of Western blot data...... 71 Figure 4.6. Effect of MPTP on dopamine and its metabolites as a function of age and Pdyn genotype...... 72 Figure 4.7. Effect of MPTP on NE, 5-HT, and HIAA...... 73 Figure 4.8. Effect of MPTP on dopamine and serotonin turnover ratios...... 74 Figure 5.1. Evaluation of gene dosage on dynorphin peptide levels...... 83 Figure 5.2. RIA comparison of dynorphin peptides with age in striatum and cortex...... 84 Figure 5.3. RIA measurement of dynorphin peptides 6 h after a single sublethal injection of MPTP...... 85 Figure 5.4. RIA quantitation of dynorphin peptides 7 days following intraperitoneal MPTP...... 86

CHAPTER 1: Introduction

Dynorphins, endogenous opioid peptides found in specific pathways in the brain, have recently received increased attention for their roles in a variety of physiological and pathological processes. This dissertation will seek to examine the role of these peptides in the brain during aging with regard to two major behavioral functionalities that are compromised in normal aging and in age-related disorders. The first is learning and memory, a major aspect of cognition that has been exensively studied across multiple species and technical disciplines, but for which little is known about mechanisms contributing to age-related impairment. The second is motor function, a prominent feature of both normal aging and pathological states such as Parkinson's disease. In both these areas, dynorphin peptides have been suggested to influence behavioral endpoints. This chapter will summarize the existing literature on the known properties and actions of dynorphins, with particular attention to their roles in the above functional domains, and establish the rationale for studies described in subsequent chapters.

Introduction to dynorphins First isolated in 1979 (Goldstein et al., 1979), the dynorphins were described originally in terms of their opiate properties, namely as opioid receptor (KOR) agonists. Numerous other studies since have identified other mechanisms by which these peptides influence the many neurobiological processes in which they have been implicated. Historically, when these peptides were first identified, they were given the name “dynorphin” to indicate their relatively high potency in an ileal tissue contractility inhibition assay when compared to morphine (Goldstein et al., 1979). The term dynorphin generally refers to peptides obtained by proteolytical cleavage of a polypeptide precursor encoded by the prodynorphin (Pdyn) gene and most commonly refer to dynorphin A (Dyn A 1-17) or dynorphin B (Dyn B 1-16) [reviewed in (Simonato and Romualdi, 1996; Hauser et al., 2005)]. Other peptides obtained from this prodynorphin precursor include alpha- neoendorphin (α-NE), "Big Dynorphin" (Dyn-32), leumorphin (Dyn B 1-29), Leucine- enkephalin-arginine (Leu-Enk-Arg), and Leucine-enkephalin (Leu-Enk) (Hauser et al.,

2005). Biochemically, dynorphin peptides are C-terminal extensions of Leu-Enk in that they share the same five amino-terminal residues as Leu-Enk, thereby explaining some the pharmacologic similarities between these classes of opiate peptides. Table 1.1 details the amino acid composition of selected prodynorphin-derived peptides. Notably, these peptides have a relatively high proportion of the basic amino acid residues lysine and arginine, resulting in their high positive charge, which imparts certain biophysical properties that can affect their biochemical interactions (Naqvi et al., 1998; Tan-No et al., 2001; Marinova et al., 2005).

Table 1.1. Amino acid composition of selected prodynorphin-derived peptides.

Peptide Sequence Dynorphin A (1-17) YGGFLRRIRPKLKWDNQ Dynorphin A (2-17) GGFLRRIRPKLKWDNQ Dynorphin A (1-13) YGGFLRRIRPKLK Dynorphin B YGGFLRRQFKVVT Dynorphin 32 (Big Dynorphin) (Dyn A)-KR-(Dyn B) -neo-endorphin YGGFLRKYP -neo-endorphin YGGFLRKYPK

Anatomical distribution of dynorphin peptides Immununohistochemistry and in situ hybridization studies have demonstrated that dynorphin peptides are widely distributed in the central nervous system (CNS) (Khachaturian et al., 1982; Watson et al., 1982a; Watson et al., 1982b; Hurd et al., 1999). High Pdyn mRNA levels have been noted in hypothalamus, nucleus accumbens, neostriatum, dentate gyrus, hippocampus, cerebral cortex, and spinal cord (Khachaturian et al., 1993; Yakovleva et al., 2006b). One particular area of interest is the hippocampal formation, where dynorphins are prominent in mossy fiber terminals (McGinty et al., 1983; Zamir et al., 1985). Prodynorphin precursors are synthesized in granule cells of the dentate gyrus and transported along the axons of the mossy fiber pathway to terminate at synapses in the CA3 region of the hippocampus. In addition to the mossy fiber-CA3 pyramidal cell synapses, dynorphins have also been identified at perforant path-granule cell synapses and mossy fiber recurrent synapses in the dentate gyrus (Wagner et al., 1993; Terman et al., 1994; Terman et al., 2000).

Dynorphins are also distributed in the striatonigral system. Within the striatum, dynorphins are found predominantly colocalized with substance P and GABA within dopamine D1-receptor (D1R)-expressing striatonigral projection neurons [reviewed in (Steiner and Gerfen, 1998)]. Striatal dynorphinergic neurons project to the substantia nigra (SN) and contact tyrosine hydroxylase (TH)-positive dopaminergic neurons (Pickel et al., 1993). Since opioid receptors are found on presynaptic terminals in striatum and on glutamatergic neurons projecting from the subthalamic nucleus to the globus pallidus (Abou-Khalil et al., 1984), the neuroanatomical distribution of dynorphins favors a opioid-mediated mechanism by which dynorphins can regulate dopamine (DA) release and/or function in the striatum. A recent study employing electron microscopy with dual labeling of prodynorphin and either D1R or TH in the rat nucleus accumbens showed that a majority (63%) of prodynorphin-containing dendrites colocalized with the D1R (Hara et al., 2006). Moreover, axon terminals containing Pdyn frequently made synaptic connections with D1R-containing dendrites and TH-containing axon terminals. The ultrastructural evidence thus supports a mechanism whereby dynorphin peptides can readily modulate dopamine release or its postsynaptic effects. Similarly, neuroanatomical colocalization of dynorphins with gonadotropins (Knepel et al., 1985) and with vasopressin (Sherman et al., 1986) has also been reported in the anterior pituitary and hypothalamus, respectively. Outside of the CNS, dynorphin peptides are found in a wide range of tissue and organ types, including adrenal glands, heart, lung, and reproductive organs (Simonato and Romualdi, 1996). Additionally, dynorphin-responsive opioid receptors have been found on dendritic cells of mice (Kirst et al., 2002), and dynorphin is found in sympathetic neurons in the liver in close approximation to -receptor-expressing lymphocytes (Kaiser et al., 2003), suggesting a role in neuroimmunomodulation.

Regulation of dynorphin expression The control of dynorphin expression occurs at a number of levels. The molecular biology of the Pdyn gene shows the presence of various sites that may be modulated by transcription factor binding (Simonato and Romualdi, 1996). Its promoter contains a

3 functional AP-1 element (Naranjo et al., 1991; Bronstein et al., 1994), and antisense c-fos mRNA has been found to reduce Pdyn mRNA levels in vitro (Hunter et al., 1995), which suggests that basal expression of dynorphins are at least partially mediated via the AP-1 site. The rat Pdyn promoter contains four distinct cAMP response elements (CREs), and when bound by CREB, these elements suppress Pdyn transcription (Collins-Hicok et al., 1994) and can potentially enhance memory processes (Josselyn et al., 2001; Nguyen, 2001). In the nucleus accumbens, CREB has been shown to increase Pdyn mRNA expression (Carlezon et al., 1998). The Downstream Regulatory Element Antagonistic Modulator (DREAM) transcription factor has been shown to block Pdyn expression in spinal cord (Cheng et al., 2002), and this inhibition also occurs in pancreatic islet cells during low-calcium states (Jacobson et al., 2006). Nitric oxide and cyclic GMP (cGMP) have also been implicated in suppressing Pdyn gene expression. In the hippocampus, injection of the nitric oxide donor Sin-1 or the cGMP analogue 8-bromo-cGMP reduced Pdyn mRNA levels (Johnston and Morris, 1994, 1995), as did injection of NMDA (Johnston and Morris, 1995). Moreover, the nitric oxide synthase inhibitor L-NAME can block the amphetamine-induced increase in Pdyn mRNA expression in the striatum and amygdala (Przewlocka et al., 1996). Dynorphin gene expression is also enhanced in embryonic stem cell nuclei by exposure to Dyn B, suggesting the presence of a feedforward control system via nuclear receptors (Ventura et al., 2003). Environmental and biological factors that are associated with Pdyn gene expression have also been identified. Dynorphin peptides are increased by various types of cellular stressors. Traumatic spinal cord injury causes localized increases in spinal cord dynorphins beginning within hours of the insult and lasting weeks, without a corresponding change in enkephalin (Faden et al., 1985a, b). Inoculation of mycobacteria to induce joint swelling and inflammation in rats as a model for chronic pain resulted in increased dynorphin levels in anterior pituitary, hypothalamus, and spinal cord (Millan et al., 1985), as does a Freund’s adjuvant model of peripheral inflammation (Iadarola et al., 1986). The increase in dynorphin expression in spinal cord following peripheral inflammation is enhanced in aged animals (Zhang et al., 2004a). In addition, dynorphin expression in the hippocampus increases following systemic administration of kainic acid, particularly during the process of mossy fiber sprouting (Simpson et al., 1997).

Surgical lesioning of the perforant path also mimics this effect (Thai et al., 1996). Normal aging is also associated with changes in the dynorphinergic system. Age-related changes in Dyn peptide or Pdyn mRNA expression have been noted in the rodent hippocampus and cortex (Jiang et al., 1989; Zhang et al., 1991; Gallagher and Nicolle, 1993). Alterations in CREB expression (Collins-Hicok et al., 1994) or binding at the AP- 1 site in the Pdyn promoter (Naranjo et al., 1991; Collins-Hicok et al., 1994; Hunter et al., 1995; Kitamura et al., 1995) could increase transcription and thus represent feasible molecular mechanisms for increased Pdyn with age. Prior studies had shown significant increases of approximately 50% in dynorphin immunoreactivity in the hippocampus and frontal cortex of aged rats, especially aged rats who were also memory-impaired (Jiang et al., 1989; Gallagher and Nicolle, 1993). Elevated levels of dynorphin A(1-8) in the hippocampus and frontal cortex were correlated with deficits in Morris water maze performance (Jiang et al., 1989) and with decreased glutamate release (Zhang et al., 1991). Increased dynorphin immunoreactivity was also observed in the supragranular layer and mossy fiber system of amyloid precursor protein (APP)-overexpressing mice (Diez et al., 2000). In humans, dynorphin A is markedly elevated in the frontal cortex of patients with Down syndrome or Alzheimer’s disease (AD) (Risser et al., 1996). In models of PD, several studies have observed decreases in Dyn following experimental manipulations that lesion the nigrostriatal dopaminergic pathway (Gerfen et al., 1991), produce dopamine depletion as in suppression of the vesicular monoamine transporter type 2 (VMAT2) gene (Colebrooke et al., 2007), or cause decreased dopamine receptor activity as in D1-receptor knockout mice (Wong et al., 2003).

Pharmacology of dynorphins Because the known pharmacological activity of dynorphins is not limited to their interaction with the KOR, their effects in vivo have been difficult to fully characterize, and many effects of the prodynorphin-derived peptides may vary from one organ system to the next [reviewed in (Hauser et al., 2005)]. The interaction at the opioid receptor was the first effect of dynorphins to be characterized, with their particularly high affinity for the κ opioid receptor (Goldstein et al., 1979; Chavkin et al., 1982) (Goldstein and

Naidu, 1989) that requires the N-terminal tyrosine residue. Although the affinity of dynorphin for κ opioid receptors is within the picomolar range, physiological concentrations of dynorphins are still capable of binding to μ and δ opioid receptors (Garzon et al., 1984). Interestingly, dynorphins at supraphysiological levels demonstrate significant interactions with the NMDA receptor (Chen et al., 1995a, b; Chen and Huang, 1998; Tang et al., 1999), presumed to be mediated by C-terminal residues. Depending on the particular experimental conditions, the dynorphin peptides can bind to the NMDA receptor and either block or potentiate NMDA receptor activation (Brauneis et al., 1996; Shukla et al., 1997; Zhang et al., 1997; Caudle and Dubner, 1998; Tang et al., 1999). Mass spectrometry studies have confirmed that the positively charged residues of dynorphin A 1-17 and its non-opioid counterpart dynorphin A 2-17 allow the peptides to form stable complexes with a conserved epitope on the NR1 subunit of the NMDA receptor that consists of five adjacent glutamate residues and an aspartate residue (Woods et al., 2006). Dynorphins have also been reported to act at sites that are not affected by either opioid or NMDA antagonists (Tang et al., 2000), including direct non-covalent interactions with the excitatory amino acids glutamate and aspartate (Woods and Zangen, 2001), likely a result of their highly basic amino acid constituency. Additionally, dynorphin peptides have also been observed to possess cell-penetrating capabilities and interact directly with phospholipid membranes (Lind et al., 2006). The complex pharmacologic activity of dynorphin peptides presents a difficulty when using specific pharmacological antagonists to evaluate the role of dynorphins in a particular experimental system, since the use of even highly selective opioid agonists or antagonists may not be sufficient to distinguish clearly the extent to which dynorphins contribute to a given physiological effect.

Review of physiological and pathological processes involving dynorphins The fairly broad interactions of dynorphins across multiple receptor and organ systems have hindered attempts to elucidate fully the functional significance of dynorphinergic activity in vivo. The literature, however, does support a role of dynorphin peptides in

6 certain neurobiological functions, which include learning and memory, motivation, pain and nociception, modulation of seizure threshold, motor activity, and a variety of other physiological processes. Regarding the role of dynorphins in learning and memory, several behavioral studies point toward a detrimental effect of excess dynorphin. Aged rats, for instance, exhibit increases in hippocampal dynorphin that correlate with spatial memory impairment (Jiang et al., 1989; Gallagher and Nicolle, 1993). Microinjection of dynorphin B into the rat hippocampus similarly showed impaired performance in the Morris water maze task (Sandin et al., 1998), and this effect was blocked by the selective κ opioid antagonist nor-binaltorphimine, while nor-binaltorphimine alone had no effect (Sandin et al., 1998). Dynorphin was also found to dose-dependently impair retention in an inhibitory avoidance task in mice (Introini-Collison et al., 1987), and direct infusion of dynorphin A(1-8) into the hippocampus dose-dependently impairs spatial memory in rats (McDaniel et al., 1990). Electrophysiological studies support an ability of dynorphin to inhibit neural plasticity by blocking long-term potentiation (LTP), by mechanisms that may include presynaptic inhibition of glutamate release via κ opioid receptors or postsynaptic modulation of LTP as a retrograde inhibitory neurotransmitter (Wagner et al., 1992; Wagner et al., 1993; Weisskopf et al., 1993). From human studies, AD patients have been noted to exhibit large increases in κ opioid receptor levels (Hiller et al., 1987) and Dyn A (Yakovleva et al., 2006a), and cognitive performance in AD patients can be improved by the nonselective opioid antagonist naloxone (Reisberg et al., 1983). On the other hand, dynorphin activity in the cholinergic system has been noted to be more favorable to learning in memory, as it reduces passive avoidance task deficits induced by mecamylamine (Hiramatsu et al., 1997; Hiramatsu and Watanabe, 2006) or galanin (Hiramatsu et al., 1996). Pain control is a common clinical indication for opiate use in general, although the opioid receptor is seldom specifically targeted. In contrast to other classes of opioids, dynorphins have minimal analgesic effects in the brain (Friedman et al., 1981). While dynorphin peptides can antagonize analgesic effects in opiate-naïve subjects, they may also potentiate analgesia in morphine-tolerant animals [reviewed in (Smith and Lee, 1988)] while exhibiting minimal withdrawal effects (Khazan et al., 1983), suggesting a modulatory role of dynorphins in the pain processing in the brain. In spinal cord,

7 intrathecal dynorphin administration at low concentrations produces analgesia mediated by opioid receptors, but at higher concentratrions can cause allodynia (Vanderah et al., 1996; Laughlin et al., 1997), flaccid hindlimb paralysis (Faden and Jacobs, 1984; Stevens and Yaksh, 1986; Long et al., 1988a; Long et al., 1988b), and neuronal and axonal injury (Caudle and Isaac, 1987; Long et al., 1988a; Long et al., 1988b; Hauser et al., 2001). In fact, a growing body of literature supports a role of dynorphins in exacerbating neuropathic pain, particularly in spinal cord injury models (Lai et al., 2001). Moreover, these latter effects are not mitigated by an opioid antagonist nor by substituting peptides lacking the N-terminal tyrosine residue needed for opiate receptor interactions. This implicates the involvement of non-opioid-mediated mechanisms (Bakshi and Faden, 1990; Vanderah et al., 1996; Shukla et al., 1997; Hauser et al., 2001), such as those involving the NMDA receptor (Caudle and Isaac, 1988; Bakshi and Faden, 1990). Blocking dynorphin by administering synthetic peptides containing the NR1 epitope of the NMDA receptor prevents neurotoxicity in cell culture experiments and in in vivo models of dynorphin-induced allodynia and paralysis (Woods et al., 2006). Additionally, the effects of dynorphins on glutamate and aspartate release are believed to contribute to excitotoxic mechanisms that may exacerbate cellular injury under appropriate circumstances. Intrahippocampal dynorphin injections produce elevations in excitatory amino acids in a non-opioid-dependent manner (Faden, 1992). Potential mechanisms include dynorphin-induced increases in intracellular calcium, as previously observed in cultured cortical neurons and dorsal root ganglion cells (Tang et al., 2000; Lai et al., 2001). In contrast to the excitotoxicity roles noted above, dynorphins also have known anticonvulsant actions (Przewlocka et al., 1983; Siggins et al., 1986) and inhibitory effects on neuronal activity and neurotransmission (Chavkin, 2000). Dynorphins can block glutamate release via κ opioid receptors and post-synaptically modulate LTP (Wagner et al., 1992; Wagner et al., 1993; Weisskopf et al., 1993). Pdyn deletion decreases seizure threshold in the kainic acid model (Loacker et al., 2007). Dynorphins have also been implicated in inflammation, with in vitro administration producing a non-opioid-dependent histamine release from rat mast cells (Sydbom and Terenius, 1985) and in vivo experiments showing dynorphin-induced

8 plasma extravasation in rat skin (Chahl and Chahl, 1986). In models of chronic pain and peripheral inflammation, dynorphin immunoreactivity increases in conjunction with appearance of chronic nociceptive behavioral patterns and immediate-early gene expression (Noguchi et al., 1991; Gillardon et al., 1994; Zhang et al., 2005), and these changes can be suppressed by anti-sense oligonucleotides against c-fos (Gillardon et al., 1994; Zhang et al., 2005). While Pdyn-derived peptides are not required for initiation of neuropathic pain in a spinal nerve ligation model, they are elevated by approximately 10 days post-ligation and are needed for maintenance of chronic neuropathic pain (Wang et al., 2001). In models of spinal cord hyperalgesia, intrathecal dynorphin injection causes release of prostaglandin and excitatory amino acids in cyclooxygenase-dependent and opiate-independent mechanisms (Koetzner et al., 2004). The role of dynorphin peptides in the motor system, including the dopaminergic nigrostriatal pathway, has also begun to receive increased attention in the literature. In the striatum and nucleus accumbens, dynorphins and exogenous agonists have clearly been shown to inhibit dopamine release (Mulder et al., 1984; Herrera-Marschitz et al., 1986; Di Chiara and Imperato, 1988; Reid et al., 1988; Spanagel et al., 1992; Ronken et al., 1993; Schlosser et al., 1995), while antagonists can elevate basal DA release (Spanagel et al., 1992). Dynorphins also reduce glutamate release in output regions of basal ganglia (Maneuf et al., 1995). Striatal neurons are also susceptible to dynorphin- induced apoptosis in cell culture models (Tan-No et al., 2001). Moreover, striatal neurons transfected with Dyn-32 (a.k.a. “big dynorphin”) exhibit cytotoxic effects (Tan- No et al., 2001). In terms of functional behavioral manifestations, administration of high doses of dynorphin peptides into brain regions were demonstrated in early studies to induce barrel rotation in rats in a naloxone-independent manner (Smith and Lee, 1988), while intrathecal dynorphin administration induces flaccid paralysis (Smith and Lee, 1988). More recent studies have shown that activation of the opioid receptor tends to suppress locomotion; for instance, in rats, the agonist CI-977 decreased locomotion in a naloxone-reversible manner (Fernandez et al., 1999). -opiate receptor deficient mice show similar spontaneous locomotor activity compared to WT but are resistant to the decreased locomotor activity that accompanies administration of agonists (Simonin et al., 1998).

The endocrine system is also affected by endogenous dynorphins. Dynorphin elevates serum prolactin presumably by lowering dopaminergic tone in the tuberoinfundibular system (Kreek et al., 1999), thereby removing the inhibitory effect of dopamine on prolactin release. Estrogens also appear to control Pdyn expression in the anterior pituitary (Spampinato et al., 1995). Also, dynorphin A(1-17), via a opioid receptor mechanism, can stimulate glucagon release from pancreatic alpha cells (Jacobson et al., 2006). Dynorphin in the hypothalamus colocalizes with vasopressin (Sherman et al., 1986), and opioid agonists and dynorphin analogues have been demonstrated to induce water diuresis via negative regulation of antidiuretic hormone in a opioid-dependent manner (Currie et al., 1994; Ohnishi et al., 1994). Furthermore, dynorphins increase serum levels of adrenocorticotropic hormone (ACTH) in both ovine and murine models (Cheng et al., 2000; Nardo et al., 2002). A putative role of dynorphins in negative feedback of gonoadotropin acticvity has also been proposed, as progesterone increases Dyn A in cerebrospinal fluid and Pdyn mRNA in hypothalamus, with resultant inhibition of pulsatile gonatotropin-releasing hormone activity (Goodman et al., 2004; Foradori et al., 2005). Other effects of dynorphin peptides that have been reported include decreased blood pressure and heart rate, respiratory depression, body temperature regulation, and appetite stimulation (Smith and Lee, 1988). Fasting increases Pdyn expression, while Pdyn suppression increases fatty acid oxidation and weight loss (Sainsbury et al., 2007).

Aging and the dynorphinergic system Since dynorphins have been implicated in aging, the neurobiology of aging will be briefly reviewed here. Aging is associated with a number of physical and behavioral alterations, often with distinct neurobiological and neurochemical changes in the CNS and commonly associated with functional declines in nervous system function that lead to sensory, motor, cognitive, and numerous other types of deficits. Cognitive changes represent an important component of brain aging that has clear relevance to an individual’s functional status. Rat behavioral studies in particular have documented significant aging declines in cognitive tasks (Barnes, 1987; Greene and Naranjo, 1987). Recent evidence suggests that cognitive decline in most cases cannot be explained solely

10 by decreased neuronal numbers in the aged brain. Age-related deficits in learning and memory are typically attributed not only to varying degrees of degenerative neuronal loss but also to functional decreases in neurotransmission or neural plasticity. One of the most widely studied regions related to cognition is the hippocampus. Numerous studies have attempted to assess whether hippocampal size, volume, or neuronal number is altered with age (Miller and O'Callaghan, 2003). While earlier studies concluded widespread decreases in neuronal density in hippocampus, more recent unbiased stereological counting techniques revealed only small or nonsignificant declines. For instance, one study found stereological estimates of hippocampal volume to decrease only 2.16% compared to normal adult mice (von Bohlen und Halbach and Unsicker, 2002). Neuronal density in hippocampus in this study was found to demonstrate a nonsignificant decrease of 4-7%. While studies in humans have noted progressive decline in certain tasks beginning in early young adulthood, the mechanisms underlying these changes have been difficult to identify. Motor impairments with normal aging have been found across multiple species and are likely due to impairments in the dopamine neurotransmitter system in basal ganglia motor pathways (Ingram, 2000). Dopamine, synthesized by neurons in the SN projecting to the striatum, has been strongly implicated in aging (Reeves et al., 2002). For instance, some early studies have reported substantial declines in normal aged humans from young adult levels (Kish et al., 1992). One study (Kish et al., 1992) observed a statistically significant 60% loss of dopamine in human postmortem striatal specimens in normal 84-year-olds compared to normal 22-year-olds. However, recent studies have concluded that the decreases typically attributed to aging are of much smaller magnitude (Friedemann and Gerhardt, 1992; Burwell et al., 1995; Hebert and Gerhardt, 1998; Stanford et al., 2003). One study in mice (Greenwood et al., 1991) found no age-related change in DA and its metabolite 3,4-dihydroxyphenylacetic acid (DOPAC) relative to striatal wet weight. However, these monoamine levels were increased 5- to 7-fold at 21 months of age compared to at 2 months when expressed relative to the proportion of remaining TH-positive neurons, suggesting the existence of compensatory effects. In pathological states such as Parkinson’s disease, clinically manifested by symptoms of bradykinesia, gait and posture disturbances, rigidity, and

11 resting tremor (Hornykiewicz and Kish, 1987; Forno, 1996), severe dopaminergic neuronal loss in SN is observed (Graybiel et al., 1990), with age as the most identifiable risk factor for sporadic cases of PD. In addition, dopamine dysfunction may also affect cognition as well (Nieoullon, 2002; Marie and Defer, 2003; Nieoullon and Coquerel, 2003; Pillon et al., 2003; Remy and Samson, 2003). The mechanisms by which dopaminergic neuron loss occurs either in normal aging or in pathological conditions are currently unknown. The role of dynorphin peptides in Parkinson’s disease is controversial with at least one study reporting decreased dynorphins in SN of PD patients (Waters et al., 1988) and some reporting no change (Taquet et al., 1985; Baronti et al., 1991). Although agonists have been evaluated for therapy in a small clinical trial of PD patients based on hypothesized effects on glutamatergic neurons, they were not found to have any discernable effects of motor performance and in fact exhibited dose-limiting adverse effects, including behavioral aberrations (Giuffra et al., 1993).

Overview of experimental aims The experimental studies described in this dissertation primarily involve behavioral, biochemical, and molecular biological studies in mice. Specifically, we make use of a strain of knockout (KO) mice lacking the coding exons of the Pdyn gene (Sharifi et al., 2001). These mice do not exhibit any gross developmental, morphological, or overt behavioral deficits (Sharifi et al., 2001). To minimize effects related to genetic background variation, these mice were maintained within the 129SvEv background and compared to 129SvEv WT mice as controls. The use of Pdyn KO mice offers several distinct advantages over studies employing exogenously administered agents. First, acute stresses related to experimental manipulations are minimized. Second, complexities associated with the pharmacodynamics and pharmacokinetics of peptide agents, including limitations on in vivo half-lives and distribution or delivery issues, are circumvented. Third, because potential dynorphin interactions include more than the opioid receptor, the approach of employing knockout mice enables analysis of effects of Pdyn suppression at the gene expression level without necessitating assumptions regarding the types of

12 receptors involved. In addition, long-term effects and compensatory mechanisms, such as those associated with aging, can be more readily evaluated. The experiments described in this dissertation therefore represent novel approaches that would complement and enhance existing knowledge of dynorphin peptides in cognition and motor function. To address the effect of Pdyn deletion on learning and memory impairment with age, a series of behavioral experiments will be described in which WT and Pdyn KO mice are tested in both spatial and nonspatial learning tasks. Spatial learning and memory was assessed using a Morris water maze paradigm (Morris et al., 1982), and non-spatial tasks include both a passive avoidance task, which uses an aversive foot- shock stimulus, and a paired object recognition assay. The role of dynorphins in motor function was evaluated using open-field analysis of spontaneous locomotion and quantitation of striatal content of dopamine and related monoamines under basal conditions and following systemic administration of the neurotoxin 1-methyl-4-phenyl- 1,2,3,6-tetrahydropyridine (MPTP). Next, we will discuss quantitaion of dynorphin peptides in varying regions of the aging or MPTP-treated mouse brain using radioimmunological methods. Finally, gene expression changes with aging and Pdyn deletion were systematically analyzed to identify patterns of upregualted or downregulated genes that may shed light on potential mechanisms to explain the behavioral and neurochemical findings reported in this dissertation.

CHAPTER 2: Prodynorphin knockout mice demonstrate diminished age- associated impairment in spatial water maze performance

Introduction The endogenous opioid peptide dynorphin, expressed widely in the central nervous system (Khachaturian et al., 1982; Watson et al., 1982a; Watson et al., 1982b; Hurd et al., 1999) and particularly prominent in mossy fiber terminals of the hippocampal formation (McGinty et al., 1983; Zamir et al., 1985), has been implicated in modulating learning and memory. Dynorphin peptides are pharmacological agonists at opioid receptors (Chavkin et al., 1982), and to a lesser extent and receptors (Garzon et al., 1984), but have been observed to interact with other non-opioid targets like the NMDA receptor (Chen et al., 1995a, b; Chen and Huang, 1998; Tang et al., 1999) and even the excitatory amino acids themselves (Woods and Zangen, 2001). The effects of endogenous dynorphins in learning and memory are supported by reports of their ability to inhibit neural plasticity by blocking long-term potentiation (LTP) at mossy fiber-CA3 pyramidal cell synapses, perforant path-granule cell synapses, and mossy fiber recurrent synapses in the dentate gyrus (Wagner et al., 1993; Terman et al., 1994; Terman et al., 2000). The mechanisms involved include presynaptic inhibition of glutamate release via κ opioid receptors and postsynaptic modulation of LTP as a retrograde inhibitory neurotransmitter (Wagner et al., 1992; Wagner et al., 1993; Weisskopf et al., 1993). Behavioral studies have documented an inhibitory effect of this neuropeptide on learning and memory. In vivo approaches in rats have demonstrated impairment of spatial memory upon intrahippocampal injection of dynorphin A(1-8) (McDaniel et al., 1990) or dynorphin B (Sandin et al., 1998). Dynorphin was also found to dose- dependently impair retention in an inhibitory avoidance task in mice (Introini-Collison et al., 1987). The findings that dynorphin is elevated in hippocampus and frontal cortex of aged rats, especially memory-impaired rats (Jiang et al., 1989; Gallagher and Nicolle, 1993), further suggested a possible involvement of dynorphins in age-related memory deficits. Of particular relevance to human disease, dynorphin A is markedly elevated in the frontal cortex of patients with Down syndrome or Alzheimer’s disease (AD) (Risser

14 et al., 1996) and in mice overexpressing human amyloid precursor protein (Diez et al., 2000). Large increases in κ opioid receptor levels have also been noted in brains of AD patients (Hiller et al., 1987), and cognitive performance in AD patients can be improved by the nonselective opioid antagonist naloxone (Reisberg et al., 1983). A definitive causative role of increases in dynorphin in producing memory deficits associated with normal aging has not been clearly established, and cellular and genetic mechanisms related to its effect in the hippocampus and other parts of the brain have not been carefully explored. To investigate the mechanisms underlying age- dependent changes in spatial learning and memory and the potential role of dynorphins in mediating these changes, we examine in this study the effects of aging in knockout mice lacking the coding exons for the precursor prodynorphin (Pdyn) (Sharifi et al., 2001).

Materials and Methods Animal subjects All animals in this study were wild-type (WT) or Pdyn KO mice that were maintained in an approved animal facility at the University of Kentucky. WT mice were derived from the 129SvEv-Tac line, and Pdyn KO mice were originally obtained by targeted deletion of the coding exons of the prodynorphin gene and bred into a 129SvEv-Tac background as previously described (Sharifi et al., 2001). These KO mice demonstrate no serious structural, developmental, or behavioral phenotypic impairment (Sharifi et al., 2001). To minimize effects of genetic background variation, the WT and Pdyn KO mice used in this study were obtained from offspring of multiple matings between Pdyn KO and WT mice. The aged mice included in this analysis ranged from 13 to 17 months at the time of testing, averaging 15.8 ± 2.0 (mean ± S.D.) months for WT and 15.2 ± 1.8 months for KO. Young mice were 3 to 6 months, averaging 4.1 ± 1.8 months for WT and 3.9 ± 1.5 months for KO. Animals were housed in groups of up to 4 per cage with food and water ad libitum in a well-ventilated room with a 12 h: 12 h light: dark cycle. Prior to weaning, tail specimens were collected from each animal, and DNA was extracted for subsequent genotyping to confirm the identity of the Pdyn locus by the polymerase chain reaction (PCR) using primer pairs specific for each genotype as described previously (Sharifi et

15 al., 2001). The following primers were used in the PCR assay to detect deletion of the coding exons of Pdyn: 5'-ATCCAGGAAACCAGCAGCGGCTAT-3' and 5'- ATTCAGACACATCCCACATAAGGACA-3'. The primers used to detect WT alleles at the Pdyn locus were 5'-CAGGACCTGGTGCCGCCCTCAGAG-3' and 5'- CGCTTCTGGTTGTCCCACTTCAGC-3'. The products were amplified in a Perkin- Elmer thermal cycler at 94°C for 5 min, followed by 30 cycles of 94°C for 30 sec, 65°C for 30 sec, and 72°C for 30 sec. Samples were then incubated at 72°C for 5 min, followed by electrophoresis on 1% agarose gels containing ethidium bromide and visualization under ultraviolet (UV) lighting. Only male mice were used in the behavioral experiments. All procedures involving animals are approved by the Institutional Animal Care and Use Committee at the University of Kentucky.

Water maze spatial navigation testing The Morris water maze protocol used in this study was an adaptation of that used previously in the detection of subtle spatial learning and memory differences (Fugger et al., 1998). Water maze tests were conducted in a 1-m-diameter circular pool within an isolated well-lit room containing a number of two- and three-dimensional cues external to pool. The water was opacified by toxic white powdered paint and held at a constant temperature of 27±2oC. The water maze testing procedure involves one day of the cued phase of the water maze task, followed by three consecutive days of the spatial (hidden platform) phase. Twelve trials were performed per day, grouped into four blocks of three trials each. An interval of 20 sec was placed between trials of the same block, with a 15- to 20-minute interval between blocks. During the cued phase, a visible rod was affixed to the platform, and the platform location and release points were varied across trials to ensure that the vector from the release point to the platform changes between each successive pair of trials. Animals that did not escape more than three times during the last six trials of cue discrimination training were considered to have deficits in motivation or sensorimotor function and were removed from the study. During spatial water maze testing, the platform location was fixed at a constant position 23 cm from the perimeter of the pool in the center of a predetermined quadrant, and the release points were selected according to a predetermined pseudorandom schedule from the three non-target

16 quadrants. In both the cued and spatial tasks, the mouse was allowed to search for the platform for a maximum of 60 seconds, after which it was guided to the platform and allowed to remain on the platform for 10 seconds. Video monitoring and tracking were performed for each trial using Water Maze v3.44 software interfaced with a Videomex recording system (Columbus Instruments, Columbus, Ohio) to record swim distances, quadrant preferences, platform crossings, and average proximity to platform. The penultimate trial of each day of spatial water maze testing was an acquisition probe trial to assess intra-day spatial learning, and the first trial of Days 2 and 3 of the spatial phase served as 24-hour retention probe trials to record how accurately the mice retain the correct location of the target platform from the previous day of training.

Latency and path length analysis Latency to platform was recorded using a stopwatch or digital timer to assess the total time from release of the subject into the pool until the subject successfully climbed onto the platform. Mice failing to locate the platform within the allotted minute were assigned a latency of 60 sec. Path length was recorded automatically by the video monitoring software based on the visual contrast between the mouse and the surrounding water. Occasional light artifacts were automatically removed by the video processing software, and additionally, all trajectories were screened manually in a blinded fashion to eliminate path length values for trials containing an unacceptably high content of light artifacts that were not removed by the automated artifact removal algorithm.

Probe trial analysis The total time of each 60-second probe trial spent in each of the four quadrants was recorded by the video processing software and was used to calculate the percentage of time the subject spent in the target or opposite quadrants. During each probe trial, the number of platform crossings during the 60-second interval was also recorded automatically by the video processing software as detailed below. A circular area 10 cm in diameter was defined in the center of the target quadrant at the position of the target platform, and an automated counter recorded the number of times the animal subject traversed the defined area. To minimize the effects of differences in motor activity or of

17 random search strategies, equivalent areas were similarly defined in each of the other three quadrants, and the number of these non-target crossings were averaged and subtracted from the number of target platform crossings to obtain a net platform crossings index. The expected value of this index in the absence of a spatial memory-dependent search strategy is therefore taken to be zero. Average proximity to platform for each probe trial was also similarly recorded by the water maze analysis software.

Swim speed measurement Average swim speed was estimated for each mouse by taking the mean swim distances recorded by the tracking software in each of the five probe trials and dividing by 60 sec, the probe trial duration.

Dynorphin A extraction and radioimmunoassay (RIA) Frontal cortical and hippocampal tissue were isolated from 3-month (n = 8) and 12-month

(n = 5) male WT mice following rapid asphyxiation by CO2 and were frozen in liquid nitrogen prior to being processed for radioimmunoassay as previously described (Christensson-Nylander et al., 1985; Christensson-Nylander and Terenius, 1985; Nylander et al., 1991). Briefly, tissues were incubated in 1 M hot acetic acid containing 0.01% mercaptoethanol, homogenized by sonication, heated again to 95°C, and subsequently centrifuged. The resulting supernatant was applied onto ion exchange columns, and the resulting fractions dried using a vacuum centrifuge and redissolved in methanol/0.1 M HCl prior to radioimmunoassay. Dynorphin A (Dyn A) was radiolabeled with 125I using chloramine T, and the labeled peptide was purified by high pressure liquid chromatography (HPLC). Samples or standards were incubated with antiserum overnight, and antibody-bound peptide was separated from free peptide by the non- immune globulin precipitation method prior to quantitation. One hippocampal specimen in the 3-month group was lost due to technical reasons.

Statistical analysis Statistical analyses were performed using Systat 10 software (SPSS Inc., Chicago, Illinois). Three-way ANOVA with repeated measures was used in conjunction with

18 selected a priori comparisons to determine statistical significance between groups and across trials or days in the behavioral experiments. To determine whether the distribution of a particular measured variable, such as net platform crossings index, differs from that expected by chance, a one-sample Student’s t-test was used for comparison with the appropriate reference mean. The effect of age on dynorphin levels as measured by radioimmunoassay was analyzed by the Student’s t-test. Statistical significance was determined based on a p-value less than 0.05.

Results Water maze A total of 18 aged Pdyn KO mice, 17 aged WT mice, 11 young Pdyn KO mice, and 13 young WT mice were initially tested on the cue discrimination task to assess visual acuity, sensorimotor function, and general motivation. Two animals, both aged Pdyn KO mice, failed to consistently locate the visible platform and were removed from the study. Based on the remaining mice, all four groups showed significant learning effects across the four blocks of the cued task (Figure 2.1), as determined by a repeated-measures ANOVA on either block-averaged latencies [F(3, 159) = 53.3, p<0.0005] or path lengths to platform [F(3, 132) = 40.2, p<0.0005]. Not unexpectedly, latencies for the younger groups approached optimal values earlier than for the aged groups, resulting in a significant age effect [F(1, 53) = 9.90, p<0.003] and learning-age interaction [F(3, 159) = 5.19, p<0.002] (Figure 2.1). Individual contrasts revealed that aged mice reached the platform more slowly among both Pdyn KO (p<0.003 in Block 2; p<0.05 in Block 3) and WT (p<0.02 in Blocks 2 and 3) mice. However, age effects between groups vanish when analyzing path length [F(1, 44) = 3.59, NS] (Figure 2.1), suggesting that the higher latencies in the aged groups are largely attributable to differences in swim speed. Indeed, average swim speeds recorded during the five probe trials were significantly higher in young mice than aged mice [F(1, 53) = 24.7, p<0.0005], and this effect was apparent among both the Pdyn KO (p<0.0005) and WT (p<0.004) genotypes. Significant genotype effects were not observed in either swim speed [F(1, 53) = 1.59, NS] or cued task performance [latency: F(1, 53) = 0.007, NS; path length: F(1, 44) = 0.596, NS].

young WT 50 900 *** young Pdyn -/- *** 45 aged WT 800 20 aged Pdyn -/- 40 *** 700 35 600 15 30 * 500 25 ** 10 400 20 ** 300

15 [cm/sec]speed Swim 5

10 200

Cued task latency to platform [sec] latency task Cued 5 100 0 Cued task path length path to platform [cm] task Cued 0 0

1 2 3 4 1 2 3 4 aged WT young WT aged Pdyn -/- Block number Block number young Pdyn -/-

Figure 2.1. Cued water maze task performance and swim speed. Left: Latency measurements during the cue discrimination task show a highly significant trend of learning across trials and a significant age effect during blocks 2 and 3, but no Pdyn genotype effect. Right: The age effect is absent in the analysis of cued task path length, and all four groups demonstrate comparably strong learning across trials. Far right: Young mice swim significantly faster than aged mice, but no significant Pdyn genotype effect is seen. *p<0.05, **p<0.02, ***p<0.005 relative to young controls.

To test whether the mice show increased efficiency in locating the submerged platform across time, the spatial water maze swim task was performed across three consecutive days. Daily averages for latency to platform showed a monotonic decrease across the three days of spatial testing for all groups tested (Figure 2.2). When analyzing individual group performance by repeated-measures ANOVA on daily mean latencies, significant learning effects were seen across days for young Pdyn KO and young WT groups (p<0.0005) and for the aged Pdyn KO group (p<0.05), but not for the aged WT group (p=0.11). However, ANOVA found no genotype effect [F(1, 53) = 0.376, NS] even though a significant aging effect [F(1, 53) = 28.2, p<0.0005], learning effect [F(2, 106) = 34.6, p<0.0005], and age-learning interaction [F(2, 106) = 5.81, p<0.004] were observed. Significantly higher latencies to platform were evident in aged mouse groups compared to young mice within both the WT (p<0.03 on Day 1; p<0.0005 on Days 2 and 3) and Pdyn KO (p<0.001 on Day 2; p<0.004 on Day 3) genotypes.

Because individual differences in swim speed may account for some of the variance in latencies, the mean path lengths for non-probe trials in each day were also analyzed by repeated measures ANOVA to obtain a speed-independent measure of spatial memory task performance. A linear trend across days is clearly evident in the daily averages for swim distance (Figure 2.2). The young Pdyn KO, young WT, and aged Pdyn KO groups all demonstrated highly significant learning effects across days p<0.0005), as did the aged WT group (p<0.004). As observed in the latency analysis, ANOVA showed a significant age effect [F(1, 53) = 15.7, p<0.0005], learning effect [F(2, 106) = 56.2, p<0.0005], and age-learning interaction [F(2, 106) = 3.74, p<0.03], but no significant main effect of Pdyn genotype [F(1, 53) = 0.380, NS]. Aged Pdyn KO mice on average use a consistently shorter path to the platform than for the aged WT mice, and the difference reaches significance on Day 3 (p<0.05). Average latencies for the aged Pdyn KO mice were also less than for aged WT mice but failed to reach significance.

35 600 young WT * young Pdyn -/- *** 30 aged WT aged Pdyn -/- *** 500 25 *** 400 *** 20 *** 300 ** # 15 **

200

Latency to platform [sec] Latency

PathPathlengthlength platformplatformtoto [cm][cm] 5 100

0 0 Day 1 Day 2 Day 3 Day 1 Day 2 Day 3

Figure 2.2. Spatial water maze navigation performance. Left: Daily means of latency to platform show a monotonically decreasing pattern for all four groups with significant age effects. *p<0.025, **p<0.005, ***p<0.001 relative to respective young controls. Right: Mean path lengths similarly show significant age differences on Days 2 and 3, but in addition, a significant difference between aged Pdyn KO and WT mice is observed on Day 3. *p<0.05, **p<0.02, ***p<0.001 relative to young controls; #p<0.05 between Pdyn KO and WT.

Age effects on path length were also significant for WT mice (p<0.001 on Day 2; p<0.005 on Day 3) and for Pdyn KO mice (p<0.02 on Day 2). To detect differences among groups in the extent of learning accomplished during each day or in the amount of spatial information retained from the previous day’s training, three acquisition and two retention probe trials were performed over the course of the spatial phase of water maze testing. Quadrant dwell times, platform crossings, and average proximity to target platform were measured. ANOVA of percent time spent in the target quadrant during acquisition probes (Figure 2.3) showed a significant age effect [F(1, 53) = 23.4, p<0.0005], learning effect across days [F(2, 106) = 18.3, p<0.0005], learning-genotype interaction [F(2, 106) = 4.32, p<0.02], and learning-age interaction [F(2, 106) = 3.15, p<0.05]. Target quadrant dwell times for retention probe trials also showed a significant age effect [F(1, 53) = 13.7, p<0.001) and learning effect [F(1, 53) = 14.7). p<0.0005] but no Pdyn genotype effect [F(1, 53) = 0.174, NS]. An analogous analysis of opposite quadrant dwell times revealed a significant age effect [F(1, 53) = 13.6, p<0.001], learning effect [F(2, 106) = 7.79, p<0.001], and learning-genotype interaction [F(2, 106) = 3.14, p<0.05] during acquisition and only an age effect [F(1, 53) = 10.5, p<0.002] and a learning effect [F(1, 53) = 7.77, p<0.007] during retention. Measured averages for target quadrant dwell times were higher in the aged Pdyn KO mice than aged WT mice on all three acquisition probe trials, but the difference as indicated by ANOVA was significant only at the end of Day 1 (p<0.02). The acquisition trial on Day 1 also indicated the time spent in the opposite quadrant was significantly lower for the aged Pdyn KO compared to aged WT mice (p<0.03). Interestingly, young Pdyn KO mice performed approximately as well as young WT mice on most days; however, a significant difference (p<0.03) was noted for the Day 3 acquisition probe trial with young WT exhibiting relatively increased goal quadrant search behavior. Genotype differences within the young age group were not observed when analyzing opposite quadrant dwell times. Aged WT mice spent significantly less time in the target quadrant than young WT mice on all three acquisition trials (p<0.001 on Day 1; p<0.001 on Days 2 and 3) and on both retention trials (p<0.005 on Day 2; p<0.001 on Day 3). The age effect on target quadrant dwell times within the Pdyn KO mice was not significant on either of the retention probes and was significant only on the acquisition probe trial on

Day 2 (p<0.02). Analysis of opposite quadrant dwell times also failed to show significant age effects within the Pdyn KO genotype except on Day 3 of acquisition (p<0.03), whereas aged WT mice consistently spend significantly more time in the

young WT young Pdyn -/- *** 80 80 aged WT * aged Pdyn -/- *** *** 70 70 * ** 60 *** 60

50 * 50

40 40

30 30

20 20

% Time quadrant in target % Time inquadrant target

10 10 50 50 0 0 45 Day 1 Day 2 Day 3 45 Day 2 Day 3 40 40 * * 35 * 35 * 30 * * 30 ** 25 25

20 20

15 15 % Time in opposite quadrant % Time in opposite quadrant 10 10

5 5

0 0 Day 1 Day 2 Day 3 Day 2 Day 3 Acquisition Retention probe probe Figure 2.3. Quadrant analysis of probe trial performance in the water maze task. Left: Acquisition probe trial analysis suggests aged Pdyn KO mice spend significantly more time in the correct quadrant and less time in the opposite quadrant at the end of Day 1 than aged WT mice. Moreover, the age effect for the WT mice is quite pronounced all three days, whereas the Pdyn KO mice show a significant age effect only on Day 2. Interestingly, young Pdyn KO mice perform more poorly than young WT mice on Day 3. Right: Retention probe trials demonstrated an age difference only within the WT mice when examining either the target or opposite quadrant. *p<0.03; **p<0.005; ***p<0.001. The dashed line denotes the reference value expected by chance.

opposite quadrant than young WT mice on all acquisition (p<0.02 on Day 1; p<0.03 on Day 2; p<0.003 on Day 3) and both retention (p<0.02 on Day 2; p<0.03 on Day 3) probes. While quadrant dwell times provide crude indications of spatial learning and memory, platform crossings and proximity measures were also recorded to offer a more refined assessment. The net platform crossings index during the probe trials (Figure 2.4) showed a significant overall age effect [F(1, 53) = 42.7, p<0.0005 for acquisition; F(1, 53) = 36.2, p<0.0005 for retention] and a significant repeated measures effect across days [F(2, 106) = 4.33, p<0.02 for acquisition; F(1, 53) = 4.09, p<0.05 for retention]. WT mice showed highly significant declines in platform crossings with age in acquisition (p<0.003 on Day 1; p<0.0005 on Days 2 and 3) and retention (p<0.001 on Day 2; p<0.0005 on Day 3) probes. For Pdyn KO mice, the age differences were smaller but still significant during both acquisition (p<0.004 on Days 1 and 2; p<0.04 on Day 3) and

young WT 10 young Pdyn -/- 10 aged WT # aged Pdyn -/- # 8 *** 8 * ** #

6 ** 6 ** *** *** 4 4 * **

2 * 2

Net platform indexcrossings Net Net platform indexcrossings Net 0 0 Day 1 Day 2 Day 3 Day 2 Day 3

-2 Acquisition probe -2 Retention probe

Figure 2.4. Platform crossings in water maze probe trials. Left: The number of times aged Pdyn KO mice cross the target platform location during the acquisition probe trials is significantly different from chance as early as the end of Day 2, whereas aged WT mice achieve a comparable level of significance much later. Right: In the retention probe trials, although aged Pdyn KO mice have on average more platform crossings than aged WT mice, the difference fails to reach statistical significance. In both acquisition and retention, age effects are significant, but for clarity, age differences are not depicted on these graphs. *p<0.025, **p<0.01, ***p<0.005, #p<0.0005 relative to chance.

24 retention (p<0.04 on Days 2 and 3). While no explicit genotype effect was seen in the ANOVA model for platform crossings in either the acquisition or retention probes, acquisition probe performance for aged WT mice attained levels significantly different from chance one day later than aged Pdyn KO mice. The higher swim speeds observed for the younger mice could partially account for a higher number of platform crossings especially for mice showing strong preferences for the target quadrant. When analyzing average proximity to the target platform location during the acquisition probe trials (Figure 2.5), a significant learning effect [F(2, 106) = 13.3, p<0.0005], learning-genotype interaction [F(2, 106) = 4.34, p<0.02], and learning-age interaction [F(2, 106) = 3.85, p<0.03] were observed, with no significant main effect of age [F(1, 53) = 1.40, NS] or Pdyn genotype [F(1, 53) = 0.00, NS]. Aged Pdyn KO mice

young WT 35 young Pdyn -/- 35 aged WT aged Pdyn -/- 30 30 *

25 25

20 20

15 15

10 10

Average proximity [cm] Average Average proximity [cm] Average

5 5

0 0 Day 1 Day 2 Day 3 Day 2 Day 3 Acquisition probe Retention probe

Figure 2.5. Average proximity measurements during water maze probe trials. Left: Average proximity to the target platform location during the acquisition probe trials was slightly but not significantly lower in aged Pdyn KO compared to aged WT. However, young WT mice performed better than aged WT mice on Day 3. *p<0.02. Right: No significant differences were detected in average proximity measures during retention trials.

25 were on average closer to the target platform location than aged WT mice on each day, with the largest difference at the end of Day 1 approaching statistical significance (p=0.057). The only significant age difference was observed on Day 3 between young WT and aged WT mice (p<0.02). During retention probes, however, the proximity measurements were not found to demonstrate any significant effects.

Measurement of Dyn A in brain To test whether dynorphin levels are altered with aging in normal mice, Dyn A in frontal cortex and hippocampus was measured by radioimmunoassay in young and aged male WT mice. A two-tailed t-test detected quantitatively higher Dyn A levels in frontal cortex [t(11) = 2.83, p<0.02] but not in hippocampus [t(12) = 0.051, NS] of 12-month WT mice compared to 3-month controls (Figure 2.6). To assess whether additional elevations in dynorphin may occur at more advanced ages, Dyn A was also measured in a small number (n = 3) of 24-month male WT mice. Dyn A levels for the 24-month age point [9.20 ± 1.22 fmol/mg (mean ± S.E.M.)] were comparable to the 12-month group

3-mo.

10 12-mo.

9 *

3 Dyntissue]A[fmol/mg 2

0 Frontal cortex Hippocampus

Figure 2.6. Dyn A quantitation by RIA in mouse brain. Dynorphin A levels show significant age-dependent increases in frontal cortex but not hippocampus of WT mice. *p<0.02 relative to 3-mo. mice.

26 and were also significantly different from the 3-month controls [t(9) = 5.34, p<0.0005]. In hippocampus, Dyn A levels were relatively unchanged across all three age groups. Dyn A was also measured in females aged 12 months (n = 4) and was found to be present in the same amount as in 12-month males in both frontal cortex and hippocampus (data not shown).

Discussion The present study confirms a significant impairment with aging in the spatial water maze task (Gage et al., 1984; Rapp et al., 1987; Magnusson et al., 2003). The age-related impairment is clearly observable for the 13- to 16-month mice used in this study as demonstrated by the significant differences shown in the vast majority of the quantitative indicators of learning and memory. Although aging confers an approximately 15 to 20% decrease in swim speed in our study, motor activity by itself is unlikely to account for most of the age-associated decline in performance since speed-independent measurements, such as path length and target quadrant preference, also exhibit highly significant differences with age. Differences in general sensorimotor function, motivation to escape the pool, and visual acuity, all of which could potentially be affected by age, can be excluded because all groups performed equally well in the cued version of the water maze. In addition, swim stress-induced immobility, in which dynorphins have been implicated (Pliakas et al., 2001), cannot account for the variations in water maze performance since the trials do not reveal significant reductions in mean trajectory length in poorly performing animals. The data also suggest that the age-related impairment is partially alleviated by the absence of dynorphin. In most of the indices recorded, young mice show little or no Pdyn genotype differences, whereas among aged mice, the absence of dynorphin was associated with superior learning and memory scores. Compared to WT mice, aged knockout mice attained modestly greater proficiency in locating the hidden platform and appeared to employ spatial search strategies earlier and with greater accuracy than aged WT mice. The results imply that prodynorphin-derived peptides are contributing in part to age-related impairment in performance on a spatial memory task, but it is not clear

27 whether this effect is attributable to an effect of dynorphin on cognition or alternatively to differences in stress or reward responses that can influence cognitive outcome. It is possible that the differences in water maze performance may be related to a hypothesized role of dynorphin in mediating stress responses and the fact that aged animals are more susceptible to the effects of stress in disrupting performance in the water maze (Foster, 1999). Indeed, disruption of dynorphin transmission is associated with interference of stress-induced behavioral responses (McLaughlin et al., 2003). Moreover, treatments such as neonatal handling, which reduce neuroendocrine stress responses, are able to decrease dynorphin in the frontal cortex and facilitate performance on the water escape task in aged animals (Meaney et al., 1988; Ploj et al., 2001). Thus, a reduction in stress influences could explain why differences emerged with age and were more apparent early in training when stress effects would be expected to be greatest. Although markers of stress were not explicitly assayed in animals performing the water maze task, it seems plausible that stress itself or the effects of stress on cognition could account for some of the observed effects. A detrimental role of dynorphins in disrupting spatial learning is consistent with a number of studies (Jiang et al., 1989; McDaniel et al., 1990; Sandin et al., 1998; Hauser et al., 1999; Tan-No et al., 2001); however, other recent publications have concluded that a rise in dynorphin following brain injury may be beneficial for recovery following various insults (Heron et al., 1996; Redell et al., 2003). Interestingly, young Pdyn KO mice perform somewhat more poorly than young WT mice during later probe trials for reasons that are not immediately clear. On the final day of spatial testing, young Pdyn KO mice performed significantly worse than young WT mice in the probe trial quadrant preference assessment, despite appearing to have shown strong preference for the target quadrant on Days 1 and 2, and despite showing latency and path length measures quantitatively indistinguishable from young WT mice on all three days. The present data suggest that dynorphins may potentiate the extent of spatial learning at later time points. The ability of normal aging to unmask the cognitive functional differences between Pdyn KO and WT mice can be explained by the findings that dynorphin peptides in WT mice increases over 2-fold with aging in the present study. However, it is not

28 possible to conclude from the data that dynorphin is exerting a direct detrimental effect on cognition, since the poorer performance in aged WT mice could in theory be due to other mechanisms that independently lead to increased dynorphin levels. The significant increases in Dyn A observed in the frontal cortex of both 12- and 24-month mice suggest that abnormal expression of endogenous dynorphin peptides may at least partially contribute to spatial learning and memory with age but not in younger animals, in which endogenous dynorphins are present at substantially lower quantities. This hypothesis is consistent with earlier published data showing increased dynorphins in both frontal cortex and hippocampus of aged memory-impaired rats (Jiang et al., 1989) and a more recent study showing elevated Pdyn mRNA in aged rats (Kotz et al., 2004). In contrast to the above studies, our results in hippocampus do not show increased dynorphin expression with age. Interestingly, another immunohistochemical study reported a slight decrease in dynorphin immunoreactivity in age-impaired rats (Croll et al., 1999). However, the lack of an age-associated increase in hippocampal dynorphins does not preclude a possible harmful effect of these neuropeptides on hippocampal function with aging, since other age-related changes may alter susceptibility to endogenous dynorphins in the absence of overt changes in dynorphin expression. Nonetheless, the results in frontal cortex showing the marked age-related increase in dynorphin, in combination with the finding that aged prodynorphin knockout mice appear relatively spared in a spatial memory task, are compellingly suggestive of a more prominent role of frontal cortical dynorphin in learning and memory impairment. Despite the fact that the water maze task has long been regarded as a hippocampal-dependent task based on studies documenting spatial impairment upon selective lesioning of the hippocampus (Jarrard, 1978; Morris et al., 1982), a growing body of literature now supports the claim that spatial learning may also be critically dependent on other regions of the brain, such as frontal cortex (Owen et al., 1995; Kessels et al., 2000). In fact, in frontal cortex of Alzheimer’s disease or Down’s syndrome patients, dynorphin A is increased more than 2-fold compared to elderly controls (Risser et al., 1996). The increases in dynorphin immunoreactivity in aged mice could therefore reasonably underlie some of the cognitive deficits observed in water maze performance in aged WT mice. Further experiments, e.g., those involving κ-selective

29 agonists and antagonists, are needed, however, to determine causality and to elucidate the mechanistic basis of the apparent adverse effects of elevated endogenous dynorphins.

CHAPTER 3: Effects of aging and Pdyn deletion on nonspatial learning and memory and spontaneous locomotion in mice

Introduction As detailed in preceding chapters, aging is associated with neurobiological and neurochemical changes in the brain and is commonly associated with functional declines in neuronal function that lead to sensory, motor, cognitive, and numerous other types of deficits. The mechanisms by which the aging process contributes to these deficits have not been elucidated. Opiate neuropeptides derived from the prodynorphin (Pdyn) gene and expressed in hypothalamus, nucleus accumbens, neostriatum, dentate gyrus, hippocampus cerebral cortex, and spinal cord (Khachaturian et al., 1993), have been implicated in age-related cognitive dysfunction (Jiang et al., 1989; Zhang et al., 1991; Nguyen et al., 2005). Findings of increased dynorphin peptides in hippocampus of normal aged rats (Jiang et al., 1989; Zhang et al., 1991; Gallagher and Nicolle, 1993; Kotz et al., 2004), in cortex and hippocampus of amyloid precursor protein- overexpressing mice (Diez et al., 2000), in frontal cortex of Down’s syndrome patients (Risser et al., 1996), and in frontal and parietal cortex of Alzheimer’s disease patients (Risser et al., 1996; Yakovleva et al., 2006a) strongly implicate dynorphin dysregulation in age-related pathophysiological processes. The existing literature contains conflicting reports regarding the role of dynorphin peptides in learning and memory. While impairment in spatial learning and memory is seen upon intrahippocampal administration of Dyn A(1-8) (McDaniel et al., 1990) or dynorphin B (Sandin et al., 1998), some studies have reported that Dyn A(1-13) can ameliorate cognitive impairment induced by anticholinergic agents (Itoh et al., 1993a, b; Hiramatsu et al., 1996; Hiramatsu et al., 1997; Ukai et al., 1997) through mechanisms involving the opioid receptor, though recent studies have also reported non-opioid- dependent beneficial effects on memory (Hiramatsu and Watanabe, 2006). Dynorphin was also found to dose-dependently impair retention in an inhibitory avoidance task in mice (Introini-Collison et al., 1987). We have already shown that aging Pdyn knockout (KO) mice demonstrate improved performance in the water maze task (Nguyen, 2001).

We therefore hypothesize that Pdyn deletion similarly ameliorates age-related impairment in other forms of learning and memory. In the present study, we examine two additional learning and memory assays, the passive avoidance test and the paired object recognition test. To evaluate the possibility of differences in nociception potentially confounding interpretation of these studies, a tail-flick assay was performed on moderately aged Pdyn KO and WT mice. In addition, because the interpretation of these tests also depends on general exploratory activity in the animal subjects, we also examine open-field activity to examine effects of both aging and Pdyn deletion. Moreover, because of the clinical importance of the dopaminergic system in pathological motor dysfunction and given the reported dopamine-inhibitory effects of dynorphins and exogenous agonists in striatum and nucleus accumbens (Mulder et al., 1984; Herrera- Marschitz et al., 1986; Di Chiara and Imperato, 1988; Reid et al., 1988; Spanagel et al., 1992; Ronken et al., 1993; Schlosser et al., 1995), we also evaluated open-field activity following systemic administration of the neurotoxin 1-methyl-4-phenyl-1,2,3,6- tetrahydropyridine (MPTP). The resulting selective loss of dopaminergic neurons (Arai et al., 1990; Sundstrom et al., 1990) is expected to produce functional impairments in locomotor activity in these mice, which are hypothesized to be alleviated by Pdyn deletion.

Methods Animal subjects All animals in this study were wild-type (WT) or Pdyn KO mice that were maintained in an approved animal facility at the University of Kentucky. WT mice were derived from the 129SvEv-Tac line, and Pdyn KO mice were originally obtained by targeted deletion of the coding exons of the prodynorphin gene and bred into a 129SvEv-Tac background as described previously (Sharifi et al., 2001). Animals were housed in groups of up to 4 per cage with food and water ad libitum in a well-ventilated room with a 12 h: 12 h light: dark cycle. Prior to weaning, tail specimens were collected from each animal, and DNA was extracted for subsequent genotyping to confirm the identity of the Pdyn locus by the polymerase chain reaction (PCR) using primer pairs specific for each genotype. All

32 procedures involving animals are approved by the Institutional Animal Care and Use Committee at the University of Kentucky.

Inhibitory avoidance An inhibitory avoidance paradigm was used to evaluate passive recall of an aversive stimulus in male 3-month WT (n = 4), 3-month Pdyn KO (n = 4), 13-month WT (n = 8), 13-month Pdyn KO (n = 8), 24-month WT (n = 6), and 24-month Pdyn KO (n = 6) mice. The task tests the mouse subject's ability to employ situational recall to passively avoid entry into a dark chamber that is associated with an aversive electrical stimulus. A WinLinc software program (Coulbourn Instruments, Allentown, PA) was used to automate the sequence of events during each trial. In the acquisition trial, the subject is placed in the brightly lit environment of the first chamber for 90 seconds, followed by automated opening of a door leading to a dark chamber. If the mouse traverses the threshold into the dark chamber, an automated trigger is activated leading to closure of the door mechanism, and after a 10-second delay, a 3-second electrical shock at 0.1 mAmp, which was confirmed to produce visible but not excessive discomfort to the mouse, is applied via the conducting medium of the floor of the dark chamber. A retention trial was performed 72 hours after the acquisition trial in a similar manner, except that the door to the dark compartment remained open for the duration of the trial and no electrical shock was applied in the retention trials. In both trials, the step-through latency, defined as the time from door-opening until the animal crosses into the dark chamber, was recorded manually using a stopwatch to a maximum of 900 seconds.

Tail flick assay Tail withdrawal to a heat stimulus was assessed using a tail-flick assay in 16-month WT (n = 5) and Pdyn KO (n = 5) mice. The mouse was placed in a restraining device, and the tail was positioned within a groove of the heating mechanism. Trials were performed in triplicate to determine the latency from the time of initiation of heat application until the animal voluntarily withdraws its tail from the heating mechanism.

Open-field assay Spontaneous open-field behavioral activity was recorded using an apparatus of four activity chambers (Model RXYZCM-8, Accuscan Instruments, Columbus, OH), each consisting of a 41 × 41 × 31-cm Plexiglas box with two grids of infrared beams mounted at horizontal intervals of 2.5 cm and at a vertical interval of 4.5 cm. Monitors were connected to a Digiscan Analyzer (Model DCM-8, Accuscan Instruments) that converts the number and pattern of beam breaks to generate various data parameters, including horizontal activity, distance traveled, vertical activity, movement time, and stereotypy counts. During operation, the data were recorded separately for each of six contiguous 5- min intervals during each 30-minute trial. Horizontal activity was determined by the number of photobeam breaks in a horizontal grid, and vertical activity was quantified as number of photobeam interruptions in a more superiorly positioned grid. Distance traveled and movement time was also calculated by the Accuscan software. Stereotypy counts were detected as a set of repeated breaks in the same photobeam or array of photobeams. Movement speed was calculated as distance traveled divided by movement time to provide an estimate of speed during periods of active locomotion. Animals that yielded a movement time of zero were excluded from the movement speed analysis, as were two speed outliers that were unrealistically high due to aberrantly low movement times recorded. Mice used in this study include 3-month WT (n = 20), 3-month Pdyn KO (n = 30), 3-month Pdyn HET (n = 22), and 10-month WT (n = 6) mice. The 3-month WT group contains 10 males and 10 females, and the 3-month Pdyn KO group has 11 males and 19 females. The HET group consists of 11 males and 11 females. Among 10-month WT mice, there are 2 males and 4 females. The group sizes listed above include mice pooled from saline controls from the MPTP experiments that were tested prior to any intervention. In the experiments evaluating effects of MPTP, a regimen of four saline or 15 mg/kg MPTP intraperitoneal injections at 2-hour intervals was administered to mixed male and female 3-month mice, and behavioral testing performed 7 days later. The saline control group contains 2 WT, 4 HET, and 3 KO mice, while the MPTP group contains 3 WT, 4 HET, and 5 KO mice. These sample sizes reflect the removal of 4 saline controls due to failure of the automated recording system during a post-injection trial. For

34 analysis, logarithmic transformation of the post- to pre-injection ratios of the relevant behavioral parameters was applied to preserve linearity of the underlying distribution.

Object recognition A paired visual object recognition assay was used to test the animal subject's inclination to preferentially explore novel objects and to recognize familiar objects in an enclosed environment. The objects used in this experiment were non-noxious inanimate objects typically found in a laboratory setting and of a size similar to the mouse subject. Animals were habituated to the empty testing chamber for 15 minutes one day prior to testing. During each trial, two objects are positioned at predetermined locations equidistant from the site at which the mouse is to be placed, and the subject's spontaneous exploratory activity is monitored for 15 minutes either by real-time video transmission to an investigator located in a different room or by videotape recording. An investigator blinded to the identity of the mice recorded the total number of seconds during each trial the subject expends on active exploration of each object as demonstrated by specific behavior patterns such as facing the object in close proximity, moving from side to side while facing the object, placing one or more forelimbs on the object, or encircling the object in close proximity. Between trials, the testing chamber is cleaned to remove any odor cues that may influence subsequent exploratory activity. Results are expressed as a proportion of total exploration time spent on the novel object. To account for individual differences in baseline exploratory activity, total novel object exploration time is also presented as a ratio to total cumulative exploration time. Two different protocols were used in implementing the timing of the testing trials. In the Protocol 1, 6-month WT (n = 12), 6-month Pdyn KO (n = 3), 12-month WT (n = 8), 12-month KO (n = 8), 24-month WT (n = 5), and 24-month KO (n = 6) male mice were monitored in three acquisition trials before new target objects are successively introduced in two retention trials 4 and 24 hours later. In Protocol 2, testing consists of an acquisition trial, a retention trial immediately afterward, and a second retention trial with a different novel object 4 hours after the start of the first trial. Animal subjects in Protocol 2 include 4-month WT (n = 3), 14-month WT (n = 7), 14-month Pdyn KO (n = 10), 24-month WT (n = 5), and 24- month KO (n = 5) male mice. These protocols differ in that Protocol 1 permits a greater

35 amount of time to be spent in the acquisition phase prior to introduction of the first novel object, which is thought to enhance the distinction between novel and familiar objects in subsequent trials. The shorter acquisition phase of Protocol 2, on the other hand, may enable more subtle differences in object recognition to become manifest.

Results Inhibitory avoidance An inhibitory avoidance assay was performed on WT and Pdyn KO mice of varying ages (Figure 3.1). Acquisition step-through latencies were recorded in experimentally naïve

# Acquisition 1000 †† Retention 900 †† 800 †† 700 †† * 600 ‡ 500 400 ### 300 200 100 Step-through latency(sec)Step-through *** 0 *** 3-mo. WT 3-mo. Pdyn 13-mo WT 13-mo. Pdyn 24-mo. WT 24-mo. Pdyn KO KO KO Figure 3.1. Inhibitory avoidance assay results in mice. Effects of age and Pdyn deletion on step-through latency before and 72 hours after an electrical foot-shock stimulus were measured in the passive avoidance task. Significant learning effects are seen independently of age and Pdyn genotype. Acquisition step- through latencies, which may reflect exploratory activity in naïve animals, are significantly increased in Pdyn KO mice, but only at 3 months. Retention latencies in the 3-month Pdyn KO group are also increased, consistent with improved recall of the aversive stimulus. Note the ceiling effect in which all four young Pdyn KO mice achieved the predetermined maximum latency of 900 sec. *p<0.05, ***p<0.001 compared to respective 3-month controls; #p<0.05, ###p<0.001 compared to corresponding WT mice; ‡p<0.01, ††p<0.001 compared to the corresponding acquisition trial.

3.0

2.5

2.0

1.5

1.0

0.5 Tail-flick latency (seconds) latency Tail-flick 0.0 WT Pdyn KO Figure 3.2. Tail-flick latencies as a function of Pdyn genotype. This assay demonstrates no effect of Pdyn deletion on nociception or subsequent reflex withdrawal to painful stimuli. mice, while retention latencies at 72 hours were obtained to indicate the subject's ability to demonstrate situational recall of the aversive stimulus. Three-way ANOVA with repeated measures was used to evaluate age, Pdyn, and training effects. An overall training effect was highly significant [F(1, 30) = 64.08, p<0.0005], and an age-Pdyn interaction was also significant [F(2, 30) = 3.93, p<0.03]. Acquisition scores showed a significant effect of age [F(2, 30) = 6.75, p<0.004], Pdyn [F(1, 30) = 6.12, p<0.019], and age-Pdyn interaction [F(2, 30) = 5.94, p<0.007]. The 3-month Pdyn KO mice latencies exhibit a ceiling effect due to all four mice in the group achieving the predetermined maximum latency of 900 sec. With the exception of 3-month Pdyn KO mice showing only a mild tendency toward a significant learning effect [F(1, 30) = 3.29, p<0.08] as a result of the elevated pre-stimulus step- through latency, step-through latencies 72 hours after the initial stimulus were consistently increased in all other groups, including 3-month Pdyn KO [F(1, 30) = 14.21, p<0.001], 13-month WT [F(1, 30) = 12.79, p<0.001], 13-month Pdyn KO [F(1, 30) = 15.53, p<0.0005], 24-month WT [F(1, 30) = 17.99, p<0.0005], and 24-month Pdyn KO mice [F(1, 30) = 7.93, p<0.009]. Among 3-month mice, Pdyn KO mice show significantly higher acquisition [F(1, 30) = 14.62, p<0.001] and retention [F(1, 30) = 4.51, p<0.042] scores. The 3-month Pdyn KO retention latency was also significantly

37 elevated when compared to that of 24-month Pdyn KO mice [F(1, 30) = 4.21, p<0.049], and the acquisition latency means were similarly higher than those of both 13-month Pdyn KO [F(1, 30) = 19.29, p<0.0005] and 24-month Pdyn KO mice [F(1, 30) = 21.70, p<0.0005]. Beyond the findings associated with the 3-month Pdyn KO mice, no clear age-associated trend in retention latencies were observed in this passive avoidance task at 72 hours. Because interpretation of the passive avoidance task as an indication of memory function assumes the subjects exhibit comparable perception of and response to the inciting stimulus, potential effects of dynorphin on pain modulation or perception were evaluated by the tail-flick assay. Tail-flick response measurements show no differences between Pdyn KO and WT mice (Figure 3.2).

Open-field analysis Spontaneous exploratory and locomotor behavior patterns were analyzed by open-field analysis in young adult (3-month) WT, Pdyn KO, and HET mice, with automated

1200 3500 3-mo. WT 3000 1000 3-mo. Pdyn HET 2500 3-mo. Pdyn KO 800 2000

600 1500

Horizontal activity Horizontal activity activity Horizontal Horizontal Horizontal activity Horizontal 1000 400 500 200 0

0 1 2 3 4 5 6 3-mo. WT

Recording Interval 3-mo. Pdyn HET3-mo. Pdyn KO

Figure 3.3. Horizontal activity measurements during open-field analysis as a function of Pdyn genotype. Left: The vertical axis depicts the number of interruptions in the horizontal photobeam grid of the activity chamber. Intra-trial changes in horizontal activity show a gradual habituation effect across recording intervals. Right: Summed totals of horizontal activity show no effect of Pdyn genotype.

recording of a series of measured variables during a trial consisting of 6 contiguous 5- minute intervals (Figure 3.3 through Figure 3.8). Two-factor repeated measures ANOVA shows a significant intra-trial habituation effect as manifested in a significant decline across intervals in horizontal activity [F(5, 345) = 36.44, p<0.0005], distance traveled [F(5, 345) = 36.58, p<0.0005], movement time [F(5, 345) = 29.76, p<0.0005], stereotypy [F(5, 345) = 14.63, p<0.0005], and movement speed [F(5, 165) = 3.90, p<0.005]. The one variable analyzed that did not show such an effect was vertical activity, which also showed substantial intra-group variability likely related to limited sample size (Figure 3.6). No significant effects of Pdyn or Pdyn-habituation interactions were observed for any of these variables. One-way ANOVA performed on the average trial movement speed and the summed intra-trial total for all the other variables found no significant effects of Pdyn deletion. To determine effects of aging on these parameters, a similar analysis was applied in 3-month and 10-month WT mice (Figure 3.9 through Figure 3.14). Similar to the preceding ANOVA results, habituation effects were again seen in horizontal activity

800 2500 3-mo. WT 700 3-mo. Pdyn HET 2000 600 3-mo. Pdyn KO 500 1500

400

1000

Distance (cm) Distance (cm) (cm) Distance Distance Distance (cm) Distance 300 500 200

100 0

0 1 2 3 4 5 6 3-mo. WT

Recording Interval 3-mo. Pdyn HET3-mo. Pdyn KO

Figure 3.4. Distance measurements during open-field analysis as a function of Pdyn genotype. Left: Intra-trial changes in distance traveled show a gradual habituation effect across recording intervals. Right: Summed totals of distance measurements show no effect of Pdyn genotype.

[F(5, 120) = 14.18, p<0.0005], distance traveled [F(5, 120) = 19.39, p<0.0005], movement time [F(5, 120) = 15.83, p<0.0005], and stereotypy [F(5, 120) = 13.74, p<0.0005]. Additionally, vertical activity showed a significant habituation effect in this analysis [F(5, 120) = 2.29, p<0.05], while movement speed showed no significant intra- trial changes. Increased stereotypy counts were observed in older mice [F(1, 24) = 8.95, p<0.006] with an age-interval interaction [F(5, 120) = 3.81, p<0.003]. Total stereotypy activity during the 30-minute trial was significantly higher in 10-month WT mice compared to 3-month WT mice [F(1, 24) = 8.95, p<0.006]. Only a minor age-associated decline in average movement speed was observed with aging [F(1, 24) = 5.28, p<0.031], with a significant intra-interval age effect noted only near the end of the trial [F(1, 9) = 7.66, p<0.022]. No other age effects or age-interval interactions were seen in the other variables analyzed.

80 250 3-mo. WT 70 3-mo. Pdyn HET 200 60 3-mo. Pdyn KO 50 150

40 100

Movement time (sec) time Movement (sec)(sec) time time Movement Movement Movement time (sec) time Movement 50 20

10 0

0 1 2 3 4 5 6 3-mo. WT

Recording Interval 3-mo. Pdyn HET3-mo. Pdyn KO

Figure 3.5. Movement time during open-field analysis as a function of Pdyn genotype. Left: The vertical axis depicts the amount of time spent in active locomotion. Intra-trial changes in movement time show a gradual habituation effect across recording intervals. Right: Summed totals of movement time show no effect of Pdyn genotype.

8 3-mo. WT 30 3-mo. Pdyn HET 7 25 3-mo. Pdyn KO 6 20 5 15 4

Vertical activity Vertical activity activity Vertical Vertical Vertical activity Vertical 3

2 5

1 0

0 1 2 3 4 5 6 3-mo. WT

Recording Interval 3-mo. Pdyn HET3-mo. Pdyn KO

Figure 3.6. Vertical activity during open-field analysis as a function of Pdyn genotype. Left: The vertical axis depicts the number of occurrences of breaks in the more superiorly placed photobeams. Vertical activity shows minimal effects across recording intervals. Right: Summed totals of vertical activity show no effect of Pdyn genotype.

200 500

180 3-mo. WT 450 400 160 3-mo. Pdyn HET 350 140 3-mo. Pdyn KO 300 120 250 100 200

Stereotypy Stereotypy Stereotypy Stereotypy 150 60 100 40 50 20 0 0 1 2 3 4 5 6 3-mo. WT

Recording Interval 3-mo. Pdyn HET3-mo. Pdyn KO

Figure 3.7. Stereotypy during open-field analysis as a function of Pdyn genotype. Left: The vertical axis depicts the number of occurrences of stereotyped behavior as recorded by repeated breaks in the same photobeam or array of photobeams. Intra-trial changes in stereotypy show a gradual habituation effect across recording intervals. Right: Summed totals of stereotypy events show no effect of Pdyn genotype.

18 14

16 12

14 10 12 8 10 6

Speed (cm/sec) Speed (cm/sec)(cm/sec) Speed Speed Speed (cm/sec) Speed 4 6 3-mo. WT

4 3-mo. Pdyn HET 2 3-mo. Pdyn KO 2 0

0 1 2 3 4 5 6 3-mo. WT

Recording Interval 3-mo. Pdyn HET3-mo. Pdyn KO

Figure 3.8. Movement speed during open-field analysis as a function of Pdyn genotype. Left: Movement speed is minimally altered across recording intervals. Right: Average movement speed shows no effect of Pdyn genotype.

As the above results were performed on groups containing both male and female mice, a subsequent analysis was performed to assess effects of gender on spontaneous open-field activity (Figure 3.15 through Figure 3.20). ANOVA on summed data from the entire 30-minute trial period reveals significant main effects of gender on movement time [F(1, 64) = 6.92, p<0.011] and movement speed [F(1, 64)=16.28], p<0.0005], with gender-Pdyn interactions observed in horizontal activity [F(2, 64) = 4.14, p<0.02], distance traveled [F(2, 64) = 5.85, p<0.005], and movement time [F(2, 64) = 6.19, p<0.003]. Extending a similar approach to the aging comparisons in WT mice finds sex effects in horizontal activity [F(1, 22) = 10.08, p<0.004], distance traveled [F(1, 22) – 24.97, p<0.0005], movement time [F(1, 22) = 27.82, p<0.0005], and vertical activity [F(1, 22) = 6.41, p<0.019]. Significant interactions between gender and Pdyn genotype were noted in recorded distance [F(1, 22) = 11.45, p<0.003] and movement time [F(1, 22) = 10.46, p<0.004].

1400 4000

3500 1200 3-mo. WT 3000 1000 10-mo. WT 2500 800 2000

600

Horizontal activity Horizontal activity activity Horizontal Horizontal Horizontal activity Horizontal 1500

400 1000

200 500

0 0 1 2 3 4 5 6

3-mo. WT Recording Interval 10-mo. WT

Figure 3 . 9. Horizontal activity measurements during open-field analysis as a function of aging.

Left: The vertical axis depicts the number of interruptions in the horizontal photobeam grid of the activity chamber. Habituation effects are seen across recording intervals, with minimal aging effect. Right: Total horizontal activity is unchanged with age.

1400 4000

3500 1200 3-mo. WT 3000 1000 10-mo. WT 2500 800 2000

600

Distance (cm) Distance (cm) (cm) Distance Distance Distance (cm) Distance 1500

400 1000

200 500

0 0 1 2 3 4 5 6

3-mo. WT Recording Interval 10-mo. WT Figure 3.10. Distance traveled during open-field analysis as a function of aging. Left: Habituation effects are seen across recording intervals, with minimal aging effect. Right: Total distance traveled is unchanged with age.

140 400

350 120 3-mo. WT 300 100 10-mo. WT 250 80 200 60

150

Movement time (sec)(sec) time time Movement Movement (sec)(sec) time time Movement Movement 40 100

20 50

0 0 1 2 3 4 5 6

3-mo. WT Recording Interval 10-mo. WT Figure 3.11. Movement time during open-field analysis as a function of aging. Left: The vertical axis depicts the amount of time spent in active locomotion. Habituation effects are seen across recording intervals, with minimal aging effect. Right: Total movement time is unchanged with age.

10 35 3-mo. WT 9 30 8 10-mo. WT 7 25

6 20 5 15

Vertical activity activity Vertical Vertical activity activity Vertical Vertical

3 10 2 5 1 0 0 1 2 3 4 5 6

3-mo. WT Recording Interval 10-mo. WT

Figure 3.12. Vertical activity during open-field analysis as a function of aging. Left: The vertical axis depicts the number of occurrences of breaks in the more superiorly placed photobeams. Minimal habituation effects are seen across recording intervals. Right: Total vertical activity is unchanged with age.

600 1400

3-mo. WT 1200 500 ** **

10-mo. WT 1000 400 800 300

600

Stereotypy Stereotypy Stereotypy Stereotypy * 200 400

100 200

0 0 1 2 3 4 5 6

3-mo. WT Recording Interval 10-mo. WT Figure 3.13. Stereotypy during open-field analysis as a function of aging. Left: The vertical axis depicts the number of occurrences of stereotyped behavior as recorded by repeated breaks in the same photobeam or array of photobeams. Habituation effects are seen across recording intervals, with significantly increased stereotypy seen in 10-month mice. Right: Total stereotypy counts are increased with age. *p<0.05, **p<0.01 compared to 3-month mice. 16 12

14 10 * 12 8 10

8 * 6

Speed (cm/sec) Speed (cm/sec)(cm/sec) Speed Speed Speed (cm/sec) Speed 6 4 4 3-mo. WT 2 2 10-mo. WT

0 0 1 2 3 4 5 6

3-mo. WT Recording Interval 10-mo. WT Figure 3.14. Movement speed during open-field analysis as a function of aging. Left: Movement speed is decreased in 10-month mice across recording intervals. Right: Mean speeds are mildly decreased with age. *p<0.05 compared to 3-month mice.

†

‡

Horizontal activity activity Horizontal Horizontal

Figure 3.15. Gender analysis of horizontal activity in open-field task. Left: The vertical axis depicts the number of interruptions in the horizontal photobeam grid of the activity chamber. Aging analysis shows increased horizontal activity in 10- month male mice compared to 10-month females. Right: Pdyn analysis shows male Pdyn HET mice exhibit less activity than female HET mice or male Pdyn KO mice. †p<0.05, ‡p<0.01 compared to corresponding male mice. #p<0.05 Pdyn genotype effect.

Horizontal activity in 3-month mice (Figure 3.15) was mildly decreased in male Pdyn HET mice compared to either female Pdyn HET mice [F(1, 64) = 4.68, p<0.034] or male Pdyn KO mice [F(1, 64) = 5.17, p<0.026]. Female 10-month WT mice also showed less horizontal activity than male 10-month WT mice [F(1, 22) = 8.58, p<0.008], but a similar gender difference among 3-month WT mice failed to reach significance [F(1, 22) = 4.15, p<0.054]. Similar findings were noted in distance (Figure 3.16) and movement time (Figure 3.17), with 10-month male WT mice accumulating more distance [F(1,22) = 17.02, p<0.0005] and spending more time in motion [F(1, 22) = 18.62, p<0.0005] than 3- month male WT mice. Ten-month female WT mice showed comparatively less distance traveled [F(1, 22) = 22.25, p<0.0005] and movement time [F(1, 22) = 22.92, p<0.0005] than their male WT counterparts. Movement time was also significantly less in female 3- month mice than males [F(1, 22) = 4.95, p<0.037]. Distance traveled by male Pdyn KO mice were significantly higher than for male Pdyn HET mice [F(1, 64) = 10.73, p<0.002]

## ***

‡

††

Distance (cm) (cm) Distance Distance

Figure 3.16. Gender analysis of distance traveled in open-field task. Left: Aging analysis shows significantly increased distance traveled in 10-month males. Right: Pdyn analysis shows male Pdyn KO mice exhibit greater traversed distance than male HET mice or female Pdyn KO mice. ***p<0.001 compared to corresponding 3- month mice. ‡p<0.01, ††p<0.001 compared to corresponding male mice. ##p<0.01 Pdyn genotype effect.

or female Pdyn KO mice [F(1, 64) = 10.29, p<0.002]. Similarly time in motion was significantly higher in Pdyn KO males than Pdyn HET males [F(1, 64) = 14.15, p<0.0005] or Pdyn KO females [F(1, 64) = 16.44, p<0.0005]. No significant contrasts were found in vertical activity (Figure 3.18). Stereotypy assessments were not significantly different among 3-month mice (Figure 3.19), but increased stereotypy behavior was noted at 10 months age in both male [F(1, 22) = 4.57, p<0.044] and female [F(1, 22) = 5.37, p<0.03] WT mice. Movement speed calculations (Figure 3.20) were consistently higher in females than males for WT [F(1, 64) = 5.40, p<0.023], Pdyn HET [F(1, 64) = 6.02, p<0.017], and Pdyn KO mice [F(1, 64) = 4.90, p<0.03]. In the aging analysis, 3-month female WT showed significantly higher movement speed than 3-month male WT [F(1, 22) = 8.62, p<0.008] or 10-month female WT mice [F(1, 22) = 12.55, p<0.002].

### ***

†† ††

†

Movement time (sec)(sec) time time Movement Movement (sec)(sec) time time Movement Movement

Figure 3.17. Gender analysis of movement time in open-field task. Left: The vertical axis depicts the amount of time spent in active locomotion. Aging analysis shows decreased activity in female mice compared to males at each age point, with 10-month males showing significantly increased time in active motion. Right: Pdyn analysis shows male Pdyn KO mice exhibit greater movement time than male HET mice or female Pdyn KO mice. ***p<0.001 compared to corresponding 3-month mice. ††p<0.001 compared to corresponding male mice. ###p<0.001 Pdyn genotype effect.

Effects of systemic MPTP or saline on open-field behavior in 3-month mice were also examined by calculating intrasubject log-ratios of post-injection to pre-injection values of each analyzed variable. ANOVA found an MPTP effect on average speed [F(1, 15) = 6.97, p<0.019] and significant Pdyn-MPTP interactions in movement time [F(2, 15) = 4.15, p<0.037] and movement speed [F(2, 15) = 5.73, p<0.014]. While most mice showed mild habituation-related decreases in horizontal activity, distance, and movement times upon re-exposure to the open-field environment (Figure 3.21), WT mice show substantial post-MPTP reduction in exploratory activity, with statistical significance demonstrated in the amount of time spent in active locomotion [F(1, 15) = 7.23, p<0.017]. WT mice also exhibit a greater post-MPTP decline in traversed distance compared to Pdyn KO [F(1, 15) = 5.15, p<0.038] and Pdyn HET mice [F(1, 15) = 6.32, p<0.024]. Time spent actively moving was also reduced to a greater extent in post-MPTP

WT mice compared to post-MPTP Pdyn KO [F(1, 15) = 9.90, p<0.007] or HET mice [F(1, 15) = 12.89, p<0.003]. Stereotypy assessments revealed mild differences (Figure 3.22), primarily in the saline-treated Pdyn HET group, which demosntrated significantly less stereotypic behavior in comparison to either WT saline controls [F(1, 15) = 4.99, p<0.041] or MPTP-treated HET mice [F(1, 15) = 4.97, p<0.041]. Estimates of average movement speed were mildly increased in the post-injection trial for most groups (Figure 3.22), but MPTP-injected WT mice demonstrate a significantly larger increase in speed compared to saline-injected WT mice [F(1, 15) = 13.94, p<0.002], MPTP-injected Pdyn KO mice [F(1, 15) = 9.47, p<0.008] and to MPTP-injected HET mice [F(1, 15) = 14.74, p<0.002], although this apparent increase in the calculated average speed of MPTP- injected WT mice is likely artifactual due to the markedly lower amount of time spent in

active locomotion.

Vertical activity activity Vertical Vertical activity activity Vertical Vertical

Figure 3.18. Gender analysis of vertical activity in open-field task. Left: The vertical axis depicts the number of occurrences of breaks in the more superiorly placed photobeams. Aging analysis shows similar trends as other variables but with substantial intra-group variability. No significant effects are seen. Right: Pdyn analysis shows no significant effects in vertical activity.

* *

Stereotypy Stereotypy

Figure 3.19. Gender analysis of stereotypy in open-field task. Left: The vertical axis depicts the number of occurrences of stereotyped behavior as recorded by repeated breaks in the same photobeam or array of photobeams. Aging analysis shows increased stereotypy with age in both male and female mice. Right: Pdyn analysis shows no significant effects. *p<0.05 relative to corresponding 3-month mice.

‡ † † †

Speed (cm/sec) (cm/sec) Speed Speed (cm/sec) (cm/sec) Speed Speed

Figure 3.20. Gender analysis of movement speed in open-field task. Left: Aging analysis shows mild increase in movement speed of female mice at 3 months but a decrease at 10 months. Right: Pdyn analysis shows consistently elevated movement speed in females of all three Pdyn genotypes. **p<0.01 compared to † ‡ corresponding 3-month mice. p<0.05, p<0.01 compared to corresponding male mice.

Horizontal activity Distance Movement time

0.0 0.2 # 0.2 ## -0.1 0.1 0.0 0.0

-0.2

injection) injection) injection) injection) injection)

injection) -0.1

------0.2 -0.3 -0.2 ## # -0.4 -0.4 -0.3 WT saline WT saline -0.4 WT saline -0.5 -0.6 HET saline HET saline -0.5 HET saline

-0.6

injection / pre / injection pre pre / / injection injection pre pre / / injection injection

injection / pre / injection KO saline KO saline -0.8 KO saline

------0.6 -0.7 WT MPTP -0.7 WT MPTP WT MPTP -1.0 -0.8 HET MPTP -0.8 HET MPTP † HET MPTP

-0.9 KO MPTP -0.9 KO MPTP -1.2 KO MPTP

Log Log Log (post (post Log Log (post (post Log Log (post (post

Figure 3.21. Effects of MPTP and Pdyn deletion on locomotor activity. Open-field data were obtained from 3-month mice before and 7 days after intraperitoneal injection of saline or MPTP. Within-subject log-ratios of post-injection to pre-injection horizontal activity, distance, and movement time are shown. WT mice show substantial decreases in post-MPTP locomotor indices, while Pdyn KO and HET mice are minimally changed. #p<0.05, ##p<0.01 compared to corresponding WT mice. †p<0.05 compared to saline controls. Stereotypy Speed

0.6 0.3 WT saline HET saline 0.4 ‡ KO saline

0.2 †

injection) injection) injection) injection)

- - - - 0.2 WT MPTP 0.0 HET MPTP -0.2 ## KO MPTP 0.1 WT saline -0.4 HET saline

-0.6

injection / pre pre / / injection injection pre pre / / injection injection

- - - - KO saline 0.0 -0.8 WT MPTP ##

-1.0 HET MPTP

Log Log (post Log Log (post (post Log Log (post -1.2 # KO MPTP -0.1 Figure 3.22. Effects of MPTP and Pdyn deletion on stereotyped behavior and motor speed. Open-field data were obtained from 3-month mice before and 7 days after intraperitoneal injection of saline or MPTP. Within-subject log-ratios of post-injection to pre-injection stereotypy counts and movement speed are shown. #p<0.05, ##p<0.01 compared to corresponding WT mice.

Object recognition assay Performance in a paired object recognition task was evaluated by measuring the amount of time spent by the mouse exploring each of two different objects during a series of 5- minute trials. Two sets of experiments were performed using slightly different protocols. In the first set of experiments, animals were monitored in three habituation trials before new target objects are introduced in retention trials 4 and 24 hours later (Figure 3.23). The proportion of time spent exploring the new target object was recorded as ratios to total intra-trial exploratory time and to total cumulative exploratory time. While the former ratio represents the relative changes in object exploration time during each trial, the latter ratio was selected to provide an indication of absolute time spent on the novel object while controlling for baseline differences in general exploratory activity. The proportion of time in each trial spent exploring the novel object is found by ANOVA to demonstrate significant effects of Pdyn genotype [F(1, 31) = 4.80, p<0.036], trial number

Figure 3.23. Effects of age and Pdyn deletion on object recognition using Protocol 1. Left: Novel object preference is expressed as the fraction of total trial exploration time spent interacting with the novel object or, in the case of Trials 1-3, the object in the position where the novel object of the subsequent trial is to be placed. Aged Pdyn KO mice spend a decreased proportion of exploration time interacting with the novel object in the 24-hour retention trial (Trial 5) compared to similarly aged WT mice. Right: Total time spent on the novel object is expressed relative to total cumulative object exploration time to control for individual differences in baseline exploratory tendencies. Total time spent on the novel object relative to cumulative exploratory activity is increased in 24- month Pdyn KO mice. #p<0.05, ##p<0.01 compared to corresponding WT mice. *p<0.05 compared to corresponding 6-month mice. **p<0.01 compared to corresponding 12-month mice.

[F(2, 62) = 9.36, p<0.0005], and a trial-Pdyn interaction [F(2, 62) = 4.30, p<0.029]. The almost equal division in the amount of time spent on each of the two objects during these acquisition trials indicates that the mice show no baseline preference for either object location. While there were no significant aging contrasts in this analysis, significantly decreased novel object exploration at 24 hours was noted in Pdyn KO mice aged 12 months [F(1, 31) = 7.64, p<0.01] or 24 months [F(1, 31) = 5.10, p<0.031]. When novel object exploration is expressed relative to cumulative exploratory activity, ANOVA reveals significant trial effects [F(1, 36) = 16.09, p<0.0005], a trial-Pdyn interaction [F(1, 36) = 5.19, p<0.029], and a trial-age interaction [F(2, 36) = 3.58, p<0.038]. Significant contrasts are seen only at the 4-hour retention trial, in which 24-month Pdyn KO mice show elevated novel-object exploratory activity, which was significantly different from 24-month WT [F(1, 36) = 5.12, p<0.03], 6-month Pdyn KO [F(1, 36) = 6.18, p<0.018], and 12-month Pdyn KO mice [F(1, 36) = 7.60, p<0.009]. Total exploratory times were also recorded during each trial and analyzed by ANOVA (Figure 3.24), with significant Pdyn effects [F(1, 36) = 4.62, p<0.038], trial effects [F(4, 144) = 13.92, p<0.0005], and trial-Pdyn-age interactions [F(8, 144) = 2.49, p<0.015]. In the first trial, exploration time in young adult Pdyn KO mice is markedly elevated, resulting in significant differences in comparison to 6-month WT [F(1, 36) = 8.77, p<0.005] and 24- month Pdyn KO mice [F(1, 36) = 5.46, p<0.025]. In contrast, total exploratory time in Trial 1 is increased at 24 months in WT mice [F(1, 36) = 7.05, p<0.012]. In Trials 2 and 3, exploration times in most groups show progressive decline due to habituation effects, but 6-month Pdyn KO mice continued to exhibit increased exploratory behavior that is statistically significant compared to 6-month WT [F(1, 36) = 20.86, p<0.0005], 12-month Pdyn KO [F(1, 36) = 10.90, p<0.002], and 24-month Pdyn KO mice [F(1, 36) = 19.20, p<0.0005]. In Trials 4 and 5, with addition of the novel stimulus, most groups show increased or unchanged exploration times; however, the only significant inter-group difference was found in 24-month Pdyn KO mice, which showed increased total exploratory activity in Trial 4 compared to 12-month Pdyn KO mice [F(1, 36) = 5.60, p<0.023]. In the second object recognition protocol, only three trials were used, with Trials 2 and 3 serving as retention trials at 1 and 4 hours, respectively (Figure 3.25). Young

Figure 3.24. Effects of age and Pdyn deletion on total exploration time during object recognition assay using Protocol 1. Pdyn KO mice demonstrate greater exploratory activity that WT mice, especially among young adult mice. *p<0.05, **p<0.01, ***p<0.001 compared to corresponding 6-month mice, or where indicated, to corresponding 12-month mice. ##p<0.01, ###p<0.001 compared to corresponding WT mice.

WT mice appear to show decreased baseline preference for one of the objects during the acquisition trial, resulting in a significant age effect compared to 14-month [F(1, 12 = 7.86, p<0.016] and 24-month WT mice [F(1, 12) = 12.17, p<0.004]. In Trial 2, the proportion of exploratory time on the target object is increased in all groups without any between-group differences. However, in Trial 3, the 4-month WT group shows a diminished response to the novel stimulus, resulting in a statistically significant difference compared to 14-month WT mice [F(1, 12) = 7.87, p<0.016]. Additionally, 24- month Pdyn KO mice in Trial 3 also show diminished novel object exploratory activity, with findings of statistical significance in comparison to 14-month Pdyn KO mice [F(1, 23) = 4.425, p<0.047]. Assessment of ratios of novel object exploration times to cumulative exploration time under Protocol 2 revealed no significant effects of either Pdyn deletion or age.

Discussion The above results show that multiple behavioral endpoints are affected by Pdyn deletion with some effects showing significant age dependence. First, in the inhibitory avoidance assay, which applies Pavlovian fear conditioning in an operative learning paradigm, Pdyn KO and WT mice of varying ages were assessed to determine whether the animal subject's motivation to exit a brightly lit chamber exceeds the adverse reaction to a negatively reinforcing stimulus. Our results show that young adult mice lacking Pdyn exhibit increased step-through latencies during both acquisition and retention passive avoidance trials. The increased retention step-through latency in Pdyn KO mice may signify relatively improved 72-hour recall of the aversive stimulus compared to WT counterparts. Since none of the mice in the 3-month Pdyn KO group traversed into the dark chamber in the retention trial, the ceiling effect imposed by the 15-mnute maximum likely underestimates the magnitude of this Pdyn genotype effect, and it is plausible that longer retention intervals may elicit differences in maximal recall interval.

Figure 3.25. Effects of age and Pdyn deletion on object recognition using Protocol 2. Left: Novel object preference is expressed as a fraction of total object exploration time spent interacting with the novel object or, in the case of Trial 1, the object in the position where the novel object of the subsequent trial is to be placed. An age-related decline in novel object interaction is seen in Pdyn KO mice in Trial 3. Right: Total time spent on the novel object is expressed relative to total cumulative object exploration time to control for individual differences in baseline exploratory tendencies. *p<0.05, **p<0.01 compared to corresponding 4-month mice, or where indicated, to corresponding 14- month mice.

Dynorphins have been found to dose-dependently impair retention in an inhibitory avoidance task in mice (Introini-Collison et al., 1987). It is plausible that unknown compensatory effects in Pdyn KO mice may render such effects less evident in the present study. However, other studies using the passive avoidance task have found Dyn A, Dyn B, and Dyn-32 to facilitate retention (Kuzmin et al., 2006) or to prevent impairment induced by other pharmacologic agents (Hiramatsu et al., 1996; Ukai et al., 1997). The significance of increased acquisition latency in 3-month Pdyn KO mice is unclear, given the lack of similar findings in other age groups in this study. One interpretation is that Pdyn KO mice used in the assay have less basal exploratory behavior, which may explain the increased step-through latencies both before and after the negatively reinforcing stimulus. However, the open-field data do not support this assertion, since young male Pdyn KO mice showed mildly increased basal exploratory behavior. Alternatively a delayed latency may also suggest increased anxiety in Pdyn KO mice, as agonists have been reported to show anxiolytic properties in rodents (Privette and Terrian, 1995). While the passive avoidance task is often interpreted in the context of operant learning, the role of opiates in pain processing pathways may potentially undermine the internal validity of studies assuming uniform nociceptive response to adverse stimuli. For instance, intrathecal Dyn B and Dyn-32 can induce nociceptive behavior (Tan-No et al., 2002). However, infusion of exogenous Dyn A and B into cerebrospinal fluid increases nociceptive latencies in a hot-plate assay (Kuzmin et al., 2006), indicating potential analgesic effects of these peptides. The absence of a Pdyn deletion effect on the tail flick assay in the present study decreases the likelihood of altered perception of painful stimuli influencing the animals' performance in the passive avoidance task. This absent pronociceptive effect of dynorphins has been replicated by other investigators using Pdyn KO mice (Wang et al., 2001). Open-field activity was used in this study to measure spontaneous locomotion as a function of age and Pdyn genotype. While our results show a reproducible intra-trial habituation effect, significant Pdyn effects were observed only when analyzing groups subclassified by gender. Male young adult Pdyn KO mice show mild increases in markers of baseline exploratory activity, with significant gender differences manifested

56 primarily among KO mice. An inhibitory action of dynorphins on exploratory behavior has been previously documented in rats (Tilson et al., 1986). While the existing literature contains few studies on a gender interaction with Pdyn, anti-estrogen agents have been noted to enhance Pdyn expression (Spampinato et al., 1995). Females may be more responsive to dynorphins, since at least one human study found that basal prolactin levels and dynorphin-induced prolactin release are significantly higher in females (Kreek et al., 1999). Since prolactin release is under tonic inhibition by dopamine, these findings suggest a potential mechanism whereby gender effects contribute to differences in motor activity. Interestingly, female mice consistently demonstrated increased movement speed across all three Pdyn genotypes in this study. Aging in this study yielded a significant but mild decrease in motor speed and an increase in stereotyped behavior. This is consistent with previous studies that have shown age-associated decreases in open-field locomotor activity and increases in susceptibility to stimulant-induced stereotypic motor behaviors in rats (Dorce and Palermo-Neto, 1994). Interestingly, aging male mice show increases in multiple indicators of exploratory behavior that are not evident in aged females. Some of these measures may be attributable to anxiety, based on associated increases in stereotypy counts. Open-field indicators of locomotor activity are significantly reduced in WT mice systemically injected with MPTP, while MPTP-injected Pdyn KO and HET mice show no substantial declines in motor activity beyond the habituation effect that is also seen in saline controls. The hypokinetic effect of MPTP, which induces dopaminergic cell loss as a model of parkinsonism (Heikkila et al., 1984), is similar to that observed previously in C57BL mice (Sedelis et al., 2001; Vijitruth et al., 2006), other mouse strains (Sedelis et al., 2001), or rats (Prediger et al., 2006). Our findings that Pdyn deletion produces a relative sparing of MPTP-induced motor dysfunction may be explained by inhibitory effects of dynorphin peptides on the striatal dopaminergic system, since dynorphins and exogenous agonists inhibit dopamine release in striatum and nucleus accumbens (Mulder et al., 1984; Herrera-Marschitz et al., 1986; Di Chiara and Imperato, 1988; Reid et al., 1988; Spanagel et al., 1992; Ronken et al., 1993; Schlosser et al., 1995), while antagonists enhance dopamine release (Spanagel et al., 1992). A mild speed increase is

57 observed in the MPTP-injected WT groups; however it is probable that this finding is artifactual due to the significantly decreased movement time and the decreased accuracy of distance measures in the setting of substantially depressed motor activity. It should be recognized that the observed changes in open-field behavior may be related to altered mechanisms of reward or arousal rather than motor impairment per se. Object recognition tasks, which in contrast to the inhibitory avoidance assay or water maze tasks permit evaluation of learning in the absence of an aversive stimulus, were used to determine effects of aging and Pdyn deletion on recognition and passive recall. In both protocols, average object recognition parameters in Pdyn KO mice are slightly lower than their WT counterparts in most comparisons; however the most reliable finding is that 24-hour recall in Pdyn KO mice is significantly impaired among 12-month and 24-month mice. However, the proportion of time spent on novel object exploration in these cases was substantially less than 50%, suggesting an altered response to novel stimuli rather than a recognition memory failure may account for these observations. In fact, despite spending a smaller percentage of time on the novel objects during retention trials, the total of time spent exploring either object is substantially elevated in the Pdyn KO mice. The above conjecture is supported by our further findings that young Pdyn KO mice show significantly greater total exploration times. While most groups show declining total exploratory activity when presented with only familiar objects during the acquisition phase of Protocol 1, the young adult Pdyn KO mice exhibited substantially less habituation-related change, resulting in a sustained elevation in total exploratory times across all trials. These findings are internally consistent with the results of our open-field experiments showing markedly elevated exploratory indices in young Pdyn KO mice. The present findings demonstrate subtle changes in learning and memory and motor function attributable to Pdyn deletion. While the improved performance of Pdyn KO mice in the passive avoidance trial may appear on initial observation to conflict with the minimally affected object recognition parameters, it has previously been proposed that the role of dynorphins in memory may depend on the nature of the reinforcing stimulus (Yakovleva et al., 2006a). The variation in the effects of Pdyn deletion on performance in the different learning and memory paradigms may be related to the

58 differing neuroanatomical substrates involved in each type of learning task. For instance, a recent study found that subchronic exposure to phencyclidine, which increases striatal dynorphin expression, causes impairment of novel object recognition without affecting passive avoidance (Shirayama et al., 2007). Since dopamine mediates the reinforcement of addictive behaviors (Bozarth and Wise, 1983) and since the opioid system is believed to modulate dopaminergic tone (Di Chiara and Imperato, 1988; Spanagel et al., 1992; Kreek, 1996; Kreek et al., 2002), dynorphinergic dysregulation is likely to interfere preferentially with tasks involving activation of dopamine-dependent reward pathways. In the case of water maze and passive avoidance performance, it is conceivable that the reward mechanisms associated with escape from the water maze and avoidance of an electrical shock, respectively, are less subject to inhibition in Pdyn KO mice, resulting in improved performance. Given the fairly diverse pharmacological profile of the various Pdyn-derived peptides and their potential roles in not only learning and memory, but also reward mechanisms, motor activity, and nociception, further studies are need to clarify the mechanisms by which Pdyn deletion produces the effects seen in the present study. However, the effects of dynorphins on dopaminergic pathways represent a plausible site of action of dynorphins in influencing performance in the behavioral assays discussed above. The following chapter will discuss changes in dopamine with aging and Pdyn deletion.

CHAPTER 4: Effects of prodynorphin deletion on striatal dopamine in mice during normal aging and in response to MPTP

Introduction Alterations in dopamine neurotransmission in basal ganglia pathways produce motor deficits in normal aging (Ingram, 2000) and pathological states like Parkinson’s disease (PD) (Hornykiewicz and Kish, 1987; Graybiel et al., 1990; Forno, 1996). Dopamine, synthesized by substantia nigra (SN) neurons projecting to the striatum, has been strongly implicated in aging (Carlsson and Winblad, 1976; Kish et al., 1992; Reeves et al., 2002; Haycock et al., 2003). More severe dopaminergic neuronal loss is observed in idiopathic Parkinson’s disease (PD), for which advanced age is the most identifiable risk factor. However a causal link between aging and these processes is yet unclear. Accumulating evidence of interactions between dynorphin and dopamine pathways in the basal ganglia suggests that these endogenous neuropeptides may play important roles in age-related dopamine regulation with relevance to mechanisms of age-related motor dysfunction or susceptibility to PD. Dynorphin peptides are obtained by proteolytic cleavage of a precursor encoded by the prodynorphin (Pdyn) gene (Simonato and Romualdi, 1996) and are expressed in striatum, hypothalamus, nucleus accumbens, dentate gyrus, hippocampus, cerebral cortex, and spinal cord (Khachaturian et al., 1982; Watson et al., 1982a; Watson et al., 1982b; Khachaturian et al., 1993; Hurd et al., 1999). Within striatum, dynorphin is colocalized with substance P and GABA within dopamine D1-receptor-expressing striatonigral projection neurons (Steiner and Gerfen, 1998). Though originally identified as an opiate (Goldstein et al., 1979) with picomolar affinity for opioid receptors (Chavkin et al., 1982), dynorphin has a complex pharmacological profile and can participate in or opioid (Mansour et al., 1995) and non-opioid mechanisms (Brauneis et al., 1996; Caudle and Dubner, 1998; Tang et al., 1999; Tang et al., 2000; Woods and Zangen, 2001). Dynorphins and exogenous agonists inhibit dopamine release in striatum and nucleus accumbens (Mulder et al., 1984; Herrera-Marschitz et al., 1986; Di Chiara and Imperato, 1988; Reid et al., 1988; Spanagel et al., 1992; Ronken et al., 1993; Schlosser et al., 1995), while antagonists result in enhanced dopamine release

(Spanagel et al., 1992). Conversely, dopamine promotes dynorphin expression via the D1-receptor (Li et al., 1986; Li et al., 1988; Li et al., 1990; Bordet et al., 1997; Wong et al., 2003). Aging has varied effects on dynorphin expression, with age-related elevations in dynorphin noted in the hippocampus and frontal cortex (Jiang et al., 1989; Zhang et al., 1991; Gallagher and Nicolle, 1993) and in spinal cord (Iadarola et al., 1986; Zhang et al., 2004a). The molecular and biochemical causes and consequences of altered Pdyn expression are unclear, but given the inhibitory roles of dynorphins on dopaminergic activity and reports of dynorphin-mediated neurotoxicity (Hauser et al., 1999; Tan-No et al., 2001; Goody et al., 2003; Singh et al., 2003), we hypothesized that suppression of Pdyn expression would mitigate effects of aging on striatal dopaminergic activity. In this study, we examine the chronic effects of Pdyn deletion on dopamine biosynthesis in mice during normal aging by quantitating striatal dopamine content in wild-type (WT) and Pdyn knockout (KO) mice.

Methods Animal subjects Mice used in this study were mixed male and female WT or Pdyn KO mice that were maintained in an approved animal facility at the University of Kentucky. WT mice were derived from the 129SvEv-Tac line, and Pdyn KO mice, provided by Dr. Ute Hochgeschwender (Oklahoma Medical Research Foundation, Oklahoma City, OK), were obtained by targeted deletion of the coding exons of the Pdyn gene and maintained in a 129SvEv-Tac background (Sharifi et al., 2001). Animals were housed in groups of up to 4 per cage with food and water ad libitum in a well-ventilated room with a 12 h: 12 h light:dark cycle. Prior to weaning, tail specimens were collected from each animal, and DNA was extracted for subsequent genotyping to confirm the identity of the Pdyn locus by the polymerase chain reaction (PCR) using primer pairs specific for each allele. For experiments involving 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), 3-month (n = 3 - 5 per group) and 10-month (n = 5 - 8 per group) mice were injected with 0.9% saline or MPTP (Sigma, St. Louis, MO) at a dose of 15 mg/kg intraperitoneally four

61 times at 2-h intervals and euthanized at 7 days post-injection. A 20-month age point was not selected in the MPTP experiments due to logistical reasons pertaining to availability of mice of substantially advanced age. Striatal tissue specimens were dissected on ice from CO2-asphyxiated mice, weighed, frozen in liquid nitrogen, and stored at -80°C until needed for protein extraction or high-performance liquid chromatography (HPLC) analysis. All procedures involving animals were approved by the Institutional Animal Care and Use Committee at the University of Kentucky.

HPLC analysis HPLC analysis was performed based on methods described previously (Hall et al., 1989). Sensitive detection of 3,4-dihydroxyphenylacetic acid (DOPAC), dopamine (DA), serotonin (5-HT), norepinephrine (NE), 5-hydroxyindole-3-acetic acid (5-HIAA), and 4- hydroxy-3-methoxyphenylacetic acid (homovanillic acid; HVA) were performed using an isocratic HPLC system (Beckman, Inc., Fullerton, CA) at a flow rate of 2 ml/min coupled to a dual-channel electrochemical array detector (model 5100A, ESA, Inc., Chelmsford,

MA), E1 = +0.35 mV and E2 = -0.25 mV, using an ESA model 5011 dual analytical cell. The compounds of interest were separated by reverse-phase chromatography, using a C18 column (4.6 mm x 75 mm, 3- m particle size, Shiseido CapCell Pak UG120, Shiseido Co., LTD., Tokyo, Japan) with a pH 4.1 citrate-acetate mobile phase containing 4.0% methanol and 0.341 mM 1-octane-sulfonic acid. By comparison to a set of analytical standards, quantities of the species of interest were calculated as pmol per mg wet tissue, and the results expressed as means ± standard errors of the mean (SEM). Samples exhibiting aberrantly high quantities of degradation products consistent with post-mortem degradation or poor tissue quality were identified and excluded from the analysis based on the presence of one or more turnover ratios exceeding 5 standard deviations above the sample mean. Results were pooled from multiple HPLC experiments after applying a proportional scaling factor to normalize data to the mean of the 10-mo WT control group in each experiment to minimize run-to-run variability in quantitation.

Protein extraction and western blotting Whole striatal specimens from a separate group of 3- (n = 6 per group) and 21-month (n = 8 per group) WT and Pdyn KO mice were dissected and homogenized in lysis buffer containing 10 mM Tris-HCl pH 7.4, 150 mM NaCl, 0.5% Triton X-100, 1 mM Na orthovanadate, and protease inhibitor cocktail (1:1000 dilution; Sigma, St. Louis, MO) using a Brinkman Kinematic sonicator. Following a 30- to 60-minute incubation on ice, samples were centrifuged to remove cell debris, and protein content was determined by a modification of the Lowry assay using the DC Protein Assay kit (catalog #500-0115; Bio-Rad Laboratories, Hercules, CA). Prior to loading, samples were normalized to a concentration of 1.0 mg/ml and subsequently diluted in loading buffer containing 5% 2-mercaptoethanol (BME; Sigma, St. Louis, MO), boiled for 5 minutes, and cooled on ice. Samples containing 7 mg total protein were subjected to electrophoresis in 12% polyacrylamide gels, followed by transfer onto nitrocellulose membranes. Membranes were then washed in a Tris-buffered saline wash buffer containing 0.05% Tween-20, blocked with 5% blotting-grade non-fat milk, and incubated at room temperature with primary antibody for a minimum of 2 h. The primary antibodies used are as follows: tyrosine hydroxylase (TH, 1:3000 dilution; Pel-Freez Biologicals, Rogers, AK), phospho-Ser40-TH (1:2500 dilution; Chemicon), phospho-Ser31-TH (1:2500 dilution; Chemicon, Billerica, MA), dopamine transporter (DAT C-20, 1 g/ml; Santa Cruz Biotech, Santa Cruz, CA), synaptophysin (SY38, 1:200 dilution; Research Diagnostics, Inc., Flanders NJ), and actin (1:3000 dilution, Sigma, St. Louis, MO). Following repeated washes in wash buffer, membranes were then incubated for 2 h with horseradish peroxidase-conjugated secondary antibody. After at least three successive rinses in wash buffer, chemiluminescent detection by the ECL-plus system (Amersham, Piscataway, NJ) was performed, with subsequent exposure on Amersham Hypermax film. For multiple antibody probings on a single blot, membranes were stripped by a 40-minute incubation in 0.7% BME at 55°C. Film exposures were scanned using an Epson Expression 1640XL scanner under transparency mode, and the resulting images were recorded as uncompressed 8-bit grayscales at 300 dpi resolution. Optical densities (ODs) of the band of interest in each

63 lane were measured using Scion Image software (release beta 4.0.2, Scion Corporation, Houston, TX) by recording the average pixel grayscale value within rectangular areas of a fixed size for a given image. Background ODs were obtained at multiple points along the width of the image, and the average background value was subtracted from each band OD prior to normalizing to the respective actin band OD and expressing the results relative to the mean value for the image. Data were obtained by combining results from two different gels containing samples from distinct sets of mice and are expressed as mean relative ODs ± standard errors of the mean (SEM).

Statistical analysis Statistical analyses were performed using Systat 10 software (SPSS, Inc., Chicago, Illinois). Two-way ANOVA was used for evaluating effects of aging and Pdyn, and in experiments involving an MPTP component, three-way ANOVA was applied. Individual contrasts were obtained by the least significant difference method.

Results Striatal dopamine is elevated in aged Pdyn KO mice Quantitative HPLC analysis of dopamine from mouse striatal samples demonstrated that dynorphin-deficient mice on average exhibited markedly higher dopamine concentrations in striatum than WT mice, especially at increasing age (Figure 4.1). Two-way ANOVA of dopamine levels revealed a significant main effect of age [F(2, 24) = 4.77, p<0.018] and of Pdyn genotype [F(1, 24) = 5.42, p<0.029]. Striatal dopamine concentrations in Pdyn KO mice were significantly higher at 10 months [F(1, 24) = 8.48, p<0.02] and 20 months [F(1, 24) = 6.44, p<0.008], when compared to 3-month Pdyn KO mice. However, the Pdyn genotype effect on total striatal dopamine levels was only statistically significant in the 20-month group [F(1, 24) = 4.50, p<0.045]. Notably, there was no appreciable age-related decline in total striatal dopamine content in the WT mice of this study.

No significant differences with age or Pdyn genotype were noted in the metabolites DOPAC or HVA despite an increase in mean DOPAC levels with age. However, two-way ANOVA of molar turnover ratios revealed subtle but statistically significant differences (Figure 4.2). With increasing age, the mean DOPAC/DA ratio exhibited an increasing trend whereas the mean HVA/DA ratio decreased. The increase in the DOPAC/DA ratio was most prominent among the WT mice with an approximately 50% elevation at 20 months that was nonetheless short of significance [F(1, 24) = 3.76, p<0.064 (NS)]. In contrast, the Pdyn KO mice showed minimal change in the DOPAC/DA ratio with age, resulting in a significant Pdyn genotype effect at 20 months when examining either DOPAC/DA [F(1, 24) = 4.51, p<0.044] or (DOPAC+HVA)/DA [F(1, 24) = 4.91, p<0.036]. A statistically significant age effect was seen only in the

140 WT DA 30 DOPAC Pdyn -/- *# 120 * 25 100 20 80 15 60 10

40 Concentration(pmol/mg) 20 Concentration(pmol/mg) 5

0 0 3-mo. 10-mo. 20-mo. 3-mo. 10-mo. 20-mo. Age Age 10 Figure 4.1. Dopamine (DA) and its HVA 9 metabolites DOPAC and HVA as a function of age and Pdyn 8 genotype. 7 6 These results were obtained by HPLC 5 analysis of whole striatum homogenates of 4 WT and Pdyn KO mice aged 3, 10, and 20 months. Note the age-dependent increase 3 2 in dopamine among Pdyn KO mice and Concentration(pmol/mg) the significantly increased dopamine in 1 aged Pdyn KO mice in comparison to 0 aged WT mice. Data represent means 3-mo. 10-mo. 20-mo. SEM. *p<0.05 relative to corresponding Age 3-month mice; #p<0.05 relative to corresponding WT mice.

HVA/DA ratio, with significant declines noted among WT mice at 10 months [F(1, 24) = 4.49, p<0.045] and among Pdyn KO mice at 10 months [F(1, 24) = 5.86, p<0.023] and at 20 months [F(1, 24) = 6.57, p<0.017]. Analysis of norepinephrine, serotonin, and the serotonin metabolite HIAA resulted in no significant contrasts (Figure 4.3). The turnover ratio HIAA/5HT also did not show significant age or Pdyn genotype effects (Figure 4.2). Although the present study was not initially designed to optimally assess effects of gender, inclusion of gender as an additional ANOVA factor reveals a main effect of gender on dopamine quantities [F(1, 19) = 5.49, p<0.03] with males averaging 14%

0.35 WT 0.14 0.30 Pdyn -/- 0.12 * 0.25 0.10 * * 0.20 # 0.08

0.15 0.06

Molar ratios Molar ratios Molar HVA / DA

0.10 HVA / DA 0.04

DOPACDOPAC / / DADA

0.05 0.02

0.00 0.00 3-mo. 10-mo. 20-mo. 3-mo. 10-mo. 20-mo. Age Age 0.45 1.00 0.40 0.90

0.35 0.80 HT HT 0.70

0.30 #

- - 0.60 0.25 0.50 0.20

0.40 Molar ratios Molar Molar ratios Molar 0.15

0.30

HIAAHIAA / / 5 5 0.10 0.20

0.05 0.10

(HVA + (HVA(HVA+ + DOPAC)DOPAC)// DADA 0.00 0.00 3-mo. 10-mo. 20-mo. 3-mo. 10-mo. 20-mo. Age Age

Figure 4.2. Dopamine and serotonin turnover ratios as a function of age and Pdyn genotype. These results were obtained by HPLC analysis of whole striatum homogenates of WT and Pdyn KO mice aged 3, 10, and 20 months. Only mild decreases in dopamine turnover ratios are seen in aged Pdyn KO mice, whereas no trends in serotonin turnover were observed. Data represent means SEM. *p<0.05 relative to corresponding 3- month mice; #p<0.05 relative to corresponding WT mice.

NE 5-HT HIAA

4 WT 7 6 Pdyn -/- 3 6 5

3 5 4 2 4 3 2 3 2

1 2

Concentration(pmol/mg) Concentration(pmol/mg) Concentration(pmol/mg) 1 1 1

0 0 0 3-mo. 10-mo. 20-mo. 3-mo. 10-mo. 20-mo. 3-mo. 10-mo. 20-mo. Age Age Age

Figure 4.3. HPLC analysis of norepinephrine (NE), serotonin (5-HT), and the serotonin metabolite HIAA as a function of age and Pdyn genotype. These results were obtained by HPLC analysis of whole striatum homogenates of WT and Pdyn KO mice aged 3, 10, and 20 months. No significant effects were noted for these variables. Data represent means SEM.

higher dopamine levels than females irrespective of Pdyn or age (males: 90.5 6.8 pmol/mg tissue; females: 79.3 5.9 pmol/mg tissue). In addition, a Pdyn-gender interaction was also significant [F(1, 19) = 5.47, p<0.03], as gender differences were less apparent among Pdyn KO mice. A male gender effect on dopamine levels was observed in WT mice [males: 84.5 7.9 pmol/mg tissue; females: 72.8 6.2 pmol/mg tissue; F(1, 19) = 7.49, p<0.013], but was smaller and not significant in Pdyn KO mice (males: 93.1 9.4 pmol/mg tissue; females: 88.5 11.0 pmol/mg tissue). Significant gender effects were not observed with other amines or any turnover ratios examined in this set of mice.

Pdyn deletion enhances TH phosphorylation at Ser40 Western blotting analyses were performed on young (3-month) and aged (21-month) mice to detect TH, the rate-limiting enzyme in dopamine synthesis that catalyzes the conversion of tyrosine to the dopamine precursor L-dopa (Figure 4.4). Other dopaminergic and neuronal markers were also examined, including DAT, SY38, and actin (Figure 4.4). In addition, because TH activity can be modulated by phosphorylation at specific serine sites (Haycock and Haycock, 1991; Sutherland et al., 1993; Harada et al., 1996; Lindgren et al., 2000; Lindgren et al., 2001), phospho-Ser40-TH and phospho-

3-mo WT 3-mo KO 21-mo WT 21-mo KO

P-Ser40-TH

P-Ser31-TH

DAT

SY38

actin

Figure 4.4. Representative Western blot of mouse striatal homogenates. Immunoblot analysis of whole striatum homogenates of WT and Pdyn KO mice aged 3 and 21 months was performed, and bands representing immunoreactivity to TH, phospho-Ser40-TH (P-Ser40-TH), phosphor-Ser31-TH (P-Ser31-TH), dopamine transporter (DAT), synaptophysin (SY38), and actin as a function of age and Pdyn genotype are shown. Depicted are blots from a representative gel from one of two separate experiments that are pooled for statistical analysis.

Ser31-TH immunoreactivity were also examined. Two-way ANOVA to evaluate for effects of age and Pdyn genotype revealed significant main effects of age [F(1, 24) = 5.51, p<0.027] and Pdyn genotype [F(1, 24) = 10.40, p<0.004] on TH immunoreactivity (Figure 4.5). A statistically significant age effect was noted only in the WT mice, with 21-month mice displaying approximately 50% greater TH immunoreactivity than 3- month mice [F(1, 24) = 4.67, p<0.041]. In both age groups, mean TH immunoreactivity in Pdyn KO mice was approximately 60% that of the corresponding WT mean; however statistical significance was seen only among the aged mice [F(1, 24) = 9.03, p<0.006]. Phospho-Ser40-TH also exhibited a significant main effect of aging [F(1, 24) = 12.69, p<0.002], with aged Pdyn KO mice demonstrating an increase by a factor of 3.6 compared to young Pdyn KO mice [F(1, 24) = 10.97, p<0.003]. In contrast, phospho- Ser31-TH showed a significant Pdyn effect [F(1, 24) = 4.76, p<0.039] similar to that exhibited by total TH, but there was no statistically significant aging effect. A

68 significant main effect of age was also noted for the dopamine transporter [F(1, 24) = 5.41, p<0.029], although individual contrasts were not significant in this analysis. Correlation analysis confirmed that DAT levels correlated very strongly with TH (Pearson's correlation coefficient r = 0.889, Bonferroni-corrected p<0.0005). No significant differences were evident in synaptophysin levels in this study. To assess the extent to which the above differences in immunoreactivity to phosphorylated TH reflect posttranslational effects as opposed to changes in dopaminergic neuron mass, a separate analysis was also performed on phospho-Ser40- TH: total TH and phospho-Ser31-TH: total TH ratios (Figure 4.5). This analysis revealed no age effect but demonstrated a significant main effect of Pdyn genotype on only the phosphor-Ser40-TH: total TH ratio [F(1, 24) = 5.09, p<0.033], which was approximately 2-fold higher in Pdyn KO mice compared to the respective WT counterparts, with a significant difference observed only among the aged groups [F(1, 24 = 4.55, p<0.043].

Table 4.1. ANOVA results from analysis of effects of MPTP, age, and Pdyn genotype on dopamine and related monoamines. Three-way ANOVA was performed on results of monoamine quantitation and their respective turnover ratios. Only variables for which a statistically significant effect or interaction is noted are listed below along with corresponding F statistics and p-values.

MPTP effect DA F(1, 37) = 54.93 p<0.0005 DOPAC F(1, 37) = 9.88 p<0.003 HVA F(1, 37) = 11.76 p<0.002 5HT F(1, 37) = 22.93 p<0.0005 HVA / DA F(1, 37) = 11.65 p<0.002 DOPAC / DA F(1, 37) = 12.15 p<0.001 (DOPAC + HVA) / DA F(1, 37) = 12.76 p<0.001 HIAA / 5HT F(1, 37) = 9.13 p<0.005 Pdyn effect 5HT F(1, 37) = 4.46 p<0.042 MPTP x age DA F(1, 37) = 9.72 p<0.004 DOPAC F(1, 37) = 4.61 p<0.038 HVA F(1, 37) = 4.86 p<0.034

MPTP-induced dopamine loss is unaffected by Pdyn deletion We also examined dopaminergic changes induced by systemic MPTP, which mimics in mice many features of human parkinsonism via toxic degeneration of SN dopaminergic neurons (Arai et al., 1990; Sundstrom et al., 1990). Intraperitoneal MPTP was administered to 3- or 10-month Pdyn KO or WT mice, with corresponding saline controls. A 3-way ANOVA to evaluate the effects of MPTP administration, age, or Pdyn genotype on dopamine and its metabolites found significant main effects of treatment and an MPTP-age interaction but no main effect of age or genotype in this data set (Table 4.1). Levels of dopamine in the striatum significantly decreased 7 days following intraperitoneal MPTP administration (Figure 4.6). Among 10-month mice, MPTP resulted in a reduction of dopamine levels by 71% in WT mice [F(1, 37) = 34.64, p<0.0005] and 73% in Pdyn KO mice [(F(1, 37) = 44.48, p<0.0005] when compared to respective saline-injected controls. The MPTP-associated loss of dopamine among the 3- month mice was less robust, with MPTP-injected WT mice exhibiting a nonsignificant reduction of 40%, and Pdyn KO mice demonstrating a statistically significant 46% reduction [(F(1, 37) = 5.66, p<0.023]. Consistent with the preceding analysis, saline- injected Pdyn KO mice at 10 months showed significantly higher dopamine levels compared to 3-month Pdyn KO mice [(F(1, 37) = 8.96, p<0.005]. The dopamine metabolites DOPAC and HVA also showed significant MPTP- induced changes, most prominent at 10 months (Figure 4.6). DOPAC levels decreased by over 50% following MPTP administration in 10-month WT [F(1, 37) = 9.98, p<0.003] and 10-month Pdyn KO mice [F(1, 37) = 10.08, p<0.003]. Similarly, MPTP caused a decrease in HVA by 33% and 44% in 10-month WT [F(1, 37) = 6.95, p<0.012] and 10- month Pdyn KO mice [F(1, 37) = 16.15, p<0.0005], respectively. Serotonin levels also responded significantly to MPTP (Figure 4.7). MPTP administration was associated with an approximately 25% decline in striatal serotonin levels in each group with statistical significance noted in 3-month Pdyn KO [F(1, 37) = 5.70, p<0.022], 10-month WT [F(1, 37) = 12.45, p<0.001], and 10-month Pdyn KO mice [F(1, 37) = 6.01, p<0.019]. No significant contrasts were observed in levels of HIAA or NE. However, the serotonin turnover ratio (HIAA / 5HT) was found to increase

following MPTP administration in 10-month WT mice to 1.7 times that of saline controls [F(1, 37) = 7.40, p<0.01, Figure 4.8].

(a) 1.8 (b) 2.0 (c) 2.0 1.6 * 1.8 ** 1.8 1.4 1.6 1.6 1.2 1.4 1.4 1.0 ## 1.2 1.2 1.0 0.8 1.0 0.8 0.8 0.6 0.6 0.6 0.4 0.4 0.4 0.2 0.2 0.2

0.0 0.0 0.0

TH TH immunoreactivityOD) (relative pSer31-TH immunoreactivityOD) pSer31-TH (relative 3-mo. WT immunoreactivityOD) pSer40-TH (relative 3-mo. WT 3-mo. WT 21-mo. WT 21-mo. WT 21-mo. WT 3-mo. Pdyn -/- 3-mo. Pdyn -/- 3-mo. Pdyn -/- 1.8 21-mo. Pdyn -/- 21-mo. Pdyn -/- 21-mo. Pdyn -/- (d) 1.6 (e) 2.5 (f) 2.0 # 1.8 1.4 2.0 1.6 1.2 1.4 1.0 1.5 1.2 0.8 1.0 0.6 1.0 0.8 0.6 0.4 0.5 0.4 0.2 0.2 0.0 pSer31-TH: total TH (relative OD) TH (relative total pSer31-TH: 0.0

0.0 OD) TH (relative total pSer40-TH: DAT immunoreactivityOD) DAT (relative

3-mo. WT 3-mo. WT 3-mo. WT 21-mo. WT 21-mo. WT 21-mo. WT 3-mo. Pdyn -/- 3-mo. Pdyn -/- 3-mo. Pdyn -/- 21-mo. Pdyn -/- 21-mo. Pdyn -/- 21-mo. Pdyn -/- 1.8 1.4 (g) 1.6 (h) 1.2 1.4 1.0 1.2 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2

0.0 0.0

SY38 immunoreactivityOD) (relative Actin immunoreactivityActin OD) (relative

3-mo. WT 3-mo. WT 21-mo. WT 21-mo. WT 3-mo. Pdyn -/- 3-mo. Pdyn -/- 21-mo. Pdyn -/- 21-mo. Pdyn -/- Figure 4.5. Densitometric quantitation of Western blot data. Immunoreactivity to (a) tyrosine hydroxylase (TH), (b) phospho-Ser40-TH (pSer40-TH), (c) phospho-Ser31-TH (pSer31-TH), (d) dopamine transporter (DAT), (e) phospho- Ser40-TH: total TH ratio, (f) phospho-Ser31-TH: total TH ratio, (g) synaptophysin (SY38), and (h) actin are shown as a function of age and Pdyn genotype. Pdyn KO mice demonstrate increased phosphorylation of TH at Ser40 but not Ser31, an effect most prominently seen at 21 months. However, total TH immunoreactivity was decreased in Pdyn KO mice compared to WT mice. Data were pooled from two separate gels and represent mean relative optical densities (ODs) SEM. *p<0.05, **p<0.01 relative to corresponding 3-month mice; #p<0.05, ##p<0.01 relative to corresponding WT mice.

DA DOPAC HVA 10 140 WT 25 9 *# 120 Pdyn -/- ** 20 8 100 7 15 6 ‡ 80 † ‡ 5 ‡ 60 ‡ 10 4 ‡ 3 40 ‡ 5 2

20 (pmol/mg) Concentration 1

Concentration (pmol/mg) Concentration (pmol/mg) Concentration 0 0 0

3-mo. control3-mo. MPTP 3-mo. control3-mo. MPTP 3-mo. control3-mo. MPTP 10-mo. control10-mo. MPTP 10-mo. control10-mo. MPTP 10-mo. control10-mo. MPTP

Figure 4.6. Effect of MPTP on dopamine and its metabolites as a function of age and Pdyn genotype. These results reflect HLPC measurements of dopamine (DA) and its metabolites in whole striatum from 3- and 10-month mice obtained 7 days post-injection of saline or MPTP. Note the decreases in dopamine with milder reductions in DOPAC and HVA in response to MPTP. Data represent means SEM. †p<0.05, ‡p<0.01 relative to corresponding saline controls; *p<0.05, **p<0.01 relative to corresponding 3-month mice; #p<0.05 relative to corresponding WT mice.

Dopamine turnover ratios also increased upon MPTP administration, more prominently among the 10-month mice (Figure 4.8). MPTP caused DOPAC/DA to increase in 10-month WT and Pdyn KO mice to 2.5 [F(1, 37) = 9.57, p<0.004] and 3 times [F(1, 37) = 10.41, p<0.003] the saline control values, respectively, and HVA/DA increased 3.6-fold [F(1, 37) = 9.03, p<0.005] and 3.9-fold [F(1, 37) = 7.89, p<0.008], respectively. Consequently, the (HVA+DOPAC)/DA ratio was increased by a factor of 2.8 and 3.3 in 10-month WT [F(1, 37) = 9.98, p<0.003] and Pdyn KO mice [F(1, 37) = 9.99, p<0.003], respectively. This analysis also revealed modest but statistically significant elevations in HVA levels in saline-injected 10-month Pdyn KO mice compared to saline-injected 10-month WT mice [F(1, 37) = 4.56, p<0.039] or 3-month Pdyn KO mice [F(1, 37) = 5.57, p<0.024]. Inclusion of gender in the ANOVA analysis found no significant main effects of gender in any of the monoamine levels examined in the above MPTP experiment. However, a significant MPTP-sex interaction was evident in dopamine [F(1, 29) = 10.22, p<0.003] and HVA levels [F(1, 29) = 4.17, p<0.05).

4 NE WT 7 5-HT 6 HIAA Pdyn -/- 6 5 3 5 † † ‡ 4 4 2 3 3 2 2 1

1 1

Concentration (pmol/mg) Concentration (pmol/mg) Concentration (pmol/mg) Concentration 0 0 0

3-mo. control3-mo. MPTP 3-mo. control3-mo. MPTP 3-mo. control3-mo. MPTP 10-mo. control10-mo. MPTP 10-mo. control10-mo. MPTP 10-mo. control10-mo. MPTP

Figure 4.7. Effect of MPTP on NE, 5-HT, and HIAA. NE and HIAA show no significant changes in response to MPTP, but 5-HT shows mild decreases 7 days after MPTP injection. Data represent means SEM. †p<0.05, ‡p<0.01 relative to corresponding saline controls.

MPTP-induced loss of dopamine in males averaged across all groups was 81% compared to 41% in females. MPTP-sex interactions were also noted for the turnover ratios HVA/DA [F(1, 29) = 5.69, p<0.024], DOPAC/DA [F(1, 29) = 4.47, p<0.043], HIAA/5HT [F(1, 29) = 4.82, p<0.036], and (HVA+DOPAC)/DA [F(1, 29) = 5.55, p<0.026], for which female turnover ratios averaged 40-60% less than the corresponding male ratios. The gender discrepancy in turnover ratios was more pronounced among 10- month mice, as manifested by significant age-MPTP-sex interactions for DOPAC/DA [F(1, 29) = 5.46, p<0.027], HIAA/5HT [F(1, 29) = 8.11, p<0.008], and (HVA+DOPAC)/DA [F(1, 29) = 4.56, p<0.041].

Discussion In this study, quantitative HPLC was used to identify changes in dopamine, other biogenic amines, and related metabolites in mouse striatum with aging, Pdyn deletion, or MPTP. Interestingly, no decline in dopamine levels with aging alone was apparent among WT mice. While several studies have documented declines in striatal dopamine with aging in rats (Moretti et al., 1987; Marshall and Rosenstein, 1990), monkeys (Collier

0.7 0.4 WT ‡ ‡ 0.6 Pdyn -/- ‡ ‡ 0.5 0.3 0.4 0.2

0.3

Molarratios

Molarratios HVA / HVA/ DA

0.2 HVA/ DA 0.1 DOPAC / DA DOPAC / DA 0.1 0.0 0.0

3-mo. control3-mo. MPTP 3-mo. control3-mo. MPTP 10-mo. control10-mo. MPTP 10-mo. control10-mo. MPTP 1.0 ‡ 1.8 0.9 ‡ 1.6 ‡

0.8 1.4 HT

0.7 HT - - 1.2 0.6 1.0 0.5 0.8 0.4

0.3 0.6 Molarratios

Molarratios 0.4 HIAA / HIAA/ 5 0.2 HIAA/ 5 0.1 0.2

0.0 0.0

(HVA + (HVA(HVA+ + DOPAC)DOPAC)// DADA

3-mo. control3-mo. MPTP 3-mo. control3-mo. MPTP 10-mo. control10-mo. MPTP 10-mo. control10-mo. MPTP

Figure 4.8. Effect of MPTP on dopamine and serotonin turnover ratios. MPTP produces increases in dopamine turnover and serotonin turnover 7 days post- injection, findings which are statistically significant only among older (10-month) mice. Data represent means SEM. †p<0.05, ‡p<0.01 relative to corresponding saline controls.

et al., 2007), and humans (Carlsson and Winblad, 1976; Kish et al., 1992; Haycock et al., 2003), there have also been studies in rodents describing whole-striatum dopamine content to demonstrate minimal or no decline with age (Friedemann and Gerhardt, 1992; Burwell et al., 1995; Hebert and Gerhardt, 1998; Stanford et al., 2003). While the

74 relatively preserved dopamine content among WT mice in this study may be attributable to species and strain differences, it also lends support to the view that normal aging alone is unlikely to produce the pathological dopamine deficits seen in disorders such as Parkinson's disease. The use of Pdyn KO mice offers distinct advantages over studies employing traditional pharmacologic antagonists as this study does not presume a specific receptor interaction and permits evaluation of chronic changes in the presence of compensatory effects, which would likely not be feasible when using only exogenously applied agents. We show in this study that striatal dopamine is elevated in aging Pdyn KO mice compared to aging WT mice. These findings could be interpreted as a chronic inhibitory effect of dynorphin on dopamine biosynthetic activity, or alternatively, a harmful effect of dynorphins on dopaminergic cell survival in the SN. Despite published evidence that dynorphin peptides may exhibit toxic or pro-apoptotic effects in both neuronal and non- neuronal cells under various experimental conditions (Hu et al., 1996; Hauser et al., 1999; Hu et al., 1999; Hu et al., 2000; Hauser et al., 2001; Tan-No et al., 2001; Goody et al., 2003; Singh et al., 2003), the latter interpretation is less likely because in comparison to their WT counterparts, 21-month Pdyn KO mice in this study show less immunoreactivity to TH, a surrogate marker for dopaminergic neuron mass. Moreover, a putative neurotoxic role of dynorphin is further challenged by reports that dynorphin peptides can show neuroprotective properties in some situations (Kong et al., 1997; Liu et al., 2001; Gabrilovac et al., 2003). Differences in total TH levels are therefore unlikely to account for the altered dopamine levels seen with aging in Pdyn KO mice. One study examining post-mortem human striata found an aging-associated decline in striatal dopamine content without loss of TH (Haycock et al., 2003), suggesting that impaired dopaminergic neuron function rather than cell death could account for the relative dopamine depletion in aged WT mice relative to aged Pdyn KO mice in this study. This assertion is supported by reports that infusion of dynorphin into the striatum decreases dopamine levels by as much as 60% (Zhang et al., 2004b). Consequently, we addressed the effect of aging and Pdyn deletion on the enzymatic activity of TH, which is subject to physiological regulatory control via phosphorylation at Ser31 or Ser40 by different kinase pathways, resulting in increases in

75 the intrinsic rate of catalysis (Haycock and Haycock, 1991; Sutherland et al., 1993; Harada et al., 1996; Lindgren et al., 2000; Lindgren et al., 2001). Other investigators, for instance, have noted increased Ser31 phosphorylation of TH by neurotrophic factors (Salvatore et al., 2004). Although we did not directly measure enzyme activity, our results show that a greater proportion of TH molecules in aged Pdyn KO mice are phosphorylated at the Ser40 site compared to aged WT mice. Because activation of striatal dopamine D2 receptors selectively inhibits TH phosphorylation at Ser40 via inhibition of adenylate cyclase (Lindgren et al., 2001), the increased phospho-Ser40-TH in aged Pdyn KO mice may be attributable to decreased D2 receptor activity. Alternatively, glutamate acting via NMDA receptors can also decrease TH phosphorylation at Ser40 in striatum (Lindgren et al., 2000), and since dynorphin peptides have been observed to potentiate NMDA receptor activation (Zhang et al., 1997; Woods et al., 2006), the increased Ser40 phosphorylation in dynorphin-deficient mice in the present study is consistent with the above observations. While the precise mechanisms underlying the age-associated increases in dopamine in Pdyn KO mice remain to be elucidated, the available data suggest a role of dynorphin peptides in suppressing dopaminergic neuron function via regulation of dopamine biosynthesis. Other potential mechanisms include cellular or metabolic dysfunction caused by toxic effects of specific dynorphin peptides as supported by evidence of dynorphin toxicity in striatal neurons (Goody et al., 2003; Singh et al., 2003), spinal neurons (Caudle and Isaac, 1987; Qu and Isaac, 1993; Hauser et al., 1999), and non-neuronal cells (Tan-No et al., 2001), primarily via non-opioid mechanisms. In the present study, the Pdyn-associated differences in dopamine levels are more pronounced and are statistically significant at increased age. The findings that advanced age unmasks these effects may reflect age-related alterations in the dynorphinergic system. Increased dynorphin was observed in hippocampus of normal aged rats (Jiang et al., 1989; Zhang et al., 1991; Gallagher and Nicolle, 1993; Kotz et al., 2004), the cortex and hippocampus of amyloid precursor protein-overexpressing mice (Diez et al., 2000), frontal cortex of Down’s syndrome patients (Risser et al., 1996), and frontal and parietal cortex of Alzheimer’s disease patients (Risser et al., 1996; Yakovleva et al., 2006a), thus implicating dynorphin in age-related pathophysiological processes. Peripheral

76 inflammation also upregulates Pdyn in spinal cord (Iadarola et al., 1986), and aging enhances this response (Zhang et al., 2004a). In striatum, Pdyn expression with aging has been less well studied, with some studies showing no changes in aging rats (Zhang et al., 1991; Kotz et al., 2004) and one study finding decreased Pdyn mRNA in striatum of normal aged humans (Backman et al., 2007). In addition, dynorphin expression is also markedly influenced by the dopaminergic system. In models of PD, decreases in dynorphin have been noted following experimental manipulations that lesion the nigrostriatal dopaminergic pathway (Gerfen et al., 1991). Similarly, striatal dynorphin expression is increased by dopamine, its biochemical precursors, or exogenous D1 agonists (Li et al., 1986; Li et al., 1988; Li et al., 1990; Bordet et al., 1997) and is reduced in D1-receptor knockout mice (Wong et al., 2003). The above studies point to a possible dysregulation of Pdyn expression with aging that may lead to or be caused by perturbations in the dopaminergic system. In this study, the potential effects of dynorphin on the dopaminergic system are abolished in the Pdyn KO mice, permitting evaluation of one component of this bidirectional interaction during the aging process. However, the molecular and cellular events influencing Pdyn expression during aging have not been fully identified and remain an interesting avenue for further research. MPTP, in this study, produced declines in striatal dopamine content 7 days after injection of a standard sublethal dose, consistent with prior studies using the C57BL6 strains (Arai et al., 1990; Sundstrom et al., 1990; Ali et al., 1993). Although the extent of dopamine loss is less than the 80% originally reported (Arai et al., 1990; Sundstrom et al., 1990), the lower magnitude observed in the present study may be attributed to differential susceptibility among different mouse strains, which has previously been noted (German et al., 1996). Dopamine turnover, as reflected in the ratio of levels of dopamine metabolites to levels of dopamine, was elevated following MPTP, consistent with activation of compensatory mechanisms to replenish dopamine following dopaminergic neuronal loss (Zigmond, 1997), while aging alone yielded no changes in turnover except for mild decreases with age in Pdyn KO mice that may reflect local negative feedback mechanisms. The MPTP-induced loss of dopamine was more prominent among aged mice, suggesting that these compensatory mechanisms are less

77 effective with advanced age, consistent with other studies finding aged-associated increases in susceptibility to MPTP (Faden, 1992). The gender effects observed in the above MPTP experiment are consistent with previous studies in mice. Attenuation of MPTP-induced dopamine depletion in females compared to males has been previously reported in other mouse strains (Freyaldenhoven et al., 1996; Miller et al., 1998), suggesting neuroprotective actions of estrogen. Also, increased opioid receptor sensitivity has been noted in females in human studies (Kreek et al., 1999), which may explain the mild gender difference in basal striatal dopamine present in the WT mice but absent in the Pdyn KO mice of this study. Other neurotransmitters, such as norepinephrine and serotonin, were relatively unaffected by age or Pdyn genotype in this study, signifying a relatively selective effect on the dopaminergic system. However our data show mild reductions in striatal serotonin content by MPTP, in contrast to prior studies documenting unchanged serotonin levels following MPTP administration (Chang et al., 1993; Callier et al., 2000). This finding is of uncertain significance but may be explained by strain-specific differences in MPTP metabolism, since an MPTP analogue 2'-NH2-MPTP has been shown to produce serotonergic deficits (Luellen et al., 2006). Interestingly, there was no significant effect of Pdyn genotype on post-MPTP dopamine levels, implying that Pdyn deletion confers no protection against MPTP toxicity. This further supports the contention that the effect of Pdyn deletion on striatal dopamine content is dependent on survival of MPTP-sensitive dopaminergic neurons, consistent with a mechanism involving modulation of dopamine biosynthetic activity rather than prevention of neurotoxicity. In the context of aging and neurodegenerative disease, the results of this study offer insight into one potential means by which striatal dopaminergic activity can be modulated by dynorphin peptide neurotransmitters in physiological aging or neurodegenerative disease. Additional studies are needed to elucidate further the mechanisms involved and to clarify the relevance of these findings to clinical entities such as Parkinson's disease.

CHAPTER 5: Radioimmunological quantitation of prodynorphin-derived peptides in mice in aging and an MPTP model

Introduction Expression of endogenous dynorphin peptides during aging and age-related disorders has received increased attention in the neurobiological literature. Initial localization studies have shown that dynorphins are widely distributed in the central nervous system (Khachaturian et al., 1982; Watson et al., 1982a; Watson et al., 1982b; Hurd et al., 1999), with high Pdyn mRNA levels in hypothalamus, nucleus accumbens, neostriatum, dentate gyrus, hippocampus, cerebral cortex, and spinal cord (Khachaturian et al., 1993). Within the striatum, dynorphins colocalize with substance P and GABA within dopamine D1-receptor-expressing striatonigral projection neurons [reviewed in (Steiner and Gerfen, 1998)], where they appear to regulate dopamine release and function in a negative feedback manner. Since dynorphins or exogenous agonists inhibit dopaminergic activity (Mulder et al., 1984; Herrera-Marschitz et al., 1986; Di Chiara and Imperato, 1988; Reid et al., 1988; Spanagel et al., 1992; Ronken et al., 1993; Schlosser et al., 1995), modulation of dynorphin expression by dopamine represents a biologically plausible mechanism by which levels of this important neurotransmitter are regulated. Lesioning of the nigrostriatal dopaminergic pathway as in some models of PD results in decreases in dynorphin levels (Gerfen et al., 1991). Similarly, striatal dynorphin expression is increased by dopaminerigc activity at the D1-receptor (Li et al., 1986; Li et al., 1988; Li et al., 1990; Bordet et al., 1997) and is reduced in mice lacking the D1- receptor (Wong et al., 2003). Given the intrinsic variability in the aging process and the diverse environmental and species-dependent factors contributing to neuropeptide expression during aging, it is not surprising that published reports show significantly varied effects on dynorphin expression. Age-related elevations in dynorphin have been noted in hippocampus and frontal cortex (Jiang et al., 1989; Zhang et al., 1991; Gallagher and Nicolle, 1993) and in spinal cord (Iadarola et al., 1986; Zhang et al., 2004a). Pdyn mRNA expression with aging in one study was decreased in arcuate nucleus and amygdala, increased in

79 hippocampus, and unchanged in striatum or cortex of male rats (Kotz et al., 2004). We previously noted no change in Dyn A with age in hippocampus but mild elevations in frontal cortex (Nguyen et al., 2005). Although the cellular events influencing Pdyn expression during aging have not been fully identified, changes in CREB expression (Collins-Hicok et al., 1994) or binding at the AP-1 site in the Pdyn promoter (Naranjo et al., 1991; Collins-Hicok et al., 1994; Hunter et al., 1995; Kitamura et al., 1995) could potentially increase transcription and thus represent feasible molecular mechanisms for increased Pdyn with age. Upregulation of Pdyn in spinal cord by peripheral inflammation (Iadarola et al., 1986) is enhanced with advancing age (Zhang et al., 2004a) by mechanisms that have not been fully elucidated. Recently, increased dynorphin levels were observed in the cortex and hippocampus of amyloid precursor protein- overexpressing mice (Diez et al., 2000) and in frontal cortex of Down’s syndrome or Alzheimer’s disease patients (Risser et al., 1996), thus further implicating dynorphins in pathophysiological neurological processes. The above studies point to a possible dysregulation of Pdyn expression with aging that may lead to pathological manifestations. Because of the relative abundance of dynorphin in striatum and the importance of this region of the brain during aging, we measure in this study Dyn A and B in striatum of mice with particular attention to changes associated with aging. In addition, we also evaluate striatal expression of dynorphin peptides following systemic exposure to the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), which produces selective dopaminergic neuronal death in the nigrostriatal system (Arai et al., 1990; Sundstrom et al., 1990; Ali et al., 1993). Since dynorphins have been found to be upregulated during aging in frontal cortex of rats as discussed above, we also assess dynorphin peptide expression in frontal cortical tissue of these mice.

Methods Animal subjects All animals in this study were wild-type (WT) or Pdyn KO mice that were maintained in an approved animal facility at the University of Kentucky. WT mice were derived from

80 the 129SvEv-Tac line, and Pdyn KO mice were originally obtained by targeted deletion of the coding exons of the prodynorphin gene and bred into a 129SvEv-Tac background as described previously (Sharifi et al., 2001). Animals were housed in groups of up to 4 per cage with food and water ad libitum in a well-ventilated room with a 12 h: 12 h light: dark cycle. Prior to weaning, tail specimens were collected from each animal, and DNA was extracted for subsequent genotyping to confirm the identity of the Pdyn locus by the polymerase chain reaction (PCR) using primer pairs specific for each genotype. All procedures involving animals are approved by the Institutional Animal Care and Use Committee at the University of Kentucky. The aging studies in frontal cortex used 3-month (n = 6), 10-month (n = 6), and 20-month (n = 6) male and female WT mice. Striatal data were pooled from two separate experiments such that sample sizes were as follows: 3-month (n = 15), 10-month (n = 15), and 20-month (n = 10). Pdyn KO mice were included as qualitative controls consisting of one mouse per group in each experiment. Pdyn-heterozygous (HET) mice (n = 5) were included in the 3-month age group only. Group sizes listed above reflect inclusion of mice pooled from saline control groups from the MPTP studies, as preliminary analyses found no significant differences among the saline controls. Two separate experimental MPTP protocols were used in this study. In experiments examining short-term post-MPTP effects, 3-month WT (n = 4) and 10- month WT (n = 5) mice were injected intraperitoneally with a single dose of 30 mg/kg MPTP, and striatal specimens were collected 6 hours later. The second MPTP protocol, involving a standard MPTP regimen of four 15 mg/kg intraperitoneal injections at 2-hour intervals with striatal tissue collection 7 days post-MPTP, used 3-month WT (n = 3), 3- month HET (n = 4), 10-month WT (n = 8) mice, and 1-2 Pdyn KO mice per age point as qualitative negative controls.

Dynorphin extraction and radioimmunoassay (RIA)

Frontal cortical and striatal tissue were isolated following rapid asphyxiation by CO2 and were frozen in liquid nitrogen prior to being processed for radioimmunoassay as previously described (Christensson-Nylander et al., 1985; Christensson-Nylander and Terenius, 1985; Nylander et al., 1991). Briefly, tissues were incubated at 95°C in 1 M

81 acetic acid containing 0.01% mercaptoethanol for 5 minutes, homogenized using a Brinkman Kinematic sonicator, heated again to 95°C, and subsequently centrifuged at 20,000 x g. The supernatant was then stored at -80°C until chromatographic separation. Ion exchange chromatography was performed on Sephadex C-25 columns. Samples were diluted in Buffer I (0.1 M formic acid, 0.18 M pyridine, pH 3.0) in a 1:1 (v/v) ratio and applied onto the Sephadex column. Elution fractions were obtained by serial application of 3 ml of Buffer I, Buffer II (0.1 M formic acid, 0.1 M pyridine, pH 4.3), Buffer III (0.35 M formic acid, 0.35 mM pyridine, pH 4.3), and Buffer IV (1.6 M formic acid, 1.6 M pyridine, pH 4.3). The eluent from Buffer III was used for Leu-Enk- Arg quantitation, and the eluent from Buffer IV was used for Dyn A and B. The fractions of interest were lyophilized and resuspended in a 1:1 (v/v) mixture of methanol and 0.1 M HCl and subsequently stored at 4°C. The radiolabeled peptide was generated by the chloramine T method. Radioimmunoassays for Dyn A and B were performed by incubating 25 L aliquots of each sample in duplicate or triplicate with 100 L of 125I-labeled Dyn A or Dyn B peptide and 100 L of the primary antiserum for the peptide of interest at 4°C overnight, followed by addition of 100 L of a secondary antibody, centrifugation at 1000 rpm, and scintillation counting of the resultant pellet. A similar method was used in the case of Leu-Enk-Arg, except that 200 L of charcoal solution was used instead of a secondary antibody, and 200 L of the supernatant obtained after centrifugation was used for scintillation counting to detect the 125I signal in each sample. Each experiment included a set of standards for the peptide of interest in addition to positive and negative controls, and sample values were confirmed to be within the linear range of the assay. Peptide concentrations were calculated based on the standard curve and expressed as fmol per mg of wet tissue.

Statistical analysis Using Systat 10 software (SPSS, Inc., Chicago, Illinois), effects of age, Pdyn genotype, and MPTP were assessed by ANOVA, with individual contrasts obtained by the least significant differences method. One Dyn B result in a 20-month cortical specimen was excluded due to processing error.

Results Evaluation of gene dosage on Dyn A and B expression in young adult mice reveals proportional declines in peptide levels in Pdyn HET and KO mice (Figure 5.1). The Pdyn KO specimen, included in this data set as a qualitative negative control, demonstrated absent or near-absent levels. HET mouse peptide levels are approximately half the levels seen in WT mice, with a statistical significant difference observed in both Dyn A [F(1, 9) = 6.20, p<0.034] and Dyn B [F(1, 9) = 6.26, p < 0.034]. The gene dosage effects are not apparent in frontal cortex, where expression levels even in WT mice are substantially lower than in striatum. The assay for Leu-Enk-Arg demonstrates some nonspecificity as small amounts are detected in Pdyn KO mice. In addition, HET levels of Leu-Enk-Arg are not significantly different from WT, suggesting the presence of mechanisms to compensate for the decreased gene dosage.

Striatum Cortex 3-mo WT 25 35 3-mo HET 30 3-mo KO 20 25 15 20

15 # 10 fmol/mg tissue fmol/mg fmol/mg tissue fmol/mg 10 # 5 5

0 0 Dyn A Dyn B Leu Enk- Dyn A Dyn B Leu Enk- Arg Arg

Figure 5.1. Evaluation of gene dosage on dynorphin peptide levels. Radioimmunoassay was performed on young adult mouse striatum and frontal cortex. The Pdyn KO group consists of only one specimen and is used only as a qualitative control. Statistical analysis was performed only on WT and HET mice to evaluate effects of gene dosage on peptide levels. HET mice show proportionately decreased levels of Dyn A and Dyn B in striatum, but amounts of Leu-Enk-Arg are not significantly changed. In cortex, peptide quantities are minimally altered in Pdyn HET mice. #p<0.05 compared to WT mice.

RIA analysis of dynorphin B shows a statistically significant main effect of age in striatum [F(2, 15) = 7.93, p<0.004] and a tendency toward significance in an aging main effect in cortex [F(2, 15) = 3.36, p<0.064], while Dyn A shows no significant change (Figure 2.3). Dyn B declines progressively with advancing age with striatal levels in 10- month and 20-month mice approaching approximately 50% [F(1, 15) = 6.24, p<0.025] and 20% [F(1, 15) = 15.49, p<0.001] of 3-month values, respectively. In cortex, Dyn B does not decrease significantly at 10 months, but by 20 months, levels are approximately 10% of 3-month controls [F(1, 14) = 6.72, p<0.021]. The Dyn A results contrast our

Dyn A Striatum Dyn B 35 WT KO 25 30 20 25 20 15 15 10 * 10

5 fmol/mg fmol/mg tissue fmol/mg fmol/mg tissue 5 *** 0 0 3-mo 10-mo 20-mo 3-mo 10-mo 20-mo

Cortex Dyn A Dyn B 7 WT KO 3 6 5 2 4 3

2 1 fmol/mg fmol/mg tissue 1 fmol/mg tissue * 0 0 3-mo 10-mo 20-mo 3-mo 10-mo 20-mo

Figure 5.2. RIA comparison of dynorphin peptides with age in striatum and cortex. The Pdyn KO group consists of only one specimen and is used only as a qualitative control. Dyn B showed a significant age-dependent decline with age in both striatum and frontal cortex. Dyn A shows a trend toward an age-dependent increase in cortex that was not statistically significant. *p<0.05, ***p<0.001 compared to 3-month WT controls.

84 previous experimental finding of elevated Dyn A in striatum at 24 months that was performed on a smaller sample size (n = 4). Upon pooling the previous experimental results with the newer data, this effect was no longer significant. The Dyn KO control specimen again shows near-undetectable levels of Dyn A and Dyn B. To assess short-term changes in striatal dynorphin following neurotoxic injury, Dyn A and B were quantified by RIA following a single 30 mg/kg dose of intraperitoneal MPTP and found to increase in an age-dependent manner (Figure 5.3). Two-way ANOVA for Dyn A shows a significant main effect of MPTP [(F(1, 35) = 11.16, p<0.002] but no age main effect or age-MPTP interaction. Post-MPTP Dyn A is higher in both 3-month [F(1, 35) = 4.00, p<0.053] and 10-month mice [F(1, 35) = 7.64, p<0.009], with relative increases of 58% and 85%, respectively, although the 3-month difference is slightly short of significance. Dyn B showed absent main effects of age and MPTP, but an age-MPTP interaction was noted trending toward significance [F(1, 17) =

control MPTP Dyn B Dyn A † 50 35 ‡ 30 40 25 30 20 20 15

10 fmol/mgtissue fmol/mgtissue 10 5 0 0 3-mo WT 10-mo WT 3-mo WT 10-mo WT

Figure 5.3. RIA measurement of dynorphin peptides 6 h after a single sublethal injection of MPTP. Mean Dyn A levels are increased in both young and aging mice following MPTP, but Dyn B is increased only at 10 months. †p<0.05, ‡p<0.01 compared to corresponding saline control.

4.00, p<0.062]. MPTP produced an increase in Dyn B only among 10-month mice [F(1, 17) = 6.23, p<0.023], yielding an approximately 3.3-fold elevation compared to 10- month saline controls. RIA was also used to evaluate dynorphin peptide expression 7 days after a standard intraperitoneal MPTP regimen (Figure 5.4) that has been previously noted to produce significant dopamine depletion. Pdyn KO mice, which were included in this study only as qualitative controls and were not incorporated into the ANOVA model, showed absent to near-absent levels of Dyn A and B. Three-month HET mice were also used to evaluate the effect of gene dosage on dynorphin peptide levels post-MPTP. Three-way ANOVA showed significant omnibus effects for both Dyn A [F(5, 44) = 3.09, p<0.018] and Dyn B [F(5, 26) = 5.23, p<0.002]. WT mice showed a decrease in Dyn A levels to approximately half that of saline controls, but the difference reached statistical significance only among 3-month mice [F(1, 44) = 4.85, p<0.033], with only near- significant differences noted in 10-month mice [F(1, 44) = 3.94, p<0.053]. As noted in the previous analysis, the gene dosage effect was also apparent when comparing 3-month WT and HET saline controls, but no genotype or age effect was seen in Dyn A among MPTP-injected mice.

Dyn A Dyn B 30 25 control MPTP 25 20 20 15 15 † # 10 ## **

10 ‡ fmol/mgtissue fmol/mgtissue 5 5 0 0 3-mo 3-mo 3-mo 10-mo 10-mo 3-mo 3-mo 3-mo 10-mo 10-mo WT HET KO WT KO WT HET KO WT KO

Figure 5.4. RIA quantitation of dynorphin peptides 7 days following intraperitoneal MPTP. Dyn A and B are decreased after MPTP. Age and Pdyn genotype effects among saline controls in this analysis are consistent with findings noted in the preceding figures. #p<0.05, ##p<0.01 compared to WT mice; *p<0.05, **p<0.01 compared to 3-month mice; †p<0.05, ‡p<0.01 compared to corresponding saline controls.

Striatal Dyn B in WT mice also exhibited a substantial decline 7 days following MPTP [F(1, 26) = 12.20, p<0.02], to levels approaching 24% of saline controls. Gene dosage effects and age effects were also present among saline-injected mice and consistent with the preceding analysis, with 3-month HET mice showing decreased Dyn B compared to 3-month WT mice [F(1, 26) = 10.23, p<0.004] and 10-month WT showing declines in Dyn B compared to 3-month controls [F(1, 26) = 8.26, p<0.008]. However, among MPTP-injected mice, no significant effects were observed.

Discussion In this study, a radioimmunological assay was used to assess dynorphin peptide expression in striatum and frontal cortex of mice. The internal validity of this study is supported by our results confirming the analytical specificity of the assay and the essentially linear quantitative relationship between gene dosage and peptide concentration for the Dyn A and Dyn B peptides in striatum, although these properties are less apparent in the data on Leu-Enk-Arg and the data in frontal cortical specimens. In examining dynorphin peptide expression in striatum and frontal cortex with aging, we find that aging produces progressive decline in Dyn B while Dyn A shows no significant change. The existing literature on dynorphin alterations with aging contain numerous reports of increased dynorphin in aging or aging-associated pathological processes, including elevated mRNA or peptide concentrations in the hippocampus of normal aged rats (Jiang et al., 1989; Zhang et al., 1991; Gallagher and Nicolle, 1993; Kotz et al., 2004), the cortex and hippocampus of amyloid precursor protein- overexpressing mice (Diez et al., 2000), the frontal cortex of Down’s syndrome patients (Risser et al., 1996), and the frontal and parietal cortices of Alzheimer’s disease patients (Risser et al., 1996; Yakovleva et al., 2006a). In striatum, Pdyn expression with aging has been less well studied, with some studies showing no changes in aging rats (Zhang et al., 1991; Kotz et al., 2004) and one study finding decreased Pdyn mRNA in striatum of normal aged humans (Backman et al., 2007). We previously had noted no significant aging effect of Dyn A in hippocampus (Nguyen et al., 2005), but did observe an increase in Dyn A in cerebral cortex in a small number of male aged WT mice (Nguyen et al.,

2005). Our current results, which incorporated data pooled from the previous study and which contain a larger sample size, show only a nonsignificant trend toward increasing Dyn A with aging. Although complete characterization of dynorphin expression during aging awaits further studies, our data and that of previous studies clearly implicate a dysregulation of dynorphin expression during aging, the consequences of which remain to be elucidated. Numerous studies have found dynorphin to be elevated following acute and chronic stressors such as inflammation (Millan et al., 1985; Sydbom and Terenius, 1985; Chahl and Chahl, 1986; Iadarola et al., 1986; Noguchi et al., 1991; Gillardon et al., 1994; Zhang et al., 2005) and neuronal injury (Faden et al., 1985a, b; Wang et al., 2001). It is plausible that the age-related dynorphin elevations seen in other studies may be a consequence of chronic subclinical episodes of neuronal injury or a manifestation of an acute stress response during the period immediately prior to euthanasia. In the present study, HPLC quantitation of dynorphin peptides after a single 30 mg/kg bolus of intraperitoneal MPTP injection in WT mice showed increased dynorphin expression 6 hours following the neurotoxic insult. While Pdyn mRNA was not explicitly measured in this study, one putative mechanism for the increased dynorphins would be upregulation of Pdyn gene expression via activation of immediate-early genes and related transcription factors (Naranjo et al., 1991; Collins-Hicok et al., 1994; Hunter et al., 1995; Kitamura et al., 1995) during the initial phases of neuronal injury following MPTP. While Dyn A is increased by MPTP in both young and middle-aged mice, the MPTP-induced increase in Dyn B is seen only among 10-month mice, favoring an age-dependent mechanism in which short-term dysregulation of Dyn B preferentially affecting aged individuals may lead to impairments in dopaminergic function. These observations are corroborated by existing literature showing that aging enhances Pdyn upregulation in response to inflammation in spinal cord models (Zhang et al., 2004a). The above findings remarkably contrast with the decreased expression of both Dyn A and Dyn B seen at a longer time interval of 7 days post-MPTP. Since this MPTP regimen in these mice produces substantial loss of striatal dopamine and because dopamine promotes dynorphin expression via the D1-receptor (Li et al., 1986; Li et al., 1988; Li et al., 1990; Bordet et al., 1997; Wong et al., 2003), the observed effects are

88 likely a consequence of decreased dopamine D1-receptor activity in a manner analogous to that seen in experimental nigrostriatal lesions (Gerfen et al., 1991) or in D1-receptor knockout mice (Wong et al., 2003). The significance of the observed changes in dynorphin peptide levels in aging is still unclear. While compensatory responses or changes in the enzymatic systems responsible for preprodynorphin processing are likely contributing factors, further studies are needed to elucidate the causes and consequences of the observed alterations in dynorphin peptide levels. Age-related patterns of gene expression, such as those described in the following chapter, may help to explain why these neuropeptide transmitters are altered by advanced age.

CHAPTER 6: Microarray analysis of hippocampal gene expression changes with aging and prodynorphin deletion

Introduction Endogenous dynorphins derived from the product of the prodynorphin (Pdyn) gene have been found to have diverse functional roles in a variety of organ systems (Simonato and Romualdi, 1996; Hauser et al., 2005). Dynorphin peptides present in the hypothalamus, dentate gyrus, striatum, cortex, and hippocampus exert a multitude of diverse biological actions, including effects on pain transmission, seizure activity, learning and memory, endocrine function, excitotoxic injury, stress responses, and the immune system [reviewed in (Smith and Lee, 1988)]. The mechanisms through which dynorphin and related peptides exert many of their effects involve the opioid receptor (Goldstein et al., 1979; Chavkin et al., 1982) (Goldstein and Naidu, 1989) and other opioid (Garzon et al., 1984) and non-opioid receptors (Chen et al., 1995a, b; Chen and Huang, 1998; Tang et al., 1999). However, the detailed mechanisms for many of the downstream mediators of dynorphin activity have not been fully elucidated. The inhibitory actions of dynorphin peptides on neurotransmission have been reported to be mediated by depressing calcium entry into neurons (Cherubini and North, 1985) via inhibition of voltage-gated calcium channels in a manner similar to and opioid agonists [reviewed in (Satoh and Minami, 1995)]. However dynorphins can increase calcium influx via interaction with the bradykinin receptor in models of neuropathic pain (Lai et al., 2006). There have also been varying reports in the literature regarding the ability of dynorphins and agonists in general to inhibit adenylate cyclase activity in a manner analogous to that of agonists (Satoh and Minami, 1995). Other downstream effects of dynorphins include activation of NMDA receptors and inflammatory signaling cascades in the case of spinal hyperalgesia (Koetzner et al., 2004). In addition, dynorphnis interact closely with multiple non- neuronal cells and with the endocrine system (Sherman et al., 1986; Currie et al., 1994; Ohnishi et al., 1994; Spampinato et al., 1995; Kreek et al., 1999; Goodman et al., 2004; Foradori et al., 2005; Jacobson et al., 2006), suggesting that Pdyn deletion is likely to

90 produce fairly broad gene expression changes, especially in a process as pervasive and multifactorial as normal aging. Early reports that prodynorphin is elevated in the hippocampus and frontal cortex in cognitively impaired subpopulations of aged rats have implicated Pdyn-derived peptides in various aspects of aging (Jiang et al., 1989; Zhang et al., 1991). In vivo studies by other laboratories had demonstrated impaired cognitive function in response to administration of various dynorphin peptides (Introini-Collison et al., 1987; Sandin et al., 1998), but the mechanistic basis for this phenomenon is currently unclear and may range from suppressive effects of dynorphin at the glutamatergic synapse to direct toxic effects of some of the dynorphin peptides (Hauser et al., 2005). Based on our experimental findings that aged knockout (KO) mice lacking the Pdyn gene perform relatively better than comparably aged wild-type (WT) mice (Nguyen et al., 2005) and other observations as noted in preceding chapters, we seek in this study to evaluate patterns of alterations in hippocampal gene expression with aging and with Pdyn deletion to explore possible mechanisms to explain the role of dynorphins in cognitive decline. Because of the broad range of interactions of prodynorphin-derived peptides with a variety of cell types and biochemical pathways, we initiated the current microarray study to enable systematic characterization of patterns of differential gene expression without a priori assumptions on any particular gene or set of genes. Microarray approaches have previously been used to assess gene expression profiles with aging and cognitive impairment (Blalock et al., 2003; Blalock et al., 2004). By comparing Pdyn KO mice to similarly aged wild-type (WT) mice, we expect to identify the molecular mechanisms and intermediary processes by which the chronic absence of Pdyn-derived peptides may contribute to the observed functional manifestations in mice.

Methods Animal subjects Pdyn KO mice, generated by deletion of exons 3 and 4 of the Pdyn gene (Sharifi et al., 2001), and wild-type (WT) 129SvEv-Tac were bred and housed in an approved animal facility. After weaning, genotyping was performed by the polymerase chain reaction

(PCR) from DNA isolated from mouse tail tissue using primer pairs specific for each genotype. Microarray analyses were conducted in two separate experiments. Experiment A included young adult (3-month) WT and Pdyn KO male mice in addition to a separate set of 22-month WT and Pdyn KO male mice (n = 3 per group). In Experiment B, only aged (22- to 26-month) WT (n = 3) and aged Pdyn KO (n = 3) mice were used, with each group consisting of 2 females and 1 male. In each experiment, fresh hippocampal specimens were dissected from CO2-asphyxiated mice and stored at –80°C until processing. All procedures involving animals are approved by the Institutional Animal Care and Use Committee at the University of Kentucky.

Microarray processing Total RNA was extracted using the Trizol reagent (Gibco BRL) and reconstituted in de- ionized water treated with diethyl pyrocarbonate (DEPC). RNA specimens were quantitated by optical density and normalized, followed by assessment of molecular integrity by electrophoresis. Samples were then submitted to the University of Kentucky Microarray Core Facility for labeling and hybridization onto individual microarray chips. Murine MOE430A cDNA microarray chips, with 22,690 probe sets each, were used in Experiment A, while murine whole-genome MG U74A version 2 chips, each containing 12,488 probe sets, were used for Experiment B. Automated detection of fluorescent intensity using an Affymetrix GeneArray scanner was performed according to standard manufacturer protocols, yielding a quantitative score for each probe set.

Probe set list selection The automated detection system assigns each probe set in each chip a numerical signal value and an absent-present designation of absent (A), present (P), or marginal (M). For each probe set, the absent-present score for each group of 3 replicates was calculated by first assigning a value of 0 for each A designation, 1 for each P designation, and 0.5 for each M designation and summing the total for the group. Within a given experiment, probe sets for which all groups have an absent-present score of less than 1.5 were deemed to be absent in a majority of replicates in each group and were excluded from further analysis. Next, lists of probe sets were generated to reflect increases or decreases in

92 mean signal values in each of the planned comparisons. Experiment A includes the following comparisons: young Pdyn KO vs. young WT, aged Pdyn KO vs. aged WT, aged WT vs. young WT, and aged Pdyn KO vs. young Pdyn KO. In Experiment B, aged Pdyn KO mice are compared to aged WT mice. An ANOVA p-value was computed for each comparison in Experiment A and a t-statistic p-value was computed for each comparison in Experiment B. Probe set lists for each of the planned comparisons were subsequently generated by selecting non-absent probe sets in which the p-value for the associated comparison is less then 0.05.

Functional categorization of differentially regulated genes A software program known as Expression Analysis Systematic Explorer (EASE) (Hosack et al., 2003) was used to systematically identify functional categories of genes, based on Biological Process and Molecular Function annotation data in existing Gene Ontology (GO) databases (Ashburner et al., 2000), that are over-represented within a given list of Affymetrix probe sets. The software program evaluates for each categorical grouping in the GO classification system, the number of probe sets in the input list that under the existing annotation database belong to the category being queried, thereby obtaining the number of list hits. A similar process is applied to a reference population, in this case, the entire list of probe sets available on the gene chip used, to calculate the number of population hits. By comparing the proportion of list hits to that of population hits, the program computes a conservative modification of the Fisher's exact statistic known as the EASE score to quantify the probability that over-representation of the identified category within the list might occur due to chance. For each of the planned experimental comparisons, functional groupings that were noted to be significantly over-represented based on an EASE score of less than 0.05 are presented. Functional annotations and other descriptors for each probe set were obtained using NetAffx Analysis Center software (Affymetrix) in conjunction with GenBank, Swiss-Prot, and related databases, but only GO Biological Process and Molecular Function annotations were used in the systematic over-representation analysis. All p- values reported in the Results section refer to the EASE score for each specified categorical identifier.

Results Gene expression in WT and Pdyn KO mice were compared to identify genes that show altered expression with either aging or Pdyn deletion. Table 6.1 describes the results of the filtering process by which probe set lists were generated. The results of the over- representation analysis using biological process annotations are listed in Table 6.2 through Table 6.15, and the results of an analogous analysis applied to molecular function identifiers are presented in Table 6.16 through Table 6.29.

Effects of aging We first examined probe sets that show a significant main effect of age in Experiment A and analyzed them initially without regard to Pdyn genotype using biological process (Table 6.2 and Table 6.3) and molecular function identifiers (Table 6.16 and Table 6.17). We find that 44 of the 251 genes downregulated with aging are involved in nervous system development, which is significantly higher than the approximately 10% prevalence of this identifier in the murine genomic background (p<0.0004). Similarly other downregulated gene categories include those involved in neurogenesis (p<0.017), mesenchymal cell differentiation and development (p<0.025), stem cell division (p<0.042), cell growth (p<0.047), potassium transport (p<0.012), DNA binding (p<0.003), and zinc-mediated transcription factor activity (p<0.034). The prevalence of genes identified as involved in learning and memory is also disproportionately high in the age-downregulated probe set list (p<0.004), as are processes such as action potential regulation (p<0.026), synaptic transmission (p<0.026), and dopamine receptor signaling (p<0.029). In contrast, gene categories upregulated by aging include a number of processes associated with immune response (p<0.00006), complement activation (p<0.002), acute- phase response (p<0.024), response to pathogens (p<0.005), and antigen presentation or processing (p<0.003). Cell development genes are also found to be over-represented among age-upregulated probe sets (p<0.03). Age effects, when evaluated among WT mice only, provide similar results with regard to biological process identifiers (Table 6.4 and Table 6.5) and molecular function

(Table 6.18 and Table 6.19). Downregulated gene categories include nervous system development (p<0.004) and neurite morphogenesis (p<0.001), as observed in the preceding analysis. In addition, we also find downregulation of genes involved in protein phosphorylation (p<0.04), nucleosome assembly (p<0.038), cellular morphogenesis (p<0.003), protein transporter activity (p<0.004), and phospho-protein binding (p<0.004). Aging in WT mice is also associated with downregulation of numerous cellular processes (p<0.026) that broadly include such functions as cell cycle control, protein transport, exocytosis, and chromosome organization. Genes upregulated in aged WT mice compared to young WT mice include those involved in neuron development (p<0.023), cell death (p<0.046), cell maturation (p<0.043), cell growth (p<0.021), protein transport (p<0.007), and calcium binding (p<0.0001). In addition, upregulated genes also have a disproportionately high prevalence of genes identified as being involved in negative regulation of cellular processes (p<0.022). There are relatively fewer genes downregulated in aged Pdyn KO mice relative to young KO mice (Table 6.6 and Table 6.20), such as those involved enzyme-linked receptor signaling pathways (p<0.036) and potassium binding (p<0.005), while age-upregulated genes in Pdyn KO mice (Table 6.7 and Table 6.21) are associated with MHC I-mediated endogenous antigen processing (p<0.007), protein binding (p<0.019), RNA binding (p<0.016), and RNA splicing (p<0.042).

Effects of Pdyn deletion When Experiment A is analyzed without regard to age, main effects of Pdyn deletion are present in multiple probe sets. Genes downregulated with Pdyn deletion (Table 6.8 and Table 6.22) are involved in macromolecule metabolism (p<0.0006), including processes related to RNA binding (p<0.0001), protein synthesis (p<0.015), protein folding (p<0.019). Electron transport-related processes (p<0.04) and ubiquinol-cytochrome C reductase activity (p<0.008) are also over-represented among genes downregulated by Pdyn deletion. Biological process categories for genes upregulated by Pdyn deletion (Table 6.9) include neuron differentiation (p<0.026) and regulation of action potential (p<0.039), while upregulated molecular function categories (Table 6.23) include ester- hydrolase activity (p<0.016) and phospho-transferase activity (p<0.036).

When probe sets were selected based on significant group contrasts in Experiment A, Pdyn deletion among young mice was associated with changes in expression of a relatively small number of biological process (Table 6.10 and Table 6.11) and molecular function categories (Table 6.24 and Table 6.25). The only statistically significant biological process category over-represented among genes downregulated by Pdyn deletion is protein metabolism (p<0.043), but upregulated biological process identifiers include cellular biosynthesis (p<0.044) and negative regulation of apoptosis (p<0.021). Calmodulin-dependent protein kinase activity is also downregulated by Pdyn deletion in young mice (p<0.010), whereas calcium ion binding (p<0.015) and tyrosine kinase activity (p<0.043) are among the molecular function identifiers of upregulated genes. Among the aged mice of Experiment A, Pdyn deletion is associated with downregulation of genes associated with cell growth (p<0.038), vesicle-mediated transport (p<0.038), cell differentiation (p<0.044), nervous system development (p<0.050), calcium binding (p<0.0001), and potassium binding (p<0.013) (Table 6.12 and Table 6.26). In addition, Pdyn deletion was also associated with decreased expression of genes identified in the category of negative regulation of cellular processes (p<0.046). Pdyn deletion among aged mice causes upregulation of genes involved in neurite morphogenesis (p<0.002) (Table 6.13). In addition, aged Pdyn KO mice also show increased expression of genes involved in positive regulation of transcription (p<0.028), phosphoprotein binding (p<0.039), transcription coactivator activity (p<0.044), and sequence-specific DNA binding (p<0.006) compared to aged WT mice (Table 6.13 and Table 6.27). Experiment B, which compares aged WT and Pdyn KO mice in a separate group of animals, found upregulation by Pdyn deletion of genes involved in Golgi organization and biogenesis (p<0.045) and organelle localization (p<0.025) (Table 6.15). Furthermore, 14 of the 117 annotated genes upregulated by Pdyn deletion possess ion channel activity (p<0.003), with a majority of these specifically noted to demonstrate voltage-gated ion channel activity (p<0.021) (Table 6.29). The only downregulated biological process category was the proteasome activator complex (p<0.033) (Table 6.14), and molecular function identifiers among genes downregulated in Pdyn KO mice

96 similarly include proteasome activator activity (p<0.031) and NADH dehydrogenase activity (p<0.047) (Table 6.28).

Discussion In this study, we implement a modification of the Fisher exact analysis to identify functional categories of genes upregulated or downregulated by either aging or Pdyn deletion using the EASE program. Analysis of microarray data often is problematic because statistical concerns can hinder interpretation of individual results when obtained among a very large number of potential comparisons (Blalock et al., 2003). However, when co-regulated groups of genes are assessed, microarray data may yield abundant information on patterns of altered gene expression, which tend to be more robust and less dependent on individual gene alterations (Ashburner et al., 2000). Our findings of increased immune response genes and decreased neurogenesis or axonogenesis genes with age are corroborated by previously published microarray studies in rats (Rowe et al., 2007). These observations may reflect impairments in regulation of various pathways that may increase susceptibility to various pathological outcomes with advancing age. Genes involved in dopamine receptor signaling are also downregulated with aging, consistent with known functional declines in the dopaminergic system with age (Kish et al., 1992; Ingram, 2000; Reeves et al., 2002). In this study, we use two separate microarray experiments to evaluate effects of Pdyn deletion on gene expression. While the biological process identifiers associated with up- and downregulated genes vary between the two experiments, Pdyn-deletion- associated findings consistently seen in both experiments include a decrease in protein metabolism genes, an increase in cellular organization and biogenesis genes, and an increase in genes related to ion transport. The effects of Pdyn deletion in aged mice, which include increased expression of genes associated with neurite development and general cell metabolic processes such as transcription, suggest a more active and functional neuronal environment in the aging hippocampus that may explain previously noted amelioration of memory deficits by Pdyn deletion (Nguyen et al., 2005). Similar changes were observed in young mice, in which Pdyn deletion produces upregulation of

97 genes involved in a broad range of developmental processes, cellular physiological processes, signal transduction, and suppression of apoptosis. These observations support a generally suppressive effect of dynorphin peptides on cellular functioning in the hippocampus. One downstream consequence of dynorphin activity is opiate-receptor-mediated inhibition of voltage-gated calcium channels (Satoh and Minami, 1995), resulting in decreased intracellular calcium (Cherubini and North, 1985). Our findings in Experiment B that Pdyn deletion increases expression of genes involved in ion channel activity, with over half of these specifically listed as voltage-gated ion channel genes, suggest that control of intracellular calcium may represent an important mechanism responsible for tonic opioid receptor-mediated suppression of calcium-dependent neurotransmitter release by dynorphin peptides, consistent with electrophysiological studies (Spadoni et al., 2004). Experiment A also reveals that, among genes downregulated by Pdyn deletion in aged mice, close to 20% are involved in calcium ion binding. In young mice, Pdyn deletion produces downregulation of genes with calcium-dependent protein kinase activity. Altered calcium regulation therefore appears to be an important reproducible consequence of Pdyn deletion. This microarray analysis was performed on whole hippocampal homogenates. Therefore, results are based on expression profiles of tissue specimens comprised of a hetereogeneous cellular composition. Differences in the relative proportion of neurons or glial cells within the hippocamps or variations in expression patterns among the different subpopulations or cell types may potentially skew the resulting data. The sensitivity with which gene expression differences may be detected may similarly be reduced for genes whose expression is altered only in a small subpopulation of cells within the hippocampus. Even with these limitations, we were able to detect a substantial number of genes that, within a specified degree of statistical certainty, show altered expression with either aging or Pdyn deletion. Importantly, this analysis evaluates groups of genes with similar patterns of changes in expression and circumvents the need to generate conclusions regarding specific probe sets. Although identification of individual genes may prove helpful in focused experiments with well-planned a priori hypotheses, attempts at independent

98 validation of microarray results would be unlikely to yield a high concordance rate unless very large numbers of replicates are tested and exceedingly stringent statistical criteria are used in probe set selection. Therefore, in light of the statistical obstacles associated with individual probe set data, we focused exclusively in this study on gene categories. Furthermore, the results in the tables below contain all statistically significant annotation groupings identified in an automated unbiased manner solely using existing annotation categories within an established ontological database, thereby avoiding subjective biases associated with manual data-mining approaches. It should be emphasized that the accuracy of this type of analysis is limited by the completeness of the available annotation systems and may be expected to improve as bioinformatic databases evolve with further additions to existing knowledge on the biochemical, molecular, and cellular processes associated with specific genes.

Table 6.1. Summary of probe set filtering and selection results. Probe sets were excluded based on insufficient presence-absence calls or inadequate statistical significance of inter-group differences. Only non-absent probe sets with significant inter-group differences by ANOVA (Experiment A) or t-test (Experiment B) were included.

Experiment A – Aging effects Total number of probe sets in chip 22690 Non-absent probe sets 14412 Non-absent probe sets with aged mice > young mice 7374 Non-absent probe sets with aged mice < young mice 7034 Non-absent probe sets with aged WT > young WT 7428 Non-absent probe sets with aged WT < young WT 6983 Non-absent probe sets with aged Pdyn KO > young Pdyn KO 7372 Non-absent probe sets with aged Pdyn KO < young Pdyn KO 7034 Non-absent probe sets with aged mice > young mice and p<0.05 by ANOVA 313 Non-absent probe sets with aged mice < young mice and p<0.05 by ANOVA 313 Non-absent probe sets with aged WT > young WT and p<0.05 by ANOVA 226 Non-absent probe sets with aged WT < young WT and p<0.05 by ANOVA 266 Non-absent probe sets with aged Pdyn KO > young Pdyn KO and p<0.05 by ANOVA 54 Non-absent probe sets with aged Pdyn KO < young Pdyn KO and p<0.05 by ANOVA 92

Experiment A – Pdyn effects Total number of probe sets in chip 22690 Non-absent probe sets 14412 Non-absent probe sets with Pdyn KO > WT 7510 Non-absent probe sets with Pdyn KO < WT 6898 Non-absent probe sets with young Pdyn KO > young WT 7291 Non-absent probe sets with young Pdyn KO < young WT 7115 Non-absent probe sets with aged Pdyn KO > aged WT 7472 Non-absent probe sets with aged Pdyn KO < aged WT 6938 Non-absent probe sets with Pdyn KO > WT and p<0.05 by ANOVA 187 Non-absent probe sets with Pdyn KO < WT and p<0.05 by ANOVA 152 Non-absent probe sets with young Pdyn KO > young WT and p<0.05 by ANOVA 113 Non-absent probe sets with young Pdyn KO < young WT and p<0.05 by ANOVA 107 Non-absent probe sets with aged Pdyn KO > aged WT and p<0.05 by ANOVA 189 Non-absent probe sets with aged Pdyn KO < aged WT and p<0.05 by ANOVA 186

Experiment B – Pdyn effects in aged mice Total number of probe sets in chip 12488 Non-absent probe sets 7487 Nonabsent probe sets upregulated in Pdyn KO 4064 Nonabsent probe sets downregulated in Pdyn KO 3423 Nonabsent probe sets upregulated in Pdyn KO and p<0.05 by t-test 149 Nonabsent probe sets downregulated in Pdyn KO and p<0.05 by t-test 116

100

Table 6.2. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated with aging in Experiment A. Terms used in the following tables reflect those used in the Gene Ontology Biological Process database. Arrows precede categories that are subsets of one or more categories listed above them, with increasing numbers of arrows denoting more specific categorizations. List hits and population hits are presented along with the corresponding totals. Note that not all entries within a given probe set list are reflected in the total, since some probe sets do not have a corresponding annotation in the ontological system. List hits / Population hits / EASE Biological process category identifier List total Population total score SYSTEM DEVELOPMENT 44 / 251 1114 / 10541 0.0009 → NERVOUS SYSTEM DEVELOPMENT 44 / 251 1078 / 10541 0.0004 →→ CENTRAL NERVOUS SYSTEM DEVELOPMENT 19 / 251 468 / 10541 0.0297 ANATOMICAL STRUCTURE FORMATION 10 / 251 169 / 10541 0.0193 → FORMATION OF PRIMARY GERM LAYER 10 / 251 167 / 10541 0.0180 MESODERM DEVELOPMENT 10 / 251 198 / 10541 0.0463 → MESODERM MORPHOGENESIS 10 / 251 167 / 10541 0.0180 →→ MESODERM FORMATION 10 / 251 166 / 10541 0.0174 CELL DEVELOPMENT 25 / 251 628 / 10541 0.0141 → NEUROGENESIS 24 / 251 605 / 10541 0.0170 →→ NEURON DIFFERENTIATION 22 / 251 562 / 10541 0.0261 →→→ NEURON MATURATION 11 / 251 204 / 10541 0.0235 →→ NEURON MIGRATION 5 / 251 51 / 10541 0.0323 → GERM CELL DEVELOPMENT 11 / 251 182 / 10541 0.0114 MESENCHYMAL CELL DIFFERENTIATION 9 / 251 149 / 10541 0.0255 → MESENCHYMAL CELL DEVELOPMENT 9 / 251 149 / 10541 0.0255 HINDLIMB MORPHOGENESIS 8 / 251 139 / 10541 0.0474 → EMBRYONIC HINDLIMB MORPHOGENESIS 8 / 251 138 / 10541 0.0459 GROWTH 19 / 251 494 / 10541 0.0464 CELLULAR MORPHOGENESIS 28 / 251 815 / 10541 0.0478 REGULATION OF CELL SIZE 16 / 251 396 / 10541 0.0500 → CELL GROWTH 16 / 251 393 / 10541 0.0473 CORTICAL CYTOSKELETON ORGANIZATION AND BIOGENESIS 3 / 251 11 / 10541 0.0268 STEM CELL DIVISION 8 / 251 135 / 10541 0.0416 LOCALIZATION 99 / 251 3265 / 10541 0.0032 → ESTABLISHMENT OF LOCALIZATION 98 / 251 3243 / 10541 0.0038 →→ TRANSPORT 84 / 251 2865 / 10541 0.0196 →→→ ION TRANSPORT 32 / 251 864 / 10541 0.0131 →→→→ CATION TRANSPORT 25 / 251 624 / 10541 0.0131 →→→→→ METAL ION TRANSPORT 24 / 251 504 / 10541 0.0020 →→→→→ MONOVALENT INORGANIC CATION TRANSPORT 17 / 251 399 / 10541 0.0282 →→→→→→ POTASSIUM ION TRANSPORT 14 / 251 269 / 10541 0.0118 COFACTOR BIOSYNTHESIS 13 / 251 295 / 10541 0.0485 → COENZYME BIOSYNTHESIS 13 / 251 278 / 10541 0.0332 RIBONUCLEOTIDE BIOSYNTHESIS 12 / 251 249 / 10541 0.0351 → PURINE NUCLEOTIDE BIOSYNTHESIS 12 / 251 245 / 10541 0.0318 PURINE NUCLEOTIDE METABOLISM 12 / 251 252 / 10541 0.0378 RIBONUCLEOTIDE METABOLISM 12 / 251 260 / 10541 0.0456 REGULATION OF GENE EXPRESSION, EPIGENETIC 10 / 251 200 / 10541 0.0488 NEUROPHYSIOLOGICAL PROCESS 42 / 251 1182 / 10541 0.0078 → TRANSMISSION OF NERVE IMPULSE 25 / 251 653 / 10541 0.0215 →→ REGULATION OF ACTION POTENTIAL 10 / 251 178 / 10541 0.0259 CELL COMMUNICATION 101 / 251 3338 / 10541 0.0030 → SIGNAL TRANSDUCTION 90 / 251 3032 / 10541 0.0105 → SYNAPTIC TRANSMISSION 23 / 251 597 / 10541 0.0265 →→ NEUROMUSCULAR SYNAPTIC TRANSMISSION 8 / 251 133 / 10541 0.0389 → ENZYME LINKED RECEPTOR PROTEIN SIGNALING PATHWAY 21 / 251 558 / 10541 0.0427

101

List hits / Population hits / EASE Biological process category identifier List total Population total score (continued from previous page) → DOPAMINE RECEPTOR SIGNALING PATHWAY 9 / 251 153 / 10541 0.0292 BEHAVIOR 25 / 251 567 / 10541 0.0042 → LEARNING AND/OR MEMORY 8 / 251 84 / 10541 0.0037 RHYTHMIC PROCESS 10 / 251 200 / 10541 0.0488 → CIRCADIAN RHYTHM 9 / 251 162 / 10541 0.0391

Table 6.3. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated with aging in Experiment A. List hits / Population hits / EASE Biological process category identifier List total Population total score DEVELOPMENT 77 / 229 2974 / 10541 0.0499 → CELL DIFFERENTIATION 44 / 229 1382 / 10541 0.0082 →→ CELL DEVELOPMENT 22 / 229 628 / 10541 0.0310 CELLULAR MACROMOLECULE METABOLISM 77 / 229 2930 / 10541 0.0367 RESPONSE TO STIMULUS 75 / 229 2723 / 10541 0.0142 → RESPONSE TO STRESS 49 / 229 1635 / 10541 0.0148 → RESPONSE TO BIOTIC STIMULUS 49 / 229 1400 / 10541 0.0007 →→ RESPONSE TO OTHER ORGANISM 34 / 229 984 / 10541 0.0072 →→→ RESPONSE TO PEST, PATHOGEN OR PARASITE 34 / 229 965 / 10541 0.0054 → IMMUNE RESPONSE 48 / 229 1231 / 10541 0.0001 →→ HUMORAL IMMUNE RESPONSE 16 / 229 409 / 10541 0.0321 →→→ COMPLEMENT ACTIVATION 6 / 229 43 / 10541 0.0022 →→→→ COMPLEMENT ACTIVATION, CLASSICAL PATHWAY 4 / 229 31 / 10541 0.0288 → DEFENSE RESPONSE 48 / 229 1346 / 10541 0.0005 → RESPONSE TO EXTERNAL STIMULUS 32 / 229 978 / 10541 0.0194 →→ RESPONSE TO WOUNDING 25 / 229 774 / 10541 0.0467 →→→ ACUTE-PHASE RESPONSE 5 / 229 51 / 10541 0.0241 ANTIGEN PRESENTATION 7 / 229 65 / 10541 0.0027 → ANTIGEN PRESENTATION, ENDOGENOUS ANTIGEN 4 / 229 31 / 10541 0.0288 ANTIGEN PROCESSING 7 / 229 52 / 10541 0.0008 → EXOGENOUS ANTIGEN PROCESSING VIA MHC CLASS II 3 / 229 12 / 10541 0.0267 ORGANISMAL PHYSIOLOGICAL PROCESS 70 / 229 2537 / 10541 0.0181 → SMOOTH MUSCLE CONTRACTION 7 / 229 106 / 10541 0.0273 CELL-CELL SIGNALING 31 / 229 964 / 10541 0.0266 REPRODUCTION 24 / 229 733 / 10541 0.0457 → SEXUAL REPRODUCTION 21 / 229 608 / 10541 0.0402

Table 6.4. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated in aged WT mice compared to young WT mice in Experiment A. List hits / Population hits / EASE Biological process category identifier List total Population total score SYSTEM DEVELOPMENT 35 / 210 1114 / 10541 0.0069 → NERVOUS SYSTEM DEVELOPMENT 35 / 210 1078 / 10541 0.0042 NEUROGENESIS 22 / 210 605 / 10541 0.0088 →NEURON DIFFERENTIATION 20 / 210 562 / 10541 0.0160 → →NEURON DEVELOPMENT 19 / 210 443 / 10541 0.0031 →→→ NEURON MORPHOGENESIS DURING DIFFERENTIATION 18 / 210 377 / 10541 0.0013 →→→→ NEURITE MORPHOGENESIS 18 / 210 377 / 10541 0.0013

102

List hits / Population hits / EASE Biological process category identifier List total Population total score (continued from previous page) →→→→→ AXONOGENESIS 15 / 210 348 / 10541 0.0094 →→→→→ DENDRITE MORPHOGENESIS 8 / 210 157 / 10541 0.0367 CELLULAR PROCESS 199 / 210 9577 / 10541 0.0261 → REGULATION OF CELLULAR PROCESS 83 / 210 3535 / 10541 0.0469 →→ POSITIVE REGULATION OF CELLULAR PROCESS 32 / 210 1111 / 10541 0.0313 →→ REGULATION OF TRANSPORT 10 / 210 217 / 10541 0.0289 → REGULATION OF PROTEIN STABILITY 6 / 210 52 / 10541 0.0036 →→ PROTEIN STABILIZATION 3 / 210 16 / 10541 0.0391 → MITOTIC CELL CYCLE 18 / 210 549 / 10541 0.0454 → INTERPHASE 12 / 210 304 / 10541 0.0397 →→ INTERPHASE OF MITOTIC CELL CYCLE 12 / 210 303 / 10541 0.0389 →→→ G1/S TRANSITION OF MITOTIC CELL CYCLE 11 / 210 214 / 10541 0.0102 → CELLULAR LOCALIZATION 34 / 210 1109 / 10541 0.0112 →→ ESTABLISHMENT OF CELLULAR LOCALIZATION 34 / 210 1102 / 10541 0.0103 →→→ EXOCYTOSIS 7 / 210 129 / 10541 0.0432 →→→ INTRACELLULAR TRANSPORT 33 / 210 1079 / 10541 0.0131 →→→ PROTEIN TRANSPORT 30 / 210 1000 / 10541 0.0236 →→→→ INTRACELLULAR PROTEIN TRANSPORT 24 / 210 788 / 10541 0.0394 →→→ NUCLEOCYTOPLASMIC TRANSPORT 14 / 210 386 / 10541 0.0436 →→→→ REGULATION OF NUCLEOCYTOPLASMIC TRANSPORT 5 / 210 62 / 10541 0.0343 → PROTEIN LOCALIZATION 30 / 210 1063 / 10541 0.0471 →→ ESTABLISHMENT OF PROTEIN LOCALIZATION 30 / 210 1034 / 10541 0.0347 → NUCLEOTIDE METABOLISM 15 / 210 386 / 10541 0.0215 →→ NUCLEOTIDE BIOSYNTHESIS 13 / 210 309 / 10541 0.0202 → CHROMOSOME ORGANIZATION & BIOGENESIS (SENSU EUKARYOTA) 16 / 210 467 / 10541 0.0448 →→ ESTABLISHMENT AND/OR MAINTENANCE OF CHROMATIN ARCHITECTURE 15 / 210 381 / 10541 0.0195 → CELL-CELL ADHESION 10 / 210 239 / 10541 0.0485 → INTRACELLULAR SIGNALING CASCADE 43 / 210 1545 / 10541 0.0188 → PROTEIN COMPLEX ASSEMBLY 15 / 210 410 / 10541 0.0336 → CELL ORGANIZATION AND BIOGENESIS 69 / 210 2289 / 10541 0.0002 →→ CELLULAR MORPHOGENESIS 28 / 210 815 / 10541 0.0057 → CELL DIFFERENTIATION 38 / 210 1382 / 10541 0.0336 →→ CELL DEVELOPMENT 24 / 210 628 / 10541 0.0032 →→→CELLULAR MORPHOGENESIS DURING DIFFERENTIATION 18 / 210 407 / 10541 0.0030 GAMETOGENESIS 18 / 210 555 / 10541 0.0494 PRIMARY METABOLISM 133 / 210 6037 / 10541 0.0471 BIOPOLYMER METABOLISM 67 / 210 2737 / 10541 0.0387 → BIOPOLYMER MODIFICATION 54 / 210 1800 / 10541 0.0014 →→ PROTEIN MODIFICATION 50 / 210 1728 / 10541 0.0049 →→→ PROTEIN AMINO ACID PHOSPHORYLATION 22 / 210 708 / 10541 0.0415 →→ BIOPOLYMER METHYLATION 9 / 210 197 / 10541 0.0424 → DNA PACKAGING 15 / 210 390 / 10541 0.0233 →→ CHROMATIN ASSEMBLY OR DISASSEMBLY 7 / 210 112 / 10541 0.0239 →→→ NUCLEOSOME ASSEMBLY 5 / 210 64 / 10541 0.0379 POSITIVE REGULATION OF PHYSIOLOGICAL PROCESS 32 / 210 1150 / 10541 0.0466 NEGATIVE REGULATION OF ENZYME ACTIVITY 11 / 210 236 / 10541 0.0192 → NEGATIVE REGULATION OF TRANSFERASE ACTIVITY 10 / 210 220 / 10541 0.0312 NEGATIVE REGULATION OF PROTEIN KINASE ACTIVITY 10 / 210 216 / 10541 0.0282

103

Table 6.5. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated in aged WT mice compared to young WT mice in Experiment A. List hits / Population hits / EASE Biological process category identifier List total Population total score SYSTEM DEVELOPMENT 28 / 167 1114 / 10541 0.0150 → NERVOUS SYSTEM DEVELOPMENT 28 / 167 1078 / 10541 0.0100 →→ BRAIN DEVELOPMENT 11 / 167 319 / 10541 0.0293 →→ NEURON DIFFERENTIATION 16 / 167 562 / 10541 0.0316 →→→ NEURON DEVELOPMENT 14 / 167 443 / 10541 0.0227 HEART DEVELOPMENT 7 / 167 154 / 10541 0.0348 DEATH 27 / 167 1131 / 10541 0.0307 → CELL DEATH 26 / 167 1121 / 10541 0.0461 CELL MATURATION 9 / 167 249 / 10541 0.0432 LOCALIZATION 72 / 167 3265 / 10541 0.0008 → LOCALIZATION OF CELL 19 / 167 717 / 10541 0.0331 →→ CELL MOTILITY 19 / 167 717 / 10541 0.0331 CELL ORGANIZATION AND BIOGENESIS 55 / 167 2289 / 10541 0.0008 → CELLULAR LOCALIZATION 31 / 167 1109 / 10541 0.0021 → ESTABLISHMENT OF LOCALIZATION 72 / 167 3243 / 10541 0.0007 →→ ESTABLISHMENT OF CELLULAR LOCALIZATION 31 / 167 1102 / 10541 0.0019 → PROTEIN LOCALIZATION 28 / 167 1063 / 10541 0.0084 →→ ESTABLISHMENT OF PROTEIN LOCALIZATION 27 / 167 1034 / 10541 0.0109 →→ TRANSPORT 62 / 167 2865 / 10541 0.0042 →→→ VESICLE-MEDIATED TRANSPORT 22 / 167 621 / 10541 0.0007 →→→→ EXOCYTOSIS 8 / 167 129 / 10541 0.0042 →→→ PROTEIN TRANSPORT 27 / 167 1000 / 10541 0.0072 →→→ INTRACELLULAR TRANSPORT 29 / 167 1079 / 10541 0.0054 →→→→ INTRACELLULAR PROTEIN TRANSPORT 21 / 167 788 / 10541 0.0228 →→→→→ PROTEIN TARGETING 18 / 167 544 / 10541 0.0053 →→ SECRETION 14 / 167 457 / 10541 0.0283 →→→ SECRETORY PATHWAY 12 / 167 343 / 10541 0.0197 GROWTH 15 / 167 494 / 10541 0.0240 MORPHOGENESIS 36 / 167 1494 / 10541 0.0093 → CELLULAR MORPHOGENESIS 23 / 167 815 / 10541 0.0089 →→ REGULATION OF CELL SIZE 13 / 167 396 / 10541 0.0226 →→→ CELL GROWTH 13 / 167 393 / 10541 0.0214 → MORPHOGENESIS OF AN EPITHELIUM 11 / 167 292 / 10541 0.0171 → EMBRYONIC MORPHOGENESIS 11 / 167 311 / 10541 0.0252 TUBE DEVELOPMENT 11 / 167 285 / 10541 0.0147 → TUBE MORPHOGENESIS 9 / 167 235 / 10541 0.0324 DETECTION OF BIOTIC STIMULUS 4 / 167 42 / 10541 0.0281 CELL-CELL SIGNALING 23 / 167 964 / 10541 0.0486 FIBROBLAST GROWTH FACTOR RECEPTOR SIGNALING PATHWAY 4 / 167 33 / 10541 0.0148 PROTEIN KINASE CASCADE 18 / 167 677 / 10541 0.0378 AROMATIC COMPOUND METABOLISM 8 / 167 181 / 10541 0.0243 SULFUR COMPOUND BIOSYNTHESIS 4 / 167 51 / 10541 0.0461 NEGATIVE REGULATION OF BIOLOGICAL PROCESS 33 / 167 1454 / 10541 0.0294 → NEGATIVE REGULATION OF PHYSIOLOGICAL PROCESS 30 / 167 1297 / 10541 0.0316 → NEGATIVE REGULATION OF CELLULAR PROCESS 31 / 167 1366 / 10541 0.0357 →→ NEGATIVE REGULATION OF CELLULAR PHYSIOLOGICAL PROCESS 30 / 167 1259 / 10541 0.0222 REGULATION OF DEVELOPMENT 14 / 167 463 / 10541 0.0310 CELL ION HOMEOSTASIS 10 / 167 284 / 10541 0.0356 SENSORY PERCEPTION OF MECHANICAL STIMULUS 8 / 167 208 / 10541 0.0462 LOCOMOTION 19 / 167 726 / 10541 0.0368 MUSCLE CONTRACTION 11 / 167 317 / 10541 0.0283

104

Table 6.6. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated in aged Pdyn KO mice compared to young Pdyn KO mice in Experiment A. List hits / Population hits / EASE Biological process category identifier List total Population total score AROMATIC COMPOUND METABOLISM 5 / 73 181 / 10541 0.0351 ENZYME LINKED RECEPTOR PROTEIN SIGNALING PATHWAY 9 / 73 558 / 10541 0.0358 COENZYME A METABOLISM 3 / 73 47 / 10541 0.0408 → COENZYME A BIOSYNTHESIS 3 / 73 46 / 10541 0.0393

Table 6.7. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated in aged Pdyn KO mice compared to young Pdyn KO mice in Experiment A. List hits / Population hits / EASE Biological process category identifier List total Population total score ANTIGEN PROCESSING 3 / 46 52 / 10541 0.0206 ANTIGEN PRESENTATION 3 / 46 65 / 10541 0.0313 → ANTIGEN PRESENTATION, ENDOGENOUS ANTIGEN 3 / 46 31 / 10541 0.0077 →→ ANTIGEN PROCESSING, ENDOGENOUS ANTIGEN VIA MHC CLASS I 3 / 46 30 / 10541 0.0072 METABOLISM 35 / 46 6509 / 10541 0.0367 → PRIMARY METABOLISM 33 / 46 6037 / 10541 0.0399 →→ RNA METABOLISM 8 / 46 653 / 10541 0.0197 →→→ RNA SPLICING 4 / 46 181 / 10541 0.0417 RESPONSE TO STIMULUS 19 / 46 2723 / 10541 0.0260

Table 6.8. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated with Pdyn deletion in Experiment A. List hits / Population hits / EASE Biological process category identifier List total Population total score METABOLISM 88 / 118 6509 / 10541 0.0026 → CELLULAR METABOLISM 84 / 118 6194 / 10541 0.0041 →→ CELLULAR BIOSYNTHESIS 24 / 118 1439 / 10541 0.0435 → PRIMARY METABOLISM 80 / 118 6037 / 10541 0.0145 → MACROMOLECULE METABOLISM 64 / 118 4082 / 10541 0.0006 →→MACROMOLECULE BIOSYNTHESIS 20 / 118 966 / 10541 0.0098 →→→ PROTEIN BIOSYNTHESIS 18 / 118 865 / 10541 0.0145 →→ PROTEIN METABOLISM 48 / 118 3122 / 10541 0.0092 →→ CELLULAR MACROMOLECULE METABOLISM 46 / 118 2930 / 10541 0.0077 →→→ CELLULAR PROTEIN METABOLISM 46 / 118 2888 / 10541 0.0058 →→→→ PROTEIN FOLDING 8 / 118 244 / 10541 0.0188 CELLULAR PHYSIOLOGICAL PROCESS 106 / 118 8620 / 10541 0.0126 → ATP SYNTHESIS COUPLED ELECTRON TRANSPORT 3 / 118 30 / 10541 0.0434 → ATP SYNTHESIS COUPLED ELECTRON TRANSPORT (SENSU EUKARYOTA) 3 / 118 29 / 10541 0.0408

105

Table 6.9. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated with Pdyn deletion in Experiment A. List hits / Population hits / EASE Biological process category identifier List total Population total score SYSTEM DEVELOPMENT 24 / 149 1114 / 10541 0.0376 → NERVOUS SYSTEM DEVELOPMENT 23 / 149 1078 / 10541 0.0464 → CELL DEVELOPMENT 16 / 149 628 / 10541 0.0303 →→ NEUROGENESIS 16 / 149 605 / 10541 0.0227 →→→ NEURON DIFFERENTIATION 15 / 149 562 / 10541 0.0264 REGULATION OF ACTION POTENTIAL 7 / 149 178 / 10541 0.0394 RESPONSE TO WOUNDING 18 / 149 774 / 10541 0.0434 CIRCULATION 8 / 149 178 / 10541 0.0127 GAMETOGENESIS 14 / 149 555 / 10541 0.0482

Table 6.10. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated in young Pdyn KO mice compared to young WT mice in Experiment A. List hits / Population hits / EASE Biological process category identifier List total Population total score PROTEIN METABOLISM 36 / 92 3122 / 10541 0.0432

Table 6.11. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated in young Pdyn KO mice compared to young WT mice in Experiment A. List hits / Population hits / EASE Biological process category identifier List total Population total score SYSTEM DEVELOPMENT 15 / 82 1114 / 10541 0.0434 → ORGAN DEVELOPMENT 21 / 82 1489 / 10541 0.0078 → NERVOUS SYSTEM DEVELOPMENT 15 / 82 1078 / 10541 0.0342 →→ CENTRAL NERVOUS SYSTEM DEVELOPMENT 11 / 82 468 / 10541 0.0030 →→→ BRAIN DEVELOPMENT 10 / 82 319 / 10541 0.0007 TUBE DEVELOPMENT 7 / 82 285 / 10541 0.0222 → TUBE MORPHOGENESIS 6 / 82 235 / 10541 0.0344 MORPHOGENESIS 22 / 82 1494 / 10541 0.0037 → EMBRYONIC MORPHOGENESIS 8 / 82 311 / 10541 0.0098 → CELLULAR MORPHOGENESIS 12 / 82 815 / 10541 0.0466 → MORPHOGENESIS OF AN EPITHELIUM 7 / 82 292 / 10541 0.0246 →→ MORPHOGENESIS OF EMBRYONIC EPITHELIUM 6 / 82 220 / 10541 0.0270 LOCALIZATION 36 / 82 3265 / 10541 0.0132 CELL ORGANIZATION AND BIOGENESIS 31 / 82 2289 / 10541 0.0012 → MEMBRANE ORGANIZATION AND BIOGENESIS 5 / 82 174 / 10541 0.0448 →→ MEMBRANE FUSION 5 / 82 96 / 10541 0.0063 → VESICLE ORGANIZATION AND BIOGENESIS 3 / 82 39 / 10541 0.0360 → CELLULAR LOCALIZATION 20 / 82 1109 / 10541 0.0006 → ESTABLISHMENT OF LOCALIZATION 36 / 82 3243 / 10541 0.0118 →→ SECRETION 9 / 82 457 / 10541 0.0238 → PROTEIN LOCALIZATION 17 / 82 1063 / 10541 0.0063 →→ ESTABLISHMENT OF PROTEIN LOCALIZATION 16 / 82 1034 / 10541 0.0114 →→ ESTABLISHMENT OF CELLULAR LOCALIZATION 20 / 82 1102 / 10541 0.0005 →→→ SECRETORY PATHWAY 8 / 82 343 / 10541 0.0161 →→ TRANSPORT 33 / 82 2865 / 10541 0.0105

106

List hits / Population hits / EASE Biological process category identifier List total Population total score (continued from previous page) →→→ NUCLEOBASE, NUCLEOSIDE, NUCLEOTIDE AND NUCLEIC ACID TRANSPORT 4 / 82 85 / 10541 0.0275 →→→→ NUCLEIC ACID TRANSPORT 4 / 82 73 / 10541 0.0185 →→→ INTRACELLULAR TRANSPORT 19 / 82 1079 / 10541 0.0012 →→→→ NUCLEOCYTOPLASMIC TRANSPORT 8 / 82 386 / 10541 0.0286 →→→ REGULATION OF TRANSPORT 6 / 82 217 / 10541 0.0256 →→→ VESICLE-MEDIATED TRANSPORT 16 / 82 621 / 10541 0.0001 →→→→ VESICLE FUSION 4 / 82 64 / 10541 0.0130 →→→→ ENDOSOME TRANSPORT 5 / 82 105 / 10541 0.0086 →→→→ ENDOCYTOSIS 10 / 82 334 / 10541 0.0010 →→→→ RECEPTOR MEDIATED ENDOCYTOSIS 4 / 82 85 / 10541 0.0275 →→→→→ REGULATION OF ENDOCYTOSIS 5 / 82 107 / 10541 0.0092 →→→→ EXOCYTOSIS 5 / 82 129 / 10541 0.0173 →→→ PROTEIN TRANSPORT 16 / 82 1000 / 10541 0.0085 →→→→ INTRACELLULAR PROTEIN TRANSPORT 14 / 82 788 / 10541 0.0067 →→→→→ PROTEIN TARGETING 10 / 82 544 / 10541 0.0233 → RNA LOCALIZATION 4 / 82 76 / 10541 0.0206 →→ ESTABLISHMENT OF RNA LOCALIZATION 4 / 82 73 / 10541 0.0185 →→→ RNA TRANSPORT 4 / 82 73 / 10541 0.0185 →→→→ RNA EXPORT FROM NUCLEUS 4 / 82 72 / 10541 0.0178 → ACTIN FILAMENT-BASED PROCESS 7 / 82 318 / 10541 0.0353 →→ ACTIN CYTOSKELETON ORGANIZATION AND BIOGENESIS 7 / 82 291 / 10541 0.0243 →→→ ACTIN CYTOSKELETON REORGANIZATION 3 / 82 35 / 10541 0.0295 SIGNAL TRANSDUCTION 32 / 82 3032 / 10541 0.0406 → RHO PROTEIN SIGNAL TRANSDUCTION 4 / 82 69 / 10541 0.0159 → ENZYME LINKED RECEPTOR PROTEIN SIGNALING PATHWAY 11 / 82 558 / 10541 0.0101 →→ TRANSMEMBRANE RECEPTOR PROTEIN TYROSINE KINASE SIGNALING PATHWAY 7 / 82 316 / 10541 0.0344 →→→ EPIDERMAL GROWTH FACTOR RECEPTOR SIGNALING PATHWAY 4 / 82 98 / 10541 0.0395 →→→ FIBROBLAST GROWTH FACTOR RECEPTOR SIGNALING PATHWAY 3 / 82 33 / 10541 0.0264 →→ TRANSMEMBRANE RECEPTOR PROTEIN SERINE/THREONINE KINASE SIGNALING PATHWAY 6 / 82 250 / 10541 0.0431 →→→ TRANSFORMING GROWTH FACTOR BETA RECEPTOR SIGNALING PATHWAY 4 / 82 98 / 10541 0.0395 → PROTEIN KINASE CASCADE 11 / 82 677 / 10541 0.0340 →→ POSITIVE REGULATION OF I- B KINASE/NF- B CASCADE 5 / 82 178 / 10541 0.0480 CELLULAR BIOSYNTHESIS 18 / 82 1439 / 10541 0.0443 PROTEIN MODIFICATION 21 / 82 1728 / 10541 0.0353 → PROTEIN UBIQUITINATION 7 / 82 319 / 10541 0.0358 VITAMIN METABOLISM 4 / 82 96 / 10541 0.0375 REGULATION OF NEUROTRANSMITTER LEVELS 5 / 82 173 / 10541 0.0440 MUSCLE CONTRACTION 7 / 82 317 / 10541 0.0348 → SMOOTH MUSCLE CONTRACTION 4 / 82 106 / 10541 0.0480 TELOMERASE-DEPENDENT TELOMERE MAINTENANCE 3 / 82 44 / 10541 0.0448 DEATH 15 / 82 1131 / 10541 0.0483 → NEGATIVE REGULATION OF PROGRAMMED CELL DEATH 8 / 82 364 / 10541 0.0216 →→ NEGATIVE REGULATION OF APOPTOSIS 8 / 82 363 / 10541 0.0213

107

Table 6.12. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated in aged Pdyn KO mice compared to aged WT mice in Experiment A. List hits / Population hits / EASE Biological process category identifier List total Population total score DEVELOPMENT 55 / 142 2974 / 10541 0.0058 → MORPHOGENESIS 32 / 142 1494 / 10541 0.0075 →→ CELLULAR MORPHOGENESIS 19 / 142 815 / 10541 0.0238 →→→ REGULATION OF CELL SIZE 11 / 142 396 / 10541 0.0397 →→→→ CELL GROWTH 11 / 142 393 / 10541 0.0380 NERVOUS SYSTEM DEVELOPMENT 22 / 142 1078 / 10541 0.0499 LOCALIZATION 55 / 142 3265 / 10541 0.0375 → ESTABLISHMENT OF LOCALIZATION 55 / 142 3243 / 10541 0.0332 →→ VESICLE-MEDIATED TRANSPORT 15 / 142 621 / 10541 0.0381 ION HOMEOSTASIS 10 / 142 341 / 10541 0.0394 → CELL ION HOMEOSTASIS 9 / 142 284 / 10541 0.0368 CELL DIFFERENTIATION 27 / 142 1382 / 10541 0.0437 NEGATIVE REGULATION OF BIOLOGICAL PROCESS 29 / 142 1454 / 10541 0.0281 → NEGATIVE REGULATION OF CELLULAR PROCESS 27 / 142 1366 / 10541 0.0387 →→ NEGATIVE REGULATION OF CELLULAR PHYSIOLOGICAL PROCESS 25 / 142 1259 / 10541 0.0458 NEGATIVE REGULATION OF PHYSIOLOGICAL PROCESS 26 / 142 1297 / 10541 0.0373 AROMATIC COMPOUND METABOLISM 7 / 142 181 / 10541 0.0345 INTERACTION BETWEEN ORGANISMS 9 / 142 298 / 10541 0.0466

108

Table 6.13. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated in aged Pdyn KO mice compared to aged WT mice in Experiment A. List hits / Population hits / EASE Biological process category identifier List total Population total score CELL ORGANIZATION AND BIOGENESIS 48 / 147 2289 / 10541 0.0020 → CELLULAR MORPHOGENESIS 22 / 147 815 / 10541 0.0041 CELL DEVELOPMENT 16 / 147 628 / 10541 0.0273 → CELLULAR MORPHOGENESIS DURING DIFFERENTIATION 14 / 147 407 / 10541 0.0042 NERVOUS SYSTEM DEVELOPMENT 23 / 147 1078 / 10541 0.0408 → NEUROGENESIS 16 / 147 605 / 10541 0.0203 →→ NEURON DIFFERENTIATION 15 / 147 562 / 10541 0.0238 →→→ NEURON DEVELOPMENT 14 / 147 443 / 10541 0.0085 →→→→ NEURON MORPHOGENESIS DURING DIFFERENTIATION 14 / 147 377 / 10541 0.0022 →→→→→ NEURITE MORPHOGENESIS 14 / 147 377 / 10541 0.0022 →→→→→→AXONOGENESIS 11 / 147 348 / 10541 0.0228 →→→→→→DENDRITE MORPHOGENESIS 7 / 147 157 / 10541 0.0219 BLOOD VESSEL MORPHOGENESIS 10 / 147 333 / 10541 0.0418 G1/S TRANSITION OF MITOTIC CELL CYCLE 8 / 147 214 / 10541 0.0292 REGULATION OF BIOLOGICAL PROCESS 67 / 147 3907 / 10541 0.0257 REGULATION OF PHYSIOLOGICAL PROCESS 62 / 147 3524 / 10541 0.0207 → REGULATION OF CELLULAR PROCESS 63 / 147 3535 / 10541 0.0145 →→ POSITIVE REGULATION OF CELLULAR PROCESS 29 / 147 1254 / 10541 0.0069 → REGULATION OF CELLULAR PHYSIOLOGICAL PROCESS 59 / 147 3309 / 10541 0.0195 →→ POSITIVE REGULATION OF CELLULAR PHYSIOLOGICAL PROCESS 28 / 147 1111 / 10541 0.0025 → POSITIVE REGULATION OF BIOLOGICAL PROCESS 31 / 147 1400 / 10541 0.0091 → POSITIVE REGULATION OF PHYSIOLOGICAL PROCESS 28 / 147 1150 / 10541 0.0041 →→ POSITIVE REGULATION OF METABOLISM 16 / 147 598 / 10541 0.0185 →→→ POSITIVE REGULATION OF CELLULAR METABOLISM 16 / 147 575 / 10541 0.0134 →→→→ POSITIVE REGULATION OF GLUCONEOGENESIS 2 / 147 3 / 10541 0.0410 →→→→ POSITIVE REGULATION OF NUCLEOBASE, NUCLEOSIDE, NUCLEOTIDE & NUCLEIC ACID METABOLISM 13 / 147 469 / 10541 0.0295 →→→→→ POSITIVE REGULATION OF TRANSCRIPTION 13 / 147 465 / 10541 0.0278 NEGATIVE REGULATION OF ENZYME ACTIVITY 8 / 147 236 / 10541 0.0457

Table 6.14. Complete list of Biological Process identifiers over-represented among probe sets significantly downregulated in aged Pdyn KO mice compared to aged WT mice in Experiment B. List hits / Population hits / EASE Biological process category identifier List total Population total score PROTEASOME ACTIVATOR COMPLEX 2 / 79 3 / 7111 0.0326

109

Table 6.15. Complete list of Biological Process identifiers over-represented among probe sets significantly upregulated in aged Pdyn KO mice compared to aged WT mice in Experiment B. List hits / Population hits / EASE Biological process category identifier List total Population total score CELL ORGANIZATION AND BIOGENESIS 38 / 115 1667 / 7176 0.0148 → MICROTUBULE CYTOSKELETON ORGANIZATION AND BIOGENESIS 6 / 115 123 / 7176 0.0457 → GOLGI ORGANIZATION AND BIOGENESIS 4 / 115 50 / 7176 0.0446 LOCALIZATION 48 / 115 2265 / 7176 0.0178 → ORGANELLE LOCALIZATION 4 / 115 41 / 7176 0.0269 → ESTABLISHMENT OF LOCALIZATION 47 / 115 2250 / 7176 0.0253 →→ ION TRANSPORT 16 / 115 563 / 7176 0.0323 →→ ESTABLISHMENT OF ORGANELLE LOCALIZATION 4 / 115 40 / 7176 0.0252 MUSCLE CONTRACTION 9 / 115 229 / 7176 0.0289

Table 6.16. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated with aging in Experiment A. Terms used in the following tables reflect those used in the Gene Ontology Molecular Function database. Arrows precede categories that are subsets of one or more categories listed above them, with increasing numbers of arrows denoting more specific categorizations. List hits and population hits are presented along with the corresponding totals. Note that not all entries within a given probe set list are reflected in the total, since some probe sets do not have a corresponding annotation in the ontological system. List hits / Population hits / EASE Molecular function category identifier List total Population total score STRUCTURE-SPECIFIC DNA BINDING 14 / 247 240 / 11005 0.0029 → DOUBLE-STRANDED DNA BINDING 12 / 247 185 / 11005 0.0029 ZINC-MEDIATED TRANSCRIPTIONAL ACTIVATOR ACTIVITY 8 / 247 137 / 11005 0.0340 TRANSPORTER ACTIVITY 55 / 247 1853 / 11005 0.0212 → ION TRANSPORTER ACTIVITY 30 / 247 884 / 11005 0.0234 CYTOSKELETAL PROTEIN BINDING 23 / 247 641 / 11005 0.0297 METALLOEXOPEPTIDASE ACTIVITY 9 / 247 174 / 11005 0.0416 SULFURIC ESTER HYDROLASE ACTIVITY 8 / 247 137 / 11005 0.0340 NUTRIENT RESERVOIR ACTIVITY 8 / 247 146 / 11005 0.0455

Table 6.17. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated with aging in Experiment A. List hits / Population hits / EASE Molecular function category identifier List total Population total score PROTEIN BINDING 134 / 241 5244 / 11005 0.0089 ANTIGEN BINDING 6 / 241 65 / 11005 0.0135 ENZYME INHIBITOR ACTIVITY 14 / 241 287 / 11005 0.0102 STRUCTURAL CONSTITUENT OF RIBOSOME 11 / 241 244 / 11005 0.0413

110

Table 6.18. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated in aged WT mice compared to young WT mice in Experiment A. List hits / Population hits / EASE Molecular function category identifier List total Population total score PROTEIN TRANSPORTER ACTIVITY 10 / 212 162 / 11005 0.0040 BINDING 182 / 212 8519 / 11005 0.0014 → PROTEIN BINDING 119 / 212 5244 / 11005 0.0092 →→ PHOSPHOPROTEIN BINDING 6 / 212 56 / 11005 0.0043 →→→ PROTEIN PHOSPHORYLATED AMINO ACID BINDING 4 / 212 40 / 11005 0.0408 →→ CYTOSKELETAL PROTEIN BINDING 20 / 212 641 / 11005 0.0389 →→ SNARE BINDING 4 / 212 43 / 11005 0.0490 → SEQUENCE-SPECIFIC DNA BINDING 18 / 212 544 / 11005 0.0324 → STRUCTURE-SPECIFIC DNA BINDING 10 / 212 240 / 11005 0.0417 → ZINC ION BINDING 40 / 212 1546 / 11005 0.0418 LIGASE ACTIVITY 23 / 212 652 / 11005 0.0070 → LIGASE ACTIVITY, FORMING CARBON- NITROGEN BONDS 15 / 212 403 / 11005 0.0232 →→ CTP SYNTHASE ACTIVITY 2 / 212 2 / 11005 0.0380 LYASE ACTIVITY 9 / 212 205 / 11005 0.0438 KINASE REGULATOR ACTIVITY 8 / 212 149 / 11005 0.0246 → PROTEIN KINASE REGULATOR ACTIVITY 7 / 212 102 / 11005 0.0136 →→ CAMP-DEPENDENT PROTEIN KINASE REGULATOR ACTIVITY 3 / 212 17 / 11005 0.0412

Table 6.19. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated in aged WT mice compared to young WT mice in Experiment A. List hits / Population hits / EASE Molecular function category identifier List total Population total score ION BINDING 63 / 177 2881 / 11005 0.0047 → METAL ION BINDING 63 / 177 2881 / 11005 0.0047 → CATION BINDING 59 / 177 2668 / 11005 0.0053 →→ CALCIUM ION BINDING 35 / 177 988 / 11005 <0.0001 ACTIN BINDING 11 / 177 314 / 11005 0.0294 RECEPTOR SIGNALING COMPLEX SCAFFOLD ACTIVITY 5 / 177 65 / 11005 0.0201 RHO GTPASE BINDING 4 / 177 28 / 11005 0.0098 GUANYL-NUCLEOTIDE EXCHANGE FACTOR ACTIVITY 6 / 177 122 / 11005 0.0460

Table 6.20. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated in aged Pdyn KO mice compared to young Pdyn KO mice in Experiment A. List hits / Population hits / EASE Molecular function category identifier List total Population total score TRIPEPTIDYL-PEPTIDASE I ACTIVITY 2 / 74 2 / 11005 0.0132 PROTEIN TRANSPORTER ACTIVITY 5 / 74 162 / 11005 0.0225 ALKALI METAL ION BINDING 5 / 74 168 / 11005 0.0253 → POTASSIUM ION BINDING 5 / 74 106 / 11005 0.0053

111

Table 6.21. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated in aged Pdyn KO mice compared to young Pdyn KO mice in Experiment A. List hits / Population hits / EASE Molecular function category identifier List total Population total score LIGASE ACTIVITY 9 / 48 652 / 11005 0.0059 → LIGASE ACTIVITY, FORMING CARBON- NITROGEN BONDS 6 / 48 403 / 11005 0.0279 MHC CLASS I RECEPTOR ACTIVITY 3 / 48 30 / 11005 0.0072 BINDING 43 / 48 8519 / 11005 0.0297 → PROTEIN BINDING 31 / 48 5244 / 11005 0.0186 → RNA BINDING 9 / 48 772 / 11005 0.0155

Table 6.22. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated with Pdyn deletion in Experiment A. List hits / Population hits / EASE Molecular function category identifier List total Population total score STRUCTURAL CONSTITUENT OF RIBOSOME 12 / 123 244 / 11005 0.0001 OXIDOREDUCTASE ACTIVITY, ACTING ON DIPHENOLS AND RELATED SUBSTANCES AS DONORS 3 / 123 12 / 11005 0.0075 → OXIDOREDUCTASE ACTIVITY, ACTING ON DIPHENOLS AND RELATED SUBSTANCES AS DONORS, CYTOCHROME AS ACCEPTOR 3 / 123 12 / 11005 0.0075 MONOVALENT INORGANIC CATION TRANSPORTER ACTIVITY 8 / 123 280 / 11005 0.0361 → HYDROGEN ION TRANSPORTER ACTIVITY 8 / 123 274 / 11005 0.0327 →→ UBIQUINOL-CYTOCHROME-C REDUCTASE ACTIVITY 3 / 123 12 / 11005 0.0075 NUCLEIC ACID BINDING 42 / 123 2711 / 11005 0.0157 → RNA BINDING 22 / 123 772 / 11005 0.0001 RAN GTPASE BINDING 3 / 123 20 / 11005 0.0203 UNFOLDED PROTEIN BINDING 8 / 123 196 / 11005 0.0061

Table 6.23. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated with Pdyn deletion in Experiment A. List hits / Population hits / EASE Molecular function category identifier List total Population total score ANTIGEN BINDING 5 / 151 65 / 11005 0.0118 HYDROLASE ACTIVITY, ACTING ON ESTER BONDS 19 / 151 766 / 11005 0.0163 GUANYL NUCLEOTIDE BINDING 11 / 151 375 / 11005 0.0324 TRANSFERASE ACTIVITY, TRANSFERRING PHOSPHORUS-CONTAINING GROUPS 24 / 151 1141 / 11005 0.0357 CYSTEINE-TYPE ENDOPEPTIDASE ACTIVITY 9 / 151 289 / 11005 0.0441

Table 6.24. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated in young Pdyn KO mice compared to young WT mice in Experiment A. List hits / Population hits / EASE Molecular function category identifier List total Population total score CALCIUM- AND CALMODULIN-DEPENDENT PROTEIN KINASE ACTIVITY 3 / 91 19 / 11005 0.0103 LIGASE ACTIVITY, FORMING CARBON-NITROGEN BONDS 8 / 91 403 / 11005 0.0467

112

Table 6.25. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated in young Pdyn KO mice compared to young WT mice in Experiment A. List hits / Population hits / EASE Molecular function category identifier List total Population total score RHO GTPASE BINDING 3 / 92 28 / 11005 0.0222 CALCIUM ION BINDING 16 / 92 988 / 11005 0.0150 COPPER ION BINDING 4 / 92 93 / 11005 0.0417 CARBOHYDRATE BINDING 10 / 92 468 / 11005 0.0151 → SUGAR BINDING 6 / 92 229 / 11005 0.0411 ACTIN BINDING 8 / 92 314 / 11005 0.0151 PROTEIN BINDING, BRIDGING 5 / 92 146 / 11005 0.0327 PATTERN BINDING 7 / 92 301 / 11005 0.0383 PROTEIN-TYROSINE KINASE ACTIVITY 8 / 92 392 / 11005 0.0434 OXIDOREDUCTASE ACTIVITY 12 / 92 752 / 11005 0.0453 EXONUCLEASE ACTIVITY 4 / 92 99 / 11005 0.0486

Table 6.26. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated in aged Pdyn KO mice compared to aged WT mice in Experiment A. List hits / Population hits / EASE Molecular function category identifier List total Population total score ION BINDING 52 / 147 2881 / 11005 0.0115 → METAL ION BINDING 52 / 147 2881 / 11005 0.0115 → CATION BINDING 48 / 147 2668 / 11005 0.0175 →→ CALCIUM ION BINDING 29 / 147 988 / 11005 0.0001 →→ POTASSIUM ION BINDING 6 / 147 106 / 11005 0.0133 RECEPTOR SIGNALING COMPLEX SCAFFOLD ACTIVITY 5 / 147 65 / 11005 0.0108 GUANYL-NUCLEOTIDE EXCHANGE FACTOR ACTIVITY 6 / 147 122 / 11005 0.0231 TRANSPORTER ACTIVITY 34 / 147 1853 / 11005 0.0427 TRIPEPTIDYL-PEPTIDASE I ACTIVITY 2 / 147 2 / 11005 0.0264

Table 6.27. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated in aged Pdyn KO mice compared to aged WT mice in Experiment A. List hits / Population hits / EASE Molecular function category identifier List total Population total score BINDING 133 / 149 8519 / 11005 0.0002 → PROTEIN BINDING 88 / 149 5244 / 11005 0.0040 →→ CYTOSKELETAL PROTEIN BINDING 18 / 149 641 / 11005 0.0056 →→ PROTEIN DIMERIZATION ACTIVITY 13 / 149 523 / 11005 0.0493 →→ PHOSPHOPROTEIN BINDING 4 / 149 56 / 11005 0.0394 →→ TRANSCRIPTION COACTIVATOR ACTIVITY 11 / 149 401 / 11005 0.0441 → PEPTIDE BINDING 9 / 149 214 / 11005 0.0083 → SEQUENCE-SPECIFIC DNA BINDING 16 / 149 544 / 11005 0.0064 → STRUCTURE-SPECIFIC DNA BINDING 9 / 149 240 / 11005 0.0157

113

Table 6.28. Complete list of Molecular Function identifiers over-represented among probe sets significantly downregulated in aged Pdyn KO mice compared to aged WT mice in Experiment B. List hits / Population hits / EASE Molecular function category identifier List total Population total score PROTEIN KINASE C BINDING 3 / 79 26 / 7363 0.0306 PROTEASOME ACTIVATOR ACTIVITY 2 / 79 3 / 7363 0.0314 NADH DEHYDROGENASE ACTIVITY 3 / 79 33 / 7363 0.0473 → NADH DEHYDROGENASE (QUINONE) ACTIVITY 3 / 79 33 / 7363 0.0473 →→ NADH DEHYDROGENASE (UBIQUINONE) ACTIVITY 3 / 79 33 / 7363 0.0473 SODIUM ION TRANSPORTER ACTIVITY 3 / 79 34 / 7363 0.0500

Table 6.29. Complete list of Molecular Function identifiers over-represented among probe sets significantly upregulated in aged Pdyn KO mice compared to aged WT mice in Experiment B. List hits / Population hits / EASE Molecular function category identifier List total Population total score TRANSPORTER ACTIVITY 29 / 117 1277 / 7363 0.0380 → ION TRANSPORTER ACTIVITY 16 / 117 575 / 7363 0.0357 → CHANNEL OR PORE CLASS TRANSPORTER ACTIVITY 14 / 117 377 / 7363 0.0062 →→ ALPHA-TYPE CHANNEL ACTIVITY 14 / 117 364 / 7363 0.0046 →→→ ION CHANNEL ACTIVITY 14 / 117 346 / 7363 0.0030 →→→→ VOLTAGE-GATED ION CHANNEL ACTIVITY 8 / 117 176 / 7363 0.0208 PROTEIN BINDING 70 / 117 3732 / 7363 0.0343 → PROTEIN DOMAIN SPECIFIC BINDING 8 / 117 190 / 7363 0.0301 → DYNEIN BINDING 4 / 117 41 / 7363 0.0263

114

CHAPTER 7: Discussion and Conclusions

Summary of effects of aging and Pdyn deletion Behavioral effects Aging produced a significant decrease in water maze performance, with mild decreases in movement speed in both the water maze and open-field experiments and increased stereotypical behavior in open-field analyses. The effects of Pdyn deletion on learning and memory observed in this dissertation include relative sparing of age-related impairment in spatial navigational memory and a subtle enhancement of passive avoidance retention in young mice, with minimal effect on object recogntion. No significant age-associated deficits in performance were noted in the object recognition and passive avoidance tasks of this study. It is noteworthy that both of the learning and memory assays in which Pdyn deletion confers a beneficial effect are tasks in which performance is driven in large part by a response to an aversive stimulus in contrast to the relative neutral stimuli required for object recognition. A likely interpretation of these findings is that Pdyn-derived peptides alter the subject's reward or motivational response, possibly in addition to any direct inhibitory effects of dynorphins on hippocampal neurotransmission previously reported (Drake et al., 1994; Terman et al., 1994). Consistent with known effects of dopamine and dynorphin in reward pathways (Carlezon et al., 1998; Hurd et al., 1999; Zhang et al., 2004b), our data on dopamine alterations in Pdyn KO mice support such a role of dynorphins. In the object recognition experiments, Pdyn deletion produces a relative avoidance of novel objects, despite demonstrating slightly increased total exploratory behavior following introduction of a new object. Spontaneous locomotor activity in the open-field assay was also mildly elevated in male Pdyn KO mice in relation to the average activity of other groups of mice tested, arguing in favor of differentially regulated arousal or motivational processes in these mice. These behavioral changes are likely influenced by significant elevations in striatal dopamine content, such as those noted in aged Pdyn KO mice.

115

Molecular changes A major finding of this dissertation was the markedly enhanced striatal dopamine in aged Pdyn KO mice, while aging alone in WT mice produced no significant changes in dopamine. With evidence of increased phosphorylation of TH at Ser40 in Pdyn KO mice, it is reasonable to conclude that dynorphin peptides, through mechanisms that have not yet been fully delineated, suppress dopamine synthesis via modulation of TH activity. Aging also was associated with altered dynorphin expression, with declines in Dyn B in striatum and frontal cortex, while Dyn A showed a tendency to increase with age in frontal cortex. These findings may account for some of the age dependence noted in the behavioral functional comparisons between WT and Pdyn KO mice. Moreover, the patterns of hippocampal gene expression from the microarray experiments suggest that aged Pdyn KO mice are metabolically and transcriptionally more active than aged WT mice, with increased expression of genes involved in neuronal development and basic cellular processes. Furthermore, calcium binding and transport appears to be an important downstream consequence of Pdyn deletion based on microarray gene expression results. This observation leads to the conclusion that prolonged suppression of dynorphin may induce significant changes in calcium regulation and other neuronal processes, including dopamine synthesis, which may significantly enhance neurotransmission within the parts of the brain where dynorphin is prevalent, such as striatum and hippocampus.

Role of dynorphins in MPTP neurotoxicity In the data reported in the preceding chapters, MPTP induced decreases in locomotor activity that was ameliorated by Pdyn deletion or heterozygosity, but in a separate experiment on dopamine quantitation, Pdyn deletion conferred no protection against MPTP-induced dopamine depletion despite the finding that dopamine was elevated in Pdyn KO mice in an age-dependent manner. These results suggest that although certain Pdyn-derived peptides can demonstrate direct neurotoxicty in various experimental situations (Hauser et al., 1999; Goody et al., 2003; Singh et al., 2003), their absence does not protect against striatal neurotoxicity by MPTP. Interestingly, in the acute phase of MPTP injury, striatal dynorphin peptides are increased but eventually return to below

116 pre-injection levels, likely as a consequence of subsequently decreased dopaminergic tone. Our results showing improved post-MPTP open-field activity in Pdyn KO mice in the absence of elevated dopamine are consistent with presumed effects of residual dynorphins on dopaminergic activity beyond that of maintaining total striatal dopamine content.

A working model for the role of dynorphins in aging Early observations on the ability of dynorphins to antagonize morphine-induced analgesia in opiate-dependent subjects while potentiating it in opiate-naïve subjects have led to the proposal that dynorphins act in a modulatory capacity (Smith and Lee, 1988). The results of the preceding chapters support such a role of physiological quantities of dynorphins, and, in an attempt to establish a working model to explain our findings, we propose the following framework to explain the role of dynorphins. In the proposed model, basal dynorphinergic input is needed to maintain striatal dopaminergic activity within normal functional levels (Steiner and Gerfen, 1998) and to prevent excessive synaptic excitability in hippocampus (Simonato and Romualdi, 1996; Loacker et al., 2007), physiological functions which are necessary for homeostasis and have ample evolutionary advantages. In the Pdyn KO mice characterized in this dissertation, the absence of these basal regulatory functions of dynorphins may be offset by appropriate counter-regulatory elements. In young adult mice, these compensatory mechanisms may be sufficient to mask functional effects of Pdyn deletion. However, with advanced age, these mechanisms may be inadequate, resulting, for instance, in elevations in dopamine as noted in the preceding chapters. The elevated dopamine may be sufficient to produce increased motivational drive, increased alertness, or increased learning and memory functioning as manifested in the relative amelioration of age-related water maze performance impairment in the aged Pdyn KO relative to aged WT mice. The absence of similar Pdyn effects among aged mice in the other learning and memory paradigms could be explained by the relatively greater reward and motivational component present in the water maze task. Dynorphin peptides are an important component in the biological mechanisms of reward, as

117 supported by studies in which upregulation of prodynorphin by overexpressing CREB substantially decreases the rewarding effects of cocaine (Carlezon et al., 1998). In addition to the tonic dynorphinergic effect, there is likely another role of dynorphins in which Pdyn expression is inducible following various types of neuronal injury (Iadarola et al., 1986; Thai et al., 1996; Simpson et al., 1997; Zhang et al., 2004a). The short-term increases in dynorphin expression following a neurotoxic injury by MPTP in our studies raise the possibility that immediate-early genes activated during various neurologic stressors (Naranjo et al., 1991; Noguchi et al., 1991; Bronstein et al., 1994; Hunter et al., 1995) could induce episodic expression of supraphysiological levels of dynorphins. Our findings that aging increases the amount of Dyn A relative to Dyn B may indicate alterations in prodynorphin processing that favor the more neurotoxic Dyn A product (Hauser et al., 2005). While the episodic dynorphin upregulation may provide protective effects under some circumstances (Przewlocka et al., 1983; Simonato and Romualdi, 1996; Liu et al., 2001; Loacker et al., 2007), such as anticonvulsant activity during seizures or traumatic injury, the increased likelihood of direct neurotoxic or pro- excitotoxic effects, mostly mediated by non-opioid mechanisms (Hauser et al., 2005), may over time lead to repetitive injury that may manifest as worsening of cognitive impairment.

Potential areas of further inquiry One topic that was not directly addressed by this series of experiments is the potential differences in activity and mechanisms of action among the prodynorphin-derived peptides. The knockout model, while effectively suppressing expression of the preprodynorphin precursor, does not address the potential differences between Dyn A, Dyn B, Dyn-32, and multiple other peptide derivatives of prodynorphin. Previous studies have demonstrated significant variations in the biological effects and relative concentrations of the different Pdyn-derived peptides in various physiological and pathological states [reviewed in (Hauser et al., 2005)]. Given the discordant expression profiles of Dyn A and B during aging seen in the WT mice of this study, further characterization of the mechanisms controlling post-translational processing of the Pdyn

118 gene product would be needed to supplement existing data on dynorphin peptide and mRNA quantitation. While the experiments described in this dissertation require no assumption regarding the type of receptor through which any of the observed findings are mediated, it would be interesting to assess whether similar observations may be made in opioid receptor knockout mice. Such studies, in conjunction with the results presented here, may help elucidate the extent to which the observed Pdyn deletion effects are due to opioid-mediated mechanisms. Additional mechanistic studies will also be needed to dissect the components of the cellular pathways in which dynorphins interact during aging or neurotoxic injury.

Extrapolation to human disease The findings of this dissertation may also be evaluated in the context of existing evidence of dynorphins in specific human disorders. Information of potential clinical relevance to human disease has been obtained by laboratory or pathological investigations in patients with specific neurological disorders such as Alzheimer's disease (AD) or Parkinson's disease (PD). One of the earliest studies of dynorphins in AD reported a 40% decline in Dyn A(1-8) levels in cerebrospinal fluid (CSF) compared to age-matched controls (Sunderland et al., 1991). However, increased levels of Dyn A were later reported in frontal cortex (Risser et al., 1996) and parietal cortex (Yakovleva et al., 2006a) in AD patients post-mortem, with no similar increase observed for Dyn B (Yakovleva et al., 2006a). Another study reported that Pdyn mRNA expression is decreased with aging, but no differences were noted between normally aged subjects and PD patients (Backman et al., 2007). Dynorphin peptide levels in most studies were similarly unchanged in various brain regions and in CSF of Parkinson's disease patients compared to controls (Taquet et al., 1985; Baronti et al., 1991; Yakovleva et al., 2006a), although one study of PD patients reported a decline in dynorphin-like immunoreactivity in substantia nigra but not basal ganglia (Waters et al., 1988). In addition, neither aging alone nor L-dopa therapy induced any change in Dyn A in CSF in one clinical study (Baronti et al., 1991). The complexity of the aging process and the heterogeneity of subject populations, in addition to published reports of inconsistencies between animal and human studies

119

(Baronti et al., 1991), make it challenging to identify consistent trends in dynorphin expression. The studies presented in this dissertation involve exclusively the 129SvEv mouse strain. As presented in the preceding chapters, Dyn A shows some evidence of upward trend with age in striatum and frontal cortex of WT mice, but Dyn B exhibits an age-dependent decline. Strain- and species-specific differences may render it difficult to generalize these findings to other rodents or to primates. Nonetheless, many results from animal studies have been replicated in the few clinical trials involving direct administration of dynorphins or related agents. Although few studies directly address the effects of pharmacologically applied dynorphins in cognition or motor function in human subjects, dynorphins have been evaluated with a broad array of other endpoints. Intravenous Dyn A(1-13) infusion in healthy human subjects produces mild, transient mood and euphoric effects (King et al., 1999) and has varying efficacy in opiate withdrawal (Greenwald et al., 1997; Specker et al., 1998). A dynorphin A derivative has also been found to be effective in treatment of post-herpetic neuralgia (Wallace et al., 2006). In addition, an earlier trial reported dose-limiting adverse behavioral effects when administering the agonist spiradoline to Parkinson’s disease patients (Giuffra et al., 1993) without effects on motor symptoms. Prodynorphin- derived peptides have also been noted to exhibit endocrine effects in healthy humans. Dyn A(1-13) was found to increase prolactin (Kreek et al., 1999; Bart et al., 2003), and a dynorphin analogue was noted to inhibit vasopressin (Ohnishi et al., 1994), consistent with findings in rodent studies. These studies support the contention that many of the major properties of dynorphin peptides discovered in animal studies will likely be generalizable to humans. However, the existing clinical literature unfortunately offers little data on effects of dynorphinergic activity or suppression during normal aging in humans. While the results of this dissertation addressing memory and motor effects of Pdyn deletion are felt to be relevant to disorders such as Alzheimer’s and Parkinson’s disease, the potential applicability of further research on dynorphins is clearly not limited to aging-associated disorders. The role of dynorphinergic input on dopamine levels and its potential effects on reward pathways have significant implications for the pathophysiology of addiction, which has been aptly described as a maladaptive

120 interaction of reward pathways and learning mechanisms (Hyman, 2005). The prevalence of addictive behavior and its effect of health care costs confirm the potential significance of the reward-mediating actions of dynorphins in human disease. While further research is needed to address the mechanisms involved and to further characterize the effects of dynorphins on human disease, the results of this dissertation support an important role of dynorphins in age-related memory impairment and in modulation of the nigrostriatal dopaminergic system. Further research into the roles dynorphins play in the processes discussed in this dissertation would be helpful in assessing the applicability and appropriateness of potential clinical interventions.

121

REFERENCES

Abou-Khalil B, Young AB, Penney JB (1984) Evidence for the presynaptic localization of opiate binding sites on striatal efferent fibers. Brain Res 323:21-29. Ali SF, David SN, Newport GD (1993) Age-related susceptibility to MPTP-induced neurotoxicity in mice. Neurotoxicology 14:29-34. Arai N, Misugi K, Goshima Y, Misu Y (1990) Evaluation of a 1-methyl-4-phenyl- 1,2,3,6-tetrahydropyridine (MPTP)-treated C57 black mouse model for parkinsonism. Brain Res 515:57-63. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, Harris MA, Hill DP, Issel-Tarver L, Kasarskis A, Lewis S, Matese JC, Richardson JE, Ringwald M, Rubin GM, Sherlock G (2000) Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet 25:25-29. Backman CM, Shan L, Zhang Y, Hoffer BJ, Tomac AC (2007) Alterations in prodynorphin, proenkephalin, and GAD67 mRNA levels in the aged human putamen: correlation with Parkinson's disease. J Neurosci Res 85:798-804. Bakshi R, Faden AI (1990) Competitive and non-competitive NMDA antagonists limit dynorphin A-induced rat hindlimb paralysis. Brain Res 507:1-5. Barnes CA (1987) Neurological and behavioral investigations of memory failure in aging animals. Int J Neurol 21-22:130-136. Baronti F, Conant KE, Giuffra M, Davis TL, Brughitta G, Iadarola MJ, Berrettini WH, Chase TN, Mouradian MM (1991) Opioid peptides in Parkinson's disease: effects of dopamine repletion. Brain Res 560:92-96. Bart G, Borg L, Schluger JH, Green M, Ho A, Kreek MJ (2003) Suppressed prolactin response to dynorphin A1-13 in methadone-maintained versus control subjects. J Pharmacol Exp Ther 306:581-587. Blalock EM, Geddes JW, Chen KC, Porter NM, Markesbery WR, Landfield PW (2004) Incipient Alzheimer's disease: microarray correlation analyses reveal major transcriptional and tumor suppressor responses. Proc Natl Acad Sci U S A 101:2173- 2178. Blalock EM, Chen KC, Sharrow K, Herman JP, Porter NM, Foster TC, Landfield PW (2003) Gene microarrays in hippocampal aging: statistical profiling identifies novel processes correlated with cognitive impairment. J Neurosci 23:3807-3819. Bordet R, Ridray S, Carboni S, Diaz J, Sokoloff P, Schwartz JC (1997) Induction of dopamine D3 receptor expression as a mechanism of behavioral sensitization to levodopa. Proc Natl Acad Sci U S A 94:3363-3367. Bozarth MA, Wise RA (1983) Neural substrates of opiate reinforcement. Prog Neuropsychopharmacol Biol Psychiatry 7:569-575. Brauneis U, Oz M, Peoples RW, Weight FF, Zhang L (1996) Differential sensitivity of recombinant N-methyl-D-aspartate receptor subunits to inhibition by dynorphin. J Pharmacol Exp Ther 279:1063-1068. Bronstein DM, Ye H, Pennypacker KR, Hudson PM, Hong JS (1994) Role of a 35 kDa fos-related antigen (FRA) in the long-term induction of striatal dynorphin expression in the 6-hydroxydopamine lesioned rat. Brain Res Mol Brain Res 23:191-203.

122

Burwell RD, Lawler CP, Gallagher M (1995) Mesostriatal dopamine markers in aged Long-Evans rats with sensorimotor impairment. Neurobiol Aging 16:175-186. Callier S, Morissette M, Grandbois M, Di Paolo T (2000) Stereospecific prevention by 17beta-estradiol of MPTP-induced dopamine depletion in mice. Synapse 37:245-251. Carlezon WA, Jr., Thome J, Olson VG, Lane-Ladd SB, Brodkin ES, Hiroi N, Duman RS, Neve RL, Nestler EJ (1998) Regulation of cocaine reward by CREB. Science 282:2272-2275. Carlsson A, Winblad B (1976) Influence of age and time interval between death and autopsy on dopamine and 3-methoxytyramine levels in human basal ganglia. J Neural Transm 38:271-276. Caudle RM, Isaac L (1987) Intrathecal dynorphin(1-13) results in an irreversible loss of the tail-flick reflex in rats. Brain Res 435:1-6. Caudle RM, Isaac L (1988) A novel interaction between dynorphin(1-13) and an N- methyl-D-aspartate site. Brain Res 443:329-332. Caudle RM, Dubner R (1998) Ifenprodil blocks the excitatory effects of the opioid peptide dynorphin 1-17 on NMDA receptor-mediated currents in the CA3 region of the guinea pig hippocampus. Neuropeptides 32:87-95. Chahl LA, Chahl JS (1986) Plasma extravasation induced by dynorphin-(1-13) in rat skin. Eur J Pharmacol 124:343-347. Chang FW, Wang SD, Lu KT, Lee EH (1993) Differential interactive effects of gliotoxin and MPTP in the substantia nigra and the locus coeruleus in BALB/c mice. Brain Res Bull 31:253-266. Chavkin C (2000) Dynorphins are endogenous opioid peptides released from granule cells to act neurohumorly and inhibit excitatory neurotransmission in the hippocampus. Prog Brain Res 125:363-367. Chavkin C, James IF, Goldstein A (1982) Dynorphin is a specific endogenous ligand of the kappa opioid receptor. Science 215:413-415. Chen L, Huang LY (1998) Dynorphin block of N-methyl-D-aspartate channels increases with the peptide length. J Pharmacol Exp Ther 284:826-831. Chen L, Gu Y, Huang LY (1995a) The mechanism of action for the block of NMDA receptor channels by the opioid peptide dynorphin. J Neurosci 15:4602-4611. Chen L, Gu Y, Huang LY (1995b) The opioid peptide dynorphin directly blocks NMDA receptor channels in the rat. J Physiol 482 ( Pt 3):575-581. Cheng HY, Pitcher GM, Laviolette SR, Whishaw IQ, Tong KI, Kockeritz LK, Wada T, Joza NA, Crackower M, Goncalves J, Sarosi I, Woodgett JR, Oliveira-dos-Santos AJ, Ikura M, van der Kooy D, Salter MW, Penninger JM (2002) DREAM is a critical transcriptional repressor for pain modulation. Cell 108:31-43. Cheng PY, Birk AV, Gershengorn MC, Szeto HH (2000) Dynorphin stimulates corticotropin release from mouse anterior pituitary AtT-20 cells through nonopioid mechanisms. Neuroendocrinology 71:170-176. Cherubini E, North RA (1985) Mu and kappa opioids inhibit transmitter release by different mechanisms. Proc Natl Acad Sci U S A 82:1860-1863. Christensson-Nylander I, Terenius L (1985) Dynorphin peptides in human substantia nigra. Neuropeptides 6:391-396.

123

Christensson-Nylander I, Nyberg F, Ragnarsson U, Terenius L (1985) A general procedure for analysis of proenkephalin B derived opioid peptides. Regul Pept 11:65- 76. Colebrooke RE, Chan PM, Lynch PJ, Mooslehner K, Emson PC (2007) Differential gene expression in the striatum of mice with very low expression of the vesicular monoamine transporter type 2 gene. Brain Res 1152:10-16. Collier TJ, Lipton J, Daley BF, Palfi S, Chu Y, Sortwell C, Bakay RA, Sladek JR, Jr., Kordower JH (2007) Aging-related changes in the nigrostriatal dopamine system and the response to MPTP in nonhuman primates: diminished compensatory mechanisms as a prelude to parkinsonism. Neurobiol Dis 26:56-65. Collins-Hicok J, Lin L, Spiro C, Laybourn PJ, Tschumper R, Rapacz B, McMurray CT (1994) Induction of the rat prodynorphin gene through Gs-coupled receptors may involve phosphorylation-dependent derepression and activation. Mol Cell Biol 14:2837-2848. Croll SD, Chesnutt CR, Greene NA, Lindsay RM, Wiegand SJ (1999) Peptide immunoreactivity in aged rat cortex and hippocampus as a function of memory and BDNF infusion. Pharmacol Biochem Behav 64:625-635. Currie IS, Gillies G, Brooks AN (1994) Modulation of arginine vasopressin secretion from cultured ovine hypothalamic cells by glucocorticoids and opioid peptides. Neuroendocrinology 60:360-367. Di Chiara G, Imperato A (1988) Opposite effects of mu and kappa opiate agonists on dopamine release in the nucleus accumbens and in the dorsal caudate of freely moving rats. J Pharmacol Exp Ther 244:1067-1080. Diez M, Koistinaho J, Kahn K, Games D, Hokfelt T (2000) Neuropeptides in hippocampus and cortex in transgenic mice overexpressing V717F beta-amyloid precursor protein--initial observations. Neuroscience 100:259-286. Dorce VA, Palermo-Neto J (1994) Behavioral and neurochemical changes induced by aging in dopaminergic systems of male and female rats. Physiol Behav 56:1015-1019. Drake CT, Terman GW, Simmons ML, Milner TA, Kunkel DD, Schwartzkroin PA, Chavkin C (1994) Dynorphin opioids present in dentate granule cells may function as retrograde inhibitory neurotransmitters. J Neurosci 14:3736-3750. Faden AI (1992) Dynorphin increases extracellular levels of excitatory amino acids in the brain through a non-opioid mechanism. J Neurosci 12:425-429. Faden AI, Jacobs TP (1984) Dynorphin-related peptides cause motor dysfunction in the rat through a non-opiate action. Br J Pharmacol 81:271-276. Faden AI, Molineaux CJ, Rosenberger JG, Jacobs TP, Cox BM (1985a) Endogenous opioid immunoreactivity in rat spinal cord following traumatic injury. Ann Neurol 17:386-390. Faden AI, Molineaux CJ, Rosenberger JG, Jacobs TP, Cox BM (1985b) Increased dynorphin immunoreactivity in spinal cord after traumatic injury. Regul Pept 11:35- 41. Fernandez B, Antelo MT, Kitchen I, Viveros MP (1999) Effects of neonatal naltrindole treatment on antinociceptive and behavioral responses to mu and kappa agonists in rats. Pharmacol Biochem Behav 62:145-149. Foradori CD, Goodman RL, Adams VL, Valent M, Lehman MN (2005) Progesterone increases dynorphin a concentrations in cerebrospinal fluid and preprodynorphin

124

messenger ribonucleic Acid levels in a subset of dynorphin neurons in the sheep. Endocrinology 146:1835-1842. Forno LS (1996) Neuropathology of Parkinson's disease. J Neuropathol Exp Neurol 55:259-272. Foster TC (1999) Involvement of hippocampal synaptic plasticity in age-related memory decline. Brain Res Brain Res Rev 30:236-249. Freyaldenhoven TE, Cadet JL, Ali SF (1996) The dopamine-depleting effects of 1- methyl-4-phenyl-1,2,3,6-tetrahydropyridine in CD-1 mice are gender-dependent. Brain Res 735:232-238. Friedemann MN, Gerhardt GA (1992) Regional effects of aging on dopaminergic function in the Fischer-344 rat. Neurobiol Aging 13:325-332. Friedman HJ, Jen MF, Chang JK, Lee NM, Loh HH (1981) Dynorphin: a possible modulatory peptide on morphine or beta-endorphin analgesia in mouse. Eur J Pharmacol 69:357-360. Fugger HN, Cunningham SG, Rissman EF, Foster TC (1998) Sex differences in the activational effect of ERalpha on spatial learning. Horm Behav 34:163-170. Gabrilovac J, Balog T, Andreis A (2003) Dynorphin-A(1-17) decreases nitric oxide release and cytotoxicity induced with lipopolysaccharide plus interferon-gamma in murine macrophage cell line J774. Biomed Pharmacother 57:351-358. Gage FH, Dunnett SB, Bjorklund A (1984) Spatial learning and motor deficits in aged rats. Neurobiol Aging 5:43-48. Gallagher M, Nicolle MM (1993) Animal models of normal aging: relationship between cognitive decline and markers in hippocampal circuitry. Behav Brain Res 57:155-162. Garzon J, Sanchez-Blazquez P, Gerhart J, Loh HH, Lee NM (1984) Dynorphin1-13: interaction with other opiate ligand bindings in vitro. Brain Res 302:392-396. Gerfen CR, McGinty JF, Young WS, 3rd (1991) Dopamine differentially regulates dynorphin, substance P, and enkephalin expression in striatal neurons: in situ hybridization histochemical analysis. J Neurosci 11:1016-1031. German DC, Nelson EL, Liang CL, Speciale SG, Sinton CM, Sonsalla PK (1996) The neurotoxin MPTP causes degeneration of specific nucleus A8, A9 and A10 dopaminergic neurons in the mouse. Neurodegeneration 5:299-312. Gillardon F, Beck H, Uhlmann E, Herdegen T, Sandkuhler J, Peyman A, Zimmermann M (1994) Inhibition of c-Fos protein expression in rat spinal cord by antisense oligodeoxynucleotide superfusion. Eur J Neurosci 6:880-884. Giuffra M, Mouradian MM, Davis TL, Ownby J, Chase TN (1993) Dynorphin agonist therapy of Parkinson's disease. Clin Neuropharmacol 16:444-447. Goldstein A, Naidu A (1989) Multiple opioid receptors: ligand selectivity profiles and binding site signatures. Mol Pharmacol 36:265-272. Goldstein A, Tachibana S, Lowney LI, Hunkapiller M, Hood L (1979) Dynorphin-(1-13), an extraordinarily potent opioid peptide. Proc Natl Acad Sci U S A 76:6666-6670. Goodman RL, Coolen LM, Anderson GM, Hardy SL, Valent M, Connors JM, Fitzgerald ME, Lehman MN (2004) Evidence that dynorphin plays a major role in mediating progesterone negative feedback on gonadotropin-releasing hormone neurons in sheep. Endocrinology 145:2959-2967.

125

Goody RJ, Martin KM, Goebel SM, Hauser KF (2003) Dynorphin A toxicity in striatal neurons via an alpha-amino-3-hydroxy-5-methylisoxazole-4-propionate/kainate receptor mechanism. Neuroscience 116:807-816. Graybiel AM, Hirsch EC, Agid Y (1990) The nigrostriatal system in Parkinson's disease. Adv Neurol 53:17-29. Greene E, Naranjo JN (1987) Degeneration of hippocampal fibers and spatial memory deficit in the aged rat. Neurobiol Aging 8:35-43. Greenwald MK, Stitzer ML, Haberny KA (1997) Human pharmacology of the opioid neuropeptide dynorphin A(1-13). J Pharmacol Exp Ther 281:1154-1163. Greenwood CE, Tatton WG, Seniuk NA, Biddle FG (1991) Increased dopamine synthesis in aging substantia nigra neurons. Neurobiol Aging 12:557-565. Hall ME, Hoffer BJ, Gerhardt GA (1989) Rapid and sensitive determination of catecholamines in small tissue samples by high performance liquid chromatography coupled with dual-electrode coulometric electrochemical detection. LCGC 7:258-265. Hara Y, Yakovleva T, Bakalkin G, Pickel VM (2006) Dopamine D1 receptors have subcellular distributions conducive to interactions with prodynorphin in the rat nucleus accumbens shell. Synapse 60:1-19. Harada, Wu J, Haycock JW, Goldstein M (1996) Regulation of L-DOPA biosynthesis by site-specific phosphorylation of tyrosine hydroxylase in AtT-20 cells expressing wild- type and serine 40-substituted enzyme. J Neurochem 67:629-635. Hauser KF, Foldes JK, Turbek CS (1999) Dynorphin A (1-13) neurotoxicity in vitro: opioid and non-opioid mechanisms in mouse spinal cord neurons. Exp Neurol 160:361-375. Hauser KF, Knapp PE, Turbek CS (2001) Structure-activity analysis of dynorphin A toxicity in spinal cord neurons: intrinsic neurotoxicity of dynorphin A and its carboxyl-terminal, nonopioid metabolites. Exp Neurol 168:78-87. Hauser KF, Aldrich JV, Anderson KJ, Bakalkin G, Christie MJ, Hall ED, Knapp PE, Scheff SW, Singh IN, Vissel B, Woods AS, Yakovleva T, Shippenberg TS (2005) Pathobiology of dynorphins in trauma and disease. Front Biosci 10:216-235. Haycock JW, Haycock DA (1991) Tyrosine hydroxylase in rat brain dopaminergic nerve terminals. Multiple-site phosphorylation in vivo and in synaptosomes. J Biol Chem 266:5650-5657. Haycock JW, Becker L, Ang L, Furukawa Y, Hornykiewicz O, Kish SJ (2003) Marked disparity between age-related changes in dopamine and other presynaptic dopaminergic markers in human striatum. J Neurochem 87:574-585. Hebert MA, Gerhardt GA (1998) Normal and drug-induced locomotor behavior in aging: comparison to evoked DA release and tissue content in fischer 344 rats. Brain Res 797:42-54. Heikkila RE, Cabbat FS, Manzino L, Duvoisin RC (1984) Effects of 1-methyl-4-phenyl- 1,2,5,6-tetrahydropyridine on neostriatal dopamine in mice. Neuropharmacology 23:711-713. Heron A, Traiffort E, Allix M, Dimitriadou V, Schwartz JC (1996) Prodynorphin mRNA expression in the rat dentate gyrus after cerebral ischemia. Neuropeptides 30:355-358. Herrera-Marschitz M, Christensson-Nylander I, Sharp T, Staines W, Reid M, Hokfelt T, Terenius L, Ungerstedt U (1986) Striato-nigral dynorphin and substance P pathways in the rat. II. Functional analysis. Exp Brain Res 64:193-207.

126

Hiller JM, Itzhak Y, Simon EJ (1987) Selective changes in mu, delta and kappa opioid receptor binding in certain limbic regions of the brain in Alzheimer's disease patients. Brain Res 406:17-23. Hiramatsu M, Watanabe E (2006) Dynorphin A (2-13) improves mecamylamine-induced learning impairment accompanied by reversal of reductions in acetylcholine release in rats. Neuropeptides 40:47-56. Hiramatsu M, Mori H, Murasawa H, Kameyama T (1996) Improvement by dynorphin A (1-13) of galanin-induced impairment of memory accompanied by blockade of reductions in acetylcholine release in rats. Br J Pharmacol 118:255-260. Hiramatsu M, Sasaki M, Nabeshima T, Kameyama T (1997) Effects of dynorphin A (1- 13) on carbon monoxide-induced delayed amnesia in mice. Pharmacol Biochem Behav 56:73-79. Hornykiewicz O, Kish SJ (1987) Biochemical pathophysiology of Parkinson's disease. Adv Neurol 45:19-34. Hosack DA, Dennis G, Jr., Sherman BT, Lane HC, Lempicki RA (2003) Identifying biological themes within lists of genes with EASE. Genome Biol 4:R70. Hu WH, Lee FC, Wan XS, Chen YT, Jen MF (1996) Dynorphin neurotoxicity induced nitric oxide synthase expression in ventral horn cells of rat spinal cord. Neurosci Lett 203:13-16. Hu WH, Li F, Qiang WA, Liu N, Wang GQ, Xiao J, Liu JS, Liao WH, Jen MF (1999) Dual role for nitric oxide in dynorphin spinal neurotoxicity. J Neurotrauma 16:85-98. Hu WH, Qiang WA, Li F, Liu N, Wang GQ, Wang HY, Wan XS, Liao WH, Liu JS, Jen MF (2000) Constitutive and inducible nitric oxide synthases after dynorphin- induced spinal cord injury. J Chem Neuroanat 17:183-197. Hunter JC, Woodburn VL, Durieux C, Pettersson EK, Poat JA, Hughes J (1995) c-fos antisense oligodeoxynucleotide increases formalin-induced nociception and regulates preprodynorphin expression. Neuroscience 65:485-492. Hurd YL, Svensson P, Ponten M (1999) The role of dopamine, dynorphin, and CART systems in the ventral striatum and amygdala in cocaine abuse. Ann N Y Acad Sci 877:499-506. Hyman SE (2005) Addiction: a disease of learning and memory. Am J Psychiatry 162:1414-1422. Iadarola MJ, Douglass J, Civelli O, Naranjo JR (1986) Increased spinal cord dynorphin mRNA during peripheral inflammation. NIDA Res Monogr 75:406-409. Ingram DK (2000) Age-related decline in physical activity: generalization to nonhumans. Med Sci Sports Exerc 32:1623-1629. Introini-Collison IB, Cahill L, Baratti CM, McGaugh JL (1987) Dynorphin induces task- specific impairment of memory. Psychobiology 15:171-174. Itoh J, Ukai M, Kameyama T (1993a) Dynorphin A-(1-13) markedly improves scopolamine-induced impairment of spontaneous alternation performance in mice. Eur J Pharmacol 236:341-345. Itoh J, Ukai M, Kameyama T (1993b) Dynorphin A-(1-13) potently prevents memory dysfunctions induced by transient cerebral ischemia in mice. Eur J Pharmacol 234:9- 15.

127

Jacobson DA, Cho J, Landa LR, Jr., Tamarina NA, Roe MW, Buxbaum JD, Philipson LH (2006) Downstream regulatory element antagonistic modulator regulates islet prodynorphin expression. Am J Physiol Endocrinol Metab 291:E587-595. Jarrard LE (1978) Selective hippocampal lesions: differential effects on performance by rats of a spatial task with preoperative versus postoperative training. J Comp Physiol Psychol 92:1119-1127. Jiang HK, Owyang VV, Hong JS, Gallagher M (1989) Elevated dynorphin in the hippocampal formation of aged rats: relation to cognitive impairment on a spatial learning task. Proc Natl Acad Sci U S A 86:2948-2951. Johnston HM, Morris BJ (1994) Nitric oxide alters proenkephalin and prodynorphin gene expression in hippocampal granule cells. Neuroscience 61:435-439. Johnston HM, Morris BJ (1995) N-methyl-D-aspartate and nitric oxide regulate the expression of calcium/calmodulin-dependent kinase II in the hippocampal dentate gyrus. Brain Res Mol Brain Res 31:141-150. Josselyn SA, Shi C, Carlezon WA, Jr., Neve RL, Nestler EJ, Davis M (2001) Long-term memory is facilitated by cAMP response element-binding protein overexpression in the amygdala. J Neurosci 21:2404-2412. Kaiser MJ, Tiegs G, Neuhuber WL (2003) Close apposition of dynorphin-positive nerve fibres to lymphocytes in the liver suggests opioidergic neuroimmunomodulation. Histochem Cell Biol 120:213-221. Kessels RP, Postma A, Wijnalda EM, de Haan EH (2000) Frontal-lobe involvement in spatial memory: evidence from PET, fMRI, and lesion studies. Neuropsychol Rev 10:101-113. Khachaturian H, Schaefer MKH, Lewis ME (1993) Anatomy and function of the endogenous opioid system. In: Opioids (Herz A, ed), pp 471-497. Berlin: Springer- Verlag. Khachaturian H, Watson SJ, Lewis ME, Coy D, Goldstein A, Akil H (1982) Dynorphin immunocytochemistry in the rat central nervous system. Peptides 3:941-954. Khazan N, Young GA, Calligaro D (1983) Self-administration of dynorphin-[1-13] and D-ala2-dynorphin-[1-11] (kappa opioid agonists) in morphine (mu opioid agonist)- dependent rats. Life Sci 33 Suppl 1:559-562. King AC, Ho A, Schluger J, Borg L, Kreek MJ (1999) Acute subjective effects of dynorphin A(1-13) infusion in normal healthy subjects. Drug Alcohol Depend 54:87- 90. Kirst A, Wack C, Lutz WK, Eggert A, Kampgen E, Fischer WH (2002) Expression of functional kappa-opioid receptors on murine dendritic cells. Immunol Lett 84:41-48. Kish SJ, Shannak K, Rajput A, Deck JH, Hornykiewicz O (1992) Aging produces a specific pattern of striatal dopamine loss: implications for the etiology of idiopathic Parkinson's disease. J Neurochem 58:642-648. Kitamura Y, Uemura S, Sato T, Yamada N, Morimoto K, Mori A, Kimura H (1995) Induction of c-fos and reduction of dynorphin in dentate granule cells of a rat model of epilepsy produced by systemic administration of kainic acid: an immunohistochemical study. Psychiatry Clin Neurosci 49:S213-216. Knepel W, Schwaninger M, Dohler KD (1985) Corelease of dynorphin-like immunoreactivity, luteinizing hormone, and follicle-stimulating hormone from rat adenohypophysis in vitro. Endocrinology 117:481-487.

128

Koetzner L, Hua XY, Lai J, Porreca F, Yaksh T (2004) Nonopioid actions of intrathecal dynorphin evoke spinal excitatory amino acid and prostaglandin E2 release mediated by cyclooxygenase-1 and -2. J Neurosci 24:1451-1458. Kong LY, McMillian MK, Hudson PM, Jin L, Hong JS (1997) Inhibition of lipopolysaccharide-induced nitric oxide and cytokine production by ultralow concentrations of dynorphins in mixed glia cultures. J Pharmacol Exp Ther 280:61- 66. Kotz CM, Weldon D, Billington CJ, Levine AS (2004) Age-related changes in brain proDynorphin gene expression in the rat. Neurobiol Aging 25:1343-1347. Kreek MJ (1996) Opiates, opioids and addiction. Mol Psychiatry 1:232-254. Kreek MJ, LaForge KS, Butelman E (2002) Pharmacotherapy of addictions. Nat Rev Drug Discov 1:710-726. Kreek MJ, Schluger J, Borg L, Gunduz M, Ho A (1999) Dynorphin A1-13 causes elevation of serum levels of prolactin through an opioid receptor mechanism in humans: gender differences and implications for modulation of dopaminergic tone in the treatment of addictions. J Pharmacol Exp Ther 288:260-269. Kuzmin A, Madjid N, Terenius L, Ogren SO, Bakalkin G (2006) Big dynorphin, a prodynorphin-derived peptide produces NMDA receptor-mediated effects on memory, anxiolytic-like and locomotor behavior in mice. Neuropsychopharmacology 31:1928-1937. Lai J, Ossipov MH, Vanderah TW, Malan TP, Jr., Porreca F (2001) Neuropathic pain: the paradox of dynorphin. Mol Interv 1:160-167. Lai J, Luo MC, Chen Q, Ma S, Gardell LR, Ossipov MH, Porreca F (2006) Dynorphin A activates bradykinin receptors to maintain neuropathic pain. Nat Neurosci 9:1534- 1540. Laughlin TM, Vanderah TW, Lashbrook J, Nichols ML, Ossipov M, Porreca F, Wilcox GL (1997) Spinally administered dynorphin A produces long-lasting allodynia: involvement of NMDA but not opioid receptors. Pain 72:253-260. Li S, Sivam SP, Hong JS (1986) Regulation of the concentration of dynorphin A1-8 in the striatonigral pathway by the dopaminergic system. Brain Res 398:390-392. Li SJ, Sivam SP, McGinty JF, Jiang HK, Douglass J, Calavetta L, Hong JS (1988) Regulation of the metabolism of striatal dynorphin by the dopaminergic system. J Pharmacol Exp Ther 246:403-408. Li SJ, Jiang HK, Stachowiak MS, Hudson PM, Owyang V, Nanry K, Tilson HA, Hong JS (1990) Influence of nigrostriatal dopaminergic tone on the biosynthesis of dynorphin and enkephalin in rat striatum. Brain Res Mol Brain Res 8:219-225. Lind J, Graslund A, Maler L (2006) Membrane interactions of dynorphins(,). Biochemistry 45:15931-15940. Lindgren N, Xu ZQ, Herrera-Marschitz M, Haycock J, Hokfelt T, Fisone G (2001) Dopamine D(2) receptors regulate tyrosine hydroxylase activity and phosphorylation at Ser40 in rat striatum. Eur J Neurosci 13:773-780. Lindgren N, Xu ZQ, Lindskog M, Herrera-Marschitz M, Goiny M, Haycock J, Goldstein M, Hokfelt T, Fisone G (2000) Regulation of tyrosine hydroxylase activity and phosphorylation at Ser(19) and Ser(40) via activation of glutamate NMDA receptors in rat striatum. J Neurochem 74:2470-2477.

129

Liu B, Qin L, Yang SN, Wilson BC, Liu Y, Hong JS (2001) Femtomolar concentrations of dynorphins protect rat mesencephalic dopaminergic neurons against inflammatory damage. J Pharmacol Exp Ther 298:1133-1141. Loacker S, Sayyah M, Wittmann W, Herzog H, Schwarzer C (2007) Endogenous dynorphin in epileptogenesis and epilepsy: anticonvulsant net effect via kappa opioid receptors. Brain 130:1017-1028. Long JB, Mobley WC, Holaday JW (1988a) Neurological dysfunction after intrathecal injection of dynorphin A (1-13) in the rat. I. Injection procedures modify pharmacological responses. J Pharmacol Exp Ther 246:1158-1166. Long JB, Petras JM, Mobley WC, Holaday JW (1988b) Neurological dysfunction after intrathecal injection of dynorphin A (1-13) in the rat. II. Nonopioid mechanisms mediate loss of motor, sensory and autonomic function. J Pharmacol Exp Ther 246:1167-1174. Luellen BA, Szapacs ME, Materese CK, Andrews AM (2006) The neurotoxin 2'-NH2- MPTP degenerates serotonin axons and evokes increases in hippocampal BDNF. Neuropharmacology 50:297-308. Magnusson KR, Scruggs B, Aniya J, Wright KC, Ontl T, Xing Y, Bai L (2003) Age- related deficits in mice performing working memory tasks in a water maze. Behav Neurosci 117:485-495. Maneuf YP, Mitchell IJ, Crossman AR, Woodruff GN, Brotchie JM (1995) Functional implications of kappa opioid receptor-mediated modulation of glutamate transmission in the output regions of the basal ganglia in rodent and primate models of Parkinson's disease. Brain Res 683:102-108. Mansour A, Hoversten MT, Taylor LP, Watson SJ, Akil H (1995) The cloned mu, delta and kappa receptors and their endogenous ligands: evidence for two opioid peptide recognition cores. Brain Res 700:89-98. Marie RM, Defer GL (2003) Working memory and dopamine: clinical and experimental clues. Curr Opin Neurol 16 Suppl 2:S29-35. Marinova Z, Vukojevic V, Surcheva S, Yakovleva T, Cebers G, Pasikova N, Usynin I, Hugonin L, Fang W, Hallberg M, Hirschberg D, Bergman T, Langel U, Hauser KF, Pramanik A, Aldrich JV, Graslund A, Terenius L, Bakalkin G (2005) Translocation of dynorphin neuropeptides across the plasma membrane. A putative mechanism of signal transmission. J Biol Chem 280:26360-26370. Marshall JF, Rosenstein AJ (1990) Age-related decline in rat striatal dopamine metabolism is regionally homogeneous. Neurobiol Aging 11:131-137. McDaniel KL, Mundy WR, Tilson HA (1990) Microinjection of dynorphin into the hippocampus impairs spatial learning in rats. Pharmacol Biochem Behav 35:429-435. McGinty JF, Henriksen SJ, Goldstein A, Terenius L, Bloom FE (1983) Dynorphin is contained within hippocampal mossy fibers: immunochemical alterations after kainic acid administration and colchicine-induced neurotoxicity. Proc Natl Acad Sci U S A 80:589-593. McLaughlin JP, Marton-Popovici M, Chavkin C (2003) Kappa opioid receptor antagonism and prodynorphin gene disruption block stress-induced behavioral responses. J Neurosci 23:5674-5683.

130

Meaney MJ, Aitken DH, van Berkel C, Bhatnagar S, Sapolsky RM (1988) Effect of neonatal handling on age-related impairments associated with the hippocampus. Science 239:766-768. Millan MJ, Millan MH, Pilcher CW, Colpaert FC, Herz A (1985) Chronic pain in the rat: selective alterations in CNS and pituitary pools of dynorphin as compared to vasopressin. Neuropeptides 5:423-424. Miller DB, O'Callaghan JP (2003) Effects of aging and stress on hippocampal structure and function. Metabolism 52:17-21. Miller DB, Ali SF, O'Callaghan JP, Laws SC (1998) The impact of gender and estrogen on striatal dopaminergic neurotoxicity. Ann N Y Acad Sci 844:153-165. Moretti A, Carfagna N, Trunzo F (1987) Effect of aging on monoamines and their metabolites in the rat brain. Neurochem Res 12:1035-1039. Morris RG, Garrud P, Rawlins JN, O'Keefe J (1982) Place navigation impaired in rats with hippocampal lesions. Nature 297:681-683. Mulder AH, Wardeh G, Hogenboom F, Frankhuyzen AL (1984) Kappa- and delta-opioid receptor agonists differentially inhibit striatal dopamine and acetylcholine release. Nature 308:278-280. Naqvi T, Haq W, Mathur KB (1998) Structure-activity relationship studies of dynorphin A and related peptides. Peptides 19:1277-1292. Naranjo JR, Mellstrom B, Achaval M, Sassone-Corsi P (1991) Molecular pathways of pain: Fos/Jun-mediated activation of a noncanonical AP-1 site in the prodynorphin gene. Neuron 6:607-617. Nardo L, Soong Y, Wu D, Young IR, Walker D, Szeto HH (2002) Site and mechanism of action of dynorphin A-(1-13) and N-methyl-D-aspartate on ACTH release in fetal sheep. Am J Physiol Endocrinol Metab 282:E1301-1307. Nguyen PV (2001) CREB and the enhancement of long-term memory. Trends Neurosci 24:314. Nguyen XV, Masse J, Kumar A, Vijitruth R, Kulik C, Liu M, Choi DY, Foster TC, Usynin I, Bakalkin G, Bing G (2005) Prodynorphin knockout mice demonstrate diminished age-associated impairment in spatial water maze performance. Behav Brain Res 161:254-262. Nieoullon A (2002) Dopamine and the regulation of cognition and attention. Prog Neurobiol 67:53-83. Nieoullon A, Coquerel A (2003) Dopamine: a key regulator to adapt action, emotion, motivation and cognition. Curr Opin Neurol 16 Suppl 2:S3-9. Noguchi K, Kowalski K, Traub R, Solodkin A, Iadarola MJ, Ruda MA (1991) Dynorphin expression and Fos-like immunoreactivity following inflammation induced hyperalgesia are colocalized in spinal cord neurons. Brain Res Mol Brain Res 10:227- 233. Nylander I, Sakurada T, Le Greves P, Terenius L (1991) Levels of dynorphin peptides, substance P and CGRP in the spinal cord after subchronic administration of morphine in the rat. Neuropharmacology 30:1219-1223. Ohnishi A, Mihara M, Yasuda S, Tomono Y, Hasegawa J, Tanaka T (1994) Aquaretic effect of the stable dynorphin-A analog E2078 in the human. J Pharmacol Exp Ther 270:342-347.

131

Owen AM, Sahakian BJ, Semple J, Polkey CE, Robbins TW (1995) Visuo-spatial short- term recognition memory and learning after temporal lobe excisions, frontal lobe excisions or amygdalo-hippocampectomy in man. Neuropsychologia 33:1-24. Pickel VM, Chan J, Sesack SR (1993) Cellular substrates for interactions between dynorphin terminals and dopamine dendrites in rat ventral tegmental area and substantia nigra. Brain Res 602:275-289. Pillon B, Czernecki V, Dubois B (2003) Dopamine and cognitive function. Curr Opin Neurol 16 Suppl 2:S17-22. Pliakas AM, Carlson RR, Neve RL, Konradi C, Nestler EJ, Carlezon WA, Jr. (2001) Altered responsiveness to cocaine and increased immobility in the forced swim test associated with elevated cAMP response element-binding protein expression in nucleus accumbens. J Neurosci 21:7397-7403. Ploj K, Roman E, Bergstrom L, Nylander I (2001) Effects of neonatal handling on nociceptin/orphanin FQ and opioid peptide levels in female rats. Pharmacol Biochem Behav 69:173-179. Prediger RD, Batista LC, Medeiros R, Pandolfo P, Florio JC, Takahashi RN (2006) The risk is in the air: Intranasal administration of MPTP to rats reproducing clinical features of Parkinson's disease. Exp Neurol 202:391-403. Privette TH, Terrian DM (1995) Kappa opioid agonists produce anxiolytic-like behavior on the elevated plus-maze. Psychopharmacology (Berl) 118:444-450. Przewlocka B, Stala L, Lason W, Przewlocki R (1983) The effect of various opiate receptor agonists on the seizure threshold in the rat. Is dynorphin an endogenous anticonvulsant? Life Sci 33 Suppl 1:595-598. Przewlocka B, Turchan J, Machelska H, Labuz D, Lason W (1996) Nitric oxide synthase inhibitor L-NAME prevents amphetamine-induced prodynorphin gene expression in the rat. Prog Neuropsychopharmacol Biol Psychiatry 20:1229-1237. Qu ZX, Isaac L (1993) Dynorphin A (1-13) potentiates dynorphin A (1-17) on loss of the tail-flick reflex after intrathecal injection in the rat. Brain Res 610:340-343. Rapp PR, Rosenberg RA, Gallagher M (1987) An evaluation of spatial information processing in aged rats. Behav Neurosci 101:3-12. Redell JB, Moore AN, Dash PK (2003) Expression of the prodynorphin gene after experimental brain injury and its role in behavioral dysfunction. Exp Biol Med (Maywood) 228:261-269. Reeves S, Bench C, Howard R (2002) Ageing and the nigrostriatal dopaminergic system. Int J Geriatr Psychiatry 17:359-370. Reid M, Herrera-Marschitz M, Hokfelt T, Terenius L, Ungerstedt U (1988) Differential modulation of striatal dopamine release by intranigral injection of gamma- aminobutyric acid (GABA), dynorphin A and substance P. Eur J Pharmacol 147:411- 420. Reisberg B, Ferris SH, Anand R, Mir P, Geibel V, De Leon MJ, Roberts E (1983) Effects of naloxone in senile dementia: a double-blind trial. N Engl J Med 308:721-722. Remy P, Samson Y (2003) The role of dopamine in cognition: evidence from functional imaging studies. Curr Opin Neurol 16 Suppl 2:S37-41. Risser D, You ZB, Cairns N, Herrera-Marschitz M, Seidl R, Schneider C, Terenius L, Lubec G (1996) Endogenous opioids in frontal cortex of patients with Down syndrome. Neurosci Lett 203:111-114.

132

Ronken E, Mulder AH, Schoffelmeer AN (1993) Interacting presynaptic kappa-opioid and GABAA receptors modulate dopamine release from rat striatal synaptosomes. J Neurochem 61:1634-1639. Rowe WB, Blalock EM, Chen KC, Kadish I, Wang D, Barrett JE, Thibault O, Porter NM, Rose GM, Landfield PW (2007) Hippocampal expression analyses reveal selective association of immediate-early, neuroenergetic, and myelinogenic pathways with cognitive impairment in aged rats. J Neurosci 27:3098-3110. Sainsbury A, Lin S, McNamara K, Slack K, Enriquez R, Lee NJ, Boey D, Smythe GA, Schwarzer C, Baldock P, Karl T, Lin EJ, Couzens M, Herzog H (2007) Dynorphin Knockout Reduces Fat Mass and Increases Weight Loss during Fasting in Mice. Mol Endocrinol 21:1722-1735. Salvatore MF, Zhang JL, Large DM, Wilson PE, Gash CR, Thomas TC, Haycock JW, Bing G, Stanford JA, Gash DM, Gerhardt GA (2004) Striatal GDNF administration increases tyrosine hydroxylase phosphorylation in the rat striatum and substantia nigra. J Neurochem 90:245-254. Sandin J, Nylander I, Georgieva J, Schott PA, Ogren SO, Terenius L (1998) Hippocampal dynorphin B injections impair spatial learning in rats: a kappa-opioid receptor-mediated effect. Neuroscience 85:375-382. Satoh M, Minami M (1995) Molecular pharmacology of the opioid receptors. Pharmacol Ther 68:343-364. Schlosser B, Kudernatsch MB, Sutor B, ten Bruggencate G (1995) Delta, mu and kappa opioid receptor agonists inhibit dopamine overflow in rat neostriatal slices. Neurosci Lett 191:126-130. Sedelis M, Schwarting RK, Huston JP (2001) Behavioral phenotyping of the MPTP mouse model of Parkinson's disease. Behav Brain Res 125:109-125. Sharifi N, Diehl N, Yaswen L, Brennan MB, Hochgeschwender U (2001) Generation of dynorphin knockout mice. Brain Res Mol Brain Res 86:70-75. Sherman TG, Civelli O, Douglass J, Herbert E, Burke S, Watson SJ (1986) Hypothalamic dynorphin and vasopressin mRNA expression in normal and Brattleboro rats. Fed Proc 45:2323-2327. Shirayama Y, Yamamoto A, Nishimura T, Katayama S, Kawahara R (2007) Subsequent exposure to the choline uptake enhancer MKC-231 antagonizes phencyclidine- induced behavioral deficits and reduction in septal cholinergic neurons in rats. Eur Neuropsychopharmacol. Shukla VK, Prasad JA, Lemaire S (1997) Nonopioid motor effects of dynorphin A and related peptides: structure dependence and role of the N-methyl-D-aspartate receptor. J Pharmacol Exp Ther 283:604-610. Siggins GR, Henriksen SJ, Chavkin C, Gruol D (1986) Opioid peptides and epileptogenesis in the limbic system: cellular mechanisms. Adv Neurol 44:501-512. Simonato M, Romualdi P (1996) Dynorphin and epilepsy. Prog Neurobiol 50:557-583. Simonin F, Valverde O, Smadja C, Slowe S, Kitchen I, Dierich A, Le Meur M, Roques BP, Maldonado R, Kieffer BL (1998) Disruption of the kappa-opioid receptor gene in mice enhances sensitivity to chemical visceral pain, impairs pharmacological actions of the selective kappa-agonist U-50,488H and attenuates morphine withdrawal. Embo J 17:886-897.

133

Simpson JN, Zhang WQ, Bing G, Hong JS (1997) Kainic acid-induced sprouting of dynorphin- and enkephalin-containing mossy fibers in the dentate gyrus of the rat hippocampus. Brain Res 747:318-323. Singh IN, Goody RJ, Goebel SM, Martin KM, Knapp PE, Marinova Z, Hirschberg D, Yakovleva T, Bergman T, Bakalkin G, Hauser KF (2003) Dynorphin A (1-17) induces apoptosis in striatal neurons in vitro through alpha-amino-3-hydroxy-5- methylisoxazole-4-propionate/kainate receptor-mediated cytochrome c release and caspase-3 activation. Neuroscience 122:1013-1023. Smith AP, Lee NM (1988) Pharmacology of dynorphin. Annu Rev Pharmacol Toxicol 28:123-140. Spadoni F, Martella G, Martorana A, Lavaroni F, D'Angelo V, Bernardi G, Stefani A (2004) Opioid-mediated modulation of calcium currents in striatal and pallidal neurons following reserpine treatment: focus on kappa response. Synapse 51:194-205. Spampinato S, Canossa M, Campana G, Carboni L, Bachetti T (1995) Estrogen regulation of prodynorphin gene expression in the rat adenohypophysis: effect of the antiestrogen tamoxifen. Endocrinology 136:1589-1594. Spanagel R, Herz A, Shippenberg TS (1992) Opposing tonically active endogenous opioid systems modulate the mesolimbic dopaminergic pathway. Proc Natl Acad Sci U S A 89:2046-2050. Specker S, Wananukul W, Hatsukami D, Nolin K, Hooke L, Kreek MJ, Pentel PR (1998) Effects of dynorphin A(1-13) on opiate withdrawal in humans. Psychopharmacology (Berl) 137:326-332. Stanford JA, Vorontsova E, Surgener SP, Gerhardt GA, Fowler SC (2003) Aged Fischer 344 rats exhibit altered orolingual motor function: relationships with nigrostriatal neurochemical measures. Neurobiol Aging 24:259-266. Steiner H, Gerfen CR (1998) Role of dynorphin and enkephalin in the regulation of striatal output pathways and behavior. Exp Brain Res 123:60-76. Stevens CW, Yaksh TL (1986) Dynorphin A and related peptides administered intrathecally in the rat: a search for putative kappa opiate receptor activity. J Pharmacol Exp Ther 238:833-838. Sunderland T, Berrettini WH, Molchan SE, Lawlor BA, Martinez RA, Vitiello B, Tariot PN, Cohen RM (1991) Reduced cerebrospinal fluid dynorphin A1-8 in Alzheimer's disease. Biol Psychiatry 30:81-87. Sundstrom E, Fredriksson A, Archer T (1990) Chronic neurochemical and behavioral changes in MPTP-lesioned C57BL/6 mice: a model for Parkinson's disease. Brain Res 528:181-188. Sutherland C, Alterio J, Campbell DG, Le Bourdelles B, Mallet J, Haavik J, Cohen P (1993) Phosphorylation and activation of human tyrosine hydroxylase in vitro by mitogen-activated protein (MAP) kinase and MAP-kinase-activated kinases 1 and 2. Eur J Biochem 217:715-722. Sydbom A, Terenius L (1985) The histamine-releasing effect of dynorphin and other peptides possessing Arg-Pro sequences. Agents Actions 16:269-272. Tang Q, Lynch RM, Porreca F, Lai J (2000) Dynorphin A elicits an increase in intracellular calcium in cultured neurons via a non-opioid, non-NMDA mechanism. J Neurophysiol 83:2610-2615.

134

Tang Q, Gandhoke R, Burritt A, Hruby VJ, Porreca F, Lai J (1999) High-affinity interaction of (des-Tyrosyl)dynorphin A(2-17) with NMDA receptors. J Pharmacol Exp Ther 291:760-765. Tan-No K, Cebers G, Yakovleva T, Hoon Goh B, Gileva I, Reznikov K, Aguilar- Santelises M, Hauser KF, Terenius L, Bakalkin G (2001) Cytotoxic effects of dynorphins through nonopioid intracellular mechanisms. Exp Cell Res 269:54-63. Tan-No K, Esashi A, Nakagawasai O, Niijima F, Tadano T, Sakurada C, Sakurada T, Bakalkin G, Terenius L, Kisara K (2002) Intrathecally administered big dynorphin, a prodynorphin-derived peptide, produces nociceptive behavior through an N-methyl- D-aspartate receptor mechanism. Brain Res 952:7-14. Taquet H, Javoy-Agid F, Giraud P, Legrand JC, Agid Y, Cesselin F (1985) Dynorphin levels in parkinsonian patients: Leu5-enkephalin production from either proenkephalin A or prodynorphin in human brain. Brain Res 341:390-392. Terman GW, Wagner JJ, Chavkin C (1994) Kappa opioids inhibit induction of long-term potentiation in the dentate gyrus of the guinea pig hippocampus. J Neurosci 14:4740- 4747. Terman GW, Drake CT, Simmons ML, Milner TA, Chavkin C (2000) Opioid modulation of recurrent excitation in the hippocampal dentate gyrus. J Neurosci 20:4379-4388. Thai L, Hong JS, Wiley RG, Gallagher M (1996) The regulation of hippocampal dynorphin by neural/neuroendocrine pathways: models for effects of aging on an opioid peptide system. Neuroscience 70:661-671. Tilson H, McLamb R, Hong J (1986) Behavioral effects of centrally administered dynorphin and [D-ala2-D- leu] enkephalin (DADLE) in rats. Neuropeptides 8:193- 206. Ukai M, Itoh J, Kobayashi T, Shinkai N, Kameyama T (1997) Effects of the kappa-opioid dynorphin A(1-13) on learning and memory in mice. Behav Brain Res 83:169-172. Vanderah TW, Laughlin T, Lashbrook JM, Nichols ML, Wilcox GL, Ossipov MH, Malan TP, Jr., Porreca F (1996) Single intrathecal injections of dynorphin A or des- Tyr-dynorphins produce long-lasting allodynia in rats: blockade by MK-801 but not naloxone. Pain 68:275-281. Ventura C, Zinellu E, Maninchedda E, Maioli M (2003) Dynorphin B is an agonist of nuclear opioid receptors coupling nuclear protein kinase C activation to the transcription of cardiogenic genes in GTR1 embryonic stem cells. Circ Res 92:623- 629. Vijitruth R, Liu M, Choi DY, Nguyen XV, Hunter RL, Bing G (2006) Cyclooxygenase-2 mediates microglial activation and secondary dopaminergic cell death in the mouse MPTP model of Parkinson's disease. J Neuroinflammation 3:6. von Bohlen und Halbach O, Unsicker K (2002) Morphological alterations in the amygdala and hippocampus of mice during ageing. Eur J Neurosci 16:2434-2440. Wagner JJ, Caudle RM, Chavkin C (1992) Kappa-opioids decrease excitatory transmission in the dentate gyrus of the guinea pig hippocampus. J Neurosci 12:132- 141. Wagner JJ, Terman GW, Chavkin C (1993) Endogenous dynorphins inhibit excitatory neurotransmission and block LTP induction in the hippocampus. Nature 363:451- 454.

135

Wallace MS, Moulin D, Clark AJ, Wasserman R, Neale A, Morley-Forster P, Castaigne JP, Teichman S (2006) A Phase II, multicenter, randomized, double-blind, placebo- controlled crossover study of CJC-1008--a long-acting, parenteral opioid analgesic-- in the treatment of postherpetic neuralgia. J Opioid Manag 2:167-173. Wang Z, Gardell LR, Ossipov MH, Vanderah TW, Brennan MB, Hochgeschwender U, Hruby VJ, Malan TP, Jr., Lai J, Porreca F (2001) Pronociceptive actions of dynorphin maintain chronic neuropathic pain. J Neurosci 21:1779-1786. Waters CM, Peck R, Rossor M, Reynolds GP, Hunt SP (1988) Immunocytochemical studies on the basal ganglia and substantia nigra in Parkinson's disease and Huntington's chorea. Neuroscience 25:419-438. Watson SJ, Khachaturian H, Akil H, Coy DH, Goldstein A (1982a) Comparison of the distribution of dynorphin systems and enkephalin systems in brain. Science 218:1134- 1136. Watson SJ, Khachaturian H, Coy D, Taylor L, Akil H (1982b) Dynorphin is located throughout the CNS and is often co-localized with alpha-neo-endorphin. Life Sci 31:1773-1776. Weisskopf MG, Zalutsky RA, Nicoll RA (1993) The opioid peptide dynorphin mediates heterosynaptic depression of hippocampal mossy fibre synapses and modulates long- term potentiation. Nature 365:188. Wong JY, Clifford JJ, Massalas JS, Finkelstein DI, Horne MK, Waddington JL, Drago J (2003) Neurochemical changes in dopamine D1, D3 and D1/D3 receptor knockout mice. Eur J Pharmacol 472:39-47. Woods A, Zangen A (2001) A direct chemical interaction between dynorphin and excitatory amino acids. Neurochem Res 26:395-400. Woods AS, Kaminski R, Oz M, Wang Y, Hauser K, Goody R, Wang HY, Jackson SN, Zeitz P, Zeitz KP, Zolkowska D, Schepers R, Nold M, Danielson J, Graslund A, Vukojevic V, Bakalkin G, Basbaum A, Shippenberg T (2006) Decoy Peptides that Bind Dynorphin Noncovalently Prevent NMDA Receptor-Mediated Neurotoxicity. J Proteome Res 5:1017-1023. Yakovleva T, Marinova Z, Kuzmin A, Seidah NG, Haroutunian V, Terenius L, Bakalkin G (2006a) Dysregulation of dynorphins in Alzheimer disease. Neurobiol Aging. Yakovleva T, Bazov I, Cebers G, Marinova Z, Hara Y, Ahmed A, Vlaskovska M, Johansson B, Hochgeschwender U, Singh IN, Bruce-Keller AJ, Hurd YL, Kaneko T, Terenius L, Ekstrom TJ, Hauser KF, Pickel VM, Bakalkin G (2006b) Prodynorphin storage and processing in axon terminals and dendrites. Faseb J 20:2124-2126. Zamir N, Quirion R, Segal M (1985) Ontogeny and regional distribution of proenkephalin- and prodynorphin-derived peptides and opioid receptors in rat hippocampus. Neuroscience 15:1025-1034. Zhang L, Peoples RW, Oz M, Harvey-White J, Weight FF, Brauneis U (1997) Potentiation of NMDA receptor-mediated responses by dynorphin at low extracellular glycine concentrations. J Neurophysiol 78:582-590. Zhang RX, Lao L, Qiao JT, Ruda MA (2004a) Effects of aging on hyperalgesia and spinal dynorphin expression in rats with peripheral inflammation. Brain Res 999:135- 141.

136

Zhang WQ, Mundy WR, Thai L, Hudson PM, Gallagher M, Tilson HA, Hong JS (1991) Decreased glutamate release correlates with elevated dynorphin content in the hippocampus of aged rats with spatial learning deficits. Hippocampus 1:391-397. Zhang Y, Qiao JT, Dafny N (2005) c-Fos antisense oligodeoxynucleotide offsets behavioral nociceptive responses and both up-regulations of c-Fos protein and dynorphin a (1-8) in dorsal horn: a study using the formalin test in rats. Int J Neurosci 115:935-948. Zhang Y, Butelman ER, Schlussman SD, Ho A, Kreek MJ (2004b) Effect of the endogenous kappa opioid agonist dynorphin A(1-17) on cocaine-evoked increases in striatal dopamine levels and cocaine-induced place preference in C57BL/6J mice. Psychopharmacology (Berl) 172:422-429. Zigmond MJ (1997) Do compensatory processes underlie the preclinical phase of neurodegenerative disease? Insights from an animal model of parkinsonism. Neurobiol Dis 4:247-253.

137

VITA

Date of Birth: March 8, 1978 Place of Birth: Ho Chi Minh City, Vietnam

Education: University of Kentucky College of Medicine Doctor of Medicine, May 13, 2006

University of Oklahoma Health Sciences Center Oklahoma Center for Neurosciences

Princeton University Department of Chemical Engineering Bachelor of Science in Engineering, June 2, 1998

Professional Positions: Resident Physician Department of Diagnostic Radiology University of Kentucky, Lexington, KY July 1, 2007 —

Resident Physician and Clinical Instructor Department of Internal Medicine University of Cincinnati, Cincinnati, OH July 1, 2006 – June 30, 2007

Scholastic & Professional Honors: National Institutes of Health Ruth L. Kirschstein National Research Service Award 2001-2006 University of Kentucky M.D.-Ph.D. Fellowship 2001-2006 Alpha Omega Alpha Medical Honor Society 2005 Beale Scholarship, College of Medicine, University of Kentucky 2005 Glenn Foundation / American Federation for Aging Research Scholarship for Research in Biology of Aging 2004 Dissertation Enhancement Award, Graduate School, University of Kentucky 2004 Promoted with High Distinction, College of Medicine, University of Kentucky 2002, 2003, 2005, 2006 Howard Hughes Medical Institute Predoctoral Fellowship 1999-2001 Phi Phi Honor Society 2000 Alpha Epsilon Lambda Honor Society 2000 National Science Foundation Graduate Research Fellowship Award 1999 Sigma Xi Book Award, Princeton, NJ chapter 1998 Hoechst-Celanese Corporation Senior Thesis Grant Award 1997 Robert C. Byrd Scholar 1994-1998 Academic All-State Scholar 1994 Presidential Scholar Semi-Finalist 1994 National Merit Scholar 1994 Professional Publications:

138

Peer-reviewed Publications

Arimoto T, Choi DY, Lu X, Liu M, Nguyen XV, Zheng N, Stewart CA, Kim HC, and Bing G. (2007) Interleukin-10 protects against inflammation-mediated degeneration of dopaminergic neurons in substantia nigra. Neurobiol Aging 28: 894-906. Vijitruth R, Liu M, Choi DY, Nguyen XV, Hunter RL, and Bing G. (2006) Cyclooxygenase-2 mediates microglial activation and secondary dopaminergic cell death in the mouse MPTP model of Parkinson's disease. J Neuroinflammation 3: 6. Bing G, Nguyen XV, Liu M, Markesbery WR, and Sun A. (2006) Biophysical and biochemical characterization of intrinsic fluorescence from neurofibrillary tangles. Neurobiol Aging 27: 823-830. Zhang J, Stanton DM, Nguyen XV, Liu M, Zhang Z, Gash D, and Bing G. (2005) Intrapallidal lipopolysaccharide injection increases iron and ferritin levels in glia of the rat substantia nigra and induces locomotor deficits. Neuroscience 135: 829-838. Nguyen XV, Masse J, Kumar A, Vijitruth R, Kulik C, Liu M, Choi D, Foster TC, Usynin I, Bakalkin G, and Bing G. (2005) Prodynorphin knockout mice demonstrate diminished age-associated impairment in spatial water maze performance. Behav Brain Res 161: 254-262. Sun A, Liu M, Nguyen XV, and Bing G. (2003) p38 MAP kinase is activated at early stages in Alzheimer’s disease brain. Exp Neurol 183: 394-405. Sun A, Nguyen XV, and Bing G. (2002) Comparative analysis of an improved thioflavin-S stain, Gallyas silver stain, and immunohistochemistry for neurofibrillary tangle demonstration on same sections. J Histochem Cytochem 50: 463-472. Sun A, Nguyen XV, and Bing G. (2001) A novel method for direct visualization of neurofibrillary pathology in Alzheimer’s disease. J Neurosci Methods 111: 17-27. Robinson KA, Stewart CA, Pye QN, Nguyen X, Kenney L, Salzman S, Floyd RA, and Hensley K. (1999) Redox-sensitive protein phosphatase activity regulates the phosphorylation state of p38 protein kinase in primary astrocyte culture. J Neurosci Res 55: 724-732. Hensley K, Floyd RA, Zheng N-Y, Nael R, Robinson KA, Nguyen X, Pye QN, Stewart CA, Geddes J, Markesbery WR, Patel E, Johnson GVW, and Bing G. (1999) p38 kinase is activated in the Alzheimer's disease brain. J Neurochem 72: 2053-2058. Schneider JE Jr, Tabatabaie T, Maidt L, Horning-Smith R, Nguyen X, Pye Q, and Floyd RA. (1998) Potential mechanisms of photodynamic inactivation of virus by methylene blue: I. RNA-protein crosslinks and other oxidative lesions in Q bacteriophage. Photochem Photobiol 67: 350-357.

139

Abstracts, Book Chapters, and Presentations

Nguyen XV, Liu M, Hunter R, Choi D, Bakalkin G, and Bing GY. (2004) Deletion of prodynorphin prevents age-dependent loss of striatal dopamine in mice. #905.8. The 34th Annual Meeting of Soc for Neurosci, San Diego, CA. October 27. Liu M, Hunter R, Nguyen XV, Kulik C, Kim H, and Bing GY. (2004) Microsomal epoxide hydrolase deficiency enhances tyrosine hydroxylase phosphorylation in mice. #562.26. The 34th Annual Meeting of Soc for Neurosci, San Diego, CA. October 25. Nguyen XV and Bing GY. (2004) Radioimmunological quantitation of prodynorphin- derived peptides in mice in aging and an MPTP model. Experimental Alcohol and Drug Addiction Research Seminar, Department of Clinical Neuroscience, Karolinska Institute, Stockholm, Sweden. September 29. Nguyen XV, Liu M, Choi D, Kulik C, Usynin I, Bakalkin G, and Bing GY. (2003) Dynorphin A is elevated in striatum and frontal cortex of aged mice and may underlie age-related alterations in immune function. #302.9. The 33rd Annual Meeting of Soc for Neurosci, New Orleans, LA. November 9. Nguyen XV and Bing GY. (2003) Amelioration of age-related memory impairment in prodynorphin knockout mice. Howard Hughes Medical Institute 2003 Meeting of Predoctoral and Physician Postdoctoral Fellows, Chevy Chase, MD. September 16. Nguyen XV, Kulik CM, Vijitruth R, Peng X, Stromberg AJ, Kim H-C, and Bing GY. (2002) Microarray analysis of hippocampi from aged prodynorphin knockout mice reveals altered expression of genes involved in inflammation and neuronal function. #294.11. The 32nd Annual Meeting of Soc for Neurosci, Orlando, FL. November 4. Vijitruth R, Feng Z, Kulik CM, Nguyen XV, Kim YC, and Bing GY. (2002) Cox-2 deficient mice are resistant to MPTP neurotoxicity. #690.10. The 32nd Annual Meeting of Soc for Neurosci, Orlando, FL. November 6. Nguyen XV, Masse J, Kumar A, Foster TC, Kim H-C, and Bing GY. (2001) Role of dynorphin in age-related impairment of spatial learning. #312.6. The 31st Annual Meeting of Soc for Neurosci, San Diego, CA. November 12. Sun AY, Nguyen XV, and Bing GY. (2001) A novel method for direct visualization of neurofibrillary pathology in Alzheimer’s disease. #429.14. The 31st Annual Meeting of Soc for Neurosci, San Diego, CA. November 12. Arimoto T, Kennedy S, Nguyen XV, and Bing GY. (2001) The role of inducible nitric oxide synthase and nitric oxide in lipopolysaccharide-induced neurodegeneration in rat substantia nigra. #379.12. The 31st Annual Meeting of Soc for Neurosci, San Diego, CA. November 11. Hensley K, Gabbita SP, Nguyen X, Salsman S, Mou S, Szweda L, Humphries K, Floyd RA, and Markesbery WR. (2000) Characterization of 4-hydroxynonenal-reactive proteins in the Alzheimer's disease brain. #P08-5. The 10th Biennial Meeting of the Society for Free Radical Research International, Kyoto, Japan. October 2000.

140

Nguyen XV, Hensley K, Stewart CA, Zheng N-Y, Jin L, Zhu M, Williamson K, Floyd RA, and Bing G. (1999) Oxidant-sensitive signal transduction pathways in hippocampal excitotoxicity. Free Radic Biol Med 27: S142, #436. The 6th Annual Meeting of the Oxygen Society, New Orleans, LA. Hensley K, Bing GY, Nael R, Zheng NY, Robinson KA, Nguyen X, Patel E, Markesbery WR, and Floyd RA. (1998) Redox sensitive p38 kinase is activated in the Alzheimer brain. Free Radic Biol Med 25: S111, #316. The 5th Annual Meeting of the Oxygen Society, Washington, DC. Robinson KA, Hensley K, Stewart CA, Pye QN, Nguyen X, and Floyd RA. (1998) Effects of interleukin 1 , H2O2 and antioxidants on p38 kinase phosphorylation in primary astrocyte culture. Free Radic Biol Med 25: S112, #320. The 5th Annual Meeting of the Oxygen Society, Washington, DC. Robinson KA, Hensley K, Stewart CA, Pye Q, Nguyen X, and Floyd RA. (1997) Global protein phosphorylation state is decreased by PBN and NAC in rat brain and primary rat glial cell culture. Oxygen ’97, p. 69, #2-9. The 4th Annual Meeting of the Oxygen Society, San Francisco, CA. Hensley K, Pye Q, Maidt L, Liu G-J, Nguyen X, Robinson K, and Floyd RA. (1996) Altered DNA-binding behavior and oxyradical production by nonenzymatically glycated histone and model DNA-binding polypeptides. Oxygen ’96, p. 33, #1-21. The 3rd Annual Meeting of the Oxygen Society, Miami, FL. Stewart CA, Tabatabaie T, Pye Q, Nguyen X, and Floyd RA. (1996) Inhibition by - phenyl-tert-butyl nitrone (PBN) of gp120-mediated induction of nitric oxide and TNF in rat brain mixed cell cultures. Oxygen ’96, p. 78, #2-42. The 3rd Annual Meeting of the Oxygen Society, Miami, FL. Schneider JE Jr, Pye QN, Maidt ML, Tabatabaie T, Nguyen X, Robinson K, and Floyd RA. (1996) Spectrum of oxidative damage in RNA, DNA and protein, including nucleic acid-protein crosslinkage formation following exposure to the photoactive agent methylene blue. Proc Am Assoc Cancer Res 37: 121, #841. The 87th Annual Meeting of the American Assn for Cancer Research, Washington, DC. Floyd RA, Maidt ML, Kotake Y, Sang H, Tabatabaie T, Pye Q, Schneider JE Jr, Stewart C, and Nguyen X. (1996) Nitric oxide reaction with 8-hydroxyguanine in free nucleosides and macromolecules. Proc Am Assoc Cancer Res 37: 144, #998. The 87th Annual Meeting of the American Assn for Cancer Research, Washington, DC. Nguyen XV and Floyd RA. (1994) Methylene blue-induced photoinactivation of Q bacteriophage. In Thurman WG (Ed.), Fleming Scholars Scientific Reports, pp. 65- 77. Oklahoma Medical Research Foundation, Oklahoma City, OK.

Xuan V. Nguyen

141