This item is only available as the following downloads:
1 NEUROBIOLOGICAL BASIS OF SELF INJURIOUS BEHAVIOR: FOLLOWING CONVERGENT PATHWAYS TOWARD NOVEL PHARMACOTHERAPY By AMBER M. MUEHLMANN VAN MATRE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTI AL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 201 1
2 201 1 Amber M. Muehlmann Van Matre
3 To doing things the right way, m aybe Ill try that next time
4 ACKNOWLEDGMENTS I would like to thank my advisor, Dr. Darragh Devine, for the opportunity to study in his laboratory and to pursue my interests in genetic vulnerability and behavioral pathologies. Throughout the years of this pursuit he has e xhibited limitless patience, guidance, and hard work. I also thank other faculty at the University of Florida, including my committee members who have mentored me in all aspects of academia: scholarship, research, and service. This incl udes Dr. Mark Lewis, Dr. Timothy Vollmer, Dr. Joanna Peris, Dr. Neil Rowland, Dr. Robin West, Dr. Alan Spector, and Dr. Sue S empleRowland. I would also like to thank current and past members of the lab and the behavioral neuroscience program, Dr. Catherine Marcinkiewcz, Dr. Megan Green, Jennifer Wilkinson, Shannon Wolfman, Kanita Beba, Nathan We instock, Brandon Brown, Dr. Cheryl Vaughan, and Kim Robertson for their assistance with the completion of these projects, both scientifically and psychologically. Finally, as I reflect on the relationships Ive made, lost and somehow managed to retain along this journey, I d like thank you all for your support, humor, and love.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 9 LIST OF FIGURES ........................................................................................................ 10 LIST OF ABBREVIATIONS ........................................................................................... 12 ABSTRACT ................................................................................................................... 14 CHAPTER 1 INTRODUCTION .................................................................................................... 17 Self Injurious Behavior ............................................................................................ 17 Clinical Phenomenolog y of Self Injurious Behavior .......................................... 18 Emergence of self injurious behavior ......................................................... 19 Maintenance of self injurious behavior ....................................................... 20 Assessment of self injurious behavior ........................................................ 22 Comorbid behaviors and related behavioral pathologies ........................... 24 Effects of self injurious behavior on self and others ................................... 25 Clinical Syndromes ........................................................................................... 26 Lesch Nyhan syndrome ............................................................................. 26 Rett syndrome ............................................................................................ 28 Prader Willi syndrome ................................................................................ 28 Fragile X syndrome .................................................................................... 29 Smith Magenis syndrome ........................................................................... 30 Cornelia de Lange syndrome ..................................................................... 31 Autism ........................................................................................................ 32 Neurobiology and Neurochemistry of the Clinical Syndromes Associated with Self Injurious Behavior ........................................................................... 33 Therapeutic Strategies for Sel f Injurious Behavior ........................................... 42 Pharmacotherapies .................................................................................... 43 Behavioral therapies .................................................................................. 46 Animal Models of Self Injurious Behavior ............................................................... 46 Nonhuman Primate Models .............................................................................. 47 6 Hydroxydopamine Model ............................................................................... 49 Bay K 8644 Model ............................................................................................ 51 Methamphetamine Model ................................................................................. 52 Muscimol Model ............................................................................................... 54 Pemoline Model ................................................................................................ 54 Summary .......................................................................................................... 55
6 2 CHANGES IN PAIN RESPONSIVENESS DURING PEMOLINE ADMIN IS TRA TION ................................................................................................. 57 Background ............................................................................................................. 57 Methods .................................................................................................................. 60 Experiment 1: Escalating Thermal Pain Testing in Drug Nave Rats ................ 60 Animals ...................................................................................................... 60 Hot plate testing ......................................................................................... 61 Experiment 2: Thermal Pain Testing in Pemolineand VehicleTreated Rats .. 61 Animals ...................................................................................................... 61 Hot plate testing habituation ....................................................................... 61 Drug ........................................................................................................... 62 Drug treatment ........................................................................................... 62 Assays of self injury, stereotypies, and locomotion .................................... 62 Hot plate testing ......................................................................................... 64 Statistical analyses .................................................................................... 64 Resul ts .................................................................................................................... 64 Experiment 1 .................................................................................................... 64 Experiment 2 .................................................................................................... 65 Discussion .............................................................................................................. 68 3 EFFECTS OF REPEATED STRESS ON PEMOLINEINDUCED SELF INJURIOUS BEHAVIOR ......................................................................................... 70 Background ............................................................................................................. 70 Methods .................................................................................................................. 71 Animals ............................................................................................................. 71 Surgery ............................................................................................................. 72 Social Defeat Procedure ................................................................................... 72 Drugs ................................................................................................................ 73 Drug Treatment ................................................................................................ 73 Assays of Self Injury and S tereotypy ................................................................ 74 Protein Analyses .............................................................................................. 75 Statistical Analyses .......................................................................................... 76 Results .................................................................................................................... 77 Discussion .............................................................................................................. 81 4 INDIVIDUAL DIFFERENCES IN VULNERABILITY FOR PEMOLINE INDUCED SELF INJURIOUS BEHAVIOR IS PREDICTED BY STRESS RESPONSIVENESS ............................................................................................... 86 Background ............................................................................................................. 86 Methods .................................................................................................................. 87 Animals ............................................................................................................. 87 Drug ................................................................................................................. 88 Stress Responsiveness Screening ................................................................... 88 Drug Treatment ................................................................................................ 89 Assays of Self Injury, Stereotypy, and Related Behaviors ................................ 89 Statistical Analyses .......................................................................................... 91
7 Results .................................................................................................................... 92 Discussion .............................................................................................................. 97 5 CHARACTERIZATION OF DRUG TITERS, MONOAMINES, MONOAMINE METABOLITES, AND AMINO ACIDS IN THE PEM OLINE MODEL OF SELF INJURIOUS BEHAVIOR ....................................................................................... 102 Background ........................................................................................................... 102 Methods ................................................................................................................ 102 Animals ........................................................................................................... 102 Drug ............................................................................................................... 103 Drug and Vehicle Treatment ........................................................................... 103 Assays of Self Injury ....................................................................................... 103 Decapitation Schedule ................................................................................... 104 Sample Collection .......................................................................................... 105 High Pressure Liquid Chromatography (HPLC) System ................................. 105 Pemoline Titer Analyses ................................................................................. 106 Plasma pemoline extraction ..................................................................... 106 Brain pemoline extraction ......................................................................... 106 Pemoline standards ........................................................................................ 107 Ultraviolet detection ................................................................................. 107 Corticosterone Radioimmunoassay ................................................................ 107 Monoamine Analyses ..................................................................................... 107 M onoamine and metabolite standards ..................................................... 108 Electrochemical detection of monoamines and metabolites .................... 108 Amino Acid Analyses ...................................................................................... 108 Amino acid derivatization ......................................................................... 108 Amino acid standards .............................................................................. 109 Electrochemical detection of amino acids ................................................ 109 Protein Content Analyses ............................................................................... 109 Pemoline, Monoamine, Metabolite, and Amino Acid Concentration Analyses 109 Statistical Analyses ........................................................................................ 110 Results .................................................................................................................. 110 PemolineInduced Self Injurious Behavior ...................................................... 110 Pemoline Analyses ......................................................................................... 111 Corticosterone Analysis .................................................................................. 113 Monoamine, Metabolite, and Amino Acid Analyses ........................................ 115 Striatum .................................................................................................... 115 Ventral tegmentum ................................................................................... 118 Cortices .................................................................................................... 121 Amygdala ................................................................................................. 124 Hippocampus ........................................................................................... 126 Septum .................................................................................................... 128 Discussion ............................................................................................................ 131 6 NEUROTENSIN PLAYS A MODULATORY ROLE IN PEMOLINE INDUCED SELF INJURIOUS BEHAVIOR ............................................................................. 137 Background ........................................................................................................... 137
8 Methods ................................................................................................................ 140 Experiment 1: Neurotensin Concentrations in Vehicle and Pemoline Treated Rats ............................................................................................... 140 Animals .................................................................................................... 140 Drug ......................................................................................................... 141 Dru g treatment ......................................................................................... 141 Assay of self injury ................................................................................... 141 Neurotensin radioimmunoassay ............................................................... 141 Statistical analyses .................................................................................. 142 Experiment 2: Effects of NTS1 Antagonist, SR48692, on PemolineInduced Self Injurious Behavior ................................................................................ 142 Animals .................................................................................................... 142 Drugs ....................................................................................................... 142 Drug treatments ....................................................................................... 142 Assays of self i njury ................................................................................. 143 Monoamine analyses ............................................................................... 144 Statistical analyses .................................................................................. 144 Experi ment 3: Effects of NTS1 Agonist, PD149163, on PemolineInduced Self Injurious Behavior ................................................................................ 145 Animals .................................................................................................... 145 Drugs ....................................................................................................... 145 Drug treatments ....................................................................................... 145 Assays of self injury, monoamines, metabolites, and protein content ...... 145 Statistical analyses .................................................................................. 146 Results .................................................................................................................. 146 Experiment 1 .................................................................................................. 146 Experiment 2 .................................................................................................. 147 Experiment 3 .................................................................................................. 152 Discussion ............................................................................................................ 156 7 GENERAL DI SCUSSION ..................................................................................... 159 LIST OF REFERENCES ............................................................................................. 1 63 BIOGRAPHICAL SKETCH .......................................................................................... 191
9 LIST OF TABLES Table page 5 1 Decapitation schedule for pemolineand vehicletreated rats. ......................... 105 5 2 Summary of significant effects of pemoline on regional concentrations of monoamines, metabolites, amino acids, and of monoamine synthesis and turnover ratios.. ................................................................................................. 131
10 LIST OF FIGURES Figure page 2 1 Hot pla te test behaviors of drug nave rats during stimulation at escalating temperatures ...................................................................................................... 66 2 2 Pemolineinduced self injurious behavior ........................................................... 67 2 3 Rates of homecage behaviors of pem oline and vehicletreated rats ................ 67 2 4 Effects of pemolineand vehiclet reatment on hot plate behaviors .................... 68 3 1 Effects of repeated social defeat stress on pemolineind uced self injury and stereotypy ........................................................................................................... 79 3 2 Effects of pemoline treatment and repeated stress on surface (S) and intracellular (I) levels of GluR1, GluR2, NTS1, and OCT3 in the nucleus accumbens. ........................................................................................................ 80 4 1 Differences between HR and LR rats in the circular corridor .............................. 94 4 2 Effects of stress responsiveness on pemoline induced self injury ...................... 95 4 3 Effect of stress responsiveness on pemolineinduced stereotypy, grooming, locomotion, and rearing. ..................................................................................... 96 4 4 Effects of pemoline on body weight, gland mass, an d corticosterone concentrations .................................................................................................... 97 5 1 Pemoline induced self i njurious behavior ......................................................... 111 5 2 Plasma and brain pemoline concentrations across days of pemoline administration ................................................................................................... 113 5 3 Effects of pem oline or vehicle injections on plasma corticosterone .................. 114 5 4 Effects of pemoline or vehicle injections on striatal m onoamine and metabolite levels ............................................................................................... 116 5 5 Effects of pemoline or vehicle injections on striatal monoamine synthesis and turnover ratios ................................................................................................... 117 5 6 Effects of pemoline or vehicle injections on striatal amino aci d levels. ............. 118 5 7 Effects of pemoline or vehicle injections on ventral tegmental monoamine and metabolite levels ........................................................................................ 119 5 8 Effects of pemoline or vehicle injections on ventral tegmental monoamin e synthesis and turnover ratios ............................................................................ 120
11 5 9 Effects of pemoline or vehicle injections on ventral tegmental amino acid levels. ............................................................................................................... 121 5 10 Effects of pemoline or vehicle injections on cortical monoamine and metabolite levels ............................................................................................... 122 5 11 Effects of pemoline or vehicle injections on cortical monoamine synthesis and turnover ratios. ........................................................................................... 123 5 12 Effects of pemoline or vehicle injections on cortical amino acid levels. ............ 124 5 13 Effects of pemoline or vehicle injections on amygdalar monoamine and metabolite levels.. ............................................................................................. 125 5 14 Effects of pemoline or vehicle injections on amygdalar monoamine s ynthesis and turnover ratios. ........................................................................................... 126 5 15 Effects of pemoline or vehicle injections on hippocampal 5HT and 5 HIAA. ... 127 5 16 Effect s of pemoline or vehicle injections on hippocampal 5HT synthesis and turnover ratios. .................................................................................................. 128 5 17 Effects of pemoline or vehicle injections on hippocampal amino acid levels. ... 128 5 18 Effects of pemoline or vehicle injections on septal monoamine and metabolite levels. ............................................................................................................... 129 5 19 Effects of pemoline or vehicle injections on septal monoamine synthesis and turnover ratios. .................................................................................................. 130 6 1 Pemolineinduced self injury and neurotensin concentrations in discrete brain regions. ............................................................................................................. 147 6 2 Effects of the NTS1 antagonist, SR48692, on pemolineinduced self injurious behavior.. .......................................................................................................... 150 6 3 Effects of pemoline and SR48692 on striatal monoamines and metabolites.. .. 150 6 4 Effects of pemoline and SR48692 on striatal monoamine biosynthesis and turnover ratios.. ................................................................................................. 151 6 5 Effects of the NTS1 agonist, PD149163, on pemolineinduced self injurious behavior ............................................................................................................ 154 6 6 Effects of pemoline and PD149163 on striatal monoamines and metabolites.. 154 6 7 Effects of pemoline and PD149163 on striatal monoamine biosynthesis and turnover ratios.. ................................................................................................. 155
12 LIST OF ABBREVIATION S 3MT 3 methyoxytyramine 5 HIAA 5 hydroxyindoleacetic acid 5 HT serotonin 6 OHDA 6 hydr oxydopamine ACTH adrenocorticotropic hormone BS3 bis(sulfosuccinimidyl) suberate COMT catechol O methyl transferase CRH corticotrophin releasing hormone CSF cerebrospinal fluid DA dopamine DHPG 3 methoxy 4 hydroxyphenylglycol DOPAC 3,4 dihydroxyphenylacet ic acid EDTA ethylenedraminetetraacetic acid Fmr1 Fragile X mental retardation 1 FMRP Fragile X mental retardation protein GABA gamma aminobutyric acid GPe external segment of the globus pallidus GPi internal segment of the globus pallidus HPA axis hypothalamic pituitary adrenal axis HPLC high pressure liquid chromatography HPRT hypoxanthineguanine phosphoribosyltransferase HR high responder HVA homovanillic acid I intracellular LE Long Evans
13 LR low responder LSD least significant differences MAO monoamine oxidase MeCP2 methyl CpG binding protein MHPG 3 methoxy 4 hydroxyphenylglycol NE norepinephrine NT S 1 neurotensin receptor 1 OCD obsessive compulsive disorder OCT3 organic cation transporter 3 OPA o phthalaldehyde PVN paraventricular nucleus Rai1 retinoic acid induced 1 RM ANOVA repeated measures analysis of variance S surface SEM standard error of the mean SNc substantia nigra pars compacta SNr substantia nigra pars reticulata SNRI serotonin and norepinephrine reuptake inhibitor SSRI selective serotonin reuptake inhibitor VTA ventral tegmental area
14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy NEUROBIOLOGICAL BASIS OF SELF INJURIOUS BEHAVIOR: FOLLOWING CONVERGENT PATHWAYS TOWARD NOVEL PHARMACOTHERAPY By Amber M. Muehlmann Van Matre August 2011 Chair: Darragh P. Devine Major: Psych ology Self injury is a debilitating feature of neurodevelopmental disorders, includi ng autism spectrum disorders and genetic disorders like LeschNyhan, Prader Willi, Rett, and Fragile X syndromes Self injury is a devastating disorder that involves not only bodily injury (even as severe as broken bones and amputation of tongues, lips, and fingers) but also leads to exclusion of the self injurer from social activities and cognitive therapies and causes severe distress for family and caretakers. As such, reducing self injury in these populations would positively impact overall health, socialization, cognitive function, and relationships with family, friends, and caregivers. Furthermore, understanding the neurobiological basis of this behavior disorder will help to elucidate genetic variables that predispose vulnerability and geneenvironment interactions that promote the development of self injurious patterns of behavior. To elucidate neurobiological changes associated with the induction and expression of self injurious behavior, an analysis of several brain regions and circuits was conduct ed using a rat model of self injury. In this model, pemoline, an indirect monoamine agonist, is administered repeatedly across days, which causes the development of repetitive behavior and self biting in rats. Repeated stress exposure increased the severit y of pemolineinduced self injury, suggesting that the biochemical
15 stress response may cross sensitize with the stimulant properties of pemoline to potentiate self injury. Relatedly, endogenous stress responsiveness correlated with the vulnerability to dev elop pemolineinduced self injury. Pain thresholds of pemolinetreated rats were also evaluated and were found to be significantly greater than the pain thresholds of vehicletreated rats. The pemolineinduced elevation in pain thresholds did not correspon d to the development of pemolineinduced self injury, which indicates that altered sensory processing is not necessary or sufficient for pemolineinduced self injurious behavior, though painrelated brain pathways may be disregulated by pemoline treatment. An extensive analysis of the neurochemistry of several brain regions was undertaken and revealed significant depletion of intracellular dopamine in the striatum that corresponded temporally with the onset and maintenance of pemolineinduced self injury. P arallel increases in extracellular monoamine metabolites were also found, suggesting that pemoline administration caused elevations in synaptic monoamine content. Taken together, these projects have implicated the important role of dopamine depletion in th e striatum for the development of pemolineinduced self injury, the impact of pemoline administration on pain responsiveness, and the consequence of stressful experiences on pemolineinduced self injury. Neurotensin is a modulatory neuropeptide that causes release of stress hormones, as well as striatal dopamine, and mediates a form of morphineindependent analgesia. Considering the forgoing findings, neurotensin appeared to be a likely candidate to contribute to pemolineinduced brain chang es and self inju rious behavior. Examinations of striatal neurotensin content and the efficacy of pharmacological challenges on pemolineinduced self injury using neurotensinrelated drugs confirmed this hypothesis. Intracellular neurotensin levels were significantly higher in the striata of pemolinetreated rats. Additionally, a neurotensin agonist was
16 shown to increase the severity of pemolineinduced self injury, whereas an antagonist lessened all measures of pemolineinduced self injury.The convergence of findings sugge sts that neurotensin is an important modulator of self injurious behavior and may be an important target for pharmacotherapy.
17 CHAPTER 1 INTRODUCTION Self Injurious Behavior Many abnormal behaviors are exhibited by people with psychiatric, neurological, and developmental disorders, but none are as maladaptive and devastating as self injurious behavior. In the context of intellectual handicaps these behaviors are typically repetitive, cause mild to moderate injury, and are overtly expressed. These differ somewhat from the self harm behaviors enacted by individuals with psychiatric or emotional problems that are more likely to be performed using rituals (Favazza, 1998), in times of heavy emotional burden (Klonsky, 2007), or where the self injurious behaviors and injuries are usually hidden (Madge et al., 2008). Though the self injury in developmentally disabled populations and in individuals with normal intelligence differs in some ways, the occurrence of self injury during times of negative affect and arousal and its function to reduce these feelings (Klonsky, 2007; Kemp et al., 2008) are quite consistent across the two groups. Our work describes the neurobiological basis of self injury in an animal model that is presumptively a more closely related analogue to the self injury seen in individuals with developmental disorders, rather than to those exhibited by individuals with psychiatric and emotional disturbances (e.g. borderline personality disorder, schizophrenia, post traumatic stress disorder). As such, w e will mainly focus on the phenomenology and neurobiology of the self injurious behavior expressed by those with intellectual and developmental disabilities. It has recently been suggested that the relationship between genotype and phenotype may be stronger in intellectual and developmental disabilities compared to psychiatric disorders (Ernst et al., 2010). The
18 onset of psychiatric disorders and the consequent self injurious behaviors are heavily influenced by severe emotional and environmental events (e.g sexual, drug, and/or alcohol abuse) that may complicate the direct relationship between genotype and phenotype. Some researchers are beginning to postulate that understanding the genetic underpinnings of self injury in the intellectually disabled may lead to advances in the genetics of vulnerability for self injury and suicide in other populations (Ernst et al., 2010). So although our focus is predominately on the self injurious behaviors of those with intellectual and development disabilities, we are enc ouraged by the possibility that our work may also elucidate the neurobiological basis of all self injury. Clinical P henomenology of Self Injurious Behavior Self injurious behavior is described as any behavior directed toward the self that causes tissue d amage or has the potential to cause damage. This definition is quite general but has been widely used, though also scrutinized, over the last several decades (Schroeder et al., 1980). In general, estimates of the population prevalence of self injurious behavior range from 1.7% to 65.9% of the intellectually disabled. These estimates vary considerably because definitions of self injury are inconsistent across studies; some include mild self injury whereas others only report the incidence of moderate to severe self injury (for review see Rojahn and Esbensen, 2002). In a study of 596 self injurers receiving services for being intellectually disabled (IQ less than 70) 40% of the self injurious sample was profoundly intellectually disabled (IQ below 20), 49% were severely intellectually disabled (IQ between 2050), and 11% were mildly disabled (IQ between 5070). Furthermore, 51% lived in hospitals, 28% lived in small health service agencies or residential schools, an d 21% lived at home with their families. Over half of the cases showed more than one type of self injurious behavior
19 and skin picking/scratching, self biting, head punching/slapping/banging w ere seen in over 30% of all cases (Oliver et al., 1987). Other self injurious behaviors include hair pulling, eye gouging, pinching, rubbing and stuffing orifices (Schroeder et al., 1980). The skin directed behaviors are typically focused on the head (mostly face), hands (usually the back of the hands), and the legs (Symons and Thompson, 1997; Symons et al., 1999) Self injurious behaviors occur in both distinct, singular episodes or as contin ued bouts (Kroeker et al. 2004). It is the repetitiveness of the self injurious behavior (i.e. continued bouts that can last hours with only very short pauses; Thompson and C aruso, 2002) and the intensity of these behaviors that contribute to the devastating nature of the behavioral phenomenon. For example, when the biomechanical dynamics of head hitting of several individuals were analyzed, the impact of the self hitting blow s was found to be similar to the force generated by a boxers jab (Newell et al., 1999; Newell et al., 2002). Considering that these behaviors can occur in continued bouts every day for years, the devastating impact of these behaviors on the health and wel l being of the self injurer and their loved ones cannot be overstated. Emergence of self injurious behavior The emergence, maintenance, and assessment of self injurious behavior have been described most thoroughly from a behavioral perspective. Several r esearchers (Guess and Carr, 1991 ; Kennedy, 2002) suggest that the emergence of self injury stems from normal rhythmic behavior that is displayed by normally developing infants (e.g. body rocking and head banging ). These benign behaviors typically fade with the development of more functional motor skills (Ryan 1983) Because these behaviors are exhibited by normally developing infants, the authors describe these rhythmic behaviors as biologically based and internally regulated. In a later stage of development these
20 stereotyped rhythmic behaviors in children with intellectual disabilities may continue to be expressed, and function as an adaptive correction for an environment that is eit her under or over stimulating. Repetitive stereotyped behaviors at this l evel come under the control of operant conditioning by being paired with positive or negative reinforcers, and lead to the emergence of self injurious behavior. A recent experimental analysis of repetitive behavior, potentially injurious behaviors, and self injurious behaviors in children with intellectual disabilities found a close temporal relationship between these behaviors (Petty et al., 2009) which suggests a developmental relationship between them Maintenance of self injurious behavior Carr and Smith (1995) expanded upon these behavi oral theories by further outlining variables that impact the expression of self injurious behavior. Namely, these variables are 1) the consequences of the behavior (i.e. conditioned positive or negative reinforcers), 2) antecedent stimuli that predict whether or not the consequences of the self injurious behavior will follow, and 3) setting events, including biological setting events. These authors focused solely on physical ailments as the potential biological setting events. Taken together, according to this behavioral theory, self injury is likely to occur during a period of physical discomfort (the biological setting event; when the individual is experiencing an ear infection, fatigue, allergies, etc.), in the presence of a trigger stimulus (the antecedent; when presented with a task, for example), and when self injury has previously been associated with reinforcement (the consequences; for example, escape from the demand to do the task). This behavioral model of sel f injurious behavior is stimula ting and lays the foundation for the neurobiological
21 concomitants of these three variables to be mapped onto the expression of self injurious behavior. Although the emergence and maintenance of self injurious behavior have m ostly been described in behavioral terms, several authors have proposed theories wherein neurobiology can interact with social consequences to affect the expression of self injury. Oliver and colleagues (1993) argued that operant conditioning may be the pr ocess that maintains the self injury, while neurobiological or neurochemical factors may predispose the individual to develop this behavioral pathology. Furthermore, Thompson et al. (1995) propose that self injurious behavior is an endogenous means of neur ochemical self administration and have outlined several parallels between self injury and drug self administration These parallels include the efforts exhausted to exhibit the behavior and the neurochemicals involved in both drug taking and self injury (d opamine, opioids, and stress hormones ). The emergence and development of stereotypy and self injury in children have received much more attention than has the continued expression of these behaviors or their potential natural remission throughout developme nt. Very few studies have focused on prevalence rates of self injury in adults and there have been no examinations of potential changes in self injury frequency, intensity, location, or maintaining variables throughout the adult years. Cooper and colleagues (2009) found that almost a third of their adult self injurious sample had stopped self injuring after a two year follow up of their initial assessment. On the contrary, using a cross sectional design, Cohen and colleagues (2010) found the prevalence of s elf injury was constant in their sample of adults with intellectual handicaps in their twenties, thirties, forties,
22 fifties, and sixties, suggesting no significant remission in their sample of 2,345 adults. Another cross sectional study of nearly 700 peopl e with a diagnosis of autism found a significant negative correlation between prevalence of self injury and age, although this correlation may have been skewed by the high prevalence of self injury in kids 2 4 years old and the low incidence of self injury in adults 41+. Self injury prevalence rates of their autistic sample between 540 years of age appear relatively stable (Esbensen et al., 2009). Although there are very few studies of self injurious behavior across the lifespan, it appears that the preval ence is quite stable and that this behavioral pathology is not just a disorder of early childhood. Assessment of self injurious behavior The preponderance of evidence suggests that social consequences have a significant impact on the expression of self in jurious behavior (Iwata et al., 1994, Kahng et al., 2002). As such, assessments of self injury, its functions, and the benefit of behavioral treatments have been developed by applied behavior analysts who study the effects of social consequences on human behavior (Vollmer et al., 2009). For therapists trying to reduce self injury it is tremendously important to assess the function of the behavior, not just the incidence of the behavior. Several assessment methods have been developed, each of which assay the conditions in which the self injury is expressed and the subsequent events that follow the self injury. Indirect methods of assessment are sometimes used in which questionnaires or checklists are completed by parents, caregivers, or teachers and do not i nvolve any direct concurrent observations of the self injurer. Descriptive analyses of self injury involve many sessions of direct observations in several different, but naturally occurring environments to correlate the presence or absence of particular antecedent or
23 subsequent events with the expression of self injury. Functional analysis, as opposed to descriptive analyses and indirect assessments, is an experimental manipulation of the environment, which allows for systematic testing of different anteced ent events and reinforcers and can elucidate the functional relationships between the behavior and the social environment. During these assessments, self injurious behavior may be provoked by different testing situations, which helps to determine the antec edent and consequent events that maintain self injury in the natural environment. The testing conditions in a functional analysis of self injury include an attention condition, an escape condition, and an alone condition (Iwata et al., 2002). A large review of many functional analyses has provided a general characterization of the proportion of self injury that is maintained by different consequences. In a sample of 152 self injurers, Iwata and colleagues (1994) found that self injury was maintained b y social negative reinforcement in 38% of self injurers (measured in the escape condition), social positive reinforcement controlled self injury in 26% of their sample (assayed in the attention condition), while another 26% of the sample showed automatical ly reinforced self injury (revealed in the alone condition). Multiple reinforcers controlled the self injury of 5% of the sample, and the data from the remaining 5% of self injurers could not be interpreted. These data suggest that functional analyses prov ide interpretable results on the functions of self injury for a large proportion of individuals and that behavior can be clearly mapped on to particular social or automatic consequences when assessed in systematic, experimental conditions. Additionally, the information provided by the functional analyses aid in treatment selection and efficacy. The effectiveness of reinforcement based interventions is significantly improved when the treatment selection is based on the precise
24 information provided by a funct ional assessment (Kahng et al., 2002; Mace and Mauk, 1995). In summary, theories on the emergence and maintenance of self injurious behavior, as well as the assessment methodologies have been predominantly developed by behavioral researchers. They have set forth theories of self injury that are behavioral in nature, but are also compatible with the hypothesis that neurobiological processes may contribute to the emergence, development, and maintenance of self injurious behavior. Comorbid behaviors and relat ed behavioral pathologies Self injury is considered part of a class of behaviors termed destructive or challenging behaviors, which also includes aggression and property destruction. The expression of self injury, aggression, and destruction correlate well with each other in individuals with intellectual handicaps (Matson and Rivet, 2008), have been found to be maintained by similar social functions (Petty et al., 2009), and show close temporal association when both challenging behavior and self injury are part of the individuals behavioral repertoire (Petty et al., 2009). In fact, pharmacological trials typically demonstrate overlapping effects on both self injurious behavior and aggression, be it significant reductions of both behaviors or no effect on either behavior (Parikh et al., 2008). These data suggest that common genetic vulnerability, social functions, and neurobiology are involved in the expression of this class of challenging behaviors. This group of challenging behaviors has been propos ed to be part of a spectrum of abnormal repetitive behaviors, including stereotypies, compulsions, impulsions, and tics (Bodfish and Lewis, 2002). When compared to individuals who do not exhibit self injury, those that do exhibit self injury are more likel y to also express stereotypy
25 (Bodfish et al., 1995), compulsions (Bodfish et al., 1995), and severe tics (Mathews et al., 2004). Moreover, a compulsive behavior hypothesis of self injury has been proposed, based on the similarities between the self injurio us behaviors exhibited by individuals with intellectual handicaps and the compulsive behaviors seen in those with Obsessive Compulsive Disorder (OCD; King, 1993). This hypothesis is framed around the fact that a significant proportion of self injurers engage in self restraint (Powell et al., 1996) in an apparent attempt to keep from self injuring. Kings compulsive behavior hypothesis (1993) outlines parallels between the attempts of people with OCD to refrain from engaging in their compulsions to the self restraint behaviors of self injurers. Consistent with this hypothesis, Powell and colleagues (1996) found that self injurers who exhibit self restraint are more likely to display compulsions than are self injurers who do not attempt self restraint. Interes tingly, self restraint attempts did not prevent eventual self injurious behavior, which is consistent with the inevitable occurrence of compulsive behavior in people with OCD. Effects of selfinjurious behavior on self and others Self injurious behavior can cause significant physical harm to the self injurer. The most severe injuries include broken bones, amputation (e.g. tongue, lips, and fingers), and disfigurement (e.g. cauliflower ear). Less severe self injury can also be quite harmful causing bruisi ng, swelling, and secondary infection of the injury site. Self injury can also be functionally impairing, leading to exclusion from educational and socialization activities and thereby further limiting the opportunities of self injurers. Family members and caregivers are also significantly impacted by the self injurious nature of these behaviors (Tausig, 1985). These people report high rates of stress (Sarimski, 1997) and feelings of sadness and despair (Bromley and Emerson, 1995).
26 Dealing with self injury and the comorbid challenging behaviors is also a common factor for staff turnover in residential facilities (George and Baumeister, 1981). Furthermore, society pays a significant cost for the support and treatment of self injurers; at last estimate the fed eral burden was approximately $3 billion per year (National Institutes of Health Consensus Development Conference Statement, 1989). Clinical Syndromes Numerous syndromes associated with mental retardation and neurobiological dysfunction exhibit self injur y, but many others generally do not. For instance, self injury is not a common phenotypic trait of Down syndrome or Williams syndrome. This suggests that self injury is not due to just any aberration in brain development, but may be due to specific deficit s in normal brain function. The developmental syndromes that are most commonly associated with self injurious behavior are LeschNyhan, Rett, Prader Willi, Fragile X, Smith Magenis, and Cornelia de Lange syndromes as well as autism These disorders differ in terms of genetic causes and comorbidities, but share the self injurious phenotype and may provide clues to a common neural dysfunction. Lesch Nyhan syndrome The prevalence of LeschNyhan syndrome is approximately 1 in 380,000 live births (Crawhall et al., 1972) and is considered a rare disease by the Office of Rare Diseases at the National Institutes of Health. Lesch Nyhan syndrome is caused by a mut ation in the Hprt gene (located at Xq26), that causes little to no functioning of the hypoxanthineguani ne phosphoribosyltransferase (HPRT) enzyme in any tissue type in the body, including the forebrain nuclei that comprise the basal ganglia, where HPRT enzyme activity is normally the highest (Kelley, 1968). HPRT is responsible for purine salvage, and without proper functioning of this enzyme purine recycling is disregulated
27 and purines are instead converted into uric acid. Diagnostic criteria for LeschNyhan syndrome include virtually com plete HPRT deficiency, hyperuricemia ( high l evels of uric acid in the blood) in assoc iation with hyperuricosuria ( excessive amounts of uric acid in urine; Nyhan, 1968). Phenotypically, LeschNyhan syndrome is associated with severe motor problems (dystonia, choreoathetoid movements), intellectual disabilities, aggression, and self injury. Nearly all individuals with LeschNyhan syndrome bite some part of their body; the most commonly injured sit es are the tongue, lips, fingers and/or arms (Anderson and Ernst, 1994). This self injury may begin as early as the primary teeth grow in. In an attempt to prevent the s elf biting, many people with LeschNyhan are fitted with mouth guards (Cauwels and Martens, 2005) or the teeth are excised (Hall et al., 2001). Anderson and Ernst (1994) completed one of the most thorough analyses of sel f injury in LeschNyhan syndrome. They found individuals with LeschNyhan syndrome self injured in multiple different ways, with the average person exhibiting seven different forms of self injury. Self biting was the most common, followed by hitting their arm, leg or head on a doorway, snapping their head backwards, head banging, placing their feet under the wheels of their moving wheelchair, placing their fingers in the moving spokes of their wheelchair, or eye gouging. These alarming findings were replicate d by Robey et al. (2003). Most individuals are fitted for arm splints or other r estraints to decrease these self injurious behaviors (Ball et al. 1985; Letts and Hobson, 1975; Torres and Puig, 2007). Even t hough the individuals with LeschNyhan syndrome have a seemingly unending compulsion to self injure, it has been shown that they have normal pain
28 sensitivities and that they will actually beg and plead for restraints to be returned if they are removed (Nyhan, 2002). Rett syndrome Rett syndrome is an X linked dominant disorder that occurs in girls at a rate of between 1:10,000 (Hagberg, 1985) and 1:22,000 (Kozinetz et al., 1993). Rett syndrome is associated with a mutation in the methyl CpG binding protein (MeCP2) gene. The MeCP2 protein binds to methylated CpG nucleotides in the DNA and is involved in epigenetic regulation of gene expression. Rett syndrome is the most common cause of intellectual disability in girls. Rett syndrome involves normal development for the first 9 to 18 months of life, followed by severe regression including deceleration of head growth, social withdrawal, loss of any acquired speech, and motor and gait problems. There is also loss of appropriate hand usage and an onset of hand stereotypies characterized by midline, chest level clasping, wringing, and washing like hand motions. Self injury occurs in about half of the girls with Rett syndrome and hand biting and mouthing are the most common types of self injurious behaviors (Sansom et al., 1993), though the hand stereotypies can also be so extreme as to produce lesions. Prader Willi syndrome Prader Willi syndrome occurs in approximately 1:15,000 births and is caused by abnormal expression of genes around the 15q1113 chromosome region. In humans, the maternal copy of this chrom osomal region is hypermethylated and therefore inactive. Approximately 70% of Prader Willi cases are caused by a large deletion in the 15q11q13 region of the paternally derived chromosome. In another 20% of Prader Willi syndrome patients, two maternal chr omosomes are transmitted (uniparental disomy). The normal imprinting of the maternal alleles remain, leaving no transcription of genes
29 in this imprinted region. Other cases of Prader Willi syndrome are caused by microdeletion of the imprinting center near the 15q1113 region or the translocation of genetic material affecting the transcription of genes in the critical locus. Prader Willi syndrome is associated with muscle hypotonia during infancy and later in development hyperphagia, obesity, irritability, anxiety, anger, obsessivecompulsive symptoms, intellectual disability, and self injury are exhibited. Self injury is seen in approximately 80% of individuals with Prader Willi syndrome and it most commonly consists of skin picking that targets the legs, head, and arms (Symons et al., 1999). The severity of self injurious behavior is significantly different between individuals with the paternal deletion and the uniparental disomy. Those with the deletion subtype exhibit more severe self injury compared to those with uniparental disomy (Dykens et al., 1999; Hartley et al., 2005; Symons et al., 1999). The genetic cause of these differences has yet to be elucidated. Fragile X syndrome The prevalence of Fragile X syndrome is approximately 1:4,000 males and 1:9 ,000 females. Fragile X syndrome is caused by a genetic mutation on the X chromosome at Xq27.3. This region is associated with the Fragile X mental retardation 1 gene ( Fmr1 ). This mutation involves CGG trinucleotide repeats, the number of which increases t hrough maternal generational transmission. These CGG repeats are found in the 5 untranslated region of the first exon, which leads to hypermethylation of the promoter region and to little or no transcription of the FMR1 protein (FMRP; Feng et al., 1995). FMRP is a RNA binding protein and is important for gene translation repression. FMRP controls local protein synthesis that is initiated with cellular activity and without
30 normal functioning of FMRP protein the amount of gene translation is significantly in creased, leading to alterations in synaptic homeostasis. Fragile X syndrome is the most common cause of intellectual handicaps and the most common genetic cause of autism (described below). Fragile X boys typically show moderate intellectual disabilities, while girls vary between mild intellectual handicaps to normal intellectual functioning. The mutation in FMRP is associated with abnormal brain development and Fragile X individuals show eye gaze aversion, hyperactivity, social anxiety, aggression, and/or self injurious behavior (Reiss and Dant, 2003; Hall et al., 2008). Nearly half of all individuals with Fragile X syndrome exhibit self injury. These self injurers typically express more than one form of self injurious behavior, which include self biting, head hitting, skin picking, hair pulling, head banging, and skin scratching. Interestingly, the area of self injured surface is negatively correlated with FMRP protein levels (Symons et al., 2003a). Smith Magenis syndrome Smith Magenis syndrome is quite r are, affecting 1:25,000 people. This syndrome is caused by a sporadic deletion involving the retinoic acidinduced 1 (Rai 1) gene on the short arm of chromosome 17. Two types of genetic mutations have been characterized in individuals with Smith Magenis syndrome. Ninety percent of cases involve large deletions at 17p11.2 that involves many genes. The remaining 10% of individuals with Smith Magenis syndrome have small deletions of 17p11.2 that only includes the Rai1 gene. RAI1 protein is proposed to be part o f the transcription machinery (Elsea and Girirajan, 2008). The behavioral phenotypes associated with Smith Magenis syndrome include intellectual disability, hypotonia, attention deficit hyperactivity disorder, seizures,
31 attention seeking, stereotypy, and self injury. The stereotyped behaviors exhibited by individuals with Smith Magenis syndrome are not associated with other common genetic disorders and consist of self hugging and repetitive page turning (Dykens and Smith, 1998). Nearly all people with Smit h Magenis syndrome exhibit self injurious behaviors, which do resemble typical self injury topographies. These include self biting, head banging/slapping, skin picking, hair pulling, pulling finger and toenails, and inserting objects into orifices (Finucane et al., 2001). Cornelia de Lange syndrome Cornelia de Lange syndrome occurs in approximately 1:10,000 people and is associated with several types of mutations in any one of three genes that code for cohesionassociated proteins. Sixty percent of cases are caused by mutations in the Nipbl gene located on chromosome 5. Another 5% of cases are due to mutations in one of two genes, Smc1a (on the X chromosome) or Smc3 (on chromosome 10). The molecular bases of the other 35% of Cornelia de Lange cases have not been identified. The process of cohesion is responsible for the proper segregation of chromosomes during mitosis and meiosis and researchers are beginning to understand the critical role of these and other cohesionassociated proteins in the expression of many different genes (Liu and Krantz, 2009). Individuals with Cornelia de Lange syndrome exhibit intellectual disability (80% of individuals exhibit moderate to profound handicap), autistic like behaviors, compulsions, hyperactivity, attention deficits, depression, sleep disturbances, aggression, and self injurious behavior (Berney et al., 1999). Self injury is expressed by 5060% of individuals with Cornelia de Lange syndrome and the behaviors typically involve self biting, skin picking, banging self against objects, and body poking (Oliver et al., 2009).
32 Autism As opposed to many of the disorders listed above, autism is a syndrome that is defined solely by the expression of particular behaviors, not by molecular characterization of genetic mutations. The three behavioral domains that are diagnostic for autism are impairments in social interaction, communication deficits, and restricted repetitive behaviors. The most recent estimates indicate that 1 in every 110 children (and 1 in every 70 boys) has an autism diagnosis. Although autism is not diagnostically defined as being linked to one particular genetic cause, it is highly heritable with heritability estimates at nearly 90%. Five to fifteen percent of individuals with autism have a single gene disorder (e.g. Fragile X syndrome, Rett syndrome) and rare mutations and copy number variants have been found in small percentages of autistic cases. The genes most often found to be associated with autism cluster into categories of genes involved in cell growth, proliferation, and differentiation and preand post synaptic proteins that maintain synaptic connectivity (Pinto et al., 2010). This genetic heterogeneity may be responsible for the clinical heterogeneity (described below), although environmental insult s and gene environment interactions probably also contribute to the highly variable autism phenotype (Herbert, 2010). Variability is characteristic of all the features of autism. In regards to the three diagnostic behavioral domains, social interaction di fficulties can range from impairments in nonverbal behaviors (e.g. eye gaze, gestures) to failure to develop appropriate peer relationships. Communicative abilities vary greatly between people with autism. Many people with autism are completely nonverbal, whereas others exhibit incessant stereotyped and repetitive language. Level of intelligence is also variable among people with autism. Some exhibit above average intelligence whereas others are intellectually
33 disabled. Restricted repetitive behaviors and i nterests also vary greatly and range from insistence on sameness, repetitive language, circumscribed interests, object attachments, repetitive facial movements, tics, motor stereotypies, repetitive manipulations of objects, and self injurious behavior (Tur ner, 1999). Self injurious behavior is exhibited by approximately half of autistic individuals and about 15% express severe self injury (Baghdadli et al, 2003). The prevalence of autism is much higher in boys, however in regards to self injury, autistic fe males are more likely to express self injury and self injure more severely than boys with autism (Cohen et al., 2010; Esbensen et al., 2009). Neurobiology and Neurochemistry of the Clinical Syndromes Associated with Self I njurious Behavior As detailed above, there is no obvious similarity in the genetic basis between any of the syndromes that exhibit self injurious behavior. Of the syndromes where the genetic and molecular pathologies have been identified, the deficits range from dysfunctions of DNA replic ation, DNA methylation, mRNA transcription, protein translation, and purine salvage. This range of molecular pathologies is quite striking considering the consistency of the phenomenology of the self injurious behaviors across syndromes. With the understanding that the neurobiological basis of adaptive behavior is predicated on the normal functioning of entire brain circuits, it is probable that the differing molecular pathologies across these syndromes contribute to a similar disregulation of the output of the brains behavioral circuits and thusly share common behavioral dysfunction. The neural circuit most likely involved in the expression of self injurious behavior is the corticobasal ganglia loop because of its role in cognition, emotion, and movement
34 (Bodfish and Lewis, 2002; Cromwell and King, 2004). The basal ganglia has several different circuits that mediate motor behavior, oculomotor movements, cognition, personality, and motivation/emotions (Visser et al., 2000). Each of these circuits differs sl ightly by the area of cortex that innervates the particular subcortical regions of the basal ganglia input nuclei, the pallidal output nuclei, and the thalamic projections. Though in general the neuroanatomy of the corticobasal ganglia loop consists of co rtical inputs to the striatum (caudate, putamen, and nucleus accumbens), which project to the internal segment of the globus pallidus (GPi) and substantia nigra pars reticulata (SNr). These two nuclei, the GPi and SNr, are functionally equivalent and are g enerally grouped together when describing striatal output regions. The outputs from the striatum to the GPi/SNr take one of two paths. One path is direct from the striatum to the GPi/SNr and is termed the direct pathway. The other path is indirect and is termed the indirect pathway. The indirect pathway includes relay projections to the external segment of the globus pallidus (GPe) and on to the subthalamic nucleus. From the GPi/SNr, projections go to the thalamus and then to the cortex. The neurochemis try of the cortico basal ganglia loop consists mainly of glutamate, gamma aminobutyric acid (GABA), and dopamine. Descending cortical projections to the striatum are glutamatergic. The direct pathway neurons from the striatum to the GPi/SNr are GABAergic, as are the indirect pathway neurons from the striatum to the GPe. The indirect pathway neurons from the GPe release GABA onto the relay neurons of the subthalamic nucleus, and those neurons that project from the subthalamic nucleus to the GPi/SNr are gluta matergic. The output from the GPi/SNr to the different nuclei of the thalamus is GABAergic, whereas the neurons that project from the thalamus to the
35 cortex release glutamate. This neuroanatomical and neurochemical organization is responsible for differing effects of the direct and indirect pathways on cortical activation. Activation of the direct pathway causes disinhibition of the thalamic neurons and results in excitation of the cortex. Likewise, activation of the indirect pathway increases the inhibitio n of the thalamic neurons, which results in a reduction of glutamate signaling to the cortex, and therefore, decreased activation of the cortical neurons. The corticobasal ganglia loops can be modulated by dopamine signaling from neurons originating in th e substantia nigra pars compacta (SNc). Dopamine has different effects on direct and indirect pathway neurons, which is mediated by the different dopamine receptor subtypes. Direct pathway neurons in the striatum contain mostly D1type dopamine receptors ( D1 and D5 subtypes). Dopamine binding to these receptors activates stimulatory G proteins, which then stimulate adenylate cyclase and its second messenger systems. Indirect pathway neurons in the striatum contain mostly D2type dopamine receptors (D2, D3, D4 subtypes). Dopamine binding to these receptors activates inhibitory G proteins, which then suppresses adenylate cyclase and its second messenger systems. Furthermore, basal ganglia function can also be modulated by other neurotransmitters (e.g. serotoni n and acetylcholine) and neuropeptides (Angulo and McEwen, 1994) The direct and indirect pathways are parallel pathways. Proper expression of basal ganglia mediated behaviors depends on the appropriate balance of activity from these two antagonistic tract s. Neuroimaging techniques and postmortem brain analyses suggest that the development and/or functioning of the corticobasal ganglia pathways may be disregulated in individuals with neurodevelopmental disorders associated with
36 self injury and potentially mediate the vulnerability to develop self injurious behavior and its continued expression (Boddaert et al., 2004; Casanova et al., 1991; Eliez et al., 2001; Ernst et al., 1996; Gothelf et al., 2008; Haas et al., 2009; Harris et al., 1998; Hoeft et al., 08, Hollander et al., 2005; Jellinger et al., 1998; Lloyd et al., 1981; Menon et al., 2004; Rassin et al., 1982 ; Saito et al., 1999; Sears et al., 1999; Subramanium et al., 1997; Wenk et al., 1991; Wong et al., 1996; Wong et al., 1998). Unfortunately these st udies were not designed to look at the neurobiological differences between self injurers and noninjurers. As such, these neuroimaging and postmortem brain studies reveal abnormalities in corticobasal ganglia circuits in neurodevelopmental disorders where self injury is commonly expressed, but the mechanisms that underlie the actual vulnerability or the etiology of self injurious behavior has not be systemically examined. Since the prevalence of self injurious behavior is highest in LeschNyhan syndrome, n early 100%, analyses of this population provide the most potentially relevant information regarding the neurobiological basis of self injury that is currently available. Neuroimaging is difficult to perform in individuals with neurodevelopmental disorders, but several research groups have succeeded in performing this technique in self injurious populations. Wong et al. (1996) found that the volume of the caudate nucleus was smaller in five subjects with LeschNyhan syndrome. No volumetric differences were f ound in the put amen. Harris et al. (1998) replicated this finding in the caudate and were able to find reduced putamen and total brain volume using magnetic resonance imaging in individuals with LeschNyhan syndrome. Dopaminergic dysfunction has been the most prevalent finding in analyses of post mortem brain tissue, neuroimages (i.e positron emission tomography binding
37 studies), cerebrospinal fluid (CSF), and blood plasma in individuals with LeschNyhan syndrome. These deficits have been seen from the levels of the amino acid precursor to dopamine, tyrosine, all the way to the dopamine receptor expression level. Lower tyrosine levels have been found in almost all brain areas that have been assayed (e.g. putamen, several regions of cortex; Rassin et al. 19 82). Decreased levels of synthesizing enzymes, tyrosine hydroxylase and dopa decarboxylase, that are responsible for the conversion of tyrosine to dopamine have been found in the caudate and putamen of individuals with Lesch Nyhan syndrome (Lloyd et al., 1981). Ernst et al. (1996) also found lower enzymatic activity of dopa decarboxylase in the caudate, putamen, frontal cortex and ventral tegmental area (VTA) Not surprisingly, dopamine levels are lower in the caudate, putamen, external segment of the globus pallidus, and the nucleus accumbens in post mortem LeschNyhan syndrome brains (Lloyd et al., 1981). This finding in the caudate was later replicated by Saito et al. (1999). Levels of the dopamine metabolite homovanillic acid (HVA) have also been found t o be lower in Lesch Nyhan syndrome brain (Lloyd et al., 1981) and in CSF (Jankovic et al., 1988), compared to cont rols. These lower levels of dopamine metabolite are associated with increased activity of the dopamine catabolizing enzyme, monoamine oxidase (Lloyd et al., 1981). All of these differences were found in dopaminergic projection areas (i.e. regions with axon terminals) and not in dopaminergic cell body regions (Lloyd et al., 1981; Saito et al., 1999). Moreover a reduction in binding to the dopamine transporter was identified in the caudate and putamen (Wong et al., 1996). These findings suggest that there is a functional loss of dopaminergic axon terminals and not a loss of dopaminergic cell bodies in Lesch Nyhan syndrome. Conversely, increases in
38 dopamine receptor expression (D1 and D2) have been identified in the caudate and putamen in individuals with Lesch Nyhan syndrome, which suggests there is dopaminergic receptor supersensivity (Saito et al., 1999). The roles of the other monoamines in LeschNyhan syndrome have not been fully characterized. Lloyd et al. (1981) found elevated levels of serotonin and its metabolite, 5 hydroxyindoleacetic acid (5HIAA) in the putamen. Jankovic et al. (1988) also found increased 5HIAA levels in the CSF. No diffe rences in norepinephrine have been identified in LeschNyhan syndrome versus controls (Ernst et al., 2000). However, decreased norepinephrine metabolite 3methoxy 4 hydroxyphenylglycol ( M HPG) has been found in plasma (Ernst et al., 2000). Additionally, mas sive increases in epinephrine have been found in the plasma of individuals with Lesch Nyhan syndrome, which may indicate disregulation of the hypothalamic pituitary adrenal ( HPA ) axis or the adrenal glands response to HPA axis activation (Ernst et al., 2000). There are no reports of other markers of HPA axis functioning (i.e. adrenocorticotropic hormone or cortisol) in LeschNyhan syndrome. In Rett syndrome, reduced caudate volume has been shown consistently (Casanova et al., 1991; Subramanium et al., 1997), and significant reductions in gray and white matter in the cortical dopaminergic projection area, the prefrontal cortex, have also been found (Subramanium et al., 1997). Postmortem findings of dopaminergic dysfunction have been inconsistent (Roux and Vi llard, 2010). Tyrosine hydroxylase immunoreactivity is lower in the dopaminergic cell body regions in the substantia nigra (Jellinger et al., 1988), suggesting reduced monoamine synthesis. Wenk et al (1991) found lower dopamine concentrations in the putam en and cortex of girls with Rett
39 syndrome (Wenk, 1995). Furthermore, several other postmortem studies have found lower concentrations of dopamine, serotonin, norepinephrine, HVA, and 5HIAA in several brain regions (Brucke et al., 1987; Riederer et al., 1986). However, in a follow up study with additional subjects, Wenk and colleagues were unable to find significant differences in any of their measures of dopaminergic functioning, including dopamine and HVA concentrations (Wenk 1996), and Lekman et al. (1989) found no differences in dopamine, serotonin, norepinephrine or their metabolites in the caudate or putamen in four women with Rett syndrome. Obtaining consistent results regarding dopamine transporter levels and functioning has also been a challenge. We nk found decreased density of dopamine transporter sites in the caudate and putamen in one set of patients (Wenk, 1995), but was unable to replicate this finding in another group of patients (Wenk, 1996). Wong et al. (1998), however, found a significant reduction in dopamine transporter binding in the caudate and putamen. Amino acid receptor levels are another inconsistent result in Rett syndrome. Blue et al. (1999) reported significant reduction in the iontropic glutamatergic AMPA receptor levels in the putamen, and increased GABAA receptor in the caudate, while Wenk et al. (1993) found no differences in any ionotropic glutamate receptor subtype in the cortex or basal ganglia structures of Rett and control postmortem brains. Neuroimaging studies of individuals with Prader Willi syndrome are few. One study reported small orbitofrontal cortex volume (Ogura et al., 2010), while findings of no differences in caudate volume is consistent across a couple studies (Ogura et al., 2010; Yamada et al., 2006). Furtherm ore, the only monoamine result reported in Prader Willi patients indicates higher HVA and 5HIAA concentrations in the CSF of
40 Prader Willi patients compared to healthy controls (Akefeldt et al., 1998). Interestingly, the deleted region of the paternal 15q11q13 chromosome associated with Prader Willi syndrome contains genes for three subunits (GABRA5, GABRB3, GABRG3) of the GABAA receptor (Meguro et al., 1997). This suggests that GABAergic functioning may be reduced in the corticobasal ganglia pathways of P rader Willi syndrome patients. Significantly elevated GABA plasma levels have been found in individuals with Prader Willi syndrome patients, which may be a homeostatic response to receptor hypofunctioning (Ebert et al., 1997). Several neuroimaging studies have revealed larger caudate volumes in patients with Fragile X syndrome compared to controls (Eliez et al., 2001; Gothelf et al., 2008; Hoeft et al., 2008). Caudate volume is negatively correlated with FMRP levels and positively correlated with the presence of aberrant behaviors as assessed by the stereotypy subscale of the Aberrant Behavior Checklist (Gothelf et al., 2008) in those with Fragile X syndrome. Functional brain imaging analyses have also shown reduced activation of the basal ganglia during a response inhibition task in individuals with Fragile X syndrome (Hoeft et al., 2007; Menon et al., 2004). Activation of the basal ganglia during this task is positively associated with FMR1 gene expression (Menon et al., 2004). Additionally, diffusion tens or imaging has shown increased fiber density in the neuronal tracts from the ventral frontal cortex to the striatum in males with Fragile X syndrome (Haas et al., 2009). These neuroimaging findings again suggest disregulation of the cortico basal ganglia l oop in a population with high prevalence of self injurious behavior.
41 There is very limited information regarding the neuroanatomical and neurochemical basis of Smith Magenis and Cornelia de Lange syndromes. The only relevant study to date found less gray matter in regions of the putamen and globus pallidus, as well as lower blood flow to these regions in individuals with Smith Magenis syndrome (Boddaert et al., 2004). The anatomical and functional deficits in the Smith Magenis brain may mediate self injurious behavior in this population, though much more work will need to be done to confirm this role. Research and publication regarding Cornelia de Lange syndrome have not yet moved beyond description of phenotype and genotype, so the neurobiological basis of this disorder or the associated aberrant behaviors like self injury are completely unknown. The only physiological finding associated with self injury in autism was made by Kolevzon and colleagues (2010). They found that whole blood serotonin levels were inversely correlated with the expression of self injurious behavior. It is unclear what this peripheral effect has on brain pathways or if it is a more general marker of altered development of serotonergic systems, both central and peripheral (Whitaker Az mitia, 2005). Additionally, increased caudate volume is a consistent finding in autism compared to normal controls, and caudate size correlates positively with repetitive behavior (Hollander et al., 2005; Sears et al., 1999). This is a result of continued caudate enlargement throughout development in the autistic brain whereas caudate size begins to decline in the normal controls beginning in adolescence (Langen et al., 2009). Autism is an incredibly heterogenous disorder and obtaining consistent results ac ross subgroups (high functioning/low functioning) and development is a challenge. In
42 respect to elucidating the neurobiological basis of self injurious behavior, autism may not be the ideal population in which to study. In summary, a lthough a single parti cular neurobiological dysfunction has not been elucidated in the different disorders that exhibit self injury, there is some evidence to suggest that overactivity of the corticobasal ganglia loops is a common pathology. Fewer dopamine terminals and striat al postsynaptic receptor supersensitivity has been suggested in LeschNyhan and Rett syndromes. Decreased GABAA receptor functioning in Prader Willi syndrome may cause disregulation of direct versus indirect pathway activation and lead to over excitation o f the corticobasal ganglia loops. Conversely, enhanced cortical glutamatergic input to the basal ganglia may be important for the expression of self injury in Fragile X syndrome. Furthermore, inconsistent results in postmortem studies of each of these syndromes are common and may be due to age differences between subjects, and the methods used to collect, dissect, or process the brain tissue. Therapeutic Strategies for Self Injurious Behavior The therapeutic strategies for the reduction of self injurious behavior have historically involved either behavioral therapies or pharmacological interventions. Assessment of combined treatment plans has rarely been documented (Durand, 1982; Luiselli, 1986) and a large scale clinical trial of any combined treatment for self injury has never been reported. The understanding that psychotropic medication can alter an individuals sensitivity to environmental events is beginning to receive notice (Kennedy et al., 2001). This suggests that pharmacotherapy for self injury ca n not only change the disregulated neurochemistry but may also be able to change the impact of reinforcers within the environment that contribute to the expression of self injurious behavior. This
43 is an interesting idea that will hopefully receive signific ant attention by both basic and behavioral researchers. Pharmacotherapies The pharmacological targets of the drug therapies that have been evaluated in self injurers are consistent with neurochemical pathways that are disregulated in self injurious popul ations. Targets of these drugs include dopaminergic, serotonergic, glutamatergic, and GABAergic systems, including their synthesis enzymes, neuronal transporters, and receptors. There are several limitations to the reports of pharmacotherapeutic efficacy of these drugs. Very few have been evaluated in clinical trials; most include very small sample sizes, many report no placebo controls, the patients are often taking several other psychotropic medications, and data are provided by researchers who are not bl ind to the drug status of the patients and often include only retrospective analyses of the patients charts. Across studies the results are often inconsistent and the positive response rate is variable. Despite this, these studies still contribute to our understanding of the neurobiological basis of self injury and these drugs have the potential to reduce self injury is some sufferers. The class of atypical antipsychotics has shown the greatest efficacy in reducing destructive behaviors, including self in jury. Olanzapine is a common atypical antipsychotic with a 5HT2/D2 receptor antagonism profile and has shown reasonable efficacy in reducing self injurious behavior in individuals with intellectual disabilities (Janowsky et al., 2003a; McDonough et al., 2000) and pervasive developmental disorders (Potenza et al., 1999). Aripiprizole is a newer atypical antipsychotic and exerts 5HT2A antagonism and D2 partial agonism. Aripiprizole has been evaluated in a large double blind, randomized, placebo controlled study with people with autism and
44 provided significant improvement of several behaviors including self injury, as measured by the irritability subscale of the Aberrant Behavior Checklist (Owen et al., 2009). Furthermore, risperidone (a 5HT2A/D2 antagonist) reduces irritability (including aggression and self injury) in children with autism (McCracken et al., 2002) and has shown efficacious effects on self injurious behavior in individuals with LeschNyhan syndrome (Allen and Rice, 1996) and intellectual disabil ities (Cohen et al., 1998). A number of selective serotonin reuptake inhibitors (SSRIs) have been evaluated for their efficacy at reducing self injurious behavior. Fluoxetine (Ricketts et al., 1993), paroxetine (Snead et al., 1994), sertraline (Hellings et al., 1996; Luiselli et al., 2001), and the serotonin and norepinephrine reuptake inhibitor (SNRI), venlafaxine (Carminati et al., 2005), have shown some efficacy at reducing self injury in small studies. Interestingly, these same drugs (King et al., 199 1; Denys et al., 2003) have all been reported to increase self injurious behavior in OCD patients who did not self injure before the SSRI/SNRI treatment. A sufficient explanation of these paradoxical effects has yet to be uncovered but they suggest a very important role of serotonin modulation on basal ganglia function (Kelland and Chiodo, 1996; Di Matteo et al., 2008) and its consequences on the expression of self injurious behavior. Antiepileptic drugs have also been used to reduce self injurious behavior with some success. In several small studies, valproate reduced aggression and self injury in patients with intellectual disabilities (Kastner et al., 1993; Ruedrich et al., 1999). Valproate is generally regarded as a GABA agonist (increasing GABA levels by inhibiting metabolism and increasing synthesis), but also dampens neuronal excitability through actions on ion channels, alters several intracellular signaling pathways, and
45 regulates gene expression through epigenetic modifications (Terbach and William s, 2009). Topiramate also reduces self injury in individuals with intellectual handicaps (Janowsky et al., 2003b) and those with Prader Willi syndrome (Shapira et al., 2002). Like valproate the pharmacology of topiramate is broad and includes inhibition of sodium and Ltype calcium channels, as well as increases in GABAA receptor activity and antagonism of glutamatergic AMPA receptors (Shank and Maryanoff, 2008). The complicated pharmacology of these antiepileptic drugs does not elucidate the particular neurobiological basis of self injurious behavior but does support the notion that increased basal ganglia output to the cortex plays a significant role in the production of self injury. Naltrexone, an opioid receptor antagonist, has been used to treat self i njury for several decades. Its utility has been disputed, but a recent review of studies evaluating the effectiveness of naltrexone on self injurious behavior revealed a significant number of reported cases (nearly 80%) were responders to naltrexone treatm ent (Symons et al., 2004). The efficacy of naltrexone corresponds to the theory that individuals engage in these maladaptive behaviors for the release of endogenous opioids that follows the self injurious episode and that recurrent self injurious behavior is a way in which the self injurers can get a "fix" for their own addicted endogenous opioid systems (Sandman, 1990). According to this opioid theory, administration of naltrexone blocks the effects of the self injury mediated opioid release and the frequency of self injury then fades. Opioids and dopamine interact in many ways within basal ganglia structures, including modulation of presynaptic neurotransmitter release, neuronal activity, and by way of their postsynaptic receptors (Churchill and Kalivas, 1996). It is unclear which of these
46 mechanisms is important for the reduction of self injury, but a role for endogenous opioidergic systems in self injurious behavior is indicated. Behavioral therapies Studies of behavioral therapies are unfortunately plagued by some of the same limitations as their pharmacotherapeutic counterparts. These studies use very small samples, typically do not include information regarding treatment generalization, and rarely describe the continued efficacy of the therapy upon follow up (DeLeon et al., 2002). Despite the limitations of these studies, behavior therapies offer significant reduction in behavior in a large majority of cases, at least while they remain under the care of the behavioral therapist. These behavior therapies range from communication training (Carr and Durand, 1985), extinction procedures (Zarcone et al., 1993), differential reinforcement strategies (Vollmer and Iwata, 1992), and noncontingent reinforcement treatments (Fischer et al., 1997). Animal Models of Self Injurious Behavior Animal models of self injurious behavior have provided important information regarding the neurobiological basis of self injury. These models allow researchers to decipher which neurobiological dysfunctions found in the human popul ations are central to the expression of self injury and which are not. These animal models utilize a wide array of techniques to induce self injury including environmental deprivation, neonatal lesions, and pharmacological manipulations. A major strength o f these animal models is that phenomenologically they are quite similar to the human behavioral pathology; they involve repetitive behaviors, typically self biting, that can cause injury. Although in most of the animal models of self injury the behavior i s induced by drugs, self injury is commonly seen in animals living in labs, farms, zoos, and human
47 households. In this respect, these animal models represent species typical behaviors, which make them ideal to model human pathology (Insel, 2007). Furthermore, it is interesting that many of the mouse models that have been developed to model the genetic neurodevelopmental syndromes do not exhibit self injury. For example, the mouse models of LeschNyhan syndrome ( Hprt knockout; Kasim and Jinnah, 2002), Fragil e X syndrome ( Fmrp1 knockout; Price et al., 2007), and Prader Willi syndrome (imprinting center deletion; Relkovic et al., 2010) do not exhibit spontaneous self injurious behavior. Consequently, the mouse models that typically receive the most attention fo r construct validity in regards to the human disorders do not adequately model the behavioral pathology. In light of this, other animal models must be studied in order to further our understanding of the neurobiological basis of self injurious behavior. N onhuman Primate M odels Th ere are two nonhuman primate models of self injury. One involves social isolation rearing for the first several months of life; the other allows for social rearing until weaning, followed by single housing in laboratory facilities for use in biomedical research. In both models only a subset of the rhesus monkeys demonstrates self injurious behavior. Several aberrant behaviors are associated with the isolation rearing model, including stereotypy, pacing, circling behavior, aggression, apathy, indifference, withdrawal, fear, and self direct behaviors, including huddling and self injury (Goosen, 1981). Adult rhesus monkeys raised in isolation have significant loss of dopaminergic axon terminals and dopamine depletion (Martin et al., 1991). Dopamine receptor supersensitivity has also been suggested by their exaggerated behavioral response to the D1 and D2 receptor agonist, apomorphine (Lewis et al., 1990). No differences in CSF measures of serotonin metabolite (5HIAA), dopamine metabolit e (HVA), or
48 norepinephrine metabolite (MHPG) have been found between self injurious and noninjurious rhesus monkeys (Lewis et al., 1990). A subset of the rhesus monkeys raised socially, but single housed as adults also exhibit self injurious and aggress ive behaviors, though these monkeys typically do not demonstrate a lot of the other aberrant behaviors seen in the isolation reared monkeys. Again, these self injurious rhesus monkeys do not differ from noninjurious monkeys on measures of 5HIAA, HVA, or M HPG. There is some evidence that the self injury in this model is mediated by a hyposerotonergic state. Administration of the serotonin precursor, Ltryptophan, reduces self injury (Weld et al., 1998) as does treatment with fluoxetine (a SSRI) and a 5HT1a receptor agonist, buspirone (Fontenot et al., 2005). In addition, a specific polymorphism in the tyrptophan hydroxylase 2 gene, which codes for an enzymatic protein involved in serotonin synthesis, was recently found to correlate with a history of self injurious behavior (Chen et al., 2010). The roles of stress and anxiety have received a lot of attention in this rhesus monkey model. The prevalence of self injury in these singlehoused monkeys is related to the occurrence of several stressors including the age in which they were isolation housed and the number of minor veterinary procedures theyve undergone (Lutz et al., 2003). Furthermore, self injurious rhesus monkeys have increased stress induced hormone release and less negative feedback of the bodys stress response system, the HPA axis (Tiefenbacher et al., 2004). In addition, stressors, such as footshock (Gluck et al., 1985) and housing relocation (Davenport et al., 2008), significantly exacerbate the incidence of self injurious behavior in monkeys. R hesus monkeys with a history of self injury also vocalize more and exhibit a greater number of threat behaviors (Novak et al.,
49 2002). There is also evidence that the self injury seen in a subset of self injurious monkeys is anxiety related. Self injuriou s behavior is increased upon administration of an anxiogenic drug, FG7142 (Major et al., 2009), and reduced by an anxiolytic drug, diazepam (Tiefenbacher et al., 2005) in a subset of self injurious monkeys. Both of these drugs bind to the benzodiazepine si te on the GABAA receptor. Taken together, the neurobiological findings from the nonhuman primate models of self injurious behavior further implicate dopamine and serotonin depletion as well as a potential role for disregulated GABAergic signaling. The se nsitized stress pathways in self injurious rhesus monkeys might also potentiate these disregulated systems to increase the incidence of self injurious behavior in stressful situations. 6 Hydroxydopamine Model As described previously, significant dopamine depletion is found in post mortem striatal tissue of individuals with Lesch Nyhan syndrome (Lloyd et al., 1981). To model this early dopamine depletion, the dopaminergic systems of neonatal rats are lesioned with the neurotoxin, 6hydroxydopamine (6OHDA). As compared to rats that are lesioned as adults, the adult rats lesioned as neonates display self injurious behavior when given a dopamine agonist like Ldopa (Breese et al., 1984). Consistent with most of the syndromes associated with self injurious behav ior, both dopamine depletion and activation of sensitized postsynaptic receptors are required for the induction of self injury. Experiments using specific dopamine receptor agonists reveal that the D1 receptors mediate the expression of self injury (Criswe ll et al., 1992; Sivam, 1989). Vulnerability to develop dopamine agonist induced self injury in the lesioned rats seems to be due to the magnitude of the dopaminergic lesion and the adaptive changes in D1 receptor binding in the midbrain. Self injurious 6OHDA treated rats have more
50 substantial depletion of striatal dopaminergic neurons and increased D1 receptor binding in the SNr compared to noninjurious 6 OHDA treated rats (Yokoyama and Okamura, 1997). Interestingly, there are no differences in D1 recept or mRNA levels (Duncan et al., 1993), protein levels (Sivam et al., 2008), or binding rates (Yokoyama and Okamura, 1997) in the striatum of self injurious versus noninjurious rats. However, significant differences in phosphorylation status of intracellular signaling molecules (p38MAPK) and transcription factors (CREB) have been found, suggesting that the self injurious behavior is mediated by changes in gene transcription and cellular functioning within the striatum (Sivam et al., 2008), which are induced through activation of the D1 receptors. Several characteristics of the 6OHDA model closely resemble characteristics of Lesch Nyhan syndrome. Ldopa induces self injury in 6 OHDA rats and exacerbates self injury in individuals with LeschNyhan syndrome (Ja nkovic et al., 1988). Also, levels of serotonin are increased in the striatum of 6 OHDA rats ( Sivam et al., 2008; Towl e et al., 1989), similar to Lesch Nyhan syndrome brains (Lloyd et al., 1981). Another similarity between LeschNyhan syndrome and the 6OH DA lesioned rat is that the self injurious behavior seen in both is potentiated by stress (Anderson and Ernst, 1994; Stodgell et al., 1998). Furthermore, risperidone reduces self injury in the 6OHDA model (Allen et al., 1998) and in LeschNyhan syndrome ( Allen and Rice, 1996). Another interesting finding is that dopamine levels in the lesioned striatum are increased with a procedure (fixed ratio discrimination) that is similar to the behavioral interventions used for human self injurious behavior (Stodgell et al., 1996; Loupe et al., 2002). This
51 suggests that behavioral interventions can rescue dopaminergic deficiencies, which may be the reason for their therapeutic efficacy in self injury Bay K 8644 Model Bay K 8644 is an Ltype calcium channel activator. L type calcium channels are predominately located in the striatum, cortex and hippocampus (Jinnah et al., 2003; Tanaka et al., 1995) and are voltage gated, indicating that only after the cell membrane is depolarized do the L type channels open to allow a transient influx of calcium. Bay K 8644 increases the calcium influx during neuronal membrane depolarization. Mice given Bay K 8644 show impaired locomotion, dystonia, spasticity, self injurious behavior, and aggression (Jinnah et al., 1999). These Bay K 8644 induced behaviors are relatively short lived. They present within 10 minutes of Bay K 8644 subcutaneous administration, peak at 2030 minutes and are absent by 50120 minutes after injection (Jinnah et al., 1999). The behavioral effects (Jinnah et al., 1999) and c fos expression (Jinnah et al., 2003) induced by Bay K 8644 are prevented by preadministration of the Ltype calcium channel antagonist nifedipine which suggests that all effects are due specifically to activation of the Ltype calcium channel. The roles of serotonin and dopamine on Bay K 8644induced self injury have been studied. Generally speaking, drugs that increase synaptic levels of serotonin and dopamine increase self injury in the Bay K 8644treated mice and drugs that deplete serotonergic or dopaminergic systems decrease self injury (Kasim et al., 2002; Kasim and Jinnah, 2003). To determine whether the effect of Bay K 8644 was related to presynaptic or postsynaptic mechanisms, Kasim et al. (2006) used microdialysis to measure the effects of Bay K 8644 on dopamine release in the striatum. They found no increases in extracellular dopamine concentrations in the striatum of Bay K 8644treated mice. Likewise, they found that antagonists for the dopamine D1/5 and
52 D3 receptors attenuated the self injurious behavior in Bay K 8644treated mice. This suggests that the self injury inducing effects of Bay K 8644 are due to postsynaptic calcium channels that are associated with D1/5 and D3 receptors, and not due to presynaptic effects of Bay K 8644. Calcium influx through Ltype calcium channels, which is enhanced by Bay K 8644 administration, activates calcium/calmodulin kinase and MAPK pathways and stimulates CREB mediated gene expression (Rajadhyaksha and Kosofsky, 2005). These pathways were found to mediate self injurious behavior in the 6OHDA model and may also cause the self biting seen in Bay K 8644treated mice. In addition, D1 receptor activation enhances Ltype calcium channel functioning (Surmeier et al., 1995). This suggests that supersensitive D1 receptors in human self injurers may lead to a potentiation of Ltype calcium channel function and enhanced intracellular signaling and CREB mediated gene expression. In summary, t his model reconfirms that dopamine, and perhaps serotonin, are inv olved in the expression of self injurious behavior and that there may be some therapeutic promise for the use of nifedipine in reduci ng self injury in humans (Blake et al., 2007). Because of the short lived behavioral effects of Bay K 8644, however, studies of chronic L type calcium channel activation and its effects on neurotransmitter content and function have not been completed. Methamphetamine M odel Methamphetamine, an indirect monoamine agonist, which reverses the monoamine transporters and causes sig nificant release of monoamines into the synapse, while depleting vesicular stores (Sora et al., 2009), dose dependently induces self injurious behavior in mice (Kita et al., 2000; Halladay et al., 2003). Self injury in this model is acute, mostly occurring only after the first injection of methamphetamine, not
53 after any subsequent injection (Kita et al., 2000; Kuroda et al., 2010). The induction of self injury in the methamphetamine model corresponds to significant reduction in dihydroxyphenylacetic acid ( D OPAC ) levels and increases in HVA levels in the striatum (Kita et al., 2000). Methamphetamineinduced self injury in mice (particularly the BALB/c strain) is lessened by coadministration of a D1 antagonist, a glutamatergic NMDA receptor antagonist, and a serotonin precursor (Shishido et al., 2000). These authors also found that a D2 antagonist and naloxone did not reduce self injurious behavior in this model. These results suggest that self injury in this model is mediated by increased synaptic dopamine and glutamate, and decreased serotonin. More specifically, D1 receptor functioning, as opposed to D2 receptor functioning, appears to play a larger role in mediating methamphetamineinduced self injury. Interestingly, the efficacy of the serotonin precursor is consistent with the reduction in self injurious behavior in the nonhuman primate model (Weld et al., 1998), and is in accord with the findings that SSRIs are effective pharmacotherapies for some self injurers (Hellings et al., 1996; Luiselli et al., 200 1; Ricketts et al., 1993; Snead et al., 1994). Furthermore, consistent with the dopaminergic terminal loss exhibited by individuals with LeschNyhan syndrome (Wong et al., 1996) methamphetamine causes significant reduction in dopaminergic terminals, which may be caused by a combination of dopaminemediated (Kita et al., 2000) and glutamatemediated (Mori et al., 2007) oxidative stress related neuronal damage. Information from the methamphetamine model of self injury supports the significant role of dopamine terminal damage in the expression of self injurious behavior, as well as critical mediation by D1 receptors and glutamatergic and serotonergic functioning.
54 Muscimol M odel Bilateral injections of the GABAA agonist, muscimol, into the SNr induce stereotypy and self injurious behavior in intact rats (Baumeister and Frye, 1984) and in rats that have neonatal 6OHDA lesions (Breese et al., 1987). These effects support our hypothesis that increased excitation of the cortex from the corticobasal ganglia loops i s involved in the expression of self injury. Enhancing the inhibition of the SNr, through GABAA agonism, may potentially decrease the inhibition of the thalamic relays, and could lead to increased excitation of the cortex. This mechanism, though consistent with the neurobiology and neurochemistry of the corticobasal ganglia loop, is only hypothetical because the effect of muscimol on self injury expression is not reduced by a GABAA receptor blocker and is actually potentiated by a GABAA antagonist (Baumeis ter and Frye, 1984). This suggests that nonGABAArelated effects of muscimol, or the GABAA antagonists, may be responsible for the expression of self injurious behavior in this model. Pemoline M odel Pemoline is an indirect monoamine agonist (Cromwell et al., 1996; Fuller et al., 1978; Gilbert et al., 1978) that induces self injurious biting in rats when administered in one high dose (Muller and Hsaio, 1980) or with repeated lower doses (Kies and Devine, 2004; Mueller et al., 1986). Although the pharmacol ogical targets of pemoline are the monoamine transporters, pemolineinduced self injury is not associated with changes in dopamine, serotonin, or norepinephrine transporter levels (Mueller et al., 1986). Consistent with data from human self injurers, dopam ine, serotonin, glutamate, and GABAergic signaling are involved in the expression of pemolineinduced self injurious behavior, as evaluated by pharmacological challenges. Haloperidol (a D2 antagonist),
55 risperidone (a 5HT2/D2 antagonist), and valproate (an antiepileptic) each reduce pemolineinduced self injury (Muehlmann et al., 2008; Mueller and Nyhan, 1982) and human self injury. Pemolineinduced self injury is also reduced by the glutamate antagonist MK 801 (King et al., 1995; Muehlmann and Devine, 2008) Interestingly, the SSRI paroxetine significantly potentiates self injury in the pemoline model, suggesting an important role for synaptic serotonin in the expression of self injury. Consistent with other animal models of self injury there are differences in vulnerability to develop pemolineinduced self injury wherein some rats exhibit self biting, and others do not. Cromwell and colleagues (1997) showed that the rats that did develop pemolineinduced self injury had differential changes in postsynaptic r esponsivity to dopaminemodulated glutamatergic signaling. Self injurious rats showed increased striatal depolarizing postsynaptic potentials induced by dopamine application, whereas pemolinetreated non injurious rats and control rats had decreased depolarizing postsynaptic potentials. These differential postsynaptic responses were normalized with the application of D1 and D2 antagonists, as well as by MK 801. Taken together, the pemoline model appears to demonstrate predictive validity and the individual differences in vulnerability that are seen in the human syndromes that exhibit self injurious behavior. The individual differences in behavioral responses to pemoline also seem to be based on specific cellular neuroplastic changes that are mediated by neur otransmitter systems within the corticobasal ganglia pathways that are altered in human self injurers. Summary Each of these animal models has contributed significantly to our understanding of the neurobiological basis of self injurious behavior. Informat ion from each model has
56 supported the hypothesis that corticobasal ganglia pathways are dysfunctional in self injurious animals, and most have indicated a role of over activation of the direct basal ganglia pathway that is modulated by D1 dopamine receptors. The pursuit of elucidating the particular neurobiological disregulation responsible for the expression of self injury can be undertaken using any of these models. We have chosen to use the pemoline model in our investigations. Previously in the lab a dose response analysis of the self injury inducing effects of pemoline was completed (Kies and Devine, 2004), and we have demonstrated its predictive validity (Muehlmann et al., 2008) and generalizability (Blake et al., 2007). The projects described herein continue our characterization of the model including the roles of pain processing, stress responsiveness, and stress exposure on the expression of pemolineinduced self injurious behavior. We have also elucidated the neurochemical changes associated with the onset and maintenance of self injury in several key brain structures. Lastly we will describe our investigation into the role of a neuropeptide, neurotensin, in pemolineinduced self injurious behavior and the efficacy of neurotensinergic drugs on self biting behavior. These projects have expanded our understanding of the pemoline model of self injury and have elucidated a novel pharmacotherapy for the treatment of self injury in humans.
57 CHAPTER 2 CHANGES IN PAIN RESP ONSIVENESS DURING PE MOLINE ADM INISTRATION Background Behavioral theories of self injurious behavior contend that self injury is maintained by positive or negative reinforcement (Favell et al., 1982). This suggests that these reinforcers outweigh the aversive aspects of self injuring, including any pain involved with the injurious actions and the resulting tissue damage. Pain is difficult to assess in people with self injurious behavior as the prevalence of self injury is inversely associated with level of intellectual ability in indivi duals with neurodevelopmental disorders (i.e. self injury is most common in individuals with severe and profound intellectual disability; Saloviita, 2000). Research on a large cohort of individuals with intellectual disabilities found that those with more severe intellectual handicap were most likely to exhibit pain insensitivity or pain indifference, according to their caregiver (Biersdorff, 1994), though measuring pain thresholds in those who are cognitively impaired is problematic (Biersdorff, 1991). Un fortunately, research from individuals of normal intelligence has not clarified the relationship between the perception of pain and self injurious behavior. Paradoxically, self injury is exhibited by individuals with average intelligence both in the absenc e of pain perception and in the presence of chronic pain. Individuals, who have suffered severe spinal cord injuries causing loss of sensory input and motor control of their limbs, sometimes self injure to the point of amputation. The self injurious behavi or in these individuals with normal intelligence consists of self biting and typically initiates long after the spinal cord injury (Frost et al., 2008). Conversely, self injurious behavior
58 has been reported in people of average intelligence with chronic pain where the self injury was directed toward the painful site (Mailis, 1996). Studies of pain responsiveness in individuals with neurodevelopmental disorders associated with self injurious behavior have also not clarified our understanding of these issues. In general, girls with Rett syndrome have reduced pain reactivity (Downs et al., 2010). Rett syndrome patients also have lower levels of substance P in CSF compared to normal controls (Matsuishi et al., 1997) Substance P is a neuropeptide involved in the transmission of pain information from the periphery and decreased CSF substance P is associated with centrally mediated analgesia (Z ubrzycka and Janecka, 2002). Furthermore, individuals with Prader Willi syndrome consistently show lower pain sensitivity in response to thermal and painful stimuli (Priano et al., 2009), compared to normal controls. Severe self injury is also seen in people with congenital insensitivity to pain. A description of their self injurious actions has not been published and this genetic syndrome is associated with intellectual disabilities, which may also contribute to the expression of self injury (Zafeiriou et al., 2004). Pain reactivity measures in kids with autism have been reported as heightened (Nader et al., 2004) and diminis hed (Tordjman et al., 2009) in response to the same stimulus (venipuncture). These inconsistent reports may be due to inadequate sample sizes, subjective measurement of facial reactivity, or to the heterogeneity of nearly all autism phenotypes. In a study of children with severe intellectual handicaps, Breau and colleagues (2003) found that the prevalence of self injurious behavior was equal in children with and without chronic pain. They also found no differences in the expression of pain in self injurer s and noninjurers. These findings suggest that the expression of self -
59 injurious behavior is not specifically associated with painful experience, though many of the clinical syndromes associated with self injury have reduced pain reactivity, as measured by subjective testing. Furthermore, in an effort to better characterize potential aberrant pain processing in individuals with intellectual disabilities, Symons and associates have moved from only studying subjective pain reactivity in these potentially nonverbal populations to adding analyses of peripheral markers of pain fibers and inflammatory responses. In comparison to normal controls, Symons et al. (2008) found further spacing between epidermal nerve fibers and higher substance P levels in the uninjure d forearm of self injurers. However, when later compared to cognitively impaired noninjurers, self injurers had more epidermal nerve fibers and comparable substance P fiber density in a noninjured site (Symons et al., 2009). Furthermore, self injurers were more responsive to sensory stimulation than the noninjurers were and sensory reactivity was positively associated with epidermal nerve fiber density in all subjects. Self injurers, however, had higher levels of mast cell degranulation, compared to IQ m atched noninjurers, though mast cell density was equal between the two groups. Mast cell degranulation is initiated by substance P release from sensory fibers and causes release of secretory granules that contain high levels of proteases, lysosomal enzymes, and cytokines, which mediate immune and inflammatory responses (Pejler et al., 2010). These results suggest that among intellectually disabled individuals, self injurers are more reactive to sensory stimulation than their noninjurious counterparts, whi ch may be a result of higher levels of epidermal nerve fibers and mast cell degranulation.
60 In summary, most studies of pain in intellectually disabled populations suggest decreased sensitivity relative to normal controls; however peripheral markers of sensory systems indicate self injurers are more similar to controls than noninjurers are. This suggests that pain processing and reactivity may not play a significant role in the expression of self injurious behavior and further supports our hypothesis that central mechanisms, such as disregulated corticobasal ganglia function, are responsible for the development and maintenance of self injury. To further investigate the relationship between pain responsiveness and self injurious behavior, we evaluated pain thresholds using the hot plate test in pemolinetreated rats during the noninjurious and self injurious phases of repeated pemoline administration. Methods This project consisted of two experiments. The first experiment was designed to test Long Evans (LE ) rats on the hot plate apparatus at escalating temperatures to evaluate which temperature was most appropriate for repeated testing. The rats used in this experiment did not receive pemoline or vehicle injections. After choosing an appropriate temperature and test duration, we initiated the second experiment wherein we analyzed pain thresholds in pemolineand vehicletreated rats throughout the five days of injections. Experiment 1: Escalating Thermal Pain Testing in Drug Nave Rats Animals Twenty four m ale LE rats weighing 200225g upon delivery (Charles River Laboratories, Raleigh, NC) were housed in a climate controlled vivarium with a 12h/12h light/dark schedule (lights on at 7:00a.m.). Standard laboratory rat chow (Lab Diet 5001) and tap water were available ad libitum The rats were pair housed in standard
61 polycarbonate cages (43 cm x 21.5 cm x 25.5 cm) during six days of acclimation to the housing facility All of the experimental procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals, and were pre approved by the Institutional Animal Care and Use Committee at the University of Florida. Hot plate testing The hot plate apparatus (10 x 10 x 12in) was enclosed by clear Plexiglas sides and a lid. After six days of acclimation to the housing facilities, rats were habituated to the hot plate apparatus (Hot Plate Analgesia Meter, Columbus Instruments) each day for four days. Beginning at approximately 4:00 p.m., each rat was individually placed on the hot plate at 36C for 10 min. Three days after the habituation period ended, the hot plate testing began in which the rats were tested once every third day at one of the following ascending temperatures: 36C for 10 min, 44C for 10 min, 47C for 3 min, 49C for 3 min, and 51C for 1 min. All hot plate sessions were videotaped at two camera angles and scored for licking and guarding behavior. Licking was defined as any oral contact on the hindpaws. Guarding was defined as either a withdrawal of the hindleg with a protrac ted return or quickly stamping the hindleg up and down. Experiment 2: Thermal Pain Testing in Pemoline and Vehicle Treated R ats Animals Thirty six male LE rats weighing 200225g upon delivery (Charles River Laboratories, Raleigh, NC) were housed in the s ame conditions as the rats in Experiment 1. Hot plate testing habituation After six days of acclimation to the housing facilities, rats were habituated to the hot plate apparatus each day for four days (as in Experiment 1). Beginning at
62 approximately 4:00 p.m., each rat was individually placed on the hot plate at 36C for 10 min. Drug Pemoline (2amino 5 phenyl 1,3 oxazol 4 one; Spectrum Chemicals, New Brunswick, New Jersey) was suspended at a concentration of 50 mg/ml in peanut oil. In order to get the pemoline into suspension the solution was kept stirring at room temperature overnight. Drug treatment Beginning the morning after the last hot plate habituation session, the rats were weighed and injected subcutaneously with either pemoline at 150 mg/kg (n=18) or peanut oil vehicle (n=18) each day for five days. These injections were administered at the nape of the neck and either flank on a rotating basis. Assa ys of self injury, stereotypies, and locomotion The rats were visually inspected each time they wer e injected (approximately 9:00 a.m.) and then again every evening (approximately 4:00 p.m.) and the inspections were videotaped. During these inspections, each rat was held in front of a video camera and the head, forepaws, hindpaws, ventrum and tail were displayed. The presence or absence of tissue injury (denuded skin, erythema, edema, or open lesion) was noted for each rat. Any rat with an open lesion was immediately euthanized. N ight vision cameras were focused on the cages of the rats (one camera per cage), and 5minute time samples were recorded once every three hours over the entire day. The duration of self injurious oral contact rearing, locomotion, and stereotypy were quantified during each videotaped interval by a trained observer. Self injurio us oral contact was defined as all oral contact that stayed fixed on any one body part for longer
63 than 2 seconds. This was differentiated from grooming, which is oral contact with any part of the body that continues to move from site to site on the body (e.g. oral contact with the forepaws, that then moves up each forelimb and continues to the ventrum, and the contact is not sustained at any spot on the body for longer than 2 seconds). Rearing was counted each time a rat lifted both forepaws off the cage fl oor for at least 3 seconds. A subsequent rear was only counted after the rat resumed and maintained contact with the cage floor for at least 1 sec. Locomotion was counted by sectioning the video image of the cage into thirds (i.e. drawing lines on the television monitor dividing the cage into three equal parts along its length) and tallying the number of times the rat s forepaws crossed these lines. However, a subsequent crossing of that line was only counted after the rat crossed the other line. Thus, eac h locomotor count represents a minimum of 14.3 cm traversing of the cage. Since individual rats differed in the specific forms of stereotypy they expressed, and some stereotyped behaviors were expressed only at very low levels, t he stereotypy measures were summed. Scored stereotypic behaviors included duration of stereotyped bob/lick, wherein the rat either bobbed its head in a stereotyped manner or licked the side or floor of the cage, and the duration of stereotyped digging, sniffing, or burrowing through the bedding. The durations (in seconds) of rearing, self injurious oral contact, and stereotypy were summed over the entire day (i.e. from the eight video samples) and divided by the total number of seconds recorded, 2400 seconds. For example, the duration measures on day 1 cover the percent of time the rat was exhibiting that behavior during the recording periods between the first pemoline injection and just prior to the second injection, thereby covering the entire 24 hours following the first pemoline i njection.
64 Hot plate testing Each afternoon (approximately 4:00 p.m.) during pemoline or vehicle administration, each rat was tested on the hot plate apparatus at 47C for 3 min. The test sessions were recorded and licking and guarding behaviors were scor ed in the same manner as in Experiment 1. Statistical analyses Differences between the pemoline group and the vehicle group in latency and frequency of licking/guarding, and duration of oral contact, rearing, locomotion, and stereotypies were each evaluate d using repeated measures analyses of variance (RM ANOVA). The frequency of licking and guarding was combined together as is standard in hot plate testing (Mogil et al., 1996). Betweengroups differences in all these dependent measures were treated as stat istically reliable when the p values were less than 0.05. S ignificant effects for the lick/guard latency and frequency measures were further analyzed with preplanned Fishers least significant difference (LSD) post tests. Three pemolinetreated rats were euthanized before the end of the experiment because they had open lesions. In these cases, the missing data were replaced by repeating the final score that was attained for each dependent measure through the end of the experiment. This strategy was used to avoid the potential that the group means would over or under estimate the licking and guarding latency and frequency when the most severe self injurers were terminated. Results Experiment 1 In drug nave rats, lick/guard latency decreased with increasing temperatures (Fig. 2 1A). Frequency of lick/guards was also related to temperature (Fig. 21B),
65 though the duration of testing was different for most temperatures making direct com parisons difficult After scoring these data we chose 47C for our hot plate testing temperature for repeated evaluations. This intermediate temperature was selected because it allowed for potential pain threshold changes in either direction without the complication of either ceiling or floor effects. We were also confident that repeated hot plate testing at this temperature would not cause tissue injury to the rats paws. Experiment 2 Pemolineinduced self injury was first exhibited on day 3, and by day 6 all rats had injuries (Fig. 22A). Self injurious oral contact was initially seen on day 2 and remained high throughout days 3, 4, and 5 (time effect: F(4,136) = 28.54, p < 0.0001; drug effect: F(1,136) = 86.08, p < 0.0001; time x drug interaction effect: F(4,136) = 28.53, p < 0.0001Fig. 22B). Vehicle treated rats showed very low levels of rearing, locomotion, and stereotypies and exhibited no self injurious oral contact or self injured tissue. In the pemolinetreated rats, rates of rearing and locomotion were highest on day 1 (Fig. 23A,B), followed by peak in the rate of ster eotypies on day 2 (Fig. 23C). RM ANOVA effects for drug, time, and drug x time interactions were significant for all measures of ancillary behaviors, in comparison to vehicletreated rats (rearing: time effect: F(4,136) = 36.28, p < 0.0001; drug effect: F(1,136) = 42.60, p < 0.0001; time x drug interaction effect: F(4,136) = 27.79, p < 0.0001; locomotion time effect: F(4,136) = 28.17, p < 0.0001; drug effect: F(1,136) = 60.51, p < 0.0001; time x drug interaction effect: F(4,136) = 22.42, p < 0.0001; ster eotypies time effect: F(4,136) = 8.061, p < 0.0001; drug effect: F(1,136) = 162.9, p < 0.0001; time x drug interaction effect: F(4,136) = 8.453, p < 0.0001). Repeated pemoline administration at 150mg/kg/day diminished the expression of painrelated behav ior on a 47C hot plate. Lick/guard latency at 47C was significantly
66 increased with pemoline treatment (Fig. 24A). The RM ANOVA revealed a significant effect of drug on lick/guard latency (F(1,136) = 6.424, p < 0.05), but no significant effect of time or time x drug interaction. Consistent were the effects of pemoline found on the frequency measure of licking/guarding behavior (Fig. 24B). The RM ANOVA showed a significant effect of drug (F(1,136) = 18.87, p < 0.001), without significant effects of time or time x drug interaction effects. In fact, these nonsignificant effects that were suggested by the RM ANOVAs that failed to reach statistical significance for time or time x drug interactions reveal that pain thresholds at 47C changed relatively little during the five days of daily testing in either group. To examine the potential relationship between elevated pain thresholds and pemolineinduced self injurious behavior, we analyzed the mean lick/guard latencies and frequencies on the third experimental day when exactly half of the pemolinetreated rats exhibited self injury and the other half did not (Fig. 22A). Two tailed t tests found no significant differences in lick/guard latency (t(16) = 0.827, p = 0.4204; Fig. 24C) or frequency (t(16) = 0.4024, p = 0.6927; Fig. 24D) between self injurers and noninjurers. These data suggest that there is no direct relationship between pain sensitivity and the induction of self injury in the pemoline model. 36 44 47 49 51 0 100 200 300 400 500 600 0 30 60 90 120 150 180 lick/guard latency (sec) 10 min testlick/guary latency (sec) 1 or 3 min test 36 44 47 49 51 0 5 10 15 20 25 30 35 40 45 10 min test 3 min test 1 min test # of lick/guardsA B Figure 2 1. Hot plate test behaviors of drug nave rat s during stimulation at escalating temperatures. Latency to lick/guard in response to the temperature of the hot
67 0 1 2 3 4 5 6 0 25 50 75 100pemoline vehicle daypercentage of rats exhibiting tissue injury 0 1 2 3 4 5 0 20 40 60 80 100dayoral contact(% of total time recorded) plate decreased as temperatures were increased (A). Frequency of lick/guard behaviors increased as temperatures were increased (B). All values are expressed as group means standard error of the mean (SEM). Figure 2 2. Pemoline induced self injurious behavior. Incidence (A) and duration (B) of pemolineinduced self injurious behavior increased across days. rearing 0 1 2 3 4 5 0 10 20 30 40 50pemoline vehicle day% of total time recorded stereotypies 0 1 2 3 4 5 0 10 20 30 40 50day% of total time recordedA C B locomotion 0 1 2 3 4 5 0 20 40 60 80 100daynumber of line crossings Figure 2 3. Rates of homecage behaviors of pemolineand vehicletreated rats. Pemolinetreated rats exhibited more rearing (A), locomotion (B), and
68 stereotypies (C), compared to vehicletreated rats. All values are expressed as group means SEM. 0 1 2 3 4 5 0 30 60 90 120 150 180* * *daylick/guard latency (sec) 0 1 2 3 4 5 0 5 10 15 20* * *day# of lick/guards injurers non-injurers 0 30 60 90 120 150 180Day 3lick/guard latency (sec) injurers non-injurers 0 5 10 15 20 25Day 3# of lick/guardsA D C B Figure 2 4. Effects of pemoline and vehicle treatment on hot plate behaviors. Lick/guard latency (A) and frequency (B) stayed relatively stable across days of pemoline and vehicle administration. Pemolinetreated rats had higher pain thresholds and exhibited fewer painrelated behaviors compared to vehicletreated rats. On day 3 of pemoline administration there were no differences in lick/guard latency (C) or frequency (D) between pemolinetreated injurers or noninjurers. All values are expressed as group means SEM. Significant differences between pemoline and vehicle treated rats are depicted with an asterisk. Discussion In this experiment we evaluated the effects of pemoline on pain thresholds and painrelated behaviors using the hot plate test. Although all pemoline treated rats exhibi ted self injury during the later days of the experiment, when we analyzed pain responsiveness on a day that half of the rats exhibited injury and half did not, we saw no differences in lick/guard latency and frequency. Furthermore, pain responsiveness did
69 not change across days, whereas self injurious behaviors did. On average, the pemolinetreated rats had longer latencies and lower numbers of lick/guard behaviors, which suggests that pemolinetreated rats have reduced pain sensitivity. This is redolent of findings from some (but not all) clinical syndromes associated with self injury. These data suggest that reduced pain sensitivity may be a requisite condition for the vulnerability to develop self injury, but that other neurobiological conditions are nec essary for the induction of self injury. Reduced pain sensitivity has been found in other animal models of self injury. In the muscimol model where injections into the SNr cause some rats to self injure, the self injurious rats failed to show any pain resp onsiveness on a 55C hot plate. Rats that did not self injure after injection of muscimol in the SNr did react to the painful stimulus of the hot plate. Furthermore, baclofen injected into the SNr also decreased pain sensitivity but did not induce self injury (Frye et al., 1986). This suggests that decreased pain sensitivity is necessary for the induction of muscimol induced self injury, but that other neurobiological changes (induced by muscimol, but not by baclofen) are also required. Additionally, amphet amine induced self injury is associated with increased pain thresholds, though like pemoline treated rats, there were no differences in lick/guard latency in amphetaminetreated self injurers and noninjurers (Jain and Sharma, 2002). These findings indicat e that decreased pain sensitivity may be a contributing factor to a vulnerability to develop pemolineinduced self injury, but that additional neurobiological changes may also be required for the induction of self injury.
70 CHAPTER 3 EFFECTS OF REPEATED STRESS ON PEMOLINE INDUCED SELF INJURIOUS BEHAVIOR Background Stress and novelty appear to play important roles in mediating the expression of self injurious behavior ; however these roles have not received much empirical study. A survey of parents of males with Lesch Nyhan syndrome reported that self injury was exacerbated by the presence of new people or the stress of an illness (Anderson and Ernst, 1994) A dditionally, a r ecent study of nonhuman primates found that the stress of relocation to a novel environment with unfamiliar neighbor animals increased the self injury of rhesus monkeys with a prior history of social deprivation and self wounding, but had no self injury in ducing effects on monkeys without a history of self wounding (Davenport et al., 2008) Furthermore, assays of stress hormones, ACTH and cortisol, reveal that are t hey are disregulated in c linical populations in which self injury is a phenotypic trait (Curin et al., 2003; Hall et al., 2008; Hessl et al., 2002; Sandman et al., 2008; Symons et al., 2003 b ; Verhoeven et al., 1999) however the direction of the disregulation (i.e. findings of increases or decreases) has been inconsistent. These findings from human and nonhuman primate self injurers suggest that the neurobiological response to stress may induce a self injurious episo de in those with a disr egulated HPA axis and a vulnerability to self injure. Using the pemoline model of self injury we examined the effect of repeated social/emotional stress on the expression and severity of pemolineinduced self injurious behavior R ep eated social defeat stress alters basal levels of corticosterone (Covington and Miczek, 2001) thereby inducing similar endoc rine dysfunction that has been reported in some self injurers. In light of this, we investigated the effect of
71 repeated social defeat stress on the ex pression of pemolineinduced self injury We also analyzed the levels of AMPA receptors, neurotensin receptors, and the organic cation transporter 3 (OCT3) to evaluate potential neurotransmitter systems that may mediate an exacerbation of pemolineinduced self injury by repeated social defeat stress. Methods Animals Twenty six male LE rats weighing 150175g up on delivery (Charles River Laboratories, Raleigh, NC) were housed in a climate controlled vivarium with a 12h/12h light/dark schedule (lights on at 6:00 a.m. ). Twenty two of t hese were the experimental rats and were used as the intruders (n=11) and handl ed controls ( n=11; see Social Defeat section). Four additional rats were used as vehicle controls. They were neither used as intruders nor handled controls. Standard laboratory rat chow (Lab Diet 5001) and tap water were available ad libitum The se rats we re pair housed in standard polycarbonate cages (43 cm x 21.5 cm x 25.5 cm) during 6 days of acclimation to the housing facility and throughout either the social defeat or handling regimens The vehicle control rats were left pair housed and undisturbed unt il the injection protocol began. An additional six male LE rats weighing 300325g and an additional six female LE rats weighing 200225g were pair housed in standard polycarbonate cages with rats of their same sex for seven days. These male rats were later used as residents (see Social Defeat section). Male and female pairs were housed together (after the males were vasectomized) in a separate housing room from the intruders with opposite reverse 12h/12h light/dark schedule (lights off at 6:00 a.m.). All the experimental procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals, and were preapproved by the Institutional Animal Care and Use
72 Committee at the University of Florida. All efforts were made to minimize ani mal suffering and to reduce the number of animals used. Surgery The male resident rats were vasectomized under ketaminexylazine anesthesia (ketamine, 62.5 mg/kg; xylazine, 12.5 mg/kg; i.p.) and Ketorolac analgesia (2 mg/kg, s.c.) Supplemental surgical anesthesia (isoflurane) was administered as necessary. Each anesthetized rat was shaved from the rostral edge of the scrotal area to the caudal abdomen A midline incision was made rostral to the penis, the vas deferens was isolated with forceps, and a 0.5 cm section was removed from each duct using a miniature cautery utensil. The internal incision was sutured with absorbable 40 Ethilon monofilament vicryl suture (Ethicon Inc.) and the external incision was closed with 9 mm stainless steel wound clips (Wor ld Precision Instruments Inc.) which were removed seven days after surgery. Social D efeat P rocedure At least two weeks prior to the onset of the social defeat sessions the vasectomized male rats were pair housed with cycling female rats. Social defeat sessions were run during the resident rats dark schedule starting at approximately 7:00 a.m. At the beginning of each social defeat session each female resident was removed from her home cage and placed in a similar cage nearby. Ten minutes after removal of the female resident an intruder rat was placed into the home cage that the male and female resident rats shared. Each male resident was trained to exhibit dominance behavior. The resident male and intruder rat were allowed to interact for five minutes or until the intruder displayed a submissive, defeated posture three times. An intruder rat was considered to be defeated when it lay motionless in a supine posture, with the
73 resident male rat on top of it, for a period of at least two seconds. Following this direct interaction phase, each intruder was removed from the home cage of the resident, placed in to a separate 10 cm x 15 cm x 10 cm doublelayered wire mesh cage and returned to the home cage of the resident male rat. This indirect interaction phase allow ed the intruder rat to be out of direct contact of the male resident rat but still able to experience visual, auditory, and olfactory interactions with the resident male. The intruder was maintained in the wire mesh cage until 10 min had elapsed from the start of the direct interac tion phase. After the entire 10 min interaction had concluded both the female resident rat and the intruder were returned to their respective home cages. The intruder rats were put through the social defeat procedure each day for twelve days, seeing each resident male twice in that period, six days apart. Nond efeated control rats were handled for 2 min each day for 12 days. Drugs Pemoline (2amino 5 phenyl 1,3 oxazol 4 one; Spectrum Chemicals, New Brunswick, New Jersey) was susp ended at a concentration of 50 mg/ml in peanut oil. In order to get the pemoline into suspension, the solution was kept stirring overnight. Ketamine/ xylazine, isoflurane, and ketorolac tromethamine were purchased from Henry Schein Inc. (Melville, New York) Drug T reatment Following the 12 day social defeat regimen or the equivalent days for the handling control rats the rats were weighed and injected with pemoline at 150 mg/kg (s.c.) each day for five days. These injections were administered at the nape of the neck and either flank on a rotating basis. For protein expression controls, four rats were administered peanut oil vehicle for five days.
74 Assays of S elf I njury and S tereotypy The rats were visually inspected each time they were injected and then again every evening, and the inspections were videotaped. During these inspections, each rat was held in front of a video camera and the head, forepaws, hindpaws, ventrum and tail were displayed. The presence or absence of tissue injury (denuded skin, erythema, edema, or open lesion) was noted for each rat. Any rat with an open lesion was immediately euthanized. Still images of the injured tissue were taken from the videotapes, and MCID software (Imaging Research Inc., St. Catherines, ON, Canada) was used to dr aw outlines around the injured tissue, and to calculate the area of injury in mm2. Additionally, night vision cameras were focused on the cages of the stress and handled rats (one camera per cage), and 5minute time samples were recorded once every three hours over each entire day. The duration of self injurious oral contact and stereotypy were quantified during each videotaped interval by a trained observer. Self injurious oral contact was defined as all oral contact that stayed fixed on any one body part for longer than 2 seconds. This was differentiated from grooming, which is oral contact with any part of the body that continues to move from site to site on the body (e.g. oral contact with the forepaws, that then moves up each forelimb and continues to t he ventrum, and the contact is not sustained at any spot on the body for longer than 2 seconds). The stereotypy measure is a compilation of the duration of stereotyped bob/lick, wherein the rat either bobbed its head in a stereotyped manner or licked the side or floor of the cage, and the duration of stereotyped digging, sniffing, or burrowing through the bedding. The duration (in seconds) of self injurious oral contact and stereotypy were each summed over each day (i.e. from the eight video samples) and
75 di vided by the total number of seconds recorded, 2400 seconds. For example, the duration measures on day 1 cover the percent of time the rat was exhibiting that behavior during the recording periods between the first pemoline injection and just prior to the second injection, thereby covering the entire 24 hours following the first pemoline injection. A second blind observer scored the duration of self injurious oral contact and stereotyped behavior for a sample of the recordings (one day of recordings for eac h rat) and inter observer reliability was evaluated by Pearson correlation. Twenty four hours after the last pemoline or vehicle injection, the rats were rapidly decapitated. Brains were immediately removed, frozen in 2methylbutane kept on dry ice at 40 C, and stored at 80C. Protein A nalyses The frozen brains were cut into 1mm slices in the coronal plane using a stainless steel rat brain matrix (Braintree Scientific, MA). Micropunches were taken from the core and shell of the nucleus accumbens and stor ed at 80C. The methods for protein crosslinking and western blot analyses were adapted from Nelson et al. (2009). Punches were incubated with protein crosslinking reagent bis(sulfosuccinimidyl) suberate (BS3) at 2uM in artificial extracellular fluid for 30 min at 4C with gentle agitation. Chemical crosslinking of proteins in intact cells allows for analysis of surface expressed proteins (which form high molecular weight aggregates when crosslinked, as opposed to the intracellular proteins that are not ac cessible to the cross linking reagent). Crosslinking was terminated by quenching the reactions with 500ul of 300uM glycine for 10 min at 4C. Tubes containing the cross linking reactions were centrifuged at 20,800g for 2 min at 4C. Supernatant was discarded and the protein pellet was sonicated in 25ul of ice cold lysis buffer containing protease and phosphatase inhibitors.
76 The tubes were again centrifuged at 20,800g for 2 min at 4C. Supernatant was separated and the total protein content was assayed using the Lowry method (Total Protein Kit, Sigma). Equal amounts of protein samples were loaded on to 38% Tris Acetate gels and run under reduced conditions. Proteins were then transferred onto polyvinylidene fluoride membranes. Membranes were incubated with primary antibody at 4C. The following primary antibodies were used: polyclonal rabbit anti GluR1 (1:500; Millipore), polyclonal rabbit anti GluR2 (1:1000; Millipore), polyclonal rabbit anti NTS1 (1:500; Abcam), polyclonal rabbit anti OCT3 (1:1000; Abcam) Membranes were exposed to secondary antibody (horseradish peroxidaselinked anti rabbit IgG, 1:5000; Cell Signaling) for 1 hr at room temperature. Proteins were detected using the chemiluminescence method and analyzed by densitometry using MCID Image Analysis software. Stripped membranes were reprobed for bactin (rabbit polyclonal, 1:2000; Cell Signaling) and secondary antibodies. Protein expression levels for GluR1, GluR2, NTS2, and OCT3 are normalized to bactin protein levels. Statistical A nalyses Differences between the stressed group a nd the unstressed group in duration of self injurious oral contact, size of the injured tissue, and duration of stereotypy were each evaluated using RM ANOVA. Proteins were analyzed according to their localization: sur face, intracellular, surface + intracellular (S+I), and surface/intracellular. Differences between the vehicle group, the pemoline + stress group, and the pemoline + no stress group in these measures were each analyzed by oneway ANOVAs. Betweengroups dif ferences in all dependent measures were treated as statistically
77 reliable when the pvalues were less than 0.05. All significant effects were further analy zed with preplanned Fishers LSD post tests. Six rats were euthanized before the end of the experime nt (four stressed rats and two control rats) because they had open lesions. In these cases, the missing behavioral and tissue injury data were replaced by repeating the final score that was attained for these dependent measures through the end of the exper iment. This strategy was used to avoid the potential that the group means would underestimate the area of tissue damage and self injurious oral contact scores, and to avoid the potential that the group means would over or under estimate the stereotypy scores when the most severe self injurers were terminated. Brains were collected from the rats that were euthanized early and their tissues are included in the protein expression analyses. Results Repeated social defeat stress did not affect the incidence of pemoline induced self injury (Fig. 1A). The dose of pemoline used (150mg/kg/day) caused all the nonstressed rats and all the stressed rats to show some sign of self injury. The severity of the injur ies however, was significantly increased by repeated social defeat stress. Specifically, the rats that were repeatedly stressed showed significantly larger areas of tissue damage than the handled control rats did (Fig. 1B). The RM ANOVA revealed a significant main effect of time (F(10,210) = 39.91, p < 0.01) and a significant stress x time interaction (F(10,210) = 2.095, p < 0.05), wherein rats first displayed injured tissue around day 2 or 3, the size of injured tissue began to asymptote on day 4 or 5, and the size of injured tissue was greater in the stressed rats. There were no significant differences in the amount of time spent engaging in self injurious oral contact between stressed and unstressed rats (Fig. 1C). There was a significant main effect of time (F(4,84) = 46.47, p <
78 0.01) as all rats began to sh ow self injurious oral contact on day 2, which peaked on day 3, and continued throughout the experiment, however there was no significant main effect of stress nor a stress x time interaction effect. Although the statistical test was not significant, the s tressed rats did appear to engage in more oral contact on days 4 and 5 when the rates of oral contact of the nonstressed rats begin to decline. There were also no differences in the duration of other pemolineinduced stereotypies (Fig. 1D). The RM ANOVA r evealed a significant main effect of time (F(4,84) = 14.35, p < 0.01), as all rats showed moderate amounts of stereotypy on days 1 and 2, which lessened on days 3 5 as self injurious oral contact duration increased. No significant main effect of stress or a stress x time interaction effect was found for stereotypy duration. Inter observer reliability, as determined by Pearson correlation, was r = 0.9647 for the duration of oral contact and r = 0.9306 for the duration of stereotypy. The significant effect o f stress on the severity of pemolineinduced self injurious behavior was not a result of altered neurotransmitter receptor or transporter expression or localization in the nucleus accumbens. We analyzed levels of two glutamatergic AMPA receptor subunits, G luR1 (Fig. 3 2A,B) and GluR2 (Fig. 32C,D), a neurotensin receptor (NTS1; Fig. 32D,E), and the OCT3 (Fig. 32F,G). We found that repeated social defeat stress, which exacerbated the pemolineinduced self injury, did not change expression or localization o f GluR1, GluR2, NTS1, or OCT3 in the nucleus accumbens. Total expression (i.e. surface + intracellular levels) of NTS1 was the only effect to approach significance (p = 0.087), and a t test comparing the vehicle group and the pemoline + no stress group rev ealed a significant effect (t(13) = 2.412, p < 0.05).
79 1 2 3 4 5 6 0 50 100 150* * BDaysize of injured tissue (mm2) 1 2 3 4 5 6 0 20 40 60 80 100stress no stress ADay% of rats that exhibited injury 0 1 2 3 4 5 0 20 40 60 80 100CDayoral contact duration(% of total time recorded) 0 1 2 3 4 5 0 20 40 60 80 100DDaystereotypies duration(% of total time recorded)Figure 3 1. Effects of repeated social defeat stress on pemolineinduced self injury and stereotypy. All rats exhibited pemolineinduced self injury (A), however the rats with a history of repeated soc ial defeat stress had larger areas of tissue damage (B). Repeated social defeat stress did not have any significant effect on pemolineinduced oral stereotypy (C) or whole body stereotypies (D). All values are expressed as group means S EM Significant di fferences between stressed and nonstressed rats are depicted with an asterisk
80 NTR1 surface intracellular S+I 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75OD/b-actin OD OCT3 surface intracellular S+I 0.0 0.5 1.0 1.5 2.0 2.5OD/b-actin OD GluR1 surface intracellular S+I 0.0 0.4 0.8 1.2pemoline stress pemoline no stress vehicle OD/b-actin OD GluR2 surface intracellular S+I 0.0 0.5 1.0 1.5 2.0 2.5OD/b-actin ODA D C B vehicle pem no stress pem stress 0.00 0.25 0.50 0.75 1.00GluR2 surface/intracellular ratio vehicle pem no stress pem stress 0.00 0.25 0.50 0.75 1.00OCT3 surface/intracellular ratio vehicle pem no stress pem stress 0.00 0.25 0.50 0.75 1.00GluR1 surface/intracellular ratio vehicle pem no stress pem stress 0.00 0.25 0.50 0.75 1.00NTS1 surface/intracellular ratioH G F E Figure 3 2. Effects of pemoline treatment and repeated stress on surface (S) and intracellular (I) levels of GluR1, GluR2, NTS1, and OCT3 in the nucleus accumbens. There were no betweengroups differences in measures of surface levels, intracellular levels, total expression levels (S+I), or localization
81 ratios (S/I) of the AMPA receptor subunits, GluR1 (A,B) and GluR2 (C,D), NTS1 (E,F), or OCT3 (G,H). All values are expressed as group means S EM Discussion In this experiment we found that the dose of pemoline used caused every rat to exhibit some pemoline induced self injury. This incidence measure does not address the component characteristics of pemolineinduced self biting, which were evaluated by our oral contact duration and size of injured tissue measures. Those assessments revealed that repeated social defeat stress did not increase the amount of self injurious oral contact but did significantly increase the severity of the self biting, which resulted in larger areas of injured tissue. The oral contact duration results (i.e. the amount of oral stereotypy ) are consistent with our other finding that social defeat stress had no impact on the expression of whole body stere otypies like head bobbing and burrowing. These results suggest that the history of repeated stress increases the severity of pemolineinduced self injurious behavior which is consistent with reports t hat stressors can exacerbate self injury in humans (Anderson and Ernst, 1994) We utilized social defeat as our stressor because it is a potent, nonhabituating, processive stress that offers clear parallels to the psychosocial stressors that humans experience each day (Huhman, 2006; Tornatzky and Miczek, 1994) Additionally, repeated social defeat alters the normal functioning of the HPA axis, which is responsible for the phy siological responses to stress. Specifically, levels of the stress h ormone corticosterone is disregulated after repeated social defeat (Covington and Miczek, 2001) and this is consistent with HPA axis dysfunction in the patient populations that exhibit self injurious behavior (Curin et al., 2003; Hall et al., 2008; Hessl et al., 2002; Sand man et al., 2008; Symons et al., 2003 b ; Verhoeven et al., 1999)
82 Behavioral sensitization has been described in relation to several psychostimulants and involves increased behavioral responsiveness with repeated injection of drug (Pierce and Kalivas, 1997). Pemolineinduced self injury fits the criteria of the behavioral sensitization phenomenon because the initial injections of pemoline do not induce self injurious behavior, but repeated injections result in the expression of self injury after days of treatment. The induction and maintenance of pemolineinduced self injury appear to be caused in part by glutamatemediated neuroplastic changes (Muehlmann and Devine, 2008), and similar changes appear to also be responsible for behavioral sensitization triggered by cocaine and amphetamine (Wolf, 1998). Furthermore, social defeat stress cross sensitizes with cocaine and amphetamine (Covington and Miczek, 2001; Dietz et al., 2008). Our finding that stress potentiates pemolineinduced self injury suggests that so cial defeat stress also cross sensitizes with pemoline, as measured by self injury severity. Stress (Kalivas and Duffy, 1995; RougePont et al., 1993) and psychostimulants similar to pemoline (RougePont et al., 1995) cause increases in dopamine release in the nucleus accumbens, and stress (Campioni et al., 2009) and psychostimulants induce glutamatemediated plasticity (Wolf et al., 2004) in the nucleus accumbens. This suggests that the effects on dopaminergic and glutamatergic signaling in the nucleus acc umbens may mediate the cross sensitization between psychostimulants and stress (Pacchioni et al., 2007). We investigated the expression levels of several proteins within the nucleus accumbens to evaluate their role in the potentiation of pemolineinduced s elf injury by social defeat stress. We found no differences in either surface or intracellular levels of the AMPA receptor subunits, GluR1 or GluR2, in comparisons between vehicletreated
83 rats, pemolinetreated nonstressed rats, and pemolinetreated stres sed rats. These negative results are consistent with other behavioral sensitization experiments wherein one day after the last repeated cocaine injection there were no changes in surface or intracellular levels of GluR1 or GluR2 in the nucleus accumbens (B oudreau and Wolf, 2005; Boudreau et al., 2007). If AMPA receptor subunit expression and localization are pursued in this model in the future, it is recommended that the nucleus accumbens be separated into shell and core regions before analysis, as these subregions differ in their expression of dopamine receptors and neuropeptides (Meredith, 1999), and may reveal differential changes in response to pemoline and/or stress. Neurotensin is a neuromodulatory peptide found throughout the brain. Neurotensin increases the release of dopamine in the nucleus accumbens, through actions in the VTA (Kalivas and Duffy, 1990) and by activating presynaptic receptors in the nucleus accumbens terminal regions that inhibit dopaminergic autoreceptor functioning (Fawaz et al., 2 009). Furthermore, following stress, neurotensin levels are elevated in the VTA (Deutch et al., 1987) and neurotensin binding sites in the nucleus accumbens are increased (Xing et al., 1998). For these reasons we evaluated whether the exacerbation of pemol ine induced self injury may be a result of changes in neurotensin neurotransmission associated with changes in NTS1 expression or localization. Our statistical tests found no significant change in NTS1 levels between vehicle controls, pemolinetreated stre ssed and pemolinetreated nonstressed rats. However, a t test comparison of the vehicle controls and pemolinetreated nonstressed rats revealed a significant difference in total NTS1 expression. This suggests that changes in
84 neurotensin neurotransmission and receptor expression may be induced upon pemoline administration. Finally, we evaluated the expression and localization of a monoamine transporter called OCT3. The OCT3 is a low affinity, high capacity transporter of monoamines, including dopamine, which is highly expressed throughout the striatum and nucleus accumbens (Amphoux et al., 2006; Gasser et al., 2009). There is evidence that OCT3s aid in transport of psychostimulants into the brain and partially regulate the development of behavioral sensiti zation and drug induced dopamine release. More specifically, rats that develop methamphetamineinduced behavioral sensitization have lower expression of OCT3 (Kitaichi et al., 2003) and reducing OCT3 function through antisense administration or with geneti c manipulations increases methamphetamineinduced elevations in dopamine levels in the nucleus accumbens and methamphetamineand cocaineinduced locomotion (Kitaichi et al., 2005; Nakayama et al., 2007; Vialou et al., 2008). Additionally, OCT3 transport o f monoamines is blocked by the stress hormone corticosterone (Gasser et al., 2006). In regards to the current experiment, we hypothesized that the exacerbation of pemolineinduced self injury may be a result of increased dopamine release in the nucleus acc umbens resulting from the cumulative effects of repeated stress and pemoline administration. Moreover, both social defeat stress and pemoline administration are associated with elevations in corticosterone levels (Covington and Miczek, 2001; Kies and Devine, 2004). Though not directly measured in this experiment, the potential concomitant rise in circulating corticosterone following repeated social defeat stress and pemoline administration might further increase dopamine levels in the nucleus accumbens by blocking transport by OCT3.
85 This convergence of dopamine and corticosterone involvement in pemolineinduced self injury, and the localization of OCT3 in the nucleus accumbens (the brain region responsible for the expression of behavioral sensitization) made OCT3 an interesting target of investigation. However, we found no change in expression or localization of OCT3 in our comparisons of vehicleand pemolinetreated groups, or pemolinetreated stressed and nonstressed groups. This suggests that changes in OCT3 expression or function are not responsible for the potentiation of pemolineinduced self injury by repeated social defeat stress.
86 CHAPTER 4 INDIVIDUAL DIFFERENCES IN VULNERABILITY FOR PEMOLINE INDUCED SELF INJURIOUS BEHAVIOR I S PREDICT ED BY STRESS RESPONSIVENESS Background The prevalence of self injurious behavior differs among the disorders in which self injurious behavior is expressed. All, or nearly all, individuals with LeschNyhan syndrome self injure ( Anderson and Ernst, 1994). The incidence of self injurious behavior in Prader Willi syndrome is estimated around 5080% (Hiraiwa et al., 2007; Symons et al., 1999) and the incidence of self injury in autism is roughly 50% (Baghdadli et al., 2003 ) Taken toget her, in the disorders that exhibit symptom variability ( e.g. Prader Willi syndrome, autism ) there is also variability in individual vulnerability to self injure. In Chapter 3 we described the effect of stress on the self injury exhibited by humans and ani mals (e.g. Anderson and Ernst, 1994, Davenport et al., 2008). Findings from human self injurers especially highlight the importance of responsiveness to stress in the expression of self injurious behavior, though the potential that innate responsiveness to stressful stimuli may predispose individuals toward self injury has not been explored. A very interesting rodent model of stress responsiveness elucidated individual differences in neuronal, hormonal, and behavioral responses to stress. In this model, rat s that exhibit high rates of locomotor reactivity, or exploratory behavior, in a novel environment are compared to rats that exhibit low rates of locomotor reactivity. The rats that show more exploratory behavior in a novel environment are said to be high responder (or HR) rats and the rats that show less exploratory behavior in the novel environment are termed low responder (or LR) rats. The HR rats have increased and prolonged corticosterone release after a mild stress (Kabbaj et al., 2000) they show
87 less anxiety like behavior in tests of anxiety (Dellu et al., 1996; Kabbaj et al., 2000), and they differ in expression levels of many gene products that regulate the HPA axis, or stress pathway, as compared to the LR rats (Kabbaj et al., 2000) Early studies of these rats also revealed that HR rats were more likely to self administer low doses of amphetamine, which suggested that the HR rats were more vulnerable to developing dr ug abuse ( Piazza et al., 1989). In the initial characterization of the pemoline model (Kies and Devine, 2004) our lab described a doseresponse analysis wherein incidence of pemolineinduced self injury was related to dose. At moderate doses of pemoline some failed to show any pemolineinduced self injury, while other rats injured quite severely. In this study we inves tigated the potential that individual differences in stress responsiveness predict individ ual vulnerability to develop self injurious behavior in the pemoline model Methods Animals Eighteen male LE rats weighing 150175g upon delivery (Charles River Laboratories, Raleigh, NC) were housed in a climate controlled vivarium with a 12h/12h light/dark schedule (lights on at 7:00a.m.). Standard laboratory rat chow (Lab Diet 5001) and tap water were available ad libitum The rats were pair housed in standard polycarbonate cages (43 cm x 21.5 cm x 25.5 cm) during 6 days of acclimation to the housing facility and for eight days following circular corridor screening. The rats were then singly housed in identical polycarbonate cages upon initiation of pemoline treatment. All of the experimental procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals, and were preapproved by the Institutional Animal Care and Use Committee at the University of Florida.
88 Drug Pemoline (Spectrum Chemicals, New Brunswick, New Jersey) was suspended at a concentration of 50 mg/ml in warm peanut oil (held at approximately 36C), with constant stirring. Stress Responsiveness S creening The novel environment of a circular corridor was used to screen for locomotor responsiveness to novelty. Circular corridors were constructed using a small plastic cylind er placed inside of a larger plastic cylinder (forming the inner and outer walls of the corridor, respectively). Both cylinders were washed with 4% bleach and standard bedding was placed on the path (7 cm wide, 44 cm outer diameter) of the circular corrido r. The lights in the experimentation room were turned off and dim illumination was provided by small lamps placed at the floor beside each circular corridor. Rats were placed in this novel environment between one to two hours after lights on (i.e. between 8 and 9 a.m.). Video cameras secured to the ceiling were used to record locomotor behavior. Locomotion was scored for the first 60 minutes that the rats were in the circular corridor. A trained observer quantified locomotion by dividing the video of the co rridor into quadrants. One count was recorded each time the rat crossed a line, but no further counts from that line were made until the rat crossed another line. Rats were determined to be either HRs or LRs based on whether their locomotor scores were above or below, respectively, the median of the distribution of a large sample of rats (n=336). The overall median of the entire population of rats s creened in our laboratory is 223 line crossings.
8 9 Drug Treatment An experimenter blind to the HR or LR designat ion of the rats weighed and injected each rat with pemoline at 150 mg/kg (s.c.) each day for ten days. These injections were administered at the nape of the neck and either flank on a rotating basis. Assays of Self I njury Stereotypy, and Related Behaviors The rats were visually inspected each time they were injected and then again every evening, and the inspections were videotaped. During these inspections, each rat was held in front of a video camera and the head, forepaws, hindpaws, ventrum and tail were displayed. The presence or absence of tissue injury (denuded skin, erythema, edema, or open lesion) was noted for each rat. Any rat with an open lesion was immediately euthanized. Still images of the injured tissue were taken from the videotapes, and MCI D software (Imaging Research Inc., St. Catherines, ON, Canada) was used to draw outlines around the injured tissue, and to calculate the area of injury in mm2. Additionally, night vision cameras were focused on the cages of the rats (one camera per cage), and 5minute time samples were recorded once per hour for 8 hours each night. The duration of self injurious oral contact, stereotypy, grooming, rearing and the amount of locomotion were quantified during each videotaped interval by a trained observer who was blind to the HR or LR designation. Self injurious oral contact was defined as all oral contact that stayed fixed on any one body part for longer than 2 seconds. This was differentiated from grooming, which is oral contact with any part of the body that continues to move from site to site on the body (e.g. oral contact with the forepaws, that then moves up each forelimb and continues to the ventrum, and the contact is not sustained at any spot on the body for longer than 2 seconds A second
90 trained but experimentally blind observer rescored a subset of the videorecordings to evaluate inter observer reliability for the duration of self injurious oral contact (one randomly selected night of recordings for each rat). The duration of stereotyped behaviors w as also quantified from the video time samples. These behaviors included episodes of stereotyped headbobbing, cagelicking, and digging/sniffing/burrowing through the bedding that exceeded 3 sec duratio n (Cromwell et al., 1999). Since individual rats exhi bited substantial differences in the expression of stereotypy (e.g. some primarily exhibited headbobbing or cagelicking, whereas others exhibited digging/sniffing/burrowing) the stereotypy scores are reported as aggregate scores that compile the total duration of all these stereotyped behaviors. Home cage rearing and locomotion were also scored from the video recordings. Rearing was counted each time a rat lifted both forepaws off the cage floor for at least 3 seconds. A subsequent rear was only counted after the rat resumed and maintained contact with the cage floor for at least 1 sec. Locomotion was counted by sectioning the video image of the cage into thirds (i.e. drawing lines on the television monitor dividing the cage into three equal parts along i ts length) and tallying the number of times the rats forepaws crossed these lines. However, a subsequent crossing of that line was only counted after the rat crossed the other line. Thus, each locomotor count represents a minimum of 14.3 cm traversing of the cage. On the morning of the 11th day each rat was visually inspected. After this final inspection, each rat was rapidly decapitated. Trunk blood (6 ml) was collected into chilled polyp EDTA at 20 mg/ ml. The tubes were centrifuged at 1000g at 4C for 5 min. The plasma fraction was aliquotted and frozen at
91 80C. Radioimmunoassay was performed to quantify the levels of corticosterone in plasma using a kit from Diagnostic Products Corporation (Los Angeles, CA). Additionally, after termination the thymus and adrenal glands were removed in order to assess whether pemoline administration caused differential endocrine or immune responses in the HR or LR rats. Statistical A nalyses Betweengroups differences in circular corridor line crossings, duration of self injurious oral contact, number of injured sites, size of the injured tissue, duration of stereotypy, grooming, rearing, the amount of locomotion, and body weight were each evaluated using repeated measures analyses of variance (RM ANOVA). Between groups differences in all these dependent measures were treated as statistically reliable when the pvalues were less than 0.05. All significant effects were f urther analyzed with pre planned Fishers least significant difference (LSD) post tests. Betweengroups differences in circular corridor totals, thymus weight, adrenal gland weight, and corticosterone levels were each analyzed by 2tailed t tests. Two HR r ats were euthanized before the end of the experiment because they had open lesions. In these cases, the missing data were replaced by repeating the final score that was attained for each dependent measure through the end of the experiment. This strategy was used to avoid the potential that the group means would underestimate the area of tissue damage and self injurious oral contact scores, and to avoid the potential that the group means would over or under estimate the scores of the other behaviors when th e most severe self injurers were terminated.
92 Results HR rats (n = 9 ) exhibited more locomotion in the circular corridor at each of the six, ten minute i ntervals than did the LR rats (n = 9; Fig. 4 1A). The RM ANOVA revealed a significant main effect of ti me (F(5,80) = 68.19, p < 0. 00 01), wherein all rats locomoted more at the beginning of the circular corridor screening and decreased as time passed, and a main effect of stress responsiveness (F(1,80) = 11.01, p < 0.01). Cumulatively, HR rats had higher tot al rates of locomotion compared to the LR rats (Fig. 4 1B; t(16) = 3.319 p < 0.01). More of the HR rats exhibited pemolineinduced self injurious behavior compared to the LR rats (Fig. 4 2A). HR rats also spent more time self injuring (Fig. 4 2B), had mo re sites of tissue injury (Fig. 4 2C), and larger total size of injured tissue (Fig. 4 2D) than did the LR rats For the self injurious oral contact measure, significant main effects of time (F(9,144) = 3.106 p < 0.01) and stress responsiveness (F(1,144) = 5.354, p < 0.05) were found as HR rats exhibited self injurious oral contact throughout nights 3 through 10, whereas the LR rats typically did not. The HR rats also had significantly more injured sites than the LR rats did. The RM ANOVA revealed a signif icant main effect of time (F(10,160) = 8.905 p < 0.01), a significant main effect of stress responsiveness (F(1,160) = 6.805, p < 0.05), and a significant time x stress responsiveness interaction (F(10,160) = 3.604 p < 0.00 1). HR rats also displayed larg er areas of tissue damage t han did the LR rats. There were significant main effect s of time (F(10,160) = 5.440, p < 0.0 00 1) and stress responsiveness (F(1 ,160) = 4.924, p < 0.05) and a significant time x stress responsiveness interaction (F(10,160) = 3.147 p < 0.01). Taken together, the HR rats exhibited a greater vulnerability to develop pemolineinduced self injury and as a result
93 spent more time self injuring and injured more tissue area and at more body sites than the LR rats did. LR rats expressed hig her rates of other pemolineinduced behaviors than HR rats did. Specifically, LR rats spent more time exhibiting pemolineinduced stereotypies than did HR rats (Fig. 4 3A). The RM ANOVA analyses revealed a significant main effect of stress responsiveness ( F(1,144) = 7.730, p < 0.05), and the LSD post tests confirmed that LR rats exhibited higher rates of stereotyped behavior at the beginning and toward the end of the ten day pemoline administration period. Grooming was a low rate behavior and did not differ in duration between the HR and LR rats (Fig. 4 3B), although a significant main effect of time was found by the RM ANOVA (F(9,144) = 2.107, p < 0.05). The rate of locomotion did not differ between HR and LR rats (Fig. 4 3C). A significant main effect of t ime (F(9,144) = 5/048 p < 0. 00 01) was found, wherein all rats exhibited high rates of locomotion on nights 1 and 2, and then again on nights 5 through 10. The duration of rearing behavior was not different between HR and LR rats (Fig. 4 3D), although the rates of rearing did change across days, as revealed by a significant main effect of time (F(9,144) = 5.048 p < 0.0 00 1). The body weight s of all the rats declined throughout the first four days of pemoline administration, rebounded to baseline levels by day 6, and continued to increase until the end of the experiment (Fig. 4 4A). These changes contributed to a significant main effect of time (F(10,160) = 17.8 3, p < 0. 00 01) in the RM ANOVA analysis. No significant differences in thymus (Fig. 5 4B) or adrenal gland mass (Fig. 5 4C) were found, repli cating previous findings (Kies and Devine, 2004). Circulating corticosterone
94 levels were not different between HR and LR rats following ten days of pemoline administration (Fig. 4 4D). 0 1 2 3 4 5 6 0 25 50 75 100HR LR * * *10 min intervalline crossings HR LR 0 100 200 300*total line crossingsA B Figure 4 1. Differe nces between HR and LR rats in the circular corridor. HR rats exhibited more novelty induced locomotion. Data are represented across the six, ten minute intervals (A) and as cumulative totals (B). All values expressed are group means SEM Significant differences between HR and LR rats are depicted with an asterisk
95 0 1 2 3 4 5 6 7 8 9 10 11 0 25 50 75 100HR LR day% of rats that exhibited injury 0 1 2 3 4 5 6 7 8 9 10 0 10 20 30 40 50* * * *nightoral contact duration (% of total time recorded) 0 1 2 3 4 5 6 7 8 9 10 11 0.0 0.5 1.0 1.5 2.0* * * * *day# of injured sites 0 1 2 3 4 5 6 7 8 9 10 11 0 25 50 75 100 125 150 175* * * *daysize of injured tissue (mm2)A B C D Figure 4 2. Effects of stress responsiveness on pemolineinduced self injury. HR rats were more likely to exhibit pemolin e induced self injury (A), spent more time injuring (B), injured more body s ites (C), and had larger total area of tissue damage (D) than the low responder (LR) rats did. All values expressed are group means SEM Significant differences between HR and LR rats are depicted with an asterisk
96 0 1 2 3 4 5 6 7 8 9 10 0 10 20 30 40 50* *HR LR nightstereotypies duration (% of total time recorded) 0 1 2 3 4 5 6 7 8 9 10 0 10 20 30 40 50nightgrooming duration (% of total time recorded) 0 1 2 3 4 5 6 7 8 9 10 0 25 50 75 100nightlocomotion(line crossings) 0 1 2 3 4 5 6 7 8 9 10 0 10 20 30 40 50nightrearing duration (% of total time recorded)D C B A Figure 4 3. Effect of stress respon siveness on pemolineinduced stereotypy, grooming, locomoti on, and rearing. LR rats exhibited more pemolineinduced ster eotypy than the HR rats did (A). Duration of grooming was not different betw een HR and LR rats (B). LR and HR rats had equal locomotor r ates (C). There were no differences between HR and LR rats on measurements of rearing duration. All values expressed are group means SE M. Significant differences between HR and LR rats are depicted with an asterisk
97 0 1 2 3 4 5 6 7 8 9 10 11 0 100 200HR LR 275 300 325 350 375daybody weight (g) HR LR 0 25 50 75 100 125 150 175thymus weight (mg/100g body weight) HR LR 0 5 10 15 20adrenal gland weight (mg/100g body weight) HR LR 0 50 100 150 200 250CORT (ng/ml)B C A D Figure 4 4. Effects of pemoline on body weight, gland mass, and corticosterone con centrations. HR and LR rats exhibited the same amount of weight loss and eventual weight gain during pemoline administration (A). There were no differences in thymus (B) or adrenal gland (C) mass between HR an d LR rats. There were also no statistically significant differences in circulating corticosterone concentrations between HR and LR rats (D). All values e xpressed are group means SEM Discussion The HR/LR model has been used to elucidate differential exp ression of emotionality (Blanchard et al., 2009), antidepressant drug response (Jama et al., 2008), and stress induced pathologies, including drug abuse ( Piazza et al., 1989, Kabbaj et al., 2004). Since self injurio us behavior is also stressinduced or stress responsive, we hypothesized that the individual differences in stress responsiveness that are assayed in the HR/LR model would also predict which rats would be particularly vulnerable to pemolineinduced self in jury We found that HR rats are more susceptible to develop pemolineinduced self injurious behavior and in doing so exhibit longer durations of self -
98 injurious oral contact, more sites of injury, and larger areas of tissue damage than the LR rats do. The body and gland weight data from the HR and LR rats confirm our previous findings that repeated pemoline administration generally does not negatively i mpact the health of the rats (Kies and Devine, 2004). Interestingly, we found a trend toward higher concen trations of corticosterone in the HR rats, although this did not reach statistical significance. Previous studies have shown that mild stressors cause increases in circulating corticosterone in HR rats relative to LR rats (Kabbaj et al., 2000). In light of this, there may potentially be significant differences in corticosterone levels between the HR and LR rats at the beginning of the pemoline treatment regimen that may have lessened as the repeated treatment persisted. There are many interesting basal and drug induced differences in dopaminergic and stress pathways between the HR and LR rats, which may explain the differential vulnerability to develop pemolineinduced self injurious behavior Since pemoline induced self injury is a product of disregulation of dopaminergic systems (Cromwell et al., 1997), the most relevant finding is that HR rats have more basal extracellular dopamine in the nucleus accumbens compared to LR rats (Hooks et al., 1992). Dopaminergic and non dopaminergic systems may be responsible for this differential dopaminergic tone. First, there is less tyrosine hydroxylase protein in the VTA of the HR rats (Lucas et al., 1998). Less tyrosine hydroxylase in this dopaminergic cell body region leads to less dopamine release from somatodendriti c regions of the neuron, which causes less autoreceptor activation and precipitates increased dopamine release from axonal regions (i.e. the nucleus accumbens; White and Wang, 1984). Furthermore,
99 there is less dopamine D2 receptor binding in the nucleus ac cumbens of the HR rats (Hooks et al., 1994), which further reduces autoreceptor activation and allows dopamine release to continue. Nondopaminergic mechanisms may also be responsible for the increase in dopamine release in the nucleus accumbens of HR rats Lower levels of cholecystokinin mRNA has been found in the substantia nigra and VTA of HR rats compared to those of LR rats (Lucas et al., 1998, Ballaz et al., 2008). Cholecystokinin in these brain regions exerts inhibitory control over dopamine release in do paminergic projection nuclei (Xie et al., 2001). Additionally, more preproenkephalin mRNA has been found in the nucleus accumbens of HR rats compared to LR rats (Lucas et al., 1998). Nucleus accumbens to VTA enkephalinergic projections exert excitator y cont rol over dopamine release (Gysling and Wang, 1983; Kalivas, 1993). B asal differences in dopaminergic tone between the HR and LR rats cause differential responsiveness to psychostimulant drugs. HR rats show behavioral sensitization to low doses amphe tamine, whereas LR rats do not (Dietz et al., 2005; Hooks et al., 1991a). Moreover, HR rats have an increased locomotor response to acute cocaine injection, which is associated with increased dopamine release from that cocaine administration (Hooks et al., 1991b ). Our finding that HR rats are more vulnerable to develop pemolineinduced self injury is consistent with our previous work, which suggested that pemolineinduced self injurious behavior is also a form of behavioral sensitization that is produced through glutamatemediated neuroplastic changes (Muehlmann and Devine, 2008). There are also very interesting differences between HR and LR rats in the stress pathway, or HPA axis. HR rats have more corticotropin releasing hormone (CRH)
100 mRNA in the paravent ricular nuc leus of the hypothalamus compared to LR rats (Kabbaj et al., 2000). Increased CRH mRNA causes increased HPA axis activation. CRH release from the PVN causes release of ACTH from the anterior pituitary, which then stimulates release of corticosterone from the adrenal gland. Corticosterone binding to glucocorticoid receptors in the brain causes several biological reactions, the most important being negative feedback to the PVN from the hippocampus to reduce HPA activation ( Herman and Cullinan, 1997). HR rats have less glucocorticoid receptor mRNA in the CA1 region of the hippocampus, which resul ts in less negative feedback (Kabbaj et al., 2000). In summary, HR rats have increased activation and less negative feedback of the HPA axis than LR rats do. This biological variation causes increased corticosterone release in the HR rats after stress (Kabbaj et al., 2000). Corticosterone produces dopamine release in the nucleus accumbens (Piazza and LeMoal, 1996) and H R rats have increased dopamine release in the nucleus accumbens than LR rats do after stress (Dellu et al., 1996, RougePont et al., 1998 ) or aft er corticosterone injections (Piazza and LeMoal, 1996). In summary, these findings indicate that rats with hig her dopaminergic and HPA axis tone are more vulnerable to develop pemolineinduced self injurious behavior Very little attention has been given to dopaminergic and stress interactions in clinical populations, however using an animal model with individual vulnerability may help to elucidate specific sites of interaction between dopaminergic and stress systems that confer heighte ned vulnerability to develop self injury These results suggest that we can begin to use the HR/LR model in order to determine the basis for the genetic or epigenetic predisposition to develop pemolineinduced self injurious behavior by
101 comparing HR and LR rats. Furthermore, we can also use the pemoline model to determine which pemolineinduced cellular changes are specifically responsible for causing the induction of self injury by comparing pemolinetreated, non injurious LR rats and pemolinetreated, self injurious HR rats. These comparisons will undoubtedly help to unveil the currently unknown biological bas e s of this greatly destr uctive behavior disorder.
102 CHAPTER 5 CHARACTERIZATION OF DRUG TITERS, MONOAMI NES, MONOAMINE METABOLITES, AND AMI NO ACIDS IN THE PEMO LINE MODEL OF SELF INJURIOUS BEHAVIOR Background Only a limited number of studies have investigated the ef fect of pemoline on catecholaminergic systems; most report some efficacy of pemoline to block reuptake of dopamine (Fuller et al., 1978) and norepinephrine (Molina and Orsingher, 1981) and to induce dopamine (Fuller et al., 1978) and norepinephrine (Gilber t et al., 1978) release. No study to date has characterized changes in the dynamics of monoamine neurotransmission in relation to the induction and maintenance of pemolineinduced self injurious behavior. Furthermore, the role of pemoline titers in relation to induction and maintenance of self injurious behavior with repeated pemoline administration has also not been investigated. To this end we analyzed monoamine, monoamine metabolite, and amino acids levels in different brain regions across days of repeat ed vehicle or pemoline administration. We also measured plasma and brain pemoline levels throughout five days of repeated pemoline treatment. Finally, as a follow up to our investigations of the role of stress in pemoline induced self injurious behavior, w e measured the changes in corticosterone levels across days of repeated pemoline or vehicle injections. Methods Animals Sixty four LE rats weighing 200225g upon delivery (Charles River Laboratories, Raleigh, NC) were housed in a climate controlled vivari um with a 12h/12h light/dark schedule (lights on at 7:00a.m.). Standard laboratory rat chow (Lab Diet 5001) and tap
103 water were available ad libitum The rats were pair housed in standard polycarbonate cages (43 cm x 21.5 cm x 25.5 cm) during six days of ac climation to the housing facility. All of the experimental procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals, and were preapproved by the Institutional Animal Care and Use Committee at the University of Flor ida. Drug Pemoline (2amino 5 phenyl 1,3 oxazol 4 one; Spectrum Chemicals, New Brunswick, New Jersey) was suspended at a concentration of 50 mg/ml in peanut oil. In order to get the pemoline into fine suspension it was kept stirring overnight. Drug and Vehicle Treatment Following six days of habituation to the housing facility, the rats were separated and singly housed in cages identical to the cages they were habituated in. Each morning (approximately 8 am) the rats were weighed and injected subcutaneously with either pemoline at 150 mg/kg (n=32) or peanut oil vehicle (n=32). These injections were administered at the nape of the neck and either flank on a rotating basis. Assays of Self Injury The rats were visually inspected each time they were injected and immediately before they were decapitated and the inspections were videotaped. During these inspections, each rat was held in front of a video camera and the head, forepaws, hindpaws, ventrum and tail were displayed. The presence or absence of tissue injury (denuded skin, erythema, edema, or open lesi on) was noted for each rat. Still images of the injured tissue were taken from the videotapes, and MCID software (Imaging Research Inc., St. Catherines, ON, Canada) was used to draw outlines around the injur ed tissue, and to calculate the area of injury in mm2.
104 Night vision cameras were focused on the cages of the rats (one camera per cage), and 5minute time samples were recorded once every three hours over the entire day. The duration of self injurious oral contact was quantified during each videotaped interval by a trained observer. Self injurious oral contact was defined as all oral contact that stayed fixed on any one body part for longer than 2 seconds. This was differentiated from grooming, which is oral contact with any part of the body that continues to move from site to site on the body (e.g. oral contact with the forepaws, that then moves up each forelimb and continues to the ventrum, and the contact is not sustained at any spot on the body for longer than 2 seconds). T he duration (in seconds) of oral contact behavior was summed over the entire day (i.e. from the eight video samples) and divided by the total number of seconds recorded, 2400 seconds. For example, the duration measures on day 1 cover the percent of time the rat was exhibiting oral contact during the recording periods between the first injection and just prior to the second injection, thereby covering the entire 24 hours following the first injection. Decapitation Schedule Four pemolin e treated rats and four vehicletreated rats were decapitated at each of the following eight time points (Table 51): two hours after the first injection, 12 hours after the first injection, 24 hours after the first injection (i.e. day 2), 24 hours after t he second injection (i.e. day 3), 24 hours after the third injection (i.e. day 4), 12 hours after the fourth injection (i.e. middle of day 4), 24 hours after the fourth injection (i.e. day 5), and 24 hours after the fifth injection (i.e. day 6).
105 Group na me on graphs Experimental day the rats were decapitated Number of injections received Hours since last injection 1 1 1 2 1.5 1 1 12 2 2 1 24 3 3 2 24 4 4 3 24 4.5 4 4 12 5 5 4 24 6 6 5 24 Table 51. Decapitation schedule for pemolineand vehicletreated rats. Sample Collection The rats were rapidly decapitated and 5ml of truck blood was collected in chilled tubes containing 500ul disodium ethylenediaminetetraacetic acid (EDTA; 20ug/ul). Plasma was separated by centrifugation at 1000g for 5 min at 4C. Brains were immediately removed and dissected on ice in to the following regions: septum, striatum, hippocampus, amygdala with overlying temporal cortex, cortex (from frontal, parietal and occipital lobes), thalamus, and ventral tegmentum (including VTA and SN). The d issection technique was adapted from Glowinski and Iversen (1966). Plasma and brain regions were frozen on dry ice and stored at 80C until processed. High Pressure Liquid Chromatography (HPLC) System Our HPLC system consisted of a Waters 510 pump and a Machery Nagel analytical column ( 125 x 4 mm, C18, 120A pore, 5um particle), which was protected by a guard column. An ultraviolet detector was used for pemoline analyses. Electrochemical detection was used for monoamine, metabolite, and amino acid analyses. For this configuration we added a guard cell, a dual electrode analytical cell, and a C oulochem electrochemical detector.
106 Pemoline Titer Analyses The protocol for pemoline analyses was adapted from Aoyama et al. (1988). The mobile phase was 80:20 distilled water: acetonitrile, pH 5 adjusted with 15uM phosphoric acid, filtered, degassed, and pumped at a flow rate of 0.7 ml/min. Plasma pemoline extraction Plasma (0.1ml) was added to 0.5ml 1M carbonate buffer (pH 10) and 4ml methylene c hloride. This solution was vortexed for 10 min, followed by centrifugation at 1680g for 5 min. The upper aqueous phase was removed and the organic phase was transferred to a fresh tube containing 1 ml of methylene chloride containing 1ug/ml of the internal standard, 5 methyl 5 phenylhydantoin. This solution was dried down in a Speed Vac overnight. The next morning 50ul of mobile phase was added to the tube, heated at 70C for 30 min, centrifuged at 1000g for 2 min, sonicated for 5 min, and centrifuged again at 1000g for 2 min. 40ul of sample was then immediately injected in to the HPLC system. Brain pemoline extraction Brain regions were weighed and sonicated in 2 volumes (using the formula, 1mg = 0.84ul) 0.05M ice cold perchloric acid. Whole homogenate ( 0.1ml) was transferred to a fresh tube and extracted like the plasma was. After the mobile phase was added to the dried down pemoline pellet, sonicated, and centrifuged at 1000g, the supernatant was transferred to a fresh tube and centrifuged at 15,300g for 5 min to remove any remaining particulates from the sample. 40ul of sample was injected in to the HPLC system.
107 P emoline standards Pemoline standards were made up in methanol (10, 20, 50, 200, 1000, 2000, 4000, and 10000 ng/ml). Samples (0.1ml) of each w ere dried down in the Speed Vac. Either 0.1ml blank plasma or 0.1ml blank brain homogenate was added to the dried down standard and extracted as described above. Ultraviolet detection The ultraviolet detector was set at 215nm for pemoline and internal standard detection. Corticosterone Radioimmunoassay The corticosterone radioimmunoassay was purchased from Diagnostic Products Corporation. Monoamine Analyses The protocol used for measuring brain monoamines (norepinephrine (NE), dopamine (DA), and serotonin (5HT) and their metabolites (HVA, DOPAC, and 5HIAA) was adapted from Saito et al. (1992). Mobile phase consisted of 0.02M sodium acetate, 12.5mM citric acid, 10% (v/v) methanol, 0.042% heptanesulfonic acid, and 0.01mM disodium EDTA, pH 2.9. The mobil e phase was filtered through 0.45um filter, degassed, and pumped at 2.0 ml/min. Brain homogenate samples prepared for the pemoline analyses were centrifuged at 12,000g for 15 min at 4C. Supernatant was removed, added to 0.45um filter, and centrifuged at 20,000g for 10 min at 4C. 40ul of this filtered supernatant was injected in to the HPLC system.
108 Monoamine and metabolite standards Monoamine and metabolite standards (3.91, 7.8, 15.625, 31.25, 62.5, 125, 250, 500, and 1000 ng/ml) were made in 0.1M per chloric acid containing 0.1mM disodium EDTA. Standards were stored at 80C. Electrochemical detection of monoamines and metabolites For monoamine and metabolite analyses the guard cell was held at +550mV, cell 1 of the analytic cell was held at 100mV, 1 0nA, 1V output, and cell 2 was held at +450mV, 5uA for all analytes except for serotonin, which was analyzed at 2uA, with 1V output and a 5sec filter. For amino acids analyses, the guard cell was held at +700mV, cell 1 was held at +400mV, and cell 2 was held at +600mV. Amino Acid Analyses The protocol for analyzing brain amino acid concentrations (glutamate and GABA) was adapted from Murai et al. (1992). The mobile phase consisted of 0.05M NaH2PO4, 10% methanol, 9% tetrahydrofuran, and 0.1mM disodium EDTA, pH 6.0. The mobile phase was filtered, degassed, and pumped at 1.2ml/min. Amino acid derivatization In order to be analyzed by HPLC, amino acids were derivatized using ophthalaldehyde (OPA) and betamercaptoethanol. The prepared OPA/betamercaptoet hanol stock solution was diluted each day of HPLC analyses. The stock solution was made with 20mg of OPA, dissolved in 1ml of ethanol. 30ul of betamercaptoethanol and 9ml of 0.1M sodium tetraborate (pH 9.1) was added. OPA/betamercaptoethanol stock soluti on was stored at 4C. Each day 1ml of stock solution was diluted with 9ml of 0.1M sodium tetraborate to make the working solution.
109 Ninety seconds before injection in to the HPLC system, 5ul of brain supernatant sample and 45ul of the OPA/betamercaptoethanol working solution were combined and vortexed. 40ul of this sample was injected in to the HPLC system. Amino acid standards Stock amino acid standards (1mg/ml) were prepared in 50% methanol and stored at 4C for one month. Working standards (3.91, 7.8, 15.625, 31.25, 62.5, 125, 250, 500, and 1000 ng/ml) were diluted in 0.05M perchloric acid. Working standards were stored at 4C for five days. Electrochemical detection of amino acids For amino acids analyses, the guard cell was held at +700mV, cell 1 was held at +400mV, and cell 2 was held at +600mV. Protein Content Analyses Protein content was analyzed by the Bradford method for each of the dissected brain regions. Pemoline Monoamine, Metabolite, and Amino Acid Concentration A nalyses Area under the peak for each analyte was measured using Logger Pro software. For pemoline analyses this area measurement was divided by that of the internal standard. Concentration values were interpolated using the standard curve produced for each analyte and each sampl e type (e.g. plasma, brain regions). Concentration values were then corrected for protein content of each sample. The biosynthesis ratios of the monoamine were analyzed by calculating DA+DOPAC, DA+HVA, 5HT+5 HIAA. The turnover ratios of the monoamines were analyzed by calculating DOPAC/DA, HVA/DA, 5 HIAA/5 HT (Commissiong, 1985; Miura et al., 2002).
110 Statistical Analyses Measures of self injury (size of injured tissue and duration of oral contact) were analyzed by twoway ANOVA. For pemoline analyses, bet weengroups comparisons of measures from rats killed at the two times on days 1 and 4 were analyzed by t test. For an analysis of plasma pemoline across days 26 a oneway ANOVA was used. For comparisons between striatum and cortex pemoline levels, a twow ay ANOVA was used with Bonferroni post tests. For the corticosterone analyses, day 1 levels were compared between groups using a twoway ANOVA; a two way ANOVA was also used for comparisons between pemolineand vehicletreated groups across days 2 through 6; and a t test was used to compare corticosterone levels between the pemoline and vehicle groups for the one time point 12 hours after the fourth injection. Monoamine, metabolite, and amino acid levels were analyzed by 2way ANOVAs, as were the biosynthesis and turnover ratios Results Pemoline Induced Self Injurious Behavior Some of the pemolinetreated rats began to exhibit self injury on day 3 of the experiment. All rats that received 5 daily injections of 150mg/kg/day pemoline exhibited self injurious behavior (Fig. 51A), which is consistent with our previous studies (Chapters 2 and 3). The size of injured tissue (Fig. 51C) and oral contact duration (Fig. 5 1E) measures are slightly lower than the measures from our previous experiments, but this is m ost likely caused by the small numbers of rats that were included at each time point. Twoway ANOVAs revealed significant main effects of drug (F(1,36) = 13.64, p<0.001), time (F(5,36) = 4.812, p<0.01), and a significant drug x time interaction (F(5,36) = 4.812, p<0.01) for size of injured tissue and for oral contact duration (drug: F(1,36) =
111 29.47, p<0.0001; time: F(5,36) = 5.193, p<0.01; drug x time interaction: F(5,36) = 5.193, p<0.01). 1 2 3 4 5 6 0 25 50 75 100pemoline vehicle day% of rats that exhibited injury 1 2 3 4 5 6 0 50 100 150 200daysize of injured tissue (mm2) 1 2 3 4 5 6 0 25 50 75 100dayoral contact duration (% of total time recorded)A F E D C B 8 a.m. 8 p.m. 0 25 50 75 100day 4% of rats that exhibited injury 8 a.m. 8 p.m. 0 25 50 75 100day 4oral contact duration (% of total time recorded) 8 a.m. 8 p.m. 0 50 100 150 200day 4size of injured tissue (mm2) Figure 5 1. Pemoline induced self injurious behavior. Discontinuous data of the incidence (A,B), size (C,D), and duration (E,F) of self injury in the rats decapitated on the respective day is represented on the x axis. Data from the group of rats that were terminated 12 hours after the 4th injection (i.e. at 8 p.m.) are shown separately (B, D, F). Data from rats killed a t 8 a.m. on day 4 are included in graphs B, D, F for comparison. Data points represent the data for all rats that were terminated in the experiment on the day represented on the x axis. All values are expr essed as group means SEM. Pemoline Analyses No chromatogram peaks were seen at the time pemoline was to be resolved from the column in any of the analyses of vehicletreated rats, therefore all scores from
112 vehicle treated rats are zero. Plasma and brain pemoline titers followed a similar pattern across days (Fig. 5 2B,E). Anecdotally, both plasma and brain pemoline levels were higher two hours after the first injection than they were twelve hours post injection (Fig. 5 2A,D), though neither analyses (t t est in plasma comparisons, twoway ANOVA in comparisons of brain regions) reached statistical significance. By 24 hours following the first injection (the day 2 time point) pemoline was still present in plasma and brain. At this time the rats that continued in the experiment received another pemoline injection, meaning that pemoline is still on board when the rats receive their next pemoline injection. This remains true for the injections given after the second, third, and fourth injections as well (Fig. 5 2B,E). A one way ANOVA revealed significant differences in pemoline levels between the groups killed on days 2 through 6 (F(4,19) = 12.79, p < 0.0001), which indicates a significant change in plasma pemoline levels across days. A two way ANOVA revealed a s ignificant main effect of time in the brain pemoline analyses (F(4,29) = 7.612, p < 0.001). There were no significant effects found for region or region x time interaction. For analyses 12 hours after the fourth injection (Fig. 52C,F), only the pemoline l evels in the cortex were significantly greater 12 hours after injection, compared to 24 hours after the previous injection.
113 2 3 4 5 6 0 10 20 30 40 50 60 70 80dayplasma pemoline (ug/ml) 2 3 4 5 6 0 5 10 15 20 25 30 35daypemoline (ug/ml)A B 9 a.m. 9 p.m. 0 10 20 30 40 50 60 70 80day 4plasma pemoline (ug/ml) 10 a.m. 8 p.m. 0 10 20 30 40 50 60 70 80vehicle pemoline day 1plasma pemoline (ug/ml) 10 a.m. 8 p.m. 0 5 10 15 20 25 30 35striatum cortex day 1pemoline (ug/ml) 8 a.m. 8 p.m. 0 5 10 15 20 25 30 35 *day 4pemoline (ug/ml)F E D C Figure 5 2. Plasma and brain pemoline concentrations across days of pemoline administration. Plasma pemoline levels change drasti cally over the course of the day following the first pemoline injection (A), and titers remain high as repeated daily pemoline administrations continue (B). This is also the case for striatal and cortical pemoline concentrations (D,E). Twelve hours followi ng the fourth injection, pemoline levels are elevated in plasma (C), striatum (F), and cortex (F), though differences from preinjection on day 4 only reach statistical significance in the cortex measures. No pemolineassociated peaks were seen on the chro matograms from vehicletreated rats. All values are expressed as group means SEM. Significant differences are depicted with an asterisk. Corticosterone Analysis Two hours following the first injection (the 10 a.m. time point on day 1), corticosterone levels were significantly elevated in pemolinetreated rats compared to those of vehicletreated rats (Fig. 5 3A). Conversely, near the zenith of the circadian cycle on day 1(the 8 p.m. time point), corticosterone levels were blunted in the pemolinetreated rats, while the vehicletreated rats showed the normal corticosterone peak (Fig. 53A). A two way ANOVA for day 1 corticosterone levels revealed a significant main effect of time (F(1,12) = 5.671, p < 0.05) and a significant drug x time interaction (F(1,12) = 5.583, p < 0.05), but no significant main effect of drug. This
114 suggests that the ongoing elevation of corticosterone by pemoline caused a blunting of the diurnal corticosterone rhythm on day 1. At most morning time points analyzed, pemolinetreated ra ts had higher circulating corticosterone concentrations relative to those of vehicletreated rats (Fig. 5 3B). This was revealed by the twoway ANOVA which found a significant drug effect (F(1,30) = 8.179, p < 0.01), but there were no significant main effects of time or drug x time interaction, and Bonferroni posttests did not reveal a significant difference between the pemoline and vehicle groups at any particular time point. There was also no significant difference of corticosterone levels between the pem oline and vehicle groups 12 hours following the fourth injection (Fig. 53C), which suggests that the normal circadian rhythmicity of corticosterone may be restored in the pemolinetreated group by this time. 10 a.m. 8 p.m. 0 50 100 150 200 250pemoline vehicle day 1CORT (ng/ml) 2 3 4 5 6 0 50 100 150 200 250dayCORT (ng/ml) pemoline vehicle 0 50 100 150 200 250day 4 8p.m.CORT (ng/ml)A C B Figure 5 3. Effects of pemoline or vehicle injections on plasma corticosterone. Two hours following the first injection, corticosterone levels are elevated in the pemolinetreated rats, though these levels are later blunted during the peak of the daily circadian cycle (A). Circulating corticosterone levels at the nadir of the daily cycle are also slightly raised in pemolinetreated rats, compared to vehicle treated rats (B). Circadian rhythmicity is reestablished by the fourth day of injections as there are no longer differences between peak corticosterone levels in pemolineand vehicletreated rats. All values are expressed as group means SEM.
115 Monoamine, Metabolite, and Amino Acid Analyses Data from the groups that were killed 12 hours after the injections on days 1 and 4 were left out of the gr aphs and analyses for the monoamine, metabolite, and amino acid measures. To focus specifically on stable pemolineinduced changes in these measures, we focused on the samples that were taken at approximately the same time each morning (810am). Striatum The striatum consists mostly of GABAergic medium spiny neurons that receive dopaminergic and glutamatergic inputs. In this experiment we found that as pemolineinduced self injury developed across days, intracellular dopamine levels in the striatum began to deplete (F(1,33) = 4.623, p < 0.05; Fig. 5 4A) and the extracellular dopamine metabolite, HVA, increased (F(1,33) = 25.77, p < 0.0001; Fig. 54B). There were no changes in the levels of the intracellular dopamine metabolite, DOPAC (Fig. 54C). These res ults suggest that pemoline causes significant increases in extracellular dopamine in the striatum, either by blocking reuptake, increasing release, or both, which is subsequently metabolized. Pemoline did not have any significant effect on dopamine synthes is ( i.e. measures of DA+HVA and DA+DOPAC; Fig. 5 5A,B) but turnover of extracellular dopamine was significantly enhanced ( HVA/DA, F(1,33) = 44.38, p < 0.0001; Fig. 5 5C), whereas no changes in intracellular dopamine turnover were found ( DOPAC/DA; Fig. 5 5 D) We also found some reductions in serotonin levels (Fig. 5 4D), though these did not reach statistical significance. Striatal 5 H IAA was higher in pemolinetreated rats starting on day 2 and continuing throughout the experiment (F(1,33) = 27.77, p < 0.0 001; Fig. 5 4E). The synthesis of striatal serotonin was not different between the pemolineand vehicletreated groups (Fig. 5 5E), though serotonergic
116 turnover was significantly increased in the pemolinetreated group (F(1,33) = 92.31, p < 0.0001; Fig. 5 5F). Amino acids analyses revealed no differences in striatal glutamate levels between pemolineand vehicletreated rats (Fig. 56A), but significant increases in GABA levels were found (F(1,35) = 7.312, p < 0.05, Fig. 56B). striatal DA 0 1 2 3 4 5 6 0 10 20 30 40 50 60pemoline vehicle dayDA (ng/mg protein) striatal HVA 0 1 2 3 4 5 6 0 2 4 6 8 10dayHVA (ng/mg protein) striatal DOPAC 0 1 2 3 4 5 6 0 5 10 15dayDOPAC (ng/mg protein) striatal 5HIAA 0 1 2 3 4 5 6 0 1 2 3 4 5day5HIAA (ng/mg protein) striatal 5HT 0 1 2 3 4 5 6 0 1 2 3 4 5 6 7day5HT (ng/mg protein)A E D C B Figure 5 4. Effects of pem oline or vehicle injections on striatal monoamine and metabolite levels. Intracellular DA levels were reduced (A) as the extracellular
117 metabolite of DA, HVA, was increased (B) following repeated pemoline injections. No changes in the intracellular DA metabolite, DOPAC (C), or in 5 HT levels were found. Levels of the 5HT metabolite (5HIAA) was also significantly elevated (E) in pemolinetreated rats, compared to those of vehicle treated rats. All values are expressed as group means SEM. striatal DA turnover 0 1 2 3 4 5 6 0.00 0.05 0.10 0.15 0.20dayHVA/DA striatal DA turnover 0 1 2 3 4 5 6 0.0 0.1 0.2 0.3dayDOPAC/DA striatal DA synthesis 0 1 2 3 4 5 6 35 40 45 50 55 60 65pemoline vehicle dayDA + HVA striatal DA synthesis 0 1 2 3 4 5 6 0 10 20 30 40 50 60 70 80dayDA + DOPAC striatal 5HT turnover 0 1 2 3 4 5 6 0.00 0.25 0.50 0.75 1.00 1.25day5HIAA/5HT striatal 5HT synthesis 0 1 2 3 4 5 6 0 1 2 3 4 5 6 7 8 9 10day5HT + 5HIAAA F E D C B Figure 5 5. Effects of pemoline or vehicle injections on striatal monoamine synthesis and turnover ratios. There were no differences in dopamine synthesis rates in
118 pemolineor vehicletreated rats (A,B). Dopamine turnover, as quantified by HVA/DA, was significantly incr eased in pemolinetreated rats (C), whereas DOPAC/DA was not (D). Serotonin synthesis was not different between pemolineand vehicletreated groups (E), however turnover of 5HT was increased in pemolinetreated rats (F). All values are expressed as group means SEM. Figure 5 6. Effects of pemoline or vehicle injections on striatal amino acid levels. No differences in glutamate concentrations were found between pemolineand vehicle treated rats (A), however pemolinetreated rats had significantly elevat ed GABA concentrations compared to those of vehicletreated rats (B). All values are expressed as group means SEM. Ventral tegmentum We analyzed the monoamine and amino acid concentrations in the ventral tegmentum because it contains the cell bodies (in the VTA and SN) for the two main dopaminergic projections, the nigrostriatal and mesolimbic pathways. These nuclei also contain GABAergic cell bodies and receive glutamatergic projections. Our analyses revealed no significant changes in ventral tegmental norepinephrine (Fig. 57A), dopamine (Fig. 57B), serotonin (Fig. 57E), glutamate (Fig. 5 9A), or GABA (Fig. 5 9B) levels in pemolinetreated rats as compared to those of vehicletreated rats There were also no significant changes in the dopamine metabol ite concentrations (HVA, Fig. 5 7C; DOPAC, Fig. 5 7D) in pemolinetreated rats, and no differences in any measure of dopamine synthesis (Fig. 5 8A,B) or turnover (Fig. 5 8C,D). 5 HIAA levels were 0 1 2 3 4 5 6 0 250 500 750pemoline vehicle striatal glutamatedayGLU (ng/mg protein) 0 1 2 3 4 5 6 0 50 100 150 200 250striatal GABAdayGABA (ng/mg protein)A B
119 significantly elevated in pemolinetreated rats, compared to vehicle treated rats (F(1,36) = 20.87, p < 0.0001; Fig. 57F) and this corresponded to significant increases in serotonin synthesis (F(1,36) = 8.669, p < 0.01; Fig. 58E) and turnover ratios (F(1,36) = 76.28, p < 0.0001; Fig. 58F). ventral tegmental NE 0 1 2 3 4 5 6 0.0 2.5 5.0 7.5 10.0pemoline vehicle dayNE (ng/mg protein) ventral tegmental DA 0 1 2 3 4 5 6 0 1 2 3 4 5 6dayDA (ng/mg protein) ventral tegmental DOPAC 0 1 2 3 4 5 6 0.0 0.4 0.8 1.2dayDOPAC (ng/mg protein) ventral tegmental HVA 0 1 2 3 4 5 6 0.0 0.1 0.2 0.3 0.4 0.5dayHVA (ng/mg protein) ventral tegmental 5HIAA 0 1 2 3 4 5 6 7 0 5 10 15kill time5HIAA (ng/mg protein) ventral tegmental 5HT 0 1 2 3 4 5 6 0 5 10 15 20day5HT (ng/mg protein)A B C D E F Figure 5 7. Effects of pemoline or vehicle injections on ventral tegmental monoamine and metabolite levels. There were no pemolineassociated changes in ventral tegmental monoamine concentrations, including NE (A), DA (B), or 5 HT (E). There were also no differences in dopami ne metabolites, HVA (C) or DOPAC (D). Ventral tegmental 5HIAA, however, was significantly higher in pemoline
120 versus vehicletreated rats (F). All values are expressed as group means SEM. ventral tegmental DA turnover 0 1 2 3 4 5 6 0.000 0.025 0.050 0.075 0.100 0.125dayHVA / DA ventral tegmental DA synthesis 0 1 2 3 4 5 6 0 1 2 3 4 5 6dayDA + DOPAC ventral tegmental DA synthesis 0 1 2 3 4 5 6 0 1 2 3 4 5 6pemoline vehicle dayDA + HVA ventral tegmental DA turnover 0 1 2 3 4 5 6 0.0 0.1 0.2dayDOPAC / DA ventral tegmental 5HT turnover 0 1 2 3 4 5 6 0.00 0.25 0.50 0.75day5HIAA / 5HT ventral tegmental 5HT synthesis 0 1 2 3 4 5 6 0 10 20 30day5HT + 5HIAAA B C D E F Figure 5 8. Effects of pemoline or vehicle injections on ventral tegmental monoamine synthesis and turnover ratios. Pemoline did not significantly alter dopamine synthesis (A,B) or turnover (C,D). 5 HT synthesis and turnover ratios, however, were significantly increased in pemolinetreated rats compared to the ratios in vehicle treated rats. All values are expressed as group means SEM.
121 ventral tegmental glutamate 0 1 2 3 4 5 6 0 100 200 300 400 500 600pemoline vehicle dayGLU (ng/mg protein) ventral tegmental GABA 0 1 2 3 4 5 6 0 50 100 150 200 250dayGABA (ng/mg protein)A B Figure 5 9. Effects of pemoline or vehicle injections on ventral tegmental amino acid levels. No differences in either glutamate (A) or GABA (B) concentrations were found between pem oline and vehicletreated rats. All values are expressed as group means SEM. Cortices Cortical regions receive inputs from the dopaminergic mesocortical pathway, as well as from GABAergic and glutamatergic projections. The cortex also contains the glut amatergic cell body regions that project to limbic and basal ganglia structures. For these reasons we analyzed the cortical content of monoamines and amino acids. We found no differences between pemolineand vehicletreated rats in measures of cortical dopamine (Fig. 5 10A), serotonin (Fig. 510D), glutamate (Fig. 512A), or GABA (Fig. 5 12B). There were also no changes in DOPAC (Fig. 510C) or 5HIAA (Fig. 5 10E) levels in the cortical samples from pemolinetreated rats. Cortical HVA was significantly hig her in pemoline treated rats (F(1, 36) = 11.98, p < 0.01; Fig. 510B), which contributed to a significant difference in cortical HVA/DA turnover ratio (F(1,36) = 21.05, p < 0.0001; Fig. 5 11C). There were no differences in dopamine (Fig. 511A,B) or serotonin (Fig. 5 11E) synthesis ratios, or in DOPAC/DA (Fig. 5 11D) or 5 HIAA/5 HT (Fig. 5 11F) turnover ratios.
122 cortical DA 0 1 2 3 4 5 6 0 1 2 3vehicle pemoline dayDA (ng/mg protein) cortical DOPAC 0 1 2 3 4 5 6 0.00 0.25 0.50 0.75 1.00dayDOPAC (ng/mg protein) cortical HVA 0 1 2 3 4 5 6 0.0 0.1 0.2 0.3 0.4 0.5dayHVA (ng/mg protein) cortical 5HT 0 1 2 3 4 5 6 0 1 2 3 4 5day5HT (ng/mg protein) cortical 5HIAA 0 1 2 3 4 5 6 0 1 2 3day5HIAA (ng/mg protein)A E D C B Figure 5 10. Effects of pemoline or vehicle injections on cortical monoamine and metabolite levels. There were no pemolineassociated changes in c ortical DA (A) or 5 HT (D) concentrations. Cortical HVA levels were increased in pemolinetreated rats, compared to those of vehicletreated rats (B). No significant differences in DOPAC (C) or 5HIAA (E) were found. All values are expressed as group means SEM.
123 cortical DA synthesis 0 1 2 3 4 5 6 0 1 2 3 4pemoline vehicle dayDA + HVA cortical DA synthesis 0 1 2 3 4 5 6 0 1 2 3dayDA + DOPAC cortical DA turnover 0 1 2 3 4 5 6 0.00 0.05 0.10 0.15 0.20 0.25dayHVA/DA cortical DA turnover 0 1 2 3 4 5 6 0.0 0.1 0.2 0.3 0.4dayDOPAC/DA cortical 5HT synthesis 0 1 2 3 4 5 6 0.0 2.5 5.0 7.5day5HT + 5HIAA cortical 5HT turnover 0 1 2 3 4 5 6 0.00 0.25 0.50 0.75day5HIAA/5HTA F C B D E Figure 5 11. Effects of pemoline or vehicle injections on cortical monoamine synthesis and turnover ratios. There were no differences in dopamine synthesis rates between pemolineor vehicletreated rats (A,B). Dopamine turnover, as quantified by HVA/DA, was significantly increased in pemolinetreated rats (C), whereas DOPAC/DA was not (D). Neither 5HT synthesis (E) nor turnover (F) was significantly different between pemolineand vehicletreated groups. All values are expressed as group means SEM.
124 cortical glutamate 0 1 2 3 4 5 6 0 100 200 300 400pemoline vehicle dayGLU (ng/mg protein) cortical GABA 0 1 2 3 4 5 6 0 10 20 30 40 50 60 70 80dayGABA (ng/mg protein)A B Figure 5 12. Effects of pemoline or vehicle injections on cortical amino acid levels. No differences in either glutamate (A) or GABA (B) concentrations were found between pemolineand vehicletreated rats. All values are expressed as group means SEM. Amygdala The dopaminergic mesolimbic pathway projects to several limbic sites, including the amygdala. We examined dopamine, DOPAC, serotonin, and 5HIAA levels in the amygdalar samples to determine pemolines effects on this projection area of the mesolimbic pathway (HVA levels in this brain region were consistently near our lower detection limit, so they are not presented here). The amygdala also contains and receives glutamatergic and GABAergic projections, though because of technical difficulties we were unable to analyze amgdalar samples for amino acid concentrations. In these analyses we found no significant differences between pemolineand vehicletreated rats in measures of amygdalar dopamine (Fig. 513A), DOPAC (Fig. 5 13B), or serotonin (F ig. 5 13C). Pemolinetreated rats, however, did have higher 5HIAA levels, though this effect appears transient, only occurring on days 3 and 4 (F(1,36) = 9.765, p < 0.01; Fig. 5 13D). We found no changes in dopamine synthesis (Fig. 514A) or turnover (Fig 5 14B) ratios in pemolinetreated rats. Serotonin synthesis rates (Fig. 514C) were
125 also not different between groups, whereas serotonin turnover was significantly elevated in pemolinetreated rats, but mostly only on days 3 and 4 (F(1,36) = 26.11, p < 0 .0001; Fig. 514D). amygdalar DOPAC 0 1 2 3 4 5 6 0.0 0.2 0.4 0.6 0.8dayDOPAC (ng/mg protein) amygdalar DA 0 1 2 3 4 5 6 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5pemoline vehicle dayDA (ng/mg protein) amygdalar 5HIAA 0 1 2 3 4 5 6 0 2 4 6day5HIAA (ng/mg protein) amygdalar 5HT 0 1 2 3 4 5 6 0 5 10 15 20 25 30day5HT (ng/mg protein)A B C D Figure 5 13. Effects of pemoline or vehicle injections on amygdalar monoamine and metabolite levels. There were no pemolineassociated changes in amygdalar DA (A), DOPAC (B), or 5 HT (C) concentrations. 5HIAA concentrations were sign ificantly elevated in pemolinetreated rats as compared to those of vehicle treated rats. All values are expressed as group means SEM.
126 amygdalar DA synthesis 0 1 2 3 4 5 6 0 1 2 3 4 5pemoline vehicle dayDA + DOPAC amygdalar DA turnover 0 1 2 3 4 5 6 0.0 0.1 0.2 0.3 0.4 0.5dayDOPAC/DA amygdalar 5HT synthesis 0 1 2 3 4 5 6 0 10 20 30 40day5HT + 5HIAA amygdalar 5HT turnover 0 1 2 3 4 5 6 0.000 0.025 0.050 0.075 0.100 0.125 0.150 0.175day5HIAA/5HTD C B A Figure 5 14. Effects of pemoline or vehicle injections on amygdalar monoamine synthesis and turnover ratios. There were no differences in dopamine synthesis (A) or turnover (B) rates between pemolineor vehicletreated rats. 5 HT synthesis was also not changed by pemoline treatment (C). 5 HT turnover, however, was significantly higher at most time points in pemolinetreated rats versus vehicle treated rats (D). All values are expressed as group means SEM. Hippocampus We were interested in studying the effects of pemoline on the neurochemistry of the hippocampus for several reasons. The hippocampus receives dopaminergic, sero tonergic, and glutamatergic inputs from midbrain nuclei, and there is considerable evidence suggesting an important role for these neurotransmitters in self injurious behavior. Furthermore, our lab previously found changes in cellular metabolism (as assaye d by cytochrome oxidase immunohistochemistry) in hippocampal regions in
127 pemolinetreated rats (data not published). Unfortunately, we were unable to measure dopamine and its metabolites in the hippocampus with our HPLC and detection settings. Serotonin lev els were unchanged in hippocampal samples from pemolinetreated rats (Fig. 5 15A), while 5HIAA concentrations rose significantly across days of pemoline administration (F(1,36) = 15.91, p < 0.001; Fig. 515B). These results contributed to significantly higher serotonin turnover ratios (F(1,36) = 26.11, p < 0.0001; Fig. 516B), but no significant changes in serotonin synthesis (Fig. 516A). We also found no changes in hippocampal glutamate (Fig. 517A) or GABA (Fig. 5 17B) in pemolinetreated rats. hippocampal 5HT 0 1 2 3 4 5 6 0 1 2 3 4 5 6pemoline vehicle day5HT (ng/mg protein) hippocampal 5HIAA 0 1 2 3 4 5 6 0 1 2 3 4 5day5HIAA (ng/mg protein)B A Figure 5 15. Effects of pemoline or vehicle injections on hippocampal 5HT and 5HIAA. There were no pemolineassociated changes in hippocampal 5HT (A), whereas hippocampal 5HIAA was significantly increased in pemolinetreated rats (B). All values are expressed as gr oup means SEM. hippocampal 5HT synthesis 0 1 2 3 4 5 6 0 4 8 12pemoline vehicle day5HT + 5HIAA hippocampal 5HT turnover 0 1 2 3 4 5 6 0.00 0.25 0.50 0.75 1.00day5HIAA/5HTA B
128 Figure 5 16. Effects of pemoline or vehicle injections on hippocampal 5HT synthesis and turnover ratios. Pemoline did not significantly alter serotonin synthesis ratios (A), but did significantly enhance 5HT turnover (B). All values are expressed as group means SEM. hippocampal glutamate 0 1 2 3 4 5 6 0 100 200 300 400 500 600 700pemoline vehicle dayGLU (ng/mg protein) hippocampal GABA 0 1 2 3 4 5 6 0 50 100dayGABA (ng/mg protein)A B Figure 5 17. Effects of pemoline or vehicle injections on hippocampal amino acid levels. No differences in either glutamate (A) or GABA (B) concentrations were found between pemolineand vehicletreated rats. All values are expressed as group means SEM. Septum The septum is another limbic brain structure that is innervated by the dopaminergic mesolimbic pathway. It was also an area in which our previous investigation found significant reduction in cellular metabolism with repeated pemoline administration (data not published). We analyzed norepinephrine (Fig. 518A), dopamine (Fig. 518B), serotonin (Fig. 518E), and metabolites (Fig. 518C,D,F) in the septum of pemoline and vehicletreated rats and found no significant differences in any measure. There were also no differences found in dopamine and serotonin synthesis or turnover ratios (Fig. 519A F). There was not enough septal sample to also run amino acid analyses.
129 septal NE 0 1 2 3 4 5 6 0 1 2 3 4 5pemoline vehicle dayNE (ng/mg protein) septal DA 0 1 2 3 4 5 6 0 1 2 3 4 5 6 7 8dayDA (ng/mg protein) septal DOPAC 0 1 2 3 4 5 6 0.0 0.5 1.0 1.5 2.0 2.5dayDOPAC (ng/mg protein) septal HVA 0 1 2 3 4 5 6 0.0 0.1 0.2 0.3 0.4 0.5dayHVA (ng/mg protein) septal 5HT 0 1 2 3 4 5 6 0 1 2 3 4 5 6day5HT (ng/mg protein) septal 5HIAA 0 1 2 3 4 5 6 0.0 2.5 5.0 7.5 10.0day5HIAA (ng/mg protein)A B C D E F Figure 5 18. Effects of pemoline or vehicle injections on septal mon oamine and metabolite levels. There were no pemolineassociated changes in septal NE (A) DA (B), or 5 HT (E) concentrations. There were also no significant differences in HVA (C), DOPAC (D), or 5 HIAA (F) between pemolineand vehicle treated rats. All val ues are expressed as group means SEM.
130 septal DA turnover 0 1 2 3 4 5 6 0.000 0.025 0.050 0.075 0.100 0.125 0.150 0.175dayHVA/DA septal DA turnover 0 1 2 3 4 5 6 0.0 0.1 0.2 0.3 0.4 0.5 0.6dayDOPAC/DA septal DA synthesis 0 1 2 3 4 5 6 0 1 2 3 4 5 6 7 8pemoline vehicle dayDA + HVA septal DA synthesis 0 1 2 3 4 5 6 0 1 2 3 4 5 6 7 8dayDA + DOPAC septal 5HT turnover 0 1 2 3 4 5 6 0 1 2 3day5HIAA/5HT septal 5HT synthesis 0 1 2 3 4 5 6 5.0 7.5 10.0 12.5 15.0day5HT + 5HIAAA B C D E F Figure 5 19. Effects of pemoline or vehicle injections on septal monoamine synthesis and turnover ratios. Pemoline did not significantly alter dopamine (A,B,C,D) or 5 HT (E,F) synthesis or turnover ratios. All values are expressed as group means SEM.
131 The significant drug effects from this experiment are summarized in Table 52. Striatum Ventral tegmentum Cortices Amygdala Hippocampus Septum DA n.s. n.s. n.s. n/a n.s. HVA *** n.s. ** n/a n/a n.s. DOPAC n.s. n. s. n.s. n.s. n/a n.s. DA synthesis DA + HVA n.s. n.s. n.s. n/a n/a n.s. DA synthesis DA + DOPAC n.s. n.s. n.s. n.s. n/a n.s. DA turnover HVA/DA *** n.s. *** n/a n/a n.s. DA turnover DOPAC/DA n.s. n.s. n.s. n.s. n/a n.s. 5 HT n.s. n.s. n.s. n.s. n.s. n .s. 5 HIAA *** *** n.s. ** ** n.s. 5 HT synthesis 5 HT + 5 HIAA n.s. ** n.s. n.s. n.s. n.s. 5 HT turnover 5 HIAA/5 HT *** *** n.s. *** *** n.s. NE n/a n.s. n/a n/a n/a n.s. GLU n.s. n.s. n.s. n/a n.s. n/a GABA n.s. n.s. n/a n.s. n/a Table 52. Sum mary of significant effects of pemoline on regional concentrations of monoamines, metabolites, amino acids, and of monoamine synthesis and turnover ratios. Acronyms and symbols: n.s., not significant; n/a, not analyzed; p < 0.05, ** p < 0.01, ***p < 0.0001. Discussion This study was designed to document pemoline titers in the brain and plasma, to examine potential changes in corticosterone circadian regulation, and to give a global impression of changes in monoamine and amino acid neurotransmission during repeated administration of pemoline. We harvested samples three times within 24 hours of the first injection, then 24 hours after each subsequent injection. We also took samples 12 hours after the fourth injection to assess whether changes during the first day of pemoline administration remained consistent during the subsequent injections. The data from the pemoline titer analyses reveal that pemoline is present in both plasma and brain throughout the six day experimental protocol. As seen from the
132 sam ples taken 24 hours after each injection (Fig. 52B), pemoline is never fully cleared from the plasma or brain when the subsequent injection is given. This long lasting bioavailability of pemoline is likely due to suspension of pemoline in peanut oil and i ts depot delivery under the skin. Without knowing the pharmacodynamics of pemoline or its dopamine transporter occupancy thresholds, it is difficult to deduce whether pemoline administration in this model constitutes intermittent or continuous stimulation of the majority of presynaptic transporter target(s) or the preor post synaptic receptors consequently activated by enhanced neurotransmission. For most stimulants, cellular and behavioral responsiveness differs depending on whether stimulation is inter mittent versus continuous (Post, 1980). The increased incidence, duration, and severity of pemolineinduced self injurious behavior across days of pemoline administration suggest that this altered behavioral responsivity exemplifies behavioral sensitization, though the role for pemoline accumulation in the expression of pemolineinduced self injury cannot be ruled out. Analyses of corticosterone levels using radioimmunoassay indicate very low levels of corticosterone in the samples collected from vehiclet reated rats in the morning hours (8 10am, roughly 1 to 3 hours after lights on). This is consistent with the normal circadian rhythm of corticosterone, which reaches its nadir at the beginning of the inactive period (Seggie et al., 1985). In pemolinetreat ed rats, corticosterone levels were elevated at most morning testing times, the greatest of which was revealed two hours after the first pemoline injection. Corticosterone levels were also elevated 24 hours after the first, second, third, and fourth inject ions, which suggests that pemoline causes marked changes to the circulating corticosterone levels throughout the day. The
133 consequence of repeated pemoline injection does not appear to be additive across days, which may indicate that the elevation in cortic osterone may not be a primary contributor to the development of pemolineinduced self injury. Though our results from other experiments (e.g. chapters 3 and 4) indicate that stress, and perhaps corticosterone and its intracellular actions, may play a modul atory role in the expression of pemolineinduced self injurious behavior. Other psychostimulants also cause an elevation in circulating corticosterone (Moldow and Fischman, 1987), and blocking this by adrenalectomy or pharmacological challenge, reduces the psychostimulant induced dopamine release and behavioral sensitization (Barrot et al., 2000; De Vries et al., 1996, Rivet et al., 1989). These results suggest that blocking the corticosterone elevation after pemoline injection may also reduce pemolineinduced self injurious behavior. Disregulation of dopaminergic systems has been the most common finding in studies regarding the neurobiological basis of human (Brucke et al., 1987; Ernst et al., 1996; Lloyd et al., 1981; Muller Vahl et al., 2000; Rassin et al., 1982; Riederer et al., 1986; Saito et al., 1999; Singer et al., 1991; Wenk, 1995; Wolf et al., 1996; Wong et al., 1996; Wong et al., 1997) and animal (Criswell et al., 1992; Cromwell et al., 1997; Kasim and Jinnah, 2003; Lewis et al., 1990; Martin et al., 1991; Mueller and Nyhan, 1982; Shishido et al., 2000; Sivam, 1989) self injurious behavior. In this experiment we found the significant changes in intracellular dopamine concentrations were only in the striatum. Reduction of intracellular dopamine corr esponds with elevations in pemoline titers in the striatum and the induction of pemolineinduced self injury. This dopamine effect is consistent with clinical findings of dopaminergic depletion in the striatum.
134 Future analyses of the neurobiological basis of self injury should continue to focus on the striatum and the direct and indirect basal ganglia pathways within it. Intracellular dopamine depletion has also been used as a preliminary indicator of dopamine terminal loss (Cadet et al., 1994). We did not directly study the status of the striatal dopaminergic terminals (analyses of this would include assays of tyrosine hydroxylase levels and autoradiographic assays of DAT number and binding characteristics), however loss of dopaminergic terminals in the caudate and putamen have been reported in LeschNyhan and Rett patients ( Wong et al., 1996; Wong et al., 1998) as well as several other animal models of self injurious behavior (Martin et al., 1991; Kita et al., 2000). A corresponding rise in HVA levels wer e also found in the striatum. Dopamine is metabolized into HVA by two distinct enzymatic pathways (Kopin, 1985). The first utilizes catechol O methyl transferase (COMT), a nonneuronal and/or extracellular enzyme, to convert dopamine into 3methoxytyramine (3MT), then monoamine oxidase (MAO) converts 3MT to HVA. MAO is an intracellular enzyme located on the membrane of mitochondria (Westlund et al., 1993). The other pathway uses MAO to convert dopamine to DOPAC, which is converted to HVA by COMT. The lack of differences in DOPAC levels between pemolineand vehicletreated rats suggests that the increase in HVA is a result of higher levels of extracellular dopamine metabolized by the nonneuronal and/or extracellular COMT in the pemolinetreated rats. This finding should be confirmed using microdialysis of the striatum, but is consistent with the action of pemoline at the dopamine transporter.
135 GABA levels in the striatum were elevated in pemolinetreated rats, in comparison to vehicletreated rats. This findin g, and the role of striatal GABAergic functioning in pemoline induced self injury, are difficult to interpret for several reasons. First, the differences in GABA levels between pemolinetreated and vehicletreated rats are most striking on days 24, which does not correspond to the days in which pemolineinduced self injurious behavior was most severe. Second, there is considerable variability in the GABA measurements in the pemolinetreated rats, which may indicate these differences represent changes in only a few samples. Third, both the direct and indirect pathway cells of the striatum are GABAergic, so this analysis does not give any insight into how the release of GABA in either pathway may change with pemoline administration. Although difficult to interpret, these results further support our rationale for focusing on striatal function in the pemoline model. Finally, significant changes in 5HIAA were found in striatum, VTA, amygdala, and hippocampus, which contributed to significant differences in serotonin turnover. Given that there are no significant increases in 5HIAA when serotonin transporters are blocked by SSRIs (Chertkow et al., 2007) or when they are knocked down in genetic models (Fox et al., 2008), suggests that pemolines main effects on the serotonin system is not through its actions on the serotonin transporter, but that the changes in serotonin turnover (as measured by the 5HIAA/5HT ratio, Karstaedt et al., 1994) may be a result of potential indirect actions and modulation by dopaminergic systems (Sivam, 1995). These increases in 5HIAA levels in pemoline treated rats are also consistent with increased 5HIAA levels in brain (Lloyd et al., 1981) and CSF (Jankovic et al., 1988) of LeschNyhan patients.
136 This project had a very wide focus and was designed to identify the time course and the most critical regions of the brain for pemolineinduced changes. We were also interested in pemoline levels in plasma and brain and the effect of pemoline on circulating corticosterone levels. Results from this experiment suggest that our six day experimental protocol is sufficient for studying changes in pemolineinduced behaviors and neurochemistry. These results also justify our continued concentration on striatal functioning in pemoline induced self injurious behavior.
137 CHAPTER 6 NEUROTENSIN PLAYS A MODULATORY ROLE IN P EMOLINE INDUCED SELF INJURIOUS BEHAVIOR Background The neurobiological basis of pemolineinduced self injurious behavior appears to be mediated by several neurochemical syst ems within the corticobasal ganglia pathway. Our previous work suggests that dopaminergic signaling is increased in the striatum (which leads to a depletion of intracellular dopamine; Chapter 4) and that glutamatergic neurotransmission is required for the induction of pemolineinduced self injurious behavior (Muehlmann et al., 2008). We have also found that a history of stressful experience worsens the severity of pemolineinduced self injury (Chapter 3) and that pemoline causes modest increases in circulating corticosterone levels (Chapter 4). Additionally, pain sensitivity is diminished in pemolinetreated rats (Chapter 2). Studying how these separate neurochemical systems interact, and finding pharmacological tools that modulate these systems, may signif icantly improve our understanding of the neurobiological basis of self injury and its treatment. Neurotensin is a modulatory neurotransmitter that binds to three different receptors. The two G protein coupled receptors, NTS1 and NTS2, predominantly control the central nervous system effects of neurotensin. NTS1 is a high affinity receptor localized in high abundance in the VTA, substantia nigra, striatum, and nucleus accumbens, as well as other limbic and cortical areas (Binder et al., 2001), and is most cl osely co localized with dopamine and glutamate systems. NTS2 is a low affinity receptor found mainly in cortical and limbic regions, as well as painrelated areas like the periaqueductal gray and superior colliculus (Asselin et al., 2001). The NTS3 is a si ngle transmembrane domain receptor that is not associated with a G protein. It is also
138 found in many cortical and limbic regions, though the functioning of this receptor in the CNS when neurotensin is bound is not well understood (Sarret et al., 2003). Neu rotensin strongly colocalizes with dopaminergic pathways. Neurotensin st imulates dopaminergic neurons in the cell body regions of these pathways, the VTA and SNc (Myers and Lee, 1983; Binder et al., 2001). Neurotensin receptors (namely NTS1) complex with both preand post synaptic D2 receptors in the striatum and nucleus accumbens (Antonelli et al., 2007; Boudin et al., 1996; Delle Donna et al., 1996; Fuxe et al., 1992) and neurotensin release in these axon terminal regions causes D2 autoreceptor inhibiti on and increased dopamine release (Fawaz et al., 2009). Consequently, neurotensin administration increases dopamine metabolite concentrations in these dopaminergic projection areas (Widerlov et al., 1982). Postsynaptically, neurotensin activation of NTS1 also antagonizes D2 receptor function, but in addition stimulates the phospholipase 3/IP3/calcium mobilization pathway (Yin et al., 2008). Neurotensin receptors are also located on glutamatergic terminals of corticostriatal neurons (Goedert et al., 1984). T he activation of these presynaptic neurotensin receptors increases glutamate release (Ferraro et al., 1995; Ferraro et al., 2000). Stimulation of the postsynaptic neurotensin receptors of the striatum amplifies the signaling of glutamates NMDA receptor ( Antonelli et al., 2004). Additionally, glutamate agonists increase neurotensin release, suggesting significant cross talk between these two neurochemical systems (Radke et al., 2001). Reciprocal relationships have also been identified between neurotensin and the HPA axis. Stressors, such as footshock, initiate the circulation of corticosterone as well
139 as increase neurotensin content in the VTA (Deutsch et al., 1987; Kilts et al., 1992). The promoter region of the neurotensin/neuromedin N gene contains a glucocorticoid response element (Harrison et al., 1995), which suggests a relationship between stress pathway activation and neurotensin/neuromedin N gene transcription and neurotensinergic tone. This suggests that an increase in neurotensin content in the VT A may be due to increased synthesis, instead of decreased release. Moreover, intracerebroventricular injections of neurotensin increase circulating corticosterone concentrations (Gudelsky et al., 1989) and NTS1 receptor antagonism blocks diurnal elevations as well as restraint stress induced corticosterone release (Rowe et al., 1997). Neurotensin is also involved in pain processing; specifically, nonopioid mediated forms of analgesia. Intracisternal injection of neurotensin significantly increases the la tency to respond to a hot plate, and decreases writhing in the acetic acid test (Clineschmidt & McGuffin, 1977). These effects are not blocked by the opioid antagonist, naloxone (Clineschmidt et al., 1979). Nonopioid mediated analgesia is also responsible for stressinduced antinociception that occurs with high intensity stressors (Grisel et al., 1993). Neurotensin knockout mice do not exhibit stress induced analgesia (Gui et al., 2004), and the NTS2 most likely mediates this form of analgesia, considering NTS2 knockouts show lower stress induced corticosterone release and decreased stressinduced analgesia (Lafrance et al., 2010). The localization of the neurotensin systems, which closely parallel the disregulated systems involved in pemolineinduced self injurious behavior, suggests that neurotensin may enhance the progressive disregulation of the dopaminergic and glutamatergic systems that we think are involved in the etiology of SIB. To evaluate this
140 possibility, we compared neurotensin levels in brain regions containing dopaminergic cell bodies and terminal fields of pemolineand vehicletreated rats. We also evaluated the effects of a neurotensin agonist and antagonist on pemolineinduced self injury and their effects on the concentrations of presynaptic monoamines and metabolites. Methods This project consisted of three experiments. The first experiment assessed neurotensin concentrations in the striatum and tegmentum in pemolineand vehicletreated rats following six daily injections. The second experiment evaluated the efficacy of a NTS1 antagonist (SR48692) to reduce pemolineinduced self injury and its effects on striatal monoamine and metabolite concentrations. The third experiment tested the effects of a NTS1 agonist (PD149163) on pemolineinduc ed self injury and striatal monoamines and metabolites. Experiment 1: Neurotensin Concentrations in Vehicle and Pemoline Treated Rats Animals Twenty male LE rats weighing 200225g upon delivery (Charles River Laboratories, Raleigh, NC) were housed in a c limate controlled vivarium with a 12h/12h light/dark schedule (lights on at 7:00 a.m.). Standard laboratory rat chow (Lab Diet 5001) and tap water were available ad libitum The rats were pair housed in standard polycarbonate cages (43 cm x 21.5 cm x 25.5 cm) during six days of acclimation to the housing facility All of the experimental procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals, and were pre approved by the Institutional Animal Care and Use Committee at the University of Florida. One rat was excluded from the analyses because it exhibited substantial denuding on it s forepaws when administered vehicle.
141 Drug Pemoline (2amino 5 phenyl 1,3 oxazol 4 one; Spectrum Chemicals, New Brunswick, New Jersey) was suspended at a concentration of 50 mg/ml in peanut oil. The pe moline was left stirring in peanut oil over night in order to achieve an even suspension. Drug treatment Following six days of acclimation, stress responsiveness was assessed as described in Chapter 5. HR and LR rats were balanced between the vehicle and pemoline treatment groups. The rats were weighed and injected with peanut oil vehicle (n=9) or pemoline (n=10) at 150 mg/kg (s.c.) each day for five days. These injections were administered at the nape of the neck and either flank on a rotating basis. Assay of self injury The rats were visually inspected each time they were injected and immediately before they were decapitated, and the inspections were videotaped. During these inspections, each rat was held in front of a video camera and the head, forepaws, hindpaws, ventrum and tail were displayed. The presence or absence of tissue injury (denuded skin, erythema, edema, or open lesi on) was noted for each rat. Still images of the injured tissue wer e taken from the videotapes, and MCID software (Imaging Research Inc., St. Catha rines, ON, Canada) was used to draw outlines around the injured tissue, and to calculate the area of injury in mm2. Neurotensin r adioimmunoassay The rats were rapidly decapitat ed 24 hours after the last injection. Brains were immediately removed and the striatum and ventral tegmentum were dissected, frozen on dry ice, and stored at 80C until processed. The neurotensin radioimmunoassay
142 was purchased from Phoenix Pharmaceuticals Inc (Burlingame, CA) and performed according to their instructions. Statistical a nalyses Betweengroup differences in size of injured tissue were evaluated using RM ANOVA. Two tailed t tests were used to analyze betweengroup differences in neurotensin content in the striatum and tegmentum. Experiment 2: Effects of NTS1 Antagonist, SR48692, on PemolineInduced Self Injurious Behavior Animals Thirty three male LE rats weighing 200225g upon delivery (Charles River Laboratories, Raleigh, NC) were housed i dentically to the ones described for Experiment 1. Drugs Pemoline was prepared as described in Experiment 1. SR48692, a NTS1 antagonist was provided by the National Institute on Mental Health Chemical Synthesis and Drug Supply Program. SR48692 was suspen ded in 0.9% saline with 0.1% Tween 80. Drug treatments The rats were weighed and injected with either peanut oil vehicle or pemoline (150 mg/kg s.c.) at approximately 8:00 a.m. on each of five consecutive days. These injections were administered at the nape of the neck and either flank on a rotating basis. The rats were also injected twice daily with either 0.1% Tween 80 vehicle or SR48692 (1.0 mg/kg s.c.). These injections were administered at approximately 8:00 a.m. (immediately after the pemoline injec tion) and approximately 6:00 p.m and also rotated
143 between the nape of the neck and either flank always administering the SR48692 or Tween 80 vehicle injections at a different site from the pemoline or peanut oil injections on any given day The drug and v ehicle groups were as follows: pemoline + SR48692 (n=11), pemoline + 0.1% Tween 80 (n=10), peanut oil + SR48692 (n=6), and peanut oil + 0.1% Tween 80 (n=6). Rats were prescreened for stress responsiveness and balanced across groups. Assays of self injury T he rats were visually inspected each time they were injected and immediately before they were decapitated and the inspections were videotaped. During these inspections, each rat was held in front of a video camera and the head, forepaws, hindpaws, ventrum and tail were displayed. The presence or absence of tissue injury (denuded skin, erythema, edema, or open lesi on) was noted for each rat. Still images of the injured tissue were taken from the videotapes, and MCID software (Imaging Research Inc., St. Catha rines, ON, Canada) was used to draw outlines around the injured tissue, and to calculate the area of injury in mm2. Night vision cameras were focused on the cages of the rats (one camera per cage), and 5minute time samples were recorded once every three hours over the entire day. The duration of self injurious oral contact was quantified during each videotaped interval by a trained observer. Self injurious oral contact was defined as all oral contact that stayed fixed on any one body part for longer than 2 seconds. This was differentiated from grooming, which is oral contact with any part of the body that continues to move from site to site on the body (e.g. oral contact with the forepaws, that then moves up each forelimb and continues to the ventrum, and t he contact is not sustained at any spot on the body for longer than 2 seconds). T he duration (in seconds)
144 of oral contact behavior was summed over the entire day (i.e. from the eight video samples) and divided by the total number of seconds recorded, 2400 seconds. Monoamine analyses Monoamine concentrations, metabolite concentrations, and protein content were analyzed as described in Chapter 5, using the same HPLC system and electrochemical detector. A fresh set of monoamine and metabolite standards were made and concentration values were interpolated using that new standard curve. Concentration values were then corrected for protein content of each sample. The biosynthesis ratios of the monoamine were analyzed by calculating DA+DOPAC, DA+HVA, 5HT+5HIAA; tu rnover ratios of the monoamines were analyzed by calculating DOPAC/DA, HVA/DA, 5 HIAA/5 HT (Commissiong, 1985; Miura et al., 2002). Statistical analyses Betweengroups differences in size of injured tissue and duration of self injurious oral contact were each evaluated using RM ANOVA. Monoamines, metabolites, monoamine synthesis ratios, and monoamine turnover ratios were analyzed by oneway ANOVAs. Betweengroups differences in all dependent measures were treated as statistically reliable when the pvalues were less than 0.05. Pre planned Fishers LSD post tests were used to further analyze betweengroups differences between pemoline + vehicle and pemoline + SR48692 groups, when RM ANOVAs revealed a significant effect.
145 Experiment 3 : Effects of NTS1 A gonist, PD149163, on Pemoline Induced Self Injurious Behavior Animals Twenty six male LE rats weighing 200225g upon delivery (Charles River Laboratories, Raleigh, NC) were housed identically to the ones described for Experiments 1 and 2. Drugs Pemoline was prepared as described in Experiment 1. PD149163, a NTS1 agonist was provided by the National Institute on Mental Health Chemical Synthesis and Drug Supply Program. PD149163 was dissolved in saline. Drug treatments The rats were weighed and injected with either peanut oil vehicle or pemoline (150 mg/kg s.c.) at approximately 8:00 a.m. on each of five consecutive days. These injections were administered at the nape of the neck and either flank on a rotating basis. The rats were also injected twice daily with either saline or PD149163 (0.01 mg/kg i.p.). These injections were administered at approximately 8:00 a.m. (immediately after the pemoline injection) and approximately 6:00 p.m. The drug and vehicle groups were as follows: pemoline + PD149163 (n=8), pemoline + saline (n=8), peanut oil + PD149163 (n=5), and peanut oil + saline (n=5). Rats were prescreened for stress responsiveness and balanced across groups. Assays of self injury, monoamines, metabolites, and protein content Measures of size of injured tissue, s elf injurious oral contact, monoamines, metabolites, and protein content were as described above for Experiment 2.
146 Statistical analyses Betweengroups differences were analyzed as described above for Experiment 2. Results Experiment 1 Rats administered five daily injections of pemoline at 150mg/kg/day exhibited significant self injury. There were significant main effects of drug (F(1,170) = 37.58, p<0.0001), time (F(10,170) = 19.93, p<0.0001), and a significant drug x time interaction (F(10,170) = 19.91, p<0.0001; Fig. 61A). Significantly higher concentrations of neurotensin were found in the striatum of pemolinetreated rats as compared to vehicletreated controls using RIA (Fig. 61B; t(17)=2.385, p<0.05). There were no significant differences in neurotensin levels in the tegmentum of pemolineand vehicletreated rats (Fig. 6 1C).
147 striatum pemoline vehicle 0 2 4 6 8 10 12*neurotensin (pg/mg protein) tegmentum pemoline vehicle 0 10 20 30 40 50neurotensin (pg/mg protein) 0 1 2 3 4 5 6 0 50 100 150 200daysize of injured tissue (mm2)C A B Figure 6 1. Pemoline induced self injury and neurotensin concentrations in discrete brain regions. Five daily pemoline injections produced self induced injury that peaked on day 4 and stayed at an asymptote until the end of the experiment on day 6 (A). Neurotensin levels within the striatum were significantly elevated after five daily pemoline injections (B), but were not significantly changed in the tegmentum (C), as compared to neurotensin levels of vehicletreated rats. All values are expressed as group means SEM. Experiment 2 The NTS1 antagonist, SR48692, significantly reduced the severity of pemolineinduced self injurious behavior, as measured by size of injured tissue and self injurious oral contact (Fig. 62A,B). For the size of injured tissue measure, the RM ANOVA revealed significant main effects of drug (F(3,290) = 6.536; p<0.01) and time (F(10,290) = 11.26, p<0.0001), and a significant drug x time interaction effect ( F(30,290) = 4.875, p < 0.0001). For the self injurious oral contact measure, an analysis using RM ANOVA also found significant main effects of drug (F(3,116) = 30.54, p<0.0001) and time (F(4,116) =
148 26.87, p<0.0001), as well as a significant drug x time int eraction effect (F(12,116) = 9.770, p<0.0001). There were very few significant betweengroups differences in the measures of monoamines, metabolites, monoamine biosynthesis ratios, and monoamine turnover ratios. The oneway ANOVA did not find any signific ant differences between groups for dopamine concentrations (F(3,29) = 2.375, p=0.091); however, striatal dopamine content was lower in the pemoline + vehicle group, compared to the peanut oil + vehicle group (t(14) = 2.247, p<0.05), which is similar to our findings of dopamine depletion in Chapter 5 (Fig. 6 3A). Interestingly, the pemoline + SR48692 group had similar dopamine concentrations to that of the vehicle groups, but the overall ANOVA was not significant. Levels of HVA did differ between the groups as revealed by the oneway ANOVA (F(3,29) = 9.764, p=0.0001). Both pemolinetreated groups had significantly higher levels of HVA compared to those of the vehicletreated groups, and LSD posttests revealed that the pemoline + SR48692 groups had even higher levels of HVA than did the pemoline + vehicle group (Fig. 63B). There were no significant betweengroups differences for DOPAC levels (Fig. 6 3C). Significant between groups differences were found for serotonin levels (F(3,29) = 3.305, p<0.05). Serotonin levels in the pemoline + vehicle group were lower than the peanut oil + vehicle group, which is consistent with our findings in Chapter 5. Furthermore, LSD posttests revealed a significant difference between the pemoline + vehicle group and the pemoline + SR48692 group, wherein the serotonin concentrations of the latter group were significantly higher and more similar to the peanut oil control groups (Fig. 63D). There were no significant betweengroups differences for 5HIAA levels, which is inconsistent w ith our findings from Chapter 5 (Fig.
149 6 3E), although the trend in these data does match the trend in the Chapter 5 data, where 5HIAA content was higher in the pemolinetreated rats than in the vehicle controls. Pemoline and SR48692 administration had more significant effects on monoamine turnover rates, than they had on monoamine biosynthesis ratios. Neither measure of dopamine biosynthesis, DA+HVA or DA+DOPAC, reached statistical significance using oneway ANOVA; though in both measures the pemoline + S R48692 group more closely resembled the peanut oil controls groups than they did the pemoline + vehicle group. Oneway ANOVAs revealed significant differences in both measures of dopamine turnover, HVA/DA (F(3,29) = 33.60, p<0.001) and DOPAC/DA (F(3,29) = 5.968, p<0.01), though these differences were between pemolinetreated groups and the peanut oil treated groups, with no significant effect of SR48692 administration. Our HPLC analysis in Chapter 4 also found a significant difference in HVA/DA in pemoline versus vehicle, though DOPAC/DA was not significant in this analysis. The data on day 6 of that experiment clearly show that the DOPAC/DA measures are higher in the pemoline group, compared to that of the vehicle group, whereas they are not different at any other time point. The monoamine measures in this experiment are only taken on day 6. This suggests that effects on dopamine turnover, which involve intracellular dopamine metabolism into DOPAC, may only occur after several days of pemoline administration One way ANOVAs did not find statistical differences in the measures of serotonin biosynthesis or turnover, though both measures approached significance (p=0.057 and p=0.066, respectively). The lack of differences in the measure of serotonin biosynthesis is consistent with our previous findings (see Chapter 4), though
150 the measure of serotonin turnover ratio was significantly higher in pemolinetreated rats in our previous experiment. In this experiment, serotonin turnover ratios were higher in pemoline + v ehicle rats than they were in peanut oil + vehicle rats, and pemoline + SR48692 rats were intermediate between the two groups. 0 1 2 3 4 5 6 0 50 100 150 200 250Pem + Veh Pem + 1.0 SR * * *Oil + veh Oil + 1.0 SR daysize of injured tissue (mm2) 0 1 2 3 4 5 0 20 40 60 80 100* * *dayself-injurious oral contact (% of total time recorded)A B Figure 6 2. Effects of the NTS1 antagonist, SR48692, on pemolineinduced self injurious behavior. SR48692 significantly lessened the size of injured tissue and time spent self injuring in pemolinetreated rats. dopamine Pem + veh Pem + SR Oil + veh Oil + SR 0 20 40 60DA (ng/mg protein) HVA Pem + veh Pem + SR Oil + veh Oil + SR 0 5 10 15 * *HVA (ng/mg protein) DOPAC Pem + veh Pem + SR Oil + veh Oil + SR 0 5 10 15 20DOPAC (ng/mg protein) serotonin Pem + veh Pem + SR Oil + veh Oil + SR 0 1 2 3 4 *5HT (ng/mg protein) 5HIAA Pem + veh Pem + SR Oil + veh Oil + SR 0 1 2 3 45HIAA (ng/mg protein)A B C D E Figure 6 3. Effects of pemoline and SR48692 on striatal monoamines and metabolites. No significant between groups differences were found for measures of dopamine (A), DOPAC (C), or 5 HIAA (E). Levels of HVA were significantly
151 increased in both pemolinetreated groups, and SR48692 further enhanced HVA levels relative to the pemoline + vehicle group (B). Serotonin levels were also different between groups (D). The pemoline + vehicle group had significantly lower serotonin levels, compared to the pemoline + SR48692 group, which more closely resembled those of the peanut oil control groups. Pem + veh Pem + SR Oil + veh Oil + SR 0 10 20 30 40 50 60DA + HVA Pem + veh Pem + SR Oil + veh Oil + SR 0 10 20 30 40 50 60 70DA + DOPAC Pem + veh Pem + SR Oil + veh Oil + SR 0 1 2 3 4 5 65HT + 5HIAA Pem + veh Pem + SR Oil + veh Oil + SR 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 *HVA/DA Pem + veh Pem + SR Oil + veh Oil + SR 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 *DOPAC/DA Pem + veh Pem + SR Oil + veh Oil + SR 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.15HIAA/5HTA B C D E F Figure 6 4. Effects of pemoline and SR48692 on striatal monoamine biosynthesis and turnover ratios. There were no significant betweengroups differences on measures of monoamine biosyntheses, DA+HVA (A), DA+DOPAC (B), or 5HT+5HIAA (E). Significant betweengroups differences were found for both measures of dopamine turnover ratios, HVA/DA (C) and DOPAC/DA (D), though these differences were between pemoline and peanut oil groups, and
152 not due to SR48692induced effects. No betweengroups differences were found for serotonin turnover ratio, 5HIAA/5HT (F). Experiment 3 The NTS1 agonist, PD149163, significantly incr eased the size of injured tissue and the duration of self injurious oral contact in pemoline treated rats (Fig. 65A, B). For the size of injured tissue measure, a RM ANOVA revealed a significant effect of time (F(10,220) = 6.124, p<0.0001) and a significant drug x time interaction (F(30,220) = 2.280, p<0.001), as pemoline + PD149163treated rats began to exhibit pemolineinduced self injury earlier and consistently showed greater injury size throughout the experiment. For the duration of self injurious oral contact measure, significant effects of drug (F(3,88) = 23.71, p<0.0001), time (F(4,88) = 16.47, p < 0.0001), and drug x time interaction (F(12,88) = 6.312, p<0.0001) were found. Pemoline + PD149163treated rats were the first to exhibit considerable amounts of self injurious oral contact, and rates stayed elevated compared to pemoline + vehicletreated rats on days 24. Durations of self injurious oral contact were no different between these groups on day 5. Pemoline and PD149163 caused significant changes in monoamine metabolites, without significantly affecting dopamine (Fig. 66A) or serotonin (Fig. 66D) levels. Striatal d opamine content s were lower in pemoline + vehicle treated rats, compared to those of peanut oil + vehicletreated rats (t(11) = 3 .274, p<0.01), which is consistent with our findings from Experiment 2 and Chapter 5. The pemoline + PD149163 group had dopamine contents intermediate between these two groups, which is similar to the effects we found in pemoline + SR48692 rats. There were no betweengroups differences in serotonin levels, which is also consistent with our findings from Chapter 4. Significant betweengroups differences in HVA levels were revealed by oneway
153 ANOVA (Fig. 6 6B). Both pemolinetreated groups had higher concentr ations of HVA than the peanut oil treated groups did (F(3,22) = 11.57, p<0.0001), and the pemoline + PD149163 group had even greater HVA levels than the pemoline + saline group. These results are also consistent with our findings from Experiment 2 and Chapter 4. DOPAC levels were also significantly different between groups (F(3,22) = 3.894, p<0.05; Fig. 66C), however the pemoline + PD149163 group was the only group different from the other drug treated groups; the DOPAC levels in this group were significantly elevated. One way ANOVA revealed significant differences in 5HIAA levels, though this effect was due to a consistent elevation in 5HIAA levels in the pemolinetreated groups compared to those of the peanut oil treated control groups (Fig. 66E). This effect is also consistent with our findings in Chapter 4. Also similar to the results of Experiment 2, PD149163 administration did not have significant effects on monoamine biosynthesis ratios, but did significantly alter all measures of monoamine turnover rates. Both measures of dopamine biosynthesis, DA+HVA and DA+DOPAC, failed to show betweengroups differences (Fig. 67A, B). The measure of serotonin biosynthesis, 5HIAA+5HT, was also not different between groups (Fig. 67E). One way ANOVAs revealed sig nificant betweengroups differences for both measures of dopamine turnover rates, HVA/DA (F(3,25) = 25.90, p<0.0001) and DOPAC/DA (F(3,25) = 7.334, p<0.01), though these differences were each influenced by the effects of pemoline, as both pemoline treated groups had higher dopamine turnover rates than the peanut oil control groups did. There were no differences in either measure between pemoline + saline and pemoline + PD149163 groups. A similar effect was found in the serotonin turnover measure, 5HIAA/5HT, wherein a significant
154 betweengroups difference was found (F(3,25) = 9.331, p<0.001; Fig. 67F), though these differences were guided mainly by the pemolinetreated groups, and not by PD149163 administration. 0 1 2 3 4 5 6 0 50 100 150 200 250Pem + veh Pem + 0.01 PD Oil + saline Oil + 0.01 PD * *daysize of injured tissue (mm2) 0 1 2 3 4 5 0 20 40 60 80 100* *dayself-injurious oral contact (% of total time recorded) Figure 6 5. Effects of the NTS1 agonist, PD149163, on pemolineinduced self injurious behavior. PD149163 significantly increased the size of injured tissue and time spent self injuring in pemoline treated rats. Pem + sal Pem + PD Oil + sal Oil + PD 0 20 40 60dopamineDA (ng/mg protein) Pem + sal Pem + PD Oil + sal Oil + PD 0 5 10 15HVA *HVA (ng/mg protein) Pem + sal Pem + PD Oil + sal Oil + PD 0 5 10 15 20DOPAC *DOPAC (ng/mg protein) Pem + sal Pem + PD Oil + sal Oil + PD 0 1 2 3 4serotonin5HT (ng/mg protein) Pem + sal Pem + PD Oil + sal Oil + PD 0 1 2 3 45HIAA5HIAA (ng/mg protein)A B C D E Figure 6 6. Effects of pemoline and PD149163 on striatal monoamines and metabolites. No significant betweengroups differences were found for measures of dopamine (A), serotonin (D), or 5HIAA (E). Levels of HVA were significantly increased in both pemolinetreated groups, and PD149163 further enhanced HVA levels relative to the pemoline + saline group ( B). DOPAC levels were not different between pemoline + saline and peanut oil +
155 saline (C). PD149163 increased DOPAC levels in the pemoline + PD149163 relative to the pemoline + saline group. Pem + sal Pem + PD Oil + sal Oil + PD 0 10 20 30 40 50 60DA + HVA Pem + sal Pem + PD Oil + sal Oil + PD 0 10 20 30 40 50 60DA + DOPAC Pem + sal Pem + PD Oil + sal Oil + PD 0.0 0.1 0.2 0.3 0.4 *HVA/DA Pem + sal Pem + PD Oil + sal Oil + PD 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 *DOPAC/DA Pem + sal Pem + PD Oil + sal Oil + PD 0 1 2 3 4 5 65HT + 5HIAA Pem + sal Pem + PD Oil + sal Oil + PD 0.0 0.5 1.0 1.5 2.0 2.5 *5HIAA/5HTA B C D E F Figure 6 7. Effects of pemoline and PD149163 on striatal monoamine bi osynthesis and turnover ratios. There were no significant betweengroups differences on measures of monoamine biosyntheses, DA+HVA (A), DA+DOPAC (B), or 5HT+5HIAA (E). Significant betweengroups differences were found for both measures of dopamine turnover ratios, HVA/DA (C) and DOPAC/DA (D), though these differences were between pemoline and peanut oil groups, and not due to PD149163induced effects. A similar pattern of betweengroups differences were also found for serotonin turnover ratio, 5HIAA/5HT (F)
156 Discussion In this study we found that repeated pemoline administration increases neurotensin content in dopaminergic axon terminal regions and that an NTS1 antagonist reduces pemolineinduced self injury, and conversely, an NTS1 agonist potentiates pemoline induced self injury. Taken together these results suggest that there is an increase in synthesis and release of neurotensin in the striatum and that by reducing the stimulation of NTS1 receptors we can reduce the severity of pemolineinduced self in jurious behavior. Findings of pemolineinduced increases in neurotensinergic signaling fit well with our previous data. Neurotensin signaling is associated with increased dopamine (Fawaz et al., 2009) and glutamate release (Ferraro et al., 1995; Ferraro et al., 2000), heightened HPA axis activation (Gudelsky et al., 1989), and reduced pain sensitivity (Clineschmidt & McGuffin, 1977). These effects are all consistent with our previous characterization of the effects of repeated pemoline and their role in pro ducing self injurious behaviors. We focused our investigations on presynaptic monoamine and metabolite concentrations because of our earlier findings of intracellular dopamine depletion and increased HVA levels in pemolinetreated rats. Additionally, NTS 1 receptors are predominantly located on presynaptic dopaminergic and glutatmatergic axon terminals. Interestingly, although the NTS1 antagonist and agonist had opposite effects on pemolineinduced self injury, they caused remarkably similar effects on presynaptic measures of monoamines and their metabolite concentrations. This suggests that the converse effects that these neurotensinergic drugs had on pemoline induced self injury were perhaps mediated by postsynaptic mechanisms that were not evaluated in t his study. Another potential explanation for the lack of differences in presynaptic markers of
157 monoamine function is the internalization of NTS1 receptors upon agonist activation. With agonist bound to NTS1 receptors, the complex rapidly internalizes, the receptor is marked for degradation, and the cell desensitizes (Mazella and Vincent, 2006; Souaze, 2001; Souaze and Forgez, 2006 ; Vandenbulcke et al., 2000). This may have the same biological effect of repeated antagonist administration, and perhaps explains the similar presynaptic effects of both the NTS1 agonist and antagonist. The effects of pemoline on striatal neurotensin content are similar to those found with other psychostimulants. Cocaine and methamphetamine each significantly increase neurotensinlike immunoreactivity in the striatum (Hanson et al., 1992). The increased neurotensin content in the striatum upon methamphetamine exposure is blocked by administration of the NMDA antagonist, MK801 (Hanson et al., 1992). Pemolineinduced self injury is s ignificantly reduced by MK801 (Muehlmann and Devine, 2008), which suggests that this effect may be partially due to an attenuation of pemolineinduced increase of neurotensin levels in the striatum. Contrary to this hypothesis, however, are the effects o f neuroleptics, such as haloperidol, pimozide, and risperidone. These drugs each significantly decrease measures of pemolineinduced self injurious behavior (Muehlmann et al., 2008; Mueller and Nyhan, 1982) but also increase neurotensin content in the stri atum (Govoni et al., 1980; Kinkead et al., 2000). Studies of these neuroleptics on striatal neurotensin content have not been completed in stimulant administered animals, so it is difficult to assess what effects they might have on neurotensin levels in st imulant induced disregulated neurons of the striatum.
158 Although we were unable to identify the particular mechanism that mediates the significant effects of the NTS1 antagonist and agonist to decrease and increase, respectively, pemolineinduced self injur ious behavior, these data suggest that neurotensin antagonists may offer pharmacotherapeutic efficacy to human self injurers. Interestingly, a recent report showed significantly higher levels of serum neurotensin in children with autism (Angelidou et al., 2010). This report did not include rates of self injurious or other repetitive behaviors, but suggests that an NTS1 antagonist, as opposed to an NTS1 agonist, may be an effective treatment for autistic like traits.
159 CHAPTER 7 GENERAL DISCUSSION Self injurious behavior is a devastating behavior disorder associated with several neurodevelopmental, neurological, and psychiatric syndromes. Our understanding of the genetic basis of vulnerability and the neurobiological mechanisms responsible for the expression of self injury has only improved marginally since the study of self injurious behavior began. The development of relevant animal models is critical for the advancement of self injury research. Using these tools, researchers are now beginni ng to understand the neurobiological basis of self injury and can begin to elucidate genetic mechanisms that confer vulnerability and to develop logical drug treatments. The pemoline model is an especially useful tool for these examinations. Work in our l aboratory has begun to demonstrate the validity of the pemoline model to represent human self injurious behavior. The pemoline model offers face, construct, and predictive validity, which are essential for any relevant model of human pathology. There are also many other characteristics of the pemoline model that make it a compelling animal model of self injurious behavior. We propose that these characteristics, which are specific to the pemoline model, will contribute significantly to the advancements in t he field of self injurious behavior research. Firstly, the development of pemolineinduced behaviors is highly patterned. Over days of pemoline administration, the pemolineinduced behaviors evolve from high rates of locomotion (day 1), to the onset of stereotyped, repetitive behaviors (e.g. head bobbing, burrowing through bedding on day 2), to high rates of self biting behavior (days 3 and 4; as demonstrated in Chapter 2). This predictable time course allows for evaluation of the neurobiological basis of t hese different behaviors. Understanding the subtle differences
160 in neurochemistry or structural biology that are associated with these different behaviors will advance our understanding of the ways in which stereotypy may be related to self injury and more broadly, how the brain controls behavior. Secondly, our finding that glutamatemediated neuroplasticity is involved in the development of pemolineinduced self injury (Muehlmann and Devine, 2008) highlights an interesting research opportunity for studying the behavioral based theories of self injurious behavior in an animal model. Over the past few decades, neuroscientists have begun to elucidate long term brain changes that are responsible for learning and memory and the pairing of operant contingencies (McKee et al., 2010). A preponderance of evidence suggests that these involve glutamatebased changes (though acetylcholine and serotonin probably also contribute) and are mediated by the specific properties of the glutamate receptors (Peng et al., 2011). No animal model to date has been able to incorporate the principles of the behavioral perspective of self injury, and as described in the Introduction, behavior analysts are the most productive researchers of human self injury. Therefore, the pemoline model is well situated to bring together behavioral and neurobiological theories of self injury to create a more cohesive understanding of the mediators of self injurious behavior. Thirdly, a predominant role for dopamine in human and pemolineinduced self in jury has been documented. This point of convergence supports our contention that the pemoline model has construct validity, but also offers a useful parallel between pemolineinduced self injury and other stimulant (i.e. indirect dopamine agonist) induced behaviors. Our work suggests that the development of pemolineinduced self injury is a behaviorally sensitized response to the stimulant properties of pemoline.
161 Researchers in the area of drug abuse have been studying the neurobiological basis of behavioral sensitization and its role in drug taking, drug craving, and relapse for several decades now. Following the research path of this area of study may allow the investigation of self injurious behavior to progress much faster than it would without this parallel line of research. Another special characteristic of the pemoline model that makes it a useful model of human self injury is the individual differences in vulnerability that are exhibited in the pemolinetreated rats. These differences allow us to anal yze the general effects of pemoline on the brain (in pemoline versus vehicle comparisons), as well as the self injury specific effects of pemoline on the brain (in comparisons of pemolinetreated self injurious versus pemolinetreated noninjurious rats). These different comparisons will allow researchers to identify self injury related brain dysfunction more specifically, which may lead to quicker advancements in our understanding of the neurobiological basis of self injury. Furthermore, weve identified a behavioral correlate to the vulnerability to develop pemolineinduced self injury (stress responsiveness in a novel environment), so we can use this information to analyze genetic vulnerability to develop pemolineinduced self injury by comparing nonpemo line treated HR and LR rats. Research on behavioral pathologies associated with neurodevelopmental disorders highlights the importance of early intervention for success in reducing problem behaviors and the ability to identify vulnerable populations early through genetic screening or behavioral testing of novelty responsiveness may enhance the probability of successful therapy. Lastly, our laboratory has recently found that pemolineinduced self injury can also be established in mice. There are many geneti c research tools that have been best
162 established in mice (e.g. gene knockouts and knock in mice). The addition of these research tools to our analysis of the pemoline model will undoubtedly enhance our understanding of self injurious behavior. Overall, the se projects have significantly increased our confidence in the utility of pemoline to model human self injury. They have also improved our understanding of the neurobiological basis of self injurious behavior and potential drug therapies to reduce this maladaptive behavior in individuals with neurodevelopmental and perhaps even neurological, and psychiatric conditions.
163 LIST OF REFERENCES Akefeldt A, Ekman R, Gillberg C, Mansson JE (1998) Cerebrospinal fluid monoamines in Prader Willi syndro me. Biol Psychiat 44: 13211328. Allen SM, Freeman JN, Davis WM (1998) Evaluation of risperidone in the neonatal 6hydroxydopamine model of LeschNyhan syndrome. Pharmacol Biochem Behav 59: 327330. Allen SM, Rice SN (1996) Risperidone antagonism of self mutilation in a LeschNyhan patient. Prog Neuropsychopharmacol Biol Psychiatry 20: 793800 Amphoux A, Vialou V, Drescher E, Bruss M, Mannoury La Cour, C, Rochat C, Millan MJ, Giros B, Bonisch H, Gautron S (2006) Differential pharmacological in vitro pro perties of organic cation transporters and regional distribution in rat brain. Neuropharmacol 50: 941952. Anderson LT, Ernst M (1994) Self injury in LeschNyhan disease. J Autism Dev Disord 24: 6781. Angelidou A, Francis K, Vasiadi M, Alysandratos K D, Zhang B, Theoharides A, Lykouras L, Sideri K, Kalogeromitros D, Theoharides TC (2010) Neurotensin is increased in serum of young children with autistic disorder. J Neuroinflammation 7: 4852. Angulo JA, McEwen BS (1994) Molecular aspects of neuropeptide regulation and function in the corpus striatum and nucleus accumbens. Brain Res Brain Res Rev 19: 128. Antonelli T, Ferraro L, Fuxe K, Finetti S, Fournier J, Tanganelli S, De Mattei M, Tomasini MC (2004) Neurotensin enhances endogenous extracellular glut amate levels in primary cultures of rat cortical neurons: involvement of neurotensin receptor in NMDA induced excitotoxicity. Cereb Cortex 14: 466473. Antonelli T, Tomasini MC, Fuxe K, Agnati LF, Tanganelli S, Ferraro L (2007) Receptor receptor interacti ons as studied with microdialysis. Focus on NTR/D2 interactions in the basal ganglia. J Neural Transm 114: 105113. Aoyama T, Kotaki H, Saitoh Y, Nakagawa F (1988) Determination of pemoline in plasma, urine and tissues by reversedphase highperformance l iquid chromatography. J Chromat 430: 351360. Asselin M L, Dubuc I, Coquerel A, Costentin J (2001) Localization of neurotensin NTS2 receptors in rat brain, using [3H]levocabastine. Neuroreport 12: 10871091.
164 Baghdadli A, Pascal C, Grisi S, Aussilloux C ( 2003) Risk factors for self injurious behaviours among 222 young children with autistic disorders. J Intell Disabl Res 47: 622627. Ball TS, Datta PC, Rios M, Constantine C (1985) Flexible arm splints in the control of a Lesch Nyhan victims finger biting and a profoundly retarded clients finger sucking. J Autism Dev Disord 15: 177184. Ballaz SJ, Akil H, Watson SJ (2008) The CCK system underpins novelty seeking behavior in the rat: Gene expression and pharmacological analyses. Neuropeptides 42: 245 253. Barrot M, Marinelli M, Abrous DN, RougePont F, Le Moal M, Piazza PV (2000) The dopaminergic hyper respnsiveness of the shell of the nucleus accumbens is hormonedependent. Eur J Neurosci 12: 973979. Baumeister AA, Frye GD (1984) Self injurious behavior in rats produced by intranigral microinjection of GABA agonists. Pharmacol Biochem Behav 21: 8995. Berney TP, Ireland M, Burn J (1999) Behavioural phenotype of Cornelia de Lange syndrome. Arch Dis Child 81: 333336. Biersdorff KK (1991) Pain insensiti vity and indifference: alternative explanations for some medical catastrophes. Ment Retard 29: 352362. Biersdorff KK (1994) Incidence of significantly altered pain experience among individuals with developmental disabilities. Am J Ment Retard 98: 619631 Binder EB, Kinkead B, Owens MJ, Nemeroff CB (2001) Neurotensin and dopamine interactions. Pharm Rev 53: 453486. Blake BL, Muehlmann AM, Egami K, Breese GR, Devine DP, Jinnah HA (2007) Nifedipine suppresses self injurious behaviors in animals. Dev Neur osci 29: 241250. Blanchard MM, Mendelsohn D, Stamp JA (2009) The HR/LR model: Further evidence as an animal model of sensation seeking. Neurosci Biobehav Rev 33: 11451154. Blue ME, Naidu S, Johnston MV (1999) Development of amino acid receptors in frontal cortex from girls with Rett syndrome. Ann Neurol 45: 541545. Boddaert N, De Leersnyder H, Bourgeois M, Munnich A, Brunelle F, Zilbovicius (2004) Anatomical and functional brain imaging evidence of lenticuloinsular anomalies in Smith Magenis syndrome. Neuroimage 21: 10211025.
165 Bodfish JW, Crawford TW, Powell SB, Parker DE, Golden RN, Lewis MH (1995) Compulsions in adults with mental retardation: Prevalence, phenomenology, and comorbidity with stereotypy and self injury. Am J Ment Retard 100: 183192. Bodfish JW, Lewis MH (2002) Self injury and comorbid behaviors in developmental, neurological, psychiatric, and genetic disorders. In: Self injurious behavior: genebrainbehavior relationships (Shroeder SR, Oster Granite ML, Thompson T, eds), pp 2339. Washington D.C.: American Psychological Association. Boudin H, Pelaprat D, Rostene W, Beaudet A (1996) Cellular distribution of neurotensin in rat brain: Immunohistochemical study using an antipeptide antibody aginst the cloned high affinity receptor. J C omp Neurol 373: 7689. Boudreau AC, Reimers JM, Milovanovic M, Wolf ME (2007) Cell surface AMPA receptors in the rat nucleus accumbens increase during cocaine withdrawal but internalize after cocaine challenge in association with altered activation of mit ogenactivated protein kinases. J Neurosci 27: 1062110635. Boudreau AC, Wolf ME (2005) Behavioral sensitization to cocaine is associated with increased AMPA receptor surface expression in the nucleus accumbens. J Neurosci 25: 91449151. Breau LM, Camfield CS, Symons FJ, Bodfish JW, MacKay A, Finley GA, McGrath PJ (2003) Relation between pain and self injurious behavior in nonverbal children with severe cognitive impairments. J Pediatr 142: 498503. Breese GR, Baumeister AA, McCown TJ, Emerick SG, Frye G D, Crotty K, et al. (1984) Behavioral differences between neonatal and adult 6hydroxydopaminetreated rats to dopamine agonists: relevance to neurological symptoms in clinical syndromes with reduced brain dopamine. J Pharmacol Exp Ther 231: 343354. Bree se GR, Hulebak KL, Napier TC, Baumeister A, Frye G, Mueller RA (1987) Enhanced muscimol induced behavioral responses after 6OHDA lesions: relevance to susceptibility for self mutilation behavior in neonatally lesioned rats. Psychopharmacol 91: 356362. B romley J, Emerson E (1995) Beliefs and emotional reactions of care staff working with people with challenging behaviour. J Intellect Disabl Res 39: 341352. Brucke T, Sofic E, Killian W, Rett A, Riederer P (1987) Reduced concentrations and increased metabolism of biogenic amines in a single case of Rett syndrome: A postmortem brain study. J Neural Transm 68: 315324. Cadet JL, Sheng P, Ali S, Rothman R, Carlson E, Epstein C (1994) Attenuation of methamphetamine induced neurotoxicity in copper/zinc superox ide dismutase transgenic mice. J Neurochem 62: 380383.
166 Campioni MR, Xu M, McGehee DS (2009) Stress induced changes in nucleus accumbens glutamate synaptic plasticity. J Neurophysiol 101: 31923198. Carminati GG, Deriaz N, Bertschy G (2005) Low dose venl afaxine in three adolescents and young adults with autistic disorder improves self injurious behavior and attention deficit/hyperactivity disorders (ADHD) like symptoms. Prog Neuropsychopharmacol Biol Psychiatry 30: 312315. Carr EG, Durand VM (1985) Reducing behavior problems through functional communication training. J Appl Behav Anal 18: 111126. Carr EG, Smith CE (1995) Biological setting events for self injury. Ment Retard Dev Disabl Res Rev 1: 94 98. Casanova MF, Naidu S, Goldberg TE, Moser HW, Khoromi S, Kumar A, Kleinman JE, Weinberger DR (1991) Quantitative magnetic resonance imaging in Rett syndrome. J Neuropsychiatry Clin Neurosci 3: 6672. Cauwels RG, Martens LC (2005) Self mutilation behaviour in LeschNyhan syndrome. J Oral Pathol Med 34: 5 73575. Chen G L, Novak MA, Meyer JS, Kelly BJ, Vallender EJ, Miller GM (2010) TPH2 5 and 3 regulatory polymorphisms are differentially associated with HPA axis function and self injurious behavior in rhesus monkeys. Genes Brain Behav 9: 335347. Cher tkow Y, Weinreb O, Youdim MBH, Silver H (2007) Dopamine and serotonin metabolism in response to chronic administration of fluvoxamine and haloperidol combined treatment. J Neural Transm 114: 14431454. Churchill L, Kalivas PW (1996) Dopamineopioid interactions in the basal forebrain. In: The modulation of dopaminergic neurotransmission by other neurotransmitters (Ashby CR, ed), pp 5586. Boca Raton, Florida: CRC Press. Clineschmidt BV, McGuffin JC (1977) Neurotensin administered intracisternally inhibits responsiveness of mice to noxious stimuli. Eur J Pharmacol 46: 395396. Clineschmidt BV, McGuffin JC, Bunting PB (1979) Neurotensin: antinociceptive action in rodents. Eur J Pharmacol 54: 129139. Cohen IL, Tsiouris JA, Flory MJ, Kim S Y, Freedland R, H eaney G, Pettinger J, Brown WT (2010) A large scale study of the psychometric characteristics of the IBR Modified Overt Aggression Scale: Findings and evidence for increased self destructive behaviors in adult females with autism spectrum disorder. J Autis m Dev Disord 40: 599609.
167 Cohen SA, Ihrig K, Lott RS, Kerrick JM (1998) Risperidone for aggression and self injurious behavior in adults with mental retardation. J Autism Dev Disord 28: 229233. Commissiong JW (1985) Monoamine metabolites: Their relationship and lack of relationship to monoaminergic neuronal activity. Biochem Pharmacol 34: 11271131. Cooper S A, Smiley E, Allan LM, Jackson A, Finlayson J, Mantry D, Morrison J (2009) Adults with intellectual disabilities: prevalence, incidence and remissi on of self injurious behaviour, and related factors. J Intellect Disabl Res 53: 200216. Covington HE, Miczek KA (2001) Repeated social defeat stress, cocaine or morphine: Effects on behavioral sensitization and intravenous cocaine self administrat ion "bi nges". Psychopharmacol 158: 388 398. Crawhall JC, Henderson JF, Kelley WN (1972) Diagnosis and treatment of the LeschNyhan syndrome. Pediatr Res 6: 504513. Criswell HE, Mueller RA, Breese GR (1992) Pharmacological evaluation of SCH 39166, A 69024, NO 0 756, and SCH 23390 in neonatal 6 OHDA lesioned rats. Further evidence that self mutilatory behavior induced by Ldopa is related to D1 dopamine receptors. Neuropsychopharmacol 7: 95103. Cromwell HC, King BH (2004) The role of the basal ganglia in the expression of stereotyped, self injurious behaviors in developmental disorders. Int Rev Res Ment Retard 29: 119158. Cromwell HC, King BH, Levine MS (1997) Pemoline alters dopamine modulation of synaptic responses of neostriatal neurons in vitro. Dev Neurosc i 19: 497504. Cromwel l HC, Levine MS, King BH (1999) Cortical damage enhances pemoline induced self injurious behavior in prepubertal r ats. Pharmacol Biochem Behav 62: 223 227. Cromwell HC, Witte EA, Crawford CA, Ly HT, Maidment NT, King BH (1996) Pemol ine produces ipsilateral turning behavior in unilateral 6OHDA lesioned rats. Prog Neuropsychopharmacol Biol Psychiatry 20: 503514. Curin J, Terzic J, Petkovic Z, Zekan L, Terzic I, Susnjara I ( 2003) Lower cortisol and higher ACTH levels in individuals with autism. J Autism Dev Disord 33: 443 448. Davenport MD, Lutz CK, Tiefenbacher S, Novak MA, Meyer JS (2008) A rhesus monkey model of self injury: Effects of relocation stress on behavior and neuroendocrine f unction. Biol Psychiatry 63: 990 996.
168 De Vri es TJ, Schoffelmeer ANM, Tjon GHK, Nestby P, Mulder AH, Vanderschuren LJMJ (1996) Mifepristone prevents the expression of long term behavioural sensitization to amphetamine. Eur J Pharmacol 307: R3R4. DeLeon IG, Rogriguez Catter, Cataldo MF (2002) Treatm ent: Current standards of care and their research implications. In: Self injurious behavior: genebrainbehavior relationships (Shroeder SR, Oster Gra nite ML, Thompson T, eds), pp 8191. Washington D.C.: American Psychological Association. Delle Donne KT, Sesack SR, Pickel VM (1996) Ultrastructural immunocytochemical localization of neurotensin and the dopamine D2 receptor in the rat nucleus accumbens. J Comp Neurol 371: 552566. Dellu F, Piazza PV, Ma yo W, Le Moal M, Simon H (1996) Novelty seeking in rat sbiobehavioral characteristics and possible relationship with the sensationseeking trait in man. Neuropsychobiology 34: 136145. Denys D, van Megen HJGM, Westernberg HGM (2003) Emerging skinpicking behaviour after serotonin reuptake inhibitor treatment in patients with obsessive compulsive disorder: Possible mechanism and implications for clinical care. J Psychopharmacol 17: 127129. Deutch AY, Bean AJ, Bissette G, Nemeroff CB, Robbins RJ, Roth RH (1987) Stress induced alterations in neurotensin, somat ostatin and corticotrophinreleasing factor in mesotelencephalic dopamine system regions. Brain Res 417: 350 354. Di Matteo V, Pierucci M, Esposito E, Crescimanno G, Benigno A, Di Giovanni G (2008) Serotonin modulation of the basal ganglia circuitry: ther apeutic implication for Parkinsons disease and other motor disorders. Prog Brain Res 172: 423463. Dietz DM, Tapocik J Gaval Cruz M, Kabbaj M (2005) Dopamine transporter, but not tyrosine hydroxylase, may be implicated in determining individual differences in behavioral sensitization to amphetamine. Physiol Behav 86: 347 355. Dietz DM, Dietz K, Moore S, Ouimet C, Kabbaj M (2008) Repeated social defeat stress induced sensitization to the locomotor activating effects of damphetamine: role of individual d ifferences. Psychopharmacol 198: 5162. Downs J, Geranton SM, Bebbington Am, Jacoby P, Bahi Buisson N, Ravine D, Leonard H (2010) Linking MECP2 and pain sensitivity: The example of Rett syndrome. Am J Med Genet Part A 152A: 11971205.
169 Duncan GE, Breese G R, Criswell HE, Johnson KB, Schambra UB, Mueller RA, Caron MG, Fremeau RT Jr (1993) D1 dopamine receptor binding and mRNA levels are not altered after neonatal 6hydroxydopamine treatment: Evidence against dopaminemediated induction of D1 dopamine receptors during postnatal development. J Neurochem 61: 12551262. Durand VM (1982) A behavioral/pharmacological intervention for the treatment of severe self injurious behavior. J Autism Dev Disord 12: 243251. Dykens EM, Cassidy SB, King BH (1999) Maladaptive behavior differences in Prader Willi syndrome due to paternal deletion versus maternal uniparental disomy. Am J Ment Retard 104: 6777. Dykens EM, Smith ACM (1998) Distinctiveness and correlates of maladaptive behaviour in children and adolescents with S mith Magenis syndrome. J Intellect Dis Res 42: 481489. Ebert MH, Schmidt DE, Thompson T, Butler MG (1997) Elevated plasma gammaaminobutyric acid (GABA) levels in individuals with either Prader Willi syndrome or Angelman syndrome. J Neuropsychiat 9: 7580. Eliez S, Blasey CM, Freund LS, Hastie T, Reiss AL (2001) Brain anatomy, gender and IQ in children and adolescents with fragile X syndrome. Brain 124: 16101618. Elsea SH, Girirajan S (2008) SmithMagenis syndrome. Eur J Human Genet 16: 412421. Erns t C, Morton CC, Gusella JF (2010) Self injurious behaviours in people with and without intellectual delay: implications for the genetics of suicide. Int J Neuropsychopharmacol 13: 527528. Ernst M, Zametkin AJ, Matochik JA, Pascualvaca D, Jons PH, Hardy K et al. (1996) Presynaptic dopaminergic deficits in LeschNyhan disease. N Engl J Med 334: 15681572. Ernst M, Zametkin AJ, Pascualvaca D, Matochik JA, Eisenhofer G, Murphy DL, et al. (2000) Adrenergic and noradrenergic plasma levels in LeschNyhan disease. Neuropsychopharmacol 22: 320326. Esbensen AJ, Mailick Seltzer M, Lam KSL, Bodfish JW (2009) Agerelated differences in restricted repetitive behaviors in autism spectrum disorders. J Autism Dev Disord 39: 5766. Favazza AR (1998) The coming of age o f self mutilation. J Nerv Ment Dis 186: 259268.
170 Favell JE, Azrin NA, Baumeister AA, Carr EG, Dorsey MF, Forehand R, Foxx RM, Lovaas OI, Rincover A, Risley T, Romanczyk RG, Russo DC, Schroeder SR, Solnick JV (1982) The treatment of self injurious behavior Behav Ther 13:529554. Fawaz CS, Martel P, Leo D, Trudeau LE (2009) Presynaptic action of neurotensin on dopamine release through inhibition of D2 receptor function. BMC Neurosci 10: 96104. Feng Y, Zhang F, Lokey LK, Chastain JL, Lakkis L, Eberhart D Warren ST (1995) Translational suppression by trinucleotide repeat expansion at FMR1. Science 268: 731734. Ferraro L, Tanganelli S, OConnor WT, Bianchi C, Ugerstedt U, Fuxe K (1995) Neurotensin increases endogenous glutamate release in the neostriatum of the awake rat. Synapse 20: 362364. Ferraro L, Tomasini MC, Siniscalchi A, Fuxe K, Tanganelli S, Antonelli T (2000) Neurotensin increases endogenous glutamate release in rat cortical slices. Life Sci 66: 927936. Finucane B, Haines Dirrigl K, Simon E W (2001) Characterization of self injurious behaviors in children and adults with SmithMagenis syndrome. Am J Ment Retard 106: 5258. Fischer SM, Iwata BA, Mazaleski JL (1997) Noncontingent delivery of arbitrary reinforcers as treatment for self injuriou s behavior. J Appl Behav Anal 30: 239249. Fontenot MB, Padgett EE 3rd, Dupuy AM, Lynch CR, De Petrillo PB, Higley JD (2005) The effects of fluoxetine and buspirone on self injurious and stereotypic behavior in adult male rhesus macaques. Comp Med 55: 6774. Fox MA, Jensen CL, French HT, Stein AR, Huang S J, Tolliver TJ, Murphy DL (2008) Neurochemical, behavioral, and physiological effects of pharmacologically enhanced serotonin levels in serotonin transporter (SERT) deficient mice. Psychopharmacol 201: 203218. Frost FS, Mukkamala S, Covington E (2008) Self inflicted finger injury in individuals with spinal cord injury: An analysis of 5 cases. J Spinal Cord Med 31: 109116. Frye GD, Baumeister AA, Crotty K, Newman KD, Kotrla KJ (1986) Evaluation of the role of antinociception in self injurious behavior following intranigral injection of muscimol. Neuropharmacol 25: 717726.
171 Fuller R, Per ry K, Bymaster F, Wong D (1978) Comparative effects of pemoline, amfonelic acid and amphetamine on dopamine uptake and release in vitro and on brain 3,4dihydroxyphenylacetic acid concentration in spiperonetreated rats. J Pharm and Pharmacol 30: 197198. Fuxe K, OConnor WT, Antonelli T, Osborne PG, Tanganelli S, Agnati LF, Ungerstedt U (1992) Evidence for a substrate o f neuronal plasticity based on preand postsynaptic neurotensindopamine receptor interactions in the neostriatum. Proc Natl Acad Sci USA 89: 55915595. Gasser PJ, Lowry CA, Orchinik M (2006) Corticosteronesensitive monoamine transport in the rat dorsom edial hypothalamus: Potential role of organic cation transporter 3 in stressinduced modulation of monoaminergic neurotransmission. J Neurosci 26: 87588766. Gasser PJ, Orchinik M, Raju I, Lowry CA (2009) Distribution of organic cation transporter 3, a corticosteronesensitive monoamine transporter, in the rat brain. J Comp Neurol 512; 529 555. George MJ, Baumeister AA (1981) Employee withdrawal and job satisfaction in community residential facilities for mentally retarded persons. Am J Ment Defic 85: 639647. Gilbert J, Allen J, Townsend B, Wyllie M (1978) Drugs affecting the central nervous system: effects of pemoline and tricyclic antidepressants on nerve terminal adenosine triphosphatase activities and neurotransmitter release. Neuropharmacology 17: 4 19421. Glowinski J, Iversen LL (1966) Regional studies of catecholamines in the rat brain. 1. The disposition of [3H]norepinephrine, [3H]dopamine and [3H]dopa in various regions of the brain. J Neurochem 131: 655669. Gluck JP, Otto MW, Beauchamp AJ (19 85) Respondent conditioning of self injurious behavior in early socially deprived rhesus monkeys (Macaca mulatta). J Ab Psychol 94: 222226. Goosen C (1981) Abnormal behavior patterns in rhesus monkeys: Symptoms of mental disease. Biol Psychiatry 16: 697716. Gothelf D, Furfaro JA, Hoeft F, Eckert MA, Hall SS, OHara R, Erba HW, Ringel J, Hayashi KM, Patnaik S, Golianu B, Kraemer HC, Thompson PM, Piven J, Reiss AL (2008) Neuroanatomy of Fragile X syndrome is associated with aberrant behavior and the Fragi le X Mental Retardation Protein (FMRP). Ann Neurol 63: 4051.
172 Grisel JE, Fleshner M, Watkins LR, Maier SF (1993) Opioid and nonopioid interactions in two forms of stress induced analgesia. Pharmacol Biochem Beahv 45: 161172. Gudelsky GA, Berry SA, Meltz er HY (1989) Neurotensin activates tuberoinfundibular dopamine neurons and increases serum corticosterone concentrations in the rat. Neuroendocrinology 49: 604609. Guess D, Carr E (1991) Emergence and maintenance of stereotypy and self injury. Am J Ment Retard 96: 299319. Gui X, Carraway RE, Dobner PR (2004) Endogenous neurotensin facilitates visceral nociception and is required for stress induced antinociception in mice and rats. Neurosci 126: 10231032. Gysling K, Wang RY (1983) Morphineinduced acti vation of A10 dopamine neurons in the rat. Brain Res 277: 119127. Haas BW, BarneaGoraly N, Lightbody AA, Patnaik SS, Hoeft F, Hazlett H, Piven J, Reiss AL (2009) Early whitematter abnormalities of the ventral frontostriatal pathway in fragile X syndrom e. Dev Med Child Neurol 51: 593599. Hagberg BA (1985) Retts syndrome: prevalence and impact on progressive severe mental retardation in girls. Acta Paediatr 74: 405408. Hall S, Oliver C, Murphy G (2001) Self injurious behaviour in young children with Lesch Nyhan syndrome. Dev Med Child Neurol 43: 745749. Hall SS, Lightbody AA, Reiss AL (2008) Compulsive, self injurious, and autistic behavior in chidren and adolescents with Fragile X syndrome. Am J Ment Retard 113: 4453. Halladay AK, Kusnecov A, Mic hna L, Kita T, Hara C, Wagner GC (2003) Relationship between methamphetamineinduced dopamine release, hyperthermia, self injurious behaviour and long term dopamine depletion in BALB/c and C57BL/6 mice. Pharmacol Toxicol 93: 3341. Hanson GR, Singh N, Mer chant K, Johnson M, Bush L, Gibb JW (1992) Responses of limbic and extrapyramidal neurotensin systems to stimulants of abuse. In: The neurobiology of neurotensin (Kitabgi P, Nemeroff CB, eds.) pp: 165172. Annals of the New York Academy of Sciences, 668. Harris JC, Lee RR, Jinnah HA, Wong DF, Yaster M, Bryan RN (1998) Craniocerebral magnetic resonance imaging measurement and findings in LeschNyhan syndrome. Arch Neurol 55: 547553.
173 Harrison RJ, McNeil GP, Dobner PR (1995) Synergistic activation of neurot ensin/neuromedin N gene expression by c Jun and glucocorticoids: novel effects of Fos family proteins. Mol Endocrinol 9: 981983. Hartley SL, MacLean Jr WE, Butler MG, Zarcone J, Thompson T (2005) Maladaptive behaviors and risk factors among the genetic s ubtypes of Prader Willi syndrome. Am J Med Genet 136A: 140145. Hellings JA, Kelley LA, Gabrielli WF, Kilgore E, Shah P (1996) Sertraline response in adults with mental retardation and autistic disorder. J clin Psychiatry 57: 333336. Herbert MR (2010) C ontributions of the environment and environmentally vulnerable physiology to autism spectrum disorders. Current Op in Neurol 23: 103110. Herman J, Cullinan W (1997) Neurocircuitry of stress: Central control of the hypothalamopituitary adrenocortical axi s. Trends Neurosci 20: 7884. Hessl D, Glaser B, Dyer Friedman J, Blasey C, Hastie T, Gunnar M, Reiss A ( 2002) Cortisol and behavior in fragile X sy ndrome. Psychoneuroendocrinol 27: 855872. Hiraiwa R, Maegaki Y, Oka A, Ohno K (2007) Behavioral and psychiatric disorders in Prader Willi syndrome: A population study in Japan. Brain Dev 29: 535 542. Hoeft F, Hernandez A, Parthasarathy S, Watson CL, Hall SS, Reiss AL (2008) Frontostriatal dysfunction and potential compensatory mechanisms in male adolescents with Fragile X syndrome. Human Br Map 28: 543554. Hollander E, Anagnostou E, Chaplin W, Esposito K, Haznedar MM, Licalzi E, Wasserman S, Soorya L, Buchsbaum M (2005) Striatal volume on magnetic resonance imaging and repetitive behaviors in autism. Biol Psychiatry 58: 226232. Hooks MS, Colvin A, Juncos JL, Justice JB (1992) Individual differences in basal and cocainestimulated extracellular dopamine in the nucleus accumbens using quantitative microdialysis. Brain Res 587: 306 312. Hooks MS, Jones GH, Smith AD, Neill DB, Justice JB (1991a ) Individual differences in locomotor activity and sensitization. Pharmacol Biochem Behav 38: 467470. Hooks MS, Jones GH, Smtih AD, Neill DB, Justice JB (1991b ) Response to novelty predicts the locomotor and nucleus accumbens dopamine response to cocaine. Synapse 9: 121 128 Hooks MS, Juncos JL, Justice JB, Jr., Meiergerd SM, Povlock SL, Schenk JO, Kalivas PW (1994) Individual locomotor response to novelty predicts selective alterations in D1 and D2 receptors and mRNA s. J Neurosci 14: 61446152.
174 Huhman KL (2006) Social conflict models: Can they inform us about human psycho pathology? Horm Behav 50, 640 646. Insel TR (2007) From animal models to model animals. Biol Psychiatry 62: 13371339. Iwata BA, Pace GM, Dorsey MF Zarcone JR, Vollmer TR, Smith RG, Rodgers TA, Lerman CD, Shore BA, Mazaleski JL, Goh H L, Edwards Cowdery G, Kalsher MJ, McCosh KC, Willis KD (1994) The functions of self injurious behavior: An experimental epidemiological analysis. Journal of Applied Be havior Analysis 27: 215240. Iwata BA, Roscoe EM, Zarcone JR, Richman DM (2002) Environmental determinants of self injurious behavior. In: Self injurious behavior: genebrainbehavior relationships (Shroeder SR, Oster Granite ML, Thompson T, eds), pp 931 03. Washington D.C.: American Psychological Association. Jain S, Sharma R (2002) Analgesia in phasic and tonic pain tests in a pharmacological model of autotomy. Ind J Exp Biol 40: 12691274. Jama A, Cecchi M, C alvo N, Watson S, Akil H (2008) Inter indiv idual differences in novelty seeking behavior in rats predict differential responses to desipramine in the forced swim test. Psychopharmacology 198: 333340. Jankovic J, Caskey TC, Stout JT, Butler IJ (1988) LeschNyhan syndrome: a study of motor behavior and cerebrospinal fluid neurotransmitters. Ann Neurol 23: 4666469. Janowsky DS, Barnhill LJ, Davis JM (2003a) Olanzapine for self injurious, aggressive, and disruptive behaviors in intellectually disabled adults: A retrospective, openlabel naturalistic trial. J Clin Psychiatry 64: 12581265. Janowsky DS, Kraus JE, Barnhill J, Elamir B, Davis JM (2003b) Effects of topiramate on aggressive, self injurious, and disruptive/destructive behaviors in the intellectually disabled: An openlabel retrospective st udy. J Clin Psychopharmacol 23: 500504. Jellinger K, Armstrong D, Zoghbi HY, Percy AK (1988) Neuropathology of Rett syndrome. Acta Neuropathol 76: 142158. Jinnah HA, Egami K, Rao L, S hin MY, Kasim S, Hess EJ (2003) Expression of cfos in the brain after activation of Ltype calcium channels. Dev Neurosci 25: 403411. Jinnah HA, Yitta S, Drew T, Kim BS, Visser JE, Rothstein JD (1999) Calcium channel activation and self biting in mice. Proc Natl Acad Sci USA 96: 1522815232.
175 Kabbaj M, Devi ne DP, Savage VR, Akil H (2000) Neurobiological correlates of individual differences in novelty seeking behavior in the rat: Differential expression of stress related m olecules. J Neurosci 20: 6983 6988. Kabbaj M, Evans S, Watson SJ, Akil H (2004) The search for the ne urobiological basis of vulnerability to drug abuse: using microarrays to investigate the role of stress and individual differences. Neuropharmacology 47: 111 122. Kahng S, Iwata BA, Lewin AB (2002) The impact of functional assessment on the treatment of s elf injurious behavior. In: Self injurious behavior: genebrainbehavior relationships (Shroeder SR, Oster Granite ML, Thompson T, eds), pp 119131. Washington D.C.: American Psychological Association. Kalivas PW (1993) Neurotransmitter regulation of dopa mine neurons in the ventral tegmental area. Brain Res Rev 18: 75 113. Kalivas PW, Duffy P (1990) Effect of acute and daily neurotensin and enkephalin treatments on extracellular dopamine in the nucleus accumbens. J Neurosci 10: 29402949. Kalivas PW, Duf fy P (1995) Selective activation of dopamine transmission in the shell of the nucleus accumbens by stress. Brain Res 675: 325328. Karstaedt PJ, Kerasidis H, Pincus JH, Meloni R, Graham J, Gale K (1994) Unilateral destruction of dopamine pathways increases ipsilateral striatal serotonin turnover in rats. Exp Neurol 126: 2530. Kasim S, Blake BL, Xueliang F, Chartoff E, Egami K, Breese GR, et al. (2006) The role of dopamine receptors in the neurobehavioral syndrome provoked by activation of L type calcium channels in rodents. Dev Neurosci 28: 505517. Kasim S, Egami K, Jinnah HA (2002) Self biting induced by activation of Ltype calcium channels in mice: serotonergic influences. Dev Neurosci 24: 322327. Kasim S, Jinnah HA (2002) Pharmacologic thresholds for self injurious behavior in a genetic mouse model of LeschNyhan disease. Pharmacol Biochem Behav 73: 583392. Kasim S, Jinnah HA (2003) Self biting induced by activation of Ltype calcium channels in mice: dopaminergic influences. Dev Neurosci 25: 2025. Kastner T, Finesmith R, Walsh K (1993) Long term administration of valproic acid in the treatment of affective symptoms in people with mental retardation. J Clin Psychopharmacol 13: 448451.
176 Kelland MD, Chiodo LA (1996) Serotonergic modulation of midbrain systems. In: The modulation of dopaminergic neurotransmission by other neurotransmitters (Ashby CR, ed), pp 87122. Boca Raton, Florida: CRC Press. Kelley WN (1968) Hypoxanthineguanine phosphoribosyltransferase deficiency in the Lesch Nyhan syndrom e and gout. Federation Proceedings 27: 1047 1052. Kemp A, Fillmore P, Lenjavi M, Lyon M, C hicz DeMet A, Touchette P, Sandman CA (2008) Temporal patterns of self injurious behavior correlate with stress hormone levels in the developmentally disabled. Psych iatry Res 157: 181189. Kennedy CH (2002) Evolution of stereotypy into self injury. In: Self injurious behavior: genebrainbehavior relationships (Shroeder SR, Oster Granite ML, Thompson T, eds), pp 133143. Washington D.C.: American Psychological Associ ation. Kennedy CH, Caruso M, Thompson T (2001) Experimental analyses of genebrainbehavior relations: Some notes on their application. J Appl Behav Anal 34: 539549. Kies SD, Devine DP (2004) Self injurious behaviour: a comparison of caffeine and pemoli ne models in rats. Pharmacol Biochem Behav 79: 587 598. Kilts CD, Coco ML, Ely TD, Bissette G, Nemeroff CB (1992) Differential effects of conditioned and unconditioned stress on the neurotensin content of dopamine cell body groups of the ventral mesenceph alon. Ann NY Acad Sci 668: 266276. King BH (1993) Self injury by people with mental retardation: A compulsive behavior hypothesis. Am J Ment Retard 98: 93112. King BH, Au D, Poland RE (1995) Pretreatment with MK 801 inhibits pemolineinduced self bitin g behavior in prepubertal rats. Dev Neurosci 17: 4752. King RA, Riddle MA, Chappell PB, Hardin MT, Anderson GM, Lombroso P, Scahill L (1991) Emergence of self destructive phenomena in children and adolescents during fluoxetine treatment. J Am Acad Child Adolesc Psychiatry 30: 179186. Kita T, Matsunari Y, Saraya T, Shimada K, OHara K, Kubo K, Wagner GC, Nakashima T (2000) Methamphetamineinduced striatal dopamine release, behavior changes and neurotoxicity in BALB/c mice. Int J Dev Neurosci 18: 521530. Kitaichi K, Fukuda M, Nakayama H, Aoyama N, Ito Y, Fujimoto Y, Takagi K, Hasegawa T (2005) Behavioral changes following antisense oligonucleotideinduced reduction of organic cation transporter 3 in mice. Neurosci Let 382: 195200.
177 Kitaichi K, Morishita Y, Doi Y, Ueyama J, Matsushima M, Zhao Y L, Takagi K, Hasegawa T (2003) Increased plasma concentration and brain penetration of methamphetamine in behaviorally sensitized rats. Eur J Pharmacol 464: 3948. Klonsky ED (2007) The functions of deliberate sel f harm: A review of the evidence. Clin Psychol Rev 27: 226239. Kolevzon A, Newcorn JH, Kryzak L, Chaplin W, Watner D, Hollander E, Smith CJ, Cook Jr EH, Silverman JM (2010). Relationship between whole blood serotonin and repetitive behaviors in autism. P sychiatry Res 175: 274276. Kopin IJ (1985) Catecholamine metabolism: basic aspects and clinical significance. Pharmacol Rev 37: 333364. Kozinetz CA, Skender ML, MacNaughton N, Almes MJ, Schultz RJ, Percy AK, Glaze DG (1993) Epidemiology of Rett syndrom e: A populationbased registry. Pediatrics 91: 445450. Kroeker R, Touchette PE (2004) Quantifying temporal distributions of self injurious behavior: Defining bouts versus discrete events. Am J Ment Retard 109: 18. Kuroda KO, Ornthanalai VG, Kato T, Mur phy NP (2010) FosB null mutant mice show enhanced methamphetamine neurotoxicity: Potential involvement of FosB in intracellular feedback signaling and astroglial function. Neuropsychopharmacol 35: 641655. Lafrance, M, Roussy G, Belleville K, Maeno H, Bea udet N, Wada K, Sarret P (2010) Involvement of NTS2 receptors in stress induced analgesia. Neurosci 166: 639652. Langen M, Schnack HG, Nederveen H, Bos D, Lahuis BE, de Jonge MV, van Engeland H, Durston S (2009) Changes in the developmental trajectories of striatum in autism. Biol Psychiatry 66: 327333. Lekman A, Witt Engerstrom I, Gottfries J, Hagberg BA, Percy AK, Svennerholm L (1989) Rett syndrome: Biogenic amines and metabolites in postmortem brain. Pediatr Neurol 5: 357362. Letts RM, Hobson DA (1975) Special devices as aids in the management of child self mutilation in the Lesch Nyhan syndrome. Pediatrics 55: 852855. Lewis MH, Gluck JP, Beauchamp AJ, Keresztury MF, Mailman RB (1990) Long term effects of early social isolation in Macaca mulatta: Changes in dopamine receptor function following apomorphine challenge. Brain Res 513: 6773.
1 78 Liu J, Krantz ID (2009) Cornelia de Lange syndrome, cohesion, and beyond. Clin Genet 76: 303314. Lloyd KG, Hornykiewcz O, Davidson L, Shannak K, Farley I, Golds tein M, Shibuya M, Kelley WN, Fox IH (1981) Biochemical evidence of dysfunction of brain neurotransmitters in the LeschNyhan syndrome. N Engl J Med 305: 11061111. Loupe PS, Zhou X, Davies MI, Schroeder SR, Tessel RE, Lunte SM (2002) Fixed ratio discrimination training increases in vivo striatal dopamine in neonatal 6OHDA lesioned rats. Pharmacol Biochem Behav 74: 6171. Lucas LR, Angulo JA, Moal ML, McEwen BS, Piazza PV (1998) Neurochemical characterization of individual vulnerability to addictive drug s in rats. Eur J Neurosci 10: 31533163. Luiselli JK (1986) Behavior analysis of pharmacological and contingency management interventions for self injury. J Behav Ther Exp Psychiatry 17: 275284. Luiselli JK, Blew P, Thibadeau S (2001) Therapeutic effect s and long term efficacy of antidepressant medication for persons with developmental disabilities. Behavioral assessment in two case of treatment resistant aggression and self injury. Behav Modif 25: 6278. Lutz C, Well A, Novak M (2003) Stereotypic and s elf injurious behavior in rhesus macaques: A survey and retrospective analysis of environment and early experience. Am J Primat 60: 115. Mace FC, Mauk JE (1995) Biobehavioral diagnosis and treatment of self injury. Ment Retard Dev Disabl Res Rev 1: 104110. Madge N, Hewitt A, Hawton K, Jan de Wilde E, Corcoran P, Fekete, S, van Heeringen K, De Leo D, Ystgaard M (2008) Deliberate self harm within an international community sample of young people: Comparative findings from the Child & Adolescent Self harm in Europe (CASE) Study. J Child Psychol Psychiatry 49: 667677. Major CA, Kelly BJ, Novak MA, Davenport MD, Stonemetz KM, Meyer JS (2009) The anxiogenic drug FG7142 increases self injurious behavior in male rhesus monkeys (Macaca mulatta). Life Sci 85: 753 758. Mailis A (1996) Compulsive targeted self injurious behaviour in humans with neuropathic pain: A counterpart of animal autotomy? Four case reports and literature review. Pain 64: 569578.
179 Martin LJ, Spicer DM, Lewis MH, Gluck JP, Cork LC (1991) Social deprivation of infant rhesus monkeys alters the chemoarchitecture of the brain: I. Subcortical regions. J Neurosci 11: 33443358. Mathews CA, Waller J, Glidden D, Lowe TL, Herrera LD, Budman CL, Erenberg G, Naarden A, Bruun RD, Freimer NB, Reus VI (2004) Self injurious behaviour in Tourette syndrome: Correlates with impulsivity and impulse control. J Neurol Neurosurg Psychiatry 75: 11491155. Matson JL, Rivet TT (2008) Characteristics of challenging behaviours in adults with autistic disorder, PDD NO S, and intellectual disability. J Intellect Dev Disabl, 33: 323329. Matsuishi T, Nagamitsu S, Yamashita Y, Murakami Y, Kimura A, Sakai T, Shoji H, Kato H, Percy AK (1997) Decreased cerebrospinal fluid levels of substance P in patients with Rett syndrome. Ann Neurol 42: 978981. Mazella J, Vincent J P (2006) Internalization and recycling properties of neurotensin receptors. Peptides 27: 24882492. McCracken JT et al. (2002) Risperidone in children with autism and serious behavioral problems. N Engl J Med 347: 314321. McDonough M, Hillery J, Kennedy N (2000) Olanzapine for chronic, stereotypic self injurious behaviour: A pilot study in seven adults with intellectual disability. J Intellect Disabl Res 44: 677684. McKee BL, Kelley AE, Moser HR, Andrzeiew ski ME (2010) Operant learning requires NMDA receptor activation in the anterior cingulate cortex and dorsomedial striatum, but not in the orbitofrontal cortex. Behav Neurosci 124: 500 509. Meguro M, Mitsuya K, Sui H, Shigenami K, Kugoh H, Nakao M, Oshimura M (1997) Evidence for uniparental, paternal expression of the human GABAA receptor subunit genes, using microcell mediated chromosome transfer. Human Mol Genet 6: 21272133. Menon V, Leroux J, White CD, Reiss AL (2004) Frontostriatal deficits in fragil e X syndrome: Relation to FMR1 gene expression. Proc Natl Acad Sci USA 101: 36153620. Meredith GE (1999) The synaptic framework for chemical signaling in nucleus accumbens. Ann N Y Acad Sci 29: 140156. Miura H, Qiao H, Ohta T (2002) Influence of aging and social isolation on changes in brain monoamine turnover and biosynthesis of rats elicited by novelty stress. Synapse 46: 116124.
180 Mogil JS, Kest B, Sadowski B, Belknap JK (1996) Differential genetic mediation of sensitivity to morphine in genetic models of opiate antinociception: influence of nociceptive assay. J Pharmacol Exp Ther 276: 532544. Moldow RL, Fischman AJ (1987) C ocaine induced secretion of ACTH, betaendorphin, and corticosterone. Peptides 8: 819822. Molina V, Orsingher O (1981) Effects of Mg pemoline on the central catecholaminergic system. Arch Int P harmacodyn Ther 251: 66 79. Mori T, Ito S, Kita T, Narita M, Suzuki T, Matsubayashi K, Sawaguchi T (2007) Oxidative stress in methamphetamineinduced self injurious behavior in mice. Behav Pharmacol 18: 239249. Muehlmann, AM, Brown BD, Devine DP (2008) Pemoline (2 Amino 5 phenyl 1,3 oxazol 4 one) induced self injurious behavior: A rodent model of pharmacotherapeutic efficacy. J Pharmacol Exp Ther 324: 214 223. Muehlmann AM, Devine DP (2008) Glutamate mediated neuroplasticity in an animal model of self injurious behaviour. Behav Brain Res 189: 3240. Mueller K, Hollingsworth E, Pettit H (1986) Repeated pemoline produces self injurious behavior in adult and weanling rats. Pharmacol Biochem Behav 25: 933938. Mueller K, Hsaio S (1980) Pemoline induced self biting in rats and self mutilation in th e deLange syndrome. Pharmacol Biochem Behav 13: 627631. Mueller K, Nyhan WL (1982) Pharmacologic control of pemoline induced self injurious behavior in rats. Pharmacol Biochem Behav 16: 957963. Murai S, Saito H, Abe E, Masuda Y, Itoh T (1992) A rapid assay for neurotransmitter amino acids, aspartate, glutamate, glycine, taurine and gammaaminobutyric acid in the brain by highperformance liquid chromatography with electrochemical detection. J Neural Transm Gen Sect 87: 145 153. Myers RD, Lee TF (1983) In vivo release of dopamine during perfusion of neurotensin in substantia nigra of the unrestrained rat. Peptides 4: 955961. Nader R, Oberlander TF, Chambers CT, Craig KD (2004) Expression of pain in children with autism. Clin J Pain 20: 8897. Nakayama H, Kitaichi K, Ito Y, Hashimoto K, Takagi K, Yokoi T, Takagi K, Ozaki N, Yamamoto T, Hasegawa T (2007) The role of organic cation transporter 3 i n methamphetamine disposition and its behavioral response in rats. Brain Res 1184: 260269.
181 National Institutes of Health Consensus Development Conference Statement (1989) Treatment of destructive behaviors in persons with developmental disabilities. NIH Consensus Statement Online (http://consensus.nih.gov/1989/1989DestructiveBehaviorsDevelopment075html.ht m), 7: 1 15. Accessed June 2010. Nelson CL, Milovanovic M, Wetter JB, Ford KA, Wolf ME (2009) Behavioral sensitization to amphetamine is not accompanied by changes in glutamate receptor surface expression in the rat nucleus accumbens. J Neurochem 109: 3551. Newell KM, Challis JH, Boros RL, Bodfish JW (2002) Further evidence of the dynamics of self injurious behaviors: Impact forces and limb motions. Am J Ment Retard 107: 6068. Newell KM, Sprague RL, Pain MT, Deutsch KM, Meinhold P (1999) Dynamics of self injurious behaviors. Am J Ment Retard 104: 1121. Novak MA, Crockett CM, Sackett GP (2002) Self injurious behavior in captive macaque monkeys. In: Se lf injurious behavior: genebrainbehavior relationships (Schroeder SR, Oster Granite ML, Thompson T, eds), pp 151161 Washington, D.C.: American Psychological Association. Nyhan WL (1968) Summary of clinical features. Federation Proceedings 27: 1034104 1. Nyhan WL (2002) Lessons from LeschNyhan syndrome. In: Self injurious behavior: genebrainbehavior relationships (Schroeder SR, Oster Granite ML, Thompson T, eds), pp 251 267. Washington, D.C.: American Psychological Association. Ogura K, Fujii T, Ab e N, Hosokai Y, Shinohara M, Takahashi S, Mori E (2010) Small gray matter volume in orbitofrontal cortex in Prader Willi syndrome: a voxel based MRI study. Hum Brain Mapp (epub). Oliver C, Murphy GH, Corbett JA (1987) Self injurious behaviour in people wi th mental handicap: A total population study. J Ment Defic Res 31: 147162. Oliver C, Murphy G, Crayton L, Corbett J (1993) Self injurious behavior in Rett syndrome: Interactions between features of Rett syndrome and operant conditioning. J Aut Dev Disord 23: 91109. Oliver C, Sloneem J, Hall S, Arron K (2009) Self injurious behaviour in Cornelia de Lange syndrome: I. Prevalence and phenomenology. J Intellect Disabl Res 53: 575589.
182 Owen R, Sikich L, Marcus RN, Corey Lisle P, Manos G, McQuade RD, Carson WH, Findling RL (2009) Aripiprazole in the treatment of irritability in children and adolescents with autistic disorder. Pediatrics 124: 15331540. Pacchioni AM, Cador M, Bregonzio C, Cancela LM (2007) A glutamatedopamine interaction in the persistent en hanced response to amphetamine in nucleus accumbens core but not shell following a single restraint stress. Neuropsychopharmacol 32: 682692. Parikh MS, Kolevzon A, Hollander E (2008) Psychopharmacology of aggression in children and adolescents with autism : A critical review of efficacy and tolerability. J Child Adolesc Psychopharmacol 18: 157178. Pejler Gn, Ronnberg E,Waern I, Wernersson S (2010) Mast cell proteases: multifaceted regulators of inflammatory disease. Blood 115: 49814990. Peng S, Zhang Y, Zhang J, Wang H, Ren B (2011) Glutamate receptors and signal transduction in learning and memory. Mol Biol Rep 38: 453460. Petty J, Allen D, Oliver C (2009) Relationship among challenging, repetitive, and communicative behaviors in children with severe intellectual disabilities. Am J Intellect Dev Disabil 114: 356368. Piazza P, Deminiere J, LeMoal M, Simon H (1989) Factors that predict individual vulnerability to amphetamine self administration. Science 245: 1511 1513. Piazza PV, Le Moal M (1996) Path ophysiological Basis of Vulnerability to Drug Abuse: Role of an Interaction Between Stress, Glucocorticoids, and Dopaminergic Neurons. Annu Rev Pharmacol Toxicol 36: 359 378. Pierce RC, Kalivas PW (1997) A circuitry model of the expression of behavioral sensitization to amphetaminelike psychostimulants. Brain Res Rev 25: 192216. Pinto D et al. (2010) Functional impact of global rare copy number variation in autism spectrum disorders. Nature epub. Post RM (1980) Intermittent versus continuous stimulation: Effect of time interval on the development of sensitization or tolerance. Life Sci 26: 12751282. Potenza MN, Holmes JP, Kanes SJ, McDougle CJ (1999) Olanzapine treatment of children, adolescents, and adults with pervasive developmental disorders: An o pen label pilot study. J clin Psychopharmacol 19: 3744. Powell SB, Bodfish JW, Parker D, Crawford TW, Lewis MH (1996) Self restraint and self injury: Occurrence and motivational significance. Am J Ment Retard 101: 4148.
183 Priano L, Miscio G, Grugni G, Mil ano E, Baudo S, Sellitti L, Picconi R, Mauro A (2009) On the origin of sensory impairment and altered pain perception in Prader Willi syndrome: A neurophysiological study. Euro J Pain 131: 829835. Price TJ, Rashid MH, Millecamps M, Sanoja R, Entrena JM, Cervero (2007) Decreased nociceptive sensitization in mice lacking the Fragile X mental retardation protein: Role of mGluR1/5 and mTOR. J Nuerosci 27: 1395813967. Radke JM, Owen MJ, Nemeroff CB (2001) The effects of glutamate receptor agonists on neurotensin release using in vivo microdialysis. Eur J Pharmacol 411: 129134. Rajadhyaksha AM, Kosofsky BE (2005) Psychostimulants, protein phosphorylation and gene expression: A growing role of Ltype calcium channels. Cellsci 2: 127144. Rassin DK, Lloyd KG, Kelley WN, Fox I (1982) Decreased amino acids in various brain areas of patients with LeschNyhan syndrome. Neuropediatrics 13: 130134. Reiss AL, Dant CC (2003) The behavioral neurogenetics of Fragile X syndrome: Analyzing genebrainbehavior relationships in child developmental psychopathologies. Dev Psychopath 15: 927968. Relkovic D, Doe CM, Humby T, Johnstone KA, Resnick JL, Holland AJ, Hagan JJ, Wilkinson LS, Isles AR (2010) Behavioural and cognitive abnormalities in an imprinting centre deletion m ouse model for Prader Willi syndrome. Eur J Neurosci 31: 156164. Ricketts RW, Goza AB, Ellis CR, Singh YN, Singh NN, Cooke JC 3rd (1993) Fluoxetine treatment of severe self injury in young adults with mental retardation. J Am Acad Child Adolesc Psychiatr y 32: 865869. Riederer P, Weiser M, Wichart I, Schmidt B, Killian W, Rett A (1986) Preliminary brain autopsy findings in progredient Rett syndrome. Am J Med Genet Suppl 1: 305315. Rivet JM, Stinus L, Le Moal M, Mormede P (1989) Behavioral sensitization to amphetamine is dependent on corticosteroid receptor activation. Brain Res 498: 149153. Robey KL, Reck JF, Giacomini KG, Barabas G, Eddey GE (2003) Modes and patterns of self mutilation in persons with LeschNyhan disease. Dev Med Child Neurol 45: 167171. Rojahn A, Esbensen AJ (2002) Epidemiology of self injurious behavior in mental retardation: A review. In: Self injurious behavior: genebrainbehavior relationships (Shroeder SR, Oster Granite ML, Thompson T, eds), pp 4177. Washington D.C.: America n Psychological Association.
184 RougePont F, Deroche V, Le Moal M Piazza PV (1998) Individual differences in stressinduced dopamine release in the nucleus accumbens are influenced by corticosterone. Eur J Neurosci 10: 3903 3907. RougePont F, Marinelli M Le Moal M, Simon H, Piazza PV (1995) Stress induced sensitization and glucocorticoids. II. Sensitization of the increase in extracellular dopamine induced by cocaine depends on stress induced corticosterone secretion. J Neurosci 15: 71897195. RougePon t F, Piazza PV, Kharouby M, Le Moal M, Simon H (1993) Higher and longer stressinduced increase in dopamine concentrations in the nucleus accumbens of animals predisposed to amphetamine self administration. A microdialysis study. Brain Res 602: 169174. R oux JC, Villard L (2010) Biogenic amines in Rett syndrome: The usual suspects. Behav Genet 40: 5975. Rowe WB, Nicot A, Sharma S, Gully D, Walker CD, Rostene WH, Meaney MJ, Quirion R (1997) Central administration of the neurotensin receptor antagonist, SR 48692, modulates diurnal and stress related hypothalamic pituitary adrenal activity. Neuroendocrinology 66: 7585. Ruedrich S, Swales TP, Fossaceca C, Toliver J, Rutkowski A (1999) Effect of divalproex sodium on aggression and self injurious behaviour in adults with intellectual disability: A retrospective review. J Intellect Disabl Res 43: 105111. Ryan, NM (1983) Body rocking, head banging, and head rolling: An analysis of rhythmic motor activities in normal infants. Pediatr Nurs 9: 281285. Saito H, M urai S, Abe E, Masuda Y, Itoh T (1992) Rapid and simultaneous assay of monoamine neurotransmitters and their metabolites in discrete brain areas of mice by HPLC with coulometric detection. Pharmacol Biochem Behav 42: 351356. Saito Y, Ito M, Hanaoka S, Ohama E, Akaboshi S, Takashima S (1999) Dopamine receptor upregulation in LeschNyhan syndrome: a postmortem study. Neuropediatrics 30: 66 71. Saloviita T (2000) The structure and correlates of self injurious behavior in an institutional setting. Res Dev Disabl 21: 501511. Sandman CA (1990) The opiate hypothesis in autism and self injury. J Child Adolesc Psychopharmacol 1: 237248.
185 Sandman CA, Touchette PE, Marion SD, Chicz DeMet A (2008) The role of proopiomelanocortin (POMC) in sequentially dependent self i njurious behavior. Dev Psychobiology 50: 680689. Sansom D, Krishnan VH, Corbett J, Kerr A (1993) Emotional and behavioural aspects of Rett syndrome. Dev Med Child Neurol 35: 340345. Sarimski K (1997) Behavioural phenotypes and family stress in thre e mental retardation syndromes. Eur Child Adolesc Psychiatry 6: 2631. Sarret P, Krzywkowski P, Segal L, Nielsen MS, Petersen CM, Mazella J, Stroh T, Beaudet A (2003) Distribution of NTS3 receptor/sortilin mRNA and protein in the rat central nervous system. J Comp Neurol 461: 483505. Schroeder SR, Mulick JA, Rojahn J (1980) The definition, taxonomy, epidemiology, and ecology of self injurious behavior. J Autism Dev Disord 10: 417432. Sears LL, Vest C, Mohamed S, Bailey J, Ranson BJ, Piven J (1999) An M RI study of the basal ganglia in autism. Prog NeuroPsychopharmacol & Biol Psychiat 23: 613624. Seggie J, Werstiuk ES, Joshi M (1985) Lithium and twenty four hour rhythms of serum corticosterone, prolactin and growth hormone in pigmented eye rats. Prog N euroPsychopharmacol & Biol Psychiat 9, 755758. Shank RP, Maryanoff BE (2008) Molecular pharmacodynamics, clinical therapeutics, and pharmacokinetics of topiramate. CNS Neurosci Ther 14: 120142. Shapira NA, Lessig M, Murphy TK, Driscoll DJ, Goodman WK (2002) Topiramate attenuates self injurious behaviour in Prader Willi syndrome. Int J Neuropsychopharmacol 5: 141145. Shishido T, Watanabe Y, Kato K, Horikoshi R, Niwa S I (2000) Effects of dopamine, NMDA, opiate, and serotoninrelated agents on acute methamphetamineinduced self injurious behavior in mice. Pharmacol Biochem Behav 66: 579583. Sivam SP (1989) D1 dopamine receptor mediated substance P depletion in the striatonigral neurons of rats subjected to neonatal dopaminergic denervation: Implicatio ns for self injurious behavior. Brain Res 500: 119130. Sivam SP (1995) Dopaminergic modulation of serotonin metabolism in rat striatum: a study with dopamine uptake inhibitor GBR 12909. Life Sci 56: 467472. Sivam SP, Pugazhenthi S, Pugazhenthi V, Brown H (2008) LDOPA induced activation of striatal p38MAPK and CREB in neonatal dopaminergic denervated rat: Relevance to self injurious behavior. J Neurosci Res 86: 339349.
186 Snead RW, Boon F, Presberg J (1994) Paroxetine for self injurious behavior. J Am Ac ad Child Adolesc Psychiatry 33: 909910. Sora I, Li B, Fumushima S, Fukui A, Arime Y, Kasahara Y, Tomita H, Ikeda K (2009) Monoamine transporter as a target molecule for psychostimulants. Int Rev Neurobiol 85: 2933. Souaze F (2001) Maintaining cell sens itivity to G protein coupled receptor agonists: Neurotensin and the role of receptor gene activation. J Neuroendocrinol 13: 473479. Souaze F, Forgez P (2006) Molecular and cellular regulation of neurotensin receptor under acute and chronic agonist stimul ation. Peptides 27: 24932501. Stodgell CJ, Loupe PS, Schroeder SR, Tessel RE (1998) Cross sensitization between footshock stress and apomorphine on self injurious behavior and neostriatal catecholamines in a rat model of LeschNyhan syndrome. Brain Res 783: 10 18. Stodgell CJ, Shroeder SR, Tessel RE (1996) Fr discrimination training reverses 6hydroxydopamineinduced striatal dopamine depletion in a rat model of LeschNyhan syndrome. Brain Res 713: 246252. Subramanium B, Naidu S, Reiss AL (1997) Neuroanatomy in Rett syndrome: Cerebral cortex and posterior fossa. Neurol 48: 399407. Surmeier DJ, Bargas J, Hemmings HC Jr, Nairn AC, Greengard P (1995) Modulation of calcium currents by a D1 dopaminergic protein kinase/phosphatase cascade in rat neostriatal neurons. Neuron 14: 385397. Symons FJ, Butler MG, Sanders MD, Feurer ID, Thompson T (1999) Self injurious behavior and Prader Willi syndrome: behavioral forms and body locations. Am J Ment Retard 104: 260269. Symons FJ, Clark RD, Hatton DD, Skinner M, Bailey Jr DB (2003a) Self injurious behavior in young boys with Fragile X syndrome. Am J Med Genet 118A: 115121. Symons F, Sutton K, Walker C, Bodfish J (2003b) Altered diurnal pattern of salivary substance P in adults with developmental disabilities and chronic self injury. Am J Ment Retard 108: 1318. Symons FJ, Thompson A, Rodriguez MC (2004) Self injurious behavior and the efficacy of naltrexone treatment: A quantitative synthesis. Ment Retard Dev Disabl 10: 193200.
187 Symons FJ, Thompson T (1997) Self injurious behaviour and body site preference. J Intellect Disabl Res 41: 456468. Symons FJ, Wendelschafer Crabb G, Kennedy W, Hardrict R, Dahl N, Bodfish JW (2008) Evidence of altered epidermal nerve fiber morphology in adults with self injurious behavior and neurodevelopmental disorders. Pain 134: 232 237. Symons FJ, Wendelschafer Crabb G, Kennedy W, Heeth W, Bodfish JW (2009) Degranulated mast cells in the skin of adults with self injurious behavior and neurodevelopmental disorders. Brain Behav Im munity 23: 365370. Tanaka O, Sakagami H, Kondo H (1995) Localization of mRNAs of voltagedepedent Ca2+ channels: four subtypes of a1and bsubunits in developing and mature rat brain. Brain Res Mol Brain Res 30: 116. Tausig M (1985) Factors in family decisionmaking about placement for developmentally disabled individuals. Am J Ment Defic 89: 352361. Terbach N, Williams RS (2009) Structurefunction studies for the panacea, valproic acid. Biochem Soc Trans 37: 11261132. Thompson T, Caruso M (2002) S elf injury: Knowing what were looking for. In: Self injurious behavior: genebrainbehavior relationships (Shroeder SR, Oster Granite ML, Thompson T, eds), pp 321. Washington D.C.: American Psychological Association. Thompson T, Symons F, Delaney D, Eng land C (1995) Self injurious behavior as endogenous neurochemical self administration. Ment Retard Dev Disabl Res Rev 1: 137148. Tiefenbacher S, Fahey MA, Rowlett JK, Meyer JS, Pouliot AL, Jones BM, Novak MA (2005) The efficacy of diazepam treatment for the management of acute wounding episodes in captive rhesus macaques. Comp Med 55: 387392. Tiefenbacher S, Novak MA, Marinus LM, Chase WK, Miller JA, Meyer JS (2004) Altered hypothalamic pituitary adrenocortical function in rhesus monkeys (Macaca mulatta) with self injurious behavior. Psychoneuroendocrinol 29: 501 515. Tordjman S, Anderson GM, Botbol M, Brailly Tabard S, Perez Diaz, Graignic R, Carlier M, Schmit G, Rolland A C, Bonnot O, Trabado S, Roubertoux P, Bronsard G (2009) Pain reactivity and plas ma b endorphin in children and adolescents with autistic disorder. PLoS One 4: e5289. Tornatzky W, Miczek KA (1994) Behavioral and autonomic responses to intermittent social stress: differential protection by clonidine and metoprolol. Psychopharmacol 116: 346 356.
188 Torres RJ, Puig JG (2007) Hypoxanthineguanine phosophoribosyltransferase (HPRT) deficiency: LeschNyhan syndrome. Orphanet J Rare Dis 2: 4857. Towle AC, Criswell HE, Maynard EH, Lauder JM, Joh TH, Mueller RA, et al. (1989) Seotonergic innervation of the rat caudate following a neonatal 6hydroxydopamine lesion: an anatomical, biochemical and pharmacological study. Pharmacol Biochem Behav 34: 367374. Turner M (1999) Annotation: Repetitive behaviour in autism: A review of psychological researc h. J Child Psychol Psychiatry 40: 839849. Vandenbulcke F, Nouel D, Vincent J P, Mazella J, Beaudet A (2000) Ligandinduced internalization of neurotensin in transfected COS 7 cells: Differential intracellular trafficking of ligand and receptor. J Cell S ci 113: 29632975. Verhoeven W, Tuinier S, van den Berg Y, Coppus A, Fekkes D, Pepplinkhuizen L, Thijssen J (1999) Stress and self injurious behavior; hormonal and serotonergic parameters in mentally retarded subjects. Pharmacopsychiatry 32: 1320. Vialo u V, Balasse L, Callebert J, Launay J M, Giros B, Gautron S (2008) Altered aminergic neurotransmission in the brain of organic cation transporter 3deficient mice. J Neurochem 106: 14711482. Visser JE, Bar PR, Jinnah HA (2000) LeschNyhan disease and the basal ganglia. Br Res Rev 32: 449475. Vollmer TR, Iwata BA (1992) Differential reinforcement as treatment for behavior disorders: Procedural and functional variations. Res Dev Disabl 13: 393417. Vollmer TR, Sloman KN, Borrero CSW (2009) Behavioral ass essment of self injury. In: Assessing Childhood Psychopathology and Developmental Disabilities (Matson JL, Andrasik F, Matson ML, eds), pp 341369. New York, New York: Springer Science+Business Media. Weld KP, Mench JA, Woodward RA, Bolesta MS, Suomi SJ, Higley JD (1998) Effect of tryptophan treatment on self biting and central nervous system serotonin metabolism in rhesus monkeys (Macaca mulatta). Neuropsychopharmacol 19: 314321. Wenk GL (1995) Alterations in dopaminergic function in Rett syndrome. Neur opediatrics 26: 123125. Wenk GL (1996) Rett syndrome: Evidence for normal dopaminergic function. Neuropediatrics 27: 256259.
189 Wenk GL, Naidu S, Casanova MF, Kitt CA, Moser H (1991) Altered neurochemical markers in Retts syndrome. Neurology 41: 1753175 6. Wenk GL, OLeary M, Nemeroff CB, Bissette G, Moser H, Naidu S (1993) Neurochemical alterations in Rett syndrome. Brain Res Dev Brain Res 74: 6772. Westlund KN, Krakower TJ, Kwan SW, Abell CW (1993) Intracellular distribution of monoamine oxidase A in selected regions of rat and monkey brain and spinal cord. Brain Res 612: 221230. Whitaker Azmitia, PM (2005) Behavioral and cellular consequences of increasing serotonergic activity during brain development: a role in autism? Int J Devl Neuroscience 23: 75 83. White FJ, Wang RY (1984) Pharmacological characterization of dopamine autoreceptors in the rat ventral tegmental area: microiontophoretic studies. J Pharmacol Exp Ther 231: 275280. Widerlov E, Kilts CD, Mailman RB, Nemeroff CB, Mc Cown TJ, Prang e AJ Jr, Breese GR (1982) Increase in dopamine metabolites in rat brain by neurotensin. J Pharmacol Exp Ther 223: 16. Wolf ME (1998) The role of excitatory amino acids in behavioral sensitization to psychomotor stimulants. Prog Neurobiol 54: 679 720. Wo lf ME, Sun X, Mangiavacchi S, Chao SZ (2004) Psychomotor stimulants and neuronal plasticity. Neuropharmacol 47 Suppl 1: 6179. Wong DF, Harris JC, Naidu S, Yokoi F, Marenco S, Dannals RF, Ravert HT, Yaster M, Evans A, Rousset O, Bryan RN, Gjedde A, Kuhar MJ, Breese GR (1996) Dopamine transporters are markedly reduced in LeschNyhan disease in vivo Proc Natl Acad Sci USA 93: 55395543. Wong DF, Ricaurte G, Grunder G, Rothman R, Naidu S, Singer H, Harris J, Yokoi F, Villemagne V, Szymanski S, Gjedde A, Kuh ar M (1998) Dopamine transporter changes in neuropsychiatric disorders. Adv Pharmacol 42: 219223. Xie J X, Tang M, Zhang C (2001) Effect of cholecystokinin 8 microinjection into ventral tegmental area on dopamine release in nucleus accumbens of rats: an in vivo voltammetric study. Prog Neuropsychopharmacol Biol Psychiatry 25: 427 434. Xing L, Karinch AM, Kauffman GL Jr (1998) Mesolimbic expression of neurotensin and neurotensin receptor during stress induced gastric mucosal injury. Am J Physiol Regulatory Integrative Comp Physiol 274: 3845.
190 Yamada K, Matsuzawa H, Uchiyama M, Kwee IL, Nakada T (2006) Brain developmental abnormalities in Prader Willi syndrome detected by diffusion tensor imaging. Pediatrics 118: e442e448. Yin HH, Adermark L, Lovinger D M (2008) Neurotensin reduces glutamatergic transmission in the dorsolateral striatum via retrograde endocannabinoid signaling. Neuropharmacology 54: 7986. Yokoyama C, Okamura H (1997) Self injurious behavior and dopaminergic neuron system in neonatal 6h ydroxydopaminelesioned rat: 1. Dopaminergic neurons and receptors. J Pharmacol Exp Ther 280: 10161030. Zabek M, Sobstyl M, Koziara H, Dzierzecki S (2008) Deep brain stimulation of the right Zafeiriou DI, Vargiami E, Economou M, Gombakis N (2004) Self mutilation and mental retardation: Clues to congenital insensitivity to pain with anhidrosis. J Pediatr 144: 284. Zarcone JR, Iwata BA, Vollmer TR, Jagtiani S, Smith RG, Mazaleski JL (1993) Extinction of self injurious escape behavior with and without ins tructional fading. J Appl Behav Anal 26: 353360. Z u brzycka M, Janecka A (2002) Substance P content in the cerebrospinal fluid and fluid perfusing cerebral ventricles during elicitation and inhibition of trigeminohypoglossal reflex in rats. Brain Res 941: 29 33.
191 BIOGRAPHICAL SKETCH Amber M. Muehlmann received her Bachelor of Arts degree in psychology from San Diego State University in May 2002. She began her graduate work at the University of Florida in 2003 and received her Master of Science degree fr om the behavioral neuroscience program in the Department of Psychology in August 2005. Her postdoctoral research training will be with Dr. Mark Lewis in the Department of Psychiatry at the University of Florida focusing on the neurobiological basis of rest ricted repetitive behaviors.