xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001478813
007 cr mnu|||uuuuu
008 040811s2004 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000460
Husband, Scott Alan.
Anatomy and function of the nucleus accumbens in the pigeon (Columba livia)
h [electronic resource] /
by Scott Alan Husband.
[Tampa, Fla.] :
University of South Florida,
Thesis (Ph.D.)--University of South Florida, 2004.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 167 pages.
ABSTRACT: Relatively little is known about the existence and traits of a possible nucleus accumbens (Acc) region in non-mammals. The current project investigated a likely candidate for such a structure in pigeons, the medioventral (mvMSt) and mediodorsal (mdMSt) parts of avian medial striatum (MSt). The methods employed were threefold: 1) tract-tracing to determine anatomical connections of the MSt; 2) lesion studies to assess MSt's role in a cognitive task (reversal learning); and 3) measuring an immediate-early gene induced protein, ZENK, in striatal regions during courtship behavior in male pigeons. The MSt was found to have many forebrain (amygdala, hippocampus, dorsal thalamus) and midbrain (ventral tegmental area, substantia nigra) connections similar to those of Acc. In addition, differences in connection patterns between mvMSt and mdMSt indicated that mvMSt was comparable to the shell of Acc, while the mdMSt showed characteristics of Acc core. Effects of MSt lesions on pattern discrimination and reversal learning were assessed. Both lesion subjects and controls performed similarly on original discrimination. Furthermore, there were no significant differences in MSt lesioned birds compared to controls. However, there was a tendency for the two groups to make different types of errors. Error patterns indicated that sham-lesioned birds had deficits due to key preference, whereas lesioned birds had fixation on previous reward contingencies (perseverative errors). The performance of the lesioned birds was consistent with Acc lesion effects on reversal learning in mammals. The expression of ZENK in the mvMSt, mdMSt, lateral MSt, and lateral striatum of male birds exposed to either an empty cage or a live female pigeon was quantified. Higher ZENK expression was found in the live pigeon condition for all the striatal structures. However, the degree of difference between live and empty was much higher in the mvMSt and mdMSt than in the other areas. Therefore, mvMSt and mdMSt appear to play a role in anticipatory sexual behaviors, as has been shown in Acc. The anatomical and functional data from the current study indicate that avian mMSt has numerous similarities with mammalian Acc. These findings will contribute to understanding the evolution of mammalian Acc and identifying the functional significance of avian MSt.
Adviser: Toru Shimizu.
t USF Electronic Theses and Dissertations.
Anatomy and Function of the Nucleus Accumbens In the Pigeon ( Columba livia ) Scott Alan Husband A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychology College of Arts and Sciences University of South Florida Major Professor: Toru Shimizu, Ph.D. Gary Arendash, Ph.D. Cynthia Cimino, Ph.D. Cheryl Kirstein, Ph.D. Paul Sanberg, Ph.D. Date of Approval: July 6, 2004 Keywords: avian, cognition, limbic, striatum, ZE NK Copyright 2004, Scott Alan Husband
i T able of Contents List of Figures v List of Tables v ii i List of Abbreviations ix Abstract 1 Chapter One: Introduction 3 Specific Aims 3 S ignificance of the Study 4 Background for Specific Aim #1 Anatomy of Nucleus Ac c umbens 5 Ascending Projections from Midbrain 8 Projections from Thalamus and Hypothalamus 9 Projections from Amygdala, Hippocampus, and Cortex 9 Descending Projections to Midbrain and Ventral Pallidum 12 Probable Location o f the Avian Nucleus Accumbens 13 Ascending Projections from Midbrain 14 Projections from Thalamus and Hypothalamus 16 Projections from Amygdala, Hippocampus, and Pallium 16 Descending Projections to Midbrain and Ventral Pallidum 20 Background for Specific Aim #2 Func tion of Nucleus Accumbens 23 Reversal Learning and Acc 24 P igeons and Dopaminergic Drugs 27 Pigeons and Disruption of MSt Function 28
ii Learning in Chicks and MSt Circuits 30 Avian Reversal Learning Studies 31 Background for Specific Aim #3 A ccumbens and Sexual Behaviors 33 Anticipatory vs. Consummatory Sexual B ehaviors 33 Role of Dopamine in Sexual Behaviors 34 Immediate Early Genes (IEGs): Neu ral Activation / Plasticity 34 Courtshi p Behaviors and IEGs in Birds 37 Chapter Two: Methods 38 Anatomical Study Methods (Tract tracing and CaBPs) 38 Anterograde Tracer In jection 40 Retrograde Tracer Injection 40 Immunohistochemistry for CaBPs 41 Immunohistochemistry for Tract tracing 42 Data Analysis 43 B eh avioral Study Methods 4 4 Apparatus and Pretraining Procedures 44 Surgery and Post operative Testing 45 Discrimination and Reversal Sessions 46 Histology and Lesion Reconstruction 4 7 Behavioral Data Analysis 47 Gene Expression Methods 50 Apparatus and Behavioral Procedures 50 Histology and Immunohistochemistry for ZENK 51 Analysis of ZENK Protein Levels 52
iii Chapter Three: R esults 54 CaBP Results 54 Tract tracing Results 54 Summary of Anatomy Experiments 54 Summary of CTb Injections 5 7 CTb Injections into mvMSt 57 CTb I njections into mdMSt 63 Summary of BDA Injections 6 6 BDA Injection into mvMSt 66 BDA Injection into mdMSt 68 BDA Injection into mM and mN 77 Lesion Effects on Pattern Discrimi nation & Reversal Learning 84 Summary of Behavioral Experiments 84 Experimental Group: Lesion & Wulst Damage Reconstruction 86 Control Group: Wulst Damage Reconstruction 90 Performance on D iscrimination and Reversal Learning 95 Key Preference Analysis 95 Response to Previous S+ Analysis 96 ZENK Protein Expression Results 99 Summary of ZENK Experiments 9 9 Quantitative Analysis of ZENK ir Cells 9 9 md MSt: Analysis of ZENK ir Cells 99 m v MSt: Ana lysis of ZENK ir Cells 101 Lateral MSt: Analysis of ZENK ir Cells 101
iv LSt: ZENK ir Cells 105 Chapter Four: Discussion 107 Comparing MSt and Acc Connections 107 Midbrain, Ventral Pallidum, and Thalamus 108 Amygdala, Hippocampus, and Other Forebrain Areas 11 1 Does MSt Have a Role in Cognitive Functions? 114 MSt: Activation and Plasticity in Sexual Behavior 118 Conclusions 121 Literature Cited 123 Appendix A Neurological Exam 147 Appe n dix B Additional CTb Cases 148 Append ix C Additional BDA Cases 150 Abo ut the Author 154
v List of Figures Figure 1. Location of the dorsal and ventral basal ganglia in pigeon (A) and rat (B). 7 Figure 2. The zinc finger IEG, krox 24. 37 Figure 3. Configuration of the operant chamber and stimulus display/response pan el. 49 Figure 4. A reas sampled for quantitative analysis of ZENK 53 Figure 5. Patterns of CaBP in the MSt. 5 5 Figure 6. Drawings from the stereotaxic atlas of the pigeon (Karten & Hodos, 1967) showing the brain regions of interest in the current stu dy. 5 8 Figure 7 CTb injection into the ventral MSt in subject Pg220. 59 Figure 8 CTb injection into the ventral MSt in subject Pg225. 60 Figure 9 CTb injection into medial MSt in subject Pg194. 61 Figure 10 CTb injection into dorsal MSt in subject Pg18 3. 62 Figure 1 1 CTb injection into posterior MSt in subject Pg180. 65 Figure 1 2 BDA injection into ventral MSt in subject Pg191. 6 7 Figure 1 3 BDA injection into ventral MSt in subject Pg195. 6 9 Figure 1 4 Darkfield photomicrographs of BDA labeled fibers in t he ventral and dorsal VP. 70 Figure 1 5 BDA injection into dorsal MSt in subject Pg199. 71 Figure 1 6 BDA injection into dorsal MSt in subject Pg193. 73
vi Figure 1 7 BDA injection into posterior MSt in subject Pg181 (right hem). 75 Figure 1 8 BDA injection into posterior MSt in subject Pg181 (left hem). 76 Figure 1 9 BDA injection into mM for subject Pg242. 7 8 Figure 20 BDA injection into mM for subject Pg213. 7 9 Figure 2 1 BDA injection into mM for subject Pg215. 80 Figure 2 2 BDA injection i nto mN for subject Pg241 (right hem). 83 Figure 2 3 BDA injection into mN for subject Pg241 (left hem). 84 Figure 2 4 Lesion and Wulst damage reconstruction for subject Pg177. 87 Figure 2 5 Lesion and Wulst damage reconstruction for subject Pg179. 8 8 Figure 2 6 Lesion and Wulst damage reconstruction for subject Pg224. 8 9 Figure 2 7 Lesion and Wulst damage reconstruction for subject Pg228. 91 Figure 2 8 Wulst damage reconstruction for subject Pg223. 92 Figure 2 9 Wulst damage reconstruction for s ubject Pg226. 93 Figure 30 Wulst damage reconstruction for subject Pg233. 94 Figure 3 1 Individual performance on errors to criterion. 95 Figure 3 2 Key preference measures for lesion and control subjects on original learning and the first five reversa ls. 9 7 Figure 3 3 Graph of errors to criteria attributable to responses on the
vii previous sessions S+ stimulus. 9 8 Figure 3 4 Boxplot graphs showing the quantitative analysis of EGR ir cells in areas of the striatum in the Live and Empty Cage conditio ns. 100 Figure 3 5 Photomicrographs of ZENK protein expression in the mdMSt a fter exposure to two different conditions. 102 Figure 3 6 Photomic rographs of ZENK protein expression in the m v MSt a fter exposure to two different conditions. 103 Figure 3 7 Photomicrographs of ZENK protein expression in the lateral MSt a fter exposure to two different conditions. 104 Figure 3 8 Photomicrographs of ZENK protein expression in the LSt a fter exposure to two different conditions. 10 6
viii List of Tables Table 1. Summary of Acc c onnections in mammals. 10 Table 2. Summary of afferents and efferents of mv and md MSt. 56
ix L ist of Abbreviations 2 DG 2 deoxyglucose 6 OHDA 6 hydroxydopamine Aa Arcopallium, pars anterior Acc Nucleus accumbens septi AcC Nucle us accumbens septi, core part AcS Nucleus accumbens septi, shell part Ai Arcopallium intermedium Aid Arcopallium intermedium, pars dorsalis Aiv Arcopallium intermedium, pars ventralis Ap Arcopallium posterior APH Parahippocampal area Av Arcop allium, pars ventralis BDA Biotinylated dextran amine BSTl Bed nucleus of the stria terminalis, pars lateralis CaBP Calcium binding proteins CB Calbindin CDL Corticoid area, pars dorsolateralis CPi Pyriform cortex CPu Caudate putamen CR Cal retinin CTb Cholera toxin, subunit B
x DMA Nucleus dorsomedialis anterior thalami DMP Nucleus dorsomedialis posterior thalami E Entopallium ENK Enkephalin GP Globus pallidus HA Hyperpallium apicale HD Hyperpallium densocellulare HI Hyperpal lium intercalatum IHA Hyperpallium apicale, pars interstitialis Hp Hippocampus IEG Immediate early genes LS Lateral septum LSt Lateral striatum LoC Locus coeruleus MSt Medial striatum mdMSt Medial striatum mediale, pars dorsalis mM Mesopa llium, pars medialis mvMSt Medial striatum mediale, pars ventralis N Nidopallium NCL Nidopallium caudale, pars lateralis NFF Nidopallium frontale, pars frontalis NFL Nidopallium frontale, pars lateralis NFM Nidopallium frontale, pars medialis m N Nidopallium, pars medialis
xi NT Neurotensin OMPFC Orbital and medial prefrontal cortex PFC Prefrontal cortex PoA Nucleus posterior amygdalopallialis PV Parvalbumin R Raphe nuclei SP Substance P SNc Substantia nigra, pars compacta SNr Su bstantia nigra, pars reticulata TO Olfactory tubercle Tn Nucleus taenae TPO Area tempo parieto occipitalis VP Ventral pallidum VTA Ventral tegmental area
1 Abstract Relatively little is known about the existence and traits of a possible nucleus accumbens (Acc) region in non mammals. The current project investigated a likely candidate for such a structure in pigeons, the medioventral (mvMSt) and mediodorsal (mdMSt) parts of avian medial striatum (MSt). The methods employed were threefold: 1) tract tracing to determine anatomical connections of the MSt; 2) lesion studies to assess MSts role in a cognitive task (reversal learning); and 3) measuring an immediat e early gene induced protein, ZENK, in striatal regions during courtship behavior in male pigeons. The MSt was found to have many forebrain (amygdala, hippocampus, dorsal thalamus) and midbrain (ventral tegmental area, substantia nigra) connections similar to those of Acc. In addition, differences in connection patterns between mvMSt and mdMSt indicated that mvMSt was comparable to the shell of Acc, while the mdMSt showed characteristics of Acc core. Effects of MSt lesions on pattern discrimination and reve rsal learning were assessed. Both lesion subjects and controls performed similarly on original discrimination. Furthermore, there were no significant differences in MSt lesioned birds compared to controls. However, there was a tendency for the two groups to make different types of errors. Error patterns indicated that sham lesioned birds had deficits due to key preference, whereas lesioned birds had fixation on previous reward contingencies ( perseverative errors) Th e performance of th e lesioned birds was consistent with Acc lesion effects on reversal learning in mammals.
2 The expression of ZENK in the mvMSt, mdMSt, lateral MSt, and lateral striatum of male birds exposed to either an empty cage or a live female pigeon was quantified. Higher ZENK expression w as found in the live pigeon condition for all the striatal structures. However, the degree of difference between live and empty was much higher in the mvMSt and mdMSt than in the other areas. Therefore, mvMSt and mdMSt appear to play a role in anticipatory sexual behaviors, as has been shown in Acc. The anatomical and functional data from the current study indicate that avian mMSt has numerous similarities with mammalian Acc. These findings will contribute to understanding the evolution of mammalian Acc and identifying the functional significance of avian MSt.
3 Chapter One Introduction This studys main hypothesis was that the avian basal forebrain contains a region anatomically and functionally equivalent to the mammalian nucleus accumbens (Acc). Although Acc is considered to play an important role in associative learning, attention, and the evaluation of contextual stimuli, the equivalent structure in the avian brain had not been identified precisely. Input from dopaminergic midbrain nuclei (e.g., substan tia nigra and ventral tegmental area) and numerous limbic forebrain areas (e.g., hippocampus, amygdala, and prefrontal and orbitofrontal cortices) anatomically characterize the mammalian Acc. The targets of its primary output are the ventral pallidum (VP ) and the dopaminergic midbrain nuclei. Functionally, lesions of the Acc are known to cause significant deficits in tasks requiring attention to the environment, such as changing reward contingencies associated with reversal learning tasks. The Acc is also characterized by the expression of immediate early genes and their associated proteins under conditions of arousal, indicating that neurons in the Acc demonstrate synaptic plasticity and play a role in learning and memory. Specific Aims The goal of the pr oject was to use anatomical, behavioral, and neurochemical approaches to establish the detailed connections and function of the avian Acc region. To identify an Acc structure in birds, the project had three specific aims: 1) SPECIFIC AIM # 1 Determine th e specific location and extent of the avian Acc by using tract tracing techniques;
4 2) SPECIFIC AIM # 2 Identify the function of the avian Acc by using targeted brain lesions and subsequent sensory and cognitive behavioral tests; and 3) SPECIFIC AIM # 3 Determine patterns of immediate early gene induced protein expression under conditions of arousal in the Acc and hodologically related areas using immunohistochemical techniques. Significance of the Study This project provides the foundation for future st udies of an avian model of Acc function. Given the controversy over what precise role the Acc plays in learning and behavior, a comparative approach can be applied to understand its basic function in amniotes (reptiles, birds, and mammals). Of particular i nterest to the current project was to identify the telencephalic zones providing visual information to an avian Acc area. Circuits of the avian visual pathways have been shown to have an independent and parallel system that may subserve different aspects o f visual function (Wang et al., 1993; Husband & Shimizu, 1999; Laverghetta & Shimizu, 2001). However, there is little information on areas able to provide visual input to striatum, whereby visual input may influence attention, motivation, and behavior. Pre liminary data from our lab show several possible areas of the visual system that may contribute information to the avian basal forebrain. Finally, the findings of this project provide important insights regarding the evolution of the basal ganglia. If the presence of Acc in birds is confirmed, then the limbic circuits of the ventral basal ganglia can be traced back at least as far as the common reptilian ancestor of both birds and mammals approximately 300 million years ago. Therefore, Acc circuits (having been selected for and maintained in various species
5 of amniotes) are an efficient way to turn limbic, motivational information from the external and internal milieu into adaptive behavioral responses. Background for Specific Aim #1 Anatomy of Nucleus Acc umbens In mammals, the Acc is a prominent part of the ventral striatum, which encompasses a large ventromedial region extending from the anterior commissure to the rostral pole of the caudate nucleus (see Figure 1B). There is no readily apparent boundary b etween the dorsal and ventral striatal territories, either upon gross anatomical inspection or in Nissl stained brain sections. However, some differing morphological characteristics between these striatal regions have been described. For example, the ventr al striatum contains smaller, more densely packed neurons and lacks the patch matrix organization that is a prominent feature of the dorsal striatum (Haber & McFarland, 1999 ). The Acc does have some patches of strong opiate receptor binding similar to that found in the caudate putamen (CP) but there is no clear relationship of such patches to the organization of similar structures in the dorsal striatum (Gerfen, 1992). Other differences between the ventral and dorsal striatum include the lack of prominent fiber bundles in the Acc that are seen in the CP and the ventral striatums demonstration of pallidal elements that the dorsal striatum lacks ( Heimer et al., 1995). Projection neurons make up 90 95% of striatal neurons in the rat ( Heimer et al., 1995) The large number and high packing density of these medium sized, spiny neurons contributes to the striatums homogenous appearance. These neurons possess perikarya ranging between 12 and 18um and have both axons leaving the striatum and local collaterals t hat remain near the axon of origin ( Heimer et al., 1995) Most all of these projection neurons are probably GABAergic, co localized with either SP or ENK (Gerfen, 1992). These different populations of projection neurons also appear to have
6 differing anatom ical connections. The SP/GABA neurons tend to be striatonigral, while ENK/GABA neurons tend to be striatopallidal (Gerfen, 1992). Intrinsic or interneurons comprise the remaining approximately 10% of striatum (Gerfen, 1992). These neurons are either large or medium sized aspiny neurons. The large aspiny neurons use acetylcholine as a neurotransmitter. There are several varieties of medium aspiny neurons based on their peptide and protein content (e.g., somatostatin, neuropeptide Y, parvalbumin) (Gerfen, 19 92). These substances are most often co localized with GABA ( Heimer et al., 1995; Reiner et al., 1998). Just ventral to the Acc is the medium celled portion of the olfactory tubercle (TO), another component of the ventral striatum. Since both the Acc and T O get extensive projections from hippocampus, amygdala, and limbic related cortical and thalamic areas, they have been referred to as limbic striatum (Nauta & Domesick, 1984; Groenwegen et al., 1999). In mammals, the Acc can be differentiated into AcS an d AcC territories with different anatomy and neurochemistry (Zahm, 2000). The AcS area (AcS) is the medial and ventral aspect of the nucleus, whereas the AcC (AcC) is located more dorsolaterally, adjacent to the CP (see Figure 1B). In general, the AcC shar es more characteristics with the adjacent dorsal striatum territories than the AcS, and is often characterized as a transitional zone between the more limbic (AcS) and motor (caudate putamen) domains of striatum. Calcium binding proteins are widely used ma rkers for differentiating the AcS and AcC across species (Haber & McFarland, 1999). The AcS is characterized by light to moderate calbindin like immunoreactivity, while the AcC exhibits more intense immunoreactivity (Groenwegen et al., 1999). Conversely, a nother calcium binding
7 protein, calretinin, shows stronger immunoreactivity in the AcS than AcC (Tan et al., 1997). There are also several other differences in neurochemical distributions between AcS and AcC, including substance P (SP), neurotensin (NT), a nd enkephalin (ENK), and several receptor subtypes for glutamate, GABA, and opioids (Haber & McFarland, 1999; Zahm, 1999). The AcS is particularly dense in SP and NT, whereas ENK is found at higher levels in the AcC (Haber & McFarland, 1999; Zahm, 1999). T he AcS is denser in the glutamate receptor subtype GluR1 (i.e., ampakine; AM PA ). Conversely, the AcS exhibits weaker immunoreactivity for GluR4 than the AcC (Haber & McFarland, 1999). The AcC is denser than AcS in the GABA A receptor subtype (Haber & McFarl and, 1999; Zahm, 1999). For the mu opioid receptor subtype, receptor binding studies show patchy Figure 1. Location of the dorsal and ventral basal ganglia in pigeon (A) and rat (B) Dorsal or somatomotor striatum and pallidum are represented in lig ht gray ; ventral or limbic striatum and pallidal areas are represented in dark gray. The precise boundaries of somatomotor and limbic MSt regions are still controversial, and were investigatged in the current project.
8 patterns in most of Acc, but with p articularly high densities within parts of the AcS (Haber & McFarland, 1999). Given the large dopaminergic input to Acc, it is not surprising that there are s everal types of dopamine (DA) receptor expressed in this region. These receptor subtypes also show differences in density between AcS and AcC. There are two broad classes of DA receptors; D 1 receptors consisting of the D 1 and D 5 subtypes, and D 2 receptors consisting of the D 2 D 3 and D 4 subtypes (Missale et al., 1998). The D 1 receptor subtype is heavi ly expressed in both AcS and AcC of mammalian Acc. The D 3 subtype has higher densities in the AcS (Zahm, 1999), while the D 2 tends to be higher in the AcC (Missale et al., 1998). In contrast, the D 5 receptor has been shown to exhibit extremely low levels, at least in rodent brain (Missale et al., 1998). The mammalian Acc receives input from the brainstem, dopaminergic midbrain nuclei, and numerous areas of the diencephalon and telencephalon. The afferent connections of mammalian AcS and AcC are detailed bel ow, and are summarized in Table 1. Ascending Projections from Midbrain The Acc receives prominent dopaminergic mesencephalic projections from the substantia nigra, pars compacta (SNc), the ventral tegmental area (VTA), and the retrorubral group (A8) (see F igure 1A). In terms of Acc subterritories, the AcS receives projections from SNc, but proportionally much more input from VTA (Groenwegen et al., 1999). The input from VTA is particularly concentrated in the medial and ventral parts of the AcS. In contrast the AcC receives lighter projections from VTA compared to the AcS, with progressively more input from the SNc as one goes from AcS to the AcC. In
9 addition, the A8 cell group sends efferents primarily to the AcS (Groenwegen et al., 1999). From brainstem, both the serotonergic raphe nuclei and noradrenergic locus coeruleus (LoC) send projections to the Acc. The median raphe nuclei send efferents primarily to the AcS, while the dorsal raphe nuclei send fibers to both AcS and AcC (Zahm, 2000). The projections originating in the LoC terminate primarily in the AcS (Zahm, 2000). Projections from Thalamus and Hypothalamus In the dorsal thalamus, nuclei of the medial nuclear group (medialis dorsalis and parataenial nuclei) and medial parts of the intralaminar nucle i send efferents to both the AcS and AcC (Zahm, 2000). Thalamic projections to AcS and AcC also arise from many of the midline thalamic nuclei, including the paraventricular, interomediodorsal, reuniens, rhomboid, and posteromedian nuclei (Elena Erro et al ., 2002). In addition, medial portions of the centromedian parafascicular complex also send projections to AcC (Haber & McFarland, 1999). In the ventral thalamus, a minor projection from the medial subthalamic nucleus (usually associated with motor functio n) has also been reported to the AcC (Groenewegen & Berendse, 1990). Finally, the lateral hypothalamus also sends efferents to Acc, specifically to the AcS (Zahm, 2000). Projections from Amygdala, Hippocampus, and Cortex The mammalian Acc receives projecti ons from various nuclei of the amygdala. These projections primarily arise in the basolateral nucleus, with some projections also coming from the cortical and posterior nucleus (Swanson & Petrovicha, 1999; Zahm, 2000). Rostral areas of the basolateral nucl eus send efferents to both AcS and AcC, with somewhat more dense projections to the AcC than the AcS ( Kita & Kitai, 1990). Caudal
10 areas of the basolateral nucleus send projections primarily to the AcS, with only minor projections to the AcC (Heimer et al., 1997). The medial zone of the cortical nucleus and the posterior nucleus send projections to both the AcC and AcS. The posterior nucleus (also referred to as the cortico amygdaloid transition area) is a major source of efferent TABLE 1. Summary of Acc c onnections in mammals. Connections of the Acc AcS and AcC areas are represented in the table below. Symbols indicate relative density of projections: ++ dense, + moderate, none. Acc Acc Afferents (Projections to Acc) A cS AcC Efferents (Projections from Acc) AcS AcC VP ventromedial ++ ++ VP ventromedial ++ VP dorsolateral ++ VP dorsolateral ++ BnST ++ Lateral Hypothalamus ++ + Cortical n. + + POA ++ + Posterior n. + + VTA (A10) ++ + Hippocampus ++ + SNc (A9) ++ ++ Septum ++ SNr + La teral Hypothalamus ++ + Retrorubral (A8) + POA ++ Subthalamic n ++ VTA (A10) ++ + Peri aqueductal Grey + + SNc (A9) + ++ Orbitofrontal Ctx + + SNr ++ Prefrontal Ctx + + Retrorubral (A8) ++ + Medial thalamic n. ++ ++ Midline thalamic n. + + ++ Centromedian PF + Subthalamic n ++ LoC ++ Raphe ++ ++ Orbitofrontal Ctx ++ + Prefrontal Ctx + + Piriform Ctx ++ Anterior Cingulate Ctx ++ + Entorhinal Ctx ++ +
11 projections through the stria terminalis, a branch of which is associated with several limbic areas, including Acc, ventral lateral septal nucleus, TO, and infralimbic areas of the p refr ontal cortex (Canteras et al., 1992).In mammals, hippocampal regions also send projections throughout the Acc. The subicular area, t he primary output center of the hippocampus, sends major projections to the AcS, with minor projections extending into the AcC ( Kelley & Domesick, 1982; Groenewegen et al., 1987) The same pattern generally holds for projections from the CA1 area ( Kelley & Domesick, 1982; Groenewegen et al., 1987) Tracer studies employing injections into the fimbria tract result in dense anterograde labeling along the entire length of the Acc, with the AcS showing particularly dense terminal labeling (Kelley & Domesick, 19 82). There are numerous cortical areas, usually associated with the medial prefrontal and orbitofrontal cortices, which send efferents to the mammalian Acc. These include the entorhinal cortex (Broadmans area 28), perirhinal cortex (area 35), and a number of regions that are collectively referred to as the orbital and medial prefrontal cortex (OMPFC). The OMPFC includes the anterior cingulate area (area 24), medial orbital (areas 14, 25, and 32), and orbital prefrontal cortices (areas 11, 12, and 13) (Naut a & Domesick, 1984; Haber & McFarland, 2001). The entorhinal cortex is associated with both AcC and AcS, with both medial and lateral entorhinal cortex projections to the AcS, and somewhat denser projections from the lateral part to just the AcC (Zahm, 200 0). The medial orbital cortex and the anterior cingulate area send projections to all of the Acc, with particular concentration of anterior cingulate area efferents in the AcS (Haber & McFarland, 1999; Haber et al., 2000; Zahm, 2000). Much of the orbital p refrontal regions projections overlap those of the medial orbital cortex and anterior cingulate
12 (Haber et al., 2000). Dorsal and ventral parts of agranular insular cortex send projections to either the AcC or AcS, respectively (Zahm, 2000). In addition to being a primary recipient of Acc efferents, the VP also sends projections back to the Acc. These connections show a general topographic relationship. Ventromedial portions of the VP send efferents to both the AcS and AcC, while the dorsolateral VP primary sends projections to the AcC (Zahm, 2000). Desecending Projections to Midbrain and Ventral Pallidum Compared to the numerous areas sending projections to the Acc, there are only a few structures in receipt of Acc efferents. Neurons of the Acc send project ions primarily to the midbrain dopamine (DA) areas and to the VP. The efferent connections of mammalian AcS and AcC are summarized in Table 1. The AcS sends efferents to both the VTA and SNc (Haber et al., 2000). These projections show a topographic relati onship, with portions of the medial AcS projecting to medial VTA, and the ventral portion of the AcS to lateral VTA. In addition, the AcS sends efferents to the retrorubral area and the periaqueductal gray (Haber et al., 2000; Zahm, 2000). The AcC sends ef ferents primarily to the substantia nigra, with most of these terminating in the SNc and somewhat fewer in the reticulata (SNr) (Haber et al., 2000; Zahm, 2000). A primary recipient of Acc efferents in mammals is the VP. A reciprocal, topographic relations hip exists between the Acc and VP (Groenwegen et al., 1999; Zahm, 2000). Efferents from the medial and lateral AcS project to ventromedial and ventrolateral VP, respectively. The AcC neurons send projections to the dorsal part of VP (Groenwegen et al., 199 9). Projections from Acc to the VP are primarily GABAergic (Pycock & Horton, 1976; Jones & Mogensen, 1980). The existence of a GABAergic pathway between the Acc and
13 VP is substantiated by the fact that stimulation of Acc results in suppression of many pal lidal neurons (Mogensen et al., 1980). In addition, this effect can be attenuated or blocked by iontophoretic application of picrotoxin (a GABA antagonist). This treatment can increase firing rates in approximately 60% of pallidal neurons when picrotoxin i s injected into Acc (Jones & Mogenson, 1980). Furthermore, there are smaller projections to the preoptic area, lateral hypothalamus, entopeduncular nucleus, retrorubral area, and midbrain reticular formation (Zahm, 2000). The preoptic area and lateral hypo thalamus receive projections that come from the AcS, while the AcC almost exclusively is the source of projections to the entopeduncular nucleus (Heimer et al., 1991). The AcC is the primary source of inputs to the retrorubral area and midbrain reticular f ormation; the AcS sends minor, diffuse projections to these areas (Heimer et al., 1991). As detailed above, the neural connections in mammals have been extensively studied because of its demonstrated importance in learning and behavior. The convergence of numerous projections from a variety of brain regions from midbrain, thalamus, cortex, and many subtelencephalic limbic regions, in addition to its relatively small number of output structures (e.g., midbrain and VP), indicate that Acc serves an important i ntegrative function and uses this information to direct motor outcomes/behavior. Probable Location of the Avian Nucleus Accumbens The portion of the basal forebrain called the Acc in the standardized atlas of the pigeon brain (Karten & Hodos, 1967) is pro bably incorrect. At the time the atlas was produced, few connectional studies had been done on this region, and many areas were identified based on an estimate of its location comparable to mammalian brain. The structure labeled Acc in the atlas has subseq uently been shown to have relatively poor
14 DA innervation (Reiner et al., 1994), making it an unlikely Acc equivalent. This area probably corresponds to the lateral bed nucleus of the stria terminalis (BSTl) in mammals, because the BSTl of mammals and the Acc structure in the Karten and Hodos (1967) atlas are distinguished by a scarcity or low density of SP, ENK, and NT fibers and neurons (Reiner et al., 1983; Reiner et al., 1984b; Reiner & Carraway, 1987; Anders on & Reiner, 1990). In addition, minimal inp ut has been found to this area in birds from the lateral hypothalamic area (Berk & Hawkin, 1985). Ascending Projections from Midbrain The two primary components of the avian striatum are the medial striatum (MSt) and the lateral striatum (LSt); the LSt is the avian equivalent of the mammalian CP In Nissl stained sections it is difficult to distinguish MSt and LSt. However, there is a tendency for the MSt to contain fewer large and medium sized cells like those found in LSt (Karten & Dubbeldam, 1973). The a vian Acc is most likely present along the medial aspect of the MSt, which is located at the rostromedial part of the basal telencephalon. The MSt itself appears relatively homogenous in Nissl stained sections. However, w ithin the MSt t wo general groups of projection neurons have been identified based on their spine density. In an extensive Golgi study in the chick (Tmbl, 1995) a population of aspiny neurons were found throughout the MSt Among spiny neurons, there w ere three main types The first were lar ge projection neurons with angular appearing perikarya of around 20 25um in diameter. These neurons had four to six primary dendrites possessing a high spine density. The second type w as medium sized neurons with a round or ovoid appearance possessing diam eters of 18 20um. These neurons only had a moderate number of spines. The third type was small, spiny projection neurons that had ovoid perikarya with diameters of 12 15um
15 (Tmbl, 1995) The p rojection neurons in both mammals and birds contain GABA; appro ximately half of these neurons have GABA co localized with SP or dynorphin, while the other s contain both GABA and ENK (Reiner et al., 1998). In both birds and mammals about 5 10% of striatal neurons are interneurons (Reiner et al., 1998). In the MSt, two primary types of interneurons have been described (Tombol, 19 95 ). One type is larger with varicose axonal branches, while the second type is moderately sized and possesses fine axonal branches with few spines. In terms of neurochemical contents, there are generally three types recognized in both the striatum of birds and mammals: 1) large, aspiny cholinergic neurons; 2) medium sized, aspiny neurons which contain somatostatin or neuropeptide Y; or 3) medium sized, aspiny neurons which contain both GABA and either the calcium binding protein PV or LANT 6 (a peptide related to neurotensin) (Reiner et al., 1998). The medial MSt is characterized by intense DA, tyrosine hydroxylase, SP, ENK, and NT, that are also abundant in the mammalian Acc. However, detailed c onnections of the medial MSt have not been well studied. As in the mammalian striatum, the avian MSt receives projections from the avian homologues of the VTA and SNc. In the pigeon, Kitt and Brauth (1986a, b) showed ascending projections to MSt from LoC, VTA, and the SNc. Projections from the substantia nigra may have also included those from SNr, which is not as well defined in birds as in mammals (Karten & Dubbeldam, 1973). These projections were heavier to the ipsilateral side of the brain, with sparse labeling observed in the contralateral hemisphere. Efferents from the VTA and SNc were also shown to terminate in several other basal telencephalic structures (such as the LSt, the lateral and medial septal nuclei, TO, and VP). As in mammals, projections t o the avian caudate putamen were particularly dense from SNc, while rostromedial portions of
16 caudate putamen and MSt received proportionally more fibers from the VTA, although a significant overlap of these projections was noted (Kitt & Brauth, 1986b). Pro jections from Thalamus and Hypothalamus The avian dorsal thalamus is a prominent source of thalamostriate projections, most of which are ipsilateral. A distinction between medial and lateral MSt is supported on the basis of different projection patterns fr om the dorsal thalamus to striatum. The medial MSt receives projections from more medially located dorsal thalamic nuclei, namely the anterior and posterior dorsomedial nuclei (DMA and DMP; Wild, 1987). More evidence for a limbic characterization of medi al, rather than lateral, MSt is that only medial MSt receives projections from neurons of the lateral hypothalamus (Berk & Hawkin, 1985). In avian brain there is a mammilary like region of hypothalamus named the lateral mammilary nuclei. Fibers from this a rea ascend through the medial forebrain bundle and terminate primarily in the VP and TO, with some fibers also found within rostral MSt (Berk & Hawkin, 1985). Projections from Amygdala, Hippocampus, and Pallium Another major input to MSt is from a telence phalic structure called the arcopallium. In birds, current data indicate that the medial and lateral MSt both share projections from various aspects of the arcopallium. The avian arcopallium, either partially or in toto is thought to be comparable to the mammalian amygdala (Zeier & Karten, 1971; Veenman et al., 1995). The arcopallium can be cytoarchitectonically divisible into six divisions: anterior (Aa), medial (Am), intermediate (Ai), posterior nucleus of the pallial amygdala (PoA), and the nucleus taen ae (Tn) (Zeier & Karten, 1971). Both the LSt and MSt receive extensive projections from the arcopallium, particularly the Ai and Aid (Zeier & Karten, 1971; Veenman et al., 1995), which have been considered
17 somatic parts of arcopallium (Zeier & Karten, 1971 ). The Am, Aiv, and Ap project to medial MSt, BSTl, and TO (Veenman et al, 1995). The TO receives more projections from Aiv, while the medial MSt region and BSTl received more from the Am, Aiv, and Ap (Veenman et al., 1995). In addition, the Tn has been sh own to send projections to a restricted portion of the anterior and medial MSt (Cheng et al., 1999). These regions of arcopallium have been associated with visceral and limbic function due to its extensive connections with the hypothalamus (Zeier & Karten, 1971; Cheng et al., 1999). Similar patterns of connections have been found in the duck and the chick, further suggesting different limbic and somatomotor regions of arcopallium (Davies et al., 1997; Dubbeldam et al, 1997). The MSt also receives a projecti on from the avian hippocampal formation. The dorsomedial pallium in birds contains a V shaped structure of d ensely packed neurons designated as the hippocampus (Hp) and a more diffusely scattered area of neurons called the parahippocampal area (APH) (But ler & Hodos, 1996). The avian hippocampus plays an important role in spatial memory and landmark based navigation tasks, similar to findings in mammalian hippocampus (Bingman et al., 1984; Bingman et al., 1990; Gagliardo et al., 1999; White et al., 2002). The avian Hp is probably homologou s to Ammons horn and the APH region to dentate gyrus, with adjacent parts of the pallium probably homologous to the subicular and entorhinal cortices (Butler & Hodos, 1996). Projection patterns of the pigeon hippocampal fo rmation show efferents to the more medial than lateral areas of MSt (Veenman et al., 1995). Injections of a retrograde tracer into medial MSt showed labeled cells in the hippocampal formation, especially in the ventral APH (Veenman et al., 1995). More late rally placed injections into MSt found no
18 labeled cells in the hippocampus (Veenman et al., 1995). Similar patterns of connectivity have been found in the zebra finch (Szekely & Krebs, 1996). Veenman and colleagues carried out the most extensive studies o f avian afferent cortical or pallial sources to striatum (Veenman et al., 1995). In birds, the pallial region includes the structure formally called neostriatum, which was erroneously named over 100 years ago based on an incorrect comparison of most of t he avian forebrain to mammalian corpus striatum (Karten & Dubbeldam, 1973; Karten, 1991). Today, this structure has been renamed nidopallium (nested pallium) and it has been clearly shown to be a pallial structure based on hodology, neurochemistry, and d evelopmental gene expression patterns (Karten & Dubbeldam, 1973; Karten & Shimizu, 1991; Veenman et al., 1995; Puelles et al., 2000). The MSt receives projections from areas in the frontal nidopallium, caudal nidopallium, and a region at the edge of the pa llium, the area temporo parieto occipitalis (TPO). Several of these regions are potential sources of visual input to the MSt. The frontal (NFF), medial (NFM), and lateral (NFL) portions of the frontal nidopallium send projections to the medial MSt (Veenman et al., 1995). However, these connections were not exclusive to medial MSt, having also been found after injections of retrograde tracers involving the lateral MSt (Veenman et al., 1995). The NFL area has been shown to receive projections from the entopal lium, the primary telencephalic station of the collothalamic visual pathway (Veenman et al., 1995; Husband & Shimizu, 1999). The collothalamic visual pathway follows a route from the retina, optic tectum (TeO), thalamic nucleus Rotundus (Rt), and the entop allium (Karten & Revzin, 1966; Karten & Hodos, 1970; Butler & Hodos, 1996). At the dorsolateral edge of the avian pallium, the TPO has been shown to send a diffuse input to the med ial MSt (Veenman et
19 al., 1995). The TPO has also been shown to receive visual input from the entopallium and immediately surrounding nidopallium (Husband & Shimizu, 1999). Efferents to MSt also come from the nidopallium caudale, pars lateralis (NCL), a reg ion thought to be analogous to mammalian PFC. This comparison is based on hodological, neurochemical, and behavioral similarities with the mammalian PFC ( Kr ner & G nt rk n, 1999) Thus, NCL receives multiple sensory inputs from telencephalic and thalamic areas (Ritchie, 1979; Husband & Shimizu, 1999; Wild et al., 1993; Leutgeb et al., 1996; Kr ner & G nt rk n, 1999), dense DA innervation (Divac et al., 1985; Waldmann & G nt rk n 1993), and plays a role in working memory, reversal learning, and go/no tasks (Gagliardo et al., 1996; Gagliardo et al., 1997; G nt rk n 1997; Hartmann & G nt rk n 1998). The NCL is another structure receiving visual input from the entopallium (Husband & Shimizu, 1999). Krner and Gntrkn (1999) described injections of an anter ograde tracer into the rostral NCL that densely labeled lateral parts of MSt and large parts of the LSt. Injections into the caudal NCL, however, resulted in labeling in medial MSt and BSTl. Finally, regions adjacent to the NCL in the caudal nidopallium su ch as the dorsolateral corticoid area (CDL) and pyriform cortex (CPi), also send projections to MSt. The CDL sends efferent fibers to both medial and lateral MSt (Veenman et al., 1995), whereas the CPi projects extensively to medial MSt, TO, BSTl, and VP ( Bingman et al., 1994; Veenman et al., 1995). Another important source of projections, with special relevance to finding sources of visual input to medial MSt, is the area called the Wulst (German for bulge), an enlarged portion of the dorsal avian pall ium comparable to mammalian dorsal pallium. The Wulst consists of four primary subdivisions. These are (from dorsal to ventral): 1)
20 hyperpallium apicale (HA), 2) nucleus interstitialis hyperpallium apicale (IHA), 3) hyperpallium intercalatum (HI), and 4) h yperpallium densocelluare (Reiner & Karten, 1983). Specifically the ventral aspect of HA sends projections to the medial MSt (Veenman et al., 1995). In contrast, more dorsal HA and rostromedial HA project toward the lateral aspects of MSt (Shimizu et al., 1995). The HD, in particular the medial and ventromedial areas, was also found to project to medial but not lateral MSt (Veenman et al., 1995). Portions of NFL receive projections from parts of the Wulst that also receive thalamic input from the principal optic nucleus of the dorsal thalamus as part of the lemnothalamic visual pathway (Shimizu et al., 1995; Husband & Shimizu, 1999). In addition, the mesopallium (M) also sends projections to MSt (Veenman et al., 1995). Past studies have indicated that numero us structures are afferent sources of the avian MSt. The limbic character of many of these regions (e.g., avian hippocampal formation, amygdala, lateral hypothalamus, et cetera ) provides strong evidence that the avian striatum, and MSt in particular, pro bably contributes to more than just somatomotor functions. However, most of the data concerning input to the MSt has been indirect, i.e, anterograde tracer injections of other structures. Relatively few studies have directly investigated MSt connections, e specially in terms of refining different functional regions with the structure. Descending Projections to Midbrain and Ventral Pallidum In birds, projections from MSt back to the midbrain DA centers of the VTA and SNc have been described (Kitt & Brauth, 19 81; Hall et al., 1984; Anderson & Reiner, 1991; Szekely et al., 1994; Mezey & Csillag, 2002). In chick, striatonigral and striatotegmental descending pathways originate in overlapping regions of the MSt, while the LSt and GP demonstrated projections exclus ively to the SNc (Mezey & Csillag,
21 2002). There are, however, different patterns of these striatonigral and striatotegmental projections to MSt. Medial MSt injections of BDA in the chick show labeled fibers in the VTA, with very few in the SNc. In contrast more lateral injections revealed much more labeling of fibers in the SNc with no fibers in VTA (Mezey & Csillag, 2002). Portions of the MSt project ventrally to a previously unnamed region of the basal forebrain in the pigeon stereotaxic atlas (Karten & Hodos, 1967). This region, based on connectional and histochemical criteria, has been argued to be analogous to mammalian VP (Medina & Reiner, 1999). The VP is a major recipient of MSt efferents. The MSt, as well as TO and BSTl, sends efferents to the VP ( Kitt & Brauth, 1981; Hall et al., 1984; Anderson & Reiner 1990, 1991). An early study by Karten & Dubbeldam (1973) investigated connections of LSt, GP, and MSt using lesions and Fink Heimer staining methods to look at patterns of degenerating axons and ter minals. Axons were found which took a caudally projecting course, passing ventral to the inferior margin of MSt, and contributing to the formation of the avian medial forebrain bundle, called the fasciculus prosencephali medialis in birds. These fibers ter minate in the anterior hypothalamus and lateral preoptic area. An Acc equivalent structure in birds has only been indirectly indicated from connection studies of avian midbrain, thalamus, striatum, and telencephalon. Some of these studies have used older g enerations of tract tracers or employed large injections of the entire MSt region. For example, the injections of MSt performed by Veenman and colleagues (1995) included most of the probable Acc region, and the BSTl, with spread of tracer to NFM. Similarly the lateral MSt injections included some spread to TO, globus pallidus (avian dorsal pallidum), VP, and medial parts of the nidopallium intermedium (Veenman et al., 1995). Therefore, no systematic attempt has been made
22 to study medial versus lateral MSt connections or identify potential AcS and AcC subregions of the presumed avian Acc. Detailed analysis of medial versus lateral MSt connections may delineate such differences (Mezey & Csillag, 2002). Of particular interest to the current study was to identi fy the telencephalic zones providing visual information to avian Acc. Circuits of the avian collothalamic visual pathway have been shown to have a rostro caudal topography that may subserve different aspects of visual function (Wang et al., 1993; Husband & Shimizu, 1999; Laverghetta & Shimizu, 2001). Physiological data suggest segregation and parallel processing of different attributes of visual stimuli within Rt. Functionally distinct subpopulations of Rt neurons have been identified, with a dorsal anterio r portion of Rt having high proportions of color sensitive cells, a small central zone with neurons responsive to overall luminance, and the posterior aspect containing neurons that were primarily motion sensitive (Wang et al., 1993). This parallel, functi onal segregation may extend to higher areas in the processing of visual information, as indicated by projection patterns of ectostriatal neurons to other telencephalic areas (NFL, TPO, NCL) (Ritchie, 1979; Husband & Shimizu, 1999). Preliminary data from ou r lab show several potential areas of both collothalamic and lemnothalamic systems that may contribute visual input to MSt, including Wulst, HV, NFL, and NCL. Since the Acc in mammals receives highly processed sensory input, i.e., from structures several synapses removed from the sensory apparatus and primary sensory processing regions in the brain (Zahm, 2000), it is important to identify regions of visual input to the presumptive avian Acc region. The first specific aim of this project clarified the sou rces of input to the medial MSt, to enable further study of its role in behavior and comparability to mammalian Acc.
23 Background for Specific Aim # 2 Function of Nucleus Accumbens The characterization of the Acc as the pleasure center of the brain has attracted wide attention. Indeed, the Acc plays an important role in evaluating the reinforcing properties of natural reinforcers such as drinking, eating, and sexual behavior, as well as artificially conditioned stimuli (Olds & Forbes, 1981). Natural rein forcers such as food, water, or sex elevate DA efflux in Acc, as do artificial ones like drugs of abuse (Blackburn et al., 1992; Koob & LeMoal, 1997; Koob & Nestler, 1997). Many species will self administer DA agonists like amphetamine and cocaine, and i ntra Acc injections of DA itself have reinforcing effects (Hoebel et al., 1983; Goeders et al., 1985). However, there is accumulating evidence to challenge the pleasure center conception of Acc function and its role in the reward side of instrumental lea rning. Changes in DA efflux in Acc have been reported with animals exposed to aversive stimuli, similar to the increase seen upon exposure to rewarding stimuli (Cabib et al., 1988; Salamone, 1994; Rada & Hoebel, 2001; Becerra et al., 2001). In addition, no vel (i.e., initially neutral) stimuli increase DA levels in Acc; such elevations are not seen in response to familiar stimuli that have not attained any direct significance (Blackburn et al., 1992; Rebec et al., 1997). The Acc may play a role in enabling the formation of stimulus behavior reward associations via an attentional gating function (Braff & Geyer, 1990; Gray, 1995). Changes in accumbal DA may increase sensory awareness to both naturally appetitive stimuli and to novel stimuli which could have s ubsequent survival value. The DA tone in Acc may regulate thalamo cortical information processing. One hypothetical example of the working of this circuitry is as follows: 1) elevations of DA in the Acc would inhibit its GABAergic projections to the VP, re leasing VP from normally tonic inhibition, 2) this
24 disinhibition of VP neurons would allow its own GABAergic projections to suppress thalamic neurons, possibly the reticular nucleus, and 3) this suppresses activity of reticular thalamic neurons, which norm ally have a tonic inhibitory effect on other thalamic nuclei (Ferry et al., 2000). The end result of increased DA in Acc, therefore, would be to release dorsal thalamic nuclei from inhibition, enhancing sensory flow to cortex of important environmental sti muli. Reversal Learning and Acc Reversal learning tasks are useful in studying attentional processes isolated from sensory deficits. In reversal learning, two stimuli (e.g., two colors, two shapes) are presented and one is arbitrarily assigned as the rewa rded stimulus, or S+, while the other is non rewarded or punished, the S stimulus. This original learning requires perceptually discriminating the stimuli, and then associating these stimuli with either reward or non reward (MacKintosh, 1974). Success ful performance in this stage indicates a functioning sensory system and intact ability to learn the relationship between stimuli and instrumental outcomes. Upon achieving the discrimination criteria, the relationship between S+ and S are reversed, the nu mber and type of errors is analyzed to determine successful performance of the reversed contingencies. If subjects have prior mastery of perceptually discriminating stimuli during original acquisition, analysis of reversal learning allows the study of stim ulus reward learning under conditions in which perceptual deficits can largely be ruled out (Jones & Mishkin, 1972). The Acc appears to be an important structure in a neural circuit involved in reversal learning. Numerous studies indicate that lesions of t he Acc and anatomically related structures (medialis dorsalis of thalamus or the VP) cause marked deficits in reversal learning with only minor effects on discrimination tasks in monkeys (Roberts et
25 al., 1990) and rats (Everitt et al., 1987a; Evenden et al ., 1989). In addition, an event related fMRI study in human subjects showed that the Acc area and ventrolateral PFC were involved in reversal learning, showing activation especially just prior to switching to the new correct stimulus (Cools et al., 2002) Reversal learning deficits have been demonstrated in tasks as varied as spatial, visual, and olfactory discriminations. Deficits on spatial discrimination have been shown after Acc lesions with 6 hydroxydopamine (6 OHDA) (Taghzouti et al., 1985). In a sp atial discrimination task such as a T maze, subjects with 6 OHDA lesions did show transient discrimination deficits, but only for the first three days of the five day testing period; in the spatial reversal, however, there was significantly reduced perform ance across all five days of testing (Taghzouti et al., 1985). In monkeys, ibotenic acid lesions of the Acc did not affect original spatial discrimination or visual discriminations of either two shapes or two visually distinct cans, but did impair spatial reversal learning (Stern & Passingham, 1995). There was not a significant deficit, however, for the visual reversal task of the two can stimuli (Stern & Passingham, 1995). This failure to find visual reversal deficits could either be a species difference b etween rats and monkeys, or as the authors suggest, a function of the extent of the lesion. Their representative lesion reconstruction shows small lesions that do not involve the whole Acc, primarily only showing damage to the AcS of Acc. In a study of odo r discrimination and reversal learning, subjects with lesions of Acc did not show effects on the ability to learn stimulus reward associations for novel odors, but did show deficits in the reversal task (Ferry et al., 2000). These errors were represented b y increases in the number of false alarms after the significance of the
26 associated stimuli were reversed. That is, lesioned animals demonstrated perseverative responding by choosing the formerly rewarded but now unrewarded stimulus. Although most studies do not report deficits in original learning in reversal tasks after Acc lesions (i.e., the discrimination phase), a study by Reading and colleagues (1991) did show such a discrimination deficit. Bilateral ibotenic acid lesions of Acc in the rat disrupt ac quisition of a visual discrimination of slow versus fast flashing lights (i.e., differences in temporal). However, the performance of the Acc lesioned group could be improved by manipulating the inter stimulus interval or the stimulus duration. This sugges ts that it was not a sensory deficit, but possibly one of attention. In addition, an attentional deficit was also implied in that Acc lesioned animals were more resistant to extinction under conditions of non reward (Reading et al., 1991). A primary input to the Acc comes from the medialis dorsalis nucleus of the thalamus, and lesions to this structure cause deficits in reversal learning. Excitotoxic lesions of the MD were found to have no effect on acquisition of a visual discrimination of shape (e.g., a w hite rectangle or white cross). However, when stimulus reward contingencies were reversed, animals with lesions of the MD made more errors and took a larger number of sessions to reach criteria (Chudasama et al., 2001). In the rat, lesions of MD, or asymme trical lesions of MD and orbitofrontal cortex, produce deficits in odor reversal tasks (McBride & Slotnik, 1997). These results suggest that Acc is part of a distinct thalamocortical pathway important for stimulus reward learning, since Acc, MD, and medial frontal cortex lesions produce similar learning deficits. Thalamic damage in human patients can cause impairments in response inhibition (Leng & Parkin, 1988; Joyce & Robbins, 1991) and attentional deficits (Rafal & Posner, 1987).
27 In marmosets, excitotoxi c lesions of the basal forebrain that included VP impaired reversal of shape discrimination, but not the original discrimination (Roberts et al., 1990). Using similar procedures in the rat, Everitt and colleagues (1987b) also showed deficits in the reversa l of a simple visual discrimination, without effects on the original learning (discrimination). Pigeons and Dopaminergic Drugs Pigeons have been used to stud y various drugs of abuse, including those which act primarily on the dopaminergic system (e.g., co caine, amphetamine). A variety of paradigms showing the effects of such drugs on performance of response sequences (Thompson, 1977), different reinforcement schedules (Nadar et al., 1988), and delayed matching to sample tasks (Branch & Dearing, 1982; Spetc h & Treit, 1984) have helped elucidate the basic effects of altering normal dopaminergic tone in birds. Birds exposed to dopaminergic drugs like their mammalian counterparts, develop place preferences (Burg et al., 1989), have increases in perseverative e rrors (Thompson, 1977), and can be trained to perform in self administration studies (e.g., cocaine; Winsauer & Thompson, 1991). The se similarities demonstrate how the basic mechanisms of dopaminergic action affect both motor and cognitive behaviors. Indeed the behavioral effects of drugs across a variety of species are usually more quantitative (in terms of dosages) than qualitative (McMillan, 1990). An intriguing use of pigeons in drug studies has that of drug discrimination paradigms. In such tasks pige ons are typically trained in an operant chamber with some number of keys. Each key is then associated with the delivery of a certain drug. For example, in four choice drug discrimination, birds learned to distinguish amphetamine, pentobarbital, morphine, a nd saline (Li & McMillan, 2001). In essence this type of testing
28 gives the experimenter access to the birds subjective state when affected by a particular drug. Moreover, the behavior of birds to novel drugs can be assessed. For instance some drug with s tructural similarities to amphetamine (i.e., methylenedioxymethamphetamine, MDMA) can be tested to see how well pigeons will generalize or discriminate between its effects and their previously learned drug response key associations. Drug studies are one im portant example in which comparative studies of different animals can help elucidate the function of important neural systems. Pigeons and Disruption of MSt Function Watanabe (2001) recently showed that MSt is associated with cognitive rigidity similar t o the perseverative errors seen in mammals after Acc lesions A repeated acquisition of a spatial and visual discrimination in pigeons was employed using a three key operant chamber. Responses to one of the three keys were rewarded. Upon reaching criterion subjects were trained with a previously incorrect key as the new correct key. Some tasks were administered with a unique color assigned to each of the three keys (an additional visual discrimination element). The MSt lesions disrupted the acquisition of both the spatial discrimination and the discrimination with added color cues. Subjects with MSt lesions showed high amounts of errors in the first trial, indicative of the preservation of the previously correct response. These results were presented in ter ms of the contribution of MSt to higher cognitive functions, especially cognitive rigidity that is often found in patients with basal ganglia disorders. However, the lesions were extensive and damaged both medial and lateral areas of MSt. It is unclear wha t role the more somatomotor or limbic territories of MSt had in the performance in this experiment. One of the few studies to purport to study accumbens function in birds and visual properties was Gargiulo and colleagues (1997), who demonstrated that glu tamate
29 blockade in the avian basal forebrain resulted in deficits which they interpreted as a cognitive disturbance of attention or visual perception. Stimuli consisting of lighted diodes in a matrix, lit in either a p like or q like configuration were presented. The lighted p, or its 180 degree rotated match (similar to a d) served as the correct stimuli (S+), and the q, or its 180 degree rotated counterpart (similar to a b) served as the incorrect stimuli (S ). After training was complete, sub jects underwent surgical implantation of cannulae into the basal forebrain. Subjects were exposed to lidocaine, apormorphine, 7 aminophosphonoheptanoic acid (AP 7), or saline. Lidocaine or apomorphine was used to create a conduction blockade or stimulation of the Acc, respectively. The substance AP 7 was employed to block glutamatergic activation (via NMDA receptors) presumed to come from pallial efferents. No apparent motor or motivational effects in executing the instrumental task were observed. Only bila teral infusions of AP 7 significantly reduced visual discrimination of the shapes to near chance level. The authors attributed AP 7 induced deficits to a disturbance in cognition, of either attention or visual perception; given that the Acc area is far rem oved from any know primary avian visual processing areas (Gargiulo et al., 1997). However, the authors figure showing the cannulae placements indicate that their injections were centered on the region that is likely the avian BnSt, and which also could ha ve affected VP function. The effects measured on visual discrimination, therefore, are likely not related to the probable Acc area. The disruption in visual discrimination also may be attributable to disruption of the fiber tract that adjoins the area, the fasciculus prosencephali lateralis, which conducts visual information in the collothalamic pathway from the Rt to the entopallium.
30 Learning in Chicks and MSt Circuits In chicks, MSt has been shown to be part of a circuit necessary for passive avoidance learning (Csillag, 1999; Mezey & Csillag, 2002). A natural, robust behavior of young chicks is to peck at small objects present in the visual field. In the passive avoidance learning paradigm, a small object, usually a shiny or conspicuously colored bead i s coated with a bitter tasting substance (e.g., methylanthranilate). Upon pecking, the chicks demonstrate a disgust response, including retreat from the object and head shaking. On subsequent presentation of the bead, chicks show avoidance behavior, having acquired a learned suppression of pecking to that particular visual object. The MSt is one brain area necessary for acquisition and/or maintenance of this learned passive avoidance behavior. As described before, MSt receives dopaminergic input from SNc an d VTA, and sends projections back to these structures. These reciprocal connections are thought to maintain normal dopaminergic tone in these systems (Mezey & Csillag, 2002). As in mammals, i n birds D 1 receptors are especially high in striatum and some of the densest areas are seen in MSt (Mezey & Csillag, 2002 ). Chicks that have learned to avoid the bitter bead have bilateral up regulations of D 1 but not D 2 receptors compared to controls (Stewart et al., 1996). In addition to regulating DA tone in the midb rain striatal loops, the MSt also sends long, descending projections to the central grey and LoC, with branchings into the brainstem reticular formation (Szekely et al., 1994). A pecking center for organizing motor components of pecking likely exists in the brainstem reticular formation, receiving inhibitory input from MSt via GABAergic efferents (Hall et al., 1984; Reiner & Anderson, 1990).
31 Avian Reversal Learning Studies Prior to the current study, no evaluation of the potential contribution of the avi an MSt on reversal learning tasks had been conducted. However, many studies have l ooked at the effects of Wulst or NCL lesions on reversal learning. Interestingly, these areas may be important sources of visual input to the avian Acc area. The Wulst may ha ve functions related to attention and/or cognitive flexibility. Reversal learning deficits in bir ds have most often been observed after lesions of the Wulst, although such lesions do not usually cause deficits in simple visual discriminations (Shimizu & Ho dos, 1989). Wulst lesions also have effects on spatial learning, delayed response, and both spatial and visual reversals (Macphail & Reilly, 1985). Pigeons with Wulst lesions have impaired reversal of position discrimination (Macphail 1971; 1976 ; Macphail and Reilly 1985 ) Such lesions also impair o rientation and color discrimination reversals (Macphail & Reilly 1983 ; Reilly 1987; Macphail 1976 ; Shimizu & Hodos 1989 ). Impaired reversal learning has also been found in other avian species with Wulst les ions (chick; B enowitz & Lee Teng 1973; quail ; S tettner & S chultz 1967) In another task requiring attention, pigeons with lesions of the Wulst had difficulty in a dimensional shift task (Powers et al., 1982). In this task, the birds were trained on a s imultaneous discrimination in which one dimension is relevant (e.g., color or pattern) while the other dimension is present but irrelevant for reward. After learning, the contingency is then reversed, so that the relevant dimension of the stimuli shifts to the previously irrelevant one. Lesions of the entopallium interfered with acquisition of the original learning, presumably due to perceptual difficulties. Wulst lesions, however, did
32 not affect original learning, but did create deficits when the relevant stimulus dimension was shifted from color to pattern. Lesions of the NCL have a variety of affects on avian cognitive functions including reversal tasks (Hartmann & G nt rk n, 1998; Lissek et al., 2002). Contributions of NCL and the adjacent nidopallium ca udale and CDL were analyzed for their effects on a go/no go, visual discrimination, and reversal tasks. The detrimental effects of the lesions were correlated with NCL damage but not to damage of the other caudal telencephalic structures. Lesions did not a ffect the go/no go task and visual discrimination, but reversal learning of the discrimination was adversely affected. The role of the NCL in behavioral or cognitive flexibility was recently investigated in pigeons in a color discrimination and reversal ta sk (Lissek et al., 2002). The study compared subjects undergoing amino 5 phosphonovalerate (AP 5) blocking of NMDA receptors (Lissek et al., 2002) versus saline treated controls. NMDA receptor blockade subjects made significantly more errors, without impai rment in the original color discrimination. These subjects also showed stronger preservation in their learning strategy, maintaining responding to the old S+ color stimulus after reward contingencies were reversed. There have been very few studies of MSt lesions on any type of cognitive behavior. In light of the similarities in connections to mammalian Acc, it is logical to expect that this region may play a role in functions attributed to mammalian Acc. The reversal learning paradigm has been used extensi vely in pigeons, but has been confined mainly to studies of Wulst and NCL. Given that both of these structures appear to send input to the MSt, it may be that the deficits seen in lesions of these structures have as a common mechanism, at least in part, th eir interaction with the MSt.
33 Background for Specific Aim # 3 Accumbens and Sexual Behaviors Reversal learning is performed in an artificial setting and is designed to test the cognitive ability of animals. Natural situations would also be expected to r ecruit the involvement of the Acc, since it plays a role in associative learning, attention, and the evaluation of contextual stimuli. In particular, sexual situations, i.e., finding a suitable conspecific with which to mate, would be expected to induce an aroused state and require attention to specific visual cues. Anticipatory vs. Consummatory Sexual Behaviors The process of mate choice and sexual behavior are of primary importance for a species survival. A variety of complex courtship behaviors have ar isen to allow potential mates to attract the attention of other conspecifics, and to ascertain their reproductive fitness and their ability to offer resources or parental support for the successful raising of progeny. Sexual behavior, like many behaviors, can be divided into both anticipatory (also called appetitive) and consummatory aspects. Often the appetitive and consummatory aspects of sexual behavior are neurochemically and anatomically dissociable. The Acc has been implicated in the appetitive and co nsummatory (i.e., intercourse) phases of sexual behavior ( Pfaus & Heeb, 1997 ). For instance, an f MRI study in humans showed increased activity in both males and females in a number of areas, including the Acc, when subjects viewed excerpts from erotic fil ms (Karama et al., 2002). Under these conditions, other active brain areas included the anterior cingulate, medial prefrontal, orbitofrontal, insular, occipitotemporal cortices and the amygdala, all of which have hodological relationship with the Acc. Thes e findings are consistent with the theory that the Acc is involved in arousal or attentional processes.
34 Role of Dopamine in Sexual Behaviors Manipulations of accumbal DA, with 6 OHDA lesions to Acc or systemic injections of DA receptor antagonists, reduce anticipatory aspects of behavior. These changes include rates of responding directed towards both food and, in male rats, a sexually receptive female; such manipulations largely leave consummatory behavior unaffected (Blackburn et al., 1987; Everitt, 1990) There is evidence that DA plays a role in appetitive and consummatory sexual behaviors. In a bi level chamber in which male rats had previous copulatory experience, systemic injections of haloperidol reduced the number of level changes, a measure of appe titive sexual behavior (Pfaus & Phillips, 1991). These males also showed increases in measures of consummatory sexual aspects (latencies for intromission and mounting). However, bilateral infusions of haloperidol directly into the Acc reduced anticipatory level changes but not the consummatory copulatory measures. An in vivo microdialysis study of DA and related metabolites in Acc showed different levels of DA change than in dorsal striatum. Significant increases in DA levels were seen when male rats saw a sexually receptive female behind a screen, and these levels increased further during copulation. An increase was also seen in DA levels in the dorsal striatum, but with lower magnitude than in Acc, and only during copulation (Damsma et al., 1993). Similar findings have also been reported in female rats (Pfaus et al., 1995). Immediate Early Genes (IEGs): Neural Activation / Plasticity A necessary complement to anatomical studies in determining brain/behavior relationships is measuring neural activation. Demo nstrating that a given brain area has hodological relationship with other areas known to be involved in a behavior is not
35 sufficient to allow one to assume that the area also participates in that behavior. The ability to measure neural activity in vivo in the active brain, or measure persistent products of that activation in vitro has added a valuable dimension to understanding brain function. In animals, the most common measures of activity are the uptake of radioactively labeled glucose (the 2 deoxy gluc ose or 2 DG method), electrophysiological techniques, and, more recently, the use of in situ hybridization or immunohistochemical detection of genes or proteins coded by various genes induced in the brain. Immediate early genes are one such class of gene s that transcribe proteins in many kinds of cells, including neurons. Immediate early genes (IEGs) are part of a broad family of inducible transcription factors, which are thought to underlie neural activation and plasticity. The IEGs were so named because of their rapid induction, with mRNAs detectable within 5 to 10 minutes; this occurs in the absence of de novo protein synthesis that is characteristic of these genes (Herdegen & Leah, 1998). There are two primary classes of IEGs, the zinc finger and le ucine zipper families. Four cysteine residues form the zinc finger of the first class of IEGs, and include the krox 20 (in mice, called egr 2 in humans) and krox 24 types (also called nerve growth factor 1 A ( NFG1 A ), zif 268 egr 1 or tis 8 ) (Herdegen & Leah, 1998) (see Figure 2). The leucine zipper class of IEGs contains an alpha helix with a leucine in every seventh position, and includes the c jun Jun B c fos and Fos B types (Herdegen & Leah, 1998). The IEG zenk is a member of the zinc finger fa mily of genetic transcription factors. ZENK is the term used for the avian homologue of these genes and the proteins they induce, being an acronym for the various members of this gene transcription family in different species: z if268 e gr 1 N GFI A and k rox24 (Mello &
36 Clatyon, 1994; Long & Salbaum, 1998). The IEGs and their associated proteins are remarkably conserved in mammals and birds. The area for the zinc finger motif, which functions as a DNA binding domain in the zinc finger family, is 100% ident ical between birds and mammals (Long & Salbaum, 1998). The IEG transcription factors are also remarkably conserved across diverse phylogentic groups, being found in human, rodent, canary, and zebrafish, among other animals (Long & Salbaum, 1998). Clayton ( 2000) compares the induction of IEGs to the action potential, in that they are both neural events that serve to code for changes in either the internal or external environment. However, the genomic action potential represented by IEG induction integrates information over much longer time frames, and thus may have a more profound effect on brain function (Clayton, 2000). If the Acc area were involved in cognitive functions related to either attention or arousal, one would expect that activation of this are a would induce IEGs. The IEG protein product Fos is induced in numerous regions, including Acc, in rodents after exposure to sex related stimuli (Pfaus & Heeb, 1997). This response is also seen in a conditioning context where neutral stimuli are paired wit h sex related stimuli. Presentation of a neutral almond odor which had previously been paired with sexually receptive female induced significantly higher levels of Fos within the Acc, as well as piriform cortex, medial preoptic area, and VTA in conditioned male rats compared to controls (Kippen et al., 2001). The IEGs c fos and c jun are induced in male rat brains medial amygdala, preoptic area, and BSTl during consummatory sexual behaviors. In contrast, the Acc is activated just prior to the consummatory aspects (the preparatory phase), in addition to TO, dorsal striatum, and basolateral amygdala (Everitt, 1995).
37 Figure 2. The zinc finger IEG, krox 24 The image shows the molecular surface for the DNA in the protein DNA complex. From: http://www.ysbl.yo rk.ac.uk/~mjh/molviewer/molviewer.html Courtship Behaviors and IEGs in Birds There are relatively few studies on avian courtship and IEG expression. However, expression of Fos in the zebra finch MSt was shown to increase under conditions of general arous al. For instance, after a birds first courtship experience, or being chased around its home cage, induced higher Fos expression in MSt compared to controls that were either socially isolated or housed in an aviary (Sadananda & Bischof, 2002). In addtion, the degree of attention, arousal, or vigilance affects the selectivity for neurons in the songbird brain for conspecific song (Park & Clatyon, 2002). Other than the song control nuclei in songbirds, the primary data regarding IEGs and avian courtship behav ior comes from studies of quail sexual behaviors. Avian ventral striatal regions (i.e., MSt, VP, and TO) and anatomically associated structures (e.g, arcopallium) have been implicated in the anticipatory and consummatory phases of avian sexual behavior. In Japanese quail, ( Cotunix coturnix ) both the appetitive and consummatory aspects of male sexual behavior have been shown to
38 increase expression of the Fos protein (Tlemcani et al., 2000; Ball & Balthazart, 2001). Expression of Fos increases in the anticipa tory phase of sexual behavior in the MSt, as well as nucleus striae terminalis (nST), and arcopallium (Tlemcani et al., 2000). High induction of IEGs has been shown to occur in hypothalamic and limbic areas, such as the medial preoptic area, BSTl, and arc opallium (Aid and Aiv), of birds who had copulated and/or expressed a social proximity response, a measure of anticipatory sexual behavior (Ball & Balthazart, 2001). Unfortunately, this study did not mention or show figures regarding Fos or ZENK expression in the MSt. In addition, these authors used the Acc definition from the atlas of the chick brain that, similar to the pigeon, may actually be the avian equivalent of the mammalian BSTl.
39 Chapter Two Methods Anatomical Study Methods (Tract tracing and Ca BPs) The anatomical studies employed fourteen subjects. The birds had free access to water and a balanced diet of mixed grains, with free feeding weights typically ranging from 500 to 600 grams. These subjects are weighed weekly to ensure their health and maintenance of a stable weight. All subjects are housed in individual cages in an American Association for Accreditation of Laboratory Animal Care certified housing facility maintained by the University of South Florida. Subjects were food deprived for at least 8 hours prior to surgery with free access to water. The birds were deeply anesthetized with ketamine (40mg/kg of body weight, i.m.) and xylazine (10mg/kg of body weight, i.m.). Feathers were then trimmed from around the head and ear area. The subject was placed in a modified stereotaxic apparatus (David Kopf Instruments, Tujunga, CA) and injection site co ordinates set per the atlas of the pigeon brain (Karten and Hodos, 1967). The exposed skin over the skull was then wiped with alcohol and a scalpel used to make one incision (approximately 2cm) at the midline of the skull. The skin was retracted with hemostats and kept moist during the operation. A small area of the skull (approximately 8 10mm in diameter) was removed with a dental drill until the dur a is exposed. This was gently teased from the surface of the brain using forceps to minimize surface tissue damage. Attached to the stereotaxic apparatus, an injection pipette was then centered over the skull opening and cleaned with water. Following the p rocedure, subjects were returned to their home cages
40 after surgery with free access to food and water. Survival periods ranged 3 7 days depending on the length of pathway to be traced and tracer efficacy. Anterograde Tracer Injection Micropipettes with ope nings of 10 20 m m in diameter were created using thin glass capillary tubes and a micropipete puller (Kopf Model 720 Vertical Pipette Puller). These micropipettes were used for both iontophoretic and pressure injections. Most injections involved bilateral i njections of both hemispheres if no inter hemispheric transfer of tracers is indicated. Biotinylated Dextran Amine (BDA) is an amine conjugated dextran with an attached biotin compound. The 10,000 molecular weight (MW) of BDA can be used as an anterograde tracer in the pigeon, as described by Veenman and colleagues (1992). BDA appears to travel via slow axonal transport mechanisms and provides good filling of cells for visualization of somatic and dendritic morphologies (Veenman et al, 1992). Precise BDA in jections with the best anterograde transport are best achieved by iontophoretic injection. In iontophoretic injection, a thin metal wire is inserted into a micropipette, which has been back filled with tracer by vacuum extraction. Current is then applied t o the wire from a Stoetling Constant Current Source Model allowing electrostatic pressure to drive tracer molecules into the tissue The electric current parameters were 3.5 5.0 m A on a 7 sec on / 7 sec off pulse cycle for 10 20 minutes. Retrograde Tracer Injection Retrograde tracer injections were performed with Cholera toxin subunit B (CTb). Stoeckel and associates first described in detail the use of CTb as an anatomical tracer (Stoeckel et al., 1977). CTb is the non poisonous subunit B of the cholera disease toxin. It is taken up preferentially by terminals near the injection site and delivered
41 axonally to the soma (Ericson & Blomqvist, 1988). CTb is characterized by high contrast between signal and noise (i.e., specific cell staining and background la beling) making it ideal for tracing and the study of dendritic and terminal morphology. Only short survival periods are necessary as the toxin typically travels from 80 240mm per day (Ericson & Blomqvist, 1988). It degrades slowly in tissue making it ideal for tracing longer pathways. The main disadvantage of CTb labeling is some labeling of passing fibers in axons damaged by the injection procedure. For this reason, confirmation of regions indicated with retrograde experiments was verified with anterograde experiments into these areas. T racing of retrograde fibers using the low salt formulation of CTb is achieved by iontophoretic injection. These injections used similar parameters to those mentioned above for BDA injections. Immunohistochemistry for CaBPs S ubjects were deeply anesthetized via ketamine and xylazine, (i.m.) and then sacrificed by transcardial perfusion with a 0.9% saline solution followed by fixation with 4% ice cold paraformaldehyde in phosphate buffer (PB, pH 7.4). Brains were post fixed 6 t o 12 hours at 4 C followed by immersion in a 30% sucrose solution for 24 hours at 4 C. Brains were then frozen in dry ice and cut in 40 m m transverse sections on a sliding microtome. Sections were then washed three times at 10 minutes each in room temper ature PB in preparation for immunohistochemical procedures. For visualization of the CaBPs calbindin (CB), calretinin (CR), and parvalbumin (PV), the appropriate antibodies against them were used in the primary incubation. The antibodies were a monoclonal mouse anti CB (dilution 1:10000; Sigma, St. Louis, MO), monoclonal mouse anti PV (dilution 1:10000; Sigma) and a polyclonal rabbit anti CR (dilution 1:10000; Chemicon, Temecula, CA). The tissues were then washed in PB
42 followed by 1 hour of incubation in an appropriate biotinylated secondary antibody (Vector Laboratories) and 0.3% Triton X 100 at a dilution of 1:200. Tissues were then washed and processed according to the avidin biotin complex method (ABC Elite Kit, Vector Labs, Burlingame, CA) and visualize d with a 0.025% solution of 3 diaminobenzadine (DAB) and hydrogen peroxide as described elsewhere (Husband & Shimizu, 1999; Laverghetta & Shimizu, 2002). Tissue controls were created by the omission of the primary antibody. Following immunohistochemical procedures, tissue sections were mounted on glass slides. Slides were then progressively dehydrated in an ethanol series. Following clearance with xylenes, slides were coverslipped with Permount (Fisher Scientific, Fair Lawn, NJ). Immunohistochemistry fo r Tract tracing General histology procedures for tract tracing were the same as for the CaBP procedures above. Tissues from BDA anterograde experiments was first be incubated in avidin biotin reagent (ABC Elite Kit, Vector Labs) in 0.3% Triton X 100 in PB with 2% NaCl for one hour at room temperature. The tissues were then reacted in a solution of 0.025% DAB and 0.025% nickel ammonium sulfate with 0.3% hydrogen peroxide for approximately 20 minutes. This procedure yields a dark blue to black oxidative prod uct within the area of the injection and any labeled cells and/or fibers. Cholera toxin subunit B is immunohistochemically visualized with overnight incubation at 4 C in a solution of goat antiserum against CTb (1:10,000; List Labs) in 0.3% Triton X 100 i n PB. This is followed by the washing procedure previously described and then incubation in a solution of biotinylated anti goat antiserum (1:200; Vector Labs) in 0.3% Triton PB at room temperature. This is followed by incubation with an avidin biotin reag ent (ABC Elite Kit, Vector Labs) in 0.3% Triton X 100 in 2% NaCl for one hour
43 at room temperature. The tissues were reacted with DAB and hydrogen peroxide as described previously for the anterograde experiments. CTb labeled cells and fibers appear light to dark brown after such reactions Tis sue sections were mounted on glass slides in either PB or gelatin mounting solution. Slides were then dehydrated in an ethanol series consisting of approximately three minutes each in 70% ETOH, 95% ETOH, and 100% ETOH. Following clearance with xylenes, slides were coverslipped with Permount (Fisher Scientific, Fair Lawn, NJ). To control for non specific binding or artifactual reactivity within cells or fibers, tissue controls were employed. For each experiment, some tiss ues cut from the same brain are processed by the standard procedures with the exception of the primary antibody incubation. Comparing this control series to regularly processed tissues precludes the possibility that some resident substance in the brain mig ht stain positive for the chosen lectin (BDA) or antibody (CTb). In this way, cells or fibers not associated with the injected area do not create false positives for immunoreactive soma/fibers. Alternating sections of brain tissue from each experiment were stained either with Cresyl violet (Sigma Chemical Co., St. Louis, MO) or Giemsa dye (EM Diagnostic Systems, Gibbstown, NJ) to visualize major nuclear groups and subdivisions (Iiguez et al., 1985). Data Analysis The tissues were examined using a macroscop e (Wild Makroskop) and a microscope (Nikon Microphot FX). The results were photographed and/or recorded through camera lucida drawings. The photographs presented in the preliminary figures are digital images acquired using a Leaf Lumina scanning CCD camera (Leaf System Inc., Southborough, MA) mounted on either a macroscope or a microscope. Digital images were loaded to Adobe Photoshop software as black and white positive images
44 (Adobe Systems Inc., Mountain View, CA) using a Dell Optiplex PC. Brightness and contrast were adjusted for the final images. No additional filtering or manipulation of the images was performed. For cases with BDA injections, although the 10,000 MW of this tracer is transported primarily in an anterograde fashion, there may be some r etrogradely labeled fibers evident after the injection. All labeled fibers were recorded and appear in the drawings. Distinction between anterograde and retrograde fibers was primarily based on the presence of retrogradely labeled cell bodies. Behavioral S tudy Methods These experiments employed eight adult White Carneaux pigeons ( Columba livia ) of both sexes as subjects (obtained from Double T Farms, Glenwood, IA). These subjects were divided randomly into three groups: Unoperated Controls (1 subject), Sham Surgery (3 subjects), and MSt Lesions (4 subjects). The birds had free access to water and a balanced diet of mixed grains. Birds intended for behavioral experiments were on gradual food reduction until attaining between 80 85% of their free feeding weigh t. Behavioral subjects were weighed daily to ensure their health and maintenance of a stable weight. Apparatus and Pretraining Procedures A two key operant chamber (Leigh Valley Pigeon Test Chamber) were used, the interior dimensions of which are 36.8cm hi gh X 35.6cm wide X 50.8cm long (see Figure 3A). The front panel (35cm 2 ) consists of: two translucent response keys each measuring 2.5cm in diameter placed 16.5cm apart from each other (measured from the center of each key), a house light located centrally at the top, and a feeder opening. The feeder
45 opening is located 10.2cm above the floor of the chamber and measures 6.4cm wide by 5.1cm high (see Figure 3B). Two stimulus lamp projectors (One Plane Readout; IEE, Inc.) was project on the rear surface of two plastic pecking keys. The stimuli were two patterns, vertical or horizontal black stripes on a white background. The experiments were controlled with an IBM compatible computer (Intel 80286 microprocessor) through an analog digital interface/converter. A p rogram written in the Microsoft Q basic language was control the stimulus presentation, response/reward contingencies, and collects response data for each trial. Birds were trained to eat out of the hopper, and then autoshaped to peck either of the two re sponse keys lit by white light. Subjects achieving stable performance during autoshaping was then enter the pretraining phase. They were trained to peck whichever of two keys is lit by a white light, which were randomly presented on either the left or righ t key using a Fellows sequence (Fellows, 1967). The seventh response was reinforced with a 3 sec access to grain reward from a lighted feeder. Training continued until the birds complete 30 trials in a 30 minute period. Surgery and Post operative Testing Within 72 hours after reaching the pretraining criterion, subjects were food deprived for at least 8 hours prior to surgery with free access to water. They were then anesthetized with ketamine and xylazine (i.m.). General surgical procedures were identical to those described in the above section on anatomical experiments. The lesion electrode, in a holder attached to the stereotaxic apparatus, was centered over the skull opening. Each lesion was performed with two penetrations into each hemisphere, in an at tempt to lesion as much of the MSts extent as possible without affecting surrouding
46 structures. Three control subjects received sham lesions, in which the electrode was lowered into the brain in the same locations as lesion subjects, but no current was ap plied. Following the procedure, subjects were returned to their home cages after surgery with free access to food and water. After surgery, all subjects were given a seven day recovery period, during which there were no behavioral testing sessions. Subject s were maintained at 80 85% of free feeding weights during the recovery period. On the first day after the recovery period, a neurological exam is given to each surgical subject to compare with prior observations in the home cage. Neurological signs such as h ead carriage, muscle tone, posture, et cetera are assessed to ensure no motor damage from the lesions or surgery has occurred (see Appendix A). A pretraining retention test is then administered to ensure experimental subjects are still pecking accurate ly. An identical procedure and criterion are used to the presurgical pretraining sessions (success defined as completion of 30 trials within 30 minutes). The procedure is repeated daily until this criterion is met. Discrimination and Reversal Sessions On the day after the birds pass the retention test, they are given a pattern discrimination task. A black vertical striped pattern and a similar horizontal pattern were presented simultaneously on the two keys (see Figure 3B). The position of each pattern wa s randomized according to a Fellows (1967) discrimination sequence. The seventh response on the correct key is reinforced with a 4 sec access to grain reinforcement. The fourth response to the incorrect key (S ) is punished with a 5 sec timeout. Any key pec ks during the intertrial interval (ITI) reset the ITI counter and delayed the next trial. The delivery of either a reward or punishment completes one trial, and no correction method or trials were used. After a 15sec ITI, the stimuli were presented and the next
47 trial began. The session ended when the subject either a) completed the criteria of 15 consecutive correct trials, b) if a total of 80 trials were reached, or c) if 2 hours had passed without the subject reaching either of the aforementioned criteria The day after achieving criteria (15 consecutive correct trials in one session), the first reversal discrimination was conducted. In the reversal learning session, the stimuli and procedures were identical to the original learning except that the positiv e and negative stimuli (S+ and S respectively) were reversed. On each subsequent reversal, the stimuli were reversed again. A total of 20 reversals were carried out after which the animal wa s sacrificed. Histology and Lesion Reconstruction Within 72 hou rs after completion of the 20 reversals of discrimination, the birds were deeply anesthetized via an injection of ketamine and xylazine, (i.m.) and then sacrificed by transcardial perfusion with a 0.9% saline solution followed by a fixative of 4% ice cold paraformaldehyde in PB (pH 7.4). The brains were then processed and cut as described for the anatomy experiments. Lesion reconstruction was carried out on a series of standard plates derived from the atlas of the pigeon brain of Karten and Hodos (1967). Gl iosis and necrosis in the telencephalon was recorded. The extent of each lesion was determined from measurements of reconstructions made with camera lucida and graphics software (Canvas 7.0, Deneba Systems). Behavioral Data Analysis The primary measures used in the analysis of behavioral results were errors and trials to criterion. Additionally, analysis of the types of errors made by lesion and control subjects was conducted. Two primary kinds of errors subjects can make are when they either a) develop a key preference or, b) continue to choose the previously correct S+ in
48 a reversal session (perseverative errors). To separate key preference effects from perseverative errors, a key preference index was calculated for all subjects in the original learning and each of the reversal sessions (Shimizu & Hodos, 1989). Key preference was calculated with the following formula: Key Preference Index = ABS ((L R) / (L + R)) where the absolute value of left key (L) responses minus those to the right key (R), is the n divided by the total number of responses. A score of 1.00 would indicate a complete reliance on one key over the other for all trials. To assess perseverative responding, each subjects response patterns were analyzed to calculate the cumulative number o f responses to the previous sessions S+. Only three or more blocks of successive responses to the previous S+ were counted, since one or two such responses could be attributable to random error or key preference. Differences between groups in total errors total trials, and error types were assessed using parametric statistics and appropriate post hoc tests in SPSS version 12 (SPSS, Inc.).
49 Figure 3. Configuration of the operant chamber and stimulus display/response panel. A) The operant chamber with in terior dimensions shown. B) The stimulus display/response panel, showing the stimuli to be used in the proposed study and the relevant dimensions mentioned in the text.
50 Gene Expression Methods This phase of the project employed a total of eight subjects These subjects were randomly assigned into two groups, the Live condition (4 subjects) and the Empty condition (4 subjects). Apparatus and Behavioral Procedures An open area testing chamber (90cm long X 90cm wide X 60cm high) with a wire mesh ceiling was used. The front wall of the chamber has a large opening in which a clear Plexiglas stimulus chamber (41cm long X 41cm wide X 31cm high) can be attached. A white divider can be placed between the testing chamber and stimulus chamber, to limit visual access of the subject. Attached at the back wall of the chamber is a Plexiglas door that opens to a small waiting room. A video camera for observation and video recording of subjects is affixed above the chamber to record the movements and vocalizations of the s ubjects. Prior to testing, each bird was acclimated to the testing chamber for at least five days to prevent novelty effects in ZENK expression. These acclimation sessions consisted of 10 minutes of darkness, then 10 minutes with an overhead light on. The 10 minute lights on condition was broken down into alternating 2 minute periods in which the light was turned on and off with a switch located in the observation room. This was then followed by 10 full minutes of darkness. This was done to insure that neit her the part of session in darkness nor the lighted part of the sessions were novel to the subject (see testing procedure, below). On testing day, subjects were put into the testing chamber for a minimum of 2 hours (7 subjects) to a maximum of 12 hours (1 subject) prior to the testing session. The testing session lasted a total of 120 minutes. The stimulus chamber contained either no
51 stimulus, i.e., the Empty condition, or contained a female pigeon, i.e., the Live condition. Courtship behaviors from subject s were visual and audio recorded by the video camera positioned above the chamber. The testing session was conducted beginning with 40 minutes of darkness. This was followed by 40 minutes of exposure to either the empty cage or the live female pigeon. This 40 minute segment consisted of alternating 2 minute periods in which the light was turned on and off via a remote switch in the observation room. The alternation of the light was performed in order to prevent the subjects from habituating to the stimulus, especially if they ascertain that due to the barrier they would not be allowed physical contact with the female. The final 40 minutes of the session were again in full darkness. The time course for the experiments and stimulation periods were designed to allow maximum detection of ZENK protein levels. Upon stimulation, the expression of IEGs like the zenk gene peaks around 10 minutes, while the protein products which they transcribe have been shown to peak around 60 minutes in both birds and mammals (Herde gen & Leah, 1998). After the two hour test session, the subject was immediately given an overdose of anesthesia, sacrificed, and perfused as described previously. Histology and Immunohistochemistry for ZENK Immunoreactivity for the ZENK protein was detecte d by using the primary antibody that recognizes a portion of the EGR 1 peptide sequence similar in both birds and mammals. Tissues were reacted with the ABC method for peroxidase labeling to visualize EGR positive and negative cell bodies. Adjacent tissue sections were stained with a Nissl stain to show major cell groups and fiber tracts. As a control for the specificity of the secondary antibody, adjacent tissue sections were processed at the same time and in the same way but the primary antibody was omitt ed.
52 Analysis of ZENK Protein Levels Using a light microscope, the distribution of ZENK labeled neurons in areas of the MSt and LSt were mapped. For quantitative analysis, four sections through the rostro caudal extent of the structure were chosen for analy sis of lateral MSt, mdMSt (A12.50, 11.50, 10.50, 9.50), and LSt (A10.50, 10.00, 9.50, 9.00). For the mvMSt, since it is a smaller area overall than the previously mentioned structures, only three sections were analyzed from each subject (A12.50, 11.50, and 10.50). A target region of 165 X 165 mm2 within these sections were photographed and used to calculate the number of ZENK positive neurons within the section. Figure 4 shows schematic drawings illustrating the areas chosen for quantitative analysis of the ZENK expression. Counts were performed blind as to condition. Two counts were performed for each section and the average used as the count for that section. The number of ZENK immunoreactive (ZENK ir) neurons was counted using NIH Image software. These t otals from each subject were then used to calculate summary statistics and perform subsequent statistical tests.
53 FIGURE Figure 4. Locations of areas sampled for quantitative analysis of ZENK ir cells Anterior coordinates of the sections are: A12.50, 1 1.50, 10.50, 10.00, 9.50, 9.00, and 8.50. OES HERE Landscape Figure
54 Chapter Three Results CaBP Results Patterns of immunoreactivity for CaBP in the basal forebrain demonstrated that the MSt is a chemically heterogeneous structure (Figure 5). Bas ed on the distribution patterns of CaBPs, there were at least two subregions within m MSt, which for the current study are identified as the mediodorsal and medioventral divisions (mdMSt and mvMSt). For PV immunoreactivity (Figure 5 B ), PV ir neuropil was st rongest in the mdMSt, and very light in mvMSt and BSTl. For CR (Figure 5 C ) immunoreactivity was most dense in mvMSt, and lighter in the mdMSt. A complimentary pattern for CB was seen (Figure 5 D ), where mdMSt showed moderate intensity in mdMSt and lighter i ntensity within the mvMSt. Based on these patterns, tracer injections were placed in each of these possible MSt subdivisions to determine whether they also had hodological heterogeneity. Tract tracing Results Summary of Anatomy Experiments The MSt was foun d to have many midbrain (ventral tegmental area, substantia nigra) and forebrain (amygdala, hippocampus, dorsal thalamus ventral pallidum ) connections similar to those of Acc. Comparing mv and mdMSt, several patterns of different connections were found. The m vMSt has connections primarily with VTA in the midbrain, DMA and LHy in the thalamus Hp/APH and PoA in the telencephalon, and ventral VP in the basal ganglia The mdMSt has connections primarily with SNc in the
55 Figure 5 Patterns of CaBP expressi on for PV, CR, and CB in pigeon striatum. A) Schematic drawing of transverse sections of the pigeon forebrain. B D) CaBP patterns for B) parvalbumin; C) calretinin; and D) calbindin. See list of abbreviations for terms. FIGURE 5 CaBPs TO BE INSERTED HERE Landscape Figure
56 TABLE 2 Summary of mv and mdMSt connections found in the current study. Symbols indicate relative density of projections: ++ dense, + moderate, none. Projections to MSt mvMSt mdMSt Projections from MSt mvMSt mdMSt VP ventr al + VP ventral ++ VP dorsal VP dorsal ++ BSTl Lateral Hypothalamus + Aa + VTA (A10) ++ Ad ++ ++ SNc (A9) + ++ Ai ++ ++ Raphe + Am + ++ PoA ++ Tn + ++ Hp ++ APH ++ + Lateral Septum + Lhy + VTA (A10) ++ SNc (A9) ++ ++ Raphe DMA (lateral) ++ + DMP (lateral) + ++ HA/HIS ++ + HD + + mM + ++ NFL ++ mN ++ ++ NCm ++ NCL ++ + Nd + TPO / CDL ++ ++ CPi +
57 midbrain, DMP in the thalamus, Aa in the telencephalon, and dor sal VP in the basal ganglia. A summary of these findings is represented in Table 2 (p. 5 6 ). Summary of CTb Injections Injections of the retrograde tracer CTb demonstrated numerous telencephalic and subtelencephalic structures that are a source of input to the medial MSt. CTb injections of mv MSt and mdMSt resulted in labeled cells in several common structures, such as SNc in the midbrain and mM and mN in the telencephalon. However, some notable differences were labeled cells in VTA, DMA and LHy of the thal amus, the hippocampus (Hp, APH), and PoA of the arcopallium after m vMSt injection s. Injections of CTb into mdMSt labeled more cells in the SNc, and cells in DMP of the thalamus, and Aa of the arcopallium Injection sites for these experiments can be divide d into ventromedial (mvMS t; Pg220, Pg225, and Pg194) and dorsomedial (mdMSt; Pg183; Pg180) targeted injections. CTb I njection s into mvMSt A key to the brain regions discussed in the following sections is shown in Figure 6. A representative case for more ventromedially targeted injections (subject Pg220) is illustrated in Figure 7 The injection site was in a more anterior part of the structure (Figure 7 B), centered on the mvMSt The tracer deposit was adjacent to the lateral cerebral ventricle, with no v isible spread outside of the MSt and a slight injection track visible in a portion of HA. In other parts of the hyper and mesopallium there were a moderate number of cells in the mM and the medial HD dorsal to it (Figure 7 B D). The few cells seen in the H A could have been attributable to the small tracer leakage in this
58 region. In nidopallium, retrogradely labeled cells were seen in the far anterior portion of the NFL (Figure 7A). A dense region of CTb ir cells was seen in the mN (Figure 7B D), CTb ir cel ls (Figure 7E F). In addition, there were cells in the LHy (Figure 7F, H). In the arcopallium, cells were restricted to the PoA division (Figure 7G). In the diencephalon, Figure 6 Drawings from the stereotaxic atlas of the pigeon (Karten & Hodos, 196 7) showing the brain regions of interest in the current study. See list of abbreviations for terms. The anterior coordinates of the sections of the atlas plates are: first and second rows, A12.50, 11.50, 10.50, 9.50, 7.50, 6.50, and 6.00; third row, A4.00, 3.00, and 2.00.
59 Figure 7 CTb injection into mvMSt in subject Pg220. See text for description The anterior coordinates of the sections shown in the figure are: first and second rows, A13.50, 12.75, 11.50, 11.00, 8.50, 7.50, and 5.50; third row, A6 .50, 4.00, 3.50, and 3.00.
60 Figure 8 CTb Injection into the mvMSt in subject Pg225. See Appendix B for details The anterior coordinates of the sections shown in the figure are: first and second rows, A13.50, 12.50, 10.50, 9.00, 8.00, 7.00, 6.25; th ird row, A7.00, 4.00, 3.00, and 2.00.
61 Figure 9 Injection of CTb into mMSt in subject Pg194. See Appendix B for details. The anterior coordinates of the sections shown in the figure are: first and second rows, A12.75, 12.00, 10.75, 9.50, 7.50, 6.75 and 6.00; third row, A6.50, 6.00, 4.00, and 3.00.
62 Figure 10 CTb injection into mdMSt in subject Pg183. See text for details. The anterior coordinates of the sections shown in the figure are: first and second rows, A12.00, 11.00, 10.50, 9.25, 7.5 0, 7.00, and 6.50; third row, A5.75, 5.00, 4.50, 3.25, and 2.50.
63 l abeled neurons were found in the dorsal thalamus, specifically the DMA (Figure 7H). In the mesencephalon, both the VTA (Figure 7I J) and the SNc (Figure 7K) showed labeled neurons. T he injection cases of Pg225 and Pg194 had largely similar patterns of CTb ir cells as that of Pg220. Drawings of Pg225 and Pg194 are shown in Figures 8 and 9, respectively. Because the labeling patterns were largely similar to those of Pg220, these cases w ill not be described in further detail here but are available in Appendix B. CTb Injections into mdMSt The injection site for subject Pg183 was located in the mdMSt (Figure 10A). There were a few cells in the HA and HD (Figure 10A D), but these may be attr ibutable to injection artifact. A dense band of CTb ir cells was found in the mM (Figure 10A E), with a smaller band of cells in the lateral HV (Figure 10C). In neostriatal areas, the mN contained numerous CTb ir neurons (Figure 10B E), while the Nd and NC m showed only a few scattered cells (Figure 10F G), with none visible in the NCL. No cells were found in the Hp proper and very few were evident in the APH region (Figure 10F G). A few retrograde cells were seen in the posterior LSt (Figure 10E). In the ar copallium, CTb ir labeling was seen in the Aa, Am, and Tn (Figure 10E). There were also larger amounts of cell labeling in Ad and Ai (Figure 10 F G). In the diencephalon, CTb ir cells were found primarily in the DMP (Figure 10 H I) Ventral and lateral to DM P there were also labeled cells in the pretectal area and substantia grisea centralis (Figure 10 I). Notably, no retrogradely labeled cells were seen in the VTA, unlike the prior cases with injections into ventromedial MSt. A few cells were also seen in the SNc (Figure 10 J), and some in the area of the LoC (Figure 10 K).
64 For subject Pg180 (Figure 1 1 ), the injection site was centered in mdMSt but more posterior, with spread into mvMSt and the BSTl (Figure 1 1 C D). In the hyperpallium and mesopallium there were CTb ir cells in the HA, HD, and mM (Figure 1 1 B E). While there were relatively few in the HA and HD, a moderate number of cells in the mM formed a band along the dorsal part of the structure (Figure 1 1 B D). In the nidopallium, there was a dense field of c ells in the mN (Figure 1 1 B E), extending through both anterior and posterior sections away from the injection site. The NFL (Figure 1 1 A) exhibited a dense field of cells, primarily anterior to the injection site. These cells could have potentially been lab eled as artifact from some damage and tracer leakage in the HA. The PE/TPO region also showed a band of cells in more posterior regions of the telencephalon (Figure 1 1 E F). In the NCm labeled neurons were scattered throughout the structure, with very few s een in the NCL region (Figure 1 1 G). In the arcopallium, CTb ir labeling was seen in the Aa, Tn, (Figure 1 1 F), Ad, Ai, (Figure 1 1 G), and PoA regions (not shown). The highest number of cells was found in the Ad and Ai, where they tended to cluster in the lat eral part of the structure, largely avoiding the Am (Figure 1 1 G). In the thalamus, cells were found in the DMA (Figure 1 1 H) and to a lesser extent the DMP (not shown). Some labeled cells were seen ventral to the DMA in the LHy and the area of the nucleus m edialis hypothalami posterioris (Figure 11H). In the midbrain, numerous cells were found in the VTA as well as the SNc (Figure 11I and J, respectively). The LoC exhibited a large number of cells (Figure 11K).
65 Figure 1 1 CTb injection into pMSt in subjec t Pg180. See text for details. The anterior coordinates of the sections shown in the figure are: first and second rows, A12.75, 12.00, 11.00, 9.50, 9.00, 7.50, and 6.25; third row, A6.25, 4.00, 3.00, and 2.00.
66 Summary of BDA Injections Injectio ns of the anterograde tracer BDA revealed several areas that receive efferents from the MSt. The primary output of the MSt areas appears to be the VP and fibers returning to the midbrain (VTA and SNc). The mvMSt sends projections primarily to the ventral V P, LHy, and VTA; in contrast the mdMSt sends its projections mainly to dorsal VP and the SNc. As in the retrograde cases, these injections can be roughly divided into mvMSt (Pg191 and Pg195) and mdMSt (Pg199 Pg193 and Pg181right and left hemisphere) targ et sites. BDA Injection into mvMSt Subject Pg191 (Figure 1 2 ) had an injection of BDA centered on the mvMSt in its more anterior aspect (Figure 1 2 B). In the hyperpallium and mesopallium, fibers were seen in the HA, HD, and mM (Figure 1 2 A E). It is possible that the HA and HD fiber labeling was artifactual due to the injection site. Fibers in the mM were found in proximity to retrogradely labeled cells (Figure 1 2 C E), so it is probable that these fibers represent retrogradely and not anterogradely labeled fi bers. In the nidopallium, both the NFL (Figure 1 2 A B) and mN (Figure 1 2 A E) demonstrated labeled fibers, again in proximity to labeled cells. Numerous fibers were seen in the Hp and APH regions (Figure 1 2 F G) accompanied by labeled cells. The Aa, Ad, Ai, a nd PoA of the arcopallium contained labeled fibers. The few fibers in Aa were concentrated in its dorsal area (not shown). The Ai exhibited more fibers than either the Aa or Ad (Figure 1 2 F), and those in PoA tended to cluster in its lateral portion (Figure 1 2 G). The fibers in Ad, Ai, and PoA were in proximity to labeled cells. Numerous anterogradely labeled fibers were seen in the ventral VP (Figure 1 2 D E). In addition, a minor field of fibers was also seen in the lateral
67 Figure 1 2 BDA injection into mv MSt in subject Pg191. See text for details. The anterior coordinates for the sections shown in the figure are: first and second rows, A12.75, 12.25, 11.50, 10.25, 9.00, 6.50, and 5.50; third row, A6.50, 4.50, 4.25, and 3.75.
68 septum (not shown). I n the diencephalon, a small number of fibers were seen in the DMA, with a much higher number visible in the DMP (Figure 12I J). Ventral to the DMA, in the posterior part of the hypothalamic area, labeled fibers were seen in the LHy (Figure 1 2 H). In the mes encephalon, the VTA (Figure 1 2 I J) contained numerous fibers and cells. No such labeling was seen in either the SNc or LoC. A few retrogradely labeled cells were seen in the area of the raphe nuclei (Figure 12K). The injection site of subject Pg195 was al so in the mvMSt (Figure 1 3 B), but avoided the area directly lateral to ventricle. This case had similar patterns to the previous case already described (Pg 191 ) therefore details are not included here but are available in Appe n dix C One primary finding th at is well illustrated in this case, however, is the projection to the VP and midbrain. The ventral aspect of the VP showed numerous BDA labeled fibers (Figure 1 3 D E). A photomicrograph of the BDA labeled fiber projection to the ventral VP is shown in Figu re 1 4 A. In the midbrain, fibers were found in the VTA (Figure 1 3 I), with a notable lack of labeling in either the SNc or LoC (not shown). BDA Injection into mdMSt In subject Pg199 (Figure 1 5 ), the tracer injection deposit was located in mdMSt (Figure 1 5 C). As in other cases, the mM exhibited fibers, in this case accompanied by retrogradely labeled cells (Figure 1 5 A E). Some fibers were also present in the HD, immediately adjacent to the lamina mesopallialis (Figure 1 5 A B). In the nidopallium, a small number of fibers were seen in the NFL (Figure 1 5 A B) and limited portion of the NCm (Figure 1 5 F). The mN exhibited a dense plexus of fibers along with labeled cells (Figure 1 5 A E). Notably there were no labeled fibers in the Hp and APH in this case. A
69 Figure 1 3 BDA injection into mvMSt in subject Pg195. See Appendix C for details. The anterior coordinates for the sections shown in the figure are: first and second rows, A12.75, 12.50, 11.00, 10.50, 9.00, 7.50, and 7.00; third row, A7.00, 4.50, and 3.50.
70 Figure 1 4 Darkfield photomicrographs of BDA labeled fibers in the ventral and dorsal VP. Fibers and terminal like varicosities found in the ventral VP (A) after BDA injection into the mvMSt in subject Pg195 or dorsal VP (B) a fter BDA injection into the mdMSt of subject Pg199. A B
71 Figure 1 5 BDA injection into mdMSt in subject Pg199. See text for details. The anterior coordinates of the sections shown in the figure are: first and second rows, A12.50, 12.25, 10.00, 9.00, 8.50, 7.25, and 6.50; third row, A6.50, 6.00, 4.75, and 3.75.
72 few scattered fibers were seen in the posterior LSt region (Figure 1 5 E F). A dense plexus of fibers was found in the dorsal VP (Figure 1 5 C E). A photomicrograph of this projection to dorsal VP is shown in Figure 1 4 B. In the arcopallium, the Ai contained a few, faint fibers (Figure 1 5 F), while the Ad and Am contained more numerous fibers which were especially den s e in the Ai (Figure 1 5 F G). In the thalamus, the DMP, and to a lesser extent the DMA, c ontained fibers in proximity to labeled cells (Figure 1 5 H and 1 5 I J, respectively). There were a moderate number of fibers in the area of the LHy ( Figure 1 5 H I ). In the midbrain, no fibers were seen in the VTA; however a dense plexus was located in the SNc in conjunction with BDA filled neurons (Figure 1 5 K). For subject Pg193 (Figure 1 6 ), the injection was more centrally located, covering the middle third of MSt and mdMSt (Figure 1 6 C). Since the patterns of labeled fibers for this case were similar to thos e of Pg199 (described above), details for this case are included in the appendicies ( Appendix C) The injection sites for the right and left hemispheres of Pg181 (Figures 17 and 18 respectively ) were similar, hence only the right hemisphere injection of case Pg181 will be discussed in detail. A description of the left hemisphere injection of Pg181 can be found in Appendix C. The right hemisphere injection (Figure 17) was located in the mdMSt, but in its more posterior part It appears to have missed the B STl, which is just ventral and medial to the MSt in these sections surrounding the ventral tip of the ventricle. The only fibers seen in hyper and mesopallium were in the HA and mM. The fibers in HA (Figure 17D) were located in sections posterior to the i njection site, but these are likely artifact from the injection. A moderate number of fibers were found in the
73 Figure 1 6 BDA injection into mdMSt in subject Pg193. See Appendix C for details. The anterior coordinates of the sections shown in the figu re are: first and second rows, A12.00, 10.50, 10.25, 9.25, 7.75, 6.50, and 6.00; third row, A6.50, 6.00, 4.50, and 3.25.
74 mM, accompanied by retrogradely labeled cells (Figure 1 7 A C). In the nidopallium, the mN and NCL both contained fibers. A more lateral region of the mN than had been seen in other cases contained the most fibers (Figure 1 7 A C); these cells again were in proximity to filled cell bodies. The fibers in the NCL were very sparse, being found only in a small lateral area of posterior s ections (Figure 1 7 G). The PE/TPO contained some dispersed fibers in sections posterior to the injection site (Figure 1 7 D F). Fibers in the Hp/APH were somewhat dense (Figure 1 7 E F), although these fibers could be attributable to injection artifact at the b rain surface. In sections of MSt anterior to the injection site, fibers were dispersed over most of the structure (Figures 1 7 A C). Fibers from the injection site formed a prominent band toward the VP, covering most of its extent (Figure 1 7 D E). There were relatively few fibers in the arcopallium, with a small number seen in the Aa (Figure 1 7 F), Ad, and Ai (Figure 1 7 G). A few fibers were also evident in the medial and lateral aspects of the LSt and GP (Figure 1 7 E F). In the thalamus, fibers and cells were fo und in the DMP (Figure 1 7 H), but very few were seen in the DMA (not shown). In the midbrain a few fibers were found in the VTA (Figure 1 7 I). A larger number of fibers and cells were seen in the SNc (Figure 1 7 J); in this section were also a number of fiber s and cells in the area of the lateral reticular formation. Some fibers and cells were also found in the LoC area (Figure 1 7 K).
75 Figure 1 7 BDA injection into pMSt in subject Pg181 (right hem). See text for details. The anterior coordinates for the secti ons shown in the figure are: first and second rows, A12.00, 11.00, 10.50, 9.50, 9.00, 7.75, and 6.50; third row, A5.50, 4.00, 3.25, and 1.50.
76 Figure 1 8 BDA injection of BDA into pMSt in subject Pg181 (left hem). See Appendix C for details. The sections are presented as right hemisphere sections for consistency with other figures. The anterior coordinates for the sections shown in the figure are: first and second rows, A12.00, 11.00, 10.00, 9.50, 9.00, 7.75, and 7.00; third row, A7.00, 4.00, 3.0 0, and 1.50.
77 BDA Injection into mM and mN T he major telencephalic input from mM and mN to MSt revealed by the CTb experiments was confirmed by injecting BDA into both the mM and the mN in several cases. Injections into the mM were conducted in subje cts Pg242, Pg213, and Pg215. Two cases for the mN condition were derived from the left and right hemisphere injections of subject Pg241. Subject Pg242 received an injection of BDA into the mM (Figure 1 9 ). The patterns of efferents were largely similar to t hose found for case Pg213 (Figure 20 ) and Pg215 (Figure 2 1 ); therefore, only Pg242 will be discussed in detail. Details on Pg213 and Pg215 are included in Appendix C The injection site for Pg242 was located in the mM (Figure 1 9 C) with some minor leakage o ver the LH to the mN; there was also a small injection track in the medial HD. In the hyper and mesopallium labeled fibers were found in the HA, HD, and mM. Fibers were found in the HA and HD, most of them anterior to the site of injection (Figure 1 9 A C). Dense fields of labeled fibers and cells were found throughout the anterior posterior extent of the mM (Figure 1 9 B E). In the nidopallium the mN, Nd, NCL, and NCm contained varying numbers of BDA labeled fibers. The fibers in the mN extended throughout mo st of the structure (Figure 1 9 B F). Dense fields of fibers In the Nd and NCm were found (Figure 1 9 F G); the NCL was
78 Figure 1 9 BDA injection into mM for subject Pg242. See text for details. The anterior coordinates for the sections shown in the figure are: first and second rows, A13.50, 12.25, 11.75, 11.00, 9.50, 7.50, and 7.00; third row, A6.50, 6.00, and 5.50.
79 Figure 20 BDA injection into mM for subject Pg213. See Appendix C for details. The anterior coordinates for the sections shown in the figure are: first and second rows, A12.50, 12.00, 11.00, 10.50, 9.50, 9.00, and 7.00; third row, A6.50 and 5.25.
80 Figure 2 1 BDA injection into mM for subject Pg215. See Appendix C for details. The anterior coordinates for the sections shown in the figure are: first and second rows, A12.50, 12.00, 11.00, 10.50, 9.50, 9.00, and 7.00; third row, A6.50 and 5.25.
81 devoid of fibers (Figure 19H) until more posterior sections (Figure 19I J). The MSt exhibited numerous fibers that were distrib uted over most of the structure, with particularly high concentrations in its medial two thirds (Figure 19B E). In the arcopallium, BDA labeled fibers were relatively dense in the following main structures: Ad, Av (Figure 1 9 H I), and Ap (Figure 1 9 J). An a rea of dispersed fibers was also seen in the LSt (Figure 1 9 E G). No fibers were visible in any subtelencephalic structures. I njection s of BDA into the right or left hemisphere mN of subject Pg241 are depicted in Figures 2 2 and 2 3 respectively. Only t he ri ght hemisphere injection will be discussed in detail, as both right and left hemisphere cases had similar efferent patterns. A detailed description for Pg241 left hemisphere is included in Appendix C The right hemisphere injection site was located in mN, with no spread outside the structure (Figure 2 2 C). In the hyper and mesopallium, labeled fibers were found in the HD and mM. Minor fiber labeling was seen in the HD (Figure 2 2 A D). A denser band of fibers and associated retrogradely labeled cells were see n in the mM (Figure 2 2 B E). In the nidopallium, the NFL, mN, Nd, and NCL exhibited labeled fibers to differing degrees. Fibers in the NFL were limited to a few sections anterior to the injection site (Figure 2 2 A). The mN showed dense fields of BDA labeled fibers in sections both anterior (Figure 2 2 B) and posterior (Figure 2 2 D E) to the injection site. Fibers in the Nd were restricted to a small band at its medial edge (Figure 2 2 G), while those in NCL only appeared in more posterior sections (compare Figure 2 2 G and H). In MSt, fibers tended to cluster in the most medial section, with few scattered fibers in more lateral areas of MSt (Figure 2 2 A E). No fibers were seen in Hp proper, while those few seen in the APH (Figure 2 2 F) may have been injection track art ifact. In the arcopallium, a dense plexus of
82 fibers were seen in the medial Ap (Figure 2 2 G H). A few scattered fibers were seen in the VP (Figure 2 2 E F). In the diencephalon, fibers in proximity to retrogradely labeled cells were seen in the DMA and DMP ( Figure 22I J). In the mesencephalon, a small number of fibers were located in the VTA (Figure 22K), with no other areas showing detectable BDA fibers or cells.
83 Figure 2 2 BDA injection into mN for subject Pg241 (right hem). See text for details The anterior coordinates for the sections shown in the figure are: first and second rows, A13.00, 12.00, 11.25, 10.00, 9.25, 8.50, and 6.00; third row, A5.00, 7.00, 6.00, and 4.50.
84 Figure 2 3 BDA injection into mN for subject Pg241 (left hem). See Appendix C for details. The sections are presented as right hemisphere sections for consistency with other figures. The anterior coordinates for the sections shown in the figure are: first and second rows, A12.50, 12.00, 11.00, 10.00, 9.00, 5.75, and 5.00 ; third row, A6.50, 5.75, and 3.00.
85 Lesion Effects on Pattern Discrimination & Reversal Learning Summary of Behavioral Experiments All l esion subjects and controls performed similarly on original discrimination. Unexpectedly, the MSt lesioned birds and controls did not show significant differences on performance in the reversal sessions. However, e rror patterns indicated that sham lesioned birds had deficits due to position preference, whereas lesioned birds had fixation on previous reward contingen cies (perseverative errors) Th is difference in the source of errors for lesioned birds was consistent with Acc lesion effects on reversal learning in mammals. Electrolytic lesions were performed on four subjects (Pg177, Pg179, Pg224, and Pg228). L esion d amage was centered either in the more dorsal MSt (two subjects; Pg177 and Pg179) or more ventral MSt (two subjects; Pg224 and Pg228). In addition, the dorsal lesions tended to be smaller overall than that of the ventral cases. An additional four subjects s erved as controls; one was an unoperated control (Pg178) and three sham lesioned subjects (Pg223, 226, and Pg233). For sham lesions the electrode was lowered to the appropriate locations but no current was applied. Most subjects in both lesion and sham gro ups received varying degrees of damage in the Wulst, which overlies the electrodes approach to arrive at the lesion targets. This damage is described along with the lesion reconstructions for the experimental group, and in a separate section for the contr ol group. In addition, other structures were damaged in the lesion cases to some degree. These structures included the mN, mM, BSTl, and the VP. Any impact from Wulst or other damage could have potentially had on subject peformance is detailed in the Discu ssion.
86 Experimental Group: Lesion and Wulst Damage Reconstruction Figure 2 4 illustrates the degree of lesion damage for subject Pg177. The damage overall was moderate. It extended from anterior coordinate A11.50 through A10.00 (Figure 2 4 D F) and was cente red in the dorsal MSt (Figure 2 4 D G). The lesion was largely symmetrical, with similar amounts of damage in all but the most anterior and posterior sections (Figure 2 4 D and G, respectively). Damage to other structures was restricted to some moderate damage in the overlying mN (Figure 2 4 D F). Some slight damage to the mM in one hemisphere of one section (Figure 2 4 D, left hemisphere) was also evident. Damage at the lesion electrodes point of entry consisted of minor unilateral damage in the left hemispheres Wulst, restricted to the HA (Figure 2 4 A B, F). Bilateral damage to the Wulst was seen in more posterior sections (Figure 2 4 G I), again restricted to the HA. Subject Pg179 had minor damage overall (Figure 2 5 ). Lesion damage was restricted to the intermedia te parts of the MSt in sections A11.00 through A9.50. The lesion was relatively small and consisted of damage confined to the MSts dorsal region (Figure 2 5 E H). Lesion damage was also unilateral in some sections (Figure 2 5 F, H), while asymmetrical in size in others (Figure 2 5 E, G). Damage to structures outside the MSt included a small electrode track in the hyperpallium (Figure 2 5 E) and minor damage to the mN (Figure 2 5 E, G) and mM (Figure 2 5 G). Overall Wulst damage was minor and localized in the HA; this damage was also unilateral with the exception of that seen in A11.00 (Figure 2 5 E). Figure 2 6 shows lesion reconstruction for subject Pg224. This case had much more extensive lesions than the previous two cases described above. The lesion
87 Figure 2 4 L esion and Wulst damage reconstruction for subject Pg177. Lesion damage in MSt and surrounding region is indicated by black areas; ablated or damaged tissue in Wulst is indicated by hatched areas.
88 Figure 25 Lesion and Wulst damage reconstruc tion for subject Pg179. Lesion damage in MSt and surrounding region is indicated by black areas; ablated or damaged tissue in Wulst is indicated by hatched areas.
89 Figure 2 6 Lesion and Wulst damage reconstruction for subject Pg224. Lesion da mage in MSt and surrounding region is indicated by black areas; no damage in this case was seen in the Wulst.
90 e xtended from A11.00 to A9.50, including most of the medial and ventral part of MSt (Figure 2 6 E H) with some unilateral damage in dorsal MSt (Figure 2 6 E F). The lesion was largely symmetrical in all sections. Damage to other structures included minor unilateral damage to the mN (Figure 2 6 F), around the area of the VP (Figure 2 6 G), and some encroach ment on the dorsal margin of the BSTl (Fig ure 2 6 H). No Wulst damage was visible in this case. Subject Pg228 showed the largest degree of lesion damage, illustrated in Figure 2 7 Lesion damage extended from sections A12.50 through A9.50. The lesions in MSt were largely symmetrical in size, although the left hemisphere lesion was concentrated on the ventral and medial aspect, while that of the right hemisphere was more somewhat ventral and lateral (Figure 2 7 C F). There was a small electrode track in the hyperpallium in the most anterior section (A12. 50; Figure 2 7 B). Unlike the other cases described above, no damage was seen in the overlying mN. Damage to other structures included some encroachment on the olfactory tubercle (Figure 2 7 D, G), damage in the area of the VP (Figure 2 7 G), and in the most pos terior section bilateral lesion damage in the BSTl (Figure 2 7 H). Surgical damage to the Wulst was limited to the right hemisphere. However, it did penetrate deeper into the Wulst than in the previously described cases, affecting more of the HA with some sl ight damage in the HD (Figure 2 7 B F). Control Group: Wulst Damage Reconstruction No visible damage to structures other than the Wulst (e.g., mN or MSt) was found. Since subject Pg178 was an unoperated control, there is no figure or description for this cas e in lesion reconstruction. Figure 2 8 illustrates the damage to the Wulst seen in subject Pg223. This damage extended from sections A10.50 to A8.50 (Figure 2 8 E I). Damage was restricted to the HA and was unilateral in all sections. For subject Pg226,
91 F igure 2 7 Lesion and Wulst damage reconstruction for subject Pg228. Lesion damage in MSt and surrounding region is indicated by black areas; ablated or damaged tissue in Wulst is indicated by hatched areas.
92 Figure 2 8 Wulst damage reconstruct ion for subject Pg223. Ablated or damaged tissue in the Wulst is indicated by hatched areas.
93 Figure 2 9 Wulst damage reconstruction for subject Pg226. Ablated or damaged tissue in the Wulst is indicated by hatched areas.
94 Figure 30 Wulst damage reconstruction for subject Pg233. Ablated or damaged tissue in the Wulst is indicated by hatched areas.
95 damage to the Wulst is illustrated in Figure 2 9 Wulst damage was largely bilateral (Figure 2 9 A E) with the exception of one more p osterior section (Figure 2 9 F). Damage was restricted to the HA division of hyperpallium. Figure 30 shows the degree of Wulst damage for subject Pg233. Damage in this case was completely unilateral to the left hemisphere, and did not go beyond the HA (Figur e 30 B G). Performance on Visual Discrimination and Reversal Learning A graph of individual subject performance in terms of the number of errors to criterion for each session is shown in Figure 3 1 In original learning, there was no significant difference f ound in a one way ANOVA between lesioned subjects and controls (F = 0.347, p = 0.57 ). Among lesioned subjects, Pg224 and Pg228 tended to do worse in early r eversals; these subjects also had the largest lesions. There was a high degree of variability in per formance among control subjects. For example, two control subjects (Pg223 and Pg226) made more errors on early reversals than even some of the lesion subjects. Subject Pg223 did especially poorly, performing worse than any subject for Reversals 2 through 5 and show ed erratic performance even in later reversals. Across all subjects, however, there were no significant differences found in a one way ANOVA across the first five reversals (F = 0.612 p = 0.44 ). Key Preference Analysis A common type of error in reversal learning studies is the development of key preference (Lissek et al., 2002). This occurs when subjects fail to respond based on the intended discriminative stimuli but instead develop preference for one of the two operant keys. Since the current s tudies hypothesis was that MSt lesions would increase errors due to perseverative responding (i.e., continued responses to the S+ of the previous reversal), an analysis of error patterns was conducted for all subjects.
95 Errors to Criterion by Subject 0 25 50 75 100 125 150 175 200 225 250 275 OL R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 Session # of Errors Pg 178 Unop Ctrl Pg 223 Sham Ctrl Pg 226 Sham Ctrl Pg 233 Sham Ctrl Pg 177 Elec Lesion Pg 179 Elec Lesion Pg 224 Elec Lesion Pg 228 Elec Lesion Figure 3 0 Graph of individual s ubject performance as measured by errors to criterion. Control subject data is indicated in green, while that of experimental subjects is in red.
96 Figure 3 2 is a graph of the key preference index values for original learning and ten subsequent reversals. I n original learning and the first reversal there is little difference between the control and experimental groups. However by the second reversal and up to reversal seven there was a strong key preference in the control group. For example, the control gro up on reversals 2 and 3 showed an average key preference of 0.42 and 0.44, respectively. Contrast this with the lesion groups score s of 0. 14 on reversals 2 and 3. In addition, the degree of key preference seen in the control group was almost entirely due to the scores of two subjects, Pg223 and 226. For example on reversal 3, t he i ndividual key preference scores for Pg223 and Pg226 were 0.84 and 0.76, respectively. Compare this to the two other control birds (Pg178 and Pg233) that scored 0.03 and 0.12, res pectively. None of these trends, however, were statistically significant in a one way ANONVA done across the first five reversals (F = 2.46 p = 0.13 ). Response to Previous S+ Analysis The graph in Figure 33 shows analysis of the degree to which subjects responded to the previous sessions S+ (perseverative errors) over original learning and ten subsequent reversals. There was a trend in the first three reversal sessions in which experimental subjects had a stronger tendency to make errors by responding to the old S+. This difference was not statistically significant, however, in a one way ANOVA analysis of reversals one through five ( F = 1.21, p = 0.30)
97 Key Preference 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 OL R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 Sessions Index Value CONTROL LESION Figure 3 2 Graph of the key preference index values on original learning and ten reversals for co ntrol and lesion subjects. Mean values and standard error are shown. Data from control subjects are indicated by gray bars, while that of experimental subjects are in black.
98 Errors: Responses to Previous S+ 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 OL R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 Sessions # of Errors CONTROL LESION Figure 33 Graph of the errors attributable to responses on the pre vious sessions S+ stimulus (perseverative errors). Data from control subjects are indicated by gray bars, while that of experimental subjects are in black. Perseverative Errors
99 ZENK Protein Expression Results Summary of ZENK Experiments The expression of ZENK in the MSt of male birds exposed to either an empty cage (Empty condition) or one with a live female pigeon (Live condition) was quantified in four subregions of interest (mvMSt, mdMSt, lMSt, and LSt) Significantly h igher ZENK expression was found in the Live co ndition for all these structures. However, the degree of difference between L ive and E mpty was higher in the mvMSt and mdMSt than in the other areas (lateral MSt and LSt) Quantitative Analysis of ZENK ir Cells The expression of ZENK protein in regions of the MSt and LSt was assessed for subjects in two different behavioral conditions. All of the striatal brain areas in birds exposed to the Live condition expressed more ZENK immunoreactive (ZENK ir) cells than those in the Empty condition. This was confirme d by a 2x4 between subjects analysis of variance, where condition (2) and brain area (4) were compared (F=2.652, p=.035). In addition, there was a tendency for those cells that expressed the protein to show darker reaction product in the Live than the Empt y condtion. Subsequent planned post hoc tests (d etailed below) demonstrated that all areas showed a significant difference between the Live and Empty conditions. However, the proportion of these differences did differ among the brain areas. mdMSt: Analysis of ZENK ir Cells In the mdMSt, a larger number of more intense ZENK ir cells w ere seen in the Live condition versus the Empty condition. Figure 34 is a boxplot graph comparing FIGURE 34 GOES HERE Landscape Figure
100 Figu re 3 4 Boxplot graphs showing the quantitative analysis of ZENK ir cells in the Live and Empty groups ZENK ir cell counts in the md MSt (A), the mv MSt (B), the lateral MSt (C), and the LSt (D). B A C D
101 summary statistics (maximum, minimum, mean, median, and the 1 st and 3 rd quartiles) for the number of cells expressing ZENK ir protein between birds in the Live and Empty g roups in different areas of the striatum. Figure 3 4A s hows summary statistics for number of ZENK ir protein expressing cells in the mdM St. The number of ZENK ir cells in the Live condition (M = 67.2, SD = 25.69) was significantly higher than in the Empty group (M = 34.6, SD = 17.66; t(30) = 4.186, p = 0.0001, two tailed). The Live condition had 49% more ZENK ir cells than the Empty condit ion based on a comparison of means from each group. Figure 35 shows low power magnification (10X) photomicrographs of the analyzed regions from mdMSt in both the Empty and Live conditions. mvMSt: Analysis of ZENK ir Cells In the mvMSt area a similar trend was seen in the number of ZENK ir cells in the Live versus the Empty condition. The graph in Figure 3 4B shows summary statistics for the number of ZENK ir cells in the Live and Empty groups. A significantly higher number of ZENK ir cells was found in the Live condition (M = 72.3, SD = 33.02) than in the Empty group (M = 38.6, SD = 13.36, t(30) = 3.283, p = .003, two tailed). Comparing the means between the two groups, the Live condition showed 47% more ZENK ir cells in the mvMSt than the Empty condition. F igure 36 shows photomicrographs of the analyzed regions from the mvMSt in both the Empty Cage and Live conditions. Lateral MSt: Analysis of ZENK ir Cells In the lateral MSt, a larger number of more intense ZENK ir cells were seen in the Live condition than the Empty condition. For the lateral MSt (Figure 3 4C ) the number of ZENK ir cells in the Live condition (M = 66.3, SD = 21.3) was significantly higher than in the Empty group (M = 41.9, SD = 20.2; t(30) = 3.299, p = .003, two tailed). Comparing FIGURE 3 5
102 Figure 3 5 ZENK expression in the mdMSt after exposure to two different behavioral conditions. A) Photomicrograph of ZENK expression in the mdMSt of subject Pg210 after exposure to the Empty condition. B) Photomicrograph of ZENK expression in the md MSt of Pg245 after exposure to the Live condition. A EMPTY B LIVE
103 Figure 3 6 ZENK expression in the mvMSt after exposure to two different behavioral conditions. A) Photomicrograph of ZENK expression in mvMSt of subject Pg231 after exposure to the Empty condition. B) Photomicrograph of ZENK expression in the mvMSt of Pg186 after exposure to the Live condition. A EMPTY B LIVE
104 Figure 3 7 ZENK expression in the lateral MSt after exposure to two different behavioral conditions. A) Photomicrograph of ZE NK expression in the lateral MSt of subject Pg210 after exposure to the Empty condition. B) Photomicrograph of ZENK expression in the lMSt of Pg245 after exposure to the Live condition. A EMPTY B LIVE
105 the means for the Live and Empty groups, the number of Z ENK ir cells in the Live condition was 37% higher than in the Empty condition. Figure 37 shows low power magnification (10X) photomicrographs of the analyzed regions from lMSt in both the Empty and Live conditions. LSt: Patterns of ZENK ir Cells In the LSt a somewhat larger number of ZENK ir cells were seen in the Live versus the Empty condition. Figure 3 4 D is a boxplot graph showing summary statistics on the number of ZENK ir protein expressing cells in the two groups. Specifically, the number of cells in the Live c ondition (M = 59.8, SD = 16.80) was found to be significantly different than the number in the Empty condition (M = 41.3, SD = 19.06 ; t(30) = 2.912, p = .003, two tailed). The proportion of cells in the Live versus Empty group was the lowest out of the four striatal areas; based on a comparison of means the Live group had 31% more ZENK ir expressing cells than the Empty group. Figure 3 8 shows photomicrographs of the analyzed regions from the LSt
106 Figure 3 8 ZENK expression in the LS t area after exposure to two different behavioral conditions. A) Photomicrograph of ZENK expression in the LSt of subject Pg210 after exposure to the Empty condition. B) Photomicrograph of ZENK expression in the LSt of Pg245 after exposure to the Live cond ition. A EMPTY B LIVE
107 Discussion Th e three lines of evidence from the anatom ical, functional and neurochemical studies presented here indicate that avian medial MSt has numerous similarities with mammalian Acc. First, t he avian medial MSt and mammalian Acc share many common midbrain and forebrain connections Furthermore, d ifferences in anatomical connections between mvMSt and mdMSt indicated that these two structures are anatomically distinct, and possibly comparable to the shell and core of Acc, respectively S econd b irds with lesions in the MSt showed a te ndency to make perseverative errors on reve rsal learning without effects on the original visual discrimination However, these differences were not statistically significant. The lack of effect of MSt lesions on reversal behavior could potentially be due to several factors, which are discussed in more detail below. Third, t he expressio n of the immediate early gene protein ZENK was examined after male birds were exposed to either an empty cage or live female pi geon. ZENK expression was found to be particularly high in the mdMSt and mvMSt in the males exposed to females. Comparing MSt and Acc Connections The MSt showed many similar connections with those found in the mammalian Acc. This included receipt of affe rents from midbrain dopamine centers, dorsomedial thalamus, and limbic areas of the telencephalon. Efferents from the MSt traveled primarily back to midbrain and to the ventral pallidum. Moreover, the mvMSt and mdMSt
108 showed hodological patterns that sugg est they are anatomically comparable to the mammalian AcS and AcC, respectively. Midbrain, Ventral Pallidum, and Thalamus The current results confirmed that the avian MSt receives projections from the VTA and SNc ( Karten & Dubbeldam, 1973; Kitt & Brauth, 1 981; Hall et al., 1984; Kitt & Brauth, 1986a, b ; Anderson & Reiner, 1991; Szekely et al., 1994; Mezey & Csillag, 2002 ). The new data, however, extends our understanding of avian striatal regions as defined by their differences in midbrain connections. In p articular, the mvMSt was associated with both VTA and SNc, receiving afferents from both of these structures. Its efferents primarily travel back to the VTA, with some minor input to SNc. This pattern is consistent with midbrain connections of the mammalia n AcS (Zahm, 2000). The mdMSt, on the other hand, was primarily associated with SNc. The mdMSts afferent source from midbrain was primarily the SNc, and it sent efferents back to the SNc. The AcC of mammals demonstrates these types of reciprocal connectio ns (Zahm, 2000). The findings of the present study in pigeon are consistent with an anterograde tracing study in chicks (Mezey & Csillag, 2002), which employed injections into either the SNc or VTA and mapped retrogradely labeled neurons within the MSt. In jections into the SNc resulted in labeled cells in dorsomedial and lateral MSt, with only a few labeled cells in the medioventral portion, near the ventricle. In contrast, injections into the VTA resulted in scattered cells within the mdMSt, with larger cl usters of cells in the mvMSt; the lateral part of MSt and LSt were devoid of labeled cells (Mezey & Csillag, 2002). These tegmento striatal tegmental connections support the idea that different functional domains exist in the avian striatum. As in mammalia n dorsal striatum, midbrain projections to the lateral portions of avian striatum (i.e., avian lateral MSt and
109 LSt) are particularly dense from SNc. Mammalian ventral striatum, conversely, receives proportionally more fibers from the VTA, similar to patter ns seen for medial portions of avian striatum (i.e., medial MSt). This suggests that the LSt and lateral portions of MSt are more involved in sensori motor function, whereas the medial portion is associated with viscero limbic functions. For other midbrain areas, the present study found somewhat small projections from the LoC and raphe nuclei. Both the LoC and the raphe nuclei send noradrenergic and serotonergic projections, respectively, to the mammalian Ac c (Zahm, 2000). In the case of avian MSt, both mvM St and mdMSt demonstrated small fields of retrogradely labeled cells near the LoC after injections into these regions. Somewhat denser fields of cells, however, were seen in LoC after injections into the posterior M St In mammals, the LoC projections termi nate primarily in the AcS (Groenwegen et al., 1999), suggesting that perhaps some of the posterior MSt region is shell like. Alternatively, the cells seen in LoC after injections into posterior areas of MSt could be the result of some tracer deposit encroa ching on the BSTl, which is known to also have connections with the LoC in mammals ( Dong & Swanson, 2003) There was little or no labeling in the raphe nuclei in any of the cases analyzed, although one case of BDA injection into the mvMSt did demonstrate s ome labeled cells in the area of the raphe (Figure 1 2 K). The dorsal and median raphe nuclei send efferents to the AcS, with the dorsal raphe also contributing fibers to the AcC (Zahm, 2000). This connection alone, of course, would do little to clarify the correspondence of avian mvMSt with mammalian AcS. The present series of experiments revealed a differential output from mdMSt or mvMSt to the VP (compare photomicrographs in Figure 1 4 A and 1 4 B). This is the first evidence of potential anatomically distingu ishable subdivisions within the avian VP
110 Injections centered in the mvMSt resulted in a band of fibers that traveled to the more ventral VP, where these fibers exhibited terminal like varicosities. The dorsal VP, on the other hand, exhibited fibers and p resumptive terminals when tracer injections were located in the mdMSt. This is similar to the connections found in mammals for the projection from Acc to VP. The AcS in mammals sends its projections primarily to the medial VP, while those of AcC travel pri marily to the l ateral VP (Zahm 2000 ). Although the shell and core have often been considered together, it has been suggested that functional networks involving them are distinguishable based in part on their different connections with VP (Zahm, 2000). The current study confirmed previous findings that thalamostriate projections to medial MSt arise from the more medially located dorsal thalamic nuclei DMA and DMP (Wild, 1987). The dorsomedial nucleus, in particular the DMA, consistently showed a dense field of labeled cells after retrograde tracer injections into the medial MSt. In addition, the current study found that the DMA contained denser fields of cells after injections of tracer into the mvMSt; conversely, the DMP contained somewhat more cells after injections of the mdMSt. The lateral parts of the avian DMA and DMP have been compared to the mammalian medialis dorsalis (MD) thalamic nucleus, which sends prominent projections to Ac c This comparison between DMA/DMP and MD is based on several lines of e vidence: 1) both are much poorer in SP and ENK fibers than the rest of the dorsal thalamus, 2) both receive projections from visceral and limbic related striatum and VP, and 3) both send projections back to ventral striatum (i.e., medial MSt, TO of birds a nd Ac of mammals) and viscero limbic portions of dorsal striatum (Wild, 1987; Veenman et al., 1997). In contrast, more lateral dorsal thalamic nuclei have been shown to send afferents to more lateral parts of the LSt and MSt. These nuclei are comparable
111 to some intralaminar nuclei like the pa rafascicular nucleus, in that both are caudally located in the diencephalon (near the fasciculus retroflexus ) and contain only moderate amo unts of GABA, ENK, SP (Wild, 1987; Veenman et al., 1997). The present results in dicated that the lateral hypothalamus (LHy) is also a likely source of input to parts of the avian MSt. After injections into the mvMSt, but not the mdMSt, a limited zone of labeled cells was found in the LHy. This is in accord with a previous study showin g that only medial MSt receives projections from neurons of the lateral hypothalamus (Berk & Hawkin, 1985). These findings further indicate the viscero limbic characterization of medial, rather than lateral, MSt. In addition, this supports the idea that th e mvMSt area of MSt is comparable to AcS, since in mammals only the shell region receives hypothalamic input (Zahm, 2000). Amygdala, Hippocampus, and Other Forebrain Areas The present results on arcopallium connections largely agree with findings from prev ious studies and also extend our knowledge of the specificity of arcopallial striatal connections (Zeier & Karten, 1971; Veenman et al., 1995). The medial MSt region and BSTl have previously been shown to receive more input from the Am, Ai, and PoA than ot her arcopallium divisions (Veenman et al., 1995). In addition, the Tn has been shown to send projections to a restricted portion of the anterior and medial MSt in starlings and ring doves (Cheng et al., 1999). Similar patterns of connections have been foun d in the duck and the chick, further suggesting the existence of different limbic and somatomotor regions of arcopallium (Davies et al., 1997; Dubbeldam et al, 1997). In the present study, the Ad, Ai, Am, and Tn regions of the arcopallium all demonstrated labeled cells after injections of mvMSt and mdMSt. The degree of labeling in Ad and Ai was similar across all of the injection sites. Labelling in the Tn was more prominent after injections of
112 mdMSt than mvMSt. In addition, injections of mdMSt showed the m ost labeling in the Am. Furthermore, only injections into mdMSt resulted in CTb ir cells in the Aa region. The Aa region has been associated with somatomotor function (Zeier & Karten 1971). Another prominent difference was that only mvMSt appears to receiv e projections from the PoA. The PoA is extensively connected with the hypothalamus (Zeier & Karten, 1971) and is associated with visceral limbic function. This region of avian arcopallium receives prominent forebrain projections (Veenman et al., 1995). Bas ed on the divisions of amygdala and associated territories put forth by Swanson and Petrovich (1998), the PoA of birds could be comparable to the basolateral amygdalar nuclei of mammals The current study confirmed that the avian hippocampus is a source of input to the MSt (Veenman et al., 1995). This is an important criterion if some region of the MSt is indeed comparable to the Acc. In the present experiments, injections of retrograde tracer into the mvMSt resulted in labeling in both the Hp and APH. Thos e centered on mdMSt primarily demonstrated a few cells in the APH and none in the Hp itself. These patterns are consistent with the idea that mvMSt is shell like and mdMSt is core like, given that in mammals the hippocampus is a more prominent source of in put to AcS than AcC (Zahm, 2000). In addition, a recent study of intratelencephalic hippocampal projections reported projections to the ventral basal ganglia, including the region called mvMSt in the current study. Th e se authors also noted that the APH was a more prominent source of such projections than the Hp itself (Atoji et al., 2002). A word of caution is warranted, however. While some studies have reported hippocampal striatal projections in pigeon (Veenman et al., 1995; Atoji et al., 2002; present re sults) and zebra finch (Szekely & Krebs, 1996), some have failed to find such a connection (Casini et al., 1986). While these contradictory findin gs are likely due to differences in tracing
113 technique, they will have to be reconciled to determine if, as in mammals, this important limbic structure is a source of input to the MSt. Both of the regions in the current study (mvMSt and pMSt) showed some common forebrain connections. For example, they all showed prominent connections with the mN and mM. Some differ ences were noted, however, for the mM and NFL connections. It appeared that the mM sent somewhat more projections to the mdMSt than the other regions. In addition, while projections from the NFL to MSt have been described previously (Veenman et al., 1995), the present study provides more detail s on these patterns. Injections into the mvMSt yielded retrogradely labeled cells in the NFL. However, injections into the mdMSt did not demonstrate connections from NFL to MSt. The connection of NFL with MSt may serv e as a conduit to send highly processed visual information from the lemnothalamic and collothalamic visual pathways (Husband & Shimizu, 1999). More caudal regions of the avian pallium (Nd, NCL, and NCm) also showed differences in projections to MSt. For in stance, the Nd was shown only to project to the mvMSt, which had not been previously reported (Kroner & Gntrkn, 1999). The projection to NCL, on the other hand, was more limited than expected, given the results of previous studies. Kroner and Gntrkn (1999) found that caudal NCL sent projections to the medial MSt, based on retrograde injections of CTb into the NCL. T he present study did find that NCL sent projections to MSt, particularly the mvMSt But most of the labeled cells from CTb injections into MSt tended to avoid the dorsolateral TH rich region identified as the NCL and were found in the Nd or NCm. In addition, the CPi contained retrogradely labeled cells only after injections of the pMSt.
114 In the Wulst, projections to the MSt were found to com e from several areas, as had been shown in previous studies (Shimizu et al., 1995; Veenman et al., 1995). The HD sent minor projections to all areas of MSt, while the HA and HIS sent more projections to the mvMSt. In mammals, the anterior cingulate sends m ore projections to the AcS than the AcC (Zahm, 2000). Based on some anatomical and functional similarities, there is speculat ion that some portions of the Wulst are comparable to the anterior cingulate, a region which plays a role in executive functions ( Johnson et al., 2004 ). The Wulst could also serve as a source of visual input to the MSt as the telencephalic station of the lemnothalamic visual pathway (Shimizu et al., 1995; Husband & Shimizu, 1999). The differences in forebrain connections to regions of the MSt found in the current study extend our understanding of avian corticostriatal circuits. Using such differences to identify potential homologies with mammalian forebrain areas which project to the Acc (e.g., orbitofrontal cortex) however, is pr oblematic. T hese comparisons are more difficult than those for mesencephalic and diencephalic structures due to the nuclear groupings of avian brain compared to the characteristic layering of mammalian cerebral cortex (Karten & Shimizu, 1991). Does MSt Hav e a Role in Cognitive Function s? On the original visual discrimination there were no discernible effects of lesions. T he present studys hypothesis predicted that lesions would have little if any effect on visual discrimination. This was based on the lack of sensory deficits after Acc lesions in mammals ( Taghzouti et al., 1985; Stern & Passingham, 1995; Ferry et al., 2000), as well as there being no primary visual areas that project directly to the MSt. On reversal learning, there were no significant diffe rences between lesioned subjects and controls.
115 Several factors could have affected the findings of the current study on reversal learning, including the choice of behavioral task and lesion location and extent. Task choice appears to be important, as th ere are even discrepancies in the Acc and reversal literature in mammals Many of these studies have used spatial and visual tasks. In a T maze, rats with ibotenic acid lesions of Acc showed impairments in both the learning of a spatial discrimination and its reversal (Annett et al., 1989) These authors also noted that lesioned rats did not perseverate in choosing the previously reinforced arm. This contrasts with another T maze study in which DA depletion of Acc with 6 OHDA did create deficits in lesioned an imals on revers als of pre vious ly learned habits (Taghzouti et al., 1985) Reading and Dunnett (1991) reported in a somewhat different spatial discrimination task (delayed matching to position) that bilateral ibotenic acid lesions of Acc significantly impai red switching response strategies in a series of reversals In monkeys, ibotenic acid lesions of the Acc did not affect a spatial or visual discrimination Such lesions did impair reversals in the spatial task, but without a significant effect on the visu al reversal (Stern & Passingham, 1995). Th e authors suggested that one reason they failed to find deficits in visual reversal deficits could be a species difference between rats and monkeys The lack of a strong effect in the current study is difficult to reconcile with literature in birds as well. For example, Watanabe (2001) showed that MSt lesions affected the ability of birds to learn a three key spatial discrimination task. Once learned, one of the two previously incorrect keys became the new S+. This change in key reward contingencies would presumably require the same kind of cognitive and motor functions as the two key reversal task employed in the present study. In Watanabe (2001) p igeons made perseverative errors more often than controls The study also found significant
116 effects on both the acquisition phase and later when key reward relationships were changed (similar to a reversal session). Watanabe explained such effects as a form of cognitive rigidity, like that seen in humans with basal ganglia disorders (Watanabe, 2001). However, the present study failed to find large effects on the reversal sessions, even though there was a trend on the first few reversals for lesion subjects to make more perseverative errors. The task used in Watanabe (2001) m ay have been more complex since it involved three key choices. Indeed, in the same study when a color cue was added (i.e., each key was lit with a different color) performance improved. The two key choice of vertical and horizontal patterns used in the cur rent study may have been too simplistic a task to reveal profound deficits after MSt lesions. A different type of study where the avian MSt and mammalian Acc appear comparable is that on impulsivity. Perturbations in Acc function may play a role in atten tion deficit hyperactivity disorder (ADHD), which is characterized by impulsiveness and inattention (Cardinal et al., 2001; King et al., 2003). N ormal animals will tolerate some degree of delay in reward if the subsequent reward is larger or of higher qual ity In animal models of ADHA, an imals with Acc lesions act impulsively; that is they choose stimuli or perform actions associated with lower effort for smaller or lower quality rewards (Cousins et al.,1996; Salamone et al., 1997). A recent study reported similar findings on the effects of MSt lesions in the chick (Izawa et al., 2003). They found that chicks with MSt lesions preferred small rewards with short delay than larger ones requiring a longer delay. This is unlike normals, which showed preference fo r some amount of delay in order to receive larger rewards. They suggested that like the Acc that MSt processes anticipation of reward proximity and helps suppress impulsiveness or habitual actions for immediate gains. Obviously the task employed by Izawa and
117 colleagues is very different than that of reversal learning, but both impulsivity effects and reversal learning appear to involve the Acc. Hence variations in the task in both mammals and birds can lead to very different conclusions about the function of a particular region in the behavior of interest. In any study of neuropathology and behavior, lesion extent and location are crucial factors to consider. T he very short lived effects of the lesions in experimental subjects were not expected. There will be some sparing of neural circuits if lesion size is not sufficient to destroy the entire structure. Potentially, enough of the underlying circuit may be available to perform the task after a brief period of adjustment to the lesion effect. Alternatively, some degree of plasticity may be at play, allowing recovery of the neural networks to accomplish the task after some period. In fact, it is even common in visual discrimination studies where animals with lesions in primary visual structures (e.g., nucleus rotundus of the thalamus) show some recovery of function (Chaves et al., 1993). Subjects with the smallest lesions generally had the least amount of errors, while those with larger lesions showed more deficits in performance. Another factor, which was con founded with lesion size, was lesion location. S maller lesions were more dorsally located, whereas larger ones tended to be more ventral whlile covering more of the MSt in general. In the study of spatial and visual reversal s in monkeys (mentioned above) l esion size may have been a factor for the lack of Acc lesion affects on visual reversal (Stern & Passingham, 1995) Their representative lesion reconstruction shows small lesions that do not involve the whole Acc, primarily only showing damage to the shell of Acc. In Watanabe (2001) the lesions employed were of the entire MSt and much larger than the selective lesions in the current study. In the impulsivity study of Izawa and colleagues
118 (2003), they noted that only lesions to caudal and not rostral MSt wer e effective in creating the impulsivity effect. Interestingly, there is evidence that the AcC and not AcS is more involved in the impulsivity effects mediated by Acc (Cardinal et al., 2001). The lesions employed in the present study tended to be more rostr al; no subject showed much damage in more caudal parts of the MSt. There is no information on differences in AcS versus AcC involvement in reversal learning However it is possible that, like impulsivity effects the AcC is more involved in reversal tasks as well. Hence the lack of strong reversal deficits in the current study could be attributable to the small amount of damage in mdMSt (hypothesized in this study to be more Acc core like) MSt: Activation and Plasticity in Sexual Behavior The present stud y demonstrated that male b irds exposed to females have ZENK upregulation in the MSt. Moreover, this difference was more pronounced in the medial areas of MSt (i.e., the mvMSt and mdMSt ) wh ich show anatomical similarities to the mammalian AcS and AcC, respe ctively The involvement of MSt in anticipatory sexual behavior is likely due to dopaminergic action in this region, as appears to be the case in mammalian Acc. It is important to remember, however, that dopamines participation is not specific to sexual b ehaviors, but reflect its involvement in a wide range of motivated behaviors and reward related processes (Giuliano & Allard, 2001). Specifically, motivated actions like feeding and sexual behavior have been shown to involve the mesolimbic dopamine system, particularly the Acc Also, th is r egion is implicated in long term changes in behavior via neural plasticity mechanisms. Dopamine has long been implicated as playing an important role in motivated behaviors (e.g., sexual behaviors). For example, substance s that enhance dopaminergic tone (e.g., D 1 /D 2 receptor agonists apomorphine) have been used for the treatment of
119 erectile dysfunction (Giuliano & Allard, 2001). Neurons in the VTA are activated by both anticipatory and consummatory sexual behaviors. For ex ample, mu opiod receptors which act on VTA neurons show increased activity when rats had been exposed to sex related environmental cues or actually engaged in copulation (Balfour et al., 2004). In addition, the same study showed more directly that the perc entage of activated DA and non DA neurons in VTA (as measured by Fos levels) increased under the same condition s As a critical part of the mesolimbic system, it is not surprising that this study also found that mating and sex related cues increased activa tion of neurons in the AcC and AcS (Balfour et al., 2004). Lesion studies in male rats also support the involvement of Acc in both anticipatory and consummatory sexual behavior. Male rats with NMDA lesions into the Acc displayed impaired non contact erecti ons compared to controls (an anticipatory measure) as well as reductions in copulations to ejaculation and lower intromissions (consummatory measures) (K ippin et al, 2004). Acc dopamine is also involved in the formation of social attachments and pair bond s in animals. Some species of p rairie voles form monogamous pair s with their partners after mating. Dopamine has been implicated in forming such pair bonds; m ating increases Acc DA turnover by 33% (Aragona et al., 2003). H aloperidol injected into Acc block s partner preference formation, and pair bonds can even be artificially induced in the absence of mating by manipulating DA with apormorphine (Aragona et al., 2003). These effects may rely on the D 2 receptors, since the D 2 antagonist eticlopride injected i nto Acc block s partner preference, whereas the D 2 agonist quinpirole facilitate s partner preference formation in the absence of mating. (Gingrich et al., 2000).
120 There are several lines of research that the avian MSt region plays a similar role in motivated behavior in birds as Acc does in mammals. Studies in avian species like Japanese quail suggest that similar to rodents, increases in DA facilitate anticipatory and consummatory aspects of male sexual behavior. Castagna and collegues (1997) showed that ma nipulations of DA had a variety of effects on anticipatory and consummatory sexual behaviors. The DA receptor agonist apomorphine inhibited both the anticipatory and consummatory aspects of male sexual behavior while amfonelic acid (an indirect DA agonist ) increased aspects of both appetitive and consummatory behaviors. However, the dopamine re uptake inhibitor nomifensine increased mount attempts ( a measure of consummatory behavior ) but decreased a social proximity response (an appetitive sexual behavior ) This dissociation suggests that DA may act on different neural systems for the anticipatory versus consummatory aspects of male sexual behavior. In terms of receptor sub types, male copulatory behavior in quail is stimulated by DA acting on D 1 receptors, but inhibited by activation of the D 2 receptor subtype (Balthazart et al., 1997) In addition to activation via DA and other systems, the Acc may demonstrate neural plasticity, encoding learning about stimuli that subsequently have importance in activati ng motivational and reward systems. Experience dependent effects on mesolimbic DA response to sexual situations have been shown in other mammals. Female Syrian hamsters that experienced more encounters with a male (e.g., six, three, or zero nave) had sig nificant elevations in DA and a prolonged increase in such elevations (Kohlert & Meisel, 1999). So i t may be that repeated experience can have long term effects on dopaminergic tone via plasticity in the mesolimbic system. Lopez and Ettenberg (2002) expose d male rats to an empty arena, one housing a non estrous
121 female, or one containing an estrous female. Expression of the immediate early gene protein Fos i n AcS and AcC showed decreasing levels for the estrous, non estrous, and empty arena conditions. In ad dition, sexually experienced males demonstrated higher Fos expression in response to estrous versus non estrous females (Lopez & Ettenberg, 2002) The authors suggest that long term changes in Acc neurons may serve to enhance motivational responses to fema les and related stimuli. There is evidence that plasticity may also be occurring in the avian MSt Both appetitive and consummatory aspects of male sexual behavior have been shown to increase expression of the Fos protein in male Japanese quail (Tlemcani e t al., 2000; Ball & Balthazart, 2001). In the anticipatory phases of sexual behavior, Fos expression increases in the MSt, as well as the BSTl, and arcopallium (Tlemcani et al., 2000). In zebra finches, a male birds first courtship experience induce s high er Fos expression in MSt compared to c ontrols that were socially isolated or house d in an aviary (Sadananda & Bischof, 2002). Also in songbirds, the degree of attention, arousal, or vigilance affects the selectivity for neurons in the songbird brain for co nspecific song (Park & Clatyon, 2002). It seems likely that dopaminergic transmission in MSt, like that in mammalian Acc, enables birds to ascertain and remember experiences associated with courtship and copulation Plastic changes in the MSt region may se rve to hard wire the context, stimuli, and actions performed to allow more efficient and successful subsequent interactions with conspecifics. C onclusions Within the amniotes one finds numerous similarities in basal ganglia organization, connections, a nd neurochemistry. In particular, there have been many studies about the dorsal (sensori motor) striatum, which has well characterized
122 similarities across all amniotes ( Hall et al., 1984; Reiner et al., 1984a; Reiner & Anderson, 1990; Veenman & Reiner, 199 4) However, th e present study is one of the first that clearly characterize the ventral (limbic) striatum in terms of connections, lesion effects, and gene expression after exposure to conspecific stimuli. Th is study, along with the previous results on dorsal striatum, emphasizes the similarities in the basal ganglia as a whole in different amniotes ( Medina & Reiner, 1997; Medina et al., 1999). Many aspects of both motor and limbic basal ganglia found in vertebrates today probably existed as far back as the ancestral amphibians, and have remained especially conserved within the amniote line (Butler & Hodos, 1996; Reiner et al., 1984a).
123 Literature Cited Anderson KD, and Reiner A (1990) Extensive co occurrence of substance P and dynorphin in striatal projection neurons: an evolutionarily conserved feature of basal ganglia organization. J Comp Neurol 295: 339 69. Anderson KD, and Reiner A (1991) Striatonigral projection neurons: a retrograde labeling study of the percentages that contain substance P or enkephalin in pigeons. J Comp Neurol 303: 658 673. Annett LE, McGregor A, and Robbins TW (1989) The effects of ibotenic acid lesions of the nucleus accumbens on spatial learning and extinction in the rat. Behav Brain Res 31(3): 231 242. Atoji Y, Wil d JM, Yamamoto Y, and Suzuki Y (2002) Intratelencephalic connections of the hippocampus in pigeons ( Columba livia ) J Comp Neurol 447(2): 177 199. Ball GF, and Balthazart J (2001) Ethological concepts revisited: immediate early gene induction in response to sexual stimuli in birds. Brain Behav Evol 57(5): 252 270. Becerra L, Breiter HC, Wise R, Gonzalez RG, and Borsook D (2001) Reward circuitry activation by noxious thermal stimuli. Neuron 32(5): 927 946.
124 Benowitz L, and Lee Teng, E (1973) Contrastin g effects of three forebrain ablations on discrimination learning and reversal in chicks. J Comp Physiol Psychol 84(2): 391 397. Berk ML, and Hawkin RF (1985) Ascending projections of the mammilary region in the pigeon: Emphasis on telencephalic connecti ons. J Comp Neurol 239: 330 340. Bingman VP, Bagnoli P, Ioale P, and Casini G (1984) Homing behavior of pigeons after telencephalic ablations. Brain Behav Evol 24(2 3): 94 108. Bingman VP, Ioale P, Casini G, and Bagnoli P (1990) The avian hippocampus: evidence for a role in the development of the homing pigeon navigational map. Behav Neurosci 104(6): 906 11. Bingman VP, Casini G, Nocjar C, and Jones TJ (1994) Connections of the piriform cortex in homing pigeons ( Columba livia ) studied with fast blue and WGA HRP. Brain Behav Evol 43: 206 18. Blackburn JR, Phillips AG, and Fibiger HC (1987) Dopamine and preparatory behavior: I. Effects of pimozide. Behav Neurosci 101(3): 352 360. Blackburn JR, Pfaus JG, and Phillips AG (1992) Dopamine functions in appetitive and defensive behaviours. Prog Neurobiol, 39(3): 247 279.
125 Braff DL, and Geyer MA (1990) Sensorimotor gating and schizophrenia. Human and a nimal model studies. Arch Gen Psychiatry 47(2): 181 188. Branch MN, and Dearing ME (1982) Effects of acu te and daily cocaine administration on performance under a delayed matching to sample procedure. Pharmacol Biochem Behav 16(5): 713 71 8. Burg B, Haase C, Lindenblatt U, and Delius JD (1989) S ensitization to and conditioning with apomorphine in pigeons. P harmacol Biochem Behav 34(1): 59 64. Bussey TJ, Everitt BJ, and Robbins TW (1997) Dissociable effects of cingulate and medial frontal cortex lesions on stimulus reward learning using a novel pavlovian autoshaping procedure for the rat: Implications for t he neurobiology of emotion. Beh Neurosci 111: 1 12. Butler AB, and Hodos W (1996) Comparative Vertebrate Neuroanatomy: Evolution and Adaptation. New York: Wiley Liss. Cabib S, Kempf E, Schleef C, Oliverio A, and Puglisi Allegra S (1988) Effects of immob ilization stress on dopamine and its metabolites in different brain areas of the mouse: role of genotype and stress duration. Brain Res 441(1 2): 153 160. Canteras NS, Simerly RB, and Swanson LW (1992) Connections of the posterior nucleus of the amygdala J Comp Neurol 324(2): 143 179.
126 Cheng M F, Chaiken M, Zuo M, and Miller H (1999) Nucleus taeniae of the amygdala in birds: Anatomical and functional studies in ring doves ( Streptopelia risoria ) and European starlings ( Sturnus vulgaris ). Brain Behav Evol 53: 243 270. Chudasama Y, Bussey TJ, and Muir JL (2001) Effects of selective thalamic and prelimbic cortex lesions on two types of visual discrimination and reversal learning. European J Neurosci 14: 1009 1020. Clayton DF (2000) The genomic action pot ential. Neurobiol Learn Mem 74(3): 185 216. Cools R, Clark L, Owen AM, and Robbins TW (2002) Defining the neural mechanisms of probabilistic reversal learning using event related functional magnetic resonsance imaging. J Neurosci 22: 4563 4567. Damsma G, Pfaus JG, Wenkstern D, Phillips AG, and Fibiger HC (1992) Sexual behavior increases dopamine transmission in the nucleus accumbens and striatum of male rats: Comparison with novelty and locomotion. Behav Neurosci 106(1):181 191. Dias R, Robbins TW, an d Roberts AC (1996) Dissociation in prefrontal cortex of affective and attentional shifts. Nature 380: 69 72.
127 Dong HW, and Swanson LW (2003) Projections from the rhomboid nucleus of the bed nuclei of the stria terminalis: implications for cerebral hemisp here regulation of ingestive behaviors. J Comp Neurol 463(4): 434 472. Dubbeldam JL, den Boer Visser AM, and Bout RG (1997) Organization and efferent connections of the arcopallium of the mallard, Anas platyrhynchos L.: an anterograde and retrograde trac ing study. J Comp Neurol 388: 632 57. Elena Erro M, Lanciego JL, and Gimenez Amaya JM (2002) Re examination of the thalamostriatal projections in the rat with retrograde tracers. Neurosci Res 42(1): 45 55. Ericson H, and Blomqvist A (1988) Tracing of n euronal connections with cholera toxin subunit B: light and electron microscopic immunohistochemistry using monoclonal antibodies. J Neurosci Methods 24(3): 225 235. Evenden JL, Marston HM, Jones GH, Giadini V, Lenard L, Everitt BJ, and Robbins TW (1989) Effects of lesions of the substantia innominata, ventral and dorsal globus pallidus on visual discrimination acquisition, performance and reversal in the rat. Beh Brain Res 32: 129 149. Everitt BJ, Robbins TW, Evenden JL, Marston HM, Jones GH, and Sirki a TE (1987a) The effects of excitotoxic lesions of the substantia innominata, ventral and dorsal globus pallidus on the acquisition and retention of a conditional discrimination task: Implications for cholinergic hypothesis of learning and memory. Neurosci ence 22: 441 469.
128 Everitt BJ, Fray P, Kostarczyk E, Taylor S, Stacey P (1987b). Studies of instrumental behavior with sexual reinforcement in male rats ( Rattus norvegicus ): I. Control by brief visual stimuli paired with a receptive female. J Comp Psychol 101(4): 395 406. Everitt BJ (1990) Sexual motivation: a neural and behavioural analysis of the mechanisms underlying appetitive and copulatory responses of male rats. Neurosci Biobehav Rev 14(2): 217 232. Everitt BJ (1995) Neuroendocrine mechanisms un derlying appetitive and consummatory elements of masculine sexual behavior. In J. Bancroft (ed): The Pharmacology of Sexual Function and Dysfunction Elsevier, Amsterdam: pp. 15 35. Fellows BJ (1967) Chance stimulus sequences for discrimination tasks. Ps ychological Bulletin 67: 87 92. F erry AT, Lu X CM, and Price JL (2000) Effects of excitotoxic lesions in the ventral striatopallidal thalamocortical pathway on odor reversal learning: Inability to extinguish an incorrect response. Exp Brain Res 131: 32 0 335. Fuster JM (2000) Prefrontal neurons in networks of executive memory. Brain Res Bull 52(5): 331 336.
129 Gagliardo A, Bonadonna F, and Divac I (1996) Behavioural effects of ablations of the presumed 'prefrontal cortex' or the corticoid in pigeons. Be hav Brain Res, 78(2): 155 162. Gagliardo A, Mazzotto M, and Divac I (1997) Memory of radial maze behavior in pigeons after ablations of the presumed equivalent of mammalian prefrontal cortex. Behav Neurosci 111(5): 955 962. Gagliardo A., Ioale P, and B ingman VP (1999) Homing in pigeons: The role of the hippocampal formation in the representation of landmarks used for navigation. J Neurosci, 19: 311 315. Gargiulo LSt, Siemann M, and Delius JD (1998) Visual discrimination in pigeons impaired by glutamate rgic blockade of nucleus accumbens. Physiology & Behavior 63: 705 709. Goeder s NE, Guerin GF, Dworkin SI, and Smith JE (1985) Reinforcing stimulus properties of endocoids. Prog Clin Biol Res 192: 63 69. Gray JA (1995) DA release in the nucleus accumbens : the perspective from aberrations of consciousness in schizophrenia. Neuropsychologia 33: 1143 1153.
130 Groenewegen HJ, Vermeulen Van der Zee E, te Kortschot A, and Witter MP (1987) Organization of the projections from the subiculum to the ventral stri atum in the rat. A study using anterograde transport of Phaseolus vulgaris leucoagglutinin. Neuroscience 23: 103 120. Groenewegen HJ, and Berendse HW (1990) Connections of the subthalamic nucleus with ventral striatopallidal parts of the basal ganglia in the rat. J Comp Neurol 294: 607 622. Haber SN, and McFarland NR (1999) The concept of the ventral striatum in nonhuman primates. Ann N Y Acad Sci 877: 33 48. Haber SN, Fudge JL, and McFarland NR (2000) Striatonigrostriatal pathways in primates form an ascending spiral from the shell to the dorsolateral striatum. J Neurosci 20: 2369 2382. Haber SN, and McFarland (2001) The place of the thalamus in frontal cortical basal ganglia circuits. The Neuroscientist, 7(4): 315 324. Hall K, Brauth SE, and Kitt CA (1984) Retrograde transport of [ 3 H]GABA in the striatotegmental system of the pigeon. Brain Res 310: 157 163.
131 Hartmann B, and G nt rk n O (1998) Selective deficits in reversal learning after nidopallium caudolaterale lesions in pigeons: possible behav ioral equivalencies to the mammalian prefrontal system. Behav Brain Res 96(1 2): 125 133. Heimer L, Zahm DS, Churchill L, Kalivas PW, and Wohltmann C (1991) Specificity in the projection patterns of accumbal AcC and AcS in the rat. Neurosci 41: 89 125. Heimer L, Zahm DS, and Alheid GF (1995) Basal Ganglia. In G Paxinos (ed) : The Rat Nervous System, 2 nd edition San Diego, CA: pp. 589 610 Heimer L, Alheid GF, de Olmos JS, Groenewegen HJ, Haber SN, Harlan R, and Zahm DS (1997) The accumbens: beyond the AcC AcS dichotomy. J Neuropsych Clin Neurosci 9: 354 381. Herdegen T, and Leah JD (1998) Inducible and constitutive transcription factors in the mammalian nervous system: control of gene expression by Jun, Fos and Krox, and CREB/ATF proteins. Brain Res R ev 28(3): 370 490. Hodos W, Karten HJ, and Bonbright JC (1973) Visual intensity and pattern discrimination after lesions of the thalamofugal visual pathway in pigeons. J Comp Neurol: 148(4), 447 467. Hodos W, and Bobko P (1984) A weighted index of bila teral brain lesions. J Neurosci Methods 12: 43 47.
132 Hoebel BG, Monaco AP, Hernandez L, Aulisi EF, Stanley BG, and Lenard L (1983) Self injection of amphetamine directly into the brain. Psychopharmacology ;81(2): 158 163. Husband SA, and Shimizu T (1999) Efferent projections of the ectostriatum in the pigeon ( Columba livia ). J Comp Neurol 406(3): 329 345. Iniguez C, Gayoso MJ, and Carreres J (1985) A versatile and simple method for staining nervous tissue using Giemsa dye. J Neurosci Methods, 13: 77 86 Izawa E, Zachar G, Yanagihara S, and Matsushima T (2003) Localized lesion of caudal part of lobus parolfactorius caused impulsive choice in the domestic chick: evolutionarily conserved function of ventral striatum. J Neurosci 23(5):1894 902. Jones B, and Mishkin M (1972) Limbic lesions and the problem of stimulus reinforcement associations. Exp Neurol 36: 362 377. Jones DL, and Mogensen GJ (1980) Nucleus accumbens GABA projection: Electrophysiological and iontophoretic investigations. Brain Res 188: 93 105. Joyce EM, and Robbins TW (1991) Frontal lobe function in Korsakoff and non Korsakoff alcoholics: Planning and spatial working memory. Neuropsychologia 29: 709 723.
133 Karama S, Lecours AR, Leroux J M, Bourgouin P, Beaudoin G, Joubert S, and Beaur egard M (2002) Areas of brain activation in males and females during viewing of erotic film excerpts. Human Brain Mapping, 16: 1 13. Karten HJ (1991) Homology and evolutionary origins of the neocortex. Brain Behav Evol 38: 264 272. Karten HJ, and Dubb eldam JL (1973) The organization and projections of the paleostriatal complex in the pigeon ( Columba livia ). J Comp Neurol 148: 61 90. Karten HJ, and Hodos W (1967) A Stereotaxic Atlas of the Brain of the Pigeon ( Columba livia ) The Johns Hopkins Press: Baltimore, MD. Karten HJ, and Hodos W (1970) Telencephalic projections of the nucleus rotundus in the pigeon ( Columba livia ). J Comp Neurol 140(1): 35 51. Karten HJ, and Shimizu T (1991) The origins of neocortex: Connections and lamination as distinct e vents in evolution. J Cogn Neurosci 1(4): 291 301. Kelley AE, and Domesick VB (1982) The distribution of the projection from the hippocampal formation to the nucleus accumbens in the rat: an anterograde and retrograde horseradish peroxidase study. Neuro science 7: 3221 3335.
134 Kippin TE, Samaha AN, Sotiropoulos V, and Pfaus JG (2001) The development of olfactoryconditioned ejaculatory preferences in the male rat. II. Parametric manipulation of conditioning session number and duration. Physiol Behav 73(4) : 471 485. Kita H, and Kitai ST (1990) Amygdaloid projections to the frontal cortex and the striatum in the rat. J Comp Neurol 298(1): 40 49. Kitt CA, and Brauth SE (1981) Projections of the paleostriatum upon the midbrain tegmentum in the pigeon. Neuro science 6: 1551 1566. Kitt CA, and Brauth SE (1986a) Telencephalic projections from midbrain and isthmal cell groups in the pigeon. I. Locus coeruleus and subcoeruleus. J Comp Neurol 247(1): 69 91. Kitt CA, and Brauth SE (1986b) Telencephalic projectio ns from midbrain and isthmal cell groups in the pigeon. II. The nigral complex. J Comp Neurol 247(1): 92 110. Koob GF, and Le Moal M (1997) Drug abuse: hedonic homeostatic dysregulation. Science 278: 52 58. Koob GF, and Nestler EJ (1997) The neurobiolo gy of drug addiction. J Neuropsychiatry Clin Neurosci 9: 482 497.
135 Kr ner S, and G nt rk n O (1999) Afferent and efferent connections of the caudolateral neostriatum in the pigeon ( Columba livia ): A retro and anterograde pathway tracing study. J Comp Ne urol 407(2): 228 260. Laverghetta AV and Shimizu T (1999) Visual discrimination in the pigeon ( Columba livia ): Effects of selective lesions of the nucleus rotundus. Neuroreport 10(5): 981 985. Leng NCR, and Parkin, AJ (1988) Double dissociation of dysf unction in organic amnesia. British J Clinical Psych 27: 359 362. Leutgeb S, Husband S, Riters LV, Shimizu T, and Bingman VP (1996) Telencephalic afferents to the caudolateral n eostriatum of the pigeon. Brain Res 730: 173 181. Li M, and McMillan DE (20 01) Four choice drug discrimination in pigeons. Behav Pharmacol 1 2(8): 621 62 8. Lissek S, Diekamp B, and G ntrk n O (2002) Impaired learning of a color reversal task after NMDA receptor blockade in the pigeon ( Columba livia ) associative forebrain (n eost riatum caudolaterale). Behav Neurosci 116(4): 523 529. Long KD, and Salbaum JM (1998) Evolutionary conservation of the immediate early gene ZENK. Mol Biol Evol 15(3): 284 192. MacKintosh NJ (1974) The Psychology of Animal Learning. Academic Press, Lond on.
136 Macphail EM, and Reilly S (1985) Hyperstriatal lesions and short term retention of non visual information in pigeons. Quarterly J Experimental Psych 37B: 121 153. McBride SA, and Slotnik B (1997) The olfactory thalamocortical system and odor reversa l learning examined using an asymmetrical lesion paradigm in rats. Beh Neurosci 111: 1273 1284. McMillan DE (1990) The pigeon as a model for comparative behavioral pharmacology and toxicology. Neurotoxicol Teratol 12(5): 523 52 9. Medina L, and Reiner A (1997) The efferent projections of the dorsal and ventral pallidal parts of the pigeon basal ganglia, studied with biotinylated dextran amine. Neurosci, 81(3): 773 802. Medina L, Jiao Y, and Reiner A (1999) The functional anatomy of the basal ganglia of birds. European J Morphol, 37: 160 165. Mello CV, and Clayton DF (1994) Song induced ZENK gene expression in auditory pathways of songbird brain and its relation to the song control system. J Neurosci 14(11): 6652 6666. Mezey S, and Csillag A (2002) Sel ective striatal connections of midbrain dopaminergic nuclei in the chick (Gallus domesticus). Cell Tissue Res, 308: 35 46.
137 Missale C, Nash SR, Robinson SW, Jaber M, and Caron MG (1998) Dopamine receptors: from structure to function. Physiol Rev, 78: 189 2 25. Mogensen GJ, Jones DL, and Yim CY (1980) From motivation to action: Functional interface between the limbic and motor system. Pr og Neurobiol 14: 69 97. Nauta WJ, and Domesick VB (1984) Afferent and efferent relationships of the basal ganglia. Ciba F ound Symp 107: 3 29. Nader MA, Hoffmann SM, and Barrett JE (1989) Behavioral effects of (+ ) 3,4 methylenedioxyamphetamine (MDA) and (+ ) 3,4 methylenedioxymethamphetamine (MDMA) in the pigeon: interactions with noradrenergic and serotonergic systems. Ps ychopharm (Berl) 98(2):183 18 8. Olds ME, and Forbes JL (1981) The central basis of motivation: intracranial self stimulation studies. Annu Rev Psychol 32: 523 574. Park KH, and Clayton DF (2002) Influence of restraint and acute isolation on the selec tivity of the adult zebra finch zenk gene response to acoustic stimuli. Behav Brain Res 136(1): 185 194. Pfaus JG, and Phillips AG (1991) Role of dopamine in anticipatory and consummatory aspects of sexual behavior in the male rat. Behav Neurosci 105(5) : 727 743.
138 Pfaus JG, Damsma G, Wenkstern D, and Fibiger HC (1995) Sexual activity increases dopamine transmission in the nucleus accumbens and striatum of female rats. Brain Res 693(1 2): 21 30. Pfaus JG, and Heeb MM (1997) Implications of immediate ear ly gene induction in the brain following sexual stimulation of female and male rodents. Brain Res Bull 44(4): 397 407. Powers AS, Halasz F, and Wasiams S (1982) The effects of lesions in telencephalic visual areas of pigeons on dimensional shifting. Phys iology & Behavior 29: 1099 1104. Puelles L, Kuwana E, Puelles E, Bulfone A, Shimamura K, Keleher J, Smiga S, and Rubenstein JL (2000) Pallial and subpallial derivatives in the embryonic chick and mouse telencephalon, traced by the expression of the gene s Dlx 2 Emx 1 Nkx 2.1 Pax 6 and Tbr 1 J Comp Neurol, 424(3): 409 438. Pycock C, and Horton R (1976) Evidence for an accumbens pallidal pathway in the rat and its possible gabaminergic control. Brain Res 110: 629 634. Rada PV, and Hoebel BG (2001) A versive hypothalamic stimulation releases acetylcholine in the nucleus accumbens, and stimulation escape decreases it. Brain Res 888(1): 60 65.
139 Rafal RD, and Posner MI (1987) Deficits in human visual spatial attention following thalamic lesions. Proceedi ngs of the Academy of Science USA 84: 7349 7353. Rebec GV, Christensen JR, Guerra C, and Bardo MT (1997) Regional and temporal differences in real time dopamine efflux in the nucleus accumbens during free choice novelty. Brain Res 776(1 2): 61 67. Read ing PJ, and Dunnett SB (1991) The effects of excitotoxic lesions of the nucleus accumbens on a matching to position to task. Behav Brain Res 46(1): 17 29. Reading PJ, Dunnett SB, and Robbins TW (1991) Dissociable roles of the ventral, medial and lateral striatum on the acquisition and performance of a complex visual stimulus response habit. Behav Brain Res 45(2): 147 161. Reilly S (1987) Hyperstriatal lesions and attention in the pigeon. Behav Neurosci 101(1): 74 86. Reiner A, and Anderson KD (1990) The patterns of neurotransmitter and neuropeptide co occurance among striatal projection neurons: Conclusion based on recent findings. Brain Res Rev 15: 251 265. Reiner A, and Carraway RE (1987) Immunohistochemical and biochemical studies on Lys8 Asn9 ne urotensin8 13 (LANT6) related peptides in the basal ganglia of pigeons, turtles, and hamsters. J Comp Neurol 257(3): 453 476.
140 Reiner A, and Karten HJ (1983) The laminar source of efferent projections from the avian Wulst. Brain Res 275(2): 349 354. Rei ner A, Brauth SE, and Karten HJ (1984) Evolution of the amniote basal ganglia. Trends in Neuroscience 7: 320 325. Reiner A, Davis BM, Brecha NC, and Karten HJ (1984b) The distribution of enkephalin like immunoreactivity in the telencephalon of the adult and developing domestic chicken. J Comp Neurol 228: 245 262. Reiner A, Karten HJ, and Solina AR (1983) Substance P: localization within paleostriatal tegmental pathways in the pigeon. Neurosci 9(1): 61 85. Reiner A, Karle EJ, Anderson KD, and Medina L (1994) Catecholaminergic perikarya and fibers in the avian nervous system. In Smeets WJAJ and Reiner A (eds): Phylogeny and Development of Catecholamine Systems in the CNS of Vertebrates 135 181, Cambridge University Press. London. Ritchie TLC (1979) Int ratelencephalic visual connections and their relationship to the arcopallium in the pigeon ( Columba livia ), Department of Physiology (pp. 217). Charlottesville: University of Virginia.
141 Robbins TW, and Everitt BJ (1996) Neurobehavioural mechanisms of rewar d and motivation. Curr Opin Neurobiol 6: 228 36. Roberts AC, Robbins TW, Everitt BJ, Jones GH, Sirikia TE, Wilkinson J, and Page K (1990) The effects of excitotoxic lesions on the basal forebrain on the acquisition, retention, and serial reversal of visu al discriminations in marmosets. Neuroscience 34: 311 329. Sadananda M, and Bischof HJ (2002) Enhanced fos expression in the zebra finch ( Taeniopygia guttata ) brain following first courtship. J Comp Neurol 448(2): 150 164 Salamone JD (1994) The involve ment of nucleus accumbens dopamine in appetitive and aversive motivation. Behav Brain Res 61(2): 117 133. Salamone JD, Cousins MS, and Snyder BJ (1997) Behavioral functions of nucleus accumbens dopamine: empirical and conceptual problems with the anhedon ia hypothesis. Neurosci Biobehav Rev 21(3): 341 59. Shimizu T, and Hodos W (1989) Reversal learning in pigeons: Effects of selective lesions of the Wulst. Beh Neurosci 103: 262 272. Shimizu T, Woodson W, Karten HJ, and Schimke JB (1989) Intratelenceph alic connections of the visual areas in birds. Soc Neurosci Abstr 15, 1398.
142 Shimizu T, Karten HJ, and Schimke JB (1990) Intratelencephalic projections of the visual Wulst in birds ( Columba livia ): A Phaseolus vulgaris leucoagglutinin study. Soc Neurosci Abstr 16: 246. Shimizu T, Cox K, and Karten HJ (1995) Intratelencephalic projections of the visual wulst in pigeons ( Columba livia ). J Comp Neurol 359(4): 551 572. Spetch ML, and Treit D (1984) T he effect of d amphetamine on short term memory for time in pigeons. Pharmacol Biochem Behav 21(4): 663 66 6. Stern CE, and Passingham RE (1995) The nucleus accumbens in monkeys ( Macaca fascicularis ). III. Reversal learning. Exp Brain Res 106(2): 239 247 Stewart MG, Kabai P, Harrison E, Steele RJ, Kossut M, G ierdalski M, and Csillag A (1996) The involvement of dopamine in the striatum in passive avoidance training in the chick. Neurosci 70: 7 14. Stoeckel K, Schwab M, and Thoenen H (1977) Role of gangliosides in the uptake and retrograde axonal transport of cholera and tetanus toxin as compared to nerve growth factor and wheat germ agglutinin. Brain Res 132(2): 273 285. Swanson LW, and Petrovicha GD (1999) What is the amygdala? A comparative approach. Trends Neurosci 22 (5): 323 331.
143 Szekely AD, Boxer MI, Steward MG, and Csillag A (1994) Connectivity of the lobus parolfactorius of the domestic chicken ( Gallus domesticus ): An anterograde and retrograde pathway tracing study. J Comp Neurol 348: 374 393. Szekely AD, and Krebs JR (1996) Efferent connectivity of the hippocampal formation of the zebra finch ( Taenopygia guttata ): An anterograde pathway tracing study using Phaseolus vulgaris leucoagglutinin. J Comp Neurol 368(2): 198 214. Taghzouti K, Louilot A, Herman JP, LeMoal M, and Simon H (1985) Alternati on behavior, spatial discrimination, and reversal disturbances following 6 hydroxydopamine lesions in the nucleus accumbens of the rat. Beh Neurobiol 44: 354 363. Tan Y, Wasiams ES, and Zahm DS (1997) Calbindin D 28kD, but not calretinin, gluR1 or gluR3/ 2 immunoreactivities in retrogradely labeled ventral mesencephalic neurons following injections of Fluoro Gold in accumbal subterritories. Soc Neurosci Abstr 23: 1282. Thompson, DM (1977) Development of tolerance to the disruptive effects of cocaine on r epeated acquisition and performance of response sequences. J Pharmacol Exp Ther 203: 294 302. Tlemcani O, Ball GF, D'Hondt E, Vandesande F, Sharp PJ, and Balthazart J (2000). Fos induction in the Japanese quail brain after expression of appetitive and co nsummatory aspects of male sexual behavior. Brain Res Bull 2000 Jul 1;52(4): 249 262.
144 Tmbl T ( 1995 ) Lobus parolfactorius and nucleus accumbens. In: The G olgi S tructure of T elencephalon of C hicken ( Gallus domesticus ) II St. Georges Hospital Medical Sc hool, London: pp. 227 242. Veenman CL, Reiner A., and Honig MG (1992) Biotinylated dextran amine as an anterograde tracer for single and double labeling studies. J Neurosci Meth 41: 239 254. Veenman CL, and Reiner A (1994) The distribution of GABA cont aining perikarya, fibers, and terminals in the forebrain and midbrain of pigeons, with particular reference to the basal ganglia and its projection targets. J Comp Neurol 339: 209 250. Veenman CL, Wild JM, and Reiner A. (1995) Organization of the avian corticostriatal" projection system: A retrograde and anterograde pathway tracing study in pigeons. J Comp Neurol 354: 87 126. Veenman CL, Medina L, and Reiner A (1997) Avian homologues of mammalian intralaminar, mediodorsal and midline thalamic nuclei: i mmunohistochemical and hodological evidence. Brain Behav Evol 49(2): 78 98. Waldmann C, and Gntrkn O (1993) The dopaminergic innervation of the pigeon caudolateral forebrain: Immunocytochemical evidence for a prefrontal cortex in birds? Brain Res 6 00: 225 234.
145 Wang YC, Jiang S, and Frost BJ (1993) Visual processing in pigeon nucleus rotundus: luminance, color, motion, and looming subdivisions. Vis Neurosci, 10(1): 21 30. Watanabe S (2001) Effects of lobus parolfactorius lesions on repeated acquisi tion of spatial discrimination in pigeons. Brain Behav Evol 58(6):333 342. White AR, Strasser R, and Bingman VP (2002) Hippocampus lesions impair landmark array spatial learning in homing pigeons: a laboratory study. Neurobiol Learn Mem 78: 65 78. Wild JM (1987) Thalamic projections to the paleostriatum and nidopallium in the pigeon ( Columba livia ). Neurosci 20(1): 305 327. Wild JM, Karten HJ, and Frost BJ (1993) Connections of the auditory forebrain in the pigeon ( Columba livia ). J Comp Neurol 337(1 ): 32 62. Winsauer PJ, and Thompson DM (1991) Cocaine self administration in pigeons. Pharmacol Biochem Behav 40(1): 41 52. Zahm DS (1999) Functional anatomical implications of the nucleus accumbens AcC and AcS subterritories. Ann N Y Acad Sci 877: 113 129.
146 Zahm DS (2000) An integrative neuroanatomical perspective on some subcortical substrates of adaptive responding with emphasis on the nucleus accumbens. Neurosci Biobehav Rev 24: 85 105. Zeier H, and Karten HJ (1971) The arcopallium of the pigeon: Organization of afferent and efferent connections. Brain Res 31: 313 326.
147 APPENDIX A Neurological Examination CHECKLIST SUBJECT #: xxx ++ Sign Normal / Reflex Present + Present, but weak or abnormal Date of Lesion: Absent Date of Exam: + Questionable Lesion Target: OBSERVATIONS Pre Op Post Op COMMENTS Head Carriage (Retrocollis,Torticollis) Muscle Tone (Hyper, Hypoflexion) Posture (e.g., Tremor) Static Ocular Signs (M ydriasis, Miosis) Piloerection Alertness (e.g., Drowsy, Aggressive) REFLEXES Corneal Pupillary (e.g. Aniscoria) Grasp Contralateral Thrust Visual Righting Explanations: Retrocollis is holding the head back, Torticollis is holding the head to one side (coud be damage to tectum, vestibular system, or spinal accessory nerve) Limb & neck muscles; Hyperflexion (unusual rigidity) or Hypoflexion (flaccidity) Visible, low amplitude Tremor (like PD; discrup tion of nigrostriatal pathway) Mydriasis (chronic, extreme dilation of the pupil) and Miosis (excessive pupillary constriction) (mydriasis could be nerve III parasympathetic and miosis the sympathetic branch of same); Ptosis is chronic drooping of the ey elid Piloerection means fluffed out or ruffled feathers Corneral Reflex ; touch eye with wisp of cotton; trigeminal and facial nerve Pupillary Light Reflex ; adapt to low light for several minutes (damage could be optic nerve, ciliary ganglia, pretecta l complex, occulomotor nerve); Aniscoria is uneven pupil size (injury to pretectal nuclei) Extensor Grasp & Thrust Reaction ; wooden rod 1cm in diameter slapped against plantar surface of outstretched foot, stimulated foot should grab rod and contralatera l leg is suddenly extended
148 APPENDIX B Additional CTb Cases Pg 225 CTb injection into mvMSt In Pg225 (Figure 8), the injection site is centered in an anterior area of MSt, with a dense injection deposit in the mvMSt, with a secondary one in the mdMSt (Figure 8B). There was also an additional, lighter deposit along the injection track in the medial HA. As was seen in Pg220, CTb ir cells were found in the NFL but were far denser in this case (Figure 8A B). Several divisions of the hyperpallium and mesop allium, including the HA, HI, HD, and mM exhibited labeled cells. The CTb ir cells in HA, HI, and HD (Figure 8A E) may be attributable to tracer leakage at the injection site. The HI tended to show very few cells, with the majority of labeling being found in the HA and clustered in the medial HD (Figure 8B D). In mM a narrow band of cells running dorsolaterally from the ventricle were visible (Figure 8B C). In the mN labeled cell bodies were again, as in similar cases, restricted to a dense field adjacent to the cerebral ventricle. Cells in the Nd were restricted to a dorsal band at the upper edge of the telencephalon (Figure 8G, with a smaller number visible in the NCL. A number of cells were seen in the ventral region of VP (Figure 8C D). Numerous cells w ere visible in the APH and Hp (Figure 8C F), although this may have be en artifact from leakage at the injection site into HA. There were also a few CTb ir cells in the LHy (Figure 8E). In the arcopallium, a few CTb ir cells were found in the Am, with large r numbers in the Ad and especially the Ai (Figure 8F G). T raveling caudally the PoA exhibited a moderate number of labeled cells
149 (not shown). In the thalamus, a large number of cells were found in the DMA, DLM, (Figure 8H) and DMP (not shown). In the midbr ain, a compact zone of labeled cells was found in the VTA; an area containing fewer cells was seen in the SNc (Figure 8I and 8J, respectively). There was also a small field of CTb ir cells in the LoC (Figure 8K). Pg 194 CTb injection into mdMSt For subj ect Pg194 (Figure 9), the injection site was narrow, as opposed to the spherical shape that is typical, and was located more toward the middle third of the structure (Figure 9C). In nidopallium and mesopallium, cell bodies labeled with CTb were seen in the NFL and mN (Figure 9A D), with much fewer cells seen in the mM (Figure 9B C) than other injection cases. The cells found in the HA (Figure 9D) may be injection artifact. A few cells were in the area of the VP (Figure 9D), with a few also found in Hp/APH ( Figure 9D E), and medial PE/TPO (Figure 9E). A few labeled cells were seen in the Ad of arcopallium (Figure 9F), with much denser fields of CTb ir cell bodies seen in the Ai (Figure 9F G). In the diencephalon, numerous cells were seen both in DMA and DMP ( Figure 9H I). In the mesencephalon, a moderate number of retrogradely labeled cells were seen in both the VTA and the SNc (Figure 9J and K, respectively).
150 APPENDIX C Additional BDA Cases Pg195 BDA injection into mvMSt The injection site of subject Pg195 was also in the mvMSt (Figure 13B), but avoided the area directly lateral to ventricle. This case tended to show fewer fibers in various brain regions; this was more likely a function of the volume of tracer deposit rather than the injection location itself. The fibers seen in the HA were likely due to artifact from the injection track (Figure 13A B). Fibers reactive for the presence of BDA were also found in the HD and HV (Figure 13A C). In the nidopallium, BDA labeled fibers were seen in th e NFL, where they were limited in number (Figure 1 3 A B). The mN contained fibers as well (Figure 13B C), and were accompanied by retrogradely labeled cells. No fibers were seen in the Hp or the APH. In the arcopallium only the lateral portion of Ad and Ai contained a few, faint fibers (Figure 13G). The ventral aspect of the VP showed numerous BDA labeled fibers (Figure 13D E). In the thalamus, unlike other cases, no fibers or cells were seen in the DMA or DMP. In the midbrain, fibers were found in the VTA ( Figure 13I), with a notable lack of labeling in either the SNc or LoC (not shown). Pg19 3 BDA injection into mdMSt For subject Pg193 (Figure 16), the injection was more centrally located, covering the middle third of MSt and mdMSt (Figure 16C). A small n umber of fibers were seen in
151 the HD and mM of mesopallium, with both structures also containing labeled cells (Figure 16A C). In the nidopallium, numerous labeled fibers and cells were seen in the mN (Figure 16A C). A small number of dispersed fibers were seen in the posterior PE/TPO (Figure 16E). In the caudal nidopallium, stained fibers for BDA were concentrated in a small portion of the NCm (Figure 16F), the Nd (Figure 16G), and a small region of the NCL (Figure 16G). In the hippocampal formation, no fib ers were seen in the Hp, with only a few fibers present in the APH (Figure 16E). In the arcopallium, fibers were found in the Aa, Ad, Ai, and Am (Figure 16E and F G, respectively). The densest fields of such fibers were found in the Ai and Am (Figure 16F G ), with a somewhat lower number of fibers in the Ad (Figure 16G). A small number of fibers were dispersed over the VP area (Figure 16D E). In addition, the LSt also contained a moderate number of fibers (Figure 16D E). In the thalamus, fibers were seen in the DMA and DMP in conjunction with labeled cell bodies (Figure 16H I). Ventral to this thalamic area, both the FPM (Figure 16H) and OM (Figure 16I) fiber tracts contained bundles of BDA labeled fibers. In the midbrain, no fibers or cells were found in the VTA, although the SNc exhibited numerous fibers and cells (Figure 16J K). Pg1 81 (left hemisphere) BDA injection into mdMSt The BDA injection into the left hemisphere of Pg181 (Figure 18) was also in the md MSt. This injection mostly avoided the BSTl, a lthough the dorsal edge of this structure may have received a small deposit of the tracer. This injection (Figure 18D) was also more medial than that of the Pg181 right hemisphere case. In the hyper and mesopallium fibers were mainly seen in the HD and mM Fibers in the HD were found primarily along its ventral border, both at the site of injection and in sections anterior and posterior to it (Figures 18A B, D). As compared to other cases, the mM (Figure 18A C)
152 showed relatively moderate numbers of fibers extending into sections away from the injection site; retrogradely labeled cells again accompanied these mM fibers. In the nidopallium, the mN and a small portion of the NCL were found to have BDA labeled fibers. The mN fiber s positive for BDA (Figure 1 8 A E) w ere dense and in proximity to retrogradely labled cells. Sections of the NCL (Figure 1 8 G) showed a limited fiber field just dorsal to the arcopallium. No fibers were seen in the Hp or the APH. The MSt showed fibers mostly anterior to the injection site which tended to cluster in the medial part of the structure (Figure 1 8 A C). A dense band of fibers was seen traveling to the VP, especially its dorsal portion (Figure 1 8 D E). The arcopallium showed fibers in both the Aa (Figure 1 8 F), Ad, and Ai (Figure 1 8 G), with there being somewhat more labeling in the Ai than other structures (Figure 1 8 G). Both the DMA (Figure 1 8 H) and DMP (not shown) divisions of the dorsal thalamus exhibited fibers and cells. The OM fiber tract ventral to the thalamus (Figure 1 8 H) co ntained a bundle of labeled fibers. Numerous BDA labeled fibers in the VTA and SNc were seen (Figures 1 8 I and J, respectively). Some fibers with retrogradely labeled cells were also seen in the LoC (Figure 1 8 K). Pg 241 (left hemisphere) BDA injection int o m N In the left hemisphere of subject Pg241 (Figure 21) the injection site was centered on the mN, with no spread beyond the lamina separating it from the mM or MSt (Figure 23C). In the hyper and mesopallium, labeled fibers were largely restricted to the mM and HI. The mM fibers were accompanied by filled cells, which were presumably retrogradely labeled (Figure 23A E). Fibers in HI were fewer in numbers and tended to cluster along the lamina separating it from the overlying mM (Figure 23B C). For areas o f the nidopallium, BDA containing fibers were seen in the NFL, mN, Nd, NCm, and NCL. The NFL fibers were sparse (Figure 23A) and restricted to its far lateral edge. In
153 contrast, the mN contained numerous stained fibers and cells (Figure 23A C); in more pos terior sections these fibers became more sparse and extended to more lateral parts of the nidopallium (Figure 23D E). In the NCm, Nd, and NCL fibers tended to be fewer in number and spread over most of the structure (Figure 23F G). Fibers extending from th e injection site to the MSt tended to accumlate in the medial two thirds of the structure in anterior sections (Figure 23B C) while those in more posterior sections were more dispersed (Figure 23D E). A small number of fibers were found in the Ap (Figure 2 3F). In the VP a few dispersed fibers were found (Figure 23D E). In the thalamus, a moderate number of BDA filled fibers and cell bodies were seen in both the DMA (Figure 23H) and DMP (Figure 23I) subdivisions. The only midbrain fibers detected were a smal l number in the SNc (Figure 23J).
154 About the Author Scott Husband received a Bachelors Degree in Psychology from the University of South Florida in 1992 and a M.A. in Experimental Psychology from the same institution in 1998. He taught sev eral lower and upper level undergraduate courses in psychology during his graduate student training. In addition he worked in several neuroscience labs at the University of South Florida. During this period he presented his research on several occasions at international conferences, including the Society for Neuroscience and the Conference on Comparative Cognition. He also published several scholarly works, including one based on his Masters Thesis work on the anatomy of the visual system in birds, and als o a World Wide Web based book on the evolution of the avian visual system. He then entered the Ph.D. program in the Cognitive and Neural Sciences division of the Psychology department at the University of South Florida in 2002 Whle in the Ph.D. program at th e University of South Florida, Mr. Husband continued to teach and perform laboratory research in the neurosciences. He is also a member of several scientific societies, including the American Association for the Advancement of Science, the Comparative Cogn ition Society (Animal Cognition), the J.B. Johnston Club (Comparative and Evolutionary Neurobiology), and the Society for Neuroscience