<%BANNER%>

In Vivo and ex Vivo MRI Studies of an Animal Model of Mesial Temporal Lobe Epilepsy Correlated with Histological Analysis

Permanent Link: http://ufdc.ufl.edu/UFE0021884/00001

Material Information

Title: In Vivo and ex Vivo MRI Studies of an Animal Model of Mesial Temporal Lobe Epilepsy Correlated with Histological Analysis
Physical Description: 1 online resource (69 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: epilepsy, histology, mri
Biomedical Engineering -- Dissertations, Academic -- UF
Genre: Biomedical Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Epilepsy is one of the most common neurological conditions worldwide and a significant cause of morbidity in modern society, especially in children and young adults. Mesial temporal lobe epilepsy (MTLE) is the most common form of symptomatic localization-related human epilepsy and the most common epilepsy syndrome with pharmacologically intractable partial-onset seizures. Patients often have an initial precipitating injury during early childhood, followed by a latent period of up to several years before the emergence of complex partial seizures. Understanding the mechanisms involved during epileptogenesis as well as the consequences of initial injuries is crucial in the diagnosis of epilepsy and development of effective strategies for therapeutic intervention. Whether seizures cause permanent damage to the brain and lead to an epileptic lesion resulting in long-term epilepsy is a subject of controversy. Several cross-sectional magnetic resonance imaging (MRI) and histology studies have reported changes suggestive of seizure-associated damage in animal models. Very few longitudinal studies have been performed to describe acute seizure-associated changes and their evolution over time. Since epileptic conditions in humans develop over long periods of time and have effects extending over many years, clinical studies seeking the causes and mechanisms underlying the development of epilepsy are very difficult and expensive to perform. Animal models allow for experiments that investigate the full dynamic course of the epileptic state. In vivo MRI permits the non-invasive detection of structural alterations in living animals, the dynamics of involvement and interactions between all the brain regions, as well as their progression in a single individual over the entire time-course of the disease. The animal model of chronic limbic epilepsy used in this study is characterized by recurrent spontaneous hippocampal seizures following an episode of status epilepticus (SE) induced by a period of electrical hippocampal stimulation. Spontaneous seizures begin several weeks later and physiologically resemble seizures seen in human MTLE. High-resolution in vivo MRI was used to evaluate the progression of brain pathological alterations in the same animal at different time points, followed by ex vivo MRI, three-dimensional magnetic resonance microscopy at high field strength and histological analysis. The findings demonstrated that development of seizures was associated with significant changes in relaxation measurements following SE in the hippocampus, lateral ventricle regions, thalamus and parahippocampal gyrus, visible or not as changes in signal intensity on T1 or T2-weighted in vivo MRI images. These in vivo changes were due to underlying pathological tissue alterations evident with ex vivo MRI and histology, including iron deposition, neuronal degeneration and microglia aggregation. Correlation of longitudinal in vivo imaging data with behavioral observations and post-mortem ex vivo MRI and histopathological analysis can help provide a more complete understanding of the causes and mechanisms underlying epileptogenesis. Continued improvement and application of these MRI techniques, particularly the development of diffusion-weighted and diffusion tensor imaging, will allow additional insights into the mechanisms involved in the epileptogenic process.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Mareci, Thomas H.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021884:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021884/00001

Material Information

Title: In Vivo and ex Vivo MRI Studies of an Animal Model of Mesial Temporal Lobe Epilepsy Correlated with Histological Analysis
Physical Description: 1 online resource (69 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: epilepsy, histology, mri
Biomedical Engineering -- Dissertations, Academic -- UF
Genre: Biomedical Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Epilepsy is one of the most common neurological conditions worldwide and a significant cause of morbidity in modern society, especially in children and young adults. Mesial temporal lobe epilepsy (MTLE) is the most common form of symptomatic localization-related human epilepsy and the most common epilepsy syndrome with pharmacologically intractable partial-onset seizures. Patients often have an initial precipitating injury during early childhood, followed by a latent period of up to several years before the emergence of complex partial seizures. Understanding the mechanisms involved during epileptogenesis as well as the consequences of initial injuries is crucial in the diagnosis of epilepsy and development of effective strategies for therapeutic intervention. Whether seizures cause permanent damage to the brain and lead to an epileptic lesion resulting in long-term epilepsy is a subject of controversy. Several cross-sectional magnetic resonance imaging (MRI) and histology studies have reported changes suggestive of seizure-associated damage in animal models. Very few longitudinal studies have been performed to describe acute seizure-associated changes and their evolution over time. Since epileptic conditions in humans develop over long periods of time and have effects extending over many years, clinical studies seeking the causes and mechanisms underlying the development of epilepsy are very difficult and expensive to perform. Animal models allow for experiments that investigate the full dynamic course of the epileptic state. In vivo MRI permits the non-invasive detection of structural alterations in living animals, the dynamics of involvement and interactions between all the brain regions, as well as their progression in a single individual over the entire time-course of the disease. The animal model of chronic limbic epilepsy used in this study is characterized by recurrent spontaneous hippocampal seizures following an episode of status epilepticus (SE) induced by a period of electrical hippocampal stimulation. Spontaneous seizures begin several weeks later and physiologically resemble seizures seen in human MTLE. High-resolution in vivo MRI was used to evaluate the progression of brain pathological alterations in the same animal at different time points, followed by ex vivo MRI, three-dimensional magnetic resonance microscopy at high field strength and histological analysis. The findings demonstrated that development of seizures was associated with significant changes in relaxation measurements following SE in the hippocampus, lateral ventricle regions, thalamus and parahippocampal gyrus, visible or not as changes in signal intensity on T1 or T2-weighted in vivo MRI images. These in vivo changes were due to underlying pathological tissue alterations evident with ex vivo MRI and histology, including iron deposition, neuronal degeneration and microglia aggregation. Correlation of longitudinal in vivo imaging data with behavioral observations and post-mortem ex vivo MRI and histopathological analysis can help provide a more complete understanding of the causes and mechanisms underlying epileptogenesis. Continued improvement and application of these MRI techniques, particularly the development of diffusion-weighted and diffusion tensor imaging, will allow additional insights into the mechanisms involved in the epileptogenic process.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Mareci, Thomas H.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021884:00001


This item has the following downloads:


Full Text

PAGE 1

IN VIVO AND EX VIVO MRI STUDIES OF AN ANIMAL MODEL OF MESIAL TEMPORAL LOBE EPILEPSY CORRELATE D WITH HISTOLOGICAL ANALYSIS By LAN HOANG MINH A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 Lan Hoang Minh 2

PAGE 3

To my family and friends for their love and encouragement. 3

PAGE 4

4 ACKNOWLEDGEMENTS First, I would like to thank my committee memb ers, Drs. Thomas Mareci, Paul Carney and Steve Blackband for their guidanc e and encouragement. I would like to thank my mentor, Dr. Mareci, for his continuous support and for shap ing my knowledge of MR imaging, as well as teaching me how to operate the magnet system s; Dr. Carney, for providing his expertise on epilepsy and guiding the project by emphasizing the clinical rele vance of our research; Dr. Blackband, for providing me a found ation on the basic principles of MR imaging. I would also like to thank Dr. Michael King for hi s help with histology experiments. I am indebted to all of the current and former members of the Mareci and Carney labs that I have had the opportunity to work with over the past two years. In particular, I would like to thank my lab mate, Hector Sepulveda for buildin g the surface coils, the stereotaxic frame and helping with MR imaging data acquisition; Mans i Parekh for contributing to MR imaging data acquisition and some of the histology experiments; Dr. Wendy No rman, for taking care of the animals and monitoring them over the entire course of in vivo imaging; Angela Hadlock, for performing the surgical implantati on of electrodes; Chad Durgin, for improving the data analysis software; and Jessica Meloy and Matthew Feldman for providing assistance in data analysis. I also appreciate other current and former lab members, Garrett Astary, Nelly Volland, Chris Taylor, Min Sig Hwang, for thei r help and moral support. I would like to give credit to the A dvanced Magnetic Resonance Imaging and Spectroscopy (AMRIS) facility and staff at the McKnight Brain In stitute of the University of Florida. I thank Xeve Silver and Gary Blaskowski for technical support w ith the magnet systems; Barbara Beck and Kelly Jenkins, for their expert ise in coil setup and design; and Jim Rocca and Dan Plant for making sure that all the machines we re running properly. I am also grateful for the

PAGE 5

financial support provided by the University of Florida Al umni Foundation, the McKnight Brain Institute, Wilder Epilepsy Research Center and National Institutes of Health. I would also like to thank my family in France and in the Un ited States for their continuous love, support and encouragement throughout the year s; my friends in Gainesville, especially Berenger, for their moral support, particularly during the last months while finishing my experiments and writing my thesis. 5

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGEMENTS .............................................................................................................4 LIST OF FIGURES .........................................................................................................................8 ABSTRACT .....................................................................................................................................9 CHAPTER 1 INTRODUCTION ..................................................................................................................11 Mesial Temporal Lobe Epilepsy .............................................................................................11 Overview .........................................................................................................................11 Pathophysiology ..............................................................................................................12 Animal Model of Mesial Temporal Lobe Epilepsy .........................................................15 In vivo Magnetic Resonance Imaging .............................................................................16 Ex vivo Magnetic Resonance Imaging ............................................................................18 2 MATERIALS AND METHODS ...........................................................................................21 Animal Preparation .................................................................................................................21 Implantation of Hippocampal Electrodes ........................................................................21 Induction of Status Epilepticus, Video Monitoring and Analysis of Spontaneous Seizures ........................................................................................................................22 In vivo MR Imaging ................................................................................................................22 In vivo High Resolution MR Im aging Data Acquisition ........................................................23 Ex vivo MR Imaging ...............................................................................................................24 Histological Procedures ..........................................................................................................25 Fixation ............................................................................................................................25 Nissl Staining ...................................................................................................................25 Perls Staining .................................................................................................................26 Fluoro-Jade C ..................................................................................................................26 Timm Staining .................................................................................................................26 Black Gold Staining ........................................................................................................27 Data Analysis ...................................................................................................................27 3 RESULTS ...............................................................................................................................31 Animal Behavior .....................................................................................................................31 In vivo High Resolution MR Imaging ....................................................................................32 Ex vivo MR Microscopy .........................................................................................................33 Histologic Findings .................................................................................................................34 4 DISCUSSION................................................................................................................... ......54

PAGE 7

LIST OF REFERENCES. ..............................................................................................................61 BIOGRAPHICAL SKETCH .........................................................................................................69 7

PAGE 8

8 LIST OF FIGURES Figure page 1-1 Hippocampal Sclerosis .......................................................................................................20 2-1 Stimulation and MR imaging timeline ...............................................................................29 2-2 Coronal T2-weighted image of implanted control excised brain showing anatomical regions of interest ...............................................................................................................30 3-1 Coronal in vivo T2 MRI in implanted control animal ........................................................36 3-2 Coronal in vivo T1-weighted MRI scans in epileptic animal ............................................37 3-3 Coronal in vivo T2-weighted MRI scans in epileptic animal ............................................38 3-4 Average T2 relaxation times in epileptic an imals before stimulation, 3 to 10 days and 20 to 60 days after stimulation ...........................................................................................39 3-5 T2 relaxation times before and after implantation of electrodes and 3, 7, 10, 20, 40 and 60 days after stimula tion in epileptic animal ..............................................................40 3-6 Ventricular volume and T2 relaxation times in lateral ventricle regions for injured animal .................................................................................................................................41 3-7 Coronal 17.6 T T2-weighted images of injured excised brain...........................................42 3-8 3D images of implanted control ex cised brain and epile ptic excised brain .......................43 3-9 Average ex vivo T2 values for epileptic animals ...............................................................44 3-10 Cresyl-Violet brain sect ions of epileptic animal ................................................................45 3-11 Timmstained sections of epileptic animal .......................................................................46 3-12 Fluoro-Jade C secti on of epileptic animal .........................................................................47 3-13 Perls staining in the dorsal thalamic nuclei ......................................................................48 3-14 Perls reaction counter-stained with Cresyl-Violet in the dentate gyrus ...........................49 3-15 Perls reaction counter-stained with Cresyl-Violet in the piriform cortex ........................50 3-16 Sections stained for myelin with Black Gold ....................................................................51 3-17 Perl sections counter-stained with Cresyl-Violet of animal sacr ificed day 9 after SE ......52 3-18 Fluoro-Jade C sections of an imal sacrificed day 9 after SE ..............................................53

PAGE 9

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science IN VIVO AND EX VIVO MRI STUDIES OF AN ANIMAL MODEL OF MESIAL TEMPORAL LOBE EPILEPSY CORRELATE D WITH HISTOLOGICAL ANALYSIS By Lan Hoang Minh May 2008 Chair: Thomas Mareci Major: Biomedical Engineering Epilepsy is one of the most common neurological conditions worldwide and a significant cause of morbidity in modern society, especially in children and young adults. Mesial temporal lobe epilepsy (MTLE) is the most common form of symptomatic loca lization-related human epilepsy and the most common epilepsy syndrome with pharmacologically intractable partialonset seizures Patients often have an initial precipitating injury during early childhood, followed by a latent period of up to several years before the emergence of complex partial seizures. Understanding the mechanisms involved during epile ptogenesis as well as the consequences of initial injuries is crucial in the diagnosis of ep ilepsy and development of effective strategies for therapeutic intervention. Whether seizures cause permanent damage to the brain and lead to an epileptic lesion resulting in long-term epilepsy is a subject of controversy. Se veral cross-sectional magnetic resonance imaging (MRI) and histology studies ha ve reported changes suggestive of seizureassociated damage in animal models. Very fe w longitudinal studies have been performed to describe acute seizure-associated changes a nd their evolution over time. Since epileptic conditions in humans develop over long periods of time and ha ve effects extending over many years, clinical studie s seeking the causes and mechanis ms underlying the development of 9

PAGE 10

epilepsy are very difficult and expensive to perf orm. Animal models allow for experiments that investigate the full dynamic course of the epileptic state. In vivo MRI permits the non-invasive detection of structural alterations in living animals, the dynamics of involvement and interactions between all the brain regions, as well as their progression in a single i ndividual over the entire time-course of the disease. The animal model of chronic limbic epileps y used in this study is characterized by recurrent spontaneous hippocampal seizures follo wing an episode of status epilepticus (SE) induced by a period of electrical hippocampal st imulation. Spontaneous seizures begin several weeks later and physiologically resemble seizures seen in human MTLE. High-resolution in vivo MRI was used to evaluate the progression of brain pathological alterations in the same animal at different time points, followed by ex vivo MRI, three-dimensional magnetic resonance microscopy at high field strength and histological analysis. The findings demonstrated that development of seizures was associated with significant changes in relaxation measurements following SE in the hippocampus, lateral ventri cle regions, thalamus and parahippocampal gyrus, visible or not as changes in signa l intensity on T1 or T2-weighted in vivo MRI images. These in vivo changes were due to underlying pathol ogical tissue altera tions evident with ex vivo MRI and histology, including iron deposit ion, neuronal degeneration and microglia aggregation. Correlation of longitudinal in vivo imaging data with behavioral observations and postmortem ex vivo MRI and histopathological analysis can help provide a more complete understanding of the causes and mechanis ms underlying epileptogenesis. Continued improvement and application of these MRI t echniques, particularly the development of diffusion-weighted and diffusion tensor imaging, will allow additional insights into the mechanisms involved in th e epileptogenic process. 10

PAGE 11

CHAPTER 1 INTRODUCTION Mesial Temporal Lobe Epilepsy Overview Epilepsy is a chronic pathologic condition of th e nervous system characterized by recurrent epileptic seizures. It is one of the most comm on neurological conditions affecting at least 50 million people worldwide (Loscher and Ebert 1996) with an incidence rate of about 50 per 100,000 in developed countries (Sander 2003). It is a significant cause of morbidity in modern society, affecting mostly children and young adults (Sundqvist 2002). Th e various types of epilepsy are characterized by simultaneous activa tion of a large population of neurons resulting in abnormal electrographic and/or behavioral seizure activity (Pitkanen, et al. 2002). Mesial temporal lobe epilepsy (MTLE) is the most common form of symptomatic (secondary) localization-related human epilepsy and the most common epilepsy syndrome with pharmacologically intractable par tial-onset seizures (Cavazos a nd Cross 2006). It affects 20% of patients with epilepsy, with a cost of approxima tely $5 billion in the United States. MTLE is characterized by seizures arising from the hippocampus, parahippocampal gyrus and amygdala. Patients often have an initial precipitating injury during early childhood, such as status epilepticus, complicated febrile seizures, head tr auma, etc., followed by a latent period of up to several years during which neurobiol ogical changes take place that eventually lead to the onset of spontaneous seizures (Berg, et al. 1999, Knudsen and A uk 2000, Mathern, et al. 1997). Status epilepticus (SE) has been defined as a seizure that shows no clinical signs of arresting after a duration encompassing the great ma jority of seizures of that type (usually 30 minutes), or recurrent seizures without interict al resumption of baseline central nervous system function (Blume, et al. 2001). SE can lead to th e onset of chronic spontaneous seizures after a 11

PAGE 12

latent period of epileptogenesis that can last several years. Understanding the mechanisms involved during epileptogenesis as well as the cons equences of initial injuri es is crucial in the diagnosis of epilepsy and the development of e ffective strategies for therapeutic intervention. Pathophysiology The syndrome of MTLE is the most common form of symptomatic localization-related epilepsy, due to a localized structural brain lesion, and is characterized by epileptogenic abnormalities in mesial temporal limbic structures (Engel 2001). Epileptic seizures arise from synchronous and sustained discharges from gr oups of neurons. Seizure discharges in MTLE patients initiate from mesial limbic structures su ch as the hippocampus, entorhinal cortex and amygdala (Bartolomei, et al. 2001, Blumenfel d, et al. 2004, Maillard, et al. 2004). This hyperexcitability may arise due to a number of different reasons that remain only partially understood (Liu, et al. 1995). These mesial limbic areas display the most prominent morphological changes. The most prominent pathological manifest ation associated with MTLE is usually hippocampal sclerosis (Figure 1-1) (Blumcke et al. 1999, Liu, Mi kati and Holmes 1995, Mathern, et al. 1996), as part of a typical pattern of brain damage in MTLE patients known as mesial temporal sclerosis (MTS) (Gloor 1997, Ja ckson, et al. 1998). However, other discrete structural lesions can also be found alone or in association with hippocampal sclerosis. Whether hippocampal sclerosis is a result of spontaneous se izures or causes seizur e activity in the region is under debate. The hallmark of MTS is a selective loss of neurons in the entorhinal cortex and the neighboring amygdala, in the dentate hilus, and in the CA3 and CA1 areas of the hippocampus. However, the granule cells of the dentate gyru s and the CA2 area of th e hippocampus as well as the subiculum are relatively spared (Du, et al. 1993, Gloor 1997). The dense gliosis that 12

PAGE 13

accompanies the loss of neurons causes shrinkage and harden ing of tissue (sclerosis). Many studies have focused on the cell damage and co nsequent reorganization that characterize the dentate gyrus of MTLE patients. Normally, input to the hippocam pus comes from the entorhinal cortex to the dentate granule cells through the perfor ant path (Chang and Lowenstein 2003). The dentate granule cells normally sprout mossy-fiber a xons, which extend to CA3 pyramidal neurons as part of the hippocampal output path way (Engel 2001). In hippocampal sclerosis, however, these cells sprout mossy -fiber axons that are directed back into the inner molecular layer (Babb, et al. 1991, Sutula et al. 1989), possibly because the neurons to which they usually extend have been lost. These aberrant mossy fibers instigate a recurrent excitatory circuit by forming synapses on the dendrites of neighboring dentate granule cells. Excitation interneurons, which normally activate inhibitory interneur ons acting on dentate granule cells, may be selectively lost, resulting in hyperexcitability (Sloviter 1991). Moreover, neurogenesis of new dentate granule cells was shown to be increase d by seizures, and these neurons may integrate themselves into abnormal circuits (Parent, et al. 1997, Scharfman, et al. 2000). In addition to morphological changes in the hippocampus, the most prominent changes at the molecular level were alterations in the composition and expressi on of inhibitory GABA receptors on the surface of hippocampal dentate gra nule cells, weeks before the onset of spontaneous seizures (Brooks-K ayal, et al. 1998). These alte red receptors become highly sensitive to zinc released by mossy-fiber termin als after repetitive synaptic stimulation (during seizure initiation), leading to fa ilure of inhibition (Buhl, et al 1996). Some studies also suggest that neuronal injury following seizures may be mediated by an excitotoxic mechanism, with excessive excitability secondary to the release of excitatory amino-acids, primarily glutamate (Liu, et al. 1995). 13

PAGE 14

In addition to the hippocampus, other mesiotempo ral regions such as the entorhinal cortex, the amygdala and the piriform cortex (PC) ha ve been associated w ith seizure generation (Bartolomei, et al. 2005, Bernasconi, et al. 1999, Bertram 1997, von Bohlen und, et al. 2004, Yilmazer-Hanke, et al. 2000). Studies have demons trated the propagation of seizures from the entorhinal cortex to the CA1subicu lum region of the hippocampus through the temperoammonic pathway (Cossart, et al. 2001, So ltesz 1995, Wozny, et al 2005). The PC sends projections to different limbic structures such as the entorhinal cortex as well as the periamygdaloid cortex and cortical amygdaloid nucl ei (Goncalves, et al. 2005). It was shown to have a critical role in th e generation of epileptogenic and ictogenic networks in experimental TLE (Bertram 1997, Loscher and Ebert 1996, McIntyre and Kelly 2000), and is associated with the development of spontaneous focal-onset seizures after SE (Roch, et al. 2002b). The amygdala is heavily inte rconnected with the hippocampus and parahippocampal region and is prone to neuronal damage both in clinic al and experimental TLE (Aliashkevich, et al. 2003, Hudson LP, et al. 1993, Kemppainen, et al. 2002, Tuunanen, et al. 1999). One of the major projections from the amygdala to the hippocampu s terminates in the temporal aspect of the CA1/subiculum border region in rat (Pikkarainen, et al. 1999), which is affected in most patients with MTLE. The seizure activity that is initia ted by the stimulation of the amygdala has been reported to cause secondary damage in the hippocampus. For example, loss of hilar neurons and sprouting of mossy fibers in the dentate gyrus occurred in rats that developed spontaneous seizures after a self-sustained status epilepticus was induced by electrical stimulation of the lateral nucleus of the amygdala (Pitkanen, et al. 2002). Stimulation studies demonstrate that action potentials are con ducted from the amygdala to the ento rhinal cortex in approximately 25 ms and from the amygdala to the temporal hippocampus in 20 ms (Buser and Bancaud 1983, 14

PAGE 15

Wilson, et al. 1990), suggesting that several regions of the temporal lobe ar e readily recruited to participate in the seizure activ ity after its onset (Kemppainen and Pitkanen 2004). These data suggest a critical role for the entorhinal cort ex, amygdala and PC in epileptogenesis as well as in the symptomatology of TLE, and suggest the necessi ty of looking at the dynamics of the entire limbic circuit when investigating the mechanisms of epileptogenesis. Animal Model of Mesial Temporal Lobe Epilepsy Animal models that reproduce the behavior al and neurophysiologic features of human epilepsy make them well suited for studyi ng the causes and mechanisms underlying the development of epilepsy. They are particularly useful in studying the progressive changes over the entire time-course of the disease within a relatively short time frame, while those developments usually take several years in human patients. In animal models, the epileptogenic process has been initiate d by chemically or electrically induced status epilepticus. However, in studies using systemically injected chemical convul sants such as pilocarpine or kainate (King, et al. 1991a, Niessen, et al. 2005), the mortality rates are high, and those models do not replicate the focal brain injury that usually initiate s the epileptogenic pro cess in human MTLE. The rat chronic limbic epilepsy (CLE) model is a well-studied and accepted animal model of acquired TLE. CLE induced by hippocampal electri cal stimulation is considered to be a model of epileptogenesis and spontaneous seizures provoked by an initia l brain insult, and this model mimics many of the histopathol ogic characteristics of MTLE. Unilateral continuous electrical stimulation of the ventral hippocampus induces an episode of SE, followed by the onset of chronic epilepsy several weeks later (Bertram 1997, Bertram and Cornett 1994, Lothman, et al. 1989). In this model, neuronal loss and gliosis are seen throughout the limbic system and the thalamus, along with mossy fiber sprouting in the dentate hilus (Bertram and Lothman 1993, Bertram, et al. 1990, Bertram and Scott 2000), as seen in human MTLE patients. This model is 15

PAGE 16

also refractory to anticonvulsant drugs, as is commonly seen in human MTLE. Moreover, the electrographic seizures resemble the seizures recorded in humans (Bertram and Cornett 1993, Bertram and Cornett 1994, Lothman, et al. 1990). Th e latent period lasting several weeks before the onset of spontaneous seizures makes it an ideal model to study epileptogenesis. No studies to date have used MRI to invest igate the epileptogenic and spontaneous seizure changes in this animal model of MTLE indu ced by hippocampal electrical stimulation. The similarities between the physiopathology in this model and the clinic al pathology of human MTLE support the use of this animal model in the study of the mechanisms of epileptogenesis and epilepsy. Another advantage of this m odel is that it does not produce the mortality, morbidity and challenges that result from the use of systemic chemical convulsants. In vivo Magnetic Resonance Imaging Magnetic resonance imaging (MRI) ha s been the definitive modality for diagnosing most disorders of the central nervous system (CNS) ov er several years. MRI provides a key advantage in that it allows the noninvasive observation of many of the pathophysiological manifestations of CNS injury and disease, as we ll as their dynamic progression. Several cross-sectional magnetic resonance imaging (MRI) and histology studies ha ve reported changes suggestive of seizureassociated damage in animal models. Very fe w longitudinal studies have been performed to describe acute seizure-associated changes a nd their evolution over time. Since epileptic conditions in humans develop over long periods of time and ha ve effects extending over many years, clinical studie s seeking the causes and mechanis ms underlying the development of epilepsy are very difficult and expensive to perform. In vivo MRI permits the non-invasive detection of structural alterations in living animal models of epilepsy, the dynamics of involvement and interactions betwee n all the brain regions and their progression in a single individual over the entire time-c ourse of the disease. When 16

PAGE 17

living tissue is placed into a high static magnetic field, signal from hydrogen nuclei can be detected by disturbing those nucle i in different energy states with a radio frequency (RF) electromagnetic pulse. When the system return s to equilibrium, it produces a detectable NMR signal. The application of magnetic field gradients allows the localization of the signal source. The contrast in the image is determined mainly by the amount and mobility of water molecules in the tissue (Grohn and Pitkanen 2007). Relaxation times (T2, tran sverse; T1, longitudinal) are influenced by the dynamics of water and reflec t how hydrogen nuclei return toward equilibrium after excitation by a RF pulse. Modifications of the pulse sequence (excitation and acquisition parameters) allow the modulation of the contrast based on the relaxation and diffusion of water, magnetic coupling between free and bound water, etc. (Grohn anf Pitkanen 2007). In previous animal MRI studies of CLE, initial increase in T2 signal wi thin a few days after chemical induction of SE was measured in earl y studies and indicative of edema, cell swelling and increased extracellular water as a result of the injury (Bouilleret et al. 2000, King, et al. 1991b, Zhong, et al. 1993). This signal increase wa s followed by normalization after a few days and secondary increase during epile ptogenesis (Roch, et al. 2002a). However, other models show a T2 decrease a few hours after SE induction (Bha gat, et al. 2001, van Eijsden, et al. 2004). Diffusion-weighted imaging (DWI) utilizes th e measurement of Brownian motion of water molecules to generate contrast with varying diffusion rates. A decrease in apparent diffusion coefficient (ADC) has been seen in the hi ppocampus, piriform cortex and amygdala and indicative of early cell loss and ongoing neuronal death only minutes after onset of seizures during SE (Engelhorn, et al. 2007). This initial decrease normalizes after a few days (Wall, et al. 2000, Wang, et al. 1996, Zhong, et al 1995), at different times for the various brain regions. 17

PAGE 18

Recent volumetric studies in chemically indu ced SE reported a decrease in hippocampal volume and increased ventricle vo lume 10 days into epileptogenesis (Wolf, et al. 2002), up to two months after SE (Niessen, et al. 2005), consistent with neur onal loss, hippocampal sclerosis and atrophy. Other longterm studies show decreased temporal lobe volume and progressive decrease in the piriform cortex and amygdala thickness from 9 days up to 6 months after SE induction (Nairismagi, et al. 2004). Initial changes in T2 affecting the limbic st ructures usually normalize by day 10 after SE and reappear at day 20 (Dube et al. 2004, Nairismagi, Grohn Kettunen, Nissinen, Kauppinen and Pitkanen 2004, Roch, Leroy, Nehlig and Namer 2002a). Care must be exerted when interpreting these relaxation time changes, as T2 increase has been associated with the presence or absence of neurodegeneration (Dube, Y u, Nalcioglu and Baram 2004, Jupp, et al. 2006, Nairismagi, Grohn, Kettunen, Nissinen, Kauppinen a nd Pitkanen 2004). With the importance and many uses of in vivo MRI, continued research is needed to enhance this technique and expand its uses to provide a better unde rstanding of the temporal prof ile of the pathological changes accompanying epileptogenesis. Ex vivo Magnetic Resonance Imaging Three-dimensional magnetic resonance micros copy (MRM) can also be used to study animal models of epilepsy. This technique allows for ex vivo measurements where motion artifacts are eliminated (Lemaire, et al. 1990). This allows the signalto-noise ratio to be optimized. Moreover, prolonged scan times and higher field strengths are not limiting factors as with in vivo measurements (Guilfoyle, et al. 2003). This ex vivo MRI technique allows greater sensitivity in visualizing patholog ical changes in the brain. Previ ous studies have shown a strong correlation between T2-weighted signal change s in high-resolution MRI of paraformaldehydefixed rat brains and the histologic al damage in the kainate-induced model of SE (Tiina-Riikka, et 18

PAGE 19

al. 2001). These studies have also shown that MRI ex vivo is sensitive enough to detect layerspecific damage in the rat brain af ter SE, which is not possible with in vivo imaging. 19

PAGE 20

20 [D] [C] [F] [A] [B] [E] Figure 1-1. Hippocampal Sclerosis. Hippocampal sclerosis is the most common identified pathological feature in cases of mesial temporal-lobe epilepsy. Normally, input to the hippocampus comes from the entorhinal cort ex to the dentate granule cells through the perforant path [A]. Dentate granule cells project to the CA3 sector as the first step in the hippocampal output pathway [B]. A cl ose-up of the dentate granule-cell layer reveals several morphologic changes characte ristic of hippocampal sclerosis that may play a part in epileptogenesis. Newly spr outed mossy fibers from dentate granule cells can synapse on dendrites of neighbori ng dentate granule cel ls, resulting in a recurrent excitatory circuit [C]. They can al so sprout onto inhibitory interneurons [D]. Excitation interneurons, which normally ac tivate inhibitory in terneurons, may be selectively vulnerable to brain insults [E]. Finally, neurogenesis of new dentate granule cells continue s into adult life, and these neurons may integrate themselves into abnormal circuits [F] (f rom Chang and Lowenstein 2003).

PAGE 21

CHAPTER 2 MATERIALS AND METHODS Animal Preparation The present study was carried out in accordan ce with National Institutes of Health guidelines for the care and use of animals and a pproved by the Institute of Animal Care and Use Committee (IACUC) of the University of Florida. Sprague Dawley (Harlan, Indianapolis, IN) male 50-day-old rats, weighing 304 g 77 g, were used in the experiments. Three groups of animals were studied. One group of experimental animals received chronic microwire electrodes to stimulate the hippocampus and induce SE. An implanted control group consisted of animals that were subjected to the same anesthesia, surg ical time and electrode implantation as the first group of injured animals but were not stimulated into status epilepticus. A third group consisted of animals which were not implanted with electrodes and served as nave controls. Implantation of Hippocampal Electrodes Animals were pre-medicated with a 2 mg s ubcutaneous injection of Xylazine (Phoenix Pharmaceutical, St. Joseph, MO), anestheti zed with 1.0 2.0 % Isoflurane (MINRAD, Bethlehem, PA) and positioned in a Kopf (David Kopf Instruments, Tujunga, CA ) stereotactic frame. A midline incision was made in the s calp, and all soft tissue was loosened from the dorsum of the skull. A craniotomy was drilled for electrode placement such that the long axis extended from 1.7 mm lateral to 3.5 mm lateral fr om bregma and the dura was removed. A pair of 50 m gold-plated tungsten wires (Plastics One, RoanokeVA) were stereotactically implanted in the ventral hippocampus (-5.3 mm posterior, 4.9 mm lateral, right of Bregma, 5 mm ventral) in each rat brain. The electrodes were chronica lly secured with Methylmethacrylate (Plastics One, RoanokeVA) that was anchored to 4 plastic micro screws driven in to the skull (Sanchez, 2006). 21

PAGE 22

Induction of Status Epilepticus, Video Monito ring and Analysis of Spontaneous Seizures One to two weeks after surgery, the implanted wires were used to attempt the induction of self-sustained status epilepticus (SE, Bertram et al., 1993). A s uprathreshold stimulus of 334 159 A was delivered to the implanted electrodes fo r 88 26 minutes using 10 s pulse trains of 20 ms period, with 1 ms duration, biphasic square waves with a 2 s delay between pulse trains. The behavior of the animals was video monitore d starting 15 days after stimulation and over a period of eight weeks. Seizure detection was performed visually using behavioral (video) monitoring by a veterinarian and be havioral neuroscientist. A behavioral seizure score (BSS) out of 5 was determined for the SE animals in th is study using the standa rd Racine scale (0, no change; 1, wet dog shakes; 2, head bobbing; 3, forelimb clonus; 4, forelimb clonus and animal rearing; 5, rearing and falling). Seizure activity was assessed usi ng the standard Racine scale; only grades 3 through 5 seizures could be determined from video recordings alone. In vivo MR Imaging Animals were imaged before and after electrode implantation to serve as their own controls and 3, 5, 7, 10, 20, 40, 60 days after SE to monitor temporal changes (Figure 2-1). Animals were initially anesthetized with 4 % isoflurane in O2 at 2.0 L/min and injected subcutaneously with 0.8 mg of Xylazine and 2 mL of lactated Ringers solution (Hospira, Lake Forrest, IL) before imaging in the magnet to maintain the physical condition of the rats during the extended MRI scan sessions. Each animal was placed in a supine pos ition, in a custommade MRI compatible stereotaxic frame and cradle, to allow repeat able positioning and prevent motion artifacts. A custom-built linear surface coil was used to measure the MR images. A 130 degree arc, 3.5 cm rectangular linear coil was cons tructed on a 4.0 cm diameter half cylinder to fit animals with 22

PAGE 23

implanted electrodes. This coil features dist ributed capacitance, bala nced impedance matching and a cable trap Anesthesia for each rat in the magnet was main tained through a nose cone with a mixture of 1.0 2.0 % isoflurane with oxygen at a flow rate of 1 L/ min. Respiration was monitored during data acquisition and kept between 30 and 50 breaths/ min by modulating the concentration of isoflurane. Core body temperat ure was also monitored during experiments by using a rectal probe and kept between 36oC and 37oC; physiological temperature was maintained using heater air flow (SA Instruments, Stony Brook, NY). In vivo High Resolution MR Imaging Data Acquisition All in vivo MRI measurements were perf ormed on an 11.1 T, 40 cm horizontal bore magnet at 470 MHz (Magnex Scientific, Abi ngdon, UK) by using a Bruker Avance console (Bruker NMR Instruments, Inc., Billerica, MA) and 200 mT/m actively shie lded gradients. All data acquisition was performed with Bruker ParaVision software. To adjust the position of the image field of view, three-axis pilot images were collected with a T1-weighted coronal spin echo scan using a RARE factor of 4, TR = 1500 ms, TE = 7 ms, a field of view (FOV) of 30 x 30 x 15 mm, matrix size of 128 x 128 with 15, 1 mm thic k slices, spectral width of 90 kHz and 4 averages. To visualize the pathology in the brai n, a T2-weighted coronal multiple-slice spin echo image data set was acquired over 15 slices of 1 mm thickness and no gap between slices, with a TR = 3000 ms, TE = 12.5 ms, 8 averages, RARE enc oding factor of 8, spectral width of 40 kHz, FOV of 30 x 30 x 15 mm and a matrix of 256 x 256. To quantify T1 relaxation times, multiple-slice spin echo coronal T1-weighted images were acquired over 9 slices of 1 mm thickness with 1. 5 mm distance between slic e centers, with TR = 4000, 2000, 1000, 500, 250 ms, TE = 15 ms, 2 averages, sp ectral width of 20 kHz, a FOV of 30 x 30 x 13.5 mm and a matrix size of 100 x 100. To quantify T2 relaxation times, a series of T223

PAGE 24

weighted spin echo coronal scans were acquire d over 9 slices of 1 mm thickness with 1.5 mm distance between slice centers, with TR = 2000 ms, TE = 15, 30, 45, 60, 75 ms, 2 averages, spectral width of 70 kHz, a FOV of 30 x 30 x 13.5 mm and a matrix of 100 x 100. The total acquisition time was 25.83 min fo r T1-weighted images and 33.33 min for T2-weighted scans. Ex vivo MR Imaging High-field, high-resolution MR im aging was used to visualized the locations of pathology in the excised fixed animal brains. After finishing the in vivo MRI study (8 weeks after SE), the animals were deeply anesthetized, sacrificed and transcardially perfused according to the Timm fixation protocol: 0.37% sulf ide solution (30 ml/min) for 10 min followed by 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4 (30 ml/min ), +4C, for 10 min (Nairismagi et al, 2004). The electrodes were removed, and th e intact brains extracted from the skulls and post-fixed in 10% formalin. Prior to ex vivo MR imaging, the intact fixed brains were soaked in phosphate buffer solution (PBS) for about 24 hours to wash out residual fixative. The excised brains were then placed in an 18 mm tube containing fluorinated oil and imaged at 750 MHz in a 17.6 T, 89 mm bore Bruker Avance MR instrument (Bruker NMR Instruments, Billerica, MA). All data acquisition was perfor med with Bruker ParaVision software. Three-dimensional MR microscopy (MRM ) images were acquired with a 3D gradient echo pulse sequence with TR = 150 ms, TE = 15 ms The image field-of-view was 30 mm 18 mm 15 mm in a matrix of 400 240 200, and the data were acquired in a total data acquisition time of 2 h per signal average. Thro ugh experimentation, 2 av erages (4 h) provided the optimum compromise between signal-to-noi se ratio and measurement time. All the MRM images were acquired with a resolution of 75 75 75 m3. A 3D Fourier transformation was applied to the acquired data matr ix to produce the 3D image, whic h was then interpolated using a 24

PAGE 25

bilinear interpolation, by a fact or of 2 in each dimension to produce an image with a display resolution of 37.5 m. To quantify T1 relaxation times, multiple-slice spin echo coronal T1-weighted images were acquired over 15 slices of 0.5 mm thickness with 0.6 mm distance between slice centers, with TR = 5000, 2000, 1000, 500, 300 ms, TE = 10 ms, 2 averages, spectral width of 38 kHz, a FOV of 19 x 18 mm and a matrix size of 128 x 128. To qua ntify T2 relaxation times, a series of T2weighted spin echo coronal scans were acquired over 15 slices of 0.5 mm thickness with 0.6 mm distance between slice centers, with TR = 3000 ms, TE = 15, 30, 45, 60, 75 ms, 2 averages, spectral width of 44 kHz, a FOV of 36 x 18 mm and a matrix of 128 x 128. The total acquisition time was 37 min for T1-weighted images and 1 h fo r T2-weighted scans. Brains were placed in fixative right after ex vivo imaging. Histological Procedures Fixation Following ex vivo MR imaging, the excised brains were cryoprotected in 30% sucrose in PBS and sodium azide before they were processed for histology. The brai ns were then blocked, frozen in dry ice and sectioned in the co ronal plane into 50 m-thick sections on a freezing microtome. Sections were collected in PBS and mounted on gelatin-coated slides. Adjacent series of sections were used for Nissl, Perl, Fluoro-Ja de C, Timm and Black Gold staining. Nissl Staining Cellular morphology of the different brain ar eas was assessed using Nissl staining. The slide-mounted brain sections were soaked in Cresyl Violet solution for 10 min, dehydrated through a graded series of ethanolwater soluti ons, cover-slipped, and analyzed under a bright field microscope. 25

PAGE 26

Perls Staining For pathological iron visualizat ion, the brain sections were incubated in Perls solution (1:1, 5% potassium ferrocyanide and 5% HCl) for 45 minutes, washed in distilled water, and incubated again in 0.5% diamine benzidine tetrahydrochloride for 60 mi nutes (Hill, 1984). Half of the slides were then counter -stained with Cresyl Violet. Fluoro-Jade C An adjacent series of sections was stained with Fluoro-Jade C for detection of degenerating neurons. Slides were first immersed in a basi c alcohol solution consisting of 1% sodium hydroxide in 80% ethanol for 5 min. They were then rinsed for 2 min in 70% ethanol, for 2 min in distilled water, and then incubated in 0.06% potassium permanganate solution for 10 min. Following a 1-2 min water rinse, the slides we re then transferred for 10 min to a 0.0001% solution of Fluoro-Jade C dissolved in 0.1% ac etic acid vehicle. The proper dilution was accomplished by first making a 0.01% stock solution of the dye in distilled water and then adding 1ml of the stock solution to 99ml of 0.1% acetic acid vehicle. The working solution was used within 2 h of preparation. The slides were then rinsed through thr ee changes of distilled water for 1 min per change. Excess water was drai ned onto a paper towel, and the slides were then air dried for at least 5 min. The air dried slides were then cleared in Xylene for at least 1 min and then cover-slipped with non-fluorescent mounting media. Timm Staining Synaptic reorganization through mossy fiber sprouting was anal yzed from sections stained with the Timm sulfide silver method (Danscher, 1981). Half of the slides were counter-stained with Cresyl Violet to stain for cytoplasmic RNA-rich ribosomes, as well as the nuclei and nucleoli of the nerve cells. 26

PAGE 27

Black Gold Staining In order to localize the presence of normal or pathological myelin, the slides were stained with haloaurophosphate complex, called Black Go ld. A 0.2% solution of Black Gold was made by adding 100mg of Black Gold to 50 ml of 0.9% NaCl and then heating it to 60 degrees C. The slide mounted tissue sections were transferred to this warm Bl ack Gold impregnating solution in the oven for 12-18 minutes. At this point, the stai n was intensified by incu bating the sections for 10-15 minutes in a 60 degrees C solution of 0.2% potassium tetrachloroaureate (Alrich Chem., Milwaukee, WI) dissolved in 0.9% saline. The sections were then rinsed for 2 minutes in distilled water, fixed for 3 minutes in a sodium thiosulfate soluti on, and then rinsed in tap water for at least 15 minutes (three 5 minute change s). Slides were dehydrated through gradated alcohols. The dehydrated sections were cleared in Xylene (Fisher Scientific, Pittsburgh, PA) for at least 2 minutes and then cover-slipped with plastic mounting media (Schmued, 1999). Data Analysis I n vivo and ex vivo general MR image processing a nd analysis was performed by using custom software written in the Interactive Data Language software (IDL, from Research Systems, Boulder, CO). Using a ra t brain anatomical reference atlas (Paxinos, 1997), regions of interest (ROIs) were chosen around the following structures in both the right and left brain hemispheres: hippocampus and two hippocampal s ubregions (CA1 and dentate gyrus regions), fimbriae/ lateral ventricles, entorhinal corte x, piriform cortex, amygdala, dorsal thalamus and retrosplenial cortex (Figure 2-2) All slices containing these ROIs were chosen for quantitative T1 and T2 relaxation measurements using in ho use MRI Analysis Software (MAS) written in IDL. For SE animals, in vivo measurements were averaged w ithin three time periods: before stimulation into SE, between 3 and 10 days after SE (latent period) and 20 to 60 days after SE 27

PAGE 28

(beginning of spontaneous seizures). In vivo T1 and T2 measurements for each ROI were assessed for statistically significant differences between time periods using paired Student's t tests (R, Boston, MA). A p value < 0.05 was considered to be significant. For ex vivo measurements, statistically significant differences between the SE and control groups were assessed using unpaired Student's t test between the two groups. 28

PAGE 29

Pre-implantation in vivo MRI (11.1T) Histology Electrode implantation Electrical Ex vivo MRI (17.6 T) Video recordings Post-implantation in vivo MRI Day 3 post-SE in vivo MRI Day 5 post-SE in vivo MRI Day 7 post-SE in vivo MRI Day 10 post-SE in vivo MRI Day 20 post-SE in vivo MRI Day 40 post-SE in vivo MRI Day 60 post-SE in vivo MRI stimulation into SE Figure 2-1. Stimulation and MR imaging timeline. 29

PAGE 30

30 1 2 4 5 6 3 7 8 9 10 11 13 14 12 1 5 16 1 7 18 Figure 2-2. Coronal T2-weighted image of implan ted control excised brain showing anatomical regions of interest: 1, 2retrosplenial co rtex; 3, 4 -hippocampus; 5, 6 CA1; 7, 8 dentate gyrus; 9, 10 dorsal thalamus; 11, 12 lateral ventricles/ fimbriae; 13, 14entorhinal cortex; 15, 16piri form cortex; 17, 18amygdala.

PAGE 31

CHAPTER 3 RESULTS Animal Behavior Control animals that were not implanted with electrodes or implanted but not stimulated did not exhibit any behavioral alteration. Thus electrode implantation alone did not induce epileptogenesis. Of the animals that were stim ulated in the study, one did not undergo SE and was found to have electrodes implanted in the medial geniculate nuclei instead of the hippocampus, as verified with MRI after electrod e implantation. One of the animals that reached SE died 12 days after stimulation from unknown cau ses. Another animal that attained SE was sacrificed at 9 days because of health complications and was kept for histological analysis. The animals that attained SE immediately following the onset of stimulation demonstrated wet dog shakes and seized several times during the procedure. After the stimulus was ended, hyperactivity and seizing was observed for approximately 8 hours afterwards. One animal started seizing for the first time 12 days after stimulation. Video analysis of seizures indicated that 4 rats had spontaneous seizures at 20 days after SE, and one had seizures beginning at 29 days after SE. At 20 days after SE in spontaneously seizing rats, the number of daily spontaneous seizures varied between 1 and 3 (mean, 2 1; n = 3) and at 29 days between 0.7 and 1.4 (mean, 1 0.2; n = 5). The spontaneously seizing animals had an average of less than 1 seizure per day and continued seizing until the end of the video recording period (60 days after stimulation). However, one animal seized with hi gh frequency (greater than 2 seizures per day) for the first 2 weeks, but no seizures were obser ved 31 days after SE. The mean score indicating the behavioral severity of spontaneous seizures was 3.2 0.5 at 20 days and 3.3 0.7 at 60 days. Only seizures grades 3 to 5 could be obs erved by the veterinarian and were reported. 31

PAGE 32

In vivo High Resolution MR Imaging All relaxation time data from pre-implantation MRI scans were compared to determine biological variability between animals, and rela xation times in the same region of the frontal cortex, away from the site of electrode implantation, were measured over time to confirm repeatability of the measurements for each animal. The relaxation time variability between animals pre-implantation was low; average st andard deviation was 4.5 2.1 % between all animals for all regions of interest. No signifi cant changes were observed in nave control and implanted control animals (Figure 3-1) and nonseizing injured animal s over the 60 days of imaging (standard deviation 2.9 0.8 %). Cont rol animals did not show any significant difference in their T1 and T2 values between ri ght and left hemispheres (p < 0.05, using paired ttest). T1 measurements followed the same pattern of variation as the T2 measurements for the various regions of interest, excep t that T1 changes were less significant than T2 changes (Figure 3-2). The average T1 relaxation times measured from control animals were 3.1 0.05 s (piriform cortex), 3.0 0.01 s (amygdala), 2.7 0.9 s (thalamus), and 2.9 0.01 s (hippocampus). Figure 3-3 shows T2-weighted MRI scans acquired over the course of in vivo imaging for an epileptic animal. Quantit ative MRI relaxation time measurements were performed in vivo in control and stimulated animals at 3, 7, 10, 20, 40 and 60 days after stimulation. These T2 measurements were averaged over the pre-stimulati on period, the latent peri od (3 to 10 days after SE) and the spontaneous seizure pe riod (20 to 60 days after SE) fo r ipsilateral and contralateral regions of interest (Figure 3-4). The average T2 relaxation times measured from control animals were 38.7 2.1 ms (piriform cortex), 38.2 1.7 ms (amygdala), 34.2 1.4 ms (thalamus), and 37.6 1.2 ms (hippocampus). The rat that survived SE but did not develop epilepsy presented no modification of the signal on T2-weight ed images over the entire course of in vivo imaging, and the T2 relaxation times remained in the range of values for control animals at all time points. In 32

PAGE 33

epileptic animals, the retrosplenial cortex s howed a significant T2 decrease on the ipisilateral side to stimulation 3 to 10 days after stimulati on (p < 0.05) compared with control values (Figure 3-4). The hippocampus and the CA1 subregion also showed significant T2 d ecrease bilaterally (p < 0.05) during the latent period. On the other hand, the lateral ventricle/ fimbriae areas, piriformic cortices and the amygdala showed a statis tically significant T2 in crease bilaterally 3 to 10 days after stimulation (p < 0.05), which peaked at 3 days after SE and returned to baseline before 10 days (Figure 3-5). This T2 increase in the lateral ventricle/ fimbriae areas correlated with a bilateral increase in ventricular volume observed and quantified over time on T2-weighted images using the Slicer software (Figure 3-6). The ventricle/fimbriae areas showed further significant T2 increas e 20 to 60 days after stimulation bilaterally, while the amygdala on the contralateral side to stimulation and piriformic cortex on both sides showed a significant decrease in T2 values after 20 days compared to 3 to 10 days after SE (p < 0.05). Th e retrosplenial cortex, hippocampus (including dentate gyrus and CA1 subregions), entorhinal cortex and dorsa l thalamus showed a significant T2 decrease bilaterally 20 to 60 days after st imulation compared to control va lues (p < 0.05) (Figure 3-4). The epileptic animals showed interindi vidual differences in T2 variati ons at each time point and in their seizing patterns, possibly due to the slight differences in location of the stimulating electrodes implanted in the ventral hippocampus. Ex vivo MR Microscopy Ex vivo MR images had greater resolution and more defined areas of pathology (i.e. neuronal loss, cavitation, iron deposition) in the epileptic brains compared with in vivo images (Figures 3-7,8). Areas of hypointensities were visible in the CA1 subregions of the hippocampus and dorsal thalamus bilaterally, as well as the piriform cortex. Atrophy of the hippocampus and enlargement of the ventricles bilaterally were also evident. Quantification of T2 measurements 33

PAGE 34

for ex vivo data (Figure 3-9) supported the qualitative observation of the MR images. T2 values were decreased in the hippocampus bilaterally, especially in the CA1 subfield, in animals that underwent seizures when compared with control animals (p < 0.05, t test). Average T2 was also significantly decreased in the piri form cortex and dorsal thalamus bilaterally. On the other hand, T2 was significantly increased in the amygdala on the contralateral side to stimulation. T2 values were dramatically increased in the lateral vent ricle/ fimbriae areas bila terally (p < 0.05). No significant differences in T2 measurements were observed between seizin g and control animals in the cortex. Histologic Findings Data obtained from histological staining help ed elucidate pathologi cal changes observed and quantified with in vivo and ex vivo MR imaging. Electrode tracts were visible on histological sections of implanted animal s. Nissl staining showed decr eased neuronal density in the hippocampus, particularly in the dentate hilu s, and the CA1 and CA3 pyramidal cell layers (Figure 3-10) in epileptic animals sacrificed after 60 days post-SE. D ecreased neuronal density was also observed in the amygdala, piriform cortex and the dorsal thalamic nuclei. Timm staining showed increased density of moss y fiber sprouting bilaterally in the inner molecular layer of the stratum granulosum in epileptic animals (Figure 3-11). Abnormal presence of astroglial cells and degenerating neur ons was observed in the CA1 subregion of the hippocampus after Fluoro-Jade C stai ning in one of the epileptic an imals (Figure 3-12) sacrificed after 60 days post-SE. Strong iron deposition was observed in the dorsal thalamic nuclei bilaterally (Figure 3-13) as well as at the site of electrode im plantation. Iron deposition was also observed in the inner molecular layer of the stra tum granulosum of the dentate gyrus (Figure 314). Perl sections counter-stained with Cresyl Violet revealed cavities in the piriform cortex bilaterally with iron deposition and aggregation of microglia around the cavities (Figure 3-15). 34

PAGE 35

Decreased myelin staining in the amygdala, the piriform cortex and the CA1 subfield of the hippocampus were observed after Black Gold staining (Figure 3-16). Enlargement of the ventricles was also visible bila terally, more pronounced on the contralateral side to stimulation, as well as extensive lo ss of tissue and decreased thickness in the amygdala and piriform cortex regions. In one rat sacrificed at day 9 after SE, Perl staining counter-stained with Cresyl Violet showed diffuse iron deposition in the dentate gyrus, as well as cavitation in the piriform cortex with microglia aggregation (Figure 3-17). Fluor o-Jade C staining reve aled ongoing degenerating neurons throughout the CA1, CA2 a nd CA3 pyramidal cell layers of the hippocampus, the dorsal thalamic nuclei and the dentate hi li bilaterally (Figure 3-18). 35

PAGE 36

5 days post-stimulation 3 days post-stimulation 7 days post-stimulation Pre-implantation Post-implantation A 20 days p os t st im u l at i o n 40 days post-stimulation 10 days post-stimulation 60 days post-stimulation Implanted control T2 vs. time0.00E+00 1.00E-02 2.00E-02 3.00E-02 4.00E-02 5.00E-02 6.00E-02 7.00E-02 8.00E-02 -10010203040506070Time (days)T2 (sec) ctx_contra ctx_ipsi hippo_contra hippo_ipsi ca1_ipsi ca1_contra dg_contra dg_ipsi lv_contra lv_ipsi amy_contra amy_ipsi pir_contra pir_ipsi ent_contra ent_ipsi thal_contra thal_ipsi B electrical stimulation Figure 3-1. T2 MR imaging in implanted control. (A) Coronal in vivo T2-weighted (TR/TE = 3000/12.5 ms; NA = 8) MRI scans before and after implantation of electrodes and at 3, 7, 10, 20, 40 and 60 days in implanted control animal (MR Rat 9), (B) T2 relaxation measurements versus time after stimulation for implanted control animals. No significant differences were observed in T2 values over time after implantation compared with control values for regions of interest. T0 = electri cal stimulation. Error bars represent one standard deviation. Regions of interest: retrosplenial cortex (ctx), hippocampus (hippo), ca1, dentate gyrus (dg), lateral ventricl es (lv), amygdale (amy), piriform cortex (pir), entorhinal cortex (ent), thalamus (thal). 36

PAGE 37

Pre-implantation Post-implantation 3 days post-stimulation 5 days post 7 days post 10 days post 20 days post 40 days post 60 days post Figure 3-2. Coronal in vivo T1-weighted (TR/TE = 1500/7 ms; NA = 4) MRI scans before and after implantation of electrodes and at 3, 7, 10, 20, 40 and 60 days after stimulation in epileptic animal (MR Rat 4). 37

PAGE 38

Figure 3-3. Coronal in vivo T2-weighted (TR/TE = 3000/12.5 ms; NA = 8) MRI scans before and after implantation of electrodes a nd at 3, 7, 10, 20, 40 and 60 days after stimulation in epileptic animal (MR Rat 4). 38

PAGE 39

0.07 A B Figure 3-4. Average T2 relaxation times in epileptic animals before stimulation 3 to 10 days (latent period) and 20 to 60 days after stimulation (s pontaneous seizures), (A) ipsilateral and (B) contralateral to stim ulation. Error bars re present one standard deviation. Regions of interest: retrosplenial cortex (Ctx), hippocampus (Hipp), CA1, dentate gyrus (Dent), lateral ventricles/f imbriae regions (LV/fim), amygdala (Amy), piriform cortex (Pir ctx), entorhinal cortex (Ento ctx), thalamus (Thal). 0.00 0.01 0.02 0.03 0.04 0.05T2 (seconds 0.06 ) Ctx Hipp CA1 Dent Lv/ fim Amy Pir ctx Ento. Thal ctx before stimulation 3-20 days after stimulation 20-60 days after stimulation* * * * * * * 39

PAGE 40

electrical stimulation Ctx_contra Ctx_ipsi Hippo_contra Hippo_ipsi LV/fim_contra LV/fim_ipsi CA1_contra CA1_ipsi dg_contra dg_ipsiTime (days)70 60 40 50 20 30 10 0 0.00E+00 2.00E-02 4.00E-02 6.00E-02 8.00E-02 1.00E-01 1.20E-01 -10 T2 (sec) A electrical stimulation Thal_ipsi Thal_contra Pir_ipsi Pir_contra Ento_ipsi Ento_contra Amyg_ipsi Amyg_contraTime (days)70 60 50 40 30 20 10 0 1.20E-01 1.00E-01 6.00E-02 8.00E-02 T2 (sec) 4.00E-02 2.00E-02 0.00E+00 -10 B Figure 3-5. T2 relaxation times before and af ter implantation of electrodes and 3, 7, 10, 20, 40 and 60 days after stimulation in epileptic animal (MR Rat 6),in ipsilateral and contralateral ROIs: (A) cortex (Ctx), hippo campus (Hippo), lateral ventricle/ fimbriae (LV/fim), CA1 and dentate gyrus (DG), (B ) amygdala (Amyg), entorhinal cortex (Ento), piriform cortex (Pir) and thalamus (Thal). T0 = electrical stimulation. Error bars represent one standard deviation. 40

PAGE 41

Figure 3-6. Ventricular volume and T2 relaxation times in lateral ventricle regions across time for one epileptic animal (MR Rat 4). LV/F i = lateral ventricle/ fimbriae region. 41

PAGE 42

* Figure 3-7. Coronal (17.6 T) T2-weighted images of injured excised brain (MR Rat 1), showing hypointense regions in CA1of the hippocam pus (black arrows) and the dorsal thalamic nuclei bilaterally (red arrows), hypointense signa l in the piriform cortex, increased ventricle areas bilaterally (*) a nd decreased thickness of piriform cortex and amygdala. 42

PAGE 43

A B Figure 3-8. (A) 17.6 T sagittal (a), transverse (b) and coronal (c) 3D images of implanted control excised brain (R101). (B) 17.6 T sagittal (a), transverse (b) an d coronal (c) 3D images of epileptic excised brain (MR Rat 1). Note the ar eas of hypointen sity within the CA1 subregion of the hippocampus bilate rally (black arrows), as well as the dorsal thalamic nuclei (red arrows) in the epileptic brain. a b a b c C c c 43

PAGE 44

A B Figure 3-9. Average T2 values are presented for various regions for both hemispheres, (A) ipsilateral and (B) contralate ral to the side of stimulation. Injury-induced changes are apparent in the contralateral hemisphere and to a lesser degree in the ipsilateral side. The T2 values were found to be significan tly different between the control and the injured brains in the CA1 subregion of the hippocampus, as well as the thalamus, amygdala and piriform cortex. (*) p < 0.05. Error bars represent one standard deviation. 0 10 20 30 40 50 60 T2 (m70 80 90 100 Control Injured Ctx Hipp CA1 Dent Thal Entor. ctx Pir. ctx Amyg. Lat. Vent. s ) 0 10 20 30 40 50 60 T2 (ms) 70 80 90 100 Ctx Hipp CA1 Dent Thal Entor. ctx Pir. ctx Amyg. Lat. Vent. Control Injured 44

PAGE 45

(2X) (4X) (10X) *** Figure 3-10. Cresyl-Violet stai ning shows decreased neuronal de nsity in CA3 pyramidal cell layer (arrowhead), CA1 (arrows) and dentat e hilus (*) of hippo campus in epileptic animal (MR Rat 5, left) compared to control (MR Rat 9, right). 45

PAGE 46

(2X) (20X) (20X) A Control (10X) (20X) B Figure 3-11. (A)Timm staining of epileptic brains 60 days after SE shows mossy fiber terminals in the inner (supragranular) molecular layer of the dentate gyrus bilaterally (arrows). Brown staining shows zinc precipitation al ong the mossy fibers in the hippocampal formation. This represents aberrant sprou ting of the mossy fibers to a region not normally innervated by mossy fibers. Hi gh magnification of the granule cell and molecular layers of the dentate gyrus in a control rat brain show little reaction deposits in the molecular layers (from www.fdneurotech.com). (B) Timm sections counter-stained with Cresyl-Vio let. White arrows point to mossy fibers stained black. 46

PAGE 47

Figure 3-12. Fluoro-Jade C section of one epilep tic brain (MR Rat 1) showing the presence of degenerating neurons in the CA1 pyramidal cell layer of the hippocampus (white arrow), as well as the abnormal presence of astroglial cells (red arrow), suggesting possible gliosis in that area. 47

PAGE 48

A B Figure 3-13. (A) Perls staining re veals iron deposition in the dorsal thalamic nuclei bilaterally in epileptic animal (MR Rat 5). (B) Perl sect ion counter-stained w ith CresylViolet shows iron deposition in the dorsal thal amic nuclei. The corresponding region is enclosed in the rectangle on a T2-weighted ex vivo MR image of the same epileptic animal and appears as an area with hypointense signal. 48

PAGE 49

(2X) (20X) Figure 3-14. Perls reaction counter -stained with Cresyl-Violet show s iron deposition in the inner molecular layer of the stratum granulosum of the dentate gyrus in epileptic brain (MR Rat 4). 49

PAGE 50

A (4X) (10X) B (2X) (10X) C Figure 3-15. Perls reaction count er-stained with Cresyl-Violet in (A) control (MR Rat 9), and (B) epileptic brain. The epileptic brain shows cavitation in the piriform cortex bilaterally with iron deposition (red arrow), as well as aggr egation of microglia (black arrows) around the cavities at hi gh magnification. (C) T2-weighted ex vivo MR scan showing piriform cortex region of interest. 50

PAGE 51

A 60 days after SE (4X) Control (4X) B (4X) Figure 3-16. Sections stained for myelin with Blac k Gold show (A) loss of myelin staining in the amygdala (black arrow). Note also the enlargement of the ventricles (star), secondary to loss of tissue and cavitat ion in the amygdala and pirifo rm cortex (blue arrow). (B) The CA1 subfield of the hippocampus (green arrow) also lacked myelin staining. 51

PAGE 52

52 A (20X) (20X) B (10X) (10X) Figure 3-17. Perl sections counter-stained with Cr esyl-Violet in animal s acrificed day 9 after SE (MR Rat 14, right). (A) granule cell layer of dentate gyrus showing diffuse iron desposition in SE animal, (B) piriform cortex showing cavitation and microglia aggregation (dark blue cells), compar ed to control (MR Rat 9, left).

PAGE 53

A B C D E Figure 3-18. Fluoro-Jade C section of one epilepti c brain sacrificed 9 days after SE (MR Rat 14) shows the presence of ongoing degenerating neurons in (A) the CA1 pyramidal cell layer, (B) CA3, (C) dentate hilus of the hippocampus, (D) thalamus and (E) piriform cortex. Neurons undergoing degeneration app ear fluorescent green compared to the background. These areas showed significant T2 increase between 3 and 10 days after SE. 53

PAGE 54

CHAPTER 4 DISCUSSION Clinically, MR imaging has b een useful in diagnosing the various types of epilepsy and detecting the secondary complicatio ns associated with seizures (Bernasconi, et al. 2005, Duncan 1997, Kimiwada, et al. 2006, Wieshmann 2003). In an imal research, the CLE model has been shown to produce pathophysiologic changes si milar to those observed in human MTLE. Considering the lack of understand ing of the causes and mechanisms leading to the recurrence of spontaneous seizures in this common type of dr ug-refractory epilepsy, it may be possible to gain insight into the mechanisms of epileptogenesis by using in vivo MR imaging. As of now, it is not possible to study the evol ution of damage from SE until the occurrence of spontaneous seizures in a single animal with a technique other than MRI. Continued research is needed to enhance this tec hnique and expand its uses to prov ide a better understanding of the mechanisms responsible for the pathologic cha nges associated with MTLE. For instance, an important question related to MTLE is the tempor al profile of pathologic changes following SE. Using in vivo MR imaging, it is possible to monitor thes e changes in the same animal over time. In the present study, a combination of in vivo ex vivo and histological methods was used to study an experimental model of MTLE. This is the first longitudinal imaging study using an animal model of MTLE with electrical stimulation in the hippocampus inducing SE, follo wed by a latent period before the appearance of spontaneous seizures (Bertram and Cornett 1994). The findings demonstrated that development of seizures was associated with significant changes in relaxation measurements following electrically induced SE, visible or not as signal intensity changes on T1 or T2weighted in vivo MRI images. These in vivo changes were due to unde rlying pathological tissue alterations evident with ex vivo MRI and histology. 54

PAGE 55

Significant changes in relaxation measurements were present three days after SE in the hippocampus, lateral ventricles, amygdala and piriform cortex (PC) regions. T2 relaxation measurements changed more significantly than T1 values in these regions. However, no signal changes in intensity were appa rent on T1 or T2-weighted sc ans in the hippocampus. The dorsal thalamus showed significant changes in rela xation times later duri ng the process, once spontaneous seizures had begun. These results show that limbic st ructures, as well as the dorsal thalamic areas, play a critic al role in epileptogenesis. The T2 signal is dependent on the relations hip between bound and free water in tissue, which is dependent on the macromolecular environm ent. Disruption in tiss ue integrity can result in increased free water in tissue and increased in T2 relaxation time. Changes in T2 in the brain have been related to pathologi cal processes like glio sis, demyelination, edema and neuronal loss (Armstrong 1993, Eriksson, et al. 2007, Petropoulos, et al. 2006). Those underlying pathological processes were assessed by histologi cal analysis following MR imaging. Here we could demonstrate that the first stru ctures injured after SE were the PC, amygdala, lateral ventricles, hippocampus and thalamus. Increased in vivo T1 and T2 measurements in the lateral ventricle/ fimbriae ar eas correlated with an enlarg ement of the ventricles. This enlargement was probably due to atrophy of the hippocampus and the piriform cortex/ amygdala regions visible on ex vivo MR images and histology sections (Wolf, et al. 2002). Decreased relaxation time measurements in the hippocampus mi ght be due to neuronal loss, gliosis and iron deposition observable in our hi stological results and describe d as hippocampal sclerosis in previous studies (Briellmann, et al. 2002, Fu erst, et al. 2003, Lewis 2005). Whereas MRI changes in the hippocampus, amygdala and lateral ventricles have been widely described in 55

PAGE 56

animal and human epilepsy research studies in the past, these do not emphasize changes in the PC and thalamus. The PC showed an initial increase in in vivo T1 and T2 values which peaked at 3 days after SE and normalized by 10 days, before the appear ance of spontaneous seizures, followed by a significant decrease between 20 and 60 days, and significantly lo wer T1 and T2 values with ex vivo MRI. Histological analysis re vealed neuronal loss resulting in cavitation bilaterally, as well as abnormal presence of microglia, as early as 9 days after SE. These structural changes were even more severe in epileptic animals at 60 days after SE, with st rong iron deposition and microglia aggregation around the caviti es. These results correlate with previous studies that have shown significant neuronal loss and g lial infiltration in the PC of p ilocarpine-treated rats as early as one week after SE (Ber tram and Cornett 1994, Chen, et al. 2007, Druga, et al. 2003) Recently, significant decreases in the numbers of pyramidal cells and interneurons in all layers of all parts of the PC were found in animals stimulated in the ventral hippocampus (Bolkvadze, et al. 2006). Several neurochemical studies have suggest ed that the PC is critically involved in the process of epileptogenesis. The PC contains the most suscep tible neural circuits of all forebrain regions for electrical (or chem ical) induction of limbic seizures (Loscher and Ebert 1996). During electrical stimulation of other limbic brain regions, larg e afterdischarges can be observed in the ipsilateral PC indicating that the PC is ac tivated early during the kindling process. The interictal discharge originates in the PC, independent of which structure serves as the kindled focus (McIntyre and Kelly 2000). The PC is the most sensitive brain structure to brain damage by continuous or frequent stim ulation of the amygdala or hippocampus. Amygdala kindling leads to a circumscribed loss of inhibitory GABA-immunoreactive neurons in the ipsilateral PC, which is likely to explain the in crease in excitability of PC pyramidal neurons 56

PAGE 57

during kindling (Lehmann and Lscher 1998). Kindling of the amygdala or hippocampus induces astrogliosis in the PC, indicating neuron al death in this brain region. Furthermore, activation of microglia is seen in the PC after amygdala kindling (Khurgel and Ivy 1996, Khurgel, et al. 1995). The decrease in relaxation time values obser ved 10 days after SE in our study correlates with the neuronal loss and probabl e gliosis reported in our hist ological results and previous research studies. The initial increase in relaxati on times peaking at day 3 after SE can probably be attributed to edema preceding neuronal lo ss and gliosis in the amygdalo-piriform region (Briellmann, et al. 2005, Meierkord, et al. 1997, Roch, et al. 2002). For the first time in this CLE model of MTLE, the pathological processes in the piriform cortex could be followed with in vivo MRI, in the same animal, from SE to the a ppearance of spontaneous seizures, confirming the important role of high resolution in vivo MRI in studying the progr ession of epilepsy. This study has also shown strong iron depositi on in the dorsal thalamic nuclei in animals undergoing SE and subsequent chro nic seizures. Iron deposition in the thalamus was observed as early as 9 days after st imulation after histologic al analysis. These iron deposits were seen as areas of decreased T2 and T1 measurements on MR I in our results as well as previous studies (Brittenham and Badman 2003, Clem ent, et al. 2007, Michaeli, et al. 2007, Neema, et al. 2007), due to the paramagnetic nature of iron particles. In a ddition to the limbic structures involved in the excitatory circuit of the entorhinalhippocampal loop (Lot hman, et al. 1991, Pare, et al. 1992, Rafiq, et al. 1993), subcortica l changes have been described in patients (Sperling, et al. 1990) and animal models (Clifford, et al. 1987), s uggesting that limbic epile psy can be viewed as a disorder of the limbic system as a whole with involvement of extra limbic structures (Bertram 1997, Bertram and Cornett 1994). 57

PAGE 58

In particular, several studies have suggested the midline thalamus to be importantly involved in seizure activity. Tr act tracing has shown reciproc al connections between the thalamus and limbic structures (DollemanVa nderWeel and Witter 1996, Wouterlood, et al. 1990). Other studies demonstrate that the thalam ic nuclei have an excitatory effect on the hippocampus (Bertram and Scott 2000, Dolleman-Van der Weel, et al. 1997). These nuclei also undergo specific pathologic changes in temporal epilepsy (Clifford, Olney, Maniotis, Collins and Zorumski 1987), especially in th e medial dorsal thalamus (Juhas z, et al. 1999). In an animal model of chronic limbic epilepsy stimulated in the hippocampus, the seizure onset in the thalamus was linked to seizure onset in the hi ppocampus; there was also consistent neuronal loss in the midline thalamus similar to neuronal loss and atrophy seen in hippocampal sclerosis (Bertram, et al. 2001, Druga, et al. 2005). Signifi cant changes in the physiology of the medial dorsal thalamic neurons from epileptic rats re sulted in hyperexcitability, and inhibition of neuronal activity in the midlin e thalamus could reduce seizure activity in the hippocampus Khurgel, et al. 1995). These changes were not ob served after multiple ki ndled seizures, which suggests that those changes in the midline thalamus are involv ed in the genesis of chronic epilepsy and are not a consequence of induced seizures. In our study, the presence of iron deposits in the dorsal thalamic nuclei suggests neuronal or microglial degeneration, which might take place as early as 9 days after SE, according to our histological data. Iron-mediated oxidative stress has been widely described in neurodegenerative diseases (Shoham 2001; Berg 2006; Brass 2006 (see ot her ref.); Stankiewicz 2007). It is not clear whether iron deposition contribu tes to the pathophysiology or is a result of tissue damage in these neurological disorders or epilepsy, but it has been present in many neurological disorders associated with neuronal degeneration. Iron in the dorsal thalamus, although not observable 58

PAGE 59

visually on T1 or T2-weighted in vivo MR images, was detectable by a significant decrease in T2 relaxation time in vivo after 10 days post-SE. High resolutio n MRI at 17.6 T revealed areas of hypointensities and decreased T2 values in the dorsal thalamic region bilaterally, which were correlated with iron deposition afte r histological analysis. This de position of iron in the thalamus detectable with MRI techniques, suggestive of neuronal degenera tion, has not yet been reported in previous epilepsy studies. In conclusion, we identified biologically relevant correlations between variations in MR relaxation time measurements and pathological changes occurring during epileptogenesis and spontaneous seizures. For instance, an increase in T2 relaxation time co rrelated with increased ventricular volume, associated with atrophy of the hippocampus and the PC/ amygdala regions in the brains of epileptic animals. Decreased T2 relaxation time was correlated with neuronal loss, gliosis and iron deposition in the dorsal thalam us and the hippocampus, particularly in the dentate gyrus and CA1 subregions. These in vivo MR changes were detect ed after quantification of relaxation times, but not always vi sible as signal intensity changes on in vivo images. Even though in vivo MRI was necessary for the longitudinal observation of the progression of the disease in the same animal, high reso lution MR microscopy and histology allowed for a more accurate determination of underlying pathologi cal changes. In order to obtain satisfactory signaltonoise ratios with in vivo imaging, many averages have to be acquired, and scan times are very long, raising the pr oblem of motion artifacts. Ex vivo MR imaging does not suffer from these drawbacks and higher spatial resolution, signal-to-noise ratio and sensitivity can be obtained with long acquisition times without trouble. Clinically, newer MRI methods with higher field strength, improved signal-to-noise ratios and newer sequences could detect cerebral 59

PAGE 60

abnormalities not identified on routine imaging in epileptic patients whose optimal MRI is said to be normal. 60

PAGE 61

61 LIST OF REFERENCES Aliashkevich AF, Yilmazer-Hanke D, Van Ro ost D, Mundhenk B, Schramm J, Blumcke I. (2003) Cellular pathology of amygdala ne urons in human tem poral lobe epilepsy. Acta Neuropathologica 106:99-106. Armstrong DD. (1993) The neuropathol ogy of temporal lobe epilepsy. J Neuropathol Exp Neurol 52:433-443. Babb TL, Kupfer WR, Pretorius JK, Crandall PH, Levesque MF. (1991) S ynaptic reorganization by mossy fibers in human epileptic fascia dentata. Neuroscience 42:351-363. Bartolomei F, Khalil M, Wendling F, Sontheimer A, Regis J, Ranjeva JP, Guye M, Chauvel P. (2005) Entorhinal cortex involvement in hum an mesial temporal lobe epilepsy: an electrophysiologic and volumetric study. Epilepsia 46:677-687. Bartolomei F, Wendling F, Bellanger J-J, Regis J, Chauvel P. (2001) Neural networks involving the medial temporal structures in temporal lobe epilepsy. Clinical Neurophysiology 112:1746-1760. Berg AT, Levy SR, Testa FM, Shinnar S. (1999) Treatment of Newly Diagnosed Pediatric Epilepsy: A Community-Based Study. pp. 1267-1271. Bernasconi N, Bernasconi A, Andermann F, Dubeau F, Feindel W, Reutens DC. (1999) Entorhinal cortex in temporal lobe epilepsy: a quantitative MRI study. Neurology 52:18701876. Bernasconi N, Natsume J, Bernasconi A. ( 2005) Progression in temporal lobe epilepsy: differential atrophy in mesi al temporal structures. Neurology 65:223-228. Bertram EH, Lothman EW, Lenn NJ. (1990) Th e Hippocampus in Experimental Chronic Epilepsy a Morphometric Analysis. Ann Neurol 27:43-48. Bertram EH, 3rd, Lothman EW. (1993) Morphometri c effects of intermitte nt kindled seizures and limbic status epilepticus in the dentate gyrus of the rat. Brain Res 603:25-31. Bertram EH, Cornett J. (1993) The ontogeny of se izures in a rat model of limbic epilepsy: evidence for a kindling process in the deve lopment of chronic spontaneous seizures. Brain Res 625:295-300. Bertram EH, Cornett JF. (1994) The evolution of a rat model of chronic spontaneous limbic seizures. Brain Res 661:157-162. Bertram EH. (1997) Functional anatomy of spon taneous seizures in a rat model of limbic epilepsy. Epilepsia 38:95-105. Bertram EH, Scott C. (2000) Th e pathological substrate of limbi c epilepsy: neuronal loss in the medial dorsal thalamic nucleus as the consistent change. Epilepsia 41 Suppl 6:S3-8.

PAGE 62

Bertram EH, Mangan PS, Zhang D, Scott CA, Williamson JM. (2001) The Midline Thalamus: Alterations and a Potential Role in Limbic Epilepsy. pp. 967-978. Bhagat YA, Obenaus A, Hamilton MG, Kendall EJ (2001) Magnetic resonance imaging predicts neuropathology from soman-mediat ed seizures in the rodent. NeuroReport 12:1481-1487. Blumcke I, Beck H, Lie AA, Wiestler OD. ( 1999) Molecular neuropath ology of human mesial temporal lobe epilepsy. Epilepsy Res Suppl 36:205-223. Blumenfeld H, McNally KA, Vanderhill SD, Paige AL, Chung R, Davis K, Norden AD, Stokking R, Studholme C, Novotny EJ, Zuba l IG, Spencer SS. (2004) Positive and negative network correlations in temporal lobe epilepsy. Cerebral Cortex 14:892-902. Bolkvadze T, Dzhaparidze N, Zhvaniya M, Kotariya N, Tsitsishvili A. (2006) Cellular Composition of the Piriform Cortex of th e Rat Brain in Experimental Epilepsy. Neuroscience and Behavioral Physiology 36:271-274. Bouilleret V, Nehlig A, Marescaux C, Namer IJ. (2000) Magnetic resonance imaging follow-up of progressive hippocampal changes in a mouse model of mesial temporal lobe epilepsy. Epilepsia 41:642-650. Briellmann RS, Kalnins RM, Berkovic SF, J ackson GD. (2002) Hippocampal pathology in refractory temporal lobe epile psy: T2-weighted signal change reflects dentate gliosis. Neurology 58:265-271. Briellmann RS, Wellard RM, Jackson GD. (2005) Se izure-associated abnormalities in epilepsy: evidence from MR imaging. Epilepsia 46:760-766. Brittenham GM, Badman DG. (2003) Noninvasive measurement of iron: report of an NIDDK workshop. Blood 101:15-19. Brooks-Kayal AR, Shumate MD, Jin H, Rikhter TY, Coulter DA. (1998) Selective changes in single cell GABAA receptor subunit expression a nd function in temporal lobe epilepsy. Nat Med 4:1166-1172. Buhl EH, Otis TS, Mody I. (1996) Zinc-induced collapse of augmented inhibition by GABA in a temporal lobe epilepsy model. Science 271:369-373. Buser P, Bancaud J. (1983) Unilateral connec tions between amygdala and hippocampus in man. A study of epileptic patients with depth electrodes. Electroencephalogr Clin Neurophysiol 55:1-12. Cavazos JE, Cross DJ. (2006) The role of synap tic reorganization in me sial temporal lobe epilepsy. Epilepsy Behav 8:483-493. Chang BS, Lowenstein DH. (2003) M echanisms of disease Epilepsy. New England Journal of Medicine 349:1257-1266. 62

PAGE 63

Chen S, Kobayashi M, Honda Y, Kakuta S, Sato F, Kishi K. (2007) Preferential neuron loss in the rat piriform cortex following pi locarpine-induced status epilepticus. Epilepsy Res Suppl 74:1-18. Clement F, Devos D, Moreau C, Coubes P, De stee A, Defebvre L. (2007) Neurodegeneration with brain iron accumulation: clinical, ra diographic and genetic heterogeneity and corresponding therapeutic options. Acta Neurol Belg 107:26-31. Clifford DB, Olney JW, Maniotis A, Collins RC, Zorumski CF. (1987) The functional anatomy and pathology of lithium-pilocarpine and high-dose pilocarpine seizures. Neuroscience 23:953-968. Cossart R, Dinocourt C, Hirsch JC, Merchan-Perez A, De Felipe J, Ben-Ari Y, Esclapez M, Bernard C. (2001) Dendritic but not somatic GABAergic inhibition is decreased in experimental epilepsy. Nat Neurosci 4:52-62. Dolleman-Van der Weel MJ, Lopes da Silva FH Witter MP. (1997) Nucleus Reuniens Thalami Modulates Activity in Hippocampal Field CA1 through Excitatory and Inhibitory Mechanisms. pp. 5640-5650. DollemanVanderWeel MJ, Witter MP. (1996) Project ions from the nucleus reuniens thalami to the entorhinal cortex, hippocampal field CA1, and the subiculum in the rat arise from different populations of neurons. Journal of Comparative Neurology 364:637-650. Druga R, Kubova H, Suchomelova L, Haugvic ova R. (2003) Lithium/pilocarpine status epilepticus-induced neuropathology of piriform co rtex and adjoining stru ctures in rats is age-dependent. Physiological Research 52:251-264. Druga R, Mares P, Otahal J, Kubova H. ( 2005) Degenerative neuronal changes in the rat thalamus induced by status epilepticu s at different developmental stages. Epilepsy Res Suppl 63:43-65. Du F, Whetsell WO, Abou-Khalil B, Blume nkopf B, Lothman EW, Schwarcz R. (1993) Preferential neuronal loss in laye r III of the entorhinal cortex in patients with temporal lobe epilepsy. Epilepsy Res Suppl 16:223-233. Dube C, Yu H, Nalcioglu O, Baram TZ. (2004) Se rial MRI after experime ntal febrile seizures: altered T2 signal without neuronal death. Ann.Neurol. 56:709-714. Duncan JS. (1997) Imaging and epilepsy. Brain 120 ( Pt 2):339-377. Engel J, Jr. (2001) Mesial Temporal Lobe Epilepsy: What Have We Learned? 7:340-352. Engelhorn T, Weise J, Hammen T, Bluemcke I, Hufnagel A, Doerfler A. (2007) Early diffusionweighted MRI predicts regional neuronal damage in generalized status epilepticus in rats treated with diazepam. Neurosci Lett 417:275-280. 63

PAGE 64

Eriksson SH, Free SL, Thom M, Martinian L, Symms MR, Salmenpera TM, McEvoy AW, Harkness W, Duncan JS, Sisodiya SM. ( 2007) Correlation of quantitative MRI and neuropathology in epilepsy surgical resec tion specimens--T2 correlates with neuronal tissue in gray matter. neuroimage 37:48-55. Fuerst D, Shah J, Shah A, Watson C. (2003) Hippocampal sclerosis is a progressive disorder: a longitudinal volumetric MRI study. Ann.Neurol. 53:413-416. Gloor P. (1997) The Temporal Lobe and Limbic System. Oxford University Press, Oxford. Goncalves PPM, Insausti R, Artacho-Perula E, Sa lmenpera T, Kalviainen R, Pitkanen A. (2005) MR volumetric analysis of the piriform cortex and cortical amygdala in drug-refractory temporal lobe epilepsy. ANJR Am J Neuroradiol. 26:319-332. Guilfoyle DN, Helpern JA, Lim KO. (2003) Diffusion tensor imaging in fixed brain tissue at 7.0 T. NMR Biomed 16:77-81. Hudson LP, Munoz DG, Miller L, McLachlan RS, Girvin JP, Blume WT. (1993) Amygdaloid sclerosis in temporal lobe epilepsy. pp. 622-631. Jackson G, McIntosh A, Briellmann R, Berkovi c S. (1998) Hippocampal sclerosis studied in identical twins. Neurology 51:78. Juhasz C, Nagy F, Watson C, da Silva EA, Muzi k O, Chugani DC, Shah J, Chugani HT. (1999) Glucose and [11C]flumazenil positron emission tomography abnormalities of thalamic nuclei in temporal lobe epilepsy. pp. 2037-40. Jupp B, Williams JP, Tesiram YA, Vosmansky M, O'Brien TJ. (2006) Hippocampal T2 signal change during amygdala kindling epileptogenesis. Epilepsia 47:41-46. Khurgel M, Ivy GO. (1996) Astrocytes in kindling: Relevance to epileptogenesis. Epilepsy Res Suppl 26:163-175. Khurgel M, Switzer RC, Teskey GC, Spiller AE Racine RJ, Ivy GO. (1995) Activation of Astrocytes During Epileptogenesis in the Absence of Neuronal Degeneration. Neurobiol Dis 2:23-35. Kimiwada T, Juhasz C, Makki M, Muzik O, Chugani DC, Asano E, Chugani HT. (2006) Hippocampal and thalamic diffusion abnormalities in children with temporal lobe epilepsy. Epilepsia 47:167-175. Kemppainen S, Jolkkonen E, Pitkanen A. (2002) Pr ojections from the posterior cortical nucleus of the amygdala to the hippocampal formation and parahippocampal region in rat. Hippocampus 12:735-755. Kemppainen S, Pitkanen A. (2004) Damage to the amygdalo-hippocampal projection in temporal lobe epilepsy: a tract-tracing st udy in chronic epileptic rats. Neuroscience 126:485-501. 64

PAGE 65

King M, Bruggen NV, Ahier RG, Cremer JE, Hajnal JV, Doran SRWM. (1991a) Diffusionweighted imaging of kainic acid lesions in the rat brain 20:158-164. King MD, Vanbruggen N, Ahier RG, Cremer JE, Hajnal JV, Williams SR, Doran M. (1991b) DIFFUSION-WEIGHTED IMAGING OF KAI NIC ACID LESIONS IN THE RATBRAIN. Magnetic Resonance in Medicine 20:158-164. Knudsen FU, Auk I. (2000) Clinical audit in the management of children with epilepsy. pp. 502504. Lehmann H, Lscher UEW. (1998) Amygdala-kin dling induces a las ting reduction of GABAimmunoreactive neurons in a discrete area of the ipsilateral piriform cortex 29:299-309. Lemaire C, Duncan EG, Solsberg MD, Armstrong RL. (1990) In vitro magnetic resonance microimaging of the spinal cord. Magn.Reson.Med. 14:97-104. Lewis DV. (2005) Losing neurons: selective vuln erability and mesial temporal sclerosis. Epilepsia 46 Suppl 7:39-44. Liu Z, Mikati M, Holmes GL. (1995) Mesial Temporal Sclerosis Pathogenesis and Significance. Pediatric Neurology 12:5-16. Loscher W, Ebert U. (1996) The role of the piriform cortex in kindling. Progress in Neurobiology 50:427-481. Lothman EW, Bertram EH, Bekenstein JW, Perlin JB. (1989) Self-sustaining limbic status epilepticus induced by 'continuous' hippocampal s timulation: electrographic and behavioral characteristics. Epilepsy Res 3:107-119. Lothman EW, Bertram EH, Kapur J, Stringer JL (1990) Recurrent Spontaneous Hippocampal Seizures in the Rat as a Chronic Sequ ela to Limbic Status Epilepticus. Epilepsy Res Suppl 6:110-118. Lothman EW, Bertram EH, Stringer JL. (1991) Functional anatomy of hippocampal seizures. Progress in Neurobiology 37:1-82. Maillard L, Vignal J-P, Gavaret M, Guye M, Biraben A, McGoni gal A, Chauvel P, Bartolomei F. (2004) Semiologic and Electrophysiologic Correlations in Temporal Lobe Seizure Subtypes. pp. 1590-1599. Mathern GW, Babb TL, Leite JP, Pretorius JK, Yeoman KM, Kuhlman PA. (1996) The pathogenic and progressive features of chronic human hippocampal epilepsy. Epilepsy Res Suppl 26:151-161. 65

PAGE 66

Mathern GW, Bertram EH, 3rd, Babb TL, Pretoriu s JK, Kuhlman PA, Spradlin S, Mendoza D. (1997) In contrast to kindled seizures, the fre quency of spontaneous epilepsy in the limbic status model correlates with greater aberrant fascia dentata excitato ry and inhibitory axon sprouting, and increased st aining for N-methyl-D-aspar tate, AMPA and GABA(A) receptors. Neuroscience 77:1003-1019. McIntyre DC, Kelly ME. (2000) The Parahi ppocampal Cortices and Kindling. pp. 343-354. Meierkord H, Wieshmann U, Niehaus L, Lehmann R. (1997) Structural c onsequences of status epilepticus demonstrated with se rial magnetic resonance imaging. Acta Neurol.Scand. 96:127-132. Michaeli S, Oz G, Sorce DJ, Garwood M, Ugurbil K, Majestic S, Tuite P. (2007) Assessment of brain iron and neuronal integrity in patient s with Parkinson's disease using novel MRI contrasts. Movement Disorders 22:334-340. Nairismagi J, Grohn OH, Kettunen MI, Nissi nen J, Kauppinen RA, Pitkanen A. (2004) Progression of brain damage after status epilepticus and its association with epileptogenesis: a quantitative MRI study in a rat model of tem poral lobe epilepsy. Epilepsia 45:1024-1034. Neema M, Stankiewicz J, Arora A, Dandamudi VSR, Batt CE, Guss ZD, Al-Sabbagh A, Bakshi R. (2007) T1-and T2-based MRI measures of diffuse gray matter and white matter damage in patients with multiple sclerosis. Journal of Neuroimaging 17:16S-21S. Niessen HG, Angenstein F, Vielha ber S, Frisch C, Kudin A, Elger CE, Heinze HJ, Scheich H, Kunz WS. (2005) Volumetric magnetic res onance imaging of functionally relevant structural alterations in chronic epilepsy after pilocarpine-ind uced status epilepticus in rats. Epilepsia 46:1021-1026. Pare D, deCurtis M, Llinas R. (1992) Role of the hippocampal-entorhinal loop in temporal lobe epilepsy: extraand intracellu lar study in the isolated guinea pig brain in vitro. pp. 18671881. Parent JM, Yu TW, Leibowitz RT, Geschwind DH, Sloviter RS, Lowenstein DH. (1997) Dentate Granule Cell Neurogenesis Is Increased by Seizures and Contributes to Aberrant Network Reorganization in the Adult Rat Hippocampus. pp. 3727-3738. Petropoulos H, Friedman SD, Sh aw DWW, Artru AA, Dawson G, Dager SR. (2006) Gray matter abnormalities in autism spectrum disorder revealed by T2 relaxation. Neurology 67:632636. Pikkarainen M, Ronkko S, Savander V, Insausti R, Pitkanen A. (1999) Projections from the lateral, basal, and accessory basal nuclei of the amygdala to the hippocampal formation in rat. Journal of Comparative Neurology 403:229-260. 66

PAGE 67

Pitkanen A, Nissinen J, Nairismagi J, Lukasi uk K, Grohn OH, Miettinen R, Kauppinen R. (2002) Progression of neuronal damage after status ep ilepticus and during spontaneous seizures in a rat model of temporal lobe epilepsy. Prog.Brain Res. 135:67-83. Rafiq A, DeLorenzo RJ, Coulter DA. (1993) Generation and propagation of epileptiform discharges in a combined entorhin al cortex/hippocampal slice. pp. 1962-1974. Roch C, Leroy C, Nehlig A, Namer IJ. (2002) Magnetic resonance imaging in the study of the lithium-pilocarpine model of temporal lobe epilepsy in adult rats. Epilepsia 43:325-335. Roch C, Leroy C, Nehlig A, Namer IJ. (2002b) Predictive value of cor tical injury for the development of temporal lobe epilepsy in 21-day-old rats: an MRI approach using the lithium-pilocarpine model. Epilepsia 43:1129-1136. Sander JW. (2003) The epidemio logy of epilepsy revisited. Curr Opin Neurol 16:165-170. Scharfman HE, Goodman JH, Sollas AL. (2000) Gra nule-Like Neurons at the Hilar/CA3 Border after Status Epilepticus and Their Synchr ony with Area CA3 Pyramidal Cells: Functional Implications of Seizure-Induced Neurogenesis. pp. 6144-6158. Sloviter R. (1991) Permanently altered hippocampa l structure, excitability, and inhibition after experimental status epilepticus in the rat: The ?dormant basket cell? hypothesis and its possible relevance to tempor al lobe epilepsy 1:41-66. Soltesz I. (1995) Brief history of cortico-hippocampal time with a sp ecial reference to the direct entorhinal input to CA1. Hippocampus 5:120-124. Sperling MR, Gur RC, Alavi A, Gur RE, Resnick S, Oconnor MJ, Reivich M. (1990) Subcortical Metabolic Alterations in Partial Epilepsy. Epilepsia 31:145-155. Sundqvist A. (2002) Epilepsy: a clinical diagnostic overview. European Journal of Pain 6:21-25. Sutula T, Cascino G, Cavazos J, Parada I, Ramirez L. (1989) Mossy fiber synaptic reorganization in the epileptic human temporal lobe. pp. 321-330. Tiina-Riikka M, Pitknen A, Tuunanen J, Ka uppinen RA. (2001) Ex vivo MR microimaging of neuronal damage after kainate-induced status epilepticus in rat: Correlation with quantitative histology. pp. 946-954. Tuunanen J, Lukasiuk K, Halonen T, Pitkanen A. (1999) Status epilep ticus-induced neuronal damage in the rat amygdaloid complex: distribution, time-course and mechanisms. Neuroscience 94:473-495. van Eijsden P, Notenboom RG, Wu O, de Graan PN, van Nieuwenhuizen O, Nicolay K, Braun KP. (2004) In vivo 1H magnetic resonan ce spectroscopy, T2-weighted and diffusionweighted MRI during lithium-pilocarpine-i nduced status epilepticus in the rat. Brain Res. 1030:11-18. 67

PAGE 68

68 von Bohlen und H, Schulze K, Albrecht D. (2004 ) Amygdala-kindling induces alterations in neuronal density and in dens ity of degenerated fibers. Hippocampus 14:311-318. Wall CJ, Kendall EJ, Obenaus A. (2000) Rapid al terations in diffusion-we ighted images with anatomic correlates in a rodent model of status epilepticus. AJNR Am.J.Neuroradiol. 21:1841-1852. Wang Y, Majors A, Najm I, Xue M, Comair Y, Modic M, Ng TC. (1996) Postictal alteration of sodium content and apparent diffusion coeffici ent in epileptic rat brain induced by kainic acid. Epilepsia 37:1000-1006. Wieshmann UC. (2003) Clinical applic ation of neuroimaging in epilepsy. J Neurol Neurosurg Psychiatry 74:466-470. Wilson CL, Isokawa M, Babb TL, Crandall PH. (1990) Functional Connections in the Human Temporal-Lobe .1. Analysis of Limbic Sy stem Pathways Using Neuronal Responses Evoked by Electrical-Stimulation. Experimental Brain Research 82:279-292. Wolf OT, Dyakin V, Patel A, Vadasz C, de Leon MJ, McEwen BS, Bulloch K. (2002) Volumetric structural magnetic resonance imaging (MR I) of the rat hippocampus following kainic acid (KA) treatment. Brain Res. 934:87-96. Wouterlood FG, Boon ME, Kok LP. (1990) Immunocyt ochemistry on Free-Floating Sections of Rat-Brain Using Microwave Irradiation During the Incubation in the Primary Antiserum Light and Electron-Microscopy. Journal of Neuroscience Methods 35:133-145.

PAGE 69

BIOGRAPHICAL SKETCH Lan Hoang Minh was born and grew up in Pa ris, France, in 1978. After graduating from high school baccalaureate with honors in 1996, Lan worked for one year as a administrative assistant for a medical clinic, before coming to the United States to attend the undergraduate program at the University of Florida, in the hope of integrating medi cal school after college. During her undergraduate educati on, she also gained valuable research experience as a UF Undergraduate Research Scholar in the microbi ology laboratory of Dr. Ke nneth Rand at Shands Hospital. Lan graduated with a Bachelor of Science in micr obiology in 2001 and entered the College of Medicine program at the University of Florida that Fall. After a couple of years in medical school, she joined the Biomedical Engin eering program that she deemed suited more her interests. She joined the laboratory of Dr. Th omas Mareci, in collaboration with Dr. Paul Carney's lab, in studying evolution into epilepsy using magnetic resonance imaging. 69