modeling fragile x syndrome

The cause of this syndrome is a triplet repeat expansion in the promoter oftheFMR1 gene, which ultimately results in the loss of an essential RNA-bindingprotein known as the fragile X me

Series Editors

Dietmar Richter, Henri Tiedge

.

Modeling Fragile X Syndrome

Robert B Denman

New York State Institute for Basic Research

Forest Hill Road 1050

Springer Dordrecht Heidelberg London New York

Library of Congress Control Number: 2011937964

# Springer-Verlag Berlin Heidelberg 2012

This work is subject to copyright All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks Duplication of this publication

or parts thereof is permitted only under the provisions of the German Copyright Law of September 9,

1965, in its current version, and permission for use must always be obtained from Springer Violations are liable to prosecution under the German Copyright Law.

The use of general descriptive names, registered names, trademarks, etc in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Series Editors

Dietmar Richter

Center for Molecular Neurobiology

University Medical Center

Department of Physiology andPharmacology

Department of NeurologySUNY Health Science Center at BrooklynBrooklyn, New York 11203

USAhtiedge@downstate.edu

and especially those relating to

understanding fragile X syndrome, we the authors dedicate this book to

Dr William T Greenough.

.

In the beginning of 2005, after finishing the galley proofs for “The Molecular Basis

of Fragile X Syndrome, Research Signpost” earlier that fall, I was invited toparticipate in a conference on fragile X syndrome This was one of the famedBanbury conferences which were held on the picturesque campus of Cold SpringHarbor Laboratory I had attended the inaugural one in 2000, where I met a childhoodidol, Dr James Watson As with all conferences there are highlights, the things thatleave an indelible impression on your memory, and there is the rest, which you, inshort order, forget For this particular Banbury conference, there was one talk whichbears on the creation of this book that I will never forget

The talk was given by Dr Richard Paylor of Baylor University and it concernedthe recent new behavioral tests that were being used in his laboratory to assess theseveral different fragile X mouse model strains that currently existed His group’swork definitively showed that specific behaviors and particular phenotypes pro-duced by the loss of the fragile X mental retardation protein were significantlyaffected by the mouse strain under investigation He summarized his findings byconstructing the behavior equivalent of a gene expression heat map and put forththe provocative thesis that in order to understand fragile X syndrome one mustassess phenotypes in a variety of model strains I remember afterwards thinking, intrue Darwinian fashion, that if strains could produce such profound effects, howmuch more so the species So to tease out the true fragile X phenotype, we may need

to examine behaviors in several species and would not that make an interestingbook project to edit

Except perhaps for the closing fragment in that last sentence such an idea wasnot novel because the Drosophila dFmr1 / model of fragile X syndrome wasalready well established in the literature and work characterizing the fragile X genefamily member expression in frogs and zebra fish had just been published Never-theless, it took a few more years before an opportunity arose to gestate this project.That opportunity came by way of an inquiry from Dr Henri Tiedge, co-editor of

“Results and Problems in Cell Differentiation”, as to whether I would be interested

in editing a volume on fragile X syndrome for the series I jumped at the chance and

vii

could not have been more pleased with the outcome I hope that you, the readerand especially those who are my colleagues in the fragile X field, agree with thisassessment.

It should be self-evident that like a symphony conductor an editor’s role in thebook-making process is mainly one of preparation and coordination; although oftenthe focus of the audience’s attention, a conductor should merely serve as a bridge,accepting the audience’s applause on behalf of the orchestra The real kudos belong

to the individual members for their performances This differentiates the roles ofeditors and conductors, as editors are often unheralded, anonymous fellows and that

is how it should be In contrast, authors are utterly like their orchestral counterparts

in deserving praise Therefore, I humbly and gratefully acknowledge my immensedebt to each of the chapter authors: first for doing the majority of the primaryresearch that enabled this project to be initiated and second for their willingness tocogently distill and disseminate their results here in these next pages They havetruly turned my dream into reality and collaborating with them has been one of thehighlights of my short editing career

2011

1 Introduction: Reminiscing on Models and Modeling 1Robert B Denman

Part I Ex Vivo Models

2 Probing Astrocyte Function in Fragile X Syndrome 15Shelley Jacobs, Connie Cheng, and Laurie C Doering

3 Neural Stem Cells 33Maija Castre´n

4 Fragile X Mental Retardation Protein (FMRP)

and the Spinal Sensory System 41Theodore J Price and Ohannes K Melemedjian

5 The Role of the Postsynaptic Density in the Pathology

of the Fragile X Syndrome 61Stefan Kindler and Hans-Ju¨rgen Kreienkamp

Part II Non-mouse Eukaryote Models

6 Behavior in a Drosophila Model of Fragile X 83Sean M McBride, Aaron J Bell, and Thomas A Jongens

7 Molecular and Genetic Analysis of the Drosophila Model

of Fragile X Syndrome 119Charles R Tessier and Kendal Broadie

ix

8 Fragile X Mental Retardation Protein and Stem Cells 157Abrar Qurashi, Xuekun Li, and Peng Jin

9 Manipulating the Fragile X Mental Retardation Proteins

in the Frog 165Marc-Etienne Huot, Nicolas Bisson, Thomas Moss,

and Edouard W Khandjian

10 Exploring the Zebra Finch Taeniopygia guttata as a Novel

Animal Model for the Speech–Language Deficit of Fragile X

Syndrome 181Claudia Winograd and Stephanie Ceman

Part III Novel Mouse Models

11 Neuroendocrine Alterations in the Fragile X Mouse 201Abdeslem El Idrissi, Xin Yan, William L’Amoreaux,

W Ted Brown, and Carl Dobkin

12 Taking STEPs Forward to Understand Fragile X Syndrome 223Susan M Goebel-Goody and Paul J Lombroso

13 Fmr-1 as an Offspring Genetic and a Maternal

Environmental Factor in Neurodevelopmental Disease 243Bojana Zupan and Miklos Toth

14 Mouse Models of the Fragile X Premutation and the

Fragile X Associated Tremor/Ataxia Syndrome 255Michael R Hunsaker, Gloria Arque, Robert F Berman,

Rob Willemsen, and Renate K Hukema

15 Clinical Aspects of the Fragile X Syndrome 273

18 The Fragile X-Associated Tremor Ataxia Syndrome 337Flora Tassone and Randi Hagerman

Part V Missing Models

19 Vignettes: Models in Absentia 361Robert B Denman

Index 385

.

Introduction: Reminiscing on Models

and Modeling

Robert B Denman

“From man or angel the great Architect did wisely to conceal, and not divulge his secrets to

be scanned by them who ought rather admire; or if they list conjecture, he his fabric of the heavens left to their disputes, perhaps to move his laughter at their quaint opinions wide hereafter, when they come to model heaven calculate the stars how will they wield the mighty frame, how build, unbuild, contrive to save appearances, how gird the sphere with centric and eccentric scribbled o’er, and epicycle, orb in orb.”

John Milton – Paradise Lost

“Models are to be used, not believed.”

H Theil – Principles of Econometrics

Abstract This chapter answers three basic questions, which are: (1) Why buildmodels, (2) why build models of fragile X syndrome, and (3) what has been learnedfrom the models of fragile X syndrome that have been made? The first question isused to frame the other two questions, providing the appropriate context by whichthe rest of the book should be examined Of necessity the last two questions are onlyaddressed briefly, and from one man’s point of view, as they contain the subjectmatter of the entirety of the book Thus, the reader is introduced to the varioustopics under review and urged to read for him/herself their contents, drawing suchconclusions as he/she thinks are warranted

R.B Denman ( * )

Biochemical Molecular Neurobiology Laboratory, Department of Molecular Biology, New York State Institute for Basic Research in Developmental Disabilities, 1050 Forest Hill Road, Staten Island, NY 10314, USA

e-mail: rbdenman@yahoo.com ; robert.denman@opwdd.ny.gov

R.B Denman (ed.), Modeling Fragile X Syndrome,

Results and Problems in Cell Differentiation 54,

DOI 10.1007/978-3-642-21649-7_1, # Springer-Verlag Berlin Heidelberg 2012

1

1.1 Origins and Necessity of Modeling

“Be fruitful and multiply, fill the earth and subdue it and have dominion over it.”The human imperative found in the biblical account of creation, and so to fulfilltheir destiny men, became natural scientists, observing the world and itsinhabitants, cataloging their observations, and distilling the data into useful theoriesthat ultimately transformed their environs and so too, themselves But despite thefact that this process has gone on since the beginning of historical time, humanityhas yet to obtain the complete dominion it quests for All you have to do is lookaround or watch a “year in review” on the news to know that this is true Ravagesdue to earthquakes, hurricanes, tornados and floods, not to mention global warming,species extinction, famine, pestilence, disease, and strife all testify to the fact that

we as a species have not mastered the world, its inhabitants, or ourselves

Why is this so? Quite simply, because the world and its inhabitants are complexand men are limited Our poets have rightly asked, “What is man that thou art mindful

of him”? Our songwriters have declared us to be, “dust in the wind,” and ourphilosophers often despair of humans knowing much of anything at all (Russell1912)

If we do not wish to acquiesce to our poets, our songwriters and our philosophershow then are we to proceed to total global dominion in the face of our limitations?One of the tools in the scientist’s arsenal is the model, a set of simplifying featuresthat allow them to clarify underlying problems and extract potentially usefulconclusions Rendered in this sense a model is a theoretical construct that uniquelyimpinges or corresponds to the natural world via its assumptions But do not take

my word for it The mathematician, John von Neumann, eerily echoes this when hestates, “the sciences do not try to explain, they hardly even try to interpret; theymainly make models By a model is meant a mathematical construct, which, withthe addition of certain verbal interpretations describes observed phenomena.” I canstill remember sitting in my undergraduate physical chemistry class and beingmesmerized by Professor Rolf Steinmann’s description of the particle in the box,that is, a particle, which is set in a well of length L whose sides have infinitepotential As you will recall, the particle can move in any of the three Cartesiandirections, but for simplicity is constrained to move only along theX axis Thesolution to this problem is found in solving the Schr€odinger wave equation:

dc

dx2þ4p2

l2 ðcÞ ¼ 0wherec is the wave function, and l is the wavelength

In solving this equation, one remarkably finds that the kinetic energy of theparticle:h2/mL2(whereh is Planck’s constant and m is the mass) has solutions, ifand only if,L¼ n l/2; that is, the energy of the system can only take on certaindiscrete values (n ¼ 1, 2, 3 .) or is “quantized.” It is just a short, albeit mathemat-ically intensive jaunt into polar coordinates to describe flesh and blood atoms interms of their three quantum numbers n, l, and m and begin to understand their

intrinsic properties.Thus, by first simplifying and then by adding complexity modelshelp us understand our universe and have dominion over it.

1.2 Utilitarian Features of Modeling

The above description of a model differs significantly from what you will find inliterature and the social sciences There models are often rather artificial constructs ofdubious worth that are added to a work as an organizing principle In his recentbestseller entitled “Genius,” the noted Yale literary critic Harold Bloom discourses

on the work of 100 literary geniuses (Bloom 2002) Although he explicitlyacknowledges the randomness of both genius and organizational principles, he never-theless endeavors to group the geniuses into ten sets of ten and assigns each set aspecific “Divine attribute” from the Kabbalah, which supposedly exemplifies the basiccharacteristics of those writers Within each set, the group of ten is broken down furtherinto two groups of five and each group of five is ordered from the genius bestexemplifying the Divine attribute to the one that least represents it How theseassignments are actually made is never explicitly stated, and so one perceives at theoutset a sense of arbitrariness in this framework Moreover, neither comparisons withineach set of geniuses nor contrasts among the sets are ever made, so as an analyticaltool this model of how to examine literary genius is more bluster than it is science.What is most disconcerting to me about this type of modeling is that it fails the test ofutilitarianism, i.e., for a model to be worthwhile, it must have some predictive power.Channeling von Neumann again, “the justification of a mathematical construct (model)

is solely and precisely that it is expected to work.” However, if you are looking for aharangue about how Shakespeare is a superior writer to the Bible’s authors or why Iago

is the most authentically human character ever written it is good read

Of course, physical scientists and social scientists are not the only ones thatconstruct and use models In the realm of medicine and disease we who seek curesoften model human maladies with other mammals, particularly primates (Fiandacaand Bankiewicz2010) and rodents (Cryan and Holmes2005) For example, thereare currently mice that model Alzheimer’s disease, Parkinsonism, prion diseases,hypercholesterolemia, Crohn’s disease, Down syndrome, and Prader–Willi syn-drome to name a few Here the model is less the theoretical framework describedabove than an actual physical entity, which has been designed and engineered tomimic a particular disease, something a logical positivist would love.The utility ofthese models is that they phenocopy one or more aspects of a disease and in doing

so can be used as tools to understand the molecular bases of a disease as well as ascreening agent for particular remedies

Along with these complex, “big brain” models (primates and rodents), there has alsobeen the concomitant development of “small brain” models like those found in smallvertebrates [zebrafish (Steenbergen et al 2011; van Tijn et al.2011; Peal et al.2010)and frogs (Pienaar et al.2010)] as well as other eukaryotes such as worms, flies, andhoneybees (Burne et al.2011) In fact, this trend to smallness can be seen in the recent

development of induced pluripotent stem cells (iPSs) disease models for spinal cular atrophy and other diseases (Vogel2010) These less complex models allow us

mus-to more easily or more precisely define anamus-tomical pathways, and or social behaviors,turn genes on and off, dissect molecular pathways, screen and test compounds, andevaluate theories of a disease’s etiology in the hope that this will lead to an under-standing that, in due course, yields a cure.In short, by modeling human disease inanimals we seek and sometimes acquire dominion of a small slice of the world

1.3 Modeling Fragile X Syndrome

To those inquisitive minds that have picked up this volume, you will find thecollected efforts of a small community to model fragile X syndrome (FXS) FXS

is, as all of our authors routinely say in their published work, “the most commoninherited cause of mental impairment and the most common known cause ofautism.” In the United States, FXS is as common as muscular dystrophy and cysticfibrosis The cause of this syndrome is a triplet repeat expansion in the promoter oftheFMR1 gene, which ultimately results in the loss of an essential RNA-bindingprotein known as the fragile X mental retardation protein (FMRP)

In actuality, the situation is more complex than the simple loss of a protein.Normal individuals have between 30 and 55 copies of a CGG triplet repeat in the

50untranslated region of theirFMR1 gene Individuals with more than 55 copies of

the repeat are classified as having the premutation (Bat et al.1997; Cunningham

et al 2011) As the number of repeats increases from 55, there is an increasedtranscription of FMR1 mRNA but a corresponding lag in protein production(Kenneson et al 2001) When the number of repeats reaches 200, the genebecomes hypermethylated and subject to transcriptional silencing (Verkerk et al

1991) Individuals with 200 or more repeats are classified as having the fullmutation and exhibit all of the hallmarks of FXS However, premutation maleshave been shown to suffer from a neurodegenerative disorder termed FragileX-associated Tremor/Ataxia Syndrome (FXTAS) (Hagerman and Hagerman

2004), while premutation females exhibit premature ovarian failure (POF) (Oostraand Willemsen2003)

But while FXS is that sterile clinically white-washed diagnosis, it is also muchmore It is families struggling for answers and coping with learning and behavioralproblems that seem baffling and at times insurmountable More than anythinghowever, FXS represents the loss of potential of our collective human soul, andthe sadness that is usually reserved for those great ones among us that die too soon.That is why we model FXS; that is why we will not rest until we find a cure andclaim dominion over a small niche in this wide world

So, how far have we come on our journey? While the reader ultimately willrender the final judgment on this matter, allow me as the first reader of this book toprovide an initial assessment, affixing some guideposts along the way At the time

of this writing, there is currently no cure for FXS; thus, most boys and many girls

afflicted with the disease remain significantly affected throughout their lives,although appropriate education and medications can help minimize the effectsand maximize the potential of each child Nevertheless, the cost to society fortreatment, special education, and lost income is staggering (Clapp and Tranfaglia

2011) While this is quite sobering I believe there is reason to hope Many of thesereasons lie in the research accomplishments that you will find in perusing thistome

First of all, it should be self-evident from simply theTable of Contents that thisbook tends to concentrate on animal models of FXS Seven of the chapters detail animpressive total of 22 distinct animal models ranging from the “large brain” mouseknockouts (Chaps.4,11,12–14) and double knockouts (Chaps.12,19) to the well-known small brain models ofDrosophila (Chaps.6 8),zebrafish (Chap.19) and themore exotic Xenopus (Chap 9), zebra finch (Chap 10), Cnidarian (Chap 19),Ciona (Chap.19),Aplysia (Chap.19), andGryllus (Chap.19) models From thisdiverse group, certain general and species-independent functions of Fmrp areextracted namely that it is a dendritic- and axonal-localized RNA-binding proteincomplexed in large neuronal granules that are involved in translational regulation atthe synapse and coupled to several receptor-mediated signaling pathways Corre-spondingly, its loss results in both subtle and profound changes in neuronalarchitecture, neuronal networking, circadian rhythm, and synaptic plasticity

A second feature emanating from the multiple models is presented here, and this

is especially true of the double knockout models described is a focus on remediatingFXS through both engineering and the development of small molecule therapeutics,which is explored in Chaps.6,12, and19 These efforts have paved the way for aninitial round of drug trials which are described by Hagerman et al in Chap.17.The molecular alterations in gene expression, translational regulation, andsignaling that derive from the loss of the fragile X mental retardation protein giverise to anatomical and neuroanatomical defects that result in networking or wiringabnormalities, which in turn produce the distinct behavioral phenotypes that weclassify as FXS The allure of stem cell biology is that by introducing “corrected”self-renewing neurons in a particular location at a precise moment in the developingbrain, all of these defects, molecular anatomical, networking, and behavioral can becorrected Here, Castren et al (Chap.3) and Qurashi et al (Chap.8) explore whathas been learned about the role Fmrp plays in the developing nervous system usingmouse, human, and Drosophila stem cell models The results show points ofconvergence and points of divergence between the big brain and small brainmodels, much like the functional similarities and dissimilarities observed betweenthe molecular aspects of each system Moreover, the recent findings that Fmrp isexpressed in astrocytes (Pacey and Doering2007) have allowed Jacobs et al (Chap.2)

to begin to outline the important role these cells play in shaping synaptogenesis.Declaring that the fly,Drosophila melanogaster, has been a workhorse of fragile

X research is either an oxymoron or aspeciespomorphism, but nonetheless it is also

a truism As a foil for the mouse, allowing us to understand with greater clarity theessential features produced by the loss of Fmrp, as a unique behavioral model, and

as a drug screening tool, the humble fly has driven and illuminated research on FXS

These and other qualities of theDrosophila model of FXS are mapped out by Bell

et al (Chap.6), Tessier et al (Chap.7), and Qurashi et al (Chap.8)

While the quintessential “small brain” model, Drosophila, has garnered thelion’s share of attention, fragile X researchers are nothing if not extremely innova-tive and having not been content with a single counterpoint to the mouse have gone

on to model in a variety of other species, which are summarized in (Chap.19).However, two of the more recent and unique animal models deserve specialattention They are the Xenopus model (Chap 9) and the zebra finch model(Chap.10)

As an experimental model of early development, the frog clearly rivals themouse (Kay and Peng 1991) Moreover, the ease by which frog eggs can

be extracted and manipulated have fostered work that ranges from examining theeffect cancer genes have on development (Sung et al.1996), to determining the roletranslational regulation plays in said process (Luo et al.2011) Previous studies byHuot et al demonstrated the use of this system in studying the fragile X homolog,FXR1P (Huot et al.2005), [reviewed in Denman (2008)] Here (Chap.9) Huot et al.expand their analyses to include work by Gessert et al., which tantalizingly shows apotential link between FXR1P, FMRP, and specific micro-RNAs (miRNAs) inregulating developmental programs involving the eye and connective tissues(Gessert et al.2010)

A constellation of fragile X pathologies little spoken of outside the clinic involvespeech Fragile X patients exhibit several speech-related deficits, both in speechproduction (articulation, perseveration) and in language competence (syntax, prag-matics) reviewed in Hagerman and Cronister (Bennetto and Pennington1996) Buthow to model these deficits? Flies and zebrafish do not speak and mice, high-pitched squeaks notwithstanding, only talk in cartoons The song bird,Taeniopygiaguttata, however, sings and as Winograd et al persuasively argue (Chap.10) boththe way zebra finch learn to sing and the anatomical makeup of their song nuclei areextremely similar to that of humans Therefore, they hypothesize that zebra finchwill be an extremely valuable model for understanding fragile X speech defects.The only caveat is that it still must be made, which could be accomplished usingviral vectors used by Zeier et al to locally ablate Fmr1 gene expression (Zeier et al

2009)

The role environment and epigenetics play in producing our phenomes and howthey combine to produce a disease state is an area that is just beginning to beexplored (Houle et al.2010) Here Zuppan and Toth (Chap 13) recount a novelbreeding strategy that allows them to tease out the genotypic effects of a particularphenotype from nongenetic, maternal-genotype-dependent ones They demonstratethat heterozygous wild-type offspring, i.e., Fmr1 heterozygous females bred towild-type males are more active than wild type–wild type offspring Moreover,male Fmr1 knockout mice born to heterozygous or Fmr1 knockout mothers areeven more active than the heterozygous wild-type offspring Thus, hyperactivity inthe Fmr1 knockouts is the result of the combination of the maternal and theoffspring genotype effects On the other hand, hyperreactivity, audiogenic seizuresusceptibility, and macroorchidism do not have a maternal-genotype component

As mentioned above, instability in CGG-repeat sizes above 55 in the FMR1promoter lead to its expansion and for repeats above 200 chromatin hyper-methylation ensues followed by transcriptional silencing Intermediate repeats,i.e., between 55 and 200 however, are associated with an RNA gain-of-functiontoxicity syndrome, FXTAS, whose etiology is outlined by Tassone and Hagerman(Chap 18) Efforts to model FXTAS in mice are described by Hunsaker et al.(Chap 14) Interestingly, while many of the features of the human disease arefaithfully recapitulated in mice expressing CGG-repeats in the intermediate range,the hallmark symptom, gait ataxia, is either absent, or occurs only in advanced agedepending on the model assessed This along with the difficulties of modeling some

of the more complex cognitive and behavioral phenotypes in mice lead theseauthors to call for the development of novel tasks that can be used to properlytest for subtler FXTAS phenotypes

1.4 Modeling Within the Model

As mentioned above, an attribute of the animal models described here is theirability to specifically test whether a putative molecular mechanism operates tocause all or some of the phenotypic features of the disease The synaptic plasticitymodel is perhaps the most all-encompassing of the mechanistic models of FXS as itsubsumes the most telling neuroanatomical anomaly of the disease, namely altereddendritic spines, with the host of molecular mechanisms by which this defect mightarise (Waung and Huber2009) From the careful quantitative work of Comery et al.(Comery et al.1997) showing that the mouse model of FXS recapitulates the spinedefects observed in human fragile X patients (Hinton et al.1991) to the more recenttranscranial two-photon imaging techniques demonstrating that the long immaturedendritic spines found in fragile X result from increased turnover, leading to anincrease in the number short-lived spines that are insensitive to sensory experience,providing a mechanistic basis for these observations has been a consuming interest

of fragile X researchers Here, Kindler and Kreienkamp (Chap.5) explore the rolethe postsynaptic density plays in FXS Their analyses link the activity-dependentprotein synthesis-dependent pathway, which utilizes a variety of kinases andphosphatases to control the synthesis of Fmrp target mRNAs directly to expressionand function of particular postsynaptic density proteins Similarly, Goebel-Goodyand Lombroso (Chap.12) add a new and important player in this process, striatal-enriched protein tyrosine phosphatase (STEP), an enzyme that regulates AMPAreceptor recycling and whose synthesis is under the direct control of Fmrp.Transcranial two-photon imaging requires a fluorescently labeled population ofneurons to examine, and the development of transgenic mouse lines that selectivelyexpress autofluorescent proteins (AFPs) has been instrumental in the success of thistechnology (Evanko 2007; Sigler and Murphy 2010) It turns out that culturedneurons expressing yellow fluorescent protein (YFP) provide an appropriate back-drop to perform a variety of imaging analyses, for example, allowing one to

determine when and where a particular protein is locally synthesized Importantly,Kao et al have shown that the Fmrp target mRNA, CamKIIa, is selectivelytranslated in mGluR5-containing dendritic spines following stimulation with S-3,5-dihdroxyphenylglycine (DHPG) (Kao et al.2010) These data confirm the criticalrole Fmrp plays in the activity-dependent regulation of target message expression(discussed in Chap.19).

Most researchers, psychiatrists, and clinicians tend to view FXS through thewindow of its cognitive impairments It is after all an example of an X-linkedmental retardation isn’t it (Raymond2006; Koukoui and Chaudhuri2007)? How-ever, the more visionary among us would point out that FXS is much more than justthe cognitive impairments we tend to focus on El Idrissi et al (Chap 11) haveexpanded their groundbreaking work on the GABAergic dysfunction in the Fmr1knockout mouse by showing that organs such as the pancreas that utilize GABA are

as substantially misregulated as the brain This results not only from deficits in theGABA system, but also decreases in voltage sensitive calcium channels (VSCCs)and in somatostatin production, which, in turn, give rise to their own set ofanatomical defects They put forward the startling hypothesis that FXS may best

be viewed as a “channelopathy” disease Similarly alterations in the spinal sensorysystem in FXS are just beginning to be considered Based on detailed behavioralanalyses of Fmr1 knockout mice, Price and Melemdjian (Chap.4) argue that theself-injurious behavior (SIB) characteristic of fragile X patients may be related to adecreased ability to sense painful stimuli Their results not only highlight that theloss of FMRP leads to profound changes in the peripheral nervous system, they alsoprovide additional behavioral tests to conduct when we examine strategies aimed atcorrecting the fragile X phenotype (discussed in Chap.19)

1.5 The Human Model

The development of the various fragile X animal models described here ispredicated upon their ability to recapitulate the phenotype(s) associated with thedisease That is why we must always go back to the one true model, humans There,along with the patients and their families, we encounter the other cast members ofour mortality play, the psychiatrists and clinicians Regarding this book, the distin-guished actors playing these roles are Dr Michael Tranfaglia, the chief scientificofficer of the FRAXA Research Foundation (FRAXA) and Dr W T Brown, thedirector of the New York State Institute for Basic Research in DevelopmentalDisabilities (IBR) In his chapter, Dr Tranfaglia details the cognitive and behav-ioral features that uniquely identify fragile X patients and provides an extensive list

of all of the medications that are used to treat the symptoms of the disease (Chap.16).Importantly, he predicts a future in which the symptomatic approach to treatmentthat is currently used will be supplanted by a disease-specific approach Whereaspsychiatrists deal with patient profiling and treatment clinicians deal with diseasediagnoses Here Dr Brown describes the methods used to diagnose FXS (Chap.14)

As the Southern blotting approach has given way to the information-rich precision

of RT-PCR for determining CGG-repeat size, Brown looks forward to the ment of a rapid quantitative sandwich ELISA assay for “low-cost newborn screen-ing to identify affected males.”

develop-1.6 Modeling FXS: Future Promises, Future Challenges

As the reader will discover in the ensuing chapters, FMRP is a broad specificityRNA-binding protein and as such regulates a host of neuronal and nonneuronalmRNAs In turn, FMRP is also regulated by a variety of stimulus-induced receptorpathways These two features have combined to yield a plethora of druggabletargets some of which are beginning to be evaluated in various sorts of drug trials.From antagonists to the mGluR5 receptor such as fenobam, AFQ056 and STX107,

or arbaclofen, which targets the GABABreceptor to the AMPA modulator, CX516,

or the matrix metalloproteinase inhibitor, minocycline, a spate of small-scale trialshave been undertaken and some candidates like donapezil have graduated to largertrials As Hagerman et al point out in their chapter (Chap.17), and this is echoed byTranfaglia (Chap.16), it is likely that combined treatments with multiple drugs willprobably be a necessary feature of future targeted FXS therapy

It would be very tempting to close this Introduction on modeling FXS here onthe fruits of our current efforts, these drug trials But to do so would be a gravedisservice to the reader as it would leave her/him with the false impression that ourlabors were both complete and successful In many respects however, we, thefragile X community of researchers, are really near the beginning of our quest

On a molecular level, we have barely scratched the surface of what we need to know

in order to develop targeted and effective medications Most of the FMRP-regulatedmRNAs obtained via high throughput assays (Sung et al.2000; Brown et al.2001;Darnell et al.2001; Chen et al.2003; Miyashiro et al.2003) have yet to be validated

by alternative means much less understood within a framework of FMRP tion Similarly, though we have identified three unique FMRP–RNA interactionmotifs, i.e., G-quartets, the kissing complex, and U-rich stretches (Schaeffer et al

interac-2001; Chen et al.2003; Darnell et al.2005), we know very little about how FMRPaccesses them in vivo in an intact full-length mRNA, whether multipleRNA–protein interactions can occur and if they can what the spacing and otherstructural requirements for the interactions are, and how such interactions may bemodified by other members of the FXRP family or other FMRP-interactingproteins Finally, FMRP–messenger ribonucleoprotein particles (mRNPs) are notindependent entities, rather they form heterogeneous higher order complexes with avariety of auxiliary and RNA-binding proteins that are collectively termed neuronalgranules These granules are translationally inactive and are used for transport andstimulus-induced translocation into dendritic spines where they play an intimaterole in shaping the postsynaptic density via local translation However, whileneuroscientists have worked out this scenario over the last 30 years (Rao and

Steward1991; Krichevsky and Kosik2001; Kanai et al.2004; Elvira et al.2006),the remodeling, which forms the molecular basis of this mechanism is still a blackbox, although it is likely governed, at least in part, by posttranslationalmodifications (Dolzhanskaya et al.2006; Xie and Denman 2011) Thus, we arecurrently years behind the story of the NOVA RNA network model (Zhang et al.

2010), and fragile X researchers have the added burden of dealing with a variety ofmutually expressed isoforms that likely modulate both RNA binding andprotein–protein interactions Nevertheless, the success of our modeling efforts todate has brought us closer to our goal of understanding and treating FXS andprovides hope for our future endeavors If I were a reader, I would be anxiouslyawaiting the publication of the sequel to this volume

Robert B Denman, editor

Staten Island, NY, 2011

References

Bat O, Kimmel M, Axelrod DE (1997) Computer simulation of expansions of DNA triplet repeats

in the fragile X syndrome and Huntington’s disease J Theor Biol 188:53–67

Bennetto L, Pennington BF (1996) The neuropsychology of fragile X syndrome In: Hagerman RJ, Cronister A (eds) Fragile X syndrome: diagnosis, treatment and research The Johns Hopkins University Press, Baltimore, pp 210–248

Bloom H (2002) Genius: a mosaic of one hundred exemplary creative minds Warner Books, New York

Brown V, Jin P, Ceman S, Darnell JC, O’Donnell WT, Tenenbaum SA, Jin X, Feng Y, Wilkinson

KD, Keene JD (2001) Microarray identification of FMRP-associated brain mRNAs and altered mRNA translational profiles in fragile X syndrome Cell 107:477–487

Burne T, Scott E, van Swinderen B, Hilliard M, Reinhard J, Claudianos C, Eyles D, McGrath J (2011) Big ideas for small brains: what can psychiatry learn from worms, flies, bees and fish? Mol Psychiatry 16:7–16

Chen L, Yun S-W, Seto J, Liu W, Toth M (2003) The fragile x mental retardation protein binds and regulates a novel class of mRNAs containing U-rich target sequences Neuroscience 120:1005–1017

Clapp K, Tranfaglia M (2011) Diagnosis and Treatment www.fraxa.org

Comery TA, Harris JB, Willems PJ, Oostra BA, Irwin SA, Weiler IJ, Greenough WT (1997) Abnormal dendritic spines in fragile X knockout mice: maturation and pruning deficits Proc Natl Acad Sci USA 94:5401–5404

Cryan JF, Holmes A (2005) The ascent of mouse: advances in modelling human depression and anxiety Nat Rev Drug Discov 4:775–790

Cunningham CL, Martı´nez Cerden˜o V, Navarro Porras E, Prakash AN, Angelastro JM, Willemsen

R, Hagerman PJ, Pessah IN, Berman RF, Noctor SC (2011) Premutation CGG-repeat sion of the Fmr1 gene impairs mouse neocortical development Hum Mol Genet 20:64–79 Darnell JC, Jensen KB, Jin P, Brown V, Warren ST, Darnell RB (2001) Fragile X mental retardation protein targets G quartet mRNAs important for neuronal function Cell 107:489–499

expan-Darnell JC, Fraser CE, Mostovetsky O, Stefani G, Jones TA, Eddy SR, expan-Darnell RB (2005) Kissing complex RNAs mediate interaction between the fragile-X mental retardation protein KH2 domain and brain polyribosomes Genes Dev 19:903–918

Denman RB (2008) FXR1P: the fragile X family member involved in muscle development In: Denman RB (ed) RNA binding proteins in development and disease India: Research Signpost, Trivandrum, pp 123–138

Dolzhanskaya N, Merz G, Aletta JM, Denman RB (2006) Methylation regulates FMRP’s lular protein–protein and protein–RNA interactions J Cell Sci 119:1933–1946

intracel-Elvira G, Wasiak S, Blandford V, Tong X-K, Serrano A, Fan X, del Rayo Sanchez-Carbente M, Servant F, Bell AW, Boismenu D, Lacaille J-C, McPherson PS, DesGroseillers L, Sossin WS (2006) Characterization of an RNA granule from developing brain Mol Cell Proteomics 5:635–651

Evanko D (2007) Windows on the brain Nat Meth 4:474

Fiandaca MS, Bankiewicz KS (2010) Gene therapy for Parkinson’s disease: from non-human primates to humans Curr Opin Mol Ther 12:519–529

Gessert S, Bugner V, Tecza A, Pinker M, K €uhl M (2010) FMR1/FXR1 and the miRNA pathway are required for eye and neural crest development Dev Biol 341:222–235

Hagerman PJ, Hagerman RJ (2004) Fragile X-associated tremor/ataxia syndrome (FXTAS) Ment Retard Dev Disabil Res Rev 10:25–30

Hinton V, Brown WT, Wisniewski K, Rudelli R (1991) Analysis of the neocortex of three male with the fragile X syndrome Am J Med Gen 41(3):289–294

Houle D, Govindaraju DR, Omholt S (2010) Phenomics: the next challenge Nat Rev Genet 11:855–866

Huot M-E, Bisson N, Davidovic L, Mazroui R, Labelle Y, Moss T, Khandjian EW (2005) The RNA-binding protein fragile X-related 1 regulates somite formation in Xenopus laevis Mol Biol Cell 16(9):4350–4361

Kanai Y, Dohmae N, Hirokawa N (2004) Kinesin transports RNA: isolation and characterization

of an RNA-transporting granule Neuron 43:513–525

Kao D-I, Aldridge GM, Weiler IJ, Greenough WT (2010) Altered mRNA transport, docking, and protein translation in neurons lacking fragile X mental retardation protein Proc Natl Acad Sci USA 107:15601–15606

Kay BK, Peng HB (eds) (1991) Xenopus laevis: practical uses in cell and molecular biology Harcourt Brace Jovanovich, New York

Kenneson A, Zhang F, Hagedorn CH, Warren ST (2001) Reduced FMRP and increased FMR1 transcription is proportionally associated with CGG repeat number in intermediate-length and premutation carriers Hum Mol Genet 10:1449–1454

Koukoui SD, Chaudhuri A (2007) Neuroanatomical, molecular genetic, and behavioral correlates

of fragile X syndrome Brain Res Rev 53:27–37

Krichevsky AM, Kosik KS (2001) Neuronal RNA granules: a link between RNA localization and stimulation-dependent translation Neuron 32:683–696

Luo X, Nerlick S, An W, King ML (2011) Xenopus germline nanos1 is translationally repressed by

a novel structure-based mechanism Development 138:589–598

Miyashiro KY, Beckel-Mitchener A, Purk TP, Becker KG, Barret T, Liu L, Carbonetto S, Weiler

IJ, Greenough WT, Eberwine J (2003) RNA cargoes associating with FMRP reveal deficits in cellular functioning in Fmr1 null mice Neuron 37:417–431

Oostra BA, Willemsen R (2003) A fragile balance: FMR1 expression levels Hum Mol Genet 12: R249–R257

Pacey LKK, Doering LC (2007) Developmental expression of FMRP in the astrocyte lineage: implications for fragile X syndrome Glia 55:1601–1609

Peal D, Peterson R, Milan D (2010) Small molecule screening in Zebrafish J Cardiovasc Transl Res 3:454–460

Pienaar IS, G €otz J, Feany MB (2010) Parkinson’s disease: Insights from non-traditional model organisms Prog Neurobiol 92:558–571

Rao A, Steward O (1991) Evidence that protein constituents of postsynaptic membrane specializations are locally synthesized: analysis of proteins synthesized within synaptosomes.

J Neurosci 11:2881–2895

Raymond FL (2006) X-linked mental retardation: a clinical guide J Med Genet 43(3):193–200 Russell B (1912) Problems of philosophy Bookjungle, Champaign

Schaeffer C, Bardoni B, Mandel J-L, Ehresmann B, Ehresmann C, Moine H (2001) The fragile X mental retardation protein binds specifically to its mRNA via a purine quartet motif EMBO J 20:4803–4813

Sigler A, Murphy TH (2010) In vivo 2-photon imaging of fine structure in the rodent brain: before, during, and after stroke Stroke 41:S117–S123

Steenbergen PJ, Richardson MK, Champagne DL (2011) The use of the zebrafish model in stress research Prog Neuropsychopharmacol Biol Psychiatry (In Press, Corrected Proof)

Sung Y-J, Hwang M-CC, Hwang Y-W (1996) The dominant negative effects of H-Ras harboring a Gly to Ala mutation at position 60 J Biol Chem 271:30537–30543

Sung Y-J, Conti J, Currie JR, Brown WT, Denman RB (2000) RNAs that interact with the fragile

X syndrome RNA binding protein FMRP Biochem Biophys Res Commun 275:973–980 van Tijn P, Kamphuis W, Marlatt MW, Hol EM, Lucassen PJ (2011) Presenilin mouse and zebrafish models for dementia: focus on neurogenesis Prog Neurobiol 93(2):149–164 Verkerk AJ, Pieretti M, Sutcliffe JS, Fu YH, Kuhl DP, Pizzuti A, Reiner O, Richards S, Victoria

MF, Zhang FP et al (1991) Identification of a gene (FMR-1) containing a CGG repeat coincident with a breakpoint cluster region exhibiting length variation in fragile X syndrome Cell 65:905–914

Vogel G (2010) Diseases in a dish take off Science 330:1172–1173

Waung MW, Huber KM (2009) Protein translation in synaptic plasticity: mGluR-LTD, Fragile X Curr Opin Neurobiol 19:319–326

Xie W, Denman RB (2011) Protein methylation and stress granules: Post-translational modifier or innocent bystander? Mol Biol Int pp 1–14, Article ID: 137459

Zeier Z, Kumar A, Bodhinathan K, Feller JA, Foster TC, Bloom DC (2009) Fragile X mental retardation protein replacement restores hippocampal synaptic function in a mouse model of fragile X syndrome Gene Ther 16:1122–1129

Zhang C, Frias MA, Mele A, Ruggiu M, Eom T, Marney CB, Wang H, Licatalosi DD, Fak JJ, Darnell RB (2010) Integrative modeling defines the Nova splicing-regulatory network and its combinatorial controls Science 329:439–443

.

Probing Astrocyte Function in Fragile

X Syndrome

Shelley Jacobs, Connie Cheng, and Laurie C Doering

Abstract Astrocytes have been recognized as a class of cells that fill the spacebetween neurons for more than a century From their humble beginnings in theliterature as merely space filling cells, an ever expanding list of functions in theCNS now exceeds the list of functions performed by neurons In virtually alldevelopmental and pathological conditions in the brain, astrocytes are involved insome capacity that directly affects neuronal function Today we recognize thatastrocytes are involved in the development and function of synaptic communica-tion Increasing evidence suggests that abnormal synaptic function may be aprominent contributing factor to the learning disability phenotype With the dis-covery of FMRP in astrocytes, coupled with a role of astrocytes in synapticfunction, research directed to glial neurobiology has never been more important.This chapter highlights the current knowledge of astrocyte function with a focus ontheir involvement in Fragile X syndrome

2.1 Historical Synopsis of Astrocytes

The term astrocyte is first mentioned in 1891 by Michael von Lenhossek in theGerman journal “Verhandlungen der Anatomischen Gesellschaft” (Transactions ofthe Anatomical Society) He indicated in his writings that glial cells should beconsidered to consist of more than one cell type Lenhossek wrote “I would suggestthat all supporting cells be named spongiocytes And the most common form invertebrates be named spider cells or astrocytes, and use the term neurogliaonly cumgrano salis (with a grain of salt), at least until we have a clearer view”(von Lenhossek1891)

S Jacobs • C Cheng • L.C Doering ( * )

Department of Pathology and Molecular Medicine, McMaster University, Hamilton, ON, Canada L8N 3Z5,

e-mail: sjacobs@mcmaster.ca ; chengc8@mcmaster.ca ; doering@mcmaster.ca

R.B Denman (ed.), Modeling Fragile X Syndrome,

Results and Problems in Cell Differentiation 54,

DOI 10.1007/978-3-642-21649-7_2, # Springer-Verlag Berlin Heidelberg 2012

15

Today, we use the term neuroglia as a broad inclusive term for all cells inthe nervous system that are not neurons Most articles cite Rudolph Virchow(1821–1902) as introducing the term neuroglia (often translated as nerve glue,cement or putty) to the scientific community In 1856, Virchow used the termneuroglia or “nervenkitt” to describe the connective or interstitial tissue substance

in the brain, recognizing that it differed in appearance and consistency from otherorgans Kettenman and Ransom (2005) point out that (although not highlighted inmost historical accounts) Virchow only used the term neuroglia to refer to theinterstitial substance and not to the cellular elements contained within the sub-stance The brain tissue that Virchow was likely referring to was the neuropil orthe astrocyte processes that we readily identify today As an interesting side note,

it was a student of Rudolph Virchow who stenographed a lecture series given byVirchow at Berlin University It was this stenographing that placed the neurogliaconcept into print for the first time with subsequent dissemination around theworld (Kettenmann and Verkhratsky 2008) Somjen (1988) provides furtherinsight into the historical evolution of the broad separation of glial cells fromneurons by scientists including Santiago Ramon y Cajal, Camillo Golgi, and OttoDeiters

2.2 Developmental Origin of the Astrocyte

Research on the glial lineage has expanded dramatically in recent years Withadvances in microscopic technology and improved methods of cellular identifica-tion, the complexity of astrocyte differentiation and function becomes increasinglyapparent each year

Astrocytes develop from precursor cells in at least three different regions of thebrain (Goldman2003) In the developing cerebrum, these include precursor cells inthe subventricular zone (SVZ), the radial glia of the ventricular zone (VZ), andfrom glial precursors not necessarily confined to the SVZ or the VZ layer.The SVZ, located just beneath the ventricular walls, is a principal source ofneural stem cells in the adult mammalian brain Stem cells in this zone generatethousands of progenitors that form neurons and glial cells each day during devel-opment (Alvarez-Buylla et al 2001; Doetsch et al 1999) In the presence ofexogenous mitogens like fibroblast growth factor, SVZ-derived neural stem cellsgrown in vitro self-renew and differentiate into all three lineages includingastrocytes (Reynolds and Weiss1992; Chiasson et al.1999; Gritti et al.2002)

In addition to the stem cells of the SVZ, another distinct group of cells referred

to as radial glia give rise to astrocytes Radial glia were repeatedly identified at theend of the nineteenth century (Magini1888; Retzius1893; von Lenhossek1895;Ramon y Cajal1899) Consistent with what is known today, the illustrations inthese early papers accurately showed these cells as having their cells bodies in the

VZ with processes spanning the complete thickness of the developing cerebralcortex

Differentiated astrocytes that reenter the cell cycle can also serve as precursors

to astrocytes and other cell types (Ganat et al 2006) These astrocytes locatedthroughout all regions of the brain represent a form of plasticity in the brain andthey respond in cases of injury or dysfunction

Freeman (2010) reviews various intrinsic epigenetic mechanisms that convergewith extrinsic signals to generate astrocyte differentiation from neural precursorcells Signaling through the Wnt and JAK-STAT pathways are prominent in drivingprecursor cells to an astrocyte fate

Our initial concept of just one or two different types of astrocytes is graduallybeing redefined For example, many subtypes of astrocytes are now being identifiedaccording to their location in the brain and on the spatial domain or microniche ofthe astrocyte environment Three subpopulations of astrocytes in the white matter

of the spinal cord have been identified on the basis of a combinatorial expression ofthe guidance molecules Reelin and Slit1 The positional identities of these astrocytesubtypes are specified by the homeodomain transcription factors Pax6 and Nkx6.1(Hochstim et al.2008) With a transcriptome database now available (Cahoy et al

2008), insight into new astrocyte subtypes from developmental and functionalperspectives will continue to emerge

2.3 Astrocytes Link Developmental Form with Function

Studies from recent years have revealed that astrocytes perform a significantlywider range of functions than previously appreciated Technological advances inmolecular approaches continue to reveal an ever expanding list of functions for theastrocyte in all ages of the nervous system Beyond the more commonly describedfunctions of modulating neurovascular blood flow (Attwell et al.2010) and theregulation of the extracellular ionic milieu (homeostasis) (Walz1989), research hasshown that astrocytes shape the synaptic environment (Ullian et al 2004) andgenerate signaling mechanisms within neural networks via calcium waves (Volterraand Meldolesi 2005) Table 2.1 outlines several astrocyte functions andcorresponding review papers The most recent comprehensive reviews of astrocytefunction include Wang and Bordey (2008) and Kimelberg (2010)

2.3.1 Astrocyte Cytoarchitecture

The classical Golgi impregnation techniques and immunocytochemical methodsused to identify glial fibrillary acidic protein (GFAP) reveal protoplasmic astrocyteswith a relatively simple stellate appearance In contrast to the classical descriptions

of astrocyte shape, the work of Bushong et al (2002), Halassa et al (2007), Ogataand Kosaka (2002), and Oberheim et al (2006) reveal astrocytes with exceedinglydense arrays of processes that radiate in a symmetrical fashion from the cell body

Table 2.1 Listing of various astrocyte functions in the CNS

Astrocyte function Sub-function Description

Suggested papers/ reviews

Clearance by uptake of excess extracellular K+ions; distribution of the ions through the astrocytic syncytium

Walz ( 2000 )

Neurotransmitter reuptake and release

High-affinity uptake of glutamate and GABA mediated by plasma membrane transporters

Kimelberg ( 2007 )

Release of glutamate or ATP

in a vesicular, Ca2+ dependent manner

-Schousboe and Waagepetersen ( 2004 ) Metabolic

support

Uptake and metabolism of glutamate into glutamine for re-distribution to neurons

Schousboe and Waagepetersen ( 2004 ) Uptake of glucose via

GLUT1 transporter found

in astrocyte endfeet surrounding capillaries

Porras et al ( 2008 )

Regulate neuronal metabolic responses to activity via:

(1) astrocytic glycogen (short term repository for glucose in the brain) and (2) lactate (released to neurons as energy substrate)

Pellerin et al ( 2007 )

Blood brain barrier (BBB)

Regulate induction, maintenance and permeability of BBB (tight junction formation, expression of various transport systems and secretion of molecules)

Abbott ( 2000 ), Abbott

et al ( 2006 ), Haseloff et al ( 2005 )

Neural development Neurogenesis GFAP-expressing cells in the

SVZ or SGZ can contribute to cell genesis both as stem cells and as neural components of the neurogenic niche

Barkho et al ( 2006 ), Lie et al ( 2005 )

Synaptic regulation Modulate

synaptic transmission and neural activity

Astrocytic glutamate release modulates synaptic transmission by activating presynaptic and postsynaptic glutamate receptors

Paixao and Klein ( 2010 )

Generate signaling mechanisms within neural networks via calcium waves

Volterra and Meldolesi ( 2005 )

(continued)

Table 2.1 (continued)

Astrocyte function Sub-function Description

Suggested papers/ reviews

Synaptogenesis Promote synaptogenesis

between CNS neurons by release of diffusible molecules

Christopherson et al (2005), Ethell and Pasquale ( 2005 ), Ullian et al ( 2004 ) Synaptic

plasticity

Modulate synaptic function through their role in glutamate re-uptake at the synapse by action of excitatory amino acid transporters (EAATs)

Paixao and Klein ( 2010 )

Modulate intensity and duration of postsynaptic activation (eg release of glutamate prompting LTP, preservation of synaptic strength by release of TNF- a, etc)

Barker and Ullian ( 2010 ), Beattie

et al ( 2002 ), Bergami et al ( 2008 )

Other Vasomodulation Coordinate blood flow to the

brain (functional hyperaemia)

Iadecola ( 2004 )

Control of blood glucose and

O2by neurotransmitter mediated signaling (predominantly by glutamate)

Attwell et al ( 2010 ), Iadecola and Nedergaard ( 2007 )

Secrete vasoactive agents to induce vasoconstriction

or vasodilation (correlated with increased intracellular Ca2+)

Zonta et al ( 2003 )

Detoxification Prevent excitotoxic neuronal

death by capturing excess ammonia and glutamate (converting them to glutamine) Uptake of toxic or heavy metals

Chung et al ( 2008 ), Struzynska et al ( 2001 )

Immune activation

Bridges CNS and immune system:

(1) Express MHC II and costimulatory molecules important for T-cell activation and antigen presentation

Dong and Benveniste ( 2001 )

(2) Express receptors involved innate immunity

Farina et al ( 2007 ) (3) Secrete a wide variety of

chemokines and cytokines that act as immune mediators

The complexity of astrocyte morphology is further highlighted by the facts that asingle astrocyte (in the rodent) can contact 300–600 dendrites (Halassa et al.2007),and each astrocyte oversees in excess of 100,000 synapses (Bushong et al.2002) Inthe human, these values are further increased due to the larger size of the protoplas-mic astrocytes Oberheim et al (2006) estimate that a single astrocyte in the humanbrain contacts on the order of two million synapses Clearly, while difficult toconceptualize the numerical value of astrocyte contacts with neurons, the morpho-logical arrangement of astrocyte processes at the synaptic level is critical to properfunction.

Specialized anatomical tracing techniques have revealed that matureastrocytes occupy distinct, nonoverlapping domains in the brain (Bushong et al

2002; Halassa et al.2007) During development, the astrocyte processes appearquite ragged and display overlapping zones with adjacent astrocytes By the 3rd

to 4th week after birth, neighboring astrocytes occupy very distinct spatialdomains with no overlapping processes Remarkably, there is a striking similarity

of these modern images to a figure published by von Lenhossek in 1893 (seeFig.2.1) It is believed that astrocytes “tile” with one another through a mecha-nism that is similar to dendritic tiling (reviewed by Freeman 2010) Thisanatomical arrangement has been taken one step further in theory, with thesuggestion that the defined domains of astrocytes function as synaptic islands(Halassa et al.2007) This concept proposes that all the synapses confined withinthe boundaries of an individual astrocyte are modulated by the gliotransmitterenvironment of the same astrocyte

Stützzellen (Astroeyten) aus dem Rückenmarke eines 3/4 jährgen Kindes, mit der Golgi’schen Methode dargestellt.

Rückenmark eines 14 cm langen menschlichen Embryos, nach Golgi behandelt, mit imprägnierten Stützzellan Links Ependymgerüst, rechts Vorläufer der Spinnenzellen (Astroblasten).

Fig 2.1 Reproduction of the diagram from the 1893 article by Michael von Lenhossek (a) Golgi impregnation of astrocytes in the spinal cord of a 9 month old child Note the exquisite pattern of astrocyte “tiling” that is observed by modern day methods of cell filling (b) Spinal cord prepara- tion of a 14 cm human embryo reveals patterns of radial glial spanning the entire thickness of the cord Golgi impregnation

2.3.2 Gliotransmitters in Astrocytes

Astrocytes are now recognized as “excitable” cells When astrocytes are activated

by internal or external signals, they communicate with neighboring cells in the form

of gliotransmission Astrocytes release various transmitters and factors such asglutamate, GABA, acetylcholine, noradrenaline,D-serine, ATP, nitric oxide, andbrain-derived neurotrophic factor (BDNF) (reviewed by Volterra and Meldolesi

2005) In concert with the release of transmitters, many different receptors forneurotransmitters are expressed on the astrocyte cell membrane These receptorsrespond with a particular form of excitability involving Ca2+oscillations (Porterand McCarthy1997)

Modulation of neuronal excitability and synaptic transmission by astrocytes wasfirst shown to be mediated by glutamate release (Haydon2001) With astrocytesproviding local neuronal excitation via glutamate, they provide a source of neuronalactivation that may be critical in controlling the synchronous depolarization ofneurons (Fellin et al.2006) At the same time, astrocytes can also suppress synaptictransmission by releasing purines Through these coordinated actions, the astrocyte

is thought to provide balanced excitation and inhibition mediated by two distincttransmitter systems

Astrocyte excitation, which is chemically encoded, can be detected tally by assays of Ca2+transients and oscillations Two main forms of astrocyteexcitation are well-documented: one that is generated by chemical signals inneuronal circuits (neuron-dependent excitation) and one that occurs independently

experimen-of neuronal input (spontaneous excitation) Numerous studies highlight the release

of glutamate from astrocytes in response to neuronal activity In the case ofglutamate, synaptic-like glutamatergic microvesicles have been identified inastrocytes and these vesicles are released via Ca2+-dependent exocytosis (Bezzi

et al.2004)

2.3.3 Astrocytes Modulate Synapse Development and Function

Various insect and vertebrate animal models indicate that glial cells and neuronsfunction together to guide axons during development (Chotard and Salecker2004).When axons reach their target, glial cells contribute to the specification of theappropriate synaptic connections The importance of the neuron–astrocyte interac-tion in synaptic development and function has been highlighted in several papers(Haydon 2001; Ullian et al 2001; Slezak and Pfrieger 2003; Schipke andKettenmann 2004) Astrocytes secrete diffusible factors, such as cholesterol(Mauch et al.2001), tumor necrosis factor-a (Beattie et al.2002), activity-dependentneurotrophic factor (Blondel et al 2000), and thrombospondins–extracellularmatrix glycoproteins (Ullian et al.2004; Christopherson et al.2005) to promotesynapse formation Other classes of cell adhesion molecules such as theg-protocadherins, a family of neuronal adhesion molecules that are critical for

synaptogenesis, are expressed by astrocytes (Garrett and Weiner 2009) Directastrocytic contacts also upregulate synapse formation in a protein kinase C-dependentmanner (Hama et al.2004).

Astrocyte contacts may induce local structural and functional modifications ofdendritic segments or individual synapses Membrane-bound ligands on astrocytes,such as ephrin-A3, have been shown to regulate spine morphology in the hippo-campus (Murai et al.2003), suggesting local activation of EphA receptors on spines

by astrocytic ephrin-A3 Using organotypic hippocampal slice preparations, Haber

et al (2006) showed that astrocytes can rapidly extend and retract fine processes toengage and disengage from postsynaptic dendritic spines These dynamic structuralchanges in astrocytes possibly control the degree of neuron–glia communication atthe synapse With two-photon time-lapse imaging methodology (Nishida andOkabe2007), they revealed that astrocyte motility in the form of protrusive activityacts as a key local regulator for stabilization of individual dendritic protrusions andsubsequent maturation into spines

2.3.4 The Neurovascular Unit

Considering the contacts made between astrocytes and blood vessels, it has beenestimated that in excess of 99% of the brain vasculature is ensheathed by astrocyticprocesess (Takano et al.2006) This active interaction between the neuron, astro-cyte, and blood vessel has been termed the neurovascular unit and is essential forthe regulation of blood flow (Takano et al 2006; Koehler et al 2009) Theimportance of regulating blood volume in the brain is highlighted by the fact thatthe brain consumes approximately 20% of the energy produced by the body at rest.The control of blood glucose and O2 are tightly controlled by neurotransmittermediated signaling (predominantly by glutamate) and this control is modulated byastrocytes (see review by Attwell et al.2010) The increase in glia research andevolution of the importance of astrocytes to normal neuronal and vasculaturefunction is also highlighted by numerous reviews (Attwell et al.2010; Freeman

2010; Pfeiffer and Huber2010; Eroglu and Barres2010; Barker and Ullian2010)

2.4 Astrocytes in Neurological Disorders

With an evident role of astrocytes in normal neural function at all cellular andmolecular levels, it is not surprising that astrocytes have been implicated in virtuallyall pathological conditions in the nervous system Dysregulated astrocyte functionhas been linked with the progressive pathology of stroke and to a number ofneurodegenerative diseases including Alzheimer’s disease, Huntington’s disease,and Parkinson’s disease (Maragakis and Rothstein 2006) While a comprehensivereview of astrocytes in the various pathologies is beyond the scope of this chapter, theinvolvement of astrocytes in the development of Rett syndrome (RTT) is very

applicable Recently, Ballas and colleagues (2009) demonstrated that astrocytes andastrocyte-conditioned media from the RTT mouse model failed to support normaldendritic morphology Taken together with our findings in Fragile X (discussed in thenext section), and the consistent synaptic alterations seen in Fragile X, learningimpairments and autism spectrum disorders, the possibility of an astrocyte involve-ment in multiple childhood neurodevelopmental disorders certainly becomes evident.

2.5 The Fragile X Astrocyte

With overall synaptic function standing as a prominent link to the expression of thedisease phenotype in a number of neurodevelopmental disorders, and knowing thatastrocytes influence synapse development and function, our lab initiatedexperiments to evaluate the role of astrocytes in Fragile X neurobiology Theseexperiments were preceded by the observation that astrocytes, in addition toneurons, also express the Fragile X Mental retardation Protein (FMRP) (Paceyand Doering2007) At the time of this finding, FMRP expression in the brain wasconsidered to be primarily neuronal FMRP had been reported in oligodendrocyteprecursor cells, but not mature oligodendrocytes by Wang et al (2004) Whenstudying stem and progenitor cells from the brains of wildtype and knockout Fragile

X mice, approximately 50% of the cells in culture coexpressed FMRP and GFAP.Parallel immunocytochemical studies in vivo also showed the coexpression ofFMRP and GFAP in the embryonic and adult developing hippocampus

With the identification of FMRP in astrocytes and knowledge of their role insynaptogenesis, we initiated experiments to explore neuronal development andsynapse formation in the Fragile X mouse A coculture design was used to selec-tively combine cells from theFmr1 KO mouse and its wild-type (WT) counterpart(Jacobs and Doering 2009) With this tissue culture approach, neurons andastrocytes were independently isolated to explore four different combinations ofneuronal-astrocyte cultures (WT neurons + WT astrocytes, WT neurons +Fmr1

KO astrocytes,Fmr1 KO neurons + WT astrocytes and Fmr1 KO neurons + Fmr1

KO astrocytes) The cells were grown for 7, 14, or 21 days and then processed forimmunocytochemistry to analyze the morphological and synaptic profiles.The first set of experiments focused on neurons in each of the four combinations,cultured for 7 days (Jacobs and Doering2010) The neurons were studied with anantibody directed against microtubule-associated protein-2, (MAP2; a dendriticmarker) and the pre- and postsynaptic proteins, synaptophysin and postsynapticdensity protein-95 (PSD-95), respectively The WT neurons grown on theFmr1 KOastrocytes had significantly altered dendritic arbor morphologies, with a shifttoward a more compact and highly branched dendritic tree These neurons alsodisplayed a significant reduction in the number of pre- and postsynaptic proteinaggregates However, when the Fmr1 KO neurons were cultured with the WTastrocytes, the alterations in dendritic morphology and synaptic protein expressionwere prevented In fact, their morphological characteristics and synaptic protein

expression approached the appearance of the normal neurons grown with WTastrocytes These experiments were the first to suggest that astrocytes contribute

to the abnormal dendritic morphology and the dysregulated synapse developmentseen in Fragile X syndrome

In the next phase of this research, we wanted to determine if these alteredcharacteristics represented a developmental delay imparted by the Fmr1 KOastrocytes (Jacobs et al 2010) Focusing on WT neurons grown in the presence

of WT orFmr1 KO astrocytes, we evaluated the dendritic arbor morphology andsynaptic protein expression at 7, 14, and 21 days in culture If we considered thedevelopmental pattern of the WT neurons on the WT astrocytes to reflect a normalpattern of dendrite and synaptic protein development, we found a significantalteration to these patterns when WT neurons were grown with Fmr1 KOastrocytes Our results revealed that the WT neurons grown with Fmr1 KOastrocytes displayed significantly altered morphological and synaptic proteinprofiles at 7 days (when compared to the WT condition); however, by 21 days inculture these differences were no longer significantly different from normal On thebasis of this research at this time, it appears that the astrocytes in the Fragile Xmouse may contribute to the altered characteristics of neurons seen in Fragile Xsyndrome, in a developmentally regulated manner

In preliminary studies to examine if neuronal subsets are preferentially affected,

we performed Sholl analyses on the morphology of the neurons in the experimentsdescribed above Our findings suggest that there is a bias in the extent of themorphological alterations imparted by the astrocytes to a subset of neurons with astellate dendritic arbor morphology (unpublished results) However, it should benoted that in these experiments, the astrocyte involvement was assessed indepen-dently of the alterations that would be observed due to a lack of Fmr1 (andtherefore, FMRP) in the Fragile X neurons themselves Therefore, the situation

in vivo, having both neurons and astrocytes affected by a lack of FMRP may nottruly reflect the experimental results in vitro Additional experiments with a rigor-ous method of identifying subtypes of neurons (e.g., excitatory versus inhibitoryneuronal markers) should be performed to specifically address this possibility.These studies create numerous new avenues to identify and detail the role ofastrocytes in the morphological alterations of neurons seen in Fragile X.Establishing key aspects to altered molecular relationships between astrocytesand neurons in Fragile X will lead to new therapeutic possibilities (Fig.2.2) Arethe alterations due to a lack of FMRP in the astrocytes or are the astrocytesabnormal because they develop and function in a diseased microenvironment? Ifthe absence of FMRP in the astrocytes is the primary source of dysfunction, how arethese effects translated to the neurons? For example, is the astrocyte–neuronsignaling disrupted due to a lack of astrocyte-FMRP? How, where, and when dothese signals act? Is the abnormal astrocyte–neuron communication mitigated by amembrane associated or a soluble factor? Finally, can these abnormalities observed

in vitro be studied in vivo? These, and many other questions about the Fragile Xastrocyte are now important targets for Fragile X research – the answers important

in gaining a full understanding of the underlying neurobiology that contributes to

Fig 2.2 The role of FMRP in astrocytes in Fragile X syndrome (a) Historically, FMRP has only been associated with neurons (i) With FMRP present there is regulated protein synthesis, normal dendritic spine morphology and no abnormalities associated with Fragile X syndrome (FXS) (ii)

In Fragile X (in humans and in mouse models) there is a lack of FMRP in neurons leading to the dysregulation of synaptic protein synthesis, abnormal dendritic spine morphologies and features associated with FXS (iii) Recent studies indicate that FMRP is present in both neurons AND astrocytes (iv) In the Fmr1 KO mouse (and presumably in FXS) FMRP is absent from both neurons AND astrocytes, and the astrocyte FMRP plays an important role in shaping the neuron morphology and synaptic protein profiles (b) It is now important to investigate the role of astrocytes in Fragile X (i) FMRP may play a similar role in astrocytes as in neurons, functioning

as a regulator of protein translation (ii) There are a number of possibilities for how the lack of FMRP in astrocytes may contribute to the abnormal neurobiology of FXS (1) The astrocytes may

be abnormal as a consequence of developing in an abnormal environment (and therefore not due to

a direct effect of astrocyte-FMRP) (2) The neuron–glia signaling may be altered as a result of dysregulated FMRP-dependent protein synthesis, which in turn could alter astrocyte function (again, not due directly to astrocyte FMRP) (3) The translation of a subset of glial proteins may

be dysregulated in the absence of astrocyte-FMRP (4) The glia–neuron signaling may be disrupted due to an abnormal glial signaling protein profile (membrane bound or secreted) as a result of a lack of astrocyte-FMRP *Presence of astrocyte but not a key player Figure # Biomedical Illuminations, 2011

the morphological phenotype seen in Fragile X, and in the potential of a futuretreatment for individuals with Fragile X syndrome.

2.6 Astrocyte Research in the Future

With each year passing, neuroscience research continues to unfold aspects ofastrocyte involvement in health and disease Each new molecular and cellularfinding builds into the extensive functioning of how glial cells control and modifyneuronal structure and communication

Subtle changes in the connectivity patterns within subsets of neurons maysignificantly alter the output of the neuronal circuitry Interestingly, mutations inthe synaptic proteins neurexin 1 and neuroligins 3 and 4 are associated with autismspectrum disorders and mental impairment (Sudhof2008) The postsynaptic scaf-folding molecule and interacting protein of neuroligin SHANK3 (ProSAP2) is alsoassociated with autism (Durand et al 2007) Accumulating evidence illustratesroles for FMRP in synapse development and corresponding alterations in synapticmolecules in Fragile X (Pfeiffer and Huber2009) In fact, synaptic function andstructure may be the converging point of malfunction in many neurodevelopmentaldisorders such as Fragile X, RTT, and autism (Walsh et al.2008; Geshwind2008).Together, the last three decades have created a more complete image of synapticdevelopment and function both in health and in diseases of neurological dysfunc-tion – one that is highly dependent on the glial cells of the CNS Keystone papers byPfrieger and Barres (1997), Ullian et al (2001), Christopherson et al (2005), andothers revealed that astrocytes play a major role in the modulation of the develop-ment and functioning of synapses Given the recent findings of astrocyte involve-ment in neurodevelopmental disorders such as RTT and FXS, it is realistic to nowconsider astrocytes as holding the key to avenues of intervention for learningdisabilities that we previously did not appreciate

Since many aspects of CNS development involve a neuron–glial interaction,solving neurological dysfunction will require solutions that include glial cells aspart of the picture To maintain a healthy microenvironment for neurons, it will beimportant to continue research efforts that target our understanding of howastrocytes interface with neuronal circuitry at the cellular and molecular levels.Modes of pharmacological therapy should indeed concentrate on the health of theastrocyte With astrocytes as “gatekeepers” of neuronal health and function, if wecan target astrocytes, then they may in turn take care of the neurons