« Role of Nestin in Mouse Development »

« Paria Mohseni»

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy (Ph.D.) Department of Molecular Genetics University of Toronto

« Paria Mohseni »

« Doctor of Philosophy » « Department of Molecular Genetics» University of Toronto

« 2010 »

Abstract

Although nestin has served as a marker of neural stem/progenitor cells for close to twenty years, its function is still poorly understood. During development, this intermediate filament protein is expressed in many different progenitors including those of the central nervous system, heart, skeletal muscle and kidney. The adult expression of nestin is mainly restricted to the subependymal zone and dentate gyrus of the brain, the neuromuscular junction and renal podocytes. I have used two approaches of gain of function and loss of function to elucidate the role of nestin in vivo. Although I was able to generate transgenic lines in which the transgene was ubiquitously expressed at the

RNA level, over-expression of nestin at the protein level was not achieved possibly due to post transcriptional regulation of this gene. My data from loss of function approach indicates that nestin-deficient mice have impaired coordination. Balance and muscle strength are not affected and there are no apparent anatomical defects. I found that nestin deficiency is compatible with normal development of the central nervous system but results in abnormal clustering of acetylcholine receptors in the neuromuscular junctions,

similar to the phenotype described for deficiency of cyclin-dependent kinase 5 (Cdk5) a candidate downstream effector of nestin. In renal podocytes, where both nestin and Cdk5 are normally expressed, we found reduced branching and abnormally contoured podocyte processes. To further connect the phenotype of nestin deficiency to Cdk5, I demonstrated that nestin deficiency can rescue maintenance of acetylcholine receptor clusters in the absence of agrin, similar to Cdk5/agrin double knockouts, indicating that the observed nestin deficiency phenotypes are the consequence of aberrant Cdk5 activity.

iii

Acknowledgments

I would like to thank first and foremost my supervisor Dr. Andras Nagy for his support and encouragement during my Ph.D. During my time in the Nagy lab not only I was able to advance my scientific knowledge but also by being part of a diverse team, I learned many valuable life lessons for which I will always be grateful. I also would like to thank my supervisory committee members Drs. Janet Rossant and Derek van der Kooy for their valuable scientific remarks and guidance throughout my Ph.D. I wish to thank all members of the Nagy lab for their support and helpful discussions.

During my Ph.D., I had the opportunity to visit and briefly work in three other laboratories. My sincere thanks to Drs. Derek van der Kooy, Anna Wobus and Hao Ding for letting me visit their labs and teaching me valuable techniques which were critical for my project.

A special thank you goes to my brother, Piroz. Without your support I would not have been able to reach this and many other milestones of my life. I would like to thank my mom for her understanding and endless support. Thanks for always listening, encouraging and providing guidance. I also would like to thank my dad for his support. A very special thank you goes to my late grandmother (maman khoobi) for always supporting me and encouraging me to continue my education.

The biggest thank you goes to my husband Mehrdad for always listening, understanding and encouraging. Without your support I could not achieve this. Thanks for driving me to downtown during the weekends so I could take care of my cells or mice. Thanks for picking me up at odd times from the lab so I didn’t have to go to Union Station in the middle of the night all by myself. We did it! I’m graduating! ☺

Table of Contents

ABSTRACT………………………………………………………………………………ii ACKNOWLEDGEMENTS………………………………………………………..…….iv TABLE OF CONTENTS…………………………………………………………...... …..v

CHAPTER 1: GENERAL INTRODUCTION AND LITERATURE REVIEW………………………...…………………………………………………...…..1 1. Intermediate filaments…………………………………………………………..….2 2. Nestin……………………………….………………………………………………8 2.1. History, structure and expression during embryonic development…………..8 2.2. Expression of nestin in the adult mouse………………………………….....10 2.3. Expression of nestin after damage or injury………………..……………….12 2.4. The nestin locus………………………..……………………………………13 2.5. Function(s) of nestin…………….…………………………………………..14 3. Cyclin-dependent kinase 5 and agrin……...……..…...…………………………..17 4. Hypothesis…………..…………………………………...………………………..21

CHAPTER 2: METHODS AND MATERIAL………………….……………………23 1. Nestin over-expression construct and ES cells ………………………………..….24 2. LacZ staining………………….……………………………………………..……25 3. RNA isolation and RT-PCR……………………………...……………………….25 4. Immunocytochemistry……………...……………………………………………..26 5. Neurosphere assay……………………………………………..………………….26 6. In vitro differentiation …………………………...………….……………………27 7. Construction of the targeting vector to generate nestin knockout mice……...…...27 8. Generation of the Nes+/- ES cells and Nes-/- mice…………………………...…….28 9. Western blot and immunohistochemistry ………………………………...………29 10. MRI………...………………………………………………………..…………..30 11. Neuro-behavior tests …………………………………………………………….32 12. Electron microscopy ……………………………………………...……………..33 13. Generation of nestin/desmin and nestin/vimentin double knockout mice…...….33

CHAPTER 3: GAIN OF FUNCTION APPROACH: NESTIN OVER- EXPRESSION…………..……………………………………………………………35

1. Introduction ……………………...……….………………………………………36 2. Results……...…………………….………………………………………….……38 2.1. Generation of nestin over-expressing embryonic stem (ES) cell lines ….....38 2.2. Forced-expression of nestin in vivo ………………………………………...40 2.3. Forced-expression of nestin in vitro ………………..………………………44 2.4. Neurosphere assay …………………..……………………………………...46 3. Discussion………..…………………………………………………………...…..47

CHAPTER 4: LOSS OF FUNCTION APPROACH: NESTIN KNOCK-OUT……51 1. Introduction ...………………...…………………...…………………………..….52 2. Results………………...………….……………………………………….………54 2.1. Generation of nestin-deficient mice………………...…………………..…..54 2.2. Nestin is dispensable for development of the mouse…………...…...……...58 2.3. Nestin deficiency alters renal podocyte structure………..………………….61 2.4. Nestin is required for proper peripheral motor function…………...….……64 2.5. Broadening of AChR endplate band and increased number of AChR clusters in NMJ of Nes-/- mice ………………………………………….…….....……………….64 2.6. Nestin regulates activity of Cdk5 in vivo: evidence from nestin/agrin double knockout mice………………………….………………………………………………..66 3. Discussion……...………...…………………………………………...…………..68

CHAPTER 5: BEING A TEAM PLAYER………………………….………………..74 1. Introduction ……………………………………………...………..…………..….75 2. Results…………………………….………...………………………….…………76 2.1. Generation and characterization of nestin/vimentin double knockout mice..77 2.2. Generation and characterization of nestin/desmin double knockout mice….78 3. Discussion……...…………………………………………………...………...…..80

CHAPTER 6: CONCLUSION AND FUTURE DIRECTIONS……………………..82

Chapter 1

GENERAL INTRODUCTION AND LITERATURE REVIEW

1 Intermediate Filaments

In eukaryotes the cyctoskeleton consists of three major filament structures: microfilaments, microtubules and intermediate filaments (IF) [1]. The main structural element of microtubules is tubulin which can polymerize and make a thick filament structure of 25 nm. Actins make the microfilament network with relatively lower thickness of 6-7 nm. Thickness of intermediate filaments falls in between that of microtubules and microfilaments (8-12 nm) and that is why they are called “intermediate” filaments.

Despite their diversity, all members of the IF family share certain structural features. They all have a characteristic central rod domain of about 310 amino acids capable of forming α-helical coiled-coil dimers. The rod domain is relatively conserved among all IF proteins. In addition, all IF proteins also have a head domain or N-terminus and the tail domain or C-terminus [2]. Both head and tail domains are non-helical and exhibit huge variations in terms of size and amino acid composition[3]. Previous studies have shown that the head domain is essential for assembly of IF subunits into protofilaments and ultimately the filamentous network [4]. The role of the tail domain of IFs, is still not fully understood [3].

Unlike microfilaments and microtubules whose main structural elements are limited to actins and tubulins respectively, IFs can be formed from one or several different subunits. IFs of vertebrates have been divided into six classes based on their gene structure as well as their expression pattern[5]. Nestin is the only member of the sixth class of intermediate filaments. Recently it has been proposed that due to evolutionary and structural similarities, it should be categorized in the class IV of IFs [6]. Table 1 provides examples of IFs for each category.

Filament class Protein members (examples) Mainly expressed in Type I • Acidic keratins • Epithelial cells (e.g. keratin 9, 10, 12 and 18) Type II • Basic keratins • Epithelial cells (e.g. keratin 1, 2, 5 and 8) Type III • Vimentin • Mesenchymal cells

• Desmin • Muscle

• Glial Fibrillary Acidic Protein • Astrocytes (GFAP) • Neurons • Peripherin Type IV • Neurofilaments • Neurons (NF-L, NF-M, NF-H)

• Alpha-internexin • Early neurons

• Nestin (?) • Neural and muscle progenitors Type V • Lamins • Nuclear membrane Type VI (?) • Nestin (?) • Neural and muscle progenitors

Table 1. Classification of intermediate filament proteins.

Although IF proteins are structurally similar at the level of base units, once assembled each has unique characteristics. For example keratins which are mostly found in epidermis do not readily dissociate even in presence of strong solvents such as 9M urea [7]. On the other hand, lamins which line the nuclear membrane of the cell readily dissociate and reform through the cell cycle [8],[9,10]. Since IFs are expressed in tissue- specific manner, they also exhibit specialized functions depending on the site of expression. For example, neurofilaments which are abundantly expressed in differentiated neurons are important for radial growth of axons [11,12] and keratin 8-18

which are found in hepatocytes are essential for protecting liver against drug-induced damage [13].

Microtubules and microfilaments are essential for structure and function of the cell and are referred to as housekeeping elements. In contrast, cells can survive and function in the absence of intermediate filaments without exhibiting any major abnormalities both in vivo and in vitro (e.g. vimentin, glial fibrillary acidic protein (GFAP) and α-internexin). Although many cytoplasmic intermediate filaments are abundantly expressed during mouse embryogenesis, except in one case (keratin), targeted deletion of IFs does not lead to embryonic lethality. Keratins are the first intermediate filaments expressed during mouse development. Expression of keratin 8 and keratin 18 is detected as early as eight-cell stage [14,15,16]. Mice deficient in expression of keratin 8 die during embryonic development but display different phenotypes depending on their genetic backgrounds. On C57/BL6 background 98% of mutants die at E12.5 due to liver defects. However when the background is switched to FVB/N, embryonic lethality decreases to about 50% and mice exhibit colorectal hyperplasia [17]. Mice deficient of keratin 18, whose expression is detected at the same time as keratin 8, do not exhibit any early embryonic abnormalities [18]. One could argue that in the absence of keratin 18, its putative function is compensated by other intermediate filaments.

Most knockout models of cytoplasmic intermediate filaments are viable and fertile and do not exhibit any obvious abnormalities. Vimentin is a cytoplasmic intermediate filament protein predominantly expressed in mesenchymal cells in vivo. In vitro vimentin is expressed in almost any cell type grown in culture [19]. To elucidate the function of this IF protein two independent strains of mice were generated via gene targeting. Mice deficient in expression of vimentin are viable, develop and reproduce normally [20]. Detailed analysis confirmed absence of vimentin protein in these knockout strains. Moreover it was shown that absence of vimentin is not compensated by expression of other IF proteins [20]. Considering the complex expression pattern of vimentin during embryonic development and in adults, it was surprising that elimination of this protein from cytoskeleton did not result in any major abnormalities. The authors

however could not exclude existence of subtle defects or putative functions for vimentin under pathological conditions [20].

GFAP is another cytoplasmic intermediate filament whose expression is detected in cells of astroglial lineage [21]. This IF protein is abundantly expressed in mature astrocytes making it the main component of the IF network in these cells. To address the role of GFAP in vivo, a few groups generated independent GFAP knockout mouse lines via gene targeting. All groups reported that mice lacking GFAP could develop and reproduce normally without exhibiting any major abnormalities [22,23,24]. Moreover, no sign of compensation by other intermediate filaments was observed. In one study long- term potentiation in CA1 region of hippocampus was examined and GFAP mutants displayed enhanced long-term potentiation compared with the control group [23]. In another study, effect of aging on GFAP mutants was further examined [24]. After 18 months, some GFAP knockout mice had hydrocephalus associated with loss of white matter. Furthermore, the blood-brain barrier was impaired in GFAP mutants exhibiting more permeability compared with the control group [24].

Another example of attempts at revealing the role of intermediate filaments in vivo, involved α-internexin. This cytoplasmic IF protein is expressed in young neurons during development as well as in some mature neurons in adults. Mice carrying a null mutation in α-internexin gene develop and reproduce normally without any obvious abnormalities [25]. Detailed examination of neurons did not reveal any defects in axonal caliber or migration of newly formed neurons [25].

So considering the abundant and complex expression pattern of intermediate filaments during development and restricted expression in certain cell types in adults, why does elimination of IF proteins lead to very subtle or in some cases no defects at all? What is the function of IF proteins? Some IF proteins especially the ones expressed in cytoplasm are thought to be important for protecting the cell against mechanical stress. Transgenic and gene knockout studies targeting keratins, vimentin and desmin provided

evidence in support of this notion showing that tissues lacking proper IF network are more susceptible when exposed to physical stress [20,26,27].

Keratins are expressed in the skin which through everyday activities is exposed to various physical and mechanical stresses. Cells with mutated form of keratin show extreme fragility. For example mutations in keratin 5 or keratin 14 cause a disorder called epidermolysis bullosa simplex (EBS) [28,29,30,31]. In this disease when skin is exposed to mechanical trauma, keratinocytes of the epidermal basal-cell compartment fracture easily leading to accumulation of the intra-epidermal fluid and formation of blisters. Detailed analysis revealed that mutated forms of keratin 5 or keratin 14 are abnormally distributed inside of the cell and do not form the filamentous network that is observed in the wild-type cells. Indeed the mutated keratins are compacted and in some severe cases the non-filamentous keratins form aggregates that accumulate in the cell [28,30].

Another example comes from studies using desmin null mice. Desmin is abundantly expressed in mature muscle cells [32,33,34]. Gene targeting studies revealed that desmin knockout mice develop normally and are fertile [35]. However when examined under challenging conditions, desmin deficient mice performed significantly worse than their wild-type littermates. Swimming exercise led to mortality of 50% of the mutants tested while all the wild-types successfully completed the test [36,37]. Desmin- deficient mice also performed poorly on treadmill test. The maximum running speed of desmin knockout mice was significantly lower than the wild-types. Moreover in contrast to the wild-types, desmin-null mice were not able to complete the prolonged running tests and would stop at a much earlier time-point compared to the controls [38].

As mentioned before, vimentin knockout mice are also viable and fertile. However the process of wound healing is impaired in both adult mice and embryos deficient in expression of vimentin [39] supporting the hypothesis regarding the putative role of vimentin under pathological conditions. Additional studies revealed that adaptive vascular response is also impaired in vimentin-deficient mice [40]. Vimentin is highly

expressed in renal vessels but under normal physiological conditions the function of kidneys in vimentin-null mice is not affected. Reduction of renal mass is a pathological condition under which adaptive vasodilation is essential for survival of the organism. Renal mass reduction leads to 100% mortality of mice lacking vimentin within 72 hours while all the wild-type littermates are able to survive [40]. These observations indicate that vimentin is essential for modulation of vascular tone when mice are exposed to stress.

It has been proposed that there is some degree of functional overlap between certain intermediate filaments. Results from experiments on vimentin and GFAP knockout mice support the above hypothesis. Although vimentin/GFAP double knockout mice are viable and fertile and do not exhibit any gross abnormalities (similar to vimentin and GFAP single knockouts), their response to CNS injury is greatly compromised [41]. In response to CNS damages such as spinal cord lesions or stab wounds in the brain, glial scars are made in the wild-type mice. Formation of glial scars in vimentin and GFAP single knockouts is not affected, however vimentin/GFAP double knockouts exhibit extensive bleeding at the site of the injury leading to mortality of 25% of the mutants shortly after the injury [41]. Detailed analysis on the double mutants who survived the injury revealed that the density of glial scars was much less and often accompanied by bleeding. Such defects were observed in neither GFAP nor vimentin single knockouts suggesting that there is some functional overlap between these two intermediate filament proteins in the context of glial scar formation.

Emerging evidence suggests that IF proteins participate in signal transduction events [42,43,44] and act as modulators of protein kinases [6]. Studies aiming at deciphering the mechanisms which regulate phosphorylation of vimentin revealed that RhoA-binding kinase α (ROKα) is a highly specific kinase for vimentin. Once RhoA (the activator of ROKα) is activated, the vimentin network collapses as a result of being highly phosphorylated by ROKα. Consequently ROKα is released from its scaffold (vimentin) and translocated to the cell periphery [45]. Vimentin can also regulate signaling events indirectly through interaction with intracellular proteins such as 14-3-3.

Previous studies have shown that cell division cycle (cdc) 25 is essential for progression of cell cycle and its activity is required at mitotic entry for activation of cyclin-dependent kinase 1 (Cdk1). At other phases of cell cycle, cdc25 is kept inactive through association with 14-3-3 [46]. Vimentin can bind to 14-3-3 in a phosphorylation dependent manner [47]. Through this interaction it limits ability of 14-3-3 to associate with other proteins. Therefore vimentin can indirectly modulate signaling events of the cell.

1 Nestin

1.1 History, structure and expression pattern during embryonic development

Nestin was first identified in 1985 in Dr. Ronald McKay’s lab [48]. In order to better understand embryonic development of mammalian nervous system, the group aimed at generating markers that would recognize major cells types in the developing nervous system. As a result, monoclonal antibodies were generated against the spinal cord of E15 rats. One of these antibodies, Rat-401, recognizes a transient population of cells that resemble radial glial cells. Further analysis showed that Rat-401 staining appears as early as E11 right after neural tube is closed [49]. Moreover, Rat-401 does not recognize cells of adult brain as vimentin and GFAP do (later studies showed that this antibody can recognize neural stem/progenitor cells residing in the adult brain [50,51,52]). Detailed characterization of Rat-401 and expression pattern of its antigen confirmed that indeed it can recognize neural stem/progenitors of the developing nervous system. It was also shown that Rat-401’s antigen is an intermediate filament protein [49]. Hence the name “nestin” was chosen (NEST: NeuroEpithelial STem cells, IN: name of most of the intermediate filaments ends with IN) [49]. Nestin has a very short head domain and therefore is unable to form a filamentous network by itself [49,53] . It is often co-expressed with class III IF proteins such as vimentin and desmin and through

formation of heterodimers it can integrate into the network of IFs. During early embryonic development cells are nestin negative. Expression of nestin coincides with start of neurulation. As neural differentiation reaches its terminal steps, nestin’s expression is down regulated. In fully differentiated cells of nervous system, expression of nestin is replaced by other intermediate filaments such as GFAP and neurofilaments [49].

Outside of the developing nervous system, staining of Rat-401 was observed in developing muscle cells but mature muscle cells did not express the antigen for Rat-401 [48]. Detailed analysis of intermediate filament expression during muscle development revealed that vimentin is expressed early in myotubes and as cells differentiate expression of vimentin is down regulated and replaced by desmin in mature muscle cells. Interestingly nestin can form a network that is indistinguishable from that of vimentin and desmin [54]. Nestin follows a complex transient expression pattern during muscle development. It can copolymerize with vimentin in early steps of muscle differentiation and as cells mature, nestin/vimentin network disassembles and heterodimers of nestin and desmin are formed. Once muscle progenitors are fully differentiated, expression of nestin is no longer detected and desmin is the main intermediate filament present in the multinuclear muscle cells [54].

Nestin is also co-expressed with desmin in the developing heart [55]. Expression of desmin is detected in the embryonic heart at E8 [56]. One day later (E9), nestin appears in the cardiomyocytes. In the developing heart, nestin and desmin are co- expressed until E10.5 after which nestin’s expression is down regulated and desmin becomes the main IF of the heart [55].

Further analysis revealed expression of nestin in several additional sites during embryonic development. Examples of such sites are limb buds [57], hair follicles [58], eyes [59] and dental lamina [60]. Development of the limb buds start at E10 in the mouse embryo. At E11 nestin is expressed ubiquitously in mesenchymal cells of the limb bud [57]. Later during development expression of nestin is down regulated in the

mesenchymal cells that undergo chondrogenesis but remains high in the cells that give rise to myoblasts [57]. Expression of nestin in hair follicles follows a complex pattern as the cycle of hair growth progresses. During early growth phase or anagen, nestin expressing cells are detected in the permanent upper portion of hair follicle or the bulge area surrounding the hair shaft [58]. As the hair growth cycle progresses (mid and late growth phase) nestin positive cells proliferate and not only are present in the bulge but also are detected in the upper outer-root sheath [58]. It is important to note that nestin positive cells in the hair follicle also express keratin 5/8 and keratin 15, all of which are markers for hair follicle progenitor cells [61]. Expression of nestin in the developing eye also changes as the embryo grows. Nestin protein can be detected as early as E9.5 in the optic stalk of the mouse embryo [59]. Later during development (E12.5) nestin is detected in the lens epithelium as well as the optic stalk. At E17.5 high level of nestin expression is observed in the optic disk, optic nerve and lens epithelium and in adult mice nestin is only detected in the optic nerve [59].

1.2 Expression of nestin in the adult mouse

In the adult brain, nestin is detected in the subependymal cells of the lateral ventricle and in the subgranular cells of the dentate gyrus. Both of these two locations are known to contain neural stem/progenitor cells [62,63,64]. Moreover expression of nestin has also been detected in the rostral migratory system in cells that do not exhibit neuronal morphology suggesting that nestin is only expressed in neural progenitors [65]. Nestin is not expressed in areas of the brain which contain fully differentiated cells. Detailed analysis revealed that indeed expression of nestin can mark the boundary of areas in which neural progenitors reside (i.e. nestin is exclusively expressed in the subgranular cells of dentate gyrus but not in the granular cell layer which contain differentiated neurons [65]).

Neural stem cells can form neurospheres when cultured in vitro [63,65]. Neural stem cells grown in cultures are able to self renew and give rise to all major cell types of the central nervous system (CNS) (i.e. neurons, astrocytes and oligodendrocytes) when plated on the appropriate substrate. Neurospheres made from adult forebrain cells express high levels of nestin [66]. Moreover expression of nestin declines rapidly as neurospheres are plated on laminin and induced to differentiate. Potential of nestin positive cells to form neurospheres was compared to that of nestin negative cells using transgenic mice. In these transgenic mice nestin expressing cells are marked by green fluorescent protein (GFP). Subventricular zone of early postnatal brain was dissected and GFP positive cells were selected using a fluorescence-activated cell sorter. GFP positive cells and GFP negative cells were plated at the same density in separate plates. GFP positive cells gave rise to 1440 fold more neuropsheres compared with GFP negative cells. This data indicates that nestin positive cells residing in the subventricular zone are the main source of neurosphere forming cells in vitro [66].

Outside of the nervous system, expression of nestin has also been reported in mouse kidneys [67]. In the developing kidneys nestin is first detected at E12.5 in the cells surrounding the ureter. Later during development at E15.5 nestin is highly expressed in the developing as well as mature glomeruli in the cortex of the kidney. Expression of nestin in glomeruli becomes restricted to mature podocytes in the adult kidney [67]. Similar pattern of expression is observed in podocytes grown in vitro and nestin is clearly detected in the immature as well as fully differentiated cultured podocytes [67].

Expression of nestin has been reported in the muscle of adult mice and rats as well [68,69]. Unlike the wide expression during embryonic development, expression of nestin in adult muscle is very restricted and only detected at neuromuscular junctions (NMJ) and neuron-tendon junctions [69]. Perhaps that is the reason that presence of nestin in adult muscle was missed in earlier studies [48]. Immunohistochemical analysis revealed that nestin positive cells are localized right underneath the NMJ. Neuromuscular junctions were visualized using botulinum toxin, a molecule which specifically binds to

Acetylcholine receptors (AChR) of NMJ [69]. In situ hybridization revealed that mRNA of nestin is exclusively synthesized by the synaptic myonuclei in mature myofibers [68]. These observations indicate that nestin is selectively expressed at NMJ and neuron- tendon junction in the adult muscle.

1.3 Expression of nestin after damage or injury

Injuries to CNS lead to formation of glial scars at the site of injury. Glial scars are made from astrocytes which undergo transformation. Under physiological conditions astrocytes of CNS express GFAP. Upon formation of reactive astrocytes (glial scars), GFAP expression is up-regulated. Moreover nestin and vimentin are re-expressed in cells forming the scars [70]. Expression of nestin in response to an injury occurs rapidly (i.e. 48 hours after damage is made to spinal cord) [70]. Since nestin is normally not expressed in the adult CNS (with the exclusion of its restricted expression in the subventricular zone and subgranular zone of the dentate gyrus), induction of nestin is used as a marker to detect formation of reactive astrocytes and facilitates further analysis [70,71]. Expression of nestin following the injury to CNS lasts for a long time. In one study where injuries to the spinal cord were studied, nestin’s expression was detected up to 13 months following the injury [70]. In another study where cortex stab wounds were utilized to induce formation of reactive astrocytes, expression of nestin lasted for 28 days [70]. In addition to detection of reactive astrocytes, expression of nestin can also be used to track proliferating neural progenitors following an injury to the hippocampus [72]. Although under normal physiological conditions (uninjured brain) subgranular cells of dentate gyrus express nestin, following an injury the number of nestin expressing progenitors expands rapidly [72]. Subsequently these proliferating progenitor cells migrate out of the subgranular layer and differentiate. Although the majority of neural progenitors of hippocampus reside in the subgranular layer, a small population of nestin positive cells resides in the granular layer. These cells are relatively quiescent under normal physiological conditions but start proliferating in response to an injury and can

give rise to neurons and astrocytes [72]. It is not clear whether this population of nestin positive cells is able to form neurospheres when cultured in vitro.

1.4 The nestin locus

In the mouse (Mus musculus) nestin gene is located on chromosome 3 and is comprised of five exons (Figure 1). Promoter analysis revealed that nestin’s minimal promoter is located in a GC rich area between -227 and -33 bp, 5’ of the translation initiation site (ATG) [73]. Although there is a TATA-like box in the promoter area of nestin, removing it does not affect transcription efficiency of the gene, suggesting that it is not a functional TATA box [73]. Expression of nestin is cell type specific and developmentally regulated. Studies aiming at defining nestin’s regulatory elements showed that promoter region of this gene does not direct its tissue-specific expression [74]. Transgenic mice with a reporter gene under the control of a 5.8 kb fragment of nestin’s 5’ upstream region exhibited a weak and patchy expression pattern at E11. This observation led to the notion that there are regulatory elements other than the promoter which control the expression pattern of this gene. Studies aiming at identifying these regulatory elements revealed that indeed there are two independent enhancers in the first and second introns of nestin [74]. Expression of a reporter gene derived by combination of nestin’s promoter and its second intron was observed shortly after closure of neural tube at E9 in transgenic mice. Moreover the reporter gene followed the same expression pattern as the endogenous nestin gene in the CNS throughout the mouse development [74]. These results indicate that enhancers located in the second intron of nestin are responsible for regulating its expression in the CNS. Similar studies using the first intron of nestin revealed that enhancers there are responsible for myotome-specific expression of this gene. Interestingly, replacement of the endogenous promoter with a 160 bp herpes simplex virus thymidine kinase promoter did not change expression pattern of the reporter gene driven by the first or second intron further confirming that nestin’s promoter does not direct its tissue-specific expression [74]. Detailed analysis of the

second intron revealed that a 257 basepair enhancer located at the 3’ of this intron is highly conserved in mouse, rat and human [75]. Moreover this region is sufficient to regulate expression of a reporter gene throughout the mouse CNS development to produce the same expression pattern as the endogenous nestin gene [75]. Since the discovery of nestin’s regulatory elements, these enhancers especially the one active in the developing CNS (located in the second intron) have been widely used to generate transgenic mice. In these mice nestin’s enhancer is used to control the expression of enhanced green fluorescent protein (nestin-GFP) [66,76], doxycycline dependent reverse tetracycline transactivator (nestin-rtTA) [77] and Cre recombinase (nestin-cre) [78,79,80]. Using these mice one can delete (nestin-cre) or induce expression of a gene (nestin-rtTA) specifically in neural progenitors or track and select genes that are expressed in progenitors of CNS (nestin-GFP).

Figure 1. Structure of the mouse nestin gene. Nestin is located on the forward strand of chromosome 3 (87,775,015-87,784,369). The start codon (ATG) is located at 87,775, 015 and the stop codon (TAG) is located at 87,783,952. The length of the transcript is 6,122 base-pairs which is translated to form a 1,864 amino acid protein. (Source: www.ensembl.org)

1.5 Function(s) of nestin

Recent studies have shown that nestin can regulate cell signaling events through interaction with cyclin-dependent kinase 5 (Cdk5) [81,82]. Cdk5 is able to phosphorylate various components of the cell including cytoskeletal proteins such as microtubule- associated protein tau and neurofilaments [83,84,85,86,87]. In cultured neural progenitors (such as ST15A cells) Cdk5 and nestin are physically associated together and can be pulled down as a complex via coimmunoprecipitation [81]. Series of biochemical experiments revealed that Cdk5 is able to phosphorylate nestin on both serine and

threonine residues. Detailed analysis identified two sites of Thr-316 and Thr-1495 as the major phosphorylation sites of mouse nestin by Cdk5 [81]. Interestingly although nestin and vimentin are both expressed in neural progenitors and they both contain consensus sites for kinases such as Cdk5, vimentin is not phosphorylated by Cdk5. In contrast vimentin seems to be highly phosphorylated by another kinase protein cdc2, while nestin seems to be a poor substrate for this kinase [81,88]. Phosphorylation of nestin by Cdk5 leads to reorganization of the filamentous network made by this intermediate filament. When Cdk5 is over-expressed in neural progenitors in vitro, nestin’s sub-localization changes from being spread throughout the cells (from nucleus to the plasma membrane) to a concentrated mass around the nucleus [81]. Although localization of the network is altered, nestin filaments around the nucleus are still detected. The effect seems to be even more dramatic in muscle progenitors. When Cdk5 is over-expressed in myoblasts, the entire filamentous network made by nestin disassembles. In these cells nestin is detected as spots throughout the cell indicating that it is no longer engaged in a filamentous network [81]. Interestingly over-expression of Cdk5 in both cases (i.e. neural progenitors and muscle progenitors) did not affect the filamentous network of vimentin. Moreover the tubulin filaments and the actin filaments remained unaffected suggesting that the observed phenotype is not a general cytoskletal reorganization event but a nestin exclusive case [81].

As precursor cells go through the differentiation process, the intermediate filaments present in the cells also change. In the case of myogenic precursors, nestin is only expressed in undifferentiated muscle cells and once cells are fully differentiated expression of nestin is down regulated and replaced by desmin. In differentiating myoblasts as cells move toward terminal stages of differentiation, activity of Cdk5 increases which subsequently leads to phosphorylation of nestin [81]. Highly phosphorylated nestin network disassembles as its solubility increases and filamentous network of desmin starts forming. Therefore one could speculate that phosphorylation of nestin by Cdk5 during muscle differentiation is a required event for transition of IF network from nestin to desmin and maybe essential for muscle formation.

Interaction of nestin and Cdk5 has also been studied in the context of oxidant- induced cells death in neural progenitors [82]. Cells treated with hydrogen peroxide (an oxidative stress inducing agent) rapidly down regulated expression of nestin. Detailed analysis revealed that depletion of nestin was not a caspase dependent event as nestin does not posses any consensus sites for caspase cleavage. Nestin depletion in these cells was a proteasome dependent event, as treatment of cells with a proteasome inhibitor can prevent degradation of nestin [82]. Oxidative stress eventually leads to cellular death. An important question is that which one is the key event here? a) Down regulation of nestin leading to sensitization of cells to oxidative stress or b) Oxidative stress leading to cellular death and down regulation of nestin being the secondary event.

Nestin knock-down studies using RNA interference showed that down regulation of nestin in neural progenitors make them significantly more susceptible to oxidative stress compared with the wild-type cells [82]. Moreover this sensitization can be reversed by introducing nestin cDNA to the nestin depleted cells. Interestingly activity of Cdk5 also increases upon exposure of cells to oxidative stress. The increase in the level of

Cdk5 activity is even higher when nestin knock-down cells are exposed to H2O2 but down regulation of nestin by itself (in the absence of oxidative stress) does not have a significant effect on Cdk5 activity. Theses observations suggest that nestin has a protective role in neural progenitors and is able to perform this task through interaction with Cdk5. Previous studies have shown that changes in subcellular localization of Cdk5 and its activator, p35 can lead to cellular death [89]. Indeed when Cdk5/p35 complex is localized in the cytoplasm, it has a protective role but cleavage of p35 to p25 and localization of Cdk5/p25 complex in the nucleus has toxic effects [89]. Further analysis of nestin knock-down cells revealed that upon exposure to H2O2, p35 is cleaved to p25 and consequently p25 is accumulated in the nucleus leading to apoptosis. Rescue experiment in which nestin cDNA is introduced back into nestin knock-down cells proved that presence of nestin prevents formation of p25 and maintains the Cdk5/p35 complex in the cytoplasm [82]. Therefore nestin is able to regulate kinase activity of Cdk5 by forming a scaffold for this signaling molecule and stabilizing its cytoplasmic location.

Although Cdk5 is also able to phosphorylate nestin and this can cause disassembly of the nestin network [81], in the case of oxidative stress and neural progenitors it is not clear how Cdk5 can control the turn-over of its scaffold (i.e. nestin).

2 Cyclin-dependent kinase 5 (Cdk5) and agrin

Cyclin-dependent kinases are a large family of proline-directed kinases involved in cell cycle events [90]. Cdk5 was originally identified and isolated through its close homology to cdc2 [86,91]. Later studies showed that unlike other Cdks, Cdk5 is mostly expressed in post-mitotic cells [92] and is not involved in cell-cycle control. Moreover distinct from other Cdks, Cdk5 is not activated by cyclins but its activation is regulated through binding to specific regulatory subunits such as p35 and p39 [92]. During mouse embryonic development expression of Cdk5 and its kinase activity increases as neural cells differentiate and the forebrain develops. In adults high level of Cdk5 expression and activity has been reported in the brain [93]. Studies on primary cortical neurons isolated from E17-18 rats revealed a novel function for Cdk5 activity [93]. These primary cortical neurons were used as a model to study maturation of neurons in vitro. High level of Cdk5 activity was detected at growth cones of migrating neurons. Decreasing the kinase activity of Cdk5 by introducing two independent dominant-negative Cdk5 mutants had a negative effect on neurite outgrowth [93]. This phenotype can be rescued by introducing the wild-type Cdk5 protein back to the cultured cells further supporting the role of Cdk5 in neurite outgrowth and migration [94].

To assess the role of Cdk5 in vivo, mice deficient in Cdk5 kinase activity were generated by gene targeting [94]. Cdk5 deficiency leads to perinatal lethality of the mutants. More than 60% of the mutants die in utero after E16.5 and the newborn pups die within 12 hours after birth [94]. The newborn mutants exhibit severe mobility defects in all their limbs and do not respond to clamp stimulation [95]. Detailed analysis revealed

abnormal corticogenesis in Cdk5 mutants. Cdk5 deficient mice lack proper cortical laminar structure due to defective neuronal migration [96]. To delineate the role of Cdk5 in adults, Cdk5 conditional knockout mice were generated by flanking the Cdk5 gene with loxP sites [96]. These conditional mutants did not exhibit any abnormalities. Removal of the flanked area was achieved by crossing the conditional mutant mice with transgenic Cre mice in which the Cre activity is driven by the heavy neurofilament (H- NF) promoter. In most neurons of these double mutants; Cdk5 activity is not detectable after E16.5 [96]. Detailed analysis revealed that these mice exhibit neural migration defects in cerebral and cerebellar cortex, in both of which neuronal migration continues through the perinatal period.

The neuronal specific activator of Cdk5, p35 is expressed in high levels in migratory neurons [97]. Knockout mouse model of p35 are viable but exhibit severe cortical lamination defects [98]. In wild-type mice, layers of neurons are formed in an inside-out manner. In contrast, p35 deficient mice display a reverse pattern such that early generated neurons occupy the superficial layers (outside) and neurons formed later stay in the deep layers (inside) [98]. Although this phenotype resembles the phenotype of Cdk5 knockout mice (both display migrations defects of neurons), it is less severe and does not lead to perinatal lethality of the mutants. This observation suggested that besides p35, there are other activators for Cdk5.

Later studies led to identification of another activator for Cdk5, named p39. This new regulator was isolated through its close homology to p35 [99] and was shown to be the activator of Cdk5. Mice deficient in expression of p39 do not show any detectable abnormalities possibly due to compensation by p35 [100] as p35 is a more robust activator of Cdk5. Compound mutant mice (p35/p39 double knockouts) exhibit severe neuronal migration defects and perinatal lethality. A phenotype that is identical to that of Cdk5 knockout mice [100]. These observations indicate that p35 and p39 are the most important activators of Cdk5 if not the only ones.

Most of our knowledge about Cdk5 and its functions comes from studies focused on CNS development and migration of neurons as the most obvious defects were observed in these sites in Cdk5 deficient mice. Recent studies have revealed novel roles for Cdk5 outside of CNS; examples of which include renal podocytes [101] and neuromuscular junction (NMJ) [102].

Podocyte are specialized epithelial cells that cover the renal glomeruli [37]. These terminally differentiated cells are important for regulation of glomeruli function and proper renal filtration. Injury to podocytes often causes proteinuria [103]. During mouse development expression of Cdk5 is detected in maturing podocytes. Moreover p35 is also co-expressed with Cdk5 in the developing podocytes [101]. High levels of Cdk5 and p35 expression can also be detected in mature podocytes in adults. Furthermore, cultured podocytes express Cdk5 and p35 in vitro and kinase assays have shown that the resulting complex of Cdk5 and p35 is in its active form. Inhibition of Cdk5 activity either pharmacologically or through siRNA resulted in dramatic morphological changes in the growing podocytes in vitro [37]. Podocytes with reduced Cdk5 activity are smaller and lose their typical processes [101]. Structure of podocytes in Cdk5 knockout mice has not been studied in detail. Routine histological analysis of Cdk5 deficient embryos did not reveal any obvious defects [94] but defects in podocyte morphology can not be excluded as to assess such defects more sensitive assays such as electron microscopy are required. In addition, since Cdk5 knockout mice are perinatal lethal, analysis of the structure of podocytes in adult Cdk5 knockouts requires crossing of conditional knockout mice with a Cre transgenic mice in which expression of Cre is driven by a podocyte specific promoter. Such an experiment will shed light on the role of Cdk5 activity in regards to morphology of mature podocytes in the adult mice.

Another site outside of the brain in which activity of Cdk5 has been studied is the NMJ. Both Cdk5 and p35 are highly expressed at the NMJ and immunohistochemical studies revealed that they are colocalized with AChRs on the postsynaptic muscle [102]. These observations suggest that Cdk5 activity is important for proper formation of the NMJ. To further examine the role of Cdk5 in NMJ development, Cdk5 knockout

embryos were used. Diaphragm muscle is commonly used to study formation of synapses because it is easily accessible [104]. Cdk5 knockout embryos display various NMJ defects both at pre and post-synaptic sites [105]. At the pre-synaptic site, axons of Cdk5 knockout embryos extend much further across the diaphragm in comparison with their wild-type littermates and display abnormal branching patterns [105]. At the post-synaptic site, the number and size of AChRs are greatly enhanced in Cdk5 knockout embryos [105]. How can Cdk5 activity affect clustering of AChRs at NMJ? Recent studies have shown that Cdk5 regulates clustering of AChRs through interaction with well-established factors that are essential for NMJ development [106].

One of the well-studied factors essential for NMJ development is agrin. Agrin is a proteoglycan [107]. Addition of agrin to muscle cells grown in culture induces formation of AChR clusters through redistribution of the diffusely distributed AChRs [108]. Therefore agrin is able to form aggregates of AChRs and that is why the name “agrin” was chosen for it. During development, different isoforms of agrin are made. They are synthesized by both nerves [109] and muscle cells [110]. However agrin purified from neurons is much more active than the one isolated from muscle cells [111]. What is the role of agrin in NMJ development? To answer this question, agrin deficient mice were generated [112]. Agrin deficiency leads to perinatal lethality due to severe pre and post synaptic defects [112]. The number and size of AChR clusters are greatly reduced in agrin mutants at E18.5. Moreover these embryos exhibit nerve branching abnormalities at pre-synaptic sites [112]. To find out which isoform of agrin is responsible for regulation of AChR clustering, isoform-specific agrin mutants were generated [113]. The phenotype of nerve-specific agrin was the same as the phenotype of the whole agrin knockout. On the other hand, deficiency of muscle-specific agrin did not cause any detectable defects in clustering of AChRs [113].

Later studies revealed that AChR clusters do form in agrin knockout embryos but are dispersed later during development [114]. This observation gave rise to a new hypothesis; namely, that agrin is essential for maintaining the AChR clusters but another factor is initiating the formation of the clusters. Indeed later studies revealed muscle-

specific kinase (MuSK) to be the factor required for initiating the postsynaptic differentiation events [115]. Therefore it is postulated that once the initiation step is over, maintenance of AChR clusters is achieved through complex interplay of positive and negative signals. Positive signals stabilize the innervated clusters and negative signals eliminate the ones that are not in contact with the nerve terminals [116]. Agrin secreted from the nerve leads to stabilization of innervated AChR clusters and therefore is considered a “positive” signal. But what are the negative signals responsible for dispersion of non-innervated clusters? To answer this question, one group focused on examining the interaction of Cdk5 and agrin at NMJ. The phenotype of Cdk5 knockout embryos at the NMJ is the opposite of what is observed in agrin deficient mice. In vitro analysis revealed that upon removal of agrin, Cdk5 is able to disperse AChR clusters that were previously aggregated in presence of agrin [106]. Blocking the activity of Cdk5 in cultured myotubes leads to maintenance of a significant number of AChR clusters that otherwise would have been dispersed [106]. To further study the interplay of Cdk5 and agrin, double mutant mice (agrin/Cdk5) were generated and AChR clusters were analyzed at E18.5. The number of stabilized AChR clusters is significantly higher in agrin/Cdk5 double knockouts compared with those in agrin single knockouts [106]. Therefore Cdk5 deficiency can partially rescue the post-synaptic defects observed in agrin knockout mice. These results suggest that Cdk5 is a “negative” signal which is responsible for dispersing the AChR clusters that are not innervated and stabilized by agrin.

3 Hypothesis

I hypothesize that nestin is essential for normal development of the mouse.

To test this hypothesis I have used two approaches of gain and loss function: i. Gain of function: studying the effects of nestin over-expression from a transgene.

ii. Loss of function: investigating the consequences of nestin deficiency through generation of knockout lines.

In chapter three of this thesis, experiments and the outcomes of the gain of function approach are explained. In chapter four and five, experiments and results of the loss of function approach are described.

Chapter 2

MATERIALS AND METHODS

1 Nestin over-expression construct and ES cells

In order to make a conditional over-expression vector, nestin cDNA was purchased from Open Biosystems (Alabama, USA) and ligated with Clip vector [117]. Briefly the Clip vector contains the following elements: A) Chicken β-actin promoter with a CMV enhancer for expression of βgeo. B) βgeo gene whose expression makes the cells Neomycin resistant and LacZ positive. Neomycin is used as a selectable marker and LacZ expression serves as a tool for monitoring the expression level of the transgene, which in this case is nestin. C) Three poly A signals 3’ of βgeo to ensure the expression of this gene is terminated. D) Two loxP sites that flank the βgeo-pA cassette, which upon activation of Cre recombinase, will remove the intervening cassette .E) mouse nestin complementary DNA (cDNA) F) Internal ribosomal entry site (IRES) Puromycin resistance cassette. The preceding IRES sequence allows the expression of Puromycin from a bicistronic mRNA that also encodes for nestin.

The Clip vector was cut with NotI and XhoI and ligated with the nestin cDNA. Consequently the nestin/clip vector was linearized by SfiI (2 hours incubation at 50C). The linearized vector was then prepared for electroporation by ethanol precipitation. Briefly ethanol (2.5 X the volume of plasmid) and 3M sodium acetate (the volume of the plasmid /10, pH=5.2) were added to the plasmid solution and incubated in dry ice/methanol bath for 15 minutes. The tubes were then centrifuged at maximum speed for 15 minutes. Consequently the pellets were washed with 70% ethanol and air dried.

The conditional expression vector containing nestin cDNA was electroporated into ES cells. The step-by-step protocols described by Nagy et al. [118] were followed for electroporation of the construct and selection of the transgenic ES clones. The transfected cells were selected based on neo resistance. The selected colonies were then

picked and grown in 96-well plates. DNA was extracted from these 96-well plates while the replica plates were frozen for future experiments. The extracted DNA was digested with AflII for Southern blot. The sequence of neo was used as a probe in Southern blot to detect the copy number of inserted vectors for each clone. Clones with one copy of the construct were selected for further analysis.

2 LacZ staining

In order to assess the level of βgeo expression, LacZ staining was used. The cells or embryos were washed twice with phosphate buffered saline (PBS) without calcium and magnesium at room temperature. Consequently they were fixed with 0.2% glutaraldehyde for 2 minutes at room temperature. After two washes with PBS the cells/embryos were stained at room temperature overnight using the following mixture: 1- 1 ml of X-gal (20 mg/ml)

2- 500 ul of K4Fe(CN)6O3H2O* (100mM)

3- 500 ul of K3Fe(CN)6** (100mM)

4- 20 ul of MgCl2 (100mM) 5- Bring the volume to 10 ml with PBS The next day cells/embryos were washed with PBS and fixed with 10% formalin.

• *K4Fe(CN)6O3H2O: Potassium Ferrocyanide

• **K3Fe(CN)6: Potassium Ferricyanide

3 RNA isolation and RT-PCR

In order to examine the expression of the transgene, RNA was extracted from cells or embryos. The samples were homogenized using a fine needle (22G) in 0.75 ml Trizol and incubated at room temperature for 5 minutes. Consequently 0.2 ml of

chloroform was added to the tubes and shaken for 15 seconds. After 10 minute incubation at room temperature, samples were centrifuged at 9000 rpm for 15 minutes (4◦C). Next the aqueous phase was transferred to a new tube and 0.5 ml of isopropanol was added to it. After 10 minute incubation at room temperature, the samples were centrifuged at 9000 rpm for 10 minutes (4◦C). Consequently the RNA pellet was washed with 70% ethanol and re-suspended in Diethylpyrocarbonate (DEPC) water.

Following RNA isolation, cDNA was made using the SuperScript® III First- Strand Synthesis System (Invitrogen). The following primers were used to detect the expression of nestin transgene: F: 5’CTGACTGACCGCGTTACTCC3’ R: 5’GACGCAACCCTCCATGTC3’ and a 300 base-pair band was obtained.

4 Immunocytochemistry

Cells were washed with PBS and fixed with 4% paraformaldehyde (PFA) for 10 minutes at room temperature. Cells were then washed with PBS three times for 5 minutes each. Consequently the cells were premeabilized using 0.3% TritonX-100/PBS solution (10 minutes at room temperature). Cells were then washed three times with PBS. Blocking was achieved by incubation with 5% goat serum solution for 1 hour at room temperature. Consequently the primary antibody was added without washing the cells. The cells were incubated at 4C overnight. The next day, cells were washed three times with PBS and secondary antibody was added to the cells and incubated for two hours at room temperature. Following three washes with PBS, cells were mounted using VECTASHIELD® Mounting Media (Vector Laboratories).

5 Neurosphere assay

Neurospheres were made from ES cells as previously described [119]. Briefly ES cells were trypsinized and single cells were grown in serum free media [120] containing the following elements: DME/F-12 (Invitrogen) with 0.6% D-glucose, 3 mM NaHCO3, 2 mM glutamine, 20 nM progesterone , 25 µg/ml insulin, 100 µg/ml transferrin, 60 µM putrescine, 5 mM Hepes and 30 nM sodium selenite. To the serum free media 1000U/ml of LIF was added and cells were grown in low adhesion plates (Nunclon). After 7 days, spheres were analyzed. Only the ones larger than 75µm-diam were counted as neurospheres.

6 In vitro differentiation

In vitro differentiation of ES cells was carried out as previously described [121]. Briefly ES colonies were trypsinized and the dilution was adjusted to 600 cells per 20 ul. Hanging drops were made on the lid of bacterial dishes and grown for 2 days in ES media without LIF. After 2 days, the small spheres were transferred to bacterial dishes and grown for 3 days in the same media. Consequently the spheres (EBs) were transferred to gelatin coated plates and grown in Iscove media supplemented with 20% fetal calf serum. In order to be able to quantify the differentiation capacity of the cells, one EB was placed in each well of a gelatinized 24-well plate. Wells were monitored daily and cells were fed every other day.

7 Construction of the targeting vector to generate nestin knockout mice

To target the nestin locus, two homology arms were generated using standard recombineering technique [122]. The bacterial artificial chromosome (BAC) used in recombineering was chosen from the RPCI-23 mouse BAC library (female C57/BL6)

and contained the entire nestin gene. This BAC was obtained from the centre for applied genomics at SickKids hospital (Toronto). The following primers were used to generate the arms:

1-F: 5’-agagccgcggtcttgttggtcctcttcctca-3’ 1-R: 5’-agagcctaggccatggcccctggaaaaggagttaca-3’ 2-F: 5’-agagcctagggggtactggcacaggcatttaatc-3’ 2-R: 5’-agaggcggccgccttaagattttccgaaagaggcttgagatac-3’ 3-F: 5’-agagaagcttatgcatagagtcgcttagaggtgcagcag-3’ 3-R: 5’-agagagatctccatggctaggaggtctcagaattccatcc-3’ 4-F: 5’-agagagatctcatatggtggaggatggagaaggtca-3’ 4-R: 5’-agaggtcgacctcgagttccagtccagctc-3’

A plasmid containing a PGK promoter, Neomycin resistance gene and pA (PNT, [123] was cut with NotI and EcoRI. The pBSKII plasmid was also cut with NotI and EcoRI. The fragments were then ligated together. The 5’ homology arm and the PNT- pBSKII product were then cut with NotI and SacII and ligated together. Finally, the 3’ homology arm and the 5’arm-PNT-pBSKII vector were cut with HindIII and SalI and the fragments were joined to generate the final targeting vector. The vector was linearized with SalI before electroporation.

8 Generation of the Nes+/- ES cells and Nes-/- mice

The linearized targeting vector was electroporated into G4 ES cells as previously described [117]. Briefly, the ES cells were grown on mitomycin-treated mouse embryonic fibroblasts in DMEM (Invitrogen), supplemented with 15% FBS (HyClone), 2 mM l-glutamine, 0.1 mM nonessential amino acids, 0.1 mM 2- mercaptophenol 1 mM sodium pyruvate and 2,000 units/ml leukemia inhibitory factor.

Approximately 5 × 106 cells were mixed with 25–35 μg of linearized targeting vector in

electroporation buffer (Specialty Media) and electroporated at 250 V and 500 μF using a Bio-Rad Gene Pulser. The cells were then plated in 100-mm plate. The following day, G418 (166 μg/ml; Invitrogen) was added to the media to select for neomycin-resistant colonies. The colonies were picked 8-12 days later and transferred to 96-well plates. The cells in 96-well plate were passaged. Subsequently one replica plate was frozen and the other used for DNA isolation for Southern blot analysis. A 5’ external probe was amplified using the following primers: F-acctagaggcctgagattctctaaa, R- caagatttttgatgaggaagaagaa. DNA was digested with SpeI and Southern blot was done according to a previously described protocol [117]. The expected band size for wild-type and targeted allele was 20 and 14 kb respectively. Once targeted clones were identified, the corresponding clones in the 96-well plates were thawed and expanded for generation of chimeras via aggregation [124]. Briefly, two-cell stage embryos (E1.5) from superovulated ICR females were electrofused using a CF-150B Pulse Generator (BLS).

The fused embryos were then cultured overnight at 37°C in 5% CO2 in KSOM medium (Specialty Media). On the day before aggregation, ES cells were trypsinized and plated on gelatinized dishes at various densities. The following day ES cell clumps were collected by gentle trypsinization. Two tetraploid embryos at the four-cell stage were then aggregated with the clumps of 8–15 ES cells. Once born, the chimeras were allowed to reach sexual maturity and then were crossed with both outbred ICR and C57/BL6 females. The resultant progeny were genotyped and the heterozygous females were crossed with the chimeras (heterozygous males) to produce homozygous mice. All the animal studies were performed in accordance with the institutional guidelines.

9 Western blot and Immunohistochemistry

Protein for western blot was isolated from E10.5 embryos. Immediately after dissection, embryos were homogenized using a fine needle (30G) and protein was extracted as previously described [82]. Standard Western blot procedure was followed. The membrane was blocked in 2.5% milk diluted in TBS-T at 4C overnight. The next

day the membrane was incubated for 1.5 hours at room temperature in the following antibodies (i) nestin (Chemicon International); (ii) β-actin (Sigma).

Mouse tissues were pre-fixed by perfusion of 1% paraformaldehyde in PBS. The animals were anesthetized using ketamine/xylazine. The thoracic cavity of the mouse was opened and the left ventricle was given a small incision to insert the canula. The animals were first flushed with 50 ml PBS and then with 1% paraformaldehyde in PBS. Tissues from brain, heart, lung, liver, kidney and skeletal muscle were collected, fixed in 4% paraformaldehyde in PBS and embedded in paraffin for sectioning. Whole-mounted tissues (diaphragm) were collected without perfusing the animals. Fresh tissues were fixed with 1% paraformaldehyde in PBS at 4°C overnight. The next day, tissues were washed three times with 0.3% Triton X-100 in PBS (PBST) and incubated for 1 hour at room temperature with blocking solution containing 5% normal goat serum (Jackson ImmunoResearch) in PBST. Whole-mounted tissues were incubated overnight at 4°C with the following primary antibodies: (i) alpha-bungarotoxin (BTX) alexa 594 conjugate (Invitrogen); (ii) nestin, (Chemicon International); (iii) desmin (Epitomics). After several washes in PBST, tissues and sections were incubated for 2 hour at room temperature with secondary antibodies: Cy3 or FITC or Cy5- conjugated secondary antibodies were used for signal detection (1:500 dilution, Jackson ImmunoResearch). For control experiments, the primary antibody was omitted or replaced by pre-immune serum. Signals were visualized and digital images were obtained using a Zeiss LSM 510 confocal microscope equipped with two photon, argon and helium–neon lasers (Zeiss). Images were analyzed using ImageJ software to determine the area occupied by AChR clusters. The number of AChR clusters was determined by manual counting. Statistical significance was calculated by two-tailed student’s T-test. The number of samples used for characterization of AChR clusters in single knockouts was 5 for each group (Nes-/- and their WT littermates). The number of samples used for characterization of AChR clusters in double knockouts was 3 for each group (Nes-/-;Agr-/- and their Nes+/+;Agr-/- littermates).

10 MRI

The mice (5WT and 10 KO) were anesthetized with a combination of Ketamine (Pfizer, Kirkland, QC) (150 mg/kg) and Xylazine (10 mg/kg) via intraperitoneal injection. Thoracic cavities were exposed, and the animals were perfused through the left ventricle with 30 ml of phosphatebuffered saline (PBS) (pH 7.4) containing 2 mM ProHance® (gadoteridol, Bracco Diagnostics Inc., Princeton, NJ) contrast agent solution at room temperature (25°C) at a rate of approximately 1.0 ml/min. This was followed by infusion with 30 ml of iced 4% paraformaldehyde (PFA) in PBS containing 2 mM ProHance® at the same rate. Following perfusion, the heads were removed along with the skin, lower jaw, ears and the cartilaginous nose tip. The remaining skull structures were allowed to postfix in 4% PFA containing 2mM ProHance® at 4 °C for 12 h. The skulls were transferred to a PBS and 0.02% sodium azide and 2 mM ProHance® solution at 4 °C on a rotator platform until they were scanned.

A multi-channel 7.0 Tesla MRI scanner (Varian Inc., Palo Alto, CA) with a 6-cm inner bore diameter insert gradient set was used to acquire anatomical images of brains within skulls. Prior to imaging, the samples were removed from the contrast agent solution, blotted and placed into 13 mm diameter plastic tubes filled with a proton-free susceptibility-matching fluid (Fluorinert FC-77, 3M Corp., St. Paul, MN). Three custom- built, 14 mm diameter solenoid coils with a length of 18.3 mm and over wound ends were used to image three brains in parallel. Parameters used in the scans were optimized for image efficiency and grey/white matter contrast: a T2-weighted, 3D fast spin-echo sequence, with TR/TE=325/32 ms, four averages, field-of view 14×14×25 mm and matrix size=432×432×780 giving an image with 32 μm isotropic voxels. Total imaging time was 11.3 hours.

The obtained 32μm isotropic resolution T2-weighted MRI scans were non- linearly aligned to a 3D atlas of the mouse brain with 62 structures identified [125]. This process consisted of an initial step in which all of the MRI scans were non-linearly

aligned to each other using an unbiased groupwise registration algorithm [1]. Briefly, rigid body registration was carried out towards a preexisting image based on the same mouse strain. All possible pair-wise 12-parameter registrations were then carried out to create an unbiased linear average model of the entire data set. All images were subsequently non-linearly aligned towards the 12-parameter average. The resulting registered MRIs were resampled and averaged [1,126]. This iterative procedure was repeated for an additional five generations with ever finer deformation grid-point spacing. The end-result is to have all 15 (5 WT and 10 KO) scans deformed into exact alignment with each other in an unbiased fashion. This allows for the analysis of the deformations needed to take each mouse's anatomy into this final atlas space, the goal being to model how the deformation fields relate to genotype. Correspondence with the 3D atlas was obtained by non-linear alignment.

To reduce random noise and assure normality under the central limit theorem, the transformation data was blurred prior to analysis with a Gaussian kernel with a full width at half maximum of 1 mm, and the logarithm of the Jacobian was computed for univariate statistical comparison at every image point. This statistical analysis results in millions of separate statistical tests. In order to account for an inflated type I error, the False Discovery Rate (FDR) technique was applied. The volume for each anatomical structure defined in the atlas was computed for each mouse by integrating the Jacobian of the transformation mapping the atlas image to the image for that mouse. This procedure has previously been shown to provide volume estimates comparable to those obtained by standard stereological methods using tissue sections [127]. MRI and data analysis was performed by Christine Laliberte and Matthijs van Eede in Dr. Mark Henkelman’s lab at Mouse Imaging Centre of Hospital for Sick Children (Department of Medical Biophysics).

11 Neuro-behavior tests

Roto-rod test was used to examine the ability of mice to maintain their balance and coordinate their movement. Mice were tested using previously described methods [128]. Briefly, mice were placed on top of the beam facing away from the experimenter's view (4 mice tested at a time). After a brief training period, the Roto-rod starts accelerating gradually and latencies for the mice to fall from the rod were recorded. Each mouse was tested 3 times per day for 3 consecutive days (same time and location every day). To examine muscles, a grip strength test was used. The mouse was is lifted by the tail and allowed to grasp the steel grip. Then it was gently pulled backward until the grip was released. The machine records the force before the release. Each mouse was tested three times. To test the balance, mice were trained to walk on a long beam. The following day, mice were put on the same beam and the time needed to walk from start to end point was recorded.

12 Electron microscopy and quantification of podocyte gaps

For transmission electron microscopy, mouse kidneys were fixed in a mixture of 4% paraformaldehyde and 1% glutaraldehyde, postfixed in 1% osmium tetroxide, and embedded in Epon. Sections were cut at 50 nm, stained with uranyl acetate and lead citrate, and examined using a JEM1011 electron microscope (JEOL, Tokyo, Japan).

For scanning electron microscopy, kidneys were fixed as for transmission electron microscopy, dehydrated in graded ethanol, critical point dried, and sputter coated with gold. Specimens were examined using a JSM 820 scanning electron microscope.

In order to quantify the gaps between podocytes, SEM images of wild-type and knockout podocytes were aligned. Three lines were drawn randomly on the images and the distance between secondary processes (gaps) were measured manually. Two wild-

type samples and two knock-out samples were used for analysis. Statistical significance was determined using Student’s T-test.

13 Generation of nestin/desmin and nestin/vimentin double knockout mice

Generation and characterization of vimentin knockout mice were described previously [20]. These vimentin knockout mice obtained from The European Mouse Mutant Archive (EMMA) were crossed with nestin knockouts described in chapter 3 of this thesis. The resulting double heterozygous mice were subsequently intercrossed to obtain mice homozygous for both nestin and vimentin. In order to expedite generation of the animals needed for histological analysis, a different breeding strategy was used. Mice homozygous for nestin and heterozygous for vimentin (Nes -/-; Vim +/-) were intercrossed. The resulting double mutants were used as the experimental group and the litter-mates homozygous for nestin and wild-type for vimentin (Nes -/-; Vim+/+) were chosen as the control group. The following primers were used to genotype vimentin knockout mice:

Primer 1: 5’-TGTCCTCGTCCTCCTACCGC-3’ Primer 2: 5’-AGCTGCTCGAGCTCAGCCAGC-3’ Primer 3: 5’-CTGTTCGCCAGGCTCAAGGC-3’

Primers 1 and 2 would amplify a 398 base-pair wild-type band while primers 2 and 3 would generate a 530 base-pair mutant band.

Generation and characterization of desmin deficient mice were described previously. These mice were obtained from EMMA. Double knockout mice were generated by crossing nestin and desmin single knockouts and subsequently the intercross of nestin/desmin double heterozygous mice. The following primers were used to genotype desmin knockout mice:

Primer DES-F: 5'-TTGGGGTCGCTGCGGTCTAGCC-3' Primer DES-R: 5'-GGTCGTCTATCAGGTTGTCACG-3' Primer LacZ-R: 5'-GATCGATCTCGCCATACAGCGC-3'

All three primers were used in one single PCR reaction resulting in a 300 bp wild-type band and a 400 bp mutant band.

Chapter 3

GAIN OF FUNCTION APPROACH: NESTIN OVER-EXPRESSION

Contributors: Mouse chimeras were generated at the transgenic facility of Samuel Lunenfeld Research Institute.

1 Introduction

The intermediate filament nestin exhibits a complex expression pattern during embryonic development, which has been well studied specifically in the developing central nervous system. As a result, nestin has been used as a marker for detection of neural stem/progenitor cells for over twenty years [49,129,130]. The sites and pattern of nestin’s expression during development of the mouse were described in chapter 1. In summary, nestin’s expression has been detected during embryonic development in various progenitors including those of central nervous system and muscles [49,54]. Upon differentiation, nestin’s expression is down regulated and replaced by tissue-specific intermediate filaments such as desmin, α-internexin and neurofilaments [54,131].

Embryonic stem (ES) cells are pluripotent cells isolated from the blastocyst of early developing embryos. These cells can be grown in the appropriate culture media for a long period of time as pluripotent cells [132,133]. In addition upon exposure to the right factors, ES cells can be differentiated to various cell types in vitro [134,135,136]. One of the most common approaches used for differentiation of ES cells is formation of embryoid bodies (EBs). This method was first described in 1985 by Doestschman et al [135] as an efficient way of differentiating ES cells. In summary, cells are grown in suspension in the absence of leukemia inhibitory factor (LIF) either in a bacterial dish as mass culture [135] or hanging drops made on the lid of a culture dish [137,138,139] for a few days. As cells proliferate they form spheroid aggregates, loose their pluripotency and start differentiating. The cells in the outer layer of embryoid bodies differentiate to endoderm-like cells. After a few days an ectodermal rim is formed. Subsequently mesoderm specification occurs [140]. To allow complete differentiation of EBs, they are plated on adhesive substrates and grown for several more days. If no specific growth factor is added to the media, EBs differentiate spontaneously and can form cells of ectoderm, mesoderm and endoderm lineages [141]. Alternatively differentiation of EBs can be guided towards a specific cell type if the right culture condition and growth factors are used [140].

Nestin protein is expressed at low levels in mouse ES cells. Upon differentiation (i.e. in EBs and early EB outgrowths) expression of nestin is significantly elevated [121]. Once differentiation is complete and cells reach terminal stages of differentiation, expression of nestin is no longer detectable [121]. Up-regulation of nestin in differentiating EBs is not limited to any specific cell lineages. Nestin’s expression has been reported in EBs differentiating to neural cells (ectoderm), skeletal muscle cells (mesoderm) and hepatic cells (endoderm) covering all the three primary germ layers [121,142,143,144]. During the differentiation process transient co-expression of nestin was observed with lineage specific markers such as GFAP (astrocytes), albumin (hepatic cells) and desmin (muscle cells). These observations indicate that differentiation of various cell types is achieved through generation of a transient progenitor cell population which expresses nestin.

Although nestin is expressed in various progenitors in vivo and in vitro, the same principal expression pattern is observed in all cases. That is undifferentiated cells exhibit very low level of nestin expression but upon start of differentiation (in EBs as well as in various progenitor cells during embryonic development) expression of nestin is significantly up-regulated. It is not clear what role(s) nestin may play during the differentiation process.

Prompted by the sophisticated expression pattern of nestin during embryonic development as well as during in vitro differentiation and the recent notion that intermediate filaments are able to modulate cell signaling pathways, we decided to investigate the role of nestin in differentiation. We hypothesized that nestin is essential for normal development of the mouse. If this hypothesis is true, deregulated expression of nestin will cause embryonic defects in the mouse. We tested the above hypothesis via a gain of function approach by forcing the expression of nestin both in vivo and in vitro.

2 Results

2.1 Generation of nestin over-expressing ES cell lines

In order to force the expression of nestin in mouse ES cells, nestin cDNA was cloned into a Cre recombinase conditional vector (Figure 1A). In this vector, the pCAGGS promoter [145] which contains a CMV enhancer element, chicken β-actin promoter with the first exon (non-coding) plus part of intron 1 and rabbit β-globin gene, derives expression of the βgeo cassette. βgeo is a beta galactosidase (lacZ)- neomycin phosphotransferase (neo) fusion gene. Beta galactosidase activity serves as a tool for monitoring the expression level of the transgene, while neo is used as a drug selection marker. The βgeo cassette is flanked by two loxP sites [146,147]. As a result upon activation of Cre recombinase the intervening cassette is removed, putting the trasngene (in this case nestin cDNA) under the transcriptional control of the pCAGGS promoter. Puromycin gene (puro) is used as a secondary selection marker for cells with Cre- removed ßgeo. The internal ribosomal entry site (IRES) allows the expression of the puro gene from a bi-cistronic mRNA that also encodes for nestin.

The construct was electroporated into two different mouse ES cells, R1 (129 background) [124] and YV1 (F1:129xC57BL6 hybrid) [148,149,150] and several colonies were picked from each line for further investigation. Protocols used for genetic manipulation of ES cells were previously established in our laboratory and described in details by Nagy et al. [118].

Expression level of the βgeo gene was examined by staining the cells with X-gal. X-gal is an organic compound which can be cleaved by β-galactosidase to form two compounds of galactose and 5-bromo-4-chloro-3-hydroxindole. The latter is an insoluble blue product used for screening purposes. Here the blue color serves as a tool for determining significant features of a given random transgenic insertion (i.e. the level of

expression as well as its mosaic or overall nature). As expected different levels of LacZ expression were detected in the analyzed colonies (Figure 1B).

Figure 1. Over-expression of nestin. (A) In order to force the expression of nestin, its cDNA was cloned a conditional over-expression vector. Upon Cre recombinese activity the floxed βgeo cassette is removed and expression of nestin cDNA is achieved through activity of pCAGGS promoter. (B) The βgeo cassette in this construct serves as a marker to monitor the expression level of the randomly inserted transgene. Several colonies were obtained with various level of LacZ expression. Clones with strong and ubiquitous expression pattern were chosen for consequent analyses.

It is important to select the clones containing a single copy of the construct as it has been suggested that presence of multiple copies of loxP could cause chromosomal abnormalities [151]. Southern blot analysis was used to detect the copy number of the

inserted plasmids. Genomic DNA was isolated from the 96-well plates and diagnostic Southern blot was performed using the neo sequence as the probe (Figure 2). Using the combined results of LacZ staining and Southern blot, four clones were chosen for subsequent studies (three sub-clones from R1 and one sub-clone from YV1 ES lines). All of the selected clones contained a single copy of the construct and exhibited strong overall LacZ expression.

Figure 2. Southern blot analysis. The neo sequence was used as the Southern blot probe to detect the single-copy clones. Arrows point to the clones that carry one single copy of the construct randomly inserted into their genomes.

2.2 Forced-expression of nestin in vivo

In order to assess the effects of nestin forced-expression in vivo, three lines were selected for further characterization (two R1 sub-clones and one YV1 sub-clone). The two R1 sub-clones were used to generate chimeric founder mice by ES cell <-> diploid embryo complementation assay and the YV1 sub-clone was used for ES cell<-> tetraploid complementation assay [118]. Chimeras were allowed to reach sexual maturity and subsequently were crossed with wild-type out-bred females (ICR strain). All lines successfully transmitted the transgene to their offspring. More breeding pairs were set-up to assess the expression level of the transgene in vivo. The embryos were dissected at E9.5 and processed for whole-mount lacZ staining. Although all three clones exhibited strong and ubiquitous lacZ expression in vitro, three different levels of lacZ expression were observed in the embryos (Figure 3 A-C). In one case (R1-F4) the expression of lacZ was limited to the heart (Figure 3A). While the other two lines exhibited ubiquitous expression of lacZ at E9.5, the expression level was strong in one line (YV1, Figure 3C) and moderate in the other (R-B8, Figure 3B). We did not observe any obvious abnormalities in the examined embryos or the adults suggesting that in our transgenic lines, insertion of the transgene in the genome does not interfere with the normal function of the host genes.

A. B. C.

Figure 3. Lac-Z expression in nestin conditional over-expressing embryos.

(A-C) Although LacZ was expressed strongly and ubiquitously in ES cells, various levels of LacZ expression were observed in the developing embryos (E9.5, A:R1-F4, B:R1-B8, C:YV1).

To further study the effects of nestin forced-expression in vivo, two of the transgenic lines (R1-B8 and YV1) were crossed with pCAGGS-Cre transgenic females. In these transgenic mice, expression of Cre recombinase is under the control of pCAGGS promoter therefore it serves as ubiquitous Cre-deletor line [152]. Embryos were dissected at E14.5 and genotyped. As expected, 25% of the embryos were positive for both of the transgenes (nestin and pCAGGS-Cre). Removal of the βgeo cassette and expression of the nestin transgene was confirmed by PCR and RT-PCR.

By taking advantage of the unique structure of the pCAGGS promoter, RT-PCR primers were designed so that the forward primer would sit on exon-1 of the chicken β- actin gene and the reverse would recognize nestin cDNA. Using this strategy any DNA contamination can easily be recognized by presence of the 600 base-pair portion of the rabbit β-globin intron. Moreover expression of the transgene (nestin cDNA) via the pCAGGS promoter would result in a 300 base-pair fragment which can be easily amplified by RT-PCR (Figure 4A). Using this method, expression of the transgene was confirmed by RT-PCR (Figure 4B). We did not observe any abnormalities in the double transgenic embryos (forced-expressing nestin) compared with the controls (wild-type or single transgenic embryos conditionally over-expressing nestin). To verify the expression of nestin at protein level, western blot analysis was performed. Surprisingly we did not detect higher level of nestin expression in the transgenic lines compared with the controls (Figure 4 C-D). These observations suggest that although the transgene is being expressed in the analyzed embryos (judged by the level of lacZ expression and positive RT-PCR results), over-expression of nestin protein is not achieved at levels detectable by western blot analysis.

Further analysis of the adult double transgenic lines (forced-expressing nestin) revealed that these mice are viable, fertile and do not exhibit any major abnormalities.

We did not observe any significant differences in the body weight or size of the double transgenic lines when compared with the controls. Although we did not observe any abnormalities in our double transgenic lines, we can not conclude that forced-expression of nestin does not interfere with the normal development of the mouse as we could not detect significant over-expression of nestin at protein level.

C. B.

500 250 D.

Figure 4. Forced-expression of nestin cDNA at RNA level. (A) Unique structure of pCAGGS promoter allows easy detection of RNA generated by the transgene (nestin cDNA). (B) RT-PCR result of one of the analyzed transgenic litters is shown as an example. As expected, 25% of them (3 out of 12) were positive by RT-PCR as well as positive for both Cre and nestin cDNA.

(C) Western blot on wild-type and nestin forced-expressing embryos did not detect a significant difference. (D) β-actin was used as a control for western blot to show the equal amount of protein was loaded in both cases.

2.3 Forced-expression of nestin in vitro

Previous studies on intermediate filaments revealed that in some cases the complex interactions that happen in vivo, might mask the phenotype that is caused by deregulated expression of a certain intermediate filament. In those cases, in vitro studies can shed light on the function(s) of intermediate filament as experiments are carried out in a much simpler and more controllable environment. Studies on desmin null mice provide a good example for such a case [27,35]. Desmin null ES cells were not able to differentiate in vitro and form mature muscle cells. The same lines were used for production of chimeras and surprisingly desmin null mice were born with no obvious signs of abnormalities. It was concluded that desmin is not essential for differentiation of muscle cells in vivo [27].

Prompted by the idea that deregulated expression of nestin in vitro might result in a different phenotype than the one observed in vivo, we decided to further characterize our transgenic lines in vitro. In order to over-express nestin in vitro, the βgeo cassette was excised by Cre recombinase activation. Expression of Cre recombinase was achieved by electroporating a plasmid containing the Cre recombinase gene into the selected R1 sub-clones. The in vitro excision of the βgeo cassette was only performed on R1 sub- clones as the parental YV1 ES line contains the puromycin resistance gene. As a result it can not be used for secondary drug selection using puromycin. After introducing the Cre recombinase plasmid to the selected R1 sub-clones, several puromycin resistant clones were picked. Subsequently removal of the βgeo cassette was confirmed by lacZ staining. The white color indicated that the βgeo cassette was no longer active/present (Figure

5A). To further confirm the removal of the βgeo cassette and expression of the transgene, PCR and RT-PCR were used. As described before unique structure of the pCAGGS promoter allows easy detection of the RNA produced by the transgene (Figure 5B). To validate forced-expression of nestin at protein level, ES clones were stained with nestin antibody. Different levels of nestin expression were observed in the analyzed transgenic lines when compared with the controls (i.e. wild-type R1 ES cells) (Figure 5C-E).

C. D. E.

R1-F4 R1-B8

Figure 5. Forced-expression of nestin in vitro.

(A) Upon activity of Cre recombinase, the βgeo cassette is excised. Consequently the cells lose their LacZ expression. (B) Excision of the βgeo cassette and expression of the transgene is confirmed by RT-PCR. (C-E) ES clones were stained using nestin antibody (C: WT, D: R1-F4, E: R1-B8).

2.4 Neurosphere assay

In order to study the effects of nestin over-expression on the fate of ES cells, neurosphere assay was used. In this assay, ES cells are differentiated toward neural lineages. The differentiation process is not complete and cells are trapped at an intermediate stage forming floating spheres called neurosphere. Each neurosphere is made up of a combination of stem as well as progenitor cells.

We hypothesized if nestin plays a role in differentiation of ES cells towards neural lineages, the number of primary neurospheres produced by nestin forced- expressing cells should differ from that of wild-type controls. Two nestin transgenic lines (R1-B8 and R1-F4) along with two controls (i.e. wild-type R1 cells and R1-GFP expressing cells) were used in the neurosphere assay. The number of primary neurospheres made by nestin forced-expressing cells was not significantly different from that of wild-type cells (Figure 6). In order to measure significance the total number of neurospheres in the control group (WT and GFP) was compared to that of the experimental group (nestin over-expressing R1-F4 and R1-B8). Student’s T-test was used to calculate the p-value.

Figure 6. Neurosphere assay. No significant difference observed in the number of primary neurospheres generated from the transgenic lines (nestin over-expressing lines #1 and #2) compared with that of the wild-type or GFP over-expressing cells.

3 Discussion

The aim of this study was to assess the effects of nestin over-expression on embryonic development of the mouse. We utilized a gain of function approach in which nestin cDNA was forced-expressed in the selected transgenic lines. We reasoned that if nestin’s up-regulation leads to initiation of differentiation, one could expect that upon forced-expression of nestin, transgenic cells lose their pluripotent characteristics and exhibit signs of differentiation. Such events in the early embryo would interfere with the normal steps of embryonic development and may result in severe abnormalities or even embryonic lethality. The same phenomenon can also be studied in vitro as the transgenic ES cells may show altered differentiation capabilities when compared with the wild-type controls.

We used a conditional transgenic approach in which βgeo is expressed under the control of a general promoter and upon excision of loxP sites, the βgeo cassette is removed and nestin cDNA is placed under the control of the pCAGGS promoter. We were able to create ES lines in which the transgene is highly expressed as the ES colonies turned dark blue in presence of X-gal. Moreover we showed that upon excision of the βgeo cassette, the cDNA is being expressed both in vivo and in vitro as shown by RT- PCR. We however were not able to demonstrate over-expression of nestin at protein level as our western blot analysis failed to detect a significant difference in the amount of nestin protein generated in the trasngenic embryos compared with that of the controls. As a result, although our transgenic mice (after excision of the βgeo) develop and reproduce normally, we could not conclude that over-expression of nestin does not have any negative effects on the embryonic development of the mouse.

It was surprising that the amount of nestin protein in our transgenic mice was not significantly different from that of the controls. The site of insertion is critical in determining the expression level of a transgene. We however selected the lines in which the transgene was strongly expressed and we were able to confirm this by lacZ staining both in vitro and in vivo. Not all the lines which exhibited strong and ubiquitous lacZ staining in vitro behaved the same in vivo. This could be attributed to epigenetic modifications that occur during differentiation. Our in vitro experiments were done on undifferentiated ES cells while our in vivo analyses were done on embryos (differentiated cells). Epigenetic changes such as DNA methylation and histone modification occur as cells differentiate [153]. As a result certain genomic sites which were in the active mode in undifferentiated cells become silenced when the cells are differentiated. To ensure that the transgene is active in both undifferentiated as well as differentiated cells, we performed lacZ staining on both ES cells and the resulting embryos. The lines in which lacZ expression remained strong and ubiquitous were chosen for protein level assessment after removal of the βgeo cassette. Therefore transgene silencing could not be the reason for low level of nestin protein.

Although we showed that nestin RNA is being transcribed from the transgene, we currently do not have any information about mechanisms which may regulate translation of nestin protein from the RNA. Post-transcriptional regulatory mechanisms play important roles in controlling the translation of protein from RNA in the cell [154,155]. Our current knowledge about nestin protein is very limited. It is not clear whether formation of nestin protein is being regulated by post-transcriptional regulatory elements.

Once the protein is made inside of the cell, post-translational modifications may lead to degradation of the excess protein. Perhaps this is a mechanism by which the cell can control the amount of active proteins. Recent evidence suggests that nestin harbors consensus sites which make it prone to rapid degradation by mechanisms such ubiquitination (Dr. John Eriksson’s group, Finland, unpublished observations). Indeed when these sites are mutated, nestin protein is no longer detected by degradation agents and is “stable” in the cell. Moreover the group has shown that although over-expression of nestin cDNA (wild-type) was not possible due to rapid degradation of the excess protein generated from the transgene, over-expression of the mutated nestin cDNA is achievable. These observations suggest that translation of nestin is tightly regulated. Moreover they could explain why in our experiments we were not able to over-express nestin at levels detectable by western blot. More studies are required to assess the level of over-expression by the mutated nestin cDNA in the developing embryo. If over- expression can be achieved in vivo using this method, one could use it to study the effects of nestin over-expression in mouse development.

As previously mentioned, structure of nestin is unique among intermediate filament proteins. It has a very short head domain and therefore is unable to form a filamentous network by itself [53]. It is often co-expressed with class III intermediate filaments such as vimentin and desmin and forms heterodimers with them [53,156]. In the absence of a polymerization partner, nestin’s IF units can only form tetramers. It is not clear how stable these tetramer particles are. One could argue that in the absence of a suitable partner, excess nestin protein generated from the transgene remains as small particles unable to form a filamentous network. This may make them more prone to

degradation. It will be interesting to over-express nestin with one of its partners such as vimentin and study the stability and interaction of the two IF proteins. Moreover it is important to examine and compare the efficiency of network formation made by the mutated nestin cDNA (more stable) to that of wild-type nestin cDNA. One could speculate that to achieve over-expression of nestin protein at high levels, both stable (mutated) nestin cDNA and excess amount of a polymerization partner such as vimentin are required.

In summary our in vivo as well as in vitro data presented here indicate that although we were able to generate transgenic lines in which the randomly inserted transgene was strongly and ubiquitously expressed, we were not able to achieve significant over-expression of nestin at the protein level. As a result, although our transgenic lines do not exhibit major abnormalities we can not conclude that over- expression of nestin does not interfere with normal differentiation of the cell. More studies are required to shed light on mechanism(s) regulating the translation of nestin protein from mRNA. Moreover stability of nestin protein and its ability to integrate into the filamentous network made by other intermediate filaments are key issues which require further investigation.

Chapter 4

LOSS OF FUNCTION APPROACH: NESTIN KNOCK-OUT

Contributors: Mouse chimeras were generated at the transgenic facilities of Samuel Lunenfeld Research Institute and Sunnybrook Health Sciences Centre. MRI was done by Ms. Christine Laliberte at Mouse Imaging Centre of Hospital for Sick Children (Department of Medical Biophysics). Electron microscopy and image analysis of renal samples were done in collaboration with Dr. Paul Thorner and Dr. Amanda Murphy at Hospital for Sick Children (Department of Pediatric Laboratory Medicine). Neuro-behavior tests were done in collaboration with Dr. John Roder and Dr. Gang Xie at Samuel Lunenfeld Research Institute.

1 Introduction

Nestin has been extensively used as a marker for neural stem/progenitor cells both in vitro and in vivo [49,66,76,129,130,157]. Outside of the developing nervous system, the expression of nestin is detected in muscle precursors [54], and cells of the developing heart, testis and kidney [55,158,159]. As cells reach terminal stages of differentiation, nestin is down-regulated and replaced by tissue-specific intermediate filaments such as desmin in the muscle and α-internexin and neurofilaments in the neurons [131]. In adults, nestin is expressed mainly in the subependymal zone and dentate gyrus of the brain where neural stem/progenitors reside [52,64]. In addition, nestin has been detected in the neuromuscular junction (NMJ) [68,69] and renal podocytes [159].

Intermediate filaments (IFs), along with microtubules and actins, are major components of the cytoskeleton. IFs are sub-categorized into five groups based on sequence and structural similarities [160]. With the exception of lamins which are associated with the nuclear membrane [161], IFs are localized in the cytoplasm. Each IF has an alpha-helical rod domain and two non-helical domains: the N-terminus or the head, and the C-terminus or the tail. The tail domain has been suggested to interact with other components of the cytoskeleton [162,163], while the head domain is used for assembly of IF blocks to form protofilaments [160,164].

Recently it has been suggested that nestin should be placed in the fourth class of IFs (neurofilaments) due to its evolutionary relationship with members of this category and having a long C-terminus [82,165]. However, Nestin was originally categorized as the only member of the sixth class of IFs due to its unique short head and large tail [49]. As a result of its short head domain, nestin lacks the ability to form protofilaments on its

own [53] and is often co-expressed with other IFs. It has been shown that nestin readily incorporates into the vimentin network [53], and can form protofilaments with desmin in muscle progenitors [54] and α-internexin in early neurons [156].

IFs have an established role in providing mechanical strength to various cells and tissues. Moreover there is accumulating evidence that they participate in cellular organization and that they also are important determinants of key signaling pathways [6,43,44]. Expression of IFs is tissue-specific, and they are often co-expressed in characteristic combinations in different cells. Some mouse knockout models of IFs have been shown to be viable and fertile with only minor abnormalities [20,22,23,24], leading to speculations that some of the IF proteins may be essential only when the organism is exposed to stress and challenging conditions. Indeed subsequent studies have proven the above hypothesis to be true (i.e. impaired wound healing in vimentin-deficient mice [39] and cardiac failure leading to death of desmin knockout mice after prolonged swimming [166]). Moreover, double gene knockout studies (e.g. vimentin/glial fibrillary acidic protein (GFAP)) suggest that there could be some degree of functional redundancy amongst IF proteins [41]. Targeted mutations of nestin have not been reported yet; therefore, it is not clear what role(s) this IF plays during development or in adults.

Recent in vitro knockdown studies have linked this IF to the migration of cancer cells, suggesting that nestin may play a role in cancer cell invasion and mobility [167] . Other in vitro studies have shown that nestin can protect neural progenitors against oxidant-induced cell death through interaction with Cdk5 [82]. Indeed, nestin can serve as a scaffold for Cdk5 and is able to regulate its kinase activity in vitro [81,82].

Cdk5 is a serine/threonine kinase protein. Although closely related to other cyclin-dependent kinases, Cdk5 is not involved in cell-cycle control. The kinase activity of Cdk5 is regulated by binding to the highly specific regulatory sub-units p35 and p39 [99,168,169]. In vivo expression of Cdk5 has been reported in proliferative as well as post-mitotic neurons; however Cdk5 kinase activity has only been detected in post- mitotic neurons in which Cdk5 is bound to its activators [170]. Mice deficient in Cdk5

expression die perinatally due to neural migration defects observed in the cerebral cortex, hippocampus and cerebellum [94,171]. Subsequent studies have revealed a novel role for Cdk5 at the NMJ. In contrast to agrin, a nerve-derived AChR stabilizing factor [112,113], Cdk5 disperses AChR clusters that are not innervated [106]. Absence of Cdk5 can partially rescue the lack of stabilized AChR clusters in an agrin null background [106].

Outside of the nervous system, Cdk5 has been shown to be active in renal podocytes where it is co-expressed with p35. Inhibition of Cdk5 activity via either a pharmacological approach or siRNA in cultured podocytes dramatically changes the morphology of the cells, leading to loss of process formation [101]. Thus, activity of Cdk5 is essential for proper formation and for maintaining the structural integrity of renal podocytes.

In the present study, we aim to decipher the role of nestin in vivo through generation of nestin knockout mice via gene targeting. Our results indicate that although nestin-deficient mice are viable, they exhibit abnormalities in clustering of AChR at the NMJ and in their podocyte structure. Moreover, through generation of nestin/agrin double knockout mice we showed that, parallel to the case of Cdk5/agrin double knockouts, deficiency of nestin partially rescues dispersion of AChR clusters in an agrin null background. These data suggest that nestin acts through the Cdk5 pathway in these developmental processes.

2 Results

2.1 Generation of nestin-deficient mice

The mouse nestin gene (Nes) comprises four exons spread through a 12Kb genomic region on chromosome 3. To generate a null mutation, we replaced almost the entire first exon and part of the 5’ upstream region with a PGK promoter driving neomycin phosphotransferase (neo) using homologous recombination-based targeting (Figure 1A). Out of 239 neo resistant colonies screened after electroporating the target vector into the G4, 129xC57BL/6 F1 hybrid embryonic stem (ES) cells [117], four targeted clones were identified (targeting frequency 1.7%) by Southern blot analysis (Figure 1B). All four clones were used to generate chimeric founder mice by ES cell <-> tetraploid embryo complementation assay [124]. Two lines successfully resulted in completely ES cell-derived males and transmitted the mutant allele to their offspring. These two lines were used for subsequent studies. We analyzed the nestin mutation in two different genetic backgrounds; the founders (50% C57BL/6 (B6) and 50% 129/Sv) were crossed either to ICR (outbred) or B6. The latter were backcrossed 3 times to B6 to minimize 129 contributions. We have not observed any variation caused by the genetic background or between the two independent gene-targeted lines.

Figure 1. Targeting the nestin locus and identification of targeted clones by Southern blot. (A) Using two homology arms, most of exon 1 and part of the 5’ upstream region was replaced with a PGK promoter driving neomycin phosphotransferase.

(B) Using a 5’ external probe, four targeted clones were identified by Southern blot analysis. The expected size for the wild-type and targeted bands were 20kb and 14kb respectively.

Western blot analysis did not detect nestin protein in Nes-/- embryos (E10.5) obtained by crossing heterozygous parents, while the protein was observed both in Nes+/- and Nes+/+ embryos (Figure 2A) proving the null mutation nature of our targeted allele. Nestin expression was further examined at the cellular level by immunohistochemistry with both whole mount embryos (E10.5) and cryosectioned tissues (E14.5). While strong nestin signals were detected in Nes+/+ mice, signals from Nes-/- mice were the same as the background level, further supporting the absence of nestin in the homozygous mutants (Figures 2B and 2C).

Figure 2. Homozygous mutants do not express nestin. (A) When analyzed by western blot, nestin protein was absent in E10.5 homozygous embryos while the appropriate amount was detected in heterozygous and wild-type embryos of the same litter. (B-C) Confirming the western blot results, no nestin protein was detected by immunohistochemistry in (B) whole E10.5 embryos (desmin: red, nestin: green and PECAM-1 (blood vessels): blue) and (C) E14.5 sections of homozygous embryos (nestin: red).

2.2 Nestin is dispensable for development of the mouse

Since nestin is expressed as early as E7.5 in the neuroepithelium and between E9 and E11 in the developing mouse heart [55], we examined the developing CNS and cardiovascular system of Nes-/- embryos at E10.25 along with their wild-type littermates as controls. Both systems were anatomically normal (Figure 3). The interbreed of Nes+/- heterozygous parents yielded the expected Mendelian ratios of Nes+/+, Nes+/- and Nes-/- mice indicating that nestin deficiency does not cause embryonic lethality (Table 1). All genotypes reached sexual maturity and were fertile. We did not observe any differences in size or body weight between the groups (data not shown). Histological examination of various tissues (e.g. CNS, muscle, heart and kidney) did not reveal any apparent morphological abnormalities resulting from the absence of nestin expression in the examined mice (Figures 4A-A’, 4B-B’ and 4C-C’). However, these investigations could not rule out the possibility of minor changes arising from the absence of nestin. Because nestin is highly expressed in progenitors of the CNS, we decided to further examine the brain structure by Magnetic Resonance Imaging (MRI). MRI analysis suggested that the average size of the mutant brains was slightly smaller than that of the wild-types but the difference was not statistically significant with our sample size (nWT=5 and nKO=10) (Figure 4D and Table 2 at the end of this chapter). This assay routinely detects significant differences down to volume of 0.5 mm3. As an overall conclusion, our data suggests that nestin deficiency does not have a negative effect on mouse development and tissue structure at a gross anatomical level.

Nes +/+ Nes +/- Nes -/- Expected 76(25%) 152(50%) 76(25%) Actual 74 159 71

Table 1. Interbreeding of Nes+/- heterozygous parents yielded the expected Mendelian ratios of Nes+/+, Nes+/- and Nes-/- mice.

Figure 3. Embryos were dissected at E10.25. No obvious abnormalities were observed in Nes -/- embryos compared with their Nes +/+ or Nes +/- littermates.

Figure 4. Nestin deficient mice do not exhibit any apparent anatomic abnormalities. (A-C’) Histological analysis on muscle tissue (A, A’), brain (B, B’) and kidney (C, C’) did not reveal any major abnormalities in nestin deficient mice. (D) Since nestin is highly expressed during CNS development, brain was further examined with MRI (nwt=5, nKO=10). No significant differences were observed between the mutants and the control groups. The graph shows the volume of various brain structures measured by MRI. 1: cerebellar cortex; 2: cerebral cortex; 3: corpus callosum; 4: hippocampus; 5: hypothalamus; 6: midbrain; 7: olfactory bulbs; 8: striatum; 9: thalamus.

2.3 Nestin deficiency alters renal podocyte structure

Since nestin is strongly expressed in mature podocytes and in the developing kidney, we investigated whether nestin deficiency affected renal structure and function. Routine histology of embryonic and adult kidneys did not show any histological abnormality. However, electron microscopy (EM) revealed differences in renal podocyte structure in Nes-/- mice compared to wild-type controls. Scanning EM of wild-type glomeruli showed podocytes with many branching primary and secondary processes forming a well-organized structure covering capillary loops (Figures 5A and 5A’). In comparison, podocytes from Nes-/- mice had reduced numbers of primary processes, which were abnormally shaped with irregular branching (dysmorphic) (Figures 5B and 5B’). The dysmorphic processes formed a disorganized covering over glomerular capillaries, with gaps between processes rather than the tight configuration in the wild- type mouse (Figure 5C). Primary processes appeared smaller, thinner and more rudimentary in knockouts compared to controls. Foot processes along the glomerular basement membrane had dysmorphic profiles, as opposed to the usual palisading appearance. Despite these findings, the slit diaphragms appeared to be morphologically normal. These ultrastructural abnormalities were apparent in both adults as well as newborns, indicating that they were congenital defects. Surprisingly, urinalysis of Nes-/- mice (dipstick testing of mouse urine and protein:creatinine ratio) failed to identify any significant abnormalities suggesting that the glomerular filtration barrier function was not compromised (Figure 6).

Figure 5. Abnormal glomerular ultrastucture in nestin KO mice. (A-A’) Podocytes processes are disorganized and reduced in number as shown in SEM pictures. (B-B’) Abnormal processes form a disorganized lace-like network with multiple gaps. (C) Large gaps are observed between the disorganized processes of nestin KO mice (n=2 for each group, P-value <0.05).

Figure 6. Protein-Creatinine (PC) ratio was measured in nestin KO mice and their wild-type littermates (n=8 for each group). The difference was not statistically significant (P-value>0.05).

2.4 Nestin is required for proper peripheral motor function

To further characterize nestin-deficient mice, we performed a series of peripheral type neurobehavioral functional tests such as motor coordination and balance test. The Roto-Rod test [128] was used to assay coordination and function of motor-neurons. Mice were placed on a controlled rotating rod and their latency-to-fall was measured. Nes-/- mice consistently showed a lower average compared with their wild-type littermates (Figure 7A). Since lower than expected performance in a Roto-Rod test could be caused by lower grip strength, we proceeded to directly examine muscle strength in nestin- deficient mice. We did not detect any significant differences in grip strength of either forelimbs or hindlimbs between knockout and control groups (Figure 7B). Performance in the Roto-Rod test can also be influenced by the ability to maintain balance. Consequently, we tested both experimental and control groups in a long rod balance test but did not observe any significant differences between the groups (data not shown). The above experiments excluded two factors, muscle strength and balance, which could have caused the poor performance of nestin-deficient mice on the Roto-Rod system. Therefore, we postulated that the poor performance was related to abnormalities in connection of peripheral nervous system to muscles, namely the NMJ [172].

2.5 Broadening of AChR endplate band and increased number of AChR clusters in NMJ of Nes-/- mice

Both nestin and Cdk5 are expressed at the NMJ [68,69,102] and the latter has recently been shown to play a major role in the dispersion of AChR clusters in the NMJ [105,106]. It has also been shown that nestin serves as a scaffold for Cdk5 and through this interaction, can regulate Cdk5/p35 signaling in vitro [82]. Therefore, we examined the NMJ and the pattern of AChR clusters in nestin-deficient mice. We focused on the diaphragm muscle where AChR formation and maintenance has been well studied.

Alpha-bungarotoxin (alpha-BTX) has been extensively used to visualize the clusters of AChRs because it binds specifically to AChRs. Using alpha-BTX, we analyzed the diaphragm muscles of nestin-deficient mice as well as those of the wild-type controls. Interestingly, we found that nestin mutants have more AChRs compared with their wild- type littermates. Furthermore, AChRs occupy a larger muscle territory and the width of the endplate band is larger in nestin mutants (Figures 7C, 7C’, 7E and 7F). Initial clustering of AChRs occurs at E17-E18 of embryonic development. To assess whether the observed phenotype is a congenital defect, we analyzed AChR clusters of newborn mice (P0). Similar to the adult mice, the number and area of AChR clusters were enhanced in newborn nestin-deficient animals, indicating that the observed phenotype is indeed a developmental defect (Figures 7D and 7D’).

Figure 7. Abnormal NMJ in nestin deficient mice. (A) When tested on Roto-Rod, nestin deficient mice had a lower performance level when compared with wild-types littermates (nWT=5, nKO=11, number of trials=9, P-value<0.01, T-test). (B) There is no significant difference between the muscle strength of mutants and controls (nWT=5, nKO=11, number of trials=3, P-value>0.05, T-test) (C-D’) Nestin knockout mice have more AChRs compared with their wild-type littermates. AChR clusters in the KOs occupy a larger muscle territory and are dispersed. The phenotype was observed both in adults (C-C’) and newborn pups (D-D’). (E-F) Quantification of the number of AChR clusters and their occupied area (nWT=5, nKO=5, P-value<0.05, T-test)

2.6 Nestin regulates activity of Cdk5 in vivo: evidence from nestin/agrin double knockout mice

Previous studies on agrin-deficient mice have shown that AChR clusters are initially made in these mutants but are dispersed at later stages, suggesting that agrin works against a dispersion signal to maintain AChR clusters[112]. Follow up studies have identified Cdk5 as the dispersing agent responsible for disassembling the AChRs that are not innervated. Moreover, a significant number of AChR clusters are maintained in agrin/Cdk5 double knockouts whereas they would have been dispersed in the presence of Cdk5 activity in agrin single knockouts [106]. These observations indicate that the opposing roles of agrin and Cdk5 as positive and negative signals, respectively, shape and maintain innervated AChR clusters. Thus, increase in AChR clusters observed in the nestin-deficient mice could reflect reduced effects of Cdk5. To investigate this hypothesis, we employed a genetic approach. We generated nestin/agrin double knockouts and examined clustering of AChRs at E17.5-E18.5 of development. Remarkably, we observed many AChR clusters in nestin/agrin double knockout mutants whereas the number of clusters was significantly less in agrin single knockouts (Figures 8A-E). This observation indicates that lack of nestin in the agrin null background leads to maintenance of AChR clusters.

Figure 8. Nestin deficiency can rescue maintenance of AChR clusters in agrin null background.

(A-D) Number of AChR clusters is significantly higher in nestin KO; agrin KO embryos compared to their nestin WT; agrin KO littermates (A: nestin WT; agrin WT, B: nestin WT; agrin KO, C: nestin KO; agrin WT, D: nestin KO; agrin KO). (E) Quantification of AChR clusters in nestin KO; agrin KO and nestin WT; agrin KO embryos (n=3 for each group, P-value<0.05, Student’s T-test).

3 Discussion

Here we report the generation of nestin knockout mice via gene targeting. Our results indicate that nestin is not essential for development of the central nervous system but required for proper formation of glomerular podocytes and fine-tuning of the NMJ. Furthermore, our data from nestin/agrin double knockout mutants indicate that, similarly to Cdk5, lack of nestin leads to stabilization of a significant number of AChR clusters in an agrin null background, suggesting that nestin acts through the Cdk5 pathway in the NMJ.

The lack of a more severe phenotype may reflect that IFs are often co-expressed in cells suggesting there could be some degree of overlap in their functions. Since nestin is unable to make a filamentous network by itself, it is often co-polymerized with vimentin [53]. It has been shown that in the absence of the vimentin network (i.e. in vimentin-deficient mice) the nestin network is severely compromised and almost undetectable [173]. Interestingly, a new member of the IF family named synemin was recently discovered [174], the expression pattern of which is very similar to that of nestin (i.e. developing nervous system and muscle progenitors). However, nestin and synemin do not co-localize in the cells and exist in two independent filamentous networks. Hence, one could postulate that in the absence of nestin, synemin may serve similar function(s) leading to normal development of nestin knockout embryos, thereby masking any crucial function of nestin during embryonic development. Currently, there is no synemin knockout mouse model available. Future studies will shed light on the interaction and possible functional overlap between these two proteins.

Interestingly, many knockout models of IFs with no gross abnormalities [20,22,23,24] exhibit profound phenotypes when exposed to stressful conditions. A swimming work overload protocol led to mortality of almost 50% of desmin null mice tested [166]. Mice deficient in intermediate filaments GFAP and vimentin show poor outcome in the acute stages of stroke [175]. These observations suggest that IFs may be crucial for maintaining the proper function(s) of tissues when the body is challenged. Although we observed abnormal renal podocyte structures in nestin knockout mice, the mutants did not have proteinuria. However defective podocytes may make the mutants more vulnerable to challenging conditions (e.g. high protein diet, diabetes and aging). The same could apply to the adult CNS, where nestin is abundantly expressed in reactive astrocytes upon injury. Other IFs present in reactive astrocytes include vimentin and GFAP. Previous work on double knockouts (i.e. GFAP/vimentin) resulted in attenuated reactive gliosis upon CNS damage [41]. Is the formation of reactive astrocytes affected by the absence of nestin? Will the scar region be more permissive for generation and maturation of new neurons in the absence of nestin? In the next phase of investigations, it will therefore be important to study recovery of kidneys and CNS in nestin-deficient mice under challenging conditions.

The main phenotypic abnormality in our model involved the NMJ with reduced motor function associated with increased stabilization of AChR clusters. Based on the following observations, we hypothesized that this might be related to decreased activity of Cdk5. First, both nestin and Cdk5 are expressed at the NMJ [68,69,102]. Second, Cdk5 is essential for dispersion of AChR clusters in the NMJ and just as in nestin knockouts, the number of AChR clusters in the Cdk5 deficient embryos are significantly higher than the wild-type controls [105,106]. Third, nestin serves as a scaffold for Cdk5 [82]. We tested our hypothesis using nestin/agrin double knockouts. We reasoned that if nestin regulates the activity of Cdk5, similar to Cdk5 deficiency absence of nestin should rescue maintenance of AChR clusters in an agrin null background. Our studies on the nestin/agrin double knockout mutants confirmed this hypothesis. We showed that lack of nestin leads to maintenance of AChR clusters where there is no agrin activity. This is

analogous to what was found in Cdk5/agrin mutants suggesting that nestin and Cdk5 act through the same pathway. Our observations are supported by data from a recent study reported by Yang et al. (submitted) [176]. Results from this recent study indicate that nestin is essential for recruitment of p35 to the muscle membrane and activation of Cdk5 at the NMJ [176]. Furthermore, reducing the expression of nestin via RNAi significantly decreases the activity of Cdk5 in cultured myotubes [176]. Finally similar to our results, Yang et al. [176] showed that the in vivo knockdown of nestin can partially rescue the maintenance of AChR clusters in an agrin null background. Altogether these results strongly suggest that the observed NMJ phenotype in the nestin knockout mice is due to reduced Cdk5 activity. This may also explain the podocyte abnormalities, since Cdk5 is also expressed in podocytes but will require confirmation through further studies.

In summary, our data indicate that although nestin is highly expressed during CNS development and commonly used for detection of neural stem/progenitor cells, it is dispensable for formation of the CNS. Nestin deficiency leads to precise defects of AChR clustering at the NMJ and the ultrastructure of renal podocytes. Furthermore, our data from nestin/agrin double mutants signify the importance of nestin as a regulator of Cdk5 kinase activity leading to modulation of certain cell signaling events in vivo. This finding provides the in vivo evidence that nestin acts through the Cdk5 pathway and suggests a novel role for nestin in Cdk5 mediated maintenance of AChR clusters.

Mean volume of Mean volume of 3 3 T- Brain structures structure in mm for structure in mm for knock-outs wild-types statistics cerebral cortex: occipital lobe 4.97 5.54 3.403 mammillary bodies 0.53 0.46 -2.076 olfactory tubercle 3.48 3.73 2.027 subependymale zone / rhinocele 0.1 0.11 1.869 cerebral cortex: parieto-temporal lobe 71.76 75.53 1.86 pontine nucleus 0.84 0.71 -1.75 olfactory bulbs 25.16 26.53 1.745 cerebellar cortex 47.95 51.15 1.695 mammilothalamic tract 0.29 0.27 -1.687 corticospinal tract/pyramids 1.74 1.85 1.587 cerebral cortex: entorhinal cortex 8.85 9.4 1.582 third ventricle 1.23 1 -1.517 facial nerve (cranial nerve 7) 0.3 0.28 -1.48 superior olivary complex 0.67 0.64 -1.433 cuneate nucleus 0.31 0.26 -1.402 stria medullaris 0.82 0.78 -1.396 cerebellar peduncle: middle 1.39 1.33 -1.319 amygdala 13.52 14.32 1.3 midbrain 14.76 14.06 -1.291 cerebral peduncle 2.45 2.27 -1.268 nucleus accumbens 4.02 4.23 1.239 cerebral cortex: frontal lobe 44.23 45.92 1.163 ventral tegmental decussation 0.15 0.14 -1.162 arbor vita of cerebellum 12.62 13.09 1.072 pons 18.12 17.52 -1.058 fimbria 3.39 3.54 1.037 cerebellar peduncle: superior 1.12 1.08 -1.015 whole brain 461.44 472.72 0.9113 hippocampus 18.95 19.41 0.8524 anterior commissure: pars anterior 1.66 1.71 0.792 habenular commissure 0.053 0.048 -0.7835 interpedunclar nucleus 0.24 0.25 0.7072 lateral septum 2.94 3.03 0.6987 fundus of striatum 0.17 0.18 0.6393 lateral olfactory tract 1.84 1.87 0.637 pre-para subiculum 2.13 2.09 -0.602 fourth ventricle 0.42 0.4 -0.5942 medial lemniscus/medial longitudinal fasciculus 2.81 2.77 -0.5848 lateral ventricle 3.56 3.68 0.5636 hypothalamus 10.15 9.99 -0.4617 posterior commissure 0.17 0.16 -0.4435 colliculus: inferior 5.99 5.88 -0.4422 cerebral aqueduct 0.52 0.54 0.3526 stratum granulosum of hippocampus 0.97 0.98 0.3474 inferior olivary complex 0.27 0.28 0.3319 fornix 0.76 0.74 -0.3025 stria terminalis 1.08 1.07 -0.2712 globus pallidus 3.21 3.18 -0.2264 basal forebrain 5.79 5.85 0.2179 optic tract 1.64 1.65 0.2053 anterior commissure: pars posterior 0.52 0.51 -0.1911

corpus callosum 21.07 20.97 -0.152 colliculus: superior 8.98 8.92 -0.1455 medulla 26.12 26.02 -0.1388 fasciculus retroflexus 0.31 0.31 0.1245 internal capsule 3.42 3.41 -0.1097 cerebellar peduncle: inferior 0.84 0.84 0.1057 striatum 22.27 22.36 0.0958 thalamus 17.36 17.41 0.08578 periaqueductal grey 4.14 4.15 0.04877 medial septum 1.38 1.38 -0.04711 dentate gyrus of hippocampus 3.55 3.56 0.04266 bed nucleus of stria terminalis 1.37 1.37 -0.001035

Table 2. Detailed analysis of brain structures using MRI did not reveal any significant differences between nestin KO mice and their wild-type littermates (nWT=5, nKO=10).

Chapter 5

BEING A TEAM PLAYER: GENERATION AND CHARACTERIZATION OF NESTIN/VIMENTIN AND NESTIN/DESMIN DOUBLE KNOCKOUT MICE

1 Introduction

As mentioned in chapter 1, nestin has a short head domain (N-terminus) [49,53]. As a result it is unable to form protofilaments [53], a structure required for formation of the filamentous network by intermediate filaments. Nestin needs a copolymerization partner to be able to integrate into the cytoskeletal network. In the absence of a polymerization partner, nestin forms non-integrated tetramers inside of the cell [177]. It is not clear what role(s) these tetramers play. Moreover since they are not integrated in the cytoskeleton, it is not known how stable these particles are.

Copolymerization partners of nestin are mostly class III IFs such as vimentin and desmin [53,54]. As described in chapter 1, co-expression of nestin and vimentin has been reported in many cell types [53,54,70]. Previous studies showed that nestin could readily incorporate into the network made by vimentin [53,177]. As a result the filamentous network made by nestin is often identical to that of vimentin. Vimentin knockout mice are viable and fertile and do not show any obvious signs of abnormalities. These results indicate that despite the widespread expression of vimentin, this IF is not essential for normal development of the mouse [20]. Later studies shed light on the importance of vimentin under pathological conditions. For example vimentin is required for wound healing both during embryonic development and in adults [39]. Interestingly, expression of nestin was examined in vimentin deficient mice; although nestin RNA was expressed at normal levels, expression of nestin at protein level was almost undetectable. Detailed analysis of CNS in vimentin knockouts revealed that low level of nestin’s immunoreactivity can be detected in some parts but rather than forming a filamentous network it is diffused in the cytoplasm [41].

Desmin is another class III intermediate filament protein that is co-expressed with nestin [54]. As mentioned in chapter 1, in muscle precursor cells nestin and desmin form

heterodimers and together they form a filamentous structure. As cells reach terminal stages of differentiation expression of nestin diminishes and desmin becomes the main intermediate filament protein expressed in mature muscle cells [54]. In order to study the role of desmin in muscle differentiation, mice deficient in expression of this IF were generated by gene targeting [35]. Desmin knockout mice did not show any apparent abnormalities suggesting that desmin is not required for skeletal muscle developement [35]. Unlike vimentin knockout mice, expression of nestin in desmin deficient mice has not been studied. Considering that nestin relies on a polymerization partner to integrate into the filamentous network of the cytoskeleton, one could speculate that in absence of desmin in muscle progenitors, nestin protein remains as tetramer particles with unknown function.

Previous studies on intermediate filaments showed that there could be some overlap in the function of IFs. For example formation of reactive astrocytes following an injury to CNS appears normal in both vimentin and GFAP knockout mice. Normally reactive astrocytes express both vimentin and GFAP [70] but in case of these single knockouts only one of the two IFs is present (e.g. in vimentin knockout mice, reactive astrocytes only contain GFAP). Later studies revealed that in vimentin/GFAP double knockouts; formation of glial scars is greatly compromised indicating that vimentin and GFAP have some overlapping functions [41]. We hypothesized that since nestin is often co-expressed with vimentin and desmin, there could be some degree of functional overlap between them as well (i.e. nestin/vimentin and nestin/desmin). Hence one could postulate that in absence of nestin, vimentin or desmin can take on its function(s) leading to normal development of nestin knockout embryos. In order to test this hypothesis we generated nestin/vimentin and nestin/desmin double knockout mice and assessed their development.

2 Results

2.1 Generation and characterization of nestin/vimentin double knockout mice

Generation and characterization of vimentin knockout mice were described previously [20]. These vimentin knockout mice were crossed with nestin knockouts which were described in chapter 3 of this thesis. We did not observe any significant differences in the size or body weight of the double mutants compared with the control littermates (Figure 1, Table 1). Moreover fertility of the double mutants was not affected by the absence of nestin and vimentin. These observations indicate that combined deletion of nestin and vimentin does not cause embryonic lethality or severe abnormalities. Subsequently we performed a series of histological analysis to take a closer look at the organs which might have been affected by the absence of nestin and vimentin. We did not observe any major abnormalities in the brain, testis or gut epithelium of double knockouts compared with the controls. However we can not rule out existence of subtle defects caused by the absence of both nestin and vimentin in mice as more detailed analyses are needed to draw such conclusions.

15 Body weight (grams)

10 Groups tested Nes-/-; Vim+/+ Nes-/-; Vim-/-

Figure 1. No obvious defects in the body weight of nestin/vimentin double knockout mice (n=6) compared with the control litter-mates (nestin knockout/vimentin wild-type, n=6), P>0.05.

Total # of mice: Nes -/- Nes -/- Nes -/- Nes -/- 28 (litter #1) (litter #2) (litter #3) (total) Vim +/+ 4 0 2 6 (1/2) Vim +/- 7 5 4 16 (1/4) Vim -/- 4 1 1 6 (1/2)

Table 1. Nestin/vimentin double knockout mice are viable. Genotyping results of three representative litters resulting from the intercross of Nes -/-;Vim +/- mice (n=28).

2.2 Generation and characterization of nestin/desmin double knockout mice

As mentioned in chapter 3, we did not observe any major abnormalities in the structure or strength of muscle in nestin deficient mice suggesting that nestin is not required for normal development of the muscle. However it is possible that in the absence of nestin, desmin takes on its function. As a result lack of an obvious muscle phenotype in nestin deficient mice could be due to compensation by desmin. We hypothesized if nestin is essential for muscle progenitor cells and such function is masked by desmin in nestin single knockouts, nestin/desmin double knockouts should exhibit enhanced muscle abnormalities. To test this hypothesis we generated mice deficient in expression of both nestin and desmin. Generation and characterization of desmin deficient mice were described previously [35]. As mentioned in chapter 1, desmin knockout mice do not exhibit any obvious abnormalities but careful histological analysis revealed subtle defects in the structure of muscle in these mutants [35].

Double knockout mice were generated by crossing nestin and desmin single knockouts and subsequently the intercross of nestin/desmin double heterozygous mice.

Nestin/desmin double mutants are viable, fertile and do not exhibit any obvious abnormalities (Figure 2, Table 2). This observation suggests that lack of muscle phenotype in nestin knockout mice is not due to compensation by desmin. Moreover, absence of nestin and desmin does not severely affect survival and differentiation of muscle precursor cells in vivo.

Body weight (grams)Body weight 15

10 Groups tested Nes+/+;Des+/+ Nes+/-;Des+/- Nes-/-;Des-/-

Figure 2. The body weight of nestin/desmin double knockout mice (n=2) was not significantly different from that of their wild-type (n=1) or nestin/desmin heterozygous littermates (n=11). P>0.05.

Total # of mice: 29 Nes +/+ Nes +/- Nes -/- Des +/+ 1 (1/16) 5 (1/8) 1 (1/16) Des +/- 1 (1/8) 11 (1/4) 3 (1/8) Des -/- 0 (1/16) 5 (1/8) 2 (1/16)

Table 2. Nestin/desmin double knockout mice are viable. The number of Nes-/-; Des -/- resulting from the intercross of Nes+/-; Des +/- mice is in accordance with the expected Mendelian ratio. This result indicates that absence of Nestin and desmin does not cause embryonic lethality (n=29).

3 Discussion

Two intermediate filaments commonly co-expressed with nestin are vimentin and desmin. In order to examine whether there is some degree of functional overlap between these IF proteins, we generated double mutants deficient in expression of nestin along with either vimentin or desmin. In both cases double mutant mice were viable, fertile and did not exhibit any obvious abnormalities. These results indicate that lack of major embryonic defects in nestin deficient mice is not due to compensation by either vimentin or desmin. We however can not rule out existence of subtle abnormalities in the double mutants (nestin/vimentin and nestin/desmin mice).

As previously mentioned, nestin is unable to form a filamentous network without a polymerization partner. In vimentin deficient mice the filamentous structure of nestin collapses and low level of nestin’s expression is detected in the cell as diffused particles [41]. It is not clear how absence of desmin affects expression of nestin. If nestin has to be part of a filamentous network in order to be functional, then vimentin or desmin knockout mice can be considered as nestin knockout mutants in the tissues where nestin is co-expressed with vimentin or desmin. However one should keep in mind that nestin as tetramer particles may play important roles in the cell without being required to integrate into the filamentous network of the cytoskeleton. More studies are required to address the role of tetramers in the cell.

Interestingly synemin, a relatively new member of IF proteins is also expressed in many progenitors which nestin is expressed in but it is able to form a separate IF network than that of nestin. This property of synemin makes it an attractive candidate for functional redundancy studies as theoretically absence of one protein should not affect the filamentous network of the other one. Currently our knowledge about synemin is very limited. Moreover knockout mouse models of synemin have not been generated yet.

Further studies are required to shed light on the potential functional overlap of nestin and synemin.

Abundant expression of GFAP is detected in astrocytes. As mentioned in chapter 1, knockout models of this intermediate protein are viable and fertile. Furthermore, mice deficient in expression of both GFAP and vimentin are also viable and fertile without any obvious signs of abnormalities [41]. However under challenging conditions such as nervous system trauma, vimentin/GFAP double knockout mice exhibit a phenotype that is not observed in either vimentin single knockout or GFAP single knockout mice (i.e. impaired reactive astrocyte formation) [41]. This observation suggests that although lack of intermediate filaments may not result in a phenotype under normal physiological conditions, once the organism is challenged the importance of IFs becomes evident. Therefore one may speculate that the same may be true for nestin/vimentin and nestin/desmin double knockout mice. For example, following damage or injury to the muscle, rapid re-expression of nestin is detected [178]. How efficient is the healing process in the absence of nestin or the absence of both nestin and desmin? Examining the double mutants (i.e. nestin/vimentin and nestin/desmin mice) under challenging and pathological conditions will provide important information about possible functional overlap of these IFs.

In summary our data presented in this chapter suggest that lack of a severe phenotype in nestin deficient mice is not due to compensation by vimentin or desmin. Both nestin/vimentin and nestin/desmin double deficient mice are viable, fertile and do not show any obvious defects. Functional overlap of nestin with other intermediate filaments such as synemin remains to be addressed in the future. Moreover it is important to study the interaction of these intermediate filaments under pathological conditions as the IF deficient mice may exhibit a phenotype not observed under normal conditions.

Chapter 6

CONCLUSION AND FUTURE DIRECTIONS

The aim of this project was to study the role of intermediate filament protein nestin in mouse development. As described in previous chapters, expression pattern of nestin has been well studied both in adults and in embryos [50,51,52,56,57,59,60,65,68]. There are also reports explaining expression of nestin in vitro during the course of differentiation towards different lineages [121,142,143,144]. Although the complex expression pattern of nestin has been studied by various groups and reported in many different cell types, our knowledge about the function of this intermediate filament protein is very limited. Recent emerging studies suggest that intermediate filaments are capable of modulating cell signaling events [46,47,81,82]. In vitro studies on neural progenitor cells have linked nestin to Cdk5. The study suggests that nestin is able to make a scaffold for this kinase protein and can control its activity [82].

I hypothesized that nestin is essential for normal development of the mouse; a role that could be fulfilled by regulating the activity of Cdk5. To test this hypothesis I used both in vivo and in vitro assays and utilized two approaches of gain of function and loss of function. If this hypothesis was true, absence of nestin (loss of function) or its deregulated expression (gain of function) would interfere with normal development of the mouse and would lead to embryonic abnormalities.

Our results from the gain of function approach were not conclusive as we were not able to over-express nestin at protein level. We were able to generate ES lines in which the floxed transgene was highly and ubiquitously expressed. Moreover we were able to generate chimeras in which expression of the transgene was detected ubiquitously (conclusions were made based on Lac-Z expression). In addition both in vivo and in vitro we successfully removed the βgeo cassette via activation of Cre recombinase and were able to detect the expression of nestin RNA (from the transgene) by RT-PCR. However we did not observe a significant difference in the amount of total nestin protein produced in the transgenic lines compared with the control lines. As a result, although we did not observe any defects in differentiation of our transgenic lines in vitro and our nestin

forced-expressing mice were viable and fertile we could not conclude that over- expression of nestin does not have any negative effects on normal development of the mouse.

We did not pursue the experiments needed to explain why over-expression of nestin at protein level was not achieved. To answer this question a series of biochemical experiments are needed to carefully evaluate the mechanisms that regulate the translation of nestin RNA and production of nestin protein. Such experiments were beyond the scope of this project. Moreover the aim of this study was to assess the role of nestin in development not to study the mechanisms which control the expression of this protein. Recent unpublished data (personal communication, John Eriksson, Finland) may provide a hint as to why nestin protein can not be over-expressed. The results of this new study suggest that nestin contains consensus sites which make it prone to rapid degradation by mechanisms such as ubiquitination. When such sites are mutated the resulting protein is no longer detected by the degradation agents and will remain in the cell. This might be a mechanism by which the cell controls the amount of nestin protein being produced. It is important to note that in absence of a polymerization partner, nestin is unable to form a filamentous network by itself. Therefore when mutated (stable) form of nestin is being over-expressed there will be an imbalance between the amount of nestin and its polymerization partner in the cell. The excess amount of nestin protein will not be able to integrate into the filamentous network and will form tetramers which will be floating in the cytoplasm. It is not clear how stable these tetramer particles are. Moreover it is not known what roles they play in the cell. Does nestin have to form a filamentous network to be functional? More studies are required to answer these questions. Moreover it will be interesting to over-express the mutated form of nestin along with a polymerization partner such as vimentin and examine the effects which it might have on embryonic development of the mouse. Since both nestin and vimentin are structural proteins, excess amount of them might saturate the cell and ultimately kill it due to physical limitations. Therefore one needs to control and titrate the amount of over-produced proteins to be able to obtain meaningful data out of these over-expression studies.

Our results from the loss of function approach indicated that nestin is not essential for normal development of the mouse. Nestin deficient mice did not show any obvious defects so we subsequently performed a series of histological analysis to examine potential minor defects in organs such as brain, heart and kidney. No significant difference was detected in nestin deficient mice when compared with the wild-type littermates. More detailed examination of brain with MRI also failed to detect a significant abnormality in the brains of mice lacking the expression of nestin.

Since we were not able to find an anatomical abnormality in nestin deficient mice, we sought to utilize some functional tests such as neurobehavioral examination. Interestingly, nestin deficient mice had a significantly lower performance level on roto- rod compared with their wild-type littermates. Defects detected in the roto-rod test could be due to abnormalities in the central nervous system, muscle or neuromuscular junction. Consequently we tested the nestin knockout mice for grip strength and balance but did not detect any abnormalities excluding defects in the central nervous system and muscles.

Next we sought to focus on examining the structure of NMJ. NMJ was particularly an interesting candidate as previous studies have shown co-expression of nestin and Cdk5 there [69,102]. Moreover Cdk5 is a critical factor for clustering of AChRs at the NMJ [105]. Since nestin is a modulator of Cdk5 activity in vitro, we hypothesized that the same might be true in vivo at NMJ and the phenotype we had observed could be due to deregulated activity of Cdk5. If this hypothesis is true, one would expect to see abnormalities in the clustering of AChRs at NMJ of nestin deficient mice. Our immunohistochemistry results support this hypothesis. We observed more AChR clusters in nestin deficient mice compared with the wild-type controls. Moreover these receptors occupy a larger area on the muscle and are more widespread in nestin knockout mice. To further examine the relations of nestin and Cdk5, we utilized a genetic approach. Absence of Cdk5 activity can partially rescue the clustering defect that is observed in NMJ of agrin knockout embryos. We reasoned if nestin is the regulator of

Cdk5 activity, similar phenotype should be observed in nestin/agrin double knockouts. Indeed we were able to show that same as Cdk5, absence of nestin can also rescue the clustering defects at NMJ of agrin knockout embryos. This data further supports our hypothesis that the observed defect at the NMJ of nestin deficient mice is the result of deregulated Cdk5 activity. Recent unpublished data by Yang et al. [179] also support our observations. This new study shows that nestin is required for recruitment of p35 (co- activator of Cdk5) to the muscle membrane and activation of Cdk5 at the NMJ.

Co-expression of nestin and Cdk5 has also been reported in renal podocytes. Moreover lack of Cdk5 activity causes morphological defects in cultured podocytes highlighting the importance of Cdk5 in development of renal podocytes [101]. Consequently we sought to examine the structure of podocytes in nestin deficient mice. Interestingly we observed that podocytes in nestin knockout mice do not form the organized network structure that is normally seen in the wild-types. The number of primary processes is significantly decreased in the nestin deficient mice and they form irregular branches. More studies are required to examine whether the observed renal phenotype is the result of abnormal Cdk5 activity in nestin deficient mice.

Our data from nestin/vimentin and nestin/desmin mice suggest that lack of a severe embryonic abnormality in nestin knockout mice is not due to compensation by vimentin or desmin. These double knockout mice (nestin/vimentin and nestin/desmin) are also viable and fertile with no gross defects.

Absence of nestin does not cause any major embryonic defects. This outcome is not surprising as most knockout models of IF proteins are viable and fertile [20,22,23,24,25]. However when IF knockout models are exposed to stress and challenging conditions, many of them exhibit profound phenotypes [39,40]. The same might be true for nestin. Other than examples mentioned in chapter 3 involving high protein diets and its effect on renal function and aging, experiments involving muscle and CNS regeneration are also very interesting as nestin is re-expressed in both cases. Are nestin deficient mice less efficient in healing and regenerating injured muscles? What is

the role of nestin in formation of reactive astrocytes following the brain injury? In the same context, interaction of nestin with other IF proteins can also be addressed. Vimentin, GFAP and nestin are the main IF proteins expressed in reactive astrocytes. Generating double and triple knockout mouse models with these IF and characterizing formation of reactive astrocytes in each model would provide us with valuable information regarding the interaction and possible functional over-lap of these IF proteins.

A very recent study revealed that nestin is essential for brain and eye development in zebrafish [180]. In this study, nestin was knocked down via morpholino. Consequently the number of apoptotic neural progenitor increased which lead to abnormal development of brain and eyes. Although the mechanism for this phenotype was not examined in the above study, the authors speculated that the phenotype might be the result of deregulated Cdk5 activity [180]. More studies are required to shed light on the interaction of Cdk5 and nestin in zebrafish. Interestingly knockdown of nestin in mouse neural progenitors does not increase apoptosis and has no significant effects on activity of Cdk5 under normal conditions [82]. However when cells are exposed to oxidative stress, both Cdk5 activity and rate of apoptosis elevate [82]. Therefore one can speculate that through evolution nestin’s function and its interaction with other intermediate filaments has become more complex. In zebrafish simply eliminating the expression of nestin leads to increased apoptosis while in mouse, elevated apoptosis is only observed when the organism is exposed to challenging conditions.

In summary, the data described in this thesis indicate that nestin is not essential for embryonic development of the mouse. However absence of nestin leads to precise defects in clustering of AChRs in NMJ and structure of renal podocytes. Our data suggest that the observed phenotype in NMJ is the result of deregulated Cdk5 activity. Regarding the kidney more studies are required to shed light on the mechanism responsible for the observed abnormalities in the podocytes. Examining the interaction of nestin with other intermediate filaments and other signaling molecules (other than Cdk5) is also an interesting path which needs to be explored in the future. Moreover

function of nestin under pathological conditions is an important issue which remains to be addressed.

References

1. Kovacevic N, Henderson JT, Chan E, Lifshitz N, Bishop J, Evans AC, Henkelman RM, Chen XJ (2005) A three-dimensional MRI atlas of the mouse brain with estimates of the average and variability. Cereb Cortex 15: 639-645. 2. Kim S, Coulombe PA (2007) Intermediate filament scaffolds fulfill mechanical, organizational, and signaling functions in the cytoplasm. Genes Dev 21: 1581- 1597. 3. Klymkowsky MW (1996) Intermediate filaments as dynamic structures. Cancer Metastasis Rev 15: 417-428. 4. Herrmann H, Hofmann I, Franke WW (1992) Identification of a nonapeptide motif in the vimentin head domain involved in intermediate filament assembly. J Mol Biol 223: 637-650. 5. Hendrix MJ (1996) Intermediate filaments. Cancer Metastasis Rev 15: 413-416. 6. Eriksson JE, Dechat T, Grin B, Helfand B, Mendez M, Pallari HM, Goldman RD (2009) Introducing intermediate filaments: from discovery to disease. J Clin Invest 119: 1763-1771. 7. Coulombe PA, Fuchs E (1990) Elucidating the early stages of keratin filament assembly. J Cell Biol 111: 153-169. 8. Heald R, McKeon F (1990) Mutations of phosphorylation sites in lamin A that prevent nuclear lamina disassembly in mitosis. Cell 61: 579-589. 9. Goldman RD, Gruenbaum Y, Moir RD, Shumaker DK, Spann TP (2002) Nuclear lamins: building blocks of nuclear architecture. Genes Dev 16: 533-547. 10. Dechat T, Pfleghaar K, Sengupta K, Shimi T, Shumaker DK, Solimando L, Goldman RD (2008) Nuclear lamins: major factors in the structural organization and function of the nucleus and chromatin. Genes Dev 22: 832-853. 11. Lee MK, Cleveland DW (1996) Neuronal intermediate filaments. Annu Rev Neurosci 19: 187-217. 12. Liem RK, Messing A (2009) Dysfunctions of neuronal and glial intermediate filaments in disease. J Clin Invest 119: 1814-1824. 13. Ku NO, Michie SA, Soetikno RM, Resurreccion EZ, Broome RL, Omary MB (1998) Mutation of a major keratin phosphorylation site predisposes to hepatotoxic injury in transgenic mice. J Cell Biol 143: 2023-2032. 14. Brulet P, Babinet C, Kemler R, Jacob F (1980) Monoclonal antibodies against trophectoderm-specific markers during mouse blastocyst formation. Proc Natl Acad Sci U S A 77: 4113-4117. 15. Paulin D, Babinet C, Weber K, Osborn M (1980) Antibodies as probes of cellular differentiation and cytoskeletal organization in the mouse blastocyst. Exp Cell Res 130: 297-304. 16. Jackson BW, Grund C, Schmid E, Burki K, Franke WW, Illmensee K (1980) Formation of cytoskeletal elements during mouse embryogenesis. Intermediate

filaments of the cytokeratin type and desmosomes in preimplantation embryos. Differentiation 17: 161-179. 17. Baribault H, Price J, Miyai K, Oshima RG (1993) Mid-gestational lethality in mice lacking keratin 8. Genes Dev 7: 1191-1202. 18. Magin TM, Schroder R, Leitgeb S, Wanninger F, Zatloukal K, Grund C, Melton DW (1998) Lessons from keratin 18 knockout mice: formation of novel keratin filaments, secondary loss of keratin 7 and accumulation of liver-specific keratin 8-positive aggregates. J Cell Biol 140: 1441-1451. 19. Franke WW, Schmid E, Winter S, Osborn M, Weber K (1979) Widespread occurrence of intermediate-sized filaments of the vimentin-type in cultured cells from diverse vertebrates. Exp Cell Res 123: 25-46. 20. Colucci-Guyon E, Portier MM, Dunia I, Paulin D, Pournin S, Babinet C (1994) Mice lacking vimentin develop and reproduce without an obvious phenotype. Cell 79: 679-694. 21. Chiu FC, Norton WT, Fields KL (1981) The cytoskeleton of primary astrocytes in culture contains actin, glial fibrillary acidic protein, and the fibroblast-type filament protein, vimentin. J Neurochem 37: 147-155. 22. Pekny M, Leveen P, Pekna M, Eliasson C, Berthold CH, Westermark B, Betsholtz C (1995) Mice lacking glial fibrillary acidic protein display astrocytes devoid of intermediate filaments but develop and reproduce normally. Embo J 14: 1590- 1598. 23. McCall MA, Gregg RG, Behringer RR, Brenner M, Delaney CL, Galbreath EJ, Zhang CL, Pearce RA, Chiu SY, Messing A (1996) Targeted deletion in astrocyte intermediate filament (Gfap) alters neuronal physiology. Proc Natl Acad Sci U S A 93: 6361-6366. 24. Liedtke W, Edelmann W, Bieri PL, Chiu FC, Cowan NJ, Kucherlapati R, Raine CS (1996) GFAP is necessary for the integrity of CNS white matter architecture and long-term maintenance of myelination. Neuron 17: 607-615. 25. Levavasseur F, Zhu Q, Julien JP (1999) No requirement of alpha-internexin for nervous system development and for radial growth of axons. Brain Res Mol Brain Res 69: 104-112. 26. Wojcik SM, Bundman DS, Roop DR (2000) Delayed wound healing in keratin 6a knockout mice. Mol Cell Biol 20: 5248-5255. 27. Milner DJ, Weitzer G, Tran D, Bradley A, Capetanaki Y (1996) Disruption of muscle architecture and myocardial degeneration in mice lacking desmin. J Cell Biol 134: 1255-1270. 28. Coulombe PA, Hutton ME, Letai A, Hebert A, Paller AS, Fuchs E (1991) Point mutations in human keratin 14 genes of epidermolysis bullosa simplex patients: genetic and functional analyses. Cell 66: 1301-1311. 29. Bonifas JM, Rothman AL, Epstein EH, Jr. (1991) Epidermolysis bullosa simplex: evidence in two families for keratin gene abnormalities. Science 254: 1202-1205. 30. Lane EB, Rugg EL, Navsaria H, Leigh IM, Heagerty AH, Ishida-Yamamoto A, Eady RA (1992) A mutation in the conserved helix termination peptide of keratin 5 in hereditary skin blistering. Nature 356: 244-246. 31. Omary MB (2009) "IF-pathies": a broad spectrum of intermediate filament- associated diseases. J Clin Invest 119: 1756-1762.

32. Lazarides E, Hubbard BD (1976) Immunological characterization of the subunit of the 100 A filaments from muscle cells. Proc Natl Acad Sci U S A 73: 4344-4348. 33. Small JV, Sobieszek A (1977) Studies on the function and composition of the 10- NM(100-A) filaments of vertebrate smooth muscle. J Cell Sci 23: 243-268. 34. Geisler N, Weber K (1982) The amino acid sequence of chicken muscle desmin provides a common structural model for intermediate filament proteins. Embo J 1: 1649-1656. 35. Li Z, Colucci-Guyon E, Pincon-Raymond M, Mericskay M, Pournin S, Paulin D, Babinet C (1996) Cardiovascular lesions and skeletal myopathy in mice lacking desmin. Dev Biol 175: 362-366. 36. Li Z, Mericskay M, Agbulut O, Butler-Browne G, Carlsson L, Thornell LE, Babinet C, Paulin D (1997) Desmin is essential for the tensile strength and integrity of myofibrils but not for myogenic commitment, differentiation, and fusion of skeletal muscle. J Cell Biol 139: 129-144. 37. Milner DJ, Taffet GE, Wang X, Pham T, Tamura T, Hartley C, Gerdes AM, Capetanaki Y (1999) The absence of desmin leads to cardiomyocyte hypertrophy and cardiac dilation with compromised systolic function. J Mol Cell Cardiol 31: 2063-2076. 38. Haubold KW, Allen DL, Capetanaki Y, Leinwand LA (2003) Loss of desmin leads to impaired voluntary wheel running and treadmill exercise performance. J Appl Physiol 95: 1617-1622. 39. Eckes B, Colucci-Guyon E, Smola H, Nodder S, Babinet C, Krieg T, Martin P (2000) Impaired wound healing in embryonic and adult mice lacking vimentin. J Cell Sci 113 ( Pt 13): 2455-2462. 40. Terzi F, Henrion D, Colucci-Guyon E, Federici P, Babinet C, Levy BI, Briand P, Friedlander G (1997) Reduction of renal mass is lethal in mice lacking vimentin. Role of endothelin-nitric oxide imbalance. J Clin Invest 100: 1520-1528. 41. Pekny M, Johansson CB, Eliasson C, Stakeberg J, Wallen A, Perlmann T, Lendahl U, Betsholtz C, Berthold CH, Frisen J (1999) Abnormal reaction to central nervous system injury in mice lacking glial fibrillary acidic protein and vimentin. J Cell Biol 145: 503-514. 42. Helfand BT, Chou YH, Shumaker DK, Goldman RD (2005) Intermediate filament proteins participate in signal transduction. Trends Cell Biol 15: 568-570. 43. Pallari HM, Eriksson JE (2006) Intermediate filaments as signaling platforms. Sci STKE 2006: pe53. 44. Ivaska J, Pallari HM, Nevo J, Eriksson JE (2007) Novel functions of vimentin in cell adhesion, migration, and signaling. Exp Cell Res 313: 2050-2062. 45. Sin WC, Chen XQ, Leung T, Lim L (1998) RhoA-binding kinase alpha translocation is facilitated by the collapse of the vimentin intermediate filament network. Mol Cell Biol 18: 6325-6339. 46. Hermeking H, Benzinger A (2006) 14-3-3 proteins in cell cycle regulation. Semin Cancer Biol 16: 183-192. 47. Tzivion G, Luo ZJ, Avruch J (2000) Calyculin A-induced vimentin phosphorylation sequesters 14-3-3 and displaces other 14-3-3 partners in vivo. J Biol Chem 275: 29772-29778.

48. Hockfield S, McKay RD (1985) Identification of major cell classes in the developing mammalian nervous system. J Neurosci 5: 3310-3328. 49. Lendahl U, Zimmerman LB, McKay RD (1990) CNS stem cells express a new class of intermediate filament protein. Cell 60: 585-595. 50. Doetsch F, Caille I, Lim DA, Garcia-Verdugo JM, Alvarez-Buylla A (1999) Subventricular zone astrocytes are neural stem cells in the adult mammalian brain. Cell 97: 703-716. 51. Rietze RL, Valcanis H, Brooker GF, Thomas T, Voss AK, Bartlett PF (2001) Purification of a pluripotent neural stem cell from the adult mouse brain. Nature 412: 736-739. 52. Fukuda S, Kato F, Tozuka Y, Yamaguchi M, Miyamoto Y, Hisatsune T (2003) Two distinct subpopulations of nestin-positive cells in adult mouse dentate gyrus. J Neurosci 23: 9357-9366. 53. Marvin MJ, Dahlstrand J, Lendahl U, McKay RD (1998) A rod end deletion in the intermediate filament protein nestin alters its subcellular localization in neuroepithelial cells of transgenic mice. J Cell Sci 111 ( Pt 14): 1951-1961. 54. Sejersen T, Lendahl U (1993) Transient expression of the intermediate filament nestin during skeletal muscle development. J Cell Sci 106 ( Pt 4): 1291-1300. 55. Kachinsky AM, Dominov JA, Miller JB (1995) Intermediate filaments in cardiac myogenesis: nestin in the developing mouse heart. J Histochem Cytochem 43: 843-847. 56. Baldwin HS, Jensen KL, Solursh M (1991) Myogenic cytodifferentiation of the precardiac mesoderm in the rat. Differentiation 47: 163-172. 57. Wroblewski J, Engstrom M, Edwall-Arvidsson C, Sjoberg G, Sejersen T, Lendahl U (1997) Distribution of nestin in the developing mouse limb bud in vivo and in micro-mass cultures of cells isolated from limb buds. Differentiation 61: 151-159. 58. Li L, Mignone J, Yang M, Matic M, Penman S, Enikolopov G, Hoffman RM (2003) Nestin expression in hair follicle sheath progenitor cells. Proc Natl Acad Sci U S A 100: 9958-9961. 59. Yang J, Bian W, Gao X, Chen L, Jing N (2000) Nestin expression during mouse eye and lens development. Mech Dev 94: 287-291. 60. Terling C, Rass A, Mitsiadis TA, Fried K, Lendahl U, Wroblewski J (1995) Expression of the intermediate filament nestin during rodent tooth development. Int J Dev Biol 39: 947-956. 61. Lyle S, Christofidou-Solomidou M, Liu Y, Elder DE, Albelda S, Cotsarelis G (1999) Human hair follicle bulge cells are biochemically distinct and possess an epithelial stem cell phenotype. J Investig Dermatol Symp Proc 4: 296-301. 62. Kuhn HG, Dickinson-Anson H, Gage FH (1996) Neurogenesis in the dentate gyrus of the adult rat: age-related decrease of neuronal progenitor proliferation. J Neurosci 16: 2027-2033. 63. Gage FH (2000) Mammalian neural stem cells. Science 287: 1433-1438. 64. Reynolds BA, Weiss S (1992) Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science 255: 1707-1710. 65. Morshead CM, Reynolds BA, Craig CG, McBurney MW, Staines WA, Morassutti D, Weiss S, van der Kooy D (1994) Neural stem cells in the adult mammalian

forebrain: a relatively quiescent subpopulation of subependymal cells. Neuron 13: 1071-1082. 66. Mignone JL, Kukekov V, Chiang AS, Steindler D, Enikolopov G (2004) Neural stem and progenitor cells in nestin-GFP transgenic mice. J Comp Neurol 469: 311-324. 67. Wagner N, Wagner KD, Scholz H, Kirschner KM, Schedl A (2006) Intermediate filament protein nestin is expressed in developing kidney and heart and might be regulated by the Wilms' tumor suppressor Wt1. Am J Physiol Regul Integr Comp Physiol 291: R779-787. 68. Kang H, Tian L, Son YJ, Zuo Y, Procaccino D, Love F, Hayworth C, Trachtenberg J, Mikesh M, Sutton L, Ponomareva O, Mignone J, Enikolopov G, Rimer M, Thompson W (2007) Regulation of the intermediate filament protein nestin at rodent neuromuscular junctions by innervation and activity. J Neurosci 27: 5948- 5957. 69. Vaittinen S, Lukka R, Sahlgren C, Rantanen J, Hurme T, Lendahl U, Eriksson JE, Kalimo H (1999) Specific and innervation-regulated expression of the intermediate filament protein nestin at neuromuscular and myotendinous junctions in skeletal muscle. Am J Pathol 154: 591-600. 70. Frisen J, Johansson CB, Torok C, Risling M, Lendahl U (1995) Rapid, widespread, and longlasting induction of nestin contributes to the generation of glial scar tissue after CNS injury. J Cell Biol 131: 453-464. 71. Krum JM, Rosenstein JM (1999) Transient coexpression of nestin, GFAP, and vascular endothelial growth factor in mature reactive astroglia following neural grafting or brain wounds. Exp Neurol 160: 348-360. 72. Yu TS, Zhang G, Liebl DJ, Kernie SG (2008) Traumatic brain injury-induced hippocampal neurogenesis requires activation of early nestin-expressing progenitors. J Neurosci 28: 12901-12912. 73. Cheng L, Jin Z, Liu L, Yan Y, Li T, Zhu X, Jing N (2004) Characterization and promoter analysis of the mouse nestin gene. FEBS Lett 565: 195-202. 74. Zimmerman L, Parr B, Lendahl U, Cunningham M, McKay R, Gavin B, Mann J, Vassileva G, McMahon A (1994) Independent regulatory elements in the nestin gene direct transgene expression to neural stem cells or muscle precursors. Neuron 12: 11-24. 75. Lothian C, Lendahl U (1997) An evolutionarily conserved region in the second intron of the human nestin gene directs gene expression to CNS progenitor cells and to early neural crest cells. Eur J Neurosci 9: 452-462. 76. Kawaguchi A, Miyata T, Sawamoto K, Takashita N, Murayama A, Akamatsu W, Ogawa M, Okabe M, Tano Y, Goldman SA, Okano H (2001) Nestin-EGFP transgenic mice: visualization of the self-renewal and multipotency of CNS stem cells. Mol Cell Neurosci 17: 259-273. 77. Yu TS, Dandekar M, Monteggia LM, Parada LF, Kernie SG (2005) Temporally regulated expression of Cre recombinase in neural stem cells. Genesis 41: 147- 153. 78. Tronche F, Kellendonk C, Kretz O, Gass P, Anlag K, Orban PC, Bock R, Klein R, Schutz G (1999) Disruption of the glucocorticoid receptor gene in the nervous system results in reduced anxiety. Nat Genet 23: 99-103.

79. Trumpp A, Depew MJ, Rubenstein JL, Bishop JM, Martin GR (1999) Cre-mediated gene inactivation demonstrates that FGF8 is required for cell survival and patterning of the first branchial arch. Genes Dev 13: 3136-3148. 80. Imayoshi I, Ohtsuka T, Metzger D, Chambon P, Kageyama R (2006) Temporal regulation of Cre recombinase activity in neural stem cells. Genesis 44: 233-238. 81. Sahlgren CM, Mikhailov A, Vaittinen S, Pallari HM, Kalimo H, Pant HC, Eriksson JE (2003) Cdk5 regulates the organization of Nestin and its association with p35. Mol Cell Biol 23: 5090-5106. 82. Sahlgren CM, Pallari HM, He T, Chou YH, Goldman RD, Eriksson JE (2006) A nestin scaffold links Cdk5/p35 signaling to oxidant-induced cell death. Embo J 25: 4808-4819. 83. Ishiguro K, Kobayashi S, Omori A, Takamatsu M, Yonekura S, Anzai K, Imahori K, Uchida T (1994) Identification of the 23 kDa subunit of tau protein kinase II as a putative activator of cdk5 in bovine brain. FEBS Lett 342: 203-208. 84. Paudel HK (1997) Phosphorylation by neuronal cdc2-like protein kinase promotes dimerization of Tau protein in vitro. J Biol Chem 272: 28328-28334. 85. Lee KY, Johnston RN (1997) Neurofilaments are part of the high molecular weight complex containing neuronal cdc2-like kinase (nclk). Brain Res 773: 197-202. 86. Lew J, Winkfein RJ, Paudel HK, Wang JH (1992) Brain proline-directed protein kinase is a neurofilament kinase which displays high sequence homology to p34cdc2. J Biol Chem 267: 25922-25926. 87. Pant AC, Veeranna, Pant HC, Amin N (1997) Phosphorylation of human high molecular weight neurofilament protein (hNF-H) by neuronal cyclin-dependent kinase 5 (cdk5). Brain Res 765: 259-266. 88. Sahlgren CM, Mikhailov A, Hellman J, Chou YH, Lendahl U, Goldman RD, Eriksson JE (2001) Mitotic reorganization of the intermediate filament protein nestin involves phosphorylation by cdc2 kinase. J Biol Chem 276: 16456-16463. 89. O'Hare MJ, Kushwaha N, Zhang Y, Aleyasin H, Callaghan SM, Slack RS, Albert PR, Vincent I, Park DS (2005) Differential roles of nuclear and cytoplasmic cyclin- dependent kinase 5 in apoptotic and excitotoxic neuronal death. J Neurosci 25: 8954-8966. 90. Meyerson M, Enders GH, Wu CL, Su LK, Gorka C, Nelson C, Harlow E, Tsai LH (1992) A family of human cdc2-related protein kinases. Embo J 11: 2909-2917. 91. Hellmich MR, Pant HC, Wada E, Battey JF (1992) Neuronal cdc2-like kinase: a cdc2-related protein kinase with predominantly neuronal expression. Proc Natl Acad Sci U S A 89: 10867-10871. 92. Tsai LH, Takahashi T, Caviness VS, Jr., Harlow E (1993) Activity and expression pattern of cyclin-dependent kinase 5 in the embryonic mouse nervous system. Development 119: 1029-1040. 93. Nikolic M, Dudek H, Kwon YT, Ramos YF, Tsai LH (1996) The cdk5/p35 kinase is essential for neurite outgrowth during neuronal differentiation. Genes Dev 10: 816-825. 94. Ohshima T, Ward JM, Huh CG, Longenecker G, Veeranna, Pant HC, Brady RO, Martin LJ, Kulkarni AB (1996) Targeted disruption of the cyclin-dependent kinase 5 gene results in abnormal corticogenesis, neuronal pathology and perinatal death. Proc Natl Acad Sci U S A 93: 11173-11178.

95. Gilmore EC, Ohshima T, Goffinet AM, Kulkarni AB, Herrup K (1998) Cyclin- dependent kinase 5-deficient mice demonstrate novel developmental arrest in cerebral cortex. J Neurosci 18: 6370-6377. 96. Hirasawa M, Ohshima T, Takahashi S, Longenecker G, Honjo Y, Veeranna, Pant HC, Mikoshiba K, Brady RO, Kulkarni AB (2004) Perinatal abrogation of Cdk5 expression in brain results in neuronal migration defects. Proc Natl Acad Sci U S A 101: 6249-6254. 97. Delalle I, Bhide PG, Caviness VS, Jr., Tsai LH (1997) Temporal and spatial patterns of expression of p35, a regulatory subunit of cyclin-dependent kinase 5, in the nervous system of the mouse. J Neurocytol 26: 283-296. 98. Chae T, Kwon YT, Bronson R, Dikkes P, Li E, Tsai LH (1997) Mice lacking p35, a neuronal specific activator of Cdk5, display cortical lamination defects, seizures, and adult lethality. Neuron 18: 29-42. 99. Tang D, Yeung J, Lee KY, Matsushita M, Matsui H, Tomizawa K, Hatase O, Wang JH (1995) An isoform of the neuronal cyclin-dependent kinase 5 (Cdk5) activator. J Biol Chem 270: 26897-26903. 100. Ko J, Humbert S, Bronson RT, Takahashi S, Kulkarni AB, Li E, Tsai LH (2001) p35 and p39 are essential for cyclin-dependent kinase 5 function during neurodevelopment. J Neurosci 21: 6758-6771. 101. Griffin SV, Hiromura K, Pippin J, Petermann AT, Blonski MJ, Krofft R, Takahashi S, Kulkarni AB, Shankland SJ (2004) Cyclin-dependent kinase 5 is a regulator of podocyte differentiation, proliferation, and morphology. Am J Pathol 165: 1175- 1185. 102. Fu AK, Fu WY, Cheung J, Tsim KW, Ip FC, Wang JH, Ip NY (2001) Cdk5 is involved in neuregulin-induced AChR expression at the neuromuscular junction. Nat Neurosci 4: 374-381. 103. Kriz W (2002) Podocyte is the major culprit accounting for the progression of chronic renal disease. Microsc Res Tech 57: 189-195. 104. Greer JJ, Allan DW, Martin-Caraballo M, Lemke RP (1999) An overview of phrenic nerve and diaphragm muscle development in the perinatal rat. J Appl Physiol 86: 779-786. 105. Fu AK, Ip FC, Fu WY, Cheung J, Wang JH, Yung WH, Ip NY (2005) Aberrant motor axon projection, acetylcholine receptor clustering, and neurotransmission in cyclin-dependent kinase 5 null mice. Proc Natl Acad Sci U S A 102: 15224- 15229. 106. Lin W, Dominguez B, Yang J, Aryal P, Brandon EP, Gage FH, Lee KF (2005) Neurotransmitter acetylcholine negatively regulates neuromuscular synapse formation by a Cdk5-dependent mechanism. Neuron 46: 569-579. 107. Tsen G, Halfter W, Kroger S, Cole GJ (1995) Agrin is a heparan sulfate proteoglycan. J Biol Chem 270: 3392-3399. 108. Wallace BG (1989) Agrin-induced specializations contain cytoplasmic, membrane, and extracellular matrix-associated components of the postsynaptic apparatus. J Neurosci 9: 1294-1302. 109. Magill-Solc C, McMahan UJ (1988) Motor neurons contain agrin-like molecules. J Cell Biol 107: 1825-1833.

110. Fallon JR, Gelfman CE (1989) Agrin-related molecules are concentrated at acetylcholine receptor clusters in normal and aneural developing muscle. J Cell Biol 108: 1527-1535. 111. Godfrey EW, Nitkin RM, Wallace BG, Rubin LL, McMahan UJ (1984) Components of Torpedo electric organ and muscle that cause aggregation of acetylcholine receptors on cultured muscle cells. J Cell Biol 99: 615-627. 112. Gautam M, Noakes PG, Moscoso L, Rupp F, Scheller RH, Merlie JP, Sanes JR (1996) Defective neuromuscular synaptogenesis in agrin-deficient mutant mice. Cell 85: 525-535. 113. Burgess RW, Nguyen QT, Son YJ, Lichtman JW, Sanes JR (1999) Alternatively spliced isoforms of nerve- and muscle-derived agrin: their roles at the neuromuscular junction. Neuron 23: 33-44. 114. Yang X, Arber S, William C, Li L, Tanabe Y, Jessell TM, Birchmeier C, Burden SJ (2001) Patterning of muscle acetylcholine receptor gene expression in the absence of motor innervation. Neuron 30: 399-410. 115. Herbst R, Avetisova E, Burden SJ (2002) Restoration of synapse formation in Musk mutant mice expressing a Musk/Trk chimeric receptor. Development 129: 5449- 5460. 116. Lin W, Burgess RW, Dominguez B, Pfaff SL, Sanes JR, Lee KF (2001) Distinct roles of nerve and muscle in postsynaptic differentiation of the neuromuscular synapse. Nature 410: 1057-1064. 117. George SH, Gertsenstein M, Vintersten K, Korets-Smith E, Murphy J, Stevens ME, Haigh JJ, Nagy A (2007) Developmental and adult phenotyping directly from mutant embryonic stem cells. Proc Natl Acad Sci U S A 104: 4455-4460. 118. Andras Nagy MG, Kristina Vintersten and Richard Behringer (2003) Manipulating the Mouse Embryo: A Laboratory Manual, Third Edition 119. Smukler SR, Runciman SB, Xu S, van der Kooy D (2006) Embryonic stem cells assume a primitive neural stem cell fate in the absence of extrinsic influences. J Cell Biol 172: 79-90. 120. Tropepe V, Sibilia M, Ciruna BG, Rossant J, Wagner EF, van der Kooy D (1999) Distinct neural stem cells proliferate in response to EGF and FGF in the developing mouse telencephalon. Dev Biol 208: 166-188. 121. Wiese C, Rolletschek A, Kania G, Blyszczuk P, Tarasov KV, Tarasova Y, Wersto RP, Boheler KR, Wobus AM (2004) Nestin expression--a property of multi- lineage progenitor cells? Cell Mol Life Sci 61: 2510-2522. 122. Liu P, Jenkins NA, Copeland NG (2003) A highly efficient recombineering-based method for generating conditional knockout mutations. Genome Res 13: 476-484. 123. Tybulewicz VL, Crawford CE, Jackson PK, Bronson RT, Mulligan RC (1991) Neonatal lethality and lymphopenia in mice with a homozygous disruption of the c-abl proto-oncogene. Cell 65: 1153-1163. 124. Nagy A, Rossant J, Nagy R, Abramow-Newerly W, Roder JC (1993) Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc Natl Acad Sci U S A 90: 8424-8428. 125. Dorr AE, Lerch JP, Spring S, Kabani N, Henkelman RM (2008) High resolution three-dimensional brain atlas using an average magnetic resonance image of 40 adult C57Bl/6J mice. Neuroimage 42: 60-69.

126. Collins DL, Neelin P, Peters TM, Evans AC (1994) Automatic 3D intersubject registration of MR volumetric data in standardized Talairach space. J Comput Assist Tomogr 18: 192-205. 127. Lerch JP, Carroll JB, Dorr A, Spring S, Evans AC, Hayden MR, Sled JG, Henkelman RM (2008) Cortical thickness measured from MRI in the YAC128 mouse model of Huntington's disease. Neuroimage 41: 243-251. 128. Kuhn PL, Petroulakis E, Zazanis GA, McKinnon RD (1995) Motor function analysis of myelin mutant mice using a rotarod. Int J Dev Neurosci 13: 715-722. 129. Yamaguchi M, Saito H, Suzuki M, Mori K (2000) Visualization of neurogenesis in the central nervous system using nestin promoter-GFP transgenic mice. Neuroreport 11: 1991-1996. 130. Doetsch F, Garcia-Verdugo JM, Alvarez-Buylla A (1997) Cellular composition and three-dimensional organization of the subventricular germinal zone in the adult mammalian brain. J Neurosci 17: 5046-5061. 131. Kaplan MP, Chin SS, Fliegner KH, Liem RK (1990) Alpha-internexin, a novel neuronal intermediate filament protein, precedes the low molecular weight neurofilament protein (NF-L) in the developing rat brain. J Neurosci 10: 2735- 2748. 132. Evans MJ, Kaufman MH (1981) Establishment in culture of pluripotential cells from mouse embryos. Nature 292: 154-156. 133. Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, Jones JM (1998) Embryonic stem cell lines derived from human blastocysts. Science 282: 1145-1147. 134. Wobus AM, Holzhausen H, Jakel P, Schoneich J (1984) Characterization of a pluripotent stem cell line derived from a mouse embryo. Exp Cell Res 152: 212- 219. 135. Doetschman TC, Eistetter H, Katz M, Schmidt W, Kemler R (1985) The in vitro development of blastocyst-derived embryonic stem cell lines: formation of visceral yolk sac, blood islands and myocardium. J Embryol Exp Morphol 87: 27- 45. 136. Wobus AM (2001) Potential of embryonic stem cells. Mol Aspects Med 22: 149- 164. 137. Wobus AM, Wallukat G, Hescheler J (1991) Pluripotent mouse embryonic stem cells are able to differentiate into cardiomyocytes expressing chronotropic responses to adrenergic and cholinergic agents and Ca2+ channel blockers. Differentiation 48: 173-182. 138. Wobus AM, Kaomei G, Shan J, Wellner MC, Rohwedel J, Ji G, Fleischmann B, Katus HA, Hescheler J, Franz WM (1997) Retinoic acid accelerates embryonic stem cell-derived cardiac differentiation and enhances development of ventricular cardiomyocytes. J Mol Cell Cardiol 29: 1525-1539. 139. Boheler KR, Czyz J, Tweedie D, Yang HT, Anisimov SV, Wobus AM (2002) Differentiation of pluripotent embryonic stem cells into cardiomyocytes. Circ Res 91: 189-201. 140. Czyz J, Wiese C, Rolletschek A, Blyszczuk P, Cross M, Wobus AM (2003) Potential of embryonic and adult stem cells in vitro. Biol Chem 384: 1391-1409.

141. Wobus AM, Guan K, Yang HT, Boheler KR (2002) Embryonic stem cells as a model to study cardiac, skeletal muscle, and vascular smooth muscle cell differentiation. Methods Mol Biol 185: 127-156. 142. Rolletschek A, Chang H, Guan K, Czyz J, Meyer M, Wobus AM (2001) Differentiation of embryonic stem cell-derived dopaminergic neurons is enhanced by survival-promoting factors. Mech Dev 105: 93-104. 143. Blyszczuk P, Czyz J, Kania G, Wagner M, Roll U, St-Onge L, Wobus AM (2003) Expression of Pax4 in embryonic stem cells promotes differentiation of nestin- positive progenitor and insulin-producing cells. Proc Natl Acad Sci U S A 100: 998-1003. 144. Kania G, Blyszczuk P, Czyz J, Navarrete-Santos A, Wobus AM (2003) Differentiation of mouse embryonic stem cells into pancreatic and hepatic cells. Methods Enzymol 365: 287-303. 145. Niwa H, Yamamura K, Miyazaki J (1991) Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108: 193-199. 146. Sauer B (1987) Functional expression of the cre-lox site-specific recombination system in the yeast Saccharomyces cerevisiae. Mol Cell Biol 7: 2087-2096. 147. Sauer B, Henderson N (1988) Site-specific DNA recombination in mammalian cells by the Cre recombinase of bacteriophage P1. Proc Natl Acad Sci U S A 85: 5166- 5170. 148. Hadjantonakis AK, Nagy A (2000) FACS for the isolation of individual cells from transgenic mice harboring a fluorescent protein reporter. Genesis 27: 95-98. 149. Hadjantonakis AK, Macmaster S, Nagy A (2002) Embryonic stem cells and mice expressing different GFP variants for multiple non-invasive reporter usage within a single animal. BMC Biotechnol 2: 11. 150. Amano T, Gertsenstein M, Nagy A, Kurihara H, Suzuki H (2006) Nuclear transfer reprogramming does not improve the low developmental potency of embryonic stem cells induced by long-term culture. Reproduction 132: 257-263. 151. Lewandoski M, Martin GR (1997) Cre-mediated chromosome loss in mice. Nat Genet 17: 223-225. 152. Nagy A (2000) Cre recombinase: the universal reagent for genome tailoring. Genesis 26: 99-109. 153. Atkinson S, Armstrong L (2008) Epigenetics in embryonic stem cells: regulation of pluripotency and differentiation. Cell Tissue Res 331: 23-29. 154. Alberts B JA, Lewis J, Raff M, Roberts K and Wlater P (2007) Molecular Biology of the Cell. 155. Weaver R (2007) Molecular Biology: post-transcriptional events. Boston: McGraw Hill Higher Education. 156. Fliegner KH, Kaplan MP, Wood TL, Pintar JE, Liem RK (1994) Expression of the gene for the neuronal intermediate filament protein alpha-internexin coincides with the onset of neuronal differentiation in the developing rat nervous system. J Comp Neurol 342: 161-173. 157. Maslov AY, Barone TA, Plunkett RJ, Pruitt SC (2004) Neural stem cell detection, characterization, and age-related changes in the subventricular zone of mice. J Neurosci 24: 1726-1733.

158. Frojdman K, Pelliniemi LJ, Lendahl U, Virtanen I, Eriksson JE (1997) The intermediate filament protein nestin occurs transiently in differentiating testis of rat and mouse. Differentiation 61: 243-249. 159. Chen J, Boyle S, Zhao M, Su W, Takahashi K, Davis L, Decaestecker M, Takahashi T, Breyer MD, Hao CM (2006) Differential expression of the intermediate filament protein nestin during renal development and its localization in adult podocytes. J Am Soc Nephrol 17: 1283-1291. 160. van de Klundert FA, Raats JM, Bloemendal H (1993) Intermediate filaments: regulation of gene expression and assembly. Eur J Biochem 214: 351-366. 161. Franke WW (1987) Nuclear lamins and cytoplasmic intermediate filament proteins: a growing multigene family. Cell 48: 3-4. 162. Hisanaga S, Hirokawa N (1990) Molecular architecture of the neurofilament. II. Reassembly process of neurofilament L protein in vitro. J Mol Biol 211: 871-882. 163. Esue O, Carson AA, Tseng Y, Wirtz D (2006) A direct interaction between actin and vimentin filaments mediated by the tail domain of vimentin. J Biol Chem 281: 30393-30399. 164. Parry DA, Strelkov SV, Burkhard P, Aebi U, Herrmann H (2007) Towards a molecular description of intermediate filament structure and assembly. Exp Cell Res 313: 2204-2216. 165. Godsel LM, Hobbs RP, Green KJ (2008) Intermediate filament assembly: dynamics to disease. Trends Cell Biol 18: 28-37. 166. Milner DJ, Mavroidis M, Weisleder N, Capetanaki Y (2000) Desmin cytoskeleton linked to muscle mitochondrial distribution and respiratory function. J Cell Biol 150: 1283-1298. 167. Kleeberger W, Bova GS, Nielsen ME, Herawi M, Chuang AY, Epstein JI, Berman DM (2007) Roles for the stem cell associated intermediate filament Nestin in prostate cancer migration and metastasis. Cancer Res 67: 9199-9206. 168. Dhavan R, Tsai LH (2001) A decade of CDK5. Nat Rev Mol Cell Biol 2: 749-759. 169. Tsai LH, Delalle I, Caviness VS, Jr., Chae T, Harlow E (1994) p35 is a neural- specific regulatory subunit of cyclin-dependent kinase 5. Nature 371: 419-423. 170. Han SH, Morgan JI (1999) Comparison of munc-18 and cdk5 expression in the nervous system during mouse embryogenesis. Neuroscience 94: 955-964. 171. Ohshima T, Gilmore EC, Longenecker G, Jacobowitz DM, Brady RO, Herrup K, Kulkarni AB (1999) Migration defects of cdk5(-/-) neurons in the developing cerebellum is cell autonomous. J Neurosci 19: 6017-6026. 172. Hirsch NP (2007) Neuromuscular junction in health and disease. Br J Anaesth 99: 132-138. 173. Eliasson C, Sahlgren C, Berthold CH, Stakeberg J, Celis JE, Betsholtz C, Eriksson JE, Pekny M (1999) Intermediate filament protein partnership in astrocytes. J Biol Chem 274: 23996-24006. 174. Bellin RM, Sernett SW, Becker B, Ip W, Huiatt TW, Robson RM (1999) Molecular characteristics and interactions of the intermediate filament protein synemin. Interactions with alpha-actinin may anchor synemin-containing heterofilaments. J Biol Chem 274: 29493-29499. 175. Li L, Lundkvist A, Andersson D, Wilhelmsson U, Nagai N, Pardo AC, Nodin C, Stahlberg A, Aprico K, Larsson K, Yabe T, Moons L, Fotheringham A, Davies I,

Carmeliet P, Schwartz JP, Pekna M, Kubista M, Blomstrand F, Maragakis N, Nilsson M, Pekny M (2008) Protective role of reactive astrocytes in brain ischemia. J Cereb Blood Flow Metab 28: 468-481. 176. Yang J DB, Fred de Winter F, Eriksson J and Lee K-F (submitted) Nestin negatively regulates postsynaptic differentiation of the neuromuscular synapse. 177. Steinert PM, Chou YH, Prahlad V, Parry DA, Marekov LN, Wu KC, Jang SI, Goldman RD (1999) A high molecular weight intermediate filament-associated protein in BHK-21 cells is nestin, a type VI intermediate filament protein. Limited co-assembly in vitro to form heteropolymers with type III vimentin and type IV alpha-internexin. J Biol Chem 274: 9881-9890. 178. Cizkova D, Soukup T, Mokry J (2009) Nestin expression reflects formation, revascularization and reinnervation of new myofibers in regenerating rat hind limb skeletal muscles. Cells Tissues Organs 189: 338-347. 179. Yang J D, de Winter F, Eriksson J and Lee K-F (submitted) Nestin negatively regulates postsynaptic differentiation of the neuromuscular synapse. 180. Chen HL, Yuh CH, Wu KK Nestin Is Essential for Zebrafish Brain and Eye Development through Control of Progenitor Cell Apoptosis. PLoS One 5: e9318.

100