*In Vivo* **and** *In Vitro* **Models to Study Amyotrophic Lateral Sclerosis**

François Berthod and François Gros-Louis

*Centre LOEX de l'Université Laval, Centre de recherche FRSQ du Centre hospitalier affilié universitaire de Québec, Département de Chirurgie, Faculté de Médecine, Université Laval, Québec, Canada* 

### **1. Introduction**

80 Amyotrophic Lateral Sclerosis

Rumfeldt, J. A., J. R. Lepock, et al. (2009). "Unfolding and folding kinetics of amyotrophic

Shi, P., J. Gal, et al. "Mitochondrial dysfunction in amyotrophic lateral sclerosis." *Biochim* 

Stathopulos, P. B., J. A. Rumfeldt, et al. (2003). "Cu/Zn superoxide dismutase mutants

Stieber, A., J. O. Gonatas, et al. (2000). "Aggregation of ubiquitin and a mutant ALS-linked

Sumi, H., S. Nagano, et al. (2006). "Inverse correlation between the formation of

Teuchert, M., D. Fischer, et al. (2006). "A dynein mutation attenuates motor neuron

Tiwari, A., Z. Xu, et al. (2005). "Aberrantly increased hydrophobicity shared by mutants of

Trelle, S., S. Reichenbach, et al. (2011). "Cardiovascular safety of non-steroidal anti-

Tokuda, E., S. Ono, et al. (2008). "Ammonium tetrathiomolybdate delays onset, prolongs

Urushitani, M., J. Kurisu, et al. (2002). "Proteasomal inhibition by misfolded mutant

Van Damme, P., M. Leyssen, et al. (2003). "The AMPA receptor antagonist NBQX prolongs

Vukosavic, S., M. Dubois-Dauphin, et al. (1999). "Bax and Bcl-2 interaction in a transgenic

Watanabe, M., M. Dykes-Hoberg, et al. (2001). "Histological evidence of protein aggregation

Wong, N. K., B. P. He, et al. (2000). "Characterization of neuronal intermediate filament

Wood, J. D., T. P. Beaujeux, et al. (2003). "Protein aggregation in motor neurone disorders."

Xu, Z., C. Jung, et al. (2004). "Mitochondrial degeneration in amyotrophic lateral sclerosis." *J* 

superoxide-dismutase-transgenic mice." *Acta Neuropathol* 112(1): 52-63. Tanaka, F., J. Niwa, et al. (2006). "Gene expression profiling toward understanding of ALS

aggregates in vitro." *Proc Natl Acad Sci U S A* 100(12): 7021-7026.

degeneration in SOD1(G93A) mice." *Exp Neurol* 198(1): 271-274.

inflammatory drugs: network meta-analysis." *BMJ* 342: c7086.

amyotrophic lateral sclerosis." *J Neurochem* 83(5): 1030-1042.

sclerosis (ALS)." *J Neuropathol Exp Neurol* 59(11): 972-982.

278-298.

*Biophys Acta* 1802(1): 45-51.

280(33): 29771-29779.

343(2): 81-84.

2468.

apparatus." *J Neurol Sci* 173(1): 53-62.

pathogenesis." *Ann N Y Acad Sci* 1086: 1-10.

lateral sclerosis." *Exp Neurol* 213(1): 122-128.

tissues." *Neurobiol Dis* 8(6): 933-941.

*Bioenerg Biomembr* 36(4): 395-399.

*Neuropathol Appl Neurobiol* 29(6): 529-545.

lateral sclerosis-associated mutant Cu,Zn superoxide dismutases." *J Mol Biol* 385(1):

associated with amyotrophic lateral sclerosis show enhanced formation of

SOD1 protein correlate with disease progression and fragmentation of the Golgi

mitochondria-derived vacuoles and Lewy-body-like hyaline inclusions in G93A

Cu,Zn-superoxide dismutase in familial amyotrophic lateral sclerosis." *J Biol Chem*

survival, and slows progression of disease in a mouse model for amyotrophic

superoxide dismutase 1 induces selective motor neuron death in familial

survival in a transgenic mouse model of amyotrophic lateral sclerosis." *Neurosci Lett*

mouse model of familial amyotrophic lateral sclerosis." *J Neurochem* 73(6): 2460-

in mutant SOD1 transgenic mice and in amyotrophic lateral sclerosis neural

protein expression in cervical spinal motor neurons in sporadic amyotrophic lateral

Amyotrophic Lateral Sclerosis (ALS) is the most common adult-onset neurodegenerative disorder characterized by the death of large motor neurons in the cerebral cortex and spinal cord (Tandan and Bradley, 1985). Dysfunction and death of these cell populations lead to progressive muscle weakness, atrophy, fasciculations, spasticity and ultimately, paralysis and death usually within 3 to 5 years after disease onset (Mulder, 1982). The estimated worldwide incidence for this disease is around 2 per 100,000 in the general population and the life-long risk to develop ALS is approximately 1:2000. The disease occurs in sporadic (90%) and familial forms (10%) (Gros-Louis, et al., 2006). With the exception of few FALS cases in which other neurodegenerative disorders can simultaneously occur, FALS and SALS are clinically indistinguishable. To date, mutations in the Cu/Zn superoxide dismutase 1 (*SOD1*) gene have remained the major known genetic causes associated with ALS. However, the mechanism whereby mutant SOD1 causes specific degeneration of motor neurons remains unclear. Nonetheless, many neuronal death pathways have been revealed through studies with transgenic mice expressing SOD1 mutants. Other vertebrate, invertebrate and *in vitro* models of ALS have also been described. Here, we will review various animal and cellular models that have been used to study the toxicity of ALS-linked gene mutations and also to investigate pathological hallmarks of the disease.

### **1.1 Familial ALS**

Though most cases of ALS are sporadic, 10% of cases have affected relatives, some with clear Mendelian inheritance and high penetrance (Gros-Louis, et al., 2006). The landmark discovery in 1993 of missense mutations in the *SOD1* gene in subsets of familial cases directed most ALS research to elucidate the mechanism of SOD1-mediadted disease (Rosen, et al., 1993). More recently, mutations in two other genes, *TARDBP* and *FUS/TLS* have been found in ALS patients (Kabashi, et al., 2008; Sreedharan, et al., 2008; Vance, et al., 2009). Rare mutations in other genes such as *ANG*, *ALS2*, *DCTN1*, *MAPT*, *SETX* and *VAPB* have also been described (for reviews see (Gros-Louis, et al., 2006)). Various other genetic mutations in the *ELP3* ((Simpson, et al., 2009), *FIG4* (Chow, et al., 2009), *DAO* (Mitchell, et al., 2010), *OPTN* (Maruyama, et al., 2010) and *CHGB* (Gros-Louis, et al., 2009a) genes have also been

*In Vivo* and *In Vitro* Models to Study Amyotrophic Lateral Sclerosis 83

For most cases of ALS, the causes are unknown. A genetic component is also thought to contribute to the pathogenesis of sporadic ALS, which accounts for the majority of ALS cases. However, identification of gene mutations associated with SALS has met with limited success so far. Several groups have reported on gene variants and association studies found in individuals with sporadic ALS, each accounting for a small number of the total cases reflecting a complex pattern of inheritance with very low penetrance, a high degree of heterogeneity and/or the existence of environmental factors predisposing to ALS. Because familial and sporadic forms of the disease are clinically and pathologically similar, understanding the familial form will shed light on possible epidemiological and

Many divergent data generated by different research groups who have studied different populations around the world have been published. The use of an adequate size population with appropriate control individuals is needed in order to reach enough statistical power and to observe any significant results. Homogeneity regarding ethnical origin, age, age of onset, disease duration, site of the first symptoms and gender among the studied populations is essential to design a good and robust genetic study. The requirement for larger sample sizes and possibly more sensitive and efficient analytic tools allowing for reliable detection across studied populations are now becoming more and more available as

The identifications of different genes and genetic risk factors associated with ALS over the past years have highlighted common molecular pathways mainly involving intracellular cell trafficking and RNA metabolism. To what extent these pathways are implicated in ALS remain to be determined and studied further. However, it is interesting to note that genes such as *ALS2*, *VAPB*, *MAPT*, *DCTN1*, *EAAT2*, *NEFH*, *PRPH, FIG4, CNTF*, *CHGB* and *OPTN* in which mutations have been found in a subset of FALS and/or SALS patients, are all related to intracellular trafficking either via axonal transport, vesicle docking and transport, or microtubule and neurofilament stabilization. Furthermore, this common theme is also seen in other motor neuron diseases and neurodegenerative disorders such as hereditary spastic paraplegia, Parkinson's disease, spinal muscular atrophy and Charcot-Marie-tooth disease. It is also noteworthy that RNA metabolism via alternative splicing abnormalities or RNA binding is also a common theme observed amongst the different ALS related genes such as *ALS2*, *MAPT*, *EAAT2*, *PRPH*, *GluR2*, *CNTF*, *SETX*, *SMN1, SMN2, TARDBP and FUS/TLS.* The high degree of genetic heterogeneity observed in ALS suggests that an individual phenotype could result from the sum of several contributing gene defects and/or epigenetic influences, which individually do not cause disease, and may explain the difficulties in identifying genes associated with SALS. The identification of other genes associated with motor neuron disease, as well as the determination of genetic and/or environmental factors that predispose to SALS is crucial in the development of novel therapies for ALS. The generation of cellular or animal models to study SALS, in which the disease can be closely replicated, would also be the utmost importance to reach a full understanding of the molecular mechanisms and environmental

In recent years, key developments have revealed a novel neurodegenerative disease spectrum with clinical symptoms overlapping between ALS and frontotemporal lobar

**1.2 Sporadic ALS** 

pathophysiological mechanisms in SALS.

factors associated with SALS.

**1.3 ALS with cognitive impairments** 

non cost-effective genotyping strategies are being developed.

associated with the disease through several candidate gene-based association studies. Given the high degree of genetic and clinical heterogeneity seen in ALS patients, along with population-specific genetic risk factors for ALS, independent replication of these genetic association studies would be crucial in order to better understand the different mechanisms involved in these disease. As there are several excellent reviews in the literature describing these genes and their relevance to ALS, we will mainly describe the most recent and exciting advances involving *SOD1*, *TARDBP* and *FUS* in this area.

To date, family-based linkage studies and positional cloning have led to the identification of fifteen ALS-associated genes in which mutations have been identified in familial ALS cases (Table 1). Different research groups or consortiums have replicated these results in several populations. To date, a number of genes have been discovered as causative for the classical adult onset form of familial ALS with typical symptoms, namely *SOD1*, *TARDBP*, *FUS/TLS*, and *C9ORF72* (DeJesus-Hernandez, et al., 2011; Kabashi, et al., 2008; Kwiatkowski, et al., 2009; Renton, et al., 2011; Rosen, et al., 1993; Sreedharan, et al., 2008). These genes cumulatively account for about 25% of familial cases, indicating that other causative genes remain to be identified. The difficulties to identify gene responsible for FALS arise in part because large families with sufficient statistical power for linkage analysis are hard to come by, due to the late-onset and age-dependant penetrance of the disease, and the relative short survival time of affected ALS patients. Furthermore, the high degree of genetic heterogeneity, i.e. many rare variants in many different genes individually having a modest effect on the total number of ALS cases, is also certainly a limiting factor for the identification of causative ALS-predisposing genes. Four other genes, namely *DAO, OPTN, ATAX2* and *VCP* have also been included in this category, but the evidence for their role in predisposing to classical ALS needs to be confirmed in replication studies, and more families linked to these genes also need to be described (Elden, et al., 2010; Johnson, et al., 2010; Maruyama, et al., 2010; Mitchell, et al., 2010).


Table 1. Identified genes predisposing for familial ALS and/or ALS-FTLD FALS: Familial ALS, ALS-FTLD: ALS with frontotemporal lobar degeneration, AD: autosomal dominant, AR: autosomal recessive, Indels: insertions, deletions

The identification of other genes responsible for the disease and the understanding of the molecular pathways involved will bring some new insights into the mechanisms of disease pathogenesis, and may also elucidate the underlying mechanism(s) for specific motor neuron degeneration observed in ALS.

### **1.2 Sporadic ALS**

82 Amyotrophic Lateral Sclerosis

associated with the disease through several candidate gene-based association studies. Given the high degree of genetic and clinical heterogeneity seen in ALS patients, along with population-specific genetic risk factors for ALS, independent replication of these genetic association studies would be crucial in order to better understand the different mechanisms involved in these disease. As there are several excellent reviews in the literature describing these genes and their relevance to ALS, we will mainly describe the most recent and exciting

To date, family-based linkage studies and positional cloning have led to the identification of fifteen ALS-associated genes in which mutations have been identified in familial ALS cases (Table 1). Different research groups or consortiums have replicated these results in several populations. To date, a number of genes have been discovered as causative for the classical adult onset form of familial ALS with typical symptoms, namely *SOD1*, *TARDBP*, *FUS/TLS*, and *C9ORF72* (DeJesus-Hernandez, et al., 2011; Kabashi, et al., 2008; Kwiatkowski, et al., 2009; Renton, et al., 2011; Rosen, et al., 1993; Sreedharan, et al., 2008). These genes cumulatively account for about 25% of familial cases, indicating that other causative genes remain to be identified. The difficulties to identify gene responsible for FALS arise in part because large families with sufficient statistical power for linkage analysis are hard to come by, due to the late-onset and age-dependant penetrance of the disease, and the relative short survival time of affected ALS patients. Furthermore, the high degree of genetic heterogeneity, i.e. many rare variants in many different genes individually having a modest effect on the total number of ALS cases, is also certainly a limiting factor for the identification of causative ALS-predisposing genes. Four other genes, namely *DAO, OPTN, ATAX2* and *VCP* have also been included in this category, but the evidence for their role in predisposing to classical ALS needs to be confirmed in replication studies, and more families linked to these genes also need to be described (Elden, et al., 2010; Johnson, et al.,

**FALS-associated known genes (gene symbol) Inheritance Onset Mutation types References** *Superoxyde dismutase 1 (SOD1)* AD Adult Missense and truncation mutation Rosen et al., 1993 *TAR DNA-binding protein (TARDBP)* AD Adult Missense and nonsense mutations Kabashi et al., 2008; Sreedharan et al., 2008 *Fused in sarcoma/translated in liposarcoma (FUS/TL*AD and AR Adult Missense mutations, indels Kwiatkowski et al., 2009; Vance et al., 2009 *Chromosome 9 opend reading frame 72 (C9ORF72)* AD Adult Hexanucleotide GGGGCC expansion repeaenton et al., 2011, DeJesus-Hernandez et al., 201

*Ataxin 2 (ATAX2)* AD Adult Trinucleotide CAG expansion repeat Elden et al., 2010 *Ubiquitin-like protein ubiquilin2 (UBQLN2)* X-linked Adult Missense mutations Deng et al., 2011 *Optineurin (OPTN)* AD Adult Missense and nonsense mutations Maruyama et al., 2010 *D-amino acid oxidase (DAO)* AD Adult One R199W missense mutation Mitchell et al., 2010 *Amyotrophic lateral sclerosis 2 (ALS2)* AR Juvenile Missense and nonsense mutations Hadano et al., 2002, Yang et al., 2002 *Valosin-containing protein (VCP)* AD Adult Missense mutations Johnson et al., 2010 *Vesicle-associated membrane protein B (VAPB)* AD Adult Missense mutations Nishimura et al., 2004 *Microtubule-associated protein tau (MAPT)* AD Adult Missense and 5'-splice-site mutations Hutton et al., 1998 *Dynactin 1 (DCTN1)* AD Adult Missense mutations Puls et al., 2003 *Angiogenin* AD Adult Missense mutations Chen et al., 2004 *Senataxin (SETX)* AD Juvenile Missense mutations Greeway et al., 2006

Table 1. Identified genes predisposing for familial ALS and/or ALS-FTLD FALS: Familial ALS, ALS-FTLD: ALS with frontotemporal lobar degeneration, AD: autosomal dominant, AR: autosomal recessive, Indels: insertions, deletions

The identification of other genes responsible for the disease and the understanding of the molecular pathways involved will bring some new insights into the mechanisms of disease pathogenesis, and may also elucidate the underlying mechanism(s) for specific motor

advances involving *SOD1*, *TARDBP* and *FUS* in this area.

2010; Maruyama, et al., 2010; Mitchell, et al., 2010).

neuron degeneration observed in ALS.

For most cases of ALS, the causes are unknown. A genetic component is also thought to contribute to the pathogenesis of sporadic ALS, which accounts for the majority of ALS cases. However, identification of gene mutations associated with SALS has met with limited success so far. Several groups have reported on gene variants and association studies found in individuals with sporadic ALS, each accounting for a small number of the total cases reflecting a complex pattern of inheritance with very low penetrance, a high degree of heterogeneity and/or the existence of environmental factors predisposing to ALS. Because familial and sporadic forms of the disease are clinically and pathologically similar, understanding the familial form will shed light on possible epidemiological and pathophysiological mechanisms in SALS.

Many divergent data generated by different research groups who have studied different populations around the world have been published. The use of an adequate size population with appropriate control individuals is needed in order to reach enough statistical power and to observe any significant results. Homogeneity regarding ethnical origin, age, age of onset, disease duration, site of the first symptoms and gender among the studied populations is essential to design a good and robust genetic study. The requirement for larger sample sizes and possibly more sensitive and efficient analytic tools allowing for reliable detection across studied populations are now becoming more and more available as non cost-effective genotyping strategies are being developed.

The identifications of different genes and genetic risk factors associated with ALS over the past years have highlighted common molecular pathways mainly involving intracellular cell trafficking and RNA metabolism. To what extent these pathways are implicated in ALS remain to be determined and studied further. However, it is interesting to note that genes such as *ALS2*, *VAPB*, *MAPT*, *DCTN1*, *EAAT2*, *NEFH*, *PRPH, FIG4, CNTF*, *CHGB* and *OPTN* in which mutations have been found in a subset of FALS and/or SALS patients, are all related to intracellular trafficking either via axonal transport, vesicle docking and transport, or microtubule and neurofilament stabilization. Furthermore, this common theme is also seen in other motor neuron diseases and neurodegenerative disorders such as hereditary spastic paraplegia, Parkinson's disease, spinal muscular atrophy and Charcot-Marie-tooth disease. It is also noteworthy that RNA metabolism via alternative splicing abnormalities or RNA binding is also a common theme observed amongst the different ALS related genes such as *ALS2*, *MAPT*, *EAAT2*, *PRPH*, *GluR2*, *CNTF*, *SETX*, *SMN1, SMN2, TARDBP and FUS/TLS.* The high degree of genetic heterogeneity observed in ALS suggests that an individual phenotype could result from the sum of several contributing gene defects and/or epigenetic influences, which individually do not cause disease, and may explain the difficulties in identifying genes associated with SALS. The identification of other genes associated with motor neuron disease, as well as the determination of genetic and/or environmental factors that predispose to SALS is crucial in the development of novel therapies for ALS. The generation of cellular or animal models to study SALS, in which the disease can be closely replicated, would also be the utmost importance to reach a full understanding of the molecular mechanisms and environmental factors associated with SALS.

#### **1.3 ALS with cognitive impairments**

In recent years, key developments have revealed a novel neurodegenerative disease spectrum with clinical symptoms overlapping between ALS and frontotemporal lobar

*In Vivo* and *In Vitro* Models to Study Amyotrophic Lateral Sclerosis 85

Invertebrate model organisms include systems such as the simple yeast (fungi), fly, and nematode. The study of these experimental systems began with genetics and development, moved into molecular and cellular biology prior to most recently propelled into functional genomics and proteomics. These model organisms have highly manipulable genomes allowing for rapid generation of transgenic lines to provide insight on gene functions and protein network interactions. In 2002, Sydner Brenner, Robert Horvitz and John Sulston received the Nobel Prize in Physiology and Medicine for their establishment of the nematode *Caenorhadditis elegans* as a novel model organism to explore the molecular bases of organ development and cell death. Their discoveries identified key genes involved in cell division, differentiation and apoptosis. They determined that these genes were in fact highly conserved both structurally and functionally with higher organisms, including humans. Their studies provided the framework in which simpler organisms can be used to define key pathways and processes of relevance with the important benefit that the results are often

The key points into generating models for human disease using smaller invertebrate organisms should encompass high degree of conservation with mammals in order for the model to be useful for the identification of the molecular components implicated in disease pathogenesis. A fully sequenced genome of the studied organism should ideally be also easily accessible for genome comparison with higher vertebrates and mammals in order to facilitate evolutionary genomics studies and to quickly generate transgenic animals through DNA transformation. The model organism should also provide significant experimental advantages over their mammalian counterparts, including a short generation time, small size, ease and reasonable cost of maintenance. It should be also amenable to both forward (phenotype to gene) and reverse (gene to phenotype) genetic approaches, which are essential molecular tools to dissect out and understand gene function. Classic forward genetic characteristics allow for the identification of novel molecules or pathways involved in a particular cellular process. This can be one of the most powerful attributes of invertebrate models. Forward-genetic screens using chemical mutagens are one of the most universal ways to generate mutants to elucidate gene function. On the other side, reverse genetics allows the quick identification of pathways on which a particular gene acts. Genetic knockdown mutants, using RNA interference (RNAi) technology in which a gene product is dramatically reduced by introducing double-stranded RNA (dsRNA) into the organism, can

also provide invaluable information on the role a gene plays in a biological process.

The obvious disadvantage of using invertebrate model organisms such as fly or worm models is that there are evolutionarily far from mammals and that many physiological functions are not conserved, e.g. the immune system. Furthermore, their organs are extremely undeveloped and simple compared to other animals. The limited cellular

Yet, for the vast majority of neurodegenerative disorders including ALS, the mechanisms underlying neuronal dysfunction and death remain poorly understood despite the identification of relevant disease genes. Given the fact that molecular conservation in neuronal signaling pathways across vertebrate and invertebrate is relatively high and since the *Caenorhabditis elegans* nervous system contains almost all of the known signaling and

**2.1 Invertebrate models** 

directly applicable to understand human diseases.

diversity also represents a major disadvantage.

**2.1.1 Caenorhabditis elegans** 

degeneration (FTLD) (Strong and Yang, 2011). Cognitive and behavioral impairment is well described in ALS with a significant proportion of patients meeting strict diagnostic criteria for FTLD. Neuroimaging and pathological studies have shown clear involvement of nonmotor areas of the brain. More recently, distinct subtypes of frontotemporal lobar dysfunction have been identified in ALS and further work to distinguish these phenotypes and their correlate neuropathology is under way. Attention has been devoted lately to a gene coding for a DNA/RNA binding protein, which have been implicated in the pathogenesis of ALS with cognitive impairments. Indeed, dominant mutations in the *TARDBP* gene, encoding for TDP-43, were reported by several groups as a primary cause of ALS in about 3% familial cases and 1.5% sporadic cases (Chio, et al., 2011; Corrado, et al., 2009; Daoud, et al., 2009; Gitcho, et al., 2008; Kabashi, et al., 2008; Millecamps, et al., 2010; Sreedharan, et al., 2008; Van Deerlin, et al., 2008). The discovery of gene mutations linked to human ALS has provided plethora opportunities to develop model systems for investigating mechanisms of TDP-43 associated disease.

Motor neuron degeneration can also occasionally occurs in patients with Parkinson's disease and frontotemporal dementia (FTD). This disease is also called Disinhibition-dementiaparkinsonism-amyotrophy complex (DDPAC) or FTDP17 (Lynch, et al., 1994). The pathologic features distinguish this disease from the ALS-parkinsonism-dementia complex of Guam seen in the peninsula of Japan and from ALS-FTD linked to chromosome 9. Mutations in the *microtubule-associated protein tau* gene (*MAPT*) have been shown to be associated with FTD and Parkinsonism (Hutton, et al., 1998). Tau is a member of the microtubule-associated protein family, which have the principal function stabilizing microtubules and promoting their assembly by binding to tubulin. As a number of reviews (Gros-Louis, et al., 2006; Julien and Kriz, 2006) have already described the relationship between MAPT and ALS, the following sections will not discuss in details about this gene and FTDP17 models that have been generated.
