Neurologic and motor dysfunctions in APP transgenic mice

Abstract

The discovery of gene mutations underlying autosomal dominant Alzheimer’s disease has enabled researchers to reproduce several hallmarks of this disorder in transgenic mice, notably the formation of Aβ plaques in brain and cognitive deficits. APP transgenic mutants have also been investigated with respect to survival rates, neurologic functions, and motor coordination, which are all susceptible to alteration in Alzheimer dementia. Several transgenic lines expressing human mutated or wild-type APP had higher mortality rates than non-transgenic controls with or without the presence of Aβ plaques. Mortality rates were also elevated in APP transgenic mice with vascular amyloid accumulation, thereby implicating cerebrovascular factors in the precocious death observed in all APP transgenic models. In addition, myoclonic jumping has been described in APP mutants, together with seizure activity, abnormal limb-flexion and paw-clasping reflexes, and motor coordination deficits. The neurologic signs resemble the myoclonic movements, epileptic seizures, pathological reflexes, and gait problems observed in late-stage Alzheimer’s disease.

Neurobiological characteristics of Alzheimer’s disease

Neuropathology

Definite Alzheimer’s disease is diagnosed after postmortem evidence of extracellular neuritic plaques and intracellular neurofibrillary tangles (Perl, 2000). In addition, there is granulovacuolar degeneration marked by intracytoplasmic vesicles (Su et al., 2002). Neuritic plaques contain A-β-protein (Aβ) and cellular debris, while Aβ-positive neurofibrillary tangles contain hyperphosphorylated τ (Perl, 2000). Cored plaques are distinguishable from diffuse ones by their β-pleated structure and later onset (Yamaguchi et al., 1988; Thal et al., 2000). Diffuse plaques contain only Aβ₄₂, but mature plaques contain both Aβ₄₂ and Aβ₄₀ (Iwatsubo et al., 1994). Aβ plaques are surrounded by activated microglia and astrocytes as well as inflammation-related proteins (Eikelenboom et al., 2002).

Neuropsychological symptoms

Probable Alzheimer’s disease is diagnosed when progressive loss in remembering new items occurs (anterograde amnesia) in conjunction with a deficit in language (aphasia), object use (apraxia), form recognition of faces or objects (agnosia), and step-by-step planning (McKhann et al., 1984). Memory and verbal fluency deficits are often the initial symptoms (Amieva et al., 2005). Progressive amnesia is sometimes the only sign over a considerable period, even years (Weintraub and Mesulam, 1993). Anterograde amnesia reflects a loss in factual information, called declarative memory, both for verbal and spatial items (Jacobs et al., 1995; Amieva et al., 2005; Graham et al., 2004), whereas procedural memory is relatively spared, as in the pursuit-rotor test (Eslinger and Damasio, 1986; Jacobs et al., 1999), although delay and trace conditioning of the eye-blink response, a cerebellum-mediated function, is not (Woodruff-Pak and Papka, 1996; Woodruff-Pak et al., 1996). Spatial deficits also appear in the form of constructional apraxia, defined as a difficulty in copying geometric figures or reproducing them with building blocks (Graham et al., 2004). As a result, patients are more likely than controls to be disoriented in a hospital ward (Monacelli et al., 2003) and in the streets while riding a motored vehicle (Uc et al., 2004). In addition, apathy, dysphoria, social withdrawal, and depression are common neuropsychiatric features of Alzheimer dementia (Daffner et al., 1992; Frisoni et al., 1999; Chung and Cummings, 2000; Hart et al., 2003; Senanarong et al., 2004).

Neurologic symptoms

Although usually occurring late in disease progression, neurologic symptoms are a prominent feature of patients with Alzheimer dementia. These include hallucinations (Chen et al., 1991; Frisoni et al., 1999; Chung and Cummings, 2000; Hart et al., 2003; Senanarong et al., 2004), deficient postural control (Huff et al., 1987; Pettersson et al., 2002), and myoclonus (Chen et al., 1991; Förstl et al., 1992), defined as brief, jerky movements of the arms and legs, most often bilateral (Snodgrass, 1990; Defebvre, 2006; Lanska, 2010). In addition, patients are susceptible to epileptic seizures, parkinsonian symptoms, and frontal lobe-associated release signs such as grasping and sucking in response to tactile stimuli (Huff et al., 1987; Burns et al., 1991; Chen et al., 1991; Förstl et al., 1992). Posture and gait can be impaired even in the early stages of the disease (Pettersson et al., 2002), perhaps due to vermal atrophy (Sjöbeck and Englund, 2001). Myoclonus and epileptic seizures are particularly frequent in patients with the hereditary form of the disease, marked by APP and PS1 mutations (Campion et al., 1996; Furuya et al., 2003). The seizures may contribute to mental deterioration in addition to Aβ-mediated processes (Leonard and McNamara, 2007). Myoclonic twitching mostly appears in late stages, generally after extrapyramidal and psychotic symptoms (Chen et al., 1991). Unlike the sporadic form, autosomal dominant cases with myoclonus contain Aβ plaques in the cerebellum (Ishino et al., 1984; Lemere et al., 1996). Thus, anomalies of the cerebellum may contribute to both balance problems and myoclonus.

Pathological characteristics in APP mice

Introduction

Thanks to the discovery of pathological mutations in humans, several Alzheimer hallmarks have been attained in transgenic mice (Cole and Frautschy, 1997; Hsiao, 1998; Duff, 1999; Van Leuven, 2000; Janus and Westaway, 2001; Dodart et al., 2002). Most APP mutants have no neurofibrillary tangles, thereby resembling rare cases of Alzheimer dementia with amyloid deposits and hyperphosphorylated τ but without tangles (Tiraboschi et al., 2004). One exception is the transgenic model expressing APP, PS1, and Mapt (3xTg-AD) characterized by Aβ plaques and neurofibrillary tangles (Oddo et al., 2003). The genetic characteristics of mice behaviorally characterized for neurologic or motor function and expressing mutated or wild-type (WT) APP or its C-terminal fragments are presented in Table 1.

Table 1.

Description of behaviorally characterized mice expressing human mutated or WT APP transgenes or Aβ₄₂ with or without mutated PS1.

Name	Mutation	Promoter	Parenchymal Aβ plaques	References
APP695SWE Tg2576	Swedish	Hamster Prp	Yes	Hsiao et al., 1996
APP695SWE Tg2123H	Swedish	Hamster Prp	No	Hsiao et al., 1995
APP695SWEch	Swedish	Murine PrP	No	Borchelt et al., 1996, 1997
APP751SWE TgAPP23	Swedish	Murine Thy1	Yes	Sturchler-Pierrat et al., 1997
APP770SWE	Swedish	Murine Thy1	Yes	Moechars et al., 1999b
APP/SWE/Ckm	Swedish	Murine Ckm	No	Sugarman et al., 2002
APP695LD	London	Murine Thy1	Yes	Moechars et al., 1999b
APP/DU/Thy1	Dutch	Murine Thy1	No	Kumar-Singh et al., 2000
APP/FL/Thy1	Flemish	Murine Thy1	No	Kumar-Singh et al., 2000
APP695SWE/IND TgCRND8 or Tg19959	Swedish+Indiana	Hamster Prp	Yes	Chishti et al., 2001
APP/SWE/IND/J20	Swedish+Indiana	Platelet-derived PDGFβ	Yes	Mucke et al., 2000; Palop et al., 2003
APP751SWE/IND TgAPP-Sw, V717F/B6	Swedish+Indiana	Platelet-derived PDGFβ	No	Lee et al., 2004
APP751SWE/LD	Swedish+London	Murine Thy1	Yes	Blanchard et al., 2003
APP/RK	Artificial 684+687 codons	Murine Thy1	No	Moechars et al., 1996
APP695TRImyc Tg1130H	Artificial	Hamster Prp	No	Hsiao et al., 1995
APP695WT/Mt2	None	Mt2	No	Yamaguchi et al., 1991
APP695WT/Thy1	None	Murine Thy1	No	Moechars et al., 1999b
APP695WT Tg1874	None	Hamster Prp	No	Hsiao et al., 1995
APP695mycWT Tg6209	None	Hamster Prp	No	Hsiao et al., 1995
APP/C99 Tg13592	None	CMV enhancer/chicken Actb	No	Fukuchi et al., 1996
Aβ42	None	Murine Nefl	No	LaFerla et al., 1995
APP695SWE+PS1/M146L	Bigenic	Hamster Prp+PDGFβ2	Yes	Duff et al., 1996; Holcomb et al., 1998
APP695SWE+PS1/A246E	Bigenic	Hamster Prp+human THY1	Yes	Qian et al., 1998; Dineley et al., 2002
APP695SWEch+PS1/A246E	Bigenic	Murine PrP	Yes	Borchelt et al., 1996
APP695SWE/co+PS1/ΔE9	Co-Injected bigenic	Murine PrP	Yes	Jankowsky et al., 2004
APP751SWE/LD+PS1/M146L	Bigenic	Murine Thy1+human HMGCR	Yes	Blanchard et al., 2003
APP751SWE/LD+PS1/M233T+L235P	Bigenic	Murine Thy1	Yes	Casas et al., 2004
3xTg-AD	Trigenic	Murine Thy1+endogenous control+murine Thy1	Yes	Oddo et al., 2003
5xFAD	Five-time mutated	Murine Thy1	Yes	Oakley et al., 2006

APP mutations

Swedish

Several APP transgenic mutants have been generated with the Swedish mutation, defined as amino acid substitutions at codons 670 and 671 first reported in a Swedish family (Axelman et al., 1994; Haass et al., 1995). One of the better known transgenics of this type is APP₆₉₅SWE (Tg2576), containing a human APP transgene with the Swedish mutation of the neuron-specific 695-amino acid isoform driven by the hamster Prp promoter (Hsiao et al., 1996). Detergent-insoluble Congo Red-positive Aβ plaques were visible in APP₆₉₅SWE mice as early as 7 months, increasing with age (Kawarabayashi et al., 2001; Westerman et al., 2002). The Aβ plaques also surround vessel walls, causing micro-hemorrhages (Frackowiak et al., 2001; Fryer et al., 2003; Domnitz et al., 2005). In contrast, a previously generated transgenic mouse with the same mutation and promoter (Tg2123H) does not harbor Aβ plaques (Hsiao et al., 1995).

Another transgene with the Swedish mutation was designed with the inserted gene driven this time by the murine Thy1 promoter on the 751-amino acid isoform (Sturchler-Pierrat et al., 1997). These APP₇₅₁SWE transgenic mice (APP23) accumulate Aβ in the hippocampus and neocortex, with plaque-associated Aβ peptides augmenting 5-fold from 3 to 6 months of age (Van Dam et al., 2003). As seen in Alzheimer brain, neuritic plaques are surrounded by astrocytes and activated microglia and τ is hyperphosphorylated (Sturchler-Pierrat et al., 1997; Stalder et al., 1999; Bornemann et al., 2001). In addition to parenchymal plaques, amyloid angiopathy occurs, leading to weakened vessel walls, aneurysm, vasculitis, and hemorrhage (Calhoun et al., 1999; Winkler et al., 2001; Beckmann et al., 2011).

Yet another APP model with the Swedish mutation driven by the murine Thy1 promoter includes the 770-amino acid isoform (Moechars et al., 1999b). Aβ plaques were observed in the hippocampus and neocortex of these APP₇₇₀SWE mice. In contrast, in a model containing the APP transgene with the Swedish mutation driven by the endogenous PrP promoter of a chimeric human/murine 695-amino acid isoform, APP₆₉₅SWEch (Borchelt et al., 1996, 1997), few or no neuritic Aβ plaques were seen despite increased Aβ₄₂ and Aβ₄₀ levels in the brain (Savonenko et al., 2003).

Although subjects with inclusion body myopathy do not carry APP mutations, they are characterized by Aβ accumulation in muscles and muscle weakness. An APP/SWE/Ckm transgene was generated with the Swedish mutation driven by the murine Ckm (creatine kinase, muscle) promoter (Sugarman et al., 2002). As found in human muscles (Askanas and Engel, 1998, 2002), Aβ accumulated intracellularly in mouse skeletal muscles of the transgenic (Sugarman et al., 2002). As seen in sporadic but not hereditary inclusion body myopathy (Askanas and Engel, 1998), inflammation was evident, with neutrophils instead of T cells as the predominant cell type (Sugarman et al., 2002).

London

An APP model comprises the V717I London mutation on the 695-amino acid isoform driven by the murine Thy1 promoter, causing diffuse and neuritic Aβ plaques (Moechars et al., 1999b; Dewachter et al., 2000a, b; Van Dorpe et al., 2000), surrounded by activated astrocytes and microglia, the neurites being linked to hyperphosphorylated τ (Moechars et al., 1999b). The angiopathy in APP₆₉₅LD mice leads to aneurysms but not hemorrhages (Van Dorpe et al., 2000).

Dutch and Flemish

APP/DU/Thy1 and APP/FL/Thy1 mice with E693Q (Dutch) and A692G (Flemish) mutations, respectively, each driven by the murine Thy1 promoter, have cerebral amyloid angiopathy with elevated parenchymal C-terminal fragments but not Aβ plaques (Kumar-Singh et al., 2000).

Swedish/Indiana

Several models with combined Swedish and Indiana mutations are available. APP₆₉₅SWE/IND mice, also known as TgCRND8 or Tg19959, express the 695-amino acid isoform regulated by the hamster Prp promoter (Janus et al., 2000; Chishti et al., 2001), showing Aβ plaques as early as 3 months of age (Janus et al., 2000; Chishti et al., 2001; Bellucci et al., 2006), together with early-onset angiopathy (Domnitz et al., 2005) and activated microglia and astrocytes (Bellucci et al., 2006). Two APP/SWE/IND constructs driven by the platelet-derived PDGFβ promoter (J9 and J20 lines) also contain Aβ plaques (Mucke et al., 2000; Palop et al., 2003). On the contrary, the APP₇₅₁SWE/IND (APP-Sw, V717F/B6) model driven by the PDGFβ promoter did not harbor plaques up to 14 months of age (Lee et al., 2004).

Swedish/London

An APP₇₅₁SWE/LD line with Swedish and London mutations was generated on a transgene driven by the murine Thy1 promoter, causing Aβ plaques (Blanchard et al., 2003).

Artificial

APP/RK transgenic mice with an artifical double mutation at the α-secretase site were generated under the control of the murine Thy1 promoter (Moechars et al., 1996). These mice harbor no Aβ plaques, but the neocortex and hippocampus are marked by astrocytosis and DNA fragmentation (Moechars et al., 1998, 1999a). Likewise, no Aβ plaques are seen in APP₆₉₅TRImyc mice (Tg1130H), expressing a triple mutation (V717L, V721A, and M722V) with a 3′-myc tag driven by the hamster Prp promoter (Hsiao et al., 1995).

Human WT APP murine WT App

The 695 splice variant was used for generating mice expressing human WT APP₆₉₅ under the control of the metallothionein IIA (Mt2) promoter (Yamaguchi et al., 1991). A similar APP₆₉₅WT construct driven by the murine Thy1 promoter is available (APP₆₉₅WT/Thy) (Moechars et al., 1999b). Mice expressing human APP₆₉₅myc WT with a 3′-myc tag (Tg6209) and murine APP₆₉₅WT (Tg1874) driven by the hamster Prp promoter have also been generated (Hsiao et al., 1995).

APP C-terminal fragments

APP/C99 mice (Tg13592) expressing the human 99-amino acid C-terminal fragment in the brain and skeletal muscles driven by a CMV enhancer/chicken Actb (β-actin promoter) mimic inclusion body myopathy in that Aβ accumulates in muscles but not in the brain (Fukuchi et al., 1996, 1998, 2000).

Aβ₄₂

Transgenic mice overexpressing human Aβ₄₂ driven by the murine Nefl promoter have increased intracellular levels of the peptide (LaFerla et al., 1995, 1996). These mice are marked by DNA fragmentation, increased p53 levels, and morphological markers of apoptosis in the neocortex, hippocampus, amygdala, and thalamus.

APP+PS1 mutations

The above-described APP₆₉₅SWE mice (Tg2576) were interbred with mutated PS1 line 5.1, yielding the APP₆₉₅SWE+PS1/M146L bigenic, characterized by earlier Aβ plaque onset than the single transgenic (Duff et al., 1996; Holcomb et al., 1998, 1999; Westerman et al., 2002; Kurt et al., 2003), with the same features of astrocytosis and microgliosis (Holcomb et al., 1998, 1999; Gordon et al., 2002), Aβ plaques surrounding vessel walls (Christie et al., 2001; Domnitz et al., 2005), and reduced vasodilatation properties (Christie et al., 2001). The same APP₆₉₅SWE line was bred with a PS1/A246E strain crossed with a null mutant for the murine Ps1 gene (Qian et al., 1998), yielding Aβ plaques with earlier onset than the single transgenic (Dineley et al., 2002).

A co-injected bigenic contains the 695-amino acid chimeric human/murine APP transgene with the Swedish mutation combined with a PS1 transgene with the ΔE9 mutation driven by the murine Prp promoter (Jankowsky et al., 2004). These APP₆₉₅SWE/co+PS1/ΔE9 mice (line 85) harbor Aβ plaques at 6 months of age (Jankowsky et al., 2004; Garcia-Alloza et al., 2006) and possess fewer capillary segments than controls in the neocortical white matter (Lee et al., 2005).

The APP₇₅₁SWE/LD line with Swedish and London mutations on a transgene driven by the murine Thy1 promoter (section 2.2.6) was crossed with a PS1/M146L line on a transgene driven by the human 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase (HMGCR) promoter, causing Aβ plaques (Blanchard et al., 2003). The APP₇₅₁SWE/LD line was also crossed with a PS1/M233T+L235P knockin, yielding a bigenic with Aβ plaque onset at 2.5 months of age, glial activation, and pyramidal cell loss in the hippocampus (Casas et al., 2004), as well as dentate gyrus granule cells (Cotel et al., 2008). The APP₇₅₁SWE/LD+PS1/M233T+L235P mice are marked by spinal axonopathy and thoracolumbar kyphosis, i.e., spine curvature, not a feature of Alzheimer’s disease (Wirths and Bayer, 2008).

A 5-fold transgenic model exists with Swedish/Florida/London mutations on APP and M146L and L286V mutations on PS1 with the murine Thy1 promoter (Oakley et al., 2006). These 5xFAD transgenic mice have Aβ plaques as early as 2 months of age, accompanied by astrogliosis and microgliosis. As mentioned above, the only model with Aβ plaques and neurofibrillary tangles is the triple APP/SWE+PS1/M146V+Mapt/P301L mutant (Oddo et al., 2003). These 3xTg-AD mice were generated by co-injecting APP with the Swedish mutation and four-repeat Mapt with the P301L mutation into single-cell embryos of PS1/M146V knockin mice.

Neurologic symptoms and survival rates in APP mice

Although several reviews of murine Alzheimer models have appeared (Van Leuven, 2000; Ashe, 2001; Morgan, 2003; Sant’ Angelo et al., 2003; German and Eisch, 2004; Mineur et al., 2005; Spires and Hyman, 2005; McGowan et al., 2006; Gimenez-Llort et al., 2007; Duyckaerts et al., 2008; Howlett and Richardson, 2009; Kokjohn et al., 2009; Ashe and Zahs, 2010; Wirths and Bayer, 2010; Scearce-Levie, 2011; Lalonde et al., 2012), the authors mostly describe the effects of accumulating Aβ concentrations on synaptic activity or learning abilities in various spatial and non-spatial tasks. We present in-depth task-by-task effects of APP transgenesis on survival and basic neurologic functions regarding pathological reflexes, myoclonus, and epilepsy. In particular, the SmithKline-Harwell Imperial College Royal Hospital Phenotype Assessment (SHIRPA) multi-test battery, first described by Rogers et al. (1997), has been used to provide information on the general neurologic function in APP mutants (Lalonde et al., 2005). We also present the effects of APP transgenesis on motor coordination, a neglected aspect of behavior but justified on the basis of the deficient postural control reported in patients with Alzheimer’s disease (Huff et al., 1987), sometimes even during early stages (Pettersson et al., 2002).

Premature death

Premature death has been described in several APP transgenic models relative to mice of the same background strain. In most of these reports, qualitative descriptions were provided without the backing of statistical analyses. Nevertheless, such reports are of interest in providing data on overall biological functions, not being limited to the impact of Aβ on synaptic activity and learning.

King and Arendash (2002) first reported that the often used APP₆₉₅SWE transgenic mutant (Tg2576) bearing Aβ plaques dies prematurely. Relative to the WT strain of the same genetic background, fewer APP₆₉₅SWE transgenic mice survived between the ages of 6 and 12 months, dying from unknown causes. This finding was confirmed at other age ranges (Lewis et al., 2004; Kim et al., 2007). The same result was found in the APP₆₉₅SWE+PS1/A246E bigenic derived from the same monogenic (Lewis et al., 2004) as well as a second Aβ plaque-bearing bigenic derived from different monogenics, APP₆₉₅SWE/co+PS1/ΔE9 (Gimbel et al., 2010).

The susceptibility of APP transgenesis to premature death was further documented in a second monogenic with the Swedish mutation resulting in Aβ plaques, APP₇₇₀SWE (Moechars et al., 1999a). In contrast to the 4% mortality rate recorded in normal mice during the first year of life, from 21% to 70% of APP₇₇₀SWE mice (depending on specific lines) died from undetermined causes. Of note is the equally high death rate (from 53% to 68%) of APP₆₉₅WT/Thy1 mice without Aβ plaques (Moechars et al., 1999a), thus indicating that APP overexpression alone or high Aβ concentrations cause precocious death, independently of Aβ plaques. In confirmation of this hypothesis, death occurred earlier than normal in six other APP transgenic models without Aβ plaques. Indeed, while 7% of normal mice died within the first year, from 44% to 69% of APP/RK mice did so (Moechars et al., 1996). Mortality rates were augmented as well in the following mutants: APP₇₅₁SWE/IND (Lee et al., 2004), APP₆₉₅mycWT (Tg6209), APP₆₉₅WT (Tg1874), APP₆₉₅TRImycWT (Tg1130H), APP₆₉₅SWE (Tg2123H) (Hsiao et al., 1995; Carlson et al., 1997), and Aβ₄₂ (LaFerla et al., 1995).

A key insight into the underlying cause of death is provided by data showing higher mortality rates in APP transgenic mice with vascular but not parenchymal amyloid accumulation. In contrast to the 3% mortality rate found in non-transgenic controls over the first 6 months of life, from 8% to 32% of APP/DU/Thy1 mutants (depending on specific lines) and from 21% to 31% of APP/FL/Thy1 mutants died (Kumar-Singh et al., 2000). Therefore, cerebrovascular factors are likely to be at work in most if not all APP transgenic models. In confirmation of this hypothesis, the death rate among APP₆₉₅mycWT mice was potentiated by overexpression of the FGF2 gene, encoding basic fibroblast growth factor, a protein involved in vascular smooth muscle hypertrophy (Carlson et al., 1997). How vascular angiopathy causes death in rodents remains to be determined.

Myoclonus

Myoclonus refers to involuntary, jerky, large-amplitude movements distinct from lower amplitude muscle tics (Snodgrass, 1990; Defebvre, 2006; Lanska, 2010). Like seizure activity displayed in the form of clonic movements, myoclonic movements are rhythmical but of briefer duration. Myoclonus is considered as part of a pre-epileptic state. Some drug treatments of human and animal epileptic seizures are also effective anti-myoclonic agents by protecting neurons from synchronized repetitious moto-neuron firing as the main therapeutic target. Myoclonic movements may originate from abnormal neural activity of several brain regions, especially the brainstem and the cerebellum.

In a manner reminiscent of the myoclonus observed in late-stage Alzheimer’s disease (Chen et al., 1991; Förstl et al., 1992), a species-specific form of myoclonus occurring in mice, myoclonic jumping (Table 2), was described in APP₇₅₁SWE mutants (APP23) bearing Aβ plaques (Lalonde et al., 2005). APP₇₅₁SWE mutants often rear against the walls of their home-cage and hop continuously, the hindpaws are on and off the ground, the forepaws are with wall support, and the snout is facing upward. This stereotyped response was not observed in WT mice of the same C57BL/6J background strain when first compared to the mutants. However, one may occasionally see this behavior when looking at large groups of normal mice of the C57BL/6J strain, and likely other strains, although this has never to our knowledge been assessed in a systematic fashion.

Table 2.

Myoclonic and epileptic phenotypes in mice expressing human mutated or WT APP transgenes with or without mutated PS1.

Name	Myoclonic jumping	Tonic-clonic seizures	Parenchymal Aβ plaques	References to behavior
APP695SWE Tg2576	Yes (3%)	Yes (small number)	Yes	Lalonde et at., 2003; Westmark et al., 2008
APP751SWE TgAPP23	Yes	Yes	Yes	Lalonde et al., 2005
APP770SWE	Not reported	Yes	Yes	Moechars et al., 1999b
APP695LD	Not reported	Yes	Yes	Moechars et al., 1999b
APP/DU/Thy1	Not reported	Yes	No	Kumar-Singh et al., 2000
APP/FL/Thy1	Not reported	Yes	No	Kumar-Singh et al., 2000
APP695SWE/IND TgCRND8 orTg19959	Yes	Not reported	Yes	Ambrée et al., 2006
APP/SWE/IND/J20	Not reported	Yes (EEG only)	Yes	Palop et al., 2007
APP695TRImyc Tg1130H	Not reported	Yes	No	Hsiao et al., 1995
APP/RK	Not reported	Yes	No	Moechars et al., 1996
APP695WT/Thy1	Not reported	Yes	No	Moechars et al., 1999b
Aβ42	Not reported	Yes	No	LaFerla et al., 1995
APP695SWE/co+PS1/ΔE9	No	Yes	Yes	Lalonde et al., 2004; Minkeviciene et al., 2009

The neurochemical basis of myoclonic jumping likely involves 5-hydroxytryptamine (5HT). Drugs potentiating 5HT neurotransmission cause myoclonic jumping in guinea pigs, such as peripheral administration of the immediate metabolic precursor of 5HT, 5-hydroxytryptophan (Weiner et al., 1979), as well as L-tryptophan combined with pargyline, a monoamine oxidase inhibitor (Luscombe et al., 1983). The stereotyped response was antagonized by both 5HT₁ or 5HT₂ receptor antagonists (Luscombe et al., 1982; Pappert et al., 1998). However, this maladaptive behavior is more difficult to be elicited in mice and rats (Carvey et al., 1986), possibly as a result of species-specific distributions of 5HT receptors. Nevertheless, in support of the hypothesis that 5HT is involved in the mutant response, 5HT concentrations of APP₇₅₁SWE mutants were higher than those of WT controls in the neocortex (Van Dam et al., 2005) and possibly in the brainstem. Moreover, myoclonic symptoms in humans can be caused by 5HT-enhancing agents (Radomski et al., 2000).

In addition to APP₇₅₁SWE mutants, myoclonic jumping was observed in a second Aβ plaque-bearing monogenic, this time with combined Swedish and Indiana mutations, APP₆₉₅SWE/IND, also known as TgCRND8 or Tg19959 (Ambrée et al., 2006). In the same mutant, twitching and vocalizations were enhanced after peripheral administration of pentylenetetrazole, the γ-amino butyric acid (GABA) receptor antagonist (Jolas et al., 2002). The involvement of GABA in myoclonic jumping is indicated by the finding that this maladaptive behavior can be elicited after intracerebroventricular administration of Zn²⁺, inhibitor in the activity of glutamic acid decarboxylase, the GABA-synthesizing enzyme, and that the Zn²⁺-induced behavior was reversed by intracerebroventricular GABA (Itoh and Ebadi, 1982). The same effect occurred after Zn²⁺ injections inside the hippocampus but not inside the amygdala, hypothalamus, thalamus, or caudate, implicating reduced GABA transmission in the hippocampus as an important mediator of myoclonic jumping.

Despite these observations of the two Alzheimer-like mutants, myoclonic jumping is not a feature of all APP mutants. Indeed, despite elevated Aβ plaque load, it was not observed in two mutants, APP₆₉₅SWE/co+PS1/ΔE9 (Lalonde et al., 2004) and APP₆₉₅SWE (Lalonde et al., 2003), or in only 3% of the latter when larger groups are available (Westmark et al., 2008). This may be due to an insufficient increase of 5HT neurotransmission, or to an insufficient decline in GABA transmission, or to some other neurochemical anomaly to which APP₇₅₁SWE and APP₆₉₅SWE/IND mutants are susceptible but not the other two. In any event, the importance of APP overexpression in myoclonic jumping is underlined by the finding that such movements are a phenotypic feature of Ts65Dn mice, a murine model of Down syndrome, overexpressing several genes from mouse chromosome 16, analogous to human chromosome 21, one of which is APP (Turner et al., 2001). Like Alzheimer patients, late-stage Down syndrome subjects often display myoclonic symptoms (Möller et al., 2002).

Convulsive seizures

In addition to myoclonus, APP₇₅₁SWE mutants exhibit spontaneous tonic-clonic seizures (Table 2), thereby indicating that myoclonic movements may appear simultaneously with clonic movements in the same mutant (Lalonde et al., 2005). The seizures are described as being spontaneous in the sense of being unprovoked by an experimenter-controlled stimulus, often audiogenic in nature. Spontaneous tonic-clonic seizures were also observed in four other models harboring Aβ plaques: APP₆₉₅SWE (although ‘a small percentage only’ according to Westmark et al., 2008), APP₇₁₀SWE (Moechars et al., 1999b), APP₆₉₅LD mice (Moechars et al., 1999b), and APP₆₉₅SWE/co+PS1/ΔE9 (Minkeviciene et al., 2009). The same phenotype was evident in four Alzheimer-like APP mutants not harboring Aβ plaques: APP₆₉₅WT/Thy1 (Moechars et al., 1999b), APP/RK (Moechars et al., 1996), APP₆₉₅TRImyc (Hsiao et al., 1995), and Aβ₄₂ (LaFerla et al., 1995). Moreover, the same phenotype appeared in APP/DU/Thy1 and APP/FL/Thy1 mutants with vascular angiopathy (Kumar-Singh et al., 2000), thus implicating vascular factors in all the mutants, as is the case with the premature death syndrome presented in section 3.1, and perhaps revealing a relation between the two. Indeed, a high mortality rate in APP₆₉₅SWE/co+PS1/ΔE9 mutants is attributed to their epileptic seizures (Mink-eviciene et al., 2009).

In one mutant carrying Aβ plaques, APP/SWE/IND/J20, no overt seizures were observed, but the EEG showed convulsive features (Palop et al., 2007), indicating that this phenotype is likely to be more prominent than believed. Two mutants, APP₆₉₅SWE (Westmark et al., 2008) and APP/SWE/IND/J20 (Palop et al., 2007), were more susceptible than normal mice to pentylenetetrazole-induced convulsions, implicating subnormal GABA transmission as the key underlying factor behind EEG anomalies and likely the overt seizures found in other models.

The susceptibility of APP mutants to convulsions is another point of resemblance with Alzheimer patients, susceptible to epileptic seizures, as well as pathological reflexes (Huff et al., 1987; Burns et al., 1991; Chen et al., 1991; Förstl et al., 1992), described in mice in the next section.

Abnormal reflexes: limb-flexion and paw-clasping

When WT C57BL/6J mice are held by the tail and slowly lowered towards a horizontal surface, they all exhibit the normal placing response marked by extension of all four limbs. On the contrary, 8 of 17 (47%) Aβ plaque-bearing APP₇₅₁SWE mutants (APP23) on the same genetic background exhibited hindlimb-flexion and 2 of 17 (12%) exhibited the hindpaw-clasping response, in which the paws are linked together while suspended in the air (Lalonde et al., 2005). The same pathological reflexes were seen in Aβ plaque-bearing APP₆₉₅SWE mutants (Tg2576) on the C57B6/SJL genetic background, with 13 of 14 (93%) of them displaying hindlimb-flexion and 5 of 14 (36%) displaying hindpaw-clasping (Lalonde et al., 2003). However, some WT C57B6/SJL littermates displayed the same sign, hindlimb-flexion noted in 10 of 21 (48%) and hindpaw-clasping in 2 of 21 (10%), presumably as a consequence of ‘abnormal’ unidentified genes on the SJL background. Although the number of mutants showing abnormal flexion reflexes might be higher than WT, the presence of these phenotypes in ‘normal’ mice rendered them ambiguous in a theoretical sense and useless for identifying them in conjunction with genotyping of tail biopsies. Nevertheless, hindlimb-flexion and paw-clasping were reported in APP₇₅₁SWE/LD+PS1/M233T+L235P mutants on a B6/CBA/129SV background, also characterized by parenchymal Aβ plaques but unlike the other two mutants by axonal neuropathology in the brain and spinal cord (Wirths et al., 2007). On the contrary, despite the presence of Aβ plaques, APP₆₉₅SWE/co+PS1/ΔE9 mutants did not exhibit paw-clasping, indicating that amyloid pathology does not necessarily lead to this phenotype (Lalonde et al., 2004).

Because paw-clasping appears in 5htt mutants lacking the 5HT transporter (Lira et al., 2003), elevated 5HT concentrations at the synapse may be responsible for triggering it. This may be one mechanism underlying paw-clasping in mice with Alzheimer-like pathology, because, as mentioned in section 3.2 regarding myoclonus, APP₇₅₁SWE mice were reported as having higher neocortical 5HT concentrations than those of WT (Van Dam et al., 2005).

Motor coordination

In view of impaired posture and gait even in early stages of Alzheimer’s disease (Pettersson et al., 2002), it is of interest to examine APP mutants with respect to motor skills. The main measure used in motor coordination tests (Table 3) is the time elapsed before falling from various types of apparatus requiring balance and equilibrium (Lalonde and Strazielle, 1999).

Table 3.

Motor coordination in mice expressing human mutated or WT APP transgenes with or without mutated PS1 (↓=deficit or normal vs. controls).

Name	Stationary beam	Suspended bar	Rotorod	References
APP695SWE Tg2576	↓ or normal	↓ or normal	Normal	Chapman et at., 1999; Arendash et al., 2001b, 2004; Dineley et al., 2002; King and Arendash, 2002; Lalonde et al., 2003; Gil-Bea et al., 2007; Perucho et al., 2010
APP751SWE APP23	Normal	Normal	↓ or normal	Lalonde et al., 2002b, 2005; Van Dam et at., 2003; Dumont et al., 2004
APP695SWEch	↓	Normal	Not reported	Savonenko et al., 2003
APP695SWE/IND TgCRND8 or Tg19959	Not reported	Not reported	Normal	Hyde et al., 2005; Bellucci et al., 2006
APP751SWE/IND	Not reported	Not reported	↓	Lee et al., 2004
APP751SWE/LD	↓	↓	Not reported	Le Cudennec et al., 2008; Blanchard et al., 2009
APP751WT	Not reported	Normal	Normal	Moran et al., 1995
APP/SWE/Ckm	Not reported	Not reported	↓	Sugarman et al., 2002; Kitazawa et al., 2006
APP/C99	Normal	Normal	Normal	Lalonde et al., 2002a
APP695SWE+PS1/M146L	↓ or normal	↓	↓	Arendash et al., 2001a, b; Sadowski et al., 2004; Ewers et al., 2006
APP695SWE+PS1/A246E	Not reported	Not reported	Normal	Dineley et al., 2002
APP695SWE/co+PS1/ΔE9	Normal	Normal	Normal	Lalonde et al., 2004
APP751SWE/LD+PS1/M233+L235P	↓	↓	Not reported	Cotel et al., 2012
3xTg-AD	Normal	Normal	Not reported	Gimenez-Llort et al., 2007; Gulinello et al., 2009
5xFAD	↓	↓	Not reported	Jawhar et al., 2011

Stationary beam

In the stationary beam test, mice are placed on a narrow beam obstructed at either end to prevent escapes and latencies before falling are measured (Lalonde and Strazielle, 1999). Alternate measures include distance travelled and, when one obstacle is removed, the time taken to escape from the beam towards a platform. On one hand, five APP mutants have been shown to fall sooner than controls from the stationary beam: APP₆₉₅SWE (Arendash et al., 2001b, 2004; King and Arendash, 2002), APP₆₉₅SWE+PS1/M146L(Arendash et al., 2001a, b), APP₇₅₁SWE/LD (Le Cudennec et al., 2008), APP₇₅₁SWE/LD+PS1/M233+L235P (Cotel et al., 2012), and 5xFAD (Jawhar et al., 2011). The result for the latter mutant is not surprising in view of its spinal axonopathy, not part of Alzheimer symptomatology. On the other hand, latencies before falling were equivalent to those of controls in the APP₆₉₅SWE mutant in a separate study (Lalonde et al., 2003), as well as three other APP mutants bearing Aβ plaques in the brain: APP₇₅₁SWE (Lalonde et al., 2002b), APP₆₉₅SWE/co+PS1/ΔE9 (Lalonde et al., 2004), and 3xTg-AD (Gimenez-Llort et al., 2007; Gulinello et al., 2009), as well as one with Aβ plaques in skeletal muscles, APP/C99, part of inclusion body myopathy but not Alzheimer symptomatology (Lalonde et al., 2002a). When latencies before escaping from the beam are considered, APP₆₉₅SWEch (Savonenko et al., 2003) mutants were slower than controls, but such a result was not found in a mutant carrying Aβ plaques, APP₆₉₅SWE+PS1/M146L (Sadowski et al., 2004), although, as mentioned above, this bigenic was impaired in terms of falling latencies (Arendash et al., 2001a, b).

One possible reason for these discrepancies is the unequal distribution of Aβ plaques or soluble Aβ accumulation in motor-related areas such as the motor cortex, dorsal striatum, and cerebellum. In addition, different results may be due to methodological factors such as the age of mice, the genetic background, the specific measure used, and the width of the bar. In particular, the APP₆₉₅SWE mutant was impaired only when the width of the bar was narrow (Arendash et al., 2001a, 2004; King and Arendash, 2002), but the same mutant was not impaired on a wider bar, making the task easier (Lalonde et al., 2003).

Suspended bar or coat-hanger

Like the stationary beam test, the mouse locomotes along a narrow surface and its latencies before falling are measured in the suspended bar test (Lalonde and Strazielle, 1999). Unlike the previous test, the narrowness of the wire is such that the animal is suspended upside-down, thereby requiring a greater degree of muscle strength to support its body weight. In the coat-hanger version, various movement time measures are added as to the time taken to reach either end of the bar and to climb atop the diagonal bar found at each end.

Relative to controls, three APP mutants harboring Aβ plaques were found to be impaired in the suspended bar test: APP₆₉₅SWE (King and Arendash, 2002), APP₆₉₅SWE+PS1/M146L (Arendash et al., 2001a), and 5xFAD (Jawhar et al., 2011). However, no impairment was found in the APP₆₉₅SWE mutant in two other studies (Chapman et al., 1999; Arendash et al., 2004). In the study with the positive result, 14- and 19-month-old animals were used (King and Arendash, 2002), whereas in the one with a null result from the same laboratory, 5-month-old mutants with few or no Aβ plaques were used (Arendash et al., 2004). Nevertheless, a null result appeared at 16 months of age when Aβ plaques are apparent (Chapman et al., 1999), a discrepancy that might be due to the mixed genetic background of the mutant, which differs according to local husbandry practices.

In any event, two other APP mutants with Aβ plaques were reported as being impaired in the suspended bar test: APP₇₅₁SWE/LD (Le Cudennec et al., 2008; Blanchard et al., 2009) and APP₇₅₁SWE/LD+PS1/M233+L235P (Cotel et al., 2012), the latter marked by spinal axonopathy. On the contrary, there was no impairment in three others with Aβ plaques, APP₇₅₁SWE (Lalonde et al., 2002b, 2005), APP₆₉₅SWE/co+PS1/ΔE9 (Lalonde et al., 2004), and 3xTg-AD (Gimenez-Llort et al., 2007), as well as three more without Aβ plaques in the brain, APP₆₉₅SWEch (Savonenko et al., 2003), APP₇₅₁WT (Moran et al., 1995), and APP/C99 (Lalonde et al., 2002a).

As in the previous test, it remains to be determined whether different results are caused by unequal distribution of Aβ accumulation in brain regions associated with motor coordination. An age-related study must be undertaken in the same mutant performing various motor coordination tests, with analyses of the distribution of Aβ plaques in the hindbrain in addition to the forebrain, focusing on brain regions known to be involved in both suspended bar and stationary beam tests such as the cerebellum (Lalonde and Strazielle, 1999).

Inclined screen

In one of the few Alzheimer models tested to date, APP₆₉₅ SWEch mice without Aβ plaques were slower than non-transgenic controls before turning upward on an inclined screen (Savonenko et al., 2003). APP₇₅₁SWE mice were not less likely to turn and climb the inclined grid, but speed was not measured (Lalonde et al., 2005).

Rotorod

The rotorod task is equivalent to the stationary beam task, except that the beam is moving, usually at an accelerated pace, augmenting the difficulty of mice to stay upright without falling (Lalonde and Strazielle, 1999). Rotorod performance was reported to be impaired in APP₇₅₁SWE transgenic mice bearing Aβ plaques (Van Dam et al., 2003; Dumont et al., 2004). This result was found in mutants prior to Aβ plaque formation, at 3 and 6 months of age, but was not reproduced after plaques had formed, at 16 and 24 months of age (Lalonde et al., 2002b), perhaps because of the declining abilities of older normal mice, causing a floor effect (no significant decrease is possible because normal values are too low). Nevertheless, one other APP mutant with Aβ plaques was impaired in this test, APP₆₉₅SWE+PS1/M146L (Ewers et al., 2006), as well as one without Aβ plaques, APP₇₅₁SWE/IND (Lee et al., 2004). On the contrary, five models with Aβ plaques performed normally on the rotorod: APP₆₉₅SWE (Dineley et al., 2002; Lalonde et al., 2003; Gil-Bea et al., 2007; Perucho et al., 2010), APP₆₉₅SWE+PS1/A246E (Dineley et al., 2002), APP₆₉₅SWE/IND (Hyde et al., 2005; Bellucci et al., 2006), and APP₆₉₅SWE/co+PS1/ΔE9 (Lalonde et al., 2004), as well as the APP₇₅₁WT mutant without plaques (Moran et al., 1995). Thus, different results are found in mutants irrespective of whether they bear Aβ plaques. As mentioned in the previous section, these questions can only be resolved by analyzing the distribution of Aβ accumulation in brain regions known to be involved in rotorod performance, such as the cerebellum and the dorsal striatum (Lalonde and Strazielle, 1999,2007).

As expected from an experimental model of inclusion body myopathy, rotorod performance worsened in mice expressing the APP/SWE/Ckm transgene in muscles relative to WT (Sugarman et al., 2002; Kitazawa et al., 2006). However, despite Aβ plaques in muscles, APP/C99 mice performed as well as controls in this test (Lalonde et al., 2002a). If confirmed within a single study, these contrasting data offer the possibility of examining the reason why Aβ accumulation in muscles leads to motor deterioration in one model but not in the other.

Concluding remarks

Several hallmarks of Alzheimer’s disease and inclusion body moyopathy have been reproduced in transgenic mice overexpressing the human APP gene, notably Aβ plaque-related pathology in the brain or muscles, respectively, and, in the case of the former, anterograde amnesia during maze testing. In addition, APP transgenics have been examined with respect to survival rates, various neurologic functions comprising myoclonus, seizure activity, and pathological reflexes, as well as motor coordination, all susceptible to be altered in Alzheimer dementia, together with inclusion body myopathy in the case of the latter. Several APP mutants expressing mutated or WT APP had higher mortality rates than non-transgenic controls with or without the presence of Aβ plaques, seen as well in those with vascular amyloid accumulation, thereby implicating cerebrovascular factors in all of them. In addition, myoclonic jumping and seizure activity have been described in APP mutants, likely the result of deficient neuronal inhibitory processes, as well as pathological reflexes such as limb-flexion and paw-clasping when mice are suspended in the air by the tail. Moreover, motor coordination deficits have been found in several APP mutants. However, as seen with myoclonic jumping and paw-clasping, some APP mutants with Aβ plaques do not display motor deficits. These conflicting results may be resolved by examining the role of Aβ-related pathologies in hindbrain regions.

Biographies

Prof. Robert Lalonde graduated from the University of Montreal, Department of Psychology, with a PhD degree in 1982. After postdoctoral training at the Clinical Research Institute of Montreal, he obtained a position as an Assistant Professor at the Department of Medicine at the University of Montreal. In 1996, he obtained a position as a Professor at the Department of Psychology at the University of Rouen. His main research work concerns the behavioral characterization of mutant mice as well as psychopharmacological studies. He has published 214 research articles.

Dr. Ken-ichiro (Ken) Fukuchi graduated from the Osaka University Medical School with a Medical Degree in 1979 and a PhD in 1985 in the field of medical genetics. He is a neuroscientist, molecular biologist and trained as a geriatrician who has dedicated his life to biomedical research. The Fukuchi Laboratory research has been focused on studying the molecular mechanisms underlying age-associated diseases such as Alzheimer’s disease and developing new preventive and therapeutic measures for over 20 years. Dr. Fukuchi is the Principal Investigator and recipient of numerous research grants from various resources including National Institutes of Health (NIH). Dr. Fukuchi has published over 70 scientific peer reviewed articles in this research area.

Dr. Catherine Strazielle graduated from the University of Lorraine with a MSc in Nutrition and a PhD in 1995 in Neuroscience, in the field of functional neuroanatomy. After postdoctoral training in Montreal in 1995, she began to collaborate with Robert Lalonde concerning the phenotypic characterization of animal models of human neurodegenerative diseases, particularly Alzheimer’s disease and cerebellar ataxia, evaluating by means of brain cartographies, the neurochemical changes linked to behavioral performances. She has published more than 60 articles in this field.