Project no.
FP6-503382
Project acronym:
SPASTICMODELS
Title:
Genetic Models of Chronic Neuronal Degeneration Causing
Hereditary Spastic Paraplegia
Instrument:
Specific Targeted Research Project
Thematic Priority:
1
Final Report
Period covered: from Jan 2004 to Aug 2007
Date of preparation: Oct 15, 2007
Start date of project: Jan 1, 2004
Duration: 44 months
Project coordinator name:
Giorgio Casari
Project coordinator organisation name:
San Raffaele Scientific Institute
1
Project execution
Project Objectives
1. Project summary
Neurodegenerative disorders, a heterogeneous group of chronic progressive diseases,
are among the most puzzling and devastating diseases in medicine. Indeed they are
characterized by onset in adult life, distinct clinical phenotypes, and specific degeneration of
subsets of neurons and axons. Hereditary spastic paraplegia (HSP) is a disorder that results in
progressive weakness and spasticity of the lower limbs affecting approximately 1 in 10000
individuals. Heterogeneity characterizes HSP in both clinical and genetic aspects.
Electrophysiological and pathological findings point to the corticospinal tracts, dorsal
columns and the spinocerebellar fibres as the structures primarily affected by HSP.
Two main pathogenetic hypotheses for the neurodegeneration seen in HSP have recently
emerged, suggesting that impaired mitochondrial function and/or defective subcellular
transportation mechanics play a crucial role. In fact, HSP-causing mutations have been found
in gene products involved in mitochondrial function, such as paraplegin and HSP60, and
axonal trafficking, such as kinesin and, possibly, spastin and spartin.
HSP has been classified traditionally as ‘pure’ or ‘complicated’, depending on whether
spastic paraplegia is the only symptom or whether it is found in association with other
neurological abnormalities, such as optic neuropathy, retinopathy, extrapyramidal symptoms,
dementia, ataxia, mental retardation and deafness.
Neuropathological analyses of tissues from a small number of individuals with pure HSP have
shown axonal degeneration involving the more distal portions of the longest motor and
sensory axons of the central nervous system (i.e. the crossed and uncrossed corticospinal
tracts, the fasciculus gracilis and the spinocerebellar tracts). A specific pattern of degeneration
is seen in HSP, during which the cell bodies remain largely intact while the degeneration is
principally limited to the cell axon and may be a ‘dying back’ axonopathy, beginning distally
and proceeding towards the cell body.
Little is known about known autosomal HSP genes encoding spastin (SPG4), paraplegin
(SPG7), atlastin (SPG6), spartin (SPG20), maspardin (SPG21), while Hsp60 is a wellcharacterised shock-response protein (SPG13) and the neural specific kinesin gene KIF5A
2
(SPG10) is a member of the kinesin superfamily of molecular motors that transport cargoes
along microtubules.
Atlastin is a novel GTPase that has sequence homology to members of the dynamin family of
large GTPases, particularly guanylate binding protein-1. Recently, it has been demonstrated
that spastin and atlastin interact, and, consequently, participate in the same biochemical
pathway. This confirms that the molecular pathology of spastin and atlastin HSP is related.
Spartin is responsible for Troyer Syndrome, an autosomal recessive HSP, and is homologous
to proteins involved in endosome morphology and protein trafficking of late endosomal
component. Data from Contractor 5 show that spartin transiently accumulates in the transGolgi network and subsequently decorates discrete puncta along neurites in terminally
differentiated neuroblastic cells. Investigation of these spartin-positive vesicles reveals that a
large proportion co-localises with the synaptic vesicle marker synaptotagmin.
Maspardin is responsible for Mast syndrome (a spastic paraplegia, autosomal recessive with
dementia). Both spastin and paraplegin are AAA proteins, ATPases associated with different
cellular activities. Members of this superfamily of ATPases exert chaperon-like activities and
mediate assembly and disassembly of macromolecular structures involved in different cellular
processes.
Paraplegin resembled a family of mitochondrial metalloproteases well-characterised in yeast
and was shown to localize in the mitochondrion. Mutations in paraplegin cause
neurodegeneration in an autosomal recessive form of HSP, but the pathogenetic mechanism
of this disorder, in particular the role of mitochondria, is presently not understood.
Paraplegin and the homologous protein AFG3L2 constitute the m-AAA protease complex.
Mitochondria of HSP patient (SPG7) have a functional deficit that could result from impaired
protein quality control, causing an accumulation of misfolded proteins within the
mitochondrial matrix. An additional mitochondrial protein, Hsp60, has been found mutated in
a form of HSP. Hsp60 is a representative of the type I chaperonin subfamily of molecular
chaperones that are involved in folding of bacterial, mitochondrial and chloroplast proteins.
Type I chaperonins are highly conserved throughout evolution and play an important role in
quality control of proteins, which seems to be a preferential mitochondrial target for SPG13
and SPG7.
Spastin interacts with microtubules and seems to modulate microtubule dynamics, an essential
attainment for maintenance of long axons. Although this major role of spastin in the
cytoplasm, Contractor 2 recently showed evidence that spastin is also a nuclear protein.
3
Nuclear targeting of spastin is regulated through two mechanisms, the usage of alternative
translational start sites and active nuclear export to the cytoplasm.
The specificity of the axonal damage in HSP apparently contrasts with the wide distribution
and range of functions carried out by the few gene products known to cause the clinical
phenotypes. However, we think to envisage two possible pathogenetic mechanisms. The
mitochondrial way seems to affect neuronal metabolism, in particular at the axon extremity,
by reducing the available energy. On the other hand, the cellular trafficking impairment
would limit the turnover of fresh molecules and organelles to and from the periphery of
neurons. Both mechanisms would result in a dying-back effect, typical of this late progressive
neurodegeneration.
A possible unifying alternative mechanism can be hypothesized: impaired mitochondrial
function generates huge aberrant mitochondria, which engulf the peripheral cellular
trafficking.
In the proposal, we planned to answer to this and other open questions on this specific
neurodegeneration by developing a formidable set of cellular and animal models of HSP. The
workplan included the development of seven novel animal models beside the further
characterisation of the paraplegin null mutant already generated, the functional
characterisation of the HSP mitochondrial dysfunction and the impaired mitochondrial protein
quality control, and the relationship between defective trafficking dynamics and axonal
degeneration.
More specifically, we have obtained two mouse models (one null and one carrier of a
missense mutation) for the paraplegin interactor Afg3l2, the conditional mutant for the
prohibitin Phb2 and the HSP60 null mutant. Conditional knockout model for Afg3l1 and
spartin and spastin null models are being generated.
This large set of murine models has been also used to generate double mutants to study the
possible genetic interactions among the different genes and to identify redundant or
independent pathways. A fine characterisation of the mouse phenotypes have been performed
for all the available models, following a common procedure including behavioural,
neuropathological and biochemical studies.
Mutant mouse tissues and derived cell lines have been examined for mitochondrial protein
quality control, protein folding and degradation efficiency. In addition, the characterisation of
the different m-AAA complexes in human and mice and the identification of binding partners
of the human m-AAA protease have been performed.
4
Furthermore, the role of spastin and spartin has been investigated to address the transportation
hypothesis and their role in vesicular dynamics. This is accomplished by identifying
interacting partners by yeast two-hybrid studies. The different mouse models will provide an
invaluable resource for the investigation of the transportation systems in HSP and range of
primary cultures are now being used to examine the axonal transport by neuronal tracers. A
description of these relevant results is detailed in the following sections.
Contractors and SPASTICMODELS Consortium
Coordinator
Giorgio Casari, PhD
Full Professor of Medical Genetics, Vita-Salute San Raffaele University and Head of the
Human Molecular Genetics Unit, Dept. of Neuroscience, DiBit-San Raffaele Scientific
Institute, Milan, Italy
The Human Molecular Genetics Unit, coordinated by G. Casari, is part of the Department of
Neuroscience, at DiBit-San Raffaele Scientific Institute in Milan. The DiBit building, which
occupies 12,000 square meters devoted to basic and applied research, hosts several common
facilities, including a specific pathogen-free animal house, a core facility for conditional
mutagenesis, expression profiling and protein microsequencing cores, a centralised imaging
facility, Alembic (Advanced Light and Electron Microscopy BioImaging Center).
The group of GC has a long-standing experience in molecular genetics of human hereditary
diseases. Characterisation and elucidation of pathogenetic mechanisms leading to inherited
neurological diseases represent the main goals of the unit. Due to the complex nature of the
pathological patterns, generation and phenotyping of animal models of human disease
represent powerful tools of the lab. The main project of the group is mostly relevant to the
present application and focuses on the motoneuron dysfunctions caused by the mitochondrial
m-AAA complex mutations. As we reported previously, loss-of-function of a m-AAA subunit,
paraplegin, causes hereditary spastic paraplegia (SPG7). Now, the group is concentrating on
axonal developmental defects that are caused by mutations of another m-AAA subunit, Afg3l2
and, finally, on new approaches to counteract axonal dysfunctions.
5
The group comprises 2 post-doctoral researchers, 4 PhD students (3 of them in their last year
of course), 2 researchers, and 2 technicians. A total of 2 postdocs and an experienced
researcher have been working exclusively on this project.
GC coordinated the Project and acted as Project Manager, together with the collaborator
Massimiliano Meoni.
Contractor 2
Elena I. Rugarli
Associate Professor of Medical Genetics, University Milano-Bicocca, Milan, Italy
Head of the laboratory of Genetic and Molecular Pathology, National Neurological Institute
“C. Besta”, Milan, Italy
The Unit operates within the National Neurological Institute “C. Besta”, Milan, Italy.
The Istituto Nazionale Neurologico “Carlo Besta” is an internationally recognized leading
centre in neuroscience and belong to the H.P.H. (Health Promoting Hospitals) network, a
project of the World Health Organization promoting healthcare. Major areas of expertise of
the Institute include: neurological disorders in adults and children, surgical and oncological
pathologies, chronic and rare diseases. Moreover, the Institute has ultraspecialized centres
devoted to specific neurological diseases. In Italy, the Institute is the main center of references
for patients with neurodegenerative diseases, including hereditary spastic paraplegias, and
provides diagnostic services at the molecular level. In the last few years, the unit headed by
EIR has focused on functional studies of two different genes involved in hereditary spastic
paraplegia (HSP) in both cellular and animal models. These are SPG7 encoding for
paraplegin, and SPG4 encoding for spastin. The main contribution of the group in the SPG7
field has been the generation and characterisation of a mouse model for SPG7/paraplegin
deficiency. This study was followed by a successful attempt to rescue the phenotype of Spg7 /-
mice in the peripheral nerve by viral gene delivery with AAV vectors. In the spastin field,
this group was the first to identify a role of spastin as a microtubule-severing protein. Further
important contributions have been the identification of a specific centrosomal interactor of
spastin, NA14, and the recognition of two different spastin isoforms generated through
alternative initiation of translation and with a different subcellular localization.
6
Scientific and technical personnel involved in the project. At present the group is composed
of 1 post-doctoral fellow, three PhD students, and two research assistants.
Contractor 3
Peter Bross, PhD
Associate Professor, Head of Research Unit for Molecular Medicine (RUMM), Aarhus
University
Research Unit for Molecular Medicine (RUMM) is located at Aarhus University Hospital
Skejby and belongs to the Aarhus University Institute of Clinical Medicine. RUMM performs
its own research projects in collaborations with research groups in Denmark and abroad and
also functions as a research service facility for clinical departments at Aarhus University
Hospital. RUMM’s own research projects comprise cellular stress and cellular protein quality
control in connection with human genetic diseases. A major part of the research is dedicated
to the molecular mechanisms associated with misfolding of mutant proteins in mitochondria.
The unit is a worldwide known research Centre for the identification and study of mutations
in genes encoding the acyl-CoA dehydrogenase family and other ß-oxidation enzymes. This
research has identified the prime importance of mitochondrial protein quality control in the
pathogenic mechanism and the role of the Hsp60/Hsp10 chaperone system is an integrated
part of it.
Associate professor Peter Bross’ group’s main research focus is on ’MITOCHONDRIAL
HSP60 IN NEURODEGENERATIVE DISEASES’. This research program comprises a series
of related research projects funded by a number of private foundations and governmental
funds set off by our discovery of the association of a mutation in the gene encoding the
mitochondrial Hsp60 chaperone with autosomal dominant hereditary spastic paraplegia
SPG13 (Hansen et al., 2002). Our group has determined the genomic structure of the
Hsp60/Hsp10 genes (Hansen et al. 2003). We have recently identified another Hsp60
missense mutation in a Danish spastic paraplegia patient (Hansen et al., 2007) and have
presently identified and characterised altogether 14 variations in the Hsp60 coding and
promoter region (Bross et al., 2007). Scientific and technical personnel involved in the
project. Peter Bross’ research group at Research Unit for Molecular Medicine working with
7
hereditary spastic paraplegia consists presently of 2 post-doctoral fellows and one biological
laboratory analyst.
Contractor 4
Thomas Langer, PhD
Institute for Genetics and Center for Molecular Medicine (CMMC)
University of Cologne Cologne Germany
The Department for Genetical Biochemistry is part of the Institute of Genetics currently. The
group is equipped to perform standard techniques for molecular biology and protein analysis
and is supported by core facilities of the institute for protein analysis/mass spectrometry,
mouse housing and DNA sequencing analysis. A centre for mouse genetics at the Institute for
Genetics offers facilities for animal housing and help for mouse germ line manipulation as a
central service. The centre is maintained by eight centrally funded positions for technical
stuff.
The research of the group, over recent years, has focused analysed over recent years on the
protein quality control system in of mitochondria using mainly yeast as a model system. The
work of the group leads to the identification of AAA proteases and other components of the
proteolytic system in the inner membrane of mitochondria, which were subsequently
functionally characterised. More recently, the dynamics of the mitochondrial proteome is
analysed by classical two-dimensional gel electrophoresis techniques as well as
multidimensional protein identification technology (MuDPiT). Studies on the quality control
of mitochondrial membrane proteins, the pathogenesis of HSP and on signalling pathways
between mitochondria and the nucleus represent other current research priorities of the group.
The group comprises three 3 post-doctoral scientists, seven 7 PhD students, and two 2
research technicians. Currently, one PhD-student and one research technician are working on
the pathogenesis of HSP. The requested funding has been used for one post-doctoral scientist
and one PhD-student. The postdoc was required to generate and characterise the conditional
prohibitin mouse model, the PhD-student has characterised the mitochondrial proteomes of
mouse models for HSP and analyse the role of paraplegin for mitochondrial biogenesis on the
8
molecular level. Additionally, funding for consumables and mouse housing costs were
requested. The equipment to conduct the proposed studies was available in the group.
Partner 4´s experience in the characterisation of mitochondrial biogenesis in general and of
mitochondrial proteins, in particular the field of AAA proteases, on the molecular level as
well as in mitochondrial proteins has been beneficial to all groups studying the mitochondrial
way of HSP. The research expertise of the group complements those of other Partners with
strong backgrounds in human genetics and pathology. An important link between those
projects where AAA proteases have been implicated in HSP pathology (spastin, paraplegin).
Contractor 5
Andrew H Crosby
Birth Defects Foundation Reader in Medical Genetics, St George’s College, University of
London
The Medical Genetics Unit, SGUL, is an internationally recognised research facility which as
part of SGUL comprises one section of London University (London, UK). It includes eight
senior scientists, six postdoctoral fellows, four research technicians and 12 PhD/graduate
students. Research at SGUL is facilitated by core facility equipment and services including
the BIOMICS facility (genomic and post-genomic sciences and equipment in the service of
Biomedical and Clinical Research), the Imaging unit (electron microscopy and
immunoflourescent technologies) and the Biological Research Facility (animal housing and
maintenance) as well as technical staff and support provided centrally and by the national
Genetic Knowledge Park scheme. The research in partner 5’s group is principally geared
towards the elucidation of the genetic and molecular basis of neurodegenerative disease. Full
time members of partner 5’s lab currently comprise 1 Lecturer Associate, 2 postdoctoral
scientists, 1 research technician and 3 PhD students. A main focus of the group is the
hereditary spastic paraplegias and over the last four years the group has identified the genes
mutated to cause for 4 forms of the condition (SPG17, SPG20, SPG21, manuscript in
preparation), as well as a gene for another neurodevelopmental condition (GM3 synthase
deficiency). Further to this, this group has recently identified the genes mutated to cause other
developmental and cardiovascular conditions including a form of lethal bone dysplasia, an
autosomal recessive form of hypertrophic cardiomyopathy at high frequency amongst the
9
Amish, a form of arrhythmogenic right ventricular cardiomyopathy (Naxos disease), Noonan
syndrome and Robinow syndrome. The spartin mouse model is being assessed in close
collaboration with Prof Nigel Brown (SGUL) and Dr Jenny Morton (University of Cambridge
UK), who have extensive experience in the characterisation of developmental and
neurological phenotypes in mice and are in possession of a current licence (PIL 80/1597).
Contractor 6
Thomas Deufel, MD
Professor and Head, Institut für Klinische Chemie und Laboratoriumsdiagnostik
Universitätsklinikum Jena, Jena, Germany
The Institute of Clinical Chemistry and Laboratory Diagnostics, with its staff of 12 academics
and 83 technical staff, is responsible for centralised laboratory diagnostic services for the
entire University Hospital of Jena in the fields of clinical chemistry, haematology,
endocrinology, medical immunology, and molecular diagnostics. J. Schickel is a senior
researcher within the Research Division of the institute and the group leader of in charge of
functional genetics. The Division is housed in the newly built Research Centre Lobeda (FZL)
of Jena University Hospital, and it has access to all relevant equipment, including cell culture
and animal research facilities, laser scanning and electron microscopy, and a wide range of
molecular biology, clinical physiology, and biochemistry research facilities. The unit is
integrated in the “Interdisciplinary Centre for Clinical Research” (IZKF) with its focus on
“Clinical Neurosciences”.
TD has a long-standing record in studies into the molecular genetics and functional genetics
of neurological and neuromuscular diseases, concentrating on identification and
characterisation of genes underlying genetically heterogeneous autosomal dominant disorders
such as Malignant hyperthermia susceptibility (MHS), deafness, and spastic paraplegia. A
number of new disease loci, e.g. for MHS and deafness, have been identified by the group,
respective genomic disease regions have been characterised and searched for candidate genes.
Work in HSP has been conducted in the group since 1998 and is focussed on phenotype
genotype correlation studies and, more recently, studies into the biochemistry, structure
biology, and functional genetics of spastin, including on-going work to establish mouse
models of different spastin mutations known to cause disease in humans.
10
The research group comprises 2 post-doctoral researchers, 1 PhD and four medical students as
well as five technicians. A total of 1 academic (post-doc) and 2 technical staff have been
working exclusively on this project.
The group has almost completed the generation of animal models carrying mutations in the
spastin gene involved in SPG4 pathology. In addition it has established stable transfectant cell
lines as well as biochemical, morphological, and cell biological systems to study the potential
action of spastin (and potentially other HSP genes) on the microtubular system. This work has
been closely linked to that of Partner 2.
11
Work performed and results
Generation and characterisation of animal models for HSP.
Mouse models for HSP due to dysfunction of the m-AAA complex
Afg3l2 mouse model
Two different Afg3l2 mutant mouse models are available in Coordinator’s lab: a spontaneous
mutant carrying a missense in the highly preserved AAA domain and a null model generated
by random insertion mutagenesis (Taylor & Rowe, 1989; Taylor et al., 1993). Both models
show a similar and extremely severe neuromuscular phenotype and they rarely survive
beyond 16 days. Detailed morphological analysis (See Task WP1.5) and primary neuronal
cell cultures from Afg3l2 mouse model (See Task WP3.3) indicate that Afg3l2 have an
important role in axonal development and maintenance. We decided to assign the highest
priority to the characterization of this new finding, thus postponing the generation of a
conditional knockout mouse, which was proposed as a Corrective Action within last Periodic
Activity Report and before this new information were available.
Afg3l1 mouse model
Two gene trap clones with an insertion in the murine Afg3l1 have been obtained.
Unfortunately homozygous mice derived from the first clone (BayGenomic) were only
hypomorph showing the presence of residual full length Afg3l1 mRNA and protein A second
gene trap clone was obtained from the German Gene Trap Consortium but no germline
transmission has been achieved. We are currently generating a traditional conditional allele
for the Afg3l1 gene with Artemis Pharmaceutical (Cologne, Germany). We have identified 4
targeted clones for injection.
A conditional mouse model for Phb
To study the role of prohibitins within mitochondria, we have generated conditional knockout
mouse model for PHB2 gene. The ubiquitous inactivation, as well as the tissue-specific
inactivation of PHB2 (brain, muscle or liver) revealed an embryonic lethal phenotype. We
have bred the PHB2flox/flox strain with a mouse strain expressing Cre-recombinase under the
control of the calmodulin-kinase-promoter, allowing the post-natal expression of Cre-
12
recombinase (at P20, with a peak of expression after three months). This resulted in reduced
brain and body weight and death of the homozygous animals after 3-4 month. Brain sections
of these mice are now analyzed in collaboration with contractor 2.
Development of a SPG13/Hsp60 mouse model for HSP
We have developed a heterozygous Hsp60+/- knockout mouse model to substitute for the
proposed Hsp60wt/V72I mouse model (not produced due to technical problems). We have
obtained Hsp60+/- mice with 99% C57BL/6 genetic background by backcrossing using a
marker-assisted selection protocol.
Development of mouse models for SPG4/spastin
Spastin knockout mutation:
Since no positive ES clones were identified using a spastin nonsense vector for homologous
recombination, we have used a trapped spastin ES clone (German Gene Trap Consortium,
GGTC). Several chimeric animals were obtained and breeding is currently under way.
Spastin missense mutation in the AAA-domain (K388R dominant HSP mutation):
A first vector carrying the K388R mutation was generated and electroporated, with no
positive results. A new vector was constructed and electroporated. 5 positive clones injected
Several times, which resulted in more than 120 chimeric animals. Currently, breeding did not
result in germline transmission.
Spastin missense mutation at the N-terminus (analogous to human HSP-modifying mutation
S44L)
The S44L spastin variation was published as a polymorphism or modifying mutation
aggravating the human HSP phenotype. We have solid evidence including further pedigrees
and patients where the mutation co-segregates with another disease mutation in spastin, as
well as expression and functional data, confirming the modifying role of S44L. Hence, two
different vectors have been constructed and used for electroporation of ES cells. Currently ES
cell screen for homologous recombination is under way.
Development of spartin mouse model
The aim of this project was to generate a Spg20 knock-in mouse model to replicate the
mutation identified in Amish Troyer syndrome patients (1110delA). In order to achieve this,
we introduced an 1102delA mutation into exon 4 of the murine gene. To date we have
13
obtained ~40 mice that are heterozygous for the 1102delA and exon 5 point mutations and
have set up mating of these animals in order to generate homozygous knock-in mice.
Phenotypic characterisation of the mouse models
Afg3l2 mouse model
Coordinator’ lab has performed an extensive morphological and biochemical characterization
both Afg3l2 mutant mouse models. Afg3l2 mutants show an extremely severe neuromuscular
syndrome, with deficits of upper and lower motoneurons leading to lethality at P16. In
contrast with the late onset axon degeneration observed in paraplegin null model, Afg3l2
mutants show a severe defect in axonal development characterized by delayed myelination
and impairment of axonal radial growth in both CNS and peripheral nervous system (PNS).
Mitochondrial morphology abnormalities are also detected in motor and sensory neurons of
CNS and PNS, more frequently in proximity of the nucleus or the plasma membrane.
Moreover, the enzymatic activities of the respiratory chain complexes are strikingly impaired
in Afg3l2 models, resulting in highly reduced ATP production.
Spg7-/- mouse model
We went further in the neuropathological characterization of Spg7-/- mouse model, by
examining mice older than 19 months. In brain, we observed many morphological alterations,
such as enlargement of the ventricular system, thickness reduction of cerebral cortex and
corpus callosum, vacuolization of pyramidal neurons. A decrease in the total number of
neurons was apparent. In the PNS, the number of spinal motor neurons was decreased and the
remaining neurons showed signs of degeneration.
A gene therapy approach to Spg7 -/- mice
Contractor 2 attempted to design a gene therapy strategy to rescue the phenotype of the spinal
motoneurons in paraplegin-deficient mice. These cells have been targeted by AdenoAssociated Viral (AAV) vectors after intramuscular delivery. Our results suggest that AAV
gene therapy is effective both in preventing the motor phenotype of paraplegin-deficient mice
and in slowing the progression of neuropathological changes in the peripheral nerves.
Notably, a consistent rescue of the mitochondrial abnormalities was achieved.
SPG13/Hsp60 mouse model
We have shown that the presence of a GENETRAP integration in both alleles of the Hsp60
gene, as expected, result in peri-implantational (between 5.75 and 8 dpc) lethality. Hsp60+/14
mice display no severe obvious phenotype. Histological analysis of brain and spinal cord of
our oldest Hsp60+/- mice with mixed genetic background has been initiated. Heterozygous
Hsp60+/- mice have decreased Hsp60 transcript and protein levels in all tissues examined, i.e.
brain, heart, skeletal muscle and liver.
Conditional mouse model for Prohibitin
Contractor 4 has performed the phenotypic analysis of a PHB2 deletion in immortalized
mouse embryonic fibroblast cell lines homozygous for the floxed PHB2 allele. Transduction
of these cell lines with a recombinant chimeric Cre-recombinase (NLS-tat-Cre) allows
inactivation of PHB2 in these cells. Our experiments revealed an essential role of prohibitins
for cell proliferation
Mouse models for SPG4/spastin
During previous reporting periods, immunohistochemical studies on different regions and
developmental stages (E14, E18, P2, P10, P31) of the wild-type mouse brain were performed
to analyse the natural localisation of spastin. We found an overall distribution of spastin in the
brain including neuronal and glial cells. Spastin was predominantly localised in the nucleus;
however a cytoplasm tic distribution was found, too. Furthermore, in cultured neuronal mouse
brain cells an axonal localisation of the protein was observed.
Generation of double mutants
Generation of Spg7 /Afg3l2 double mutants
In this reporting period, Coordinator and Contractor 2 generated mice that are double mutants
for Spg7 and Afg3l2. Double heterozygous mice for Spg7 and Afg3l2Emv66 and for Spg7 and
Afg3l2par were obtained and intercrossed. Animals that are null for Spg7 and heterozygous for
a mutation in Afg3L2 at 9 weeks display a clear neurological phenotype characterized by
abnormal gait, loss of balance, tremor, dystonic movements and, consequently. They die
between 13 and 20 weeks of age. Neuropathological analyses showed that long spinal axons
of Spg7 -/-;Afg3L2Emv66/+ mice undergo axonal swellings much earlier than Spg7 -/- mice and,
moreover, ultrastructural analysis revealed morphological alterations in mitochondrial cristae
and structure and accumulation of cytoskeletal components in swollen axons. Histological and
immunohistochemical studies revealed also clear signs of Purkinje cells degeneration.
15
The mitochondrial way to HSP
Characterisation of the different m-AAA complexes in human and mice mouse
We generated model proteins as substrates of protease and chaperone-like activities. The
amino-terminal part of the model proteins consists of a portion of the human mitochondrial
ribosomal protein MrpL32 including the mitochondrial targeting signal and m-AAA protease
cleavage signal. The carboxy-terminal part is the wild-type form of mouse dihydrofolate
reductase (wtDHFR) or a mutated form (mutDHFR) bearing three point mutations that cause
misfolding of the protein. These fusion-proteins were cloned in eukaryotic expression vectors
and tagged with HA-epitope (MrpL32-wtDHFR-HA and MrpL32-mutDHFR-HA). Moreover
we generated two fusion proteins containing the transmembrane domains of paraplegin to
allow proper localisation of the model proteins within the inner mitochondrial membrane,
exposing DHFR portion to the matrix (MrpL32-paraTMs-wtDHFR-HA and MrpL32paraTMs-mutDHFR-HA). The mitochondrial specific processing of the fusion proteins has
been confirmed. The stability of these proteins has been tested after transfection in HSP
primary fibroblasts or paraplegin knockdown cell lines (see Task WP2.3). Preliminary results
show that the absence of functional m-AAA protease results in a decreased rate of degradation
of the model proteins carrying mutant DHFR.
By continuing the collaborative effort with Coordinator and Contractor 2, we have identified
various isoenzymes of m-AAA proteases in murine and human mitochondria which differ in
their subunit composition. Homo-oligomeric AFG3L2- and, in mice, AFG3L1- complexes as
well as hetero-oligomeric complexes containing paraplegin were identified (Koppen et al.,
2007). Moreover, complementation studies in yeast demonstrated their functional activity in
protein quality control, protein processing, and protein dislocation. Using subunit-specific
antibodies we have continued to determine the relative abundance of different m-AAA
protease subunits in different tissues and observed striking tissue-specific variations in their
expression. Accordingly, the selective loss of different m-AAA protease subunits is expected
to result in different phenotypes in mice as has been confirmed in the meantime by the
analysis of AFG3L2-/- mice by contractor 1.
16
Gene expression studies, mitochondrial stress response and mitochondrial protein
misfolding in HSP mouse models and patient cell lines
Total RNA has been isolated from embryonic fibroblast cell lines and from spinal cords of 12
month-old paraplegin-deficient mice (Spg7-/-) and appropriate controls. These reagents have
been used for microarray studies in order to identify genes whose expression levels are
changed in absence of paraplegin. At present, no significant transcriptional changes were
found.
Studies of mitochondrial function parameters in cultured fibroblast or lymphoblastoid cells
from a SPG13 patient showed that the Val98Ile mutation in Hsp60 does not significantly
affect the subcellular localisation of Hsp60, mitochondria volume, mitochondrial membrane
potential, cell vitality, and sensitivity to oxidative stress insults.
Protein and transcript studies
We designed four short hairpin RNAs (shRNA) targeting the coding sequence of paraplegin
mRNA. Cotransfection of paraplegin-myc- and shRNA- expressing vectors demonstrated
RNAi efficiency. We generated HeLa cells stably expressing these RNA hairpins and we
obtained a good efficiency of paraplegin silencing with two of them. We are currently
generating paraplegin knockdown in SH-SY5Y cell lines.
Since in Afg3l2 mutants we detect deficiencies in respiratory chain complexes I and III, we
evaluated the possible consequent oxidative damage measuring carbonyl formation, which is
an easy detectable marker of protein oxidation. Preliminary results indicate that the level of
protein carbonylation is increased in Afg3l2 mutants, both in brain and spinal cord.
Studies of purified Hsp60-(p.Val98Ile) variant protein show that it is incorporated into
chaperonin complexes, but those complexes containing the mutant chaperonin display
decreased ATPase activity and a significantly reduced capacity to promote folding of
denatured malate dehydrogenase in vitro.
Microarray analysis of a bacterial model system lacking the endogenous chaperonin genes
and expressing the mutant Hsp60 chaperonin showed no global induction of heat shock genes.
In contrast a number of metabolic enzymes and transporters were strongly induced.
Identification of binding partners of the human m-AAA protease
17
m-AAA complex-interacting proteins have been isolated by affinity purification and identified
by mass spectrometry. This analysis revealed the interaction between the human m-AAA
protease and mtHsp70, the central component of mitochondrial import machinery. We
confirmed this evidence by co-immunoprecipitation experiments in Hela cells after
transfection of paraplegin or AFG3L2 and mtHsp70 tagged with different epitopes. We
performed GST pull-down assay after either paraplegin or AFG3L2 transfection in HeLa cells
or in vitro translation and import into mitochondria, thus confirming the binding of m-AAA
protease to mtHsp70.
Role of paraplegin in mammalian mitochondria
Exogenous expression of wt paraplegin or AFG3L2 or the corresponding proteolytic variants
carrying a mutation in the metalloprotease zinc-binding site in HeLa cells did not affect
mitochondrial morphology. In contrast, exogenous expression of AFG3L2 mutants with
impaired ATPase activity (mutations in Walker A and Walker B sites of the AAA domain)
induced mitochondrial fragmentation. We can hypothesise that the ATPase activity of mAAA protease is required in the regulation of mitochondrial fusion.
No
impairment
in
the
oxidoreductase
activity
of
respiratory
complexes,
both
spectrophotometrically and using a Clark-type electrode oxygraph, is detected in synaptosome
and non-synaptic mitochondria from the brain of 20-month-old Spg7 -/- animals and agematched controls.
Our previous studies revealed murine MrpL32, a subunit of mitochondrial ribosomes, as the
first known substrate of the paraplegin-containing m-AAA protease in mammalian
mitochondria (Nolden et al., 2005). In the meantime, we have continued to analyse the
activity of the m-AAA protease using Saccharomyces cerevisiae as a model system and
focused on its role during maturation of cytochrome c peroxidase (Ccp1) in mitochondria.
The m-AAA protease was found to mediate the ATP-dependent membrane dislocation of
Ccp1 independent of its proteolytic activity. allowing intramembrane cleavage by rhomboid.
Defective subcellular transportation and HSP
18
Yeast two-hybrid studies
The identification of binding partners of spastin
We have identified NA14, a centrosomal protein, as a spastin interactor by yeast two-hybrid
screening of a human foetal brain cDNA library. Co-immunoprecipitation experiments in
HeLa cells demonstrate that NA14 and spastin interact in vivo and that binding to NA14
contributes to the microtubule localization of spastin.
The identification of binding partners of spartin
In order to investigate the function of spartin the yeast two-hybrid system has been employed
to identify binding partners. In house studies have been undertaken in parallel with an
outsourced project Proteinlinks and the German Cancer Research Centre utilising human
foetal brain libraries. The reproducibility and authenticity of positive transformants are
currently being examined.
Immunolocalisation and cell studies
We previously showed that SPG4 synthesizes two spastin isoforms by usage of different
AUGs as translational starting site. We have identified and defined a novel transcriptional
mechanism that regulates the relative abundance of spastin isoforms in tissues and cells.
To dissect the role of spastin in neurite outgrowth, we downregulated spastin with short
interfering RNA duplexes in NSC34 cells, a murine immortalized cell line. This results in an
increase in the number of cells showing neurites longer than 50 micron and in the average
length of the processes. In contrast spastin overexpression significantly decreased the amount
of acetylated tubulin, and suppressed neurite outgrowth. These results suggest that
microtubule severing may regulate neuronal length in vivo and that tight regulation of spastin
levels is necessary for normal development of neurite outgrowth.
We have performed cell studies using a large battery of constructs (see Section 2 for details),
but transfection with spartin constructs has proven problematical showing very low
efficiency. Furthermore, N-tagged molecules (wild type or mutant) look very similar
displaying a diffuse cytoplasmic staining, while C-terminal tagged constructs (wild type or
mutant) produces aggregate-like accumulations within the cytoplasm. We have also studied
the localisation of spartin in primary neurons, which showed a ubiquitous localisation both in
developing neurons (DIV6) and in mature cells (DIV14). Staining was seen both in dendrites
and axonal processes as well as in the nucleus. The nuclear localisation at DIV6 was very
19
intense but appeared to be absent in some cells by DIV14, this is in agreement with our
previous observations made in differentiated neuronal-like culture cells. In developing
neurons at DIV6 spartin co-localised with the growth cone marker GAP-43 and was also
present in axonal varicosities as evidenced by its co-localisation with the axonal marker Tau1. This was similar to published observations of the localisation of another protein mutated in
HSP, atlastin (Zhu et al 2006), in primary neurons.
We confirmed differences in subcellular localisation, showed capability for active nucleocytoplasmic shuttling, identified a mechanism involved in expression regulation, and revealed
differences as regards microtubule-severing activity. Building on our previous identification
of nuclear localisation sequences in spastin (Beetz et al., 2004, BBRC 318:1079) we also
aimed at clarifying spastins nuclear function. Extensive characterisation of sub-nuclear
localisation suggested a role in transcription. These data, obtained in collaboration with P.
Hemmerich (HKI Jena), are in preparation for publication.
Our close collaboration with clinicians had enabled us to identify SPG4 deletions as a novel
and unexpectedly frequent cause of HSP (Beetz et al., 2006. Neurology 67:1926). The
multitude of heterozygous-loss-of-spastin HSP families now described by us clearly establish
haploinsufficiency as a disease mechanism. Our patient screening also identified the first case
of a partial SPG4 duplication..
For the spastin interaction partner atlastin, itself being the product of an HSP gene (SPG3A),
we could show that haploinsufficiency is not a relevant mechanism in this form of the disease.
This was concluded from non-segregation of an SPG3A whole gene deletion in an SPG4 HSP
pedigree (Beetz et al., 2007, Neurogenetics, 2007 in press). A meaningful SPG3A animal
model, though not part of the SPASTIC MODELS project, would therefore have to be a
knock in rather than a knock out.
Finally, we identified large rearrangements in HSP patients in a number of other HSP genes
such as SPG6 (NIPA1), SPG7 (paraplegin), and SPG31 (REEP1). Testing for disease
association of these variants is under way.
Studies on transportation dynamics in vivo and in vitro
20
We have generated mouse embryonic fibroblast cell lines (MEFs) and primary dorsal root
ganglia (DRG) from Afg3l2 mouse models. Consistently with the phenotype in vivo, Afg3l2deficient primary DRG show impairment in extending and maintaining long neurites.
At present, isolated DRG are used to investigate the axon-myelin interplay, which appears
affected in both Afg3l2 models. Furthermore, MEF cell lines have been established from
paraplegin-Afg3l2 double mutant strains in collaboration with Contractor 2.
We introduce novel approaches like neuronal differentiation of murine embryonal stem cells
carrying relevant mutations and of adult human stem cells from genetically diagnosed HSP
patients.
21
Dissemination and use
Communication strategies
The SPASTICMODELS web site.
The Project website is now established (www.spasticmodels.org) and is used for communicating
the progress in knowledge on spastic paraplegia. General updated information on spastic
paraplegia and other neurodegenerative diseases are provided. It includes the following principal
sections: a homepage with the main features of the pathology, the pathogenetic hypotheses and the
aim of the project; a description of the Consortium with links to each member’s homepage; the
genetics of Hereditary Spastic Paraplegia; useful links; relevant publications of the members of the
Consortium.
Project logo
The following logo has been proposed. After reaching general consensus by
SPASTICMODELS Contractors, it is being adopted and shown in every document and public
presentation where the EU-FP6-503382 SPASTICMODELS is acknowledged.
Kick-off meeting: Milan, Italy, May 3rd, 2004.
The Coordinator organized the kick-off meeting in Milan on May 2004. In this occasion, all
22
participants were present and gave slide presentations of the main specific aims. Active
interactions among partners on specific tasks were discussed. In fact, the Coordinator’s lab hosted
a Contractor 4’s PhD student for three months to set up experiments on the functional role of the
human paraplegin/AFG3L2 complex in human control and mutant fibroblasts.
Mid-Term Review Meeting and HSP Workshop: June 13th- June 14th, 2005.
The Mid-Term Review Meeting took place on June 13-14, 2005 at San Raffaele Scientific
Institute in Milan in the presence of the External Reviewer, Dr. Olaf Riess, and the EU-FP6
Scientific Officer, Dr Jurgen Sautter. The first day of the meeting was dedicated to the update of
the current status of the 4 WorkPackages. WP Coordinators presented achievements in relation to
the WP milestones. Dr. Sautter presented new opportunities for funding and innovative
characteristics of the next Seventh FrameWork Program. On the second day, we entered the HSP
Workshop. Partners’ young collaborators gave a close look of their specific research projects, with
special emphasis on sharing experimental details of several applied technologies. The poster
session in the afternoon allowed continuing and deepening of scientific discussions.
It follows the participants list: Drs. Ries Olaf and Jurgen Sautter; Giorgio Casari, Luigia Atorino,
Luana Bardinella, Laura Cassina, Francesca Maltecca, Laura Silvestri (Coordinator’s group);
Elena Rugarli, Elena Riano, Andrea Bernacchia (Contractor 2); Peter Gerd Bross, Jakob Hansen,
Jane Hvarregaard (Contractor 3); Thomas Langer, Metodiev Metodi, Koppen Mirko (Contractor
4); Andrew Harry Crosby, Heema Patel, Johanna Reed, Michael Simpson, Dimitri Claude Robay
(Contractor 5); Beetz Christian, Schickel (Contractor 6).
3rd year SPASTICMODELS Meeting. Milan, February 5-6, 2007
The Consortium met at the San Raffaele Scientific Institute in Milan. During these two intense
days, all Contractors presented updates of specific projects with wide discussion on experimental
results. A particular attention has been spent to review the state-of-the-art of the HSP mouse
models generation, which is a focal point of the Project.
A session of the meeting was devoted to evaluate the potential of this Consortium to present an
application to the next FP7-HEALTH call for grants. Based on experimental achievements and
effective intra-Consortium collaborations, all Contractors agreed to circulate a FP7 draft proposal
in the next 2-3- weeks.
Meeting participants were:
Giorgio Casari, Laura Cassina, Francesca Maltecca, Raffaella Magnoni (Coordinator’s group);
Elena Rugarli, Elena Riano, Andrea Bernacchia (Contractor 2); Peter Gerd Bross, Jane
23
Hvarregaard (Contractor 3); Thomas Langer (Contractor 4); Andrew Harry Crosby, Heema Patel,
(Contractor 5); Thomas Deufel, Christian Beetz, Jörg Schickel (Contractor 6).
Design of standardised dissemination materials and reagents
Animal models generated within this project are available, in a first phase, to the Consortium
for scientific purposes in respect of the Intellectual Property Rights agreement accepted by the
Participants (to be signed within November 2003). After patenting and publishing relevant
data obtained through the animal models, these latter will be available to the Scientific
Community through existing European channels.
Deployment of dissemination campaign
Performed, so that these can be monitored and reported to the Commission.
Summer School
Active interactions among partners have been taking place during the Project period. In fact, the
Contractor 3 is presently hosting a Coordinator’s PhD student to exchange expertises in the
biochemical and neuropathological characterization ot the Hsp60 and Afg3l2 mouse models,
respectively.
Public awareness
Participants have direct connections to patients associations and charity organizations that
have great impact to non specialized general audience in Europe. In particular, the TomWahlig-Stiftung
(http://www.fsp-info.de/)
and
the
Italian
Telethon
Foundation
(http://www.telethon.it/) will be updated with the new results obtained within the Consortium
and translated to information comprehensible to a large audience.
We have prepared a set of information (one or several board members in charge) on
hereditary spastic paraplegia and related neurodegenerative diseases, the relevance of research
on this topic and the expertise of the FP6 funded consortium. Every consortium member can
use this basic material to inform local press and other channels on accomplishments of the
project.
24
Relevant scientific publications on international journals from Consortium members
are:
2007
Graef M, Seewald G, Langer T. Substrate recognition by AAA+ ATPases: Distinct substrate
binding modes in the ATP-dependent protease Yme1 of the mitochondrial intermembrane
space. Mol Cell Biol. 27(7):2476-85, 2007.
Tatsuta T, Augustin S, Nolden M, Friedrichs B, Langer T. m-AAAprotease-driven membrane
dislocation allows intramembrane cleavage by rhomboid in mitochondria. EMBO J. Jan;
24;26(2):325-35, 2007.
Osman C, Wilmes C, Tatsuta T, Langer T. Prohibitins Interact Genetically with Atp23, a
Novel Processing Peptidase and Chaperone for the F1FO-ATP Synthase. Mol Biol Cell.
Feb;18(2):627-35, 2007.
Koppen M, Metodiev MD, Casari G, Rugarli EI, Langer T. Variable and tissue-specific
subunit composition of mitochondrial m-AAA protease complexes linked to hereditary spastic
paraplegia. Mol Cell Biol. 27:758-67, 2007
Duvezin-Caubet S, Koppen M, Wagener J, Zick M, Israel L, Bernacchia A, Jagasia R, Rugarli
EI, Imhof A, Neupert W, Langer T, Reichert AS. OPA1 processing reconstituted in yeast
depends on the subunit composition of the m-AAA protease in mitochondria. Mol Biol Cell.
18(9):3582-90, 2007.
Bross P, Li Z, Hansen J, Hansen JJ, Nielsen MN, Corydon TJ, Georgopoulos C, Ang D,
Lundemose JB, Niezen-Koning K, Eiberg H, Yang H, Kolvraa S, Bolund L, Gregersen N.
Single-nucleotide variations in the genes encoding the mitochondrial Hsp60/Hsp10 chaperone
system and their disease-causing potential. J Hum Genet. 52(1):56-65, 2007.
Hansen J, Svenstrup K, Ang D, Nielsen MN, Christensen JH, Gregersen N, Nielsen JE,
Georgopoulos C, Bross P. A novel mutation in the HSPD1 gene in a patient with hereditary
spastic paraplegia. J Neurol. Jul;254(7):897-900, 2007.
Tsaousidou MK, Ouahchi K, Warner TT, Yang Y, Simpson MA, Laing NG, Wilkinson PA,
Madrid RE, Patel H, Hentati F, Patton MA, Hentati A, Lamont PJ, Siddique T, Crosby AH.
Sequence alterations within [gene name withheld] implicate defective cholesterol homeostasis
in motor neurone degeneration. Am J Hum Genet (in press).
Dick KJ, Al-Mjeni R, Baskir W, Koul R, Simpson MA, Patton MA, Raeburn S, Crosby AH.
A novel locus for an autosomal recessive hereditary spastic paraplegia (SPG35) maps to
16q21-q23. Neurology (in press).
Schickel J, Pamminger T, Ehrsam A, Münch S, Huang X, Klopstock T, Kurlemann G,
Hemmerich P, Dubiel W, Deufel T, Beetz C. Isoform-specific increase of spastins stability by
N-terminal missense variants including intragenic modifiers of SPG4 hereditary spastic
paraplegia. Eur J Neurol. 2007 Oct 3 [Epub ahead of print].
25
Mitne-Neto M*, Kok F*, Beetz C*, Pessoa A, Bueno C, Graciani Z, Martyn M, Monteiro
CBM, Mitne G, Hubert P, Nygren AOH, Valadares M, Cerqueira AMP, Starling A, Deufel T,
Zatz M. A multi-exonic SPG4 duplication underlies sex-dependent penetrance of hereditary
spastic paraplegia in a large Brazilian pedigree. Eur J Hum Genet. 2007 Sep 26 [Epub ahead
of print].
Beetz C, Nygren AO, Deufel T, Reid E. An SPG3A whole gene deletion neither co-segregates
with disease nor modifies phenotype in a hereditary spastic paraplegia family with a
pathogenic SPG4 deletion. Neurogenetics. 2007 Jul 27; [Epub ahead of print]
Beetz C, Zuchner S, Ashley-Koch A, Auer-Grumbach M, Byrne P, Chinnery PF, Hutchinson
M, McDermott CJ, Meijer IA, Nygren AO, Pericak-Vance M, Pyle A, Rouleau GA, Schickel
J, Shaw PJ, Deufel T. Linkage to a known gene but no mutation identified: comprehensive
reanalysis of SPG4 HSP pedigrees reveals large deletions as the sole cause. Hum Mutat 28:
739-40, 2007.
2006
Pirozzi M, Quattrini A, Andolfi G, Dina G, Malaguti MC, Auricchio A, and Rugarli EI.
Intramuscular viral delivery of paraplegin rescues peripheral axonopathy in a model of
hereditary spastic paraplegia. J.Clin.Invest Jan;116(1):202-8, 2006.
Rugarli EI, Langer T. Translating m-AAA protease function in mitochondria to hereditary
spastic paraplegia. Trends Mol Med. 12: 262-269, 2006.
Andrenacci D, Grimaldi MR, Panetta V, Riano E, Rugarli EI, Graziani F. Functional
dissection of the Drosophila Kallmann's syndrome protein DmKal-1. BMC Genet. 7:47, 2006.
Robay D, Patel H, Simpson MA, Brown NA, Crosby AH. Endogenous spartin, mutated in
hereditary spastic paraplegia, has a complex subcellular localization suggesting diverse roles
in neurons. Exp Cell Res. Sep 10; 312(15):2764-77, 2006.
Bross P, Li Z, Hansen J, Hansen JJ, Nielsen MN, Corydon TJ, Georgopoulos C, Ang D,
Lundemose JB, Niezen-Koning K, Eiberg H, Yang H, Kolvraa S, Bolund L, Gregersen N.
Single-nucleotide variations in the genes encoding the mitochondrial Hsp60/Hsp10 chaperone
system and their disease-causing potential. J Hum Genet. 52(1):56-65, 2007.
Beetz C, Nygren AO, Schickel J, Auer-Grumbach M, Burk K, Heide G, Kassubek J, Klimpe
S, Klopstock T, Kreuz F, Otto S, Schule R, Schols L, Sperfeld AD, Witte OW, Deufel T.
High frequency of partial SPAST deletions in autosomal dominant hereditary spastic
paraplegia. Neurology 67:1926-30, 2006.
Schickel J, Beetz C, Frommel C, Heide G, Sasse A, Hemmerich P, Deufel T. Unexpected
pathogenic mechanism of a novel mutation in the coding sequence of SPG4 (spastin).
Neurology 66:421-3, 2006.
2005
26
Nolden M, Ehses S, Koppen M, Bernacchia A, Rugarli EI, and Langer T. The m-AAA
Protease Defective in Hereditary Spastic Paraplegia Controls Ribosome Assembly in
Mitochondria. Cell 123:277-289, 2005.
Maggi R, Cariboni A, Zaninetti R, Samara A, Stossi F, Pimpinelli F, Giacobini P, Consalez
GG, Rugarli E, and Piva F. Factors involved in the migration of neuroendocrine hypothalamic
neurons. Arch Ital Biol. 143:171-178, 2005.
Claudiani P, Riano E, Errico A, Andolfi G, and Rugarli EI. Spastin subcellular localization is
regulated through usage of different translation start sites and active export from the nucleus.
Exp Cell Res. Oct 1;309(2):358-69, 2005.
DiSchiavi E, Riano E, Heye B, Bazzicalupo P, and Rugarli EI. UMODL1/Olfactorin is an
extracellular membrane-bound molecule with a restricted spatial expression in olfactory and
vomeronasal neurons. Eur J Neurosci. 21:3291-3300, 2005.
Gregersen N, Bolund L, and Bross P. Protein misfolding, aggregation, and degradation in
disease. Mol Biotechnol. 31:141-150, 2005.
Corydon TJ, Hansen J, Bross P, and Jensen TG. Down-regulation of Hsp60 expression by
RNAi impairs folding of medium-chain acyl-CoA dehydrogenase wild-type and diseaseassociated proteins. Mol Genet Metab. 85:260-270, 2005.
Hansen J, Gregersen N, and Bross P. Differential degradation of variant medium-chain acylCoA dehydrogenase by the protein quality control proteases Lon and ClpXP. Biochem
Biophys Res Commun. 333:1160-1170, 2005.
Kambacheld M, Augustin S, Tatsuta T, Muller S, and Langer T. Role of the novel
metallopeptidase Mop112 and saccharolysin for the complete degradation of proteins residing
in different subcompartments of mitochondria. J Biol Chem. 280:20132-20139, 2005.
Tatsuta T, Model K, and Langer T. Formation of membrane-bound ring complexes by
prohibitins in mitochondria. Mol Biol Cell. 16:248-259, 2005.
Augustin S, Nolden M, Muller S, Hardt O, Arnold I, and Langer T. Characterization of
peptides released from mitochondria: evidence for constant proteolysis and peptide efflux. J
Biol Chem. 280:2691-2699, 2005.
Reed JA, Wilkinson PA, Patel H, Simpson MA, Chatonnet A, Robay D, Patton MA, Crosby
AH, and Warner TT. A novel NIPA1 mutation associated with a pure form of autosomal
dominant hereditary spastic paraplegia. Neurogenetics 6:79-84, 2005.
Brockmann K, Simpson MA, Faber A, Bonnemann C, Crosby AH, Gartner J. Complicated
hereditary spastic paraplegia with thin corpus callosum (HSP-TCC) and childhood onset.
Neuropediatrics 36(4):274-8, 2005.
Wilkinson PA, Simpson MA, Bastaki L, Patel H, Reed JA, Kalidas K, Samilchuk E, Khan R,
Warner TT, Crosby AH. A new locus for autosomal recessive complicated hereditary spastic
paraplegia (SPG26) maps to chromosome 12p11.1-12q14. J Med Genet. Jan;42(1):80-2,
2005.
27
2004
Ferreirinha F, Quattrini A, Pirozzi M, Valsecchi V, Dina G, Broccoli V, Auricchio A,
Piemonte F, Tozzi G, Gaeta L, Casari G, Ballabio A, Rugarli EI Axonal degeneration in
paraplegin-deficient mice is associated with abnormal mitochondria and impairment of axonal
transport. J Clin Invest 113, 231-42, 2004.
Errico A., Claudiani P., D’Addio M., Rugarli E.I. Spastin interacts with the centrosomal
protein NA14, and is enriched in the spindle pole, the midbody and the distal axon. Hum Mol
Genet 13, 2121-32, 2004
Orlacchio A; Kawarai T, Totano A, Errico A, St George-Hyslop PH, Rugarli EI, Bernardi G.
Hereditary spastic paraplegia: Clinical genetic study of 15 families. Arch Neurol. 61:849-855,
2004.
Wilkinson PA, Simpson M, Bastaki L, Patel H, Reed J, Samilchuk E, Khan R, Warner TT,
Crosby AH. A new locus for autosomal recessive complicated hereditary spastic paraplegia
(SPG26) maps to chromosome 12p11.1-12q14. J. Med Genet. 36(11):1225-9, 2004.
Proukakis, C, Cross, H, Patel, H, Patton, MA, Valentine, A, Crosby, AH. Troyer syndrome
revisited: A clinical and radiological study of a complicated hereditary spastic paraplegia. J.
Neurol. 251(9):1105-10, 2004.
Sperfeld A-D, Kassubek J, Crosby AH, Winkler J, Uttner I, Ludolph AC, Hanemann CO.
Complicated hereditary spastic paraplegia with thin corpus callosum: remarkable changes in
phenotypic expression in follow up investigations. J Neurol 251(10):1285-7, 2004.
Windpassinger C, Auer-Grumbach M, Irobi J, Patel H, Petek E, Legenstein A, Malli R, Ines
Dierick I, Warner T, Proukakis C, Van Den Bergh P, Verellen C, Van Maldergem L, Merlini
L, De Jonghe P, Timmerman V, Crosby AH & Wagner K. Missense mutations in the BSCL2
gene cause distal hereditary motor neuronopathy type V as well as Silver syndrome. Nat
Genet. 36(3):271-6, 2004.
Wilkinson PA, Crosby AH, Turner C, Bradley LJ, Ginsberg L, Wood NW, Schapira AH,
Warner TT. A clinical, genetic and biochemical study of SPG7 mutations in hereditary spastic
paraplegia. Brain 127(Pt 5):973-80, 2004.
Warner TT, Patel H, Proukakis C, Reed JA, McKie L, Wills A, Patton MA, Crosby AH. A
clinical, genetic and candidate gene study of Silver Syndrome, a complicated form of
hereditary spastic paraplegia. J Neurol. 251(9):1068-74, 2004.
Beetz C, Brodhun M, Moutzouris K, Kiehntopf M, Berndt A, Lehnert D, Deufel T, Bastmeyer
M, Schickel J. Identification of nuclear localisation sequences in spastin (SPG4) using a novel
Tetra-GFP reporter system. Biochem Biophys Res Commun. 318:1079-1084, 2004.
Scientific oral and poster communications:
2007
28
G. Casari “Animal models for HSP”. GeneMove symposium, Bonn, Germany May 2007
E.I. Rugarli "The role of the mitochondrial m-AAA protease in neurodegeneration" Molecular
Mechanisms of Neurodegeneration (3rd Meeting) - Milan, Italy. May 19-21, 2007
Maltecca F, Cassina L, Magnoni R, Cox G.A., Guenet J.L., Previtali S.C., Quattrini A. and
Casari G. “Neurodegeneration and axonal development: the dual role of the mitochondrial
paraplegin-AFG3L2 complex”. Molecular Mechanisms of Neurodegeneration (3rd Meeting) Milan, Italy. May 19-21, 2007
Beetz C, Byrne P, Depienne C, Reid E, Schöls L, Nygren A, Schickel J, Deufel T. Copy
number aberrations in hereditary spastic paraplegia. 6. Symposium der Tom-Wahlig-Stiftung
in Verbindung mit der Tagung der Deutschen Gesellschaft für Klinische Neurophysiologie.
March 23rd – 24th 2007; Munich, Germany.
2006
Beetz C, Schickel J, Auer-Grumbach, Schöls L, Zuchner S, Deufel T. MLPA-based screening
for disease-causing copy number aberrations in hereditary spastic paraplegia. 12th Annual
Meeting of the German Society of Neurogenetics. October 13th – 15th 2006; Rostock,
Germany.
Schickel J, Beetz C, Deufel T. The human SPG4 variant c.131C>T: detection, prevalence and
pathology. 6.Hj.-Staudinger Symposium der DGKL. June 25th – 26th 2006; Kloster Banz,
Germany.
Beetz C, Schickel J, Nygren A, Auer-Grumbach M, Klimpe S, Kassubek J, Klopstock S, Otto
S, Schöls L, Schüle R, Sperfeld A, Deufel T. High frequency of partial SPG4 gene deletions
in hereditary spastic paraplegia. 6. Symposium der Tom-Wahlig-Stiftung in Verbindung mit
der Tagung der Deutschen Gesellschaft für Klinische Neurophysiologie. March 24th – 25th
2006; Bad Nauheim, Germany.
Elena I. Rugarli “Dissecting hereditary spastic paraplegia to rescue degenerating axons”
Gordon Conference Molecular Cell Biology, 4-7-2006
Tilton College, New Hampshire, USA
Elena I. Rugarli “Genetica della sofferenza assonale nel sistema nervoso centrale”
CORSI RESIDENZIALI DI NEUROIMMUNOLOGIA, BERGAMO,
8-11 MARZO 2006
2005
Atorino L, Cassina L, Maltecca F, Bardinella L Silvestri L, Guenet JL, Cox G, Quattrini A,
Casari G. “Hereditary Spastic Paraplegia due to mitochondrial defects: development of animal
and cellular models of pathogenesis”. Telethon Convention 2005, Salsomaggiore, Parma,
March 6-8, 2005.
29
Atorino L, Maltecca F, Cassina L, Bardinella L, Silvestri L, Quattrini A., Cox G, Guénet JL,
Casari G. Paraplegin complex defect in cellular and animal models of Hereditary Spastic
Paraplegia, II Meeting on the Molecular Mechanisms Of Neurodegeneration, Universita' degli
Studi di Milano, Milan (Italy).
Bross P, Hansen J, Kruhøffer M, Georgopoulos C, Ang D, Nielsen MN, Ørntoft T, Gregersen
N. The HSP60-V72I mutant variant detected in patients with hereditary spastic paraplegia is
incorporated into chaperonin complexes and displays impaired chaperonin function. 2nd CSSI
International Congress on Stress Responses in Biology and Medicine, Tomar, Portugal,
September 24-28 2005.
Beetz C, Schickel J, Deufel T. Functional identification of a classical nuclear export signal in
spastin (SPG4) and generation of constructs targeting over-expressed protein to the nucleus.
European Journal of Human Genetics 13 (suppl.1):292.
Ehrsam A, Beetz C, Schickel J, Deufel T. A new assay for the S44L variant in spastin (SPG4)
allows to determine its prevalence among patients with hereditary spastic paraplegia (HSP)
and controls and provides further evidence for a disease-modifying effect in SPG4-linked
HSP. Clinical Chemistry and Laboratory Medicine 43(9):A018.
Beetz C, Nygren AOH, Schickel J, Lens SI, Klimpe S, Auer-Grumbach M, Kreuz F, Heide G,
Deufel T. 2006. MLPA-based screening for disease-causing copy number alterations in the
hereditary spastic paraplegia genes SPG3A and SPG4. (to be presented at ECHG 2006 in
Amsterdam).
Beetz C, Sasse A, Schickel J, Hemmerich P, Deufel T. Functional identification of a classical
nuclear export signal in spastin (SPG4) and generation of constructs targeting over-expressed
protein to the nucleus. European Human Genetics Conference. May 7-10, 2005; Prague
Congress Center, Prague, Czech Republic.
Beetz C, Ehrsam A, Schickel J, Deufel T. The S44L variant in spastin (SPG4) and its role in
hereditary spastic paraplegia. 11th Annual Meeting of the German Society of Neurogenetics.
September 8-10, 2005; Westfälische Wilhelms Universität, Münster, Germany.
Schickel J, Beetz C, Frömmel C, Sasse A, Heide G, Hemmerich P, Deufel T. Unexpected
mutational mechanism of a single base change in SPG4 revealed by mRNA and functional
analysis. 11th Annual Meeting of the German Society of Neurogenetics. September 8-10, 2005;
Westfälische Wilhelms Universität, Münster, Germany.
Ehrsam A, Beetz C, Schickel J, Deufel T. A new assay for the S44L variant in spastin (SPG4)
allowsto determine its prevalence among patients with hereditaryspastic paraplegia (HSP) and
controls and provides furtherevidence for a disease-modifying effect in SPG4-linked HSP. 2nd
Annual Conference of the German United Society for Clinical Chemistry and Laboratory
Medicine (DGKL). October 6-8, 2005; Friedrich-Schiller-Universität, Jena, Germany.
Schickel J, Beetz C, Heide G, Sasse A, Hemmerich P, Deufel T. Functional characterisation
of spastin mutations causing hereditary spastic paraplegia. 35th annual meeting of the Society
for Neuroscience. November 8-12, 2005; Washington Convention Center, Washington, DC,
USA.
30
2004
T.Langer: 6th European Meeting on Mitochondrial Pathology (EUROMIT6), Nijmegen,
Netherlands, June 30th- July 4 th, 2004, “Regulation of mitochondrial activity by proteolysis”.
A.H. Crosby: American Society of Human Genetics 2004, “Infantile onset symptomatic
epilepsy syndrome caused by a homozygous loss of function mutation in GM3 synthase”.
A.H. Crosby: American Society of Human Genetics 2004, “Investigation of the subcellular
localisation of spartin, mutated in a complicated form of HSP, in neuronal lines NSC34 and
SH-SY5Y”.
G. Casari: International Congress on Amino Acid / Protein Metabolism on Health and
Disease, Milan, March 27, 2004, "Mitochondrial genomics and proteomics".
Bross, P., Hansen, J., Kruhøffer, M., Georgopoulos, C., Ang, D., Nielsen, M.N., Corydon,
T.J., Ørntoft, T., Bolund, L., Gregersen, N. “Investigation of the effects of the V72I mutation
in human Hsp60 that is associated with hereditary spastic paraplegia”. Cold Spring Harbor
Meeting ‘Molecular Chaperones and the Heat Shock Response’, Cold Spring Harbor, NY,
U.S.A., May 5th-9th, 2004.
T. Langer, “Proteolytic control of mitochondrial biogenesis”, Gordon Research Conference
"Mitochondria & Chloroplasts, Meriden, NH, July 24th -31st, 2004.
Cassina L., Atorino L , Silvestri L , Casari G. “Functional Role of paraplegin/AFG3L2
Complex in Mitochondria”, 3rd Dibit/IEO-IFOM campus joint PhD student Workshop 2004,
18-20 October 2004, Riva del Garda.
Maltecca F, Atorino L, Cassina L, Silvestri L, Casari G, "Study of the paraplegin/AFG3L2
complex in an Afg3l2-/- conditional mouse model". 3rd Dibit/IEO-IFOM campus joint PhD
student Workshop 2004, 18-20 October 2004, Riva del Garda.
31
Publishable executive summary
Hereditary spastic paraplegia (HSP) comprises a group of clinically and genetically
heterogeneous diseases that affect the upper motor neurons and their axonal projections. It is
characterised by progressive lower-limb weakness, spasticity and subtle impairment of the
vibratory sense. The age of onset of the disease is quite variable, generally lying between 10
and 40 years. Degeneration of the lateral corticospinal tracts is a typical pathological
presentation of HSP which increases in severity caudally.
A total of 30 chromosomal loci have been identified for autosomal dominant, recessive and
X-linked HSP. The underlying genes for 15 of these loci have been described. The molecular
dissection of the cellular functions of the related gene products has already greatly advanced
our understanding of the most critical pathways involved in HSP.
In particular, two main hypotheses for the neurodegeneration seen in HSP have been
proposed: abnormal mitochondrial function and defective subcellular transportation
mechanics. A specific mitochondrial malfunction seems to affect neuronal metabolism, in
particular at the axon ends, by reducing the available energy. The cellular trafficking
impairment would limit the turnover of fresh molecules and organelles to and from the
periphery of neurons. Both mechanisms would result in a dying-back effect, typical of this
late-onset progressive neurodegeneration.
Aim of this project was to investigate these hypotheses by developing and analysing a set of
cellular and animal models of HSP. The workplan included the development of seven novel
animal models beside the further characterisation of the paraplegin null mutant already
generated, the functional characterisation of the HSP mitochondrial dysfunction and impaired
mitochondrial protein quality control, and the relationship between defective trafficking
dynamics and axonal degeneration.
More specifically, during the project we have obtained two mouse models (one null and one
carrier of a missense mutation) for the paraplegin interactor Afg3l2, the conditional mutant
for the prohibitin Phb2 and the HSP60 null mutant. Conditional knockout model for Afg3l1
and spartin and spastin null models are being generated. A fine characterisation of the mouse
phenotypes have been performed for all the available models, following a common procedure
including behavioural, neuropathological and biochemical studies. This large set of murine
models has been also used to generate double mutants to study the possible genetic
interactions among the different genes and to identify redundant or independent pathways.
Furthermore, the therapeutic potential of Spg7 gene delivery by viral vectors was tested on
32
Spg7
-/-
mice with important successful results. Mutant mouse tissues and derived cell lines
have been examined for mitochondrial protein quality control, protein folding and degradation
efficiency. In addition, the characterisation of the different m-AAA complexes in human and
mice and the identification of binding partners of the human m-AAA protease have been
performed. New functions of the m-AAA protease have been described: the regulation of
mitochondrial protein synthesis and the ATP-dependent membrane dislocation of specific
proteins independent of its proteolytic activity. Moreover, a novel transcriptional mechanism
regulating the relative abundance of spastin isoforms in tissues and cells, and the spartin
intracellular localization and possible functions have been identified and defined.
It is hoped that in the foreseeable future this knowledge will begin to translate into novel
pharmacological approaches for this devastating disease.
33