R

Adn

MTa

b

h

•••

a

ARR2AA

KPAA

1

nnpdicew

(

h0

Neuroscience Letters 609 (2015) 223–228

Contents lists available at ScienceDirect

Neuroscience Letters

journa l homepage: www.e lsev ier .com/ locate /neule t

esearch paper

myotrophic lateral sclerosis-associated mutant profilin 1 increasesendritic arborisation and spine formation in primary hippocampaleurons

erryn Brettle a, Alexandra K. Suchowerska a, Sook W. Chua b, Lars M. Ittner b,∗,1,homas Fath a,∗,1

Neurodegeneration and Repair Unit, School of Medical Sciences, University of New South Wales, AustraliaDementia Research Unit, School of Medical Sciences, University of New South Wales, Australia

i g h l i g h t s

Mutant PFNC71G expression in cultured hippocampal neurons promotes dendrite extension and branching.PFNC71G increases dendritic spine density in cultured hippocampal neurons.PFNC71G forms cytoplasmic inclusions in hippocampal neurons.

r t i c l e i n f o

rticle history:eceived 9 June 2015eceived in revised form5 September 2015

a b s t r a c t

Amyotrophic lateral sclerosis (ALS) is the most common motor neuron disease and familial ALS accountsfor 10% of cases. The identification of familial ALS mutations in the actin-binding protein profilin 1directly implicates actin dynamics and regulation in the pathogenesis of ALS. The mechanism by whichthese mutations cause ALS is unknown. In this study we show that expression of the ALS-associated

ccepted 26 September 2015vailable online 21 October 2015

eywords:rofilin 1myotrophic lateral sclerosis

actin-binding deficient mutant of PFN1 (PFN1C71G) results in increased dendritic arborisation and spineformation, and cytoplasmic inclusions in cultured mouse hippocampal neurons.

© 2015 Elsevier Ireland Ltd. All rights reserved.

ctin cytoskeleton

. Introduction

Amyotrophic Lateral Sclerosis (ALS) is a rapidly progressiveeurodegenerative disease affecting both upper and lower motoreurons [16]. Treatments are limited, with the average life spanost-diagnosis being 3 years [30]. Although familial forms of theisease only account for approximately 10% of all cases [28], var-

ous histopathological hallmarks, such as TDP-43 inclusions, areonserved across familial and sporadic cases [24,33]. Our knowl-

dge of mechanisms in the pathogenesis of ALS are incomplete,hich is reflected in the lack of treatment options for the disease.

∗ Corresponding authors.E-mail addresses: l.ittner@unsw.edu.au (L.M. Ittner), t.fath@unsw.edu.au

T. Fath).1 These authors are equal senior authors.

ttp://dx.doi.org/10.1016/j.neulet.2015.09.034304-3940/© 2015 Elsevier Ireland Ltd. All rights reserved.

Pathogenic mutations in the profilin 1 (PFN1) gene were identi-fied in familial ALS [34]. Although it is estimated that PFN1 is onlyresponsible for less than 1% of familial ALS cases [18,28,32], identi-fying how mutant PFN1 causes ALS will further our understandingof the complexities of the disease and may aid in the development oftreatments. Although cytoskeletal dysfunction has previously beensuggested to be associated with ALS [3,33], the identification ofPFN1 mutations directly implicates the actin cytoskeleton in thepathogenesis of the disease.

PFN1 is a 12–15 kDa ubiquitous actin-binding protein [21]. Itsinvolvement in actin filament polymerization is well characterizedand it carries out both an inhibitory and a promoting role in actinfilament extension [10]. The actin cytoskeleton is essential for avariety of cellular functions, which include cell division, morpho-

genesis and motility. Dysfunction in the actin cytoskeleton has beenextensively studied in neuronal function and development as wellas in neurodegenerative diseases (reviewed in [4,12]).

2 nce Le

arofPtatA(

aiseg

2

2

(NiTpAscIp

2

mppfidtpp#(mTm

2

P(TopLctp

r

24 M. Brettle et al. / Neuroscie

Limited functional analysis has been carried out on the ALS-ssociated PFN1 mutations. It is known that mutant PFN1 has aeduced ability to bind to actin, which results in reduced axonalutgrowth in primary mouse motor neurons, and have an impairedunction in stress granules in yeast [13,34]. Due to the role thatFN1 plays in actin filament elongation, the hypothesis emergedhat ALS-associated PFN1 mutations impact the dynamics of thectin cytoskeleton, which affects neuronal morphology and func-ion; and that these effects contribute to neuronal dysfunction inLS not only in motor neurons but also central nervous system

CNS) neurons of the hippocampal region.Here, we studied the impact of the most common ALS-

ssociated PFN1 mutation (C71G) [34] on neuronal morphogenesisn mouse primary hippocampal neurons, showing that the expres-ion of PFN1C71G leads to a specific increase in the length of dendritextension and arborisation while having no obvious effect of axonalrowth.

. Materials and methods

.1. Plasmids and cloning

Human PFN1 was amplified from human embryonic kidneyHEK293T) cell cDNA with flanking restriction enzyme sites and an-terminal V5 epitope tag. The amplified PFN1 insert was ligated

nto the T-overhangs within the multiple cloning site of the pGEM- easy vector (Promega), where site directed mutagenesis waserformed as previously described (Quickchange Multi SDM kit,gilent) [19] to introduce the C71G mutation. Human and mouseequences of PFN1 at the C71 site are identical. The inserts wereonfirmed by sequencing (Macrogen) and sub-cloned into a pLVX-RES-puro (Clontech) expression vector. mNeonGreen vector wasrovided by Allele Biotechnology & Pharmaceuticals [31].

.2. Primary neuronal cell culture

All cell culture experiments in this study were carried out withouse primary hippocampal neurons. Hippocampal neurons were

repared from E16.5 embryos from C57 black 6 mice as describedreviously [11]. In brief, hippocampi were dissected out using ultra-ne micro-dissecting scissors (Fine science tools), enzymaticallyigested for 20 min at 37 ◦C with 1x trypsin (Sigma–Aldrich) andriturated using fire-polished glass Pasteur pipettes. Cells werelated in DMEM with 10% FBS at a density of 3.7 × 104 cells/cm2 onoly-D-lysine (Sigma–Aldrich) coated glass coverslips (ø 12 mm,1.0 thickness, glass; Menzel-Glaser, Germany) in 24-well plates

Corning, USA). After 2 h, the medium was changed to Neurobasaledium (Life Technologies) supplemented with 2% (v/v) B27 (Life

echnologies) and 0.25% (v/v) Glutamax (Invitrogen). The cells wereaintained at 37 ◦C and 5% CO2 in a humidified atmosphere.

.3. Transfection

Primary mouse hippocampal neurons were transfected with V5-FN1wt or V5-PFN1C71G in pLVX-IRES-puro and a control pmaxGFPLonza, Switzerland) construct using Lipofectamine® LTX (Lifeechnologies) at 3 days in vitro (DIV). Before transfection, 500 �Lf conditioned medium were set aside. The lipid complex was pre-ared by adding 200 ng of plasmid and 0.6 �L of Lipofectamine®

TX to 100 �L of Neurobasal medium and briefly vortexing. Theomplex was incubated at RT for 25 min then 100 �L were added

o the culture. Medium change with conditioned medium was com-leted after 60 min incubation (Table 1).

Double-transfections with mNeonGreen protein as transfectioneporter and either no pLVX, V5-PFN1wt or V5-PFN1C71G in pLVX-

tters 609 (2015) 223–228

IRES-puro were performed in a similar manner except 66.7 ng ofeach plasmid were added.

2.4. Antibodies

Antibodies used for immunocytochemistry were as follows:mouse anti-V5 (1:200, life sciences), chicken anti-�3 tubulin(1:500, Millipore), rabbit anti-TDP 43 (1:400, Proteintech), donkeyanti-mouse Alexa Fluor-488 (1:500, Invitrogen), donkey anti-rabbitAlexa Fluor-488 (1:500, Invitrogen), donkey anti-mouse AlexaFluor-555 (1:500, Invitrogen), goat anti-chicken Alexa Fluor-555(1:500, Invitrogen) and donkey anti-mouse Alexa Fluor-647 (1:500,Invitrogen).

2.5. Immunocytochemistry

Cells were fixed for 20 min with 4% paraformaldehyde(ProSciTech, Australia) at 5 DIV for the single transfection or at19 DIV for the double-transfection. Fixed cells were permeabilizedwith 0.1% Triton X-100 for 5 min, and blocked in 2% FBS diluted inPBS for 30 min. Cells were incubated for 1 h at RT with primary anti-bodies diluted in blocking solution (PBS with 2% FBS) followed by40 min incubation with secondary antibodies, diluted in blockingsolution. The coverslips were mounted onto glass slides (Living-stone, Australia) with Prolong Gold antifade mounting mediumwith DAPI (Life Technologies).

2.6. Imaging

Stained coverslips were imaged on a BX51 microscope (Olym-pus, Japan) with a 40× and 100× objective with the Neurolucida(MBF Bioscience, USA) program. Morphometric analysis of neuronswas performed on Neurolucida with AutoNeuron interactive andmanual neuron tracing. Five neurons per condition across 3 inde-pendent sets were analysed. The majority of the neurons in theculture (>90%) and the cells that were analysed in this study, weremultipolar with 3-8 dendrites. The data were exported to Excel(Microsoft, USA) using Neurolucida Explorer.

Immunofluorescence images of dendritic spines were acquiredwith a FluoView FV1000 confocal microscope (Olympus, USA) usingthe UplanSApo 60 × 1.35 NA oil immersion objective. Z-stacks weretaken at 0.4 �m intervals with four images taken per cell. The num-ber of spines in 20 �m long segments of dendrites at a distance of50 �m from the cell body was analysed. Spine number was analysedusing Image J (Version 1.47).

2.7. Statistical analysis

Data was processed in Excel and analyzed in GraphPadPrism (v.6.00, GraphPad Software, USA). Outliers were removedusing ROUT analysis. One way ANOVA with multiple compar-isons were used on parametric data. All values are presented asmean ± standard error.

3. Results

3.1. PFN1C71G increases dendritic arborisation of hippocampalneurons

To investigate the effects of PFN1 on neuronal development,we first generated expression constructs encoding V5-tagged wild-type (wt) and C71G mutant human PFN1. Hippocampal neurons

were transfected with V5-PFN1wt, V5-PFN1C71G or a GFP controlvector at 3 DIV, fixed at 5 DIV and then immuno-stained withantibodies directed against the V5-tag and �3-tubulin (Fig 1A-F). V5-PFN1wt and V5-PFN1C71G localised to both the axonal and

M. Brettle et al. / Neuroscience Letters 609 (2015) 223–228 225

Table 1Morphological changes in V5-PFN1C71G and V5-PFN1wt expressing hippocampal neurons.

V5-PFN1C71G V5-PFN1wt Control p-Value compared to control

V5-PFN1C71G V5-PFN1wt

Dendritic length [�m] 199.3 ± 32.9 113.6 ± 4.02 81.98 ± 15 0.013* 0.503Number of branches per dendrite 5.04 ± 0.67 1.85 ± 0.38 1.68 ± 0.11 0.001*** 0.647Axonal length [�m] 1206 ± 196.50 873.5 ± 88.65 1200 ± 50.27 0.999 0.201Number of branches per axon 25.17 ± 3.38 17.33 ± 2.46 25.38 ± 3.14 0.999 0.423Number of spines per 20 �m dendrite segment 13.11 ± 0.22 9.98 ± 0.55 9.02 ± 0.09 0.002** 0.317Percent of V5-positive aggregates 3.47

± 1.50 N/A p-Value, V5-PFN1C71G compared to V5-PFN1wt

0.034*

* P ≤ 0.05.** P ≤ 0.01.

*** P ≤ 0.001.

F euronm quanl N1wt

tAdn(f(eHle

3n

pVhmVpcw

ig. 1. The C71G mutation in PFN1 promotes dendritic arborisation. Hippocampal netrically at 5 days in vitro and dendritic and axonal extension and branching were

ength in neurons expressing PFN1C71G as compared to mNeonGreen control and PF

he somato-dendritic compartment in neurons grown for 5 days.nalysis of neuronal morphology showed significantly increasedendrite extension (2.4-fold) and branching (3-fold) of dendrites ineurons expressing V5-PFN1C71G as compared to GFP control cellsFig. 1G, H), and a significant increase in dendritic branching (2.7-old) in V5-PFN1C71G as compared to V5-PFN1wt expressing cellsFig. 1G). Overexpression of V5-PFN1wt had no effect on dendritextension or branching as compared to GFP control cells (Fig. 1G,). Interestingly, there was no significant difference in axonal tree

ength or branching between V5-PFN1wt, V5-PFN1C71G and GFPxpressing cells (Fig. 1I, J).

.2. PFN1C71G increases dendritic spine density in hippocampaleurons

As PFN1C71G had a dendrite-specific effect on hippocampal mor-hology in developing neurons, we next looked at the effect of5-PFN1wt and V5-PFN1C71G expression in long-term cultures ofippocampal neurons. Neurons were transfected at 3 DIV withNeonGreen with or without co-transfection of V5-PFN1wt or

5-PFN1C71G. mNeonGreen was used to visualise the entire mor-hology of transfected neurons, including the dendritic spineompartments. Cells were fixed at 19 DIV and immuno-stainedith an antibody directed against the V5-tag (Fig 2A-F). Spine

s expressing GFP control, V5-PFN1wt or V5-PFN1C71G (A–F) were analysed morpho-tified (G–J). Note the significant increase in dendritic branching and mean dendriteexpressing neurons. **p ≤ 0.01, *p ≤ 0.05. Scale bars = 20 �m.

morphology was predominantly mature with no overt differencesin the proportion of mature and filopodial spines between thedifferent conditions. Quantification of dendritic spines showedsignificantly higher numbers of dendritic spines in hippocampalneurons, expressing V5-PFN1C71G (13.1 ± 0.2 spines per 20 �m ofdendrite segment) as compared to neurons expressing V5-PFN1wt

(10 ± 0.6 spines per 20 �m of dendrite segment) or control neurons(9 ± 0.1 spines per 20 �m of dendrite segment) (Fig 2G). No statis-tically significant difference was found between cells expressingV5-PFN1wt and control neurons, indicating that the effect on spinedensity is linked to the C71G mutation in PFN1.

3.3. PFN1C71G expression induces PFN1 aggregation inhippocampal neurons

Insoluble protein aggregates are a hallmark of neurodegener-ative diseases. Wu et al. showed aggregates of PFN1C71G in N2aneuroblastoma cells and primary motor neurons treated withMG132 [34]. To determine whether the PFN1 mutations could causecytoplasmic aggregates in CNS neurons, hippocampal neurons were

transfected at 3 DIV, fixed at 4 or 5 DIV and then stained withV5 and �3-tubulin antibodies. Cells transfected with V5-PFN1C71G

showed sporadic aggregation (3.5 ± 1.5% of all transfected cells ana-lysed; n = 214) (Fig. 3). Staining with antibodies specific for TDP-43

226 M. Brettle et al. / Neuroscience Letters 609 (2015) 223–228

F pocaP ritic sc alized

smn

4

2

ig. 2. The C71G mutation in PFN1 leads to increased dendritic spine density. HipFN1C71G (A–F) were analysed at 19 days in vitro for dendritic spine density. Dendontrol and PFN1wt expressing neurons (G). Higher magnification of dendrites visu

how no TDP-43 immunoreactivity of the aggregates or cytoplas-ic localization of TDP43 (Fig. S1). No aggregates were found in

eurons transfected with V5-PFN1wt (n = 468) (Fig. 3).

. Discussion

Pathogenic mutations in PFN1 were identified in familial ALS in012 [34]. Wu and colleagues have found that mutant PFN1C71G has

mpal neurons expressing mNeonGreen alone, or together with V5-PFN1wt or V5-pine density is significantly increased in neurons expressing as compared to GFP

with mNeonGreen (B, D, F) *p ≤ 0.05. Scale bars = 20 �m.

reduced actin-binding in co-immunoprecipation experiments [34].A subsequent study [7] used an actin polymerisation assay to showthat the mutant PFN1 had the same inhibitory effect as PFN1wt inhigh concentrations, which highlights the need for further studies

into the impact of the disease-associated PFN1 mutation and thefunctional consequence on actin dynamics. Mutant PFN1C71G hasalso an increased susceptibility for aggregation [32]. The impactof PFN1 mutations on different CNS neurons, however, remains

M. Brettle et al. / Neuroscience Letters 609 (2015) 223–228 227

F pocamt wt (A)n

ucpf

iiaooumttrato[naesmma

fcdPtAocca11afiidwmaca

ig. 3. Hippocampal neurons form spontaneous aggregates of mutant PFN1C71G. Hiphe presence of protein aggregates. Shown are exemplary neurons expressing PFN1euron (arrow in insert). Scale bars = 20 �m.

nclear. In the current study we analysed the effect of the mostommon mutation in PFN1, C71G, on neuronal structure in mouserimary neurons from the hippocampal region. The hippocampal

ormation is impacted at late stages in ALS pathology [1,5–7,24].We show that expression of PFN1C71G results in a significant

ncrease of dendrite length and branching. We observed no changesn axon length or branching. Hence, expression of PFN1C71G has

compartment specific effect on neuronal morphology. Neuriteutgrowth is highly dependent on the regulation of the dynamicsf the actin cytoskeleton and actin-associated proteins can reg-late axonal and dendritic outgrowth in a compartment specificanner (for a review, see [15]). Therefore, it is not surprising

hat inhibition of profilin/actin interaction and thereby modula-ion of actin dynamics impacts on neurite formation. However, ouresults are in contrast to the findings by Wu et al., who showedttenuated axonal length in cultured motor neurons in responseo PFN1C71G expression [34]. Furthermore, no phenotypic changesf the somato-dendritic compartment were reported in this study34]. A possible explanation for the lack of a somato-dendritic phe-otype in Wu’s study is the early time point of analysis (3DIV)t which dendrites are poorly developed. Alternatively, the differ-nces in phenotype reported may represent a neuronal cell typepecific effect of mutations in PFN1. An in vivo study in SOD1G85R

ice by [14], showed an increase in dendritic branching in lowerotor neurons, demonstrating that mutations in other familial ALS-

ssociated genes can also result in a somato-dendritic phenotype.Concomitant with the increased dendritic arborisation, we

ound an increase in the density of dendritic spines in matureultures of hippocampal neurons. Proper control of actin filamentynamics is critical for dendritic spine stability and function [17].revious work has implicated function of the related profilin 2 inhe stabilisation of dendritic spines in hippocampal neurons [2].ccordingly, using a vector-based RNAi approach for knockdownf PFN2a in hippocampal slice cultures showed reduced dendriteomplexity and spine numbers of CA1 neurons [23], which wasompensated by PFN1 overexpression. However, spine morphologynd synaptic plasticity were not affected in a study using profilin

deficient mice. These data suggest a functional role of profilin at the synapse with partial overlap in the function of profilin 1nd profilin 2.While profilin 1 primary function is to promote actinlament assembly, it can inhibit the activity of the actin nucleat-

ng complex Arp2/3 [29]. The ability of profilin1 to inhibit Arp2/3epends on its ability to bind to actin monomers. PFN1 interactsith the proline-rich motifs of Ena/VASP and formin and the C71G

utation in the actin-binding region of PFN1is not expected to

ffect these interactions. Therefore, an increase in Arp2/3 activityould account for the significant increase in dendritic arborisationnd spine density observed in hippocampal neurons expressing

pal neurons expressing PFN1wt or PFN1C71G were analysed at 4–5 days in vitro for or PFN1C71G (B). Note the aggregates in the cytoplasm of the PFN1C71G expressing

PFN1C71G. Without the inhibition of Arp2/3, dendritic arborisa-tion could increase by de novo branch formation along the shaftof the developing dendrites. Also, Arp2/3 is an integral componentof branched actin filament populations in the dendritic spine com-partment, in particular near the post-synaptic density [20], whichcould account for the increased spine density in cells expressingPFN1C71G. Changes in the spine density, and therefore synapticconnectivity, could lead to altered excitability of neurons. Whilehippocampal excitability has not been specifically studied in ALS,altered cortical and motor neuron excitability has been reported inmouse models of ALS and human ALS cases [22,25].

In the current study, we show for the first time PFN1C71G

aggregates in cultured hippocampal neurons (Fig. 3). Protein aggre-gations are a histopathological hallmark of neurodegenerativedisease and protein aggregation in motor neurons is found in sev-eral mouse models of ALS [9,33]. Although ALS is characterisedby a loss of upper and lower motor neurons, extra-motor neuronareas are being increasingly implicated in ALS. Hippocampus-associated learning and microcircuitry deficits have been found inpre-symptomatic SOD1 mouse models of ALS [26]; and diffusiontensor MRIs on ALS patients has revealed damage to hippocampalmicrostructure [5]. An association of a reduction of hippocampalgrey matter and memory performance in non-FTD ALS patientsimplicates the hippocampus in cognitive dysfunction in late stagesof disease [27]. Furthermore, as the disease progresses, pathologybecomes widespread involving regions such as the cerebellum andhippocampus [1,6]. For example, phosphorylated TDP-43-positivelesions have been found in the hippocampus of late stages of ALS[8]. The presence of PFN1C71G positive aggregations in cultured hip-pocampal neurons is therefore of particular interest and will needto be further investigated in an in vivo model of ALS.

Taken together, our study has shown new data of an ALS-associated PFN1 mutation in hippocampal neurons. ALS is acomplex neurodegenerative disease and a more complete pictureof the pathomechanisms involved will advance the search for muchneeded therapies.

Acknowledgements

This work was supported by a Grant-in-Aid from The Motor Neu-ron Disease Institute Australia (MNDRIA) (L.M.I and T.F). L.M.I. is anNH&MRC Senior Research Fellow.

Appendix A. Supplementary data

Supplementary data associated with this article can be found,in the online version, at http://dx.doi.org/10.1016/j.neulet.2015.09.034.

2 nce Le

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

J.A. Zitzewitz, Z.S. Xu, L.H. van den Berg, J.D. Glass, G. Siciliano, E.T. Cirulli, D.B.Goldstein, F. Salachas, V. Meininger, W. Rossoll, A. Ratti, C. Gellera, D.A. Bosco,G.J. Bassell, V. Silani, V.E. Drory, R.H. Brown, J.E. Landers, Mutations in the

28 M. Brettle et al. / Neuroscie

eferences

[1] S. Abdulla, J. Machts, J. Kaufmann, K. Patrick, K. Kollewe, R. Dengler, H.J.Heinze, S. Petri, S. Vielhaber, P.J. Nestor, Hippocampal degeneration inpatients with amyotrophic lateral sclerosis, Neurobiol. Aging 35 (2014)2639–2645.

[2] M. Ackermann, A. Matus, Activity-induced targeting of profilin andstabilization of dendritic spine morphology, Nat. Neurosci. 6 (2003)1194–1200.

[3] A. Al-Chalabi, P.M. Andersen, P. Nilsson, B. Chioza, J.L. Andersson, C. Russ, C.E.Shaw, J.F. Powell, P.N. Leigh, Deletions of the heavy neurofilament subunit tailin amyotrophic lateral sclerosis, Hum. Mol. Genet. 8 (1999) 157–164.

[4] J.R. Bamburg, G.S. Bloom, Cytoskeletal Pathologies of Alzheimer Disease, CellMotil. Cytoskelet. 66 (2009) 635–649.

[5] G. Barbagallo, G. Nicoletti, A. Cherubini, M. Trotta, T. Tallarico, C. Chiriaco, R.Nistico, D. Salvino, F. Bono, P. Valentino, A. Quattrone, Diffusion tensor MRIchanges in gray structures of the frontal-subcortical circuits in amyotrophiclateral sclerosis, Neurol. Sci. 35 (2014) 911–918.

[6] P. Bede, M. Elamin, S. Byrne, R.L. McLaughlin, K. Kenna, A. Vajda, A. Fagan, D.G.Bradley, O. Hardiman, Patterns of cerebral and cerebellar white matterdegeneration in ALS, J. Neurol. Neurosurg. Psychiatry 86 (2014) 468–470.

[7] S. Boopathy, T.V. Silvas, M. Tischbein, S. Jansen, S.M. Shandilya, J.A. Zitzewitz,J.E. Landers, B.L. Goode, C.A. Schiffer, D.A. Bosco, Structural basis formutation-induced destabilization of profilin 1 in ALS, Proc. Natl. Acad. Sci. U.S. A. 112 (2015) 7984–7989.

[8] J. Brettschneider, K. Del Tredici, J.B. Toledo, J.L. Robinson, D.J. Irwin, M.Grossman, E. Suh, V.M. Van Deerlin, E.M. Wood, Y. Baek, L. Kwong, E.B. Lee, L.Elman, L. McCluskey, L.B. Fang, S. Feldengut, A.C. Ludolph, V.M.Y. Lee, H.Braak, J.Q. Trojanowski, Stages of pTDP-43 Pathology in Amyotrophic LateralSclerosis, Ann. Neurol. 74 (2013) 20–38.

[9] M. Cai, K.W. Lee, S.M. Choi, E.J. Yang, TDP-43 modification in the hSOD1amyotrophic lateral sclerosis mouse model, Neurol. Res. 37 (2015) 253–262.

10] N. Courtemanche, T.D. Pollard, Interaction of Profilin with the Barbed End ofActin Filaments, Biochemistry 52 (2013) 6456–6466.

11] T. Fath, Y.D. Ke, P. Gunning, J. Gotz, L.M. Ittner, Primary support cultures ofhippocampal and substantia nigra neurons, Nat. Protoc. 4 (2009) 78–85.

12] T. Fath, A.K. Suchowerska, Cytoskeletal changes in diseases of the nervoussystem, Front. Biol. 9 (2014) 5–17.

13] M.D. Figley, G. Bieri, R.M. Kolaitis, J.P. Taylor, A.D. Gitler, Profilin 1 associateswith stress granules and ALS-linked mutations alter stress granule dynamics,J. Neurosci. 34 (2014) 8083–8097.

14] A.A. Filipchuk, J. Durand, Postnatal dendritic development in lumbarmotoneurons in mutant superoxide dismutase 1 mouse model ofamyotrophic lateral sclerosis, Neuroscience 209 (2012) 144–154.

15] G. Gallo, The cytoskeletal and signaling mechanisms of axon collateralbranching, Dev. Neurobiol. 71 (2011) 201–220.

16] P.H. Gordon, Amyotrophic lateral sclerosis: an update for 2013 clinicalfeatures, pathophysiology, management and therapeutic trials, Aging Dis. 4(2013) 295–310.

17] P. Hotulainen, C.C. Hoogenraad, Actin in dendritic spines: connectingdynamics to function, J. Cell Biol. 189 (2010) 619–629.

18] C. Ingre, J.E. Landers, N. Rizik, A.E. Volk, C. Akimoto, A. Birve, A. Hubers, P.J.Keagle, K. Piotrowska, R. Press, P.M. Andersen, A.C. Ludolph, J.H. Weishaupt, A

novel phosphorylation site mutation in profilin 1 revealed in a large screen ofUS, Nordic, and German amyotrophic lateral sclerosis/frontotemporaldementia cohorts, Neurobiol. Aging 34 (2013) e1701–e1706, 1708.

19] L.M. Ittner, D. Koller, R. Muff, J.A. Fischer, W. Born, The N-terminalextracellular domain 23–60 of the calcitonin receptor-like receptor in

tters 609 (2015) 223–228

chimeras with the parathyroid hormone receptor mediates association withreceptor activity-modifying protein 1, Biochemistry 44 (2005) 5749–5754.

20] F. Korobova, T. Svitkina, Arp2/3 complex is important for filopodia formation,growth cone motility, and neuritogenesis in neuronal cells, Mol. Biol. Cell 19(2008) 1561–1574.

21] D.J. Kwiatkowski, G.A. Bruns, Human profilin. Molecular cloning, sequencecomparison, and chromosomal analysis, J. Biol. Chem. 263 (1988) 5910–5915.

22] P. Menon, M.C. Kiernan, S. Vucic, Cortical hyperexcitability precedes lowermotor neuron dysfunction in ALS, Clin. Neurophysiol. 126 (2015) 803–809.

23] K. Michaelsen, K. Murk, M. Zagrebelsky, A. Dreznjak, B.M. Jockusch, M.Rothkegel, M. Korte, Fine-tuning of neuronal architecture requires twoprofilin isoforms, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 15780–15785.

24] M. Neumann, D.M. Sampathu, L.K. Kwong, A.C. Truax, M.C. Micsenyi, T.T.Chou, J. Bruce, T. Schuck, M. Grossman, C.M. Clark, L.F. McCluskey, B.L. Miller,E. Masliah, I.R. Mackenzie, H. Feldman, W. Feiden, H.A. Kretzschmar, J.Q.Trojanowski, V.M. Lee, Ubiquitinated TDP-43 in frontotemporal lobardegeneration and amyotrophic lateral sclerosis, Science 314 (2006) 130–133.

25] M. Pieri, F. Albo, C. Gaetti, A. Spalloni, C.P. Bengtson, P. Longone, S. Cavalcanti,C. Zona, Altered excitability of motor neurons in a transgenic mouse model offamilial amyotrophic lateral sclerosis, Neurosci. Lett. 351 (2003) 153–156.

26] E. Quarta, R. Bravi, I. Scambi, R. Mariotti, D. Minciacchi, Increased anxiety-likebehavior and selective learning impairments are concomitant to loss ofhippocampal interneurons in the presymptomatic SOD1(G93A) ALS mousemodel, J. Comp. Neurol. 523 (2015) 1622–1638.

27] J. Raaphorst, M.J. van Tol, M. de Visser, A.J. van der Kooi, C.B. Majoie, L.H. vanden Berg, B. Schmand, D.J. Veltman, Prose memory impairment inamyotrophic lateral sclerosis patients is related to hippocampus volume, Eur.J. Neurol. 22 (2015) 547–554.

28] A.E. Renton, A. Chio, B.J. Traynor, State of play in amyotrophic lateral sclerosisgenetics, Nat. Neurosci. 17 (2014) 17–23.

29] C. J.D. Rotty, E.M. Wu, C. Haynes, J.D. Suarez, H.E. Winkelman, J.M. Johnson,D.R. Haugh, J.E. Bear Kovar, Profilin-1 serves as a gatekeeper for actin assemblyby Arp2/3-dependent and -independent pathways, Dev. Cell 32 (2015) 54–67.

30] L.P. Rowland, N.A. Shneider, Amyotrophic lateral sclerosis, N. Engl. J. Med. 344(2001) 1688–1700.

31] N.C. Shaner, G.G. Lambert, A. Chammas, Y. Ni, P.J. Cranfill, M.A. Baird, B.R. Sell,J.R. Allen, R.N. Day, M. Israelsson, M.W. Davidson, J. Wang, A brightmonomeric green fluorescent protein derived from Branchiostomalanceolatum, Nat. Methods 10 (2013) 407–409.

32] B.N. Smith, C. Vance, E.L. Scotter, C. Troakes, C.H. Wong, S. Topp, S. Maekawa,A. King, J.C. Mitchell, K. Lund, A. Al-Chalabi, N. Ticozzi, V. Silani, P. Sapp, R.H.Brown Jr, J.E. Landers, S. Al-Sarraj, C.E. Shaw, Novel mutations support a rolefor Profilin 1 in the pathogenesis of ALS, Neurobiol. Aging 36 (36 2015)e1617–e1627, 1602.

33] T.L. Williamson, D.W. Cleveland, Slowing of axonal transport is a very earlyevent in the toxicity of ALS-linked SOD1 mutants to motor neurons, Nat.Neuroci. 2 (1999) 50–56.

34] C.H. Wu, C. Fallini, N. Ticozzi, P.J. Keagle, P.C. Sapp, K. Piotrowska, P. Lowe, M.Koppers, D. McKenna-Yasek, D.M. Baron, J.E. Kost, P. Gonzalez-Perez, A.D. Fox,J. Adams, F. Taroni, C. Tiloca, A.L. Leclerc, S.C. Chafe, D. Mangroo, M.J. Moore,

profilin 1 gene cause familial amyotrophic lateral sclerosis, Nature 488 (2012)499–503.