Abstract

Cataracts are characterized by an opacification of the eye lens, often caused by protein misfolding and aggregation. The intermediate filament protein vimentin, which is highly expressed in lens fiber cells and in mesenchymal tissues, is a main structural determinant in these cells forming a membrane-connected cytoskeleton. Additional functions of vimentin remain to be identified. Here, we demonstrate that a mutation in VIM causes a dominant, pulverulent cataract. We sequenced the complete human VIM gene in 90 individuals suffering from congenital cataract and found a G596A change in exon 1 in a single individual, causing the missense mutation E151K in coil 1B of vimentin. The mutant vimentin formed an aberrant vimentin cytoskeleton and increased the proteasome activity in transfected cells. Furthermore, this mutation causes a severe kinetic defect in vimentin assembly both in vitro and in vivo. Hence, in conjunction with available mouse and cell culture models, our results reveal for the first time an important functional role for vimentin in the maintenance of lens integrity. Finally, this invites novel therapy approaches for cataracts.

INTRODUCTION

Intermediate filament (IF) proteins are encoded by a large gene family of ∼70 members in humans, the mouse and other mammals (1). According to the sequence type, they are classified into six sequence types that show differentiation-dependent and cell type restricted expression patterns. Cytoplasmic IF assemble into distinct cytoskeletal networks, associate with the plasma membrane (vimentin), cell junctions (keratins) and Z lines (desmin), whereas nuclear lamins associate with the lamina and chromatin. More than 30 diseases are caused by mutations in IF genes, including the blistering skin disorder epidermolysis bullosa simplex caused by mutations in KRT5 and KRT14, desminopathies caused by DES mutations, laminopathies arising from laminA mutations and Alexander disease caused by GFAP mutations (2–4) (http://www.interfil.org/). Analysis of these diseases has revealed major roles of IF as scaffolds, in addition to tissue-restricted functions including organelle transport, signaling and translation control (4,5). Vimentin is a type III IF protein that is highly conserved in vertebrate evolution and is expressed in all mesenchymal cells and tissues including the eye lens. It displays a typical tripartite domain structure with a central α-helical rod-domain flanked by non-α-helical amino- and carboxy-terminal domains (6). In the latter, it interacts directly with the small lens chaperone αB-crystallin (7,8). The search for its in vivo function has remained elusive for more than a decade, largely due to the absence of an obvious phenotype in vim−/− mice (9).

The majority of congenital and infantile inherited cataracts have been associated with mutations in genes coding for crystallins, connexins and the beaded filament proteins BFSP1 and BFSP2. The latter two, previously termed filensin (CP115) and phakinin (CP49), are IF proteins unique to lens fiber cells and have been assumed to be crucial for lens integrity (10–12).

Here, we report on the first mutation in VIM that gives rise to a pulverulent cataract in a 45-year-old individual and demonstrate that the underlying mutations cause an assembly defect of vimentin, in analogy to a VIM mutation we recently engineered in mice where it causes a cataract (13).

RESULTS AND DISCUSSION

To establish a link between mutations in vimentin and a specific human disease, we have recently exploited the high degree of sequence conservation among vimentin and other IF proteins in the so-called IF consensus motifs at either end of the α-helical domain (‘rod’) by generating transgenic mice expressing vimentin carrying a dominant mutation in coil 1A. As a consequence, mice developed a severe posterior cataract (13). The analogous mutation in KRT14, changing arginine 125 to cysteine (R125C) in the epidermis-specific KRT14, is the most frequent genetic alteration leading to severe epidermolysis bullosa simplex (14).

Based on our successful reverse genetics approach, we decided to screen a panel of 90 patients suffering from various types of cataracts for mutations in the human vimentin gene (15). One affected individual, a 45-year-old female who belongs to a small autosomal dominant cataract family, suffered from a pulverulant cataract which was dust-like cataract in parts of the lens. As the patient’s mother also has cataract, autosomal dominant inheritance is established. Sequencing of exon 1 revealed a G596A substitution (Fig. 1A). This mutation was absent from the 192 healthy controls in the European Collection of Cell Cultures Human Random Control DNA Panels 1 and 2 (HRC-1 and 2). The mutation is not 1 of the 51 currently recorded single nucleotide polymorphisms in vimentin, 3 of which are in the segment of exon 1 amplified by primer pair 1.2. The G596A substitution causes a glutamic acid to lysine substitution in an ‘e’ position in the first heptad of the α-helical coil 1B rod segment (Fig. 1B). This glutamic acid is evolutionarily highly conserved in vimentin from cartilaginous fish to man (Fig. 1C). In addition, it is also conserved in desmin and neuronal IF proteins, which are both co-assembly partners of vimentin during embryonic development, indicating an important functional role (6,16). An analogous mutation has so far not been reported in any other IF gene. The rod segment of vimentin is crucial for dimer formation and participates in multimerization of dimers into long vimentin filaments (17,18). We hypothesized that the position and non-conservative nature of the mutation compromise formation of ionic interactions involved in intra- and interhelical interactions, affecting higher order structures and the function of vimentin.

Figure 1.

Vimentin structure and mutation. (A) Identification of a VIM mutation in a cataract patient by DNA sequencing. Altered nucleotides in the mutant sequence are shown above the chromatogram, resulting in the amino acid change shown below. (B) Schematic representation of vimentin gene and protein domain organization. Note the position of the mutation E151K (black triangle) in coil 1B of human vimentin. Dashed line denotes the peptide (CGQVINETSQHHDDLE) which was used to raise a highly specific vimentin antiserum. (C) Alignment of partial vimentin protein sequences from various species. 1A, L1 and 1B denote coil 1A, linker and coil 1B domains of vimentin. The position of the rod domains is indicated above the sequence. Letters underneath the sequence denote relative position in α-helix. Note the species conservation of E151 which is mutated in a cataract patient (boxed in red). Numbers indicate the amino acid position in vimentin of the corresponding species.

Figure 1.

Vimentin structure and mutation. (A) Identification of a VIM mutation in a cataract patient by DNA sequencing. Altered nucleotides in the mutant sequence are shown above the chromatogram, resulting in the amino acid change shown below. (B) Schematic representation of vimentin gene and protein domain organization. Note the position of the mutation E151K (black triangle) in coil 1B of human vimentin. Dashed line denotes the peptide (CGQVINETSQHHDDLE) which was used to raise a highly specific vimentin antiserum. (C) Alignment of partial vimentin protein sequences from various species. 1A, L1 and 1B denote coil 1A, linker and coil 1B domains of vimentin. The position of the rod domains is indicated above the sequence. Letters underneath the sequence denote relative position in α-helix. Note the species conservation of E151 which is mutated in a cataract patient (boxed in red). Numbers indicate the amino acid position in vimentin of the corresponding species.

To test the biological and biochemical consequences of the mutation, cDNAs encoding E151K, wild-type and the previously engineered mutant R113C vimentin, known to disrupt filaments and to cause a cataract in transgenic mice (13), were transiently transfected into MCF7 cells. About 19% of E151K vimentin transfectants displayed cytoplasmic vimentin aggregates of various sizes, indicating that the mutation caused misfolding of vimentin (Fig. 2A). The remainder of transfectants contained filaments that ranged from near normal length, spanning the cytoplasm, to short assemblies (Fig. 2A and B). This altered vimentin organization was reminiscent of that seen following expression of the dominant-negative R113C mutant in cultured cells and in transgenic mice, but less severe (13,19) (Fig. 2C). The formation of extensive cytoskeletal arrays from wild-type vimentin in MCF7 cells demonstrated that the cells were fully competent to assemble vimentin correctly. Since, we previously have found that the R113C vimentin mutant caused an upregulation of proteasome activity in vivo which might contribute to the cataract phenotype (13), we examined whether this pathomechanism was conserved in E151K transfectants. In total cell extracts of R113C and E151K transfectants, proteasome activity was increased to a similar extent by ∼60%, compared with mock-transfected MCF-7 cells (Fig. 3). Upon cotransfection of cDNAs encoding E151K and wild-type vimentin, proteasome activity was raised still by ∼50%, suggesting that the E151K mutant, in line with genetic data, behaves dominantly in this assay. Finally, we turned to established in vitro assays to characterize the E151K mutation further, using highly purified recombinant vimentin (19). In vitro assembly of wild-type vimentin gave rise to long and smooth filaments, whereas E151K vimentin produced comparatively much shorter filaments, even after prolonged assembly (Fig. 4A and B). At early time points (10 s), normal unit-length filaments (ULFs) were present, however, the elongation reaction of ULFs proceeded much slower (data not shown). These data were corroborated by viscometric experiments, as E151K vimentin gained viscosity indeed only very slowly over time (Fig. 4C). Hence, wild-type vimentin had reached the plateau after 10 min, whereas for the mutant vimentin, the curve for the relative viscosity η leveled off only after 1 h. In particular, the value for η obtained at 1 min was 0.035 for wild-type vimentin (36% of the value reached at 1 h), whereas it was only 0.004 for E151K vimentin (4% of the 1 h value). Collectively, these in vitro data provide strong support for our hypothesis that the E151K mutation does not impede IF formation as such but that it causes a kinetic defect affecting both early and later phases of filament assembly. Probably, the E→K change interferes with ionic interactions between individual coiled coil chains necessary for the molecular rearrangements taking place both during the lateral association and the elongation phase of assembly (17). Nevertheless, it does not produce catastrophic defects as found for various disease mutations of the muscle-specific IF protein desmin (17).

Figure 2.

Immunofluorescence analysis of transiently transfected MCF7 cells. Expression of mutant E151K vimentin causes an altered organization of the vimentin cytoskeleton. (A) and (B) display a range of phenotypes, including severely disrupted vimentin organized in dense cytosolic aggregates (A), as well as near normal appearing vimentin filaments (B). (C) The previously characterized dominant vimentin R113C mutant caused more severe disruption of the cytoskeleton (13,19). (D) Formation of an extensive cytoskeleton upon transfection of wild-type vimentin. Bar: (A–D) 10 µm.

Figure 2.

Immunofluorescence analysis of transiently transfected MCF7 cells. Expression of mutant E151K vimentin causes an altered organization of the vimentin cytoskeleton. (A) and (B) display a range of phenotypes, including severely disrupted vimentin organized in dense cytosolic aggregates (A), as well as near normal appearing vimentin filaments (B). (C) The previously characterized dominant vimentin R113C mutant caused more severe disruption of the cytoskeleton (13,19). (D) Formation of an extensive cytoskeleton upon transfection of wild-type vimentin. Bar: (A–D) 10 µm.

Figure 3.

The missense mutation E151K causes increased proteasome activity in transiently transfected MCF7 cells. (A) In single transfectants, both E151K and R113C mutants cause a strong increase in proteasome activity. Mock-transfected cells show a modest increase. (B) Display of relative proteasome activity from (A) at 45 min. In comparison to mock-transfected cells, single and double transfectants show an increase of 60% and 50%, respectively.

Figure 3.

The missense mutation E151K causes increased proteasome activity in transiently transfected MCF7 cells. (A) In single transfectants, both E151K and R113C mutants cause a strong increase in proteasome activity. Mock-transfected cells show a modest increase. (B) Display of relative proteasome activity from (A) at 45 min. In comparison to mock-transfected cells, single and double transfectants show an increase of 60% and 50%, respectively.

Figure 4.

Characterization of the assembly kinetics of E151K vimentin. (A and B) Electron microscopic analysis of filaments assembled from mutant (B) and wild-type (A) vimentin for 1 h at 37°C. In agreement with other IF mutations, E151K vimentin has a dominant effect on filament formation. Although E151K vimentin forms filaments of apparently normal appearance, the assembly reaction is slowed down considerably. Arrowheads in (B) highlight unit-length filament precursors not seen in wild-type vimentin assembly reactions. Bar: 100 nm. (C) Viscosity measurement of purified E151K and wild-type vimentin. The viscosity for E151K vimentin protein is displayed for two independent experiments. Both show a very similar viscosity profile for the E151K vimentin emphasizing the extreme slow speed of assembly.

Figure 4.

Characterization of the assembly kinetics of E151K vimentin. (A and B) Electron microscopic analysis of filaments assembled from mutant (B) and wild-type (A) vimentin for 1 h at 37°C. In agreement with other IF mutations, E151K vimentin has a dominant effect on filament formation. Although E151K vimentin forms filaments of apparently normal appearance, the assembly reaction is slowed down considerably. Arrowheads in (B) highlight unit-length filament precursors not seen in wild-type vimentin assembly reactions. Bar: 100 nm. (C) Viscosity measurement of purified E151K and wild-type vimentin. The viscosity for E151K vimentin protein is displayed for two independent experiments. Both show a very similar viscosity profile for the E151K vimentin emphasizing the extreme slow speed of assembly.

The data reported here have established for the first time a link between a VIM mutation and a human cataract. They add vimentin to the growing list of IF gene mutations underlying tissue-restricted disorders and demonstrate its role in the maintenance of lens integrity (2,4). In view of its widespread expression in embryonic and adult mesenchymal cells (9,20), the question arises why VIM mutations cause a tissue-restricted disorder, namely a cataract, in the human and the mouse despite the presence of misfolded vimentin species in most mesenchymal cells (13). We have found that E151K and R113C vimentin mutations are accompanied by an increase in proteasome activity and in Hsp70 (this manuscript and 13). This might lead to a rapid turnover of mutant misfolded vimentin and its associated proteins, followed by protein biosynthesis. In contrast, differentiated lens fiber cells possess only a decreased ability to replace degraded proteins and may suffer gradual depletion of vimentin-interacting proteins (11,20). The notion that VIM mutations cause a cataract and are accompanied by an upregulation of Hsp70 and of proteasome activity suggests a novel link between IF disorders and the protein folding machinery. Should this turn out to be a more general mechanism in IF disorders, it offers a novel targeted approach for the therapy of cataracts involving the chaperone machinery. The availability of cell and animal models and the accessibility of the eye lens provide an attractive model to investigate this further. Finally, in view of the numerous putative functions of vimentin in cells of mesenchymal origin and considering the range of mutations detected in other IF disorders, our study provides the ground to explore an involvement of vimentin in additional disorders.

MATERIALS AND METHODS

Human samples and gene sequencing

A panel of 90 patients affected with various types of cataract were screened for mutations from the published sequence of vimentin (RefSeq ID NM_003380). Exons, including acceptor and donor splice sites, were amplified with primer sequences listed in Table 1. The PCR products of these Vim exons were analyzed with the use of BigDye terminator cycle sequencing kit v3.1 (Applied Biosystems) on an ABI 3100 Genetic Analyser. Chromatograms were reviewed with Sequencer 4.5 (Gene Codes Corporation). Identified mutation was confirmed by sequencing on both DNA strands.

Table 1.

Primers used for PCR amplification of Vim exons

Exon Product size (bp) Forward primer Reverse primer 
1.1 499 GAGGGGACCCTCTTTCCTAAC TCTTGAACTCGGTGTTGATGG 
1.2 357 CAGGACTCGGTGGACTTCTC CGAGGCCCAGCTACTTGCAT 
171 GAGCGAATACGTGGTGTTTG TGCGAGTGGGAAGAGAGTCTA 
157 CCTCCATGTCCTGTCTTTTCTC TCATTCCCCGAAAGTCACTC 
218 GCTGACCGTCTGTCTGTTCTT GTTGAAGCCGCACTGATTTG 
294 ACTGCTCTTTCCCTGGCTTT CCCTCTTTTACTGCAGGGTTG 
357 GGAACAGCTGGGTTTTTCTG TGACACTGCTTAGAGCCCAGT 
169 TCCCAGTGGTTGAAGTTATTTG TGCTTTGACATAAACACAGTGG 
234 TCATTTTTGGCCTGTTTGTTT GCATGAATGAAACCTGAACCT 
465 GTCTTTGGCATGTGGCATTAT AGGACCAAGTAGAGAATTGTT 
Exon Product size (bp) Forward primer Reverse primer 
1.1 499 GAGGGGACCCTCTTTCCTAAC TCTTGAACTCGGTGTTGATGG 
1.2 357 CAGGACTCGGTGGACTTCTC CGAGGCCCAGCTACTTGCAT 
171 GAGCGAATACGTGGTGTTTG TGCGAGTGGGAAGAGAGTCTA 
157 CCTCCATGTCCTGTCTTTTCTC TCATTCCCCGAAAGTCACTC 
218 GCTGACCGTCTGTCTGTTCTT GTTGAAGCCGCACTGATTTG 
294 ACTGCTCTTTCCCTGGCTTT CCCTCTTTTACTGCAGGGTTG 
357 GGAACAGCTGGGTTTTTCTG TGACACTGCTTAGAGCCCAGT 
169 TCCCAGTGGTTGAAGTTATTTG TGCTTTGACATAAACACAGTGG 
234 TCATTTTTGGCCTGTTTGTTT GCATGAATGAAACCTGAACCT 
465 GTCTTTGGCATGTGGCATTAT AGGACCAAGTAGAGAATTGTT 

Proteasome activity assay

For analysis of proteasome activity, 5 million cells were homogenized in 2 ml Earl’s balanced salt solution in a Potter-Elvehjem homogenizer (Braun, Germany) at 1200 rpm at 4°C for 3 min. Supernatant was used to measure the cleavage of the synthetic proteasome substrate III (Calbiochem, USA) by use of a Varioskan multititer plate reader (Thermo Fisher Scientific, Germany) at 37°C over a time course of 90 min. Assay was measured in triplicates and proteasome activity was normalized to 1 µg/µl protein concentration.

Cell transfections and immunofluorescence analysis

The cell lines used are human mammary carcinoma, MCF7. Cells were cultured in DMEM supplemented with 10% FCS (Invitrogen, Germany) and incubated at 37°C and 5% CO2. Human MCF7 cells were transfected with Lipofectamin2000 (Invitrogen) as indicated by the manufacturer. The consequences of protein expression were analyzed 48 h after transfection. Cells were fixed with cold methanol (5 min, −20°C) and acetone (20 s, −20°C). Antibodies were used in the dilutions indicated below. Specimen were examined with a fluorescence microscope (Axiophot 2E; Carl Zeiss, Germany), equipped with a Zeiss Plan-Apochromat 63 × /1.4 oil immersion objective and recorded with an Axiocam HR camera (Carl Zeiss). Confocal images were recorded on an LSM510 microscope (Carl Zeiss; equipped with the same objective). Image analysis and processing were performed using the AxionVision LE 4.6 (Carl Zeiss) and Corel Draw 12.0 software (Corel Corporation, USA).

A new antiserum against vimentin was generated by immunizing rabbits with the peptide (CGQVINETSQHHDDLE). The antiserum was shown to be monospecific for vimentin by western blotting and immunofluorescence microscopy, as it did not react with total protein extracts or frozen section from vim−/− mice (M.M. and T.M.M., unpublished data). In cell transfectants, it reacted with vimentin in the same way as other vimentin antisera and antibodies used before (13,19). Primary antibodies were used for immunofluorescence reactions in the following dilutions: anti-Vimentin: 1/500, anti-Flag 1/2000 (Sigma). Secondary antisera were used in a dilution of 1/800 and were either Alexa-488- or Alexa-594-conjugated (Molecular Probes, The Netherlands) or Cy3- or Cy2-conjugated (Dianova, Germany).

In vitro assembly

In vitro mutagenesis, expression and purification of recombinant proteins, as well as in vitro assembly, viscosity measurements and imaging were performed as described (21).

Conflict of Interest statement. None declared.

FUNDING

We acknowledge partial support by the Deutsche Forschungsgemeinschaft (grant numbers Ma 1316/7 and He 1853/5-1) and by the Bonner Forum Biomedizin to T.M.M. M.M. is supported by a fellowship from DFG (GRK804).

REFERENCES

1
Hesse
M.
Magin
T.M.
Weber
K.
Genes for intermediate filament proteins and the draft sequence of the human genome: novel keratin genes and a surprisingly high number of pseudogenes related to keratin genes 8 and 18
J. Cell Sci.
 , 
2001
, vol. 
114
 (pg. 
2569
-
2575
)
2
Magin
T.M.
Reichelt
J.
Hatzfeld
M.
Emerging functions: diseases and animal models reshape our view of the cytoskeleton
Exp. Cell Res.
 , 
2004
, vol. 
301
 (pg. 
91
-
102
)
3
Omary
M.B.
Coulombe
P.A.
McLean
W.H.
Intermediate filament proteins and their associated diseases
N. Engl. J. Med.
 , 
2004
, vol. 
351
 (pg. 
2087
-
2100
)
4
Kim
S.
Coulombe
P.A.
Intermediate filament scaffolds fulfill mechanical, organizational, and signaling functions in the cytoplasm
Genes Dev.
 , 
2007
, vol. 
21
 (pg. 
1581
-
1597
)
5
Betz
R.C.
Planko
L.
Eigelshoven
S.
Hanneken
S.
Pasternack
S.M.
Bussow
H.
Van Den Bogaert
K.
Wenzel
J.
Braun-Falco
M.
Rutten
A.
, et al.  . 
Loss-of-function mutations in the keratin 5 gene lead to Dowling-Degos disease
Am. J. Hum. Genet.
 , 
2006
, vol. 
78
 (pg. 
510
-
519
)
6
Herrmann
H.
Aebi
U.
Intermediate filaments: molecular structure, assembly mechanism, and integration into functionally distinct intracellular Scaffolds
Annu. Rev. Biochem.
 , 
2004
, vol. 
73
 (pg. 
749
-
789
)
7
Djabali
K.
de Nechaud
B.
Landon
F.
Portier
M.M.
AlphaB-crystallin interacts with intermediate filaments in response to stress
J. Cell Sci.
 , 
1997
, vol. 
110
 (pg. 
2759
-
2769
)
8
Nicholl
I.D.
Quinlan
R.A.
Chaperone activity of alpha-crystallins modulates intermediate filament assembly
EMBO J.
 , 
1994
, vol. 
13
 (pg. 
945
-
953
)
9
Colucci-Guyon
E.
Portier
M.M.
Dunia
I.
Paulin
D.
Pournin
S.
Babinet
C.
Mice lacking vimentin develop and reproduce without an obvious phenotype
Cell
 , 
1994
, vol. 
79
 (pg. 
679
-
694
)
10
Graw
J.
Congenital hereditary cataracts
Int. J. Dev. Biol.
 , 
2004
, vol. 
48
 (pg. 
1031
-
1044
)
11
Hejtmancik
J.F.
Congenital cataracts and their molecular genetics
Semin. Cell Dev. Biol.
 , 
2008
, vol. 
19
 (pg. 
134
-
149
)
12
Perng
M.D.
Quinlan
R.A.
Seeing is believing! The optical properties of the eye lens are dependent upon a functional intermediate filament cytoskeleton
Exp. Cell Res.
 , 
2005
, vol. 
305
 (pg. 
1
-
9
)
13
Bornheim
R.
Muller
M.
Reuter
U.
Herrmann
H.
Bussow
H.
Magin
T.M.
A dominant vimentin mutant upregulates Hsp70 and the activity of the ubiquitin-proteasome system, and causes posterior cataracts in transgenic mice
J. Cell Sci.
 , 
2008
, vol. 
121
 (pg. 
3737
-
3746
)
14
Uitto
J.
Richard
G.
McGrath
J.A.
Diseases of epidermal keratins and their linker proteins
Exp. Cell Res.
 , 
2007
, vol. 
313
 (pg. 
1995
-
2009
)
15
Ferrari
S.
Battini
R.
Kaczmarek
L.
Rittling
S.
Calabretta
B.
de Riel
J.K.
Philiponis
V.
Wei
J.F.
Baserga
R.
Coding sequence and growth regulation of the human vimentin gene
Mol. Cell Biol.
 , 
1986
, vol. 
6
 (pg. 
3614
-
3620
)
16
Schaffeld
M.
Herrmann
H.
Schultess
J.
Markl
J.
Vimentin and desmin of a cartilaginous fish, the shark Scyliorhinus stellaris: sequence, expression patterns and in vitro assembly
Eur. J. Cell Biol.
 , 
2001
, vol. 
80
 (pg. 
692
-
702
)
17
Herrmann
H.
Bar
H.
Kreplak
L.
Strelkov
S.V.
Aebi
U.
Intermediate filaments: from cell architecture to nanomechanics
Nat. Rev. Mol. Cell Biol.
 , 
2007
, vol. 
8
 (pg. 
562
-
573
)
18
Parry
D.A.
Strelkov
S.V.
Burkhard
P.
Aebi
U.
Herrmann
H.
Towards a molecular description of intermediate filament structure and assembly
Exp. Cell Res.
 , 
2007
, vol. 
313
 (pg. 
2204
-
2216
)
19
Schietke
R.
Brohl
D.
Wedig
T.
Mucke
N.
Herrmann
H.
Magin
T.M.
Mutations in vimentin disrupt the cytoskeleton in fibroblasts and delay execution of apoptosis
Eur. J. Cell Biol.
 , 
2006
, vol. 
85
 (pg. 
1
-
10
)
20
Ivaska
J.
Pallari
H.M.
Nevo
J.
Eriksson
J.E.
Novel functions of vimentin in cell adhesion, migration, and signaling
Exp. Cell Res.
 , 
2007
, vol. 
313
 (pg. 
2050
-
2062
)
21
Herrmann
H.
Haner
M.
Brettel
M.
Muller
S.A.
Goldie
K.N.
Fedtke
B.
Lustig
A.
Franke
W.W.
Aebi
U.
Structure and assembly properties of the intermediate filament protein vimentin: the role of its head, rod and tail domains
J. Mol. Biol.
 , 
1996
, vol. 
264
 (pg. 
933
-
953
)