FASEB J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published as doi: 10.1096/fj.06-7022com.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
fj.06-7022comv1
fj.06-7022comv2
21/2/497    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Knoblauch, A.
Right arrow Articles by Wixler, V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Knoblauch, A.
Right arrow Articles by Wixler, V.
(The FASEB Journal. 2007;21:497-510.)
© 2007 FASEB

The binding of Mss4 to {alpha}-integrin subunits regulates matrix metalloproteinase activation and fibronectin remodeling

Alexander Knoblauch*, Carola Will*, Grigori Goncharenko{dagger}, Stephan Ludwig* and Viktor Wixler*,1

* Institute of Molecular Virology, Muenster University Hospital Medical School, Muenster, Germany; and

{dagger} Department of Genetics, University of Gomel, Gomel, Belarus

1 Correspondence: Institute of Molecular Virology, Muenster University Hospital Medical School, Von-Esmarch-Str. 56, 48149 Muenster, Germany. E-mail: vwixler{at}uni-muenster.de


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
In four independent yeast two-hybrid screens with the integrin {alpha}-subunits {alpha}3A, {alpha}6A, {alpha}7A, and {alpha}7B, we identified the Mss4 protein, a nucleotide exchange factor for exocytic Rab GTPases, as a novel integrin interacting protein. We have previously shown that it binds to the conserved KXGFFKR region of integrin {alpha}-subunits located directly beneath the cell membrane. Here we show that the binding site for integrins on Mss4 is overlapping with those for Rab GTPases. Functional analysis of the Mss4/integrin interaction revealed its importance for activation of matrix metalloproteinases (MMPs) and remodeling of secreted extracellular matrix (ECM) proteins. The exocytosis of all the proteins analyzed, however, was unaffected. Furthermore, our data suggest that Mss4 drives the coordinated action of the MT1-MMP/integrin protein complex, thus regulating the presence and activation of MT1-MMP at newly formed filopodia and lamellipodia. This in turn facilitates the conversion of pro-MMPs to MMPs, resulting in cleavage and remodeling of ECM proteins. C2C12 myoblasts with stably down-regulated Mss4 showed a disturbed fibronectin remodeling during differentiation, resulting in malfunctioned myotube formation.—Knoblauch A., Will C., Goncharenko G., Ludwig S., Wixler V. The binding of Mss4 to {alpha}-integrin subunits regulates matrix metalloproteinase activation and fibronectin remodeling


Key Words: integrin cytodomain • integrin-binding proteins • MT1-MMP • myoblast differentiation


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
INTEGRINS ARE {alpha}/ß HETERODIMERIC RECEPTORS, whose extracellular parts determine the specificity of ligand binding and provide outside-in signaling. Their intracellular carboxy-terminal sequences transfer signals into the interior of the cell and modulate the affinity of receptors for extracellular ligands in response to environmental and developmental factors (inside-out signaling) [reviewed in (1) ]. All integrin {alpha}-subunits contain in their cytosolic part the highly conserved KXGFFKR motif, which lies directly beneath the cell membrane. Structural studies have shown that this amino acid stretch, together with the transmembrane region and at least three amino acids following this motif, forms an {alpha}-helical structure that is responsible for interaction with the ß-subunit (2 3 4) . The association between {alpha}- and ß-subunits regulates the surface presentation of heterodimers as well as the activation state of integrin receptors (5 6 7 8 9) . When the receptor is inactive, both subunits tightly interact with each other. On ligand binding they separate, providing binding sites for cytosolic proteins, thus facilitating transduction of intracellular signals (10 11 12) . The closed conformation of the receptor can also be changed to the open conformation via binding of cytosolic proteins, resulting in change in the affinity state of integrins (13 , 14) . Binding of integrins to their ligands, proteins of the ECM, regulates many fundamental cellular functions, such as adhesion, spreading, migration, proliferation, or differentiation of cells but also the assembly and remodeling of the ECM itself, which are important for adhesion, migration or invasion of cells.

Mss4 (mammalian suppressor of Sec4) is a highly conserved protein and is expressed in all mammalian tissues. Originally it was described as the mammalian counterpart of the yeast Dss4 protein, a guanine nucleotide exchange factor (GEF) for Rab GTPases (15) . More than 50 members of Rab GTPases are known, which regulate particular steps in exocytic and endocytic trafficking pathways. Mss4 stimulates GDP release mostly for exocytic Rab GTPases (16) . While functioning as a GEF, promoting exchange of GDP for GTP (16 , 17) , Mss4 has no significant sequence and structural homology with other members of the GEF family of Ras-like G proteins and represents a new class of GEF proteins, possibly having other alternative roles. In contrast to the scant functional data about Mss4, its structure has been solved by both NMR (18) and crystallographic methods (19 , 20) . Structurally, Mss4 shows a striking similarity with the translationally controlled tumor-associated proteins (21) . The fold of Mss4 is a central ß-sheet, flanked by a ß-hairpin and a small variable sheet. The four conserved regions, CR1–CR4, build a primary hydrophobic core with a Zn2+-binding site and encode a structural subdomain responsible for interaction with Rab GTPases (18 19 20) .

Recent studies have demonstrated that the ECM is undergoing continual remodeling due to different cell stimuli. This is not only especially significant during development and tissue regeneration, but also during inflammation or cancer metastasis. While the assembly of secreted ECM proteins takes place on the cell surface and is dependent on integrin activation, the matrix metalloproteinase (MMP) enzyme family is additionally responsible for cleavage and degradation of the ECM molecules (22 23 24 25 26 27 28 29) . Two groups of MMPs are known, soluble, and membrane anchored. Most of the soluble matrix metalloproteinases are secreted as nonactive prozymogens and are activated by other proteases through cleavage of their N-terminal prodomains. The membrane-type MMPs (MT-MMP) act only locally in the pericellular space. The membrane-type MT1-MMP digests a broad range of ECM molecules, including collagens and fibronectin but also pro-MMP-2 and pro-MMP-13. The MT1-MMP is already expressed as an active enzyme, but its functional activity is regulated (30) . It has been demonstrated that integrins interact directly via their extracellular parts with MT1-MMP functioning in a coordinated manner (31 , 32) to alter the immediate microenvironment of cells.

Here we report that Mss4 binds to the cytosolic membrane proximal conserved region of {alpha}-integrin chains and that this interaction is essential for MT1-MMP-mediated activation of MMP-2 and –9 as well as for ECM organization.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
DNA constructs
For yeast two-hybrid screens, cDNA fragments coding for the cytoplasmic parts of murine integrin {alpha}7A (aa 1104–1161) or {alpha}7B (aa 1104–1171) were inserted into the DNA-binding domain vector pAS2–1 and used as baits. A human placenta cDNA library cloned into pACT2 vector (MATCHMAKERTM, Clontech, Palo Alto, CA, USA) was screened as described in (33 , 34) . All integrin constructs used in this work have been described elsewhere (35) . cDNA of complete human Mss4 (aa 1–123) in pACT2 vector was obtained from yeast two-hybrid screens. For expression in mammalian cells, the Mss4 insert was released by SalI/XhoI and inserted into the pCS2+MT vector. The mutants: Mss4 del1 (aa 11–123); Mss4 EF loop (aa 49 – 55 were deleted) and Mss4 F75A (F at position 75 was exchanged for A) were derived by extrasize-overlapping polymerase chain reaction (PCR) and cloned into EcoRI/XhoI sites of the pACT2 vector. For expression in mammalian cells, Mss4 wt [wild-type (WT)] or mutants were released from pACT2 and recloned into pCDNA3 or pCS2+MT vectors. As a result, the Mss4 F75A mutant contained only five repetitive myc-tags at N terminus, while the others had six. For generation of retrovirus stocks, oligos coding for siRNA of Mss4 were inserted into the pSUPER vector and the cassette including the insert plus regulatory elements was cloned into the pBabe-neo vector. Virus-producer cells GPE + 86 were transfected with this construct, selected for G418 resistance and supernatants from confluent monolayers were used as retroviral stocks.

The modified minimal protein transduction domain (PTD) of the HIV TAT protein, YGRKKRRQRRR, (36) was used to obtain chimeric PTD-{alpha}-integrin peptides. Two PTD chimeras were derived. The first, PTD-F, included the full-length wt {alpha}3A cytosolic domain (aa 1015–1051) with an aromatic amino acid before the KXGFFKR conserved motif and bound to Mss4. In the second, PTD-D, which did not bind the Mss4 (34) , this aromatic amino acid was exchanged for an aspartic acid. For purification of PTD-chimeras from E. coli and for their detection by immunofluorescence or immunoblotting a glutathione S-transferase (GST)- and a myc-tag were inserted before and behind of the PTD sequence, respectively (Fig. 5C) . All constructs were verified by sequencing before use.


Figure 1
View larger version (24K):
[in this window]
[in a new window]

  		Figure 1. Mapping of Mss4 binding site on integrins. A) The cytosolic part of {alpha}7Aß1A integrin is shown. Bold letters show the conserved amino acids, and the gray and the dark lines mark the {alpha}-helices and the salt bridge, respectively; the shaded box, the binding site for Mss4. Amino acids essential for binding of Mss4 are marked with an asterisk. The binding site for Mss4 is based on results of Tables S1–S3 and on data published in (34) . B) HEK293 cells were transiently transfected with cDNA constructs as indicated on the left side of the upper box. After 40 h, GST-{alpha}7A-D or GST-{alpha}7A-F proteins were precipitated (P) from cell lysates and the coprecipitated Mss4 protein was detected by immunoblotting (IB) with anti-myc mAb (upper blot). The same blot was stripped and subsequently redeveloped with a rabbit anti-glutathione S-transferase Ab (middle blot). The lower panel shows the expression of myc-Mss4 proteins analyzed by immunoblotting of equal amounts of total cell lysates.


Figure 2
View larger version (27K):
[in this window]
[in a new window]

  		Figure 2. Mapping of integrin binding sites on Mss4. A) Schematic presentation of Mss4 mutants used in this work. Numbers over the diagram indicate Mss4 aa encoded by each construct. CR1–CR4 represent conserved regions. B) Yeast Y190 cells were transformed with GAL4 DNA-binding domain (BD) and GAL-4 activation domain (AD) chimeric constructs as indicated and a ß-Gal filter assay was performed as described in Materials and Methods. C) HEK293 cells were transiently transfected with cDNA constructs as indicated on the left side of the upper box. After 40 h, GST-{alpha}7A precipitates were analyzed as described in legends to Fig. 1 . The Mss4 F75A mutated protein is of smaller size than the other two tagged proteins, because it contains, due to different cloning strategy, only five instead of six repeated myc-tags.


Figure 3
View larger version (30K):
[in this window]
[in a new window]

  		Figure 3. Down-regulation of Mss4 delays spreading of cells and their motility on uncoated surfaces. A) Mss4 expressing (E1) or low-expressing (C5) C2C12 myoblasts were plated on FN-coated (20 µg/ml–lower panel) or uncoated (upper panel) glass coverslips and fixed at times indicated. The number of spread cells was scored after staining with crystal violet. Arrowheads show spread cells. B) For quantification of spread cells, at least 300 cells per coverslip and time point were scored. The experiment was repeated three times and median percentages ± SD are shown. C) The migration into a non coated cell-free cleft was monitored by an inverted microscope at different times and distances covered by cells were calculated. On FN or laminin-1 coated plates, no difference in cell migration was seen (not shown).


Figure 4
View larger version (47K):
[in this window]
[in a new window]

  		Figure 4. Mss4 is colocalized with integrins at newly formed adhesion structures. C2C12 cells were transfected with myc-tagged Mss4 and plated 24 h later on FN-coated coverslips. The cells were stained for Mss4 with anti-myc mAb (red) and for integrins with rabbit anti-{alpha}5 antibodies (green). A) 15 min after plating, the majority of cells was already adhered and started to spread and form focal contacts. In the cells shown here, clearly visible integrin clusters have already been formed (the green ring). Mss4 is abundantly present in newly forming protrusions (arrowheads) but not in already formed focal contacts (arrows). B) Upper panel: a cell is shown just beginning to spread. Note that Mss4 is abundantly present at the border of the spreading cell where it is colocalized with still poorly organized integrins (arrowheads). Inwards to the cell body, integrins began to form large protein clusters and no Mss4 is present (arrows). When cells have already spread (lower panel), Mss4 is found at the periphery only in places where cells form new filopodia or lamellipodia (arrowheads) but not in mature focal contacts (arrows). Bars, 20 µm.


Figure 5
View larger version (35K):
[in this window]
[in a new window]

  		Figure 5. Down-regulation of Mss4 as well as inhibition of Mss4/{alpha}-integrin interaction impairs FN organization. A) Adhered cells were incubated for 8 h in DMEM without FCS. Conditioned media as well as radio-immuno-precipitation assay (RIPA) lysates of the same cells were then analyzed for FN by WB. B) E1 or C5 cells were incubated on uncoated glass coverslips for 8 h or 24 h in DMEM + 10% FCS, fixed and stained without being permeabelized for FN organization (red). Nuclei of cells were counterstained with 4',6'-diam idino-2-phenylidole (DAPI). C) Schematic presentation of chimeric PTD-integrin constructs. The GST-tag served for fast and simple purification from E. coli lysates. The myc-tag can be used for recognition of chimeric peptides, when the GST-tag is cleaved off. In our experiments it was not removed, as preliminary experiments showed that it did not interfere with the binding of fused {alpha}-integrin peptide to Mss4. D) Mss4 expressing cells (E1) were incubated for 8 h with 400 µM of either PTD-D or PTD-F and stained for FN organization as in (B). Bars, 50 µm.

Direct yeast two-hybrid interaction assays
The yeast strain Y190 was cotransformed with pAS2–1 and pACT2 plasmids containing GAL4-BD and GAL4-AD fused with appropriate cDNAs as bait and prey, respectively (34) . The grown-up transformants were tested for the lacZ reporter gene activity in a ß-Gal filter assay. The interaction was scored as negative (–) when no blue colonies were visible after 8 h, and scored as: weak (+), intermediate (++), or strong (+++) when blue colonies became visible after 8 h, 4 h, or 1 h, respectively.

Cell culture
C2C12 myoblasts were cultured at low density in a 1:1 mixture of F10(Ham):Dulbecco’s modified Eagle medium [low glucose (Glc)] and 20% FCS. For differentiation, 8 x 104 C2C12 cells were plated into 6 cm Petri-dishes and cultivated until subconfluency. The growth medium was then replaced by Dulbecco’s modified Eagle medium (DMEM) (high Glc) plus 2% horse serum and was exchanged every 24 h. HEK293 cells were cultured in DMEM with 10% FCS. Cell migrations were performed as described in (35) . Briefly, 105 cells in 0.5 ml DMEM + 10% FCS were plated into 48-well plates either noncoated or precoated with laminin-1 or fibronectin (FN). To produce a cell free "window", a 1.0-mm-thick steel plate was inserted into wells before seeding of cells. These plates were removed again after the cells have attached to the well. This method has the advantage over the frequently used "scratch window" assay in that the substrate in the "window" is maintained. The migration was monitored by inverted microscopy at the times indicated.

GST coprecipitation assays and immunoblotting
For coprecipitation assays, HEK293 cells were transfected with appropriate cDNA plasmids as described in (35) . After 40 h, cells were washed with PBS and lysed in a buffer containing 137 mM NaCl; 25 mM, Tris pH 7.5; 2 mM MgCl2; 10% glycerol; and 1% Brij98 and supplemented with protease inhibitors at room temperature for 10 min. Lysates were cleared by centrifugation, rotated for 2 h at 4°C with glutathione-sepharose beads for GST precipitation analyses. To reduce any non-specific binding of proteins, sepharose beads were blocked with 1% BSA in lysis buffer before incubation with cell lysates. Samples were resolved by SDS-PAGE. The specific detection of proteins was performed by immunoblotting using following antibodies: mouse anti-myc, clone 9E10, rabbit anti-glutathione S-transferase, mouse anti-ß-actin, and anti-MyHC monoclonal antibody (mAb) (NOQ7.5.4D) (Sigma, Taufkirchen, Germany), rabbit polyclonal anti-Mss4 (a gift of P. deCamili, New Haven, CT, USA).

Purification of GST-fusion peptides
Protein purification was performed as described previously (37) . The GST-PTD-{alpha}3A fusion peptides were expressed in E. coli BL21pLysS. Cell pellets were broken by four freezing-thawing cycles in NENT buffer (100 mM NaCl; 20 mM Tris-HCl, pH 8.8; 1 mM EDTA; and 0.5% Nonidet P-40). Extracts were then incubated with glutathione-coated sepharose beads for 4 h at 4°C, washed twice with NENT buffer, and peptides were eluted by incubation of beads for 30 min in 0.5 M Tris/HCl, pH 8.9, and 0.5 M NaCl, containing 70 mM GSH. Eluted proteins were analyzed by SDS-PAGE followed by Coomassie staining and immunoblotting.

Immunofluorescence
Indirect immunofluorescence staining was done as described previously (35) . Briefly, cells grown on glass coverslips were fixed with 2% PFA in PBS at RT for 10 min, permeabilized, where indicated, by incubation with 0.2% Triton X-100 in PBS for 2 min, blocked with 1% BSA and incubated with the relevant antibodies diluted in PBS for 1 h. These were: anti-myc mAb, clone 9E10; rabbit polyclonal anti-{alpha}5-integrin subunit and mAb anti-MT1-MMP (Chemicon, Temecula, CA, USA); rabbit polyclonal anti-MT1-MMP; rabbit polyclonal anti-fibronectin (Sigma, St. Louis, MO, USA); rat monoclonal anti-laminin {gamma}1 chain (a gift of L. Sorokin, University of Muenster, Germany). Secondary antibodies labeled with Cy3 or Alexa 488 were from Biosource (Camarillo, CA, USA). Cell images were taken using an Axiovert 2000 ApoTome microscope with an AxioCam digital camera and AxioVision software (Zeiss, Jena, Germany).

Gelatin zymography
C2C12 cells (7.5x105) were plated for 1 h on fibronectin-coated 6-well dishes and after washing of the attached cells, the medium was replaced to DMEM without serum. After 4 and 8 h, conditioned media were collected and centrifuged and 100 µl of supernatants were separated under nonreducing conditions on 10% SDS-PAGE containing 1 mg/ml gelatin. Gels were washed always twice for 15 min in water, in water containing 2.5% Triton X-100, and again in water. Finally, gels were incubated in substrate buffer (50 mM Tris at pH 7.5, 5 mM CaCl2) for 16–24 h at 37°C, stained with Coomasie blue and destained until clearly visible gelatin-free bands representing active MMP enzymes appeared.

Flow cytometry
Cells (1x105) were detached either by trypsinization or by incubation with 25 mM EDTA; suspended in FACS-PBS (PBS containing 2% FCS and 0.02% NaN3) and incubated with anti-MT1-MMP mAb, anti-{alpha}7 integrin mAb, clone 3C12 (a gift of H. von der Mark, Erlangen, Germany), anti-ß1 integrin rat antibody (Ab), clone 9EG7, for 20 min on ice; washed twice with FACS-PBS; and incubated with FITC-, Cy3- or Cy5-conjugated secondary antibodies for additional 15 min. After washing of cells, measurements were performed with a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA, USA).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Analysis of Mss4/integrin interaction
In searching for cytoplasmic proteins binding to integrin {alpha}-subunits, we identified in four independent yeast-two hybrid screens [using the different {alpha}-integrin molecules {alpha}3A and {alpha}6A (33 , 34) and {alpha}7A and {alpha}7B (this work)], the Mss4 protein as a novel integrin interaction partner. A human placenta cDNA library was screened. Of 288 and 31 clones isolated as potentially {alpha}7A and {alpha}7B positive, 23 and 8 clones, respectively, coded for Mss4. Analysis of Mss4/integrin interaction revealed that Mss4 binds to {alpha}- but not ß- chains. The binding, however, was restricted to {alpha}-subunits of integrins that do not require activation prior to binding to ligands [Supplemental Table 1 and (34) ]. So, Mss4 did not interact with {alpha}L, {alpha}v, or {alpha}IIb subunits of integrin receptors that are naturally in a low-affinity state. Being in the "inactive" or "closed" conformation, the C-terminal tails of these integrin {alpha}-subunits interact with their membrane proximal parts, blocking the binding of cytosolic proteins [for details see the Discussion section and (1 , 4 , 14) ]. Deletion and point mutation analysis of {alpha}3A, {alpha}7A, {alpha}7B, and {alpha}IIb chains further showed that Mss4 binds to the {alpha}-helical structure composed of the membrane proximal conserved KXGFFKR region plus an additional three residues (Supplemental Tables 2, 3). Interestingly, the last amino acid of the transmembrane region, invariably an aromatic residue located directly in front of the first lysine of the conserved cytosolic motif, was also essential for the binding of {alpha}-integrin chains to Mss4. Direct yeast two-hybrid interaction tests showed that an exchange of this aromatic amino acid for an aspartate abrogates the binding of integrins to Mss4 [Fig. 1 A, Supplemental Tables 2, 3, (34) ].

To verify the binding specificity of Mss4 to integrins in mammalian cells, we coexpressed a myc-tagged Mss4 protein along with the cytosolic part of {alpha}7A integrin chain containing either a phenylalanine ({alpha}7A-F) or an aspartate ({alpha}7A-D) before the conserved KXGFFKR motif. These experiments confirmed the Mss4/integrin association seen in yeast cells (Fig. 1B ) and showed that the aromatic amino acid of the transmembrane region is indeed essential for interaction with Mss4, as only the {alpha}7A-F integrin molecule but not the {alpha}7A-D could precipitate the Mss4 protein (Fig. 1B , upper panel, lanes 5, 6).

To understand the molecular basis of the Mss4/integrin interaction, we wished to identify the binding domain on Mss4 for {alpha}-integrin chains. For this purpose different Mss4 mutants were created (Fig. 2 A). Two of these, Mss4 del 1 and Mss4 EF loop, had truncations of highly mobile and protruding extensions of nonconserved regions of the Mss4 molecule, while the other was a point mutation having an F75A exchange in the conserved Rab-binding core domain. Deletion of the first 8 N-terminal amino acids or the residues 49–54 of the EF loop did not perturb the nucleotide exchange activity of Mss4. The phenylalanine 75, however, is of high importance for the Mss4/Rab interaction (18 , 20) . Using these mutants we hoped to distinguish between binding sites for integrins and Rab GTPases on Mss4. Initially the Mss4 mutants were assayed for an interaction with {alpha}-integrin chains in direct yeast two-hybrid tests. As the results in Fig. 2B show, deletion of the ßE-ßF loop (Mss4 EF loop) had no effect on the binding of Mss4 to integrins, while deletion of the first 10 N-terminal aa (Mss4 del1) affected it. Surprisingly, mutation of F75A also abolished the binding of Mss4 to integrins. Studying these interactions in mammalian cells also confirmed that {alpha}7A precipitated only Mss4 wt and Mss4 EF loop but not the Mss4 F75A mutant (Fig. 2C ).

Taken together, these results showed that Mss4 associates with the highly conserved {alpha}-helical structure of {alpha}-integrin chains but only when this region is exposed. The binding site for integrins on Mss4 includes both the conserved Rab-binding core and the nonconserved N-terminal region of the molecule, indicating that the interaction of Mss4 with Rab GTPases and integrins is rather of exclusive than of simultaneous nature.

Analysis of functional consequences of Mss4/integrin interaction
As the {alpha}-helical region of integrin {alpha}-chains, to which Mss4 binds, is also responsible for association with the ß integrin subunit and required for the surface expression of heterodimeric receptors and for their affinity activation state, we analyzed whether binding of Mss4 to integrins might regulate these functions. However, neither the transport to the cell surface, the affinity state of integrins, nor attachment or migration of cells were changed, when Mss4 and its mutants were over-expressed or down-regulated (Supplemental Fig. 1 and data not shown). The inability to ascertain any role for Mss4/integrin interaction in these experiments was independent of cell type or integrins or ECM proteins used.

Another integrin function is their modulation of the assembly and remodeling of ECM networks (29) . To test whether Mss4/integrin interaction might influence the secretion or assembly of ECM proteins, we compared the spreading of four different C2C12 cell clones on uncoated glass surfaces. The first two clones expressed high amounts of endogenous Mss4, while in the other two the Mss4 expression was reduced by siRNA (Supplemental Fig. 1A). When attaching to an untreated surface cells have to organize their own secreted matrix proteins properly before being able to spread on them. Indeed, cells with low amounts of Mss4 expression spread on the noncoated glass surface less efficiently than those with high Mss4 expression. The difference was transient but significant (Fig. 3 A, B). Sixty minutes after cells have been plated, twice as many high Mss4 expressing cells had spread as cells from the low expression clones. This difference was not evident after 120 min, and no difference was seen at any time point when cells were allowed to spread on surfaces precoated either with laminin-1 or fibronectin (Fig. 3A , lower panel). We then examined whether Mss4 could influence cell motility on noncoated surfaces. The migration of cells with down-regulated Mss4 expression was reduced when compared to Mss4 expressing cells under this condition (Fig. 3C ), contrary to their migration on precoated surfaces (Supplemental Fig. 1C). These results suggested that Mss4 might be involved in integrin-mediated organization or cleavage of newly synthesized ECM proteins at forming adhesion edges. Immunofluorescence analysis of Mss4 and {alpha}-integrin distribution in C2C12 cells supported this (Fig. 4 ). These assays showed that Mss4 and {alpha}5-integrins colocalize only at the outer border of spreading cells or at very distant tips of filopodia and lamellipodia, sites where a cell forms new contacts with the support. Mss4 was never observed incorporated into already organized focal adhesion structures; indeed Mss4 was lost from these regions as soon as integrins formed clusters that were visible under the light microscope (Fig. 4) .

Mss4 is involved in integrin-dependent recruitment of fibronectin to the cell surface
We then questioned whether down-regulation of Mss4 altered recruitment of ECM proteins. Proliferating C2C12 myoblasts produce only low amounts of laminin but high amounts of fibronectin (FN). Comparison of FN in cell lysates or conditioned media of Mss4-expressing and low-expressing cells revealed equal amounts of these molecules, suggesting that expression or secretion of FN were not changed by down-regulation of Mss4 (Fig. 5 A). However, the recruitment of FN to the surface of Mss4 low-expressing cells was radically altered. Immunofluorescence staining showed that on WT cells FN was evenly distributed over the whole cell body and assembled into a well organized network. This appeared as a large and bright meshwork evident at all time points tested (4, 8, 12, 24, or 72 h after plating). The network only grew in size and complexity over the period of cultivation. Cells with suppressed Mss4 also secreted FN shortly after they had been plated, but in contrast to Mss4 expressing cells, the FN was mostly at the cell boundary, bundled into long and thick strings. This abnormal organization of the FN network was also seen 24 h later (Fig. 5B ). To ensure that the binding of Mss4 to integrins is indeed needed for proper FN assembly, we blocked this interaction by introducing peptides containing {alpha}-integrin binding sites for Mss4 into Mss4 expressing cells. This was performed with protein transduction domain fusion peptides (PTD), which cross cell membranes with high efficiency in a receptor-, transporter-, and endocytic-independent manner (38) . Two different types of PTD-chimeric peptides differing in their {alpha}-integrin parts were transduced. The first, PTD-F, which binds to Mss4, included the entire cytosolic part of integrin {alpha}3A with an aromatic amino acid before the conserved {alpha}-helical region. This phenylalanine represents the last transmembrane amino acid of the integrin {alpha}-chain. In the second chimera, PTD-D, which does not bind to Mss4, this aromatic residue was exchanged for an aspartate (see Figs. 1A and 5C for details). Cells were fixed 8 h after transduction with PTD-chimeras and the FN organization was analyzed by immunofluorescence staining (Fig. 5D ). Transduction of PTD-D peptide that does not bind to Mss4 showed no effect on FN assembly. The Mss4-binding PTD-F peptide, however, mimicked the down-regulation of Mss4, as FN on these cells was not organized into bright networks, but occurred as in Mss4 low-expressing cells mostly at the cell periphery as thick rod-like structures.

Mss4 is involved in integrin-dependent activation of MMPs
Remodeling of ECM is closely related to activation of secreted pro-MMPs. To study whether Mss4 might influence this particular function, we compared MMP activity in supernatants of Mss4 expressing and low-expressing cells. Results from gelatin zymography assays showed that activity of both MMP-2 and MMP-9 enzymes is reduced in C2C12 cells with down-regulated Mss4 protein (Fig. 6 A). The observed low MMP activity, however, was not due to an expression or a secretion defect of these cells, as immunoblot analysis of conditioned media or cell lysates showed that both cell types expressed and secreted roughly equal amounts of MMP proteins (Fig. 6B ). Furthermore, quantifying the amount of pro-MMP-9 in conditioned media of these cells by ELISA showed that C5 (low Mss4 expressing) cells contained twice as much of nonactive enzyme molecules than E1 (high Mss4 expressing) cells (Fig. 6C ). To study further whether the conversion of pro-MMPs to MMPs is dependent on Mss4/integrin association, we blocked the interaction by introducing PTD-{alpha}3A-integrin peptides. Introduction of the PTD-D chimera, which does not bind to Mss4, had no effect on MMP cleavage. In fact, the MMP pattern of cells treated with PTD-D shown in gelatin zymography gels of Mss4 expressing cells is similar to that of nontreated cells (Fig. 6A, D , upper panel). However, the Mss4-interacting PTD-F peptide affected it, decreasing the intensity of bands representing the activated forms of both MMP-2 and –9 in gelatin zymography gels of Mss4 expressing cells. Further, the reduced MMP cleavage in supernatants of Mss4 low-expressing cells was additionally decreased by introduction of the PTD-F peptide (Fig. 6D , upper panel). The dependence of MMP cleavage on the presence of Mss4 was especially obvious when we analyzed conditioned media of these cells by immunoblotting (Fig. 6D , lower panel). In the presence of the nonbinding peptide PTD-D, supernatants of the two Mss4 expressing clones (D1 and E1) contained two MMP-2 bands, with the lower, the activated form, being more abundant than the larger, the uncleaved proform. However, the presence of the Mss4-binding PTD-F peptide led to a clear shift of the lower toward the upper band (Fig. 6D , lanes 5, 6), representing the inhibition of MMP cleavage. When Mss4 low-expressing cells (E2 and C5) were incubated with PTD-D chimera, they showed as expected a less-abundant band corresponding to the activated form of MMP-2 in comparison to Mss4 expressing cells. This already weak band was further shifted to the upper band during incubation of cells with the Mss4-binding peptide PTD-F, confirming the necessity of Mss4/integrin interaction for successful cleavage of pro-MMP-2. Taken together, these experiments clearly show that the Mss4-mediated organization of FN and activation of secreted MMPs depends on its interaction with integrins.


Figure 6
View larger version (31K):
[in this window]
[in a new window]

  		Figure 6. Mss4 mediates activation of MMP-2 and MMP-9 in an integrin-dependent manner. A) Already adherent cells were incubated in serum-free medium for 8 h, and conditioned media were analyzed by gelatin zymography. B) Supernatants of cells cultivated for 8 h in medium without FCS as well as their RIPA lysates were separated by SDS-PAGE and analyzed for expression of MMP-2 by immunoblotting. C) Quantification of secreted pro-MMP-9 protein by ELISA. Supernatants of cells cultivated for 8 h were studied. D) Cells were transduced with 400 µM of PTD-F (which binds to Mss4) or PTD-D (which does not bind to Mss4) chimeric peptides for 8 h and supernatants of these cells were analyzed for MMP-2 and MMP-9 enzyme activity by gelatinase zymography (upper panel) or by immunoblotting (lower panel).

Blocking of MT1-MMP impairs FN assembly
As MMP-9 is a substrate of MMP-2, which in turn is activated by MT1-MMP, we questioned next whether inhibition of MT1-MMP will have a similar effect as inactivation of Mss4. For that, Mss4 expressing cells were incubated in the presence of functional blocking antibodies against MT1-MMP and the assembly of FN was analyzed. While several unrelated antibodies (antimyc, anti-{alpha}7-integrin, anti-glutathione S-transferase) had no effect on FN organization, anti MT1-MMP antibodies affected the FN assembly resulting in changes similar to those caused by the down-regulation of Mss4 (Fig. 7 A). To ensure that the effect in Mss4 low-expressing cells was not due to reduced expression of MT1-MMP, we carried out FACSscan analysis of endogenous MT1-MMP. As data on Fig. 7B show, both Mss4 expressing and low-expressing cells contained equal amounts of MT1-MMP on their surface, suggesting that the alteration in the FN matrix was not due to changed expression of MT1-MMP enzyme. Immunofluorescence staining of endogenous MT1-MMP in Mss4 expressing and low-expressing cells showed that it was localized at the outer cell border during spreading of cells, though to a slightly less extent in the latter (data not shown). It has been demonstrated that integrins interact directly via their extracellular parts with MT1-MMP, providing a spatially focused activity of MT1-MMP (31 , 32) . Here we showed that the conversion of MMP-2 and –9 from pro- to active forms depends on the binding of Mss4 to integrins. We then analyzed whether Mss4 is also colocalized with MT1-MMP in spreading cells. Immunofluorescent staining of C2C12 cells for Mss4 and MT1-MMP distribution, indeed, revealed that both proteins colocalize at the outer borderline of adhering cells as well as at the tips of newly formed filopodia or lamellipodia (Fig. 7C ), sites where the colocalization of Mss4 with integrins was also detected.


Figure 7
View larger version (31K):
[in this window]
[in a new window]

  		Figure 7. Inhibition of MT1-MMP by functional-blocking antibodies impairs FN organization. A) E1 Mss4 expressing cells were incubated in DMEM + 10% FCS for 8 h in the presence of indicated antibodies and stained for FN (red). Nuclei of cells were counterstained with DAPI. The 9E10 anti-myc mAb or polyclonal rabbit anti-glutathione S-transferase antibodies served as specificity controls. Bar, 50 µm. B) FACSscan analysis of MT1-MMP cell surface expression. Incubation of cells with FITC-labeled secondary antibodies only (filled graphs) served as negative controls. C) E1 cells were cotransfected with MT1-MMP and myc-tagged Mss4. 24 h later, they were plated on FN-coated coverslips for 15 or 60 min and stained for Mss4 with anti-myc mAb (green) and for MT1-MMP with rabbit anti-MT1-MMP antibodies (red). Note, Mss4 is colocalized with MT1-MMP only at the border of the spreading cell (upper panel) or at newly formed lamellipodia (arrowheads), lower panel. Bars, 20 µm.

Down-regulation of Mss4 impairs differentiation of myoblasts into myotubes
C2C12 myoblasts undergo myogenesis in vitro on deprivation of growth factors, whereupon they fuse and develop into multinucleated myotubes within few days. Changes in integrin expression pattern as well as extensive synthesis of new and remodeling of the existing ECM are needed during this time for successful adhesion of elongating myotubes (39 40 41) . To analyze the functional relevance of the Mss4/integrin dependent matrix organization, we studied the myotube formation in cells with down-regulated Mss4 expression. The Mss4 expressing WT C2C12 cells start to fuse at 2–3 d after transfer into differentiation medium and form well-developed myotubes within 6 d (Fig. 8 A, upper panels). The Mss4 low-expressing myoblasts E2 also fuse, albeit with a delay of at least two days compared to Mss4 expressing cells, but the fused cells could not elongate and developed mostly into sac-like myotubes with centrally located nuclei (Fig. 8A , lower panels). Analysis of markers for differentiated muscle cells, myosin heavy chain (MyHC) and integrin {alpha}7A subunit, confirmed the delayed beginning of E2 cell fusion. However, it also showed that the initiation of the differentiation program was not basically altered in these cells (Fig. 8B ). The expression of Mss4 itself was not changed during differentiation of myoblasts, as our Reverse transcriptase-PCR (RT-PCR) tests showed (Fig. 8C ), only the adhesion of newly formed myotubes was disturbed in cells with down-regulated Mss4. To ensure further that the pathological differentiation of Mss4 low-expressing myoblasts into myotubes was indeed due to incorrect matrix remodeling or assembly, we compared the Mss4 expressing and low-expressing cells for organization of FN and laminin ECM proteins in differentiation. During myotube formation C2C12 cells start to express increasing amounts of the {gamma}1-chain containing laminin-2 and laminin-10/11 (42) . Immunofluorescence staining of myoblasts and myotubes with a monoclonal anti-laminin-{gamma}1 Ab showed that the expression of laminins was negligible in myoblasts but increased during differentiation in both Mss4 expressing and low-expressing cells, albeit with a small delay in the latter. However, in Mss4 expressing cells laminin molecules were organized into a well-formed network, which was mostly detected on the surface of newly built myotubes (Fig. 9 A). In low-expressing cells, we usually noted a less-developed laminin network, supporting thus the effect we already observed for FN. No large difference in the FN network was observed between the two myoblast lines at day 0 (Fig. 9B ). The cells were incubated for 2 d before changing the growth medium to differentiation medium, enough time to reach a steady state in assembly and remodeling of FN for both types of cells. During myotube formation, however, the FN amount decreased in Mss4 expressing cells, being present only on nonfused cells but not on myotubes, as already described for myogenesis (43) . In sharp contrast, the FN amount in Mss4 low-expressing cells increased, when these cells were transferred into differentiation medium (Fig. 9B ), reflecting a defect in remodeling of ECM proteins in these cells. The degradation of FN was not totally disturbed, but delayed, a finding that correlated very well with the slower myoblast fusion.


Figure 8
View larger version (111K):
[in this window]
[in a new window]

  		Figure 8. Down-regulation of Mss4 impairs myoblast differentiation. A) Mss4 expressing (E1) or low-expressing (E2) C2C12 myoblasts were cultured in growth medium until 70–80% confluency (day 0) and then shifted to differentiation medium (days 1–6). Phase contrast pictures were taken by inverted microscope. B) RIPA lysates (20 µg protein/lane) from cells cultured in growth or differentiation medium were immunoblotted with anti-MyHC, anti-integrin {alpha}7A or anti-FHL2 antibodies. The latter served as a loading control. C) mRNA was isolated from proliferating or differentiating cells and RT-PCR was performed with appropriate specific primers for Mss4 and GAPDH as a control.


Figure 9
View larger version (72K):
[in this window]
[in a new window]

  		Figure 9. Down-regulation of Mss4 impairs FN remodeling during myoblast differentiation. Mss4 expressing or low-expressing myoblasts were cultured in growth medium until 70–80% confluency (day 0), shifted to differentiation medium (days 1–6) and costained for laminin {gamma}1 chain (A) and FN (B) at the days indicated. To better distinguish between FN (red) and laminin (green) distribution during differentiation, the stainings are presented on separate pictures, but the same cell images are shown. Nuclei of cells were counterstained with DAPI.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
We have identified Mss4, a nucleotide exchange factor for Rab GTPases and a protein with chaperone properties, as a novel integrin-binding protein that interacted with the majority of {alpha}-integrin subunits tested. Mss4 binds to the membrane proximal {alpha}-helical domain of {alpha}-integrin chains, which is an extension of the transmembrane {alpha}-helix and consists of the conserved KXGFFKR region with at least three following amino acids and is common to all integrins. By interacting with the ß chain this area keeps the receptor in a closed or inactive conformation (3 , 4 , 11 , 14 , 44) . Unclasping of the two integrin chains leads to activation of the receptor. The integrity of the {alpha}-helical structure with its hydrophobic and charged amino acids has been shown to be important for interaction of the two integrin chains (8 , 45) . Mss4 requires the whole {alpha}-helical stretch of {alpha}-integrin chains for binding, as well as the phenylalanins and the arginine of the KXGFFKR motif, mutation of which abolishes the interaction between the {alpha}- and ß-chains converting the receptor into a constitutively active form. Mutation of these residues also abrogates the binding of Mss4 to integrins (34) , suggesting that the intact {alpha}-helical structure is important for this interaction. Mss4 does not bind to {alpha}IIb, {alpha}L, or {alpha}v, to {alpha}-chains of integrins that have to be activated before they can bind ligands (46) . Structural and functional analyses of {alpha}IIbß3 cytosolic parts have shown that in the inactive state the acidic C-terminal end of the {alpha} chain associates with the positively charged membrane proximal conserved region (3 , 4) . Unmasking the {alpha}-helix of the {alpha}IIb integrin chain by deletion or exchanging its C-terminal part renders it to the activated state (9) and allowed in our experiments the binding of Mss4 to this integrin {alpha}-chain too. Thus, the Mss4/integrin binding data indicate that Mss4 interacts only with receptors that are in the open, activated conformation, with the {alpha}-helix being the principal binding site.

Using the yeast two-hybrid assay, we have shown previously that the aromatic amino acid residue before the first lysine of the conserved cytosolic region of integrin {alpha}-chains is essential for interaction with Mss4. Here, we confirmed this conclusion showing that Mss4 can be coprecipitated from mammalian cell lysates only with {alpha}-integrin chains containing this aromatic amino acid. Furthermore, the Mss4-mediated MMP activation and FN allocation was affected only by PTD-{alpha}-integrin chimeras with a phenylalanine before the conserved KXGFFKR region and not with those having instead aspartate at this site. These data support the supposed involvement of the integrin transmembrane region in their function (47 , 48) . We cannot exclude that the used PTD peptides might also interfere with other cytosolic integrin-binding proteins affecting thus various integrin functions. However, the Mss4-mediated MMP activation or FN assembly were troubled only when the Mss4-binding PTD-F but not PTD-D peptide was introduced into cells. Moreover, as our control experiments showed, the general integrin function was not perturbed by these peptides, at least at concentrations we used (up to 400 µM). The cells adhered and spread on fibronectin or laminin1 in the presence of either PTD-D or PTD-F, and they also formed focal adhesion structures, as judged by immunofluorescence staining of FHL2. This protein directly interacts with ß1 and {alpha}3 and {alpha}7 integrin subunits and is recruited via these interactions into focal adhesions (35) .

We show here that the binding site for integrins on Mss4 overlaps with those for Rab GTPases, as mutation of amino acids that abrogate the association of Mss4 with Rabs interferes also with its binding to integrins. This observation is in a good agreement with recently published findings describing the Mss4:Rab8 structure (19) . These showed that the {alpha}SI-ß2 region of Rab8 is part of the binding site for Mss4. More interestingly, this among exocytic Rabs most conserved region contains sequence homology to the conserved {alpha}-helical motif of {alpha}-integrin chains. This stretch has proposed to be sufficient for recognition by Mss4, providing an anchor point for Rab folding without binding a GDP or GTP nucleotide. Our results thus suggest that interaction of Mss4 with integrins and Rabs is of a competitive nature and that the functional consequence of Mss4/integrin interaction is not of exocytic character.

As Mss4 binds to the same {alpha}-helical motif of {alpha}-integrin chains that is important for membrane presentation and the affinity state of integrin receptors, we initially hypothesized that Mss4 might regulate these integrin functions. However, analyses of integrin/Mss4 interaction showed that it is rather responsible for activation of matrix metalloproteinases and remodeling of secreted matrix proteins.

The ECM network undergoes continuous remodeling with degradation of existing and introduction of newly synthesized proteins. Furthermore, to be incorporated into a functional network that facilitates cellular functions like spreading or migration, many ECM proteins have to be cleaved to be brought into a biologically active conformation (29 , 49 , 50) . The remodeling of ECM networks is intimately related to integrin function and MMP activation. Despite the high self-assembly capacity of many ECM proteins, the formation of a regular ECM network is more efficient when it occurs at the cell surface rather than when it occurs in a solution (22 , 28 , 29) . It has been shown that a coordinated action of MT1-MMP/integrin protein complex is important for cell surface directed transfer of pro-MMP to active MMP forms [for reviews see (51 52 53) ]. Moreover, a direct association between integrins and MT1-MMP via their extracellular domains has been demonstrated (31 , 32) . These authors have further assumed that the activation of MT1-MMP/integrin protein complex might be regulated by still unknown cytosolic proteins. Mss4 is very likely such a regulator. Our results showed that Mss4/integrin interaction is essential for ECM assembly and remodeling and that binding of Mss4 to integrins is needed for cleavage of pro-MMP-2 and –9 into their active forms. MT1-MMP is the major activator of MMP-2 at the cell surface (51) . Its inhibition with blocking antibodies affected the FN assembly in a similar manner to that seen with peptide competition of the Mss4/integrin interaction. In cells, Mss4 is colocalized with integrins and MT1-MMP at the adhering front or at distal ends of newly formed filopodia or lamellipodia. Considering these facts, we hypothesize that Mss4 is a member of the MT1-MMP/integrin protein complex regulating its activity (Fig. 10 ). While MT1-MMP associates with integrins extracellularly, Mss4 binds to integrins intracellularly. A direct binding of Mss4 to MT1-MMP is rather unlikely, as mutation of the complete cytosolic tail of MT1-MMP did not abrogate the ability of the enzyme to activate MMP-2 (54) . The binding of Mss4 to integrins occurs at newly formed adhesion contacts where {alpha}- and ß-chains are unclasped after ligand binding and the membrane proximal {alpha}-helical domain of {alpha}-integrin chains is exposed for interaction with cytosolic proteins. Formation of the Mss4/integrin/MT1-MMP protein complex leads to directed activation of the MT1-MMP protease, which in turn leads to conversion of pro-MMP-2 and its substrate pro-MMP-9 into their active form. These enzymes now cleave ECM proteins, opening additional binding sites for integrins, thus facilitating integrin clustering and polymerization of ECM proteins on the cell surface. However, formation of large integrin clusters releases Mss4. As Mss4 binds to the conserved region, common to all {alpha}-integrin chains, it is possible that modulation of MMP activity via interaction with Mss4 is a general property of integrin receptors. The delayed laminin assembly during differentiation of Mss4 low-expressing C2C12 cells supports this assumption.


Figure 10
View larger version (68K):
[in this window]
[in a new window]

  		Figure 10. A proposed model for regulation of MMP activation by Mss4. Integrins at newly formed cell adhesion sites bound their ligand causing the two subunits to unclasp, thus opening the membrane proximal {alpha}-helical domain of {alpha}-integrin chains for interaction with Mss4. This leads to formation of a functional Mss4/integrin/MT1-MMP protein complex resulting in spatial-directed conversion of pro-MMPs into enzymatically active molecules by MT1-MMP. Activated MMPs cleave FN molecules, opening additional integrin binding sites as well as binding sites for FN polymerization, facilitating the formation of a regular FN network. When integrins have formed large focal adhesion clusters, Mss4 is released from them through an as yet unknown mechanism.

Myogenesis is accompanied with changes in integrin receptor expression pattern as well as with remodeling of ECM proteins. While proliferating and migrating myoblasts express high amounts of the FN-binding {alpha}5ß1 integrin, they switch to the laminin-binding {alpha}7ß1 integrin during myotube formation, which is then the major integrin receptor in adult muscles. {alpha}5ß1 was reported to be a receptor supporting myoblast proliferation but not differentiation (55) . Accordingly, myoblasts secrete a large amount of fibronectin, which is replaced by laminins during myogenesis and myotubes do not have a fibronectin matrix (40 , 56) . The importance of ECM remodeling during myogenesis is also highlighted by the fact that the expression of MMP-2 and –9, the major MMPs in muscle tissue, is increased during myotube formation in vitro or muscle regeneration in vivo (39 , 41 , 57) . Interestingly, C2C12 myoblasts with down-regulated Mss4 protein showed a defect in remodeling of FN when they were exposed to differentiation medium. While the reduced activity of MMP-2 and –9 in proliferating myoblasts resulted in an insufficient assembly of the FN network, it led to accumulation of FN fibers due to a degradation defect in differentiating cells. This fact is in good agreement with the defect of these cells to convert the pro-MMP-2 and –9 into active enzymes. We suppose that this was the major reason for the delayed entry into myotube formation and for the delayed switch in expression of muscle specific integrins and laminins. Once formed, these myotubes were, however, unable to grow in the length, probably because of their inability to properly integrate the ECM at growing myotube tips to give them an adequate mechanical support for further elongation. As a result, the ongoing fusion of myoblasts leads to formation of sac-like myotubes, due to their insufficient attachment.

We have shown in the present report that Mss4 binds to the membrane proximal conserved region of {alpha}-integrin chains and regulates in an integrin-dependent manner the conversion of inactive pro-MMPs into their enzymatically active forms, probably via activation of the MT1-MMP/integrin protein complex. Inhibition of Mss4/integrin interaction leads to delayed MMP activation and to disturbed allocation and remodeling of ECM proteins. In C2C12 myoblasts this results in severe defects of myotube formation during in vitro myogenesis.


   ACKNOWLEDGMENTS
 
We thank Dr. Neil Smyth and Dr. Thomas Pap for critical reading the manuscript. This work was supported by a grant from the Mildred-Scheel-Stiftung für Krebsforschung and by the fund "Innovative Medical Research" of the University of Münster Medical School to V.W.

Received for publication August 3, 2006. Accepted for publication September 22, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Hynes, R. O. (2002) Integrins: bidirectional, allosteric signaling machines. Cell 110,673-687[CrossRef][Medline]
  2. Hynes, R. O. (2002) A reevaluation of integrins as regulators of angiogenesis. Nat. Med. 8,918-921[CrossRef][Medline]
  3. Vinogradova, O., Haas, T., Plow, E. F., Qin, J. (2000) A structural basis for integrin activation by the cytoplasmic tail of the alpha IIb-subunit. Proc. Natl. Acad. Sci. U. S. A. 97,1450-1455[Abstract/Free Full Text]
  4. Vinogradova, O., Velyvis, A., Velyviene, A., Hu, B., Haas, T., Plow, E., Qin, J. (2002) A structural mechanism of integrin alpha(IIb)beta(3) "inside-out" activation as regulated by its cytoplasmic face. Cell 110,587-597[CrossRef][Medline]
  5. Briesewitz, R., Kern, A., Smilenov, L. B., David, F. S., Marcantonio, E. E. (1996) The membrane-cytoplasm interface of integrin alpha subunits is critical for receptor latency. Mol. Biol. Cell 7,1499-1509[Abstract]
  6. De Melker, A. A., Kramer, D., Kuikman, I., Sonnenberg, A. (1997) The two phenylalanines in the GFFKR motif of the integrin alpha6A subunit are essential for heterodimerization. Biochem. J. 328(Pt 2),529-537
  7. Ginsberg, M. H., Yaspan, B., Forsyth, J., Ulmer, T. S., Campbell, I. D., Slepak, M. (2001) A membrane-distal segment of the integrin alpha IIb cytoplasmic domain regulates integrin activation. J. Biol. Chem. 276,22514-22521[Abstract/Free Full Text]
  8. Hughes, P. E., Diaz-Gonzalez, F., Leong, L., Wu, C., McDonald, J. A., Shattil, S. J., Ginsberg, M. H. (1996) Breaking the integrin hinge. A defined structural constraint regulates integrin signaling. J. Biol. Chem. 271,6571-6574[Abstract/Free Full Text]
  9. Hughes, P. E., O’Toole, T. E., Ylanne, J., Shattil, S. J., Ginsberg, M. H. (1995) The conserved membrane-proximal region of an integrin cytoplasmic domain specifies ligand binding affinity. J. Biol. Chem 270,12411-12417[Abstract/Free Full Text]
  10. Lu, C., Shimaoka, M., Zang, Q., Takagi, J., Springer, T. A. (2001) Locking in alternate conformations of the integrin alphaLbeta2 I domain with disulfide bonds reveals functional relationships among integrin domains. Proc. Natl. Acad. Sci. U. S. A 98,2393-2398[Abstract/Free Full Text]
  11. Takagi, J., Erickson, H. P., Springer, T. A. (2001) C-terminal opening mimics ‘inside-out’ activation of integrin alpha5beta1. Nat. Struct. Biol. 8,412-416[CrossRef][Medline]
  12. Yamanouchi, J., Hato, T., Tamura, T., Fujita, S. (2004) Suppression of integrin activation by the membrane-distal sequence of the integrin alphaIIb cytoplasmic tail. Biochem. J. 379,317-323[CrossRef][Medline]
  13. Calderwood, D. A., Yan, B., de Pereda, J. M., Alvarez, B. G., Fujioka, Y., Liddington, R. C., Ginsberg, M. H. (2002) The phosphotyrosine binding-like domain of talin activates integrins. J. Biol. Chem. 277,21749-21758[Abstract/Free Full Text]
  14. Travis, M. A., Humphries, J. D., Humphries, M. J. (2003) An unraveling tale of how integrins are activated from within. Trends Pharmacol. Sci. 24,192-197[CrossRef][Medline]
  15. Burton, J., Roberts, D., Montaldi, M., Novick, P., De Camilli, P. (1993) A mammalian guanine-nucleotide-releasing protein enhances function of yeast secretory protein Sec4. Nature 361,464-467[CrossRef][Medline]
  16. Burton, J. L., Burns, M. E., Gatti, E., Augustine, G. J., De Camilli, P. (1994) Specific interactions of Mss4 with members of the Rab GTPase subfamily. EMBO J. 13,5547-5558[Medline]
  17. Yu, H., Schreiber, S. L. (1995) Cloning, Zn2+ binding, and structural characterization of the guanine nucleotide exchange factor human Mss4. Biochemistry 34,9103-9110[CrossRef][Medline]
  18. Yu, H., Schreiber, S. L. (1995) Structure of guanine-nucleotide-exchange factor human Mss4 and identification of its Rab-interacting surface. Nature 376,788-791[CrossRef][Medline]
  19. Itzen, A., Pylypenko, O., Goody, R. S., Alexandrov, K., Rak, A. (2006) Nucleotide exchange via local protein unfolding-structure of Rab8 in complex with MSS4. EMBO J. 25,1445-1455[CrossRef][Medline]
  20. Zhu, Z., Dumas, J. J., Lietzke, S. E., Lambright, D. G. (2001) A helical turn motif in Mss4 is a critical determinant of Rab binding and nucleotide release. Biochemistry 40,3027-3036[CrossRef][Medline]
  21. Thaw, P., Baxter, N. J., Hounslow, A. M., Price, C., Waltho, J. P., Craven, C. J. (2001) Structure of TCTP reveals unexpected relationship with guanine nucleotide-free chaperones. Nat. Struct. Biol. 8,701-704[CrossRef][Medline]
  22. Colognato, H., Winkelmann, D. A., Yurchenco, P. D. (1999) Laminin polymerization induces a receptor-cytoskeleton network. J. Cell Biol. 145,619-631[Abstract/Free Full Text]
  23. Deryugina, E. I., Quigley, J. P. (2006) Matrix metalloproteinases and tumor metastasis. Cancer Metastasis Rev. 25,9-34[CrossRef][Medline]
  24. Jones, C. B., Sane, D. C., Herrington, D. M. (2003) Matrix metalloproteinases: a review of their structure and role in acute coronary syndrome. Cardiovasc. Res. 59,812-823[Abstract/Free Full Text]
  25. Kerkela, E., Saarialho-Kere, U. (2003) Matrix metalloproteinases in tumor progression: focus on basal and squamous cell skin cancer. Exp. Dermatol. 12,109-125[CrossRef][Medline]
  26. Leeman, M. F., Curran, S., Murray, G. I. (2003) New insights into the roles of matrix metalloproteinases in colorectal cancer development and progression. J. Pathol. 201,528-534[CrossRef][Medline]
  27. Li, S., Harrison, D., Carbonetto, S., Fassler, R., Smyth, N., Edgar, D., Yurchenco, P. D. (2002) Matrix assembly, regulation, and survival functions of laminin and its receptors in embryonic stem cell differentiation. J. Cell Biol. 157,1279-1290[Abstract/Free Full Text]
  28. Magnusson, M. K., Mosher, D. F. (1998) Fibronectin: structure, assembly, and cardiovascular implications. Arterioscler Thromb. Vasc. Biol. 18,1363-1370[Free Full Text]
  29. Wierzbicka-Patynowski, I., Schwarzbauer, J. E. (2003) The ins and outs of fibronectin matrix assembly. J. Cell Sci. 116,3269-3276[Abstract/Free Full Text]
  30. Sato, H., Takino, T., Miyamori, H. (2005) Roles of membrane-type matrix metalloproteinase-1 in tumor invasion and metastasis. Cancer Sci. 96,212-217[CrossRef][Medline]
  31. Deryugina, E. I., Ratnikov, B., Monosov, E., Postnova, T. I., DiScipio, R., Smith, J. W., Strongin, A. Y. (2001) MT1-MMP initiates activation of pro-MMP-2 and integrin alphavbeta3 promotes maturation of MMP-2 in breast carcinoma cells. Exp. Cell Res. 263,209-223[CrossRef][Medline]
  32. Galvez, B. G., Matias-Roman, S., Yanez-Mo, M., Sanchez-Madrid, F., Arroyo, A. G. (2002) ECM regulates MT1-MMP localization with beta1 or alphavbeta3 integrins at distinct cell compartments modulating its internalization and activity on human endothelial cells. J. Cell Biol. 159,509-521