QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 2 December 2003
doi: 10.1242/jcs.00845

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ilic, D. Articles by Damsky, C. H. Search for Related Content PubMed PubMed Citation Articles by Ilic, D. Articles by Damsky, C. H.

FAK promotes organization of fibronectin matrix and fibrillar adhesions

Duko Ili^1,*, Branka Kovai², Kohei Johkura³, David D. Schlaepfer⁴, Nenad Tomaevi⁵, Qin Han¹, Jae-Beom Kim¹, Kyle Howerton¹, Clark Baumbusch¹, Naoko Ogiwara³, Daniel N. Streblow⁶, Jay A. Nelson⁶, Paul Dazin⁷, Yuji Shino⁸, Katsunori Sasaki^3,9 and Caroline H. Damsky^1,10

¹ Department of Stomatology, ² Medicine, and ¹⁰ Anatomy, and the ⁷ Howard Hughes Medical Institute, University of California San Francisco, San Francisco, California 94143, USA
³ Department of Anatomy and Organ Technology, and ⁹ Institute of Organ Transplants, Reconstructive Medicine and Tissue Engineering, Shinshu University School of Medicine, Matsumoto 390-8621, Japan
⁴ Department of Immunology, The Scripps Research Institute, La Jolla, California 92037, USA
⁵ Biochemistry Department, Cytokinetics Inc., South San Francisco, California 94080, USA
⁶ Department of Molecular Microbiology and Immunology, Oregon Health Sciences University, Portland, Oregon 97201, USA
⁸ Department of Microbiology, School of Medicine Chiba University, Chiba 260, Japan

View larger version (98K): [in a new window]   Fig. 1. FN matrix organization is altered by FAK-deficiency. (A) FAK-/- and control FAK+/+ mouse endothelial cells were isolated from embryos and embryoid bodies. Endothelial cells isolated from mouse embryos were differentiated in vivo and were immortalized by p53-deficiency. Cells isolated from embryoid bodies were differentiated in vitro and were immortalized with polyoma middle T. FN staining in permeabilized FAK-/- cells has a granular appearance and an abundant disarray of short thin fibrils in comparison with FAK+/+ controls. (B) FN in wild-type and FAK-/- E8.5 mouse littermates as seen by immunofluorescence on sections through the headfold. In both embryos the most intense FN staining was between ectodermal and mesodermal layers (arrowheads). The abundance of fibrillar FN matrix was greater in headfold mesoderm of FAK+/+ mice, whereas in FAK-/- mice the FN matrix was less fibrillar and more punctate (arrows, black and white images). (C) Immunoelectron microscopy of FN matrix. FN-rich areas on sections of wild-type and FAK-/- E8.5 mouse embryos. Fn-/- embryos were used as a negative control. (D) Comparison of FN mRNA levels in embryos from FAK heterozygote crosses. Eight micrograms of total RNA of wild-type (+/+), FAK-heterozygous (+/-), and FAK-null (-/-) genotype was analyzed by northern blot. The membrane was hybridized with radiolabeled mouse FN cDNA probe. 28S and 18S are ribosomal RNAs in an ethidium bromide-stained gel to demonstrate equal loading and quality of RNA. (E) Western blot analysis of FN expression in E8.5 FAK-/- embryos. E8.5 littermates from FAK+/- crosses were genotyped, and embryos of the same genotype were pooled and lysed. Slightly stronger FN bands were detected in lysates of FAK-/- embryos. The absence of FN bands in lysates of Fn-/- embryos verifies their specificity. Skp1 was used as a loading control.

View larger version (58K): [in a new window]   Fig. 2. FAK-null cells have altered FN allocation and organization. (A) Staining of confluent live cells showed highly organized matrix on the dorsal surface of FAK+/+ cells. FAK-/- cells had almost no detectable organized fibrillar matrix at the dorsal surface. Stable reintroduction of FAK into these cells (DA2 cells) rescued matrix organization to normal. (B) FN deposits (red) followed cell-cell boundaries in confluent cultures of FAK-/- cells, as visualized with ß-catenin immunostaining (green). (C) Length distribution of FN trails left behind by migrating cells. There is a higher frequency of short FN fibrils left behind migrating FAK-/- cells. Length of fibrils was measured using a 1 µm grid laid over immunostained FN trail of migrating cells.

View larger version (21K): [in a new window]   Fig. 3. Distribution of radiolabeled FN in different fractions is comparable between FAK-/- and DA2 cells. Proteins were labeled with a pulse of 1.0 mCi/ml L-[35S]methionine and L-[35S]cysteine mixture for 15 minutes. Distribution of 35S-labeled FN was followed after 2 and 15 hours in three different fractions (DOC-soluble, DOC-insoluble SDS-soluble, and secreted into medium) and compared with a total amount of the labeled FN (time 0). Data (means from three repeats) suggest that there is only slightly more FN incorporated into the DOC-insoluble SDS-soluble fraction in DA2 cells. However, the FAK-/- cells secreted somewhat more FN into the medium.

View larger version (78K): [in a new window]   Fig. 4. FAK catalytic activity, PR domains, and localization in focal adhesion sites are required for proper allocation of FN matrix. (A) Loss-of-function experiments. FAK+/+ cells were transduced with FAK autophosphorylation mutant (Y397F FAK), kinase-dead FAK (KD FAK), PR domains mutant (PA FAK), FRNK, and L1034S FRNK using adenoviral vectors and plated at confluent density. Forty-eight hours later only non-transduced cells and cells transduced with L1034S FRNK, which does not localize in focal contacts, retained complex FN matrix organization, as assessed with staining of live cells. Western blots show the phosphorylation states of Y397 and Y861 on FAK and FRNK and expression levels of endogenous and exogenous FAK and FRNK proteins. FRNK expression could be detected only with antibodies that recognize an epitope at the C terminus of the full-length FAK protein. (B) Gain-of-function experiments. FAK-/- cells were transduced with wild-type FAK, Y397F FAK, KD FAK, PA FAK and FRNK using adenoviral vectors. Transduced cells were plated at confluent density. Only wild-type FAK was able to rescue organization of FN matrix in FAK-/- cells 48 hours later. Whole lysate western blots show the phosphorylation states of Y397 and Y861 on FAK and FRNK, and the expression levels of exogenous FAK and FRNK proteins. Arrows point to the size of FRNK, and arrowheads to the FAK band in all panels. Actin was used as a loading control in both sets of experiments.

View larger version (84K): [in a new window]   Fig. 5. Presence of PYK2/FAK chimeras in focal contacts rescues FN allocation. (A) Schematic diagram of Pyk2/FAK chimeras and properties of cell lines expressing them (Klingbeil et al., 2001). (B) Stable expression of the PYK2/FAK chimera in FAK-/- cells either at a low level (CA3 cells) or at a high level (CB4 cells) was sufficient to rescue FN matrix allocation on the dorsal surface of confluent cells, as assessed with FN and ß1-integrin staining. Mutation in the paxillin-binding domain of FAT (L1034S) prevents localization of chimeric PYK2/FAK protein in focal adhesion sites. Neither low levels (SE6 cells) nor high levels (SX4 cells) of stable L1034S PYK2/FAK chimera expression were sufficient to rescue the cells.

View larger version (99K): [in a new window]   Fig. 6. FAK is required for the formation of fibrillar adhesions. (A) Tensin immunostaining shows lack of fibrillar adhesions (arrowheads) on the dorsal surface of cells that do not have either FAK or PYK2/FAK chimeras localized in focal adhesions. (B). Anti-active ß1 integrin 9EG7 Ab-chase experiments demonstrate defects in formation of fibrillar adhesions and organized FN in FAK-/- cells. Live cells were exposed to the anti-active ß1 integrin 9EG7 mAb for 45 minutes. After the Ab was washed away, one set of cells was fixed immediately (t=0) and a second set 2 hours later (t=2). Procedures for subsequent labeling of FN, active ß1 integrin and paxillin are given in Materials and Methods. In overlaid images paxillin, a marker of focal adhesions, is colored blue, FN is red, and chased 9EG7 mAb is green. Within 2 hours ß1 integrin and FN were translocated backward to the dorsal surface in cells that have FAK in focal contacts (FAK+/+ and DA2). In FAK-/- cells only a few ß1 integrin/FN patches were detectable at the margins of spread cells. Integrin/antibody complexes are endocytosed at a much higher rate in cells that do not have either FAK or FAK/PYK2 chimera in focal adhesions. These complexes are not visualized because secondary anti-rat antibody is incubated with non-permeabilized cells.

View larger version (49K): [in a new window]   Fig. 7. Organization of actin cytoskeleton depends on localization of either FAK or FAK/PYK2 chimeras in focal adhesions. (A) The rate of actin polymerization in vitro is similar using lysates of cells that do (FAK+/+, DA2) and cell that do (FAK+/+, DA2) and cells that do not (FAK-/-) have FAK in focal contacts. (B) Cells that lacked FAK-family kinase activity in focal adhesions and had disorganized FN matrix (FAK-/-, SE6, SX4) were also unable to organize stress fibers into a parallel pattern (red, actin; blue, nuclei).