ERGIC-53 KKAA signal mediates endoplasmic reticulum retrieval in yeast.

Studies on the ERGIC-53 KKAA signal have revealed a new mechanism for static retention of mammalian proteins in the endoplasmic reticulum (Andersson, H., Kappeler, F., Hauri, H. P. (1999): Protein targeting to endoplasmic reticulum by dilysine signals involves direct retention in addition to retrieval. J. Biol. Chem. 274,15080 - 15084). To test if this mechanism was conserved in yeast, the ERGIC-53 KKAA signal was transferred on two different yeast reporter proteins. Making use of a genetic assay, we demonstrate that this signal induces COPI-dependent ER retrieval. ER retention of KKAA-tagged proteins was impaired in yeast mutants affected in COPI subunits. Furthermore, biochemical analysis of post-ER carbohydrate modifications detected on reporter proteins indicated that KKAA-tagged proteins recycle continuously within early compartments of the secretory pathway. Therefore in yeast, the KKAA signal might only function as a classical dilysine ER retrieval signal.

The ADP-ribosylation factor GTPase-activating protein Glo3p is involved in ER retrieval.

Retrograde transport of proteins from the Golgi to the endoplasmic reticulum (ER) has been the subject of some interest in the recent past. Here a new thermosensitive yeast mutant defective in retrieval of dilysine-tagged proteins from the Golgi back to the endoplasmic reticulum was characterized. The ret4-1 mutant also exhibited a selective defect in forward ER-to-Golgi transport of some secreted proteins at the non-permissive temperature. The corresponding RET4 gene was found to encode Glo3p, a GTPase-activating protein (GAP) specific for ADP-ribosylation factor (ARF). In vitro, the Glo3 thermosensitive mutant showed a reduced ARF1-GAP activity. The Glo3 protein belongs to a family of zinc finger proteins that may include additional ARF-GAPs. Gene deletion experiments of other family members showed that only GLO3 deletion resulted in impaired retrieval of dilysine-tagged proteins back to the ER. These results demonstrate that Glo3p is the main ARF-GAP specifically involved in ER retrieval.

Cell adhesion to a population of laminin isoforms isolated from normal renal tissue.

To assess whether cells react differently towards a population of several laminin isoforms, as found in vivo, vs. a single isoform, we have compared the biological activity of kidney laminins to that of pure laminin 1. The kidney laminin preparation contained laminin 1 and further isoforms. Both substrates induced adhesion of a large spectrum of cell types, with kidney laminins being the most active. Unfolding of the coil-coiled conformation of the kidney isoforms negatively affected cell adhesion-promoting activity, which indicated that conformation-dependent cell binding is a characteristic feature of many or all laminins. Cellular interactions with kidney laminins were mediated by alpha3beta1 and alpha6beta1 integrins, with the contribution of alpha3beta1 being apparently lower than that of alpha6beta1 integrins. Immunofluorescence staining of vinculin and integrin subunits decorated focal adhesions on kidney laminins which differed in morphology from those formed on laminin 1 alone, in spite of the presence of the latter in the kidney preparation. These observations collectively indicate that tissue specific but often overlapping expression of laminin isoforms might modulate cell behavior by the activation of distinct sets of integrins and by the induction of distinct molecular assemblies within the cell adhesion signaling complexes.

Cell adhesion to laminin 1 or 5 induces isoform-specific clustering of integrins and other focal adhesion components.

Laminin 1 (alpha1beta1gamma1) and laminin 5 (alpha3beta3gamma2) induce cell adhesion with different involvement of integrins: both are ligands for the alpha6beta1 integrin, while alpha3beta1 integrin has affinity for laminin 5 only. These two laminin isoforms therefore provide good models to investigate whether alpha3beta1 and alpha6beta1 integrins play different roles in signal transduction and in focal adhesion formation. Laminin 1 or 5 induced adhesion of normal human skin fibroblasts to a similar extent but promoted different overall cell shapes. On laminin 1 the fibroblasts formed mainly filopodia-like structures, while on laminin 5 they developed lamellipodias. Staining of fibrillar actin with fluorescein-phalloidin revealed a similar organisation of the actin cytoskeleton on both substrates. However, integrin subunits and several cytoskeletal linker proteins, including vinculin, talin, and paxillin, showed an isoform-specific arrangement into focal adhesions. On laminin 1 they were recruited into thick and short aggregates localized at the termini of actin stress fibers, while on laminin 5 they appeared as dots or streaks clustered on a long portion of actin microfilaments. To test whether the differing affinity of laminin 1 or 5 for alpha3beta1 integrin would explain the formation of morphologically different focal adhesions, cells were seeded on laminin 1 under conditions in which alpha3beta1 integrins were occupied by a function-blocking antibody. This resulted in the formation of focal adhesions similar to that observed on laminin 5, where the integrin is occupied by its natural ligand. These results provide the first evidence for a cross-talk between alpha3beta1 and alpha6beta1 integrins and indicate that occupancy of alpha3beta1 integrins results in a trans-dominant regulation of alpha6beta1 integrin clustering and of focal adhesions. It suggests that recruitment of integrins and cytoskeletal linker proteins are laminin isoform-specific and that tissue specific expression of laminin isoforms might modulate cell behavior by the activation of distinct sets of integrins and by the induction of distinct molecular assemblies within the cell adhesion signaling complexes.

Targeting of cytoskeletal linker proteins to focal adhesion complexes is reduced in fibroblasts adhering to laminin-1 when compared to fibronectin.

Cellular interactions with the extracellular matrix determine to a large extent cell behavior, including cell migration. These interactions take place at specialized cellular structures, the focal adhesions, which have a substrate-specific morphology. To determine the molecular and functional relevance of this observation, the composition of isolated focal adhesions developed by fibroblasts adhering to fibronectin or laminin-1 was analyzed by indirect immunofluorescence and immunoblotting with or without stabilization of the structures by cross-linking. In the absence of cross-linking, integrins, talin, vinculin and, to a lower extent, paxillin remained associated with the focal adhesions formed on both substrates, indicating a tight association of these proteins with the extracellular matrix support. By contrast, alpha-actinin, FAK, and actin were apparently loosely maintained within focal adhesions and were found associated to these structures only after stabilization by cross-linking. Interestingly, although both substrates induced clustering and aggregation of all these proteins, their relative concentration, with the exception of alpha-actinin, was lower within the focal adhesions formed on laminin-1 than in those formed on fibronectin. Moreover, as assessed in migration assays, the locomotory speed of fibroblasts was higher on laminin-1 than on fibronectin. Altogether these results indicate that integrins involved in cellular interactions with fibronectin or laminin-1 trigger the formation of focal adhesion structures which differ by molecular organization, concentration in several adhesion plaque components, and function.

Impaired mechanical stability, migration and contractile capacity in vimentin-deficient fibroblasts.

Loss of a vimentin network due to gene disruption created viable mice that did not differ overtly from wild-type littermates. Here, primary fibroblasts derived from vimentin-deficient (-/-) and wild-type (+/+) mouse embryos were cultured, and biological functions were studied in in vitro systems resembling stress situations. Stiffness of -/- fibroblasts was reduced by 40% in comparison to wild-type cells. Vimentin-deficient cells also displayed reduced mechanical stability, motility and directional migration towards different chemo-attractive stimuli. Reorganization of collagen fibrils and contraction of collagen lattices were severely impaired. The spatial organization of focal contact proteins, as well as actin microfilament organization was disturbed. Thus, absence of a vimentin filament network does not impair basic cellular functions needed for growth in culture, but cells are mechanically less stable, and we propose that therefore they are impaired in all functions depending upon mechanical stability.