They belong to the same subfamily, but the small GTPases Rap1 and K-Ras show sibling rivalry when it comes to cell adhesion. Oncogenic mutations in K-Ras cause cells to lose adhesion, whereas Rap1 has a positive influence on cell spreading and adhesion. William Arthur, Lawrence Quilliam and Jonathan Cooper used the fact that the Rap1 yeast orthologue activates a Rho protein by stimulating a guanine nucleotide-exchange factor (GEF) to explore the possibility that, in mammalian cells, Rap1 might also mediate cell spreading by activating Rho-family GTPases. Their findings are reported in The Journal of Cell Biology.

When suspended cells were re-plated onto a fibronectin surface, Rap1 and the Rho-family GTPase Rac1 were both activated. And cells expressing active forms of Rap1, Rac1 or Rac GEFs were all similarly spread. On further investigation of an interrelationship between Rap1 and Rho-family GTPases, the authors found that inhibiting Rac1, but not Cdc42 or RhoA, antagonized Rap1-induced cell spreading, which implies that Rap1 requires Rac1 for spreading. In support of this, Rap1 activation increased Rac1–GTP levels. And activating Rac1 when Rap1 function was compromised restored cell spreading.

Faced with several possibilities for how Rap1 might induce cell spreading through Rac1, the authors explored potential interactions of Rap1 with Rac GEFs. Rap1 could bind to the central Dbl-homology and pleckstrin homology (DH–PH) domains of VAV2 and Tiam1, but not to SWAP-70 or COOL-1. But although activated versions of these GEFs enhanced spreading in control cells, they couldn't do so in cells in which Rap1 activity was inhibited — unlike the case mentioned above, in which activated Rac1 could overcome impaired Rap1 activity.

Again, there were many possible reasons for the requirement of Rap1 activity for VAV2 and Tiam1 function. Active Rap1 didn't seem necessary to stimulate the catalytic activities of the GEFs, so, instead, the authors studied the potential influence of Rap1 on GEF localization. In studies using extracts from fractionated pseudopodia and by immunofluorescence, both GDP- and GTP-bound Rap1 were found in membrane protrusions that were associated with the substratum. So, too, was VAV2, but not when Rap1 activity was inhibited — indeed, this caused displacement of VAV2 from the cell edge.

It follows from these results, then, that artificially targeting Rap1-dependent Rac GEFs such as VAV2 and Tiam1 to the edge of the cell, where Rap1 is normally active, could compensate for a lack of Rap1 activity. By fusing the membrane-localization signals that are present in the Rap1 C-terminal 'CAAX' box and its adjacent N-terminal hypervariable domain to VAV2, the requirement for GTP-bound Rap1 in Rac1 activation could be bypassed.

The resulting Rac1-induced formation of productive membrane protrusions associated with the adjacent extracellular matrix thereby induces cell spreading. Such protrusive activity would, in turn, create new sites for Rap1 localization, and so the cycle should continue.