[show abstract][hide abstract] ABSTRACT: Cells require optimal substrate stiffness for normal function and differentiation. The mechanisms for sensing matrix rigidity and durotaxis, however, are not clear. Here we showed that control, Shp2-/-, integrin beta1-/-, and talin1-/- cell lines all spread to a threefold greater area on fibronectin (FN)-coated rigid polyacrylamide surfaces than soft. In contrast, RPTPalpha-/- cells spread to the same area irrespective of rigidity on FN surfaces but spread 3x greater on rigid collagen IV-coated surfaces than soft. RPTPalpha and alphavbeta3 integrins were shown previously to be colocalized at leading edges and antibodies to alphavbeta3 blocked FN rigidity sensing. When FN beads were held with a rigid laser trap at the leading edge, stronger bonds to the cytoskeleton formed than when held with a soft trap; whereas back from the leading edge and in RPTPalpha-/- cells, weaker bonds were formed with both rigid and soft laser traps. From the rigidity of the trap, we calculate that a force of 10 pN generated in 1 s is sufficient to activate the rigidity response. We suggest that RPTPalpha and alphavbeta3 at the leading edge are critical elements for sensing FN matrix rigidity possibly through SFK activation at the edge and downstream signaling.