Invasin is a key adhesin displayed on the outer membrane of Yersinia enterocolitica and Y. pseudotuberculosis that mediates the initial stages of infection. Invasin specifically targets microfold (M) cells in the small intestine by binding β1 integrins and is sufficient to trigger eukaryotic uptake of invasin-coated particles, including Yersinia, Escherichia coli, and latex beads. As a result, invasin has generated interest to mediate oral delivery of vaccines and other biologics. Integrin binding affinity has been shown to correlate with particle uptake; thus we hypothesized that invasin variants with higher affinity would confer enhanced internalization. We first performed alanine scanning of surface-exposed tyrosine residues to identify those contributing to integrin binding. We identified two residues, which, when substituted with alanine, reduced binding to soluble α5β1 integrin. Next, we constructed four targeted mutagenesis libraries spanning these and other residues known to contribute to binding, followed by enrichment of variants able to mediate Caco-2 cellular invasion and to bind soluble α5β1 integrin. We identified three amino acid substitutions that increased α5β1 integrin binding affinity as measured by flow cytometry and ELISA assays, two of which created a novel RGD motif surrounding the D911 residue critical for binding. This variant confers enhanced internalization into CHO cells but not Caco-2 cells when expressed on the E. coli surface. Further analysis showed that inclusion of an RGD expands invasin-integrin specificity, thereby impacting cellular selectivity. This work provides a molecular explanation for the lack of an RGD motif in invasin that is present in many other adhesins.