Onion (Allium. cepa L), garlic (A. sativum L.), and other members of the Allium genus produce volatile antimicrobial thiosulfinates upon cellular damage. Allicin has been known since the 1950s as the primary antimicrobial thiosulfinate compound and odorant produced by garlic. However, the roles of endogenous thiosulfinate production in host-bacterial pathogen interactions have not been described. The bacterial onion pathogen Pantoea ananatis, which lacks both the virulence Type III and Type II Secretion Systems, induces necrotic symptoms and extensive cell death in onion tissues dependent on a proposed secondary metabolite synthesis chromosomal gene cluster. We found strong correlation between the genetic requirements for P. ananatis to colonize necrotized onion tissue and its capacity for tolerance to the thiosulfinate allicin based on the presence of an eleven gene, plasmid-borne, virulence cluster of sulfur/redox genes. We have designated them ‘alt’ genes for allicin tolerance. We show that allicin and onion thiosulfinates restrict bacterial growth with similar kinetics. The alt gene cluster is sufficient to confer allicin tolerance and protects the glutathione pool during allicin treatment. Independent alt genes make partial phenotypic contributions indicating that they function as a collective cohort to manage thiol stress. Our work implicates endogenous onion thiosulfinates produced during cellular damage as mediators of interactions with bacteria. The P. ananatis-onion pathosystem can be modeled as a chemical arms race of pathogen attack, host chemical counter-attack, and pathogen resistance.Significance StatementAlliums (e.g. onion and garlic), after sustaining cellular damage, produce potent antimicrobial thiosulfinates that react with cellular thiols. The bacterial onion pathogen Pantoea ananatis, which lacks the virulence Type III and Type II Secretion Systems, induces cell death and necrotic symptoms on onions. We have identified a plasmid-borne cluster of sulfur/redox virulence genes that 1) are required for P. ananatis to colonize necrotized onion tissue, 2) are sufficient for tolerance to the thiosulfinates, and, 3) protect the glutathione pool during thiosulfinate treatment. We propose that the thiosulfinate production potential of Allium spp. governs Allium-bacterial interaction outcomes and that the P. ananatis-onion pathosystem can be modeled as a chemical arms race of attack and counterattack between the pathogen and host.