Three endophytic bacterial isolates were obtained in Italy from olive knots caused by Pseudomonas savastanoi pv. savastanoi. Phenotypic tests in combination with 16S rRNA gene sequence analysis indicated a phylogenetic position for these isolates in the genera Erwinia or Pantoea, and revealed two other strains with highly similar 16S rRNA gene sequences (>99 %), CECT 5262 and CECT 5264, obtained in Spain from olive knots. Rep-PCR DNA fingerprinting of the five strains from olive knots with BOX, ERIC and REP primers revealed three groups of profiles that were highly similar to each other. Multilocus sequence analysis (MLSA) based on concatenated partial atpD, gyrB, infB and rpoB gene sequences indicated that the strains constituted a single novel species in the genus Erwinia. The strains showed general phenotypic characteristics typical of the genus Erwinia and whole genome DNA–DNA hybridization data confirmed that they represented a single novel species of the genus Erwinia. The strains showed DNA G+C contents ranging from 54.7 to 54.9 mol%. They could be discriminated from phylogenetically related species of the genus Erwinia by their ability to utilize potassium gluconate, l-rhamnose and d-arabitol, but not glycerol, inositol or d-sorbitol. The name Erwinia oleae sp. nov. (type strain DAPP-PG 531^T = LMG 25322^T = DSM 23398^T) is proposed for this novel taxon.

Summary: In a number of compatible plant‐bacterium interactions, a rise in apoplastic Ca ²⁺ levels is observed, suggesting that Ca ²⁺ represents an important environmental clue, as reported for bacteria infecting mammalians. We demonstrate that Ca ²⁺ entry in Pseudomonas savastanoi pv. savastanoi ( Psav ) strain DAPP‐PG 722 is mediated by a Na ⁺ /Ca ²⁺ exchanger critical for virulence. Using the fluorescent Ca ²⁺ probe Fura 2‐AM, we demonstrate that Ca ²⁺ enters Psav cells foremost when they experience low levels of energy, a situation mimicking the apoplastic fluid. In fact, Ca ²⁺ entry was suppressed in the presence of high concentrations of glucose, fructose, sucrose or adenosine triphosphate (ATP). Since Ca ²⁺ entry was inhibited by nifedipine and LiCl, we conclude that the channel for Ca ²⁺ entry is a Na ⁺ /Ca ²⁺ exchanger. In silico analysis of the Psav DAPP‐PG 722 genome revealed the presence of a single gene coding for a Na ⁺ /Ca ²⁺ exchanger ( cneA ), which is a widely conserved and ancestral gene within the P. syringae complex based on gene phylogeny. Mutation of cneA compromised not only Ca ²⁺ entry, but also compromised the Hypersensitive response (HR) in tobacco leaves and blocked the ability to induce knots in olive stems. The expression of both pathogenicity ( hrpL , hrpA and iaaM ) and virulence ( ptz ) genes was reduced in this Psav ‐ cneA mutant. Complementation of the Psav ‐ cneA mutation restored both Ca ²⁺ entry and pathogenicity in olive plants, but failed to restore the HR in tobacco leaves. In conclusion, Ca ²⁺ entry acts as a ‘host signal’ that allows and promotes Psav pathogenicity on olive plants.

The endophytic bacterium Pantoea agglomerans DAPP‐PG 734 was previously isolated from olive knots caused by infection with Pseudomonas savastanoi pv. savastanoi DAPP‐PG 722. Whole‐genome analysis of this P. agglomerans strain revealed the presence of a Hypersensitive response and pathogenicity (Hrp) type III secretion system (T3SS). To assess the role of the P. agglomerans T3SS in the interaction with P. savastanoi pv. savastanoi, we generated independent knockout mutants in three Hrp genes of the P. agglomerans DAPP‐PG 734 T3SS (hrpJ, hrpN, and hrpY). In contrast to the wildtype control, all three mutants failed to cause a hypersensitive response when infiltrated in tobacco leaves, suggesting that P. agglomerans T3SS is functional and injects effector proteins in plant cells. In contrast to P. savastanoi pv. savastanoi DAPP‐PG 722, the wildtype strain P. agglomerans DAPP‐PG 734 and its Hrp T3SS mutants did not cause olive knot disease in 1‐year‐old olive plants. Coinoculation of P. savastanoi pv. savastanoi with P. agglomerans wildtype strains did not significantly change the knot size, while the DAPP‐PG 734 hrpY mutant induced a significant decrease in knot size, which could be complemented by providing hrpY on a plasmid. By epifluorescence microscopy and confocal laser scanning microscopy, we found that the localization patterns in knots were nonoverlapping for P. savastanoi pv. savastanoi and P. agglomerans when coinoculated. Our results suggest that suppression of olive plant defences mediated by the Hrp T3SS of P. agglomerans DAPP‐PG 734 positively impacts the virulence of P. savastanoi pv. savastanoi DAPP‐PG 722.