Burkholderia cenocepacia Lipopolysaccharide Modification and Flagellin Glycosylation Affect Virulence but Not Innate Immune Recognition in Plants

UNLABELLED Burkholderia cenocepacia causes opportunistic infections in plants, insects, animals, and humans, suggesting that "virulence" depends on the host and its innate susceptibility to infection. We hypothesized that modifications in key bacterial molecules recognized by the innate immune system modulate host responses to B. cenocepacia. Indeed, modification of lipopolysaccharide (LPS) with 4-amino-4-deoxy-L-arabinose and flagellin glycosylation attenuates B. cenocepacia infection in Arabidopsis thaliana and Galleria mellonella insect larvae. However, B. cenocepacia LPS and flagellin triggered rapid bursts of nitric oxide and reactive oxygen species in A. thaliana leading to activation of the PR-1 defense gene. These responses were drastically reduced in plants with fls2 (flagellin FLS2 host receptor kinase), Atnoa1 (nitric oxide-associated protein 1), and dnd1-1 (reduced production of nitric oxide) null mutations. Together, our results indicate that LPS modification and flagellin glycosylation do not affect recognition by plant receptors but are required for bacteria to establish overt infection. IMPORTANCE Virulence and pathogenicity are properties ascribed to microbes, which actually require careful consideration of the host. Using the term "pathogen" to define a microbe without considering its host has recently been debated, since the microbe's capacity to establish a niche in a given host is a critical feature associated with infection. Opportunistic bacteria are a perfect example of microbes whose ability to cause disease is intimately related to the host's ability to recognize and respond to the infection. Here, we use the opportunistic bacterium Burkholderia cenocepacia and the host plant Arabidopsis thaliana to investigate the role of bacterial surface molecules, namely, lipopolysaccharide and flagellin, in contributing to infection and also in eliciting a host response. We reveal that both molecules can be modified by glycosylation, and although the modifications are critical for the bacteria to establish an infection, they do not impact the host's ability to recognize the pathogen.