A type VI secretion system (T6SS) gene cluster has been reported in Acidovorax citrulli. Research on the activation conditions, functions, and the interactions between key elements in A. citrulli T6SS is lacking. Hcp (Hemolysin co-regulated protein) is both a structural protein and a secretion protein of T6SS, which makes it a special element. The aims of this study were to determine the role of Hcp and its activated conditions to reveal the functions of T6SS. In virulence and colonization assays of hcp deletion mutant strain Δhcp, tssm (type VI secretion system membrane subunit) deletion mutant strain Δtssm and double mutant ΔhcpΔtssm, population growth was affected but not virulence after injection of cotyledons and seed-to-seedling transmission on watermelon. The population growth of Δhcp and Δtssm were lower than A. citrulli wild type strain Aac5 of A. citrulli group II at early stage but higher at a later stage. Deletion of hcp also affected growth ability in different culture media, and the decline stage of Δhcp was delayed in KB medium. Biofilm formation ability of Δhcp, Δtssm and ΔhcpΔtssm was lower than Aac5 with competition by prey bacteria but higher in KB and M9-Fe3 medium+. Deletion of hcp reduced the competition and survival ability of Aac5. Based on the results of Western blotting and qRT-PCR analyses, Hcp is activated by cell density, competition, ferric irons, and the host plant. The expression levels of genes related to bacterial secretion systems, protein export, and several other pathways, were significantly changed in the Δhcp mutant compared to Aac5 when T6SS was activated at high cell density. Based on transcriptome data, we found that a few candidate effectors need further identification. The phenotypes, activated conditions and transcriptome data all supported the conclusion that although there is only one T6SS gene cluster present in the A. citrulli group II strain Aac5, it related to multiple biological processes, including colonization, growth ability, competition and biofilm formation.

Bacterial fruit blotch (BFB) caused by Acidovorax citrulli (Ac) is a devastating watermelon disease that severely impacts the global watermelon industry. Like other Gram-negative bacteria, the type three secretion system (T3SS) is the main pathogenicity factor of A. citrulli. The T3SS apparatus gene hrpE codes for the Hrp pilus and serves as a conduit to secret effector proteins into host cells. In this study, we found that the deletion of hrpE in A. citrulli results in the loss of pathogenicity on hosts and the hypersensitive response on non-hosts. In addition, the A. citrulli hrpE mutant showed a reduction in in vitro growth, in planta colonization, swimming and twitching motility, and displayed increases in biofilm formation ability compared to the wild type. However, when HrpE was transiently expressed in hosts, the defense responses, including reactive oxygen species bursts, callose deposition, and expression of defense-related genes, were activated. Thus, the A. Citrulli growth in HrpE-pretreated hosts was suppressed. These results indicated that HrpE is essential for A. citrulli virulence but can also be used by hosts to help resist A. citrulli. Our findings provide a better understanding of the T3SS pathogenesis in A. citrulli, thus providing a molecular basis for biopesticide development, and facilitating the effective control of BFB.

Acidovorax citrulli, the causal agent of bacterial fruit blotch, can be divided into two groups. Group I is mainly isolated from melon, and group II is mainly isolated from watermelon. In this study, group I and II strains were used to assess competition in vivo and in vitro by evaluating inhibition activity assays and by measuring population growth dynamics. Our results indicated that there is no mutual inhibition of growth between the two groups of strains in King’s B medium. The bacterial populations of A. citrulli strains were counted within 7 days after inoculation in melon and watermelon cotyledons and plotted against time to generate population growth curves. Area under the population growth curves was calculated. When the two groups of strains were inoculated separately into melon and watermelon cotyledons, the results of analysis of variance show that there was no significant difference. In this study, for the first time in an experimental setting, we inoculated two A. citrulli groups into melon and watermelon cotyledons at the same time and determined the population dynamics of each strain. The results showed that there was no significant difference between two group strains in melon cotyledons. However, in watermelon cotyledons, the area under population growth curves of group II strain were significantly higher than those of group I strain. Three-way analysis of variance results showed that there was interaction between host and grouping on the influence of strain population level (p < 0.0001). These data provide additional information on the host preference of different groups of A. citrulli.