It is possible that differences in chemotaxis might be appreciated by the use of specific attractant or repellent molecules. However, it is not trivial to identify such specific compounds and further studies should be performed in order to address this point. These unexpected and interesting results strongly suggest that the biological role of the RND-4 and RND-9 efflux pumps might not be restricted to the sole transport of toxic compounds, but also that their function might be Orbifloxacin related to motility and/or chemotaxis. To the best of our knowledge, this is the second time that the effect of RND efflux pumps mutation on motility-related phenotypes has been described. Indeed, the absence of RND components AcrB or TolC in Salmonella enterica caused widespread repression of chemotaxis and motility genes in these mutants, and for acrB mutant this was associated with decreased motility. However, why the deletion of an efflux pump should have a fallout on bacterial motility and chemotaxis remains an open question. It is conceivable that the cytoplasmic accumulation of efflux pump-specific metabolites could act as signals triggering opposite behavioural response in the two mutants. For instance, we have recently shown that RND-4 contributes to the transport of N-acyl homoserine lactone as we found a reduced accumulation of AHLs quorum sensing signal molecules in the growth medium of D4 mutant. Actually, the D4 and D4/D9 mutant produce about 30% less AHLs than the wild-type, while D9 produces almost the same level of acyl-HSL as the wild-type. In accordance with the low impact of D4 and D9 mutations on AHLs production, only few genes known to be AHLregulated are also differentially regulated in our microarray analysis. Among these, none can be directly related to chemotaxis or biofilm formation, and only BCAL0562 and BCAL3506 could be related to flagella. Overall, these observations suggest that it is unlikely that the phenotype of the D4, D9 and D4�CD9 mutants is due to an unbalance in AHLs import/export rates. However, it cannot be ruled out that other molecules acting as metabolic signals could accumulate in the D4, D9 and D4�CD9 mutants and account for the motility and biofilm phenotypes of these strains. Another possible explanation for the biological significance of the phenotype exhibited by D4 and D4�CD9 Gomisin-D strains might rely on the assumption that: i) the bacterial cell can ”sense” the concentration of toxic compounds outside and/or inside the cell and that ii) the cell itself tries to respond to the increase of the concentration of toxic compound by activating the efflux pump systems responsible for the extrusion of that compound. Accordingly, we can speculate that in the absence of these systems, the cell might somehow bypass this defect by increasing the ability to move in the environment in order to ”escape” and to explore spaces and niches where the concentration of the toxic compounds is lower. In other words, the increased ability to move might represent a sort of ”indirect protection” of the cell towards toxic compounds. Since in many bacteria flagellum could play a role in biofilm formation, the different regulation of flagellum-related genes in D4 and D9 prompted us to speculate that these strains might also have opposite biofilm phenotypes. Therefore, we performed preliminary experiments to investigate the biofilm formation ability of the wild-type and of the three mutants. Results showed, surprisingly, that all the mutants had an enhancement of biofilm formation with respect to the wild-type. Therefore, differences in flagella expression in the D4 and D9 strains, with respect to the wildtype, play a minor role in biofilm formation, at least under our experimental conditions. The increased biofilm production of the RND-mutants was unexpected since we did not identify genes obviously involved in biofilm formation among the 33 having the same expression pattern in the three microarray experiments. Actually, biofilm formation is a complex pleiotropic phenotype, strongly dependent upon experimental conditions and growth media.