In the tripartite

protein complex, MexB is the inner membrane protein and a member of the resistance–nodulation–division (RND) family, MexA is a membrane fusion protein and OprM is an outer membrane protein. Although all three proteins in the complex are necessary for drug efflux from P. aeruginosa, the substrate specificity of the complex is mediated by MexB. MexB recognizes a wide variety of chemically different compounds including antibiotics, Small molecule library cell assay detergents, dyes and molecules involved in quorum sensing (Poole, 2001). MexB bears a close resemblance to its counterpart from Escherichia coli, AcrB (70% identity), and can also functionally substitute for AcrB in the AcrAB-TolC complex (Krishnamoorthy et al., 2008; Welch et al., 2010). Recently, the crystal structure of MexB has 3-MA been solved and it was found to be an asymmetric homotrimer similar to AcrB (Sennhauser et al., 2009). Each monomer of MexB consists of 12 transmembrane α-helices constituting the inner membrane domain and a large periplasmic domain (Sennhauser et al., 2009). The periplasmic domains of the RND family of drug transporter proteins are implicated in drug recognition and transport (Elkins & Nikaido, 2002; Mao et al., 2002; Tikhonova et al., 2002; Middlemiss & Poole,

2004; Murakami et al., 2006; Seeger et al., 2006; Bohnert et al., 2007; Dastidar et al., 2007; Sennhauser & Grutter, 2008; Takatsuka et al., 2010; Nakashima et al., 2011). Based upon the asymmetric structures of the AcrB trimers, a

substrate pathway through the periplasmic domains of the individual subunits has been proposed as an alternative access mechanism with the protomers adopting binding, access and extrusion conformations, respectively (Murakami et al., 2006; Seeger et al., 2006; Sennhauser & Grutter, 2008). Recent biochemical studies have confirmed the peristaltic pump mechanism of transport (Seeger et al., 2008; Takatsuka & Nikaido, 2009), while structural, functional and computational analyses yielded an insight into the entire substrate path through the periplasmic domain of AcrB (Husain & Nikaido, 2010; Schulz et al., 2010, 2011; Yao et al., 2010; Nakashima et al., 2011). Although the drug efflux pathway through the periplasmic MRIP domains of AcrB has now been very well established and characterized, the question still remains if all drugs are effluxed from the periplasm or if substrates could also be removed directly from the cytoplasm/inner cytoplasmic membrane. In MexB and the related RND transporter MexD, mutations affecting resistance against drugs mapped to periplasmic domains affected both periplasmically and cytoplasmically acting antibiotics; therefore, the authors concluded that there are no separate binding sites for antimicrobials with periplasmic vs. cytoplasmic targets (Mao et al., 2002; Middlemiss & Poole, 2004).