2.A.6 The Resistance-Nodulation-Cell Division (RND) Superfamily

Characterized members of the RND superfamily all probably catalyze substrate efflux via an H⁺ antiport mechanism.These proteins are found ubiquitously in bacteria, archaea and eukaryotes (see Nikaido 2018 for a review of RND transporters). They fall into ten recognized phylogenetic families, three primary phylogenetic families that are restricted largely to Gram-negative bacteria (families 1-3, see below), the SecDF family (family 4) that is represented in both Gram-negative and Gram-positive bacteria as well as archaea, the HAE2 family (family 5) that is restricted to Gram-positive bacteria, one very diverse eukaryotic family (family 6), one archaeal plus spirochete family (family 7) (Tseng et al., 1999), a recently identified family that includes a probable pigment exporter in Gram-negative bacteria (TC #2.A.6.8.1; Goel et al., 2002), Dispatched (family 9), an exporter of the amino terminal portion (19 kDa) of Hedgehog which has a C-terminal cholesterol covanlent linkage that may be recognized by Dispatched (Nikaido 2018), and the uncharacterized family 10 from Actinobacteria. For mechanistic considerations, see Murakami et al. 2020 and Simsir et al. 2020.  Additionally, the structural basis of substrate recognition of RND pumps and the molecular mechanisms underlying multidrug extrusion has been reviewed (Klenotic and Yu 2024).

Clustering patterns in the Gram-negative bacterial families of the RND superfamily correlate with substrate specificity with family 1 catalyzing export of heavy metals, family 2 catalyzing export of multiple drugs, cluster 3 probably catalyzing export of lipooligosaccharides concerned with plant nodulation for the purpose of symbiotic nitrogen fixation, cluster 4 corresponding to the SecDF proteins, and cluster 8 catalyzing pigment export. Within family 2, MdtABC, consisting of an MFP (TC #8.A.1) and two RND family proteins (MdtB [TC #2.A.6.2.12] and MdtC [TC #2.A.6.2.14]) may form a complex exhibiting broader specificity than either MdtAB or MdtAC (Baranova and Nikaido, 2002; Nagakubo et al., 2002). The ActII3 protein, one of the two partially characterized member of family 5, has been implicated in drug resistance. The MmpL7 protein, also of this family, catalyzes export of an outer membrane lipid, phthiocerol dimycocerosate (PDIM) in M. tuberculosis. The SecDF proteins (family 4) function as nonessential constituents of the IISP protein secretory system (TC #3.A.5). They seem to allow coupling of substrate protein translocation to the proton motive force by facilitating deinsertion of the SecA component of the IISP system, thus rendering this system partially ATP-independent.

Nies (2003) subdivied the HME-RND proteins (2.A.6.1) into subgroups according to substrate specificity.  HME1 (Zn²⁺, Co²⁺, Cd²⁺), HME2 (Co²⁺, Ni²⁺), HME3a (divalent cations), HME3b (monovalent cations) HME4 (Cu⁺ or Ag⁺) and HME5 (Ni²⁺).  Kim et al. (2011) have proposed two models for the extrusion of heavy metals (2.A.6.1) from the periplasm to the extracellular medium, the 'switch' and the 'funnel' mechanisms. The funnel model involves the shuttling of periplasmic substrate from the membrane fusion protein to the RND transporter and further on through the outer membrane factor to the extracellular space. Conversely, the switch model requires substrate binding to the membrane fusion protein, inducing a conformational change and creating an open-access state of the tripartite protein complex. They favor the switch mechanism.

Some or all of the eukaryotic proteins (family 6) may function in cholesterol/lipid/steroid hormone transport, reception, regulation or catalysis. One such protein complex includes the RND family disease protein, Niemann-Pick C1, which may function in the export of cholesterol and lipids from lysosomes in conjunction with a soluble lysosomal protein with cholesterol binding properties, NPC2 (TC #2.A.6.6.1; Sleat et al., 2004). The disorder is typified by inhibited egress of cholesterol and glycosphingolipids from endosomal and lysosomal compartments. In the majority of NPC patients, mutations in the NPC1 gene can be identified, but about 5% of patients show mutations in the NPC2 gene. Many different mutations can cause NPC disease, and multiple variants not associated with the disease are known in both genes. There is an NPC disease gene variation database (NPC-db; http://npc.fzk.de). Non-transporter homologues possess the sterol recognition domain and do not exhibit the typical RND family internal duplication (see below). The functions of the archaeal and spirochete proteins of family 7 have not been investigated.

Water-soluble Niemann-Pick C2 (NPC2) and membrane-bound NPC1 are cholesterol-binding lysosomal proteins required for export of lipoprotein-derived cholesterol from lysosomes. The binding site in NPC1 is located in its N-terminal domain (NTD), which projects into the lysosomal lumen. Transfer of cholesterol from NPC2 to NPC1 requires three residues that form a patch on the surface of NPC2. Wang et al. (2010) previously identified a patch of residues on the surface of NPC1(NTD) that is required for transfer. They presented a model in which these two surface patches on NPC2 and NPC1(NTD) interact, thereby opening an entry pore on NPC1(NTD) and allowing cholesterol to transfer without passing through the water phase. They referred to this transfer as a hydrophobic handoff and hypothesized that this handoff is essential for cholesterol export from lysosomes (Wang et al., 2010).

Most of the RND superfamily transport systems consist of large polypeptide chains (700-1300 amino acyl residues long). These proteins possess a single transmembrane spanner (TMS) at their N-termini followed by a large extracytoplasmic domain, then six additional TMSs, a second large extracytoplasmic domain, and five final C-terminal TMSs. In the case of one system (NolGHI) the system may consist of three distinct polypeptide chains, and most of the SecDF homologues consist of two polypeptide chains. Most others probably consist of a single polypeptide chain. The first halves of RND family proteins are homologous to the second halves, and the proteins therefore probably arose as a result of an intragenic tandem duplication event that occurred in the primordial system prior to divergence of the family members. One protein homologue from Methanococcus jannaschii is of half size and has no internal duplication. It can be postulated to function as a homo- or heterodimer in the membrane. The same is true of the eukaryotic RND family homologues that do not appear to function in transport. Some of the eukaryotic proteins have hydrophilic C-terminal domains.

Crystal structures of the RND drug exporter of E. coli, AcrB (TC #2.A.6.2.2), have been solved at 3.5 Å and 2.8 Å resolution (Murakami et al., 2002, 2006). Three AcrB protomers are organized as a homotrimer in the shape of a jellyfish. Each protomer consists of a 50 Å thick transmembrane domain and a 70 Å headpiece, protruding from the external membrane surface. The top of the headpiece opens like a funnel, and this may be a site of interaction with the MFP, AcrA (TC #8.A.1.6.1) and the OMF, TolC (TC #1.B.17.1.1). A pore formed by the three α-helices connects the funnel with a central cavity at the bottom of the headpiece. The 12 TMSs in the membrane domain are visible. Substrates are presumably successively transported through the channels of AcrB and TolC (Murakami et al., 2002). An MFP such as MexF of P. aeruginosa facilitates proper assembly of the RND permease as well as stabilization of the OMF such as OprN (Maseda et al., 2002). Vestibules are part of substrate path in AcrB multidrug efflux transporter of Escherichia coli (Husain et al., 2011). Pagès et al. (2011) have described several classes of efflux pump inhibitors that counteract MDR.

The large external cavity is of 5000 cubic angstroms. Several different hydrophobic and amphipathic ligands can bind in different positions within the cavity simultaneously. Binding involves hydrophobic forces, aromatic (π) stacking and van der Waals interactions (Yu et al., 2003). Crystallographic studies of the asymmetric trimer of AcrB suggest that each protomer in the trimeric assembly goes through a cycle of conformational changes during drug export. The external large cleft in the periplasmic domain of AcrB appears to be closed in the crystal structure of one of the three protomers. Conformational changes, including the closure of the external cleft in the periplasmic domain, are apparently required for drug transport by AcrB (Takatsuka and Nikaido, 2007; Takatsuka et al., 2010).

Murakami et al. (2006) have described crystal structures of AcrB with and without substrates. The AcrB-drug complex consists of three protomers, each of which has a different conformation corresponding to one of the three functional states of the transport cycle. Bound substrate was found in the periplasmic domain of one of the three protomers. The voluminous binding pocket is aromatic and allows multi-site binding. The structures indicate that drugs are exported by a three-step functionally rotating mechanism in which substrates undergo ordered binding change. A crystal structure at 2.9 Å resolution of trimeric AcrB was reported by Seeger et al. (2006) and shows asymmetry of the monomers. This structure reveals three different monomer conformations representing consecutive states in a transport cycle. The structural data imply an alternating access mechanism and a novel peristaltic mode of drug transport by this type of transporter.

The RND members of families 1-3 function in conjunction with a 'membrane fusion protein' (MFP; TC #8.A.1) and an 'outer membrane factor' (OMF; TC #1.B.17) to effect efflux across both membranes of the Gram-negative bacterial cell envelope in a single energy-coupled step. They may also pump hydrophobic substances from the cytoplasmic membrane, and toxic hydrophilic substances (i.e., heavy metals) from the periplasm to the external medium. The large periplasmic domains of RND pumps are involved in substrate recognition and form a cavity that can accommodate multiple drugs simultaneously (Mao et al., 2002). A comprehensive review of the classes of efflux pump inhibitors from various sources, highlighting their structure-activity relationships, which can be useful for medicinal chemists in the pursuit of novel efflux pump inhibitors, has appeared (Durães et al. 2018). Pyridylpiperazine-based allosteric inhibitors of RND-type multidrug efflux pumps have been studied (Plé et al. 2022).

Symmons et al., 2009 showed that the adaptor termini assemble a beta-roll structure forming the final domain adjacent to the inner membrane. The completed structure enabled in vivo cross-linking to map intermolecular contacts between the adaptor AcrA and the transporter AcrB, defining a periplasmic interface between several transporter subdomains and the contiguous beta-roll, beta-barrel, and lipoyl domains of the adaptor. The flexible linear topology of the adaptor allowed a multidomain docking approach to model the transporter-adaptor complex, revealing that the adaptor docks to a transporter region of comparative stability distinct from those key to the proposed rotatory pump mechanism, putative drug-binding pockets, and the binding site of inhibitory DARPins. AcrA(3)-AcrB(3)-TolC(3) is a 610 KDa, 270-A-long efflux pump crossing the entire bacterial cell envelope (Symmons et al., 2009).

RND transporters such as AcrD of E. coli can capture drugs such as aminoglycosides, from the periplasm and maybe from the cytoplasm (Aires and Nikaido, 2005). The latter process has been referred to as periplasmic vacuuming where, in this case, AcrD is the periplasmic vacuum cleaner (Lomovskaya and Totrov, 2005). This allows Gram-negative bacteria to protect themselves against cell wall biosynthetic inhibitors (drugs) that act in the periplasm. It also explains why HAE1 family members are largely restricted to Gram-negative bacteria. They are rarely found in Gram-positive bacteria or archaea.

A novel member of the RND superfamily, very distantly related to other established members of the superfamily, was shown to be a pigment (xanthomonadin) exporter in Xanthomonas oryzae (Goel et al., 2002). This protein (TC #2.A.6.8.1) has close homologues in various species of Xanthomonas as well as Xylella, Ralstonia and E. coli (AAG58596). These proteins comprise the eighth recognized family in the RND superfamily.

Protein translocation across the bacterial membrane, mediated by the secretory translocon SecYEG and the SecA ATPase, is enhanced by the proton motive force and membrane-integrated SecDF, which associates with SecYEG. The role of SecDF has been shown to function in late stages of protein secretion and membrane protein biogenesis. Tsukazaki et al. (2011) determined the crystal structure of Thermus thermophilus SecDF TC# 2.A.6.4.3) at 3.3 Å resolution, revealing a pseudo-symmetrical, 12-helix transmembrane domain belonging to the RND superfamily and two major periplasmic domains, P1 and P4. Higher-resolution analysis of the periplasmic domains suggested that P1, which binds an unfolded protein, undergoes functionally important conformational changes. In vitro analyses identified an ATP-independent step of protein translocation that requires both SecDF and the proton motive force. Electrophysiological analyses revealed that SecDF conducts protons in a manner dependent on pH and the presence of an unfolded protein, with conserved Asp and Arg residues at the transmembrane interface between SecD and SecF playing essential roles in the movements of protons and preproteins. Therefore, Tsukazaki et al. (2011) proposed that SecDF functions as a membrane-integrated chaperone, powered by the proton motive force, to achieve ATP-independent protein translocation. Furukawa et al. 2018  reported the crystal structure of SecDF in a form named the super-membrane-facing form, demonstrating a beta-barrel architecture instead of the previously reported beta-sheet structure. A remote coupling model was proposed in which a structural change of the transmembrane region drives a functional, extracytoplasmic conformational transition.

AcrB actively exports a wide variety of noxious compounds using the proton-motive force for energy. AcrB adopts an asymmetric structure of three protomers with different conformations that are sequentially converted during drug export; these cyclic conformational changes during drug export are referred to as functional rotation. Using different protonation states for the titratable residues in the middle of the transmembrane domain, simulations revealed a correlation between the specific protonation states and the side-chain configurations. Changing the protonation state for Asp408 induced a spontaneous structural transition, which suggests that the proton translocation stoichiometry may be one proton per functional rotation cycle.  Simulations also demonstrate that alternating the protonation states in the transmembrane domain induces functional rotation in the porter domain, which is coupled to drug transport (Yamane et al. 2013).  A mechanism involving two remote alternating-access conformational cycles within each protomer, namely one for protons in the transmembrane region and another for drugs in the periplasmic domain, 50 A apart, has been proposed (Eicher et al. 2014). Each of these cycles entails two distinct types of collective motions of two structural repeats, coupled by flanking α-helices that project from the membrane. Cross-talk among protomers across the trimerization interface might lead to a more kinetically efficient efflux system.

The generalized transport reaction catalyzed by functionally characterized RND proteins is:

Substrates (in) + nH⁺ (out) → Substrates (out) + nH⁺ (in).

Substrates: (a) heavy metals, (e.g., Co²⁺, Zn²⁺, Cd²⁺, Ni²⁺, Cu⁺ and Ag⁺; family 1); (b) multiple drugs (e.g., tetracycline, chloramphenicol, fluoroquinolones, β-lactams, etc.; family 2); (c) lipooligosaccharides (nodulation factors; family 3); (d) unfolded proteins as for the SecDF-mediated translocation of substrate proteins (Family 4), (e) lipids and possibly antibiotic drugs in Gram positive bacteria (e.g., outer membrane mycolic acid-containing lipids in actinobacteria and  actinorhodin; family 5), (f) possibly sterols in eukaryotes (family 6), (g) fused pentacyclic ring compounds such a hopanoids in bacteria (family 7), (h) pigments (family 8), and (i) cholesterol-modified peptides such as 'hedgehog', a sterol sensor in animals (family 9).