Evolutionary origins of multidrug and drug-specific efflux pumps in bacteria

^a Department of Biology, University of California at San Diego, La Jolla, California 92093–0116, USA
^b School of Biological Sciences, University of Sydney, NSW 2006, Australia
^c Department of Molecular and Cell Biology, University of California, Berkeley, California 94720–3206, USA

	ABSTRACT

TOP ABSTRACT INTRODUCTION DRUG RESISTANCE PUMPS ENCODED... THE ATP BINDING CASSETTE... THE MAJOR FACILITATOR... THE SMALL MULTIDRUG RESISTANCE... THE RESISTANCE-NODULATION-CELL... POSSIBLE MECHANISMS OF DRUG... CONCLUSIONS AND PERSPECTIVES REFERENCES

 
The available genomic sequences of three pathogenic and three nonpathogenic bacteria were analyzed to identify known and putative drug-specific and multidrug resistance transport systems. Escherichia coli was found to encode 29 such pumps, and with the exception of the archaebacterium Methanococcus jannaschii, the numbers of multidrug efflux pumps encoded within genomes of the other organisms were found to be approximately proportional to their total numbers of encoded transport systems as well as to total genome size. The similar numbers of chromosomally encoded multidrug efflux systems in pathogens and nonpathogens suggests that these transporters have not arisen recently in pathogens in response to antimicrobial chemotherapy. Phylogenetic analyses of the four transporter families that contain drug efflux permeases indicate that drug resistance arose rarely during the evolution of each family and that the diversity of current drug efflux pumps within each family arose from just one or a very few primordial systems. However, although the ability to confer drug efflux appears to have emerged on only a few occasions in evolutionary time and was stably maintained as an evolutionary trait, modulation of the substrate specificities of these systems has occurred repeatedly. A speculative model is presented that may explain the apparent capability of these multidrug transport systems to mediate drug transport from the cytoplasm or directly from the phospholipid bilayer.—Saier, M. H., Jr., Paulsen, I. T., Sliwinski, M. K., Pao, S. S., Skurray, R. A., Nikaido, H. Evolutionary origins of multidrug and drug-specific efflux pumps in bacteria. FASEB J. 12, 265–274 (1998)

Key Words: antimicrobial agents • drugs • antibiotics • resistance • transport • evolution • phylogenetic trees

	INTRODUCTION

TOP ABSTRACT INTRODUCTION DRUG RESISTANCE PUMPS ENCODED... THE ATP BINDING CASSETTE... THE MAJOR FACILITATOR... THE SMALL MULTIDRUG RESISTANCE... THE RESISTANCE-NODULATION-CELL... POSSIBLE MECHANISMS OF DRUG... CONCLUSIONS AND PERSPECTIVES REFERENCES

 
Living organisms communicate with their environments in part via solute-specific transport systems. These systems consist of integral membrane proteins that usually span the cytoplasmic membrane of the cell multiple times as -helices. These proteins form solute-specific channels or transport pathways (1, 2). Energy coupling proteins, localized to the cytoplasmic surface of the membrane, and extracytoplasmic receptors, tethered to or localized to the external surface of the membrane, may be superimposed on the transporters to provide increased pumping capacity and high-affinity solute recognition, respectively (1, 3).

Transport proteins have been extensively studied from structural, functional, and phylogenetic standpoints and have been found to be derived from several distinct primordial sources (2, 4). Consequently, they comprise families of proteins, some of which are found ubiquitously throughout the living world; others are restricted to one of the major living kingdoms. Nearly 200 families of transport proteins have been identified, and about 100 of these have been detected in bacteria (2, 4, 5; M. H. Saier, Jr., unpublished observations).

During the past two decades, drug-specific and multidrug efflux pumps have emerged as a major problem in medicine (6, 7). Indeed, pathogenic bacteria are proving to exhibit increasing degrees of resistance to multiple drugs, presumably in response to the common use of antibiotics and other antimicrobial agents for the control of diseases associated with these bacteria (8). Due in large measure to the occurrence of multidrug resistance (MDR)² efflux pumps and to the fact that novel classes of antimicrobial agents have not been discovered or developed in recent years (9), we face the frightening probability that many, if not all, pathogenic bacteria that threaten human health will soon be resistant to all known antibiotics (7). There is therefore a need to understand the structures, functions, and origins of these transport systems.

We can pose several important questions with respect to bacterial drug resistance. For example, 1) how many drug resistance pumps are encoded within the genome of a particular bacterium, and 2) to what transport families do they belong? 3) Do all bacteria encode similar types of drug resistance pumps in their genomes, or is the distribution of the various types of pumps species-specific? 4) Did drug efflux pumps arise recently in response to drug use, or have they been encoded within bacterial genomes for hundreds of millions or even billions of years? 5) During the evolution of a family of transport proteins, did drug efflux pumps appear multiple times or did they arise just once or a few times? 6) Did drug efflux pumps readily expand and restrict their substrate specificities during the evolution of a family, or is substrate specificity a stable characteristic? These questions and their answers provide the focus of this review.

	DRUG RESISTANCE PUMPS ENCODED WITHIN BACTERIAL GENOMES

TOP ABSTRACT INTRODUCTION DRUG RESISTANCE PUMPS ENCODED... THE ATP BINDING CASSETTE... THE MAJOR FACILITATOR... THE SMALL MULTIDRUG RESISTANCE... THE RESISTANCE-NODULATION-CELL... POSSIBLE MECHANISMS OF DRUG... CONCLUSIONS AND PERSPECTIVES REFERENCES

 
In our initial studies (5), we identified all known and putative transport proteins from Escherichia coli, Haemophilus influenzae, and Mycoplasma genitalium, three pathogenic bacteria for which complete genome sequence data are available (10–12). We similarly examined three nonpathogenic organisms including Bacillus subtilis (45% sequenced), Methanococcus janneschii (100% sequenced), and Synechocystis PCC8603 (100% sequenced) (13–15). The identified transport proteins were classified according to topology, protein family, substrate specificity, bioenergetics, and distribution of homologues in other organisms (2, 4). When possible, phylogeny was correlated with function, and crypticity was sometimes evaluated. Auxilliary proteins, necessary for full function, were identified in some cases.

Bacterial drug efflux pumps proved to occur in four families. Two of these are large and ancient superfamilies known as the ATP binding cassette (ABC) superfamily (16) and the major facilitator superfamily (MFS) (17, 18). The other two are smaller and more recently developing families, called the small multidrug resistance (SMR) family (19) and the resistance-nodulation-cell division (RND) family (20, 21).

Table 1 summarizes the occurrence of known and putative multidrug efflux systems in the six bacteria for which extensive sequence data were available at the time of our analyses. In E. coli, 29 proven and putative drug pumps were identified. Of these 29 systems, 9 have been shown to be capable of pumping drugs out of E. coli cells when expressed at high levels, although most of these systems are probably expressed in wild-type E. coli K12 at levels that are insufficient to cause substantial drug resistance (8, 22). The `putative' drug pumps are suggested to be drug efflux systems because they share a high degree of sequence similarity with the established drug efflux pumps. This proposal seems justified, since all functionally characterized members of these phylogenetic clusters exclusively catalyze efflux of drugs and other toxic substances.

H. influenzae and M. genitalium possess only six and two putative chromosomally encoded drug efflux pumps, respectively. These numbers (29, 6, and 2) are approximately proportional to genome size (4.5, 1.8, and 0.46 Mb, respectively). They are also roughly proportional to the total numbers of transport systems identified in these three organisms (5).

Comparisons of the nonpathogenic organisms B. subtilis, M. jannaschii, and Synechocystis PCC6803 with the pathogenic bacteria discussed above revealed that the former organisms possess similar numbers of multidrug efflux proteins relative to the total number of encoded transporters ( Table 1). The numbers of multidrug pumps and the total number of transporters were also found to be approximately proportional to genome size in B. subtilis. However, M. jannaschii and Synechocystis PCC6803 proved to possess twofold fewer transport systems relative to genome size, possibly due to their adaptation to survival in organic nutrient-poor, mineral-rich environments. The fact that pathogenic and nonpathogenic bacteria exhibit comparable numbers of chromosomally encoded multidrug resistance efflux systems argues against the tenet that these systems have arisen recently in pathogens as a result of extensive exposure to medically relevant drugs. Instead, they may play important physiological roles in the extrusion of naturally occurring toxic substances.

Among the 29 putative drug pumps identified in E. coli, 2 are of the ABC type, 18 are members of the MFS, 5 belong to the RND family, and 4 belong to the SMR family ( Table 1). About the same distribution was observed for H. influenzae, a fairly close relative of E. coli with similar metabolic capacity. By contrast, the two putative drug efflux pumps in M. genitalium proved to be members of the ABC superfamily ( Table 1), as are a large percentage of its solute permeases. The latter findings agree with the following: ATP is the energy source that drives efflux of drugs via ABC permeases; M. genitalium, unlike E. coli and H. influenzae, lacks an electron transport chain, and only M. genitalium cannot generate a proton electrochemical gradient as a primary source of energy (11). It exclusively utilizes substrate-level phosphorylation during glycolytic sugar metabolism in order to generate energy.

Examination of the three nonpathogenic bacteria under study reveals divergent patterns. B. subtilis has a complement of MDR transporters similar to that of E. coli. Its genome encodes only one RND-type system, systems that are largely restricted to the Gram-negative bacterial kingdom. The absence of RND and SMR pumps in M. jannaschii is expected, since these systems have so far been identified only in eubacteria. Synechocystis exhibits a majority of ABC-type systems, with a surprisingly skimpy complement of MFS systems. In fact, it is the only organism examined to possess more RND than MFS systems.

Functioning in conjunction with MFS, ABC, and RND transport proteins, but not with SMR proteins, two families of auxillary proteins have been identified. These two auxillary protein families have been designated the membrane fusion protein family and the outer-membrane factor family based on their cellular locations in the Gram-negative bacterial envelope and their presumed functions (see refs 21, 23–26 for recent reviews). These proteins are believed to allow transport across both membranes of the Gram-negative bacterial cell envelope in a single energy-coupled step (23, 25). They will not be the focus of this discussion.

	THE ATP BINDING CASSETTE SUPERFAMILY

TOP ABSTRACT INTRODUCTION DRUG RESISTANCE PUMPS ENCODED... THE ATP BINDING CASSETTE... THE MAJOR FACILITATOR... THE SMALL MULTIDRUG RESISTANCE... THE RESISTANCE-NODULATION-CELL... POSSIBLE MECHANISMS OF DRUG... CONCLUSIONS AND PERSPECTIVES REFERENCES

 
Properties of the ABC superfamily are summarized in Table 2A. This superfamily consists of more than 30 families, each specific for one of a tremendous variety of substrates. These substrates include small molecules that may be taken up or expelled from the cell, depending on the transporter, and also macromolecules such as proteins and complex carbohydrates that are synthesized in the cytoplasm and secreted to the cell envelope or the external milieu. For all bacterial ABC transport systems, ATP provides the source of energy that drives transport.

ABC-type efflux transporters generally consist of 1) an integral membrane protein with six putative transmembrane -helical spanners, and 2) an energy-coupling protein localized to the cytoplasmic side of the membrane (27, 28). The latter protein may be either noncovalently associated with the former protein or covalently attached to it in a single polypeptide chain ( Fig. 1, left). These proteins are usually present as dimeric complexes, and the complete prototypical system therefore possesses 12 spanners. ABC transporters occur universally throughout the living world, and more than 300 members of the family have been sequenced. Well-characterized members include the E. coli maltose uptake permease, the mammalian multidrug resistance efflux pumps, and the human cystic fibrosis protein (16, 27, 31).

View larger version (64K): [in this window] [in a new window]   Figure 1. Generalized structural model for ATP binding cassette-2 (ABC-2) family permeases (left) and phylogenetic tree for integral membrane proteins of the ABC-2 family (right). Methods of tree construction and protein abbreviations are presented in Feng and Doolittle (29) and in Paulsen et al. (30). Branch length (provided in arbitrary units) is approximately proportional to phylogenetic distance (29). Substrate specificities of the proteins within the major clusters are as follows: A) exopolysaccharides of Gram-positive and Gram-negative bacteria; B) lipopolysaccharides of Gram-negative bacteria; TagG, teichoic acids of the Gram-positive bacterium, Bacillus subtilis; C) lipooligosaccharides of nodulating Gram-negative bacteria; D, drugs synthesized by Gram-positive bacteria. All of the functionally characterized drug resistance permeases in the ABC-2 family can be found in cluster D. The left panel was modeled after the DrrA2-B2 system of Streptomyces peucetins. Other drug resistance pumps included in cluster D are the OrrB1 and DrrC Daunorubicin resistance pumps of Mycobacterium leprae and the OleC5 Oleandomycin resistance pump of Streptomyces antibioticus. Names, substrate specificities, and organismal sources of the other proteins analyzed are presented in Paulsen et al. (30).

Phylogenetic analyses of one of the families of the ABC superfamily, termed the ABC-2 family, proved particularly informative (30, 32). Most members of this family include proteins that catalyze the export of cell surface carbohydrates synthesized within the bacterial cell. Separate phylogenetic trees were constructed for the integral membrane proteins and for the ATP-hydrolyzing, energy-coupling proteins (30). The clustering patterns for the two trees were similar, suggesting that little shuffling of constituent proteins between systems had occurred during the family's evolution.

The tree for the integral membrane proteins of the ABC-2 permeases is shown in Fig. 1, right. Each major cluster proved to be specific for a different type of substrate. Thus, all members of cluster A (see Fig. 1) catalyze capsular polysaccharide export; members of cluster B catalyze lipopolysaccharide export; members of cluster C catalyze lipooligosaccharide export; and members of cluster D apparently catalyze drug-specific efflux. It would appear that substrate specificity has been a well-conserved trait during evolution of the ABC-2 family. Moreover, drug resistance apparently evolved only once during the early evolution of this ABC subfamily, and all members of the family that catalyze drug resistance were therefore probably derived from a single primordial permease. Whether these ABC-2 drug resistance pumps are drug-specific or capable of transporting multiple drugs has not been tested.

In addition to the drug resistance permeases that fall within the ABC-2 family of the ABC superfamily, ABC-type MDR or MDR-like pumps have been found in bacteria (33–36). Preliminary evidence suggests that these bacterial ABC-type MDR pumps resemble mammalian MDR pumps. They apparently fall within one or a few closely related, closely allied families of the ABC superfamily (E. Dassa, personal communication). It therefore seems that drug efflux pumps have evolved only occasionally, but more than once, during evolution of the ABC superfamily. A detailed understanding of this important group of proteins must await further phylogenetic analysis.

	THE MAJOR FACILITATOR SUPERFAMILY

TOP ABSTRACT INTRODUCTION DRUG RESISTANCE PUMPS ENCODED... THE ATP BINDING CASSETTE... THE MAJOR FACILITATOR... THE SMALL MULTIDRUG RESISTANCE... THE RESISTANCE-NODULATION-CELL... POSSIBLE MECHANISMS OF DRUG... CONCLUSIONS AND PERSPECTIVES REFERENCES

 
Characteristics of the major facilitator superfamily are presented in Table 2B. Like the ABC superfamily, the MFS is an ancient superfamily that probably dates back through evolutionary time more than 3 billion years. It consists of more than 300 sequenced proteins that fall into 17 currently recognized, distantly related families, each in general specific for a different type of solute ( Fig. 2, bottom; 37).

View larger version (39K): [in this window] [in a new window]   Figure 2. Structural model for the 12-transmembrane segment (TMS) (A) and the 14-TMS drug efflux pumps of the MFS (top) and phylogenetic tree for representative proteins of the major facilitator superfamily (MFS) (bottom) (B). The tree was constructed according to Feng and Doolittle (29), as described in Pao et al. (37). The non-drug transporting families depicted are the oligosaccharide:H+ symporter (OHS) family, which includes the LacY lactose permease of E. coli (lower left); the organophosphate:phosphate antiporter (OPA) family, which includes the UhpT hexose-P permease of E. coli (upper left); the sugar porter (SP) family, which includes the AraE arabinose permease of E. coli. (upper right); and the metabolite:H+ symporter (MHS) family, which includes the citrate permease (CitA) of Salmonella typhimurium. The 12-TMS drug:H+ antiporter family (DHA12; bottom right) includes the TetB tetracycline resistance pump of E. coli, the Bmr multidrug resistance pump of Bacillus subtilis, and the VMAT1 multidrug resistance pump of rats. The 14-TMS drug:H+ antiporter family (DHA14, bottom left) includes the EmrB multidrug resistance pump of E. coli, the QacA multidrug resistance pump of Staphylococcus aureus, the Sge1 multidrug resistance pump of Saccharomyces cerevisiae, and the Tet tetracycline resistance pump of Bacillus subtilis. Other protein abbreviations are as described in Pao et al. (37) and in Paulsen et al. (21). As noted above, the clustering pattern generally correlates with substrate specificity. Thus, all recognized drug resistance permeases cluster together, with the 12-TMS drug efflux permeases and the 14-TMS drug efflux permeases clustering separately. The tree depicts 6 of the 17 currently recognized families within the MFS (37). The top panels were modeled after the TetB protein of E. coli (12 TMS) and the QacA protein of Staphylococcus aureus (14 TMS), both of established topology (see text).

These 17 families include 4 families specific for various types of sugars, a fifth that catalyzes uptake of phosphorylated glycolytic intermediates, a sixth that catalyzes uptake of Krebs cycle intermediates and other metabolites, 2 families that catalyze drug efflux, and several that transport organic and inorganic anions.

The two large drug resistance families within the MFS are topologically distinguishable in that one possesses 12 putative or established transmembrane spanners, like most other members of the MFS (38), whereas the other possesses 14 spanners (21, 39, 40) ( Fig. 2, top). Sequence analyses suggested that the 14-spanner drug resistance proteins arose from a primordial 12-spanner drug resistance protein early in the evolution of these two drug resistance families, before duplication and divergence of the genes that encode their many members (21; Fig. 2, bottom). Moreover, the gene segment that gave rise to the primordial 14-spanner protein from the primordial 12-spanner protein proved to have been inserted in the middle of the 12-spanner-encoding gene—between the two halves of the primordial gene—giving rise to spanners 7 and 8 in the 14-spanner protein (17, 39).

Drug-specific and multidrug efflux pumps are found interspersed fairly randomly on the phylogenetic tree for both the 12- and 14-spanner drug resistance permease families (6, 19), suggesting that once drug resistance arose, broadening and narrowing of the specificities of these systems have occurred repeatedly.

Recently, a third family in the MFS, the sugar porter family (37), has been shown to include eukaryotic multidrug transporters (21, 41). Still a fourth family in the MFS has been proposed to transport drugs, but no member of this family has yet been functionally characterized (37, 42). These facts suggest that, as for the ABC superfamily, drug resistance arose during the evolution of the MFS only a very few times. Most of the dozens of currently recognized bacterial drug resistance permeases in the MFS probably were derived from a single primordial system.

	THE SMALL MULTIDRUG RESISTANCE (SMR) FAMILY

TOP ABSTRACT INTRODUCTION DRUG RESISTANCE PUMPS ENCODED... THE ATP BINDING CASSETTE... THE MAJOR FACILITATOR... THE SMALL MULTIDRUG RESISTANCE... THE RESISTANCE-NODULATION-CELL... POSSIBLE MECHANISMS OF DRUG... CONCLUSIONS AND PERSPECTIVES REFERENCES

 
Recently, a novel family of proteins, some members of which catalyze multidrug resistance, has been described ( Table 2C; 19, 43). Subunits of these presumed homo-oligomeric permeases are unusually small. Their polypeptide chains are 100 to 110 amino acids in length, and they span the membrane four times as putative -helices ( Fig. 3, top; 19, 44, 45). Their native state in the membrane is probably that of a homotrimer. They are so hydrophobic that, unlike most membrane proteins, they are soluble in organic solvents (46, 47). Only 10 members of this small, well-conserved family have been sequenced to date, and all are from bacteria (19, 21).

View larger version (33K): [in this window] [in a new window]   Figure 3. Structural model (top) and phylogenetic tree (bottom) for members of the small multidrug resistance (SMR) family. Methods of tree construction and abbreviations of the proteins are provided in Paulsen et al. (19). All known drug resistance permeases cluster together on the right-hand side of the tree. These proteins are the QacE protein of Klebsiella pneumoniae, the EmrE protein of E. coli, and the Smr protein of Staphylococcus aureus. The specificities and functions of the Sug proteins are not known. The top figure was modeled after the Smr protein of Staphylococcus aureus. See Paulsen et al. (21) for substrate specificities of the proteins.

The SMR family consists of two phylogenetic subfamilies ( Fig. 3, bottom). Members of one subfamily (right-hand side of the tree in Fig. 3) all confer multidrug resistance and catalyze drug efflux via a drug:H⁺ antiport mechanism, as do the corresponding MFS drug resistance proteins. However, members of the other subfamily apparently do not confer drug resistance or catalyze drug:H⁺ antiport. The natural substrates of this second subfamily have not yet been identified (19).

Subdivision of the SMR family into two phylogenetic clusters and the observation that the members of only one of these clusters apparently catalyze drug extrusion (see Fig. 3) argue strongly that, for the SMR family, drug resistance permeases arose only once during its evolutionary history. The same was noted above for the ABC-2 family, and possibly also for the MFS. We must therefore conclude that for all three families, substrate specificity is a well-conserved evolutionary trait. In the next section, we shall see that the same is true for the RND family. Phylogenetic tree construction evidently provides a clear indication of the evolutionary process that gave rise to proteins exhibiting differing substrate specificities and functions (4, 48).

	THE RESISTANCE-NODULATION-CELL DIVISION FAMILY

TOP ABSTRACT INTRODUCTION DRUG RESISTANCE PUMPS ENCODED... THE ATP BINDING CASSETTE... THE MAJOR FACILITATOR... THE SMALL MULTIDRUG RESISTANCE... THE RESISTANCE-NODULATION-CELL... POSSIBLE MECHANISMS OF DRUG... CONCLUSIONS AND PERSPECTIVES REFERENCES

 
Like the SMR family, the RND family is a small, bacterial-specific family, some members of which catalyze drug:H⁺ antiport ( Table 2D; 20). However, in contrast to the SMR permease subunits, subunits of RND permeases are large proteins, generally of more than 1000 amino acyl residues. They possess an unusual putative topology characteristic of the family ( Fig. 4, top). At the amino-terminal end of each such protein, the polypeptide chain probably traverses the cytoplasmic membrane once from cytoplasm to periplasm, and this spanner is followed by a large water-soluble domain localized to the periplasmic or extracytoplasmic space. The polypeptide chain then spans the membrane six more times before it again emerges into the periplasm as another water-soluble domain of the same size as the first one. The carboxyl-terminal end of the permease is again embedded in the membrane with five additional spanners. Thus, each permease has 12 putative spanners as well as 2 large, presumably extracytoplasmic domains. Sequence analyses have shown that the first halves of these polypeptide chains are homologous to the second halves, and they therefore probably arose as a result of a tandem, intragenic duplication event (20).

View larger version (31K): [in this window] [in a new window]   Figure 4. Structural model (top) and phylogenetic tree (bottom) for representative members of the resistance-nodulation-cell division (RND) family. Methods of tree construction and protein abbreviations are provided in Saier et al. (20) and Paulsen et al. (21). The three clusters correlate with substrate specificity as follows: upper right-hand side, multiple drugs; left-hand side (CzcA, NccA, and CnrA), heavy metal ions; lower left (NolFGH), lipooligosaccharides. Of the established multidrug resistance RND-type pumps, MtrD is from Neisseria gonorrhoeae, MexB is from Pseudomonas aeruginosa, and all other proteins (AcrB, D, and F) are from E. coli. The top figure was modeled after the AcrB protein of E. coli. See Paulsen et al. (21) for substrate specificities of the proteins.

Phylogenetic analyses of the members of the RND family revealed that these proteins fall into three subfamilies that cluster on the phylogenetic tree in accordance with function ( Fig. 4, bottom). Thus, members of one subfamily are specific for divalent heavy metal ions (left-hand side of the tree shown in Fig. 4); those of the second are probably specific for lipooligosaccharides (lower right-hand side of the tree shown in Fig. 4; a single putative three-component transporter, NolFGH is represented), and those of the third subfamily all catalyze efflux of multiple drugs (upper right-hand side in the tree shown in Fig. 4; 20, 21). The current drug resistance members of the RND family probably all arose from a single primordial drug resistance protein as noted above for the ABC-2, MF, and SMR families.

	POSSIBLE MECHANISMS OF DRUG EXPORT

TOP ABSTRACT INTRODUCTION DRUG RESISTANCE PUMPS ENCODED... THE ATP BINDING CASSETTE... THE MAJOR FACILITATOR... THE SMALL MULTIDRUG RESISTANCE... THE RESISTANCE-NODULATION-CELL... POSSIBLE MECHANISMS OF DRUG... CONCLUSIONS AND PERSPECTIVES REFERENCES

 
Although the four recognized families of transport proteins that include members that extrude drugs from living cells arose independently from different primordial sources at different times in evolutionary history, no evidence is currently available to suggest that the basic mechanisms of transmembrane substrate translocation are different. However, substantial evidence suggests that both for certain ABC-type drug efflux pumps (33, 34) and for selected RND-type drug efflux pumps (25), hydrophobic and amphipathic substrate molecules may enter the channel of the permease from either the inner or the outer leaflet of the cytoplasmic membrane phospholipid bilayer and be extruded to the extracytoplasmic milieu (25). Indeed, homologous murine ABC-type MDR-2 transporters are believed to function as phospholipid flippases, catalyzing movement of phospholipids from the inner leaflet to the outer leaflet of the bilayer (49–51). Since these proteins are homologous to other ABC-type transporters that clearly catalyze efflux of molecules from the cell cytoplasm to the extracellular milieu, we must conclude that the 3-dimensional architectures of the former and latter exporters must be similar (52). The question therefore arises as to how transport proteins of similar 3-dimensional structures, driven either by ATP hydrolysis or by transmembrane ion translocation, can catalyze molecular efflux from either the cytoplasm, the inner leaflet of the phospholipid bilayer, or the outer leaflet of this bilayer.

In considering this interesting question, an analogy can possibly be drawn with the protein secretory (Sec) apparatus of E. coli and other organisms (53–55). The SecYEG complex of E. coli or the homologous Sec61 element of the eukaryotic endoplasmic reticulum can catalyze both secretion of water-soluble polypeptide chains across the cytoplasmic membrane and insertion of hydrophobic integral membrane proteins into the membrane. Thus, when a hydrophobic spanner within an integral membrane protein becomes properly aligned in the transmembrane channel of the Sec complex, release must presumably be effected by lateral diffusion of the polypeptide chain from the amphipathic channel of the protein complex to the hydrophobic milieu of the phospholipid bilayer. Indeed, early contact of the hydrophobic segment of the translocated polypeptide chain both with protein and with phospholipids has been demonstrated (56).

We suggest that some multidrug efflux pumps, possibly like Sec protein export apparatuses, may exhibit reversible intramembranous domain associations that allow transient lateral channel opening and closing. Indeed, transient channel formation due to the presence of the Sec apparatus in a phospholipid bilayer has been demonstrated (55; T. A. Rapoport, personal communication). A `breathing' motion of the protein might allow the channel to open and close laterally, thus allowing passage of either hydrophobic peptides (in the case of the Sec complex) or hydrophobic drugs (in the case of an MDR pump) to exit from the channel into the phospholipid bilayer. Further, a flexibly amphipathic nature of the channel might allow a hydrophobic or amphipathic detergent or drug (as for an amphipathic phospholipid) to `flip' from one leaflet of the bilayer to the other. Such a postulate could account for the flexibility of MDR pumps to export drugs from the cytoplasm, the inner leaflet of the cytoplasmic membrane, and the outer leaflet of the membrane in equivalent energy-coupled steps.

	CONCLUSIONS AND PERSPECTIVES

TOP ABSTRACT INTRODUCTION DRUG RESISTANCE PUMPS ENCODED... THE ATP BINDING CASSETTE... THE MAJOR FACILITATOR... THE SMALL MULTIDRUG RESISTANCE... THE RESISTANCE-NODULATION-CELL... POSSIBLE MECHANISMS OF DRUG... CONCLUSIONS AND PERSPECTIVES REFERENCES

 
In eukaryotes, all well-characterized MDR-type efflux pumps fall into two transport protein families: those of the ATP binding cassette (ABC) superfamily and those of the major facilitator superfamily (MFS). These two families are large and ancient, and are believed to date back in evolutionary time more than 3 billion years. Members of both families are distributed in all three kingdoms of living organisms (4, 20, 37, 57). In prokaryotes, MDR pumps are members not only of these two permease families, but also of two recently discovered families: the small multidrug resistance (SMR) and the resistance-nodulation-cell division (RND) families (19). Permeases of these two families have so far been identified only in prokaryotes and prove to be far less diverse in sequence than proteins of the MFS or the ABC superfamily. We suggest that they arose in bacteria after the eukaryotic and archaeal phyla split off from the bacterial phylum (57). The absence of SMR and RND proteins encoded within the complete genomes of Methanococcus jannaschii and Saccharomyces cerevisiae, and the substantial amounts of additional archaeal and eukaryotic DNA sequenced, substantiate this suggestion.

Each permease family found to include members that confer drug resistance possesses only one subfamily (or a few at most) with recognized members that catalyze drug efflux. Other subfamilies function in other transport capacities, usually either bringing nutrients into the cell or exporting biosynthetic macromolecules. We therefore suggest that in each family, specificity for drugs evolved only once, or just a few times at most. The tremendous diversity of drug resistance permeases that has evolved from each such primordial system has resulted 1) from extensive gene duplication events within a single organism, 2) as a consequence of horizontal transfer of genetic material between organisms, and 3) during speciation. Although specificity for a given class of compounds appears to be a relatively stable evolutionary trait (e.g., sugars vs. drugs vs. phosphorylated compounds in the MFS; heavy metals vs. drugs vs. lipooligosaccharides in the RND family), the apparent facile interconversion of drug-specific and multidrug transporters in the MFS is worthy of note. We infer from this observation and others that the broadening and narrowing of transporter specificity, at least as applied to MFS permeases, occurred frequently during evolutionary history, although a shift in specificity between classes of compounds occurred seldom. It is equally noteworthy that the SMR and RND families currently include only MDR-type efflux pumps. Drug-specific permeases apparently are lacking in these two families. Perhaps the nature of their channel construction differs from that of MFS or ABC permeases and precludes the acquisition of intramembranous stereospecific drug recognition.

The observations summarized in this review clearly suggest that phylogenetic analysis provides a valid but restrictive guide to function. They also provide clues concerning the existence of fundamental functional differences between independently evolving permease families. The structure–function relationships that dictate these differences represent an intriguing new area of inquiry.

	ACKNOWLEDGMENTS

 
We thank Alison M. Beness for assistance with the computational analyses reported and preparation of the figures. Mary Beth Hiller provided invaluable assistance in the preparation of this manuscript. I.T.P. was a recipient of a C. J. Martin Fellowship from the National Health and Medical Research Council of Australia. This work was supported by USPHS grant 2RO1 AI14176 from the National Institute of Allergy and Infectious Diseases and grant RO1 GM55434 from the National Institute of General Medical Science.

	FOOTNOTES

 
¹ Correspondence: Department of Biology, 0116, University of California at San Diego, 9500 Gilman Dr., La Jolla, CA 92093–0116, USA. E-mail: msaier{at}ucsd.edu

² Abbreviations: MDR, multidrug resistance; SMR, small multidrug resistance; ABC, ATP binding cassette; MFS, major facilitator superfamily; RND, resistance-nodulation-cell division; Sec, secretory.

	REFERENCES

TOP ABSTRACT INTRODUCTION DRUG RESISTANCE PUMPS ENCODED... THE ATP BINDING CASSETTE... THE MAJOR FACILITATOR... THE SMALL MULTIDRUG RESISTANCE... THE RESISTANCE-NODULATION-CELL... POSSIBLE MECHANISMS OF DRUG... CONCLUSIONS AND PERSPECTIVES REFERENCES

 

1. Nikaido, H., and Saier, M. H., Jr. (1992) . Transport proteins in bacteria: Common themes in their design. Science 258, 936–942[Abstract/Free Full Text]
2. Saier, M. H., Jr. (1994) Computer-aided analyses of transport protein sequences: Gleaning evidence concerning function, structure, biogenesis, and evolution. Microbiol. Rev. 58, 71–93[Abstract/Free Full Text]
3. Tam, R., and Saier, M. H., Jr. (1993) Structural, functional, and evolutionary relationships among extracellular solute-binding receptors of bacteria. Microbiol. Rev. 57, 320–346[Abstract/Free Full Text]
4. Saier, M. H., Jr. (1996) Phylogenetic approaches to the identification and characterization of protein families and superfamilies. Microb. Comp. Genomics 1, 129–150[Medline]
5. Paulsen, I. T., Sliwinski, M. K., and Saier, M. H., Jr. (1998) Microbial genome analyses: Global comparisons of transport capabilities based on phylogenies, bioenergetics and substrate specificities. J. Mol. Biol. In press
6. Lewis, K. (1994) Multidrug resistance pumps in bacteria: Variations on a theme. Trends Biochem. Sci. 19, 119–123[Medline]
7. Nikaido, H. (1994) Prevention of drug access to bacterial targets: Permeability barriers and active efflux. Science 264, 382–388[Abstract/Free Full Text]
8. Davies, J. (1992) Another look at antibiotic resistance. 1991 Fred Griffith Review Lecture. J. Gen. Microbiol. 138, 1553–1559
9. Parr, T. R., Jr., and Saier, M. H., Jr. (1992) The bacterial phosphotransferase system as a potential vehicle for the entry of novel antibiotics. Res. Microbiol. 143, 443–447[Medline]
10. Fleischmann, R. D., Adams, M. D., White, O., Clayton, R. A., Kirkness, E. F., Kerlavage, A. R., Bult, C. J., Tomb, J.-F., Dougherty, B. A., Merrick, J. M., McKenney, K., Sutton, G., FitzHugh, W., Fields, C., Gocayne, J. D., Scott, J., Shirley, R., Liu, L.-I., Glodek, A., Kelley, J. M., Weidman, J. F., Phillips, C. A., Spriggs, T., Hedblom, E., Cotton, M. D., Utterback, T. R., Hanna, M. C., Nguyen, D. T., Saudek, D. M., Brandon, R. C., Fine, L. D., Fritchman, J. L., Fuhrmann, J. L., Geoghagen, N. S. M., Gnehm, C. L., McDonald, L. A., Small, K. V., Fraser, C. M., Smith, H. O., and Venter, J. C. (1995) Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science 269, 496–512[Abstract/Free Full Text]
11. Fraser, C. M., Gocayne, J. D., White, O., Adams, M. D., Clayton, R. A., Fleischmann, R. D., Bult, C. J., Kerlavage, A. R., Sutton, G., Kelley, J. M., Fritchman, J. L., Weidman, J. F., Small, K. V., Sandusky, M., Fuhrmann, J., Nguyen, D., Utterback, T. R., Saudek, D. M., Phillips, C. A., Merrick, J. M., Tomb, J.-F., Dougherty, B. A., Bott, K. F., Hu, P.-C., Lucier, T. S., Peterson, S. N., Smith, H. O., Hutchison, C. A., III, and Venter, J. C. (1995) The minimal gene complement of Mycoplasma genitalium. Science 270, 397–403[Abstract/Free Full Text]
12. Blattner, F. R., Plunkett, G., III., Bloch, C. A., Perna, N. T., Burland, V., Riley, M., Collado-Vides, J., Glasner, J. D., Rode, C. K., Mayhew, G. F., Gregor, J., Davis, N. W., Kirkpatrick, H. A., Goeden, M. A., Rose, D. J., Mau, B., and Shao, Y. (1997) The complete genome sequence of Escherichia coli K-12. Science 277, 1453–1474[Abstract/Free Full Text]
13. Biaudet, V., Samson, F., Anagnostopoulos, C., Ehrlich, S. D., and Bessières, P. (1996) Computerized genetic map of Bacillus subtilis. Microbiology 142, 2669–2729[Free Full Text]
14. Bult, C. J., White, O., Olsen, G. J., Zhou, L., Fleischmann, R. D., Sutton, G. G., Blake, J. A., FitzGerald, L. M., Clayton, R. A., Gocayne, J. D., Kerlavage, A. R., Dougherty, B. A., Tomb, J.-F., Adams, M. D., Reich, C. I., Overbeek, R., Kirkness, E. F., Weinstock, K. G., Merrick, J. M., Glodek, A., Scott, J. L., Geoghagen, N. S. M., Weidman, J. F., Fuhrmann, J. L., Nguyen, D., Utterback, T. R., Kelley, J. M., Peterson, J. D., Sadow, P. W., Hanna, M. C., Cotton, M. D., Roberts, K. M., Hurst, M. A., Kaine, B. P., Borodovsky, M., Klenk, H.-P., Fraser, C. M., Smith, H. O., Woese, C. R., and Venter, J. C. (1996) Complete genome sequence of the methanogenic archaeon, Methanococcus jannaschii. Science 273, 1058–1073[Abstract]
15. Kaneko, T., Sato, S., Kotani, H., Tanaka, A., Asamizu, E., Nakamura, Y., Miyajima, N., Hirosawa, M., Sugiura, M., Sasamoto, S., Kimura, T., Hosouchi, T., Matsuno, A., Muraki, A., Nakazaki, N., Naruo, K., Okumura, S., Shimpo, S., Takeuchi, C., Wada, T., Watanabe, A., Yamada, M., Yasuda, M., and Tabata, S. (1996) Sequence analysis of the genome of the unicellular cyanobacterium Synechocystis sp. strain PCC6803. II. Sequence determination of the entire genome and assignment of potential protein-coding regions. DNA Res. 3, 109–136[Abstract]
16. Higgins, C. F. (1992) ABC transporters: from microorganisms to man. Annu. Rev. Cell Biol. 8, 67–113
17. Griffith, J. K., Baker, M. E., Rouch, D. A., Page, M. G., Skurray, R. A., Paulsen, I. T., Chater, K. F., Baldwin, S. A., and Henderson, P. J. (1992) Membrane transport proteins: Implications of sequence comparisons. Curr. Opin. Cell Biol. 4, 684–695[Medline]
18. Marger, M. D., and Saier, M. H., Jr. (1993) A major superfamily of transmembrane facilitators that catalyse uniport, symport and antiport. Trends Biochem. Sci. 18, 13–20[Medline]
19. Paulsen, I. T., Skurray, R. A., Tam, R., Saier, M. H., Jr., Turner, R. J., Weiner, J. H., Goldberg, E. B., and Grinius, L. L. (1996) The SMR family: A novel family of multidrug efflux proteins involved with the efflux of lipophilic drugs. Mol. Microbiol. 19, 1167–1175[Medline]
20. Saier, M. H., Jr., Tam, R., Reizer, A., and Reizer, J. (1994) Two novel families of bacterial membrane proteins concerned with nodulation, cell division and transport. Mol. Microbiol. 11, 841–847[Medline]
21. Paulsen, I. T., Brown, M. H., and Skurray, R. A. (1996) Proton-dependent multidrug efflux systems. Microbiol. Rev. 60, 575–608[Abstract/Free Full Text]
22. George, A. M. (1996) Multidrug resistance in enteric and other gram-negative bacteria. FEMS Microbiol. Lett. 139, 1–10[Medline]
23. Dinh, T., Paulsen, I. T., and Saier, M. H., Jr. (1994) A family of extracytoplasmic proteins that allow transport of large molecules across the outer membranes of gram-negative bacteria. J. Bacteriol. 176, 3825–3831[Abstract/Free Full Text]
24. Paulsen, I. T., Park, J. H., Choi, P. S., and Saier, M. H., Jr. (1997) A family of Gram-negative bacterial outer membrane factors that function in the export of proteins, carbohydrates, drugs and heavy metals from Gram-negative bacteria. FEMS Microbiol. Lett. 156, 1–8.[Medline]
25. Nikaido, H. (1996) Multidrug efflux pumps of gram-negative bacteria. J. Bacteriol. 178, 5853–5859[Free Full Text]
26. Holland, I. B., and Blight, M. A. (1996) Structure and function of HlyB, the ABC-transporter essential for haemolysin secretion from Escherichia coli. In Handbook of Biological Physics (W. N. Konings, H. R. Kaback and J. S. Lolkema, eds) Vol. 2, pp. 111–135, Elsevier, Amsterdam
27. Fath, M. J., and Kolter, R. (1993) ABC-transporters: The bacterial exporters. Microbiol. Rev. 57, 995–1017[Abstract/Free Full Text]
28. Pigeon, R. P., and Silver, R. P. (1994) Topological and mutational analysis of KpsM, the hydrophobic component of the ABC-transporter involved in the export of polysialic acid in Escherichia coli K1. Mol. Microbiol. 14, 871–881[Medline]
29. Feng, D.-F., and Doolittle, R. F. (1990) Progressive alignment and phylogenetic tree construction of protein sequences. Methods Enzymol. 183, 375–387[Medline]
30. Paulsen, I. T., Beness, A. M., and Saier, M. H., Jr. (1997) Computer-based analyses of the protein constituents of transport systems catalyzing export of complex carbohydrates in bacteria. Microbiology 143, 2685–2699[Abstract/Free Full Text]
31. Saurin, W., and Dassa, E. (1994) Sequence relationships between integral inner membrane proteins of binding protein-dependent transport systems: Evolution by recurrent gene duplications. Prot. Sci. 3, 325–344[Medline]
32. Reizer, J., Reizer, A., and Saier, M. H., Jr. (1992) A new subfamily of bacterial ABC-type transport systems catalyzing export of drugs and carbohydrates. Prot. Sci. 1, 1326–1332[Medline]
33. Bolhuis, H., van Veen, H. W., Molenaar, D., Poolman, B., Driessen, A. J. M., and Konings, W. N. (1996) Multidrug resistance in Lactococcus lactis: Evidence for ATP-dependent drug extrusion from the inner leaflet of the cytoplasmic membrane. EMBO J. 15, 4239–4245[Medline]
34. Gros, P., and Hanna, M. (1996) The P-glycoprotein family and multidrug resistance: An overview. In Handbook of Biological Physics (W. N. Konings, H. R. Kaback, and J. S. Lolkema, eds) Vol. 2, pp. 137–163, Elsevier, Amsterdam
35. van Veen, H. W., Bolhuis, H., Putman, M., and Konings, W. N. (1996) Multidrug resistance in prokaryotes: Molecular mechanism of drug efflux. In Handbook of Biological Physics (W. N. Konings, H. R. Kaback, and J. S. Lolkema, eds) Vol. 2, pp. 165–187, Elsevier, Amsterdam
36. van Veen, H. W., Venema, K., Bolhuis, H., Oussenko, I., Kok, J., Poolman, B., Driessen, A. J. M., and Konings, W. N. (1996) Multidrug resistance mediated by a bacterial homolog of the human multidrug transporter MDR1. Proc. Natl. Acad. Sci. USA 93, 10668–10672[Abstract/Free Full Text]
37. Pao, S. S., Paulsen, I. T., and Saier, M. H., Jr. (1998) Major facilitator superfamily. Microbiol. Mol. Biol. Rev. In press
38. Allard, J. D., and Bertrand, K. P. (1992) Membrane topology of the pBR322 tetracycline resistance protein. TetA-PhoA gene fusions and implications for the mechanism of TetA membrane insertion. J. Biol. Chem. 267, 17809–17819[Abstract/Free Full Text]
39. Paulsen, I. T., and Skurray, R. A. (1993) Topology, structure and evolution of two families of proteins involved in antibiotic and antiseptic resistance in eukaryotes and prokaryotes–an analysis. Gene 124, 1–11[Medline]
40. Paulsen, I. T., Brown, M. H., Littlejohn, T. G., Mitchell, B. A., and Skurray, R. A. (1996) Molecular characterization of the multidrug resistance proteins QacA and QacB: Membrane topology and identification of residues involved in specificity for divalent cations. Proc. Natl. Acad. Sci. USA 93, 3630–3635[Abstract/Free Full Text]
41. Grundemann, D., Gorboulev, V., Gambaryan, S., Veyhl, M., and Koepsell, H. (1994) Drug excretion mediated by a new prototype of polyspecific transporter. Nature (London) 372, 549–552[Medline]
42. Goffeau, A., Park, J., Paulsen, I. T., Jonniaux, J.-L., Dinh, T., Mordant, P., and Saier, M. H., Jr. (1997) Multidrug-resistant transport proteins in yeast: Complete inventory and phylogenetic characterization of yeast open reading frames within the major facilitator superfamily. Yeast 13, 43–54[Medline]
43. Schuldiner, S., Lebendiker, M., and Yerushalmi, H. (1997) EmrE, the smallest ion-coupled transporter, provides a unique paradigm for structure-function studies. J. Exp. Biol. 200, 335–341[Abstract]
44. Paulsen, I. T., Brown, M. H., Dunstan, S. J., and Skurray, R. A. (1995) Molecular characterization of the staphylococcal multidrug resistance export protein QacC. J. Bacteriol. 177, 2827–2833[Abstract/Free Full Text]
45. Arkin, I. T., Russ, W. P., Lebendiker, M., and Schuldiner, S. (1996) Determining the secondary structure and orientation of EmrE, a multi-drug transporter, indicates a transmembrane four-helix bundle. Biochemistry 35, 7233–7238[Medline]
46. Yerushalmi, H., Lebendiker, M., and Schuldiner, S. (1995) EmrE, an Escherichia coli 12-kDa multidrug transporter, exchanges toxic cations and H⁺ and is soluble in organic solvents. J. Biol. Chem. 270, 6856–6863[Abstract/Free Full Text]
47. Yerushalmi, H., Lebendiker, M., and Schuldiner, S. (1996) Negative dominance studies demonstrate the oligomeric structure of EmrE, a multidrug antiporter from Escherichia coli. J. Biol. Chem. 271, 31044–31048[Abstract/Free Full Text]
48. Kuan, G., Dassa, E., Saurin, W., Hofnung, M., and Saier, M. H., Jr. (1995) Phylogenetic analyses of the ATP-binding constituents of bacterial extracytoplasmic receptor-dependent ABC-type nutrient uptake permeases. Res. Microbiol. 146, 271–278[Medline]
49. Higgins, C. F., and Gottesman, M. M. (1992) Is the multidrug transporter a flippase? Trends Biochem. Sci. 17, 18–21[Medline]
50. Ruetz, S., and Gros, P. (1994) Phosphatidylcholine translocase: A physiological role for the mdr2 gene. Cell 77, 841–847[Medline]
51. van Helvoort, A., Smith, A. J., Sprong, H., Fritzsche, I., Schinkel, A. H., Borst, P., and van Meer, G. (1996) MDR1 P-glycoprotein is a lipid translocase of broad specificity, while MDR3 P-glycoprotein specifically translocates phosphatidylcholine. Cell 87, 507–517[Medline]
52. Sali, A., Overington, J. P., Johnson, M. S., and Blundell, T. L. (1990) From comparisons of protein sequences and structures to protein modelling and design. Trends Biochem. Sci. 15, 235–240[Medline]
53. Singer, S. J. (1990) The structure and insertion of integral proteins in membranes. Annu. Rev. Cell Biol. 6, 247–296
54. Singer, S. J., and Yaffe, M. P. (1990) Embedded or not? Hydrophobic sequences and membranes. Trends Biochem. Sci. 15, 369–373[Medline]
55. Rapoport, T. A., Jungnickel, B., and Kutay, U. (1996) Protein transport across the eukaryotic endoplasmic reticulum and bacterial inner membranes. Annu. Rev. Biochem. 65, 271–303[Medline]
56. Martoglio, B., Hofmann, M. W., Brunner, J., and Dobberstein, B. (1995) The protein-conducting channel in the membrane of the endoplasmic reticulum is open laterally toward the lipid bilayer. Cell 81, 207–214[Medline]
57. Olsen, G. J., Woese, C. R., and Overbeek, R. (1994) The winds of (evolutionary) change: Breathing new life into microbiology. J. Bacteriol. 176, 1–6[Free Full Text]