Membrane topology of the xenobiotic-exporting subunit, MexB, of the MexA,B-OprM extrusion pump in Pseudomonas aeruginosa

The MexA,B-OprM efflux pump assembly of Pseudomonas aeruginosa consists of two inner membrane proteins and one outer membrane protein. The cytoplasmic membrane protein, MexB, appears to function as the xenobiotic-exporting subunit, whereas the MexA and OprM proteins are supposed to function as the membrane fusion protein and the outer membrane channel protein, respectively. Computer-aided hydropathy analyses of MexB predicted the presence of up to 17 potential transmembrane segments. To verify the prediction, we analyzed the membrane topology of MexB using the alkaline phosphatase gene fusion method. We obtained the following unique characteristics. MexB bears 12 membrane spanning segments leaving both the amino and carboxyl termini in the cytoplasmic side of the inner membrane. Both the first and fourth periplasmic loops had very long hydrophilic domains containing 311 and 314 amino acid residues, respectively. This fact suggests that these loops may interact with other pump subunits, such as the membrane fusion protein MexA and the outer membrane protein OprM. Alignment of the amino- and the carboxyl-terminal halves of MexB showed a 30% homology and transmembrane segments 1, 2, 3, 4, 5, and 6 could be overlaid with the segments 7, 8, 9, 10, 11, and 12, respectively. This result suggested that the MexB has a 2-fold repeat that strengthen the experimentally determined topology model. This paper reports the structure of the pump subunit, MexB, of the MexA,B-OprM efflux pump assembly. This is the first time to verify the topology of the resistant-nodulation-division efflux pump protein.     

Complementation between nucleotide binding domains in an anion-translocating ATPase.

The catalytic component of the oxyanion-translocating ATPase of the plasmid-encoded ars operon of Escherichia coli is a homodimer of the ArsA protein. This enzyme is an oxyanion-stimulated ATPase with two consensus nucleotide binding sequences in each subunit, one in the N-terminal (A1) half and one in the C-terminal (A2) half of the ArsA protein. The two halves of both the arsA gene and the ArsA protein exhibit similar nucleotide and amino acid sequences, respectively. The two halves of the arsA gene were subcloned into compatible plasmids. Neither alone was sufficient to confer resistance, but cells in which the arsA1 and arsA2 half genes were coexpressed were resistant to arsenicals. Genetic complementation was also observed in cells bearing plasmids with point mutations in the two halves of the arsA gene and between cells with plasmids carrying combinations of the arsA1 or arsA2 subclones and point mutations. In every case, complementation was observed only when one plasmid contained a wild-type arsA1 sequence and the other contained a wild-type arsA2 sequence. These results demonstrate that both sites are required for resistance but that the two nucleotide binding domains need not reside in a single polypeptide. We propose a model in which the ArsA dimer has two catalytic units, each composed of an A1 domain from one monomer and an A2 domain from the other monomer.     

Mutational analysis of the yeast a-factor transporter STE6, a member of the ATP binding cassette (ABC) protein superfamily.

STE6, the yeast a-factor transporter, is a member of the ATP binding cassette protein superfamily, which also includes the mammalian multidrug resistance protein and the cystic fibrosis gene product. These proteins contain two homologous halves, each with six membrane spanning segments and a predicted ATP nucleotide binding domain. To assess the importance of the two halves of STE6, and to examine the functional significance of residues conserved among members of the ATP binding cassette superfamily, we introduced mutations into the nucleotide binding domains of STE6. Our analysis demonstrates that both halves of STE6 are critical for function and that some, but not all, mutations analogous to those known to result in cystic fibrosis impair STE6 activity. To examine further the functional contribution of each half of the STE6 protein, we severed the STE6 coding sequence and expressed the two halves of the transporter as separate polypeptides. Whereas 'half-molecules' are unable to provide transport function individually, co-expression of both half-molecules in the same cell leads to functional reconstitution of STE6-mediated a-factor transport.     

Deletion and insertion mutants of the multidrug transporter

The multidrug transporter is a 170,000-dalton membrane glycoprotein which confers multidrug resistance through its activity as an ATP-dependent efflux pump for hydrophobic, cytotoxic drugs. To determine the essential structural components of this complex membrane transporter we have altered an MDR1 cDNA in an expression vector by deletion and insertion mutations. The structure of the transporter deduced from its amino acid sequence suggests that it consists of two homologous, perhaps functionally autonomous, halves each with six transmembrane segments and a cytoplasmic ATP-binding domain. However, several carboxyl-terminal deletions, one involving 53 amino acids, the second removing 253 amino acids, and an internal deletion within the carboxyl-terminal half of the molecule, totally eliminate the ability of the mutant transporter to confer drug resistance. An internal deletion of the amino-terminal half, which removed residues 140-229, is also nonfunctional. Small carboxylterminal deletions of up to 23 amino acids leave a functional transporter, although the removal of 23 COOH-terminal amino acids reduces its ability to confer colchicine resistance. Insertions of 4 amino acids in a transmembrane domain, and in one of the two ATP-binding regions, have no effect on activity. These studies define some of the limits of allowable deletions and insertions in the MDR1 gene, and demonstrate the requirement for two intact halves of the molecule for a functional multidrug transporter.     

Fitting periplasmic membrane fusion proteins to inner membrane transporters: mutations that enable Escherichia coli AcrA to function with Pseudomonas aeruginosa MexB

AcrAB-TolC from Escherichia coli is a multidrug efflux complex capable of transenvelope transport. In this complex, AcrA is a periplasmic membrane fusion protein that establishes a functional connection between the inner membrane transporter AcrB of the RND superfamily and the outer membrane channel TolC. To gain insight into the mechanism of the functional association between components of this complex, we replaced AcrB with its close homolog MexB from Pseudomonas aeruginosa. Surprisingly, we found that AcrA is promiscuous and can form a partially functional complex with MexB and TolC. The chimeric AcrA-MexB-TolC complex protected cells from sodium dodecyl sulfate, novobiocin, and ethidium bromide but failed with other known substrates of MexB. We next identified single and double mutations in AcrA and MexB that enabled the complete functional fit between AcrA, MexB, and TolC. Mutations in either the α-helical hairpin of AcrA making contact with TolC or the β-barrel domain lying on MexB improved the functional alignment between components of the complex. Our results suggest that three components of multidrug efflux pumps do not associate in an “all-or-nothing” fashion but accommodate a certain degree of flexibility. This flexibility in the association between components affects the transport efficiency of RND pumps.     

Use of recombinant P-glycoprotein fragments to produce antibodies to the multidrug transporter

Multidrug-resistance of human cancer cells may result from expression of a 170,000 dalton multidrug efflux pump called P-glycoprotein. To identify this multidrug transporter, and to study its structure and function, we have generated polyclonal rabbit antibodies against the amino-terminal and carboxy-terminal halves of the molecule using recombinant protein fragments produced in Escherichia coli. Two recombinant P-glycoprotein fragments, representing amino acids 140-228 and 919-1280, were overproduced in Escherichia coli by an inducible T7 expression system, gel-purified and injected into rabbits. Both antisera specifically immunoprecipitate 3H-azidopine and 35S-methionine labeled P-glycoprotein from multidrug-resistant cells and detect P-glycoprotein on Western blots with high sensitivity. Because these antisera were raised against epitopes in the amino- and carboxy-terminal halves of P-glycoprotein, they should be useful as research tools to define the function of these two halves of the molecule.     

Identification of an additional interaction domain in transmembrane domains 11 and 12 that supports oligomer formation in the human serotonin transporter

Na+/Cl--dependent neurotransmitter transporters form constitutive oligomers. The topological arrangement is not known, but a leucine heptad repeat in transmembrane domain (TM) 2 and a glycophorin-like motif in TM6 have been proposed to stabilize the oligomer. To determine the topology, we generated versions of the human serotonin transporter (hSERT) that carried cyan or yellow fluorescent proteins at their amino and/or carboxyl terminus. Appropriate pairs were coexpressed to measure fluorescence resonance energy transfer (FRET). Donor photobleaching FRET microscopy was employed to deduce the following arrangement: within the monomer, the amino and carboxyl termini are in close vicinity. In addition, in the oligomer, the carboxyl termini are closer to each other than the amino termini. Hence, a separate interaction domain (i.e. distinct from TM2 and TM6) must reside in the carboxyl-terminal half of hSERT. This was confirmed by expressing the amino- and carboxyl-terminal halves of hSERT. These were retained intracellularly; they also retained the coexpressed full-length transporter by forming export-deficient oligomers and, when cotransfected in all possible combinations, supported FRET. Hence, both the carboxyl and amino termini contain elements that drive oligomerization. By employing fragments comprising two neighboring TM helices, we unequivocally identified TM11/12 as a new contact site by donor photobleaching FRET and beta-lactamase protein fragment complementation assay. TM1/2 was also found to self-associate. Thus, oligomerization of hSERT involves at least two discontinuous interfaces. The currently identified interaction sites drive homophilic interactions. This is consistent with assembly of SERT oligomers in an array-like structure containing multimers of dimers.     

Crystal Structure of the Multidrug Exporter MexB from Pseudomonas aeruginosa

We report here the crystal structure of the Pseudomonas aeruginosa multidrug exporter MexB, an intensively studied member of the resistance-nodulation-cell division family of secondary active transporters, at 3.0 Å. MexB forms an asymmetric homotrimer where each subunit adopts a different conformation representing three snapshots of the transport cycle similar to the recently determined structures of its close homologue AcrB from Escherichia coli, so far the sole structurally characterized member of the superfamily. As for AcrB, the conformations of two subunits can be clearly assigned to either the binding step or the extrusion step in the transport process. Unexpectedly, a remarkable conformational shift in the third subunit is observed in MexB, which has potential implications for the assembly of the tripartite MexAB-OprM drug efflux system. Furthermore, an n-dodecyl-d-maltoside molecule was found bound to the internal multidrug-binding cavity, which might indicate that MexB binds and transports detergent molecules as substrates. As the only missing piece of the puzzle in the MexAB-OprM system, the X-ray structure of MexB completes the molecular picture of the major pump mediating intrinsic and acquired multidrug resistance in P. aeruginosa.     

Two different regions of P-glycoprotein [corrected] are photoaffinity-labeled by azidopine

Cells that express P-glycoprotein are resistant to many unrelated anticancer drugs. All evidence suggests that P-glycoprotein is a plasma membrane protein that confers multidrug resistance by actively transporting these cytotoxic drugs out of cells. The objective of our work is to locate drug binding sites on P-glycoprotein. Azidopine is a photoaffinity drug analog that specifically labels P-glycoprotein. To determine the region of P-glycoprotein that binds azidopine, we labeled P-glycoprotein with azidopine and digested the labeled protein into fragments. We then identified the labeled fragments with specific antibodies. We have determined that azidopine labels two different regions of P-glycoprotein: one region is in the amino half of P-glycoprotein, and the other is in the carboxyl half of the protein. Our results suggest that P-glycoprotein contains either two binding sites for azidopine or a single site formed by the two homologous halves of the protein.     

Val133 and Cys137 in transmembrane segment 2 are close to Arg935 and Gly939 in transmembrane segment 11 of human P-glycoprotein

P-glycoprotein (P-gp; ABCB1) transports a wide variety of structurally diverse compounds out of the cell. The protein has two homologous halves joined by a linker region. Each half consists of a transmembrane (TM) domain with six TM segments and a nucleotide-binding domain. The drug substrate-binding pocket is at the interface between the TM segments in each half of the protein. Preliminary studies suggested that the arrangement of the two halves of P-gp shows rotational symmetry (i.e. "head-to-tail" arrangement). Here, we tested this model by determining whether the cytoplasmic ends of TM2 and TM3 in the N-terminal half are in close contact with TM11 in the C-terminal half. Mutants containing a pair of cysteines in TM2/TM11 or TM3/TM11 were subjected to oxidative cross-linking with copper phenanthroline. Two of the 110 TM2/TM11 mutants, V133C(TM2)/G939C(TM11) and C137C(TM2)/A935C (TM11), were cross-linked at 4 degrees C, when thermal motion is reduced. Cross-linking was specific since no cross-linked product was detected in the 100 double Cys TM3/TM11 mutants. Vanadate trapping of nucleotide or the presence of some drug substrates inhibited cross-linking of mutants V133C(TM2)/G939C(TM11) and C137C(TM2)/A935C(TM11). Cross-linking of TM2 and TM11 also blocked drug-stimulated ATPase activity. The close proximity of TM2/TM11 and TM5/TM8 (Loo, T. W., Bartlett, M. C., and Clarke, D. M. (2004) J. Biol. Chem. 279, 7692-7697) indicates that these regions between the two halves must enclose the drug-binding pocket at the cytoplasmic side of P-gp. They may form the "hinges" required for conformational changes during the transport cycle.     

Mutations in MexB that affect the efflux of antibiotics with cytoplasmic targets

Drug efflux pumps such as MexAB-OprM from Pseudomonas aeruginosa confer resistance to a wide range of chemically different compounds. Within the tripartite assembly, the inner membrane protein MexB is mainly responsible for substrate recognition. Recently, considerable advances have been made in elucidating the drug efflux pathway through the large periplasmic domains of resistance-nodulation-division (RND) transporters. However, little is known about the role of amino acids in other parts of the protein. We have investigated the role of two conserved phenylalanine residues that are aligned around the cytoplasmic side of the central cavity of MexB. The two conserved phenylalanine residues have been mutated to alanine residues (FAFA MexB). The interaction of the wild-type and mutant proteins with a variety of drugs from different classes was investigated by assays of cytotoxicity and drug transport. The FAFA mutation affected the efflux of compounds that have targets inside the cell, but antibiotics that act on cell wall synthesis and membrane probes were unaffected. Combined, our results indicate the presence of a hitherto unidentified cytoplasmic-binding site in RND drug transporters and enhance our understanding of the molecular mechanisms that govern drug resistance in Gram-negative pathogens.     

Structure-function relationships in an anion-translocating ATPase

The ArsAB ATPase is an efflux pump located in the inner membrane of Escherichia coli. This transport ATPase confers resistance to arsenite and antimonite by their extrusion from the cells. The pump is composed of two subunits, the catalytic ArsA subunit and the membrane subunit ArsB. The complex is similar in many ways to ATP-binding cassette ('ABC') transporters, which typically have two groups of six transmembrane-spanning helical segments and two nucleotide-binding domains (NBDs). The 45 kDa ArsB protein has 12 transmembrane-spanning segments. ArsB contains the substrate translocation pathway and is capable of functioning as an anion uniporter. The 63 kDa ArsA protein is a substrate-activated ATPase. It has two homologous halves, A1 and A2, which are clearly the result of an ancestral gene duplication and fusion. Each half has a consensus NBD. The mechanism of allosteric activation of the ArsA ATPase has been elucidated by a combination of molecular genetics and biochemical, structural and kinetic analyses. Conformational changes produced by binding of substrates, activator and/or products could be revealed by stopped-flow fluorescence measurements with single-tryptophan derivatives of ArsA. The results demonstrate that the rate-limiting step in the overall reaction is a slow isomerization between two conformations of the enzyme. Allosteric activation increases the rate of this isomerization such that product release becomes rate-limiting, thus accelerating catalysis. ABC transporters, which exhibit similar substrate activation of ATPase activity, can undergo similar conformational changes to overcome a rate-limiting step. Thus the ArsAB pump is a useful model for elucidating mechanistic aspects of the ABC superfamily of transport ATPases.     

Photoaffinity labeling of the adenosine cyclic 3',5'-monophosphate receptor protein of Escherichia coli with 8-azidoadenosine 3',5'-monophosphate

Photoaffinity labeling of the cAMP receptor protein (CRP) of Escherichia coli with 8-azidoadenosine 3',5'-monophosphate (8-N3cAMP) has been demonstrated. 8-N3cAMP is able to support the binding of (3H)d(I-C)n by CRP, indicating that it is a functional cAMP analogue. Following irradiation at 254 nm, (32P)-8-N3cAMP is photocross-linked to CRP. Photolabeling of CRP by (32P)-8-N3cAMP is inhibited by cAMP but not by 5'AMP. The data indicate that (32P)-8-N3cAMP is covalently incorporated following binding at the cAMP binding site of CRP. The (32P)-8-N3cAMP-CRP digested with chymotrypsin was analyzed by NaDodSO4-polyacrylamide gel electrophoresis. Of the incorporated label, one-third remains associated with the amino-proximal alpha core region of CRP [Eilen, E., Pampeno, C., & Krakow, J.S. (1978) Biochemistry 17, 2469] which contains the cAMP binding domain; the remaining two-thirds of the label associated with the beta region are digested. Limited proteolysis of the (32P)-8-N3cAMP-CRP by chymotrypsin in the presence of NaDodSO4 shows the radioactivity to be distributed between the molecular weight 9500 (amino-proximal) and 13,000 (carboxyl-proximal) fragments produced. These results suggest that a part of the carboxyl-proximal region is folded over and close enough to the cAMP binding site to be cross-linked by the photoactivated (32P)-8-N3cAMP bound at the cAMP binding site.     

Control of the Expression of AV-1 Protein, a Position-Specific Antigen, by ZPA (Zone of Polarizing Activity) Factor in the Chick Limb Bud

AV-1 protein is a molecule which shows position-specific expression during chick limb development, and is expected to have some important roles in limb pattern formation. In this study, to examine whether the ZPA (Zone of polarizing activity) effects the expression of the AV-1 protein, we have removed or grafted the ZPA in chick limb buds and observed AV-1 expression. Anterior halves of the limb buds which lack a ZPA were used as hosts. In such anterior halves, AV-1 expression was initially observed in distal mesodermal cells including the cut surface. These anterior halves were combined with ZPA fragments, anterior fragments, posterior half limb buds, or left to develop alone, and the distribution of AV-1 expression was examined. The results of these experiments show that AV-1 expression requires the ZPA, and that expression occurs in the distal mesodermal cells certain distance from the ZPA.     

Functional analysis of chimeric genes obtained by exchanging homologous domains of the mouse mdr1 and mdr2 genes.

A full-length cDNA clone for the mouse mdr1 gene can confer multidrug resistance when introduced by transfection into otherwise drug-sensitive cells. In the same assay, a full-length cDNA clone for a closely related member of the mouse mdr gene family, mdr2, fails to confer multidrug resistance. To identify the domains of mdr1 which are essential for multidrug resistance and which may be functionally distinct in mdr2, we have constructed chimeric cDNA molecules in which discrete domains of mdr2 have been introduced into the homologous region of mdr1 and analyzed these chimeras for their capacity to transfer drug resistance. The two predicted ATP-binding domains of mdr2 were found to be functional, as either could complement the biological activity of mdr1. Likewise, a chimeric molecule in which the highly sequence divergent linker domain of mdr2 had been introduced in mdr1 could also confer drug resistance. However, the replacement of either the amino- or carboxy-terminus transmembrane (TM) domain regions of mdr1 by the homologous segments of mdr2 resulted in inactive chimeras. The replacement of as few as two TM domains from either the amino (TM5-6) or the carboxy (TM7-8) half of mdr1 by the homologous mdr2 regions was sufficient to destroy the activity of mdr1. These results suggest that the functional differences detected between mdr1 and mdr2 in our transfection assay reside within the predicted TM domains.     

Functional interactions between the noncovalently associated N- and C-terminal halves of mammalian Type I hexokinase

The 100 kDa Type I isozyme of mammalian hexokinase has evolved by duplication and fusion of a gene encoding an ancestral 50 kDa hexokinase. Although the N- and C-terminal halves are similar in sequence, they differ in function, catalytic activity being associated only with the C-terminal half while the N-terminal half serves a regulatory role. The N- and C-terminal halves of rat Type I hexokinase have been coexpressed in M + R 42 cells. The halves associate noncovalently to produce a 100 kDa form that exhibits characteristics seen with the intact Type I isozyme but not with the isolated catalytic C-terminal half, i.e., characteristics that are influenced by interactions between the halves. These include a decreased K(m) for the substrate ATP and the ability of P(i) to antagonize inhibition by Glc-6-P or its analog, 1-5-anhydroglucitol-6-P. Thus, functional interactions between the N- and C-terminal halves do not require their covalent linkage.     

Crystal structure of Mil (Mth680): internal duplication and similarity between the Imp4/Brix domain and the anticodon-binding domain of class IIa aminoacyl-tRNA synthetases

Proteins of the Imp4/Brix superfamily are involved in ribosomal RNA processing, an essential function in all cells. We report the first structure of an Imp4/Brix superfamily protein, the Mil (for Methanothermobacter thermautotrophicus Imp4-like) protein (gene product Mth680), from the archaeon M. thermautotrophicus. The amino- and carboxy-terminal halves of Mil show significant structural similarity to one another, suggesting an origin by means of an ancestral duplication. Both halves show the same fold as the anticodon-binding domain of class IIa aminoacyl-tRNA synthetases, with greater conservation seen in the N-terminal half. This structural similarity, together with the charge distribution in Mil, suggests that Imp4/Brix superfamily proteins could bind single-stranded segments of RNA along a concave surface formed by the N-terminal half of their beta-sheet and a central alpha-helix. The crystal structure of Mil is incompatible with the presence, in the Imp4/Brix domain, of a helix-turn-helix motif that was proposed to comprise the RNA-binding moiety of the Imp4/Brix proteins.     

Sequences within the VP6 molecule of bluetongue virus that determine cytoplasmic and nuclear targeting of the protein.

Genome segment 9 of bluetongue virus serotype 10 encodes the minor protein VP6. The protein is abundant with basic residues particularly in two regions of the carboxy half of the molecule. A series of amino- and carboxy-terminal deletion mutants was expressed in mammalian cells by using a vaccinia virus T7 polymerase-driven transient expression system, and the intracellular fate of the products was monitored by both immunofluorescence staining and cell fractionation techniques. Data obtained indicated clearly that VP6 has nuclear transportation signals which may be correlated with positively charged domains of the molecule. In the intact molecule, though, these signals are masked and the protein is retained in the cytoplasm. The biochemical and immunofluorescence data obtained indicate that sequences in the region of residues 33 to 80 of the 328-amino acid protein are required for the retention of VP6 within the cell cytoplasm while amino acids 303 to 308 in the carboxy-terminal half of the molecule appear to possess nuclear localization capabilities.     

Gene duplication in the evolution of the two complementing domains of gram-negative bacterial tetracycline efflux proteins

The resistance of Gram- bacteria to the broad-spectrum antibiotic tetracycline (Tc) results from energy-dependent drug efflux mediated by the tet gene product, the cytoplasmic membrane Tet protein. Amino acid (aa) sequences deduced from total tet nucleotide sequences of three different resistance determinants (classes A, B and C) indicate that the protein products [Tet(A), Tet(B), and Tet(C)] share a common ancestor. Hydropathic analysis of Tet sequences predicts twelve transmembrane segments in each protein, with six occurring in each half of the molecule. More importantly, the linear distributions of these segments in the N- and C-terminal halves are nearly identical, suggesting that the two halves of each Tet protein are related by a process of tandem gene duplication and divergence. Indeed, a variable but significant conservation of sequence was detected among the N- and C-terminal halves for all possible comparisons of the three proteins. Such conservation was not observed within other prokaryotic integral membrane proteins or when other prokaryotic proteins were compared to Tet halves. Similarity, both in sequence and in predicted transmembrane structural organization, strongly suggests that a common ancestor of Tet(A), Tet(B), and Tet(C) arose by duplication of a gene reading frame specifying a transmembrane protein of approximately 200 aa residues. The two halves of Tet proteins correspond to the two domains, alpha and beta, which have distinct, complementary roles in Tc efflux. Nevertheless, selective constraints to function in the cytoplasmic membrane have apparently led to maintenance of similar patterns of secondary structural organization in these complementary domains.     

Functional Studies of Split Arabidopsis Ca2+/H+ Exchangers

In plants, high capacity tonoplast cation/H⁺ antiport is mediated in part by a family of cation exchanger (CAX) transporters. Functional association between CAX1 and CAX3 has previously been shown. In this study we further examine the interactions between CAX protein domains through the use of nonfunctional halves of CAX transporters. We demonstrate that a protein coding for an N-terminal half of an activated variant of CAX1 (sCAX1) can associate with the C-terminal half of either CAX1 or CAX3 to form a functional transporter that may exhibit unique transport properties. Using yeast split ubiquitin, in planta bimolecular fluorescence complementation, and gel shift experiments, we demonstrate a physical interaction among the half proteins. Moreover, the half-proteins both independently localized to the same yeast endomembrane. Co-expressing variants of N- and C-terminal halves of CAX1 and CAX3 in yeast suggested that the N-terminal region mediates Ca²⁺ transport, whereas the C-terminal half defines salt tolerance phenotypes. Furthermore, in yeast assays, auto-inhibited CAX1 could be differentially activated by CAX split proteins. The N-terminal half of CAX1 when co-expressed with CAX1 activated Ca²⁺ transport, whereas co-expressing C-terminal halves of CAX variants with CAX1 conferred salt tolerance but no apparent Ca²⁺ transport. These findings demonstrate plasticity through hetero-CAX complex formation as well as a novel means to engineer CAX transport.