![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of Pharmacology and Toxicology (I.J.L., S.P.C.C.) and Pathology and Molecular Medicine (A.J.S., R.G.D., S.P.C.C.) and Division of Cancer Biology and Genetics, Cancer Research Institute (I.J.L., A.J.S., R.G.D., S.P.C.C.), Queen's University, Kingston, Ontario, Canada
(Received February 28, 2007; accepted May 8, 2007)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
-32P]8N3ADP by the Pro1150Ala mutant in the absence of substrate was also greatly reduced compared with wild-type MRP1 suggesting an uncoupling of ATP hydrolysis and transport activity. To determine whether the functional importance of MRP1-Pro1150 is conserved, the analogous Pro1158 and Pro1147 residues in the MRP2 and MRP3 transporters, respectively, were mutated to Ala. Expression levels of the three mutants were unaffected; however, the vesicular transport activity of at least one organic anion substrate was significantly altered. As observed for MRP1-Pro1150Ala, LTC4 transport by MRP2-Pro1158Ala was decreased. However, E217ßG and MTX transport was comparable with that of wild-type MRP2 rather than increased as was observed for MRP1-Pro1150Ala. In the case of MRP3-Pro1147Ala, LTC4 transport was increased, whereas E217ßG transport was unaffected. MTX transport by MRP3-Pro1147Ala was also increased but to a lesser extent than for MRP1-Pro1150Ala. In contrast, all three mutants showed a marked reduction in levels of vanadate-induced trapped [-32P]8N3ADP. We conclude that MRP1-Pro1150, MRP2-Pro1158, and MRP3-Pro1147 in CL7 differ in their influence on substrate specificity but share a common role in the nucleotide interactions of these transporters.
; Deeley et al., 2006). These proteins mediate transport of their substrates across the plasma membrane using ATP binding and hydrolysis as an energy source. All three transporters are composed of three membrane spanning domains containing 5, 6, and 6 transmembrane (TM) -helices, respectively, and two functionally nonequivalent nucleotide binding domains (NBDs). We initially identified MRP1 as a transporter capable of conferring tumor cell resistance to chemotherapeutic agents including doxorubicin, etoposide, and vincristine. Subsequently, it was determined that glutathione (GSH) is required for MRP1 to transport at least some of these unconjugated drugs (Loe et al., 1996b; Leslie et al., 2005).
The related transporter, MRP2, was first characterized as a canalicular multispecific organic anion transporter (cMOAT) responsible for biliary excretion of conjugated organic anions (Paulusma et al., 1996). A third homolog, MRP3, was identified through screening of a database of human expressed sequence tags, based on its homology with MRP1 (Kool et al., 1997). Although MRP1, MRP2, and MRP3 share 48 to 58% sequence identity, they each have their own distinct, yet overlapping, physiological function, tissue distribution, and substrate specificity (Deeley et al., 2006).
MRP1 is expressed ubiquitously throughout the body, except in the liver where it is usually not detectable, and is mainly found in the basolateral membranes of polarized epithelial cells (Leslie et al., 2005). Studies of Mrp1–/– mice showed that this transporter mediates the release of leukotriene C4 (LTC4) during inflammatory responses and protects several normal tissues from cytotoxic agents (Wijnholds et al., 1997, 1998). In vitro, MRP1 transports a wide variety of endogenous and exogenous compounds, many of which are conjugated to glucuronide, GSH, or sulfate (Leslie et al., 2005). For example, in addition to LTC4, the conjugated steroids estradiol 17ß-glucuronide (E217ßG) and estrone sulfate have been identified as MRP1 substrates (Leier et al., 1994; Loe et al., 1996a; Qian et al., 2001).
As alluded to above, MRP2 and MRP1 transport a similar spectrum of substrates, but they differ in their affinity for many of these. In addition, MRP2 is expressed in a more limited number of tissues than MRP1, being found primarily in liver, intestine, lung, and kidney. Furthermore, MRP2 is found in the apical membrane of polarized cells in these tissues in contrast to the basolateral localization of MRP1 (Leslie et al., 2005).
|
; Zelcer et al., 2001; Oleschuk et al., 2003). Like MRP1, however, MRP3 is expressed on the basolateral membrane of polarized cells but in a more limited number of tissues (Leslie et al., 2005).
We previously described a functionally complex MRP1 mutant in which a conserved Pro residue located at the beginning of a cytoplasmic loop (CL7) that connects TM15 to TM16 was replaced with Ala (Koike et al., 2004). Compared with wild-type MRP1, MRP1-Pro1150Ala displayed decreased levels of LTC4, estrone sulfate, and GSH transport but substantially increased levels of E217ßG and methotrexate (MTX) transport. Accompanying the altered transport profile of this mutant, substantial changes in the interaction of MRP1 with nucleotide were observed. Although no difference in ATP binding was detected, the ability of MRP1-Pro1150Ala to trap 8N3ADP under hydrolytic conditions in the presence of sodium orthovanadate (but in the absence of substrate) was severely diminished. Because the ability to detect vanadate-inducible trapped [-32P]8N3ADP complexes is often considered an indicator of the ATPase activity of a protein (Urbatsch et al., 1995), our observations suggested that either ATP hydrolysis by the MRP1-Pro1150Ala mutant was reduced or that the release of ADP may be enhanced (Koike et al., 2004).
It was also observed that although the apparent Km(ATP) for both wild-type MRP1 and MRP1-Pro1150Ala was the same during LTC4 transport, the Km(ATP) for the mutant transporter was reduced more than 4-fold during E217ßG transport. This change in ATP dependence suggests that E217ßG, but not LTC4, interacts with the MRP1-Pro1150Ala mutant in a way that increases its apparent affinity for ATP. The complexity of the MRP1-Pro1150Ala mutant phenotype, together with the high conservation of this Pro residue among ABCC family members (Fig. 1), has prompted us to investigate whether the corresponding residues in the MRP1 homologs, MRP2 and MRP3, are also functionally important. We have therefore determined the consequences of Ala substitution of MRP2-Pro1158 and MRP3-Pro1147 on the transport properties and nucleotide interactions of these two transporters.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
-32P]8N3ATP (12.0 Ci/mmol) and [
-32P]8N3ATP (10.6 Ci/mmol) were purchased from Affinity Labeling Technologies, Inc. (Lexington, KY). LTC4 was purchased from Calbiochem (San Diego, CA). AMP, ATP, and E217ßG were purchased from Sigma-Aldrich (St. Louis, MO). Creatine kinase and creatine phosphate were obtained from Roche Diagnostics (Laval, QC, Canada). MTX sodium salt was purchased from Faulding (Vaudreuil, QC, Canada). Monoclonal antibodies M2I-4 and M3II-9 specific for MRP2 and MRP3, respectively, were purchased from Alexis Laboratories (San Diego, CA).
Site-Directed Mutagenesis. Generation of the MRP1-Pro1150Ala mutant has been described previously (Koike et al., 2004). Mutations in MRP2 and MRP3 were similarly generated using the QuickChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA). The template for generating the MRP2-Pro1158Ala mutant was created by subcloning a 2.2-kb ApaI/ClaI fragment into pGEM-7Zf(+) (Promega, Madison, WI). The mutagenesis template for the MRP3-Pro1147Ala was a pBluescript II KS(+) plasmid (Stratagene, La Jolla, CA) containing the full-length MRP3 cDNA as described previously (Oleschuk et al., 2003). Mutagenic oligonucleotide primers were obtained from IDT Inc. (Coralville, IA). Mutagenesis was performed according to the manufacturer's instructions with the following sense primers (substituted nucleotides are underlined, and in both cases, the substitution created a BstUI restriction site, which was used to confirm the successful introduction of the nucleotide substitutions): MRP2-Pro1158Ala, 5'-C ACC AGG TCC GCG ATC TAC TCT C-3' (BstUI) and MRP3-Pro1147Ala, 5'-GTC AGC CGC TCC GCG ATC TAC TCC C-3' (BstUI). The MRP2-Pro1158Ala mutation was subcloned back into pcDNA3.1(–) MRP2 as a 1.75-kb Bsu36I/SfiI fragment and the MRP3-Pro1147Ala mutation was subcloned into pcDNA3.1(+) MRP3 as a 1.24-kb AgeI/SacII fragment (Oleschuk et al., 2003). The fidelity of the fragments was then verified by sequencing (ACGT Corp., Toronto, ON, Canada).
Transfection of MRP1, MRP2, and MRP3 Expression Vectors in HEKSV293T Cells. Mutant and wild-type pcDNA3.1(–)MRP1k, pcDNA3.1(–)MRP2, and pcDNA3.1(+)MRP3 DNA expression vectors were transfected into simian virus 40-transformed human embryonic kidney (HEK293T) cells. Cells were maintained in Dulbecco's modified Eagle's medium supplemented with 4 mM l-glutamine and 7.5% fetal bovine serum. Approximately 18 x 106 cells were seeded per 150-mm plate, and 24 h later cells (50–75% confluence) were transfected with 20 µg of DNA using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. After 48 h at 37°C, the HEK293T cells were collected and stored as cell pellets at –70°C until needed.
Membrane Vesicle Preparation. Transfected cell pellets were thawed and disrupted by argon cavitation at 300 psi, and membrane vesicles were prepared as described previously (Loe et al., 1996b; Létourneau et al., 2005b). Membrane vesicles were aliquoted and stored at –70°C until required and used within 3 months of preparation. Protein concentrations were determined using a Bradford assay (Bio-Rad, Mississauga, ON, Canada) with bovine serum albumin as a standard.
Determination of MRP1, MRP2, and MRP3 Protein Levels in Transfected Cells. Levels of MRP1, MRP2, and MRP3 proteins in the membrane vesicles were determined by immunoblot analysis. Briefly, proteins were resolved on a 7% SDS-polyacrylamide gel and electrotransferred to a polyvinylidene difluoride membrane (Pall Corporation, Pensacola, FL). Immunoblots were blocked with 4% (w/v) skimmed milk powder in Tris-buffered saline containing Tween 20 for 1 h followed by incubation with the human MRP1-specific murine mAb QCRL-1 (1:10,000), the MRP2-specific murine mAb M2I-4 (1:10,000), or the MRP3-specific murine mAb M3II-9 (1:7500) (Hipfner et al., 1994; Ito et al., 2001a; Oleschuk et al., 2003; Létourneau et al., 2005a) diluted in the blocking solution. After washing, immunoblots were incubated with horseradish peroxidase-conjugated goat anti-mouse antibody (Pierce Chemical, Edmonton, AB, Canada) followed by application of Western Lightning chemiluminescence blotting substrate (PerkinElmer Life and Analytical Science) and exposed to film (Ultident, St. Laurent, QC, Canada). The films were analyzed by densitometry using Image J software (http://rsb.info.nih.gov/ij/).
MRP-Mediated Transport of 3H-Labeled Organic Anions by Membrane Vesicles. ATP-dependent uptake of 3H-labeled organic anion substrates by the MRP1-, MRP2-, and MRP3-enriched membrane vesicles was measured using a rapid filtration technique in a 96-well format as described previously (Tabas and Dantzig, 2002; Létourneau et al., 2005b). All reactions were carried out in a final reaction volume of 30 µl with 250 mM sucrose and 50 mM Tris-HCl (pH 7.5) buffer (TSB) containing AMP or ATP (2 mM), MgCl2 (10 mM), creatine phosphate (10 mM), and creatine kinase (100 µg/ml) under conditions shown previously to measure transport in the linear range of vesicular uptake (Létourneau et al., 2005a). Uptake was stopped by rapid dilution in ice-cold TSB, and the reaction mixtures were filtered using a FilterMate Harvester and Unifilter-96 GF/B filter plate apparatus (Packard BioScience, Meriden, CT). Radioactivity on the filters was quantified by liquid scintillation counting. All data were corrected for the amount of 3H-labeled substrate that remained bound to the filter, which was usually <10% of the total radioactivity. Transport in the presence of AMP was subtracted from transport in the presence of ATP to determine ATP-dependent uptake. Values for the negative controls (membrane vesicles of untransfected cells) were subtracted from uptake measured for wild-type and mutant MRPs. All transport assays were carried out in triplicate, and results are expressed as means (±S.D.).
For MRP1 activity, LTC4 uptake was measured by incubating [3H]LTC4 (50 nM, 10 nCi) with 2 µg of vesicle protein for 60 s at 23°C. E217ßG uptake was measured by incubating [3H]E217ßG (400 nM, 20 nCi) with 2 µg of vesicle protein for 60 s at 37°C. MTX uptake was measured by incubating [3H]MTX (100 µM, 125 nCi) with 5 µg of vesicle protein for 20 min at 37°C (Létourneau et al., 2005a). For MRP2 activity, LTC4 uptake was measured by incubating [3H]LTC4 (50 nM, 20 nCi) with 6 µg of vesicle protein for 3 min at 23°C. E217ßG uptake was measured by incubating [3H]E217ßG (400 nM, 100 nCi) with 6 µg of vesicle protein for 5 min at 37°C. MTX uptake was measured by incubating [3H]MTX (1 µM, 125 nCi) with 6 µg of vesicle protein for 10 min at 37°C (Ito et al., 2001a; Létourneau et al., 2005a). For MRP3 activity, LTC4 uptake was measured by incubating [3H]LTC4 (500 nM, 50 nCi) with 5 µgof protein at 37°C for 5 min. E217ßG uptake was measured by incubating [3H]E217ßG (400 nM, 70 nCi) with 5 µg of vesicle protein for 5 min at 37°C. MTX uptake was measured by incubating [3H]MTX (1 µM, 175 nCi) with 8 µg of vesicle protein for 10 min at 37°C (Oleschuk et al., 2003; Létourneau et al., 2005a).
[3H]LTC4 Photolabeling of MRP1 and MRP2. Membrane proteins were photolabeled with [3H]LTC4 as described previously (Loe et al., 1996b; Koike et al., 2004). Briefly, in a final volume of 50 µl, membrane vesicles (50 µgof protein) were incubated with [3H]LTC4 [for MRP1, 0.2 µM, (0.08 µCi) and for MRP2, 2 µM (0.8 µCi)] and 10 mM MgCl2 for 30 min at room temperature and then frozen in liquid nitrogen. Samples were then alternately irradiated at 302 nm for 1 min using a CL-1000 UV cross-linker (DiaMed, Mississauga, ON, Canada) and snap-frozen in liquid nitrogen 10 times. Radiolabeled proteins were resolved by SDS-PAGE, and proteins were fixed in a solution of water-isopropanol-acetic acid (65:25:10) for 30 min and washed in Amplify NAMP100 (Amersham, Baie d'Urfé, QC, Canada) for 30 min. After drying, the gel was exposed to Bioflex MSI film (InterScience, Markham, ON, Canada) for 5 days at –70°C. The films were analyzed by densitometry using Image J software as before. [3H]LTC4 photolabeling of MRP3 was not measured because the affinity of this transporter for this substrate is too low (Zeng et al., 2000; Zelcer et al., 2001; Oleschuk et al., 2003).
Photolabeling of MRP1, MRP2, and MRP3 by [-32P]8N3ATP. Membrane vesicle proteins were photolabeled with [-32P]8N3ATP as described previously (Leslie et al., 2001; Koike et al., 2004). Briefly, membrane vesicles (10 µg) were incubated at 4°C (nonhydrolysis conditions) with 5 mM MgCl2 and 5 µM[-32P]8N3ATP (1 µCi) in a final volume of 20 µl. Membrane vesicles prepared from untransfected HEK293T cells were included as a negative control. After 5 min of incubation on ice, samples were cross-linked at 302 nm in a 96-well plate for 8 min, washed twice with ice-cold Tris/EGTA/MgCl2 buffer (50 mM Tris-HCl, pH 7.4, 0.1 mM EGTA, and 5 mM MgCl2), and then solubilized in Laemmli buffer and subjected to SDS-PAGE. After drying, the gel was exposed to film for 2 to 12 h.
Orthovanadate-Induced Trapping of [-32P]8N3ADP by MRP1, MRP2, and MRP3. Membrane vesicle proteins (10 µg) were incubated for 15 min at 37°C (hydrolysis conditions) in 20 µl of TSB containing 5 mM MgCl2,5 µM [-32P]8N3ATP (1 µCi), and 1 mM freshly prepared sodium orthovanadate (Koike et al., 2004). For labeling MRP3, 10 µg of vesicle protein was also incubated with 5 mM CoCl2, 1 mM sodium orthovanadate, and 25 µM [-32P]8N3ATP (5 µCi). Membrane vesicles prepared from untransfected HEK293T cells were included as negative controls. The reactions were stopped by the addition of ice-cold Tris/EGTA/MgCl2 buffer and washed twice, and then resuspended membrane proteins (15 µl) were transferred to a 96-well plate and exposed to UV light at 302 nm on ice for 8 min. Membrane vesicles were then solubilized in Laemmli buffer and subjected to SDS-PAGE as before. After drying, the gel was exposed to film for 12 to 24 h.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). To determine whether this expression was also true for the corresponding MRP2-Pro1158Ala and MRP3-Pro1147Ala mutants, these mutations were created in pcDNA3.1(±)-based MRP2 and MRP3 expression vectors and then transfected into HEK293T cells. Membrane vesicles were prepared from the transfected cells, and the expression levels of each mutant were determined by immunoblotting. As shown in Fig. 2, all three Pro to Ala mutants (MRP1-Pro1150Ala, MRP2-Pro1158Ala, and MRP3-Pro1147Ala) were expressed at levels comparable to those of their corresponding wild-type proteins, confirming that the Pro residue at this position is not required for expression of the MRP proteins in the plasma membrane of mammalian cells.
|
0.1 µM), followed by MRP2 (Km 1 µM) and then MRP3 (Km 5 µM) (Leier et al., 1994; Loe et al., 1996b; Cui et al., 1999; Zeng et al., 2000). Consequently, the specific assay conditions for each transporter were modified accordingly (Loe et al., 1996b; Ito et al., 2001b; Oleschuk et al., 2003; Létourneau et al., 2005a). Furthermore, to take into account the background transport activity of HEK293T cells, uptake in vesicles from untransfected cells was subtracted from the uptake in vesicles from transfected cells.
As we reported previously, LTC4 transport by MRP1-Pro1150Ala was decreased by approximately 50% (Fig. 3A) (Koike et al., 2004). For MRP2-Pro1158Ala, LTC4 transport was also decreased by 50% whereas it was significantly increased (1.8-fold) for MRP3-Pro1147Ala (Fig. 3A). However, analysis of the latter data were complicated by the relatively high level of [3H]LTC4 transport observed in the vesicles made from untransfected cells when MRP3 assay conditions in which the initial LTC4 concentration is 10-fold higher than that for the MRP1 and MRP2 assays were used (data not shown). Thus, because of its much higher affinity for LTC4, the low level of endogenous MRP1 present in the HEK293T cells was suspected to be responsible for this non-MRP3-mediated transport. For this reason, the LTC4 uptake assays for MRP3-Pro1147Ala were repeated in the presence of the mAb QCRL-3, which specifically inhibits MRP1 transport activity (Loe et al., 1996b). Under these conditions, a 1.6-fold increase in LTC4 uptake by the MRP3-Pro1147Ala mutant was still observed compared with that for wild-type MRP3, but background LTC4 uptake by the vesicles from untransfected cells was decreased by approximately 90%. These observations confirm that, as suspected, the LTC4 transport activity in vesicles from untransfected cells was mainly due to endogenous MRP1. They also confirm that LTC4 transport by the MRP3-Pro1147Ala mutant is increased relative to that of wild-type MRP3.
|
In contrast to LTC4 uptake, E217ßG transport by MRP2 and MRP3 was unaffected by the Pro1158Ala and Pro1147Ala mutations, respectively (Fig. 3B). On the other hand, E217ßG transport by MRP1-Pro1150Ala was significantly increased (2-fold) relative to wild-type MRP1 as expected (Fig. 3B) (Koike et al., 2004).
Also as expected for MRP1-Pro1150Ala, MTX transport was significantly increased by approximately 3-fold (Fig. 3C) (Koike et al., 2004). In contrast, MTX transport by the MRP2-Pro1158Ala mutant remained the same as that of wild-type MRP2 (Fig. 3C), whereas MTX transport by the MRP3-Pro1147Ala mutant was increased relative to that of wild-type MRP3 but to a lesser extent than was observed for MRP1-Pro1150Ala (1.5-fold versus 3-fold) (Fig. 3C).
|
10-fold difference between MRP1 and MRP2 in apparent Km for this substrate. As shown previously, [3H]LTC4 photolabeling of MRP1 was not affected by the Pro1150Ala mutation (Fig. 4A) (Koike et al., 2004). Similarly, MRP2-Pro1158Ala was photolabeled by [3H]LTC4 to the same extent as wild-type MRP2 despite the reduced LTC4 transport activity of the mutant (Fig. 4B). [3H]LTC4 photolabeling of MRP3-Pro1147Ala was not measured because the affinity of this transporter for this substrate is too low (Zeng et al., 2000; Zelcer et al., 2001; Oleschuk et al., 2003).
Mutations MRP1-Pro1150Ala, MRP2-Pro1158Ala, and MRP3-Pro1147Ala Do Not Affect [-32P]8N3ATP Binding but Decrease Vanadate-Induced Trapping of [-32P]8N3ADP. We previously showed that photolabeling of MRP1-Pro1150Ala by [-32P]8N3ATP was comparable to that of wild-type MRP1 (Koike et al., 2004). When MRP2-Pro1158Ala and MRP3-Pro1147Ala were photolabeled with [-32P]8N3ATP, the intensity of the signals observed with the mutant and wild-type proteins was also comparable (Fig. 5), suggesting that ATP can bind to the Pro mutants as well as it does to their wild-type counterparts.
|
). As reported previously, when MRP1-Pro1150Ala was exposed to vanadate (1 mM) and [-32P]8N3ATP under hydrolytic conditions (37°C), the trapping of [-32P]8N3ADP was very low compared with that of wild-type MRP1 (Fig. 6A) (Koike et al., 2004). When the MRP2-Pro1158Ala and MRP3-Pro1147Ala mutants were examined under the same conditions, a significant decrease (50%) in the level of vanadate-induced 8N3ADP trapping by MRP2-Pro1158Ala was also observed (Fig. 6B), whereas no trapping of nucleotide was detected with either wild-type MRP3 or its Pro1147Ala mutant (data not shown).
|
Because sodium orthovanadate could not support trapping of [-32P]8N3ADP by MRP3 in the presence of Mg2+, other divalent cations (Mn2+ and Co2+) were tested based on previous studies of several other ABC transporters (Cai et al., 2002; Ozvegy et al., 2002; Sauna et al., 2004). The highest level of vanadate-induced [-32P]8N3ADP trapping by MRP3 was achieved with Co2+, although the signal was relatively weak. Nevertheless, in the presence of Co2+, a significant decrease in vanadate-induced 8N3ADP trapping by the MRP3-Pro1147Ala mutant relative to that by wild-type MRP3 was observed (results not shown). A band of greater intensity was observed when the concentration of [-32P]8N3ATP was increased from5to25 µM, and, under these conditions, a 60% decrease in vanadate-induced trapping of [-32P]8N3ADP by MRP3-Pro1147Ala relative to that by wild-type MRP3 was detected (Fig. 6C). Thus, 8N3ADP trapping by the MRP3 mutant was decreased as observed with the MRP1-Pro1150Ala and MRP2-Pro1158Ala mutants, although under different conditions.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
; Deber and Therien, 2002). These changes are presumably related to the distinctive chemical characteristics of Pro in which the NH2 group is unavailable to take part in any H-bonding networks. In addition, the Pro ring occupies some of the space that would be filled by the side chains of neighboring amino acids and may thus induce conformation constraints on proteins (Richardson and Richardson, 1989). The complex phenotype of the MRP1-Pro1150Ala mutant and the fact that Pro1150 is highly conserved in the ABCC subfamily suggested to us that this residue might serve an important functional role in other members of this class of ABC transporters (Koike et al., 2004). Ala, like Pro, is an aliphatic amino acid, but, unlike Pro, its NH2 group is available for H-bonding and its small side chain is unlikely to induce any conformational restraints. For this reason, the consequences of replacing the Pro residues in MRP2 and MRP3 corresponding to MRP1-Pro1150 with Ala on the transport and nucleotide binding properties of the mutant proteins were examined.
As described earlier, the MRP1-Pro1150Ala mutant showed substrate-selective changes in its transport activity (Koike et al., 2004). Thus, the increased E217ßG transport activity of this mutant was accompanied by a 4-fold reduction in Km(E217ßG) and apparent Km(ATP). Together, these changes could conceivably explain the increased E217ßG transport observed for this mutant. In contrast, the reduced LTC4 transport by MRP1-Pro1150Ala was associated with a decreased Vmax for LTC4 transport, whereas no change in Km(ATP) was observed (Koike et al., 2004). The different effects of LTC4 and E217ßG on the Km(ATP) of the mutant MRP1 indicate that the interaction of MRP1 with ATP can be selectively influenced by the substrate being transported. This substrate-selective effect on nucleotide interactions may not be exclusive to MRP1 and related transporters because the ATPase activity of P-glycoprotein (ABCB1) is also reported to be stimulated by some, but not all of its transported substrates (Sharom, 1997; Loo and Clarke, 2005). However, although the ATPase activity of P-glycoprotein is increased by vinblastine and verapamil, these drugs do not affect its Km(ATP) (Ambudkar et al., 1992; Sarkadi et al., 1992). Thus, the precise relationship between substrate-stimulated ATPase activity and substrate-induced changes in Km(ATP) for MRP1, P-glycoprotein, and other ABC transporters is not clear.
As for MRP1-Pro1150Ala, the transport activities of the MRP2 and MRP3 Pro mutants were altered in a substrate-selective manner, demonstrating the conserved functional importance of a Pro residue at the NH2-proximal end of CL7. However, there was substantial variability with respect to the extent to which the transport of individual substrates was affected (Fig. 3). The present study is not the first to report differences in the functional consequences of mutating a conserved amino acid in MRP1, MRP2, and MRP3. We had previously shown that mutations of a conserved Trp residue in TM17 of MRP1, MRP2, and MRP3 dramatically altered the transport activity of all three transporters but, as with the Pro mutants, each Trp mutant exhibited a distinct pattern of changes. Thus, mutation of MRP1-Trp1246 eliminated E217ßG and MTX transport without affecting LTC4 transport (Ito et al., 2001b), whereas in addition to eliminating E217ßG and MTX transport, mutation of MRP2-Trp1254 also significantly decreased LTC4 transport (Ito et al., 2001a). MRP3-Trp1242 mutants displayed decreased MTX and LTC4 transport, but in contrast with the MRP1-Trp1246 and MRP2-Trp1254 mutants, E217ßG transport was increased at least 2.5-fold (Oleschuk et al., 2003). Primary sequence comparisons and helical wheel projections make it clear that the molecular environments of the conserved Trp residues differ significantly in each of the three transporters. Thus, it seems likely that these differences are responsible for the marked differences in the substrate specificities of the mutants. In contrast, the immediate environments of MRP1-Pro1150, MRP2-Pro1158, and MRP3-Pro1147 are predicted to be much more similar (Fig. 1). This prediction may, at least in part, explain why the phenotypes of the three Pro mutants, although distinct, are more similar to one another than to the phenotypes of the three Trp mutants.
In addition to the substrate-selective changes in transport activity caused by the Pro mutations in MRP1, MRP2, and MRP3, the interaction of the three transporters with nucleotide was also altered. Thus, although 8N3ATP binding to MRP1-Pro1150Ala, MRP2-Pro1158Ala, and MRP3-Pro1147Ala remained unchanged, the level of vanadate-induced 8N3ADP trapping was very low for all three mutant transporters. This finding suggests that the influence of these Pro residues on the catalytic activity of the three homologs is conserved and that mutating the conserved Pro residue somehow uncouples, at least in part, the vanadate-induced ADP trapping properties of the transporters from their substrate transport. The evidence that these two activities can be uncoupled supports the idea that ATP hydrolysis is responsible for resetting the transporter in a conformation that allows a new cycle of substrate binding and transport to begin, rather than being required for the transport process itself (Higgins and Linton, 2004; Dawson and Locher, 2006).
Reduced levels of vanadate-induced trapping of 8N3ADP have been observed in other TM15/CL7 mutants of MRP1. Thus, MRP1 mutants of Arg1138 and Arg1142 exhibited reduced vanadate-induced trapping of 8N3ADP and, as shown for the Pro mutants in the present study, this was not associated with an overall reduction in transport activity, because transport of some substrates was comparable with wild-type activity (Conseil et al., 2006). However, the Pro1150, Arg1138, and Arg1142 MRP1 mutants differ from a double CL7 mutant described recently by Ren et al. (2006) in which Glu1157 and Gly1161 were mutated to Leu and Pro, respectively. Thus, the Glu1157Leu/Gly1161Pro mutant displayed both decreased vanadate-induced 8N3ADP trapping as well as decreased 8N3ATP photolabeling at both NBDs (Ren et al., 2006). On the other hand, LTC4 transport by this double mutant, like that for the Pro mutants in this study and the other CL7 mutants mentioned above, was markedly reduced. Unfortunately, the Glu1157Leu/Gly1161Pro mutant was not tested with other substrates, so it is not known whether the decreased transport activity of this mutant was substrate-selective. Nevertheless, despite some differences, these studies together with the present study all indicate that residues in CL7 can influence the activity of the NBDs of MRP1. CL7 mutations have also been shown to affect the function of other ABCC proteins, including MRP2 (ABCC2), CFTR (ABCC7), MRP6 (ABCC6), and SUR1 (ABCC8), which are responsible for the genetic disorders known as Dubin-Johnson syndrome, cystic fibrosis, pseudoxanthoma elasticum, and persistent hyperinsulinemic hypoglycemia of infancy, respectively (Conseil et al., 2005). Thus, the integrity of CL7 is important not only for the activity of the drug-transporting MRP proteins but also for the physiological functions of other ABCC proteins. These observations support the idea that this region may have a common role as a signaling conduit between the membrane spanning domains where substrate recognition and transport occurs and the NBDs where the energy for the transport process is generated.
Based on their crystal structure of BtuCD, a bacterial ABC importer of vitamin B12, Locher et al. (2002) have suggested that a cytoplasmic loop ("L loop") in the BtuC subunit is critical for transmitting signals from the membrane spanning domains to the NBDs (BtuD subunit) upon substrate binding. Amino acid sequence alignments of the L loop in BtuC suggest that it might correspond to CL4 and CL6 of human MRP1. However, sequence similarities also exist between the L loop of BtuC and the "EAA loop" in the CL4 of CFTR, which is also present in CL5 and CL7 in MRP1 (Locher et al., 2002; Ren et al., 2006). Thus, mutagenesis studies to date suggest that the proposed signaling role of the BtuC L loop could be fulfilled by CL7 in MRP1 and the analogous region in related ABCC proteins. The precise role of CL7 is not yet clear, but it is reasonable to suggest that it might interact with NBD2 because many of the MRP1 CL7 mutations affect vanadate-induced [-32P]8N3ADP trapping, which occurs primarily at NBD2 in MRP1 (Gao et al., 2000).
Finally, it could be presumed, based on the high degree of sequence similarity between MRP1/MRP2 and MRP3, that their interactions with nucleotide would be the same. Our observation that Mg2+ does not support trapping of 8N3ADP by MRP3 was therefore unexpected, particularly in view of the fact that vesicular transport and 8N3ATP photolabeling could be readily measured in the presence of this divalent cation. Thus, it seems that the differences between MRP1/ MRP2 and MRP3 may lie in their interactions with the vanadate anion because Mg2+-dependent transport and 8N3ATP photolabeling were observed for all three proteins and only the extent to which the proteins can trap 8N3ADP in the presence of vanadate was different. Several earlier reports have shown that other ABCC proteins [e.g., MRP4 (ABCC4) and rat Mrp6 (Abcc6)], as well as the more distantly related half-transporter BCRP (ABCG2), require divalent cations other than Mg2+ to detect 8N3ADP trapping (Cai et al., 2002; Ozvegy et al., 2002; Sauna et al., 2004). As observed for ABCG2 and MRP4, we found that Co2+ was more effective than Mg2+ and Mn2+ at supporting trapping of 8N3ADP by MRP3. These differences in cation dependence not withstanding, these and other data (I. J. Létourneau, A. Nakajima, R. G. Deeley, and S. P. C. Cole, manuscript in preparation) favor the conclusion that the conserved Pro residue in CL7 of all three transporters has a significant influence on the activity of NBD2.
In conclusion, our findings indicate that the role of MRP1-Pro1150, MRP2-Pro1158, and MRP3-Pro1147 is only partially conserved among the three MRP-related transporters. Thus, MRP1-Pro1150, MRP2-Pro1158, and MRP3-Pro1147 differ in their influence on the substrate specificity of these homologous transporters but share a common influence on nucleotide interactions at NBD2 that appears uncoupled from substrate transport.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
ABBREVIATIONS: MRP, multidrug resistance protein; ABC, ATP-binding cassette; TM, transmembrane; NBD, nucleotide binding domain; GSH, glutathione; CL, cytoplasmic loop; LTC4, leukotriene C4;E217ßG, 17ß-estradiol 17ß-D-glucuronide; MTX, methotrexate; kb, kilobase; HEK, human embryonic kidney; mAb, monoclonal antibody; TSB, Tris-sucrose buffer; PAGE, polyacrylamide gel electrophoresis.
1 Current affiliation: St. Jude Children's Research Hospital, Department of Pharmaceutical Sciences, Memphis, Tennessee. 
Address correspondence to: Dr. Susan P. C. Cole, Division of Cancer Biology and Genetics, Cancer Research Institute, Queen's University, Kingston, ON, Canada K7L 3N6. E-mail: spc.cole{at}queensu.ca
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. J. Slot, D. D. Wise, R. G. Deeley, T. J. Monks, and S. P. C. Cole Modulation of Human Multidrug Resistance Protein (MRP) 1 (ABCC1) and MRP2 (ABCC2) Transport Activities by Endogenous and Exogenous Glutathione-Conjugated Catechol Metabolites Drug Metab. Dispos., March 1, 2008; 36(3): 552 - 560. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
