MDR transporters as therapeutic targets in cancer.

ATP-binding cassette (ABC) transporters comprise a superfamily of structurally related proteins that include multidrug resistance--P-glycoprotein (MDR1 P-gp), the cystic fibrosis transmembrane conductance regulator (CFTR), the transporter associated with antigen processing (TAP), and breast cancer resistance protein (BCRP). These molecules are located in the cell membrane, where they scan the inner membrane leaflet for lipids/xenobiotics and flip them to the outer membrane in an ATP-dependent process. In many human cancers, intrinsic and acquired resistance to chemotherapy is due to an overexpression of MDR1 that enhances drug efflux. Clinical trials have shown promising increases in remission duration and overall survival when chemotherapy is combined with MDR1 P-gp inhibitors.

**********

Mammalian ATP-binding cassette (ABC) transporters are a large superfamily of structurally related membrane proteins that help to actively transport a range of molecules across the plasma membrane, from the cytoplasm out of the cell or into organelles. The normal physiologic function of these transporters is protective to epithelial cells of the gastrointestinal tract, liver, and kidney, as well as to the capillaries of the brain, testes, and ovaries, where they provide a barrier to the uptake of xenobiotics and promote their excretion into bile and urine.

To date, more than 50 human ABC transporter genes have been sequenced, and they are classified into eight subfamilies, ABC-A to ABC-H. These proteins perform a variety of functions, including transmembrane transport and cellular signaling. Several ABC transporters also are associated with inherited genetic diseases.

The linking of membrane transporters to drug resistance was made in the late 1970s in Victor Ling's laboratory, where a glycoprotein of 170 kD, called the P-glycoprotein (P-gp, P for permeability), was shown to limit cellular accumulation and cytotoxicty of antineoplastic agents in a Chinese hamster ovary cell line. The gene for P-gp, called MDR1 (multi-drug resistance protein 1), encodes a 1280-residue polypeptide chain organized in two homologous halves. The protein's putative function as an ATP-dependent pump that expels small molecules from the cell interior was surmised from its sequence similarity with bacterial ABC transporters.

Pathogenic bacteria also contain ABC transporters and appear to use them to pump cytotoxic drugs from the cell, thus allowing the bacteria to resist antibiotic treatment. In fact, LmrA, a multidrug transporter from Lactococcus lactis, has a specificity for chemotherapy agents that is very similar to that of P-gp and can functionally substitute for P-gp in human lung fibroblast cells.

Reversing chemotherapy resistance mediated by MDR1 P-gp is envisioned as a way to improve treatment outcomes in tumors that inherently overexpress this protein. Although clinical trials thus far have not been conclusive, several trials have shown statistically significant improvements in overall survival with the use of a P-gp inhibitor combined with chemotherapy.

MDR Transporters Consist of Three Paired Domains

ABC transporters typically contain three paired domains, which include the highly conserved ATP-binding cassette. All ABC proteins contain at least one ABC domain.

Subfamilies that play a role in molecular transport contain membrane-spanning domains (MSD). Translocation proteins like MDR1 contain two MSDs, which function as a translocation channel for substrates to be transported across the membrane, and two nucleotide-binding domains (NBD), which bind ATP and provide energy for transport.

[ILLUSTRATIONS OMITTED]

Several ABC transporters have only one of each domain and must homo- or heterodimerize to form a functional transporter.

The three-dimensional structure for human MDR1 has not yet been determined by x-ray crystallography, but the crystal structure of the highly conserved ABC transporter Escherichia coli msbA highlights the topologic features of ABC transporters and provides a model to explore the mechanism of multi-drug resistance.

To ascertain structural information for MDR1, our laboratory built a homology model that is based on the high sequence similarity observed between the three-dimensional structure of E. coli MsbA and the human MDR1 protein.

MSD: The MSD is a cone-shaped structure. It comprises two sets of six transmembrane (TM) [alpha]-helices, which are tilted between 30[degrees] and 40[degrees] from the plane of the membrane. The transporter channel has a substantial opening on the cytoplasmic face of ~45 A in the widest dimension.

All 12 TM [alpha]-helices contribute to residues lining the channel and are highly solvated. TM2 and TM5, from opposing monomers within the dimer, form the major dimerization contact and are positioned to function as a hinge, allowing the transporter to undergo structural rearrangements. The inner membrane leaflet side of the channel consists of a cluster of positively charged basic residues, while the outer membrane side is more hydrophobic.

ICD: Three distinct regions of 25 to 50 residues in length come together between the MSD and nucleotide-binding domain (NBD) to form the intracellular domain (ICD). The ICD is composed of 3 [alpha]-helical bundles (ICD 1, 2, and 3) of ~125 residues which link TM6 to the NBD.

[ILLUSTRATION OMITTED]

NBD: On the cytoplasmic face of the transporter are two NBDs. These ATP-binding domains are the most conserved domains of the MDR transporter family and form an L-shaped molecule with two domains, a RecA-like domain and a small [alpha]-helical domain.

The RecA-like domain consists of two [beta]-sheets and six [alpha]-helices, which include the Walker A (P-loop), Walker B, Q-loop, D-loop, and the switch motif, forming the ATP-binding site. The Walker A motif has the consensus sequence GXXGXGKST (where X is any residue), which forms the P-loop that binds to the [alpha]- and [beta]-phosphates of ADP and ATP.

The Walker B motif forms a [beta]-strand that consists of four aliphatic residues followed by two negatively charged residues, aspartatic acid followed by glutamatic acid, and is postulated to coordinate the [Mg.sup.2+] ion through a water molecule.

[ILLUSTRATION OMITTED]

The Q-loop, or "lid," contains a glutamine that interacts with the [gamma]-phosphate through a water molecule and may be the attacking nucleophile. The D-loop resides between the Walker B and switch motifs. The switch motif contains a conserved histidine that may polarize the attacking water molecule for ATP hydrolysis.

The second smaller helical domain comprises four [alpha]-helices and contains the signature or C motif (consensus LSGGQ) that may function as a [gamma]-phosphate sensor in the opposing molecule of the dimer. The observed hinge motion of the helical domain with respect to the RecA-like domain may be relevant in signaling between the MSD and RecA-like domain.

ATP-Switch Model Uses Conformational Changes for Drug Transport

Previous models to explain the efflux of molecules through ABC transporters proposed that ATP hydrolysis provides the energy for transport and that two NBDs, operating in sequence, transport the substrate through the membrane. However, recent biochemical and structural studies suggest an alternative ATP-switch model. Recent reporting of the three-dimensional structure of human MDR1 using electron crystallography reveals that a major reorganization of the MSDs occurs throughout the entire depth of the membrane on binding of an ATP analog.

The transport cycle is initiated by the binding of a high-affinity substrate to the MSD, which lowers the activation energy for ATP-dependent NBD dimerization. ATP binding induces large conformational changes in the MSD and NBD, leading to formation of a closed dimer and establishing the power stroke for substrate transport.

Next, autocatalytic ATP hydrolysis induces the vectorial release of the substrate from the MSD to the extracellular space. [P.sub.i] and ADP are sequentially released due to destabilization caused by repulsive forces between the ADP bound to the Walker A motif of one NBD and the [P.sub.i] coordinated to the signature motif of the other NBD. This restores the open dimer configuration.

The transporter thus is reset to its basal state, until ADP-ATP exchange occurs again on substrate binding to the MSD and the cycle repeats. This entropically driven "flip-flop" mechanism may account for the unusually broad range of hydrophobic drugs and lipids transported by members of the MDR ABC-transporter family.

The location of the substrate-binding site(s) within the MSD is unknown, despite mutagenesis and photoaffinity labeling studies, and no ABC transporter has been co-crystallized with a substrate in place to elucidate the mode of binding and mechanisms of transport. Nevertheless, it is predicted that MDR1 P-gp recognizes different substrates in multiple overlapping binding sites in the MSD.

However, some ABC transporters have a single substrate-binding site, and a variable stoichiometry is predicted depending on the specific transporter and substrate(s). Evidence also suggests that for some ABC transporters, such as MDR1 P-gp, two ATP molecules are hydrolyzed per transport cycle.

Single Nucleotide Polymorphisms in MDR1 Alter P-gp Function

Single nucleotide polymorphisms (SNPs) have been shown to affect MDR/P-gp function by altering channel properties. Systemic screens of the human MDR1 gene have identified at least 28 SNPs from healthy volunteers, in 12 of the 28 exons distributed throughout the gene.

Some of these SNPs, for example the silent mutation C3435T, have been associated with altered transporter expression and function. Individuals with the CC genotype had ~2-fold higher P-gp expression than those with the TT genotype, while persons with the CT genotype had an intermediate level of expression.

In these patients, the risk of CNS relapse in childhood acute lymphoblastic leukemia is significantly reduced in patients with the MDR1 3435 TT or CT genotype as compared to the CC genotype. This difference in risk is most likely due to lower P-gp levels in the blood-brain barrier of TT and CT genotype persons as compared to the CC genotype, resulting in better CNS permeability of P-gp substrates used in the treatment of childhood ALL. The lower P-gp levels in TT and CT genotypes result in less drug efflux, improving CNS drug levels and drug exposure times.

A recent study investigated 405 patients with acute myelogenous leukemia for somatic genotypes of the three most frequent SNPs in exons 12, 21, and 26. The homozygous wild-type allele was classified as genotype A, heterozygous as B, and homozygous as C. Real-time PCR was performed to ascertain a potential regulatory effect of these SNPs on MDR1 expression.

MDR1 expression was strongly correlated with a decreased complete remission rate (p=0.01) but failed to predict a decreased overall survival. There was a significant association of the A genotype in exons 21 (p=0.05) and 26 (p<0.05) with lower MDR1 expression, and the B variants showed the highest MDR1 values at all three loci.

Linkage disequilibrium of the SNPs was detected and indicated that a combined polymorphism also could affect the regulation of MDR1 expression. The A genotype of all SNPs demonstrated both the lowest MDR1 values and significantly decreased overall survival (p<0.05) with a high probability of relapses (p<0.01).

These allelic variants of the MDR1 gene appear to influence therapy outcome by additional mechanism(s) potentially involving pharmacokinetic effects of P-gp.

Cellular Stress Induces Expression of Human MDR1

During stressful stimuli, such as exposure to chemotherapy, several signaling pathways have been identified that upregulate MDR1 gene expression. The importance of each of these pathways and their interrelationships, however, are not well known.

[ILLUSTRATION OMITTED]

Evidence indicates that p53 and MDR1 play important roles in chemoresistance. Mutations in p53 induce MDR1 promoter transactivation, leading to resistance to chemotherapy, whereas gene therapy that introduces wild-type p53 results in reduced MDR1 promoter expression.

Reactive oxygen species also have been shown to participate in MDR1 overexpression mediated by insulin, epidermal growth factor, tumor necrosis factor-[alpha], and doxorubicin. Upregulation of MDR1 via the nuclear factor-[kappa]B (NF-[kappa]B) pathway protects cells from apoptosis induced by reactive oxygen species by preventing mitochondria-activated caspase-3 elaboration.

The orphan nuclear receptor SXR/PXR, steroid receptor, xenobiotic receptor, and hypoxia inducible factor-1[alpha] (HIF-1[alpha]) also may mediate increased MDR1 expression regulated at the transcriptional level.

Abnormal Apoptotic Signaling in Cancer Cells Enhances MDR1 Activity

Apoptosis is a common response to cellular stress and a key part of the cytotoxic effect of many anticancer agents. Ceramide, a neutral glycosphingolipid, often plays a role in activating cell death signals initiated by cytokines, chemotherapeutic agents, and ionizing radiation.

Ceramide generation occurs in response to the postreceptor action of various cytokines, hormones, and growth factors, including TNF[alpha]. TNF[alpha] binding to its receptor (or binding of ligands to other "death receptors") stimulates the generation of sphingomyelinase, which in turn mediates the production of ceramide.

TNF binding can stimulate production of two forms of sphingomyelinase, neutral (N-SMase) or acid sphingomyelinase (A-SMase), and different domains of TNF[alpha] stimulate the different SMases.

Ceramide produced by either SMase can mediate a cell growth arrest or proapototic signal, depending on the presence of specific cofactors and activators. A-SMase has been recognized as one of the molecules mediating proapoptotic signaling in cell death induced by an array of stresses such as [H.sub.2][O.sub.2], heat, ultraviolet light exposure, and radiation.

[ILLUSTRATION OMITTED]

Activity of ceramide-activated protein kinase (CAPK), also known as kinase suppressor of Ras (KSR), is the switch point in the balance between proapoptotic and antiapoptotic signals. Activation of CAPK leads to further mitochondrial stimulation (promotes apoptosis) or association with Ras and activation of the Raf1 cascade, leading to proliferation or differentiation.

Ceramide is generated in cancer cells in response to treatment with anticancer drugs, such as doxorubicin, paclitaxel, vinblastine, etoposide, and dactinomycin. Data suggest that chemotherapy resistance in cancer cells in some instances is associated with an increased ability to inactivate ceramide via glycosylation. Glucosylceramide synthase, the enzyme that catalyzes ceramide glycosylation, plays a critical role.

Results of a 5760-gene cDNA microarray analysis of the human multiple myeloma cell line RPMI-8226 and its doxorubicin-selected P-gp-expressing sublines 8226/Dox6 and 8226/Dox40 (both of which express MDR1) identified 29 genes that participate in apoptotic signaling mediated by ceramide and mitochondrial permeability transition.

However, ceramide activity can be overcome through glycosylation of ceramide by glucosylceramide synthase. In fact, glucosylceramide has been shown to accumulate in doxorubicin-resistant breast carcinoma cells and vinblastine-resistant epithelioid carcinoma cells, suggesting that MDR1 activity is enhanced by glucosylceramide synthase.

Aberrant MDR Expression Occurs in Nonmalignant and Malignant Disease

Several mammalian ABC transporters are associated with inherited genetic diseases due to mutations that affect their respective transport functions. These include Tangier's disease (cardiovascular ), cystic fibrosis (lung and gut), Stargardt's macular dystrophy (eye), adrenoleukodystrophy (neurologic), Zellweger's syndrome (cerebrohepatorenal), and familial persistent hyperinsulinemic hypoglycemia (pancreas).

Moreover, overexpression of MDR1 P-gp and MRP (MDR-related protein) transporters is clinically important in that cancer cells acquire multidrug resistance, leading to treatment failure in both solid and hematologic malignancies.

The nature of anticancer drug resistance is multifactorial. However, MDR1 P-gp and MRP1 are detected in a significant proportion of untreated human malignancies, with ~40% to 50% of cases not having a clear and consistent association with stage of disease.

Exposure to chemotherapy increases the expression of both MDR1 and P-gp. In vitro studies on primary cancer cell cultures consistently demonstrate an association between P-gp (protein) or MDR1 (mRNA) expression and resistance to chemotherapy.

Current methods for MDR1 detection, which include immunohistochemistry, real-time reverse transcriptase-PCR, and flow cytometry, have been used alone and in combination to evaluate the expression of MDR1 P-gp in several human solid and hematologic malignancies.

Inhibiting MDR1 Might Reverse Drug Resistance in Cancer

P-gp is inherently overexpressed at baseline in chemotherapy-resistant tumors such as colon and kidney cancers, and its expression is upregulated following exposure to chemotherapy in leukemia and breast cancer. Numerous chemotherapeutic agents of diverse structure are susceptible to drug efflux via the P-gp pump and include the anthracyclins, vinca alkaloids, taxols, and epipodophyllotoxins.

During the past decade, in vitro studies have identified a variety of chemosensitizers that can decrease anticancer drug resistance of MDR1-positive cancer cells and frequently restore their normal sensitivity to chemotherapy. The concept that inhibiting drug efflux can potentiate chemotherapy effectiveness and reverse drug resistance has led to the development of several clinical P-gp inhibitors.

P-gp-reversing agents belong to diverse structural classes but are essentially hydrophobic and lipophilic. They include the calcium channel blockers (verapamil, nifedipine, diltiazem), immunosuppressives (cyclosporines), steroid hormones (progesterone, tamoxifen), antibiotics (ceftriaxone), protein kinase inhibitors (staurosporin, H87), and monocolonal antibodies (MRK16, UIC-2).

Of the first-generation agents, verapamil and cyclosporine A required doses severalfold higher than therapeutic doses in order to overcome MDR1 resistance, with consequent toxicity. Of the second-generation antagonists, PSC 833 is a nonimmunosuppresive cyclosporine A analog that is a superior chemosensitizer compared to the parent compound.

The development of third-generation P-gp inhibitors, such as XR9576 (tariquidar), a novel anthranilic acid derivative, has been evaluated in mice bearing the intrinsically resistant MC26 colon tumors. Combination of XR9576 with doxorubicin potentiated antitumor activity without a significant increase in toxicity. In addition, a combination of XR9576 with paclitaxel, etoposide, and vincristine fully restored activity against two highly resistant MDR human tumor xenografts (2780AD, H69/LX4) in nude mice. The combination of paclitaxel with XR9576 did not alter its plasma pharmacokinetics.

A second third-generation P-gp chemosensitizer of potential interest is LY335979 (zosuquidar trihydrochloride). When combined with mitoxantrone and vinorelbine using a panel of HL60 leukemia cell lines or MCF-7 breast cancer transfectants, it substantially reversed mitoxantrone resistance and fully reversed vinorelbine resistance of MDR1-expressing HL60/VCR cells (vincristine resistant). LY335979 had no effect on MRP1-expressing HL60/ADR (doxorubicin resistant) or BCRP-transfected MCF-7 breast cancer cells. These results demonstrate that LY335979 is highly specific for P-gp and does not modulate MRP1 or BCRP.

Novel approaches to determining P-gp overexpression clinically will play a major role in elucidating successful reversal of drug resistance. [sup.99m]Tc-Sestamibi, an imaging agent, is a P-gp substrate, and increased uptake has been observed in normal liver and kidney after administration of PSC 833 (a cyclosporine analog), VX710 (biricodar), and XR9576 (tariquidar). These studies indicate that the concentrations of modulator achieved in patients are able to increase uptake of [sup.99m]Tc-sestamibi.

Furthermore, [CD56.sup.+] cells obtained from patients treated with PSC 833 demonstrate enhanced rhodamine retention in an ex vivo assay. Finally, a subset of patients treated with P-gp antagonist show enhanced sestamibi retention in imaged tumors.

These results suggest that P-gp modulators can increase drug accumulation in P-gp-expressing tumors and normal tissues. Using third generation MDR1 P-gp antagonists in properly designed clinical trials, it should be possible to determine the concentration of modulators needed to achieve the reversal of clinical drug resistance.

Clinical Trials of P-gp Antagonists Have Been Inconclusive

Initial clinical trials of P-gp inhibitors evaluated the efficacy and safety of first-generation agents that were already in the clinic for other indications. Nine randomized trials in P-gp inhibition have been conducted to date, but only three have shown a statistically significant improvement in overall survival with the use of a P-gp inhibitor.

The Southwest Oncology Group (SWOG) evaluated cytarabine and daunorubicin with or without cyclosporine in 226 patients with refractory, relapsed, or high-risk acute myeloid leukemia (AML). Treatment with cyclosporine led to a statistically significant improvement in rates of relapse-free survival (34% vs 9%) and overall survival (22% vs 12%). Survival and response rates were related to higher serum levels of daunorubicin observed in the cyclosporine group. The maximum benefit in the cyclosporine group occurred in those with the highest levels of P-gp expression.

A trial of verapamil in patients with breast cancer (relapse rate 11% vs 27%) and lung cancer (18% vs 41%) demonstrated a survival benefit with the addition of the P-gp inhibitor. One major limitation of these trials was the reduction in anticancer drug doses that was required with concurrent administration of the P-gp inhibitor. The reduction in chemotherapy dose likely confounded the results and contributed to the disappointing outcomes.

The "third-generation" candidates now being tested should show whether inhibition of P-gp can potentiate chemotherapy efficacy and potency in both solid and hematologic malignancies. Improved diagnostic techniques aimed at the selection of patients with tumors that express P-gp should result in more successful outcomes. Further optimism is warranted with the advent of potent, nontoxic inhibitors and new treatment strategies, including the combination of new targeted therapies with therapies aimed at the prevention of drug resistance.

An area for future investigation is the targeting of specific components of the ABC transporter molecule, such as the membrane-spanning domain (MSD) or nucleotide-binding domain (NBD), using a structure-based design approach. Targeting the MSD will most likely provide channel-specific inhibitors, and targeting the NSB will provide ATP site inhibitors that may affect several members of the ABC family simultaneously.

However, specific targeting of the MSD without reliable structural data is challenging. Due to the availability of a wealth of structural data for the NBDs of several ABC transporters, targeting these domains using a structure-based design approach provides a opportunity for lead discovery utilizing the relatively easy in silico virtual screening of high-diversity small molecular databases for potential drug candidates.

Substantial data strongly suggest that MDR1 P-gp is important in clinical drug resistance. However, early clinical trials have used inferior inhibitors to the MSD and did not determine the level of P-gp expression. With the advent of potent nontoxic third-generation inhibitors, novel treatment strategies (including combinations of novel targeted therapies) and a better understanding of signaling pathways provide for enthusiasm to continue to target ABC transporters in drug resistance. Yet, a structure-based design approach to targeting this family of multidomain proteins holds promise.

Immune Regulatory Functions of MDR1

MDR1 is expressed on hematopoietic stem cells, natural killer cells, leukocytes, dendritic cells, and T and B lymphocytes. The expression of MDR1 in hematopoietic stem cells is dependent on the level of stem cell differentiation, suggesting MDR1 may have a potential role in immune regulation.

There is precedent for involvement of the ABC family of transporters in immune regulation. The transporter associated with antigen processing (TAP) is a member of the ABC transporter superfamily that transports cytosolic peptides for loading onto MHC class I molecules.

However, MDR1 may play a different role in the immune system, in that it has been shown to increase the release of interleukins 2 and 4 and interferon-[gamma] from lymphocytes and also potentiate dendritic cell migration.

For more on TAP, see the paper by Daniel Hicklin and Soldano Ferrone, HLA Class I Antigen Loss in Cancer, in the March/April 2002 issue of Science & Medicine.

Other Drug-Resistance Transporters

Observations of functional ATP-dependent drug efflux in multidrug-resistant cancer cell lines without the overexpression of P-gp has suggested the existence of other ABC transporters capable of causing cancer drug resistance.

Identification of the MDR-related protein (MRP/ ABC-C) subfamily has provided an additional mechanism of reduced drug accumulation, and the discovery of ABC-G2 has added a new dimension to drug resistance.

In a breast cancer cell line, the overexpression of a novel member of the G subfamily of ABC transporters was found. The new transporter was termed the breast cancer resistance protein (BCRP). BCRP is a 655-amino-acid polypeptide, formally designated as ABC-G2. Like all members of the ABC-G subfamily, BCRP is a half-transporter, synthesized at the gene level as a single copy, which dimerizes to form the complete channel protein.

Overexpression of BCRP in drug-sensitive MCF-7 or MDA-MB-231 cells mirrors the drug-resistance phenotype of MCF-7/AdrVp cells. This is consistent with evidence suggesting the functional BCRP is a homo-dimer.

The spectrum of anticancer drugs effluxed by BCRP includes mitoxantrone, camptothecin-derived and indolocarbazole topoisomerase I inhibitors, methotrexate, flavopiridol, and quinazoline ErbB1 inhibitors.

Potent and specific inhibitors of BCRP are in development and should become available for clinical use. Future prospective studies will incorporate BCRP protein or mRNA quantification with functional assays in order to assess the contribution of BCRP to drug resistance in human cancers.

[ILLUSTRATION OMITTED]

Publication date: 20 June 2005

RECENT REVIEWS

Lutz Schmitt and Robert Tampe: Structure and mechanism of ABC transporters. Curr Opin Struct Biol 12(6):754-760, Dec 2002.

Daruka Mahadevan and Alan F. List: Targeting the multidrug resistance-1 transporter in AML: molecular regulation and therapeutic strategies. Blood 104(7):1940-1951, 1 Oct 2004.

Christopher F. Higgins and Kenneth J. Linton: The ATP switch model for ABC transporters. Nat Struct Molec Biol 11(10):918-926, Oct 2004.

Suresh V. Ambudkar, Chava Kimchi-Sarfaty, Zuben E. Sauna, and Michael M. Gottesman: P-glycoprotein: from genomics to mechanism. Oncogene 22(47):7468-7485, 20 Oct 2003.

Susan Bates, Clara Chen, Robert Robey, Min Kang, William D. Figg, and Tito Fojo: Reversal of multidrug resistance: lessons from clinical oncology. Novartis Found Symp 243:83-96, 2002.

ORIGINAL PAPERS

John R. Riordan, Kathryn Deuchars, Norbert Kartner, et al: Amplification of P-glycoprotein genes in multidrug-resistant mammalian cell lines. Nature 316(6031):817-819, 29 Aug 1985.

Mark F. Rosenberg, Alhaji Bukar Kamis, Richard Callaghan, et al: Three-dimensional structures of the mammalian multidrug resistance P-glycoprotein demonstrate major conformational changes in the transmembrane domains upon nucleotide binding. J Biol Chem 278(10): 8294-8299, 7 Mar 2003.

Geoffrey Chang, Christopher B. Roth: Structure of MsbA from E. coli: a homolog of the multidrug resistance ATP binding cassette (ABC) transporters. Science 293(5536):1793-1800, 7 Sep 2001.

Kaspar P. Locher, Allen T. Lee, Douglas C. Rees: The E. coli BtuCD structure: a framework for ABC transporter architecture and mechanism. Science 296:1091-1098, 10 May 2002.

Paola Vergani, Steve W. Lockless, Angus C. Nairn, and David C. Gadsby: CFTR channel opening by ATP-driven tight dimerization of its nucleotide-binding domains. Nature 433(7028):876-880, 24 Feb 2005.

LAURENCE COOKE, MISCHALA GRILL, NIKHIL SHIRAHATTI, and DARUKA MAHADEVAN are in the University of Arizona Cancer Center, in Tucson, Arizona.

[e-mail]

ABC Transporter Superfamily

               Prior
Subfamily   Nomenclature    Members         Transport Molecules

ABC-A       ABC1           1-13        Cholesterol, phospholipids,
                                       steroids, lipids, surfactant
                                       proteins, tissue-specific
                                       proteins

ABC-B       MDR, * TAP     1-9, 10 &   Xenobiotics, peptides,
                           10P, 11     phosphatidyl choline, metals,
                                       heme, phospholipids, bile
                                       salts

ABC-C       CFTR, MRP, *   1-13        Organic anions, glutathione,
            MOAT, SUR                  conjugates, cyclic
                                       nucleotides, chloride,
                                       potassium, steroids,
                                       anticancer drugs

ABC-D       ALD, PXMP,     1P(1-4),    Fatty acids, fatty AcylCoAs
            PMP            2-4

ABC-E       OABP           1           Unknown

ABC-F       GCN20          1-13        Unknown

ABC-G       White          1-5, 8      Cholesterol, xenobiotics,
                                       organic anions, steroids

ABC-H       ANSA           1-2         Unknown

* MDR1 and MRP1, when overexpressed, confer resistance
to anticancer drug treatments. MRP, MDR-related protein.

MDR1 Mutations in Human Disease States

Disease                                Mutated ABC transporter

Tangier's disease                      ABC-A1
Neonatal respiratory failure           ABC-A3
Stargardt's macular dystrophy          ABC-A4 (ABCR)
Fundus flaximaculatus                  ABC-A4 (ABCR)
Colchicine sensitivity                 ABC-B1 (MDR1)
X-linked sideroblastic anemia          ABC-B7 (ABC7)
  and ataxia
Dubin-Johnson syndrome                 ABC-C3 (cMOAT/MRP2)
Cystic fibrosis                        ABC-C7 (CFTR)
Familial persistent hyperinsulinemic   ABC-C8 (SUR1)
  hypoglycemia
Adrenoleukodystrophy                   ABC-D1 (ALDP)
Zellweger's syndrome                   ABC-D3 (PMP70/PXMP3)

MDR1 P-gp Expression in Human Malignancies

Tumor Type       Percent Expression

Bladder cancer         67-86%
AML                    18-80%
Breast                 11-80%
Sarcoma                20-64%
Myeloma                6-50%
Lung                   27-45%
Lymphoma                15%

P-gp Antagonists Tested in Randomized Clinical Trials

Pg Inhibitor             Chemotherapy      Disease

Verapamil
                         Vindesine +       NLCLC
                         ifosfamide
                         CAVE              SCLC
                         VAD               Myeloma
                         Vindesine + 5FU   Breast cancer

Quinidine
                         Epirubicin        Breast cancer

Quinine
                         Mitoxantrone +    AML
                         cytarabine

Cyclosporine
                         VAD               Myeloma
                         Daunorubicin+     AML
                         cytarabine

                                Median Survival

                         No
Pg Inhibitor             Inhibitor         Inhibitor

Verapamil
                         22 wks            41 wks *

                         44 wks            40 wks
                         10 mos            13 mos
                         209 d             323 d *

Quinidine
                         59 wks            47 wks

Quinine
                         11.5 mos          11.5 mos

Cyclosporine
                         15 mos            13 mos
                         12% (2-yr)        22% (2-yr) *

* Statistically significant difference vs no inhibitor.
VAD=vincristine, Adriamycin (doxorubicin), dexamethasone;
CAVE=cyclophosphamide, doxorubicin, vincristine, etoposide.