The P-glycoprotein multidrug transporter
Frances J. Sharom1
Department of Molecular and Cellular Biology, University of Guelph, Guelph,
Ontario Canada N1G 2W1
Key words/phrases: ABCB1; ABCB4; ABCB11; multidrug resistance; chemotherapy; drug
distribution; blood brain barrier; modulators; X-ray crystal structure; substrate binding pocket;
drug transport; hydrophobic vacuum cleaner; flippase; ATP hydrolysis; vanadate; alternating
Frances J. Sharom is currently Professor and Tier 1 Canada Research Chair in Membrane Protein
Biology in the Department of Molecular and Cellular Biology at the University of Guelph. She
studied Chemistry at the University of Guelph, and then carried out her Ph.D. studies in
membrane biochemistry at the University of Western Ontario. She was a post-doctoral fellow
with Dr. Alan Mellors in the Department of Chemistry and Biochemistry at Guelph before taking
up the position of Assistant Professor there in 1980. Her research interests focus on the
fascinating role of membrane proteins, especially ABC proteins, in health and disease.
Footnote to p1
1 To whom correspondence should be addressed (email email@example.com)
Phone: +1-519-824-4120 ext 52247
P-Glycoprotein (ABCB1) is an ATP-powered efflux pump which can transport hundreds of
structurally unrelated hydrophobic amphipathic compounds, including therapeutic drugs,
peptides and lipid-like compounds. This 170 kDa polypeptide plays a crucial physiological role
in protecting tissues from toxic xenobiotics and endogenous metabolites, and also affects the
uptake and distribution of many clinically important drugs. It forms an important component of
the blood brain barrier, and restricts the uptake of drugs from the intestine. The protein is also
expressed in many human cancers, where it likely contributes to resistance to chemotherapy
treatment. Many chemical modulators have been identified that block the action of P-
glycoprotein, and may have clinical applications in improving drug delivery and treating cancer.
P-Glycoprotein substrates are generally lipid-soluble, and partition into the membrane before the
transporter expels them into the aqueous phase, much like a “hydrophobic vacuum cleaner”. The
transporter may also act as a “flippase”, moving its substrates from the inner to the outer
membrane leaflet. An X-ray crystal structure shows that drugs interact with P-glycoprotein
within the transmembrane regions by fitting into a large flexible binding pocket, which can
accommodate two substrate molecules simultaneously. The nucleotide-binding domains of P-
glycoprotein appear to hydrolyse ATP in an alternating manner, however, it is still not clear
whether transport is driven by ATP hydrolysis or ATP binding. Details of the steps involved in
the drug transport process, and how it is coupled to ATP hydrolysis, remain the object of
It was observed many years ago that cultured cells exposed to increasing concentrations of
cytotoxic drugs can develop resistance to their killing effects by several different mechanisms.
Victor Ling’s group noted that cells selected for resistance to colchicine also displayed resistance
to several structurally unrelated compounds, a phenomenon known as multidrug resistance
(MDR). In 1976, this resistance phenotype was linked to the presence in the plasma membrane
of a 170 kDa surface glycoprotein, which they named P-glycoprotein (Pgp; also known as
MDR1) for its ability to reduce the apparent cellular permeability of drugs . It soon became
clear that Pgp is involved in energy-dependent removal of drugs from the cell; in other words, it
is an active drug efflux pump. Following the discovery of Pgp, its expression was identified in
normal tissues, as well as many cultured cells of mammalian and human origin, including 39 of
60 tumour cell lines used by the U.S. National Cancer Institute in the discovery of new anti-
cancer drugs . Pgp is now the most studied and well-understood transporter in the ABC
superfamily. Pgp genes have been sequenced from human, mouse and Chinese hamster, among
others, and homologues have been identified in many other species, including Caenorhabditis
elegans and Drosophila melanogaster.
Tissue distribution and physiological role of Pgp
Although Pgp is expressed in most human and rodent tissues at low levels, it is found in much
higher amounts at the apical surface of epithelial cells of the large and small intestine, liver bile
ductules, and kidney proximal tubules , which all have excretory roles. The transporter is also
found in the adrenal gland, the placenta, and in the apical membrane of endothelial cells lining
the capillaries of the brain. These cells form a continuous monolayer, the so-called blood brain
barrier, which stops blood components from crossing into the central nervous system. Pgp is
oriented in these cells to transport substrates towards the blood, and is thus a major factor in
limiting their entry to the brain . Similarly, it appears to play a central role in protecting the
fetus and other sensitive tissues from the toxicity of a variety of endogenous and exogenous
molecules. This proposal is supported by studies on transgenic knockout mice lacking Pgp
(rodents have two Pgp genes, abcb1a and abcb1b, so both must be inactivated). These animals
have a normal life-span, and appear healthy and fertile. However, when Pgp substrates are
administered to them, the drugs accumulate to very high levels in the brain compared to wild-
type mice, resulting in neurotoxicity . For example, Pgp knockout mice are 100-fold more
sensitive to the toxicity of ivermectin, a pesticide commonly used in mouse colonies.
Pgp is now known to affect the disposition of many clinically administered drugs (discussed
in more detail below), and makes a major contribution to ADME (Absorption, Distribution,
Metabolism, Excretion). In the intestine, Pgp is responsible for extrusion of drugs into the
lumen, thus reducing their absorption and oral bioavailability. As expected, Pgp knockout mice
display increased uptake of substrates from the digestive tract, and markedly slower elimination
from the circulation, which for some drugs, leads to dramatically increased toxicity. Pgp
knockout mice have proved very useful for identifying or confirming drugs as substrates for the
transporter in vivo. For the most part, species differences between mouse and human with
respect to Pgp substrate specificity appear to be small.
Certain lineages of collie and other dog breeds have a frameshift mutation in the Pgp gene,
rendering the protein non-functional; they are also highly sensitive to ivermectin toxicity.
Despite the widespread clinical use of drugs that are Pgp substrates, there have been no reports
so far of non-functional Pgp variants in humans.
In addition to providing protection from harmful xenobiotics, Pgp may play a role in the
transport of endogenous molecules and metabolites (see ). Possible endogenous substrates
include phospholipids and glycolipids, platelet-activating factors, amyloid -peptides, and small
cytokines such as interleukins. Pgp may also export steroid hormones such as aldosterone and
progesterone from the adrenal gland and the uterine epithelium. However, little information is
available on the extent of Pgp-mediated transport of these endogenous molecules in vivo.
Close relatives of Pgp; ABCB4 and ABCB11
The human ABCB4 and ABCB1 genes are located adjacent to each other, and arose from a gene
duplication event. The two proteins are closely related (78% sequence identity), which suggests
that they likely share similar structures and mechanisms of action. The ABCB4 protein
(formerly known as MDR3 in human and Mdr2 in rodents) is a lipid translocator, which carries
out export of phosphatidylcholine (PC) from the apical (canalicular) membrane of hepatocytes
into the bile . PC present in the bile forms mixed micelles with bile salts, which otherwise
cause cellular damage because of their detergent-like properties. Transgenic knockout mice
lacking ABCB4 do not excrete any PC into the bile . The physiological function of ABCB4 is
crucial to human health, and a deficiency leads to progressive familial intrahepatic cholestasis
type 3, a severe liver disease resulting from decreased bile flow . ABCB4 probably operates
as an outwardly-directed “flippase”, moving PC from the inner to the outer leaflet of the
hepatocyte plasma membrane, where it can be extracted into the canalicular lumen by bile salts.
As we will see below, Pgp also demonstrates lipid flippase activity.
In 1995, another close relative of Pgp was identified, and named sister of P-glycoprotein
(Spgp). This protein, which is only expressed in the liver, turned out to function as an ATP-
dependent bile salt export pump and was thus renamed BSEP (ABCB11) . Inherited
mutations in ABCB11 result in severe cholestatic liver disease arising from reduced bile salt
secretion. Certain polymorphisms in ABCB11 appear to result in the protein being more
susceptible to inhibition by drugs, thus causing acquired cholestasis.
Pgp substrates and modulators
Pgp can interact with a vast array of compounds that are structurally unrelated, including natural
products, chemotherapeutic drugs, steroids, linear and cyclic peptides, fluorescent dyes and
ionophores. Most are weakly amphipathic and relatively hydrophobic, often (but not always)
containing aromatic rings and a positively-charged N atom. However, direct measurement of
transport has been carried out for only a few of these putative substrates, and most have been
identified based on resistance of cell lines overexpressing Pgp to their cytotoxic effects. Table 1
lists some of the different classes of Pgp substrates, and Figure 1 shows representative structures.
A second group of compounds can also interact with Pgp, known as modulators (also called
inhibitors or chemosensitizers) . Modulators are able to reverse MDR in intact cells by
interfering with the ability of Pgp to efflux drugs. Like substrates, Pgp modulators are
structurally diverse (Figure 1), and many have been identified over the years (see Table 2).
Modulators may inhibit Pgp transport function in a number of ways, and their mode of action at
the molecular level is not well understood. Some are transported themselves (e.g. cyclosporin A
), and thus act as competing substrates, while others (e.g. LY335979 ) bind tightly to the
protein’s drug binding pocket for prolonged periods of time without being transported. A few
modulators appear to interact with the nucleotide binding domains (NBDs) (e.g. steroids,
disulfiram ) and interfere with ATP hydrolysis. Several membrane fluidizers and surfactants
(e.g. benzyl alcohol, Nonidet P40, and Cremaphor EL ) do not appear to directly interact
with Pgp, and likely act as non-specific modulators at the level of the membrane bilayer. Pgp
modulators may be valuable for their ability to increase delivery of therapeutic drugs to the brain
, which is desirable in the treatment of diseases such as cancer, AIDS, Alzheimer’s disease,
Parkinson’s disease, schizophrenia and epilepsy.
Structure of Pgp
Pgp is a single polypeptide made up of two homologous halves that arose from gene duplication.
Each half comprises 6 transmembrane (TM) helices and one NBD located on the cytoplasmic
side of the membrane, which are connected by a cytoplasmic linker (Figure 2A). This topology
has been confirmed by Cys mutagenesis and epitope insertion experiments. The structure of Pgp
was first studied by electron microscopy (EM), and the highest resolution structure (8 D; see
Figure 2B), confirmed the presence of 12 TM helices and two closely associated NBDs . In
2009, X-ray crystal structures (3.8-4.4 D) of mouse Pgp were reported , both in the absence
of substrate, and with two stereoisomeric cyclic hexapeptide inhibitors bound to the transporter
(Figure 2C and D). Two interdigitated bundles of -helices were observed, each made up of
portions from both the N-terminal and C-terminal halves (TMs 1-3, 6, 10, and 11, and TMs 4, 5,
7-9, and 12). The long intracellular loops of one helix bundle contact the opposing NBD. This
crossover between the two halves of the transporter is also seen in the crystal structure of the
bacterial transporters Sav1866  and MsbA .
The two bundles of TM helices enclose a very large (6000 D2) cavity which can
accommodate at least two substrate molecules simultaneously. This cavity is contained within
the membrane, and is open to the inner leaflet via “portals”, likely to allow drug entry from the
lipid bilayer. The QZ59-RRR and QZ59-SSS peptide stereoisomers bind to distinct sub-sites in
the binding cavity in different orientations, and make different sets of contacts with amino acid
side chains of the protein. QZ59-RRR binds to a “middle” site, held in place by favourable
hydrophobic interactions with Phe, Tyr, Leu and Ile residues (Figure 2D). On the other hand,
two molecules of QZ5-SSS are found in different locations, “upper” and “lower” sites, within the
binding pocket (Figure 2D), and some polar residues are involved in interactions with the latter.
Thus, Pgp appears to bind multiple drugs by having a large, highly flexible binding cavity which
can accommodate many compounds in different locations by an “induced fit” type of
mechanism. Biochemical cross-linking and fluorescence studies had already pointed to a
substrate binding region with these properties [21,22]. The polyspecific nature of the Pgp
binding pocket, and its ability to bind more than one drug molecule simultaneously, makes the
rational design of specific high affinity inhibitors a challenging problem.
Drug pumping by Pgp
Characteristics of Pgp-mediated drug transport
Because cytotoxic Pgp substrates are generally lipophilic, they can enter cells readily by passive
diffusion across the plasma membrane, and then bind to their cellular target, which could be (for
example) microtubules or DNA. Early studies using drug-selected MDR cell lines expressing
Pgp showed that cellular drug accumulation was greatly reduced compared to the parent drug-
sensitive cell line, and similar observations were also made for cells transfected with the ABCB1
gene. Lower drug accumulation was observed to require cellular energy. Treatment with Pgp
modulators reversed drug resistance and restored both higher drug accumulation and
cytotoxicity, with the drug dose-response curve often moving 50- to 100-fold lower in the
presence of a compound like verapamil. Such changes in drug uptake can be conveniently
observed in intact cells using flow cytometry with fluorescent Pgp substrates, such as
daunorubicin or rhodamine 123 (e.g. ).
Intact cell systems proved to be too complex for dissection of the biochemical details of drug
transport, and studies were soon initiated using plasma membrane vesicles isolated from MDR
cells expressing Pgp. Inside-out vesicles transport drug into the vesicle interior in vitro when
supplied with ATP in the external medium, thus generating a drug concentration gradient across
the membrane, which is collapsed by inhibitors (e.g. ). Purified Pgp reconstituted into
proteoliposomes can similarly transport drug into the interior Transport is observed to be
saturable, requires ATP hydrolysis, and is inhibited by other transport substrates and modulators
(e.g. [25,26]). However, unexpected allosteric interactions are also observed between
transported substrates. For example, Hoechst 33342 stimulates the transport of rhodamine 123
and vice versa , likely because the two drugs bind simultaneously to different sub-sites
within the binding pocket. Such complex drug-drug interactions may be an important
consideration in clinical therapy, where two Pgp substrates are often co-administered.
Hydrophobic vacuum cleaner or lipid flippase?
Compounds that interact with Pgp are relatively hydrophobic and readily soluble in lipid
bilayers, and it is now widely accepted that they partition into the membrane before interacting
with the protein. Thus, Pgp has been described a “hydrophobic vacuum cleaner”  that expels
lipophilic molecules from the membrane into the extracellular medium (Figure 3). This pumping
action gives rise to a concentration gradient across the plasma membrane, with a higher drug
concentration in the external aqueous phase. The transporter is able to intercept substrates before
they have an opportunity to enter the cytosol, thus protecting the cell from exposure to
potentially toxic molecules. These ideas are supported by the Pgp X-ray crystal structures,
which show that two peptide stereoisomers are bound deep within the transmembrane helices,
suggesting that they may gain access to the protein from within the lipid bilayer .
Pgp binds lipid-like drugs and platelet-activating factor , which may be a physiological
substrate , and it also translocates fluorescently labelled phospholipids across the membrane
in an ATP-dependent fashion . Thus, the protein may function as a translocase or “flippase”
for lipophilic molecules (Figure 3) , moving them from the cytoplasmic to the extracellular
membrane leaflet. Given the high level of sequence similarity between Pgp and its close relative
ABCB4, which is a PC flippase, it is perhaps not surprising that they share some functional
attributes. The rate of spontaneous transbilayer movement of many Pgp substrates is relatively
low, and since they appear to localise discretely to one membrane leaflet, the transporter would
be able to maintain a higher drug concentration in the outer leaflet. This flippase activity would
also give rise to a substrate concentration gradient across the membrane, since drug in the outer
leaflet would rapidly equilibrate with the external medium. However, because of the rapid
partitioning equilibria involved, it is very difficult to distinguish between flippase activity and
direct transport of drug from the membrane to the extracellular aqueous phase.
ATPase catalytic cycle
Drug transport by Pgp is powered by hydrolysis of ATP at the two cytoplasmic NBDs, which are
characterized by three highly conserved sequence motifs . The Walker A and B motifs are
found in many proteins that bind and hydrolyse ATP and GTP, and the signature C motif is
unique to the ABC superfamily. Structural studies on bacterial ABC proteins led to the now
generally accepted concept that the NBDs must dimerize in order to hydrolyse ATP . In this
so-called “ATP sandwich dimer”, the NBDs are arranged in a head-to-tail arrangement, with two
ATP molecules bound along the interface. Each nucleotide is held in place by the Walker A and
B motifs of one NBD and the signature C motif of the other NBD, which thus form a composite
binding site . The Pgp crystal structure does not contain bound nucleotide, and since the two
NBDs are separated by 30 D , it does not provide any useful information about the putative
dimerization process. However, the signature C and Walker A motifs are known to be close to
each other from cross-linking studies .
It is not well understood how ATP hydrolysis is coordinated between the two NBDs, or how
this energy is used to drive drug transport . Drug binding to Pgp alters its basal ATPase
activity, which is relatively high even in the apparent absence of substrates. Some drugs
stimulate ATPase activity, while others inhibit it, and a biphasic pattern of stimulation at low
concentrations and inhibition at higher concentrations is common. It is still not known whether
Pgp hydrolyses one or two molecules of ATP for each drug molecule it transports.
Some important mechanistic insights were provided by the use of ortho-vanadate (Vi), an
inorganic phosphate analogue. Addition of Vi and ATP to Pgp leads to very rapid loss of
ATPase activity. After a single catalytic turnover, Pi dissociates and Vi takes its place, leading to
“trapping” of the stable complex ADP·Vi·Mg2+ in one NBD . The Vi-trapped complex is
very stable, and its structure is thought to resemble that of the pentacoordinate catalytic transition
state. Since trapping of Vi in one NBD abolishes catalytic activity at the other, this suggests that
both active sites must be functional for ATP hydrolysis to occur. This observation led to the
proposal of the “alternating sites” mechanism, in which only one NBD is catalytically active at
any instant in time, and the two active sites take turns in hydrolysing ATP .
In special circumstances where nucleotide hydrolysis is blocked (for example, after inactivation
of the NBDs by mutation, or the use of the poorly hydrolysable analogue ATPS), it has been
possible to isolate a Pgp intermediate with one very tightly-bound nucleotide (“occluded”) and
one loosely-bound nucleotide [39-41]. This species likely represents the E·S intermediate
normally present immediately before ATP hydrolysis occurs. The tightly-bound ATP is assumed
to have a “closed” dimer interface (Figure 4), and in the case of an active protein with bound
ATP it would progress rapidly to the transition state, leading to ATP hydrolysis. In contrast, the
dimer interface around the loosely-bound nucleotide is likely “open”. This information led to
proposal of a model in which Pgp always exists in an asymmetric state, with one closed (high
affinity) active site and one open (low affinity) active site (Figure 4). The tightly-bound ATP
molecule is committed to enter the transition state and undergo hydrolysis. Simultaneous
switching of the affinities as a result of ATP hydrolysis would result in alternation of catalysis
between the two NBDs. Movement of drug across the membrane may be driven by the energy
released during ATP hydrolysis, and would involve switching from an inward-facing
conformation with high drug binding affinity to an outward-facing conformation with low drug
binding affinity (Figure 4). Other models for the mechanism of ABC transporters have been
proposed (see Chapter 4) in which ATP binding alone drives drug transport (ATP switch model
), or two ATP molecules are hydrolysed sequentially to open both halves of the NBD dimer
interface (processive clamp model ). Clearly, more work will be needed to clarify the
mechanistic details of how the NBDs of Pgp hydrolyse ATP, and how drug transport is powered.
Importance of Pgp in drug therapy and disease
Interactions with clinically used drugs
Pgp interacts with many drugs in widespread clinical use (see Table 1), including a large number
of anti-cancer drugs, anti-histamines, immunosuppressive agents, anti-epileptics, cardiac
glycosides, anti-hypertensives, HIV protease inhibitors, calcium channel blockers, antibiotics,
and cholesterol-lowering statins (for a review, see ). The clinical effectiveness of these
drugs is greatly affected by Pgp, which alters their absorption and tissue distribution .
Interaction of a drug with Pgp can cause poor uptake in the intestine, thus reducing oral
bioavailability, and prevent delivery of drugs to the brain, which is a serious problem in the
treatment of brain diseases. The co-administration of two drugs that are both Pgp substrates can
also lead to major pharmacokinetic effects as they compete for the transporter . Plasma drug
levels stay higher for longer, and a reduction in drug dose is often necessary to avoid toxicity.
Similar effects are observed when a Pgp substrate drug is consumed with foods or herbal
supplements containing natural products that are also Pgp substrates (e.g. plant flavonoids, St.
John’s Wort ). For these reasons, drug testing for Pgp interactions is now recommended by
the U.S. Food and Drug Administration (F.D.A.) as part of its approval process . Such tests
are usually carried out using polarized monolayers of epithelial cell lines either transfected with
human Pgp, or expressing Pgp naturally. Movement of the test compound from the medium on
the basolateral side of the monolayer to the medium on the apical side (B-A, passive diffusion
plus Pgp-mediated efflux) is compared to movement in the opposite direction (A-B, passive
diffusion only). This is a time-consuming and expensive assay, and there is currently a pressing
need for new rapid, high throughput methods.
Over 50 polymorphisms (single nucleotide polymorphisms and insertions/deletions) in the
ABCB1 gene are known , and some of them appear to change the mRNA expression, protein
expression and function of Pgp. Polymorphisms may be responsible for the variation in drug
responses observed between different individuals and populations, and in recent years, there has
been considerable interest in how they might affect the outcome of drug therapy in people
carrying them. However, there have been many conflicting reports in this field, and no clear
associations between genotype and altered response to drug treatment have emerged .
Similarly, polymorphisms have been reported to alter the susceptibility of certain individuals to
various disease states, including colon cancer, renal cancer, inflammatory bowel disease, and
Parkinson’s disease (reviewed in ). This is a rapidly developing field, and will require
substantial further investigation before firm conclusions can be reached.
Pgp in drug discovery
The presence of Pgp in the intestinal epithelium is a serious problem in drug discovery, since
new drug candidates may be poorly absorbed, making them ineffective clinically. It is especially
important to screen out Pgp substrates when developing drugs targeted to the brain, since their
efficacy depends on their ability to cross the blood brain barrier. Many pharmaceutical
companies now routinely include testing for interactions with Pgp as part of their drug discovery
process. Methods that are employed range from measurements of drug uptake in intact cells to
in vitro assays that assess the effect of a compound on Pgp ATPase or transport activity in
membrane vesicles . Predictive methods have been of limited value in determining whether
a compound is likely to be a substrate for Pgp, although this may change as we gain a better
understanding of its drug binding pocket.
Resistance to chemotherapy treatment
Resistance of tumours to the cytotoxic action of chemotherapy drugs is the single most common
cause of cancer treatment failure. Some cancers, including tumours of the kidney, liver, breast
and ovary, are drug-resistant at the outset (intrinsic resistance), whereas others, typically
leukemias, lymphomas and multiple myeloma, develop resistance after one or more rounds of
chemotherapy treatment (acquired resistance). For several cancers, including acute myelogenous
leukemia (AML), acute lymphoblastic leukemia (ALL), and ovarian tumours, high Pgp
expression levels are strongly linked to a weak response to chemotherapy treatment and poor
overall prognosis [51,52]. However, in other cases it has proved difficult to link MDR in cancer
to Pgp expression, likely because there are multiple mechanisms by which some tumours can
develop drug resistance.
There has been much interest in combining modulators with chemotherapy drugs to improve
the outcome of cancer treatment . However, only a few of the hundreds of compounds
identified as modulators in vitro proved suitable for use in clinical trials  (Table 2). First
generation modulators, which were already in use for other medical conditions (e.g. verapamil),
were tested in the 1980s and generally proved too toxic to patients. They were modified to
produce second generation compounds, which often showed adverse pharmacokinetic
interactions in which decreased clearance of anti-cancer agents led to toxicity. In recent years,
more rationally designed third generation modulators of low toxicity have been produced, which
are highly selective and potent inhibitors of Pgp transport function. One notable success has
been the combined use of modulators and chemotherapy to cure childhood retinoblastoma .
However, the results of clinical trials using Pgp modulators in cancer treatment have generally
been disappointing [51,53,54], possibly because these studies suffered from serious experimental
limitations such as poor patient selection criteria. There is current interest in exploring the use of
natural components of foods/plants as Pgp modulators (e.g. flavonoids, curcuminoids; see Table
2), and they could serve as useful leads for the development of fourth generation agents .
This review provides a brief overview of what we currently know about the structure, function
and mechanism of the Pgp drug efflux pump. This transporter is of central importance in
protecting sensitive tissues from toxic xenobiotics, and can also interfere with the delivery of
clinically used drugs. As indicated, rapid progress has been made in recent years in elucidating
the 3-dimensional structure and ATP hydrolysis cycle of Pgp. However, much remains to be
learned in understanding how so many different compounds interact with the protein’s flexible
substrate-binding pocket, and how ATP hydrolysis powers drug transport. The many molecular
tools now at our disposal should provide insights into the involvement of this fascinating protein
in both normal human physiology and drug delivery. Such information may lead to the
development of novel approaches to manipulate Pgp function in a way that will improve the
outcome of clinical therapy.
The primary role of the Pgp multidrug efflux pump appears to be protection of sensitive
tissues from the toxic effects of xenobiotics and endogenous compounds, and it may also be
involved in exporting physiologically important molecules, such as platelet-activating factor.
Pgp can interact with a vast array of molecules of divergent structure; most are relatively
hydrophobic, and many possess aromatic rings and a positively-charged N atom.
Pgp can transport many drugs that are in common use clinically, and plays a major role in
their absorption and distribution; its localisation in the intestine and blood brain barrier
reduces oral uptake of drugs and prevents their entry into the brain.
Pgp knockout mice are a very useful tool for identification of Pgp substrates in vivo.
Compounds known as modulators can block the drug efflux activity of Pgp, and may have
important clinical applications in cancer therapy and drug delivery to the brain.
The structure of the protein is known from X-ray and electron microscopic studies; two
bundles of 6 TM helices form a large flexible substrate-binding pocket with different subsites
where one or two drug molecules are held primarily by hydrophobic interactions.
Pgp is thought to operate as either a hydrophobic vacuum cleaner that pumps lipophilic
molecules out of the membrane, or a flippase that moves substrates from the cytoplasmic to
the extracellular membrane leaflet.
The catalytic cycle of Pgp is thought to involve dimerization of the NBDs to form a
nucleotide sandwich with two bound ATP molecules, followed by ATP hydrolysis at only one
active site, which drives drug transport. The two NBDs appear to alternate in hydrolysing
F.J.S. is a Tier 1 Canada Research Chair in Membrane Protein Biology, and her work on P-
glycoprotein is supported by an operating grant from the Canadian Cancer Society. Due to space
constraints, only representative references have been cited throughout this review.
1 Juliano, R.L. and Ling, V. (1976) A surface glycoprotein modulating drug permeability in
Chinese hamster ovary cell mutants. Biochim. Biophys. Acta 455, 152-162
2 Alvarez, M., Paull, K., Monks, A., Hose, C., Lee, J.S., Weinstein, J., Grever, M., Bates, S.
and Fojo, T. (1995) Generation of a drug resistance profile by quantitation of mdr-1/P-
glycoprotein in the cell lines of the National Cancer Institute Anticancer Drug Screen. J.
Clin. Invest. 95, 2205-2214
3 Thiebaut, F., Tsuruo, T., Hamada, H., Gottesman, M.M., Pastan, I. and Willingham, M.C.
(1987) Cellular localization of the multidrug-resistance gene product P-glycoprotein in
normal human tissues. Proc. Natl. Acad. Sci. U. S. A. 84, 7735-7738
4 Schinkel, A.H. (1999) P-glycoprotein, a gatekeeper in the blood-brain barrier. Adv. Drug
Deliv. Rev. 36, 179-194
5 Schinkel, A.H., Smit, J.J., van Tellingen, O., Beijnen, J.H., Wagenaar, E., van Deemter, L.,
Mol, C.A., van der Valk, M.A., Robanus-Maandag, E.C. and te Riele, H.P. (1994)
Disruption of the mouse mdr1a P-glycoprotein gene leads to a deficiency in the blood-brain
barrier and to increased sensitivity to drugs. Cell 77, 491-502
6 Eckford, P.D. and Sharom, F.J. (2009) ABC efflux pump-based resistance to chemotherapy
drugs. Chem. Rev. 109, 2989-3011
7 Ruetz, S. and Gros, P. (1994) Phosphatidylcholine translocase: a physiological role for the
mdr2 gene. Cell 77, 1071-1081
8 Smit, J.J., Schinkel, A.H., Oude Elferink, R.P., Groen, A.K., Wagenaar, E., van Deemter, L.,
Mol, C.A., Ottenhoff, R., van der Lugt, N.M. and van Roon, M.A. (1993) Homozygous
disruption of the murine mdr2 P-glycoprotein gene leads to a complete absence of
phospholipid from bile and to liver disease. Cell 75, 451-462
9 Oude Elferink, R.P. and Paulusma, C.C. (2007) Function and pathophysiological importance
of ABCB4 (MDR3 P-glycoprotein). Pflugers Arch. 453, 601-610
10 Stieger, B., Meier, Y. and Meier, P.J. (2007) The bile salt export pump. Pflugers Arch. 453,
11 Robert, J. and Jarry, C. (2003) Multidrug resistance reversal agents. J. Med. Chem. 46,
12 Saeki, T., Ueda, K., Tanigawara, Y., Hori, R. and Komano, T. (1993) Human P-
glycoprotein transports cyclosporin A and FK506. J. Biol. Chem. 268, 6077-6080
13 Dantzig, A.H., Law, K.L., Cao, J. and Starling, J.J. (2001) Reversal of multidrug resistance
by the P-glycoprotein modulator, LY335979, from the bench to the clinic. Curr. Med.
Chem. 8, 39-50
14 Loo, T.W. and Clarke, D.M. (2000) Blockage of drug resistance in vitro by disulfiram, a
drug used to treat alcoholism. J Natl. Cancer Inst. 92, 898-902
15 Woodcock, D.M., Linsenmeyer, M.E., Chojnowski, G., Kriegler, A.B., Nink, V., Webster,
L.K. and Sawyer, W.H. (1992) Reversal of multidrug resistance by surfactants. Br. J.
Cancer 66, 62-68
16 Breedveld, P., Beijnen, J.H. and Schellens, J.H. (2006) Use of P-glycoprotein and BCRP
inhibitors to improve oral bioavailability and CNS penetration of anticancer drugs. Trends
Pharmacol Sci 27, 17-24
17 Rosenberg, M.F., Callaghan, R., Modok, S., Higgins, C.F. and Ford, R.C. (2005) Three-
dimensional structure of P-glycoprotein: the transmembrane regions adopt an asymmetric
configuration in the nucleotide-bound state. J. Biol. Chem. 280, 2857-2862
18 Aller, S.G., Yu, J., Ward, A., Weng, Y., Chittaboina, S., Zhuo, R., Harrell, P.M., Trinh,
Y.T., Zhang, Q., Urbatsch, I.L. and Chang, G. (2009) Structure of P-glycoprotein reveals a
molecular basis for poly-specific drug binding. Science 323, 1718-1722
19 Dawson, R.J. and Locher, K.P. (2006) Structure of a bacterial multidrug ABC transporter.
Nature 443, 180-185
20 Ward, A., Reyes, C.L., Yu, J., Roth, C.B. and Chang, G. (2007) Flexibility in the ABC
transporter MsbA: Alternating access with a twist. Proc. Natl. Acad. Sci. U. S. A. 104,
21 Loo, T.W. and Clarke, D.M. (2005) Do drug substrates enter the common drug-binding
pocket of P-glycoprotein through "gates"? Biochem. Biophys. Res. Commun. 329, 419-422
22 Sharom, F.J. (2007) Multidrug resistance protein: P-glycoprotein. In Drug Transporters:
Molecular Characterization and Role in Drug Disposition (You, G. and Morris, M.E. Eds.),
pp. 223-262, John Wiley & Sons, Hoboken, NJ.
23 Scala, S., Akhmed, N., Rao, U.S., Paull, K., Lan, L.B., Dickstein, B., Lee, J.S., Elgemeie,
G.H., Stein, W.D. and Bates, S.E. (1997) P-glycoprotein substrates and antagonists cluster
into two distinct groups. Mol. Pharmacol. 51, 1024-1033
24 Doige, C.A. and Sharom, F.J. (1992) Transport properties of P-glycoprotein in plasma
membrane vesicles from multidrug-resistant Chinese hamster ovary cells. Biochim.
Biophys. Acta 1109, 161-171
25 Sharom, F.J., Yu, X. and Doige, C.A. (1993) Functional reconstitution of drug transport and
ATPase activity in proteoliposomes containing partially purified P-glycoprotein. J. Biol.
Chem. 268, 24197-24202
26 Lu, P., Liu, R. and Sharom, F.J. (2001) Drug transport by reconstituted P-glycoprotein in
proteoliposomes. Effect of substrates and modulators, and dependence on bilayer phase
state. Eur. J. Biochem. 268, 1687-1697
27 Shapiro, A.B. and Ling, V. (1997) Positively cooperative sites for drug transport by P-
glycoprotein with distinct drug specificities. Eur. J. Biochem. 250, 130-137
28 Higgins, C.F. and Gottesman, M.M. (1992) Is the multidrug transporter a flippase? Trends
Biochem. Sci. 17, 18-21
29 Eckford, P.D. and Sharom, F.J. (2006) P-glycoprotein (ABCB1) interacts directly with
lipid-based anti-cancer drugs and platelet-activating factors. Biochem. Cell Biol. 84, 1022-
30 Raggers, R.J., Vogels, I. and van Meer, G. (2001) Multidrug-resistance P-glycoprotein
(MDR1) secretes platelet-activating factor. Biochem. J. 357, 859-865
31 Romsicki, Y. and Sharom, F.J. (2001) Phospholipid flippase activity of the reconstituted P-
glycoprotein multidrug transporter. Biochemistry 40, 6937-6947
32 Oswald, C., Holland, I.B. and Schmitt, L. (2006) The motor domains of ABC-transporters -
What can structures tell us? Naunyn Schmiedebergs Arch. Pharmacol. 372, 385-399
33 Smith, P.C., Karpowich, N., Millen, L., Moody, J.E., Rosen, J., Thomas, P.J. and Hunt, J.F.
(2002) ATP binding to the motor domain from an ABC transporter drives formation of a
nucleotide sandwich dimer. Mol. Cell 10, 139-149
34 Jones, P.M. and George, A.M. (1999) Subunit interactions in ABC transporters: towards a
functional architecture. FEMS Microbiol. Lett. 179, 187-202
35 Loo, T.W., Bartlett, M.C. and Clarke, D.M. (2002) The "LSGGQ" motif in each nucleotide-
binding domain of human P-glycoprotein is adjacent to the opposing Walker A sequence. J.
Biol. Chem. 277, 41303-41306
36 Callaghan, R., Ford, R.C. and Kerr, I.D. (2006) The translocation mechanism of P-
glycoprotein. FEBS Lett. 580, 1056-1063
37 Urbatsch, I.L., Sankaran, B., Weber, J. and Senior, A.E. (1995) P-glycoprotein is stably
inhibited by vanadate-induced trapping of nucleotide at a single catalytic site. J. Biol.
Chem. 270, 19383-19390
38 Senior, A.E., al-Shawi, M.K. and Urbatsch, I.L. (1995) The catalytic cycle of P-
glycoprotein. FEBS Lett. 377, 285-289
39 Tombline, G. and Senior, A.E. (2005) The occluded nucleotide conformation of P-
glycoprotein. J. Bioenerg. Biomembr. 37, 497-500
40 Siarheyeva, A., Liu, R. and Sharom, F.J. (2010) Characterization of an asymmetric
occluded state of P-glycoprotein with two bound nucleotides: implications for catalysis. J.
Biol. Chem. 285, 7575-7586
41 Sauna, Z.E., Kim, I.W., Nandigama, K., Kopp, S., Chiba, P. and Ambudkar, S.V. (2007)
Catalytic cycle of ATP hydrolysis by P-glycoprotein: evidence for formation of the E.S
reaction intermediate with ATP-gamma-S, a nonhydrolyzable analogue of ATP.
Biochemistry 46, 13787-13799
42 Higgins, C.F. and Linton, K.J. (2004) The ATP switch model for ABC transporters. Nat.
Struct. Mol. Biol. 11, 918-926
43 Janas, E., Hofacker, M., Chen, M., Gompf, S., Van der Does, C. and Tampe, R. (2003) The
ATP hydrolysis cycle of the nucleotide-binding domain of the mitochondrial ATP-binding
cassette transporter Mdl1p. J. Biol. Chem. 278, 26862-26869
44 Sharom, F.J. (2008) ABC multidrug transporters: structure, function and role in
chemoresistance. Pharmacogenomics 9, 105-127
45 Zhou, S.F. (2008) Structure, function and regulation of P-glycoprotein and its clinical
relevance in drug disposition. Xenobiotica 38, 802-832
46 Staud, F., Ceckova, M., Micuda, S. and Pavek, P. (2010) Expression and function of p-
glycoprotein in normal tissues: effect on pharmacokinetics. Methods Mol Biol. 596, 199-
47 Borrelli, F. and Izzo, A.A. (2009) Herb-drug interactions with St John's wort (Hypericum
perforatum): an update on clinical observations. AAPS. J. 11, 710-727
48 Giacomini, K.M., Huang, S.M., Tweedie, D.J., Benet, L.Z., Brouwer, K.L., Chu, X.,
Dahlin, A., Evers, R., Fischer, V., Hillgren, K.M., Hoffmaster, K.A., Ishikawa, T., Keppler,
D., Kim, R.B., Lee, C.A., Niemi, M., Polli, J.W., Sugiyama, Y., Swaan, P.W., Ware, J.A.,
Wright, S.H., Yee, S.W., Zamek-Gliszczynski, M.J. and Zhang, L. (2010) Membrane
transporters in drug development. Nat. Rev. Drug Discov. 9, 215-236
49 Choudhuri, S. and Klaassen, C.D. (2006) Structure, function, expression, genomic
organization, and single nucleotide polymorphisms of human ABCB1 (MDR1), ABCC
(MRP), and ABCG2 (BCRP) efflux transporters. Int. J. Toxicol. 25, 231-259
50 Sharom, F.J. and Siarheyeva, A. (2008) Functional assays for identification of compounds
that interact with P-gp. In Multidrug Resistance: Biological and Pharmaceutical Advances
in Antitumour Treatment (Colabufo, N.A. Ed.), pp. 261-290, Research Signpost,
51 Polgar, O. and Bates, S.E. (2005) ABC transporters in the balance: is there a role in
multidrug resistance? Biochem. Soc. Trans. 33, 241-245
52 Steinbach, D. and Legrand, O. (2007) ABC transporters and drug resistance in leukemia:
was P-gp nothing but the first head of the Hydra? Leukemia 21, 1172-1176
53 Szakacs, G., Paterson, J.K., Ludwig, J.A., Booth-Genthe, C. and Gottesman, M.M. (2006)
Targeting multidrug resistance in cancer. Nat. Rev. Drug Discov. 5, 219-234
54 Coley, H.M. (2010) Overcoming multidrug resistance in cancer: clinical studies of P-
glycoprotein inhibitors. Methods Mol. Biol. 596, 341-358
55 Chan, H.S., DeBoer, G., Thiessen, J.J., Budning, A., Kingston, J.E., O'Brien, J.M., Koren,
G., Giesbrecht, E., Haddad, G., Verjee, Z., Hungerford, J.L., Ling, V. and Gallie, B.L.
(1996) Combining cyclosporin with chemotherapy controls intraocular retinoblastoma
without requiring radiation. Clin. Cancer Res. 2, 1499-1508
Table 1. Compounds that interact with P-glycoprotein (substrates)
Taxanes; paclitaxel, docetaxel
Vinca alkaloids; vinblastine, vincristine
Tyrosine kinase inhibitors
Calcium channel blockers
HIV protease inhibitors
Table 2. P-glycoprotein modulators
Figure 1. Structures of some Pgp substrates and modulators
Figure 2. Topology and structure of Pgp
(A) Topological model of Pgp, showing the two homologous halves, each with 6 TM helices and
one NBD. (B) Medium resolution cryo-EM structure of hamster Pgp bound to the non-
hydrolysable nucleotide analogue, AMP-PNP. (C) High resolution X-ray structures of mouse
Pgp bound to a single molecule of the cyclic peptide substrate QZ59-RRR (left panel, peptide in
green; generated from pdb 3G60) and two molecules of its stereoisomer QZ59-SSS (right panel,
peptides in red and yellow; generated from pdb 3G61). The approximate location of the
membrane is indicated by the coloured bar. (D) Close-up views of the peptides inside the
substrate binding pocket; QZ59-RRR interacting with the middle site (left), and two molecules of
QZ59-SSS occupying the upper and lower sites (right).
Figure 3. The role of Pgp as a hydrophobic vacuum cleaner and a flippase
Figure 4. The Pgp catalytic and transport cycle
One proposed model of the catalytic and transport cycle of Pgp. In the symmetric state, both
halves of the dimer interface are “open” with each binding an ATP molecule loosely (ATPL).
The catalytically active pump rapidly progresses to the asymmetric state in which one ATP
molecule is tightly bound (ATPT) in one NBD, where the dimer interface is “closed”. The
tightly-bound ATP molecule is committed to enter the transition state and undergoes hydrolysis,
which provides the energy for movement of drug (red sphere) from the binding pocket within the
transporter to the extracellular space. Drug transport likely involves a conformational change
from an inward-facing to an outward-facing protein conformation. After ATP hydrolysis, ADP
is loosely bound (ADPL), resulting in opening of the dimer interface. The other catalytic site
simultaneously switches to the high affinity state, resulting in tight binding of the second ATP
molecule and closure of the dimer interface in the other NBD. Pi then dissociates from the
catalytic site of the first NBD, followed by ADP, which is replaced by another molecule of
loosely-bound ATP (nucleotide exchange) to achieve the asymmetric state again. A second
round of ATP hydrolysis and drug transport then takes place at the other NBD. Pgp exists in an
asymmetric state at all stages of the catalytic cycle, thus requiring that the two NBDs alternate in
hydrolysing ATP. This figure was adapted from Ref.  © the American Society for
Biochemistry and Molecular Biology.
Pgp substratesPgp modulators
Figure 4 Download full-text
and drug transport
ATP site affinity switch
ATP hydrolysis and
ATP site affinity switch