• P gp
• References
• Slides

P-glycoprotein (PDB ID:3G61)

Created by Azhar Ahmed

Biological Significance

   There are "at least three different kinds of molecular pumps that actively transport drugs out of the cell," of which P-glycoprotein (P-gp) is the most common in humans (1). P-gp is encoded in humans by the multidrug resistance 1 gene, so the protein is sometimes referred to as MDR1 (1). It can also be referred to as ABCB1 because it is a member of the ABC transporter family. Although P-gp (3G61) expels hydrophobic toxins ranging in size from 330 to 4000 Da, its ability to drive out therapeutic drugs is a point of interest for the scientific community. When a malignant cell is treated with an anti-cancer drug, wild-type P-gp in the membrane can actively remove the drug before it even enters the cytosol, rendering the therapy ineffective. P-gp can bind to a substrate once it has entered the inner leaflet of the cell membrane en route to the inside of the cell (1). Knowledge of the structural composition of P-gp can guide the design of multidrug resistance inhibitors and therapies for cancer, HIV, and parasitic diseases (1). It is also possible to target molecules that regulate P-gp activity in order to increase the effectiveness of anticancer therapies. For example, a paper recently released by the Scotto lab at the Cancer Institute of New Jersey reported a histone methyltransferase, called MLL1, involved in regulating P-gp activity, making it a "novel target for epigenetic therapy" (2).

   P-gp has other pharmacokinetic functions, including "drug absorption, renal secretion, biliary excretion, and brain distribution" (3). It acts in membranes of "physiological barriers such as intestinal epithelium, kidney tubule cells, the liver, and the capillary endothelium of the central nervous system" (3). P-gp is also involved in an apoptotic feedback loop. General apoptotic stimuli interact with cells and upregulate P-gp production, which in turn suppresses apoptotic signaling (2).

   P-gp  binds a foreign molecule called (4R,5R,11R,12R,18R,19S)-4,11,18-tris(1-methylethyl)-6,13,20-triselena-3,10,17,22,23,24-hexaazatetracyclo[17.2.1.1~5,8~.1~12,15~]tetracosa-1(21),7,14-triene-2,9,16-trione. This ligand has a molecular formula of C24H36N6O3Se3. In humans, P-gp has a molecular weight of 141.47811 kDa and an isoelectric point of 9.4389 (4).

Structure

   The ATP Binding Cassette transporter P-glycoprotein (P-gp) is a large transmembrane protein made up of two identical subunits, A and B. Each 170 kilodalton subunit contains 1284 amino acids and is comprised of an N-terminal and C-terminal half (10). These two halves have about 43% sequence similarity, resulting in a "toroidal shape and six-fold symmetry." The two halves can be two different polypeptide chains that are connected by a nonconserved linker region of about 60-80 amino acids. This linker region has a flexible alpha-helical structure necessary for functions such as ATP hydrolysis and substrate transport. The residues in the flexible linker region are not listed in the ICM model of 3G61.

   Each half is made up of a transmembrane domain and a nucleotide binding domain. The transmembrane domain contains six membrane-spanning R-helices. The amino acid sequences between two adjacent R-helices make up extracellular or intracellular loops. This slide portrays the secondary structure of P-gp with alpha-helices in red and beta-pleated sheets in yellow.

   The nucleotide binding domain is embedded in the cytosolic leaflet of the lipid bilayer. It is a highly conserved sequence of about 220 amino acids that is responsible for binding and hydrolyzing ATP. The nucleotide binding domain contains three functionally-significant segments. The N-terminal portion is the A-loop, which contains a notable Tyr401 residue that facilitates ATP binding. Unfortunately, the residue labeling feature could not be used because of discrepancies between the residue numbering of the program and a paper that was published in March of 2010. The pi electrons in the residue's phenyl ring associate with the basal ring of the adenine portion of ATP to stabilize the transitional conformation formed by ATP and its catalytic center. Neighboring the A-loop is the Walker A region, which is about eight amino acids long and contains many basic lysine residues. It contains the bulk of the residues that bind ATP. Each P-gp transporter binds two ATP molecules in the catalytic domain of the protein, which exists between the two nucleotide binding domains. The two halves of the protein must work cooperatively to hydrolyze ATP. One half is responsible for translocation and dissociation of substrate, while the other is responsible for changing back to the physiological conformation. The C-terminal portion is the Walker B region, which is about eleven amino acids long and contains many aspartic acid residues. The Walker B region is the main Mg2+ binding sequence in P-gp. Though their functions are not quite as prominent or well understood, a short Q loop and a lengthy ABC linker region can be found between the Walker regions. In addition, a D loop and H loop have been designated on the C-terminal side of the Walker B region.

   Structurally and functionally important residues in the N-terminal half of the protein include Leu-65 in TM1 (transmembrane segment 1), Ser-222 in TM4, Ile-306 in TM6, and Ala-342, Phe-343, and Leu-339 in TM6. Structurally and functionally important residues in the C-terminal half include Phe-728 in TM7, Ile-868 and Gly-782 in TM10, Phe-942 and Thr-945 in TM11, and Leu-975, Val-982, and Ala-985 in TM12. "Of the 73 solvent accessible amino acids in the internal cavity of P-gp, [only] 15 are polar," showing why P-gp mostly recognizes hydrophobic foreign materials (1). This slide portrays the overall polarity of P-gp. The internal cavity is spacious (about 6000 Å3 in volume) and funnel-shaped, narrowing towards the cytoplasmic end of the protein. P-gp has at least three substrate-binding sites and one allosteric site (10). The twelve transmembrane segments that make up the substrate-binding site are mobile enough to accommodate different numbers and types of substrates, enabling the protein to translocate more than one foreign substance with a single conformation change. This phenomenon is in accordance with the substrate-induced fit model.

   N-linked glycosylation can be observed near the N-terminus of the protein at Asn-91, Asn-94, and Asn-99. Glycosylation is not observed near the C-terminus, supporting the speculation that P-gp glycosylation is only useful for transport to the membrane and does not play a functional role once it is embedded in the membrane. These glycosylation sites can also play a role in recognition by a monoclonal antibody for P-gp.

   P-gp can be phosphorylated at Ser-661, Ser-667, Ser-671, and Ser-683, all of which are found in the linker region. Experiments have shown that phosphorylation of P-gp is not necessary for function and the effect it has on the protein has yet to be determined.