Pgp Mutational Analysis (Loo, 1999). Mutational analysis of Pgp has led to interesting discoveries concerning its structure and function. As previously mentioned, Pgp is composed of two linked subunits each containing an ATP binding site. The two ATP binding domains of Pgp are located at both the N terminal and C terminal end of the polypeptide (Figure 4).

Figure 4

When the subunits of the protein are coexpressed (wild type Pgp), only the N-terminal nucleotide binding domain is able to actively bind and hydrolyze ATP. However, when the polypeptide linker region between the two subunits is deleted and each subunit is expressed separately, both subunits are capable of hydrolyzing ATP (Figure 5).

Figure 5

In contrast, substrate-induced ATPase activity is seen only when both halves of Pgp are present. Drug-sensitive cell lines do not use Pgp to eliminate toxic chemicals unless both halves of the protein are linked. Deletion of the linker polypeptide between the two halves of Pgp will also inactivate the protein as an ATP-dependent pump (Figure 6). Normal Pgp function can be restored if the two subunits are reconnected using a stable polypeptide chain.

Figure 6

This data leads to two important conclusions. First, ATPase function is not coupled to substrate recognition, since both ATP binding domains are capable of hydrolyzing ATP. However, the two subunits of Pgp must be allowed to interact in order for proper substrate recognition and Pgp function to occur.

Pgp and Cellular Changes (Hammerle, 2000). Over-expression of Pgp has cellular effects which can be visualized by microscopy and Western blotting. MDR (Pgp over-expressing) cells begin growing in layers in cell culture much earlier than normal cells, and additional stress fibers can be detected. Pgp expressed in MDR cells is localized to the apical side of the top layer of cells, visual evidence supporting the role of Pgp as a drug efflux agent between tissue types. MDR cells also show a more extensive network of tight junctions, both at cell-cell contacts and in the cytoplasm compared to normal cells. In Figure 7 below, MDCK cells express the wild-type phenotype, while the MDR1-MDCK cells are overexpressing Pgp. Notice the pileup of blue-stained nuclei in the MDR cells, indicating multiple cell layers. The yellow stain was used to highlight regions of cell contact called tight junctions. In the MDR cells, the tight junction proteins are found in not only at cell-cell boundaries as in the MDCK cells, but also in the cytoplasm. Finally, the green actin staining demonstrates that Pgp is localized to the apical cell regions in MDR cells. This makes sense since such regions are most likely to contact harmful toxins.

Figure 7 (figure from Hammerle, 2000)

Western blotting of cell extracts of normal and MDR cells also revealed differences. Although Pgp was expressed in all cells, relative to normal cells, MDR cells grossly overexpress Pgp. In the Western blot probing for Pgp shown below as Figure 8, notice the extremely dark bands produced by the MDR cells as compared to the normal MDCK cells.

Figure 8 (figure taken from Hammerle, 2000)

Protein Folding (Loo, 1999) A protein folding mechanism for Pgp was elucidated using Cys replacement mutations. In the absence of drug substrates, the mutant proteins are retained within the endoplasmic reticulum during the folding process and are targeted for degradation by the chaperone proteins calnexin and Hsp70. The lack of Cys and subsequent lack of disulfide bonds leads to a conformational change in the unfolded and partially folded Pgp polypeptide. This change is somehow recognized by the cell's proofreading mechanisms and the mutant proteins are destroyed. However, in an interesting twist, upon exposure to Pgp substrate, the biosynthesis of mutant Pgp proteins ends in a normal product, not a degradation pathway. The mutant proteins are restored to a wildtype status even though their polypeptide sequence has not changed (Figure 9). This extraordinary "rescuing" phenomenon has several implications for Pgp folding. Substrate binding sites must be present in equilibrium even in mutant cells. The folding pathway for Pgp is complicated enough that the presence of substrate is capable of disrupting the equilibrium so that even mutant proteins can fold properly and be excreted to the cell membrane. Without the presence of substrate interactions, the mutant Pgp does not fold correctly and is eventually eliminated.

Figure 9