Biomedical Engineering Reference
In-Depth Information
which is the glycophorin A homodimer. 33 As the helix turns, the two small glycine
residues end up on the same surface and could be involved in tight packing of the
helices in a dimer or higher-order multimer. Mutating these glycines to leucines had
a marked effect on ABCG2 function, consistent with impaired dimerization. 34 How-
ever, a role for these residues in substrate binding cannot be excluded without an
accurate model. Although in glycophorin A the GXXXG motif interacts with the
same motif in the dimerization partner, in view of the above-mentioned twofold sym-
metry seen in other ABC half-transporters, it seems unlikely that the TM1 segments
from two ABCG2 monomers would come together directly to form an interface in
the homodimer. Twofold symmetry has also been observed in the packing of trans-
membrane helices of full-length transporters, an example of which is TMs 5 and 8 of
Pgp being close together on the cytoplasmic side of the membrane in cross-linking
studies following cysteine-scanning mutagenesis. 35 Thus, an alternative explanation
for the role of GXXXG motif in TM1 of ABCG2 could be its involvement in forming
a higher-order oligomer, given that in one study, ABCG2 was suggested primarily to
form tetramers, with the possibility of dodecamers as the functional unit. 36
Another region potentially involved in dimerization is a conserved three-amino
acid sequence in TM5 (residues 552 to 554). Mutations in the corresponding residues
in the Drosophila white protein (an ortholog of ABCG2) are thought to disrupt
heterodimerization. 37 Further, the ABCG8 G574E mutant, which corresponds to
amino acid 551 of ABCG2, was also characterized as interfering with dimeriza-
tion. However, the ABCG8 G574R mutant, although demonstrating some effect on
ABCG8 maturation, did not prevent formation of the ABCG5/ABCG8 heterodimer.
In the case of ABCG2, substituting glycine 553 with either leucine or glutamic acid
resulted in decreased protein expression, impaired glycosylation, and retention in
the endoplasmic recticulum (ER). 38 Furthermore, the G553L mutant was not res-
cued from the ER when cotransfected with the wild-type protein, suggesting that no
mutant or wild-type dimers were formed. Again, these findings are consistent with
impairment in dimerization, although other explanations, such as improper folding,
cannot be excluded. On the other hand, when cotransfected with wild type, the inactive
L554P mutant partially reversed drug resistance in PA317 cells, a result that implied
that residue 554 is critical for function; yet, mutating this residue does not prevent
dimerization. 24 Altogether, these findings suggest a critical role for this region of the
fifth transmembrane helix.
Little is known about which residues of the ABCG2 TMD bind and translocate its
substrates. Similar to P-glycoprotein (ABCB1), 39 , 40 at least two drug-binding sites
within a larger drug-binding pocket have been suggested. This model is based on the
fact that certain substrates can abolish labeling with the photoaffinity analog IAAP,
whereas others have no effect. 41 Arginine 482, predicted to localize in TM3 near
the membrane-cytoplasm interface, is speculated to be part of one of these drug-
binding sites, given that replacing this residue by any nonpositively charged residue
results in wider substrate specificity. 42 , 43 The R482G and T variants were found in
drug-resistant cancer cell lines and described as gain-of-function mutants, with the
addition of substrates such as anthracyclines and rhodamine 123 to drugs transported
by the wild-type protein. 44
However, no single-nucleotide polymorphisms (SNPs)
 
Search WWH ::




Custom Search