点击显示 收起
Cardiff School of Biosciences, Cardiff University, Cardiff, United Kingdom
The brief report by Skinner-Adams et al. [1] on the antimalarial effects of HIV proteinase (referred to as "protease" by Skinner-Adams et al.) inhibitors is of great interest and potential therapeutic importance in areas where malaria and HIV infection are coendemic. We were interested by the views expressed by these authors, as well as those expressed by Savarino et al. in their correspondence [2], regarding the aspartic proteinases of Plasmodium falciparum (plasmepsins) as likely targets for inhibition by these compounds. The work of both groups in this field is valuable, but greater caution should be exercised in the discussion of structural and sequence similarities between the 2 enzymes, because an appreciation of the molecular intricacies of the aspartic proteinase family will be essential for the proper interpretation of the antimalarial effects of HIV proteinase inhibitors.
The plasmepsins belong to aspartic proteinase family A1 [3], in which the 2 catalytic aspartate residues are contributed by different domains, generally formed from a single amino acid chain in a monomeric enzyme. A1 enzymes have long active site trenches overarched with a single hairpin loop (the flap) and are underlaid by a sheet of 6 strands running approximately parallel to and beneath the active site trench. In contrast, HIV proteinase belongs to family A2 [3] and is a homodimeric enzyme that, consequently, has a symmetrical active site trench with 2 flaps. The active site trench of A2 enzymes is much shorter than that of A1 enzymes, and the dimer is held together on a base of 4 interdigitating strands, forming a sheet perpendicular to the active site trench. Thus, although there may be some important similarities in architecture between HIV proteinase and plasmepsins, to discuss them as being structurally similar is, in reality, misleading. Consequently, we do not share Savarino et al.'s surprise that no plasmodial homologue of the HIV proteinase was found, because there is no true equivalent: the 9 P. falciparum aspartic proteinases and the closely related histoaspartic proteinase [4] are members of the A1 proteinase family and are quite distinct from the A2 retroviral-type dimeric enzymes.
Furthermore, the sequence alignment of HIV proteinase and plasmepsin II presented by Savarino et al. shows no more similarity than would be expected for 2 diverse members of the aspartic proteinase superfamily. The regions surrounding the active site motifs of aspartic proteinases are highly conserved, with a characteristic hydrophobichydrophobicaspartic acidthreonine/serineglycine sequence. In fact, comparison of the leucineaspartic acidthreonineglycineserine sequence (plasmepsin II) with the isoleucineaspartic acidthreonineglycinealanine sequence (HIV proteinase) illustrates the differences between the enzymes as much as the similarities, because the serine in plasmepsin II is within hydrogen-bonding distance of the catalytic aspartic acid, whereas the alanine in the equivalent position in HIV proteinase cannot form such a bond.
The overall importance and interest of these structural distinctions is that, assuming the antimalarial activity of the antiretrovirals is mediated via the plasmepsins, the inhibitors described are actually interacting with active site trenches with very different architectureseven if the model produced by Skinner-Adams et al. indicates that some HIV proteinase inhibitors may adopt similar conformations on binding to both enzymes. These differences in the characteristics of the active site trenches are illustrated by distinct ligand-binding affinitiesfor example, the general aspartic proteinase inhibitor isovaleryl pepstatin inhibits plasmepsins with subnanomolar Ki values [5] but shows a Ki of 400 nmol/L against HIV proteinase [6].
The inhibition of plasmepsins by the HIV proteinase inhibitor ritonavir may not be surprising, because interactions between this drug and nonretroviral aspartic proteinases have been known for almost 10 years, with Ki values against human cathepsins D and E of 20 and 8 nmol/L, respectively (see Kempf et al. [7], whose work significantly predates that cited by Savarino et al. in showing such interactions). The plasmepsins are much more closely related to human cathepsin D than to HIV proteinase, so, when the development of antimalarial therapy based on HIV proteinase inhibitors is considered, the potential interactions with human aspartic proteinases may need to be minimized. For ritonavir in particular this may be of clinical relevance, because cathepsin D is a lysosomal enzyme and, thus, is present in most cell types and because cathepsin E is found in epithelial cells lining the gut (which are exposed to orally administered drugs) and is also present in red blood cellsthe clinically significant site of malarial infection. Saquinovir, by contrast, achieves potent inhibition of HIV proteinase with negligible inhibition of the human aspartic proteinases [8], demonstrating the selectivity that can be achieved between these types of enzymes. Ultimately, it may be the structural differences governing the interactions between inhibitors and the plasmepsins (on the one hand) versus the human aspartic proteinases (on the other) that hold the key to the development of existing drugs that target HIV proteinase as antimalarial agents.
References
1. Skinner-Adams TS, McCarthy JS, Gardiner DL, Hinton PM, Andrews KT. Antiretrovirals as antimalarial agents. J Infect Dis 2004; 190:19982000. First citation in article
2. Savarino A, Cauda R, Cassone A. Aspartic proteases of Plasmodium falciparum as the target of HIV-1 protease inhibitors . J Infect Dis 2005; 191:13812. First citation in article
3. Barrett AJ, Rawlings ND, Woessner JF. Handbook of proteolytic enzymes. London: Elsevier Academic Press, 2004. First citation in article
4. Coombs GH, Goldberg DE, Klemba M, Berry C, Kay J, Mottram JC. Aspartic proteases of Plasmodium falciparum and other parasitic protozoa as drug targets. Trends Parasitol 2001; 17:5327. First citation in article
5. Wyatt D, Berry C. Activity and inhibition of plasmepsin IV, a new aspartic proteinase from the malaria parasite Plasmodium falciparum. FEBS Lett 2002; 513:15962. First citation in article
6. Richards AD, Roberts R, Dunn BM, Graves MC, Kay J. Effective blocking of HIV-1 proteinase activity by characteristic inhibitors of aspartic proteinases. FEBS Lett 1989; 247:1137. First citation in article
7. Kempf DJ, Marsh KC, Denissen JF, et al. ABT-538 is a potent inhibitor of human immunodeficiency virus protease and has high oral bioavailability. Proc Natl Acad Sci USA 1995; 92:24848. First citation in article
8. Roberts NA, Martin JA, Kinchington D, et al. Rational design of peptide-based HIV proteinase inhibitors. Science 1990; 248:35861. First citation in article