Diss. (sammanfattning) Uppsala : Uppsala universitet, 2006. / Härtill 4 uppsatser.

Thesis (Ph. D.)--University of Washington, 2004. / Vita. Includes bibliographical references (leaves 127-141).

As the first structural example of this class of proteins, crystals of Pro-RIP and MOD were grown and diffracted to 2.4 and 2.5 A respectively. The structures of the two proteins are solved and found to be highly similar, with main chain RMSD of 0.519. Each protein has two domains. The N-terminal domain consists of five alpha-helices and five-stranded mixed beta-sheet. The conserved active site residues Y94, Y130, E207, R210 and W241, similar to those of other RIPs, are located at the cleft between the N-terminal and C-terminal domains. In Pro-RIP, the 25-amino acid internal inactivation region is found on the surface of the N-terminal domain and consists of a flexible loop followed by a long alpha-helix. Like bacterial ribosome-inactivating proteins, maize ribosome-inactivating protein does not have a back-up glutamate in the active site, which helps the protein to retain some activity if the catalytic glutamate is mutated. The structure of maize RIP reveals that the active site is too small to accommodate two glutamate residues and suggests that maize ribosome-inactivating protein may represent an intermediate product in the evolution of ribosome-inactivating proteins. / Pull-down assay indicated that the internal inactivation region diminished the interaction of Pro-RIP with purified ribosomes and ribosomal proteins P0, P1 and P2. Surface plasmon resonance assays showed that Pro-RIP has a slower association rate and faster dissociation rate on intact ribosomes when compared to MOD, resulting 80-fold decrease in binding affinity. These evidences strongly suggested that the reduced ribosome-inactivating activity and cytotoxicity of Pro-RIP is the result of its diminished interaction with the ribosomes. The ribosome binding site of MOD is found to be different from TCS and saporin, which are located between the anti-parallel beta-sheet in the C-terminal domain. In MOD, the positive-charged residues K158, K159, K160 and K161 that were found to be important for ribosome binding are located in the N-terminal domain, underneath the internal inactivation region. / Ribosome-inactivating proteins (RIPs) are rRNA N-glycosidases, which hydrolyze the N-glycosidic bond of A-4324 in 28S rRNA of eukaryotic ribosomes. Based on the number of subunits, RIPs are grouped into three classes. Type I RIPs (e.g. trichosanthin and saporin) are monomeric polypeptide with molecular weights of 25-32 kDa. Type II RIPs (e.g. ricin and cinnamomin) are heterodimeric proteins whose subunits are linked by a disulphide bridge, with molecular weights of 60-65 kDa. Chain A of type II RIPs is the catalytic subunit, while chain B is the lectin subunit, which facilitates the cellular entry of the protein by interacting with carbohydrates on the cell surface. Maize ribosome-inactivating protein is classified as a type III RIP, or an atypical RNA N-glycosidase. It is synthesized and stored in the kernel as a 34 kDa inactive precursor (Pro-RIP). During germination, the precursor undergoes proteolysis to generate a two-chain active RIP (MOD). Previous study has found that the 25-amino acid residues at the acidic internal inactivation region, which are removed during activation of Pro-RIP, is the major control element to suppress its in vitro protein synthesis inhibition activity. / Since the internal inactivation region of Pro-RIP controls the ribosome-inactivating activity and cytotoxicity, it provides an opportunity to engineer an on/off switch forits activity by HIV-1 protease through engineering HIV-1 protease recognition sites into the internal inactivation region of Pro-RIP. A variant that contains two HIV-1 protease recognition sites incorporated to the 25-amino acid internal inactivation region was found to be activated by HIV-1 protease in vitro. This variant entered cells more efficiently than Pro-RIP and was as cytotoxic as MOD. This switch may be applied to other RIPs such as ricin A chain and other protease recognition sequences may be used for increasing the specificity of an RIP toward viral infected cells. / The internal inactivation region of Pro-RIP greatly decreases its cytotoxicity, but not cellular uptake through alpha-2 macroglobulin receptor. On the contrary, the acidic residues within the region hinder fluid-phase endocytosis. Moreover, it is found that the internal inactivation region does not affect sub-cellular localization of the protein - MOD and Pro-RIP locate in the same cellular compartment (nucleus in JAR or cytoplasm in J774A.1 and C8166). / Mak, Nga Sze Amanda. / "July 2007." / Adviser: Shaw Pang Chui. / Source: Dissertation Abstracts International, Volume: 73-03, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (leaves 216-236). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

Palladium-catalysed coupling reactions have been embraced by synthetic chemists as one of the preferred means for smooth formation of new carbon-carbon bonds: a truly ubiquitous methodology of synthesizing complex molecules. This thesis describes the study of a series of palladium(0)-catalysed C2-arylations of a 1-cyclopentenyl ether, equipped with a chiral (S)-N-methyl-pyrrolidine auxiliary. The investigated olefin was demonstrated to undergo Si-face insertion, providing (R)-configuration of the arylated C2-carbon. In addition, the mild and novel palladium(II)-catalysed dominoHeck/Suzuki β,α-diarylation-reduction of a dimethylaminoethyl-substituted chelating vinyl ether was developed using arylboronic acids as arylating agents in combination with 1,4-benzoquinone (BQ). Further, highly regioselective palladium(II)-catalysed α-and β-monoarylation of the chelating vinyl ether was achieved using either a bidentate ligand or by employing ligand-less conditions. These studies demonstrate that the choice of ligands has a profound effect on the reaction outcome, as productive β,α-diarylation could only be obtained by suppressing the competing β-hydride elimination using BQ as the stabilising ligand and terminal reoxidant. The pivotal role of BQ in the reaction was studied using computer-aided density functional theory calculations. The calculations highlight the crucial role of BQ as a Pd(II)-ligand. In addition of serving as an oxidant of palladium, the calculations support the view that the coordination of BQ to the Pd(II)-centre in the key σ-alkyl complex leads to a low-energy pathway, aided by a strong η2 Pd-BQ donation-back-donation interaction. Furthermore, an investigation of the scope and limitations of novel stereoselective and BQ-mediated palladium(II)-catalysed domino Heck/Suzuki β,α-diarylation reactions, involving metal coordinating cyclic methylamino vinyl ethers and a number of electronically diverse arylboronic acids, conducted. In addition, a set of 4-quinolone-3-carboxylic acids, structurally related to elvitegravir and bearing different substituents on the condensed benzene ring, was designed and synthesized as potential HIV-1 integrase inhibitors. Finally, in an effort to identify a new class of HIV-1 protease inhibitors, four different stereopure β-hydroxy γ-lactam-containing inhibitors were synthesized, biologically evaluated, and co-crystallized with the enzyme. / The time 12:05 for the public defense mentioned in the thesis is incorrect. It will take place at 09:15, 2012-06-08.

Thesis (Ph. D.)--School of Pharmacy. University of Missouri--Kansas City, 2007. / "A dissertation in pharmaceutical science and pharmacology." Advisor: Ashim K. Mitra. Typescript. Vita. Title from "catalog record" of the print edition Description based on contents viewed July 16, 2008. Includes bibliographical references (leaves 231-248). Online version of the print edition.

Les patients infectés par le VIH (Virus de l'Immunodéficience Humaine) traités par les antirétroviraux (ARV) ont aujourd'hui une espérance de vie quasi normale. Cependant, ils présentent une augmentation de la prévalence de pathologies classiquement associées au vieillissement, dont l'ostéoporose, suggérant un vieillissement prématuré ou accentué. Les facteurs impliqués sont, entre autres, l'infection par le VIH, les ARV et un état d'inflammation chronique. L'ostéoporose correspond à une déminéralisation, suite à un déséquilibre entre formation (ostéoblastes) et résorption osseuse (ostéoclastes). L'infection par le VIH et les ARV, en particulier les inhibiteurs de protéase (PI), augmentent la prévalence de l'ostéoporose. Dans ce contexte, je me suis intéressée à l'effet de certaines protéines du VIH et PI sur les précurseurs ostéoblastiques, les cellules souches mésenchymateuses (MSC). Nous montrons que deux protéines du VIH, Tat et Nef, induisent une sénescence prématurée des MSC, associée à un stress oxydant et in fine à un défaut de différenciation en ostéoblastes. Les effets de Tat sont médiés par le facteur pro-inflammatoire et pro-sénescent NF-?B, et ceux de Nef sont liés à une inhibition de l'autophagie. Nous montrons également que les PI atazanavir et lopinavir associés au ritonavir induisent une sénescence et un stress oxydant du fait de l'accumulation de prélamine A farnésylée toxique conduisant à un déficit de la différentiation ostéoblastique. Ces travaux montrent que le VIH et certains PI peuvent jouer un rôle délétère sur les MSC, et mettent en lumière certains des mécanismes impliqués dans le vieillissement des patients infectés par le VIH et traités. / The efficacy of highly active antiretroviral treatment (ARV) has resulted in a considerable improvement in the life expectancy of HIV (Human Immunodeficiency Virus)-infected patients. However, many patients encounter the early occurrence of several common age-related comorbidities, such as osteoporosis, stressing for a premature or accentuated aging process. Proposed pathogenic mechanisms include HIV infection, ARV treatment and chronic inflammation. Osteoporosis is defined by a decrease in bone mineral density, resulting from the alteration of the balance between bone formation (osteoblasts) and resorption (osteoclasts). HIV infection, through the bystander effect of HIV secreted proteins, and ARV, particularly protease inhibitors (PI) increase the prevalence of osteoporosis.Our studies focused on the capacity of some HIV proteins, and of some PI to alter osteoblast precursors, namely mesenchymal stem cells (MSC). We showed that two HIV proteins, Tat and Nef, induced premature senescence of MSC, associated with an oxidative stress and a decreased osteoblastic differentiation potential. Tat triggered senescence via NF-κB activation, whereas the effect of Nef was linked to the inhibition of autophagy. We also showed that the PI, atazanavir and lopinavir boosted with ritonavir, induced senescence and oxidative stress through the accumulation of toxic farnesylated prelamin A, thus leading to a decreased osteoblastic differentiation.Overall, these data show that some HIV proteins and some PI can exert deleterious effects on MSC, resulting in senescence, and highlight several mechanisms which could be involved in the aging process of ART-controlled HIV infected patients.

HIV-1 protease is a very important drug target for AIDS therapy. Nine protease inhibitors have been proved by FDA and used in AIDS treatment. Due to the high replication rate and the lack of fidelity of the HIV-1 reverse transcriptase, HIV-1 virus developed various drug-resistant variants. Although experimental methods such as crystallography and isothermal titration calorimetry provide structural and thermodynamic data on drug-resistant variants, they are unable to discern the mechanism by which the mutations confer resistance to inhibitors. Understanding the drug-resistance mechanism is crucial for developing new inhibitors more tolerant to the drug-resistant mutations. Computational methods such as free energy calculations and molecular dynamic simulations can provide insights to the drug resistance mechanism at an atomic level. In this thesis, I have focused on the elucidation of the energetic and dynamics of key drug-resistant variants of HIV-1 protease. Two multi-drug resistant variants, in comparison with wild-type HIV-1 protease were used for the comparisons: Flap+ (L10I, G48V, I54V, and V82A) which contains a combination of flap and active site mutations and ACT (V82T, I84V) that only contains active site mutations. In Chapter II, I applied free energy simulations and decomposition methods to study the differential mechanism of resistance to the two variants, Flap+ and ACT, to the recently FDA-approved protease inhibitor darunavir (DRV). In this study, the absolute and relative binding free energies of DRV with wild-type protease and the two protease variants were calculated with MM-PB/GBSA and thermodynamic integration methods, respectively. And the predicted results are in good agreement with the ITC experimental results. Free energy decomposition elucidates the mutations alter not only its own interaction with DRV but also other residues by changing the geometry of binding pocket. And the VdW interactions between the bis-THF group of DRV is predominant even in the drug-resistant variants. At the end of this chapter, I offer suggestions on developing new inhibitors that are based on DRV but might be less susceptible to drug-resistant mutations. In Chapter III, 20-ns MD simulations of the apo wildtype protease and the apo drug-resistant protease variant Flap+ are analyzed and compared. In these studies, these mutations have been found to decrease the protease flexibility in the apo form but increase the mobility when the protease is binding with inhibitor. In Chapter IV, more details of the free energy simulation and decomposition are discussed. NMR relaxation experiments were set up as a control for the MD simulation study of the dynamics of the Flap+ variant. The difficulty of finishing the NMR experiment is discussed and the solution and some preliminary results are shown. In summary, the scope of this thesis was to use computational methods to study drug-resistant protease variants’ thermodynamic and dynamic properties to illuminate the mechanism of protease drug resistance. This knowledge will contribute to rational design of new protease inhibitors which bind more tightly to the protease and hinder the development of drug-resistant mutations.

Introdução: Uma proporção significante de pacientes infectados por HIV-1 (pacientes HIV-1+) tratados com inibidores de protease (IPs) desenvolve mutações de resistência. Estudos recentes têm mostrado que células T CD8+ de pacientes HIV- 1+ reconhecem epitopos de Pol incluindo mutações selecionadas por drogas. Nenhum epitopo CD4+ da protease foi descrito na base de dados de Los Alamos. Objetivo: Considerando que a protease de HIV-1 é alvo de terapia antiretroviral e que essa pressão pode selecionar mutações, nós investigamos se mutações selecionadas por IPs afetariam o reconhecimento de epitopos da protease de HIV-1 por células T CD4+ em pacientes tratados com IPs. Nós investigamos o reconhecimento de três regiões da protease preditas de conter epitopos de células T CD4+ bem como mutações induzidas por IPs por células T CD4+ em pacientes HIV- 1+ tratados com IPs. Materiais e Métodos: Quarenta pacientes HIV-1+ tratados com IPs foram incluídos (30 em uso de Lopinavir/ritonavir, 9 em uso de Atazanavir/Ritonavir e 1 em uso exclusivo de Atazanavir). Para cada paciente determinou-se a seqüência endógena da protease de HIV-1, genotipagem viral e tipagem HLA classe II. Utilizamos o algoritmo TEPITOPE para selecionar peptídeos promíscuos, ligadores de múltiplas moléculas HLA-DR, codificando as três regiões da protease de HIV-1 cepa HXB2 (HXB2 4-23, 45-64, e 76-95) e 32 peptídeos adicionais contidos nas mesmas regiões incorporando as mutações induzidas por IPs mais freqüentes no Brasil. Os 35 peptídeos foram sintetizados. Respostas proliferativas de células T CD4+ e CD8+ aos peptídeos foram determinadas por ensaios de proliferação com diluição do corante CFSE. Ensaios de ligação a alelos HLA classe II foram realizados para confirmar a promiscuidade desses peptídeos e avaliar a habilidade de se ligarem a moléculas HLA presentes em cada paciente. Resultados: Todos os peptídeos foram reconhecidos por pelo menos um paciente e respostas proliferativas de células T CD4+ e CD8+ a pelo menos um peptídeo da protease de HIV-1 foram encontradas em 78% e 75% dos pacientes, respectivamente. A terceira região (Protease 76 95) foi a mais freqüentemente reconhecida. Ao compararmos as respostas de células T às seqüências da protease do HIV-1 endógeno, observamos que a maioria dos pacientes não foi capaz de reconhecer peptídeos idênticos às essas seqüências, porém reconheceram peptídeos variantes diferentes das mesmas regiões. Apenas sete pacientes responderam às seqüências endógenas. Verificamos que diversos peptídeos endógenos que não foram reconhecidos apresentaram ausência de ligação a alelos HLA portados por estes pacientes, sugerindo que mutações selecionadas por pressão imune tenham levado ao escape de apresentação de antígeno e evasão de resposta de linfócitos T CD4+. Alternativamente, isso poderia ser explicado pela presença de um vírus replicante distinto presente no plasma uma vez que somente foram obtidas seqüências provirais. Conclusão: Epitopos selvagens e mutantes da protease do HIV-1 reconhecidos por células T CD4+ foram identificados. Também verificamos que a maior parte dos pacientes não reconheceu as seqüências da protease endógena enquanto que reconheceram seqüências variantes. O reconhecimento de seqüências não-endógenas poderia ser hipoteticamente conseqüência de alvo de populações HIV-1 minoritárias; protease de HERV que contém regiões de similaridade com a protease do HIV-1; ou seqüências de HIV-1 presentes apenas em parceiros virêmicos. A falha de reconhecimento de seqüências endógenas seria mais provável devido ao escape imune, do que ao nível de apresentação ou reconhecimento por células T. Isso implica em uma conseqüência patofisiológica na evasão de respostas de células T contra a protease de HIV-1 e no fato de ser tradicionalmente considerada uma proteína pouco antigênica / Introduction: A significant proportion of protease inhibitor (PI)-treated HIV-1 infected (HIV-1+) patients develop resistance mutations. Recent studies have shown that CD8+ T cells from HIV-1 patients can recognize antiretroviral drug-induced mutant Pol epitopes. No HIV-1 protease CD4 epitopes are described in the Los Alamos database. Aims: Given that the protease of HIV-1 is a target of antiretroviral therapy and this pressure may lead to the selection of mutations, we investigated whether PI-induced mutations affect the recognition of HIV-1 protease epitopes by CD4 + T cells in PI-treated patients. We investigated the recognition of three protease regions predicted to harbor CD4+ T cell epitopes as well as PI-induced mutations by CD4+ T cells of PI-treated HIV-1+ patients. Methods: Forty PI-treated HIV-1+ patients were included (30 undergoing Lopinavir/ritonavir, 9 undergoing Atazanavir/ritonavir and 1 undergoing exclusively Atazanavir treatment). For each patients, the endogenous HIV-1 protease sequence, viral genotype and HLA class II typing were determined. We used the TEPITOPE algorithm to select promiscuous, multiple HLA-DR-binding peptides encoding 3 regions of HIV-1 HXB2 strain protease (HXB2 4-23, 45-64, and 76-95) and 32 additional peptides contained in the same regions, but encompassing the most frequent PI-induced mutations in Brazil. The 35 peptides were thus synthesized. Proliferative responses of CD4+ and CD8+ T cells against peptides were determined by the CFSE dilution assay. HLA class II binding assays were made to confirm the promiscuity of these peptides and evaluate their ability to bind the HLA molecules carried by each patient. Results: All tested peptides were recognized by at least one patient and proliferative responses of CD4+ and CD8+ T cells against at least one HIV-1 protease peptide were found in 78% and 75% patients, respectively. The third region (Protease 76-95) was the most frequently recognized. By comparing T-cell responses to HIV-1 endogenous protease sequences, we found that most patients failed to recognize identical peptides of those sequences, but recognized different variant peptides of the same region. Only seven patients responded to endogenous sequences. We found that several endogenous peptides that failed to be recognized showed no binding to the HLA alleles carried by that given patient, suggesting that mutations selected by immune pressure have led to escape of antigen presentation, as well as direct escape of the CD4+ T cell response. Alternatively, it could have been due to the presence of a different replicating virus in the plasma-since we only obtained proviral sequences. Conclusion: Wild-type and mutant HIV-1 protease epitopes recognized by CD4+ T cells were identified. We also found that most patients failed to recognize their endogenous protease sequences, while they recognized variant sequences. The recognition of non-endogenous sequences could hypothetically be a consequence of targeting a minor HIV-1 population; HERV protease, that contains regions of similarity with HIV-1 protease; or HIV-1 sequences present only in viremic partners. The failure to recognize endogenous sequences is most likely due to immune escape, either at the level of presentation or direct T cell recognition. This may have a pathophysiological consequence on evasion of T cell responses against protease and the fact that it has been considered traditionally a poorly antigenic HIV-1 protein.

Introdução: Uma proporção significante de pacientes infectados por HIV-1 (pacientes HIV-1+) tratados com inibidores de protease (IPs) desenvolve mutações de resistência. Estudos recentes têm mostrado que células T CD8+ de pacientes HIV- 1+ reconhecem epitopos de Pol incluindo mutações selecionadas por drogas. Nenhum epitopo CD4+ da protease foi descrito na base de dados de Los Alamos. Objetivo: Considerando que a protease de HIV-1 é alvo de terapia antiretroviral e que essa pressão pode selecionar mutações, nós investigamos se mutações selecionadas por IPs afetariam o reconhecimento de epitopos da protease de HIV-1 por células T CD4+ em pacientes tratados com IPs. Nós investigamos o reconhecimento de três regiões da protease preditas de conter epitopos de células T CD4+ bem como mutações induzidas por IPs por células T CD4+ em pacientes HIV- 1+ tratados com IPs. Materiais e Métodos: Quarenta pacientes HIV-1+ tratados com IPs foram incluídos (30 em uso de Lopinavir/ritonavir, 9 em uso de Atazanavir/Ritonavir e 1 em uso exclusivo de Atazanavir). Para cada paciente determinou-se a seqüência endógena da protease de HIV-1, genotipagem viral e tipagem HLA classe II. Utilizamos o algoritmo TEPITOPE para selecionar peptídeos promíscuos, ligadores de múltiplas moléculas HLA-DR, codificando as três regiões da protease de HIV-1 cepa HXB2 (HXB2 4-23, 45-64, e 76-95) e 32 peptídeos adicionais contidos nas mesmas regiões incorporando as mutações induzidas por IPs mais freqüentes no Brasil. Os 35 peptídeos foram sintetizados. Respostas proliferativas de células T CD4+ e CD8+ aos peptídeos foram determinadas por ensaios de proliferação com diluição do corante CFSE. Ensaios de ligação a alelos HLA classe II foram realizados para confirmar a promiscuidade desses peptídeos e avaliar a habilidade de se ligarem a moléculas HLA presentes em cada paciente. Resultados: Todos os peptídeos foram reconhecidos por pelo menos um paciente e respostas proliferativas de células T CD4+ e CD8+ a pelo menos um peptídeo da protease de HIV-1 foram encontradas em 78% e 75% dos pacientes, respectivamente. A terceira região (Protease 76 95) foi a mais freqüentemente reconhecida. Ao compararmos as respostas de células T às seqüências da protease do HIV-1 endógeno, observamos que a maioria dos pacientes não foi capaz de reconhecer peptídeos idênticos às essas seqüências, porém reconheceram peptídeos variantes diferentes das mesmas regiões. Apenas sete pacientes responderam às seqüências endógenas. Verificamos que diversos peptídeos endógenos que não foram reconhecidos apresentaram ausência de ligação a alelos HLA portados por estes pacientes, sugerindo que mutações selecionadas por pressão imune tenham levado ao escape de apresentação de antígeno e evasão de resposta de linfócitos T CD4+. Alternativamente, isso poderia ser explicado pela presença de um vírus replicante distinto presente no plasma uma vez que somente foram obtidas seqüências provirais. Conclusão: Epitopos selvagens e mutantes da protease do HIV-1 reconhecidos por células T CD4+ foram identificados. Também verificamos que a maior parte dos pacientes não reconheceu as seqüências da protease endógena enquanto que reconheceram seqüências variantes. O reconhecimento de seqüências não-endógenas poderia ser hipoteticamente conseqüência de alvo de populações HIV-1 minoritárias; protease de HERV que contém regiões de similaridade com a protease do HIV-1; ou seqüências de HIV-1 presentes apenas em parceiros virêmicos. A falha de reconhecimento de seqüências endógenas seria mais provável devido ao escape imune, do que ao nível de apresentação ou reconhecimento por células T. Isso implica em uma conseqüência patofisiológica na evasão de respostas de células T contra a protease de HIV-1 e no fato de ser tradicionalmente considerada uma proteína pouco antigênica / Introduction: A significant proportion of protease inhibitor (PI)-treated HIV-1 infected (HIV-1+) patients develop resistance mutations. Recent studies have shown that CD8+ T cells from HIV-1 patients can recognize antiretroviral drug-induced mutant Pol epitopes. No HIV-1 protease CD4 epitopes are described in the Los Alamos database. Aims: Given that the protease of HIV-1 is a target of antiretroviral therapy and this pressure may lead to the selection of mutations, we investigated whether PI-induced mutations affect the recognition of HIV-1 protease epitopes by CD4 + T cells in PI-treated patients. We investigated the recognition of three protease regions predicted to harbor CD4+ T cell epitopes as well as PI-induced mutations by CD4+ T cells of PI-treated HIV-1+ patients. Methods: Forty PI-treated HIV-1+ patients were included (30 undergoing Lopinavir/ritonavir, 9 undergoing Atazanavir/ritonavir and 1 undergoing exclusively Atazanavir treatment). For each patients, the endogenous HIV-1 protease sequence, viral genotype and HLA class II typing were determined. We used the TEPITOPE algorithm to select promiscuous, multiple HLA-DR-binding peptides encoding 3 regions of HIV-1 HXB2 strain protease (HXB2 4-23, 45-64, and 76-95) and 32 additional peptides contained in the same regions, but encompassing the most frequent PI-induced mutations in Brazil. The 35 peptides were thus synthesized. Proliferative responses of CD4+ and CD8+ T cells against peptides were determined by the CFSE dilution assay. HLA class II binding assays were made to confirm the promiscuity of these peptides and evaluate their ability to bind the HLA molecules carried by each patient. Results: All tested peptides were recognized by at least one patient and proliferative responses of CD4+ and CD8+ T cells against at least one HIV-1 protease peptide were found in 78% and 75% patients, respectively. The third region (Protease 76-95) was the most frequently recognized. By comparing T-cell responses to HIV-1 endogenous protease sequences, we found that most patients failed to recognize identical peptides of those sequences, but recognized different variant peptides of the same region. Only seven patients responded to endogenous sequences. We found that several endogenous peptides that failed to be recognized showed no binding to the HLA alleles carried by that given patient, suggesting that mutations selected by immune pressure have led to escape of antigen presentation, as well as direct escape of the CD4+ T cell response. Alternatively, it could have been due to the presence of a different replicating virus in the plasma-since we only obtained proviral sequences. Conclusion: Wild-type and mutant HIV-1 protease epitopes recognized by CD4+ T cells were identified. We also found that most patients failed to recognize their endogenous protease sequences, while they recognized variant sequences. The recognition of non-endogenous sequences could hypothetically be a consequence of targeting a minor HIV-1 population; HERV protease, that contains regions of similarity with HIV-1 protease; or HIV-1 sequences present only in viremic partners. The failure to recognize endogenous sequences is most likely due to immune escape, either at the level of presentation or direct T cell recognition. This may have a pathophysiological consequence on evasion of T cell responses against protease and the fact that it has been considered traditionally a poorly antigenic HIV-1 protein.

L’apprêtement antigénique par les protéases intracellulaires et la présentation des épitopes sont essentiels pour la reconnaissance des cellules infectées par les lymphocytes CD8+. Ici nous avons montré que certains inhibiteurs de la protéase de la VIH (IPs) modulent l’activité de la protéasome et aminopeptidase impliqué dans l’apprêtement antigénique endogène et l’activité cathepsins importante dans l’apprêtement croisée. Deux IPs agissent directement sur les cathepsins et leurs régulateurs en inhibant les activités kinase, NOX2 et en régulant le pH phagolysosomal. Les IPs ont changé la dégradation des protéines viral et la production des épitopes de façon séquence- et cellule-spécifique, ont altéré la présentation direct et croisée des épitopes, et ont partiellement changé l’auto-peptidome des cellules primaires. La modulation par les drogues de l’apprêtement et la présentation des épitopes peut fournir une approche thérapeutique alternative pour moduler la reconnaissance immunitaire. / Antigen processing by intracellular proteases and peptidases and epitope presentation are critical for recognition of pathogen-infected cells by CD8+ T lymphocytes. Here we show that several HIV protease inhibitors (PIs) prescribed to HIV-infected persons variably modulate proteasome and aminopeptidase activities involved in endogenous antigen presentation and cathepsin activities involved in antigen cross-presentation. Two HIV PIs acted directly on cathepsins and on their regulators by inhibiting kinases, NOX2 and the regulation of phagolysosomal pH, subsequently enhancing cathepsin activities. HIV PIs modified HIV protein degradation and epitope production in a sequence- and cell-dependent manner, altered direct- and cross-presentation and T cell-mediated killing, and partly changed the self-peptidome of primary cells. Drug-induced modulation of antigen processing and peptidome may provide an alternate therapeutic approach to modulate immune recognition.