The efficiency of shCCR5 was initially evaluated in MAGI-CCR5 cells. Transient expression of shCCR5 in MAGI-CCR5 cells induced downregulation of the CCR5 coreceptor to approximately 60% (Fig. 1A), similar to the results of our previous report using U937 cells [3]. MAGI-CCR5 cells transiently expressing shCCR5 or shGFP were then challenged with HIV-1 6920B and viral p24 antigen was analysed in the supernatants by the ELISA capture assay. Assay of HIV-1 p24 concentration was selected as an estimate of viral replication by practicality since new capsid antigens are formed during viral replication. In the shCCR5-protected culture, a significantly lower production of the p24 antigen was observed on days 4, 7 and 9 after infection (p = 0.0043, p < 0.0002 and p < 0.0001, respectively) compared with that in the unprotected culture (Fig. 1B) and the inhibition of viral replication was 91, 95 and 77% on the corresponding days (Fig. 1C); these results suggest that HIV-1 infection requires a minimal threshold concentration of CCR5 on the cell surface and that even a partial reduction in CCR5 expression substantially decreases HIV-1 infection. The threshold of surface CCR5 density required for HIV-1 infection has also been shown in previous in vitro studies and in vivo in individuals heterozygous for CCR5Δ32 [15, 18].
Then, the efficiency of shCCR5 and shRev was evaluated in the U937 cells stably expressing one of these vectors (Fig. 2A). When U937-shCCR5 or U937-shRev cells were infected with HIV-1 6920B (100 TCID50) for two weeks, the comparative nonlinear regression analysis (exponential growth) of the viral replication kinetics showed an overall lower replication in the shCCR5-modified cells than that in the shRev-modified cells (95% CI; p = 0.0089; k = 0.2246 and 0.2612 and doubling time = 3.086 and 2.654 for shCCR5 and shRev cultures, respectively). The p24 concentration in the supernatants of the two cell cultures at various timepoints showed significantly lower production by the U937-shCCR5 culture on days 10 and 14 (p = 0.0042 and p = 0.0007, respectively) indicating a more sustained protective effect of the shCCR5 treatment (Fig. 2A).
Prevention of viral entry in the cells or blocking viral replication in infected cells may have a distinct protective effect in vivo because shRNA delivery can protect only a certain fraction of HIV-1 target cells. In this scenario, blocking viral infection by downregulating the CCR5 coreceptor rather than by inhibiting replication by targeting viral genes may provide a better selective survival and proliferative advantage to the modified cells by reducing the HIV-1 cytopathogenic effect mediated by the coreceptor engagement [19]. Targeting only viral genes may block replication but does not necessarily block infection; thus, the proportion of unprotected-infected cells and the production of viral particles by these cells may substantially influence the infection, proliferation and survival rate of the shRev-protected cells. Since U937-shCCR5 and U937-shRev cells are models of viral entry and inhibition of early viral replication, respectively, the kinetics of the inhibitory effect on viral p24 production in these cells were evaluated in titration experiments of HIV-1 infection of shRNA-protected (U937-shCCR5 or U937-shRev) cells mixed with various proportions of unprotected (U937-shGFP) cells (Figs. 2B and 2C).
Interestingly, a different dynamic of viral replication inhibition was observed in the two cultures. The global curve fitting by nonlinear regression analysis that defines the family of the datasets of viral replication in the U937-shCCR5/U937-shGFP cell cultures grown for 3, 7, 10 and 13 days, showed a better fit to a nonlinear exponential plateau than to a line model (AICc = 93.90 and 130.2 for exponential plateau and line fit models, respectively; evidence ratio = 7.6x107). As shown in Figs. 2B and 2C, the curve of the U937-shCCR5 cell cultures displayed saturation at approximately 75–80% of protected cells, whereas the curve of the U937-shRev cells showed an almost linear decrease of viral replication with respect to the percentage of shRev-transfected cells (AICc = 62.18 for line model; results of calculations for the exponential plateau model return “ambiguous”). In the U937-shCCR5 cell experiment, a lower number of shCCR5-modified cells conferred a higher inhibition of viral replication than that in the cultures of the U937-shRev cells. Cultures containing as low as 25% of U937-shCCR5 cells were characterized by viral replication inhibition of approximately 50%, while the cultures with 50 and 75% of shCCR5-modified cells were characterized by inhibition by approximately 75% and 90%, respectively. This observation is very relevant for the outcome of RNAi treatment of HIV-1 infection expected in a clinical setting, since it shows that in a model of cell populations composed of shRNA-protected and unprotected cells, which may happen in vivo, shCCR5 has a better protective potential than shRev. In this situation, the extent of HIV infection and cytopathogenic effect will depend on the number of unprotected and infected cells. Theoretically, CCR5-negative cells will be less susceptible to the free virus, viral cell-to-cell infection [20] and coreceptor-induced alterations in signal transduction and cytopathogenic effects [21]. Thus, apoptosis of infected cells will decrease the viral production and therefore, the number of shCCR5-expressing cells may become higher than the number of shRev-expressing cells in the corresponding cultures because of a higher survival and proliferative advantage. These distinctive responses to the inhibition of HIV-1 infection by shRNA CCR5 downregulation reinforce that the CCR5 coreceptor can be considered a strategic target for HIV-1 RNAi therapy, which is in agreement with the previous results of in vitro and in vivo studies [22, 23]. However, due to the virus ability to switch its coreceptor affinity (e.g., CXCR4 or αβ47), a strong inhibition of viral replication using RNAi against other suitable viral targets in combination with inhibition of CCR5 may enhance the protective effect against highly infectious doses and may reduce the rate of escape mutant generation.
Accordingly, we demonstrate that when cellular and viral genes are simultaneously targeted by the two shRNAs, the protection against HIV-1 infection is bolstered and the combined protective properties of these shRNAs are enhanced (Fig. 3). In cultures of U937-shCCR5, U937-shRev and U937-shCCR5-Rev cells infected with increasing doses of HIV-1 6920B for four days, double-transfected cells have significantly lower viral replication than the cells transfected with shCCR5 or shRev (p < 0.0001 and p < 0.0001, respectively) according to comparison at increasing infective doses of 40 and 70 TCID50 (Fig. 3A). The combined protective effect of the shCCR5-Rev treatment was higher than any single treatment and ranged from 95 to 98% of viral replication inhibition at increasing infective doses within the limits of 25-70 TCID50 (p = 0.0040 and 0.0003, < 0.0001 and 0.0006, < 0.0001 and < 0.0013 for shCCR5 and shRev comparisons with shCCR5-Rev at 33, 40 and 70 TCID50, respectively). In contrast, the protection by cells transfected with a single construct was substantially overcome by increased viral infectious doses (Fig. 3B).
Furthermore, when different cell lines were challenged with a high viral dose (100 TCID50) and cultured for 14 days, the double-transfected cells showed a significantly lower viral replication over the whole experimental period compared with that in the shCCR5- or shRev-modified cells (p= 0.0037 and <0.0001, <0.0001 and <0.0001 for shCCR5 and shRev comparison with shCCR5-Rev at 10 and 13 days, respectively. The levels of viral replication over time in the U937-shGFP-infected cultures were similar to those in infected human PBMC (Fig. 3C). The cell lines expressing both shRNAs showed a higher and more sustained inhibition of p24 production (100%-98%) than the cells modified with a single vector over the whole culture period (p = 0.0065 and <0.0001, 0.0044 and <0.0001, <0.0001 and <0.0001, for shCCR5 and shRev comparison with shCCR5-Rev at 7, 10 and 14 days, respectively. The inhibitory effect of the cells modified by a single vector declined after day 7 (Fig. 3D). These results are in agreement with previous in vitro and in vivo studies on simultaneous targeting of CCR5 and various viral genes by RNAi [22, 23].
In summary, the present study demonstrates that shRNAs targeting the CCR5 or Rev transcripts have a distinct impact on HIV-1 infection and that shCCR5 has additional advantages. High protective effect of partial downregulation of CCR5 expression, better performance of shCCR5 in a model of cell population composed of shRNA-protected and unprotected cells, and better sustained functionality account for efficient inhibition of HIV-1 infection by shCCR5. On the other hand, inhibition of viral replication by shRev was close to 100% although the effect was less sustained. Thus, simultaneous treatment with both shRNAs resulted in a clearly enhanced and more stable viral replication inhibition. Both properties are essential for delay or prevention of the appearance of RNAi virus escape mutants or switch of the coreceptor affinity.
Congenital absence of CCR5 appears to be well tolerated [24]; thus, considering all previous studies [25], the results of the present study reinforce the convenience of inclusion of CCR5 inhibitors in HIV-1 gene silencing treatment schedules to further reduce endogenous viral infection and replication in a clinical setting in a properly ART-treated HIV-1 infected individual.