Phenotypic properties of transmitted founder HIV-1

Abstract

Defining the virus-host interactions responsible for HIV-1 transmission, including the phenotypic requirements of viruses capable of establishing de novo infections, could be important for AIDS vaccine development. Previous analyses have failed to identify phenotypic properties other than chemokine receptor 5 (CCR5) and CD4+ T-cell tropism that are preferentially associated with viral transmission. However, most of these studies were limited to examining envelope (Env) function in the context of pseudoviruses. Here, we generated infectious molecular clones of transmitted founder (TF; n = 27) and chronic control (CC; n = 14) viruses of subtypes B (n = 18) and C (n = 23) and compared their phenotypic properties in assays specifically designed to probe the earliest stages of HIV-1 infection. We found that TF virions were 1.7-fold more infectious (P = 0.049) and contained 1.9-fold more Env per particle (P = 0.048) compared with CC viruses. TF viruses were also captured by monocyte-derived dendritic cells 1.7-fold more efficiently (P = 0.035) and more readily transferred to CD4+ T cells (P = 0.025). In primary CD4+ T cells, TF and CC viruses replicated with comparable kinetics; however, when propagated in the presence of IFN-α, TF viruses replicated to higher titers than CC viruses. This difference was significant for subtype B (P = 0.000013) but not subtype C (P = 0.53) viruses, possibly reflecting demographic differences of the respective patient cohorts. Together, these data indicate that TF viruses are enriched for higher Env content, enhanced cell-free infectivity, improved dendritic cell interaction, and relative IFN-α resistance. These viral properties, which likely act in concert, should be considered in the development and testing of AIDS vaccines.

Fig. 1.

Virion infectivity and Env content. (A–D) Infectivity values for TF and CC viruses (x axis) are expressed as relative light units (RLUs) per picogram of viral RT activity (y axis). (A) Bars indicate the median infectivity of TF (filled) and CC (open) viruses, with interquartile ranges indicated. TF viruses were 1.7-fold more infectious than CC viruses (P = 0.049). (B) Infectivity values are shown for each virus. Subtypes B and C viruses are shown in red and blue, respectively, with TF viruses indicated in dark colors and CC viruses indicated in light colors, respectively. Values represent averages from four independent experiments. (C and D) Infectivity values are shown for TF and CC viruses as in A and B, except that infections were performed in the presence of DEAE dextran. Values represent averages from three independent experiments. (E and F) Env content of TF and CC virions (x axis) is expressed as the mass ratio of Env and RT content (y axis). (E) Bars indicate the median values of Env content for TF (filled) and CC (open) viruses, with interquartile ranges indicated. TF viruses contained 1.9 times more Env per unit of RT activity than CC viruses (P = 0.048). (F) Env content is shown for each virus and color-coded as in B and D. Values represent averages from two independent experiments.

Fig. 2.

Virus binding to moDCs. The percent of captured TF and CC virus is plotted (y axis) for moDC cell preparations from six different donors labeled A through F (x axis). (A and B) Virus input was normalized by RT activity. (C and D) Virus input was normalized by p24 content. (A and C) Bars indicate median values of moDC capture for TF (filled) and CC (open) viruses, with interquartile ranges indicated. TF viruses were captured 1.7-fold more efficiently than CC viruses (P = 0.035). (B and D) Values are plotted for each virus individually (color-coding for TF and CC viruses from subtypes B and C as in Fig. 1). Subtype B viruses were captured 3.4 times more efficiently than subtype C viruses (P = 4.6 × 10⁻⁶ by GLM).

Fig. 3.

DC-mediated trans infection. (A) Virus replication expressed as picograms of RT activity per milliliter of culture supernatant (y axis) is shown after cocultivation of virus-pulsed moDC with CD4+ T cells for one representative donor (of two analyzed) over 10 d (x axis). Bars indicate median values of viral replication for TF (filled) and CC (open) viruses, with interquartile ranges indicated (there was no significant difference between TF and CC viruses). (B) Values are plotted for each virus individually (color-coding for TF and CC viruses from subtypes B and C as in Fig. 1). Averaging data from two different donors, subtype B TF viruses grew to 11.2-fold higher titers than subtype B CC viruses (P = 0.004), whereas no significant differences were observed for subtype C TF and CC viruses (P = 0.23).

Fig. 4.

Virus replication in CD4+ T cells in the presence and absence of IFN-α. (A, C, and E) The replication kinetics of TF (solid lines) and CC (broken lines) viruses are shown in CD4+ T cells from three donors in the absence (gray lines) and presence (black lines) of 500 U IFN-α. RT activity indicated as picograms per milliliter of culture supernatant (y axis) was measured every 3 d (x axis). Data points indicate median values of virus production, with interquartile ranges indicated. Averaging data from all donors and points, TF viruses grew to 24-fold higher titers than CC viruses in the presence of IFN-α (P = 0.012). (B, D, and F) The ratio of virus production in the presence and absence of IFN-α is plotted for each virus (y axis) at different time points (days) postinfection (x axis; color-coding for TF and CC viruses from subtypes B and C as in Fig. 1). Averaging data from all donors, subtype B TF viruses grew to 62-fold higher titers than subtype B CC viruses (P = 0.000013), whereas subtype C TF viruses grew only 1.7-fold more efficiently than subtype C CC viruses (P = 0.53). Note that these replication differentials represent cumulative totals for the 12-d culture period.