Impairment of endocytosis-related factors FNBP1L, ARHGAP24, and ATP6V1B1 increases HIV-1 entry into dendritic cells

Abstract

HIV-1 infects several types of CD4+ cells. Among these, dendritic cells (DCs) are considered one of the first to encounter the virus upon sexual transmission. Expression of several restriction factors, of which SAMHD1 is well known, limits productive infection. Still, DCs are essential players in shaping adaptive immune responses that contribute heavily to the pathogenesis of HIV. Here, we set out to identify other factors that potentially contribute to the resistance of dendritic cells to HIV infection. Since endocytosis and the cytoskeleton impact HIV infection, we have put special emphasis on proteins implied in these pathways. In a selective, shRNA-mediated knockdown screen in primary monocyte-derived dendritic cells (MDDCs) infected with HIV in the presence of SAMHD1-disactivating Vpx containing virus-like particles, three proteins hampering HIV-1 infection were identified: FNBP1L, ARHGAP24, and ATP6V1B1. Findings of our research indicate that upon blocking of factors involved in endocytosis, increased viral entry is observed providing supportive evidence for endocytosis mostly being a dead-end entry pathway for HIV infection of MDDCs. Additional experiments show that changes in the cytoskeleton and endosomal pH that lead to impaired fluid-phase endocytosis and phagocytosis are responsible for these shifts in the phenotype observed.IMPORTANCEUnderstanding how HIV-1 interacts with dendritic cells (DCs) is pivotal in deciphering early viral transmission and immune evasion but is subject to a long-standing controversy in HIV virology. Therefore, the identification of endocytosis-related host factors as barriers to productive infection in DCs emphasizes the role of endocytosis as a restrictive pathway for viral entry. By disrupting these processes, we highlight a shift in the cellular environment that could influence viral entry and transmission. These findings challenge existing models of HIV-1 entry into DCs. New insights into how cellular pathways limit viral spread have implications for the development of strategies aimed to curb viral dissemination and reservoir formation. Whether the knockdown of the proteins described simply augments the efficiency of infection via existing pathways or opens additional routes for HIV-1 entry remains to be investigated.

Keywords: ARHGAP24; ATP6V1B1; FNBP1L; HIV-1; cytoskeleton; dendritic cells; endocytosis; viral entry.

Fig 1

CD4 knockdown validates the screening approach. (A) Surface CD4 expression by flow cytometry on MDDC, day 5 post-transduction with vectors expressing control—non-targeting, scrambled shRNA (filled histogram), or CD4-targeting shRNA (empty histogram). Histogram depicts staining fluorescence intensity versus relative cell number. (B) Flow cytometry dot plots show MDDCs transduced with the shRNAs as indicated and infected with GFP-expressing CCR5-tropic HIV-1, on day 3 post-infection as measured by eGFP expression within the live population; figures represent the percentage of GFP+ cells (rectangle gate). (C) Flow cytometry dot plots show MDDCs transduced with the shRNAs as indicated and infected with VSV-G envelope pseudotyped HIV-1 and analyzed as in panel B. (D) Graph represents the percentage of infected cells on day 3 post-infection with CCR5 tropic HIV-1 or VSV-G pseudotyped env deleted counterpart, on either sh-scrambled (CTRL) or sh-CD4 transduced cells, as indicated. Individual data points shown, vertical lines indicate median, and standard deviation among the donors tested (n = 5, **P = 0.008).

Fig 2

shRNA knockdown identifies factors involved in HIV infection of monocyte-derived dendritic cells. (A) Bar graph represents fold-change in HIV-1 NL4-3-Bal-IRES-eGFP infection rates (percentage eGFP+ cells) on day 3 post-infection in MDDCs transduced to express shRNA targeting the genes indicated in the presence of Vpx-containing VLPs. Results were normalized to the non-targeting, control scrambled shRNA transduced cells (depicted in blue). Error bars represent the standard deviation between the two donors tested. Green bars represent the factors selected for further validation. (B) Flow cytmetry dot plots show representative HIV-1 infection rates in cells from one of the donors used in the screening transduced to express scrambled (CONTROL) or targeting shRNA constructs (FNBP1L or CD4) as indicated. (C and D) Graphs show transduction rates of MDDCs from each of the donors used in the screening as verified by parallel transduction with the use of additional control vector encoding GFP (panel C) and comparison of HIV infection rates between non-transduced cells (NTD) and control transduced (TD) cells (with control vector used throughout the screening encoding non-targeting shRNA vector without fluorescent marker gene) (panel D).

Fig 3

Experimental controls during HIV-1 infection of MDDCs cells expressing the indicated shRNA. (A) Bar graph depicting cell viability of MDDCs 5 days post-transduction with shRNA targeting CD4, FNBP1L, ARHGAP24, or ATP6V1B1. The percentage of live cells is presented, showing that transduction does not significantly affect cell viability (n > 5). (B) Bar graph showing expression of the differentiation marker DC-SIGN (CD209) in MDDCs 5 days post-transduction. Percentages represent DC-SIGN-positive cells (n > 5). (C) Bar graph displaying the percentage of CD4+ cells among MDDCs 5 days post-transduction with the indicated shRNA constructs (n > 5). (D and E) For each of the targeted factors, knockdown efficiency was quantified on protein (FNBP1L, ARHGAP24) or mRNA level (ATP6V1B1). Panel D shows a western blot analysis of FNBP1L and ARHGAP24 protein levels in MDDCs transduced with non-targeting control (sh-scrambled) or shRNA targeting each gene. Actin is shown as a loading control. Panel E shows a bar graph depicting the knockdown efficiency of ATP6V1B1 at the mRNA level, as quantified by RT-qPCR. Two shRNA vector were tested; however, only the vector depicted as sh-ATP6V1B1 1 was used for experiments. Data are presented as relative expression compared to control cells.

Fig 4

FNBP1L, ARHGAP24, and ATP6V1B knockdown in MDDCs increases HIV-1 infection, but this effect is not sustained upon VSV-G pseudotyping of the virus. (A) Flow cytometry plots show the percentage of eGFP+ MDDCs transduced to express the FNBP1L, ARHAGP24, and ATP6V1B1 shRNAs in the presence of Vpx-containing VLPs plotted against the side scatter (SSC), on day 3 post-infection with 50 ng p24 of HIV-1 NL4-3-Bal-IRES-eGFP (CCR5-tropic virus), compared to control transduced cells (CTRL) from a representative donor. (B) Flow cytometry plots show percentage of eGFP+ MDDCs transduced to express the FNBP1L, ARHAGP24, and ATP6V1B1 shRNAs in the presence of Vpx-containing VLPs plotted against the side scatter (SSC), on day 3 post-infection with 5 ng p24 of HIV-1 NL4-3-IRES-eGFP lacking WT envelope and pseudotyped with VSV-G, compared to control transduced cells (CTRL) from a representative donor. (C and D) Plots show infection levels (eGFP+ transduced MDDCs) infected with CCR5-tropic HIV-1-GFP (panel C, n ≥ 4, shARHGAP24 **P = 0.0023, shFNBP1L and shATP6V1B1 ****P < 0.0001) or with VSV-G pseudotyped HIV-1-GFP (panel D, n ≥ 4, shFNBP1L ns P = 0.2338, shARHGAP24 ns P = 0.1553, shATP6V1B1 ns P > 0.9999). Vertical lines represent the median and standard deviation among the tested donors.

Fig 5

Effects of host protein knockdown on HIV-1 infection across cell types. (A) Line graph shows the effect of host protein knockdown on HIV-1 infection in THP1 cells over a 7-day period. Data represent viral infection levels at multiple time points post-infection compared to control cells (CTRL). (B–D) Bar graph depicts the effect of host protein knockdown on HIV-1 infection on day 3 post-infection in Jurkat CD4-CCR5 cells (panel B, n = 3, shFNBP1L *P = 0.0139, shARHGAP24 ns P = 0.5227, shATP6V1B1 ns P = 0.1627); in primary CD4+ T cells (panel C, n = 4, shFNBP1L *P = 0.0225, shARHGAP24 ns P = 0.0.5440, shATP6V1B1 ns P > 0.999); in monocyte-derived dendritic cells (MDDCs) treated with protease inhibitor (1 µM ritonavir) (panel D, n = 4, shFNBP1L *P = 0.0277, shARHGAP24 **P = 0.0.0025, shATP6V1B1 ns P = 0.7022). For panels B–D, Kruskal-Wallis ANOVA screening was performed to assess any preliminary differences between the groups.

Fig 6

HIV-1 entry into transduced monocyte-derived dendritic cells. (A) Schematic representation of the BlaM-Vpr assay used to measure HIV-1 entry. Upon viral entry, host esterases cleave the AM tag, trapping the CCF2 substrate inside the cell. Once loaded, the green substrate gets cleaved by the β-lactamase (delivered by infecting the cells with HIV-1 virus that carries BlaM-Vpr fusion) causing a shift in fluorescence emission from green (520 nm) to blue (447 nm), indicating successful viral entry. (B) Gating strategy for identifying cleaved CCF2 in MDDCs. Dot plots show forward scatter (FSC) versus side scatter (SSC) and the fluorescence intensity of uncleaved (FITC channel) versus cleaved (VioBlue channel) CCF2 substrate in control, uninfected cells. MDDCs were loaded with CCF2 substrate on day 5 post-transduction and analyzed by flow cytometry after overnight incubation. (C) Representative flow cytometry plots show cleavage of CCF2 in MDDCs transduced with shRNAs targeting CD4, FNBP1L, ARHGAP24, or ATP6V1B1, following infection with BlaM-Vpr-containing HIV-1. The percentage of cells displaying viral entry is indicated, as demonstrated with a diagonal shift of the population upon cleavage of the CCF2 substrate (gated population). (D and E) Bar graphs show fold-change in viral entry compared to control (as shown in panel C) in MDDCs infected with CXCR4-tropic HIV-1 virus (panel D, n ≥ 4, shCD4 and shARHGAP24 **P = 0.0079, shFNBP1L and ATP6V1B1 **P = 0.0043) or with CCR5-tropic HIV-1 (panel E, n ≥ 4, shCD4, shFNBP1L, shARHGAP24, and shATP6V1B1 **P = 0.0022). Plots show individual data points, and error bars indicate standard deviation among the donors tested.

Fig 7

Assessment of endocytic activity in MDDCs following knockdown of host proteins. (A) Bar graph showing the effect of shRNA knockdown of CD4, FNBP1L, ARHGAP24, and ATP6V1B1 on FITC-dextran uptake by MDDCs. Data are presented as fold-change in mean fluorescence intensity (MFI) compared to control (CTRL), normalized to values obtained at 4°C to account for passive uptake (n = 4, shCD4 ns P = 0.9892, shFNBP1L *P = 0.0469, shARHGAP24 ns P = 0.0771, shATP6V1B1 ns P > 0.9999). (B and C) Graphs showing the effect of protein knockdown (panel B, shFNBP1L ns P = 0.9136, shARHGAP24 ns P = 0.4518, shATP6V1B1 *P = 0.0412) or treatment with endocytosis and cytoskeleton inhibitors as indicated (panel C, Dynasore ns P = 0.9136, NSC23766 ns P = 0.4518, Latrunculin A *P = 0.0412) on the phagocytosis of pHrodo E. coli bioparticles by MDDCs (n = 2). The fold-change in red integrated intensity, indicative of particle uptake and acidification, is calculated relative to control cells. For panels A-C, Kruskal-Wallis ANOVA screening was performed to assess any preliminary differences between the groups.