Dendritic cell maturation, but not type I interferon exposure, restricts infection by HTLV-1, and viral transmission to T-cells

Abstract

Human T lymphotropic Virus type 1 (HTLV-1) is the etiological agent of Adult T cell Leukemia/Lymphoma (ATLL) and HTLV-1-Associated Myelopathy/Tropical Spastic Paraparesis (HAM/TSP). Both CD4+ T-cells and dendritic cells (DCs) infected with HTLV-1 are found in peripheral blood from HTLV-1 carriers. We previously demonstrated that monocyte-derived IL-4 DCs are more susceptible to HTLV-1 infection than autologous primary T-cells, suggesting that DC infection precedes T-cell infection. However, during blood transmission, breast-feeding or sexual transmission, HTLV-1 may encounter different DC subsets present in the blood, the intestinal or genital mucosa respectively. These different contacts may impact HTLV-1 ability to infect DCs and its subsequent transfer to T-cells. Using in vitro monocyte-derived IL-4 DCs, TGF-β DCs and IFN-α DCs that mimic DCs contacting HTLV-1 in vivo, we show here that despite their increased ability to capture HTLV-1 virions, IFN-α DCs restrict HTLV-1 productive infection. Surprisingly, we then demonstrate that it is not due to the antiviral activity of type-I interferon produced by IFN-α DCs, but that it is likely to be linked to a distinct trafficking route of HTLV-1 in IL-4 DCs vs. IFN-α DCs. Finally, we demonstrate that, in contrast to IL-4 DCs, IFN-α DCs are impaired in their capacity to transfer HTLV-1 to CD4 T-cells, both after viral capture and trans-infection and after their productive infection. In conclusion, the nature of the DCs encountered by HTLV-1 upon primo-infection and the viral trafficking route through the vesicular pathway of these cells determine the efficiency of viral transmission to T-cells, which may condition the fate of infection.

Fig 1. IL-4 DCs, TGF-β DCs and IFN-α DCs are not equally susceptible to HTLV-1 infection.

(A) MDDCs were co-cultured with mitomycin-treated C91-PL cells, and the percentage of MDDCs expressing the viral p19^gag protein was assessed by flow cytometry after 3 h to measure viral capture or 72 h to measure de novo viral expression by productively infected MDDCs. MDDCs were discriminated from C91-PL cells by CD11c gating. The results obtained from MDDCs differentiated from the same blood donor are shown as percentage of p19^gag protein present in DCs during the time course. Results for 3 independent experiments performed on different donors are shown. Asterisks indicate statistically significant differences calculated using a paired t-test: **p<0.01; *p<0.05; ns: non significant. (B) MDDCs were exposed to purified viral biofilm or heat-inactivated biofilm. Genomic DNA was extracted 3 days or 5 days post-exposure and analyzed by real-time PCR. Results are presented as the ratio of tax/actin obtained from 3 independent experiments performed on different donors. Mean values are indicated. Asterisks indicate statistically significant differences calculated using a t-test: *p<0.05; ns: non significant. (C) MDDCs were co-cultured with mitomycin-treated C91-PL cells for 3 days and the percentage of MDDCs expressing the viral Tax protein determined by flow cytometry. The mean value of at least 4 independent experiments obtained with different blood donors is indicated. Asterisks indicate statistically significant differences calculated using one-way ANOVA followed by Bonferroni’s post-hoc test: ***p<0,001; **p<0.01.

Fig 2. Expression of HTLV-1 receptors in IL-4 DCs, TGF-β DCs and IFN-α DCs is inversely correlated to HTLV-1 capture, but correlated to HTLV-1 productive infection.

(A-C) Cells were labeled with fluorochrome-coupled antibodies (black lines) directed against (A) NRP-1, (B) DC-SIGN and (C) Glut-1. Cells stained with matched isotype antibodies were used as negative controls (solid grey graphs). Quantification of the mean fluorescence intensity is represented on the right of each panel as the Δ mean fluorescence intensity detected in stained versus unstained cells. Results obtained in 3 independent experiments performed on 3 different donors are presented as percentage of receptor expression normalized to the expression detected in IL-4 DCs. (D) IL-4, TGF-β and IFN-α DCs were co-cultured with mitomycin-treated C91-PL cells for 3 h, and the percentage of MDDCs expressing the viral p19^gag protein was assessed by flow cytometry. The mean values from at least 4 independent experiments obtained with different donors were normalized to the percentage of p19^gag protein detection in IL-4 DCs, and presented as fold change. (E) IL-4, TGF-β and IFN-α DCs from the same blood donors were co-cultured with mitomycin-treated C91-PL cells for 3 days and productive infection assessed by flow cytometry allowing viral Tax detection. The number of DCs expressing the viral Tax protein was determined and normalized to the number detected in IL-4 DCs. Results are presented as fold changes. (A-E) Asterisks indicate statistically significant differences calculated using one-way ANOVA followed by Bonferroni’s multiple comparison test: ****p<0.0001; *** p<0.001; **p<0.01; *p<0.05; ns: non significant.

Fig 3. Recombinant IFN-α or supernatant from IFN-α DCs is inefficient to restrict infection of IL-4 DCs.

(A) Supernatant from IL-4, TGF-β and IFN-α DCs derived from the same donors were collected after differentiation and type I IFN levels measured using HL116 reporter cells expressing ISRE-Luc. Asterisks indicate statistically significant differences calculated using one-way ANOVA followed by Bonferroni’s multiple comparison test: *** p<0.001; ns: non significant. (B) IL-4 DCs were treated with 500 IU/ml of recombinant IFN-α for 18 h, or left untreated as a control. Total RNA was extracted and retro-transcribed (RT). PCR was performed using Mx-1 or GADPH specific primers. Absence of DNA contamination was controlled using PCR amplification on total RNA in the absence of RT treatment. (C) IL-4 DCs were treated or not with recombinant IFN-α (500 or 2000 IU/ml) before co-culture with mitomycin-treated C91-PL cells for 3 days, and Tax expression was measured. Cells from the co-culture were first gated on Tax expression and then on CD11c expression. The Tax-expressing DC population is shown in the box, as well as the percentage of Tax expressing DC in the Tax positive population. A representative experiment is shown as zebra plots. (D) The percentage of infected CD11c⁺ Tax⁺ DCs was determined after treatment of IL-4 DCs with increasing amounts of recombinant IFN-α 2a. Results from at least 3 independent experiments with different donors are shown. Statistical analysis was performed using one-way ANOVA: ns: non significant. (E) IL-4 DCs were treated or not with the supernatant from IFN-α DCs before co-culture with mitomycin-treated C91-PL cells for 3 days, and Tax expression was measured. IFN-α DCs were included as a control. Cells were then gated and results represented as in (C). (F) The results from 4 independent experiments obtained with different donors are presented as in (D). Statistical analysis was performed using a t-test: ns: non significant.

Fig 4. DC maturation is responsible for restriction of HTLV-1 productive infection.

(A) IL-4 DCs were treated with the TLR-4 ligand LPS before co-culture with C91-PL for 3 h, and viral capture measured by flow cytometry using p19^gag detection. Untreated IL-4 DCs were used as a control. The percentage of DCs expressing p19^gag is shown for immature IL-4 DCs and LPS-treated IL-4 DCs from the same donor. Results from 5 independent experiments performed on different donors are shown. Statistical analysis was performed using a t-test: ***<0.001. (B) IL-4 DCs were treated with LPS before co-culture with mitomycin-treated C91-PL cells for 3 days, and productive infection measured by flow cytometry using Tax protein detection. Untreated IL-4 DCs were used as a control. The percentage of DCs expressing Tax is shown for immature IL-4 DCs and LPS-treated IL-4 DCs from the same donor. Paired results from 9 independent experiments performed on different donors are shown. Statistical analysis was performed using a t-test: *** p< 0.001. (C) IL-4 DCs were treated or not with the supernatant from LPS-treated DCs (middle plot) before co-culture with mitomycin-treated C91-PL cells for 3 days, and Tax protein expression was measured. LPS-treated DCs were included as a control (right plot). Cells from the co-culture were first gated on Tax expression and then the infected DC population expressing Tax was determined on CD11c expression. A representative experiment is shown as zebra plots with the infected Tax⁺CD11c⁺ DC population squared, and the percentage of Tax expressing DCs in the Tax positive population is indicated. (D) The percentage of DCs expressing Tax was measured for IL-4 DCs (control) or for IL-4 DCs treated with the supernatant from LPS-treated DCs from the same donor co-cultured with mitomycin-treated C91-PL cells for 3 days. Results from 4 independent experiments performed on different donors are shown. Statistical analysis was performed using a t-test: ns: non significant. (E) IL-4 DCs were left untreated as controls or treated with the TLR-2 ligand PAM3CSK4, the TLR-5 ligand flagellin, the TLR-3 ligand PolyI:C, the TLR-4 ligand LPS or the TLR-7/8 ligand R848 before co-culture with mitomycin-treated C91-PL cells for 3 h, and viral capture measured by flow cytometry using p19^gag staining. The percentage of DCs stained for p19^gag was determined and normalized to that of control IL-4 DCs. Results from at least 3 independent experiments performed on different donors are represented as fold increase over control. Asterisks indicate statistically significant differences calculated using one-way ANOVA followed by Bonferroni’s multiple comparison test: **** p< 0.0001, ** p<0.01, ns: non significant. (F) IL-4 DCs were left untreated or treated with the TLR-2 ligand PAM3CSK4, the TLR-5 ligand flagellin, the TLR-3 ligand PolyI:C, the TLR-4 ligand LPS or the TLR-7/8 ligand R848 before co-culture with mitomycin-treated C91-PL cells for 3 days, and productive infection measured by flow cytometry using Tax expression. The percentage of Tax expressing DCs was determined and normalized to that of control IL-4 DCs. Results from at least 3 independent experiments performed on different donors are represented as fold increase over control. Asterisks indicate statistically significant differences calculated using one-way ANOVA followed by Bonferroni’s multiple comparison test: **** p<0.0001, *** p< 0.001, ns: non significant.

Fig 5. HTLV-1 localizes in CD82⁺ vesicles after capture by DCs.

(A) IL-4 DCs, LPS-treated IL-4 DCs and IFN-α DCs were co-cultured with mitomycin-treated C91-PL cells for 3 h, and immunostained for HTLV-1 (green) and CD82 (red) before analysis by confocal microscopy. Nuclei were counterstained with DAPI (blue). White arrows show co-localization of HTLV-1 and CD82. Green arrows show vesicles stained with HTLV-1 only. Representative images obtained by confocal microscopy are shown. HTLV-1 and CD82 signals were quantified along the segment represented on the zoom panel and plotted on the histogram. Scale bars = 10 μm. (B) Co-localization of HTLV-1 signal with CD82 signal was calculated using Mander’s coefficient.

Fig 6. HTLV-1 uses macropinocytosis pathways to enter into IL-4 DCs and LPS-stimulated IL-4 DCs, but not in IFN-α DCs.

(A-B) IL-4 DCs (white bars), LPS treated IL-4 DCs (light gray bars) or IFN-α DCs (dark gray bars) were treated for 3 h with (A) clathrin-mediated endocytosis inhibitors (chloropromazine, CPZ), dynamin II inhibitors (dynasore, Dyna), actin polymerization inhibitors (latrunculin, Lat) or macropinocytosis inhibitors (amiloride, ami) or (B) constitutive macropinocytosis inhibitors (genistein, Gen), PI3K inhibitor (wortmanin, Wort), PKC inhibitors (rottlerin, Rott and Gö6976), Rac-1 inhibitors (NSC23766, NSC), Pak-1 inhibitor (IPA-3, IPA), and actin inhibitor (Cytochalasin D, Cyto D). After treatment, DCs were co-cultured with mitomycin-treated C91-PL cells for 3 h and viral capture determined by flow cytometry using p19^gag staining. DCs treated with DMSO, ethanol or medium depending on the solvent used for inhibitor solubilization (see material and methods section) were used as controls. The percentage of DCs positive for p19^gag was determined for each condition and normalized to that of the untreated control DCs. Results were obtained from 3 independent experiments performed with different blood donors. Asterisks indicate statistically significant differences between treated and untreated DCs calculated using a t-test: * p<0.05; ** p<0.01;*** p<0.001; ns: non significant.

Fig 7. Acidic pH of vesicles decreases DC infection.

(A) IL-4 DCs were treated with 10μM DPI for 3 h and incubated for 30 minutes with a pH-sensitive lysotracker fluorescent dye. Fluorescence intensity was determined using ImageJ. Results are representative of at least three independent experiments performed on different donors. Asterisks indicate statistically significant differences calculated using a t-test: *** p<0.001. (B) DPI-treated IL-4 DCs or untreated IL-4 DCs control were co-cultured with mitomycin-treated C91-PL cells for 3 h. Viral capture was measured by flow cytometry using p19^gag staining. The percentage of p19^gag-positive DCs was normalized to that of control IL-4 untreated DCs and presented as fold changes over control. Results are representative of 4 independent experiments performed on different donors. Asterisks indicate statistically significant differences calculated using a t-test: ns: non significant. (C) DPI-treated IL-4 DCs or untreated IL-4 DCs control were co-cultured with mitomycin-treated C91PL cells for 3 days. Productive infection was measured by flow cytometry using Tax staining. The number of Tax-expressing DCs was normalized to that of control IL-4 untreated DCs and presented as fold changes over control. Results are representative of 4 independent experiments performed on different donors. Asterisks indicate statistically significant differences calculated using a t-test: ** p<0.01. (D) IFN-α DCs were treated for 3 h with 150 μM chloroquine, and incubated with lysotracker fluorescent dye for 30 minutes. Fluorescence intensity determined using ImageJ was compared using a t-test: *** p<0.001. Results are representative of at least 3 independent experiments performed on different donors. (E) Chloroquine-treated IFN-α DCs or untreated IL-4 and IFN-α DCs used as controls were co-cultured with C91-PL for 3 h. Viral capture was measured by flow cytometry using p19^gag staining. The percentage of p19^gag-positive DCs was normalized to that of control IL-4 untreated DCs and presented as fold change over control. Results are representative of at least 3 independent experiments performed on different donors. Asterisks indicate statistically significant differences calculated using one-way ANOVA followed by Bonferroni’s multiple comparison test: * p<0.05 ns: non significant. (F) Chloroquine-treated IFN-α DCs or untreated IL-4 and IFN-α DCs used as controls were co-cultured with mitomycin-treated C91-PL cells for 3 days. Productive infection was measured by flow cytometry using Tax detection. The number of Tax-expressing DCs was normalized to that of control IL-4 untreated DCs and presented as fold changes over control. Results are representative of 3 independent experiments performed on different donors. Asterisks indicate statistically differences calculated using one-way ANOVA followed by Bonferroni’s multiple comparison test: ***p<0.001, **p<0.01. (G) LPS-treated IL-4 DCs were treated for 3 h with 150 μM chloroquine, and incubated with lysotracker fluorescent dye for 30 minutes. Fluorescence intensity was determined using ImageJ and compared using a t-test: *** p<0.001. Results are representative of at least 3 independent experiments performed on different donors. (H) Chloroquine-treated LPS-matured IL-4 DCs or untreated IL-4 and LPS-matured IL-4 DCs used as controls, were co-cultured with mitomycin-treated C91-PL cells for 3 h. Viral capture was measured and presented as in (E). Results are representative of 3 independent experiments performed on different donors. Asterisks indicate statistically differences calculated using one-way ANOVA followed by Bonferroni’s multiple comparison test: * p<0.05, ns: non significant. (I) Chloroquine-treated LPS-matured IL-4 DCs or untreated IL-4 and LPS-matured IL-4 DCs used as controls, were co-cultured with mitomycin-treated C91-PL cells for 3 days. Productive infection was measured and presented as in (F). Results are representative of 4 independent experiments performed on different donors. Asterisks indicate statistically significant differences calculated using one-way ANOVA followed by Bonferroni’s multiple comparison test: ***p<0.001.

Fig 8. HTLV-1 is transferred to reporter T-cells from IL-4 DCs but not from LPS-treated or IFN-α DCs.

(A) DCs were exposed to mitomycin-treated C91-PL cells for 4 h and isolated from C91-PL using positive selection with magnetic beads coated with anti BDCA-4 antibodies. Then, purified MDDCs were co-cultured with Jurkat-LTR-Luc cells and luciferase activity was assessed after 2 days. When indicated, IFN-α DCs were treated with chloroquine before their exposure to C91-PL. The graph shows the means and the standard deviations from 3 independent experiments obtained with different donors. Background signal due to leaky luciferase activity of Jurkat-LTR-Luc cells was subtracted and results were normalized to the luciferase activity of Jurkat-LTR-Luc measured after co-culture with IL-4 DCs and presented as fold change over control. Asterisks indicate statistically significant differences calculated using one-way ANOVA followed by Bonferroni’s multiple comparison test: * p<0.05, ** p<0.01. (B) DCs were mixed with GFP-transduced Jurkat-LTR-Luc cells and C91-PL cells stained with a red cell tracker, and contacts were visualized by fluorescence microscopy. Arrows indicate contacts between unstained DCs and GFP-transduced Jurkat-LTR-Luc cells (JK). Scale bar = 50μm. Contacts between DCs and GFP-transduced Jurkat-LTR-Luc cells were manually counted, normalized to the number of DCs in the picture analyzed, and represented as percentages. Results are representative of three independent experiments performed on different donors. Statistics were performed using one-way ANOVA. ns: non significant. (C) DCs were exposed to mitomycin-treated C91-PL cells for 4 h, isolated by positive selection and cultured for 3 days, in presence or absence of AZT. Then, Jurkat-LTR-Luc cells were added and luciferase activity was assessed 2 days later. When indicated, IFN-α DCs were treated with chloroquine before their exposure to C91-PL. The graph shows the means and the standard deviations from 3 independent experiments obtained with different donors. Results were normalized to the luciferase activity of Jurkat-LTR-Luc measured after co-culture with IL-4 DCs and presented as fold changes over control. Asterisks indicate statistically significant differences calculated using one-way ANOVA followed by Bonferroni’s multiple comparison test: ***p<0.001, ** p<0.01, *p<0.05.

Fig 9. Schematic representation of HTLV-1 entry in IL-4, LPS-treated IL-4, and IFN-α DCs.

(A) Entry through macropinocytosis in IL-4 and LPS-treated DCs. After binding to its receptors NRP-1 and Glut-1, HTLV-1 enters using the transduction pathway characteristic of macropinocytosis, that is common in IL-4 and LPS-treated DCs. Specific drugs are shown on the right side of the first arrow. Neutral pH of macropinosomes from immature DCs is indicated in light grey, while acidic pH from mature DCs is indicated in dark grey. The presence of CD82 is indicated below each macropinosome. (B) Entry through clathrin-mediated endocytosis (CME) in IFN-α DCs. After binding to NRP-1 and Glut-1, HTLV-1 enters using clathrin and actin pathway. Specific drugs are shown on the right side of the arrow. The pH of vesicles is indicated by the grey color. See text for details.