Fig 1.
KSHV LANA recruits MRN (Mre11-Rad50-NBS1) complex.
(A) Co-immunoprecipitation of endogenous LANA and MRN proteins in BC3 cells. Cells were lysed using TBS-T buffer and the cell lysate was incubated with benzonase. After centrifugation, supernatant was incubated overnight with anti-LANA or IgG-control beads. The precipitated complexes were analyzed for the presence of endogenous Rad50, Mre11 and NBS1 by SDS-PAGE and immunoblotting. For the input, see Materials and methods. (B) Co-immunoprecipitation of endogenous LANA and Rad50 in BC3 cells. Co-immunoprecipitation of endogenous Rad50 was performed and analysed as in (A), but with anti-Rad50-antibody-coated-beads (left) or anti-LANA coated beads (right). The arrowhead indicates the smaller LANA forms co-immunoprecipitating with Rad50 (see text). (C) Schematic representation of LANA domain structure. NLS: Nuclear Localization Signal; TR: KSHV Terminal Repeats. (D) Pull-down assay with GST-fused LANA-C (aa 931–1162) and LANA-N (aa 1–312) proteins and HEK293T cell lysates. HEK293T were lysed with TBS-T buffer and incubated 4 hours with GST-fused proteins or GST alone, as negative control. Top: immunoblot for endogenous Rad50, Mre11 and NBS1 bound to GST-fused LANA fragments. Bottom: Ponceau staining to detect GST-fused proteins. (M) for marker.
Fig 2.
KSHV LANA recruits Rad50 and Mre11 in the cytosol.
(A) Co-immunoprecipitation of endogenous LANA, Rad50, Mre11 and Brd4 in BCBL-1 cells upon cytosolic-nuclear fractionation. Cells were lysed and cytoplasmic extracts (Cyto) and nuclear extracts (Nu) were prepared using the Thermo-Fischer Nu-Cyto fractionation kit following the manufacturer‘s instructions. Cytoplasmic and nuclear fractions were incubated overnight with sepharose beads coated with LANA-antibody or IgG-control. Left (INPUT, see Materials and methods): Brd4, Lamin A/C and GAPDH immunoblots were analyzed to confirm the efficiency of the fractionation. Right (IP): immunoprecipitation with LANA-antibody or IgG-control coated-beads and immunoblot for endogenous Rad50, Mre11 and Brd4. (B) Co-immunoprecipitation of endogenous Rad50 and full-length LANA or ΔN mutants (Δ161 and Δ282) transfected into HEK293 cells. HEK293 cells were transfected with LANA constructs (or empty vector). 48 hours later cells were lysed and incubated with benzonase. After centrifugation, cells were incubated overnight with beads coated with LANA-antibody. Left (INPUT): immunoblot to check the expression of LANA constructs and the endogenous Rad50 in the cells. Right (IP from LANA-antibody-coated-beads): immunoblot for endogenous Rad50 co-immunoprecipitation. (C) Co-immunoprecipitation of endogenous LANA and Rad50, Mre11 and CARD9 in latently KSHV-infected THP-1 cells (TrK.219 cells, see Materials and methods). Cells were lysed and incubated with benzonase. After centrifugation, whole cell lysates were incubated overnight with beads coated with anti-LANA or IgG-control antibody. Precipitated complexes were analysed by SDS-PAGE and immunoblotting with the indicated antibodies.
Fig 3.
LANA Δ161 modulates the activation of NF-kB cascade triggered by cytosolic DNA.
HeLa.CNX cells were transfected with the LANA Δ161 plasmid (or empty vector, EV) for about 48 hours. Cells were then stimulated with ISD (4μg/well) using Lipofectamine2000 following the manufacturer‘s instructions for 6 hours. Afterwards cells were lysed with TBS-T buffer and phosphorylation levels of p65 were analyzed by immunoblotting.
Fig 4.
Inhibition of p65 phosphorylation and KSHV lytic reactivation upon Mre11 silencing in KSHV infected cell lines.
(A) HeLa.CNX cells were stably infected with rKSHV.219 (see Materials and methods) and later reactivated using 20% RTA (vol/vol, see Materials and methods) and sodium butyrate (1.5 mM) for 24 hours: (i) RFP signal and (ii) ORF45 and K-bZIP expression were analyzed to confirm the KSHV lytic reactivation in these cells. (B) HeLa.CNX.rKSHV cells were transfected with siRNA against Mre11 (pool of 3 siRNA sequences) or negative control siRNA (see Materials and methods) for 48 hours. Cells were lysed with TBS-T buffer and protein levels analyzed by immunoblotting. (C-D) Downmodulation of p-p65 level and KSHV lytic reactivation upon Mre11 silencing, by siRNA transfection in BCBL-1 (C) and in TrK.219 (D) cells. Cells were microporated (see Materials and methods) with siRNA (300pmol/well) against Mre11 (pool of 3 siRNA sequences) or non targetting siRNA (as negative control). After 2 days cells were lysed with TBS-T buffer and KSHV K-bZIP or ORF45 and p-p65 levels were analyzed by immunoblotting. Phospho-p65 levels in Fig 5B–5D were digitally quantified (see Materials and methods).
Fig 5.
Inhibition of canonical NF-κB cascade and KSHV lytic reactivation upon LANA Δ161 overexpression.
(A-B) HeLa.CNX.rKSHV and the parental HeLa.CNX cells were transfected with LANA Δ161 plasmid or empty vector for 48 hours and (B) treated with 5% RTA (vol/vol, see Materials and methods) for about 24 hours. Cells were lysed with TBS-T buffer and lysates resolved by immunoblotting. Phospho-p65 levels were digitally quantified (see Materials and methods). (C) HEK293 cells were transfected with the plasmid expressing full-length LANA, the truncated LANA isoforms or the empty vector (2 μg/well), together with an NF-κB reporter vector (200 ng/well). The luciferase activity was measured 48 hours later in duplicates and the statistical analysis was performed with two-tailed student’s t-test. Statistical significance of the difference between control (pcDNA3.1) and LANA (FL or Δ161 or Δ282) transfected samples is shown: (***) for p<0.005 and (ns) for not significant. (D) HeLa.CNX cells were transfected with the plasmid expressing vFLIP or the empty vector (1μg) and additionally with the plasmid expressing full-length LANA (1μg) or the truncated isoform (Δ161, 0.5–1μg) or the empty vector (2μg of plasmid DNA in total per condition), together with NF-κB reporter vector (200 ng/well). The luciferase activity was measured 48 hours later in duplicates and the statistical analysis was performed with two-tailed student’s t-test. Statistical significance of the difference between vFLIP alone and vFLIP-LANA (FL or Δ161) transfected samples is shown: (*) for p<0,05 and (ns) for not significant.
Fig 6.
Model of LANA antagonizing cytoplasmic DNA sensors.
Cytoplasmic KSHV LANA isoforms recruit and antagonize cellular DNA sensor proteins cGAS as well as the Rad50-Mre11-CARD9 complex to inhibit innate immunity responses (IFN-β and NF-κB) and support KSHV lytic reactivation from latency. During KSHV lytic reactivation, free viral DNA in the cytosol is detected by host DNA sensors, such as cGAS and the Rad50/Mre11/CARD9 complex. The cGAS-STING cascade leads to IFN-β production, whereas the Rad50/Mre11/CARD9 complex is responsible for NF-κB cascade activation. A cross-talk between these two pathways may also be possible as indicated by the dashed arrows (Fig 6). Triggering of the Rad50/Mre11/CARD9 complex leads to the activation and nuclear accumulation of NF-κB p65 and the subsequent production of chemokines and cytokines. These events would interfere with the efficient KSHV lytic replication and therefore KSHV LANA ΔN isoforms block these signalling cascades by recruiting and inhibiting the upstream activators (cGAS as well as Rad50/Mre11).