Human cytomegalovirus UL18 utilizes US6 for evading the NK and T-cell responses

Abstract

Human cytomegalovirus (HCMV) US6 glycoprotein inhibits TAP function, resulting in down-regulation of MHC class I molecules at the cell surface. Cells lacking MHC class I molecules are susceptible to NK cell lysis. HCMV expresses UL18, a MHC class I homolog that functions as a surrogate to prevent host cell lysis. Despite a high level of sequence and structural homology between UL18 and MHC class I molecules, surface expression of MHC class I, but not UL18, is down regulated by US6. Here, we describe a mechanism of action by which HCMV UL18 avoids attack by the self-derived TAP inhibitor US6. UL18 abrogates US6 inhibition of ATP binding by TAP and, thereby, restores TAP-mediated peptide translocation. In addition, UL18 together with US6 interferes with the physical association between MHC class I molecules and TAP that is required for optimal peptide loading. Thus, regardless of the recovery of TAP function, surface expression of MHC class I molecules remains decreased. UL18 represents a unique immune evasion protein that has evolved to evade both the NK and the T cell immune responses.

Figure 1. Cell surface expression of UL18 was TAP-dependent.

HeLa, HeLa-US6, and HeLa-ICP47 cells were infected with either vvUL18 or vvWT. Cells were preincubated with mAb W6/32 (anti-MHC class I), 10C7 (anti-UL18), or normal mouse IgG (negative control) and stained with FITC-conjugated anti-mouse IgG. Fluorescence profiles are shown. Dashed lines indicate the negative control. (A) The surface expression of UL18 was not influenced by US6, the self-TAP inhibitor. (B) The surface expression of UL18 was down regulated by ICP47, the HSV-derived nonself- TAP inhibitor. (C) Down-regulation of UL18 surface expression in the T2, TAP-deficient cell line. T1 was included as a TAP-positive control. Representative FACS plots are shown.

Figure 2. Peptide transport inhibited by US6 was restored upon expression of UL18.

HeLa, HeLa-US6, and HeLa-ICP47 cells were infected with either vvWT or vvUL18. Cells were permeabilized with Streptolysin O and incubated with the FITC-conjugated peptides in the presence (filled bars) or absence of ATP (open bars). Transported peptides were quantified by fluorescence measurements (excitation and emission at 485 and 520 nm, respectively). The y-axis reflects peptide transport expressed as a percentage of translocation relative to the translocation observed in control cells. Averages and S.D. were calculated from three independent experiments.

Figure 3. Molecular mechanism for the reversal of US6-mediated TAP inhibition by UL18.

(A) UL18 restored ATP binding by TAP. HeLa and HeLa-US6 infected with either vvWT or vvUL18 were extracted in 1% digitonin. Digitonin lysates were incubated with N-6 ATP-agarose for 1 h, and agarose-bound proteins were separated by 10% SDS-PAGE and analyzed by immunoblotting with TAP1-specific antibodies. As a loading control, aliquots of cell lysates were probed with TAP1 and US6-specific antibodies. (B) UL18 interfered with the association between US6 and TAP1. HeLa-US6 cells were infected with either vvWT or vvUL18, lysed in 1% digitonin, and immunoprecipitated with anti-US6 antibodies. The US6 immunoprecipitates were separated by 8% SDS-PAGE and probed with polyclonal TAP1 and TAP2 antibodies. Cell lysates were subjected to immunoblotting with anti-US6 and immunoprecipitation with anti-UL18 to verify the expression of US6 and UL18. GAPDH was included to serve as a loading control. (C) UL18 bound to both TAP1 and TAP2. HeLa cells infected with vvUL18 were lysed in 1% digitonin and immunoprecipitated with either anti-UL18 mAb, 10C7, or anti-MHC class I polyclonal Ab, H-300 followed by blotting with TAP1 antibody (top panel) and TAP2 antibody (middle panel). GAPDH was included to verify equal loading of proteins (bottom panel). (D) US6 inhibited the association of UL18 with TAP2. Identical analysis was performed as described in Figure 3C , except for the use of HeLa cells stably expressing US6. Data are representative of three independent experiments with similar results.

Figure 4. The recovery of TAP function by UL18 was not sufficient to restore the cell surface expression of MHC class I.

(A) The surface expression of MHC class I molecules was not restored despite the recovery of TAP activity mediated by UL18 and US6. HeLa-US6 cells were infected with vvUL18 and then analyzed for MHC class I surface expression by flow cytometry after preincubation with mAb W6/32. (B) Specific down-regulation of MHC class I surface expression by the vaccinia virus- expressed UL18. HeLa cells were infected with vvUL18, stained for 1 hr with the indicated antibodies, and analyzed by FACS. C) Down-regulation of MHC class I molecules by UL18 expressed by a transient transfection system. HeLa and HeLa-US6 cells were transfected with either mock vector or UL18 cDNA. Cells were analyzed by FACS using mAbs 10C7 and W6/32 as described in Figure 4C . The results are representative of three independent experiments.

Figure 5. UL18 impaired optimal peptide loading of MHC class I molecules.

HeLa (A) and HeLa-US6 cells (B) infected cells with either vvWT or vvUL18 were radiolabeled for 30 min and then lysed in 1% NP-40 lysis buffer. Equal lysate aliquots were incubated at 4°C, 37°C, or 50°C for 40 min prior to immunoprecipitation with mAb W6/32. For each temperature, the intensity of the MHC class I heavy chain radioactive band was quantified. Similar results were obtained in three independent experiments.

Figure 6. UL18 interfered with the assembly of the peptide-loading complex.

(A) The expression level of the peptide-loading complex components was not affected by UL18 or US6. HeLa (−US6) and HeLa-US6 cells (+US6) infected with either vvUL18 or vvWT were lysed with 1% NP-40 Whole cell lysates were subjected to western blot analysis with the antibodies specific to MHC class I, TAP1, or tapasin. GAPDH served as a loading control. (B) UL18 inhibited the MHC class I-TAP1 and MHC class I-tapasin interactions. Infected cells were lysed in 1% digitonin and immunoprecipitated with anti-MHC class I antibody. Eluted proteins from the immunoprecipitates were resolved by SDS-PAGE, and the blots were probed with anti-TAP1 (top panel) and anti-tapasin antibody (middle panel). The same blot was stripped and reprobed with anti-MHC class I antibody as a loading control (bottom panel). (C) In the presence of US6, UL18 disturbed the association between TAP1 and MHC class I. Analysis was performed as described in Figure 6B , except HeLa-US6 cells were used. (D) Neither UL18 nor US6 affected the interaction between tapasin and TAP1. HeLa (−) or HeLa-US6 cells (+) were infected with vvUL18. Whole cell lysates of 1% digitonin were immunoprecipitated with anti-tapasin antibody, and blots were probed with anti-TAP1 antibody. Data (A–D) are representative of at least three independent experiments.

Figure 7. Hypothetical model for immune regulation of UL18 in the presence of US6.

US6 binds to both TAP1 and TAP2, resulting in inhibition of ATP binding by TAP1 and, thereby, blocking TAP-mediated peptide translocation. Upon coexpression of UL18 and US6, TAP1 preferentially binds to UL18 relative to US6 and MHC class I, counteracting the US6-mediated inhibition of TAP. US6 preferentially binds to TAP2 over UL18, and thus, the US6-TAP2 interaction remains unaffected by UL18. As a result, even in the presence of ‘self-derived’ US6 TAP inhibitor, UL18 acquires peptides for assembly, whereas MHC class I molecules fail to load peptides due to incorrect assembly of the peptide-loading complex, despite the recovery of TAP function.