UL49.5 proteins
Verweij, M.C.
Citation
Verweij, M. C. (2010, September 29). Unmasking the masters of evasion : TAP inhibition by varicellovirus UL49.5 proteins. Retrieved from https://hdl.handle.net/1887/15995
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden
Downloaded
from: https://hdl.handle.net/1887/15995
Note: To cite this publication please use the final published version (if applicable).
Abstract
Viral infections are counteracted by virus‐specific cytotoxic T cells that recognize the infected cell via MHC I molecules presenting virus‐derived peptides. The loading of the peptides onto MHC I molecules occurs in the endoplasmic reticulum (ER) and is facilitated by the peptide loading complex. A key player in this complex is the transporter associated with antigen processing (TAP), which translocates the viral peptides from the cytosol into the ER. Herpesviruses have developed many strategies to evade cytotoxic T cells. Several members of the genus Varicellovirus encode a UL49.5 protein that prevents peptide transport through TAP. These include bovine herpesvirus 1 (BoHV‐1), pseudorabies virus, and the equine herpesvirus 1 and 4. BoHV‐1 UL49.5 inhibits TAP by preventing conformational changes essential for peptide transport and by inducing degradation of the TAP complex. UL49.5 consists of an ER‐luminal N‐terminal domain, a transmembrane domain and a cytosolic C‐terminal tail domain.
In this study, the following features of UL49.5 were deciphered: 1) the ER‐luminal and the transmembrane domain of UL49.5 are both required for its function and cannot be substituted for, 2) chimeric constructs of BoHV‐1 and VZV UL49.5 attribute the lack of TAP inhibition by VZV UL49.5 to its ER‐luminal domain, 3) the ER‐luminal and TM domains of UL49.5 are required for efficient interaction with TAP, 4) the C‐terminal RXRX sequence is essential for TAP degradation by BoHV‐1 UL49.5, and 5) in addition to the RXRX sequence, the cytoplasmic tail of BoHV‐1 UL49.5 carries a motif that is required for efficient TAP inhibition by the protein. A model is presented depicting how the different domains of UL49.5 may block the translocation of peptides by TAP and target TAP for proteasomal degradation.
Introduction
Herpesviruses establish a lifelong infection in their host, which is accompanied by intermittent reactivations. This persistent infection is characterized by a delicate balance between the host immune response and immune evasion strategies employed by the virus.
Herpesvirus’ interference with the host immune system ranges from innate to adaptive immune responses. These viruses code for specific molecules that counteract the rising immune response through the inhibition of, amongst others, chemokine, cytokine and toll‐
like receptor signaling pathways that are required for the induction of a sustained immune response. Virus‐specific cytotoxic T cells (CTL) that do arise despite these countermeasures are inhibited by immune evasion proteins that interfere with the function of major histocompatibility complex class I (MHC I) molecules (Hansen and Bouvier, 2009; Griffin et al., 2010; Koch and Tampe, 2006).
MHC I molecules present viral peptides derived from cytosolic proteins that are degraded by proteasomes. The transporter associated with antigen processing (TAP) translocates the peptides into the endoplasmic reticulum (ER) where they are loaded onto MHC I molecules. TAP consists of two multi‐membrane‐spanning subunits, TAP1 and TAP2, that constitute a channel through which viral peptides are transported in an ATP‐
dependent manner. Quality controlled loading of the peptides onto MHC I molecules is facilitated by a number of chaperone molecules that link TAP to the MHC I molecules and stabilize the resulting complex. These molecules include tapasin, ERp57 and calreticulin.
Assembled MHC I molecules travel from the ER to the cell surface to present the peptides, including those derived from viral proteins, to specific CTL (Lehner and Trowsdale, 1998;
Wright et al., 2004; Cresswell et al., 2005; Peaper and Cresswell, 2008).
Herpesviruses interfere with many steps of the MHC I antigen presentation route. A favorable target within this pathway is the TAP transporter. Currently, four families of TAP inhibitors have been identified. Herpes simplexvirus 1 and 2 code for ICP47, a cytosolic protein that inhibits TAP function by interfering with peptide binding to the transporter (Fruh et al., 1995; Hill et al., 1995; Tomazin et al., 1996; Ahn et al., 1996; Aisenbrey et al., 2006). Human cytomegalovirus (HCMV) and rhesus CMV inhibit TAP via the US6 protein.
This protein prevents ATP binding to the transporter, thereby limiting the energy supply to the complex (Ahn et al., 1997; Lehner et al., 1997; Hengel et al., 1997; Hewitt et al., 2001;
Halenius et al., 2006; Pande et al., 2005). The Epstein‐Barr virus‐encoded BNLF2a protein blocks both the binding of peptides and of ATP to TAP (Hislop et al., 2007; Horst et al., 2009). Functional homologs of this protein were also found in herpesviruses of Old World primates (Hislop et al., 2007). TAP‐inhibiting UL49.5 proteins have been identified for many varicelloviruses, these proteins appear to be remarkably heterogeneous in function.
Bovine herpesvirus 1 (BoHV‐1) UL49.5 inhibits TAP by preventing conformational changes that are required for the transport of peptides from the cytosol to the ER. Additionally, TAP complexes are sent for proteasomal degradation by the viral protein (Koppers‐Lalic et al., 2005). A strong inhibition of TAP and degradation of the transporter by UL49.5 was observed in cells of bovine, human and mouse origin (Koppers‐Lalic et al., 2005; Koppers‐
Lalic et al., 2008; van Hall et al., 2007). UL49.5 possesses a dual role in viral infection; in addition to immune evasion, UL49.5 guides the maturation of glycoprotein M (gM), which is required for proper virion formation and spread (Lipinska et al., 2006). Homologs of UL49.5, or glycoprotein N, are found in every herpesvirus sequenced to date (McGeoch et al., 2006; Davison et al., 2009). Whereas the gM chaperone function of UL49.5 appears to be conserved amongst herpesviruses (Rudolph et al., 2002; Klupp et al., 2000; Ziegler et al., 2005), the capacity to block TAP has been found for varicelloviruses only. Like BoHV‐1, equine herpesvirus (EHV) 1 and 4, and pseudorabiesvirus (PRV) encode a UL49.5 homolog that mediates immune evasion via TAP inhibition (Koppers‐Lalic et al., 2005; Koppers‐Lalic et al., 2008). Despite these similarities, the mechanism of TAP inhibition differs remarkably. While BoHV‐1, EHV‐1, EHV‐4, and PRV UL49.5 all prevent essential conformational changes within the TAP complex, only BoHV‐1 UL49.5 mediates TAP degradation. In contrast, EHV‐1 and EHV‐4 UL49.5 block ATP binding to the complex, thereby limiting the energy supply required for peptide translocation (Koppers‐Lalic et al., 2008). BoHV‐1, EHV‐1, EHV‐4, and PRV belong to the genus Varicellovirus. Outside this genus, no TAP‐inhibiting UL49.5 proteins have been found (Koppers‐Lalic et al., 2005).
However, even amongst the varicelloviruses, homologs that fail to block TAP have been identified. Varicella‐zoster virus (VZV) UL49.5 interacts with TAP, but interference with its function has not been found (Koppers‐Lalic et al., 2008). The UL49.5 homolog of canine herpesvirus 1 (CHV‐1) causes only a minor reduction in peptide transport by canine TAP to some extent (Koppers‐Lalic et al., 2008). Thus, effective UL49.5‐mediated TAP inhibition is
Fig. 1, alignment of varicellovirus‐encoded UL49.5 proteins. The amino acid sequences of of UL49.5 proteins were aligned using VectorNTI software (Invitrogen). Previous work has shown that of these varicellovirus‐encoded homologs only BoHV‐1, PRV, EHV‐1, and EHV‐4 are able to block TAP (names shown in bold), while CaHV‐1 and VZV had minor or no affect, respectively, on TAP function. N‐terminal signal sequences (dashed line) and TM regions (bold line) are indicated. The proline conserved amongst TAP‐inhibiting homologs is boxed.
limited to a subgroup of the varicelloviruses that includes BoHV‐1, EHV‐1, EHV‐4 and PRV UL49.5.
UL49.5 proteins consist of a cleavable signal sequence, an ER‐luminal domain, a transmembrane (TM) domain, and a carboxy‐terminal cytosolic tail (Fig. 1). The cytoplasmic domain of the BoHV‐1 UL49.5 protein mediates TAP degradation (Koppers‐
Lalic et al., 2005). Deletion of this domain resulted in reduced inhibition of peptide transport (Koppers‐Lalic et al., 2005). In this study, we aim to further elucidate the contribution of the different UL49.5 domains to TAP inhibition and degradation. In addition, designated point mutations are introduced to reveal the contribution of individual amino acids to the various functions of UL49.5.
Materials and methods
The construction of chimeric proteins and mutant forms of UL49.5 All PCR‐generated products were inserted into the retroviral expression vector pLZRS‐IRES‐GFP, behind the HCMV E1 promotor‐enhancer and upstream of an internal ribosome entry site (IRES) element, which is followed by GFP. The identity of the constructs was verified by DNA sequencing. The sequence of the described primers can be found in table 1.
BHVER‐VZVTM and VZVER‐BHVTM were constructed using a plasmid coding for BoHV‐1 UL49.5 (Koppers‐Lalic et al., 2005) and one coding for VZV UL49.5 (Koppers‐Lalic et al., 2008) as a template for standard PCR reactions using the polymerases Pfu (Invitrogen) and Taq (Promega). For BHVER‐VZVTM, the sequence coding for the ER‐luminal domain of BoHV‐1 UL49.5 was amplified using primer 1 and 2 and the sequence coding for the TM domain of VZV UL49.5 was amplified using primers 3 and 4. Through fusion PCR, both constructs were combined and amplified using primers 1 and 4. The fused insert was cloned into pLZRS by GATEWAY technology (pDEST‐LZRS‐IRES‐GFP was constructed by K.
Franken, Department of Immunohematology and Blood Transfusion, Leiden University Medical Center, The Netherlands, and Invitrogen) using primers 5 and 6. VZVER‐BHVTM was constructed via the same method using primers 7 and 8 for amplification of the sequence coding for the ER‐luminal domain of VZV UL49.5, primers 9 and 10 for the amplification of the sequence coding for the TM domain of BoHV‐1 UL49.5, and primers 7 and 10 for the fusion PCR. Primers 11 and 12 were used to amplify the fused insert before insertion into pLZRS via GATEWAY technology.
For BHVER‐CD3TM, the sequence coding for the ER‐luminal domain of BoHV‐1 UL49.5 was amplified by standard PCR using Pfu and Taq and the primers 13 and 14, the sequence coding for the TM domain of CD3δ was amplified using primers 15 and 16, after which both PCR products were fused using primers 13 and 16. Primers 17 and 18 were used to amplify the product before insertion into pLZRS via GATEWAY technology.
anchor
Table 1, PCR primers
#a Constructa Primer sequence a
1 BHVER‐VZVTM Fw: 5’‐ CTGCAGACCATGCCGCGG ‐3’
2 Rev: 5’‐ GTAGAAGAAAAGCAGGGCCTGCGGTGGCT ‐3’
3 Fw: 5'‐ CCGCAGGCCCTGCTTTTCTTCTACGCATCCCTTTTGG ‐3' 4 Rev: 5’‐ AGATCTAATGAACACGCATGATAAGCTAACGAAATAAG ‐3’
5 Fw: 5’‐ TCTAGATTAGAAACACGCATGATAAGCTAACGAAATAAG ‐3’
6 Rev: 5’‐ GGGGACCACTTTGTACAAGAAAGCTGGGTCCTCGAGTTAGAAACACGCATGATAAGCTAACGAAATAAG ‐3’
7 VZVER‐BHVTM Fw: 5’‐ CTGCAGACCATGGGATCAATTACCGCTTCGTTC ‐3’
8 Rev: 5’‐ GTAAAAAACAACGGTGGTGATCATTGATCCGTCG ‐3' 9 Fw: 5’‐ ATGATCACCACCGTTGTTTTTTACGTGGCCCTGCTGGACC ‐3’
10 Rev: 5’‐ TCTAGATTAAAAGCAAAGCCCGTACGCGT ‐3’
11 Fw: 5’‐ GGGGACAAGTTTGTACAAAAAAGCAGGCTGAATTCACCATGGGATCAATTACCGCTTCGTTC ‐3’
12 Rev: 5’‐ GGGGACCACTTTGTACAAGAAAGCTGGGTCCTCGAGTTAAAAGCAAAGCCCGTACGCGT ‐3’
13 BHVER‐CD3TM Fw: 5’‐ GAATTCACCATGCCGCGGTCGCCG ‐3’
14 Rev: 5'‐ GTGGCTGGCAGGGCCTGCGGTGGCTC ‐3’
15 Fw: 5’‐ GGCCCTGCCAGCCACCGTGGCTGGC ‐3’
16 Rev: 5’‐ CTCGAGTCATCCAGCAAAGCAGAAGACTCC ‐3’
17 Fw: 5’‐ GGGGACAAGTTTGTACAAAAAAGCAGGCTGAATTCACCATGCCGCGGTCGCCG ‐3’
18 Rev: 5’‐GGGGACCACTTTGTACAAGAAAGCTGGGTCCTCGAGTCATCCAGCAAAGCAGAAGACTCC ‐3’
19 CD3ER‐BHVTM+tail Fw: 5’‐ GAATTCACCATGGAACATAGCACGTTTCTCTCTGGC ‐3’
20 Rev: 5’‐ GTAAAAAACAACATCCAGCTCCACACAGCTCTGGC ‐3’
21 Fw: 5’‐ GGAGCTGGATGTTGTTTTTTACGTGGCCCTGACCG ‐3’
22 Rev: 5’‐ CTCGAGTCAGCCCCGCCCCCGC ‐3’
23 CD3ER+TM‐BHVtail Fw: 5’‐ GAGCCTTCCAGCAAAGCAGAAGACTCCC ‐3’
24 Rev: 5’‐ CTTTGCTGGAAGGCTCATGGGCGCCAGC ‐3’
27 UL49.5 P48A Fw: 5’‐ CGGTGGCTCCGAGAGGGCCACCCCGCGCGCGTA ‐3’
28 Rev: 5’‐ TACGCGCGCGGGGTGGCCCTCTCGGAGCCACCG ‐3’
29 UL49.5 P48G Fw: 5’‐ CGGTGGCTCCGAGAGTCCCACCCCGCGCGCGTA ‐3’
30 Rev: 5’‐ TACGCGCGCGGGGTGGGACTCTCGGAGCCACCG ‐3’
31 UL49.5 KK→AA Fw: 5’‐ CAGAATTCACCATGCCGCGGTCG ‐3’
32 Rev: 5’‐ GGGGAATTCTTTCAGCCCCGCCCCCGCGACTCCGCGGCATTGGGC ‐3’
33 UL49.5 SS→AA Fw: 5’‐ CGGGATCCCACCATGCCGCGGTCG ‐3’
34 Rev: 5’‐ CTTTTTATTGGGCCCGGCGGCGCCCATG ‐3’
35 Rev: 5’‐ CGCCCCCGCGCCTCCTTTTTATTGGG ‐3’
36 Rev: 5’‐ GGAATTCAGCCCCGCCCCCGCGC ‐3’
37 UL49.5 AA/AA Rev: 5’‐ GCGGCATTGGGCCCGGCGGCGCCCATG ‐3’
38 Rev: 5’‐ CGCCCCCGCGCCTCCGCGGCATTGGG ‐3’
39 UL49.5 G96A Rev: 5’‐ GGAATTCACGCCCGCCCCCGCGAC ‐3’
40 UL49.5 R95A Rev: 5’‐ GGAATTCAGCCCGCCCCCCGCGAC ‐3’
41 UL49.5 G94A Rev: 5’‐ GGAATTCAGCCCCGCGCCCGCGAC ‐3’
a = restriction sites used in bold
The BHVER‐TLR2TM construct was obtained by de novo gene synthesis (Genscript). The construct was inserted into the pLZRS via the BamHI and EcoRI restriction sites. For CD3ER‐ BHVTM+tail, the sequence coding for the ER‐luminal domain of CD3δ was amplified by standard PCR using Pfu and Taq and the primers 19 and 20, the sequence coding for the TM and tail domain of BoHV‐1 UL49.5 were amplified using primers 21 and 22, after which both PCR products were fused using primer 19 and 22. For CD3ER+TM‐BHVtail, the the sequence coding for the ER‐luminal and the TM domain of CD3δ were amplified using primers 19 and 23, the sequence coding for the tail domain of BoHV‐1 UL49.5 was amplified using primers 24 and 22, after which both PCR products were fused using primers 19 and 20. The resulting inserts were cloned into pLZRS via the BamHI and EcoRI restriction sites.
For UL49.5 P48A and P48G point mutations were introduced into UL49.5 wt by Quickchange site directed mutagenesis using Pfu and primers 27 and 28, and 29 and 30, respectively. The mutants were inserted into pLZRS via the BamHI and EcoRI restriction sites.
To create UL49.5 KK→AA point mutations were introduced into UL49.5 wt by standard PCR with Pfu using primers 31 and 32. For UL49.5 SS→AA, UL49.5 wt was used as a template for three sequential PCR reactions using KOD DNA polymerase (Novagen‐Merck).
The forward primer 33 was used in combination with primer 34, followed by two amplification steps with primers 35 and 36. For UL49.5 AA/AA the mutant UL49.5 SS→AA was used as a template. Again three sequential reactions were performed with primer 33 as forward primer and, in this order, primers 37, 38 and 36.
The alanine scan of the tail domain resulting in G96A, R95A, G94A, R93A, and AAAA was performed by introducing point mutations into UL49.5 wt by standard PCR using KOD DNA polymerase. The forward primer 33 was used in combination with primers 39, 40, 41,
Table 1, continued
#a Constructa Primer sequence a
42 UL49.5 R93A Rev: 5’‐ CGAATTCAGCCCCGCCCCGCCGACTCCG ‐3’
43 UL49.5 AAAA Rev: 5’‐ GGAATTCACGCTGCTGCTGCCGACTCCTTTTTATTG ‐3’
44 UL49.5 P87A Rev: 5’‐ GCGACTCCTTTTTATTGGCCCCGGCGG ‐3’
45 Rev: 5’‐ CGAATTCAGCCCCGCCCCCGCGACTC ‐3’
46 UL49.5 R95K Fw: 5’‐ CCGGAATTCCGGATGCCGCGGTCGCCGCTCATCGTTG ‐3’
47 Rev: 5’‐ CCGCTCGAGTCAGCCCTTCCCCCGCGACTCCTTTTTATTG ‐3’
48 UL49.5 R93K Rev: 5’‐ CCGCTCGAGTCAGCCCCGCCCCTTCGACTCCTTTTTATTG ‐3’
49 UL49.5 R93/95K Rev: 5’‐ CCGCTCGAGTCAGCCCTTCCCCTTCGACTCCTTTTTATTG ‐3’
a = restriction sites used in bold
42, and 43, respectively. P87A was created using primer 33 and two sequential amplification using primers 44 and 45.
The arginine point mutations resulting in R95K, R93K, and R93/95K were introduced into the UL49.5 wt by standard PCR using Phusion HF DNA polymerase (Finnzymes). The forward primer 46 was used in combination with primers 47, 48 or 49 (respectively).
Cell lines and recombinant viruses The human melanoma cell line Mel JuSo (MJS), MJS BoHV‐1 UL49.5, MJS BoHV‐1 UL49.5Δtail and MJS BoHV‐1 UL49.5 sol were constructed as described before (Koppers‐Lalic et al., 2005; Lipinska et al., 2006) and maintained in RPMI‐
1640 medium supplemented with 10% heat‐inactivated fetal bovine serum (FBS), 2 mM L‐glutamine (Invitrogen), 140 IU/ml penicillin and 140 mg/ml streptomycin.
Recombinant viruses were made using the Phoenix amphotropic packaging system as described before (www.stanford.edu/group/nolan/retroviral_systems/retsys.html). The retroviruses were used to transduce target cells, after which GFP positive cells were selected using a FACSAria cell sorter (Becton Dickinson). The following stable cell lines were generated: MJS expressing BHVER‐VZVTM, VZVER‐BHVTM, BHVER‐CD3TM, BHVER‐TLR2TM, CD3ER‐BHVTM+tail, CD3ER+TM‐BHVtail, UL49.5 P48A, UL49.5 P48G, UL49.5 KK→AA, UL49.5 SS→AA, UL49.5 AA/AA, UL49.5 G96A, UL49.5 R95A, UL49.5 G94A, UL49.5 R93A, UL49.5 AAAA, UL49.5 P87A, UL49.5 R95K, UL49.5 R93K, and UL49.5 R93/95K.
Antibodies The following antibodies were used for flow cytometry: anti‐human MHC I complexes mAb B9.12.1 (kindly provided by Bernard Malissen, Centre d’Immunologie Marseille‐Luminy, France) and anti‐human MHC II HLA‐DR mAb L243 (ATCC). For detection of UL49.5, previously raised rabbit polyclonal anti‐sera against synthetic peptides derived either from the N‐terminal domain (H11/nt) or from the C‐terminal domain (H19/ct) of BoHV‐1 UL49.5 were used (Lipinska et al., 2006). Anti‐VZV UL49.5 was made earlier using two synthetic peptides from both the N‐ and C‐terminal domain of the protein, as described before (Koppers‐Lalic et al., 2008). In addition, we used anti‐TAP1 mAb 148.3 (Meyer et al., 1994), anti‐TAP2 mAb 435.3 (kind gift from P. van Endert, INSERM, U580, Université Paris Descartes, Paris, France), and anti‐β‐actin mAb AC‐74 (Sigma‐Aldrich), as a control.
Flow cytometry Surface levels of MHC I and MHC II molecules were determined by flow cytometry. Cells were stained with the indicated primary antibodies and, after washing, with the secondary goat anti‐mouse allophycocyanin Ab (Leinco Technologies) or goat anti‐mouse phycoerythrin Ab (Jackson ImmunoResearch Laboratories) at 4°C. Stained cells were measured using a FACSCalibur (Becton Dickinson) and analyzed using CellQuest software.
Peptide transport assay Cells were permeabilized using Streptolysin‐O (Murex Diagnostics) at 37°C for 10 min. Permeabilized cells were incubated with 4.5 µM of the fluorescein‐
conjugated synthetic peptide CVNKTERAY (N‐core glycosylation site underlined) in the presence of 10 mM ATP or 0.125 M EDTA at 37°C for 10 min. Peptide translocation was terminated by adding 1 ml of ice‐cold lysis buffer (1% Triton X‐100, 500 mM NaCl, 2 mM MgCl2, 50 mM Tris HCl, pH 8.0). After lysis for 30 min at 4°C, cells were centrifuged at 16,000 g for 20 min at 4°C in order to obtain post‐nuclear lysates. Glycosylated peptides were isolated from these lysates by incubation with concanavalin A Sepharose beads (GE Healthcare) for 2 h at 4°C. After washing of the beads, glycosylated peptides were eluted from the beads with elution buffer (500 mM mannopyranoside, 10 mM EDTA, 50 mM Tris HCl pH 8.0) during a one hour incubation step at room temperature. Fluorescence was measured using a Mithras LB 940 multilabel reader (Berthold Technologies).
Immunoprecipitations and Western blotting For immunoprecipitations, cells were resuspended in a lysis buffer containing 1% (w/v) digitonin, 50 mM Tris HCl (pH 7.5), 5 mM MgCl2, 150 mM NaCl, 1mM leupeptin, and 1 mM 4‐(2‐aminoethyl)benzenesulfonyl fluoride. Lysates were incubated with anti‐TAP1 and protein‐G and ‐A Sepharose beads (GE Healthcare) to isolate the immune complexes. Precipitated immune complexes and 1%
Nonidet P‐40 lysates of the cells were separated by SDS‐PAGE and subsequently transferred to PVDF membranes (GE Healthcare). UL49.5 proteins were separated using 16.5%‐tricine PAGE. The blots were incubated with the indicated antibodies, followed by horseradish peroxidase (HRP)‐conjugated secondary antibodies (DAKO and Jackson ImmunoResearch Laboratories). Bound HRP‐labeled antibodies were visualized using ECL Plus (GE Healthcare).
Results
The ER‐luminal and TM domains of UL49.5 contribute essentially to TAP inhibition and degradation
To assess whether the ER‐luminal domain of UL49.5 domain inhibits TAP in the absence of a TM domain, a soluble UL49.5 protein, encompassing amino acid residues 1‐54, was constructed (Fig. 2A). Previously, it was shown that the deletion of the tail domain of UL49.5 results in intermediate MHC I downregulation and the loss of TAP degradation (Koppers‐Lalic et al., 2005). Wild‐type and recombinant proteins were expressed in human melanoma cells (MJS), after which MHC I cell surface expression was analyzed. Whereas wild‐type UL49.5 (UL49.5 wt) and UL49.5Δtail reduced MHC I expression to the expected levels, the expression of soluble UL49.5 (UL49.5 sol) did not result in decreased MHC I cell surface expression (Fig. 2B). MHC I downregulation was specific, as MHC II expression was not affected by UL49.5 (Fig. 2B, lower panel). The assessment of the steady state levels of all proteins revealed the presence of UL49.5 wt and UL49.5Δtail, but undetectable UL49.5
sol (Fig. 2C). Yet, UL49.5 mRNA transcripts were found in all cell lines, suggesting that the UL49.5 sol protein is unstable (Fig. 2D).
VZV UL49.5 has been shown to interact with TAP, but the protein does not block peptide transport by TAP (Koppers‐Lalic et al., 2008). To assess the contribution of the ER‐
luminal and the TM domains of BoHV‐1 and VZV UL49.5 to TAP binding and, in case of BoHV‐1 UL49.5, TAP inhibition, chimeras of both proteins were constructed. The ER‐
luminal domain of BoHV‐1 UL49.5 was fused to the TM domain of VZV UL49.5 (BHVER‐ VZVTM) and vice‐versa (VZVER‐BHVTM) (Fig. 3A). The chimeric proteins were expressed in MJS. Proper expression was confirmed using BoHV‐1 and VZV UL49.5‐specific antibodies (Fig. 3B). Expression of BHVER‐VZVTM resulted in a reduction in MHC I cell surface expression that was comparable to the reduction induced by UL49.5 deficient of its cytosolic domain (Fig. 3C). In contrast, VZVER‐BHVTM did not reduce MHC I levels (Fig. 3C).
None of the UL49.5 proteins reduced MHC II expression, indicating that the observed downregulation was specific for MHC I.
Fig. 2, the ER‐luminal domain of UL49.5, expressed as a soluble protein, is unstable. (A) Soluble UL49.5 (UL49.5 sol) encompasses amino acid residues 1 till 54 of UL49.5. SS: signal sequence, TM: transmembrane. (B) Surface expression of MHC I and MHC II molecules on control (graph 2) and UL49.5‐expressing cells (graph 3) was assessed through flow cytometry using the indicated antibodies. Graph 1: secondary antibody only. (C) The expression of the truncated UL49.5 proteins was evaluated by SDS‐PAGE and Western blotting (WB) using an antibody specific for the N‐terminus of the UL49.5. β‐actin was used as a loading control. (D) Transcription of the UL49.5 proteins was verified by RT‐PCR using specific primers. GAPDH expression was assessed as a control for loading. One representative experiment out of at least two independent experiments is shown.
These results described above show that the ER‐luminal domain of BoHV‐1 UL49.5 is unstable when expressed in the absence of its TM domain. However, when fused to the TM domain of VZV UL49.5, the ER‐luminal domain of BoHV‐1 UL49.5 can induce TAP inhibition. To evaluate if the ER‐luminal domain can still inhibit TAP when attached to an irrelevant membrane anchor, it was fused to the TM domain of CD3 (BHVER‐CD3TM) (Fig. 4A). When expression of the resulting BoHV‐1 UL49.5/CD3 chimera was verified, it appeared to be undetectable (Fig. 4B). Treating the cells with the proteasome inhibitor Cbz‐L3 for 3 or 6 h stabilized the chimeric protein, yet the expression level was still low compared to UL49.5Δtail (Fig. 4B). The assessment of MHC I cell surface expression levels on the BHVER‐CD3TM‐expressing cells revealed that this chimera was unable to downregulate MHC I, even after Cbz‐L3 treatment (Fig. 4C).
A second chimera was constructed in which the ER‐luminal domain of BoHV‐1 UL49.5 was fused to the TM domain of toll‐like receptor (TLR) 2 (BHVER‐TLR2TM) (Fig. 4A). In contrast to BHVER‐CD3TM, BHVER‐TLR2TM was detectable without Cbz‐L3 treatment.
Fig. 3, BHVER‐VZVTM induces MHC I downregulation. (A) UL49.5 chimeras were established by fusing the ER‐
luminal domain of BoHV‐1 UL49.5 to the transmembrane (TM) domain of VZV UL49.5 (BHVER‐VZVTM) and vice versa (VZVER‐BHVTM). BoHV‐1 UL49.5: dark gray blocks, VZV UL49.5: light gray blocks. SS: signal sequence. (B) Surface expression of MHC I and MHC II on untransduced cells (graph 2) and on cells expressing the UL49.5 proteins (graph 3) was assessed through flow cytometry using the indicated antibodies. Graph 1: background staining in the presence of secondary antibody only. (C) The expression of the UL49.5 chimeras was verified by SDS‐PAGE and Western blotting (WB) using antibodies specific for BoHV‐1 and VZV UL49.5. β‐actin was used as a loading control. One representative experiment out of at least two independent experiments is shown.
anchor
Fig. 4, the role of the UL49.5 TM domain in TAP inhibition. (A) The N‐terminal domain of BoHV‐1 UL49.5 was fused to the transmembrane (TM) domain of CD3 (light gray) or TLR2 (off‐white). SS: signal sequence. (B) BHVER‐ CD3TM expression in the presence or absence of Cbz‐L3 was evaluated by SDS‐PAGE and Western blotting (WB) using an antibody specific for the N‐terminus of the UL49.5. β‐actin was used as a loading control. (C) Surface expression of MHC I and MHC II molecules on control (graph 2) and UL49.5‐expressing cells (graph 3) was assessed through flow cytometry using the indicated antibodies. Graph 1: secondary antibody only. (D) BHVER‐ TLR2TM expression in the presence or absence of Cbz‐L3 was evaluated using the UL49.5 antibody raised against its N‐terminus (see B for legend). (E) MHC I and MHC II cell surface expression (see C for legend). (F) To visualize the interaction between TAP and BHVER‐TLR2TM, TAP complexes were immunoprecipitated (IP) from cell lysates using a TAP1‐specific antibody. The resulting complexes were stained for TAP1 or UL49.5. All experiments were repeated twice.
However, the steady state levels of this chimera were low and could only be increased slightly by treating the cells with the proteasome inhibitor (Fig. 4D). The assessment of MHC I cell surface levels revealed some BHVER‐TLR2TM‐induced downregulation, which was independent of Cbz‐L3 treatment (Fig. 4E). Cells expressing UL49.5Δtail displayed a higher degree of MHC I downregulation (Fig. 4E). To assess binding of BHVER‐TLR2TM to TAP, TAP1 was precipitated from the cells. While UL49.5Δtail co‐precipitated efficiently with the TAP complex, an interaction with BHVER‐TLR2TM was undetectable (Fig. 4F).
Summarizing the data obtained thus far, a fusion protein comprising the ER‐luminal domain of BoHV‐1 UL49.5 and the TM domain VZV UL49.5 efficiently mediates MHC I downregulation. However, when fused to the TM domain of TLR2, the ER‐luminal domain of BoHV‐1 UL49.5 had only a minor effect on MHC I expression. This indicates that TAP inhibition, and consequently MHC I downregulation, is mediated by the ER‐luminal domain of BoHV‐1 UL49.5, but the TM of UL49.5 is required for efficient inhibition.
To assess the necessity of the TM and cytosolic domains for TAP inhibition and TAP degradation, two chimeras were created in which the ER‐luminal domain or both the ER‐
luminal and TM domains of UL49.5 were replaced by the corresponding domains of CD3.
The resulting chimeras, CD3ER‐BHVTM+tail and CD3ER+TM‐BHVtail, are shown in Fig. 5A.
MHC I and MHC II levels were assessed on cells expressing UL49.5 wt and the two chimeras. While UL49.5 wt reduced MHC I levels to almost background levels, CD3ER‐ BHVTM+tail and CD3ER+TM‐BHVtail did not induce MHC I downregulation at all (Fig. 5B).
Analyzing the expression levels of the proteins revealed proper expression of CD3ER‐ BHVTM+tail, but CD3ER+TM‐BHVtail could not be detected (Fig. 5C). TAP1 and TAP2 levels appeared unaffected by the presence of CD3ER‐BHVTM+tail, while UL49.5 wt‐expressing cells did show diminished TAP levels (Fig. 5C). To investigate if CD3ER‐BHVTM+tail was able to interact with TAP, co‐immunoprecipitation studies were performed. UL49.5 wt was present in the complexes that co‐immunoprecipitated with TAP1, despite the fact that most TAP molecules were degraded in the presence of UL49.5 wt (Fig. 5C). An interaction between CD3ER‐BHVTM+tail and TAP1 appeared undetectable (Fig. 5D). These data indicate that the TM domain of UL49.5 contributes to stabilization of the protein. However, the presence of the ER‐luminal domain of UL49.5 is required for a stable interaction with TAP and the induction of its degradation.
Function of the ER‐luminal membrane‐proximal proline residue uniquely present in TAP‐inhibiting UL49.5 homologs
Comparing the protein sequences of ER‐luminal domains of the TAP‐inhibiting homologs with the non‐inhibiting homologs revealed a conserved proline at position 48 of the TAP‐
inhibiting homologs encoded by BoHV‐1, EHV‐1, EHV‐4, and PRV (Fig. 1). The fact that this
involvement of this residue in TAP inhibition. To establish the involvement of Pro48 in BoHV‐1 UL49.5‐induced TAP inhibition, this residue was substituted for an alanine or a glycine, the latter residue is also capable of introducing a kink in the structure.
MHC I cell surface expression and TAP function were determined to assess if the substitution of the Pro48 for either an alanine or a glycine influenced UL49.5‐mediated TAP inhibition. The MHC I downregulation induced by these mutants was entirely comparable to MHC I downregulation induced by UL49.5 wt (Fig. 6A). Also, UL49.5 P48A and P48G inhibited peptide transport to the same extent as UL49.5 wt (Fig. 6B). Analyses of the steady state levels of UL49.5, TAP1 and TAP2 confirmed proper UL49.5 expression and showed that UL49.5‐mediated degradation of both TAP subunits was also induced by the
Fig. 5, the UL49.5 cytoplasmic tail domain does not induce degradation of TAP in the absence of the N‐terminal domain of UL49.5. (A) Two chimeric proteins carrying either the N‐terminal or the N‐terminal domain of CD3
(dark gray blocks) fused to the tail and/or TM domain of BoHV‐1 UL49.5 (light gray blocks) were constructed. (B) Surface expression of MHC I and MHC II on untransduced cells (graph 2), on UL49.5 wt‐expressing cells (graph 3) and on cells expressing the UL49.5 mutants (graph 4) was assessed through flow cytometry using the indicated antibodies. Graph 1: secondary antibody only. (C) Steady state levels of UL49.5, TAP1, and TAP2 in control and UL49.5‐expressing MJS were determined by SDS‐PAGE and Western blotting (WB) using specific antibodies. β‐
actin was used as a loading control. (D) To visualize the interaction between TAP and the UL49.5 proteins, TAP complexes were immunoprecipitated (IP) from the cells and the resulting complexes were stained for TAP1, TAP2 or UL49.5. All experiments were repeated twice.
proline‐mutants (Fig. 6C). These data indicate that Pro48 does not contribute essentially to UL49.5‐induced TAP inhibition or degradation.
The contribution of the lysine and serine residues within the cytoplasmic domain of UL49.5 to the degradation of TAP
The cytoplasmic domain of BoHV‐1 UL49.5 is required for the degradation of TAP (Koppers‐Lalic et al., 2005). Viruses are known to utilize the ubiquitin pathway for immune evasion (Randow and Lehner, 2009). Ubiquitin is mostly attached to lysine residues, but ubiquitinylation was also observed at cysteine (Cadwell and Coscoy, 2005), serine, and threonine residues (Wang et al., 2007). The protein sequence of the BoHV‐1 UL49.5 cytoplasmic domain contains two lysine and two serine residues that are absent in the
Fig. 6, the proline kink at position 48 is not required for UL49.5‐mediated MHC I downregulation. Point mutations were introduced to replace Pro48 by alanine (P48A) or glycine (P48G). (A) Cell surface expression of MHC I and MHC II was determined via flow cytometry on control (graph 2), UL49.5 wt‐expressing cells (graph 3), and cells expressing mutant UL49.5 proteins (graph 4) using the indicated antibodies. Graph 1: secondary antibody only. (B) Peptide transport was analyzed in the presence of 10 mM ATP (black bars) or EDTA (white bars).
Peptide transport is expressed as percentage of translocation, relative to the translocation observed in control cells (set at 100%). (C) Steady state levels of UL49.5, TAP1, and TAP2 in control and UL49.5‐expressing MJS were determined by SDS‐PAGE and Western blotting (WB) using specific antibodies. β‐actin was used as a loading control. One representative experiment out of two or three independent experiments is shown.
C‐terminal domains of the other TAP‐inhibiting UL49.5 homologs that all fail to target TAP for proteasomal degradation (Fig. 1 and 7A). To investigate the role of potential ubiquitination sites of UL49.5 in TAP‐degradation, the lysine and serine residues in the tail domain were replaced by alanines. We constructed three different mutants of UL49.5, with
Fig. 7, TAP degradation does not depend on ubiquitinylation of lysine and serine residues in the cytoplasmic domain of UL49.5. (A) A schematic representation of the BoHV‐1 UL49.5 protein displaying the amino acid sequence of the cytoplasmic tail domain. Residues that were mutated throughout this study are shown in bold.
(B) The potential ubiquitin‐binding lysine and serine residues in the tail domain of UL49.5 were substituted for alanines. The effect of these mutations on the cell surface expression of MHC I and MHC II was determined on control (graph 2), and UL49.5‐expressing cells (graph 3) via flow cytometry using the indicated antibodies. Graph 1: secondary antibody only. (C) Steady state levels of UL49.5, TAP1, and TAP2 in control and UL49.5‐expressing MJS were determined by SDS‐PAGE and Western blotting (WB) using specific antibodies. β‐actin was used as a loading control. All experiments were performed in triplicate.
(1) all lysines substituted for alanines (UL49.5 KK→AA), (2) all serines substituted for alanines (UL49.5 SS→AA), and (3) both the lysines and serines substituted for alanines (UL49.5 AA/AA).
It was assessed if the mutations introduced in the tail of UL49.5 would have an effect on MHC I downregulation. The resulting phenotype was comparable to UL49.5 wt‐induced downregulation, indicating that the introduced point mutations did not affect downregulation (Fig. 7A). Western blot analysis revealed comparable expression for UL49.5 and its recombinants (Fig. 7B). TAP1 and TAP2 were degraded in cells expressing the UL49.5 wt and in cells expressing the mutated UL49.5 constructs. For reference, TAP expression was unaffected in UL49.5Δtail‐expressing cells (Fig. 7B). All together, the substitution of lysine and serine residues in UL49.5 C‐terminal tail domain did not result in impaired downregulation of MHC I nor affected UL49.5‐mediated degradation of TAP1, thereby excluding these amino acid residues as acceptor sites for ubiquitin in the context of TAP degradation.
The RGRG motif is required for UL49.5‐mediated degradation of TAP
To determine the tail‐residues involved in BoHV‐1 UL49.5‐induced TAP degradation, an alanine scan was performed on the cytosolic domain of the protein, starting at the C‐terminal residues of UL49.5. Four point mutations were introduced to construct five different mutants: UL49.5 G96A, R95A, G94A, R93A, and AAAA (Fig. 7A), in which all four residues of the RGRG sequence were replaced by alanines. The proline residue at position 66 was predicted to induce a turn in the cytosolic tail of UL49.5. To investigate if this amino acid residue is required for UL49.5‐induced TAP inhibition and degradation, Pro66 was substituted by an alanine (P87A).
Cells expressing the above described mutants and UL49.5 wt were analyzed for downregulation by examining MHC I cell surface expression. The observed downregulation of MHC I induced by the UL49.5 mutants was largely comparable to that obtained in the presence of UL49.5 wt (Fig. 8A). UL49.5 wt‐expressing cells displayed a 99% reduction in MHC I expression compared to control cells (Fig. 8B). The reduction induced by the alanine mutants ranged from 98% to 95%, implying that the introduced point mutations hardly affected UL49.5‐mediated MHC I downregulation. The observed phenotypes were specific, as MHC II expression was unaffected in all cells (Fig. 8A, lower panel). Next, TAP activity was assessed in cells expressing the alanine mutants and UL49.5 wt. TAP function was greatly reduced by all UL49.5 variants, yet the relative reduction ranged from 97% for UL49.5 wt to 80% for UL49.5 R93A (Fig. 8C). Western blot analysis of the lysates confirmed proper expression of all UL49.5 mutants (Fig. 8D). Subsequently, the steady state levels of TAP were determined. Remarkably, the UL49.5 mutants R95A, R93A, and AAAA largely lost
anchor
the capacity to induce TAP degradation, while UL49.5 G96A, G94A and P87A maintained the ability to degrade TAP1 and TAP2 (Fig. 8D).
The results described above identify Arg95 and Arg93 as amino acid residues essential for UL49.5‐induced degradation of TAP. To assess if degradation is dependent on the charge of the residues at position 95 and 93, Arg95 and Arg93 were replaced by lysines (Fig. 7A). These substitutions resulted in the UL49.5 mutants R95K, R93K, and R93/95K, with the latter having both Arg95 and Arg93 are replaced by lysines.
These point mutations did not affect MHC I downregulation, as MHC I levels on R95K, R93K, and R93/95K‐expressing cells were highly comparable to the UL49.5 wt‐expressing cells (Fig. 9A). MHC II was unaffected by the expression of UL49.5, illustrating the specificity of the downregulation (Fig. 9A, lower panel). As shown by Western blotting, all recombinant forms of UL49.5 were expressed (Fig. 9B). Evaluation of TAP1 and TAP2 steady state levels revealed that UL49.5 R93K and R93/95K lost the capacity to induce TAP degradation. In contrast, UL49.5 R95K reduced the transporter’s expression levels to the same extent as UL49.5 wt (Fig. 9B). Next, TAP was immunoprecipitated from the cells to assess if the interaction between UL49.5 and TAP was affected by the amino acid substitutions. All UL49.5 recombinants were found to co‐immunoprecipitate with TAP, indicating that the mutations did not influence the binding of UL49.5 to TAP (Fig. 9C). The amount of TAP1 and TAP2 detected in cells expressing UL49.5 wt and R95K was strikingly lower than that of cells expressing UL49.5 R93K and R93/95K (Fig. 9C), confirming that the latter two mutants have lost their capacity to mediate the degradation of TAP. These results indicate that a charged amino acid residue (Arg or Lys) at position 95 of UL49.5 is sufficient for TAP degradation. However, the arginine residue at position 93 of the protein appears to be critical.
Fig. 8 (opposite), Arg95 and Arg93 are indispensible for UL49.5‐induced degradation of TAP. The C‐terminal RGRG residues were both individually and collectively replaced by alanine residues. In addition, Pro87 was substituted for an alanine. (A) Surface expression of MHC I and MHC II was determined on untransduced cells (graph 2), on UL49.5 wt‐expressing cells (graph 3) and on cells expressing the UL49.5 recombinants (graph 4) via flow cytometry using the indicated antibodies. Graph 1: secondary antibody only. (B) Depicted in this graph: the percentage of MHC I expression on UL49.5‐expressing cells relative to the expression of MHC I on control cells (set at 100%). (C) Peptide transport was analyzed in the presence of 10 mM ATP (black bars) or EDTA (white bars).
Peptide transport is expressed as percentage of translocation, relative to the translocation observed in control cells (set at 100%). (D) Steady state levels of UL49.5, TAP1 and TAP2 in control and UL49.5‐expressing MJS were determined by SDS‐PAGE and Western blotting (WB) using specific antibodies. β‐actin was used as a loading control. *The upper band is due to non‐specific staining of the antibody. All experiments were performed in triplicate.
anchor
Fig. 9, Arg93 and a positive charge at residue 95 of UL49.5 are required for TAP degradation. By substituting Arg93 and Arg95 for lysines, the positive charge at these positions was maintained. (A) Surface expression of MHC I and MHC II on untransduced cells (graph 2), on UL49.5 wt‐expressing cells (graph 3) and on cells expressing the UL49.5 mutants (graph 4) was assessed through flow cytometry using the indicated antibodies. Graph 1:
secondary antibody only. (B) Steady state levels of UL49.5, TAP1, and TAP2 in control and UL49.5‐expressing MJS were determined by SDS‐PAGE and Western blotting (WB) using specific antibodies. β‐actin was used as a loading control. (C) To visualize the interaction between TAP and the UL49.5 proteins, TAP complexes were immunoprecipitated (IP) from the cells and the resulting complexes were stained for TAP1, TAP2 and UL49.5. All experiments were done in triplicate.
Discussion
BoHV‐1 UL49.5 is a member of the family of varicellovirus‐encoded TAP inhibitors. So far, other TAP‐inhibiting UL49.5 proteins have been found in PRV, EHV‐1 and EHV‐4 UL49.5.
These proteins efficiently block peptide transport through TAP, thereby preventing the MHC I‐mediated presentation of viral peptides and subsequent elimination by specific CTL (Koppers‐Lalic et al., 2008). In this study, we constructed multiple UL49.5 recombinants to reveal the contribution of the different domains of the protein and of specific amino acids to UL49.5‐induced immune evasion.
UL49.5 inhibits TAP by introducing structural changes that interfere with conformational rearrangements required for peptide transport and translocation (Koppers‐Lalic et al., 2005; Koppers‐Lalic et al., 2008). To assess the contribution of the ER‐
luminal domain of BoHV‐1 UL49.5 to TAP inhibition, chimeric proteins of BoHV‐1 and VZV UL49.5 were constructed. The latter protein binds to TAP without affecting its function.
The chimera consisting of the ER‐luminal domain of BoHV‐1 UL49.5 and the TM domain of VZV UL49.5 was found to block TAP and downregulate MHC I. Fusing the N‐terminal domain of BoHV‐1 UL49.5 to the TM domain of TLR2 partially preserved the observed phenotype. A chimeric protein comprising the ER‐luminal domain of BoHV‐1 UL49.5 and the TM domain of CD3δ was non‐functional. The latter protein could only be detected after treatment with a proteasome inhibitor. In contrast, detectable steady state levels of BHVER‐ TLR2TM were found in untreated cells. The lower expression levels and the decreased stability of the BHVER‐CD3TM protein might be accountable for the lack of TAP inhibition.
The ER‐luminal domain of BoHV‐1 UL49.5 expressed as a soluble protein appeared unstable. Taken together, these data indicate that the ER‐luminal domain of BoHV‐1 UL49.5 contributes essentially to TAP inhibition, provided the presence of a TM domain to retain the membrane integration and stability of the protein. An interaction between BHVER‐ TLR2TM and TAP could not be detected, while in the same experiment UL49.5Δtail was demonstrated to bind TAP. Both proteins were expressed at comparable levels. Apparently, BHVER‐TLR2TM interacts weakly with TAP. BHVER‐VZVTM did interact with TAP and induced a stronger phenotype than BHVER‐TLR2TM. These findings suggest that the chimeric proteins have to interact with sufficient affinity to efficiently inhibit TAP; this interaction is mediated by the joint action of the ER‐luminal and TM domains of UL49.5.
A chimeric construct of BoHV‐1 UL49.5 and CD3δ was also used to analyze the contribution of the C‐terminal tail domain to TAP inhibition in the absence of the ER‐
luminal domain. The ER‐luminal of region CD3δ fused to the TM and tail domain of UL49.5 was properly expressed in MJS, but did not result in either MHC I downregulation or degradation of TAP. This is in line with a previous study where the ER‐luminal domain of BoHV‐1 UL49.5 was demonstrated to be required for TAP degradation (Loch et al., 2008).
Attempts to show binding of CD3 BHV to TAP were unsuccessful. In the study
mentioned above, an N‐terminally truncated BoHV‐1 UL49.5, deficient of the first 25 amino acids following the signal sequence, was demonstrated to interact with TAP (Loch et al., 2008). However, CD3ER‐BHVTM+tail lacks the entire N‐terminal domain. The absence of the remaining 11 N‐terminal residues within the latter construct might be accountable for the observed lack of interaction. This suggests that the amino acid residues 47 to 57 of the N‐terminal domain of UL49.5 contribute essentially to the interaction between BoHV‐1 UL49.5 and TAP. Combined with the interaction studies with BHVER‐TLR2TM, these data suggest that BoHV‐1 UL49.5 binding to TAP is mediated by the last 11 amino acids of the ER‐luminal and a yet unidentified region within the TM domain of the protein.
The TAP‐inhibiting UL49.5 proteins all code for a proline residue at position 48 that is not found in the sequence of non‐inhibiting homologs (Fig. 1). Proline residues can allow the Cα‐chain to make a sharp turn in the cis‐conformation. The predicted 3D structure of BoHV‐1 UL49.5 (Fig. 10) reveals two prominent turns: a bend in ER‐luminal domain that is facilitated by Pro48 and a C‐terminal bend facilitated by Pro87. These residues might be important for the specific folding of BoHV‐1 UL49.5 and the interaction between UL49.5 and TAP and, consequently, the inhibition of peptide transport. Nevertheless, the substitution of Pro48 by either an alanine or a glycine did not abrogate TAP inhibition, implicating that the structure facilitated by Pro48 is not required for BoHV‐1 UL49.5 function. Similarly, the turn facilitated by Pro87 is dispensable for BoHV‐1 UL49.5‐
mediated TAP inhibition.
BoHV‐1 UL49.5 facilitates TAP degradation via its C‐terminal tail domain (Koppers‐
Lalic et al., 2005). This strategy has not been found for other members of the UL49.5 TAP inhibitor family (Koppers‐Lalic et al., 2008). Therefore, degradation of TAP must be mediated by a tail‐resident motif that is unique for BoHV‐1 UL49.5. Here, we show that the two lysine and two serine residues, which all are potential ubiquitin‐anchor residues, are not required for TAP degradation.
An alanine scan of the C‐terminal RGRG sequence of BoHV‐1 UL49.5 revealed that the arginine residues at position 93 and 95 are involved in BoHV‐1 UL49.5‐induced degradation of TAP. Furthermore, the replacement of Arg95 by a lysine instead of an alanine maintained TAP degradation by BoHV‐1 UL49.5. In contrast, the substitution of Arg93 by a lysine abrogated the phenotype. Concluding, UL49.5‐initiated TAP degradation requires Arg93 and Arg95, but the latter residues can be replaced by a positively charged residue.
UL49.5Δtail lacks the capacity to degrade TAP, which is accompanied by reduced TAP inhibition that results in a less efficient downregulation of MHC I cell surface expression (Koppers‐Lalic et al., 2005 and this manuscript). Unexpectedly, the reduction of MHC I expression and TAP function induced by UL49.5 R93A and R95A closely resembles the degree of UL49.5 wt‐mediated inhibition, despite the fact that these mutants fail to degrade
TAP. This result was confirmed by the UL49.5 R93K mutant. These two independent observations suggest that degradation of TAP is not necessary for UL49.5‐induced MHC I downregulation and indicate that the cytoplasmic domain of UL49.5 must harbour a hitherto unidentified motif that contributes to MHC I downregulation. The experiments described here exclude a role for Pro87 and the RGRG domain in this effect.
The C‐terminal two arginines required for the degradation of TAP might serve as a docking motif for an E3 ligase. In this way, BoHV‐1 UL49.5 could facilitate the ubiquitinylation of TAP and, subsequently, proteasomal degradation of TAP. Alternatively, the RGRG domain might attract other components of the ER‐associated protein degradation (ERAD) machinery that mediates the retrograde transport and subsequent degradation of TAP.
The results described here demonstrate that UL49.5‐mediated TAP inhibition is accomplished by motifs within the N‐, TM and C‐terminal domains of the protein, as summarized in Fig. 10. TAP inhibition, and consequently MHC I downregulation, induced by the ER‐luminal domain essentially depend on an efficient interaction with TAP. This is facilitated by residues 47 to 57 and a motif within the TM domain. The TM domain not only anchors UL49.5 in the ER membrane, but also contributes to its stability. Inhibition and degradation of TAP by the C‐terminal domain of BoHV‐1 UL49.5 requires the presence of the ER‐luminal domain of the protein. The two C‐terminal arginine residues of the tail domain are essential for TAP degradation by UL49.5. Analysis of downregulation by
Fig. 10, the predicted 3D structure of BoHV‐1 UL49.5. Predicted tertiary structure of UL49.5 (ribbon) inserted in a lipid bilayer (sticks) as determined using the I‐TASSER server. The different domains of the UL49.5 protein together with their proposed function are indicated. The proline residues 48 and 87 are shown in ball‐and‐stick in light gray, the arginine residues 93 and 95 are shown in ball‐and‐stick in dark gray.
recombinant forms of UL49.5 revealed that degradation of TAP does not contribute additionally to MHC I downregulation. Apparently, UL49.5 affects MHC I expression by inhibiting TAP through an additional sequence motif located in the cytoplasmic domain of the protein.
Acknowledgements
We thank Sjoerd van den Worm for the construction of mutant UL49.5 P27A and P27G and assisting with cloning, Aleksandra Kwasnik and Marta Matlacz for helpful technical assistance, and Guido de Roo and Menno van der Hoorn at the Flowcytometry Research Unit (LUMC) for their technical support.
Reference List
Ahn,K., Gruhler,A., Galocha,B., Jones,T.R., Wiertz,E.J., Ploegh,H.L., Peterson,P.A., Yang,Y., Fruh,K., 1997. The ER‐luminal domain of the HCMV glycoprotein US6 inhibits peptide translocation by TAP.
Immunity 6, 613‐621.
Ahn,K., Meyer,T.H., Uebel,S., Sempe,P., Djaballah,H., Yang,Y., Peterson,P.A., Fruh,K., Tampe,R., 1996.
Molecular mechanism and species specificity of TAP inhibition by herpes simplex virus ICP47.
EMBO J. 15, 3247‐3255.
Aisenbrey,C., Sizun,C., Koch,J., Herget,M., Abele,R., Bechinger,B., Tampe,R., 2006. Structure and dynamics of membrane‐associated ICP47, a viral inhibitor of the MHC I antigen‐processing machinery. J. Biol. Chem. 281, 30365‐30372.
Cadwell,K., Coscoy,L., 2005. Ubiquitination on nonlysine residues by a viral E3 ubiquitin ligase.
Science 309, 127‐130.
Cresswell,P., Ackerman,A.L., Giodini,A., Peaper,D.R., Wearsch,P.A., 2005. Mechanisms of MHC class I‐
restricted antigen processing and cross‐presentation. Immunol. Rev. 207, 145‐157.
Davison,A.J., Eberle,R., Ehlers,B., Hayward,G.S., McGeoch,D.J., Minson,A.C., Pellett,P.E., Roizman,B., Studdert,M.J., Thiry,E., 2009. The order Herpesvirales. Arch. Virol. 154, 171‐177.
Fruh,K., Ahn,K., Djaballah,H., Sempe,P., van Endert,P.M., Tampe,R., Peterson,P.A., Yang,Y., 1995. A viral inhibitor of peptide transporters for antigen presentation. Nature 375, 415‐418.
Griffin,B.D., Verweij,M.C., Wiertz,E.J., 2010. Herpesviruses and immunity: The art of evasion. Vet.
Microbiol.
Halenius,A., Momburg,F., Reinhard,H., Bauer,D., Lobigs,M., Hengel,H., 2006. Physical and functional interactions of the cytomegalovirus US6 glycoprotein with the transporter associated with antigen processing. J. Biol. Chem. 281, 5383‐5390.
Hansen,T.H., Bouvier,M., 2009. MHC class I antigen presentation: learning from viral evasion strategies. Nat. Rev. Immunol. 9, 503‐513.
Hengel,H., Koopmann,J.O., Flohr,T., Muranyi,W., Goulmy,E., Hammerling,G.J., Koszinowski,U.H., Momburg,F., 1997. A viral ER‐resident glycoprotein inactivates the MHC‐encoded peptide transporter. Immunity 6, 623‐632.
Hewitt,E.W., Gupta,S.S., Lehner,P.J., 2001. The human cytomegalovirus gene product US6 inhibits ATP binding by TAP. EMBO J. 20, 387‐396.
Hill,A., Jugovic,P., York,I., Russ,G., Bennink,J., Yewdell,J., Ploegh,H., Johnson,D., 1995. Herpes simplex virus turns off the TAP to evade host immunity. Nature 375, 411‐415.
Hislop,A.D., Ressing,M.E., van Leeuwen,D., Pudney,V.A., Horst,D., Koppers‐Lalic,D., Croft,N.P., Neefjes,J.J., Rickinson,A.B., Wiertz,E.J., 2007. A CD8+ T cell immune evasion protein specific to Epstein‐Barr virus and its close relatives in Old World primates. J. Exp. Med. 204, 1863‐1873.
Horst,D., van Leeuwen,D., Croft,N.P., Garstka,M.A., Hislop,A.D., Kremmer,E., Rickinson,A.B., Wiertz,E.J., Ressing,M.E., 2009. Specific targeting of the EBV lytic phase protein BNLF2a to the transporter associated with antigen processing results in impairment of HLA class I‐restricted antigen presentation. J. Immunol. 182, 2313‐2324.
