K3 (ORF K3; also IE1B, or MIR1, modulator of immune recognition 1) is a 333-residue Kaposi's sarcoma-associated herpesvirus membrane protein and, with its ~40% identical paralog K5/MIR2, the prototype of the MARCH family of membrane-bound RING-CH E3 ubiquitin ligases. Its architecture is a cytosolic N-terminal RING-CH (RINGv) zinc finger that binds two zinc ions in a cross-brace motif and docks the E2 ubiquitin-conjugating enzymes UbcH5A-C and UbcH13, two transmembrane helices that recognise targets through their own transmembrane segments, and a long cytosolic C-terminal tail carrying a tyrosine-based endocytosis motif and acidic tracts that direct trafficking and degradation of the modified substrate. The protein is made and resides largely in the host endoplasmic reticulum and pre-Golgi compartment and acts on its targets at or near the host plasma membrane. K3 polyubiquitinates the cytosolic tails of host immunoreceptors, marking them for accelerated endocytosis and delivery to the endolysosomal or proteasomal degradation route: MHC class I (all of HLA-A, -B, -C and -E), the gamma interferon receptor IFNGR1, CD1D, and the C-type lectins DC-SIGN (CD209) and DC-SIGNR (CLEC4M). Uniquely among characterised E3s of its era, it can transfer ubiquitin to cysteine as well as lysine residues of the substrate tail, so MHC class I molecules lacking cytoplasmic lysines are still downregulated. The net effect is multi-pronged immune evasion: loss of surface MHC class I blinds cytotoxic T lymphocytes to the infected cell, and loss of IFNGR1 blunts IFN-gamma-driven STAT1 phosphorylation. In the intact virus K3 and K5 act sequentially rather than redundantly, K5 immediately on entry into the lytic phase and K3 during later stages of lytic replication.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0005515 protein binding | IPI PMID:16339411 Herpesviral protein networks and their interaction with the ... | REMOVE | Summary: Intraviral yeast two-hybrid hit from the Uetz/Haas herpesviral interactome screen; P88941 is an uncharacterised 110-residue HHV-8 type M product with no functional follow-up and no evidence of being a K3 substrate. Reason: Bare GO:0005515 protein binding conveys no functional information and is covered by the protein-binding policy. The cached record for PMID:16339411 is abstract-only and reports only the network-level outcome of a high-throughput screen, so no more informative molecular function can be inferred for this pair, and K3's informative molecular function (GO:0061630 ubiquitin protein ligase activity) is already asserted elsewhere in this review from direct evidence. Removal of the generic term does not assert that the reported interaction is false. Supporting Evidence: PMID:16339411 The comprehensive yeast two-hybrid analysis of intraviral protein interactions |
| GO:0005515 protein binding | IPI PMID:16339411 Herpesviral protein networks and their interaction with the ... | REMOVE | Summary: Second intraviral yeast two-hybrid hit from the same screen; P88951 is the HHV-8 ribonucleoside-diphosphate reductase small subunit (RIR2), again with no functional characterisation of the pair. Reason: Same reasoning as the P88941 row: bare protein binding is uninformative and the abstract-only high-throughput screen supports no specific molecular function for K3 with this partner. A viral ribonucleotide reductase subunit is not a membrane immunoreceptor and is not among K3's documented ubiquitination substrates, so no substrate-based replacement term is warranted either. Removal does not assert the interaction is false. Supporting Evidence: PMID:16339411 revealed 123 and 173 interactions |
| GO:0006511 ubiquitin-dependent protein catabolic process | IDA PMID:23460925 Kaposi's sarcoma-associated herpesvirus K3 and K5 proteins d... | ACCEPT | Summary: CACAO row. Lang et al. show directly that K3 ubiquitylates DC-SIGN and drives its ubiquitylation-dependent degradation, which is exactly what this term describes. Reason: CACAO, assigned 2013-03-26, IDA from PMID:23460925. What the paper shows for K3: Fig. 7A, an HA-ubiquitin pull-down in which DC-SIGN is ubiquitylated only with RING-CH-intact K3 and not with the K3 mZn mutant; Fig. 7B, the same assay under MG132 and chloroquine, concluding that degradation runs primarily through a ubiquitylation-dependent proteasomal route with a lysosomal fraction; and Fig. 5A, a western blot showing that wild-type but not mZn K3 lowers total DC-SIGN and DC-SIGNR protein. The CACAO term matches that assay exactly, at the right granularity: K3's substrates are routed both to the proteasome and, for MHC class I, to the late endosome/lysosome (PMID:12006494, PMID:10859362), so the parent term is preferable to the proteasome-specific child GO:0043161. K3 performs the step that commits the substrate to this catabolic route (it is the E3 that attaches the degron), which is the participation GO intends for an E3 ligase here. This is independently corroborated for MHC class I by PMID:11756476 and PMID:12006494. Right protein, right assay, right term, correct evidence code. Supporting Evidence: PMID:23460925 We confirmed that DC-SIGN displayed ubiquitylation only in the presence of K3 or K5 with an intact RING-CH domain and not in the presence K3 mZn or K5 mZn. PMID:23460925 These results indicate that degradation of DC-SIGN occurs primarily through a ubiquitylation-dependent, proteasomal mechanism PMID:23460925 expression of either wild-type K3 or K5 resulted in decreased protein levels of both DC-SIGN and DC-SIGNR |
| GO:0006511 ubiquitin-dependent protein catabolic process | IEA GO_REF:0000117 | ACCEPT | Summary: ARBA rule reaching the same conclusion as the CACAO IDA row above, and independently grounded by the MHC class I literature. Reason: This row carries the same claim as the IDA row, so the two reinforce each other and take the same action. The claim does not rest on the rule: K3 ubiquitylates MHC class I after its export from the ER and the mark routes class I into the late endocytic pathway for degradation (PMID:12006494), and MIR1-dependent ubiquitination of endogenous MHC class I chains was shown directly (PMID:11756476). Supporting Evidence: PMID:12006494 promotes ubiquitylation of class I after export from the endoplasmic reticulum. PMID:11756476 cells expressing either MIR1 or MIR2 display ubiquitination of their endogenous MHC-I chains |
| GO:0008270 zinc ion binding | IEA GO_REF:0000120 | ACCEPT | Summary: The RING-CH domain binds two zinc ions in a cross-brace motif; this was established structurally by NMR, not merely predicted from the signature. Reason: The InterPro RING-CH signature match is good support for membership, and the claim is confirmed directly by the solution structure of the K3 N-terminal domain (PDB 1VYX, PMID:15465811), which showed it to be a variant RING rather than a plant homeodomain and retaining the canonical two-zinc cross-brace. UniProt annotates the eight coordinating residues (Cys9, Cys12, Cys24, His26, Cys34, Cys37, Cys50, Cys53). The zinc sites are not incidental: disrupting them (the mZn mutants) abolishes ubiquitination and substrate downregulation in every assay reported for K3, so this is a genuine, functionally load-bearing molecular function of the protein. Supporting Evidence: PMID:15465811 binding two zinc ions in a stable cross-brace motif PMID:15465811 it is conclusively demonstrated that the K3 N-terminal domain is a variant file:HHV8P/K3/K3-uniprot.txt ZN_FING 1..60 |
| GO:0016567 protein ubiquitination | IEA GO_REF:0000041 | ACCEPT | Summary: K3 ubiquitinates host membrane proteins; this is the central process the protein executes. Reason: The UniPathway mapping lands on the right process and is directly supported: MIR1/K3 expression produces ubiquitinated endogenous MHC class I chains (PMID:11756476), K3 ubiquitylates class I after ER export (PMID:12006494), K3 ubiquitylates DC-SIGN in a RING-CH-dependent manner (PMID:23460925), and K3 ubiquitinates IFNGR1 (PMID:17166914). The term is as precise as GO offers for the generic process; the interspecies framing of the same activity is carried by GO:0039648 elsewhere in this review. Supporting Evidence: PMID:11756476 cells expressing either MIR1 or MIR2 display ubiquitination of their endogenous MHC-I chains PMID:23460925 We confirmed that DC-SIGN displayed ubiquitylation only in the presence of K3 or K5 with an intact RING-CH domain and not in the presence K3 mZn or K5 mZn. |
| GO:0016740 transferase activity | IEA GO_REF:0000043 | TOO GENERAL | Summary: True but uninformative: K3's transferase activity is specifically ubiquitin transfer, asserted at the right level by GO:0061630 in this same review. Reason: GO:0061630 ubiquitin protein ligase activity is accepted elsewhere in this review on direct evidence (RING-CH-dependent ubiquitylation of MHC class I, DC-SIGN and IFNGR1), and it entails GO:0016740. The keyword mapping is not wrong - it judges the term, not the evidence - but it adds nothing once the specific activity is asserted. Supporting Evidence: PMID:11756476 cells expressing either MIR1 or MIR2 display ubiquitination of their endogenous MHC-I chains |
| GO:0020002 host cell plasma membrane | IEA GO_REF:0000044 | ACCEPT | Summary: The UniProt subcellular-location mapping rests on an experimental localisation (PubMed:10799607) that placed K3 near the host plasma membrane, which is also where its substrates are internalised. Reason: UniProt's SL-0375 assignment carries ECO:0000269|PubMed:10799607, i.e. it is a curated statement from confocal immunofluorescence on transfected COS-1 and stably expressing BJAB cells, not a prediction. This is the functionally relevant pool: K3 ubiquitylates MHC class I after its export from the ER and the mark triggers internalisation at the plasma membrane (PMID:12006494). It is not redundant with the host-ER row - the two locations reflect the steady-state bulk (ER) and the site of action (plasma membrane), and both are curated from experiment. Supporting Evidence: PMID:10799607 cross sections of transfected cells demonstrated that K3 and K5 were localized near the plasma membrane PMID:12006494 signal for class I internalization at the plasma membrane |
| GO:0039502 symbiont-mediated suppression of host type I interferon-mediated signaling pathway | IEA GO_REF:0000043 | MODIFY | Summary: The interferon pathway K3 is actually shown to block is type II (IFN-gamma), via degradation of IFNGR1; no evidence places K3 on the type I (IFNAR) pathway. Reason: KW-1114 ("Inhibition of host interferon signaling pathway by virus") is type-agnostic, but its GO mapping lands on the type I child. The only interferon target documented for K3 is IFNGR1: Li et al. showed that K3 and K5 induce IFNGR1 ubiquitination, endocytosis and degradation and that IFN-gamma-driven STAT1 Tyr701 phosphorylation is markedly reduced in K3-expressing BJAB cells (PMID:17166914). That is the type II pathway, and GO:0140884 is the sibling term for it. This is a lateral correction within the interferon branch rather than a refinement, and it is the same replacement proposed for the GO:0140886 row below so that the two keyword-derived interferon rows agree. Nothing in the literature reviewed here tests K3 against IFN-alpha/beta receptor signalling, so asserting the type I term overstates the evidence. Proposed replacements: symbiont-mediated suppression of host type II interferon-mediated signaling pathway Supporting Evidence: PMID:17166914 K3 and K5 both specifically target IFN-Ξ³R1 and induce its ubiquitination, endocytosis, and degradation, resulting in downregulation of IFN-Ξ³R1 surface expression PMID:17166914 tyrosine-phosphorylated STAT1 (pY701-STAT1) was markedly reduced in BJAB-K3 and BJAB-K5 cells |
| GO:0039504 symbiont-mediated suppression of host adaptive immune response | IEA GO_REF:0000117 | MODIFY | Summary: Correct, but the specific mechanism - removal of surface MHC class I - has its own term, GO:0046776, which the evidence supports. Reason: The adaptive-immune suppression K3 effects is antigen presentation via MHC class I: K3 downregulates all four HLA allotypes tested (PMID:10799607), ubiquitylates class I and signals its internalisation and TSG101-dependent degradation (PMID:12006494), and does so during lytic replication in the intact virus (PMID:24899205). GO:0046776 is an is_a descendant of GO:0039588, itself a child of GO:0039504, so this is a refinement to the term the evidence actually supports. The same replacement is proposed for the CACAO IMP row on this term so that the two rows agree. Proposed replacements: symbiont-mediated suppression of host antigen processing and presentation of peptide antigen via MHC class I Supporting Evidence: PMID:10799607 K3 downregulated all four HLA allotypes PMID:12006494 promotes ubiquitylation of class I after export from the endoplasmic reticulum. |
| GO:0039504 symbiont-mediated suppression of host adaptive immune response | IMP PMID:23460925 Kaposi's sarcoma-associated herpesvirus K3 and K5 proteins d... | MODIFY | Summary: CACAO row. The cited paper does assay K3's effect on host antigen presentation, but the readout is MHC class I surface loss, for which GO has the specific child term GO:0046776. Reason: CACAO, assigned 2013-03-26, IMP from PMID:23460925. What the paper shows for K3: in Fig. 2 and Fig. 3, 293 cells transfected with GFP-K3 wild-type and with the mZn, Y/A, Y/F, P/A and DE12 mutants were stained for endogenous MHC class I alongside the lectins, and wild-type K3 reduced surface MHC class I by about 80-90% while the RING-CH mutant reduced it by only 20-30% - a mutant-based experiment, so IMP is the right code. The CACAO term is therefore true of K3 and the paper is the right paper, but it is one level too shallow: the specific process assayed is the suppression of antigen processing and presentation via MHC class I (GO:0046776), an is_a descendant via GO:0039588. The paper's other K3 targets, the C-type lectins DC-SIGN and DC-SIGNR, are innate pattern receptors rather than adaptive effectors, so they do not themselves sustain the adaptive-immunity framing. Note the scope caveat the authors themselves record: in the recombinant-virus arm K5 is dominant and K3 acts only weakly, so the K3 lectin phenotype is an exogenous-expression result. MHC class I downregulation by K3, by contrast, is corroborated in the intact virus (PMID:24899205). Proposed replacements: symbiont-mediated suppression of host antigen processing and presentation of peptide antigen via MHC class I Supporting Evidence: PMID:23460925 In the presence of wild-type K3 or K5, surface levels of MHC I were reduced about 80β90%. PMID:24899205 while K5 acts immediately upon entry into the lytic phase, K3-mediated downregulation of MHC-I was evident during later stages of lytic replication |
| GO:0039648 symbiont-mediated perturbation of host ubiquitin-like protein modification | IEA GO_REF:0000117 | ACCEPT | Summary: Same claim as the CACAO IMP row on this term: a viral E3 that ubiquitinates host proteins alters host ubiquitin modification by definition. Reason: The ARBA rule is not the basis of the claim. K3 is a virus-encoded E3 whose substrates are host membrane proteins - endogenous MHC class I (PMID:11756476), class I after ER export (PMID:12006494), DC-SIGN (PMID:23460925) and IFNGR1 (PMID:17166914) - which is precisely what GO:0039648 describes, namely a symbiont altering ubiquitin-like protein modification of host target proteins. The action matches the CACAO row on the same term. Supporting Evidence: PMID:11756476 cells expressing either MIR1 or MIR2 display ubiquitination of their endogenous MHC-I chains PMID:17166914 K3 and K5 both specifically target IFN-Ξ³R1 and induce its ubiquitination, endocytosis, and degradation, resulting in downregulation of IFN-Ξ³R1 surface expression |
| GO:0039648 symbiont-mediated perturbation of host ubiquitin-like protein modification | IMP PMID:23460925 Kaposi's sarcoma-associated herpesvirus K3 and K5 proteins d... | ACCEPT | Summary: CACAO row, well supported: the paper shows RING-CH-dependent ubiquitylation of the host lectins DC-SIGN and DC-SIGNR by K3. Reason: CACAO, assigned 2013-03-09, IMP from PMID:23460925. What the paper shows for K3: Fig. 7A is an HA-ubiquitin co-transfection in which DC-SIGN is ubiquitylated with wild-type K3 but not with the RING-CH mutant K3 mZn, and DC-SIGNR shows the same RING-CH dependence at a lower level; Fig. 2 and Fig. 3 extend the mutant series (mZn, Y/A, Y/F, P/A, DE12), which is what makes IMP rather than IDA the fitting code. The host lectins are host proteins and the modification is ubiquitin, so the CACAO term is an exact match for the assay at the right granularity. Right protein, right assay, right term, correct evidence code. The one caveat, which the authors state, is that this activity was demonstrated by exogenous expression; in the context of infection K5 dominates. Supporting Evidence: PMID:23460925 We confirmed that DC-SIGN displayed ubiquitylation only in the presence of K3 or K5 with an intact RING-CH domain and not in the presence K3 mZn or K5 mZn. PMID:23460925 Pull-down of GST-tagged proteins followed by WB showed that both DC-SIGN and DC-SIGNR co-precipitated with both K3 and K5 |
| GO:0044165 host cell endoplasmic reticulum | IEA GO_REF:0000044 | ACCEPT | Summary: The steady-state bulk of K3 is in the host ER, shown by colocalisation with KDEL-GFP and calnexin, and the pre-Golgi pool is where K3 first engages at least some substrates. Reason: UniProt's SL-0388 assignment carries ECO:0000269|PubMed:10859362, a curated experimental statement rather than a prediction: Coscoy and Ganem found K3 colocalised with an ER-targeted GFP and with endogenous calnexin, and not with markers of later compartments. The 2013 lectin paper independently supports a pre-Golgi site of first engagement, since K3 preferentially pulls down the EndoH-sensitive immature forms of DC-SIGN and DC-SIGNR (PMID:23460925). This is kept alongside, not instead of, the host plasma membrane row: K3 resides mainly in the ER and acts at the cell surface. Supporting Evidence: PMID:10859362 Thus, at steady state the bulk of the K3 and K5 chains are localized within the ER PMID:23460925 the initial interactions between the viral proteins and DC-SIGN or DC-SIGNR is likely to occur in a pre-Golgi compartment with immature forms of these two lectins |
| GO:0046872 metal ion binding | IEA GO_REF:0000043 | TOO GENERAL | Summary: True but uninformative: the metal is zinc, and GO:0008270 zinc ion binding is accepted in this review on structural evidence. Reason: GO:0008270 is an is_a descendant of GO:0046872 and is asserted elsewhere in this review on the NMR structure of the K3 RING-CH domain (PMID:15465811) and the eight curated Zn-coordinating residues, so the broader keyword term adds nothing. This judges the term, not the keyword mapping, which is correct as far as it goes. Supporting Evidence: PMID:15465811 binding two zinc ions in a stable cross-brace motif |
| GO:0052031 symbiont-mediated perturbation of host defense response | IEA GO_REF:0000043 | TOO GENERAL | Summary: Correct, but it is the common ancestor of the specific immune-evasion terms already kept in this review. Reason: GO:0052031 is an ancestor of GO:0039504 (whose MODIFY replacement GO:0046776 is asserted here) and of GO:0052170, kept as non-core, so it is entailed by claims this review already makes at a more informative level. The keyword pair KW-0899/KW-0922 is accurate about K3; the term is simply uninformative once the MHC class I and innate arms are stated. Supporting Evidence: PMID:10799607 K3 downregulated all four HLA allotypes |
| GO:0052170 symbiont-mediated suppression of host innate immune response | IEA GO_REF:0000043 | KEEP AS NON CORE | Summary: Supported - K3 destroys IFNGR1 and the innate C-type lectins DC-SIGN and DC-SIGNR - but peripheral to the MHC class I activity that defines the protein. Reason: The claim holds on two independent lines: K3 induces IFNGR1 ubiquitination, endocytosis and degradation with a measurable loss of IFN-gamma-driven STAT1 phosphorylation (PMID:17166914), and it ubiquitylates and degrades the dendritic-cell lectins DC-SIGN and DC-SIGNR (PMID:23460925). The term is as precise as the keyword allows and is not entailed by the interferon-signalling terms, which sit under signal transduction rather than under innate immune response, so it is retained rather than called too general. It is non-core because K3's dominant, best-replicated activity in the intact virus is MHC class I removal (PMID:24899205), and because the lectin arm is an exogenous-expression result that K5 dominates during infection. Supporting Evidence: PMID:17166914 tyrosine-phosphorylated STAT1 (pY701-STAT1) was markedly reduced in BJAB-K3 and BJAB-K5 cells PMID:23460925 expression of either wild-type K3 or K5 resulted in decreased protein levels of both DC-SIGN and DC-SIGNR |
| GO:0061630 ubiquitin protein ligase activity | IEA GO_REF:0000003 | ACCEPT | Summary: The defining molecular function of K3, established by structure, by in vitro E2 docking, and by RING-CH-dependent substrate ubiquitylation in cells. Reason: The EC mapping is grounded in curated experimental work rather than inference: the K3 N-terminal domain is a variant RING that docks UbcH5A-C and UbcH13 (PMID:15465811), MIR1 expression ubiquitinates endogenous MHC class I (PMID:11756476), and every substrate effect reported for K3 is abolished by mutating the RING-CH zinc ligands (PMID:23460925). The distinguishing feature behind EC 2.3.2.36 - transfer of ubiquitin to a cysteine rather than a lysine of the substrate - is itself experimental (PMID:15994556). This is the core molecular function. Supporting Evidence: PMID:15465811 found to interact with the cellular ubiquitin-conjugating enzymes UbcH5A to -C PMID:15994556 ubiquitination required a cysteine residue in the intracytoplasmic tail of MHC I PMID:23460925 Mutation of the RING-CH domain abolished this effect |
| GO:0075509 endocytosis involved in viral entry into host cell | IEA GO_REF:0000117 | MODIFY | Summary: The endocytosis K3 drives is of host surface receptors, not of the virion; K3 is expressed after entry and cannot take part in uptake of the particle that delivered it. Reason: GO:0075509 is defined as any endocytosis involved in the uptake of a virus into a host cell, i.e. the virion is the cargo. K3 has no reported role in virion uptake; it is an immediate-early to late lytic product acting inside an already infected cell, and the endocytosis it enhances is that of host immunoreceptors - MHC class I (PMID:10799607, PMID:10859362) and DC-SIGN (PMID:23460925). The sound essence of the row is therefore symbiont-driven enhancement of host endocytosis, for which GO:0044077, a process in which a symbiont initiates, promotes or enhances host receptor-mediated endocytosis, is the appropriate interspecies term. This is a lateral correction, and the same replacement is proposed for the CACAO IMP row on this term so the two agree. Proposed replacements: symbiont-mediated perturbation of host receptor-mediated endocytosis Supporting Evidence: PMID:10859362 they act to enhance endocytosis of MHC I from the cell surface; internalized class I chains are delivered to endolysosomal vesicles, where they undergo degradation PMID:10799607 K3-expressing cells exhibited a rapid endocytosis of class I molecules |
| GO:0075509 endocytosis involved in viral entry into host cell | IMP PMID:23460925 Kaposi's sarcoma-associated herpesvirus K3 and K5 proteins d... | MODIFY | Summary: CACAO row. The paper does show K3-driven endocytosis, but of the host lectin DC-SIGN; the entry-enhancing property belongs to the lectin, and K3 removes it. The actor and the direction are both wrong for this term. Reason: CACAO, assigned 2013-03-26, IMP from PMID:23460925. What the paper shows for K3: in the dynasore washout assays of Fig. 4B (THP-1 lines) and Fig. 4C (293 lines), wild-type K3 accelerates endocytosis of DC-SIGN while the Y/A endocytosis-motif mutant does not, and no endocytosis of DC-SIGNR is seen - a mutant comparison, so IMP fits. Separately, Fig. 1 shows that DC-SIGN and DC-SIGNR enhance KSHV infectivity, but that experiment uses Bac16 virus deleted for both K3 and K5, so the entry enhancement is a property of the host lectins measured in the complete absence of K3. The CACAO term fuses the two results and attributes the host receptor's entry-enhancing role to K3, which inverts the biology: K3 destroys the co-receptor, and it is expressed only after entry has occurred. Per the curator-deference rule this is not removed - the underlying observation (K3 enhances endocytosis of a host surface receptor) is real and the paper is the right paper - so the row is redirected to the interspecies endocytosis term GO:0044077, which describes what was actually measured. Proposed replacements: symbiont-mediated perturbation of host receptor-mediated endocytosis Supporting Evidence: PMID:23460925 K3 expression also caused increased endocytosis of DC-SIGN relative to vector expressing cells PMID:23460925 It has been documented that DC-SIGN can act a co-receptor for KSHV, not sufficient for entry by itself, but capable of enhancing viral infectivity |
| GO:0140886 symbiont-mediated suppression of host interferon-mediated signaling pathway | IEA GO_REF:0000043 | MODIFY | Summary: Correct at the type-agnostic level, but the evidence identifies the pathway as type II (IFN-gamma), via IFNGR1 degradation. Reason: KW-1091 cannot distinguish interferon types, so the mapping stops at the parent. The evidence does not: K3 and K5 target IFNGR1 for ubiquitination, endocytosis and degradation, and IFN-gamma-induced STAT1 Tyr701 phosphorylation is markedly reduced in K3-expressing cells (PMID:17166914). GO:0140884 is the is_a child for type II interferon signalling and is what the assays support; the same replacement is proposed for the GO:0039502 row, which the keyword mapping sent to the type I child instead. Proposed replacements: symbiont-mediated suppression of host type II interferon-mediated signaling pathway Supporting Evidence: PMID:17166914 K3 and K5 both specifically target IFN-Ξ³R1 and induce its ubiquitination, endocytosis, and degradation, resulting in downregulation of IFN-Ξ³R1 surface expression PMID:17166914 tyrosine-phosphorylated STAT1 (pY701-STAT1) was markedly reduced in BJAB-K3 and BJAB-K5 cells |
Loading supporting contentβ¦
Download this section (compressed HTML)Q: Does K3 contribute measurably to DC-SIGN and DC-SIGNR downregulation in a natural infection, or only on exogenous overexpression? The recombinant-virus arm of PMID:23460925 shows K5 doing essentially all the work, and the answer determines whether the lectin arm belongs in K3's core function at all.
Q: Why does K3 drive endocytosis of DC-SIGN but not of DC-SIGNR, when it ubiquitylates and degrades both? Distinguishing a different ubiquitin-chain linkage from a different trafficking itinerary would say whether K3 writes one signal that the two lectins read differently.
Q: Given that the RING-CH domain docks UbcH13, does K3 build Lys63-linked chains on its substrates in cells, and is the endosomal-sorting fate of MHC class I a consequence of chain type rather than of substrate identity?
Q: Is K3 ever present at the plasma membrane in sufficient amount to act there, or does it modify substrates during a transient surface encounter? Its steady-state bulk is in the ER, it does not colocalise with internalised MHC class I, and it engages the lectins as immature, pre-Golgi forms.
Q: Does cysteine ubiquitination contribute to K3's activity against substrates other than MHC class I - for example IFNGR1 or the C-type lectins - or is the non-lysine route specific to the class I tail?
Experiment: Complement the Bac16 double-deletion recombinant with K3 alone at the K5 genomic locus (the construct already described in PMID:24899205, which raises K3 expression) and measure surface DC-SIGN and DC-SIGNR in infected THP-1 or primary monocyte-derived dendritic cells, to establish whether the K3 lectin activity is real at native expression levels or an artefact of overexpression.
Experiment: Ubiquitin-linkage-specific immunoblotting (K48 and K63 antibodies) plus mass-spectrometric chain profiling of MHC class I, IFNGR1 and DC-SIGN immunoprecipitated from K3-expressing versus K3 mZn cells, together with reconstituted in vitro assays using UbcH5 or UbcH13/Uev1a, to test whether K3 writes distinct chain types on different substrates.
Experiment: Lysine-free and cysteine-free cytoplasmic-tail variants of IFNGR1 and DC-SIGN, assayed for K3-dependent ubiquitination and surface loss under beta-mercaptoethanol-sensitive conditions, to test how far the non-lysine ubiquitination mechanism extends beyond MHC class I.
Experiment: Bimolecular fluorescence complementation or a surface-restricted proximity labelling (split-APEX at the plasma membrane) between K3 and MHC class I in live cells, to determine whether the ubiquitin mark is written at the surface or in a pre-Golgi compartment, which the current steady-state localisations leave unresolved.
Experiment: A cytotoxic T lymphocyte killing assay against KSHV-infected cells carrying wild-type, K3-deleted, K5-deleted and doubly deleted genomes at defined lytic stages, to quantify how much protection K3's late-stage MHC class I downregulation actually buys the infected cell.
Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)