Mutations in SAMHD1 cause Aicardi–Goutières syndrome (AGS), a Mendelian inflammatory disease which displays remarkable clinical and biochemical overlap with congenital viral infection. SAMHD1 (SAM domain and HD domain-containing protein 1) has also been defined as an HIV-1 restriction-factor that, through a novel triphosphohydrolase activity, inhibits early stage HIV-1 replication in myeloid-derived dendritic cells (MDDCs), macrophages and resting CD4+ T-cells. The potent activity of SAMHD1 is likely to be the subject of a variety of regulatory mechanisms. Knowledge of proteins that interact with SAMHD1 may not only enhance our understanding of the pathogenesis of AGS, but may also provide further details on the link between the regulation of cellular dNTPs and HIV-1 restriction. In the present study, we used a yeast two-hybrid screen and pull-down analysis followed by MS to identify the eukaryotic elongation factor 1A1 (eEF1A1) as a potential interaction partner of SAMHD1. This interaction was confirmed by unbiased co-immunoprecipitation and demonstrated in situ by a proximity ligation assay (PLA). We show that this interaction is enhanced in mutant SAMHD1 cell lines and suggest that eEF1A1 may mediate SAMHD1 turnover by targeting it to the proteosome for degradation through association with Cullin4A and Rbx1.
SAMHD1 was originally identified as an orthologue of the murine interferon (IFN)-γ- induced gene Mg11 and is highly expressed in cells of myeloid lineage . SAMHD1 (SAM domain and HD domain-containing protein 1) is one of the seven genes associated with the human disease Aicardi–Goutières syndrome (AGS), a severe genetic infantile encephalopathy that displays remarkable clinical and biochemical overlap with congenital viral infection and, in some cases, with systemic lupus erythematosus [2–5]. The pathogenesis of AGS is thought to result from defects in nucleic acid metabolism since all seven genes associated with AGS so far, TREX1 , RNASEH2A, RNASEH2B, RNASEH2C , SAMHD1  ADAR1  and, recently IFIH1/MDA5 , encode proteins with enzymatic activities involved in the sensing processing of RNA and DNA.
In 2009, mutations in SAMHD1 were detected in 10% of AGS patients . The SAMHD1 gene encodes a 626-amino-acid multi-domain protein that exists as an apo form monomer–dimer or, when activated with nucleotides, a tetramer [8,11]. The protein contains both a sterile α-motif (SAM) and an HD domain . The HD domain is found in proteins exhibiting phosphohydrolase activity  and, in 2011, we demonstrated that SAMHD1 is a dGTP-stimulated deoxynucleoside triphosphate triphosphohydrolase that converts dNTPs (dATP, TTP, dCTP and dGTP) into the constituent nucleoside and inorganic triphosphate . Although such triphosphohydrolase activity has been partially characterized in bacteria , this was the first reported instance of such an activity in eukaryotes.
Currently, two proteins are known to interact with SAMHD1, the HIV-2 accessory protein Vpx  and cyclin A2/cyclin-dependent kinase 1 (cdk1) [15,16]. The Vpx protein is found packaged with the HIV-2 virus and increases the permissiveness of cells to HIV-2 infection. This is achieved through an interaction with SAMHD1, which it targets for degradation via the Cullin4A-RING E3 ubiquitin ligase 4 (CRL4)–DDB1-and-Cullin4-associated factor 1 (DCAF1) E3 ubiquitin ligase complex [14,17]. SAMHD1-mediated restriction can be partially rescued in resting CD4+ T-lymphocytes by restoration of dNTP levels , suggesting that the role of SAMHD1 in nucleotide metabolism is key to viral restriction. Knockdown of SAMHD1 expression led to both an increase in HIV-1 infection, as judged by proviral gene expression, and the accumulation of reverse transcription products, suggesting that SAMHD1 acts to prevent HIV-1 infection at an early stage . These findings indicate that SAMHD1 functions to inhibit viral infection by reducing the cellular dNTP pool to a level at which viral reverse transcription is no longer supported [11,19]. Subsequent studies have shown that SAMHD1 can modulate permissiveness to infection by a number of lentiviruses , suggesting that this function is not specific to HIV-1. The restriction activity of SAMHD1 is modulated by phosphorylation. Cyclin A2/cdk1 binds to the C-terminus of SAMHD1 and phosphorylates Thr592, negatively regulating its activity towards HIV-1 [15,16].
In the present study we have used a variety of complementary techniques to identify the eukaryotic elongation factor 1A (eEF1A1) in complex with SAMHD1. Using recombinant SAMHD1 deletion products we demonstrate that eEF1A1 preferentially binds to a region containing the HD domain and that the N-terminal SAM domain negatively regulates protein binding. A proximity ligation assay (PLA)  demonstrated that SAMHD1 and eEF1A1 interact in situ in the cytoplasm and that this interaction is increased in patient fibroblast cells carrying mutant SAMHD1. We suggest that one outcome of the association of eEF1A1 is the targeted degradation of SAMHD1. In support of this possibility, we also identify Cullin4A and Rbx1 in complex with both SAMHD1 and eEF1A1.
Yeast two-hybrid screening
Yeast two-hybrid screening was performed by Hybrigenics Services, (http://www.hybrigenics-services.com). The coding sequence for full-length SAMHD1 (GenBank® accession number gi: 38016913) was PCR-amplified and cloned into pB27 as a C-terminal fusion to LexA (N-LexA–SAMHD1-C) and into pB35 as a C-terminal fusion to Gal4 DNA-binding domain (N-Gal4–SAMHD1-C). Constructs were verified by DNA sequencing and used as a bait to screen a random-primed human leucocyte and an activated mononuclear cell cDNA library constructed in pP6. The plasmids pB27 and pP6 were derived from pBTM116  and pGADGH  respectively. pB35 was constructed by inserting the Gal4 DNA-binding domain from pAS2ΔΔ  into the pFL39 backbone  under control of the MET25 promoter .
For the LexA bait construct, 50 million clones (5-fold the complexity of the library) were screened using a mating approach with YHGX13 (Tyr187 Ade2–101:loxP-kanMX-loxP, MATα) and L40ΔGal4 (MATa) yeast strains as described previously . His+ colonies were selected on a medium lacking tryptophan, leucine and histidine. For the Gal4 construct, 79 million clones (7-fold the complexity of the library) were screened using the same mating approach with Tyr187 (MATα) and CG1945 (MATa) yeast strains. His+ colonies were selected on a medium lacking tryptophan, leucine, methionine and histidine. The prey fragments of the positive clones were amplified by PCR and sequenced at their 5′ and 3′ junctions. The resulting sequences were used to identify the corresponding interacting proteins in the GenBank® database (NCBI) using a fully automated procedure. A confidence score (PBS, for Predicted Biological Score) was attributed to each interaction as described previously .
Maintenance of cell lines
THP-1 cells were maintained in RPMI (Roswell Park Memorial Institute) medium (Invitrogen) supplemented with 10% (v/v) FBS and penicillin/streptomycin antibiotic (Invitrogen). Fibroblast cell lines derived from AGS patients and normal controls were grown in DMEM (Dulbecco's modified Eagle's medium; Invitrogen) supplemented with 10% (v/v) FBS and penicillin/streptomycin antibiotic. Cell lines were maintained at 37°C in 5% CO2.
Synthesis of recombinant SAMHD1 protein
The following recombinant protein fragments were generated, expressed and purified as described previously : SAMHD1 26–626, SAMHD1 26–583, SAMHD1 115–583 and SAMHD1 34–114.
StrepTactin pull-down assay
Settled StrepTactin Sepharose resin (25 μl) (IBA) was centrifuged at 1000 g for 1 min. The resin was washed in 0.1 M Tris/HCl (pH 8.0)/5% (v/v) glycerol/50 mM NaCl/5 mM MgCl2 buffer for 3 min and centrifuged at 1000 g for 1 min before discarding the supernatant. Recombinant protein (180 nM) was added to the resin and incubated at 4°C for 1 h to allow binding to occur. The beads were washed in a Triton X-100 wash buffer containing 0.1 M Tris/HCl (pH 8.0)/1% (v/v) Triton X-100/5% (v/v) glycerol/50 mM NaCl/5 mM MgCl2, before the addition of freshly prepared cell lysate. The reaction mixture was incubated at 4°C for 2 h. The beads were washed five times in Triton X-100 wash buffer before interacting proteins were eluted from the column by three sequential washes with elution buffers of increasing stringency (Triton X-100 wash buffer containing 100, 200 or 500 mM NaCl). Samples were analysed by SDS/PAGE in 10% polyacrylamide gels and visualized using either Instant Blue (Expedeon) or Silver Staining (0.1% silver nitrate stain). Gels were stored at 4°C in 1% (v/v) acetic acid solution until processing.
Identification of interacting proteins by MS
MS analysis of stained protein bands was carried out in the Biomolecular Analysis Facility at the Faculty of Life Sciences, University of Manchester.
Proteins of interest were excised from the gel and dehydrated using acetonitrile followed by vacuum centrifugation. Dried gel pieces were reduced with 10 mM DTT and alkylated with 55 mM iodoacetamide. Gel pieces were then washed sequentially with 25 mM ammonium bicarbonate followed by acetonitrile. This was repeated and the gel pieces dried by vacuum centrifugation. Samples were digested in a trypsin solution overnight at 37°C. Peptides were extracted with one wash of 20 mM ammonium bicarbonate followed by two washes of 50% (v/v) acetonitrile/5% (v/v) formic acid (FA). Samples were then dried down in a vacuum centrifuge and resuspended in 10 μl of 5% (v/v) acetonitrile/0.1% FA.
Digested samples were analysed by LC–MS/MS using an UltiMate® 3000 Rapid Separation LC (RSLC, Dionex Corporation) coupled to a LTQ Velos Pro (Thermo Fisher Scientific) mass spectrometer.
Peptide mixtures were separated using a gradient from 92% A (0.1% FA in water) and 8% B (0.1% FA in acetonitrile) to 33% B, in 44 min at 300 nl·min−1, using a 75 mm × 250 μm internal diameter 1.7 mM BEH (ethylene-bridged hybrid) C18, analytical column (Waters). Peptides were selected for fragmentation automatically by data-dependent analysis.
Data produced were searched using Mascot (Matrix Science), against the UniProt database with taxonomy of [Human] selected. Data were validated using Scaffold (Proteome Software).
Co-immunoprecipitation of SAMHD1, eEF1A1, Rbx1 and Cullin4A
Co-immunoprecipitations were carried out using the Co-immunoprecipation kit (Thermo-Scientific Pierce) as directed by the manufacturer's instructions. Briefly, 60 μg of SAMHD1 antibody (Abcam) was immobilized to 25 μl of AminoLink Plus coupling resin before incubation with 1 mg of pre-cleared cell lysate for 4 h at 4°C. After washing in 25 mM Tris/HCl/150 mM NaCl/1% (v/v) Nonidet P40, bound fractions were eluted at low pH (pH 2.8) using elution buffer (Thermo-Scientific Pierce) and stored at–20°C until analysis by SDS/PAGE. Eluate fractions were assayed for the presence of eEF1A1, Rbx1 and Cullin4A by Western blotting. Reciprocal experiments coupling antibodies specific to eEF1A1 (Sigma–Aldrich) were also performed.
Cells were fixed by treatment with 2% (w/v) paraformaldehyde (PFA) for 20 min followed by treatment with 0.1 M glycine for 15 min. The slides were washed three times in PBS before permeabilization in 0.5% Triton X-100 solution for 20 min. The slides were then washed three times in PBS solution before blocking in 1% BSA for 1 h at room temperature. Slides were treated with primary antibody diluted in 1% BSA, 5% goat serum [eEF1A1 1:100 (Abcam), SAMHD1 1:250 (Abcam)] for 1 h at room temperature (or overnight at 4°C). The slides were again washed three times in PBS/Tween before exposure to Alexa Fluor® 488 and 594 secondary antibodies (Molecular Probes/Invitrogen; 1:1000 dilution) and DAPI (1 μg/ml) for 1 h at room temperature. The slides were finally washed three times in PBS/Tween before mounting with the anti fade reagent Mowiol 4-88 (Sigma-Aldrich). Images were viewed and analysed as described for the PLA.
Harvested cells were resuspended in 20 volumes of lysis buffer [20 mM Hepes (pH 7.9), 250 mM sucrose, 3 mM MgCl2, 0.5% Nonidet P40, 3 mM 2-mercaptoethanol and Mini Complete™ protease inhibitor (Roche)] and lysed by 15 strokes with a chilled Dounce homogenizer. An aliquot of the homogenate (1/25th volume) was retained for testing as whole-cell lysate extract (WCL). The remaining homogenate was centrifuged at 1500 g at 4°C. Both the cell pellet and the supernatant were retained. A 40 μl aliquot of the supernatant representing the crude cytoplasmic fraction (C1) was retained for testing. The remaining supernatant was centrifuged at 5000 g for 10 min at 4°C. The supernatant was removed and retained for analysis, this was the post nuclear supernatant (PNS) or clean cytoplasmic fraction. The pelleted fraction was resuspended in 20 volumes of lysis buffer before homogenization by ten strokes in a cooled Dounce homogenizer. The homogenate was centrifuged at 1500 g at 4°C for 10 min. The supernatant was removed and retained for analysis as cytoplasmic fraction 2 (C2). The pellet was suspended in 20 pellet volumes of lysis buffer and homogenized by ten strokes in a cooled Dounce homogenizer before centrifugation at 1500 g at 4°C for 10 min. The supernatant was removed and retained as cytoplasmic fraction 3 (C3). The remaining pellet was resuspended in 100 μl of 1× RIPA (radioimmunoprecipitation assay) buffer supplemented with 1 unit/ml benzonase (NEB). The homogenate was incubated for 3 h at 4°C mixing occasionally. This homogenate represents the nuclear pellet (NP) fraction. The protein concentration of each fraction was quantified using the Pierce BCA Protein Assay kit (Thermo-Fisher) and SAMHD1 protein expression in each fraction was analysed by Western blot analysis.
Western blot analysis
The eluate samples were separated by PAGE in 10% acrylamide gels and transferred to a PVDF membrane (GE Healthcare) for 90 min at 115 V. The membrane was blocked in 5% (v/v) goat serum solution before probing with an antibody specific for the antigen of interest: SAMHD1 (Abcam), eEF1A1 (Sigma–Aldrich), Rbx1 (Abcam) or Cullin4A (Abcam). After incubation with the secondary antibody, anti-mouse (SAMHD1) or -rabbit (eEF1A1, Cullin4A and Rbx1), protein was detected using the ECL Advance Chemiluminescence kit (GE Healthcare).
In situ proximity ligation assay
Cultured cells were fixed by treatment with 2% (w/v) PFA for 20 min followed by 0.1 M glycine for 15 min. The cells were washed three times in PBS before permeabilization in 0.5% Triton X-100. Cells were washed again in PBS before incubation in a 1% (w/v) BSA block for 1 h at room temperature. After blocking, the cells were treated with primary antibodies diluted in 1% (w/v) BSA/5% (v/v) goat serum solution [eEF1A1 1:100 (Abcam), SAMHD1 1:250 (Abcam)] for 1 h at room temperature. The cells were washed three times in PBS/Tween solution. Proximity ligation was carried out using the Duolink in situ proximity ligation assay (Sigma–Aldrich) as directed by the manufacturer's instructions. Images were collected on an Olympus BX51 upright microscope using a ×60/UPlanFLN (Ph 3) objective and captured using a Coolsnap HQ camera (Photometrics) through MetaVue software (Molecular Devices). Specific bandpass filter sets for DAPI, FITC and Texas Red were used to prevent bleed through from one channel to the next. Images were then processed and analysed using ImageJ (http://rsb.info.nih.gov/ij).
eEF1A1 is identified in a complex with SAMHD1
Two independent techniques were used to identify novel proteins that interact with human SAMHD1. First, a yeast two-hybrid assay was carried out using a fusion protein between the LexA DNA-binding domain and the full-length human SAMHD1 as bait. Screening of 129 million colonies across two assays enabled selection of 20 His+ clones representing interactions between the SAMHD1 bait and proteins from the prey library. Sequencing of positive clones identified a number of candidate interacting proteins; these included SAMHD1 itself and eEF1A (Table 1). As SAMHD1 can exist as a homodimer and tetramer, it is not surprising that it was identified in this screen.
|Gene name||Shared domain||5′ identity (%)||3′ identity (%)|
|Gene name||Shared domain||5′ identity (%)||3′ identity (%)|
The second technique utilized N-terminal StrepTactin-tagged recombinant SAMHD1 protein (amino acids 26–626) in a pull-down experiment using a non-differentiated THP-1 cell line-derived lysate. A number of protein bands representing potential SAMHD1-binding partners were observed (Figure 1a). MS analysis identified one of these bands as the eEF1A isoform eEF1A1 (Figure 1b), confirming the interaction identified by yeast two-hybrid screen (Table 1).
Identification of SAMHD1-interacting proteins by StrepTactin pull-down assays
Endogenous SAMHD1 interacts with eEF1A1 in the mammalian cell line THP-1
To determine whether endogenous SAMHD1 interacts with eEF1A1, we performed co-immunoprecipitation experiments in cycling THP-1 cell-derived lysates using an anti-SAMHD1 antibody. Subsequent Western blot analysis demonstrated that eEF1A1 co-precipitated with SAMHD1 (Figure 2a). This interaction was confirmed in the reverse experiments utilizing an antibody-specific to eEF1A1 and where SAMHD1 was successfully co-precipitated (Figure 2b). These results indicate that SAMHD1 and eEF1A1 interact in THP1 cells at endogenous levels of expression.
Co-immunoprecipitation of SAMHD1 and eEF1A1 in THP-1 cells
Amino acids in the HD domain of SAMHD1 mediate complex formation
To determine which functional domain(s) of SAMHD1 interact(s) with eEF1A1, a series of StrepTactin-tagged recombinant fragments representing various domains of SAMHD1 were generated (Figure 3a) and pull-down experiments performed. Western blot analysis demonstrated eEF1A1 protein binding to SAMHD1 115–583 (Figure 3b). This result suggests that the interaction occurs between amino acids 115 and 583 of SAMHD1 within the HD domain and that the region between amino acids 26 and 114, which contains the SAM domain, confers a negative effect on eEF1A1-binding activity.
Identification of SAMHD1 domains that mediate eEF1A1 binding
eEF1A1–SAMHD1 complex formation is observed in situ in the cytoplasm
We employed a PLA to determine whether SAMHD1 and eEF1A1 are able to interact in situ. This experiment involved dual recognition of eEF1A1 and SAMHD1 with antibody affinity probes which, when brought together through complex formation, generate an amplifiable DNA reporter molecule . PLA demonstrated that eEF1A1 and SAMHD1 interact in fibroblasts and that this interaction is localized to the cytoplasm (Figure 4a and b). This is surprising given the localization of SAMHD1 in the cell is largely nuclear (Figure 5a). However, biochemical fractionation of normal fibroblast cells reveals that a small proportion of SAMHD1 can also be found in the cytoplasm (Figure 5b) and is, therefore, able to interact with eEF1A1. We also compared the level of interaction in normal fibroblasts and in fibroblasts from the AGS patients with mutations in SAMHD1 (Figure 4). Patient AGS128 harbours a premature stop codon at amino acid 149 (Q149X) of SAMHD1 leading to the loss of part of the HD domain, whereas patient AGS295 carries a deletion of exons 12–16, which affects the C-terminal region of the protein. Both patients maintain an intact SAM domain. Again, the observed interactions between SAMHD1 and eEF1A1 localized to the cytoplasm, with the greater number of interactions identified in SAMHD1 mutant AGS patient fibroblast cells compared with normal fibroblast controls (Figure 4d). Retention in the cytoplasm may, therefore, explain the increase in SAMHD1–eEF1A1 complexes in these cell lines. Patient AGS128 retained the ability to form an eEF1A1–SAMHD1 complex despite carrying a Q149X truncating mutation, suggesting that a fully intact HD domain is not required for the interaction to occur.
In situ PLA of the SAMHD1–eEF1A1 interaction
Cellular localization of SAMHD1 in normal and patient fibroblasts
Interaction of SAMHD1 and eEF1A1 with protein degradation machinery
One recognized function of eEF1A is to detect mis-folded proteins and target them to the proteasome for destruction, leading us to hypothesize that eEF1A1 might be required to target SAMHD1 to the proteasome. In order to test this possibility, co-immunoprecipitation experiments were undertaken to assess whether components of E3 ligase complexes could be found in association with SAMHD1 and eEF1A1. These data demonstrate an interaction of both SAMHD1 and eEF1A1 (Figure 6) with the Rbx1 and Cullin4A components of the CRL4–DDB1 E3 ubiquitin ligase complex, indicating that these proteins do physically interact with the protein degradation machinery.
Identification of the ubiquitin ligase components Rbx1 and Cullin4A in a complex with SAMHD1 and eEF1A1
In the present study, we have employed a range of complementary techniques to identify the eEF1A1 as a novel interaction partner of SAMHD1. We demonstrate that amino acids 115–583 of SAMHD1, containing the HD domain, mediates the eEF1A1 interaction. The PLA, however, showed that a mutant SAMHD1 protein carrying a deletion leading to the loss of part of the HD domain retained the ability to bind eEF1A1. This suggests that a complete HD domain is not required for the interaction to occur and localizes crucial interaction residues to between amino acids 115 and 149 of SAMHD1. Other proteins, possibly those associated with proteosomal degradation, mediate this interaction, as pull-down assays using purified SAMHD1 and purified eEF1A1 demonstrated that the association was not direct (result not shown)
Initial yeast two-hybrid and pull-down analyses identified a number of candidate SAMHD1-interacting proteins including SAMHD1 itself and eEF1A1. The ability of SAMHD1 to interact with itself has been described previously in reports looking first at the crystal structure of the protein and secondly at the oligomerization potential of SAMHD1, with these reports concluding that oligomerization of SAMHD1 is required for its function [11,20]. The interaction with eEF1A1 is novel and our in situ studies suggest that this interaction is physiological.
Despite extensive investigation, no role has yet been identified for the SAM domain in SAMHD1. Experiments employing SAMHD1 deletion and mutant constructs demonstrate that it is the HD domain that modulates retroviral restriction and potential RNA binding [20,28]. eEF1A1 has been reported to bind to the SAM domain of the Rho GTPase-activating protein (RhoGAP) deleted in liver cancer 1 (DLC1) , suggesting that the SAMHD1–eEF1A1 interaction may be mediated by the SAM domain. It was perhaps surprising then when our pull-down studies using recombinant deletion fragments of SAMHD1 localized this interaction to a region containing the HD domain. Indeed, the interaction with eEF1A1 appears to be stronger in experiments using a deletion construct lacking the SAM domain, suggesting that it may even have an inhibitory effect on eEF1A1 binding. This finding is reminiscent of the results of earlier studies in which a SAMHD1 deletion construct encoding only the HD domain appeared to have a greater ability to restrict HIV-1 infection than the full-length construct . Taken as a whole, these data suggest that the SAM domain and modifications/interactions within it, could play a role in regulating the functions of eEF1A1 [30,31]. These include nuclear export [30,32,33], apoptosis [34,35], maintaining cytoskeleton integrity and cell morphology [31,36] and proteolysis where eEF1A1 proteins have been proposed to detect misfolded proteins and target them to the proteosome for destruction [30,37,38]. Although SAMHD1 is considered to be a nuclear protein, a number of SAMHD1 mutant proteins associated with the inflammatory disorder AGS display altered sub-cellular localization to the cytoplasm . Therefore, one potential role for eEF1A1 might be to recognize misfolded SAMHD1 molecules in the cytoplasm and help target them for degradation by the proteasome. The increase in interactions observed in the patient cell lines containing mutant SAMHD1 proteins and the co-immunoprecipitation of Rbx1 and Cullin4A, components of an E3 ubiquitin ligase complex with both SAMHD1 and eEF1A1, would support such a hypothesis.
Catherine Morrissey conducted the experiments and wrote the paper. David Schwefel and Valerie Ennis-Adeniran supplied all the purified protein constructs for the study. Ian Taylor aided the design of the study and helped write the paper. Yanick Crow and Michelle Webb designed the study and wrote the paper.
Yeast two-hybrid analysis was carried out by Hybrigenics. The Bioimaging Facility microscopes used in the present study were purchased with grants from Biotechnology and Biological Sciences Research Council, Wellcome and the University of Manchester Strategic Fund. Special thanks goes to Peter March and Steven Marsden for their help with the microscopy. We also thank Emma-Jane Keevil and Julian Selley for their assistance with MS.
We would like to dedicate this work to Dr Tina Ivanov.
This work was funded by the European Leukodystrophy Association [grant number ELA 2009-012I5]; the European Union's Seventh Framework Programme [grant number FP7/2007-2013 under grant agreement 241779]; and the UK Medical Research Council [grant number U117565647 to I.A.T].
cyclin-dependent kinase 1
Cullin4A-RING E3 ubiquitin ligase 4
eukaryotic elongation factor
proximity ligation assay
post nuclear supernatant
SAM domain and HD domain-containing protein 1