The following evidence supports the 25 directly cited annotation contexts. The natural products of 339, 152 and 206 amino acids remain distinct; none of these experiments establishes their functional equivalence.
Original PDF text lines 337–510 and 760–880; Figures 4–7 captions; plasmid Methods lines 803–831. Human ERLIN2 cloned from U937 cDNA is expressed as tagged protein in mouse NIH3T3 and human HeLa hosts. Membrane fractionation, cholesterol-sensitive intracellular detergent-resistant fractions and ER-marker colocalization support ER membrane and raft-like domain localization. Engineered N-terminal targeting probes are separate from natural products. No figure pixels or supplementary material inspected. Local PDF download failed; operative original PDF text was available through the saved primary web rendering.
Blood pressure is regulated by an alpha1D-adrenergic receptor/dystrophin signalosome.
Normal original Results lines 100–185; actual original Table 3 in saved primary rendering. TAP-alpha1D-AR in human HEK293 cells recovers ERLIN2 with 19 distinct peptides and 48% sequence coverage. This supports contextual assembly, while the functional signalosome experiments test syntrophin/dystrophin components rather than an ERLIN2-specific role. Complete original Methods, figure pixels and supplements were not recovered/read. Table 3 and the relevant Results were read.
Normal original Results lines 65–455; recovered plasmid Methods; Figures 1–7 captions. Native complex and IP3-receptor experiments use mouse alphaT3-1 or rat Rat-1 cells. Methods specify mouse SPFH1/SPFH2 constructs, including when HeLa is the expression host. Early IP3R engagement, assembly perturbations, RNAi/rescue and degradation establish a mechanistic erlin-complex contribution. Apparent masses, band ratios and the approximately 50-subunit model do not establish exact human stoichiometry. No figure pixels or supplements inspected. Human conservation is corroborated separately, not relabeled as a human assay here.
Indexed original blocks 11–65; Methods 17–45 and Results 47–65; Figures 1–5 captions. Human SPFH2/ERLIN2, gp78, HRD1 and TMUB1 constructs are tested in CHO-7 cells; endogenous human gp78–SPFH2 association is measured separately in HEK293. The tagged HMGCR construct is hamster P00347. Insig/gp78 enhance HMGCR association and TMUB1 is the experimentally tested bridge to gp78. Knockdown links ERLIN2 to sterol-accelerated HMGCR ubiquitination and turnover in this context. Original supplementary Figure 1B protein list and figure pixels were not inspected. The ERLIN1 pair is independently supported by PMID:22119785 and PMID:33961781. Discordant HMGCR-dependence reports are acknowledged in the main paper.
Defining human ERAD networks through an integrative mapping strategy.
Indexed original blocks 7–8, 10–17, 22, 26, 30, 38 and 41–42; exact source-specific IntAct records listed in recommendations. Human S-tagged ERAD baits in HEK293, digitonin/Triton extraction, affinity proteomics and CompPASS support network associations. Current source-specific records contain ERLIN2 with ERLIN1, SYVN1 and AMFR. Independent PMID:21343306 experiments corroborate the E3 interactions. Detailed Online Methods and original supplementary peptide tables were not inspected. Global validation rates are not pair-specific orthogonal validation. No product-level identity beyond the inspected source is inferred.
Actual PMC Methods lines 122–150; Results 186–218; Figure 4A caption around line 201. FLAG-RNF185 and FLAG-RNF5 each recover endogenous ERLIN2 in human HEK293T cells against control-vector comparisons. The full recovered original supplies the operative assay missing from the truncated normal cache. CFTR-function experiments perturb the ligases, not ERLIN2. No figure pixels or supplementary material inspected. The normal cache access metadata is preserved despite its truncated body.
Erlins restrict SREBP activation in the ER and regulate cellular cholesterol homeostasis.
Indexed original blocks 8–58; Figures 1–5 captions; Methods 48 and 56–58. Human ERLIN2 depletion elevates SREBP processing/nuclear accumulation, pathway transcripts and lipid pools. Fatostatin, Insig turnover and regulatory-complex association support negative pathway regulation. Some Insig and other constructs are tagged; Figure 3D/E includes hamster Scap. Figure 4 purified human binding reagent is ERLIN1; the native cholesterol-binding complex is mouse erlin1/2. Neither dissociation constant is an isolated human ERLIN2 measurement. No figure pixels or supplements inspected. Transcript/processing/lipid-pool assays are not isotope-resolved biosynthetic flux, and natural shorter ERLIN2 products were not shown equivalent.
Flotillin-1 facilitates toll-like receptor 3 signaling in human endothelial cells.
Indexed Methods block 21; Results 53; Figure 5 caption 54; Table 1 block 56; primary web Methods 126–132. Human HUVEC SILAC/sucrose-gradient detergent-resistant membrane proteomics identifies ERLIN2 O94905: score 793, shFlot1/control ratio 0.58, SEM 0.10, three preparations. HUVEC provenance is explicit. This is operational membrane-fraction association, not evidence that ERLIN2 executes TLR3 signaling or that the detected material is exclusively plasma membrane. Supplement 395_2014_439_MOESM1_ESM.pdf and figure pixels were not read. Main Table 1 was directly read.
Histone Interaction Landscapes Visualized by Crosslinking Mass Spectrometry in Intact Cell Nuclei.
Indexed original blocks 11–22, 25, 27–29, 39–41 and 70; exact IntAct EBI-20904152. U2OS nuclear preparations are cross-linked with DSSO and analyzed against a human sequence database. The exact ERLIN1–ERLIN2 pair is in the current original-specific interaction record. Nuclear-envelope/ER material can remain in nuclear preparations; the raw occurs_in nucleus extension is recorded without inferring soluble nucleoplasmic function. Original supplementary cross-linked peptide files could not be recovered: saved XLSX-named responses are not ZIP spreadsheets. No peptide-residue pair, figure pixels or product-specific boundary is asserted.
Actual official PMC search-rendered Results 3.1–3.3 and Discussion; Figure 1D and Figure 2A captions read. Human GFP-STAS/TMEM41B in HEK293 cells is affinity-purified; ERLIN2 recovery is confirmed by immunoblot with GFP/untransfected controls and SMN as a negative interaction control. Mouse embryo and NIH3T3 experiments are separate. This supports the exact association without transferring Stasimon function to ERLIN2. Complete Methods and supplementary protein list remained unavailable; direct PMC access challenged and EPMC XML returned HTTP 500. No figure pixels, exact ERLIN2 reagent sequence or purified binary affinity was inspected.
Dual proteome-scale networks reveal cell-specific remodeling of the human interactome.
Indexed original Methods blocks 184, 187–188, 195–197, 217–218 and 220–227; HCT116 directed table line 13690; 293T table line 33356. Human ORFeome C-terminal FLAG-HA baits, sequence-verified clones and CompPASS-Plus filtering yield gene-level networks. HCT116 contains ERLIN1 bait–ERLIN2 prey and 293T contains TMUB1 bait–ERLIN2 prey. These are affinity associations, not proof of binary interfaces, product equivalence or nuclear localization. No figure pixels or exact pair-specific clone sequence analysis was performed. Directed edge rows were read from actual author data.
Role of ERLINs in the Control of Cell Fate through Lipid Rafts.
Actual review Sections 2.4 and 3. This is a secondary review of ERLIN proteins, detergent-resistant domains and MAM-associated functions. Original localization support comes from PMID:16835267 and PMID:25204797; the review is not treated as a new direct experiment. No new primary assay, exclusive localization or natural-product equivalence is inferred from the review.
The four ERLIN1-partner records carrying a raw nucleus extension and the HUVEC raft cell-type extension are preserved in the raw-to-review context ledger. These extensions do not become new location assertions. Screening interaction support is distinguished from original peptide-table access and from binary binding.
Current official QuickGO definitions support ubiquitin protein ligase binding, negative regulation of SREBP signaling pathway, and negative regulation of cholesterol biosynthetic process. SREBP restraint is a regulatory refinement, not a claim that ERLIN2 executes SREBP cleavage.
These source assessments distinguish natural ERLIN2, its experimentally studied constructs, and the ERLIN2–FGFR1 cancer fusion. They do not assign a disease fusion's kinase activity or location to all natural splice products. Publication wording is summarized without adding quotations.
Nine ER-membrane assertions reach native human ERLIN2 entity 8866520, O94905 residues 1–339. Twelve plasma-membrane assertions instead reach 8853148 or its phosphorylated counterpart 8853122. Those are disease-fusion participants. Their modeled fusion location is not transferred to natural ERLIN2: the twelve native-product GOA projections are removed, while the valid fusion participants remain. Independent cell-specific membrane staining of natural ERLIN2 is documented separately.
The fusion model retains O94905 coordinates 1–186 but replaces residue 186 with the 30-residue segment MVSLKRRIELTVEYPWRCGALSPTSNCRTG and inserts FGFR1/P11362 residues 1–822 at the following junction. The resulting arithmetic is 185 + 30 + 822 = 1,037 residues, agreeing with the actually inspected original Figure 2. The model's EML4 display name conflicts with its ERLIN2 reference entity and alternative name; the original source is preserved and the naming inconsistency should be corrected upstream. Names saying 1–185 versus coordinates 1–186 reflect the replacement boundary. The original paper reports fusion activation in its main Results; its supplementary Figure S6 was not inspected. A failed supplement download contains HTML and supplies no figure evidence. Plasma-membrane location here is the curated Reactome entity model, not an original localization image.
The following paths follow actual input, output, catalyst, active-unit, physical-entity, component or member edges. An event-level compartment alone was not used to infer target location. Numbers in the first column identify the frozen source contexts used for this assessment.
| Source index | Event | Exact participant route | Target scope |
|---|---|---|---|
| 20 | R-HSA-8866542 | 8866542 → 8866531 → 8866522 → 8866520 | Natural ERLIN2, ER membrane |
| 21 | R-HSA-8866546 | 8866546 → 8866538 → 8866531 → 8866522 → 8866520 | Natural ERLIN2, ER membrane |
| 22 | R-HSA-8866551 | 8866551 → 8866531 → 8866522 → 8866520 | Natural ERLIN2, ER membrane |
| 23 | R-HSA-8866854 | 8866854 → 8866531 → 8866522 → 8866520 | Natural ERLIN2, ER membrane |
| 24 | R-HSA-8866856 | 8866856 → 8866848 → 8866531 → 8866522 → 8866520 | Natural ERLIN2, ER membrane |
| 25 | R-HSA-8866857 | 8866857 → 8866531 → 8866522 → 8866520 | Natural ERLIN2, ER membrane |
| 26 | R-HSA-9931264 | 9931264 → 8866531 → 8866522 → 8866520 | Natural ERLIN2, ER membrane |
| 27 | R-HSA-9931298 | 9931298 → 9931282 → 8866531 → 8866522 → 8866520 | Natural ERLIN2, ER membrane |
| 28 | R-HSA-9931313 | 9931313 → 8866531 → 8866522 → 8866520 | Natural ERLIN2, ER membrane |
| 29 | R-HSA-1839094 | 1839094 → 5691527 → 5655215 → 8853273 → 8853166 → 8853122 | ERLIN2–FGFR1, plasma membrane |
| 30 | R-HSA-1839098 | 1839098 → 1839061 → 5691527 → 5655215 → 8853273 → 8853166 → 8853122 | ERLIN2–FGFR1, plasma membrane |
| 31 | R-HSA-1839100 | 1839100 → 5691527 → 5655215 → 8853273 → 8853166 → 8853122 | ERLIN2–FGFR1, plasma membrane |
| 32 | R-HSA-5655240 | 5655240 → 5655216 → 5655215 → 8853273 → 8853166 → 8853122 | ERLIN2–FGFR1, plasma membrane |
| 33 | R-HSA-5655263 | 5655263 → 5655218 → 5655216 → 5655215 → 8853273 → 8853166 → 8853122 | ERLIN2–FGFR1, plasma membrane |
| 34 | R-HSA-5655266 | 5655266 → 5655216 → 5655215 → 8853273 → 8853166 → 8853122 | ERLIN2–FGFR1, plasma membrane |
| 35 | R-HSA-5655269 | 5655269 → 5655215 → 8853273 → 8853166 → 8853122 | ERLIN2–FGFR1, plasma membrane |
| 36 | R-HSA-5655278 | 5655278 → 5655296 → 5655215 → 8853273 → 8853166 → 8853122 | ERLIN2–FGFR1, plasma membrane |
| 37 | R-HSA-5655290 | 5655290 → 5655254 → 5655217 → 5655218 → 5655216 → 5655215 → 8853273 → 8853166 → 8853122 | ERLIN2–FGFR1, plasma membrane |
| 38 | R-HSA-5655326 | 5655326 → 5655316 → 5655328 → 5655216 → 5655215 → 8853273 → 8853166 → 8853122 | ERLIN2–FGFR1, plasma membrane |
| 39 | R-HSA-8853322 | 8853322 → 8853277 → 8853148 | ERLIN2–FGFR1, plasma membrane |
| 40 | R-HSA-8853325 | 8853325 → 8853271 → 8853164 → 8853148 | ERLIN2–FGFR1, plasma membrane |
In the official PTHR15351 version 19 tree, human ERLIN2 leaf PTN002495868 descends from ER-membrane/ERAD node PTN000794169 and SREBP node PTN001282580. Its ungapped MSA equals the canonical 339-residue sequence. The worm ERAD NOT node PTN002692340 is outside this target path. The current cholesterol-binding IBD, PTN002692334, lies in the ERLIN1 clade outside the ERLIN2 path, although aggregate family labels retain the cholesterol term. This is a source-version/ancestry distinction, not evidence of biological loss. Target self-WITH and the number of experimental descendants are not defects.
The earlier ERLIN2 cholesterol IBA is scientifically UNDECIDED. Recombinant human ERLIN1 binds cholesterol in the 2013 erlin study; native mouse mixed erlin complexes do not isolate ERLIN2's contribution. Independent sterol-probe proteomics positively detects human ERLIN2 in Group III but lacks threshold native-cholesterol competition. Those findings support analogue interaction without settling the exact cholesterol-binding annotation. Whether an absent historical source row is retired is separate provenance bookkeeping; its inherited source and quotation remain preserved.
InterPro IPR033294 matches O94905 residues 1–314 and carries the ER and ubiquitin-ligase-binding mappings. The actual erlin FunFam is 3.30.479.30:FF:000009. Matching ARBA disjuncts are 00028685 set 35 (ER membrane), 00028902 set 381 (complex), 00029018 set 18 (ERAD), 00084177 set 1 (fatty-acid regulation), 00085017 set 0 (cholesterol regulation), and 00087340 set 0 (SREBP). Their taxonomic conditions were checked against the target lineage. The supported complex refinement is GO:0140534. The regulatory SREBP correction is GO:2000639; it is a lateral role correction, not an is_a specialization of the signaling-pathway term. Fatty-acid regulation is retained as a supported downstream consequence, outside the defining core.
HPA ERLIN2 uses antibody HPA002025, whose 148-residue antigen matches canonical residues 190–337. A-431 and U-251MG support ER localization; raw U2OS records also report plasma membrane/cytosol. Actual A-431 and U2OS images were inspected. These images do not independently establish bilayer attachment, and they do not identify every short splice product. Human ERLIN2-GFP and independently established membrane topology support the ER-to-ER-membrane refinements. GO_REF:0000054 instead describes LIFEdb GFP-cDNA imaging; its historical target image/construct was not recovered after the resource redirected. That access limit is not used to reject the supported location.
Original article. Read: Indexed Results 8–42; Methods 44,46,48–60,62,64,66; Figures 1–3: ERLIN2 depletion, SREBP processing, endogenous complex association; Figure 4B/Methods 56–57: recombinant human ERLIN1; Figure 4C/Methods 58–60: native mouse erlin complex; Figure 5: human ERLIN2-GFP FRAP.
Human ERLIN2 restrains SREBP activation and downstream lipid accumulation. Single ERLIN2 depletion, processing assays and regulatory-complex association establish the regulatory role. The purified cholesterol-binding measurement belongs to ERLIN1; mixed native mouse complexes do not assign an ERLIN2-specific affinity. No figure pixels or supplementary material read. The tagged regulatory-partner experiment includes hamster SCAP; cellular associations do not establish every binary interaction. Lipid abundance is not synthetic flux.
Original article. Read: Indexed Methods 17,19,21–24; Results/captions 49–59,61,64–66; Figures 1B,2A–C: endogenous human ERLIN2/gp78 and human SPFH2 construct associations; Figure 3: human HEK293 HMGCR ubiquitination/degradation; Figures 4–5: TMUB1-mediated bridge.
Target-specific human ERLIN2 associates with ER ligases and contributes to sterol-regulated HMGCR ubiquitination/degradation. CHO cells are a host for human SPFH2 constructs. TMUB1-mediated association limits direct binary-binding claims. No figure pixels or supplementary material read. Cargo dependence varies between cellular contexts; no ERLIN2 ubiquitin-transfer chemistry inferred.
Original article. Read: Normal original text lines 78–215 and 270–419; Actual author-original Methods: mouse SPFH1 BC011220 and mouse SPFH2-HA constructs; Figures 1,2A/B/D,5–7: recruitment timing, membrane/topology, depletion/rescue and assembly.
Erlin complexes associate with activated IP3 receptors before substantial ubiquitination and contribute to their regulated degradation. Rat-1 and alphaT3-1 experiments and mouse tagged constructs remain distinct from human host cells. No figure pixels or supplementary material read; complete Methods and full Figure 2C context were not read. This is conserved mammalian mechanism with explicit construct boundaries, not a newly demonstrated human rescue assay.
Original article. Read: Indexed Results/Discussion 17–35,44–48; Methods 53–61,64–66; Table 1 Total profile Excel row 437 and Group III row 138; Table 3 controls row 465; vs PEA-DA & 10X Ch row 174; sterol probe comparison row 166.
Human ERLIN2 is a positive Group III sterol-probe interactor: no-UV ratio 20, PEA-DA selectivity 3.1, cholesterol-competition ratio 1.3 in Table 1 and 1.4 in Table 3; vehicle ratio 13.3. The competition values are below the stated 1.5 threshold. Cis/epi/trans probe comparisons are 0.4/2/0.9. No figure pixels read. This is living-cell analogue labeling, not purified ERLIN2 affinity measurement; absence of threshold competition does not prove absence of cholesterol binding. Supplementary Tables 1 and 3 were actually parsed, other supplements were not read.
Original article. Read: Indexed Results 43–54 and Methods 72–88; Figure 2 and caption 44–46: ERLIN2–FGFR1 junction 186 and schematic length 1,037; Main Results 51 reports active expressed ERLIN2–FGFR1 and points to Supplementary Figure S6; construct preparation Methods 87.
The cancer-associated ERLIN2–FGFR1 fusion is an experimentally reported product, distinct from natural ERLIN2 splice products. The observed Figure 2 length agrees with the explicit Reactome modified-sequence arithmetic. Supplementary Figure S6 was not read; the attempted supplement1.pdf is an HTML challenge, not a PDF. No direct plasma-membrane imaging is attributed to the paper. The specific plasma-membrane assignment is the Reactome disease-product model.
Original article. Read: Indexed main 0–36,39–44; Figure 3 and Methods; Worm ERL-1-HA, 312 residues, cDNA yk705a8; HeLa localization and HEK293 oligomerization hosts; tm2703 worm physiology and ITR-1/ER-stress tests.
Worm ERL-1 localizes to the ER when expressed in HeLa and supports a conserved membrane/assembly interpretation. Its negative physiological ERAD findings belong to the worm source and the off-target PAINT NOT branch. No figure pixels or supplementary material read. Heterologous human host cells do not make the worm construct a human ERLIN2 assay.
| Source | Original and operative locator | Supported scope and limits |
|---|---|---|
| PMID:19751772 | Original PMC2764810, model-substrate Methods; Results and Figure 6; Figure 7 HMGCR controls | Actual full normal original read. Combined erlin depletion modestly stabilizes misfolded A1AT-NHK/TTR-D18G in human HEK/HeLa, including CHX turnover; selected controls are unaffected. ERLIN2 depletion also reduces ERLIN1; rescue uses mouse SPFH2. This supplies quality-control evidence distinct from activated IP3R turnover. No figure pixels or supplements inspected. |
| PMID:21610068 | Original PMC3129221, Experimental Procedures; Interaction of RNF170 with the erlin1/2 complex; Priority of binding to activated IP3 receptors; Figures 4–6, especially 5C/D | Official primary search-rendered Methods/Results/captions read despite the immutable normal cache being abstract-only. Human mHeLa ERLIN2 depletion prevents RNF170 recruitment; reciprocal RNF170 depletion preserves ERLIN2 recruitment. Mouse cloned RNF170 and rodent cell experiments remain distinct. No figure pixels/supplements or purified binary-affinity result claimed. |
| PMID:38782601 | Original PMC11116810, Results/Figures 1–3 and 7–8; cell, interaction, lipid and AlphaFold Methods | Human HeLa double-knockout/rescue and association experiments support the erlin platform; cholesterol esterification increases after knockout. SOAT1 inhibition modifies cellular phenotypes. AlphaFold interfaces and simultaneous partner occupancy are not established atomic structure. Main text read; no pixels, supplements or raw model coordinates inspected in this synthesis. |
| PMID:38607533 | Original PMC11306645, fibroblast, velocity-gradient and blue-native PAGE Methods; assembly Results | One heterozygous V168M patient fibroblast line and two controls show comparable gross complex distribution. This does not rule out every dominant-negative or cargo-handling defect. Main text read; no pixels or supplements inspected. |
| PMID:40225166 | Original PMC11919107, cell-derivation/calcium/co-IP Methods; Figures 3–4 Results | Patient-derived nonisogenic comparisons measure IP3R1 abundance, calcium responses and RNF213-associated complexes. Direct ubiquitination, half-life and RNF213-loss rescue are not demonstrated. Main text read; no pixels or supplements inspected. |
For current scope see GO:0140378 scaffold activity, GO:0036503 ERAD quality control, GO:2000639 negative SREBP regulation, GO:0045541 negative cholesterol-biosynthesis regulation, and GO:0031625 ubiquitin protein ligase binding. Binding uncertainty concerns target and native-ligand attribution, not a general requirement for purified-protein assays.
The structural core now records GO:0005198 structural molecule activity for ERLIN2's integral contribution to oligomer integrity, and contributes_to_molecular_function: GO:0140378 for the assembly's scaffold activity. Physical assembly evidence (PMID:19240031) and human complex-dependent partner association (PMID:38782601) support this synthesis without independently resolved ERLIN2 bridging or equal/interchangeable subunits. The two slots express distinct attribution levels within one core. No NEW scaffold annotation is proposed because the synthesis is bounded to structural participation in the complex; an independently enabled ERLIN2 scaffold activity is not established. All 64 annotation/review objects, three products, 49 reference identities and 57 inherited quotation paths remain exact. No new original-access or figure-inspection claim is made by this refinement.
The twelve fusion routes in the table above support the identity and modeled location of ERLIN2–FGFR1, not native ERLIN2. WRONG_GENE_PRODUCT defects request correction of the enclosing native-product GOA projections. Separate OTHER/CHECK defects identify the EML4 display-name conflict on entities 8853148 and 8853122 without proposing a new accession or deleting a fusion event. Their Reactome reference assessments are LOW relevance to native biology and remain VERIFIED for source identity. No upstream report has been submitted.
For the alpha1D-AR complex, PMID:18468998 Table 3 reports ERLIN2 with 19 peptides and 48% coverage in HEK293 affinity material. The inherited PMID:19240031 erlin-complex quotation is historical context only; it does not establish that signalosome claim. For RNF185/RNF5, PMID:24019521 Methods and Figure 4A directly support endogenous ERLIN2 association in HEK293T. Their accepted binding fits the existing association core while CFTR recruitment, catalytic and universal ERAD claims remain unestablished.
UniProt CC INTERACTION snippets are imported interaction provenance, not independent experimental replication. The original co-immunoprecipitation sections and author-directed affinity datasets described above provide the positive partner evidence. Supported generic interactions remain non-core under the campaign binding instruction; inherited quotations remain exact under the adjacent no-new-quotation instruction. The two source-context decisions do not assume all appearances of one GO term require the same action.