FAM111B is a predominantly nuclear serine endopeptidase with intrinsic self-cleavage activity. A smaller pool associates with the nuclear periphery near nuclear pore complexes, and cytoplasmic protein is also detected. Loss of FAM111B disrupts nuclear architecture, chromosome stability and telomere maintenance in human cell models; protease-active FAM111B is required to rescue the tested ultrafine-bridge phenotype. Its physiological cleavage targets remain unidentified. Substitutions associated with hereditary fibrosing poikiloderma can increase protease activity and disrupt DNA synthesis, transcription and cell survival. The respective activities and localization of the natural 734- and 704-residue products remain unresolved.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0000785 chromatin | IBA GO_REF:0000033 | UNDECIDED | Summary: The ancestral chromatin assertion is traceable, but chromatin association of human FAM111B remains unresolved at the resolution of the inspected target experiments. Reason: The actual PANTHER v19 tree, exact human FAM111B sequence and matching chromatin IBD at PTN002683271 are verified. PMID:24561620 supplies positive FAM111A nascent-chromatin and PIP-dependent PCNA evidence; donor count is not a measure of its validity. PMID:32776417 Results on hereditary fibrosing poikiloderma reports that FAM111B lacks a recognizable PIP-box and is not enriched at DNA replication sites (Figure EV5A). Its depletion/knockout experiments also find normal basal DNA replication, PCNA chromatin loading and proliferation (Appendix Figure S2 as described in the main Results). Crucially, the paper reports the same basal dispensability for FAM111A, whose obstacle-dependent activity is independently established; this negative alone cannot refute the inherited stress-dependent claim. Only patient-associated FAM111B alleles, not wild type, produce the reported shutdown of DNA synthesis/transcription and associated toxicity. These mutant gain-of-activity phenotypes are not evidence that wild-type FAM111B normally performs the donor function. Current GO:0000785 denotes the organized chromosome-forming DNA-protein complex. PMID:37342232 Results 2.1/Figure 1 separates soluble material from a pellet that can include both chromatin and membranes, while detergent-resistant peripheral FAM111B associates with nuclear-pore components rather than lamins. These positive nuclear/peripheral observations do not resolve a chromatin-specific pool. Lack of replication-site enrichment supports a difference from the donor recruitment mechanism without excluding every chromatin interaction. Retain UNDECIDED because target-specific compartment applicability remains unresolved, not because the ancestral path is invalid or a universal loss has been demonstrated. Propagation Review Root cause: UNRESOLVED Sources checked: PANTHER:PTN002683271 UNRESOLVED The saved v19 target sequence, path and chromatin IBD match are verified. FAM111A chromatin evidence is positive; whether this compartment assertion applies to the target remains unresolved after inspecting the different recruitment and fractionation contexts. The uncertainty concerns target applicability, not node placement or donor count. |
| GO:0005515 protein binding | IPI PMID:25416956 A proteome-scale map of the human interactome network. | KEEP AS NON CORE | Summary: A controlled binary screen identifies a FAM111B–SET interaction. Reason: PMID:25416956 Table S2G explicitly reports SET–FAM111B in the HI-II-14 yeast two-hybrid map. The source ORF54777 nucleotide sequence in Table S2B translates exactly to the current 734-residue Q6SJ93-1 sequence, although this comparison does not establish the sequence of every subsequently used plasmid. The screen excludes autoactivators and requires repeated pairwise positives with sequence identification; the inspected validation tables do not establish a target-specific orthogonal validation. Retain the positive interaction as non-core under the campaign exception for supported generic protein binding. SET-domain binding is not binding to the SET protein, and SET association does not assign SET-associated phosphatase inhibition or histone-chaperone activity to FAM111B. No supported functional binding replacement was established. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | KEEP AS NON CORE | Summary: HuRI recovers a binary FAM111B–SET isoform-2 interaction. Reason: PMID:32296183 and the exact HuRI records identify Q6SJ93-1/ORF54777 with Q01105-2/ORF10104, using N-terminal Gal4 fusions in yeast. Pooling, array and validated-two-hybrid records describe the screen workflow, not three native-cell interaction assays. The historical FAM111B ORF identifier agrees with the independently sequenced HI-II-14 source construct, but current HuRI records specify partial sequence identification and do not prove complete plasmid or 704-residue product equivalence. Retain the positive interaction as non-core under the campaign exception for supported generic protein binding. Neither SET-mediated phosphatase inhibition nor a specific FAM111B adaptor/chaperone mechanism was measured for this pair. |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | KEEP AS NON CORE | Summary: NeuroNet reports a binary FAM111B–WFS1 interaction. Reason: PMID:32814053 Table S2 NN1.0 row 3208 and corresponding IntAct array, pooling and validated-two-hybrid records identify the FAM111B–O76024 pair. Human bait/prey constructs were assayed as fusions in yeast; non-autoactivating baits and multiple reporter/retest steps support the screen. The gene-level network and retrieved participant records do not establish a complete FAM111B insert sequence or equivalence of its 734- and 704-residue natural products. No individual mammalian orthogonal validation is claimed. Retain the positive interaction as non-core under the campaign exception for supported generic protein binding. The contact does not establish an ER targeting, transporter-regulation or chaperone activity for FAM111B. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | KEEP AS NON CORE | Summary: NeuroNet reports a binary FAM111B–GRN interaction. Reason: PMID:32814053 Table S2 NN1.0 row 3206 and corresponding IntAct array, pooling and validated-two-hybrid records identify the FAM111B–P28799 pair. Human bait/prey constructs were assayed as fusions in yeast; non-autoactivating baits and multiple reporter/retest steps support the screen. The gene-level network and retrieved participant records do not establish a complete FAM111B insert sequence or equivalence of its 734- and 704-residue natural products. No individual mammalian orthogonal validation is claimed. Retain the positive interaction as non-core under the campaign exception for supported generic protein binding. The contact does not establish cytokine-receptor activity or that GRN is cleaved by FAM111B. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | KEEP AS NON CORE | Summary: NeuroNet reports a binary FAM111B–HTT interaction. Reason: PMID:32814053 Table S2 NN1.0 row 3207 and IntAct records for multiple HTT fragments, including 1–504/1–511, 1221–1939 and 1932–2642, with polyglutamine-related feature annotations identify the FAM111B–P42858 pair. Human bait/prey constructs were assayed as fusions in yeast; non-autoactivating baits and multiple reporter/retest steps support the screen. The gene-level network and retrieved participant records do not establish a complete FAM111B insert sequence or equivalence of its 734- and 704-residue natural products. No individual mammalian orthogonal validation is claimed. Retain the positive interaction as non-core under the campaign exception for supported generic protein binding. A fragment-screen interaction does not establish FAM111B-mediated HTT proteolysis, chaperoning or pathological brain aggregation. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | KEEP AS NON CORE | Summary: NeuroNet reports a binary FAM111B–DNM2 isoform 2 interaction. Reason: PMID:32814053 Table S2 NN1.0 row 3204 and corresponding exact isoform-labelled IntAct records identify the FAM111B–P50570-2 pair. Human bait/prey constructs were assayed as fusions in yeast; non-autoactivating baits and multiple reporter/retest steps support the screen. The gene-level network and retrieved participant records do not establish a complete FAM111B insert sequence or equivalence of its 734- and 704-residue natural products. No individual mammalian orthogonal validation is claimed. Retain the positive interaction as non-core under the campaign exception for supported generic protein binding. No DNM2 GTPase regulation, endocytic adaptor mechanism or catalytic coupling by FAM111B was demonstrated. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:33961781 Dual proteome-scale networks reveal cell-specific remodeling... | KEEP AS NON CORE | Summary: BioPlex detects SET-associated FAM111B in human HCT116 cells. Reason: PMID:33961781 and its BioPlex HCT116 network row 18175 report SET–FAM111B with historical identifiers Q01105-2 and Q6SJ93-2. The separate directed table explicitly identifies SET as the bait and FAM111B as the prey; this direction is not inferred from undirected A/B order. C-terminal FLAG-HA bait purification and mass spectrometry establish a cellular co-association, not purified binary binding or a demonstrated FAM111B regulatory mechanism. The prey identifier alone does not resolve its isoform-specific peptides or assign the interaction uniquely to the natural 704-residue product. No corresponding pair in the inspected 293T files is not evidence of biological absence. Retain the positive interaction as non-core under the campaign exception for supported generic protein binding. The association does not transfer SET phosphatase-inhibitor or histone-chaperone function to FAM111B. Supporting Evidence: |
| GO:0005634 nucleus | IBA GO_REF:0000033 | ACCEPT | Summary: Nuclear localization is supported by both the verified ancestral assertion and endogenous human FAM111B localization. Reason: The matching nucleus IBD at PTN002683271 is on the verified v19 path to the exact 734-residue FAM111B sequence. Both FAM111B and FAM111A are legitimate experimental descendant sources; the target appearing in WITH/FROM is not circular evidence. Independently, PMID:37342232 reports that most endogenous FAM111B signal in human U2OS cells is nuclear, with a cytoplasmic pool and detergent-resistant nuclear-peripheral material also present. The nuclear assertion does not require exclusive localization, a chromatin-specific assignment, or a lamin-containing complex. Endogenous antibody staining does not distinguish the two natural products, and the tagged full-length and mutant constructs are not evidence that the 704-residue product has an identical distribution. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN002683271 SUPPORTS TRANSFER The exact target leaf descends from PTN002683271 and the nucleus IBD matches its recorded descendant sources, including the valid target self-source. Independent endogenous human target localization supports the transfer. |
| GO:0005634 nucleus | IDA PMID:36610347 FAM111B dysregulation promotes malignancy in fibrosarcoma an... | ACCEPT | Summary: Human FAM111B has a well-supported nuclear pool. Reason: The accessible PMID:36610347 abstract reports robust nuclear FAM111B in human HT1080 cells, but its operative main article could not be recovered in the bounded publisher/repository attempts. Independent PMID:37342232 Results 2.1 and Figure 1A–C provide actual endogenous U2OS imaging and fractionation evidence for a predominantly nuclear pool. These directly inspected human experiments support the broad location without attributing the independent methods to the original citation or claiming exclusive nuclear residence. The endogenous signals do not distinguish the two natural products. |
| GO:0005634 nucleus | IDA PMID:37342232 Loss of FAM111B protease mutated in hereditary fibrosing poi... | ACCEPT | Summary: Endogenous human FAM111B localizes predominantly to the nucleus. Reason: PMID:37342232 Results 2.1, Figure 1A–C and the immunofluorescence/fractionation methods show endogenous FAM111B in human U2OS nuclei, with no nucleolar staining reported by the authors, with additional positive cytoplasmic staining. Broad nuclear residence is retained without equating it with a particular DNA-repair activity, exclusive localization, or product-specific expression. The antibody localization and human cDNA constructs do not establish equivalent behavior of both natural products. Supporting Evidence: |
| GO:0005652 nuclear lamina | IDA PMID:36610347 FAM111B dysregulation promotes malignancy in fibrosarcoma an... | KEEP AS NON CORE | Summary: A smaller endogenous FAM111B pool is associated with the nuclear lamina region. Reason: The original PMID:36610347 main remains unavailable; its abstract alone does not resolve the nuclear-lamina assay. Independent PMID:37342232 Results 2.1, Figure 1C and detergent-extraction microscopy show endogenous human U2OS FAM111B at the nuclear periphery alongside Lamin A/C, supporting the retained lamina-region location. The authors find NPC associations rather than a lamin-containing FAM111B complex, so this is spatial localization, not a claim of direct lamin binding or lamina structural-subunit activity. Mutant and engineered-fragment enrichment is kept separate from the observed endogenous pool; no natural-product equivalence is inferred. Retain this positive spatial observation as non-core: protease-dependent rescue of ultrafine bridges in Results 2.4/Figure 3I does not locate the rescuing reaction at the lamina. Nuclear-pore cleavage at that site remains a Discussion working model, and normal Lamin B1 with only a small Lamin A/C ratio change does not identify lamins as substrates. The catalytic core therefore retains the broader, well-supported nuclear location without assigning its reaction to the peripheral pool. |
| GO:0005737 cytoplasm | IDA PMID:36610347 FAM111B dysregulation promotes malignancy in fibrosarcoma an... | KEEP AS NON CORE | Summary: Positive human cytoplasmic FAM111B localization is retained as a secondary pool. Reason: The main article for PMID:36610347 was not recovered, so its source-specific cytoplasmic methods cannot be verified from the abstract. Independent PMID:37342232 Results 2.1 and Figure 1A explicitly show endogenous FAM111B in both cytoplasm and nucleus in human U2OS cells, with the majority nuclear. Retain the positive cytoplasmic location as non-core rather than infer absence from predominant nuclear staining. This does not establish a separate cytoplasmic molecular activity or identify which natural product produces the signal. |
| GO:0005737 cytoplasm | IDA PMID:37342232 Loss of FAM111B protease mutated in hereditary fibrosing poi... | KEEP AS NON CORE | Summary: Endogenous FAM111B also occupies a cytoplasmic pool. Reason: PMID:37342232 Results 2.1 and Figure 1A report positive cytoplasmic as well as predominantly nuclear endogenous FAM111B staining in human U2OS cells. The secondary cytoplasmic pool is retained as non-core; predominant nuclear residence is not evidence against it. Fractionation separates soluble and insoluble material and is not by itself an isoform-specific cytoplasm assay. No cytoplasmic catalytic mechanism or equality of the 734- and 704-residue natural products is inferred. Supporting Evidence: |
| GO:0008236 serine-type peptidase activity | IBA GO_REF:0000033 | MODIFY | Summary: Refine the sound inherited serine peptidase activity to serine-type endopeptidase activity on the basis of controlled internal self-cleavage. Reason: The exact target sequence and serine-peptidase IBD are verified at PTN002683271; the human target is a valid experimental descendant source. GO:0004252 is an is_a descendant of GO:0008236 and specifies internal alpha-peptide-bond hydrolysis. In PMID:32776417, purified full-length human FAM111B produced in yeast undergoes internal fragmentation, whereas the catalytic D544N preparation does not; the time/temperature experiment and AEBSF inhibition provide additional controls (Figure 1H and Figure EV1E,G). These results support endopeptidase specificity rather than a generic peptidase assignment. The Figure 1 and EV1 images were inspected qualitatively; no kinetic parameters or exact cleavage-site sequence are inferred. The proposed FAM111B F438 site in PMID:37342232 is based on homology and fragment size, distinct from the experimentally sequenced FAM111A F334 site. No physiological trans-substrate, TOP1 cleavage, or activity equivalence of the natural 704-residue product is established by these assays. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN002683271 SUPPORTS TRANSFER The saved v19 path and target-supported IBD are correct. Controlled human FAM111B internal cleavage permits a narrower molecular-function term without alleging a propagation failure; NO_FAILURE_CORE records the sound inherited activity. Proposed replacements: serine-type endopeptidase activity |
| GO:0008236 serine-type peptidase activity | IDA PMID:32776417 FAM111 protease activity undermines cellular fitness and is ... | MODIFY | Summary: Purified human FAM111B performs controlled serine-type internal proteolysis. Reason: PMID:32776417 Results “Human FAM111A and FAM111B are active proteases,” Figure 1H, Figure EV1E/G and the recombinant-protein purification methods directly test full-length human FLAG-FAM111B expressed in yeast JEL-1. Incubation produces internal autocleavage fragments; the D544N catalytic-triad mutant and AEBSF controls support serine-protease catalysis. The inspected Figure 1A identifies the 734-residue experimental protein, while the historical insert was not independently resequenced here. Refine broad serine-type peptidase activity to serine-type endopeptidase activity, whose current definition specifies internal peptide-bond hydrolysis. Patient-mutant hyperactivity and cleavage fragments are not natural splice products; the 704-residue product and direct cleavage of RFC1, RPB1 or other candidate cellular substrates are not established by these assays. Proposed replacements: serine-type endopeptidase activity Supporting Evidence: |
| GO:0031297 replication fork processing | IBA GO_REF:0000033 | UNDECIDED | Summary: Verified inheritance of the fork-processing assertion is retained, but target-specific participation in restoring stalled replication forks is unresolved. Reason: The source IBD at PTN002683271 and the exact target path are verified. PMID:32165630 supports the FAM111A donor through replication-track progression under camptothecin or PARP-trapping stress, with rescue dependent on protease activity, PCNA interaction and ssDNA binding. GO:0031297 specifically concerns restoration and restart of a stalled replication fork. PMID:32776417 Results on hereditary fibrosing poikiloderma reports that FAM111B lacks a recognizable PIP-box and is not enriched at DNA replication sites (Figure EV5A). Its depletion/knockout experiments also find normal basal DNA replication, PCNA chromatin loading and proliferation (Appendix Figure S2 as described in the main Results). Crucially, the paper reports the same basal dispensability for FAM111A, whose obstacle-dependent activity is independently established; this negative alone cannot refute the inherited stress-dependent claim. Only patient-associated FAM111B alleles, not wild type, produce the reported shutdown of DNA synthesis/transcription and associated toxicity. These mutant gain-of-activity phenotypes are not evidence that wild-type FAM111B normally performs the donor function. PMID:37342232 Results 2.4 explicitly concludes that FAM111B is dispensable for DNA repair in the tested setting: MCF7 knockouts were not more sensitive to the tested replication/transcription inhibitors and DNA-damaging agents in a 72-hour resazurin assay. U2OS knockout growth defects prevented the same comparison in U2OS. Those survival endpoints do not directly measure fork restart. Positive U2OS mitotic DNA synthesis/ultrafine-bridge and telomere phenotypes, including wild-type versus protease-dead rescue, establish a protease-dependent genome-integrity role without identifying the stalled-fork restoration step. The target evidence warrants caution about functional divergence while leaving this specific inherited assertion UNDECIDED; it neither proves the donor mechanism in FAM111B nor establishes its absence under every obstacle or cell context. Propagation Review Root cause: UNRESOLVED Sources checked: PANTHER:PTN002683271 UNRESOLVED Target descent from PTN002683271 and the actual FAM111A-derived fork-processing IBD are verified. Donor stressed-fork evidence is sound; target assays address partly different endpoints and suggest functional differences without demonstrating universal loss. Functional applicability, not tree placement, remains unresolved. |
| GO:0106300 protein-DNA covalent cross-linking repair | IBA GO_REF:0000033 | UNDECIDED | Summary: The ancestral DNA-protein crosslink-repair assertion is real, but removal of these crosslinks by human FAM111B has not been resolved by the inspected target evidence. Reason: The matching IBD and sequence-resolved target descent from PTN002683271 are verified. PMID:32165630 supplies positive FAM111A donor evidence from TOP1 cleavage-complex accumulation and wild-type versus catalytic/PCNA-interaction mutant rescue, together with obstacle-specific fork assays. GO:0106300 concerns removal of a covalent DNA-protein crosslink. PMID:32776417 Results on hereditary fibrosing poikiloderma reports that FAM111B lacks a recognizable PIP-box and is not enriched at DNA replication sites (Figure EV5A). Its depletion/knockout experiments also find normal basal DNA replication, PCNA chromatin loading and proliferation (Appendix Figure S2 as described in the main Results). Crucially, the paper reports the same basal dispensability for FAM111A, whose obstacle-dependent activity is independently established; this negative alone cannot refute the inherited stress-dependent claim. Only patient-associated FAM111B alleles, not wild type, produce the reported shutdown of DNA synthesis/transcription and associated toxicity. These mutant gain-of-activity phenotypes are not evidence that wild-type FAM111B normally performs the donor function. PMID:37342232 Results 2.4 and Discussion report no extra MCF7 knockout sensitivity to the tested agents, including formaldehyde and camptothecin, and distinguish the target from FAM111A. This is a 72-hour metabolic-survival comparison; U2OS growth defects prevented the equivalent test there. It is genuine contextual negative evidence, but it does not measure crosslink abundance or clearance. The positive genome/telomere phenotypes and purified protease activity do not supply that missing target-specific step. Retain UNDECIDED while preserving valid ancestry and the balanced target evidence. No direct FAM111B cleavage of TOP1, trapped PARP or a nuclear-pore protein, and no demonstrated universal ancestral loss, is asserted. Propagation Review Root cause: UNRESOLVED Sources checked: PANTHER:PTN002683271 UNRESOLVED The exact target descends from the verified IBD node; the donor crosslink-repair evidence is positive. Target-specific repair remains uncertain because the inspected experiments measure proteolysis, genome integrity and context-limited drug survival rather than adduct removal. No source error or demonstrated ancestral loss is asserted. |
| GO:0016540 protein autoprocessing | IDA PMID:32776417 FAM111 protease activity undermines cellular fitness and is ... | NEW | Summary: Purified wild-type human FAM111B directly performs controlled self-cleavage. Reason: PMID:32776417 Results Human FAM111A and FAM111B are active proteases, Figure 1H, Figure EV1E/G and Purification of recombinant FAM111 proteins describe full-length human FLAG-FAM111B purified from yeast JEL-1. Fragment formation in the wild-type preparation is lost with catalytic D544N and inhibited by AEBSF, directly identifying FAM111B as the performer rather than only a substrate or a required gene. This supports protein autoprocessing under the tested biochemical conditions. The current term includes processing carried out by the protein itself; it does not require proof that the observed cleavage is an obligatory physiological maturation step. Same-role annotation checks find this term on human FAM111A and HTRA2, while CAPN1 lacks an exact-term record in the bounded query; the proposal rests on the target assay rather than assumed pathway completeness. No existing FAM111B process assertion is an ancestor or descendant of this term, and the fixed-head GO-CAM index has no target entry. The assay does not resolve intramolecular versus intermolecular cleavage, physiological regulation, exact cleavage-site sequencing or equivalence of the natural 734- and 704-residue products. Supporting Evidence: |
Loading supporting content…
Download this section (compressed HTML)Q: Which endogenous substrates are directly cleaved by FAM111B, and how do those reactions connect controlled protease activity to nuclear architecture, chromosome stability and telomere maintenance?
Q: Does FAM111B participate directly in chromatin-associated fork recovery or DNA–protein crosslink removal under particular conditions, or do the observed genome-integrity effects arise through a distinct nuclear-peripheral mechanism?
Q: How do the natural 734- and 704-residue products differ in abundance, localization, protease regulation and contribution to cellular phenotypes?
Experiment: Combine endogenous-level wild-type and catalytic-inactive FAM111B replacement with cleavage-site proteomics in matched human cells. Reconstitute cleavage of candidate substrates with purified human FAM111B, catalytic controls and inhibitor controls, measuring mapped peptide products separately from binding, protein abundance and nuclear or telomere phenotypes. Include self-cleavage as an assay control rather than assuming it identifies the physiological reaction.
Hypothesis: A direct substrate should yield reproducible FAM111B-dependent cleavage products in a defined system and show a mechanistically linked cellular phenotype; changes in abundance or co-association alone may instead reflect indirect effects.
Experiment: Compare matched human cell backgrounds with controlled FAM111B depletion and endogenous-level rescue, balancing basal proliferation and cell-cycle state. Measure chromatin-specific association, replication-track recovery and covalent DNA–protein adduct removal directly alongside nuclear-peripheral organization and the existing metabolic-survival endpoint. Use FAM111A perturbation as a mechanistic comparator without substituting its substrates or recruitment mechanism for those of FAM111B.
Hypothesis: A condition-specific fork or adduct-removal role would require a positive direct endpoint beyond drug survival; dissociation of those endpoints from peripheral or telomere defects would support a distinct mechanism without implying universal absence of an ancestral function.
Experiment: Establish product-specific endogenous expression with validated transcript and protein measurements, then compare matched levels of sequence-verified 734- and 704-residue products in localization, controlled protease and rescue assays. Distinguish natural products from engineered fragments, tags and hyperactive disease mutants, and control for differences in protein stability.
Hypothesis: The 30-residue N-terminal difference may alter regulation or distribution without requiring a different catalytic specificity; equivalent behavior is an outcome to test, not a premise of the experiment.
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)