# HISTORY 24 Mar 2016: Updated by: TOUCHUP-v1.15 14 Mar 2016: Updated by: TOUCHUP-v1.12 # molecular_function 20100629: Eukaryota_PTN000024429 has function endopeptidase activity (GO:0004175) 20100621: Euteleostomi_PTN000024458 has function cadherin binding (GO:0045296) 20140530: Euteleostomi_PTN000024458 has function beta-catenin binding (GO:0008013) # cellular_component 20100629: Eukaryota_PTN000024429 is found in integral component of plasma membrane (GO:0005887) 20140530: Eukaryota_PTN000024429 is found in nucleus (GO:0005634) 20140530: Euteleostomi_PTN000024458 is found in endoplasmic reticulum (GO:0005783) 20100623: Bilateria_PTN000024431 is found in neuronal cell body (GO:0043025) 20100623: Bilateria_PTN000024431 is found in Z disc (GO:0030018) 20100623: Bilateria_PTN000024431 is found in perinuclear region of cytoplasm (GO:0048471) 20100623: Bilateria_PTN000024431 is found in apical plasma membrane (GO:0016324) 20100623: Bilateria_PTN000024431 is found in cell cortex (GO:0005938) # biological_process 20140530: Euteleostomi_PTN000024458 participates in canonical Wnt signaling pathway (GO:0060070) 20100630: Euteleostomi_PTN000024458 participates in smooth endoplasmic reticulum calcium ion homeostasis (GO:0051563) 20100630: Bilateria_PTN000024431 participates in negative regulation of apoptotic process (GO:0043066) 20100623: Bilateria_PTN000024431 participates in membrane protein ectodomain proteolysis (GO:0006509) 20100623: Bilateria_PTN000024431 participates in amyloid precursor protein catabolic process (GO:0042987) 20100630: Bilateria_PTN000024431 participates in calcium ion transport (GO:0006816) 20100623: Bilateria_PTN000024431 participates in beta-amyloid metabolic process (GO:0050435) 20100623: Bilateria_PTN000024431 participates in Notch receptor processing (GO:0007220) # WARNINGS - THE FOLLOWING HAVE BEEN REMOVED FOR THE REASONS NOTED # NOTES . Description of phylogeny This family consists primarily of 2 paralogous vertebrate clades, PSEN1 and PSEN2. There is a worm/fly outgroup, plus unannotated outgroups with dicty and plant proteins. Two worm proteins, hop-1 and spe-4, appear very divergent; in P-POD, they are in the same Jaccard family, but not the same OrthoMCL family, as the others. Molecular Function -The "catalytic aspartate" mentioned in the evolution paper PMID 11978763 is at position 399 in this alignment; it's present in all sequences except ENSMMUP00000024373, which seems to have a 25-30 aa gap starting there. A second catalytic aspartate is at position 668, which is conserved in all sequences except ENSMMUP00000013302, which has a large gap spanning that region. -"endopeptidase activity" is found in both the PSEN1 and -2 clades. Due to the sequence conservation mentioned above, propagate to AN0. -Propagate "cadherin binding" within PSEN1 clade. Cellular Component -There are lots of "membrane fraction" annotations, which we won't propagate because they are biochemical designations, but they are consistent with membrane-spanning proteins, such as these. The most specific term is GO:0005887 "integral to plasma membrane." Due to conservation of membrane-spanning regions, propagate to AN0. -Similarly, "synaptosome" is also a biochemical designation, but it's consistent with "neuromuscular junction," which is a child of "synapse." Propagate "neuromuscular junction" to AN2. -Propagate neuro terms to metazoa (neuronal cell body, axon, dendritic shaft, neuromuscular junction, growth cone). Same for "Z disc" (part of a myofibril) and "ciliary rootlet." These can't be propagated to dicty or plants. -Do not propagate "protein complex," which is not specific enough. -PMID 9298903 (cited in OMIM 104311) says that observation of PSEN1 & 2 in ER and Golgi is expected because that's where they are synthesized. Do not propagate these terms (or their parents). However, mouse Psen1 has 5 annotations to the BP "endoplasmic reticulum calcium ion homeostasis," so propagate ER to PSEN1 clade. -There are several "cell surface"-like terms: cell surface, apical plasma membrane, cell cortex. Propagate these to AN2 and look for evidence in dicty and plants. -Remaining terms (lysosomal membrane, membrane raft, mitochondrial inner membrane, nucleus, perinuclear region of cytoplasm): Propagate to AN2 and look for evidence in dicty and plants. Biological Process -The vast majority of these annotations are IMP's or IGI's, and the vast majority of those are developmental terms. Since it's not clear how directly PSEN proteins affect development, these should be given lower priority. -Highest priority should go to protein processing and maturation terms (using the endopeptidase activity of the PSEN's), followed by signaling pathways; however, signaling pathways may need "regulation" terms rather than direct annotation. -Propagate "Notch receptor processing" to AN2. Note that there is a possible problem with the definition of "Notch receptor processing" (see SF# 3020333) and this may have to be changed. -Calcium: There are annotations to "calcium ion transport" (Psen2, PMID 16204356) and "endoplasmic reticulum calcium ion homeostasis" (Psen1, multiple annotations). These terms are not related in the ontology. Propagate ER Ca+2 ion homeostasis to the PSEN1 clade and Ca+2 ion transport to AN2. -There are "apoptosis" IDA's in both the PSEN1 and -2 clades, but the papers seem to demonstrate anti-apoptosis instead. There are IDAs to GO:0043066 "negative regulation of apoptosis" (Psen2) and GO:0006916 "anti-apoptosis" (rat Psen1). Propagate 6916 to AN2. Questions for MOD curators: Phylogeny -Worm: Do hop-1 and spe-4 belong in this family? The branch length seems pretty long, especially compared to sel-12. Some "NOT" annotations would be helpful. WB reply, Jul 7, 2010, at 2:10 PM: In the worm literature, hop-1 and sel-12 are described as presenilins and spe-4 also as a presenilin, although divergent. If we encounter any experimental evidence that spe-4 (or hop-1) have been demonstrated to have lost any presenilin-like functions, we can annotate using "NOT". MF -All: Is there justification for an annotation to GO:0004190 "aspartic-type endopeptidase activity"? See PMID 19278647 section 2 on catalysis of/by presenilin. TAIR, July 12, 2010 8:32:30 PM EDT: Not in Arabidopsis. ZFIN, July 13, 2010 7:12:41 PM EDT: I Don't see any for zebrafish presenilins CC -All: There should be an annotation to GO:0070765 "gamma-secretase complex," whose definition includes that it contains presenilin. Also look for evidence supporting GO:0070764 "gamma-secretase-Delta1 complex." TAIR, July 12, 2010 8:32:30 PM EDT: Can't find any. ZFIN, July 13, 2010 7:12:41 PM EDT: I only see ancillary evidence like the phenotype of DAPT (a "gamma secretase complex inhibitor") treated fish is similar to phenotypes observed in fish with mutated or knocked down presenilin...nothing more direct. PMID: 9632714 seems to provide some subcellular localization data for Human PS1 in a complex with beta-catenin. Human (EBI) replies, July 14, 2010 7:08:17 AM EDT: I have annotated PS1 as binding to beta-catenin. In addition there is evidence for localisation of PS1 in rough and smooth ER and the Golgi, which I have also now annotated. [ZFIN continues] PMID: 10801983 seems to provide data supporting an IDA annotation for Human PS1 to 'gamma-secretase complex"...using HeLa cells, they state in the abstract: "Upon gel exclusion chromatography, solubilized gamma-secretase activity coelutes with presenilin 1 (PS1) at an apparent relative molecular weight of approximately 2.0 x 10(6). Anti-PS1 antibody immunoprecipitates gamma-secretase activity from the solubilized gamma-secretase preparation. These data suggest that gamma-secretase activity is catalyzed by a PS1-containing macromolecular complex." Human (EBI) replies, July 14, 2010 7:08:17 AM EDT: I have made the annotation to gamma secretase complex. -Worm: Can the spe-4 "intracellular membrane-bound organelle" annotation be refined? The title of the paper begins "The presenilin protein family member SPE-4 localizes to an ER/Golgi derived organelle..." WB reply, Jul 7, 2010, at 2:10 PM: This ER/Golgi-derived organelle is called the fibrous body/membranous organelle (FB-MO) and is a sperm-specific organelle. I don't know how many gene products are currently localized to this organelle, but perhaps we could request a new, more specific term. Ranjana has been doing some annotation work on spermatogenesis and can probably comment further. Ranjana's reply, July 8, 2010 3:14:29 PM EDT: Yes, I am working on worm spermatogenesis for the gene ontology and as I recall, fibrous body/membranous organelle (FB-MO) formation is an important step in worm spermatogenesis, so I will be requesting both process and cellular component terms. -Mouse Psen2: Is "cytosol" correct? For an integral membrane protein? MGI replies, July 14, 2010 10:52:30 AM EDT: The annotation is correct. Presenilin was found in the cytosol in that paper. PAINT (MSL) responds, July 14, 2010 11:57:22 AM EDT: Actually, I think that paper (PMID 9535056) refers to a PS2-like protein, not to PS2 itself. What do you think? It doesn't make any sense for a protein with multiple membrane-spanning regions to be found in the cytosol, which has no membranes. If anything, it would have to be cytoplasm instead. -Human: OMIM 104311 and 600759 mention several CC annotations like kinetochore and centrosome. Human (EBI) replies, July 29, 2010 7:53:51 AM EDT: I have made annotations to both kinetochore and centrosome for PSEN1 and 2 from PMID:9298903 BP -dicty and Arabidopsis: Does your organism contain Notch? Is it activated by proteolysis? TAIR, July 12, 2010 8:32:30 PM EDT: Nope, no Notch in Arabidopsis. Neatly enough, there is an article called "Arabidopsis genome:Life without Notch ", from which these sentences come: "Plants appear to lack several of the most widely adopted signalling pathways found in vertebrates, flies and worms, such as the Notch, Wnt and Hedgehog pathways, or the various receptor tyrosine kinase�Ras pathways. Instead, Arabidopsis makes use of signal transduction pathways comprising different modules." (PMID: 11231172) Just for kicks, since that article was so 2001, I did a blast with Drosophila Notch on the current release of the Arab genome (TAIR9) and still got nothing. -All organisms, especially those with annotations to "Notch receptor processing," "protein processing," or "protein maturation by peptide bond cleavage": Please consider annotations to GO:0033619 "membrane protein proteolysis," GO:0035333 "Notch receptor processing, ligand-dependent," and/or GO:0035334 "Notch receptor processing, ligand-independent." ZFIN, July 13, 2010 7:12:41 PM EDT: I don't see the proper experiment that encapsulates proteolysis of a membrane protein unless I pull in knowledge from outside the paper that the notch receptor or APP is a transmembrane protein. Maybe that is an acceptable insertion of curator judgement...? -All organisms: What do presenilins do to/with calcium? ZFIN, July 13, 2010 7:12:41 PM EDT: I see no data in this regard in the zebrafish literature, but PMID: 16959576 appears to show that Human PS1 and PS2 can form Ca2+ channels in the ER. Not sure if that has been curated yet by anyone. Looking in AmiGO, I see similar annotations have been curated for Human Presenilin from other pubs. Human (EBI) replies, July 14, 2010 7:08:17 AM EDT: I have annotated to 'endoplasmic reticulum calcium ion homeostasis' and 'calcium channel activity' for human PS1 from this paper. -Mouse and rat: Should any of the annotations to GO:0006915 "apoptosis" or GO:0043066 "negative regulation of apoptosis" actually be to GO:0006916 "anti-apoptosis"? MGI replies, July 14, 2010 10:52:30 AM EDT: "anti-apoptosis" is a better term, change made in MGI. Questions for ontology curators: -SF# 3020333 Seeking clarification on GO:7220 Notch receptor processing -SF# 3020371 Protein processing: 16485 vs. 51605 Submitted July 2010, MSL Revised 2010 December 22, MSL -Added catenin binding and wnt pathway to PSEN1 clade. -propagated nucleus to the root (it was to the metazoa) HM, updated on May 30, 2014 # REFERENCE Annotation inferences using phylogenetic trees The goal of the GO Reference Genome Project, described in PMID 19578431, is to provide accurate, complete and consistent GO annotations for all genes in twelve model organism genomes. To this end, GO curators are annotating evolutionary trees from the PANTHER database with GO terms describing molecular function, biological process and cellular component. GO terms based on experimental data from the scientific literature are used to annotate ancestral genes in the phylogenetic tree by sequence similarity (ISS), and unannotated descendants of these ancestral genes are inferred to have inherited these same GO annotations by descent. The annotations are done using a tool called PAINT (Phylogenetic Annotation and INference Tool).