Semantic resources project/PRO/Tau Discussion 07-08-2010

= Notes from a Discussion On Tau Representation, 07/08/2010 =

Attendance By Phone

 * TWD
 * EW
 * GW

Discussion was centered around a set of powerpoint slides created by Gwen to describe the basic biology of Tau, including variants, isoforms, and disease associations:

Slide #1

 * Tau is derived from MAPT, a single-copy gene.

Slide #2
Features of a Tau Representation:
 * List of disease phenotypes
 * Alzheimers.
 * FTDP : frontal temporal dementia with parkinson's.
 * FTDP 17 -- from chromosome-17 encoded Tau

Slide #4

 * 6 isoforms -- all are protein coding.
 * (Later slides will suggest that there are eight, or even nine, isoforms)
 * Some have 3 microtubule binding domains, some have 4 -- these are all in exon 10

Slide #5

 * Full-length isoform shown, with sites of phosphorylation
 * "a" is the location of the phosphorylation sites
 * "b" superimposes the names of the kinases that phosphorylate these sites.
 * Block in the middle is a microtubule binding domain
 * 15 known phosphorylation sites with kinases
 * more phos. means more polar, and therefore more likely to pop off the microtubule.
 * In general, aggregate phosphorlyation is worse, although the phos. sites in the MTBD (box) are more dangerous than those outside

Slide #6

 * One intermediate stage between MT and NFT isn't shown: PHF
 * "paired helical filaments"
 * the intermediate of Tau (not shown) that form when phos.
 * Tau binds to itself, before forming the NFTs -- helical, two-stranded, and they form the intra-cellular tangles.

Slide #7

 * Six isoforms shown again
 * different patterns of exon usage define the isoforms
 * 0N, 1N, and 2N define the usage of exons 2/3
 * 4R, 3R define the pattern of Exon 10 usage. 3R means three MT binding domains, while 4R includes Exon 10, and means that there are 4 MT binding domains (R = "repeat")
 * 0N,3R is the "fetal tau" isoform, expressed only during development.
 * "Adult" Tau is probably not a category we need to represent explicitly

Slide #8

 * Again, identifies isoforms based on exon 2, 3, or 10 content
 * "Tau 69", "74", etc are designations of isoforms based on molecular weight
 * "Tau 69" is sometimes "Tau 68"
 * Some of the names are ambiguous, and they aren't useful as unique identifiers.
 * E10 +/- indicates the exon-10 usage.
 * Isoform correspondence to disease phenotype is noted on the left.

Slide #9

 * Sequence Variants listed at bottom
 * this list is partial at best.
 * phosphorylation sites listed again (with associated Kinases) above
 * Not clear what "PP2A" and "PP2B" mean.

Slide #11

 * We're not clear where the "8" isoforms (not "6" as before) comes from
 * possibly a typo? (-GW)
 * EW looks up Tau in Uniprot, notes that they list 9 isoforms, but one without experimental validation.

Final Notes

 * Normally forms complexes with tubulin, etc. as part of the microtubule complex.
 * When phosphorylated, when forming the PHFs, it has intermediate forms and pulls other proteins off the MT. These can get caught in the NFT (reference?)

Action Items

 * TWD: assemble diagram, OBO representation of
 * isoforms
 * sequence variants
 * phosphorylation sites
 * GW: draw up list of known protein interactions for different isoforms.