Tauopathies such as Alzheimer's disease (AD) feature progressive intraneuronal deposition of aggregated tau protein. The cause is unknown, but in experimental systems trans-cellular propagation of tau pathology resembles prion pathogenesis. Tau aggregate inoculation into mice produces transmissible pathology, and tau forms distinct strains, i.e. conformers that faithfully replicate and create predictable patterns of pathology in vivo. The prion model predicts that tau seed formation will anticipate neurofibrillary tau pathology. To test this idea requires simultaneous assessment of seed titer and immunohistochemistry (IHC) of brain tissue, but it is unknown whether tau seed titer can be determined in formaldehyde-fixed tissue. We have previously created a cellular biosensor system that uses flow cytometry to quantify induced tau aggregation and thus determine seed titer. In unfixed tissue from PS19 tauopathy mice that express 1 N,4R tau (P301S), we have measured tau seeding activity that precedes the first observable histopathology by many months. Additionally, in fresh frozen tissue from human AD subjects at early to mid-neurofibrillary tangle stages (NFT I-IV), we have observed tau seeding activity in cortical regions predicted to lack neurofibrillary pathology. However, we could not directly compare the same regions by IHC and seeding activity in either case. We now describe a protocol to extract and measure tau seeding activity from small volumes (.04 mm
) of formaldehyde-fixed tissue immediately adjacent to that used for IHC. We validated this method with the PS19 transgenic mouse model, and easily observed seeding well before the development of phospho-tau pathology. We also accurately isolated two tau strains, DS9 and DS10, from fixed brain tissues in mice. Finally, we have observed robust seeding activity in fixed AD brain, but not controls. The successful coupling of classical IHC with seeding and strain detection should enable detailed study of banked brain tissue in AD and other tauopathies.