Tau, alpha-synuclein and TDP-43 as antibody targets

Tau, alpha-synuclein and TDP-43 are the aggregating proteins behind most of the non-amyloid neurodegenerative diseases. All three are normally intracellular, all three form ordered aggregates with disease-specific structures, and all three have been pursued with antibodies. The targeting problem is different for each.

Tau

Tau is a microtubule-associated protein, heavily phosphorylated in disease, that forms filaments whose structures differ between Alzheimer's disease, Pick's disease, corticobasal degeneration and progressive supranuclear palsy. Cryo-electron microscopy of patient material established those differences, and they are the reason a binder raised against one preparation may not recognize another.

Antibody programs have targeted the N-terminus, the mid-domain, phospho-epitopes and the microtubule binding region. The rationale for extracellular targeting is that seeds move between cells, so intercepting them could slow propagation. Several N-terminal and mid-domain antibodies reached clinical trials and did not show clear benefit, which has shifted interest toward the aggregation-competent core and toward conformation-selective binders.

For discovery work the practical consequences are: state which tau species your binder recognizes, monomer, oligomer or filament; state which filament preparation you used; and test on material that resembles disease rather than only on heparin-induced recombinant fibrils.

Alpha-synuclein

Alpha-synuclein is a small presynaptic protein whose aggregates define Parkinson's disease, dementia with Lewy bodies and multiple system atrophy. Its filament folds also differ between those conditions.

Clinical antibodies have mostly targeted the C-terminus or aggregated forms. The trials reported so far have not established clear clinical benefit, which again points at the same open questions: whether the extracellular pool is the right target, whether the antibody reaches it in enough quantity, and whether the epitope is exposed on the species that propagates.

Synuclein is also a useful system for assay work, because seeding assays and biosensor lines are well developed for it and because the protein aggregates reproducibly in vitro.

TDP-43

TDP-43 is a nuclear RNA binding protein that mislocalizes to the cytoplasm and aggregates in most amyotrophic lateral sclerosis and in a large fraction of frontotemporal dementia. It is the hardest of the three to target with an antibody, because the disease biology is dominated by loss of nuclear function and by cytoplasmic aggregates rather than by an extracellular species. Approaches that address function, such as restoring correct splicing, have drawn more attention than extracellular clearance.

For an antibody program, TDP-43 is worth entering with a clear idea of what an antibody would do and where it would do it.

Cross-cutting issues

• Brain exposure. Only a small fraction of a peripherally dosed antibody reaches the central nervous system, which is why transport approaches are an active area.
• Target engagement. Measuring that an antibody engaged its target in brain is harder than measuring binding in a tube, and programs that cannot show engagement cannot interpret a negative result.
• Species and conformation. Monomer is abundant, oligomer is rare and probably relevant, filaments are structured and disease-specific. An antibody that binds all three is measuring something different from one that binds only the aggregate.
• Preclinical models replicate parts of the biology, not the disease. Seeding models show propagation; they do not show clinical benefit.

How we work on them

We build the assays that answer the species question: aggregation kinetics, seeded aggregation in cells, and binding against the specific preparation the program cares about, with the provenance of every preparation recorded alongside the result.