*2.2. Acetylation Changes Tau Phase Behavior*

LLPS has recently been observed for wt and hyperphosphorylated Tau, and truncation mutant (K18) [35–38]. At low salt conditions (5 mM sodium phosphate, pH 7.8), we observed near-instantaneous formation of wt Tau droplets (Figure 2A). Subsequent fusions indicate the liquid nature of the wt Tau droplets. We characterized the protein concentration (2.5–20 μM) and salt concentration (0–250 mM NaCl) dependencies of wt Tau LLPS. Consistent with the literature [37,38], higher salt concentrations disfavor LLPS and higher protein concentrations favor LLPS (Figure 2E). For the case of Ac-Tau, we observed a dramatic reduction in droplet formation (Figure 2B,F). Similar results were also observed when LLPS experiments were performed with wt Tau or Ac-Tau in the presence of a crowding agent (10% PEG 8K, 200 mM NaCl, 10 mM acetate, 10 mM glycine, 10 mM sodium phosphate, pH 7.5; Figure 2C,D). Thus, independent of the presence or absence of crowding, the hyperacetylation of Tau disfavors LLPS.

**Figure 2.** Acetylation disfavors Tau liquid-liquid phase separation (LLPS). Droplet formation of wt Tau versus hyperacetylated (Ac-Tau), respectively, in the absence (**A**,**B**) and presence of the crowding agent PEG-8K (**C**,**D**). Phase transition maps of Tau versus Ac-Tau (in the absence of crowders), respectively (**E**,**F**). The diagrams present protein and salt concentration dependencies of droplet formation. Blue open circles represent conditions of minimal droplet formation (<5 droplets/frame); small red solid circles represent 5–20 droplets, medium red solid circle 20–100 droplets, and large red circle >100 droplets. The bars in (**A**–**D**) represent 10 μm. See Section 3 for details.

Interestingly, even though both hyperphosphorylation and hyperacetylation decrease the overall pI of Tau, the two PTMs seem to have opposite effects on LLPS. In contrast to LLPS enhancement by hyperphosphorylation [36,37], hyperacetylation clearly disfavors Tau LLPS (Figure 2). Further experiments performed in identical or comparable conditions using the same Tau constructs, full-length or otherwise, are needed for a clear and direct comparison of LLPS behaviors of hyperphosphorylated, hyperacetylated, and wt Tau proteins. Nevertheless, we think that hyperacetylation disfavors full-length Tau LLPS by neutralizing the lysine positive charges, thereby affecting opposite-charge

attractions that help support Tau self- and mesoscale interactions. Our data also give direct support that electrostatics plays a major role in Tau LLPS.
