Extended Data Fig. 4: Cryo-EM data processing of TTYH2 in detergent in complex with delipidated APOE and conservation of the APOE interaction interface.

a, Representative motion and CTF-corrected micrograph (total micrographs=27,154). b, Data processing workflow. Particles were picked using a previously obtained TTYH2 map for template generation. After several rounds of 2D classification, particles of APOE in complex with TTYH2 were isolated by 3D variability analysis in cryoSPARC and used for the training of a Topaz model for more accurate particle picking. Topaz-picked particles were sorted by 2D classifications and 3D variability analysis, which yielded two similar maps of the TTYH2-APOE complex at 4.1 and 4.0 Å. c, Angular distribution of particle orientations and, d, Fourier shell correlation (FSC) plot of the final refined cryo-EM density map of the TTYH2 in complex with APOE at 4.0 Å with pronounced density of APOE. e, Final 3D reconstruction of the TTYH2 in complex with APOE colored according to its local resolution. f, Sequence conservation of residues from an alignment of 26 mammalian TTYH2 orthologs (left) and the three human paralogs TTYH1-3 (right). Shown is the tip of the extracellular domain with the approximate binding epitope for APOE indicated by a yellow line and residues located in this epitope labeled. g, h, Low resolution 3D reconstructions of a sample containing, g, a TTYH2 quadruple mutant involving residues of the putative APOE binding site (G165P/D166E/Q169R/F173R) and, h, TTYH3 in presence of APOE containing a nanogold label. A slice through the density is shown left. In contrast to the data of TTYH2 prepared under equivalent conditions (Fig. 3a) no evidence for bound APOE is detected.