Fig 2: Kinetochore layers in human kinetochores are wide and spatially overlap. (a) Layout of computational model used to predict LM appearance of human kinetochores. Randomized distribution of molecules within a specified 3D shape is convolved with wavelength-specific Gaussian point-spread function (PSF). Convolved volumes are then downscaled to the voxel size typical in conventional LM. Predicted LM appearance shown for kinetochores shaped as 300 × 250 nm rectangular prism with 50 nm-wide layers separated by 140 nm. Delta and FWHM values are measured in pseudo-LM volume constructed at 40 nm XY pixels and 200 nm Z-steps. (b) LM images and intensity profiles of a 100 nm multicolour bead shown in green (515 nm) and red (580 nm) colours. FWHM is reported for the longer axis in XY plane. (c) Delta and FWHM values measured in RPE1 cells prepared under the same conditions as IM cells. More than 100 centromeres from greater than 3 cells for each experimental condition. (d–e) LM images (d) and intensity profiles (e) of a human kinetochore with typical values of delta and FWHM. Note that both inner (CenpA-GFP) and outer (Hec1, CenpF) domains are significantly larger than a diffraction-limited spot. Delta and FWHM for CenpF spots cannot be measured in Taxol as their complex shape is not Gaussian.

Fig 3: Expression pattern of transitional cell marker genes. (a) Violin plots showing the expression of Cenpf in collecting duct cells. (b) Representative immunofluorescence staining for Cenpf in Pkd1 heterozygous and homozygous knockout mouse kidneys. Green color labels DBA and red color indicates Cenpf. (c,d) Representative immunofluorescence staining for Col1a1 (c) and fibronectin (d) in Pkd1 heterozygous and homozygous mouse kidneys. Green color labels DBA and red color indicates Col1a1 and fibronectin, respectively.