Extended Data Fig. 6: Benchmarking optimized conventional antibody staining with or without tissue clearing versus ThICK staining with SPEARs in SHIELD-protected samples.

a–d, 2 mm-thick SHIELD-protected brain slice immunostained with antibodies or SPEARs targeting phospho-S6 ribosomal protein (pSer244, pSer247) (Phospho-S6), a neuronal activity marker. a, Schematic of benchmarking experiment. The imaging and tissue processing parameters were kept identical to ensure comparability. b, 3D volume rendering of full-thickness images. Note that due to BABB clearing the sample thickness was reduced by ~60%. Experiment performed once. c, Excerpted neurons from the 3 indicated levels in b. d, Quantified cell intensities across the full thickness of the samples. Data are presented as mean values ± S.D. e–i, SHIELD-protected mouse hemibrains immunostained with antibodies or SPEARs targeting choline acetyltransferase (ChAT) at various dilutions. e, Schematic of benchmarking experiment. The imaging and tissue processing parameters were kept identical to ensure comparability. f, Appearances of the pedunculopontine nuclei (ppn) in various samples, which is located ~300 𝜇m from the tissue surface in a hemi-sectioned brain with low local ChAT fiber density. Grouped comparative experiment performed once. g, Appearances of the anterior commissure (ac) in various samples, which is in contact with the tissue surface. Note that the presence of positive stainings along the tissue surface suggest specific staining of ChAT-positive fibers in all cases but with varying penetration depths. Yellow arrows indicate a line of interest (LOI) along which the direction of signal intensity was repeatedly measured and shown in h. h, Averaged signals measured along 10 pixels-wide LOIs and were repeated 5 times per sample. Data are presented as mean values ± S.D. i, Signal intensities at the deeper portion (200–400 m 𝜇m) of the visualized anterior commissures, as sampled from the depth range in the yellow boxes in h. Whiskers: minima and maxima, bounds of box: 75th and 25th percentiles, centers: medians. P < 0.0001, two-sided Brown–Forsythe ANOVA test. ****: P < 0.0001, Tamhane’s T2 multiple comparisons test with 4 comparisons per family. j–l, iDISCO-processed mouse hemibrains immunostained with antibodies or SPEARs targeting tyrosine hydroxylase (TH). j, Schematic of benchmarking procedure. The imaging and tissue processing parameters were kept identical to ensure comparability. k, 600-𝜇m-thick maximum intensity projections along the sagittal direction of the stained mouse hemibrains. Somas of the tyrosine hydroxylase (TH)-positive neurons are clearly visible in the TH SPEAR ThICK-stained sample due to homogeneous staining along the z-direction, whereas TH antibodies in the control sample are mostly deposited as bright areas along the mid-sagittal surface of the hemibrain, impeding the visualization of deeper TH-expressing neurons. l, Enlarged view in the boxed areas in k, showing TH-positive neurons located in the hypothalamus (group A15). z-depths were sampled starting from the tissue surface. Experiment performed once.