Fig 1.
ProSAAS is expressed in multiple cell types in a mixed species meta-analysis.
Log-transformed transcriptional counts of proSAAS (Red); clusterin (CLU, Blue); and αcrystallin-β (CRYAB, Green) transcripts in human and mouse studies obtained from ARCHS4, were normalized using the ComBat batch correction algorithm. Expression distributions are represented as a continuous density function, with vertical lines representing the means of each distribution. Panel A: Human: n = 2737 samples, 79 batches; Panel B: Mouse: n = 3045 samples, 155 batches).
Fig 2.
ProSAAS is expressed in multiple cell types in the human and rodent retina. proSAAS transcript expression in retinal cell types of: Panels A-C: Human; Panels D-F: Mouse.
Panels A, B: Differential cell expression plots. Panels B-C and E-F: UMAP projections and differential cell expression plots. Specific cell types are indicated by arrows in panels B and E, and by color in panels C and F. Relative proSAAS transcript expression is greatest in horizontal cells in both species, followed by RGCs. RGC subgroup expression is identifiable in the mouse data (seen as orange in panel F). Amacrine cell interneurons are subdivided into different types in the human study (panel A), and as a single cell type in the mouse study (panel D). Panel G depicts a comparison of relative Pcsk1n mRNA expression in different RGC subtypes [data from [24,30]], while Panel H shows a comparison of changes in retinal PCSK1N expression in five disorders affecting the retina.
Fig 3.
ProSAAS is enriched in the GCL and IPL mouse retinal layers and is present in synapses and retinal nerve tissue.
Panel A: ProSAAS-ir (LS46 staining) and RBPMS-ir co-staining showing proSAAS-ir surrounding RGCs. Panel B: Quantitative analysis of proSAAS-ir in each retinal layer, derived from 28 total images of retinal slices (n = 22 for LS46 antiserum-stained images; n = 6 for proSAAS ALS16157 antiserum-stained images; the choice of proSAAS antiserum was dictated by the available marker primary antibodies. Data are reported as corrected total fluorescence (CTF) normalized to the proSAAS-ir of the no-primary control in each experiment. Panel C: Stitched tilescan image of the optic nerve, optic nerve head, and surrounding retinal layers, depicting proSAAS LS46 and IIIβ-tubulin immunostaining. (* p < 0.05, *** p < 0.0001, **** p < 0.00001, unpaired t-test). (GCL = Ganglion Cell Layer, IPL = Inner Plexiform Layer, INL = Inner Nuclear layer, OPL = Outer Plexiform Layer, ONL = Outer Nuclear Layer, PR = Photoreceptor Layer).
Fig 4.
ProSAAS is found in RGCs, amacrine, and horizontal cells in the mouse retina.
Panel A: ProSAAS and ChAT co-staining indicates that proSAAS is in both normal and displaced ChAT-positive amacrine cells, as well as in ChAT-positive synaptic layers in the IPL. Panel B: ProSAAS and GFAP co-staining depicts limited proSAAS staining in astrocytes. Panel C: ProSAAS and calbindin co-staining shows that proSAAS-ir is enriched in the cell bodies of calbindin-positive horizontal cells. Panel D: ProSAAS and Brn3a co-staining depicts proSAAS in and surrounding Brn3a-positive retinal ganglion cells. Panels A-D: All images represent maximum projection micrographs of fixed, cross-sectioned retina probed with proSAAS and cell-type marker antibodies. Brightness levels were manually adjusted for each experiment to increase contrast. LS46 proSAAS antiserum was used in this experiment.
Fig 5.
Image quantitation of RGCs reveals differences in mouse proSAAS distribution as compared to bioinformatics analysis.
Panel A: Quantitative image analysis of retinal cross sections labeled for proSAAS (LS46 antiserum) in cell-marker positive cells (GFAP, n = 4; calbindin, n = 3; CHaT, n = 6; RBPMS n = 4; Brn3a n = 4). Panel B: Quantitative image analysis of 6 retinal cross-sections, colabeled with proSAAS, Brn3a, and melanopsin antisera, showing no significant differences. Panel C: Representative image of proSAAS, melanopsin, and Brn3a staining. (ns = not significant, * p < 0.05, ** p < 0.01, *** p < 0.0001, **** p < 0.00001, using one-way ANOVA with Tukey’s post-test for Panel A; and unpaired Student’s t-test for Panel B.). For antiserum compatibility purposes, Abcepta ALS16157 goat anti-proSAAS antibody was used for the experiments in Panels B and C.
Fig 6.
Retinal proSAAS-immunoreactive species differ from those found in hippocampus.
Panel A: ProSAAS processing diagram depicting paired basic proprotein convertase cleavage sites (blue vertical lines); epitopes for the three proSAAS antisera used in this study (red, dark green, and light green horizontal lines); and theoretical sizes of each proSAAS fragment derived from intact 27 kDa proSAAS by proprotein convertase cleavage. Panel B: Western blotting of mouse retinal and hippocampal (hipp) acid extracts using LS46 antiserum. Panel C: Western blotting of retinal lysates using proSAAS antibodies LS46 and 1E9. Each lane represents pooled retinal or hippocampal extracts taken from an individual animal. Panel D: Western blotting of a human retinal extract (retina) using LS46 antiserum.