Journal: bioRxiv

Article Title: Dyrk1a gene dosage controls bipolar cell development and retinal connectivity

doi: 10.64898/2026.03.15.710015

Figure Lengend Snippet: (A – X) Representative immunohistochemical staining of coronal eye sections taken from Lhx2-Cre and Lhx2-Cre:Dyrk1a +/f adult mice (6 – 8 weeks) across four designated dorsoventral quadrants (D1, D2, V3 and V4) spanning the entire retina relative to the optic nerve head. (A – H) All quadrants in the INL of Lhx2-Cre (A – D) animals were comparably populated with bipolar cells (Chx10 + , BC, arrowheads). In contrast, the dorsoventral gradient of reduced INL thickness in Lhx2-Cre:Dyrk1a f/f animals (E – H) was accompanied by an apparent reduction in total bipolar cell number (Chx10 + , BC, arrowheads) which was most prominent in the D1 quadrant (E). (I – P) The INL and GCL of Lhx2-Cre animals (I – L) were densely populated with amacrine cells (INL, Pax6 + , AC, arrowheads) and RGCs (GCL, Pax6 + , RGC, arrowheads) across the four designated dorsoventral quadrants. In contrast, the dorsoventral gradient of reduced INL and GCL thickness in Lhx2-Cre:Dyrk1a f/f (M – P) animals led to a reduction in amacrine and RGC number (Pax6 + , RGC/AC, arrowheads) which was most prominent in the D1 quadrant (M). (Q – X) Horizontal cells (Calbindin + , HC, arrowheads) in the INL of Lhx2-Cre (Q – T) exhibited comparable numbers across the retina. In contrast, the dorsoventral gradient of reduced INL thickness in Lhx2-Cre:Dyrk1a f/f (U – X) animals led to a reduction in horizontal cell number (Calbindin + , HC, arrowheads) which was most prominent in the D1 quadrant (U). Scale bar: (A – X) 25 μm. Abbreviations: AC, amacrine cell; BC, bipolar cell; D, dorsal; GCL, ganglion cell layer; HC, horizontal cell, INL, inner nuclear layer; ONL, outer nuclear layer; RGC, retinal ganglion cell; V, ventral.

Article Snippet: The following primary antibodies and dilutions were used in this study: Calbindin (1:300, Sigma-Aldrich, #C9848); Calretinin (1:1000, Swant, #CG1); cleaved Caspase3 (Asp175) (1:500, Cell Signalling, #9661S); chAT (1:100, Millipore, #AB144P); Chx10 (1:50, SCBT, #SC365519); Dyrk1a (1:100, Novus, #NBP1-84032); Isl1/2 (recognises both Isl1 and Isl2 proteins) (1:500; DSHB, #39.4D5); M-opsin (1:500, Proteintech, #30975-1-AP); Pax6 (1:100, DSHB, #Pax6); Pikachurin (1:200, Proteintech, #14578-1-AP); Rhodopsin (1:500, Millipore, #MABN15); S-opsin (1:500, Proteintech, #24660-1-AP); TH (1:300, Millipore, #AB152); vGlut1 (1:2000, Millipore, #AB5905).

Techniques: Immunohistochemical staining, Staining

Journal: bioRxiv

Article Title: Dyrk1a gene dosage controls bipolar cell development and retinal connectivity

doi: 10.64898/2026.03.15.710015

Figure Lengend Snippet: (A – H) Representative flat mount retina images from Lhx2-Cre (A – D) and Lhx2-Cre:Dyrk1a +/f (E – H) adult mice (6 – 8 weeks) showing the mosaic distribution of bipolar cells across four defined dorsoventral quadrants (D1, D2, V3, and V4) spanning the entire central retina relative to the optic nerve head. Transgenic controls displayed an even distribution of bipolar cells across all quadrants. In contrast, heterozygous animals showed a reduced number of Chx10 + bipolar cells in the dorsal-most quadrant (E) while the remaining domains (F – H) were comparable to transgenic controls. (I – P) Representative Voronoi domain plots of bipolar cell distribution in Lhx2-Cre (I – L) and Lhx2-Cre:Dyrk1a +/f (M – P) adult mice (6 – 8 weeks). Transgenic controls showed evenly spaced bipolar cell mosaics with nearest neighbour (NNRI) and Voronoi domain regularity indices (VDRI) indicating robust spatial regularity. In contrast, the dorsal-most quadrant of heterozygous mice exhibited mosaic spacing variability reflected by irregular Voronoi domain areas and reduced regularity indices (M) while the remaining quadrants (N – P) retained regularity indices comparable to transgenic controls. (Q – T) Quantitative analysis of nearest neighbour distance in Lhx2-Cre ( n = 5) and Lhx2-Cre:Dyrk1a +/f ( n = 8) adult animals (6 – 8 weeks). Nearest neighbour distance was significantly increased in the dorsal-most quadrant of heterozygous animals owing to a reduced cell number (Q) while the other regions remained comparable to transgenic controls (R – T). (U – X) Quantitative analysis of Voronoi domain area in Lhx2-Cre ( n = 5) and Lhx2-Cre:Dyrk1a +/f ( n = 8) adult animals (6 – 8 weeks). A significant increase in Voronoi domain area was observed in the dorsal-most quadrant of heterozygous mice owing to decreased cell number (U) while the other quadrants showed no significant differences (V – X). All data represents the mean ± SEM. Statistical differences were calculated using Mann-Whitney tests. p-values are denoted as follows: **p≤ 0.01. Scale bar: (A – P) 50 μm. Abbreviations: CV, coefficient of variation; D, dorsal; N, nasal; NNRI, nearest neighbour regularity index; T, temporal; V, ventral; VDRI, Voronoi domain regularity index.

Article Snippet: The following primary antibodies and dilutions were used in this study: Calbindin (1:300, Sigma-Aldrich, #C9848); Calretinin (1:1000, Swant, #CG1); cleaved Caspase3 (Asp175) (1:500, Cell Signalling, #9661S); chAT (1:100, Millipore, #AB144P); Chx10 (1:50, SCBT, #SC365519); Dyrk1a (1:100, Novus, #NBP1-84032); Isl1/2 (recognises both Isl1 and Isl2 proteins) (1:500; DSHB, #39.4D5); M-opsin (1:500, Proteintech, #30975-1-AP); Pax6 (1:100, DSHB, #Pax6); Pikachurin (1:200, Proteintech, #14578-1-AP); Rhodopsin (1:500, Millipore, #MABN15); S-opsin (1:500, Proteintech, #24660-1-AP); TH (1:300, Millipore, #AB152); vGlut1 (1:2000, Millipore, #AB5905).

Techniques: Transgenic Assay, MANN-WHITNEY