A Effects of zebrafish Insl3 on four relaxin family peptide receptors (Rxfps) expressed in HEK293T cells transiently transfected with pcDNA3.1 and pCRE plasmids. Data are expressed as mean ± SEM (N = 3, technical replicates). Numbers in brackets indicate the EC₅₀ concentrations. AU, arbitrary units. B Expression levels of responsive Insl3 receptors in control, germ cell-depleted (by exposure to the cytostatic agent busulfan⁵¹), and testes with recovering (from busulfan) spermatogenesis, as described by Crespo et al.²⁵ (NCBI GEO data set GSE116611). Data are expressed as mean ± SEM (N = 5; *p < 0.05). C, D Localization of rxfp2a:EGFP (green; C) and rxfp2b:mCherry signal (red; D) in adult testis tissue. Confocal laser scanning microscopy analysis of whole-mount testes shows preferential expression of EGFP and mCherry in type A spermatogonia and Sertoli and myoid cells, respectively. DAPI counterstain is shown in gray. Representative germ cell types (insets 1–4) and Sertoli and myoid cells are shown (insets 5–6). A_und, type A undifferentiated spermatogonia; A_diff, type A differentiating spermatogonia; B, type B spermatogonia; SPC, spermatocytes. In D, red and white arrowheads indicate representative mCherry⁺ and mCherry⁻ somatic cells, respectively. Scale bars, 25 µm.

A–C Body weight (A), gonado-somatic indices (GSI; B) and testicular morphology (C) of wild-type (insl3^+/+) and insl3 knockout (insl3^−/−) males 9 and 12 months post-fertilization. Data are mean ± SEM (insl3^+/+ and insl3^−/−, 9 months: N = 8 and 13; insl3^+/+ and insl3^−/−, 12 months: N = 8 and 15; p < 0.05; **p < 0.01; ***p < 0.001). In C, yellow dashed lines indicate representative apoptotic germ cell cysts. Scale bars, 25 µm. D, E Quantitative analysis of spermatogenesis (D) and transcript levels of growth factors, steroidogenesis-related and Insl3 receptors (E) in 12 months-old insl3^+/+ and insl3^−/− adult testis tissue. Data are mean ± SEM (insl3^+/+ and insl3^−/−: N = 6 and 10; *p < 0.05; **p < 0.01) and, in E, expressed as relative to the wild-type group (which is set at 1; dashed line). A_und, type A undifferentiated spermatogonia; A_diff, type A differentiating spermatogonia; B, type B spermatogonia; SPC, spermatocytes; SPT, spermatids; SPZ, spermatozoa; Others, including (i) empty spaces lined by Sertoli and/or germ cells that are not part of the lumen, (ii) Sertoli cell only areas, and (iii) apoptotic cells (see Supplementary Fig. 2 for further details); SC, Sertoli cell; LC, Leydig cell.

A–C Detection (A, B) and quantification (C) of germ cell apoptosis/DNA damage in wild-type (insl3^+/+) and insl3 knockout (insl3^−/−) testes 9 and 12 months post-fertilization. In A, white arrowheads and arrows indicate representative TUNEL⁺ spermatogonia and Sertoli cells, and gray arrowheads and arrows indicate representative TUNEL⁺ spermatids and spermatocytes, respectively. In B, representative TUNEL⁺ spermatogonia (SPG), spermatocyte (SPC) or spermatid (SPT) cysts are encircled with a white dashed line, and a TUNEL⁺ Sertoli cell (SC) is indicated by a white arrowhead. TUNEL⁺ cells/cysts are shown in green and propidium iodide (PI) counterstain is red. Scale bars, 25 µm. D Transcript levels of anti- (xiap) a pro-apoptotic (casp9) genes in insl3^+/+ and insl3^−/− testis tissue 9 and 12 months post-fertilization. TUNEL quantification results are shown as mean ± SEM (insl3^+/+ and insl3^−/−, 9 months: N = 4 and 7; insl3^+/+ and insl3^−/−, 12 months: N = 3 and 10; *p < 0.05; **p < 0.01), and gene expression data as mean fold change ± SEM (insl3^+/+ and insl3^−/−, 9 months: N = 8 and 7; insl3^+/+ and insl3^−/−, 12 months: N = 6 and 10; *p < 0.05) and expressed relative to the wild-type group (which is set at 1).

A Total numbers of up-regulated and down-regulated genes (DEGs) identified by RNAseq (N = 3; p < 0.05). To generate testis samples for RNAseq, male zebrafish testes were incubated in the absence or presence of zebrafish Insl3 (100 ng/mL) for 2 days. FC fold change. B, C Insl3-regulated KEGG pathways (B) and Gene Ontology terms (C) in adult zebrafish testis tissue. KEGG pathways represented by at least three DEGs and ratio of regulated genes (down-regulated/up-regulated) higher than 2 were considered for the analysis. In C, number of identified genes is shown in brackets and enrichment significance (FDR, false discovery rate) is represented as a color gradient. D Selected DEGs identified by KEGG and GO analyses grouped by their function. Fold change values are shown with a red or green background indicating up-regulation or down-regulation, respectively.

A, B Evaluation of the proliferation activity (A) and of the proportions of spermatogonia (B) in zebrafish testes cultured for 4 days with 100 ng/mL Insl3, and in the absence or presence of the RA inhibitor DEAB (10 μM). C Ex vivo transcript levels of the RA producing (aldh1a2) and degrading enzymes (cyp26a1) in testis tissue incubated in the absence or presence of 100 ng/mL Insl3. Daldh1a2 and cyp26a1 expression levels in wild-type (insl3^+/+) and insl3 knockout (insl3^−/−) testes 9 and 12 months post-fertilization. E Read numbers (RNAseq) of RA metabolic enzymes in control, germ cell-depleted (by exposure to the cytostatic agent busulfan⁵¹), and testes with recovering (from busulfan) spermatogenesis, as described by Crespo et al.²⁵ (NCBI GEO data set GSE116611). In A, B and E, data are shown as mean ± SEM (A: N = 6; B: N = 4; E: N = 5; *p < 0.05; **p < 0.01), and in C, D as mean fold change ± SEM (C: N = 7; D: insl3^+/+ and insl3^−/− 9 months, N = 8 and 7; insl3^+/+ and insl3^−/− 12 months, N = 6 and 10; *p < 0.05) and expressed relative to the control group (which is set at 1). A_und, type A undifferentiated spermatogonia; A_diff, type A differentiating spermatogonia; B, type B spermatogonia.

A, B Evaluation of the proportions (A) and of the proliferation activity of spermatogonia (B) in zebrafish testes cultured for 4 days in the absence or presence of the Pparg antagonist T0070907 (10 μM). C Area occupied by different types of spermatogonia in wild-type (pparg^+/+) and pparg knockout (pparg^−/−) adult testes. Two different pparg^-/- mutants (alleles 1220 and 1737) were evaluated. D In vivo pparg expression levels in wild-type (insl3^+/+) and insl3 knockout (insl3^-/-) testes 9 and 12 months post-fertilization (left panel), and in control, germ cell-depleted (by exposure to the cytostatic agent busulfan⁵¹), and testes with recovering (from busulfan) spermatogenesis, as described by Crespo et al.²⁵ (NCBI GEO data set GSE116611) (right panel). In A–C and right panel in D, data are shown as mean ± SEM (A, B: N = 6; C: pparg^+/+, pparg^{−/− sa1220} and pparg^{−/− sa1737}, N = 2, 3 and 4; D: N = 5; *p < 0.05; **p < 0.01), and in the left panel in D as mean fold change ± SEM (insl3^+/+ and insl3^−/−, 9 months: N = 8 and 7; insl3^+/+ and insl3^−/−, 12 months: N = 6 and 10; *p < 0.05; **p < 0.01) and expressed relative to the control group (which is set at 1). In C, different letters indicate significant differences between groups (*p < 0.05). A_und, type A undifferentiated spermatogonia; A_diff, type A differentiating spermatogonia; B, type B spermatogonia.

Described effects are indicated by black (Fsh), gray (Insl3), blue (Pparg), and orange (RA) arrows, while germ cell development or germ cell-mediated effects are indicated in green. Gray dashed line denotes no experimental evidence reported. Fsh, follicle-stimulating hormone; Insl3, insulin-like 3; Pparg, peroxisome proliferator-activated receptor gamma; RA, retinoic acid; Cx43, connexin 43; Rxfp2a, relaxin family peptide receptor 2a; Rxfp2b, relaxin family peptide receptor 2b; A_und, type A undifferentiated spermatogonia; A_diff, type A differentiating spermatogonia; B, type B spermatogonia; SPC, spermatocytes; SPT, spermatids; SPZ, spermatozoa; Leydig, Leydig cell; Sertoli, Sertoli cell.