Left-Biased Spermatogenic Failure in 129/SvJ Dnd1Ter/+ Mice Correlates with Differences in Vascular Architecture, Oxygen Availability, and Metabolites

Abstract

Homozygosity for the Ter mutation in the RNA-binding protein Dead end 1 (Dnd1(Ter/Ter)) sensitizes germ cells to degeneration in all mouse strains. In 129/SvJ mice, approximately 10% of Dnd1(Ter/+) heterozygotes develop spermatogenic failure, and 95% of unilateral cases occur in the left testis. The first differences between right and left testes were detected at Postnatal Day 15 when many more spermatogonial stem cells (SSCs) were undergoing apoptosis in the left testis compared to the right. As we detected no significant left/right differences in the molecular pathway associated with body axis asymmetry or in the expression of signals known to promote proliferation, differentiation, and survival of germ cells, we investigated whether physiological differences might account for asymmetry of the degeneration phenotype. We show that left/right differences in vascular architecture are associated with a decrease in hemoglobin saturation and increased levels of HIF-1alpha in the left testis compared to the right. In Dnd1 heterozygotes, lower oxygen availability was associated with metabolic differences, including lower levels of ATP and NADH in the left testis. These experiments suggest a dependence on oxygen availability and metabolic substrates for SSC survival and suggest that Dnd1(Ter/+) SSCs may act as efficient sensors to detect subtle environmental changes that alter SSC fate.

Keywords: Dnd1/DND1; germ cell; hypoxia; metabolism; spermatogenic failure; spermatogonial stem cells.

Fig. 1

Spermatogenic failure is biased to the left testis of 129/SvJ Dnd1^Ter/+ mice. a) An obvious size difference was frequently observed between right and left testes in adult 129/SvJ Dnd1^Ter/+ mice. b) Histological evaluation of 129/SvJ Dnd1^Ter/+ testis stained with hematoxylin and eosin showed a progressive disruption of spermatogenic development in the left testis beginning at P15 and abundant Sertoli cell-only tubules in the left testis by P180 (stars). c) Differences in tubule diameter, quantified through morphometric analysis, showed the first significant differences between right and left testes by P15 in Dnd1^Ter/+ mice. Note that wild-type 129/SvJ mice also showed a similar trend, significant only at P30. *P ≤ 0.05, **P ≤ 0.01 (n ≥ 50).

Fig. 2

Germ cells in the right and left testes of Dnd1^Ter/+ mice were similar to wild type during fetal and perinatal stages. a) Germ cells that arrived in the gonad were visualized and counted using the Oct4-EGFP reporter and confocal microscopy. b) Similar numbers of germ cells arrived in the right and left gonads by E11.5 in 129/SvJ Dnd1^+/+ and Dnd1^Ter/+ mice (n = 7 for each genotype). Expression of the autocrine regulators of germ cell development, Nodal (c), Cripto (d), Lefty1 (e), and Lefty2 (f) were quantified by qPCR in FACS OCT4-EGFP+ male germ cells at E13.5 and E14.5. Although there was a trend toward higher expression of this pathway in Dnd1^Ter/+ mutants, no significant differences were detected between the right and left testes (n ≥ 4). g and h) Similar to wild type, at P0 (top panel), gonocytes with large pale nuclei were localized in the center of the seminiferous tubules and were negative for Ki67 (arrow). By P3 (bottom panel), gonocytes of both genotypes showed normal migration to the basement membrane of the tubules and resumed proliferation based on expression of Ki67 (arrowhead) (n = 6 for each genotype).

Fig. 3

The reduced number SSCs in the left testis of Dnd1^Ter/+ mice is due to apoptosis. a) Immunofluorescence of whole-mount tubules of Dnd1^Ter/+ testes at P12. In tubules from the left testis, more SSCs (PLZF⁺ cells) were positive for cCASPASE-3 (green, top panel) and cPARP (red, bottom panel) (n = 6 mice of each genotype). b) Quantification of the numbers of activated CASPASE-3 (cC-3)-positive cells in the right and left testes of both genotypes showed that apoptosis is more frequent in Dnd1^Ter/+ compared to wild-type mice (n = 10 of each genotype, *P ≤ 0.05). c) Fewer PLZF⁺ cells were detected in Dnd1^Ter/+ mice by P17 compared to wild-type mice. By P25, the numbers of PLZF⁺ cells in the right testis of wild-type and Dnd1^Ter/+ mice were similar. However, the number of PLZF⁺ cells continued to show a significant decline in the left testis of Dnd1^Ter/+ at P34 (n = 6 mice of each genotype, at each stage, *P ≤ 0.05). d) Immunofluorescence and confocal microscopy using antibodies against PLZF (green) and STRA8 (red), a marker of cells undergoing meiosis (used to select tubules in the same stage of spermatogenesis), revealed the reduction of SSCs and differentiating spermatogonia in the left testis of both genotypes. This difference was more severe in Dnd1^Ter/+ heterozygous mutants. The introduction of a single mutant allele of the proapoptotic gene Bax rescued the number of spermatogonia in the left testis (n = 6 mice of each genotype).

Fig. 4

The number and distribution of Sertoli cells and the expression of important paracrine signals were indistinguishable from wild type in the right and left testes. The number of Sertoli cells, stained with SOX9 (a) as well as (b) the distribution of Sertoli cell cytoplasm from the basement membrane to the lumen of the tubule, based on TUJ1 staining, was normal at P20. The frames to the right in b are higher magnification images of the box defined by the broken line in the left-most frames. However, spermatogonia (GCNA+, arrowhead in a, and γH2AX+, arrowhead in b) were reduced in some tubules (a and b, asterisk) in the left testis of heterozygous mice (n = 6 mice of each genotype). Expression of a group of growth factors produced by Sertoli cells—(c) Cxcl12, (e) Gdnf, (g) KitL (SCF), and (i) Fasl—and their receptors expressed by germ cells—(d) Cxcr4, (f) Gfrα, (h) Kit, and (j) Fas—were evaluated by qPCR in whole tubules at P8. Sertoli-expressed genes were normalized to Sox9, whereas germ cell-expressed genes were normalized to Ddx4. No significant differences were detected (n = 6 mice of each genotype).

Fig. 5

Differences in vascular architecture between the right and left testes are correlated with oxygen availability. BLUE dye was injected through the heart of P17 Dnd1 wild-type and heterozygous mice, labeling their vascular systems. a) Schematic representation of the origin of right and left spermatic arteries from the aorta, illustrating the superior divergence of the right. The boxed region of this diagram is photographed in b. In 129/SvJ mice, the right spermatic artery (red arrowhead) branches from the aorta (yellow dashed line) superior to the left (green arrowhead) (n = 8 mice of each genotype). Using spectroscopic methods, (c) total Hb and (d) Hb saturation were measured in right and left testes at P5–P15, P16–P30, P31–P45, and adult. Significant differences in the percentage of Hb saturation were detected between right and left testes at P5–P15 (n = 6 mice at each stage of each genotype, *P ≤ 0.05). Higher levels of HIF-1α (arrowhead) were detected by Western blot analysis in the left testis compared to the right testis of wild-type and Dnd1^Ter mice at (e) P0 and (f) P30; u-HIF-1α: ubiquitinated HIF-1α (n = three independent Western blots).

Fig. 6

Differences in the levels of AMP, ADP, ATP, NAD⁺, and NADH between right and left testes. Right and left testes from seven to eight wild-type and Dnd1^Ter/+ mice at stage P10 were collected and independently pooled (n = 3 biological replicates for right and left testes of each genotype). a) Tissue homogenates were analyzed using LC-MS/MS to measure the levels of AMP, ADP, and ATP, and the (b) AMP/ATP and (c) ADP/ATP ratios were calculated and expressed, for each genotype, as the fold change relative to the right testis. B) Levels of NAD+ and NADH were measured in the same samples and (e) NAD⁺/NADH ratios were calculated and expressed, for each genotype, as the fold change relative to the right testis.