Fig 2: Hallmarks of ferroptosis in Parl-/- primary spermatocytes.(A) Quantitative immunofluorescence shows severely reduced GPX4 expression in SCP-1-positive primary spermatocytes from 5-week-old Parl-/- mice compared to WT littermates (n = 3 mice for each genotype, 500–1000 SCP-1-positive spermatocytes considered for each mouse, p=0.0013). (B) Quantitative immunofluorescence shows increased HNE accumulation in Parl-/- SCP-1-positive spermatocytes compared to WT littermates (n = 3 mice for each genotype, 500–1000 SCP-1-positive spermatocytes considered for each mouse, p=0.0002). Bar graphs indicate average ± SD. Statistical significance calculated by two-sided Student’s t-test. Scale bars, 100 µm.

Fig 3: Severe loss of COX4 associated with increased expression of glucose intracellular transporter in Parl-/- spermatocytes.(A) Quantitative immunofluorescence shows decreased expression of COX4 in SCP-1-positive spermatocytes from 5-week-old Parl-/- mice compared to WT littermates (n = 3 for each genotype, 500–1000 SCP-1-positive spermatocytes for each mouse, two-sided Student’s t-test: p=0.0027). Scale bars, 100 µm. Bar graphs indicate average ± SD. (B) Normalized quantification of TOMM20 immunofluorescence in SCP-1-positive primary spermatocytes does not reveal significant differences in mitochondrial mass in the two different genotypes (n = 3 mice for each genotype, 500–1000 SCP-1-positive spermatocytes considered for each mouse; p=0.821). Scale bars, 100 µm. Bar graphs indicate average ± SD. Statistical significance calculated by two-sided Student’s t-test. (C) GLUT1 immunohistochemistry of testis from 5-week-old mice shows prominent overexpression of GLUT1 in arrested Parl-/- spermatocytes, and low levels in WT (arrowheads) (n = 3 for each genotype). Scale bars, 50 µm.

Fig 4: Severe testis atrophy in Parl-/- mice is caused by arrested spermatogenesis.(A) Reduced testicular size and weight in 5-week-old Parl-/- mice (n = 5) compared to WT littermates (n = 6; unpaired two-tailed t-test, p-value<0.0001). The reduction in testicular weight is not explained by body weight differences (p=0.0598). (B) Histological assessment of testes from 6-week-old Parl-/- and WT mice reveals reduced diameter of Parl-/- seminiferous tubules with impaired germ cell maturation and complete spermatogenesis arrest at the level of primary spermatocytes (testis HE stain, n = 10 for each genotype). Parl-/- seminiferous tubules also exhibit intraluminal exfoliation of degenerated spermatocytes often in the form of multinucleated syncytia (testis HE stain inset, arrowheads). The complete arrest of spermatogenesis leads to total absence of sperm in Parl-/- seminiferous tubules and epididymis compared to WT littermates (testis and epididymis HE stain, n = 10 for each genotype; asterisks indicate mature spermatozoa in the WT). Immunohistochemistry for synaptonemal complex protein 1, SCP-1, confirms complete spermatogenesis arrest at the level primary spermatocytes in Parl-/- testis (testis SCP-1, n = 10 for each genotype). The distribution of SCP-1 expression is confined to primary spermatocytes and is lost in postmeiotic germ cells as they undergo maturation in WT seminiferous tubules. Immunohistochemistry for allograft inflammatory protein 1, AIF-1, reveals the complete absence of spermatids in Parl-/- testis while WT seminiferous tubules are densely populated by AIF-1-positive spermatids at different levels of maturation (testis AIF-1, n = 10 for each genptype). 8-week-old mice with conditional Parl deletion driven by the Nes promoter in the nervous system and Leydig cells (Parl L/L::NesCre) display a normal testicular and epididymal histology as well as SCP-1 and AIF-1 immunohistochemistry comparable to WT mice (right column, n = 4). Scale bars, 200 µm.