Sc 28319

The 5′‐iodo‐2′‐deoxyuridine (IdU)/5′‐bromo‐2′‐deoxyuridine (BrdU) injection for calculating T_c is described schematically in Figure S1. Both IdU and BrdU were injected into the mice (100 mg/kg). Testes were embedded in paraffin and serially sectioned into 4‐μm slices. The specimens were incubated with anti‐PLZF (SC‐28319, Santa Cruz Biotechnology, Inc., USA; 1:200) to label SSCs in serial sections. In addition, the specimens were incubated with anti‐BrdU (#347580, BD Biosciences, USA; 1:100), which recognizes both IdU and BrdU, and anti‐BrdU (ab6326, Abcam, UK; 1:200) for double staining. T_c was calculated using the ratio of IdU‐only‐labelled cells (leaving cells, L_cells) and PLZF‐labelled cells (the total number of SSCs, P_cells).²⁷, ²⁸ Cells labelled with IdU only left the S phase during the interval time (12 h) between the IdU and BrdU injections. Therefore, the following formula can be used to calculate T_c: $$\frac{T_c}{\text{Interval time}}=\frac{{\mathrm{P}}_{\text{cells}}}{{\mathrm{L}}_{\text{cells}}}\text{or}{T}_c=\frac{{\mathrm{P}}_{\text{cells}}}{{\mathrm{L}}_{\text{cells}}}\times \text{interval time}.$$

Sections were prepared as described in immunohistochemical analysis except without treatment with 3% H₂O₂. Samples were incubated with primary antibodies overnight at 4°C, washed in PBS four times, and then incubated with 1∶100 dilution of Cy3 conjugated sheep anti-rabbit antibody (Sigma) or 1∶200 dilution of Alexa Fluor 488 goat anti-mouse IgG (H+L) (Molecular Probes) in 1% saline-buffered BSA for 1 h at room temperature. Following primary antibodies were used: affinity-purified anti-cyclin K, anti-Ki67, anti-CD117 and anti-Plzf (SC-28319, SantaCruz). To visualize DNA, sections were incubated with 1 µg/ml of DAPI. After washing in PBS, coverslips were mounted with Aqueous Mounting Medium (F4680, Sigma), and visualized by Nikon Eclipse Ti-U fluorescence microscope.
To visualize the localization of cyclin K and vimentin simultaneously, sections were incubated with anti-vimentin antibodies first followed by immunofluorescence detection. Both primary and secondary antibodies were then removed by incubation with 10 mM citrate buffer (pH 6.0) in a pressure cooker for 2 min. Cyclin K was then detected by immunohistochemistry. The reason for this modified method is because both primary antibodies are generated from rabbit, and therefore not suitable for the double-staining procedure.

Testis tissues of mice were dissected and immersed in 10% Formalin for overnight at 4 °C, then embedded into wax. Sections were dewaxed and treated with antigen retrieval solution (HK057-5K, BioGenex, Fremont, CA, USA) as manufacture’s guide. The germ cell markers, PLZF (1:100, sc–28319, Santa Cruz Biotechnology Inc., Dallas, CA, USA), rat anti-TRA98 (1:150, ab82527, Abcam) and VASA (1:1000, ab13840, Abcam) were stained in testis sections. Isotype IgG including rabbit (550875, BD) and rat (559072, BD) were used as the negative control as shown in Supplementary Figure S2. Secondary antibodies (1:500) were used as follows: Alexa Fluor donkey anti-rat Cy5(A10525) and Alexa Fluor goat anti-rabbit IgG 594 (A11012).