Supplementary Information for

Supporting Information

• Supporting Information    Includes:
  • Supplementary Materials and Methods
  • Supplementary Figure 1. STRA8 is cyclically expressed in both spermatogonia and spermatocytes. (A) Structure of the mouse testis comprising seminiferous tubules. In any given tubule cross-section, one observes germ cells at different steps in their development into spermatozoa. These germ-cell types are concentrically layered; undifferentiated spermatogonia lie on the basal lamina of the tubule, and germ cells move toward the tubule lumen as they differentiate (1). Germ-cell differentiation is precisely timed; hence, particular steps of development are always found together at the same time and place (Fig. 1B). Blue line indicates the orientation of testis cross-sections inB. Panel: A representative tubule cross-section in stage VII, stained with hematoxylin and periodic acid-Schiff (He-PAS).Arrowheads: spermatogonia (white) and preleptotene spermatocytes (yellow). (Scale bar, 30µm.) (B) A scheme to determine the expression pattern of STRA8in each testis tubule. One testis cross-section was immunostained for STRA8 (Right). The adjacent section was stained with He-PAS and used for staging (Left).(Scale bar, 100µm.) (CandD) Percentage of tubules containing STRA8-positive spermatogonia (C) or STRA8-positive spermatocytes (D). All data are from wild-type, unperturbed mice. Error bars, mean±SD. (E) Immunostaining for STRA8 on testis cross-sections. Roman numerals indicate stages. Arrowheads: sper-matogonia (white), and preleptotene (yellow), leptotene (green), and zygotene spermatocytes (red). Asterisks indicate adjacent containing STRA8 positive spermatogonia. (Scale bar, 30µm.)
  • Supplementary Figure 2. Gene expression patterns ofStra8-deficient large testes resemble type A undifferentiated spermatogonia. (A) Expression of 30 spermatogonialmarker genes (1) inStra8-deficient large testes relative to adult wild-type testes. We show log2ratios of normalized transcript counts (from mRNA-Seq). Errorbars, mean±SD *P=0.05, **P=0.01 (one-tailed Welch’sttest). (B) Heat map of correlations between (i) ourStra8-deficient and wild-type testis RNA-seq data,and three other publically available genome-wide datasets: (ii) microarray data from type A spermatogonia, type B spermatogonia, pachytene spermatocytes,and round spermatids; (iii) mRNA-Seq data from Id4-GFP–positive and Id4-GFP–negative spermatogonia (1) (SRA accession no. SRP042067); and (iv) mRNA-Seqdata from KIT-positive and KIT-negative spermatogonia (SRA accession no. DRP002386). Numbers are pairwise Spearman correlations between FPKMs/mi-croarray expression values. Dendrogram represents complete linkage clustering of samples based on their correlations. (C) Expression of undifferentiated anddifferentiating spermatogonial marker genes, inStra8-deficient large testes relative to adult wild-type testes. Undifferentiated spermatogonia marker genes:Gfra1;Lin28a;Oct4(a.k.a.Pou5f1);Nanos2(2, 3);Ngn3(4);Plzf(a.k.a.Zbtb16);Sall4; andUtf1(5, 6). Differentiating spermatogonia marker genes:Ccnd2,which is expressed exclusively in very early differentiating spermatogonia (7), andKit, which is expressed in all differentiating spermatogonia. We show log2ratios of normalized transcript counts (from mRNA-Seq). Error bars, mean±SD *P=0.01 (one-tailed Welch’sttest). (D) Testis cross-sections from 1-y-old mice:wild-type (Top) andStra8-deficient large (Bottom) testes. From left to right: immunostaining for PLZF, LIN28A, GFRa1, and KIT. (Scale bars, 50µm.) (E)Per-centage of genes whose expression is higher in (Left) Id4-GFP–positive spermatogonia or Id4-GFP–negative spermatogonia and (Right) KIT-positive sper-matogonia or KIT-negative spermatogonia. Black bars (control), all analyzable genes (18,936 genes). Gray bars, 100 genes most significantly up-regulated inStra8-deficient large testes relative to wild-type testes (Table S1)
  • Supplementary Figure 3. LIN28A-positive and PLZF-positive spermatogonia accumulate at p10 and p30 inStra8-deficient males. (A) Immunostaining for PLZF on postnatal day10 (p10) testis sections: wild-type (Top) andStra8-deficient (Bottom). (Scale bars, 50µm.) (B) Histogram of number of PLZF-positive cells per tubule cross-section,in p10 wild-type andStra8-deficient testes. Testes from one wild-type mouse and twoStra8-deficient mice were counted and averaged. (C) Number of PLZF-positive spermatogonia per tubule cross-section, in p10 and p30 wild-type andStra8-deficient testes. One animal was counted for each condition (except forp10Stra8-deficient, for which two animals were counted). (D) Testis cross-sections from p30 mice: wild-type (Left),Stra8-deficient (Middle), andDmc1-deficient(Right). (Top) Sections stained with He-PAS. (Bottom) Grayscale versions of top panel, with red dots indicating type A spermatogonia. Spermatogonial clusters(close packing of type A spermatogonia in multiple layers) were observed in someStra8-deficient tubules, but never in wild-type orDmc1-deficient tubules.Because tubule size is reduced in bothStra8-deficient andDmc1-deficient testes (owing to meiotic arrest), we conclude that the spermatogonial clusters are not simply a result of reduced tubule size. (Scale bars, 50µm.) (E) Immunostaining for PLZF on p30Stra8-deficient testis sections, showing clusters of PLZF-positiveundifferentiated spermatogonia. (Scale bars, 50µm.) (F) Testis cross-sections from p30 mice: wild-type (Top) andStra8-deficient (Bottom). These are repre-sentative of the tubule cross-sections used for counts of type B and LIN28A-positive spermatogonia (Fig. 3D). (Left) He-PAS stained section. Inset enlarges theboxed region. (Middle) Adjacent section, immunostained for LIN28A. (Right) Grayscale version of middle panel, with red dots indicating LIN28A-positivespermatogonia, and purple dots indicating type B spermatogonia. (Scale bars, 50 µm.)
  • Supplementary Figure 4. Injected RA induces precocious STRA8 expression, KIT expression, and S phase entry. (A) Whole-mount immunostaining of the intact wild-type testistubules shown in Fig. 4A, for PLZF (red), STRA8 (green), and GRFa1 (gray), with DAPI counterstain (gray). (PLZF and STRA8 panels are duplicated from Fig. 4A.)Dashed lines: putative interconnected chains of spermatogonia. Magenta labels: early undifferentiated type A spermatogonia (Aearly). Yellow labels: lateundifferentiated type A spermatogonia (Alate). As,Apr, and Aal:Asingle(isolated; arrowheads), Apaired(chains of 2), and Aaligned(chains of=4) spermatogonia. Blue labels: differentiating type A spermatogonia (Adiff). Aearlyidentified as PLZF-positive, GRFa1-positive Asand Apr;Alateidentified as PLZF-positive,GFRa1-negative Aal. (Scale bar, 30µm.) (B) Immunostaining for PLZF (red) and STRA8 (green), with DAPI counterstain (blue), on control and RA-injected testiscross-sections in stage IV. Arrowheads: PLZF-positive cells. (Scale bar, 30µm.) (C) PLZF is expressed only in type A spermatogonia, not in intermediate or type Bspermatogonia, in controls and 1 d after RA injection. Immunostaining for PLZF (red), with DAPI counterstain (blue), on control and RA-injected testis cross-sections in stages II and V. Arrowheads: type A (white), intermediate (yellow), and type B (green) spermatogonia. (Scale bar, 10µm.) (D) Immunostaining forSTRA8 (red), with DAPI counterstain (blue), on control and RA-injected testis cross-sections in stages II and V. Arrowheads: premeiotic germ cells,specificallyintermediate (yellow) and type B (green) spermatogonia. (Scale bar, 10µm.) (E) Percentage of type A spermatogonia that are positive for KIT in testis cross-sections, in controls or 1 d after RA injection. Error bars, mean±SD *P=0.01(one-tailed Welch’sttest). (F) Percentage of PLZF-positive cells that are alsopositive for BrdU in testis cross-sections, in controls or 1 d after RA injection. Error bars, mean±SD *P=0.01(one-tailed Welch’sttest). (G) Immunostaining forKIT, on control and RA-injected testis cross-sections in stages IV and VIII. Insets enlarge boxed regions. Arrowheads: type A (white) and intermediate (yellow)spermatogonia, and preleptotene spermatocytes (green). (Scale bar, 30µm.) (H) Immunostaining for PLZF (red) and BrdU (green), with DAPI counterstain(blue), on control and RA-injected testis cross-sections in stage IV. Arrowheads: PLZF-positive cells. (Scale bar, 30µm.
  • Supplementary Figure 5. Injected RA induces premature spermatogonial differentiation in stages II–VI and premature meiotic initiation in stage VI. (A–C) RA-injected testiscross-sections, stained with He-PAS. Roman numerals indicate stages. Arrowheads inA: type B spermatogonia. Arrowheads inB: preleptotene spermatocytes. Arrowheads inC: leptotene spermatocytes. Insets enlarge boxed regions. (Scale bars, 30µm.) (D) The entire tubule images of RA-injected testis cross-sections instages II, IV, and VI, stained with He-PAS. Insets (blue and black) enlarge boxed regions. (Left) He-PAS stained section. (Right) Grayscale version of left panel.Red, purple, and yellow dots indicate Intermediate spermatogonia, type B spermatogonia, and preleptotene spermatocytes, respectively. At 8.6 d after RAinjection, the tubules in stages II–VI contained predominantly preleptotene spermatocytes, with only a few Intermediate and type B spermatogonia. This resultindicates that, at the time of RA injection, most (although not all) undifferentiated spermatogonia in stages II–VI successfully underwent precocious sper-matogonial differentiation. (Scale bars, 50µm.) (E–G) Two RA injections induced meiotic initiation in all preleptotene spermatocytes present in stages II–VIII. At10.6 d after a single RA injection, leptotene spermatocytes were present only in stages VI–X (Fig. 5FandG). (If all preleptotene spermatocytes in stages II–VIIIhad initiated meiosis, we would expect to observe leptotene spermatocytes throughout stages V–X. See Table S2 for details.) Because RA is required for meioticinitiation, we reasoned that a second injection of RA might be required to induce meiotic initiation in all preleptotene spermatocytes present in stages II–VIII.Thus, we gave a second injection of RA, 8.6 d after the first injection, and collected testes after 10.6 d (E). We observed leptotene spermatocytes in stages V–X(FandG). (E) Diagram of germ-cell development after two injections of RA (8.6 d apart). Testes were harvested 10.6 d after the first RA injection. Aundiff,Adiff,In and B: undifferentiated type A, differentiating type A, intermediate, and type B spermatogonia. Pl and L: preleptotene and leptotene spermatocytes.(F) Percentage of tubules containing leptotene spermatocytes, in control or RA-injected testis cross sections. Mice were injected with RA once (RAx1)or twice(RAx2). Error bars, mean±SD *P=0.01 (Tukey–Kramer test). (G) Control and RA-injected testis cross-sections in stage V, stained with He-PAS. Mice wereinjected with RA once (RAx1) or twice (RAx2). Arrowheads: intermediat e spermatogonia (white) and leptotene spermatocytes (yellow). Insets enlarge boxedregions. (Scale bars, 30µm.) (H) Percentage of tubules containing type B spermatogonia, in control or RA-injected testis cross-sections. Error bars, mean±SD.(I) Immunostaining for STRA8 on testis cross-sections in stage VIII. Sections are from control mice, and from mice treated with WIN18,446 for 2 d. Dashedlines,basal laminae. Arrowheads: type A spermatogonia (white) and preleptotene spermatocytes (yellow). (Scale bar, 30µm.) (J) Testis cross-sections in stage IX,stained with He-PAS. Sections are from control mice and from mice treated with WIN18,446 for 4 d. Arrowheads: preleptotene spermatocytes (white) andleptotene spermatocytes (yellow). Insets enlarge boxed regions. (Scale bar, 30µm.) (K) Percentage of tubules containing preleptotene spermatocytes (Top)orleptotene spermatocytes (Bottom), in testis cross-sections from control or WIN18,446-treated mice. Error bars, mean±SD *P=0.01, compared with control(Dunnett’s test). See also Table S2.
  • Supplementary Figure 6. After RA injection, precocious leptotene spermatocytes undergo meiotic progression and meiotic divisions. (A) Diagram of spermatocyte de-velopment after RA injection. L, Pl, and Z: leptotene, preleptotene, and zygotene spermatocytes. (B) Control and RA-injected testis cross-sections, stained withHe-PAS. Roman numerals indicate stages. Arrowheads: leptotene (yellow) and zygotene (green) spermatocytes. Insets enlarge boxed regions. (Scalebar,30µm.) (C) Immunostaining for?H2AX (green) and SYCP3 (red), with DAPI counterstain, on control and RA-injected testis cross-sections. Roman numeralsindicate stages. Arrowheads: type A spermatogonia (white), and preleptotene (yellow) and leptotene (green) spermatocytes. D and P: diplotene and pachytenespermatocytes. Insets enlarge boxed regions. (Scale bars, 10µm.) (D) Control and RA-injected testis cross-sections, stained with He-PAS. Roman numerals in-dicate stages. Arrowheads: pachytene (white) and meiotically dividing (yellow) spermatocytes. Insets enlarge boxed regions. (Scale bar, 30µm.) (E) Percentageof tubules containing meiotically dividing spermatocytes, in control or RA-injected testis cross-sections. Error bars, mean±SD *P=0.01 (one-tailed Welch’sttest)
  • Supplementary Figure 7. Successive RA injections confirm distinct competencies for spermatogonial differentiation and meiotic initiation. (A) Percentage of tubules containingspermatids at various steps of postmeiotic differentiation, steps 2–5(Top), step 6 (Middle), or steps 7–8(Bottom), in control or RA-injected testis cross-sections.Error bars, mean±SD *P=0.05, **P=0.01, compared with control (Dunnett’s test). (B–D) Successive RA injections can drive release of spermatozoa.(B) Diagram of five successive RA injections. Testes were harvested 36.4 d after the first RA injection. (C) Percentage of tubules in stages IX–X of the semi-niferous cycle, in controls and after five RA injections. Tubules in stages IX–X have recently undergone release of spermatozoa. Error bars, mean±SD *P=0.01(two-tailed Welch’sttest). (D) Testis cross-sections, stained with He-PAS, in controls (Left) or after five RA injections (Right). The cross-section areas with thehighest frequency of stage IX–X tubules (red) were selected. Roman numerals indicate stages. Insets enlarge boxed regions. (Scale bar, 100µm.)
  • Supplementary Table 1. One hundred genes most significantly up-regulated inStra8-deficient large (large)testes, relative to wild-type (WT) testes.
  • Supplementary Table 2. Progression of germ cells through the seminiferous stages after RA-induced precocious spermatogonial differentiation.
  • Supplementary Table 3. Antibodies and conditions for immunostaining.