Distinct waves of ovarian follicles contribute to mouse oocyte production
- Howard Hughes Medical Institute and Department of Biosphere Sciences, Carnegie Institution for Science, United States
Figures
Figure 1 with 2 supplements
Wave 1 follicles are remodeled prior to peri-puberty.
(A) Schematic diagrams of fetal and postnatal ovaries at the indicated stages (E = embryonic day; P = postnatal day). Wave 1 follicles, located in the ovarian medulla (below the dashed line), retain bipotential pre-granulosa (BPG, green) cells and develop without delay. In contrast, cortical follicles, which become quiescent primordial follicles (wave 2), undergo replacement of BPGs by epithelial pre-granulosa (EPG, yellow) cells. This replacement is completed by P5. By P21 (3 wk), wave 1 follicles undergo atresia, characterized by oocyte loss. Germ cells/oocytes are shown in blue. Dashed line: medullary–cortical boundary. (B) Postnatal follicular developmental timeline: wave 1 antral follicles emerge by 3 wk (Figure 1—figure supplement 1A–C); juvenile sexual development (peri-puberty) occurs at 4–5 wk; ovulation takes place around 7–8 wk. (C) Annotation of total follicles in P5 and P10 ovaries using Imaris software (gray spheres). Each follicle with a DDX4-positive oocyte was manually counted and labeled using the Imaris spot model. Scale bars: left, 50 µm; right, 70 µm. (D) Total follicle counts in P5 and P10 ovaries (mean ± SD; N = 3–4). ns, not significant. (E) Annotation of wave 1 follicles in P5 and P10 ovaries using Imaris software (gray spheres). Follicles with oocytes larger than 20 µm in diameter, primarily located in the medulla, are classified as wave 1 follicles at the indicated times. Each gray sphere represents one follicle. Scale bars: left, 50 µm; right, 70 µm. (F) Wave 1 follicle counts in P5 and P10 ovaries (mean ± SD; N = 3–4). ns, not significant. (G) 3D projection of the ovary at P14, reconstructed using the Imaris surface model. Scale bars = 50 µm. (H) Oocyte volume increase in wave 1 follicles over time (mean ± SD; N = 10). (Note: Nuclear volume is excluded, as DDX4 is a cytoplasmic protein.) ****p < 0.0001. (I) Representative images of ovaries at 4 and 5 wk. Left (4 wk): arrowheads indicate an atretic follicle with a distorted oocyte and disorganized granulosa cells; arrow indicates a developing healthy secondary follicle. Right (5 wk): arrowheads indicate degrading follicles at various stages; arrow indicates a developing healthy secondary follicle. Scale bars = 100 µm. (J) Wave 1 follicle remodeling, indicated by oocyte loss over time (mean ± SD; N = 3–4). Oocyte loss begins around 3 wk and continues through 5 wk, accompanied by granulosa cell disorganization and cell death. ns, not significant; **p < 0.01; ****p < 0.0001.
Figure 1—figure supplement 1
Wave 1 follicle development in the early juvenile ovary.
(A–D) Representative images of wave 1 follicles stained for DDX4 (green) at P5, P7, P14 (2 wk), and P21 (3 wk). (A) Early primary follicle with theca cells (arrow); (B) late primary follicle; (C) secondary follicle; (D) antral follicle. White arrowhead: non-surrounded nucleus; red arrowhead: pyknotic nuclei of granulosa cells. Scale bars in, A: 20 µm; (B–D) 50 µm.
Figure 1—figure supplement 2
Wave 1 follicles begin to remodel by 3 wk.
(A) Electron microscopy of follicles from 5-wk ovary. Left: normal follicle. Middle: intermediate follicle, where the oocyte cytoplasm shrinks and begins to degrade, indicated by the space between the oocyte and the ZP, the absence of microvilli, and folding of the ZP (red arrow). Right: remodeled follicle, where the oocyte has disappeared but the ZP remains. Surrounding theca and interstitial gland cells further accumulate lipid droplets. The boxed region is magnified in the lower panel. Abbreviations: Oo = oocyte; ZP = zona pellucida; LD = lipid droplet; mt = mitochondria; mv = microvilli. Scale bars = 20 µm. (B) DAPI-stained DBA/2J ovaries showing remodeling wave 1 follicles (arrowheads) and normally developing follicles (arrows). Left (3 wk): early remodeling follicles appear in the central medulla (arrowheads). Right (7 wk): wave 1 follicle derivatives retain an empty structure (arrowheads), remaining visible alongside normal follicles (arrows) and corpora lutea (asterisks). Ovaries were stained and imaged as whole mounts; only a single section is shown here. Scale bars = 200 µm.
Figure 2 with 1 supplement
Wave 1 follicles lose granulosa cells during remodeling.
(A) Schematic diagram illustrating activation of Foxl2-driven EYFP expression following TAM injection at E16.5. TAM administration at E16.5 induces excision of a stop cassette, leading to EYFP expression in granulosa cells of follicles (see Methods). (B) Efficient EYFP labeling of granulosa cells in wave 1 medullary follicles of a 2-week ovary following TAM injection at E16.5, as shown in (A), magnified on the right. Scale bars: left, 100 µm; right, 30 µm. (C) EYFP-labeled follicles that reached the primary/secondary follicle stage by 2 wk were quantified using Imaris software, with annotation performed using the spot model. Left: section showing annotated labeled follicles (solid white circles). Right: 3D projection using Imaris software, with labeled follicles represented by gray spheres. Scale bars: left, 100 µm; right, 80 µm. (D) By 5 wk, EYFP-labeled follicles that reached the preantral/antral follicle stage were quantified using Imaris software and annotated using the spot model. Left: section showing annotated labeled follicles (solid white circles). Right: 3D projection using Imaris software, with labeled follicles represented by gray spheres. Scale bars = 200 µm. (E) Over 90% of wave 1 follicles contained EYFP-positive granulosa cells at 2 wk. At 5 wk, approximately 70% of normal antral follicles contained EYFP-positive granulosa cells. (F) Quantification of wave 1 follicle numbers (representative images shown in C–D) shows a decline from 2 to 5 wk. ****p < 0.0001. Each dot represents one sample. (G) Representative section from a 5-week ovary showing ongoing remodeling of wave 1 follicles. Follicles near the medullary core appear small and without oocytes (arrowheads), with few remaining EYFP-labeled granulosa cells, as seen in the follicle in rectangle ‘1’, enlarged on the right. In contrast, a normal antral follicle in rectangle ‘2’, also enlarged on the right, contains numerous EYFP-labeled granulosa cells. Right: average number of EYFP-labeled granulosa cells in medullary follicles before remodeling (3 wk) versus remodeled medullary follicles at 5 wk. Each point represents one analyzed follicle. ****p < 0.0001. Scale bars: left, 200 µm; right, 50 µm. (H) About 80% of remodeled follicles retained some EYFP-labeled granulosa cells at 5 wk, consistent with their wave 1 origin. Dashed line: follicle boundary; arrowheads: EYFP-positive granulosa cells. Each point represents one analyzed ovary. (I) Remnant granulosa cells of remodeled wave 1 follicles continued to express Foxl2 at 5 wk. TAM was injected into Foxl2-CreERT2; Rosa26-LSL-tdTomato mice at 5 wk, and ovaries were collected 1 week later. Arrowheads indicate remodeled follicles; one follicle (outlined in the rectangle) is enlarged on the right. Scale bars: left, 100 µm; right, 20 µm.
Figure 2—figure supplement 1
Lineage labeling using Foxl2-CreERT2 and Rosa26-LSL-tdTomato.
(A) Schematic diagram of the experiment. The Foxl2-CreERT2 mouse line was generated with EGFP inserted into the Foxl2 locus, resulting in a weak EGFP signal detectable in the cytoplasm using an EGFP antibody. To rule out the possibility that the observed signal originated from the Foxl2 locus rather than Rosa26-EYFP, we repeated the experiment using the Foxl2-CreERT2; Rosa26-LSL-tdTomato mouse line. (B) At 2 wk, wave 1 follicles are labeled. By 5 wk, some have remodeled, as indicated by cavities left by lost oocytes and remaining tdTomato signals (arrowheads), while others have developed into antral follicles (arrow). One developing early primary follicle in the boxed region is shown in panel C. Scale bars are included in the images. (C) The EGFP channel shows a weak cytoplasmic signal from the Foxl2 locus, distinct from the strong, evenly distributed expression of EYFP from the Rosa26-EYFP reporter shown in Figures 3 and 4. Scale bar = 20 µm. (D) Boundary follicles (enlarged on the right) were observed in the 2-wk ovary of the Foxl2-CreERT2; Rosa26-LSL-tdTomato mouse line injected with TAM at E16.5. Scale bars: left, 50 µm; right: 10 µm.
Figure 3
The labeling and quantification of boundary follicles.
(A) At 2 wk, one or more granulosa cells in some primordial follicles located at the medullary–cortical boundary are labeled with EYFP (examples enlarged in ‘1’, ‘2’, ‘3’) following TAM injection, as shown in the schematic. These follicles are referred to as boundary follicles or, in accordance with wave classification, wave 1.5 follicles. Dashed line: medullary–cortical boundary. Scale bars: left, 100 µm; right, 20 µm. (B) Quantification shows that 7% (421 ± 74) of the total 5843 ± 1256 primordial follicles contained EYFP-labeled granulosa cells in the experiment shown in (A). Each point represents one analyzed sample. Labeled PFs: labeled primordial follicles. (C) Representative images of Foxl2-CreERT2; Rosa26-LSL-EYFP mouse ovary injected with TAM at E14.5. Boundary follicles labeled with EYFP are enlarged on the right. Dashed line: medullary–cortical boundary. Scale bars: left, 100 µm; right, 20 µm. (D) Approximately 70% of secondary follicles in 5-wk ovaries contained EYFP-labeled granulosa cells, suggesting they originate from labeled primordial follicles (wave 1.5) at 2 wk. TAM was injected at E16.5. Each point represents one analyzed sample. Scale bar = 50 µm.
Figure 4 with 1 supplement
Follicles remodel to form an interconnected stromal network rich in thecal secretory cells.
(A) Schematic diagram illustrating lineage labeling of Cyp17a1-expressing theca cells, interstitial gland cells, and potentially other cell types using the Cyp17a1-iCre; Rosa26-LSL-tdTomato mouse model. (B) A 2-wk ovary showing tdTomato-labeled theca cells sheathing follicles (arrowhead) and putative interstitial gland cells between follicles (rectangle). Scale bars: left, 20 µm; right, 5 µm. (C) 3D reconstruction of follicles with surrounding labeled theca cells and interconnected interstitial gland cells in a 2-wk ovary. Left: 3D projection using Imaris software. Right: pseudo-colored annotation using the Imaris surface model. Green: oocytes. Red: Cyp17a1-expressing cells, including theca and interstitial gland cells. Scale bars = 20 µm. Representative images of 2 wk (D) and 5 wk (E) ovaries showing a dramatic increase in labeled theca cells, interstitial gland cells, and possibly other Cyp17a1-lineage cells in the ovarian medulla. Scale bars = 100 µm. Oil Red O staining (red) reveals a significant increase in lipid droplet storage in ovarian medullary cells from 2 wk (F) to 5 wk (G) ovaries. Scale bars = 100 µm. (H) Immunofluorescence staining of HSD3B1 and PLIN1 at 5 wk shows that genes required for steroidogenesis and lipid droplet storage are highly expressed in medullary cells. Scale bars = 50 µm. (I) Electron microscopy of a remodeled follicle reveals cells with abundant lipid droplets (LD) and characteristic mitochondria (mt) (boxed region magnified in right panel). Scale bar = 10 µm. (J) 3D reconstruction of a 5-wk ovary labeled by tdTomato expression activated by Cyp17a1-iCre. Left: 3D projection using Imaris software. Right: pseudo-colored using the Imaris surface model. Red: theca/interstitial network structure. Spheres: oocyte locations. Scale bars = 100 µm.
Figure 4—figure supplement 1
Expansion of steroidogenic theca cells and interstitial gland cells.
(A) Immunofluorescence staining of CYP17A1 protein in the ovary shows a significant increase in the number of theca cells and interstitial gland cells within the stroma from 3 to 5 wk. Scale bars = 50 µm. (B) Immunofluorescence staining of a Cyp17a1 lineage-traced ovary at 2 wk shows initial expression in theca cells and associated cells resembling interstitial gland cells (arrowheads). A few granulosa cells were labeled (arrows), as previously reported (Bridges et al., 2008). Scale bar = 50 µm. (C) Structurally, steroidogenic theca cells closely resemble interstitial gland cells. In the electron microscopy images, red lines delineate distinct zones enriched in interstitial gland cells (1), theca cells (2), and granulosa cells (3), with each region magnified below. Zone 1 (rectangle ‘1’) represents interstitial gland cells. Zone 2 (rectangle ‘2’) contains steroidogenic theca cells. Zone 3 (rectangle ‘3’) corresponds to granulosa cells. Abbreviations: nu = nucleus; LD = lipid droplet; mt = mitochondria. Scale bars are included in the images. (D) Cyp17a1 lineage-traced ovaries at 4 and 7 wk show a marked increase in the number of labeled theca cells and interstitial gland cells by 7 wk. Extensive signal is also observed in the rete ovarii (arrowhead), indicating Cyp17a1 expression, as well as in the ovarian epithelium (arrow). Scale bars = 200 µm.
Figure 5
Single-cell RNA sequencing of 2- to 6-wk ovaries.
(A) UMAP plot of ovaries from 2 to 6 wk, showing 31 initial clusters (c0–c30) represented in different colors. Sequencing data from different ages were merged before being analyzed in Seurat (see Methods for details; see Source data 1). (B) Ovarian cells from 2 to 6 wk contribute to almost every cluster in the UMAP plot. O2w: scRNA-seq data from 2-wk ovary. O3w: 3-wk ovary. O4w: 4-wk ovary. O5w: 5-wk ovary. O6w: 6-wk ovary. (C) Cell type groups are indicated by colored regions, with their relationship to the numbered initial clusters as shown in (A). Granulosa cells: c1, c2, c3, c5, c10, c12, c20, c21, c24, and c30; stromal cells: c0, c4, c7, c8, c9, c13, c17, and c19; theca cells: c16; hematopoietic cells: c23; endothelial cells: c22 and c27; pericytes: c18; smooth muscle cells: c26; epithelial cells: c25 and c28; low UMI clusters: c6, c11, c15, and c29; TBD (to be determined): c14. (D) Multi-violin plot of selected marker gene expression across different cell types. Y-axis: gene names with expression levels normalized for display. X-axis: cell types.
Figure 6 with 3 supplements
Further re-cluster analysis of sequencing data.
(A) UMAP plot of re-clustered granulosa cells from Figure 5A, showing 18 clusters (gc0–gc17). (B) The deduced ovarian follicle developmental stages are labeled and indicated by colored regions, with their relationship to the numbered clusters shown in panel A. Primordial: gc11; primary: gc3 and gc8; secondary: gc1, gc4, and gc9; antral: gc2, gc7, gc10, and gc13; remodeling: gc0 and gc6; mitotic: gc5, gc12, gc14, gc15, gc16, and gc17. (C) Dot plot of marker gene expression in granulosa cells across different follicle stages. Y-axis: cell types. X-axis: gene names with expression levels normalized for display. (D) Pseudotime trajectory analysis using Monocle 3 reveals two pathways in secondary follicles: Path 1 leads to further development into antral follicles, while Path 2 results in remodeling follicles. (E) UMAP plot of re-clustered theca and stromal cells from Figure 5A, showing 21 clusters (g0–g20). (F) The deduced mesodermal cell subgroups are labeled and indicated by colored regions, with their relationship to the numbered clusters shown in panel E. Hoxd9+: g0; Tagln+: g1, g8, g14, and g17; early theca cells (TCs): g2 and g11; late theca cells (TCs): g9; Cxcl12+: g3, g5, and g20; Cd24a+: g4; Col11a1+: g6 and g7; Col14a1+: g10 and g19; mitotic: g12, g13, g15, and g18; TBD (to be determined): g16. (G) Diagram of steroidogenic gene expression in theca and granulosa cells and their role in androgen and estrogen synthesis. Top: Synthesis of androgens in theca cells and estrogens in granulosa cells from cholesterol. Bottom: Estimated numbers of steroidogenic theca cells and interstitial gland cells expressing three key steroidogenic genes in the sequencing data—Cyp11a1, Hsd3b1, and Cyp17a1—are shown, along with those expressing all four essential genes (Cyp11a1, Hsd3b1, Cyp17a1, and Star). The total cell numbers are shown in the first row. From 2 to 5 wk, the theca cell population increases, followed by a decline at 6 wk. Meanwhile, the number and percentage of theca cells expressing all four essential genes progressively increase. In contrast, the estimated numbers of granulosa cells expressing Hsd3b1, Hsd17b1, and Cyp19a1 follow a fluctuating pattern, decreasing from 2 to 4 wk before rising again at 5 wk. Notably, 2 wk corresponds to the time of mini-puberty. (H) Estimated numbers of granulosa cells in follicles across developmental subclasses: primordial, primary, secondary, antral, remodeling, and mitotic cycling.
Figure 6—figure supplement 1
Analysis of single-cell RNA sequencing.
(A) nFeature_RNA of the scRNA-seq data shows that clusters 6, 11, 15, and 29 have low UMI counts. (B) Highlighting cluster 14 reveals its dispersed distribution. (C) Distribution of samples across clusters shows that cells from each sample contribute to all clusters (orig.ident). (D) Feature plots of marker gene expression from the reanalysis of mesenchymal cells, including late theca cells (Cyp11a1), early theca cells (Hhip), Tagln+ fibroblasts (Tagln), Hox9+ fibroblasts (Hox9), Col11a1+ fibroblasts (Col11a1), Cxcl12+ fibroblast (Cxcl12), Col14a1+ fibroblasts (Col14a1), Cd24a+ fibroblasts (Cd24a), and mitotic cells (Mki67).
Figure 6—figure supplement 2
Analysis of other cell types.
(A) Expression plots of representative genes highly expressed in cluster 22, identified as endothelial cells in blood vessels, including Esam, Flt1, and Cd93. Red boxes in the top panel are enlarged in the bottom panel. (B) Expression plots of representative genes highly expressed in cluster 27, identified as endothelial cells in lymphatic vessels, including Ccl21a, Lyve1, and Prox1. Red boxes in the top panel are enlarged in the bottom panel. (C) Expression plots of representative genes shared between pericytes and endothelial cells, including Ebf1, Emid1, and Apold1. Red boxes in the top panel are enlarged in the bottom panel. (D) Ridge plots of representative gene expression in epithelial cells from the ovarian surface epithelium (cluster 25) and the rete ovarii (cluster 28). (E) Further analysis of hematopoietic cells using re-clustering in Seurat identified 9 clusters across five samples. (F) The deduced hematopoietic cell subgroups are labeled and indicated by colored regions, including macrophages (clusters 0, 4, and 7), dendritic cells (cluster 3), neutrophils (cluster 5), B cells (cluster 1), T cells (cluster 2), NK cells (cluster 6), and mitotic cells (cluster 8). (G) Dot plot showing marker gene expression across different hematopoietic cell types.
Figure 6—figure supplement 3
Hormone receptor expression and XO mouse line.
(A) Expression levels of hormone receptors (Y-axis) across different cell clusters (X-axis). Androgen receptor (Ar) is expressed in nearly all cell types except hematopoietic cells. Estrogen receptor 1 (Esr1) is primarily expressed in theca cells (including interstitial gland cells), epithelial cells, and smooth muscle cells, whereas estrogen receptor 2 (Esr2) is expressed in granulosa cells as expected. No expression of follicle-stimulating hormone receptor (Fshr) is detected in any cell type, while luteinizing hormone/choriogonadotropin receptor (Lhcgr) is specifically expressed in theca cells. Progesterone receptor (Pgr) is present in smooth muscle cells but absent in granulosa cells. (B) The total follicle count in XO mice at 1 wk is significantly lower compared to control XX mice. Follicle counts were performed using Imaris software on whole-mount-stained ovaries. N = 3–5. **p < 0.01. (C) Activated follicles were fewer in XO mice than in XX mice at 1 wk. By 2 wk, the statistical difference between XO and XX mice was no longer observed. Follicle counts were performed using Imaris software on whole-mount-stained ovaries. N = 4–5. **p < 0.01; ns, not significant.
Figure 7
Model of wave 1 follicle development and function.
Wave 1 follicles undergo remodeling during the peri-pubertal period, expanding the population of androgen-producing theca and interstitial gland cells. Gradients between the medulla and cortex facilitate the sequential activation of primordial follicles, a process that continues into early adulthood. We refer to these sequentially activated follicles as boundary follicles (wave 1.5). These follicles, whose granulosa cells are derived from bipotential pre-granulosa (BPG) cells, remain dormant in the medulla and cortex until recruiting signals—either positive or negative—become available. Boundary follicles are typically recruited around 2 weeks of age and constitute the majority of developing follicles by 5 weeks. In parallel, wave 1 follicles remodel to support androgen production through their remaining theca cells and associated interstitial gland cells.
Videos
Video 1
Whole ovary from a 1-wk mouse.
The ovary was whole-mount stained, cleared, and imaged using a Leica SP8 inverted microscope with a 20× objective. The 3D projection was generated in Imaris. Gray spheres indicate wave 1 follicles annotated using the Imaris spot model. DDX4: green; DAPI: white. Scale bar = 100 µm.
Video 2
Partial ovary from a 1-wk mouse.
A field of the ovary was imaged with a Leica SP8 inverted microscope using a 40× objective after whole-mount staining and clearing. The Imaris surface model was used to reconstruct oocyte contours. DDX4: green; DAPI: white. Scale bar = 50 µm.
Video 3
Whole ovary from a 2-wk mouse.
The ovary was whole-mount stained, cleared, imaged with a Leica SP8 inverted microscope, and projected in Imaris. Gray spheres indicate wave 1 follicles using the Imaris spot model. DDX4: green; DAPI: white. Scale bar = 200 µm.
Video 4
Partial ovary from a 2-wk mouse.
The ovary was imaged with a Leica SP8 inverted microscope using a 40× objective after whole-mount staining and clearing. The Imaris surface model shows oocyte contours. DDX4: green; DAPI: white. Scale bar = 50 µm.
Video 5
Partial ovary from a 3-wk mouse.
The ovary was whole-mount stained, cleared, imaged with a Leica SP8 inverted microscope using a 10× air objective, and projected in Imaris. DDX4: green; DAPI: white. Scale bar = 200 µm.
Video 6
Partial ovary from a 3-wk mouse.
A field of the ovary was imaged using a Leica SP8 inverted microscope with a 40× objective after whole-mount staining and clearing. The Imaris surface model shows oocyte contours. DDX4: green; DAPI: white. Scale bar = 70 µm.
Video 7
Partial ovary from a 4-wk mouse.
The ovary was whole-mount stained, cleared, imaged with a Leica SP8 inverted microscope using a 20× immersion objective, and projected in Imaris. DDX4: green; DAPI: white. Scale bar = 300 µm.
Video 8
Partial ovary from a 4-wk mouse.
A field of the ovary was imaged with a Leica SP8 inverted microscope using a 40× objective after whole-mount staining and clearing. The Imaris surface model shows oocyte contours. DDX4: green; DAPI: white. Scale bar = 50 µm.
Video 9
Partial ovary from a 5-wk mouse.
The ovary was whole-mount stained, cleared, imaged with a Leica SP8 inverted microscope using a 20× immersion objective, and projected in Imaris. DDX4: green; DAPI: white. Scale bar = 300 µm.
Video 10
Partial ovary from a 5-wk mouse.
A field of the ovary was imaged using a Leica SP8 inverted microscope with a 40× objective after whole-mount staining and clearing. DDX4: green; DAPI: white. Scale bar = 50 µm.
Tables
Table 1
Summary of whole-mount analyses of C57BL/6J ovaries.
| Postnatal age | Total follicles/ovary | Antral follicles/ovary | Wave 1follicles/ovary | Reference files | Ovaries analyzed |
|---|---|---|---|---|---|
| P5–P7 | 3822 ± 838 | 0 | 278 ± 17 | Video 1.mp4 Video 2.mp4 | 4 |
| P10–P14 | 3304 ± 926 | 0 | 293 ± 36 | Video 3.mp4 Video 4.mp4 | 3 |
| 3 wk | 3089 ± 669 | 23 ± 2 | ND | Video 5.mp4 Video 6.mp4 | 3 |
| 4 wk | ND | 35 ± 2 | ND | Video 7.mp4 Video 8.mp4 | 4 |
| 5 wk | ND | 24 ± 4 | ND | Video 9.mp4 Video 10.mp4 | 5 |
-
Values are presented as mean ± SD. Data were collected from at least two litters, and the experiment was repeated once. ND, not determined.
Additional files
-
Source data 1
Summary of general information for all five samples (2wk to 6wk).
Excel file containing sample metadata, including sample code, number of reads, number of cells, average genes per cell, average UMI per cell, total detected genes, reference transcriptome used for alignment, and sample name.
- https://cdn.elifesciences.org/articles/107352/elife-107352-data1-v1.xlsx
-
Source data 2
Average gene expression in 31 clusters (0 to 30) from five samples (2wk to 6w).
Excel file with average expression values of genes generated using Seurat. Column A: gene name. Row 3: cluster and sample name (e.g., 0_O2w, where ‘0’ indicates cluster 0 and ‘O2w’ indicates the 2-wk ovary sample).
- https://cdn.elifesciences.org/articles/107352/elife-107352-data2-v1.xlsx
-
Source data 3
Average gene expression of re-clustered granulosa cells and Metascape analysis of selected clusters.
Excel file with multiple sheets: (1) Expression_2wk_6wk_Granulosa: average expression values of genes in re-clustered granulosa cells (gc0–gc17) from five samples, generated using Seurat. Column A: gene name. Row 3: cluster and sample name. (2) gc0_Metascape: Gene Ontology (GO) enrichment analysis of cluster gc0 (Metascape). (3) gc6_Metascape: GO enrichment analysis of cluster gc6 (Metascape). (4) gc11_Metascape: GO enrichment analysis of cluster gc11 (Metascape).
- https://cdn.elifesciences.org/articles/107352/elife-107352-data3-v1.xlsx
-
Source data 4
Average gene expression of re-clustered mesenchymal cells (g0–g20).
Excel file containing average expression values generated using Seurat. Column A: gene name. Row 3: cluster and sample name (e.g., 0_O2w, where ‘0’ indicates cluster 0 and ‘O2w’ indicates the 2-wk ovary sample).
- https://cdn.elifesciences.org/articles/107352/elife-107352-data4-v1.xlsx
-
Source data 5
Top 250 preferentially expressed genes in initial and re-clustered mesenchymal cell clusters.
Excel file with three sheets:(1) Top250_Initial: top 250 genes (Column G) in initial clusters c0–c30 (Column F). (2) Top250_Granulosa: top 250 genes (Column G) in re-clustered granulosa cell clusters gc0–gc17 (Column F) . (3) Top250_Mesenchymal: top 250 genes (Column G) in re-clustered mesenchymal cell clusters g0–g20 (Column F) .
- https://cdn.elifesciences.org/articles/107352/elife-107352-data5-v1.xlsx
-
Source data 6
Cell numbers in initial and re-clustered clusters from five samples (2wk to 5wk).
Excel file with three sheets: (1) Cells_in_initial_clusters: cell numbers in initial clusters (c0–c30) (Column A). (2) Cells_in_granulosa_recluster: cell numbers in re-clustered granulosa cell clusters (gc0–gc17) (Column A). (3) Cells_in_mesenchymal_recluster: cell numbers in re-clustered mesenchymal cell clusters g0–g20 (Column A). Row 3: sample names.
- https://cdn.elifesciences.org/articles/107352/elife-107352-data6-v1.xlsx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/107352/elife-107352-mdarchecklist1-v1.pdf
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Distinct waves of ovarian follicles contribute to mouse oocyte production
eLife 14:RP107352.
https://doi.org/10.7554/eLife.107352.2
