The European roe deer (Capreolus capreolus) was the first mammal in which embryonic diapause has been described. While diapause is characterized by a complete developmental arrest in some species, roe deer blastocysts show a very slow, yet continuous growth. To date, it is neither known whether this growth is accompanied by developmental progression nor whether it is uniform in both, the trophectoderm (TE) and the inner cell mass (ICM). We collected roe deer blastocysts during the regular hunting season from September 2018 to January 2019, and quantified the fraction of cells expressing the proliferation marker Ki67 by immunofluorescence and light-sheet microscopy. We found that the cell number increased from around 300 cells in September to over 20'000 cells per blastocyst in December before elongation occured. Concurrently, we observed considerable morphological changes, i.e. cavity formation and transition to a disk-like shape of the inner cell mass. During diapause, less than 10% of all cells displayed positive Ki67 staining. Strikingly, the relative increase in cell number was lower in the ICM compared to the TE, whereas the fraction of Ki67 positive cells appeared to be lower in the TE than in the ICM. Our findings thus confirm that roe deer blastocysts display developmental progression in the course of diapause. We hypothesize that while the overall duration of the cell cycle is longer in the ICM than in the TE, the fractional distribution of cell cycle phases differs, with TE cells having a longer G1 phase than cells of the ICM.

Embryonic diapause in mammals leads to a reversible developmental arrest. While completely halted in many species, European roe deer (Capreolus capreolus) embryos display a continuous deceleration of proliferation. During a 4-mo period, the cell doubling time is 2 to 3 wk. During this period, the preimplantation blastocyst reaches a diameter of 4 mm, after which it resumes a fast developmental pace to subsequently implant. The mechanisms regulating this notable deceleration and reacceleration upon developmental resumption are unclear. We propose that amino acids of maternal origin drive the embryonic developmental pace. A pronounced change in the abundance of uterine fluid mTORC1-activating amino acids coincided with an increase in embryonic mTORC1 activity prior to the resumption of development. Concurrently, genes related to the glycolytic and phosphate pentose pathway, the TCA cycle, and one carbon metabolism were up-regulated. Furthermore, the uterine luminal epithelial transcriptome indicated increased estradiol-17β signaling, which likely regulates the endometrial secretions adapting to the embryonic needs. While mTORC1 was predicted to be inactive during diapause, the residual embryonic mTORC2 activity may indicate its involvement in maintaining the low yet continuous proliferation rate during diapause. Collectively, we emphasize the role of nutrient signaling in preimplantation embryo development. We propose selective mTORC1 inhibition via uterine catecholestrogens and let-7 as a mechanism regulating slow stem cell cycle progression.

Progesterone (P4) plays a pivotal role in maintenance of pregnancy in many mammalian species. Species-specific P4 metabolites have been shown to function as primary acting progestogen and the receptor binding capacity varies between species. The European roe deer (Capreolus capreolus) displays a 4-5 month period of embryonic diapause, which decouples fertilization from implantation. The majority of roe deer have two corpora lutea that secrete P4. No changes in P4 concentrations have been observed during pre-implantation embryo development. As 5α-DHP is known to play a major role during pregnancy in elephants and horses, we hypothesized that 5α-DHP functions as additional progestogen facilitating embryo reactivation. The profile of 11 progestogens was quantified in roe deer plasma over the course of diapause and resumption of embryo development including P4, 3α- and 3β-DHP, 20α- and 20β-DHP, 5α- and 5β-DHP, 3α,5α- and 3α,5β-THP, as well as 3β,5α- and 3β,5β-THP. While P4 was most abundant during diapause and resumption of development, 20α-DHP was the most abundant P4 metabolite. This is different than in pregnant elephants, where 5α-DHP was most abundant, and the luteal phase in cattle, where 3α,5α-THP was most abundant. With the exception of a weak correlation of 3β,5α-THP, none of the progestogens significantly correlated with embryonic development in the roe deer. Thus, plasma 5α-DHP does not seem to play a role in embryo reactivation. We propose that progestogens might contribute to priming the endometrium for supporting embryo development and preparation for implantation.

Embryonic developmental arrest, known as diapause, has been reported in more than 130 species. However, its mechanisms are still not completely understood. In the roe deer, the only known ungulate that exhibits this phenomenon, diapause lasts for approximately 5 months, starting after the rut period in mid-July to early August and ending with embryo elongation and implantation in December/January. Little is known regarding oocyte characteristics during this period. Here, we analysed the roe deer oocyte transcriptome as a model to understand diapause effects on oocyte features during embryonic developmental arrest and reactivation. During regular hunting, immature oocytes were obtained by ovary slicing from diapause and nondiapause stages, and classified according to morphological characteristics. Only oocytes with >2 layers of compact cumulus cells and cytoplasm from 30 hunted females were used for analyses. Immature oocytes were denuded and snap frozen. Additional oocytes were cultured in maturation medium for 20-24h. Matured oocytes with a present polar body were snap frozen. Two pools of 10 immature and mature oocytes for both diapause and nondiapause stages were included (at least 4 donors/pool). Oocyte pools were processed using the Smart-seq 2 single-cell protocol (Illumina Inc., San Diego, CA, USA) for full-length cDNA and library preparation. We performed RNA-seq on an Illumina sequencer. The obtained Fastq files were clipped and analysed with a locally installed version of the Galaxy platform. Sequences were mapped against the roe deer transcriptome (unpublished data) and annotated against human and bovine transcripts. Differentially expressed genes (DEG; false discovery rate <1%) were identified using EdgeR (https://bioconductor.org/packages/release/bioc/html/edgeR.html) in immature (IMM) and mature (M) oocytes from diapause (D) and nondiapause (ND) stages. Additionally, to evaluate the effects of maturation on oocyte transcript abundancy, DEG between IMM and M oocytes for D and ND were identified. Multidimensional scaling resulted in clustering according to oocyte types. Gene ontology terms for biological processes were assigned using ToppCluster tools. A total of 23066, 23022, 22438, and 22532 transcripts were detected for IMM D, M D, IMM ND and M ND oocytes, respectively. Using false discovery rate filtering, we found 333 and 288 DEG in immature and mature oocytes, respectively. Furthermore, maturation changed the oocyte expression profiles during D (2233 DEG) and ND periods (2589 DEG). Gene ontology classification revealed that most of the DEG in immature oocytes were involved in macromolecule catabolic process and oxidation-reduction process. For mature oocytes, DEG were mostly related to regulation of chromosome condensation and ER and Golgi vesicle-mediated transport. Additionally, the top 20 DEG from IMM D v. M D oocytes were related to mRNA splicing, mRNA metabolic process, and mRNA processing. These processes were not identified in the top 20 DEG from IMM ND v. M ND oocytes. These preliminary results suggest that oocyte transcriptome analysis could disclose new pathways implicated in oocyte competence and embryonic developmental arrest and activation stages.