Serum anti-Müllerian hormone concentration as a diagnostic tool to identify testicular tissue in canine disorders of sexual development

Disorders of sexual development (DSD) may have their origin in alterations of the chromosomal, gonadal or phenotypic sex. Affected animals are usually presented because of ambiguous external genitalia, seldom because of reproductive disorders. Anti-Müllerian hormone (AMH) is secreted in the gonads with higher amounts in males than in females and can be used to identify gonadal tissue in sexually normally developed dogs. The aim of this study was to examine the diagnostic potential of serum AMH to identify testicular tissue in 11 dogs with DSD. The diagnostic procedures applied were: determination of the phenotypic sex (n = 11), genital ultrasound (n = 9), determination of the SRY gene (n = 11), karyogram (n = 6), gonadectomy (n = 11), pathohistology of the gonads (n = 10), serum AMH measurement (n = 11). 39 female dogs described in a previous study and 19 male dogs with a normal spermiogram served as controls for the AMH serum concentrations in sexually intact dogs. The 11 dogs with DSD were classified as 7 XY DSD and 4 XX DSD. Presumptive testes were obtained in 10 dogs and 1 dog had an ovotestis combined with a testis. Mean serum AMH values of the dogs with DSD were significantly higher (P < 0.001) than in male and female controls. The upper limit of the AMH test (≥ 23ng/ml) was reached in 6 dogs. High AMH concentrations have been described previously in cryptorchid dogs. 1 dog with a male phenotype and 2 with a female phenotype had AMH values within the range of the male controls, although all of them had cryptorchid testes. A Poodle, in which epididymis were identified but no definitive gonads, had an AMH concentration of the lower limit of the test (≤ 0.01 ng/ml), comparable to previously described castrated dogs. This study indicates that serum AMH levels are a useful diagnostic tool to identify testicular tissue in dogs with DSD and suggests the possible use of AMH to diagnose testicular dysgenesis.

202 Targeting Galactosyl-α-1,3-Galactose (αGal) Epitopes for Multi-Species Embryo Immunosurgery

Immunosurgical isolation of the inner cell mass (ICM) from blastocysts is based on complement-mediated lysis of antibody-coated trophectoderm (TE) cells. Conventionally, anti-species antisera, containing antibodies against multiple undefined TE cell epitopes, have been used as antibody source. We previously generated α-1,3-galactosyltransferase deficient (GTKO) pigs to prevent hyper-acute rejection of pig-to-primate xenotransplants. Because GTKO pigs lack galactosyl-α-1,3-galactose (αGal) but are exposed to this antigen (e.g. αGal on gut bacteria), they are expected to produce anti-αGal antibodies. In this study, we examined whether serum from GTKO pigs can be used as a novel antibody source for embryo immunosurgery. First, the presence of αGal epitopes in mouse (E3.5), rabbit (Day 4), pig (Day 6–7), and bovine (Day 7–8) blastocysts was examined by staining with fluorescein isothiocyanate (FITC)-conjugated BSI-B4 lectin (Sigma, St. Louis, MO, USA) that binds αGal. Expression of αGal epitopes on the surface of TE cells was detected in blastocysts of all examined species. Next, pig blastocysts were incubated with a medium containing GTKO pig serum. Swollen TE cells were observed in some of the blastocysts already after 2 min and, after 10 min, almost all TE cells of these blastocysts were completely destroyed. No lysis was recorded when the same experiment was done with wild-type pig serum, suggesting the presence of sufficient quantities of anti-αGal antibodies in GTKO serum to coat the TE cells and induce their complement-mediated lysis. Finally, GTKO serum was systematically tested for immunosurgery. Zona-free blastocysts of the species mentioned above were incubated with heat-inactivated GTKO pig serum for 1 h at 38°C. After washing, the blastocysts were labelled with Hoechst 33342 and TE was stained with FITC-conjugated concanavalin A (ConA) to distinguish the ICM from TE cells. Eventually, the blastocysts were individually incubated in complement solution for 30 to 40 min. Complement-mediated lysis of TE cells was efficiently induced in mouse, rabbit, pig, and bovine blastocysts (10/10, 7/7, 10/10, and 5/6, respectively), and intact ICM were successfully recovered from all species (100, 100, 60, and 80%, respectively). Double fluorescent staining with Hoechst 33342 and ConA clearly showed that the majority of isolated ICM was not contaminated with TE cells. Our study demonstrates that GTKO pig serum is a reliable source of antibodies targeting the αGal epitope of TE cells. Major advantages of using GTKO serum for embryo immunosurgery are (1) that it can be produced easily in large batches, thus reducing experimental variation; and (2) that it reacts with a large number of different species, except for humans, apes, and old world monkeys that lack αGal epitopes. Interesting applications include the preparation of TE and ICM for transcriptome profiling or chimeric embryo complementation experiments. This work is supported by the German Research Council (TR-CRC 127).

108 Single-Cell RNA Sequencing Reveals Blastomere Heterogeneity and Early Lineage Specification Events in Bovine Embryos During Major Embryonic Genome Activation

During early embryonic stages, gene products generated by the embryo acquire control over embryonic development. At the 8- to 16-cell stage, major embryonic genome activation (EGA) occurs in bovine embryos. Morphological observations, such as size of blastomeres and distribution of microvilli, suggest heterogeneity of individual cells already at this developmental stage. To study this heterogeneity on the transcriptome level, we performed single-cell RNA sequencing (scRNA-seq) of 161 blastomeres from 14 in vitro-produced bovine embryos at Day 2 and Day 3 post-fertilization. After removing the zona pellucida, blastomeres were mechanically separated in Ca2+- and Mg2+-free PBS, individually collected, and lysed. Complementary DNA libraries were prepared by the single cell RNA-barcoding and sequencing (SCRB-Seq) protocol. Exogenous RNA was added for quality control and cell specific barcodes and unique molecular identifiers (UMI) were used to enable pooling of libraries and to exclude PCR duplicates, respectively. After sequencing (Illumina HiSEqn 1500; 50 nt reads; Illumina Inc., San Diego, CA, USA), UMI were counted with the published Drop-seq pipeline (45,000 UMI on average per library) and cells with UMI count <2.000 were removed. Data were normalized based on UMI and non-supervised clustering analyses of single-cell data were performed (SC3 and M3Drop R packages). The transcriptome profiles of all individual cells were assigned to 6 clusters with specific sets of genes. Sorting cells according to their transcriptome profiles by the CellTree R package (Bioconductor; https://bioconductor.org/packages/release/bioc/html/cellTree.html) resulted in a linear pseudo-timeline. Furthermore, this tool identified 6 groups of genes (topics). Each of them showed an over-representation of distinct Gene Ontology (GO) terms; topic 1, “translation” and “cell division”; topic 2, GO terms involved in translation, RNA splicing and cell division; topic 3, “translation”; topic 4, “ATP synthesis coupled proton transport”; topic 5, “mitochondrial translational elongation”; topic 6, “organic hydroxyl compound transport”. Moreover, increased expression of PCDH10 (protocadherin 10) was observed in the biologically pseudo-ordered more advanced blastomeres. This gene is known to be predominantly expressed in the inner cell mass (ICM) at the blastocyst stage, suggesting that these cells might become ICM. In summary, our study reveals developmental heterogeneity and hints to early lineage specification events in bovine embryos at the time of major EGA.

62 BOVINE OCT4 (POU5F1) KNOCKOUT EMBRYOS FAIL DURING THE SECOND LINEAGE DIFFERENTIATION DUE TO LOSS OF NANOG

We generated a CRISPR/Cas9 mediated knockout (KO) of the OCT4 gene in adult fibroblasts, where biallelic deletion of a single nucleotide leads to a frameshift mutation. Through reconstruction of embryos by somatic cell NT, we were able to study the role of OCT4 during pre-implantation development until Day 7 in vitro. The presence of OCT4 protein was evaluated after immunofluorescent staining by confocal laser scanning microscopy of Day 5 morulae and Day 7 blastocysts; somatic cell NT embryos reconstructed from nontransfected cells of the same source served as control. Whereas control morulae expressed OCT4 in all cells, OCT4 KO morulae showed expression in only 67.8 ± 11.1% (mean ± SD, n = 6) of cells and overall intensity was decreased. By Day 7, no expression of OCT4 was detected in OCT4 KO blastocysts (n = 24), suggesting that maternal stores of the OCT4 protein had decayed. In contrast, control blastocysts (n = 20) showed OCT4 expression ubiquitously in both inner cell mass (ICM) and trophectoderm (TE). Simultaneously to the OCT4 staining, we differentially stained ICM and TE with the TE specific marker CDX2 and counterstained cell nuclei with 4′,6-diamidino-2-phenylindole. No significant differences between OCT4 KO Day 7 blastocysts and controls were detected in total cell numbers (89.6 ± 27.5 v. 96.3 ± 38) and percentage of CDX2 positive cells (50.7 ± 16.8% v. 59.0 ± 20.8%) (P > 0.05, mean ± SD, unpaired, two-tailed t-test). To analyse the role of OCT4 during the second lineage differentiation, we stained Day 5 morulae and Day 7 blastocysts for the epiblast and hypoblast specific markers NANOG and GATA6, respectively. In morulae, both markers were present and co-expressed in OCT4 KO and control embryos. By Day 7, control blastocysts (n = 6) already showed the typical salt and pepper distribution of NANOG and GATA6 positive cells, but expression was not mutually exclusive in all cells and also not restricted to ICM. OCT4 KO embryos lost all NANOG expression at Day 7 blastocyst stage (n = 8) and only stained positive for GATA6 in both TE and ICM. We conclude that OCT4 is not required for the quantitative allocation of cells to either the ICM or the TE during the first lineage differentiation, as total cell number and percentage of CDX2 positive cells was unchanged. Additionally, expression of NANOG seems to be OCT4 dependent and OCT4 KO embryos fail to establish the epiblast lineage—unlike mouse Oct4 KO embryos, where developmental failure was connected to loss of GATA6 expression during second lineage differentiation. This work was funded by the Bavarian Research Foundation (AZ-1031–12).

84 PAIRS OF BLASTOMERES FROM BOVINE DAY 5 MORULAE ARE ABLE TO CONTRIBUTE TO INNER CELL MASS AND TROPHECTODERM IN CHIMERIC EMBRYOS GENERATED BY AGGREGATION WITH TWO DAY 4 MORULAE

The multiplication of high-value embryos by chimera formation using asynchronic aggregation is a promising alternative to embryonic cell nuclear transfer. Single blastomeres from a donor embryo are aggregated with 2 host embryos, thus several chimeras can be constructed per donor embryo. Due to the advanced developmental stage, the donor blastomeres are likely to contribute to the inner cell mass (ICM) and later give rise to the embryo proper, whereas the host embryos form extra-embryonic tissues. To test if pairs of blastomeres from Day 5 morulae are able to form the ICM when aggregated with 2 Day 4 host embryos, we produced transgenic donor embryos carrying a fluorescent reporter gene (enhanced green fluorescent protein, eGFP) by using semen from an eGFP transgenic bull (Reichenbach et al. 2010 Transgenic Res. 19, 549–556) for in vitro fertilization and in vitro host embryos produced by a standard procedure. The zona pellucida of all embryos was removed by treatment with 1 mg mL–1 pronase. Donor embryos were assessed for eGFP expression by fluorescence microscopy and disaggregated by gentle pipetting after incubation in Mg2+- and Ca2+-free medium. Pairs of blastomeres were then placed between 2 host embryos and cultured individually in a well-of-the-well culture dish. On Day 6 after aggregation, fully developed blastocysts were assessed for eGFP fluorescence. In 3 replicates, n = 30 chimeras were produced by aggregation; 13 (43%) developed to blastocysts, of which 2 (15%) showed local eGFP expression in the ICM and 7 (54%) showed a generalized expression. From the results of this study we conclude that Day 5 morulae may be multiplied in an efficient manner by using the chimera formation technique, which makes this approach applicable to ex vivo-derived embryos. In future investigations we will study the effect of using donor blastomeres from either the inside or outside of the donor morula and test the use of tetraploid host embryos to increase the rate of blastocysts with the desired genotype in the ICM. Finally, we aim to introduce this multiplication approach to the production of genotyped embryos with a genomic estimated breeding value (gEBV) and intend to produce calves with identical gEBV.Funded by the Bavarian Research Foundation (AZ-1031–1).