HH5 Double-Carrier Embryos Fail to Progress through Early Conceptus Elongation

Massive genotyping in cattle has uncovered several deleterious haplotypes that cause pre-term mortality. Holstein Haplotype 5 (HH5) is a deleterious haplotype present in the Holstein Friesian population that involves the ablation of the Transcription Factor B1 mitochondrial (TFB1M) gene. The developmental stage at which HH5 double-carrier (DC, homozygous) embryos or fetuses die remains unknown and this is a relevant information to estimate the economic losses associated to the inadvertent cross between carriers. To determine if HH5 DC survive to maternal recognition of pregnancy, embryonic day (E)14 embryos were flushed from super-ovulated carrier cows inseminated with a carrier bull. DC E14 conceptuses were recovered at Mendelian rates but they failed to achieve early elongation, as evidenced by a drastic (>26-fold) reduction in the proliferation of extraembryonic membranes compared with carrier or non-carrier embryos. To assess development at earlier stages, TFB1M knockout (KO) embryos -functionally equivalent to DC embryos-were generated by CRISPR technology and cultured to the blastocyst stage -Day (D)8-and to the early embryonic disc stage -D12-. No significant effect of TFB1M ablation was observed on the differentiation and proliferation of embryonic lineages and relative mtDNA content up to D12. In conclusion, HH5 DC embryos are able to develop to early embryonic disc stage but fail to undergo early conceptus elongation, required for pregnancy recognition.


INTRODUCTION
Multiple haplotypes (alleles) involved in cattle reproductive performance have been uncovered by massive genotyping projects, including several deleterious haplotypes, recessive lethal alleles causing pre-term mortality (VanRaden et al., 2011).The cross between individuals carrying the same deleterious haplotype is avoided to evade the fertility problems associated to the one out of 4 odd of generating an unviable double carrier (DC) embryo.However, although heifers genotyping is currently under expansion (VanRaden, 2020), the carrier status of most heifers worldwide is still unknown and carrier bulls are still in use.The economic loss associated to the generation of DC embryos during the inadvertent cross between carrier individuals varies greatly depending on when DC embryos or fetuses arrest their development.If DC embryos fail to progress through maternal recognition of pregnancy, the economic loss will be similar to a failed insemination, but if they progress through maternal recognition of pregnancy, there would be a significant increase in open days and if they progress beyond implantation, the risk of uterine pathologies will increase further the economic impact (Schutz et al., 2016).
Holstein Haplotype 5 (HH5) is a deleterious allele present in Holstein Friesian population at an estimated carrier frequency of 4.4-5.5% (Cole et al., 2018;Schutz et al., 2016).HH5 consists of a large 138 kbp deletion on chromosome 9 flanked by bovine long interspersed nuclear elements (LINEs) Bov-B and L1ME3, which suggests that the haplotype was caused by a homologous recombination/deletion event (Schutz et al., 2016).The deletion ablates the gene Transcription Factor B1 Mitochondrial (TFB1M), encoding for a protein that participates in mitochondrial translation though the demethylation of 2 adjacent adenines in a stem-loop structure close to the 3′end of the small subunit rRNA (McCulloch et al., 2002), apparently required for normal function and integrity of the mammalian mitochondrial ribosome (Metodiev et al., 2009;Shoubridge, 2009).

HH5 Double-Carrier Embryos Fail to Progress through Early Conceptus Elongation
A. Pérez-Gómez, 1 JG.Hamze, 1,2 I. Flores-Borobia, 1 B. Galiano-Cogolludo, 1 I. Lamas-Toranzo, 1,2 L. González-Brusi, 1,2 P. Ramos-Ibeas, 1 and P. Bermejo-Álvarez 1 * Pregnancy losses caused by HH5 have been estimated to occur before 60 d of gestation based on the reproductive data collected at a genotyping program (Cooper et al., 2013), but given that data on embryo survival is not collected at earlier stages, it is unclear when the developmental defects associated to HH5 arise.From conception to implantation, bovine embryos undergo a series of critical developmental events (Perez-Gomez et al., 2021).Following fertilization, the zygote undergoes a series of symmetrical mitotic divisions until a first lineage differentiation event gives rise to the inner cell mass and the trophectoderm, resulting in the formation of a blastocyst around embryonic day (E) 7. The inner cell mass differentiates further into the hypoblast -which will form the extra-embryonic membranes together with the trophectoderm-and the epiblast -which will form the embryo proper-.Following blastocyst hatching (~E9) and before implantation (E19-20 (Bazer et al., 2009)), bovine embryos undergo a series of developmental processes encompassed under the term conceptus elongation.At early elongation, the hypoblast migrates to cover completely the inner surface of the trophectoderm by E11 (Maddox- Hyttel et al., 2003) and both extra-embryonic membranes grow exponentially to be able to trigger maternal recognition of pregnancy by E15-16 (Forde et al., 2011;Northey and French, 1980).Concomitantly, the epiblast will form an embryonic disc (ED) that starts gastrulation around E14-15 (van Leeuwen et al., 2015).Herein, we have conducted a series of in vivo and in vitro observations to identify when HH5 DC embryos arrest their development.

MATERIALS AND METHODS
The development of HH5 DC embryos was initially evaluated at E14, by collecting in vivo elongated bovine conceptuses.The effect of HH5 at earlier stages of development (Days (D) 8 and D12 of in vitro culture) was investigated by generating TFB1M KO embryos, functionally equivalent to HH5.

In vivo conceptus collection
Cattle embryos were collected at a commercial cattle farm following conventional protocols employed for reproductive management (Ramos-Ibeas et al., 2020), involving a superovulation protocol detailed in Figure S1.Animals -housed and managed under conventional farm conditions-were selected based on the genotyped population of CONAFE and those available for superovulation (i.e., nonlactating) at one collaborating farm were used.Three stimulated HH5 carrier cows were inseminated with semen of a carrier bull of proven fertility and embryos were collected from the uterus 14.5 d after first insemination (E14) by non-surgical flushing using a size 12 Luer catheter (Minitube, Spain) positioned in the uterine horn and BoviHold (Minitube, Spain) as flushing solution.Immediately after collection whole embryos were fixed in 4% paraformaldehyde (PFA) for 15 min at room temperature (RT) and washed and stored in PBS supplemented with 1% of bovine serum albumin (PBS-1% BSA) at 4°C until analysis.Uterine debris was collected to obtain DNA that was genotyped as described below to confirm that all carrier cows contained HH5 haplotype.

Generation of CRISPR components for oocyte microinjection
Capped polyadenylated Cas9 mRNA was in vitro transcribed by mMESSAGE mMACHINE T7 ULTRA kit® (Life Technologies) using the plasmid pMJ920 (Addgene 42234) linearized with BbsI and treated with Antarctic phosphatase (NEB) and purified using MEGAClear kit (Life Technologies) (Bermejo-Alvarez et al., 2015).One single guide RNA (sgRNA) against the first exon of TFB1M was designed using CRISPOR (Concordet and Haeussler, 2018) and synthesized and purified using Guide-it sgRNA In Vitro Transcription Kit® (Takara) using a primer containing T7 promoter and the sgRNA sequence (Table 1).

In vitro production of TFB1M KO embryos
Genome edited bovine embryos were produced in vitro following protocols previously described (Lamas-Toranzo et al., 2019).Briefly, immature cumulus-oocyte complexes (COCs) were obtained by aspirating 2-8 mm follicles from bovine ovaries collected at local slaughterhouses from animals of mixed breeds, predominantly beef cows.COCs were selected based on conventional morphological criteria and matured for 24 h in groups of 50 in 4-well dishes containing TCM-199 medium supplemented with 10% (vol/vol) fetal calf serum (FCS) and 10 ng/ml epidermal growth factor (EGF) at 38.5°C and 5% CO 2 in the air with humidified atmosphere.Matured oocytes were denuded by vortex for 3 min in 1 mL of 300 µg/ml hyaluronidase and randomly divided in 2 groups for microinjection.One group was microinjected with a solution of Cas9 mRNA and sgRNA against TFB1M (C+G group) at a concentration of 300 and 100 ng/µL, respectively, and another with Cas9 alone at 300 ng/µL, serving as an injection control (C group).Cytoplasmic microinjection was performed on denuded oocytes using a filament needle under a Leica DMi8 inverted microscope assisted by Femptojet (Eppendorf).Oocyte microinjection allows Pérez-Gómez et al.: HH5 impairs conceptus elongation for an early delivery of CRISPR components achieving edition in both parental genomes while minimizing the appearance of mosaicism (Lamas-Toranzo et al., 2019).Following microinjection, oocytes from both groups were kept separated throughout the next fertilization and culture steps.
Immediately following microinjection, in vitro fertilization (IVF) was carried out with frozen-thawed spermatozoa from a single non-carrier stud bull selected through a gradient of 40-80% Bovipure (Nidacon Laboratories AB, Göthenborg, Sweden).Spermatozoa were co-incubated with 30-50 microinjected oocytes at a final concentration of 10 6 spermatozoa/ml in TALP medium supplemented with 10 mg/ml heparin, in 4-well plates at 38.5°C in an atmosphere of 5% CO 2 and maximum humidity.At approximately 20 h postinsemination (hpi), presumptive zygotes were vortexed for 30 s to remove the spermatozoa attached to de zona pellucida and cultured in groups of 20-25 in 25 µL droplets of synthetic oviduct fluid (SOF) (Holm, Booth et al., 1999), supplemented with 5% FCS under mineral oil.Culture took place at 38.5°C in an atmosphere of 5% CO 2 , 5% O 2 , and 90% N 2 with maximum humidity.Cleavage and blastocysts rates were evaluated at 48 hpi and 7 d post-insemination, respectively.Day (D) 8 blastocysts were fixed in 4% PFA during 15 min at RT and stored in PBS-1% BSA as described above.
Post-hatching culture was carried out following an optimized protocol described in (Ramos-Ibeas et al., 2023).Briefly, D7 blastocysts were cultured in 500 µL of N2B27 medium (1:1 Neurobasal and DMEM/F12 medium supplemented with 1X penicillin/streptomycin, 2 mM glutamine, 1x N2 and 1x B27 supplements; Thermo Fisher Scientific).Post-hatching culture took place at 38.5°C in a water-saturated atmosphere of 5% CO 2 , 5% O 2 , and 90% N 2 and half of the culture medium was replaced every second day.Embryos remained in culture until D12, when embryo survival was analyzed (alive embryos were able to maintain the blastocoel, whereas dead embryos collapsed).Surviving embryos were fixed as described above.

Lineage development analysis by immunohistochemistry (IHC)
D8 and D12 (produced in vitro) and E14 (collected in vivo) embryos fixed as described above were washed in PBS-1% BSA and permeabilized in 1% Triton X-100 in PBS for 15 min at RT and blocked in 10% FCS-0.02Tween 20 in PBS for 1 h at RT. Subsequently, specimens were incubated overnight at 4°C with primary antibodies to detect trophectoderm (CDX2, Biogenex MU392A-UC 1:100 dilution), hypoblast (SOX17, R&D AF1924, 1:100 dilution), and epiblast (SOX2, Invitrogen 14-9811-80, 1:100 dilution) cells.After 4 washes in PBS-1% BSA, embryos were incubated in the appropriate secondary Alexa-conjugated antibodies (Donkey anti-rat IgG Alexa Fluor TM 488, Donkey anti-goat IgG Alexa Fluor TM 555 and Donkey anti-mouse IgG Alexa Fluor TM 647; Life Technologies) and DAPI for 1 h at RT, followed by 4 washes in PBS-1% BSA.Finally, embryos were mounted on PBS-1% BSA microdrops made by drawing circles with a PAP pen (Kisker Biotech GmbH) on a coverslide (Bermejo-Alvarez et al., 2012).Microdrops were covered by an incubation chamber (Sigma Z359467) to prevent embryo crushing.Embryos were imaged at a structured illumination equipment composed by a Zeiss Axio Observer microscope coupled to ApoTome.2 (Zeiss).Following image acquisition, embryo diameter was measured to determine embryo size and lineages development was analyzed.Total, CDX2+ and SOX2+ cell numbers were manually counted in D8 blastocysts using the ZEN software (Zeiss).In D12 embryos, hypoblast migration was determined by the ex- tension of the SOX17+ hypoblast layer through the inner surface of the CDX2+ trophectoderm, and epiblast survival was identified by the presence of SOX2+ cells in the embryo, whereas ED-like formation was identified by the presence of a compact structure containing SOX2+ cells.CDX2+ and total cells were not counted in D12 embryos due to their large size (>1000 cells), instead embryo diameter was measured as a proxy of the proliferation of extra-embryonic membranes.

Embryo genotyping
Embryo genotyping was performed following fixation and image analysis.E14 conceptuses, generated by the cross of carrier individuals, were genotyped employing the 3 primers strategy described in (Schutz et al., 2016) that employs a single reverse primer, a forward primer binding to a sequence flanking HH5 deletion and another forward primer binding to a sequence within HH5 deletion, that results in a 442 bp product in the wildtype (WT) allele and a 256 bp product in HH5 allele (Table 1).A fragment of each conceptus was dissected and placed at the bottom of a 0.2 mL PCR tube and stored at −20°C until analysis.Samples were digested with 15 µL of Arcturus Picopure DNA extraction solution (ThermoFisher Scientific) following incubation at 65°C for 1 h and inactivation at 95°C for 10 min.PCR was performed using 2 µL of the lysate in a 25 µL PCR reaction (GoTaq Flexi, Promega).PCR conditions were as follows: 94°C for 2 min; × 35 (94°C for 20 s, 60°C for 30 s, 72°C for 30 s); 72°C for 5 min; hold at 8°C PCR products were subjected to agarose gel electrophoresis to identify WT, HH5 single-carrier (SC) and doublecarrier (DC) conceptuses.
49 D8 and 25 D12 embryos, generated following CRISPR-based genome edition in C+G group, were genotyped by Deep Sequencing (miSeq, Illumina) to identify all alleles harboured by each specimen (Lamas-Toranzo et al., 2019).To that aim, specimens were stored and digested as described for E14 fragments but employing a reduced amount (8 µL) of Arcturus Picopure DNA extraction solution.A first PCR was performed to amplify the sequence containing CRISPR target site adding Illumina adaptors using the primers detailed in Table 1 and adding 3 µL of lysate in a 25 µL PCR reaction (GoTaq Flexi, Promega).PCR conditions were as follows: 96°C for 2 min; × 32 (96°C for 20 s, 60°C for 30 s, 72°C for 30 s); 72°C for 5 min; hold at 8°C.PCR product was purified using AMPure XP beads (Beckman Coulter TM ) following manufacturer's recommendations.A second 50 µL PCR was performed with Nextera XT Index Kit v2 primers (Illumina) to add barcodes identifying each specimen, using 5 µL of each purified PCR product as a template.PCR condi-tions were as follows: 95°C for 3 min; × 8 (95°C for 30 s, 55°C for 30 s, 72°C for 30 s); 72°C for 5 min; hold at 8°C.PCR products from different embryos were purified with AMPure XP beads, pooled into a 4 nM library and sequenced at miSeq (Illumina).Reads (>1000/embryo) were aligned with the WT sequence using the BWA mem aligner (v.0.7.17, (Li and Durbin, 2009)), sorted and indexed with a pipeline of different tools from the SAM tools package (v.1.16.1 (Li et al., 2009)) and variants were called with freebayes (Garrison and Marth, 2012).VCF files were processed with an in-house script to select only potential indels within the CRISPR probe editing region and passing the quality check filter; variants under Q30 or sequenced with a depth under 50 were discarded and indels should be present in both 5′-3′and 3′-5′reads.Results were confirmed visually checking each VCF file with the Integrative Genomics Viewer (Thorvaldsdottir et al., 2013).Following genotyping, embryos were classified as wild-type (WT, containing no mutated alleles, i.e., all embryos in C group), in-frame (IF, edited embryos containing at least one non-frame disrupting indel) and knockout (KO, containing only alleles formed by indels non-multiple of 3, i.e., frame-disrupting indels).

Mitochondrial DNA analysis
Mitochondrial DNA relative content was analyzed on 16 KO and 19 WT conceptuses.To that aim 2 quantitative PCRs (qPCR) were conducted on embryo samples digested by Arcturus Picopure DNA extraction solution as described above to amplify a mitochondrial sequence (COX1, cytochrome c oxidase subunit I) and a chromosomic sequence (PPIA, peptidylprolyl isomerase A) to adjust for the total amount of DNA present in the sample as described in (Martínez-Moro et al., 2022).Primer sequences are shown in Table 1.Samples were run in a 20 µL qPCR reaction (Gotaq qPCR, Promega, Madison, WI USA) in a MIC thermocycler (BioMolecular Systems Upper Comera, Australia).PCR conditions were optimized to achieve efficiencies close to I and then the comparative cycle threshold method was used to obtain relative values following 2 -ΔΔCq calculation, as described in (Schmittgen and Livak, 2008).

Statistical analyses
Data were analyzed using GraphPad Prism (Graph-Pad Software, San Diego, CA, USA) and SigmaStat (Systat Software, San Jose, CA, USA) packages.Chisquared test was used to analyze the differences in complete hypoblast migration, epiblast survival and ED-like formation between groups.Differences in blastocyst rates were analyzed by t-test and differences in Pérez-Gómez et al.: HH5 impairs conceptus elongation embryo survival, embryo diameter, and cell numbers between the different genotypes were analyzed by Oneway ANOVA.When normality test failed, statistical differences were analyzed by non-parametric t-test (Mann-Whitney) or One-way ANOVA (Kruskal-Wallis test).Sample size estimation and statistical power analysis were calculated for a minimum power analysis of 80% and a significance level of 5% using the formula 2*SD 2 *(Z α/2 +Z β ) 2 /d 2 .

HH5 DC conceptuses show impaired development at E14
A total of 28 conceptuses were recovered at E14 from 3 superovulated HH5 carrier cows inseminated with a carrier bull.All conceptuses were recovered intact, as evidenced by closed ends at both extremes.Genotyping revealed that 7/28 were DC (Table 2), following the expected 1/4 Mendelian proportion and suggesting that no DC conceptuses were lost before the collection time.However, HH5 DC conceptuses were significantly smaller than their WT and SC counterparts, evidencing a severe impairment in the development of their extraembryonic membranes: the maximum length reached by a DC conceptus was 1 mm, whereas the average conceptus length in WT and SC was above 26 mm (Table 3).Although the number of conceptuses analyzed was modest, the striking variation in size resulted in a statistical power of 92% at a significance level of 5%, exceeding the conventional 80% employed to determine sample size.Following immunohistochemistry, epiblast (SOX2+), hypoblast (SOX17+) and trophectoderm (CDX2+) cells were detected in DC embryos (Figure 1).Hypoblast cells migrated covering the entire inner surface of the TE, and epiblast cells formed an embryonic disc in all conceptuses, irrespective of their genotype.However, embryonic discs were smaller in diameter in DC conceptuses compared with those of other groups (Table 3).No differences in conceptus length or any of the other parameters analyzed were observed between SC and WT conceptuses.

TFB1M KO embryos develop normally to D12
To assess the developmental effect of HH5 before E14, TFB1M KO embryos (functionally equivalent to HH5 DC embryos as both are unable to produce TFB1M protein) were generated in vitro.To that aim, a group of oocytes were microinjected with Cas9 mRNA and sgRNA against TFB1M gene (group C+G, partially composed by TFB1M KO embryos) and another with Cas9 mRNA (group C, an injection control only composed by WT embryos).Developmental rates (cleavage at 48 hpi and blastocyst rates) were similar between C+G and C groups, suggesting that TFB1M disruption did not impair embryo development to the blastocyst stage (Table 4).
Embryo genotyping of 49 blastocyst from C+G group revealed the presence of 9 KO embryos (harboring only KO alleles) and 40 edited IF embryos (containing at least one IF allele), resulting in a 100% edition efficiency and a 18% KO generation efficiency.IHC analysis did not identified a developmental failure caused by the ablation at that stage of development: KO blastocyst showed a clear blastocoel and lineage differentiation and proliferation was not affected, as no differences   2).As embryo development to the blastocyst stage was not altered by TFB1M ablation, a post-hatching in vitro culture system was employed to assess embryo development up to D12 (Ramos-Ibeas et al., 2020).Such in vitro system allows determining if embryos are able to undergo hypoblast proliferation and migration and to form an early embryonic disc.As occurring from D0 to D7, similar survival rates from D7 to D12 were observed in C and C+G groups (Table 5).Genome edition was observed in 21/25 D12 embryos analyzed in C+G group (84% edition efficiency) and 11/25 were KO (44% KO generation efficiency).IHC analysis was conducted on the 11 KO embryos and 10 IF embryos from C+G group and in 20 WT embryos including the 4 obtained in C+G group and 16 from C group, revealing no significant differences in any of the parameters analyzed between WT, IF and KO D12 embryos (Table 6, Figure 3).Relative mtDNA, assessed in 16 KO and 19 WT embryos, was not affected by TFB1M ablation (Table 6).

DISCUSSION
In this study, we have analyzed the impact of HH5 on preimplantation embryo development.This information is crucial to evaluate the economic impact of inadvertent crosses between HH5 carriers.The economic losses caused by pregnancy failures vary greatly depending on the stage of development attained by the unvi-able embryo (Wiltbank et al., 2016).A developmental impairment occurring before maternal recognition of pregnancy is roughly equivalent to a non-fertilizing insemination, as luteolysis will occur timely leading to normal estrous cyclicity.Embryonic losses beyond that point result in delayed luteolysis (i.e., increase in open days) and an increased risk for uterine pathologies associated to abortions if the unviable embryo achieves implantation.Herein, we have observed that HH5 DC embryos fail to progress to early conceptus elongation, before maternal recognition of pregnancy and thereby causing a minimal economic loss.
Uncovering the developmental period at which DC embryos arrest their development required dedicated investigation, as it cannot be precisely identified by the reproductive data collected in the context of genomic improvement programs given that pregnancy diagnose is annotated beyond Day 60 of gestation, i.e., at least 1.5 mo beyond implantation.In agreement with our observations, fertility losses caused by HH5 were estimated to occur before 60 d of gestation (Cooper et al., 2014), but it was unclear if HH5 DC embryos progressed through elongation and implantation.The dramatic reduction in the proliferation of the extraembryonic membranes observed in HH5 DC conceptuses right before maternal recognition of pregnancy is not compatible with the production of the interferon Tau (IFNT) required to signal pregnancy to the uterus and prevent luteolysis (Helmer et al., 1989;Imakawa et al., 1987).The degree of development attained by the extra-embryonic membranes (responsible for IFNT production) in E14 HH5 DC embryos is well below (<1 mm and spherical, roughly mimicking those collected at E11) the minimum 5 mm size limit where IFNT production becomes detectable in the uterine fluid and prevents luteolysis (Mann and Lamming, 2001).In this context, even if DC conceptuses could hypothetically slowly progress to further developmental stages, such development could not be achieved in vivo, as after luteolysis the uterine environment becomes unsuitable for embryo development.
HH5 DC embryos progressed normally up to preelongation stages, as evidenced by the lack of significant differences on any parameter analyzed up to D12 (roughly equivalent to E11 (Ramos-Ibeas et al., 2020)).According to a prior transcriptional analysis (Mamo et al., 2011), TFB1M relative expression is steadily reduced from E7 to E16, being the expression level at roughly halved by E16 compared with E7, which seems counterintuitive with the late (E11-E14) appearance of the phenotype associated to TFB1M ablation.A possible failure in mtDNA replication at D12 could have explained the developmental failure observed at further stages, but mtDNA relative content was comparable between KO and WT embryos.The apparently normal development to D12 may indicate that mitochondrial translation is not essential for early bovine embryo development, given the relevant role of TFB1M in the mammalian mitochondrial ribosome (McCulloch et al., 2002;Metodiev et al., 2009).The embryonic KO approach employed eliminates the synthesis of novel functional TFB1M mRNA from the oocyte onwards, but it does not erase the protein and mRNA already present in the oocyte.This mimics the situation of the naturally produced DC embryos, which derive from Hz oocytes containing functional TFB1M mRNA and protein but unable to produce functional TFB1M mRNA.Oocyte-stored mRNA are degraded around embryonic genome activation (Yang et al. 2024) -i.e., before the morula stage-, but TFB1M protein and other oocytestored proteins generated through TFB1M-mediated mitochondrial translation remain functional to further stages.Mitochondrial translation is required to generate the 13 proteins encoded by the mitochondrial genome, all involved in the mitochondrial respiratory   chain (Anderson et al., 1982).Mitochondrial respiration increases significantly as the early bovine embryo progresses through the blastocyst stage (Thompson et al., 1996), and therefore the lack of mitochondrial translation at early developmental stages could be compensated by the mitochondrial proteins already present in the oocyte, where mitochondria are remarkably abundant (Hyttel et al., 1988;Lamas-Toranzo et al., 2018).As development progresses through early elongation, there is an increase in both the embryonic mass and the associated energetic demands (Johnson et al., 2023;Ribeiro et al., 2016;Simintiras et al., 2021), leading to an increased need of active mitochondrial translation, ultimately causing the developmental arrest of TFB1M KO embryos.The differential effects of TFB1M ablation on specific lineages observed at E14 is also coherent with an altered mitochondrial respiration, as the proliferation of trophectoderm cells -the main driver of conceptus elongation-was more affected than hypoblast and epiblast proliferation.Trophectoderm metabolic rates are higher than those of the inner cell mass -the precursor of epiblast and hypoblast-in bovine embryos (Gopichandran and Leese, 2003), and mouse trophectoderm cells consume more oxygen, produce more ATP and contain a greater number of mitochondria than inner cell mass cells (Houghton, 2006).The higher demand for mitochondrial respiration in the trophectoderm compared with the hypoblast and epiblast precursors could explain the higher phenotypic effect of the ablation on that lineage.
The stage at which bovine TFB1M KO embryos arrest their development contrasts to that of mouse Tfb1m KO embryos (Metodiev et al., 2009), which develop to fetal stages (E8.5)showing severe developmental defects -neural clef open, and lack of optic disc and heart structure-.Mouse E8.5 corresponds to Theiler Stages 12-13 (Theiler, 1989), not reached by bovine embryos at E21 (Maddox-Hyttel et al., 2003), i.e., after implantation.The delayed developmental arrest of mouse Tfb1m embryos may be due to their considerably faster development and earlier implantation timing, and are in line with the critical developmental differences observed between mice and ungulate early embryo development (Berg et al., 2011;Bermejo-Alvarez et al., 2010;Carreiro et al., 2021).
In conclusion, bovine embryos lacking TFB1M (i.e., HH5 DC) develop normally up to pre-elongation stages, but fail to develop through the early elongation development required to trigger maternal recognition of pregnancy.

Table 3 .
Conceptus length and lineages development in WT, SC and DC E14 conceptuses

Table 4 .
Developmental rates in C and C+G groups up to D8 blastocysts N Microinjected oocytes Cleavage rate (mean ± s.e.m.) Blastocyst rate (mean ± s.e.m.)

Table 5 .
Survival rates of in D12 in vitro embryos from C and C+G groups 1 ED = embryonic disc; WT = wild-type; IF = in-frame; KO = knockout.No significant differences were found in any of the parameters analyzed.Mean ± s.e.m; One-way ANOVA for embryo diameter and SOX2+ cell number, Chi-squared test for complete hypoblast migration, epiblast survival and ED-like formation rates, p>0.05.