Abstract
The hypothesis that colony-stimulating factor 2 (CSF2) plays a role in the preimplantation development of the bovine embryo was tested by evaluating consequences of inactivation of CSF2RA (the functional receptor in the embryo) for the development of embryos in utero. CRISPR/Cas9 was used to alter sequences on exon 5 and intron 5 of CSF2RA, Control embryos were injected with Cas9 mRNA only. Embryos > 16 cells at day 5 after insemination were transferred to synchronized recipient females in groups of 7–24. Embryos were flushed from the uterus 2 days later. The proportion of recovered embryos that developed to the blastocyst stage was lower for knockout embryos (39%) than for control embryos (63%). RNA sequencing of individual morulae and blastocysts indicated a total of 27 (morula) or 15 (blastocyst) differentially expressed genes (false discovery rate <0.05). Gene set enrichment analysis indicated that the knockout affected genes playing roles in several functions including cell signaling and glycosylation. It was concluded that signaling through CSF2RA is not obligatory for the development of the bovine preimplantation embryo to the blastocyst stage but that CSF2 signaling does enhance the likelihood that the embryo can become a blastocyst and result in specific changes in gene expression.
Lay summary
Development of the early embryo depends upon regulation by chemical signals produced by the uterus. One of these signals is a protein called colony-stimulating factor 2 (CSF2) that can affect the development of embryos in culture. To test whether CSF2 also regulates the embryo in the uterus, where development ordinarily occurs, we evaluated development in the uterus of embryos in which the receptor for CSF2 was disrupted. Embryos without the receptor gene were less likely to develop to the typical stage of development than control embryos and experienced some differences in the expression of specific genes. We conclude that CSF2 regulates embryonic development in the uterus.
Introduction
Development of the preimplantation embryo is regulated by cell-signaling molecules produced by the uterine endometrium called embryokines (Hansen & Tríbulo 2019). One of these molecules is colony-stimulating factor 2 (CSF2). Otherwise known as granulocyte-macrophage colony-stimulating factor, CSF2 is a pleiotropic glycoprotein produced by many cell types including T and B cells, natural killer cells, epithelial cells, endothelial cells, and fibroblasts (Ingelfinger et al. 2021). Originally described as a hematopoietic growth factor, its primary function in the immune system is as a regulator of phagocyte function (Ingelfinger et al. 2021). Secretion of CSF2 from most cells depends upon stimulation by cytokines or pathogen-associated molecular pattern molecules. However, the endometrium secretes CSF2 in the apparent absence of immune stimuli. In the cow, for example, CSF2 is present in the uterine fluid during the first 7 days after ovulation and can be localized to the endometrial epithelium and stroma (de Moraes et al. 1999, Tríbulo et al. 2018).
In vitro, CSF2 can increase the competence of the embryo to develop to the blastocyst stage in several species, including the bovine (de Moraes & Hansen 1997, Loureiro et al. 2009, Dobbs et al. 2013, Sosa et al. 2020), mouse (Robertson et al. 2001, Sjöblom et al. 2005), human (Sjöblom et al. 1999), pig (Kwak et al. 2012, Lee et al. 2013), and yak (Wen et al. 2017). Actions of CSF2 on development may be related to reducing cellular stress because CSF2 can decrease the number of apoptotic cells in the embryo of several species (Sjöblom et al. 2002, Loureiro et al. 2011, Shyam et al. 2020) and improve survivability after heat shock (Sosa & Hansen 2022) and cryopreservation (Sosa et al. 2020) in vitro. Moreover, many of the genes regulated by CSF2 in the embryo by CSF2 are associated with cellular protection (Chin et al. 2009, Zolini et al. 2020).
Exposure of the preimplantation embryo to CSF2 can impact its developmental trajectory. In the cow, for example, CSF2 treatment from day 5 to day 7 of development programmed trophoblast elongation so that embryo length at day 15 of pregnancy was affected by CSF2 in a sex-dependent manner (Dobbs et al. 2014). Treatment with CSF2 decreased the embryo length and accumulation of interferon tau in the uterus in female embryos but increased both measurements in male embryos (Dobbs et al. 2014). There have also been reports that blastocysts derived in the presence of CSF2 have increased competence to establish pregnancy after transfer to females, as reported in mouse (Sjöblom et al. 2005), human (Ziebe et al. 2013, Okabe-Kinoshita et al. 2022) and cattle (Loureiro et al. 2009, Denicol et al. 2014), although contradicting reports exist on this effect of CSF2 (Zhou et al. 2016, Tríbulo et al. 2017, Estrada-Cortes et al. 2021, Amaral et al. 2022). There are reports that CSF2 influences fetal and postnatal programming when introduced during the preimplantation period in the mouse (Sjöblom et al. 2005) and bovine (Kannampuzha-Francis et al. 2015, Siqueira et al. 2017, Estrada-Cortés et al. 2021).
Although CSF2 has been shown to affect the development of the preimplantation embryo in several species, most of these findings were obtained using embryos treated with CSF2 in culture. Such experiments do not reveal whether CSF2 regulates embryonic development in vivo, where the stresses of cell culture are absent, and many other cell-signaling molecules are present. The one experimental approach to evaluate the role of CSF2 in vivo was using mice that were genetically deficient in CSF2. Implantation rates were not reduced in CSF2-null females mated to CSF2-null or wildtype males. Nonetheless, pregnancies produced with matings of null males and females resulted in a reduction in fetal weight and increase in the proportion of implantation sites with resorbing or morphologically abnormal fetuses at day 17 (Robertson et al. 1999). In another experiment, blastocysts derived from CSF2-null parents recovered on day 4 of natural pregnancy or after superovulation had lower cell number than embryos from control mice (Robertson et al. 2001). These results would indicate that CSF2 is not essential for a successful pregnancy but does modify embryonic and placental development.
Here, we tested the hypothesis that CSF2 plays a role in the preimplantation development of the bovine embryo by evaluating whether the disruption of the CSF2 receptor gene in the embryo would alter the characteristics of the blastocyst derived from an embryo produced in vitro that developed in vivo from day 5 to day 7 of development. Unlike myeloid cells, where the CSF2 receptor is a dodecamer formed by equimolar ratios of CSF2RA and CSF2RB (Hansen et al. 2008), the preimplantation embryo does not express CSF2RB (Robertson et al. 2001, Sjöblom et al. 2002, Dobbs et al. 2013, Lee et al. 2013). Deletion of CSF2RA by CRISPR/Cas9 technology (Xiao et al. 2021) or inhibition of receptor binding by the addition of antibody to CSF2RA (Sjöblom et al. 2002) blocks the effect of CSF2 on the preimplantation embryo. Accordingly, it was reasoned that, if CSF2 regulates embryonic development in vivo, embryos in which CSF2RA was deleted using CRISPR/Cas9 would have altered patterns of development and gene expression after development in utero than control embryos. CSF2RA-deficient embryos were made in vitro and transferred to the uterus at day 5 of development. This day was chosen because the bovine embryo enters the uterus at day 4 or day 5 of development (Hackett et al. 1993).
Methods
In vitro oocyte maturation and fertilization
In vitro embryo production was carried out as described inStoecklein et al. (2021). A total of four replicates were performed. Each replicate consisted of a group of ~500 Holstein oocytes purchased from J.R. Simplot (Boise, ID, USA) that were shipped overnight in tubes containing oocyte maturation medium covered with mineral oil in a portable incubator at 38.5˚C. A total of 50 cumulus complex oocytes were placed in each tube. After maturation for 22 h, oocytes were rinsed with HEPES-TALP (Tyrode's Albumin-Lactate-Pyruvate) medium and placed in groups of up to 250 in a dish containing 1.7 mL of invitro fertilization (IVF)-TALP and fertilized in vitro with non-sorted semen from two Holstein bulls. Bulls were chosen because their spermatozoa had been shown to work well for in vitro fertilization. For each replicate, half of the oocytes were fertilized with one bull and half with the other bull. Gametes were co-incubated in a humidified atmosphere of 5% (v/v) CO2 for 18–20 h. After fertilization, putative zygotes were washed with HEPES-TALP medium and placed in a tube containing 200 μL HEPES-TALP medium to perform removal of cumulus cells from zygotes by agitation for 5 min using a vortex mixer.
CRISPR/Cas9-mediated gene targeting of CSF2RA and embryo culture
Putative zygotes (i.e. oocytes exposed to sperm) were subjected to CRISPR/Cas9-mediated gene targeting for CSF2RA, as previously described by Xiao et al. (2021). The CRISPR/Cas9 system was designed to introduce mutations in exon 5 and intron 5 of CSF2RA on BTA3 or induce the deletion of the exon 5 and intron 5. A total of 844 zygotes were injected with 40 ng/μL of each sgRNA for CSF2RA and 75 ng/μL Cas9 mRNA (knockout; KO), while another 831 zygotes (wildtype controls) were injected with 75 ng/μL Cas9 mRNA only (wildtype), respectively. Eight knockout blastocysts produced in vitro were genotyped to ensure the effectiveness of the CRISPR/Cas9 system. None of the embryos carried a wildtype allele, confirming the effectiveness of the CRISPR/Cas9 system. Specifically, embryos were either biallelically edited with an exon 5 deletion (n = 2), biallelically edited (n = 1) or edited in both alleles but where the alleles exhibited mosaicism (n = 5). Modifications introduced by the CRISPR/Cas9 system led to the formation of a premature stop codon, thereby inactivating the gene. The frequency and types of edits were consistent with our previous study using the identical CRISPR/Cas9 system (Xiao et al. 2021).
Following microinjection, embryos were washed in HEPES-TALP medium, and up to 50 putative zygotes were placed in a four-well dish containing 500 μL synthetic oviduct fluid bovine embryo 2 (Tríbulo et al. 2019) covered with 300 μL mineral oil in an incubator at 38.5˚C in a humidified atmosphere of 5% (v/v) CO2, 5% O2, and the balance N2 at 38.5°C. Embryos were cultured until day 5 after insemination when cleavage was evaluated, and embryos greater than 16 cells in number were selected for embryo transfer.
Embryo transfer and recovery
All procedures were approved by the University of Missouri Institutional Animal Care and Use Committee. Embryo transfers were performed in three replicates, with 7–8 Holstein heifers at the University of Missouri Beef Research and Teaching Farm per replicate. Heifers were examined by transrectal ultrasound to ensure cyclicity. For each replicate, one to two heifers received knockout embryos from bull 1, one to two received knockout embryos from bull 2, one to two received control embryos from bull 1, and one to two received control embryos from bull 2. Two heifers were used as recipients in two replicates (treatment was alternated between replicates), whereas the other heifers were used once only.
Ovulation for the heifers of each replicate was synchronized hormonally using a 7-day CO-Synch + CIDR protocol (Stevenson et al. 2008). Briefly, heifers received an intravaginal progesterone-releasing device (1.38 g CIDR®; Zoetis, Kalamazoo, MI, USA) and 100 µg, i.m., of gonadotrophin-releasing hormone (GnRH; Fertagyl®, Merck) on the morning of day 0. On the morning of day 7, CIDRs were removed, heifers received 25 mg of prostaglandin F2α (Lutalyse®, Zoetis), and an estrus detection aid (Estrotech®, Hermitage, TN, USA) was placed on the tail head of each heifer. A second injection of prostaglandin F2α was given on the morning of day 8. Starting on day 9, ovaries were scanned by ultrasound twice daily to monitor ovulation. Only heifers that ovulated were eligible to receive embryos. A second injection of GnRH was given on the morning of day 10 (anticipated day of ovulation and considered equivalent to day 0 of embryonic development).
Embryo transfer was performed on day 15 (i.e. day 5 after anticipated ovulation). Embryos >16 cells were selected from embryo cultures and transferred in groups ranging in number from 7 to 24 into each heifer. The number transferred depended on the number of available embryos in each replicate. Embryos were placed in the uterine horn ipsilateral to the corpus luteum using an embryo transfer pipette and a transcervical approach. A total of 107 knockout embryos (62 from bull 1 and 45 from bull 2) were transferred into 10 recipient heifers and 171 control embryos (82 from bull 1 and 89 from bull 2) were transferred into 12 recipient heifers.
Embryos were recovered non-surgically at day 7 of development (day 17 of the synchronized estrous cycle). Briefly, a 58.4 cm, 30 cc silicone Foley catheter (Agtech, Manhattan, KS, USA) was inserted into the body of the uterus and secured by inflating the cuff of the catheter with 10–15 cc of air. Uterine horns were filled with complete flush medium (BiolifeTM Advantage, Agtech) and emptied into a 75 µm mesh collection filter (WTA, São Paulo, Brazil) by massaging the uterus multiple times using the shallow method of collection and recollection (Sartori et al. 2003). Approximately 2 L of flush medium was used for each heifer. The number, stage, and grade of recovered embryos were recorded according to the International Embryo Technology Society manual (Stringfellow & Givens 2010). Individual morulae and blastocysts were incubated in 1% (w/v) proteinase from Streptomyces griseus in Dulbecco’s phosphate-buffered saline for 30 s to remove the zona pellucida, snap-frozen in liquid nitrogen and stored at –80°C until analysis of gene expression by RNA-Seq.
Statistical analysis of embryonic development data
Data were analyzed by logistic regression using the GLIMMIX procedure of SAS version 9.4 (Cary, NC, USA). Development (yes/no) was modeled as a binomial variable. Data for in vitro development were analyzed with a model that included effects of treatment, sire, and the interaction and with replicate as a random effect. Sire and the interaction were not significant, and hence the data were re-analyzed after removal of these terms. Data for development in utero were analyzed with a model that included effects of treatment, sire, and replicate and with cow nested within replicate and treatment as a random effect. Sire and replicate were not significant so the data were re-analyzed after the removal of these terms. As an additional test of the hypothesis, a two-tailed chi-square test was used to determine whether the proportion of recovered embryos that were blastocysts was affected by treatment.
RNA isolation and sequencing
RNA from individual embryos was isolated using the PicoPure® Kit (ThermoFisher) following the manufacturer’s instructions with the exception that Nano spin columns (Luna Nanotech, Toronto, Ontario, CA, USA) were used instead of the mini columns provided with the kit. High‐throughput sequencing was performed at the University of Missouri Genomics Technology Core. The number of morulae analyzed was four control (three males and one female) and four knockout embryos (one male and three females). The number of blastocysts was 17 control (eight males and nine females) and 9 knockout embryos (two males and seven females). Libraries were constructed following the manufacturer’s protocol with reagents supplied in the Takara SMART-Seq v4 Ultra Low-input RNA Kit (Takara Bio). Sample concentration was determined by Qubit fluorometer (Invitrogen) using the Qubit HS RNA assay kit, and RNA integrity was checked using the Fragment Analyzer automated electrophoresis system. Briefly, the poly-A containing mRNA was purified from total RNA, tRNA was fragmented, double-stranded cDNA was generated from fragmented RNA, and the index containing adapters were ligated to the ends. The amplified cDNA constructs were purified by the addition of Axyprep Mag PCR Clean-up beads (Corning). The final construct of each purified library was evaluated using the Fragment Analyzer automated electrophoresis system, quantified with the Qubit fluorometer using the Qubit HS dsDNA assay kit, and diluted according to Illumina’s standard sequencing protocol for sequencing on the NovaSeq 6000.
Analysis of RNA-sequencing data
The quality of fastq files of sequence data was checked with a FastQC tool (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Adapter trimming was performed using the cutadpat tool (Martin 2011). Trim Galore (https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/) was used to perform quality trimming (Phred score >30) by sliding window scan and selection of read lengths of 30 nucleotides or longer. Reads were then mapped to the bovine reference genome ARS-UCD1.2 (Rosen et al. 2020) using the program STAR (Dobin et al. 2013). Quantification of reads was performed with FeatureCounts (Liao et al. 2014). The total number of mapped reads per sample is summarized in Supplementary File 1 Table 1 (see section on supplementary materials given at the end of this article). The number of net mapped reads per sample averaged 5.67 million (range: 0.96–10.70 million reads).
The sex of each embryo was determined by quantifying the number of reads mapped to the Y chromosome. Differentially expressed genes (DEGs) between groups were determined by edgeR-robust (Zhou et al. 2014) while controlling for embryo sex. Genes were considered as differentially expressed if the false discovery rate (FDR) was <0.05. Data have been deposited in the Gene Expression Omnibus of the National Center for Biotechnology Information and are accessible through GEO Series accession number GSE222003.
Functional analysis of gene expression
Functional annotation of DEG was performed using the DAVID Bioinformatics Database (https://david.ncifcrf.gov/). Functional annotations with FDR < 0.05 were considered significant. In addition, DEGs were evaluated for overrepresentation of molecular and cellular functions as classified by Ingenuity Pathway Analysis (IPA; Qiagen) using FDR < 0.05 as a cutoff.
Gene set enrichment analysis of all genes was performed using a program called pathway analysis with down-weighting of overlapping genes (PADOG) whereby genes that frequently occur in multiple gene sets are weighed less than genes that appear in few gene sets (Tarca et al. 2012). Gene expression data from each individual embryo were used in the analysis. In cases of duplicate genes, the gene with the greatest expression was retained in the data set. The analysis was performed on the http://reactome.org website as described byGriss et al. (2020). The analysis was performed to examine the effect of treatment (knockout vs control) while holding embryo sex constant. Gene sets with FDR < 0.05 were considered significant.
Results
Development in vitro and in vivo
A total of 831 putative zygotes were injected with the Cas9 mRNA control and 844 zygotes with the RNA form of CRISPR/Cas9 to disrupt CSF2RA. Cleavage rate was greater (P = 0.0003) for control embryos (40 ± 9%) than for knockout embryos (25 ± 7%), but there was no difference (P = 0.485) in the percent of cleaved embryos that developed to >16 cells at day 5 of development (61 ± 12% for control vs 64 ± 2% for knockout).
Developmental data in utero are summarized in Fig. 1A. Recovery of embryos from the reproductive tract was low but not different among treatments (P = 0.677). A total of 56 embryos were recovered from cows receiving control embryos (33%) and 33 embryos were recovered from cows receiving knockout embryos (31%). There was no difference in the proportion of recovered embryos classified as morulae (P = 0.712), but the percent of recovered embryos classified as blastocysts tended (P = 0.067) to be greater for control embryos (63%) than for knockout embryos (39%). By chi-square analysis, the difference in percent blastocyst between treatments was P = 0.048.
Consequences for CRISPR/Cas9-mediated disruption of CSF2RA for the development of embryos in utero from day 5 to day 7 of development. (A) Number of embryos recovered from the uterus at day 7 (center) and the percent of embryos that were < morulae, morulae or blastocysts. The percent of embryos that were blastocysts was lower for knockout embryos (P = 0.048 by chi-square and P = 0.067 by logistic regression). (B, C) Volcano plot of genes expressed by morulae (B) and blastocysts (C). Genes in red were differentially expressed (false discovery rate < 0.05). (D, E) Molecular and cellular ontologies in which differentially expressed genes were overrepresented (false discovery rate < 0.05) are shown for morulae (D) and blastocysts (E).
Citation: Reproduction and Fertility 4, 2; 10.1530/RAF-23-0001
Consequences for CRISPR/Cas9-mediated disruption of CSF2RA for the development of embryos in utero from day 5 to day 7 of development. (A) Number of embryos recovered from the uterus at day 7 (center) and the percent of embryos that were < morulae, morulae or blastocysts. The percent of embryos that were blastocysts was lower for knockout embryos (P = 0.048 by chi-square and P = 0.067 by logistic regression). (B, C) Volcano plot of genes expressed by morulae (B) and blastocysts (C). Genes in red were differentially expressed (false discovery rate < 0.05). (D, E) Molecular and cellular ontologies in which differentially expressed genes were overrepresented (false discovery rate < 0.05) are shown for morulae (D) and blastocysts (E).
Citation: Reproduction and Fertility 4, 2; 10.1530/RAF-23-0001
Consequences for CRISPR/Cas9-mediated disruption of CSF2RA for the development of embryos in utero from day 5 to day 7 of development. (A) Number of embryos recovered from the uterus at day 7 (center) and the percent of embryos that were < morulae, morulae or blastocysts. The percent of embryos that were blastocysts was lower for knockout embryos (P = 0.048 by chi-square and P = 0.067 by logistic regression). (B, C) Volcano plot of genes expressed by morulae (B) and blastocysts (C). Genes in red were differentially expressed (false discovery rate < 0.05). (D, E) Molecular and cellular ontologies in which differentially expressed genes were overrepresented (false discovery rate < 0.05) are shown for morulae (D) and blastocysts (E).
Citation: Reproduction and Fertility 4, 2; 10.1530/RAF-23-0001
Gene expression
The number of genes whose expression was quantified was 14,878 for morula and 18,442 for blastocysts. Results for each gene after adjusting for sex are summarized in Supplementary File 1 Table 2 (morula) and Table 3 (blastocyst). Volcano plots illustrating the relationship between log2 fold change (knockout/control) and FDR are shown in Fig. 1B (morula) and 1C (blastocyst). Using an FDR of <0.05, there were 27 genes that were differentially expressed between knockout and control morula (17 genes upregulated in knockout embryos and 10 downregulated). Only 12 DEGs in morula were annotated. The annotated genes that were upregulated in knockout embryos were UPK3BL1 (44-fold) and PTMA (18-fold). The annotated genes that were downregulated in knockout embryos were PAGE4 (72-fold), SLC13A4 (65-fold), PLEKHO2 (62-fold), PTPRJ (60-fold), GCA (33-fold), APOA1 (25-fold), PAG2 (19-fold), PLAU (14-fold), PTGS2 (14-fold), and ARPC1B (12-fold).
Using an FDR of <0.05, there were 15 DEG for blastocysts with ten genes upregulated in knockout embryos and 5 genes downregulated. The upregulated genes were THBS1 (30-fold), PPP1R8 (28-fold), RAB33A (24-fold), ADIRF (16-fold), LAMP5 (15-fold), DNAH14 (15-fold), AREG (14-fold), UPK1B (13-fold), ATP13A4 (12-fold), and NAGPA (12-fold), while the downregulated genes were TREM2 (99-fold), VAMP5 (72-fold), KCTD14 (5-fold), UNCX (53-fold), and SLC12A17 (23-fold).
Most embryos did not have detectable transcripts for CSF2RA (Supplementary File 1). This was true for all morulae, regardless of treatment, and for 15 of 17 control blastocysts and 8 of 9 knockout blastocysts.
Properties of DEG
Analysis of DEG for morulae by DAVID indicated that there were no functional annotation terms with FDR < 0.05. Analysis by IPA indicated 19 molecular and cellular functions in which DEGs were overrepresented (Fig. 1D) including cellular movement (six genes), cell-to-cell signaling and interaction (four genes), and carbohydrate metabolism (three genes).
Analysis of DEG for blastocysts by DAVID indicated two functional annotations with FDR < 0.05 – transmembrane helix (AREG, DNAH14, LAMP5, NAGPA, SLC22A17, TREM2, UPK1B, and VAMP5; FDR = 0.0038) and EGF-like domain (AREG, NAGPA, and THBS1; FDR=0.0089). Analysis by IPA indicated 21 molecular and cellular functions in which DEGs were overrepresented (Fig. 1E), including cellular development (seven genes), cell-to-cell signaling and interaction (four genes), carbohydrate metabolism (two genes), and cell morphology (five genes).
There were no DEGs found in the current data set that were also identified as regulated by CSF2 in cultured blastocysts (Zolini et al. 2020, Xiao et al. 2021).
Pathway analysis
Gene set enrichment analysis was performed for both morula and blastocyst data sets using the program PADOG that reduces the weight of genes that frequently occur in multiple gene set annotations. Results are summarized in Supplementary File 1 Table 4 (morula) and Table 5 (blastocyst). Using an FDR of < 0.05, there were 177 regulated gene sets for morula. Of these, eight were upregulated in knockout embryos including those related to NOTCH signaling, MAPK1 signaling, and β-oxidation. There were 169 gene sets that were downregulated in knockout embryos including those related to keratan sulfate metabolism, glycosylation, programmed cell death, DNA methylation, ion transport, gluconeogenesis, and fatty acid metabolism, Several gene sets related to cell signaling were also downregulated including those associated with NOTCH, chemokines, androgen receptor, acetylcholine, olfactory receptors, glutamate, erythropoietin, kit ligand, PDGF, IL7, TNF, NOD1/2, and signaling pathways involving RAS, RAF, MAPKs, CREB, and integrins.
There were 127 regulated gene sets for blastocysts identified by PADOG. All were upregulated in knockout embryos. Like for morula, many gene sets were involved in cell signaling, including pathways involving RAC and RHO GTPases, RET proto-oncogene, erythropoietin, PI3K, insulin-like growth factor 1, insulin, fibroblast growth factors, IL38, GABA, hedgehog, VEGFA, opioids, NCAM, hemes, and IL2. Other gene sets included those associated with glycosylation, chondroitin sulfate biosynthesis, glycogen metabolism, apoptosis, and RUNX transcription factors.
Discussion
The findings in this study provide evidence that, like in the mouse (Robertson et al. 2001), CSF2 plays a physiological role in the development of the preimplantation embryo. In particular, development to the blastocyst stage in utero was compromised in CSF2RA-knockout embryos, as indicated by a reduction in the proportion of recovered embryos that developed to the blastocyst. Moreover, cellular function of the embryos that did become a morula or blastocyst was altered by the absence of CSF2 signaling because gene expression was moderately altered. One conclusion, therefore, is that CSF2 can facilitate the progression of the embryo to the blastocyst stage. It is also clear, however, that CSF2 is not obligatory for development to the blastocyst stage because some knockout embryos were able to develop to the blastocyst stage.
A similar conclusion was drawn from earlier experiments in mice. There was no difference in the proportion of embryos that were blastocysts or hatched blastocysts recovered at day 4 of pregnancy between pregnancies established by mating Csf2-null parents or wildtype parents (Robertson et al. 2001). However, embryo development was altered by the absence of CSF2 because the total cell number in blastocysts derived from Csf2-null parents was reduced compared to control. Later in pregnancy, at day 17 of gestation, there was no effect of Csf2 deletion on the number of implantation sites (Robertson et al. 1999). Fetal weight for fetuses derived from Csf2-null parents was reduced, but this effect could have involved results of embryonic or maternal deficiency in Csf2 later in gestation.
One action of CSF2 is to protect the embryo from cellular stress (Loureiro et al. 2011, Sosa et al. 2020, Zolini et al. 2020, Sosa & Hansen 2023). It is possible that the increased development in utero was caused, in part, by differences in cell stress responses to the process of transferring embryos to the uterus. Also, however, CSF2 has direct effects on the embryo beginning at day 5 of development to facilitate development to the blastocyst stage (de Moraes & Hansen 1997, Loureiro et al. 2009, Sosa et al. 2020).
The number of genes that were altered in expression in morulae or blastocysts because of the absence of a functional CSF2RA was <30. The genes were distinct from those found to be regulated by CSF2 in embryos under in vitro conditions (Zolini et al. 2020, Xiao et al. 2021). Specific aspects of the maternal environment, including nutrient supply and presence of other cell-signaling molecules, probably modify the transcriptional response of the embryo to CSF2. It is notable that a preponderance of annotated DEGs in the morula were downregulated in knockout embryos, while two-thirds of DEGs in blastocysts were upregulated in knockout embryos. Perhaps, the downregulation of genes in knockout embryos at the morula stage reflects inadequate development due to the absence of CSF2 signaling, while the upregulation of genes in knockout embryos at the blastocyst stage reflects adjustments in gene regulation in knockout embryos that allowed development to the blastocyst stage despite the absence of CSF2 signaling. The DEGs for both morulae and blastocysts were overrepresented in IPA functions associated with cell signaling including for morulae (APOA1, PLAU, PTGS2, and PTPRJ) and blastocysts (AREG, RAB33A, THBS1, and UNCX). Other DEGs associated with cell signaling in blastocysts include the receptor-associated genes SLC22A17 and UPK1B. Three of the genes upregulated in knockout blastocysts (AREG, NAGPA, and THBS1) were upregulated or exclusively expressed in inner cell mass (ICM) of blastocysts that developed in vivo (Hosseini et al. 2015). Thus, development of the ICM in vivo might be modified by CSF2. One study (Loureiro et al. 2009), but not another (Siqueira & Hansen 2016), found that CSF2 treatment in culture increased the proportion of blastomeres in the blastocyst that were ICM. CSF2 has also been reported to increase the survival of isolated ICM in culture (Dobbs et al. 2013).
More gene sets were identified as being affected by the knockout using gene set enrichment analysis than there were DEGs. The most likely explanation for this discrepancy is that gene set enrichment analysis considers all genes in the data set and that genes affected by the knockout appear in more than one gene set. In any case, similar conclusions on the impact of CSF2RA knockout can be made from gene set enrichment analysis. A preponderance of the regulated gene sets (169 of 177) were predicted to be downregulated in knockout embryos at the morula stage, while all 127 regulated gene sets for the blastocyst stage were predicted to be upregulated for knockout embryos. There were also large numbers of regulated gene sets that were associated with cell signaling for both the morula and blastocyst stage as well as those involved in glycosylation and other functions.
One unexpected finding was the low cleavage rate in microinjected embryos of both groups but particularly for those injected with both guide RNA and Cas9 mRNA. The low cleavage probably reflects random technical issues. An identical CRISPR/Cas9 system was used in our previous study, and cleavage rates after microinjection were 69% for control embryos and 70% for knockout embryos (Xiao et al. 2021).
Another unexpected finding was the low expression of CSF2RA itself, even in control embryos. In another experiment, expression of CSF2RA in the bovine preimplantation embryo was identified at all stages, and the embryo was examined up to the blastocyst stage using reverse-transcription PCR with pools of 25–30 embryos (Dobbs et al. 2013). It is likely that expression in the morula and blastocyst was too low to be detected using RNA-Seq of single embryos. Transcripts for another important gene in the blastocyst, SOX2, were also undetected. One weakness of the study was the low number of mapped reads per sample (average 5.67 million), which reflects the difficulties associated with RNA sequencing of single embryos. One result is likely to be low read counts for important genes and another result is reduced accuracy in expression values. It may also be that failure to identify CSF2RA in most embryos may mean that the effect of CSF2 on the embryo occurs before the morula stage of development. Although the expression of CSF2RA in the embryo produced in vitro was identified through the blastocyst stage (Dobbs et al. 2013), the situation could be different for the embryo developing in vivo.
In conclusion, signaling through CSF2RA is not obligatory for the development of the bovine preimplantation embryo to the blastocyst stage but CSF2 signaling does enhance the likelihood that the embryo can become a blastocyst and result in specific changes in gene expression.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/RAF-23-0001.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This research was supported by NIH R01 HD088352, USDA-NIFA 2022-67015-36371, the Southeast Milk Dairy Checkoff Program and the L.E. ‘Red’ Larson Endowment.
Author contribution statement
The experiment was conceived by PJH and designed by PJH, FS, MSO, and KL. Experimental procedures were conducted by FS, KU, JND, KSS, and KMD. Data analysis was performed by FS, MSO, and PJH. The first draft of the paper was written by FS and PJH.
Acknowledgements
The authors thank the University of Missouri Genomics Technology Core for technical support and Drs Johannes Griss, Medical University of Vienna, and Adi Tarca, Wayne State University, for assistance with interpretation of results of PADOG analysis.
References
Amaral TF, de Grazia JGV, Martinhao LAG, De Col F, Siqueira LGB, Viana JHM & & Hansen PJ 2022 Actions of CSF2 and DKK1 on bovine embryo development and pregnancy outcomes are affected by composition of embryo culture medium. Scientific Reports 12 7503. (https://doi.org/10.1038/s41598-022-11447-7)
Chin PY, Macpherson AM, Thompson JG, Lane M & & Robertson SA 2009 Stress response genes are suppressed in mouse preimplantation embryos by granulocyte-macrophage colony-stimulating factor (GM-CSF). Human Reproduction 24 2997–3009. (https://doi.org/10.1093/humrep/dep307)
de Moraes AA & & Hansen PJ 1997 Granulocyte-macrophage colony-stimulating factor promotes development of in vitro produced bovine embryos. Biology of Reproduction 57 1060–1065. (https://doi.org/10.1095/biolreprod57.5.1060)
de Moraes AA, Paula-Lopes FF, Chegini N & & Hansen PJ 1999 Localization of granulocyte-macrophage colony-stimulating factor in the bovine reproductive tract. Journal of Reproductive Immunology 42 135–145. (https://doi.org/10.1016/s0165-0378(9800075-8)
Denicol AC, Block J, Kelley DE, Pohler KG, Dobbs KB, Mortensen CJ, Ortega MS & & Hansen PJ 2014 The WNT signaling antagonist Dickkopf-1 directs lineage commitment and promotes survival of the preimplantation embryo. FASEB Journal 28 3975–3986. (https://doi.org/10.1096/fj.14-253112)
Dobbs KB, Khan FA, Sakatani M, Moss JI, Ozawa M, Ealy AD & & Hansen PJ 2013 Regulation of pluripotency of inner cell mass and growth and differentiation of trophectoderm of the bovine embryo by colony stimulating factor 2. Biology of Reproduction 89 141. (https://doi.org/10.1095/biolreprod.113.113183)
Dobbs KB, Gagné D, Fournier E, Dufort I, Robert C, Block J, Sirard MA, Bonilla L, Ealy AD, Loureiro B, et al.2014 Sexual dimorphism in developmental programming of the bovine preimplantation embryo caused by colony-stimulating factor 2. Biology of Reproduction 91 80. (https://doi.org/10.1095/biolreprod.114.121087)
Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M & & Gingeras TR 2013 STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29 15–21. (https://doi.org/10.1093/bioinformatics/bts635)
Estrada-Cortés E, Jannaman EA, Block J, Amaral TF & & Hansen PJ 2021 Programming of postnatal phenotype caused by exposure of cultured embryos from Brahman cattle to colony-stimulating factor 2 and serum. Journal of Animal Science 99 skab180. (https://doi.org/10.1093/jas/skab180)
Griss J, Viteri G, Sidiropoulos K, Nguyen V, Fabregat A & & Hermjakob H 2020 ReactomeGSA - efficient multi-omics comparative pathway analysis. Molecular and Cellular Proteomics 19 2115–2125. (https://doi.org/10.1074/mcp.TIR120.002155)
Hackett AJ, Durnford R, Mapletoft RJ & & Marcus GJ 1993 Location and status of embryos in genital tract of superovulated cows 4 to 6 days after insemination. Theriogenology 40 1147–1153. (https://doi.org/10.1016/0093-691X(9390285-D)
Hansen PJ & & Tríbulo P 2019 Regulation of present and future development by maternal regulatory signals acting on the embryo during the morula to blastocyst transition - insights from the cow. Biology of Reproduction 101 526–537. (https://doi.org/10.1093/biolre/ioz030)
Hansen G, Hercus TR, McClure BJ, Stomski FC, Dottore M, Powell J, Ramshaw H, Woodcock JM, Xu Y, Guthridge M, et al.2008 The structure of the GM-CSF receptor complex reveals a distinct mode of cytokine receptor activation. Cell 134 496–507. (https://doi.org/10.1016/j.cell.2008.05.053)
Hosseini SM, Dufort I, Caballero J, Moulavi F, Ghanaei HR & & Sirard MA 2015 Transcriptome profiling of bovine inner cell mass and trophectoderm derived from in vivo generated blastocysts. BMC Developmental Biology 15 49. (https://doi.org/10.1186/s12861-015-0096-3)
Ingelfinger F, De Feo D & & Becher B 2021 GM-CSF: master regulator of the T cell-phagocyte interface during inflammation. Seminars in Immunology 54 101518. (https://doi.org/10.1016/j.smim.2021.101518)
Kannampuzha-Francis J, Denicol AC, Loureiro B, Kaniyamattam K, Ortega MS & & Hansen PJ 2015 Exposure to colony stimulating factor 2 during preimplantation development increases postnatal growth in cattle. Molecular Reproduction and Development 82 892–897. (https://doi.org/10.1002/mrd.22533)
Kwak SS, Jeung SH, Biswas D, Jeon YB & & Hyun SH 2012 Effects of porcine granulocyte-macrophage colony-stimulating factor on porcine in vitro-fertilized embryos. Theriogenology 77 1186–1197. (https://doi.org/10.1016/j.theriogenology.2011.10.025)
Lee K, Redel BK, Spate L, Teson J, Brown AN, Park KW, Walters E, Samuel M, Murphy CN & & Prather RS 2013 Piglets produced from cloned blastocysts cultured in vitro with GM-CSF. Molecular Reproduction and Development 80 145–154. (https://doi.org/10.1002/mrd.22143)
Liao Y, Smyth GK & & Shi W 2014 featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30 923–930. (https://doi.org/10.1093/bioinformatics/btt656)
Loureiro B, Bonilla L, Block J, Fear JM, Bonilla AQ & & Hansen PJ 2009 Colony-stimulating factor 2 (CSF-2) improves development and posttransfer survival of bovine embryos produced in vitro. Endocrinology 150 5046–5054. (https://doi.org/10.1210/en.2009-0481)
Loureiro B, Oliveira LJ, Favoreto MG & & Hansen PJ 2011 Colony-stimulating factor 2 inhibits induction of apoptosis in the bovine preimplantation embryo. American Journal of Reproductive Immunology 65 578–588. (https://doi.org/10.1111/j.1600-0897.2010.00953.x)
Martin M 2011 Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal 17 10–12. (https://doi.org/10.14806/ej.17.1.200)
Okabe-Kinoshita M, Kobayashi T, Shioya M, Sugiura T, Fujita M & & Takahashi K 2022 Granulocyte-macrophage colony-stimulating factor-containing medium treatment after thawing improves blastocyst-transfer outcomes in the frozen- thawed blastocyst-transfer cycle. Journal of Assisted Reproduction and Genetics 39 1373–1381. (https://doi.org/10.1007/s10815-022-02493-1)
Robertson SA, Roberts CT, Farr KL, Dunn AR & & Seamark RF 1999 Fertility impairment in granulocyte-macrophage colony-stimulating factor-deficient mice. Biology of Reproduction 60 251–261. (https://doi.org/10.1095/biolreprod60.2.251)
Robertson SA, Sjöblom C, Jasper MJ, Norman RJ & & Seamark RF 2001 Granulocyte-macrophage colony-stimulating factor promotes glucose transport and blastomere viability in murine preimplantation embryos. Biology of Reproduction 64 1206–1215. (https://doi.org/10.1095/biolreprod64.4.1206)
Rosen BD, Bickhart DM, Schnabel RD, Koren S, Elsik CG, Tseng E, Rowan TN, Low WY, Zimin A, Couldrey C, et al.2020 De novo assembly of the cattle reference genome with single-molecule sequencing. GigaScience 9 gigaa021. (https://doi.org/10.1093/gigascience/giaa021)
Sartori R, Suárez-Fernández CA, Monson RL, Guenther JN, Rosa GJ & & Wiltbank MC 2003 Improvement in recovery of embryos/ova using a shallow uterine horn flushing technique in superovulated Holstein heifers. Theriogenology 60 1319–1330. (https://doi.org/10.1016/s0093-691x(0300147-x)
Shyam S, Goel P, Kumar D, Malpotra S, Singh MK, Lathwal SS, Chand S & & Palta P 2020 Effect of Dickkopf-1 and colony stimulating factor-2 on the developmental competence, quality, gene expression and live birth rate of buffalo (Bubalus bubalis) embryos produced by hand-made cloning. Theriogenology 157 254–262. (https://doi.org/10.1016/j.theriogenology.2020.07.022)
Siqueira LG & & Hansen PJ 2016 Sex differences in response of the bovine embryo to colony-stimulating factor 2. Reproduction 152 645–654. (https://doi.org/10.1530/REP-16-0336)
Siqueira LG, Tribulo P, Chen Z, Denicol AC, Ortega MS, Negrón-Pérez VM, Kannampuzha-Francis J, Pohler KG, Rivera RM & & Hansen PJ 2017 Colony-stimulating factor 2 acts from days 5 to 7 of development to modify programming of the bovine conceptus at day 86 of gestation†. Biology of Reproduction 96 743–757. (https://doi.org/10.1093/biolre/iox018)
Sjöblom C, Roberts CT, Wikland M & & Robertson SA 2005 Granulocyte-macrophage colony-stimulating factor alleviates adverse consequences of embryo culture on fetal growth trajectory and placental morphogenesis. Endocrinology 146 2142–2153. (https://doi.org/10.1210/en.2004-1260)
Sjöblom C, Wikland M & & Robertson SA 1999 Granulocyte–macrophage colony-stimulating factor promotes human blastocyst development in vitro. Human Reproduction 14 3069–3076. (https://doi.org/10.1093/humrep/14.12.3069)
Sjöblom C, Wikland M & & Robertson SA 2002 Granulocyte-macrophage colony-stimulating factor (GM-CSF) acts independently of the beta common subunit of the GM-CSF receptor to prevent inner cell mass apoptosis in human embryos. Biology of Reproduction 67 1817–1823. (https://doi.org/10.1095/biolreprod.101.001503)
Sosa F & & Hansen PJ 2022 Colony stimulating factor 2 protects the preimplantation bovine embryo from heat shock. Zygote 31 51–54. (https://doi.org/10.1017/S0967199422000508)
Sosa F, Block J, Xiao Y & & Hansen PJ 2020 Determinants of survival of the bovine blastocyst to cryopreservation stress: treatment with colony stimulating factor 2 during the morula-to-blastocyst transition and embryo sex. CABI Agriculture and Bioscience 1 12. (https://doi.org/10.1186/s43170-020-00012-9)
Stevenson JS, Tenhouse DE, Krisher RL, Lamb GC, Larson JE, Dahlen CR, Pursley JR, Bello NM, Fricke PM, Wiltbank MC, et al.2008 Detection of anovulation by heatmount detectors and transrectal ultrasonography before treatment with progesterone in a timed insemination protocol. Journal of Dairy Science 91 2901–2915. (https://doi.org/10.3168/jds.2007-0856)
Stoecklein KS, Ortega MS, Spate LD, Murphy CN & & Prather RS 2021 Improved cryopreservation of in vitro produced bovine embryos using FGF2, LIF, and IGF1. PLoS One 16 e0243727. (https://doi.org/10.1371/journal.pone.0243727)
Stringfellow D & & Givens M 2010 Manual of the International Embryo Transfer Society (IETS) 4th ed. Champaign, IL: IETS.
Tarca AL, Draghici S, Bhatti G & & Romero R 2012 Down-weighting overlapping genes improves gene set analysis. BMC Bioinformatics 13 136. (https://doi.org/10.1186/1471-2105-13-136)
Tríbulo P, Bernal Ballesteros BH, Ruiz A, Tríbulo A, Tríbulo RJ, Tríbulo HE, Bo GA & & Hansen PJ 2017 Consequences of exposure of embryos produced in vitro in a serum-containing medium to dickkopf-related protein 1 and colony stimulating factor 2 on blastocyst yield, pregnancy rate, and birth weight. Journal of Animal Science 95 4407–4412. (https://doi.org/10.2527/jas2017.1927)
Tríbulo P, Siqueira LGB, Oliveira LJ, Scheffler T & & Hansen PJ 2018 Identification of potential embryokines in the bovine reproductive tract. Journal of Dairy Science 101 690–704. (https://doi.org/10.3168/jds.2017-13221)
Tríbulo P, Rivera RM, Ortega Obando MS, Jannaman EA & & Hansen PJ 2019 Production and culture of the bovine embryo. Methods in Molecular Biology 2006 115–129. (https://doi.org/10.1007/978-1-4939-9566-0_8)
Wen Z, Pan Y, Cui Y, Peng X, Chen P, Fan J, Li G, Zhao T, Zhang J, Qin S, et al.2017 Colony-stimulating factor 2 enhances the developmental competence of yak (Poephagus grunniens) preimplantation embryos by modulating the expression of heat shock protein 70 kDa 1A. Theriogenology 93 16–23. (https://doi.org/10.1016/j.theriogenology.2017.01.034)
Xiao Y, Uh K, Negrón-Pérez VM, Haines H, Lee K & & Hansen PJ 2021 Regulation of gene expression in the bovine blastocyst by colony-stimulating factor 2 is disrupted by CRISPR/Cas9-mediated deletion of CSF2RA. Biology of Reproduction 104 995–1007. (https://doi.org/10.1093/biolre/ioab015)
Zhou X, Lindsay H & & Robinson MD 2014 Robustly detecting differential expression in RNA sequencing data using observation weights. Nucleic Acids Research 42 e91. (https://doi.org/10.1093/nar/gku310)
Zhou W, Chu D, Sha W, Fu L & & Li Y 2016 Effects of granulocyte-macrophage colony-stimulating factor supplementation in culture medium on embryo quality and pregnancy outcome of women aged over 35 years. Journal of Assisted Reproduction and Genetics 33 39–47. (https://doi.org/10.1007/s10815-015-0627-7)
Ziebe S, Loft A, Povlsen BB, Erb K, Agerholm I, Aasted M, Gabrielsen A, Hnida C, Zobel DP, Munding B, et al.2013 A randomized clinical trial to evaluate the effect of granulocyte-macrophage colony-stimulating factor (GM-CSF) in embryo culture medium for in vitro fertilization. Fertility and Sterility 99 1600–1609. (https://doi.org/10.1016/j.fertnstert.2012.12.043)
Zolini AM, Block J, Rabaglino MB, Tríbulo P, Hoelker M, Rincon G, Bromfield JJ & & Hansen PJ 2020 Molecular fingerprint of female bovine embryos produced in vitro with high competence to establish and maintain pregnancy†. Biology of Reproduction 102 292–305. (https://doi.org/10.1093/biolre/ioz190)
