Actions of colony-stimulating factor 3 on the maturing oocyte and developing embryo in cattle

Abstract

Colony-stimulating factor 3 (CSF3), also known as granulocyte colony-stimulating factor, is used to reduce the incidence of mastitis in cattle. Here, we tested whether recombinant bovine CSF3 at 1, 10, or 100 ng/mL acts on the bovine oocyte during maturation or on the developing embryo to modify competence for development and characteristics of the resultant blastocyst. For experiment 1, oocytes were matured with or without CSF3. The resultant embryos were cultured in a serum-free medium for 7.5 d. There was no effect of CSF3 on cleavage or on development to the blastocyst stage except that 100 ng/mL reduced the percent of putative zygotes and cleaved embryos becoming blastocysts. Expression of transcripts for 93 genes in blastocysts was evaluated by RT-PCR using the Fluidigm platform. Transcript abundance was affected by one or more concentrations of CSF3 for four genes only (CYP11A1, NOTCH2, RAC1, and YAP1). For experiment 2, cumulus-oocyte complexes (COC) were fertilized with either X- or Y-sorted semen. Putative zygotes were cultured in medium containing CSF3 treatments added at the beginning of culture. There was no effect of CSF3, sex, or the interaction on the percent of putative zygotes that cleaved or on the percent of putative zygotes or cleaved embryos becoming a blastocyst. For experiment 3, CSF3 was added from day 4 to 7.5 of development. There was no effect of CSF3 on development to the blastocyst stage. Transcript abundance of 10 genes was increased by 100 ng/mL CSF3, including markers of epiblast (NANOG, SOX2), hypoblast (ALPL, FN1, KDM2B, and PDGFRA), epiblast and hypoblast (HNF4A) and trophectoderm (TJAP1). Results are indicative that concentrations of CSF3 higher than typical after therapeutic administration can reduce oocyte competence and act on the embryo to affect characteristics of the blastocyst.

Introduction

Colony-stimulating factor 3 (CSF3), otherwise known as granulocyte colony-stimulating factor, is a hematopoietic growth factor that increases production and maturation of neutrophils from bone marrow (Tay et al., 2017). A commercial preparation consisting of CSF3 conjugated to polyethylene glycol (i.e., pegylated CSF3) can increase circulating neutrophil numbers and improve neutrophil function in dairy cattle (Kimura et al., 2014; McDougall et al., 2017; Van Schyndel et al., 2018). Moreover, pegylated CSF3 has been reported to reduce the incidence of clinical mastitis in dairy cows when administered during the periparturient period (Canning et al., 2017; Ruiz et al., 2017; see, however, Zinicola et al., 2018 for lack of effect).

Many cytokines are pleiotropic, and the effects of administration of CSF3 extend beyond those relating to neutrophil function. Treatment of cows with pegylated CSF3 can increase circulating concentrations of lymphocytes and monocytes (McDougall et al., 2017; Zinicola et al., 2018; Putz et al., 2019) and cause increased plasma concentrations of γ-glutamyl transferase, alkaline phosphatase, IL6, and IL-1β (Lopreiato et al., 2019). Two potential targets for CSF3 are the oocyte and developing embryo. Addition of 10 ng/mL recombinant human CSF3 to pig cumulus-oocyte complexes (COC) during oocyte maturation increased the proportion of oocytes becoming blastocysts after in vitro fertilization and embryo culture (Cai et al., 2015). Treatment of women undergoing in vitro fertilization (IVF) with intravaginal perfusion of CSF3 increased pregnancy rate (Li et al., 2017). There were no significant effects of periparturient treatment with pegylated CSF3 on pregnancy/AI at first service in studies by Canning et al. (2017) and Zinicola et al. (2018). In both experiments, however, pregnancy/AI at first service was numerically greater for the treated group, with values of 42.6% vs. 38.2% pregnant in the former study and 34.3% vs. 28.8% in the latter study.

Here we used the oocyte subjected to in vitro maturation (experiment 1) and the embryo produced in vitro (experiments 2 and 3) to evaluate whether CSF3 can modify competence of either the oocyte or embryo for development to the blastocyst stage and alter characteristics of the resultant blastocyst as evaluated by expression of genes important for blastocyst differentiation, epigenetic regulation, and cell signaling. Experiments with other embryokines indicate that effects on the embryo can depend on embryo sex (Dobbs et al., 2014; Siqueira and Hansen, 2016). Accordingly, effects of CSF3 added at the beginning of culture on embryonic development were tested separately for male and female embryos in experiment 2. In experiments with other embryokines, including colony stimulating factor 2 (de Moraes and Hansen, 1997), insulin-like growth factor 1 (Bonilla et al., 2011) and dickkopf-1 (Denicol et al., 2014), the effect of the growth factor on characteristics of development was exerted when the embryokine was added at day 4 or 5 of development. Accordingly, experiment 3 was conducted to test the effects of CSF3 added at day 4 of development.

Materials and Methods

IACUC approval was not sought because no animals were used.

In vitro production of embryos

Details of procedures for production of embryos in vitro are described elsewhere (Tríbulo et al., 2019). Briefly, COC were obtained from ovaries of cattle of various breeds at a local abattoir in Florida or were purchased from SimVitro (Boise, ID, USA) or Vytelle (Hermiston, OR, USA). For oocytes collected in Florida, COC were recovered by bisecting follicles in 3–8 mm diameter using a scalpel and by strongly rinsing the ovaries in a bath of oocyte collection medium (MOFA Global, Verona, WI, USA). Only oocytes with a least three layers of compact cumulus cells and a homogeneous cytoplasm were included in the experiments. The COC were matured in groups of 10 in 50 µL microdrops of a commercial oocyte maturation medium (IVF-Bioscience, Falmouth, Cornwall, UK). Maturation drops were covered with mineral oil (Sigma Aldrich, St Louis, MO, USA) and incubated for 22–24 h at 38.5 °C and 5% (v/v) CO₂ in a humidified atmosphere. Purchased oocytes were shipped overnight at 38.5 °C in proprietary maturation medium and allowed to mature for 22–24 h.

Semen for fertilization was thawed and purified using PureSperm 40/80 gradient (Nidacon International AB, Mölndal, Sweden) and resuspended in IVF-TALP. Depending on the experiment, matured oocytes were either fertilized in groups of 300 in 1,900 µL of fertilization solution or fertilized in oil-covered 60 µL microdrops in groups of up to 30. Details are described for each experiment. After fertilization, putative zygotes (i.e., oocytes exposed to sperm during fertilization) were denuded of cumulus cells by vortexing in 100 µL hyaluronidase (1,000 U/mL) and cultured in groups of 25–30 in 50-µL microdrops of a defined, serum-free culture medium called synthetic oviduct fluid – bovine embryo 2 (SOF-BE2; Tríbulo et al., 2019). Drops were overlaid with mineral oil and embryos were cultured at 38.5 °C in a humidified atmosphere of 5% CO₂, 5% O₂, and the balance nitrogen. For all experiments, cleavage was assessed on day 4 after insemination and blastocyst rate was determined at day 7.5 after insemination.

Preparation of CSF3

Recombinant bovine CSF3 (Elanco, Greenfield, IN, USA; lot RS0830) was prepared at a concentration of 4.66 mg/mL in 30 mM sodium acetate buffer, pH 4.0 containing 4.0% (wt/vol) sucrose and stored in at −80 °C in aliquots of 5 μL. To prepare for culture, one aliquot was thawed and diluted with either oocyte maturation medium (experiment 1) or SOF-BE2 (experiments 2 and 3) to either 1, 10, or 100 ng/mL (experiments 1 and 2) or 10, 100, or 1,000 ng/mL (experiment 3). Dilutions were made so that all concentrations of CSF3, as well as the vehicle control (termed 0 ng/mL), contained the same final amounts of the acetate/sucrose buffer (1:46,600, vol:vol).

Experiment 1: effects of CSF3 during maturation on developmental competence and gene expression of embryos

The experiment was performed in 10 replicates with 366 to 376 putative zygotes per treatment (1,486 total putative zygotes). The COC were obtained from abattoir-derived ovaries from Florida. Oocytes were matured for 20–22 h at 38.5 °C and 5% CO₂ in humidified air in oocyte maturation medium (IVF-Bioscience) containing either 0, 1, 10, or 100 ng/mL CSF3. After maturation, COC were transferred to fertilization drops covered with mineral oil. Each fertilization drop included up to 30 COC in 60 µL in vitro fertilization Tyrodes albumin lactate pyruvate solution (IVF-TALP) fertilization medium, 20 µL PureSperm®-purified spermatozoa (non-sexed) from a pool of three bulls (final concentration 1 × 10⁶/mL) and 3.5 µL of a penicillamine-hypotaurine-epinephrine solution (described in detail by Tríbulo et al., 2019). A different pool of bulls were used for each fertilization procedure; a total of seven bulls (Brangus, Simmental, and Limousin) were used for the experiment. After 15–18 h of co-incubation at 38.5 °C and 5% CO₂ in humidified air, putative zygotes were removed from fertilization drops, denuded of cumulus cells, and cultured in groups of 20–30 in 50-µL microdrops of SOF-BE2.

Experiment 2: effects of CSF3 during day 0 to 7.5 of embryo culture on developmental competence of female and male embryos

The experiment was conducted in six replicates with 408 to 523 putative zygotes per treatment for a total of 3,968 putative zygotes. The COC were collected from various breeds from abattoir-derived ovaries in California and Idaho by SimVitro and shipped overnight at 38.5 °C in maturation medium. After 22–24 h of maturation, fertilization was initiated by incubating half the COC with X-sorted semen from a single bull and half with Y-sorted semen from the same bull. Each fertilization drop included up to 30 COC in 60 µL fertilization medium, 20 µL PureSperm®-purified spermatozoa (final concentration 1 × 10⁶/mL), 3.5 µL of a penicillamine-hypotaurine-epinephrine solution and with the antibiotic amikacin (Sigma Aldrich) added at a final concentration of 20 µg/mL. The amikacin was added to prevent bacterial growth sometimes encountered with sex-sorted semen. A single Angus bull was used for each fertilization with a total of three Angus bulls being used for the experiment. After 19–22 h of co-incubation at 38.5 °C and 5% CO₂ in humidified air, putative zygotes (i.e., oocytes exposed to sperm) were removed from fertilization drops and denuded of cumulus cells by vortexing in 100 µL hyaluronidase (1,000 U/mL). Denuded putative zygotes were cultured in groups of 20–30 in 50-µL microdrops of SOF-BE2 as described by Tribulo et al. (2019) except with the addition of 20 µg/mL amikacin and with either 0, 1, 10, or 100 ng/mL CSF3. The CSF3 was added at day 0 and was not supplemented thereafter.

Experiment 3: effects of CSF3 added at day 4 of embryo culture on developmental competence and blastocyst gene expression

It is possible that CSF3 only acts on the embryo after day 4 of development and CSF3 added at day 0 in experiment 2 had degraded by day 4. To account for this possibility, experiment 3 was performed where CSF3 was added at day 4 of culture. The experiment was conducted in seven replicates, with 347 to 385 putative zygotes per treatment for a total of 1,452 putative zygotes. The COC were from various breeds and were obtained from abattoir-derived ovaries in Florida (n = 4 replicates) or from Vytelle (n = 3 replicates). The COC were matured for 21–22.5 h at 38.5 °C. After maturation, COC were transferred to fertilization dishes containing 1,700 µL IVF-TALP, 120 µL sperm (non-sexed) diluted in IVF-TALP, and 80 μL of penicillamine, hypotaurine, and epinephrine. The sperm was PureSperm®-purified (final concentration 1 × 10⁶/mL) and represented a pool of three bulls of various breeds. A total of nine bulls (Brangus, Polled Hereford, Limousin, Simmental, and Holstein) were used for the experiment. Fertilization proceeded for 17–18 h at 38.5 °C and 5% CO₂ in humidified air. Putative zygotes were removed from fertilization drops, denuded of cumulus cells, and cultured in groups of 20–30 in 50-µL microdrops of SOF-BE2 culture medium. Drops were overlaid with mineral oil and embryos were cultured at 38.5 °C in a humidified atmosphere of 5% CO₂, 5% O₂ and the balance nitrogen. At day 4 of culture, 5 µL culture medium was removed from each drop and replaced by 5 µL CSF3 at 10X the final concentration. The final concentration of CSF3 was 0, 1, 10, or 100 ng/mL.

PCR

Blastocysts of all stages from experiments 1 and 3 were harvested at day 7.5 for measurement of gene expression by real-time reverse transcriptase. The zona pellucida was removed with either 0.1% (w/v) proteinase from Streptococcus griseus or Acid Tyrode’s solution. Zona-denuded blastocysts were snap-frozen in groups of up to 10 in liquid nitrogen and stored at −80 °C. Subsequently, embryos were thawed and pooled into groups of 10 and RNA was purified using the RNeasy micro kit (Qiagen, Germantown, MD, USA) with the manufacturer’s protocol. Isolated RNA (15 µL) was treated with 1 µL (2 U) of DNaseI for removal of DNA contamination prior to reverse transcription. The High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA) was used for reverse transcription following the manufacturer’s instructions. A total of 40 PCR cycles were performed using the Biomark 96.96 dynamic array integrated fluidic circuit developed by Fluidigm (South San Francisco, CA, USA) and 96 primer pairs. The primers were designed against a variety of genes potentially important for embryonic development. The primers were used previously (Siquiera and Hansen, 2016; Negrón-Peréz et al., 2017; Siqueira et al., 2017), and the rationale for their inclusion is described in those papers. Among the genes are markers of blastocyst lineages, transcriptional regulators, cell signaling molecules, imprinted genes, apoptosis genes, and tight junction genes. The list of gene symbols and primer sequences is presented in Supplementary Table 1.

Statistical analysis

Analysis of effects of treatment on the proportion of putative zygotes undergoing cleavage and the percent of cleaved embryos and putative zygotes developing to the blastocyst stage for all experiments was evaluated using Proc GLIMMIX of SAS for Windows, version 9.4 (SAS Institute Inc., Cary, NC, USA). Each embryo was considered an observation with binary response (0 = not developed to blastocyst, 1 = developed to blastocyst) and analysis was performed by logistic regression fitting binary data distribution. The statistical model included the fixed effects of treatment, and random effect of replicate. Contrasts were calculated to determine concentrations of CSF3 that differed from the zero concentration (control).

For analysis of gene expression, the geometric means of the three housekeeping genes, ACTB, H2AFZ, and HPRT1, were calculated and used to obtain the delta Ct values of the other genes. Fold changes were calculated as 2^ΔCt relative to the geometric mean of the housekeeping genes. Treatment effects on gene expression (fold change) were determined by analysis of variance using the GLM procedure of SAS for Windows, version 9.4 (SAS Institute Inc., Cary, NC, USA). The statistical model included the fixed effects of treatment. When the main effect of treatment was significant, contrasts were calculated to determine concentrations of CSF3 that differed from the zero concentration (control).

Results

Experiment 1: effects of CSF3 during maturation on developmental competence of embryos

Results for developmental competence are shown in Figure 1. There was no effect of CSF3 on the percent of putative zygotes that cleaved. Similarly, there was no effect of 1 or 10 ng/mL CSF3 on the percent of cleaved embryos becoming a blastocyst or on the percent of putative zygotes becoming a blastocyst. However, putative zygotes derived from oocytes cultured with 100 ng/mL CSF3 were less likely to develop to the blastocyst stage than zygotes for other groups. Differences between the 0 and 100 ng/mL concentrations were significant for percent of cleaved embryos becoming blastocysts (P = 0.0254) and for percent of putative zygotes becoming blastocysts (P = 0.0366).

Figure 1.

Effect of addition of various concentrations of CSF3 during oocyte maturation on resultant cleavage and development of cleaved embryos and putative zygotes to the blastocyst stage. Data shown are least-squares means ± SEM. *Different from 0 ng/mL CSF3 (P < 0.05).

There were only four genes whose expression at the blastocyst stage was significantly affected by exposure to CSF3 during oocyte maturation (Figure 2). Each of these four genes was differentially regulated in blastocysts derived from oocytes treated with CSF3 during maturation. CYP11A1 was reduced by all concentrations of CSF3, RAC1 was decreased by 1 ng/mL CSF3 only, NOTCH2 was reduced by 10 ng/mL CSF3 and increased by 100 ng/mL CSF3. YAP1 was reduced in blastocysts derived from oocytes treated with 1, 10, or 100 ng/mL CSF3. Expression of the other genes evaluated were not significantly affected by G-CSF treatment.

Figure 2.

Gene expression for genes whose transcript abundance at the blastocyst stage was affected by addition of CSF3 to cumulus-oocyte complexes during maturation. Data shown are least-squares means ± SEM. Asterisks represent means that differ from untreated oocytes (i.e., 0 ng/mL CSF3). *P < 0.05; **P < 0.01.

Experiment 2: effects of CSF3 during day 0 to 7.5 of embryo culture on developmental competence of female and male embryos

Results on developmental competence are shown in Figure 3. There was no effect (P > 0.10) of CSF3, sex or the interaction on the percent of putative zygotes that cleaved, the percent of cleaved embryos becoming a blastocyst, or the percent of putative zygotes becoming a blastocyst.

Figure 3.

Effect of sex and addition of various concentrations of CSF3 at the beginning of embryo culture on cleavage rate and development of cleaved embryos and putative zygotes to the blastocyst stage. Blue represents embryos produced with Y-sorted semen (~90% male) and pink represents embryos produced with X-sorted semen (~90% female). Data shown are least-squares means ± SEM.

Experiment 3: effects of CSF3 added at day 4 of embryo culture on developmental competence and blastocyst gene expression

Results on developmental competence are shown in Figure 4. There was no effect of CSF3 on the percent of putative zygotes that cleaved, the percent of cleaved embryos becoming a blastocyst, or the percent of putative zygotes becoming a blastocyst. When gene expression was analyzed, there were 12 genes whose expression in the blastocyst was affected by CSF3 (Figure 5). With the exception of SFN, CSF3 increased expression of all genes relative to untreated controls. For SFN, the main effect of CSF3 was significant (P < 0.05), but there was no concentration of CSF3 that was significantly different from the control. For CREM1, expression was very low (fold-change relative to housekeeping genes was 0.001 to 0.004), and the only concentration affecting expression was 1 ng/mL. For the remaining 10 genes, the only concentration of CSF3 affecting gene expression was 100 ng/mL or, for the case of MKRN3, 10 and 100 ng/mL G-CSF.

Figure 4.

Effect of addition of various concentrations of CSF3 at day 4 of culture on cleavage rate and development of cleaved embryos and putative zygotes to the blastocyst stage. Data shown are least-squares means ± SEM. Effects of CSF3 were not significant.

Figure 5.

Gene expression for genes whose transcript abundance at the blastocyst stage was affected by addition of CSF3 from day 4 to 7.5 of culture. Data shown are least-squares means ± SEM. Asterisks represent means that differ from untreated oocytes (i.e., 0 ng/mL CSF3). *P < 0.05; **P < 0.01.

Discussion

Experiments reported here indicate a lack of effect of CSF3 on observed functions of the oocyte or preimplantation embryo except when the cytokine was added at 100 ng/mL. These results imply that administration of commercial preparations of CSF3 to reduce incidence of mastitis are unlikely to affect reproductive function through direct actions on the maturing oocyte or developing embryo. This conclusion is based on the anticipated concentrations of CSF3 in the blood of cows treated with pegylated CSF3. The one product approved for use in cattle, Imrestor, involves two injections of 15 mg pegylated CSF3 about a week apart approximately 7 d prior to calving and at the time of calving. Assuming a post-parturient cow weighs 600 kg (Ortega et al., 2017), this dose is equivalent to 25 μg/kg body weight. Concentrations of CSF3 in the blood of cows treated with pegylated CSF3 have not been reported. In the rat, however, concentrations of CSF3 in the blood reached above 100 ng/mL at an injection dose of 400 μg/kg, s.c. (Kang and Lee, 2013). In one experiment in the human, achieving concentrations of 100 ng/mL CSF3 in the blood required injection of between 3.6 (~52 μg/kg) and 6 mg (~89 μg/kg) of pegylated CSF3 (Ahn et al., 2013). In another experiment, injection of 50 μg/kg of pegylated CSF3 resulted in peak concentrations of CSF3 in the blood of ~90 ng/mL (Misra et al., 2018). It is also important to consider that pegylated CSF3 is less active biologically than non-pegylated recombinant CSF3 (Son et al., 2012; Kang and Lee, 2013; Kunstelj et al., 2013). Thus, not all immunoreactive CSF3 in the blood caused by administration of pegylated CSF3 is capable of regulating cellular function.

Maximum activation of proliferation of the CSF3-dependent NFS-60 cell line was achieved at a concentration of ~1 ng/mL (Son et al., 2012) and even pegylated CSF3 exhibits pro-proliferative activity at this concentration (Son et al., 2012; Wadhwa et al., 2015; Shekhawat et al., 2019). Similarly, other biological effects of CSF3 occur within the range of 0.1 to 10 ng/mL including stimulation of satellite cell formation in cultured myofibers (Hayashiji et al., 2015), neutrophil apoptosis (van Raam et al., 2008), and inhibition of NO production by lipopolysaccharide-activated neutrophils (Peng, 2017). The most likely explanation for the observation that CSF3 only affected oocytes and embryos at 100 ng/mL is that the molecule was activating cell signaling through a receptor distinct from that for CSF3. Indeed, CSF3R was not expressed in oocytes or preimplantation embryos harvested from cows (Jiang et al., 2014) and was either not expressed in blastocysts produced in vitro (Zolini et al., 2020) or was expressed at a very low level in both morula and blastocysts produced in vitro (Ozawa et al., 2012). Expression of the receptor in cumulus cells is also low (Sugimura et al., 2017). The CSF3 receptor is a member of the hematopoietin receptor superfamily (Fukunaga et al., 1990), and it is possible that one or more of these other receptors is activated at high concentrations of CSF3.

The deviation in gene expression in the blastocyst caused by CSF3 actions on the oocyte during maturation were slight. Thus, even at high concentrations, results are largely inconsistent with the idea that actions of CSF3 on the oocyte cause large changes in function of the resultant blastocyst. In contrast, exposure of the embryo to CSF3 from day 4 to 7.5 caused more extensive changes in the pattern of gene expression in the blastocyst than did addition of CSF3 during oocyte maturation. Of the 10 genes that were upregulated in blastocysts that were exposed to 100 ng/mL CSF2, seven were associated with epiblast or hypoblast. In particular, Negrón-Peréz et al. (2017) identified KDM2B and NANOG as markers of epiblast, ALPL, FN1, and PDGFRA as hypoblast markers and HNF4A as a marker for epiblast and hypoblast. Moreover, another upregulated gene, SOX2, is also expressed preferentially in the inner cell mass (Ozawa et al., 2012). Only one upregulated gene, TJAP1, was identified as a marker for a subset of trophectoderm cells by Negrón-Peréz et al. (2017). Given the large number of regulated genes that are markers of epiblast or hypoblast, both of which are derived from the inner cell mass, it is possible that CSF3 acts on the developing embryo at high concentrations to preferentially regulate the function of the inner cell mass.

The fact that CSF3 did not affect the function of the oocyte or embryo at concentrations expected to be achieved after therapeutic treatment with pegylated CSF3 does not mean that CSF3 treatment is without effect on reproductive processes. Other tissues involved in reproduction could be targets of CSF3, including the oviduct, endometrium, and leukocytes. Non-significant but positive effects of CSF3 on pregnancy/AI have been observed (Canning et al., 2017; Zinicola et al., 2018). Treatment of cows with pegylated CSF3 increase circulating numbers of neutrophils, monocytes, and lymphocytes and improves neutrophil function (Kimura et al., 2014; McDougall et al., 2017; Van Schyndel et al., 2018; Zinicola et al., 2018; Putz et al., 2019). Inflammatory disease in the postpartum period is a large risk factor for low fertility (Ribeiro et al., 2016) and it is possible that reduced incidence of mastitis (Canning et al., 2017; Ruiz et al., 2017) or other inflammatory diseases during the postpartum period achieved by treatment with pegylated CSF3 has additional positive effects on fertility.

Conflict of interest statement

The authors declare that there are no conflicts of interest.

Glossary

Abbreviations

COC

cumulus-oocyte complexes

CSF3

colony-stimulating factor 3

IVF

in vitro fertilization

IVF-TALP

in vitro fertilization Tyrode’s albumin lactate pyruvate solution

SOF-BE2

synthetic oviduct fluid – bovine embryo 2.