Animals

All procedures involving animals were carried out according to the German Animal Welfare Act, the relevant guidelines and approved by the competent authority, i.e. Niedersächsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit (LAVES; Animal Experiment Permission #33.19–42,502-04–19/3221). This study was carried out as part of the hormonal stimulation optimization protocol for a larger project consisting of multiple subprojects aiming at the production of genetically modified marmoset monkey model intended for further research.

The animals used in this study are part of the self-sustained research colony of the German Primate Center, which has originally been established in the 70-ties with C.jacchus imported from England and which now includes around 480 animals. All animals used for this study were housed according to the applicable legal guidelines and institutional practice of the German Primate Center for C. jacchus [5, 32, 37, 38]. Well-being of the animals was controlled daily by animal care personnel with additional regular health check-ups by veterinarians. Marmosets were housed at 25 ± 1 °C and 65 ± 5% relative humidity and air circulation and filtration at artificially controlled light/dark cycle (12/12 h). Vertically-oriented stainless steel cages [165 cm (H) × 65 cm (W) × 80 cm (D)] were furnished with wooden elements for environmental enrichment. Rooms and cages were cleaned with water biweekly. The animals received pelleted primate food Altromin 0633 (Altromin Spezialfutter GmbH & Co. KG, Lage, Germany) ad libitum. In addition, approx. 20 g mash (bananas, carrot juice, yogurt and quark enriched with vitamins and minerals) per animal was served in the morning and 30 g fruit or vegetables mixed with rice and gum Arabic were provided in the afternoon. Furthermore, mealworms or locusts were offered three times per week. Total daily vitamin D3 intake provided as a liquid supplement to the morning mash consisted of approx. 140 IU/animal, while additional approx. 150 IU/animal were provided with food pellets. Drinking water was always available in the cages while fruit teas were provided occasionally.

In total, 14 female marmosets were used as oocyte donors in this study (age: from 3.3 to 6.1 years (mean 4.6 ± 0.7 (SD) y), bodyweight: from 371 to 528 g (mean 442 ± 35.5 g)). Some animals have previously undergone one or more OPU procedures, but their oocytes/embryos were cultured under different conditions; therefore, data from these operations were not included in this study. Oocyte donors were housed with male partners to ensure the optimal ovarian function [39]. In addition, 5 males with proven fertility were used as sperm donors (age: from 5.6 to 7.7 years (mean 6.7 ± 0.9 years), bodyweight: from 368 to 561 g (mean 443 ± 39.1 g)). These were housed in male-male pairs to avoid copulations, which could affect the volume of the sample and/or its quality. In our colony, neither homosexual behaviour nor masturbation have been observed of yet (own observations).

Cycle monitoring and control

Ovarian cycles were monitored by plasma progesterone levels measurements performed twice weekly [9, 40] and carried out by enzyme immunoassay as described earlier [41]. In brief, 0.2 mL blood was collected from the Vena femoralis using a 1 mL syringe with a 26G heparin-primed needle. The blood samples were kept on ice and transferred to the endocrinology laboratory, where the samples were centrifuged at 1500 g for 10 min at RT to obtain the plasma for progesterone analysis. Progesterone was determined by a direct, non-extraction EIA using a monoclonal antibody to 11-hydroxyprogesterone-hemisuccinate: BSA (Quidel clone no. 425; CL425) [41]. Progesterone–horseradish peroxidase (HRP) was used as enzyme conjugate in the EIA. Samples were usually measured in a 1:30 dilution in assay buffer (TBS, pH 7.2). The assay has previously been analytically and biologically validated [41], and inter-assay coefficients of variation of high- and low-value quality controls run in each assay were both < 15% across all measurements. Progesterone concentrations < 10 ng mL−1 indicated the follicular phase, whereas concentrations > 10 ng mL−1 indicated the luteal phase of the ovarian cycle [9, 40].

To time the beginning of hormonal stimulation and to prevent natural pregnancies, luteolysis was induced in oocyte donors by prostaglandin F2α analogue (PGF-2α, Estrumate, Essex Tierarznei, Munich, Germany) given i.m. [38].

On the day following PGF-2α injection, i.e. day 1 of the new cycle, hormonal stimulation was initiated. Injections of 25–35 IU recombinant human FSH (Gonal-f, Merck Europe B.V., Amsterdam, Niederlande) were performed daily at 09:00 with the total duration of FSH-stimulation according to the study protocol (see below). In vivo ovulation was triggered with the injection of 75 IU hCG (Ovogest, MSD Animal Health, Unterschleissheim, Germany) given at 15:00 on the last day of stimulation, and the operation (OPU or ovariohysterectomy (OvH)) was performed 18–20 h later.

Study design

Two stimulation approaches were investigated in this study: stimulation with hFSH plus in vivo maturation triggering with hCG (hereafter referred to as "FSH + hCG") was compared with priming with human recombinant FSH only (hFSH; hereafter referred to as "FSH-priming"). In total, 20 FSH + hCG stimulation cycles were performed in 11 individual animals. Out of these, 16 stimulations included eight (8) daily injections of FSH, and in four (4) stimulation cycles the duration of the FSH-stimulation was individually adjusted (see below) and varied between 6 and 9 days. Eighteen stimulations of 9 individual animals were performed as FSH-priming for 6–9 days. The stimulation length chosen depended on the individual follicular phase length, which was calculated based on the data from previous cycles. Additionally, individual responsiveness to FSH in previous stimulation cycles was considered. In case of rising progesterone observed on the day of operation, the duration of FSH-stimulation was reduced in the following stimulation cycles. The operations were always planned 1–2 days before the predicted day of the ovulation (Table 1).

Table 1 Two hormonal stimulation protocols applied in Callithrix jacchus for recovery of oocytes via OPU or ovary dissection

There were no statistically significant differences in either age or body weight assessed at the moment of each operation between the animals of the two stimulation protocol groups (Fig. 1).

Fig. 1

Age (A) and bodyweight (B) of C. jacchus females in both study groups. The number of stimulations in each group is shown in square brackets

Individual adjustment of the duration of FSH-stimulation was based on the natural follicular phase (FP) length. The average length of the FP was calculated based on several cycles preceding the experiment. In addition, ovarian response to FSH was controlled by ultrasound and in the case of an unexpectedly fast growth of the follicles the current stimulation cycle was cancelled due to logistical reasons and inability to perform unscheduled surgeries. The duration of stimulation was adjusted accordingly in the next stimulation cycle.

OPU and OvH procedures

For each animal, the oocyte collection was performed in four stimulation cycles: the first three surgeries were OPU procedures, while the fourth surgery was carried out as an OvH. Each animal received Metacam (0.5 mg/mL meloxicam) 0.15 mg p.o. as premedication and 0.1 mg postoperatively over 3–4 days for pain management. In addition, all animals received antimicrobial treatment postoperatively (Hostamox LA 150 mg/mL, 15 mg/animal i.m. immediately post-OP and every second day for 4 days).

All operations were carried out under injection narcosis: to this end, ketamine 100 mg/mL at 25–50 mg/kg, medetomidine 1 mg/mL at 0.05–0.1 mg/kg, and atropine 0.513 mg/mL at 0.05 mg/kg) were administered i.m. For the last operation (OvH of one animal from the FSH-priming group), new narcosis was used, with the aim to improve the general tolerability and to shorten the postoperative recovery/waking-time. The new protocol consisted of alfaxalone 10 mg/mL at 10 mg/kg, midazolam 5 mg/mL at 0.25 mg/animal and buprenorphine 0.3 mg/mL at 0.005–0.025 mg/kg administered i.m., and bupivacaine at 2 mg/kg applied topically. The outcome of the operation carried out with the new narcosis was comparable to the outcomes seen with the old protocol. OPU was performed using an aspirator (Labotect Aspirator 3, Labotect, Göttingen, Germany). 23Gx3/4 puncture needles (BD, Allschwil, Switzerland) were connected to a sterile Heidelberger extension tube (#13,802, Labotect, Göttingen, Germany). The oocytes were collected into a 10 mL tube with pre-warmed (37 °C) handling medium (POE-CM ([Porcine Oocyte/Embryo Collection Medium], CosmoBio, Tokyo, Japan) supplemented with 5% FBS (Biochrom AG, Berlin, Germany) and 2.5 IU/mL heparin (Heparin-Natrium-5000-ratiopharm®, Ratiopharm, Ulm, Germany).

After ovariohysterectomy, the ovaries attached to the uterus were immediately placed into pre-warmed (37 °C) POE-CM + 5% FBS and transferred to the laboratory, where large ovarian follicles were mechanically dissected using 26G needles and/or forceps to recover oocytes.

Culture plate preparation

All culture steps were performed in 4-well dishes (NUNC, Thermo Fisher Scientific, Osterode am Harz, Germany, Catalogue number #179,830) in drops of the respective culture medium (see below) under light mineral oil (LiteOil®, LifeGlobal Europe, Brussels, Belgium): the drops in the first three wells were used for washing, followed by culturing in the fourth well. All culture plates (for short-term storage; in vitro maturation (IVM); in vitro fertilization (IVF); embryo culture) with drops of medium under pre-incubated light mineral oil were prepared the day before use and stored overnight in an incubator with gas conditions according to the developmental stage to assure proper pH (7.3) and dissolved oxygen concentration.

Oocyte collection and in vitro maturation

All available cumulus-oocyte complexes (COCs) and naked oocytes retrieved during the OPU procedure or dissected out of antral follicles from the removed ovaries were immediately transferred from the handling medium (see above) into hormone-free POM medium (CosmoBio, Tokyo, Japan) supplied with 5% FBS (Biochrom AG, Berlin, Germany) and left at 37.5 °C, 5.5% CO2 in air, until the collection was finished. Only healthy, non-degenerated COCs and oocytes were selected for culture. These were then transferred into culture dishes with 500 µL drops of IVM medium with hormones (POM, 5% FBS, 1 IU/mL FSH and 1 IU/mL hCG) under oil. IVM was performed at 37.5 °C, 5.5% CO2 in air, for 29-30 h. The choice of gas conditions has been described in detail in Tkachenko et al. [42]. All oocytes and embryos from different donors were always cultured separately.

Sperm collection and in vitro fertilization

At least two sperm donors were used for each experiment. Sperm was collected by penile vibrostimulation (PVS) procedure described in detail in [43]. Briefly, ejaculation was induced using a FertiCare personal vibrator (Multisept ApS, Rungsted, Denmark) fitted with a 2 cm × 0.5 cm glass tube with rounded edges serving as an artificial vagina. Sperm collection medium (HEPES-buffered Tyrode’s lactate with 0.25 mM sodium pyruvate (Sigma-Aldrich (Merck), Darmstadt, Germany) and 0.3% wt/vol Bovine Serum Albumin ((BSA, Sigma-Aldrich (Merck), Darmstadt, Germany), pH 7.3, 400 µL, 37 °C) was immediately added to the ejaculate to minimize coagulation of the seminal plasma. In the laboratory, sperm samples were washed using a density gradient centrifugation (300 g, 18 min, 37 °C, soft mode) The 40%/80% density gradient prepared with PureSperm solution (PureSpermTM 100, Nidacon, Mölndal, Sweden) and Sperm collection medium. Washed sperm samples were carefully pipetted on the bottom of glass tubes with 600 µL Capacitation medium (Tyrode’s lactate with 0.25 mM sodium pyruvate and 0.3% wt/vol BSA) and placed in the incubator with 37.5 °C, 5% O2/5.5% CO2 for 40 min for capacitation. Capacitated sperm samples were analysed for motility and the sample with the highest rapid progressive motility was chosen for IVF. On average, samples with concentration of ca. 5 000 000 sperm/mL and 80% motility were used for IVF. For IVF, the oocytes were transferred into 500 µL IVF medium (Tyrode’s lactate medium supplemented with 0.5 mM sodium pyruvate, 0.5 mM GlutaMAX (GlutaMAX™, Gibco® (Merck), Darmstadt, Germany), 2 mM CaCl2.2H2O, 1% v/v minimum nonessential amino acid solution (NEAA 100x, PAN Biotech, Aidenbach, Germany), 0.3% wt/vol BSA, 10 IU/mL penicillin, and 10 µg/mL streptomycin (Penicillin–Streptomycin, PAN Biotech, Aidenbach, Germany) under oil (37.5 °C, 5% O2/5.5% CO2). Sperm concentration was adjusted to 300.000 sperm/mL.

CRISPR injection and embryo culture

The following morning, 15–16 h after the sperm suspension was added into IVF drops, cumulus cells were removed by pipetting. Denuded oocytes were assessed for the presence of two polar bodies (PBs) and pronuclei (pn) and zygotes were injected with CRISPR/Cas9 to induce genetic modification (results will be published separately). All fertilized oocytes analysed in the present study received injection of CRISPR/Cas9, therefore no additional variation in the study groups was introduced by this manipulation. In brief, at the pronuclear stage, intracytoplasmic injection of CRISPR/Cas9 mixture (4 × 15 ng/μL sgRNAs, 100 ng/μL of Cas9 protein (NEB, Ipswich, USA), 25 ng/μL of hCas9 mRNA, 5 mM TrisHCL pH 7.4 buffer and 100 mM NaCl) was performed. The mixture was injected using 1.6 µm pronucleus injection pipettes (Type PI-1.6, BioMedical Instruments, Zöllnitz, Germany) attached to the FemtoJet 4i microinjector (Eppendorf, Hamburg, Germany). Immediately after CRISPR/Cas9 injection, zygotes were transferred into culture plates with 50 µL ORIGIO® Sequential Cleav™ (Origio, Måløv, Denmark). Starting from the 4-cell stage, embryos were cultured in ORIGIO® Sequential Blast™ (Origio, Måløv, Denmark) at 37.5 °C, 5% O2/5.5% CO2.

From the morula stage (12-cell stage or initiation of compaction), ORIGIO® Sequential Blast™ was additionally supplemented with 2.5% FBS. Throughout the embryo culture period, every two days the embryos were transferred into new culture 4-well dishes with fresh medium chosen according to the developmental stage. Pictures of all cultured embryos were taken daily, with detailed recording of the developmental stage.

Statistics

The analysis was performed using GraphPad Prism version 8.0 for Windows (GraphPad Software, Boston, Massachusetts USA, (www.graphpad.com). Differences between two groups were analysed using Student t-test for normally distributed samples; comparison of multiple groups was performed using one-way ANOVA with post-hoc Tukey multiple comparisons test. For evaluation of the relationships between variables, Pearson’s correlation coefficient was used. The results of correlation analysis are presented in the form of correlation coefficients and p-values. Differences were considered significant at p < 0.05, otherwise the differences are labelled as ‘ns’ (not significant) in the text. Unless indicated otherwise, the data are presented as mean ± SD.