Introduction
Male sub-fertility accounts for 50% of all causes of infertility in couples seeking in vitro fertilization (IVF) treatment [
1]. This can be due to low sperm count, aberrant sperm motility, or abnormal morphology [
2]. Male sub-fertility is typically addressed using intracytoplasmic sperm injection (ICSI) [
3,
4]. This procedure requires an experienced embryologist who uses a micromanipulator to inject a spermatozoon directly into an oocyte [
5].
The ICSI procedure requires a microscope equipped with dual manually controlled manipulators, one of which maneuvers a holding pipette and the other an injection pipette [
6]. The holding pipette maintains the position of the oocyte using negative pressure during microinjection. The injection pipette is used to aspirate sperm, adjust the orientation of the oocyte, and inject a spermatozoon into the oocyte [
7‐
9]. Adding to the complexity, this process occurs within a confined microliter volume with the embryologist required to consistently refocus and micromanipulate within this miniscule 3D space [
10,
11]. With such a manually intensive procedure, it is not surprising that fertilization and implantation rates are positively associated with increased embryologist experience (e.g., those who have performed < 500 ICSI cycles vs those who have performed > 1000 ICSI cycles [
12].
Intracytoplasmic sperm injection is routinely performed on multiple oocytes from a single patient [
13,
14]. As a result, multiple oocytes are injected sequentially within the same microliter volume of medium. Thus, tracing injected vs non-injected oocytes adds an additional level of difficulty to this procedure. In the instance of many oocytes requiring injection, the procedure may become more inefficient with gametes remaining outside the incubator for an extended period. In turn, this may lead to impaired embryo production [
15,
16].
Intracytoplasmic sperm injection can be a technically challenging procedure with its success being influenced by human variability [
12,
14,
15]. This procedural variation can induce mechanical stress on the oocyte, negatively impacting fertilization [
17]. Additionally, mechanical stress during ICSI may lead to compromised DNA integrity: a cause of oocyte degeneration [
18]. Thus, we hypothesized that oocyte microinjection would be simplified with a procedure that requires only one micromanipulator. Furthermore, we hypothesized that a device that houses multiple oocytes in a linear array would decrease the time required for multiple injections and improve tracing of injected vs non-injected oocytes.
In the present study, we designed and fabricated a micrometer scale device that houses the oocyte. The device minimizes oocyte manipulation, requiring only one micromanipulator to perform microinjection. Here, we investigated the use of the device by (1) assessing biocompatibility and embryo culture performance within the device; (2) using the device to microinject presumptive zygotes; and (3) investigate the potential for high-throughput microinjection and improved tracing of injected vs non-injected oocytes.
Materials and methods
Unless otherwise stated, all chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA).
Fabrication of the pod and garage
Our device is comprised of 2 components, the Pod and Garage, which were designed using 3D modeling software (SolidWorks®, Dassault Systèmes SE, Paris, France). Fabrication of the devices was performed using two-photon polymerization technology, using a Nanoscribe Photonic Professional GT printer (Nanoscribe GmBH, Eggenstein-Leopoldshafen, Germany).
The 3D design files (standard tessellation language;.STL) were imported into Describe software (Nanoscribe, Karlsruhe, Germany). The fabrication parameters (i.e., laser printing pattern, structure fill: solid or shell and scaffold) were set into a fabrication job file (.GWL). Fabrication was then performed by direct laser writing. The writing speed and laser power were set to 75 MHz and 75%, respectively.
A 25 × microscope objective, numerical aperture 0.8, was used to focus the laser beam into the sample. The Pods and Garages were fabricated onto a substrate (50 × 50 × 0.55 mm indium-tin oxide glass slide, Fluke Australia Pty Ltd., Baulkham Hills, NSW, Australia). Prior to printing, the glass substrate was rinsed with ethanol then isopropyl alcohol and dried with compressed air.
An IP-S photoresist resin (Nanoscribe GmBH) was used to fabricate the Pods and Garages. Following printing, the excess unfabricated resin was removed by washing in isopropyl alcohol for 8 min, followed by 8-min development (SU-8 developer; Nanoscribe GmBH), and then dried using compressed air. Next, the substrate was submerged in 5% 7X-O-Matic cleaning solution (MP Biomedicals, Solon, OH, USA) overnight at room temperature (RT). Following fabrication, the Pods and Garages were carefully removed from the glass substrate. Once removed, the Pods and Garages were washed three times in 5% 7X-O-Matic cleaning solution at RT. This was followed by three overnight washes in filtered phosphate-buffered saline (PBS; one tablet per 200 mL of Milli-Q water) at RT. The Pods and Garages were stored in PBS at 4 °C until use.
Animals
All experiments were approved by the University of Adelaide Animal Ethics Committee (M-2019–008) and conducted in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes.
Female (pre-pubertal, 3–4 weeks old) and male (6–8 weeks old) CBA × C57BL/6 first filial generation (F1) mice (CBAF1) were obtained from the University of Adelaide Laboratory Animal Services and maintained under 12-h light to 12-h dark cycle with rodent chow and water provided ad libitum.
All gamete and embryo culture took place in media overlaid with paraffin oil (Merck Group, Darmstadt, Germany) at 37 °C in a humidified incubator with 5% O2 and 6% CO2 balanced with N2. For the Pod and Garage treatment groups, a single Garage and 3 individual Pods were placed into a drop of culture medium using fine forceps. Culture dishes for all treatment groups were then pre-equilibrated for at least 4 h prior to use. Oocyte and embryo handling were carried out on a heated stage with the temperature set at 37 °C. Mouse tissues were collected in Research Wash Medium (IVF VET Solutions, SA, Australia) supplemented with 4 mg/mL low fatty acid bovine serum albumin (BSA; MP Biomedicals, Albumin NZ™, Auckland, New Zealand). Research Cleave Medium (IVF VET Solutions) was also supplemented with 4 mg/mL BSA and used for embryo culture (manufacturer’s recommended density = 2 μL/embryo). Embryos were cultured from the zygote to blastocyst stage in a single-step culture system (i.e., no media change occurred on day 3 of development).
Isolation of mouse cumulus oocyte complexes
Pre-pubertal female mice were injected intraperitoneally (i.p.) with 5 IU equine chorionic gonadotrophin (eCG; Folligon; Pacific Vet Pty Ltd., Braeside, VIC, Australia) followed by 5 IU (i.p.) human chorionic gonadotrophin (hCG; Pregnyl; Merck, Kilsyth, VIC, Australia) 46–48 h later. Mice were culled via cervical dislocation, and the ampullae of the oviducts dissected in warmed Research Wash Medium. Ovulated cumulus oocyte complexes (COCs) (14–16 h post-hCG) were isolated by puncturing the ampullae in warmed Research Wash Medium. The isolated COCs were briefly incubated in hyaluronidase (6.1 μM) diluted in warmed Research Wash Medium for 1 min to remove cumulus cells with the aid of gentle pipetting.
Isolation of mouse presumptive zygotes
Pre-pubertal female mice were injected with eCG (5 IU; i.p.) followed 46–48 h later by 5 IU if hCG (i.p.). Females were then paired overnight with males, with mating confirmed the following morning by the presence of a copulation plug. Female mice were culled via cervical dislocation and the ampullae dissected to isolate presumptive zygotes (PZs) (22–24 h post-hCG). Cumulus-enclosed PZs were incubated in hyaluronidase (6.1 μM) diluted in warmed Research Wash Medium for 1 min to remove cumulus cells with the aid of gentle pipetting.
Analysis of pod and garage embryo toxicity
The toxicity of the Pod and Garage was assessed using a standard mouse embryo assay (MEA) that used both negative and positive controls, with a certificate of assessment provided (IVF VET Solutions, SA, Australia) [
19]. The Pods and Garages were soaked in protein-free MEA medium and incubated overnight at 37 °C in a humidified incubator with 5% O
2 and 6% CO
2 balanced with N
2. Embryo culture drops (10 PZs/20 μL) were then prepared using the MEA medium utilized to wash the Pods and Garages. In addition to the MEA test, embryo culture from the PZ to the blastocyst stage was conducted within Pods docked in a Garage (three PZs docked individually in 3 Pods and a Garage/10 μL) using Research Cleave Medium supplemented with BSA. Fertilization rate was scored 24 h later, with embryos then allowed to develop to the blastocyst stage within Pods and a Garage. At 96 h post fertilization, embryos were considered on-time if at the blastocyst stage (i.e., having a blastocoel cavity ≥ two-thirds the size of the embryo; or expanded; or hatching).
Analysis of DNA damage/repair in cultured blastocysts (phosphorylated-histone-H2A.X; γH2A.X)
Unless otherwise stated, all immunohistochemistry procedure was carried out at RT. Following either standard embryo culture or embryo culture within the Pods and a Garage, embryos were stained with γH2A.X to assess for double-stranded DNA repair [
20]. The blastocysts were fixed in 200 μL 4% paraformaldehyde diluted in PBS (w/v) for 30 min following fixation, blastocysts were washed with PBV (0.3 mg/mL polyvinyl alcohol diluted in PBS) and permeabilized for 30 min in 0.25% (v/v) Triton X-100 in PBS. Blastocysts were then blocked for 1 h with 10% goat serum (v/v; Jackson Immuno, PA, USA) diluted in PBV. Following blocking, blastocysts were incubated overnight in the dark with anti-γH2A.X primary antibody (Cell Signaling Technology, MA, USA) at a 1:200 dilution with 10% goat serum in PBV (v/v). A negative control was included where embryos were incubated in the absence of the primary antibody. Next, embryos were washed three times in PBV before incubation for 2 h in the dark with anti-rabbit Alexa Fluor 594-conjugated secondary antibody (Life Technologies, Carlsbad, CA) at 1:500 dilution with 10% goat serum in PBV (v/v). Embryos were then counterstained with 3 mM of 4′,6-diamidino-2-phenylindole (DAPI; Thermo Fisher Scientific, MA, USA). Finally, embryos were washed three times in PBV and transferred onto a glass microscope slide with DAKO mounting medium (Dako Inc., CA, USA) and enclosed with a coverslip using a spacer (Thermo Fisher Scientific, MA, USA). Embryos were imaged using an Olympus Fluoview 3000 confocal microscope (Olympus Life Science, Tokyo, Japan). Images were captured at 60 × magnification, using the imaging channels Alexa Fluor 594 (red) for γH2A.X (591/614 nm) and DAPI (blue) for DNA (358/461 nm). A z-stack projection for each blastocyst was generated using images captured at 4-μm intervals. The same imaging parameters were kept for each replicate. The intensity of γH2A.X immunostaining was quantified using Fiji ImageJ software (National Institute of Health, MD, USA).
Comparing standard microinjection vs microinjection within the Pods and a Garage
Microinjection process was assessed to compare technical components for standard microinjection vs within the Pods and a Garage under a micromanipulator. Microinjection was performed in a 60-mm petri dish lid (Falcon, Corning, In Vitro Technologies, VIC, Australia). The microinjection drops were prepared with warmed Research Wash Medium (5 × 10 μL drops at the center of the dish overlaid with paraffin oil). The microinjection dish was pre-equilibrated on a heated stage at 37 °C for at least 4 h before use.
Mouse oocytes were loaded into the microinjection drops to compare standard microinjection (three oocytes/drop) and microinjection within the Pods and a Garage (three oocytes within Pods and a Garage/drop). For standard microinjection, the holding pipette (inner diameter: 17 μm; outer diameter: 80 μm; bevel: 30°; Cook Medical, PA, USA) and injection pipette (inner diameter: 5 μm; outer diameter: 7 μm; bevel: 20°; Cook Medical) were mounted into the micromanipulators. Conversely, for microinjection within the Pods and a Garage, only the injection pipette was used. For standard microinjection and microinjection within a Pod and Garage, the orientation of the oocyte with respect to the polar body was adjusted to either the 6 or 12 o’clock position by placing the injection pipette above the oocyte (in proximity to either the top or the bottom of the oocyte) and moving the pipette along the z-axis.
Microinjection of PZs with fluorescent microspheres
To demonstrate the application of the device for ICSI, microinjection of 4-μm fluorescent microspheres (Invitrogen, Thermo Fisher Scientific, MA, USA) was performed in PZs within the Pods and a Garage. Microinjection of PZs with fluorescent microspheres occurred under a Nikon Eclipse TE2000-E inverse microscope (Nikon Instruments Inc.) equipped with a Tokai Hit ThermoPlate set at 37.5 °C.
Only the injection pipette was loaded into the micromanipulator and used to perform microinjection. The microspheres were aspirated from a separate drop consisting of warmed Research Wash Medium and were then individually injected into each PZ.
Following microsphere microinjection, PZs were transferred from within the Pods and a Garage into pre-equilibrated 2 μL Research Wash Medium on a glass-bottomed confocal dish (Cell E&G, Houston, TX, USA) overlaid with paraffin oil and imaged under the Olympus Fluoview 10i confocal microscope. The fluorescent microspheres were then visualized using the red fluorescence channel (660/680 nm). The microinjected PZs were then cultured in Research Cleave Medium and were allowed to develop to the blastocyst stage.
Assessment of multiple oocyte microinjection capability within Pods and Garage
The feasibility of microinjecting multiple oocytes within the Pods and a Garage was then tested and compared to microinjecting multiple oocytes using the standard procedure. Microinjection was performed on a microinjection dish with three oocytes loaded per drop for standard microinjection and for microinjection in our device (three oocytes/3 Pods docked in a Garage) within a separate drop. The microinjection drop size for standard microinjection and microinjection within the Pods and a Garage was 10 μL. Under standard microinjection, oocytes were placed into the drop using a pulled glass pipette. For microinjection within the Pods and a Garage, oocytes were loaded into a Pod (1 oocyte per Pod) using a pulled glass pipette. Each Pod was then docked within a Garage using fine forceps.
During standard microinjection, the holding and injection pipettes were utilized to locate and hold individual oocytes prior to microinjection manually. Subsequent microinjection of oocytes was performed after the injected oocyte was released and separated from the non-injected oocyte cohort within the same drop. This process was also performed manually using the holding and injection pipettes. Conversely, for microinjection within the Pods and a Garage, only the injection pipette was utilized. Oocytes were arranged next to each other within our system prior to microinjection. The micromanipulator stage was controlled manually to lead the injection pipette into the non-injected oocyte through the injection pipette channel of the Pod. The micromanipulator stage was then used to facilitate microinjection. Subsequent microinjection of non-injected oocytes was also performed within adjacent oocytes within the remaining Pods docked in a Garage using the micromanipulator stage.
Comparative analysis of time taken to microinject oocytes within the Pods and Garage vs standard microinjection
Three mouse oocytes were preloaded into individual Pods that were then docked within a Garage. Quantification of time required for individual parameters of the microinjection procedure and comparison between standard and microinjection within the Pods and a Garage was performed. Key parameters considered for this experiment were setting up of pipettes (holding and injection pipettes), holding oocytes before microinjection, and injecting each oocyte. Each parameter within the entire procedure was measured individually (Fig.
5b).
Statistical analysis
All statistical analyses were performed using GraphPad Prism 8.0 for Windows (GraphPad Software, San Diego, CA). Embryo development data were arcsine transformed prior to statistical analysis. All experimental data were tested for normality to determine whether a parametric or non-parametric test should be used. Statistical analyses were performed using a Student’s t-test as described in the figure legends. A P-value < 0.05 was considered statistically significant.
Discussion
Successful fertilization following ICSI is dependent on the training and expertise of the embryologist performing the procedure [
14,
15]. In the case of a less experienced operator, oocytes may be subjected to higher levels of mechanical stress (causing oocyte lysis) and spend extended periods of time outside an incubator. This negatively impacts fertilization success [
24]. Therefore, there exists a need to simplify the procedure and reduce the dependence upon the experience of the operator. We address this need in the current study using our novel device, which we term the Pod and Garage, which minimizes oocyte handling during microinjection. The device improves oocyte microinjection by (1) removing the need for a holding pipette; (2) simplifying and reducing the duration of the procedure; (3) improving the traceability of injected vs non-injected oocytes; and (4) showing the path to high-throughput microinjection.
Following printing of the Pod and Garage using photopolymerization, we verified that the device was suitable for future clinical use. The gold-standard procedure used to determine whether products are suitable for use in the IVF laboratory is to evaluate preimplantation development of mouse embryos under defined conditions and in culture medium that has previously been in contact with the potential toxicant [
25]. This test is known as the mouse embryo assay (MEA). It is the primary quality control measure for all equipment and consumables in an IVF clinic [
26]. The Pods and Garage passed a standard MEA with an accepted blastocyst rate above 80% [
19]. Additionally, embryos cultured within the device developed to the blastocyst stage at rate comparable to those cultured in standard conditions. We also investigated more subtle effects on embryo health. We found similar levels of DNA damage/repair in embryos cultured in standard conditions compared to those cultured within our device. Taken together, these results suggest that the printed device is embryo-safe; however, further work demonstrating safety is required prior to clinical use. In a follow-up study currently underway, cell numbers within the divergent cell lineages of the blastocyst as well as implantation rates and fetal health following embryo transfer will be used to assess the safety of the device. The biocompatibility of the Pod and Garage shown in the current study is consistent with previous work. In those studies, the same polymer was used to fabricate devices for cell culture and drug delivery [
27,
28].
Microinjection of oocytes within our device removed the need for a holding pipette, requiring only one micromanipulator equipped with an injection pipette. Furthermore, by docking multiple oocytes within our device, we improved traceability—tracking injected from non-injected oocytes—which also facilitated high-throughput microinjection of multiple oocytes. This is in contrast with standard practice where operation of a second micromanipulator is required to hold the oocyte. In such standard practice, tracing of injected from non-injected oocytes can become difficult in a microliter volume of medium. The Pod and Garage brings a step-change to the procedure, simplifying the process and minimizing oocyte handling and mechanical stress during microinjection.
One crucial point is that direct embryo handling was minimized within our device. The only period when oocytes and embryos were directly handled using a pipette was when they were loaded and unloaded from the Pods. The Pods were docked and undocked into a Garage using fine forceps. Moving the docked Garage between dishes and into different drops (e.g., from the culture dish to an ICSI dish) was performed with the aid of fine forceps. This was simpler than repeated handling and transfer using a micropipette. As no further handling of the oocyte or embryo occurred, we anticipate mechanical stress induced by repeated micro-pipetting and handling in standard practice to be obviated with the use of our device. This will be investigated in a future study.
In the current study, we demonstrated capability to perform ICSI using our device. Microinjection of presumptive zygotes with fluorescent microspheres persisted in the embryo throughout development to the blastocyst stage. This demonstrated that our system supports microinjection with no impairment to the embryo following injection. Importantly, future studies will directly compare standard ICSI with ICSI performed within the Pods and Garage using a compatible animal model. These studies are currently underway. Although these results are encouraging, further examination of short- and long-term development is required before the implementation in the clinic.
In conclusion, microinjection within our device minimizes the requirement for an experienced operator for handling and manipulation. This work suggests the Pod and Garage may improve embryo production and that they may form a precursor to automated ICSI.
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.