Pregnancy establishment is a dynamic process involving complex embryo and uterine crosstalk. The precise contributions of the maternal uterine environment to these processes remain an active area of investigation. Here, detailed protocols are provided to aid in designing in vivo animal models to address these research questions.
For pregnancy to be established, a viable blastocyst must successfully interact with a receptive uterine lining (endometrium) to facilitate implantation and placenta formation and enable ongoing pregnancy. The limitations to pregnancy success caused by embryonic defects are well known and have been largely overcome in recent decades with the rise of in vitro fertilization (IVF) and assisted reproductive technologies. As yet, however, the field has not overcome the limitations caused by an inadequately receptive endometrium, thus resulting in stagnating IVF success rates. Ovarian and endometrial functions are closely intertwined, as hormones produced by the ovary are responsible for the endometrium’s menstrual cyclicity. As such, when using rodent models of pregnancy, it can be difficult to ascertain whether an observed result is due to an ovarian or uterine deficit. To overcome this, an ovariectomized mouse model was developed with embryo transfer or artificial decidualization to allow the study of uterine-specific contributions to pregnancy. This article will provide instructions on how to perform ovariectomy and offer insights into various techniques for supplying exogenous hormones to support successful artificial decidualization or pregnancy following embryo transfer from healthy donors. These techniques include subcutaneous injection, slow-release pellets, and osmotic mini pumps. The key advantages and disadvantages of each method will be discussed, enabling researchers to choose the best study design for their specific research question.
With the rising use of assisted reproductive technologies in recent decades, many barriers to conception have been overcome, allowing many couples to start families despite fertility problems1. Oocyte or sperm deficits can often be bypassed using in vitro fertilization or intracytoplasmic sperm injection; however, issues related to the uterus and endometrial receptivity remain an elusive "black box" of reproductive potential2.
Pregnancy is established when a high-quality embryo successfully interacts with a receptive endometrium (uterine lining). The chances of successful pregnancy in any given menstrual cycle are low, at around 30%3,4. Of those that are successful, only 50%-60% advance past 20 weeks of gestation, with implantation failure being responsible for 75% of pregnancies that do not reach 20 weeks3. Despite these figures dating back to the late 1990s, the field is yet to overcome the limitations caused by an inadequately receptive endometrium. This has resulted in stagnating – and sometimes declining – IVF success rates in recent years5,6.
Women with unexplained infertility often have a displaced window of receptivity or are unable to achieve receptivity for unknown reasons. Recently, the endometrial receptivity array was developed, which assesses the expression of hundreds of genes with the purpose of tailoring the timing of embryo transfer to an individual's window of receptivity7,8,9. However, the field still lacks an understanding of the pathogenesis of pregnancy complications that manifest after the implantation process is complete.
The female reproductive system is highly dynamic and under tight hormonal control. The hypothalamic-pituitary-gonadal (HPG) axis controls the release of luteinizing hormone and follicle-stimulating hormone, which regulate aspects of the ovarian cycle, including follicle maturation and estrogen and progesterone activity. In turn, the uterine menstrual cycle is regulated by estrogens and progesterone10,11. Thus, studying uterine biological mechanisms is complicated by ovarian influence. For example, when studying how cancer therapies may impact the uterus, it can be difficult to distinguish if any uterine phenotype observed (such as pregnancy loss or menstrual acyclicity) is the result of a direct insult to the uterus or a consequential effect from damage to the ovaries.
To comprehensively understand fertility, the uterine contributions to pregnancy must be characterized. Importantly, this understanding must extend beyond uterine function under ovarian control. This cannot be studied in humans; therefore, animal models are often employed. As such, ovariectomy (OVX) is commonly used to enable researchers to regulate rodent estrous cycles (analogous to the menstrual cycle) by supplying hormones exogenously. Additionally, OVX allows uterine responses to be studied independently of ovarian influence12. However, if hormones are not immediately supplied post-OVX, a menopause phenotype will eventuate, which needs to be carefully considered by the researchers.
OVX is frequently utilized in rodent models13,14,15,16,17 and is relatively easy to perform after adequate training. Methods vary depending on whether the ovary alone or the ovary and oviduct are removed, as well as depending on the age of the animal (adult, cycling animals have larger ovaries with a visible corpus luteum on their surface, meaning their ovaries are easier to visualize). Similarly, many methods of hormone supplementation exist, including subcutaneous injections14, slow-release pellets15, osmotic mini pumps18, and ovarian grafting.
In this article, detailed instructions are provided on how to perform ovariectomy and prepare three types of hormone supplementation, including subcutaneous injections, slow-release pellets, and osmotic mini pumps. Two detailed protocols are provided for experimental endpoints that benefit from OVX followed by exogenous hormone supplementation (embryo transfer and artificial decidualization). This article discusses the strengths and weaknesses of each approach with the goal of guiding researchers regarding how to perform studies to isolate the impacts on the uterus, specifically in the pregnancy and fertility fields of research.
All animals were housed in temperature-controlled, high-barrier facilities (Monash University Animal Research Laboratory) with free food and water access and a 12 h light-dark cycle. All the procedures were performed in accordance with approval from the Monash Animal Research Platform Ethics committee (#21908, 17971) and performed in accordance with the National Health and Medical Research Council Code of Practice for the care and use of animals.
1. Surgical preparation
2. Performing ovariectomy
3. Hormone preparation: Subcutaneous injection
4. Hormone preparation: Slow-release pellets
5. Hormone preparation: Osmotic mini pumps
6. Surgical procedure: Insertion of subcutaneous hormone pellets and mini pumps
7. Experimental procedure: Embryo transfer
8. Experimental procedure: Artificial decidualization
9. Surgical procedure: Post-surgical recovery, monitoring, and clip repairs
A well-characterized model of artificial decidualization is described in this protocol paper (Figure 1A). Here, young adult female mice (8 weeks old) underwent surgical ovariectomy as described in section 1 and section 2. The mice were then rested for 2 weeks to ensure that the endogenous ovarian hormones dissipated before being supported with exogenous hormones as described in sections 3-7 and section 9. Artificial decidualization was induced by an intravaginal injection of sesame oil, and then the animals were rested until tissue collection, as described in section 9. In this study, artificial decidualization was performed in C57BL6/J mice, a commonly used mouse strain. At the time of tissue collection, the body weight was recorded, and the uterus was dissected and well-trimmed before being weighed (Figure 1B). The extent of the decidual response was recorded by expressing the uterine weight as a ratio of the body weight. In this study, 80% of the C57BL6/J mice decidualized (0.01012 ± 0.001515, n = 15), while 20% of the animal uteri did not decidualize (0.002108 ± 0.0001764, n = 3) (Figure 1C).
Figure 1: Schematic and representative results. (A) Schematic timeline for experimentally inducing artificial decidualization in a mouse model. Abbreviations: OVX = ovariectomy; E2 = estradiol (100 ng days 1-3, 5 ng days 7-9); P4 = progesterone. Note: A P4 pellet was used to generate the results presented here. Alternative methods for progesterone delivery include daily subcutaneous injections and mini-osmotic pumps. (B) Representative images of non-decidualised (ND) and decidualised (D) uteri from young adult C57BL6/J mice. Scale bars = 5 mm.(C) Comparison of the uterine weight to body weight (UW:BW) ratio in non-decidualized and decidualized animals. Data are mean ± SEM; Mann-Whitney test, **p = 0.003; ND: n = 3, D: n = 15. Please click here to view a larger version of this figure.
Hormone delivery method | Strengths | Weaknesses |
Subcutaneous injections | No surgical intervention required | Repeated daily handling |
Accessible technique that does not require surgical training (as compared to pellet implantation) | Hormones in oil can leak out of the injection site, therefore amount absorbed by each animal can vary | |
Slow release pellets | No need for daily handling | Surgical procedure required |
Can be made in house | Not commercially available | |
Affordable alternative to osmotic mini pumps | ||
Small and very well tolerated by animals | ||
Osmotic mini pumps | No need for daily handling | Surgical procedure required |
Commercially available | Expensive | |
Most accurate delivery method | Much larger than slow release pellets |
Table 1: Strengths and weaknesses of the hormone delivery methods.
This article provides step-by-step instructions on how to perform OVX and provide exogenous hormones for studies focused on understanding the uterine contributions to pregnancy and fertility. Two detailed protocols are provided on two experimental applications of these methods, including performing embryo transfer and inducing decidualization artificially.
Whilst performing OVX can be challenging initially – especially for researchers new to rodent models – it is a relatively simple procedure once appropriately trained and practiced. The key steps in the procedures include closely monitoring the animals while they are under anesthesia and ensuring there is no ovarian tissue left behind. In some models, the oviduct may be left intact. However, it should be noted that the oviduct is a hormone-responsive tissue with abundant estrogen and progesterone receptors19. The surgical protocol for removing the ovary and oviduct is much simpler compared to removing just the ovary, as the former can be completed with the naked eye. To remove just the ovary and leave the oviduct in place in the latter case, a dissecting microscope is required, as this is a much more intricate procedure. Consequently, the operating time may be extended, as the animal needs to be moved between the dissecting microscope stage and the operating field for different parts of the procedure, such as suturing the internal body wall.
The analgesic protocols detailed here are standard and approved by the Monash University Animal Ethics committee, so they may vary depending on the individual institution's ethics committee requirements or preferences. It should be noted that no analgesia was provided for the artificial decidualization procedure, as typical non-steroidal anti-inflammatories interfere with the decidualization process. If researchers wish to provide analgesics at the time of artificial decidualization, this should be taken into consideration.
This work presents three methods of hormone delivery to supplement ovarian hormones following OVX, and each method has its own strengths and weaknesses (Table 1). Subcutaneous injections of hormones in oil are common in the literature14,16,17. This technique has many strengths, including the fact that no surgical procedure is required, and, thus, no formal training in rodent surgery or gas anesthesia is required. This makes subcutaneous injection an accessible option for almost all research groups. Injections are also affordable and easy to carry out. Practically, however, they have some limitations, particularly in models of pregnancy. To maintain pregnancy in an OVX animal, hormone supplementation with progesterone must be given daily to support the pregnancy. It may be possible to stop the daily injections once the placenta is sufficiently developed to take over as the main source of progesterone, though this has not been trialed in the protocols presented here. Anecdotally, it is possible for hormones in oil to leak from the injection site following subcutaneous injection. In part, this may be due to the needle size required (26 G) to easily dispense something as viscous as sesame oil. Therefore, this leakage needs to be monitored and recorded when performing injections in oil in order to correlate this with the experimental outcomes.
Slow-release pellets are preferable to subcutaneous injections, as they are cost-effective and simple to make in-house. However, they require multiple overnight steps, which should be considered when planning experimental timelines. These pellets secrete approximately 500 µg daily (as assessed during a time course incubation in cell culture medium and subsequent progesterone ELISA). It should be noted this is a lower concentration compared to the daily subcutaneous injections described above, and this is due to the consistency in the delivery of progesterone from the pellet. As aforementioned, oil injections can leak out of the injection site, thus reducing the overall concentration delivered. In previous studies, these pellets have only been active in vivo for up to 10 days after they are surgically inserted. Therefore, in pregnancy studies, it remains unclear if it may be necessary to insert a second pellet at mid-gestation or whether the placenta could sufficiently provide endocrine support for the pregnancy by that stage. These pellets are, therefore, optimal for short-term models of pregnancy, including the artificial decidualization protocol presented here, as well as pregnancy studies up to 10 days post-embryo transfer. While slow-release pellets negate the need for daily animal handling and injections, some low-dose estrogen injections are still required to balance the progesterone receptor feedback loop. This strategy has been used previously20,21.
Lastly, osmotic mini pumps are the most accurate hormone delivery method and are commercially available, but they are the most expensive option. Osmotic mini pumps can deliver a set concentration of hormone daily for up to 28 days, depending on the model selected. Similar to the slow-release pellets, while osmotic mini pumps avoid the need for daily animal handling, some low-dose estrogen injections are still required.
The artificial decidualization protocol described here allows the study of an early pregnancy milestone independent of ovarian and embryonic influence. While humans undergo decidualization with every menstrual cycle, rodents only decidualize during pregnancy establishment. Thus, this model has immense value for studying human-like pregnancy milestones in a manipulable rodent model. The procedure detailed herein is relatively non-invasive, as it uses a non-surgical embryo transfer device (NSET) to deliver sesame oil directly to the uterine horn via the vagina and cervix. Though this procedure is less invasive than other methodologies, it can become quite expensive when using commercial NSETs. In comparison, other published models of artificial decidualization require a surgical procedure to perform intrauterine injections of oil17. This requires a surgical setup similar to that described in section 1 and steps 2.1-2.11. However, in animals that have been ovariectomized previously, it can be more challenging to identify and expose the uterine horn. There can also be adhesions formed from the previous surgical procedure for ovariectomy. Thus, while it may be more cost-efficient to perform surgical intrauterine injections to induce decidualization, the surgical and anesthesia time is substantially longer than the alternative of using NSETs. There are established protocols for in-house fabricated alternatives22 to commercially available NSETs, which are much more cost-effective.
While the embryo transfer procedure is described here, we have previously published this model and its success rates across different strains of mice14. Moreover, while the embryo transfer method described here uses a surgical approach, NSETs could also be integrated into this procedure.
Future directions in this area should include studies focused on the specific uterine contributions to the establishment and maintenance of pregnancy. This knowledge is critical for furthering our understanding of idiopathic infertility, implantation failure, and pregnancy complications. Expanding our knowledge in these areas is also fundamental for improving clinical outcomes for IVF/ICSI patients, as well as our comprehension of pregnancy as a biological process.
In conclusion, OVX is a simple procedure that can be integrated into animal models to study the uterine contributions to pregnancy and fertility. Models in the future will benefit from integrating OVX and exogenous hormone delivery so that comparisons can be made between ovarian-specific and uterine-specific contributions to fertility and pregnancy.
The authors have nothing to disclose.
This work was made possible through the Victorian State Government Operational Infrastructure Support and Australian Government National Health and Medical Research Council (NHMRC) IRIISS. This work was supported by the Monash University Faculty of Medicine, Nursing and Health Science Platform Access Grant to A.L.W. (Winship-PAG18-0343) to access the Monash Reproductive Services Platform. A.L.W. is supported by DECRA funding DE21010037 from the Australian Research Council (ARC). J.N.H. and L.R.A. are supported by an Australian Government Research Training Program Scholarship. L.R.A. is supported by a Monash Graduate Excellence Scholarship. K.J.H. is supported by an ARC Future Fellowship FT190100265.
ALZET 1002 mini osmotic pumps | BioScientifica | 1002 | Delivers 0.25 µL/h for 14 days. Use for section 7 (Experimental procedure – Embryo transfer). |
ALZET 1003D mini osmotic pumps | BioScientifica | 1003D | Delivers 1 µL/h for 14 days. Use for section 8 (Experimental procedure – Artificial decidualization). |
ALZET Reflex 7 mm clips | BioScientifica | 0009971 | Either Reflex clips or Michel clips can be used for wound closure, depending on preference |
ALZET Reflex clip applicator | BioScientifica | 0009974 | Either Reflex clips or Michel clips can be used for wound closure, depending on preference |
ALZET Reflex clip remover | BioScientifica | 0009976 | Either Reflex clips or Michel clips can be used for wound closure, depending on preference |
Bupivicaine injection | Pfizer | NA | Stock 0.5%. Use at 0.05% in saline |
Estradiol | Sigma | E8875 | |
Meloxicam | Ilium | NA | Active constituent 0.5 mg/mL. Use 3.5 mL per 200 mL cage bottle, or as your institutions vet prescribes. |
Michel clips | Daniels | NS-000242 | |
Multi purpose sealant | Dow Corning | 732 | |
Non-surgical embryo transfer (NSET) device | ParaTechs | 60010 | Contains 6 mm speculum. Single use only. |
Progesterone | Sigma | P0130 | Soluble in ethanol. Use for section 3 (Hormone preparation – subcutaneous injection) and section 4 (Hormone preparation – slow-release pellets) |
Progesterone | Sigma | P7556 | Soluble in water. Use for section 5 (Hormone preparation – osmotic mini pumps) |
Refresh eye ointment | Allergan | NA | 42.5% w/v liquid paraffin, 57.3% w/v soft white paraffin |
Rimadyl Carprofen | Zoetis | NA | Stock 50 mg/mL. Use at 5 mg/kg |
Rubber tubing | Dow Corning | 508-008 | Washed in 100% ethanol and cut into 1 cm pieces. Inside diameter 1.57 mm ± 0.23 mm; outside diamater 3.18 mm ± 0.23 mm; wall 0.81 mm. |
Sesame oil | Sigma | S3547 | |
Sofsilk Silk sutures size 3-0 | Covidien | GS-832 |