Competence Classification of Cumulus and Granulosa Cell Transcriptome in Embryos Matched by Morphology and Female Age

Harnessing Artificial Intelligence to Predict Ovarian Stimulation Outcomes in In Vitro Fertilization: Scoping Review

BACKGROUND

In the realm of in vitro fertilization (IVF), artificial intelligence (AI) models serve as invaluable tools for clinicians, offering predictive insights into ovarian stimulation outcomes. Predicting and understanding a patient's response to ovarian stimulation can help in personalizing doses of drugs, preventing adverse outcomes (eg, hyperstimulation), and improving the likelihood of successful fertilization and pregnancy. Given the pivotal role of accurate predictions in IVF procedures, it becomes important to investigate the landscape of AI models that are being used to predict the outcomes of ovarian stimulation.

OBJECTIVE

The objective of this review is to comprehensively examine the literature to explore the characteristics of AI models used for predicting ovarian stimulation outcomes in the context of IVF.

METHODS

A total of 6 electronic databases were searched for peer-reviewed literature published before August 2023, using the concepts of IVF and AI, along with their related terms. Records were independently screened by 2 reviewers against the eligibility criteria. The extracted data were then consolidated and presented through narrative synthesis.

RESULTS

Upon reviewing 1348 articles, 30 met the predetermined inclusion criteria. The literature primarily focused on the number of oocytes retrieved as the main predicted outcome. Microscopy images stood out as the primary ground truth reference. The reviewed studies also highlighted that the most frequently adopted stimulation protocol was the gonadotropin-releasing hormone (GnRH) antagonist. In terms of using trigger medication, human chorionic gonadotropin (hCG) was the most commonly selected option. Among the machine learning techniques, the favored choice was the support vector machine. As for the validation of AI algorithms, the hold-out cross-validation method was the most prevalent. The area under the curve was highlighted as the primary evaluation metric. The literature exhibited a wide variation in the number of features used for AI algorithm development, ranging from 2 to 28,054 features. Data were mostly sourced from patient demographics, followed by laboratory data, specifically hormonal levels. Notably, the vast majority of studies were restricted to a single infertility clinic and exclusively relied on nonpublic data sets.

CONCLUSIONS

These insights highlight an urgent need to diversify data sources and explore varied AI techniques for improved prediction accuracy and generalizability of AI models for the prediction of ovarian stimulation outcomes. Future research should prioritize multiclinic collaborations and consider leveraging public data sets, aiming for more precise AI-driven predictions that ultimately boost patient care and IVF success rates.

Transcriptomic signature of luteinized cumulus cells of oocytes developing to live birth after women received intracytoplasmic sperm injection

OBJECTIVE

To compare the transcriptome of human cumulus cells (CCs) from oocytes with different outcomes (pregnancy yes/no, live birth [LB] yes/no), to identify noninvasive biomarkers for oocyte selection as well as new therapeutic targets to increase LB rates from assisted reproductive technologies (ART).

DESIGN

Retrospective observational study.

SETTINGS

This study was conducted at a University Hospital in Switzerland.

PATIENTS

Subfertile couples undergoing controlled ovarian superstimulation and intracytoplasmic sperm injection with subsequent unbiopsied embryo transfer below the female age of 43 years.

INTERVENTION(S)

RNA sequencing of CCs from oocytes results in a pregnancy, no pregnancy, LB, or no LB.

MAIN OUTCOME MEASURES

Differential gene expression (DEG) between CCs of oocytes results in "no pregnancy" vs. "pregnancy" and "pregnancy only" vs. "live birth."

RESULTS

Although RNA sequencing did not reveal DEGs when comparing the transcriptomic profiles of the groups "no pregnancy" with "pregnancy," we identified 139 DEGs by comparing "pregnancy only" with "live birth," of which 28 belonged to clusters relevant to successful ART outcomes (i.e., CTGF, SERPINE2, PCK1, HHIP, HS3ST, and BIRC5). A functional enrichment analysis revealed that the transcriptome of CCs associated with LB depicts pathways of extracellular matrix, inflammatory cascades leading to ovulation, cell patterning, proliferation, and differentiation, and silencing pathways leading to apoptosis.

CONCLUSION

We identified a CCs transcriptomic profile associated with LB after embryo transfer that, after further validation, could serve to predict successful ART outcomes. The definition of relevant pathways of CCs related to oocyte competency contributes to a broader understanding of the cumulus oocyte complex and helps identify further therapeutic targets for improving ART success.

The expression levels of NOS2, HMOX1, and VEGFC in cumulus cells are markers of oocyte maturation and fertilization rate

Throughout the reproductive life of women, cumulus cells (CC) protect the dormant oocyte from damage, act as sensors of the follicular microenvironment, and act as a gatekeeper for oocyte developmental potential. One such mechanism relies on the hypoxia-tolerance response, which, with age, decreases systematically, including in the ovary. We aimed to evaluate the association between gene expression related to hypoxia and aging in CC and reproductive results in in vitro fertilization cycles. We recruited 94 women undergoing controlled ovarian stimulation. Total RNA was extracted from pooled CCs collected after oocyte pick-up (OPU) and reverse-transcribed to complementary DNA using random hexamers to test 14 genes related to hypoxia response via HIF1α activation, oxidative stress, and angiogenic responses. The expression of CLU, NOS2, and TXNIP had a positive correlation with age (rs  = 0.25, rs  = 0.24, and rs  = 0.35, respectively). Additionally, NOS2 and HMOX1 expression correlated positively with the retrieval of immature oocytes (rs  = 0.22 and rs  = 0.40, respectively). Moreover, VEGFC levels decreased overall with increasing fertilization rate, independently of age (rs  = -0.29). We found that the fertilization potential of a cohort of oocytes is related to the ability of CC to respond to oxidative stress and hypoxia with age, pointing at NOS2, HMOX1, and VEGFC expression as markers for oocyte maturation and fertilization success.

mtDNA content in cumulus cells does not predict development to blastocyst or implantation

STUDY QUESTION

Is relative mitochondrial DNA (mtDNA) content in cumulus cells (CCs) related to embryo developmental competence in humans and/or the bovine model?

SUMMARY ANSWER

mtDNA content in CCs provides a poor predictive value of oocyte developmental potential, both in vitro and following embryo transfer.

WHAT IS KNOWN ALREADY

CCs are closely connected to the oocyte through transzonal projections, serving essential metabolic functions during folliculogenesis. These oocyte-supporting cells are removed and discarded prior to ICSI, thereby providing interesting biological material on which to perform molecular analyses designed to identify markers that predict oocyte developmental competence. Previous studies have positively associated oocyte mtDNA content with developmental potential in animal models and women. However, it remains debatable whether mtDNA content in CCs could be used as a proxy to infer oocyte developmental potential.

STUDY DESIGN SIZE DURATION

mtDNA content was analyzed in CCs obtained from 109 human oocytes unable to develop to blastocyst, able to develop to blastocyst but failing to establish pregnancy or able to develop to blastocyst and to establish pregnancy. mtDNA analysis was also performed on bovine cumulus samples collected from 120 oocytes unable to cleave, oocytes developing into cleaved embryos but arresting development prior to the blastocyst stage or oocytes developing to blastocysts.

PARTICIPANTS/MATERIALS SETTING METHODS

Human CCs samples were obtained from women undergoing IVF. Only unfrozen oocytes and embryos not submitted to preimplantation genetic testing were included in the analysis. Bovine samples were obtained from slaughtered cattle and individually matured, fertilized and cultured in vitro. Relative mtDNA was assessed by quantitative PCR analysis.

MAIN RESULTS AND THE ROLE OF CHANCE

mtDNA content in human and bovine CCs did not differ according to the developmental potential of their enclosed oocyte. Moreover, mtDNA content in bovine oocytes did not correlate with that of their corresponding CCs.

LARGE SCALE DATA

N/A.

LIMITATIONS REASONS FOR CAUTION

The lack of correlation found between mtDNA content in human CCs and oocytes was also assessed in bovine samples. Although bovine folliculogenesis, mono-ovulatory ovulation and early embryo development exhibit considerable similarities with that of humans, they may not be fully comparable.

WIDER IMPLICATIONS OF THE FINDINGS

The use of molecular markers for oocyte developmental potential in CCs could be used to enhance success rates following single embryo transfer. However, our data indicate that mtDNA in CCs is not a good proxy for oocyte quality.

STUDY FUNDING/COMPETING INTERESTS

This research was supported by the Industrial Doctorate Project IND2017/BIO-7748 funded by the Madrid Region Government. The authors declare no competing interests.

Cumulus cells of euploid versus whole chromosome 21 aneuploid embryos reveal differentially expressed genes

RESEARCH QUESTION

Can cumulus cells be used as a non-invasive target for the study of determinants of preimplantation embryo quality?

DESIGN

Cumulus cells were collected from monosomy 21, trisomy 21 and euploid embryos and subjected to RNA sequencing analysis and real-time polymerase chain reaction assays. The differential gene expression was analysed for different comparisons.

RESULTS

A total of 3122 genes in monosomy 21 cumulus cells and 19 genes in trisomy 21 cumulus cells were differentially expressed compared with euploid cumulus cells. Thirteen of these genes were differentially expressed in both monosomy and trisomy 21, compared with euploid, including disheveled segment polarity protein 2 (DVL2), cellular communication network factor 1 (CCN1/CYR61) and serum response factor (SRF), which have been previously implicated in embryo developmental competence. In addition, ingenuity pathway analysis revealed cell-cell contact function to be affected in both monosomy and trisomy 21 cumulus cells.

CONCLUSIONS

These findings support the use of cumulus cell gene expression analysis for the development of biomarkers evaluating oocyte quality for patients undergoing fertility preservation of oocytes.

Stratifying Cumulus Cell Samples Based on Molecular Profiling to Help Resolve Biomarker Discrepancies and to Predict Oocyte Developmental Competence

To increase the efficiency of assisted reproductive techniques (ART), molecular studies have been performed to identify the best predictive biomarkers for selecting the most suitable germ cells for fertilization and the best embryo for intra-uterine transfer. However, across different studies, no universal markers have been found. In this study, we addressed this issue by generating gene expression and CpG methylation profiles of outer cumulus cells obtained during intra-cytoplasmic sperm injection (ICSI). We also studied the association of the generated genomic data with the clinical parameters (spindle presence, zona pellucida birefringence, pronuclear pattern, estrogen level, endometrium size and lead follicle size) and the pregnancy result. Our data highlighted the presence of several parameters that affect analysis, such as inter-individual differences, inter-treatment differences, and, above all, specific treatment protocol differences. When comparing the pregnancy outcome following the long protocol (GnRH agonist) of ovarian stimulation, we identified the single gene markers (NME6 and ASAP1, FDR < 5%) which were also correlated with endometrium size, upstream regulators (e.g., EIF2AK3, FSH, ATF4, MKNK1, and TP53) and several bio-functions related to cell death (apoptosis) and cellular growth and proliferation. In conclusion, our study highlighted the need to stratify samples that are very heterogeneous and to use pathway analysis as a more reliable and universal method for identifying markers that can predict oocyte development potential.

The mitochondrial DNA copy number of cumulus granulosa cells may be related to the maturity of oocyte cytoplasm

STUDY QUESTION

Is the mitochondrial DNA (mtDNA) copy number of cumulus granulosa cells (CGCs) related to the maturation of oocyte cytoplasm?

SUMMARY ANSWER

Compared with the mtDNA copy number of CGCs from germinal vesicles (GV), CGCs from Metaphase I (MI) oocytes appear to have a lower mtDNA copy number.

WHAT IS KNOWN ALREADY

The growth and development of CGCs and oocyte are synchronised. The interaction between CGCs and the oocyte provides the appropriate balance of energy, which is necessary for mammalian oocyte development. Moreover, in the oocyte-cumulus complex (OCC), mature oocytes with higher mtDNA copy numbers tend to have corresponding CGCs with higher mtDNA copy numbers.

STUDY DESIGN, SIZE, DURATION

This is a prospective study of 302 OCCs obtained from 70 women undergoing in vitro fertilisation with intracytoplasmic sperm injection (ICSI) at the Reproductive and Genetic Hospital of CITIC-Xiangya, between 24 February 2018 and 21 December 2019. The CGCs were divided into three groups (GV, MI and MII stages) based on the maturation status of their corresponding oocyte. The sample sizes (n = 302) of CGCs in the three stages were 63 (CGCGV), 70 (CGCMI) and 169 (CGCMII), respectively. Some of the samples (n = 257) was used to quantify the mtDNA copy number, while the rest (n = 45) were used to analyse the expression level of mitochondrial genes. Furthermore, we retrieved 82 immature oocytes from among the 257 OCCs used for mtDNA copy numbers, including 36 GV oocytes and 46 MI oocytes, for analysis of oocyte mtDNA.

PARTICIPANTS/MATERIALS, SETTING, METHODS

We selected genes with high consistency of real-time PCR results to accurately measure the mtDNA copy number by testing the efficacy and the reproducibility in whole genome amplification (WGA) samples from a human embryonic stem cell line. The CGCs of each oocyte were individually isolated. The mtDNA copy number and gene expression of the CGCs were assessed using real-time PCR techniques. Mitochondrial DNA copy number of the corresponding immature oocytes was also evaluated.

MAIN RESULTS AND THE ROLE OF CHANCE

MT-ND1, MT-CO1 and β-globin genes were chosen for the assessment of mtDNA content, and mRNA expressions of MT-ND1, MT-CO1, PGC-1α and TFAM were also measured. The genome of 257 CGCs and 82 immature oocytes were amplified according to the multiple displacement amplification (MDA) protocol, and RNA was extracted from 45 CGCs. Compared with CGCGV, CGCMI had a significantly lower mtDNA copy number. In the MT-ND1 assay, the CGCGV: CGCMI was [270 ± 302]: [134 ± 201], P = 0.015. In the MT-CO1 assay, CGCGV: CGCMI was [205 ± 228]: [92 ± 112], P = 0.026. There was no statistical difference in mtDNA between CGCGV and CGCMII. In the MT-ND1 assay, CGCGV: CGCMII was [270 ± 302]: [175 ± 223], P = 0.074. In the MT-CO1 assay, CGCGV: CGCMII was [205 ± 228]: [119 ± 192], P = 0.077. No statistical difference of mtDNA copy number was observed between CGCMI and CGCMII. In the MT-ND1 assay, CGCMI: CGCMII was [134 ± 201]: [175 ± 223], P = 0.422. In the MT-CO1 assay, CGCMI: CGCMII was [92 ± 112]: [119 ± 192], P = 0.478. To verify the reliability of the above results, we further analysed the mtDNA copy number of CGCs of 14 patients with GV, MI and MII oocytes, and the results showed that the mtDNA copy number of CGCMI may be lower. The mtDNA copy number of CGCGV and CGCMI was statistically different in the MT-ND1 assay where CGCGV: CGCMI was [249 ± 173]: [118 ± 113], P = 0.016, but in the MT-CO1 assay, CGCGV: CGCMI was [208 ± 199]: [83 ± 98], P = 0.109. There was no significant difference in mtDNA between CGCGV and CGCMII. In the MT-ND1 assay, CGCGV: CGCMII was [249 ± 173]: [185 ± 200], P = 0.096. In the MT-CO1 assay, CGCGV: CGCMII was [208 ± 199]: [114 ± 139], P = 0.096. There was also no significant difference in mtDNA between CGCMI and CGCMII. In the MT-ND1 assay, CGCMI: CGCMII was [118 ± 113]: [185 ± 200], P = 0.198. In the MT-CO1 assay, CGCMI: CGCMII was [83 ± 98]: [114 ± 139], P = 0.470. Moreover, there were no statistical differences in the expression levels of MT-ND1, MT-CO1, PGC-1α and TFAM between CGCGV, CGCMI and CGCMII (P > 0.05).

LARGE SCALE DATA

N/A.

LIMITATIONS, REASONS FOR CAUTION

Due to the ethical issues, the study did not quantify the mtDNA content of MII oocytes. Thus, whether the change in mtDNA copy number in CGCs is related to the different developmental stages of oocytes has not been further confirmed. Moreover, the sample size was relatively small.

WIDER IMPLICATIONS OF THE FINDINGS

The mtDNA copy number of CGCs decreases from the GV phase to the MI phase and stays steady from the MI to MII stage. At different stages of oocyte maturation, the mtDNA of CGCs may undergo self-degradation and replication to meet the energy requirements of the corresponding oocyte and the maturation of the oocyte cytoplasm.

STUDY FUNDING/COMPETING INTEREST(S)

Funding was provided by the National Key R&D Program of China (Grant 2018YFC1003100, to L.H.), the science and technology major project of the Ministry of Science and Technology of Hunan Province, China (grant 2017SK1030, to G.L.), the National Natural Science Foundation of China (grant 81873478, to L.H.), and Merck Serono China Research Fund for Fertility Experts (to L.H.). There is no conflict of interest.

Polybrominated Diphenyl Ethers in Human Follicular Fluid Dysregulate Mural and Cumulus Granulosa Cell Gene Expression

Polybrominated diphenyl ethers (PBDEs), a major class of flame retardants incorporated into numerous consumer products, leach out into dust resulting in widespread exposure. There is evidence from in vitro and in vivo animal studies that PBDEs affect ovarian granulosa cell function and follicular development, yet human studies of their association with female infertility are inconclusive. Here, we tested the hypothesis that exposure to the PBDEs in follicular fluid is associated with dysregulation of gene expression in the mural and cumulus granulosa cells collected from women undergoing in vitro fertilization by intracytoplasmic sperm injection. The median concentration of the ∑ 10PBDEs detected in the follicular fluid samples (n = 37) was 15.04 pg/g wet weight. RNA microarray analyses revealed that many genes were differentially expressed in mural and cumulus granulosa cells. Highest vs lowest quartile exposure to the Σ 10PBDEs or to 2 predominant PBDE congeners, BDE-47 or BDE-153, was associated with significant effects on gene expression in both cell types. Mural granulosa cells were generally more sensitive to PBDE exposure compared to cumulus cells. Overall, gene expression changes associated with BDE-47 exposure were similar to those for ∑ 10PBDEs but distinct from those associated with BDE-153 exposure. Interestingly, exposure to BDE-47 and ∑ 10PBDEs activated the expression of genes in pathways that are important in innate immunity and inflammation. To the best of our knowledge, this is the first demonstration that exposure to these environmental chemicals is associated with the dysregulation of pathways that play an essential role in ovulation.

Improved clinical outcomes after non-invasive oocyte selection and Day 3 eSET in ICSI patients

BACKGROUND

Non-invasive oocyte quality scoring, based on cumulus gene expression analysis, in combination with morphology scoring, can increase the clinical pregnancy (CPR) and live birth rates (LBR) in Day 3 eSET (elective single embryo transfer) ICSI patients. This was first investigated in a pilot study and is now confirmed in a large patient cohort of 633 patients. It was investigated whether CPR, LBR and time-to-pregnancy could be improved by analyzing the gene expression profile of three predictive genes in the cumulus cells, compared to patients with morphology-based embryo selection only.

METHODS

A large interventional, non-randomized, assessor-blinded cohort study with 633 ICSI patients was conducted in a tertiary fertility center. Non-PCOS patients, 22-39 years old, with good ovarian reserve, were stimulated with HP-hMG using a GnRH antagonist protocol and planned for fresh Day 3 eSET. The cumulus cells from individually denuded oocytes were ranked by a lab-developed cumulus cell test: qRT-PCR for three predictive genes (CAMK1D, EFNB2 and SASH1) and two control genes (UBC, B2M). The embryo selected for transfer was highest ranked from the pool of morphologically transferable Day 3 embryos. Patients in the control (n = 520) and experimental arm (n = 113) were compared for clinical pregnancy and live birth, using a weighted generalized linear model, and time-to-pregnancy using Kaplan-Meier curves.

RESULTS

The CPR was 61% in the experimental arm (n = 113) vs 29% in the control arm (n = 520, p < 0.0001). The LBR in the experimental arm (50%) was significantly higher than in the control arm (27%,p < 0.0001). Time-to-pregnancy was significantly shortened by 3 transfer cycles independent of the number of embryos available on Day 3 (Kaplan-Meier, p < 0.0001). Cumulus cell tested patients < 35 years (n = 65) or ≥ 35 years (n = 48) had a CPR of 62 and 60% respectively (ns). For cumulus cell tested patients with 2, 3-4, or > 4 transferable embryos, the CPR was 66, 52, and 67% (ns) respectively, and thus independent of the number of transferable embryos on Day 3.

CONCLUSIONS

This study provides further evidence of the clinical usefulness of the non-invasive cumulus cell test over time in a larger patient cohort.

TRIAL REGISTRATION

Clinicaltrials.gov, NCT03659786 / NCT02962466 (Registered 6Sep2018/11Nov2016, retrospectively registered.

Comparison of Multivariable Logistic Regression and Other Machine Learning Algorithms for Prognostic Prediction Studies in Pregnancy Care: Systematic Review and Meta-Analysis

Background

Predictions in pregnancy care are complex because of interactions among multiple factors. Hence, pregnancy outcomes are not easily predicted by a single predictor using only one algorithm or modeling method.

Objective

This study aims to review and compare the predictive performances between logistic regression (LR) and other machine learning algorithms for developing or validating a multivariable prognostic prediction model for pregnancy care to inform clinicians’ decision making.

Methods

Research articles from MEDLINE, Scopus, Web of Science, and Google Scholar were reviewed following several guidelines for a prognostic prediction study, including a risk of bias (ROB) assessment. We report the results based on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Studies were primarily framed as PICOTS (population, index, comparator, outcomes, timing, and setting): Population: men or women in procreative management, pregnant women, and fetuses or newborns; Index: multivariable prognostic prediction models using non-LR algorithms for risk classification to inform clinicians’ decision making; Comparator: the models applying an LR; Outcomes: pregnancy-related outcomes of procreation or pregnancy outcomes for pregnant women and fetuses or newborns; Timing: pre-, inter-, and peripregnancy periods (predictors), at the pregnancy, delivery, and either puerperal or neonatal period (outcome), and either short- or long-term prognoses (time interval); and Setting: primary care or hospital. The results were synthesized by reporting study characteristics and ROBs and by random effects modeling of the difference of the logit area under the receiver operating characteristic curve of each non-LR model compared with the LR model for the same pregnancy outcomes. We also reported between-study heterogeneity by using τ² and I².

Results

Of the 2093 records, we included 142 studies for the systematic review and 62 studies for a meta-analysis. Most prediction models used LR (92/142, 64.8%) and artificial neural networks (20/142, 14.1%) among non-LR algorithms. Only 16.9% (24/142) of studies had a low ROB. A total of 2 non-LR algorithms from low ROB studies significantly outperformed LR. The first algorithm was a random forest for preterm delivery (logit AUROC 2.51, 95% CI 1.49-3.53; I²=86%; τ²=0.77) and pre-eclampsia (logit AUROC 1.2, 95% CI 0.72-1.67; I²=75%; τ²=0.09). The second algorithm was gradient boosting for cesarean section (logit AUROC 2.26, 95% CI 1.39-3.13; I²=75%; τ²=0.43) and gestational diabetes (logit AUROC 1.03, 95% CI 0.69-1.37; I²=83%; τ²=0.07).

Conclusions

Prediction models with the best performances across studies were not necessarily those that used LR but also used random forest and gradient boosting that also performed well. We recommend a reanalysis of existing LR models for several pregnancy outcomes by comparing them with those algorithms that apply standard guidelines.

Trial Registration

PROSPERO (International Prospective Register of Systematic Reviews) CRD42019136106; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=136106

Brilliant cresyl blue enhanced optoacoustic imaging enables non-destructive imaging of mammalian ovarian follicles for artificial reproduction

In the field of reproductive biology, there is a strong need for a suitable tool capable of non-destructive evaluation of oocyte viability and function. We studied the application of brilliant cresyl blue (BCB) as an intra-vital exogenous contrast agent using multispectral optoacoustic tomography (MSOT) for visualization of porcine ovarian follicles. The technique provided excellent molecular sensitivity, enabling the selection of competent oocytes without disrupting the follicles. We further conducted in vitro embryo culture, molecular analysis (real-time and reverse transcriptase polymerase chain reaction) and DNA fragmentation analysis to comprehensively establish the safety of BCB-enhanced MSOT imaging in monitoring oocyte viability. Overall, the experimental results suggest that the method offers a significant advance in the use of contrast agents and molecular imaging for reproductive studies. Our technique improves the accurate prediction of ovarian reserve significantly and, once standardized for in vivo imaging, could provide an effective tool for clinical infertility management.

Metabolic co-dependence of the oocyte and cumulus cells: essential role in determining oocyte developmental competence

BACKGROUND

Within the antral follicle, the oocyte is reliant on metabolic support from its surrounding somatic cells. Metabolism plays a critical role in oocyte developmental competence (oocyte quality). In the last decade, there has been significant progress in understanding the metabolism of the cumulus-oocyte complex (COC) during its final stages of growth and maturation in the follicle. Certain metabolic conditions (e.g. obesity) or ART (e.g. IVM) perturb COC metabolism, providing insights into metabolic regulation of oocyte quality.

OBJECTIVE AND RATIONALE

This review provides an update on the progress made in our understanding of COC metabolism, and the metabolic conditions that influence both meiotic and developmental competence of the oocyte.

SEARCH METHODS

The PubMed database was used to search for peer-reviewed original and review articles. Searches were performed adopting the main terms 'oocyte metabolism', 'cumulus cell metabolism', 'oocyte maturation', 'oocyte mitochondria', 'oocyte metabolism', 'oocyte developmental competence' and 'oocyte IVM'.

OUTCOMES

Metabolism is a major determinant of oocyte quality. Glucose is an essential requirement for both meiotic and cytoplasmic maturation of the COC. Glucose is the driver of cumulus cell metabolism and is essential for energy production, extracellular matrix formation and supply of pyruvate to the oocyte for ATP production. Mitochondria are the primary source of ATP production within the oocyte. Recent advances in real-time live cell imaging reveal dynamic fluctuations in ATP demand throughout oocyte maturation. Cumulus cells have been shown to play a central role in maintaining adequate oocyte ATP levels by providing metabolic support through gap junctional communication. New insights have highlighted the importance of oocyte lipid metabolism for oocyte oxidative phosphorylation for ATP production, meiotic progression and developmental competence. Within the last decade, several new strategies for improving the developmental competence of oocytes undergoing IVM have emerged, including modulation of cyclic nucleotides, the addition of precursors for the antioxidant glutathione or endogenous maturation mediators such as epidermal growth factor-like peptides and growth differentiation factor 9/bone morphogenetic protein 15. These IVM additives positively alter COC metabolic endpoints commonly associated with oocyte competence. There remain significant challenges in the study of COC metabolism. Owing to the paucity in non-invasive or in situ techniques to assess metabolism, most work to date has used in vitro or ex vivo models. Additionally, the difficulty of measuring oocyte and cumulus cell metabolism separately while still in a complex has led to the frequent use of denuded oocytes, the results from which should be interpreted with caution since the oocyte and cumulus cell compartments are metabolically interdependent, and oocytes do not naturally exist in a naked state until after fertilization. There are emerging tools, including live fluorescence imaging and photonics probes, which may provide ways to measure the dynamic nature of metabolism in a single oocyte, potentially while in situ.

WIDER IMPLICATIONS

There is an association between oocyte metabolism and oocyte developmental competence. Advancing our understanding of basic cellular and biochemical mechanisms regulating oocyte metabolism may identify new avenues to augment oocyte quality and assess developmental potential in assisted reproduction.

Necessity is the mother of invention and the evolutionary force driving the success of in vitro fertilization

During the last few decades, millions of healthy children have been born with the aid of in vitro fertilization (IVF). This success belies the fact that IVF treatment is comprised of a complex series of interventions starting with a customized control ovarian stimulation protocol. This is followed by the induction of oocyte maturation, the retrieval of mature oocytes and in vitro fertilization, which often involves the microinjection of a single sperm into the oocyte. After fertilization, the resulting embryos are cultured for up to 7 days. The best embryos are transferred into the uterus where the embryo implants and hopefully develops into a healthy child. However, frequently the best embryos are biopsied and frozen. The biopsied cells are analyzed to identify those embryos without chromosomal abnormalities. These embryos are eventually thawed and transferred with pregnancy rates as good if not better than embryos that are not biopsied and transferred in a fresh cycle. Thus, IVF treatment requires the coordinated efforts of physicians, nurses, molecular biologists and embryologists to conduct each of these multifaceted phases in a seamless and flawless manner. Even though complex, IVF treatment may seem routine today, but it was not always the case. In this review the evolution of human IVF is presented as a series of innovations that resolved a technical hurdle in one component of IVF while creating challenges that eventually lead to the next major advancement. This step-by-step evolution in the treatment of human infertility is recounted in this review.

Does the molecular and metabolic profile of human granulosa cells correlate with oocyte fate? New insights by Fourier transform infrared microspectroscopy analysis

STUDY QUESTION

Does the molecular and metabolic profile of human mural granulosa cells (GCs) correlate with oocyte fate?

SUMMARY ANSWER

A close relation between the metabolic profile of mural GCs and the fate of the corresponding oocyte was revealed by the analysis of selected biomarkers defined by GC Fourier transform infrared microspectroscopy (FTIRM) analysis.

WHAT IS KNOWN ALREADY

In ART, oocyte selection is mainly based on the subjective observation of its morphological features; despite recent efforts, the success rate of this practice is still unsatisfactory. FTIRM is a well-established vibrational technique recently applied to evaluate oocytes quality in several experimental models, including human.

STUDY DESIGN, SIZE, DURATION

GCs retrieved from single-follicle aspirates were obtained with informed consent from 55 women undergoing controlled ovarian stimulation for IVF treatment. GCs were analysed by FTIRM to retrospectively correlate their spectral features with the fate of the companion oocytes. The study has been conducted between March 2016 and September 2017.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Patients were selected according to the following inclusion criteria: age <40 years; non-smokers; no ovarian infertility diagnosis (only tubal, idiopathic and male infertility); regular ovulatory menstrual cycles (25-30 days) with FSH < 10 IU/I on Day 3 of the menstrual cycle; sperm sample with a total motility count after treatment ≥300.000; number of retrieved oocytes ≥8. Based on the clinical outcome of the corresponding oocyte, GCs were retrospectively classified into the following experimental groups: clinical pregnancy (CP), fertilization failure (FF), embryo development failure (EDF) and implantation failure (IF). All samples were analysed by the FTIRM technique. The spectral biomarker signature of different oocyte fates was derived by several feature selection procedures ('Leave-one-out' method on factorial discriminant analysis (FDA), variable characterization method and logistic regression method with the multinomial Logit model). ANOVA, permutational multivariate ANOVA, FDA and canonical analysis of principal co-ordinates statistical tools were also applied to validate the identified spectral biomarkers.

MAIN RESULTS AND THE ROLE OF CHANCE

In total, 284 GCs samples were retrieved and retrospectively classified as FF: (N = 92), EDF (N = 113), IF (N = 56) and CP (N = 23). From the spectral profiles of GCs belonging to CP, FF, EDF and IF experimental groups, 17 spectral biomarkers, were identified by several feature selection procedures (P < 0.0001). These biomarkers were then validated by applying multivariate tools, to evaluate their ability to segregate GCs samples into the four experimental groups. FDA showed a clear separation along the F1-axis (62.75% of discrimination) between GCs from oocytes able (CP, IF groups) or not (FF, EDF groups) to develop into embryos; the F2-axis (24.14% of discrimination) segregated the embryos that gave pregnancy (CP) from those that failed implantation (IF). The confusion matrix (total percentage of correctness = 80.25%) obtained from this analysis pinpointed that GCs from oocytes unable to develop into embryos (FF, EDF) were better characterized than those from oocytes able to give viable embryos (CP, IF). ANOVA (P < 0.05) analysis pinpointed that: each experimental group showed specific macromolecular traits, ascribable to different biological and metabolic characteristics of GCs; these metabolic features were likely associated with different oocytes fates, but not to patient characteristics, since from the same patient we obtained GCs with different metabolic profiles.

LIMITATIONS, REASONS FOR CAUTION

The study is based on a small sample size but provides proof of concept that the GCs' metabolic profile is associated with the companion oocyte fate. The generated model should be further tested on a larger cohort of patients, classified in a similar manner, to assess the potential predictive value of this approach. Ultimately, validity of the proposed approach should be tested in a RCT.

WIDER IMPLICATIONS OF THE FINDINGS

For the first time, the FTIRM analysis of human GCs has demonstrated an approach to better understand the molecular crosstalk between follicular cells and oocytes and has identified potential spectral biomarkers for improving human IVF success rate.

STUDY FUNDING/COMPETING INTEREST(S)

The study was funded by GFI 2014 grant. The authors declare that there is no conflict of interest.

The Unique Mechanisms of Cellular Proliferation, Migration and Apoptosis are Regulated through Oocyte Maturational Development-A Complete Transcriptomic and Histochemical Study

The growth and development of oocyte affect the functional activities of the surrounding somatic cells. These cells are regulated by various types of hormones, proteins, metabolites, and regulatory molecules through gap communication, ultimately leading to the development and maturation of oocytes. The close association between somatic cells and oocytes, which together form the cumulus-oocyte complexes (COCs), and their bi-directional communication are crucial for the acquisition of developmental competences by the oocyte. In this study, oocytes were extracted from the ovaries obtained from crossbred landrace gilts and subjected to in vitro maturation. RNA isolated from those oocytes was used for the subsequent microarray analysis. The data obtained shows, for the first time, variable levels of gene expression (fold changes higher than |2| and adjusted p-value < 0.05) belonging to four ontological groups: regulation of cell proliferation (GO:0042127), regulation of cell migration (GO:0030334), and regulation of programmed cell death (GO:0043067) that can be used together as proliferation, migration or apoptosis markers. We have identified several genes of porcine oocytes (ID2, VEGFA, BTG2, ESR1, CCND2, EDNRA, ANGPTL4, TGFBR3, GJA1, LAMA2, KIT, TPM1, VCP, GRID2, MEF2C, RPS3A, PLD1, BTG3, CD47, MITF), whose expression after in vitro maturation (IVM) is downregulated with different degrees. Our results may be helpful in further elucidating the molecular basis and functional significance of a number of gene markers associated with the processes of migration, proliferation and angiogenesis occurring in COCs.

Association between early embryo morphokinetics plus cumulus cell gene expression and assisted reproduction outcomes in polycystic ovary syndrome women

RESEARCH QUESTION

Can a combination of time-lapse morphokinetic parameters and cumulus cell gene expression in polycystic ovary syndrome (PCOS) women be used to predict assisted reproductive treatment outcome?

DESIGN

A total of 547 embryos from 100 intracytoplasmic sperm injection (ICSI) cycles were evaluated. Fifty women with PCOS and 50 women who were categorized as tubal factor infertility were recruited. Time-lapse records were annotated for time to pronuclear fading (tPNf), time to 2 to 8 cells (t2-t8), reverse cleavage, direct cleavage and also for the presence of multinucleation. Expression levels of three genes involved in mitotic divisions, diaphanous-related formin 2 (DIAPH2), nibrin (NBN) and NIMA-related protein kinase (NEK4), were measured in 100 associated cumulus cell samples using quantitative real-time polymerase chain reaction.

RESULTS

Expression of DIAPH2 and NBN was significantly higher in the embryos of PCOS patients that resulted in implantation, biochemical and clinical pregnancies as well as live birth compared with embryos that were negative for these outcomes (P <0.01). However, in the tubal factor group, NBN gene expression was significantly higher in embryos resulting in biochemical pregnancy, clinical pregnancy and live birth (P <0.01) only. Multivariate logistic regression analysis showed that tPNf together with DIAPH2 gene expression were independent prognostic factors of clinical pregnancy rate and live birth in both groups.

CONCLUSIONS

Some time-lapse embryo parameters may be related to cumulus gene expression and clinical outcome. Furthermore, the expressions of cumulus cell genes involved in mitotic divisions are significantly associated with ICSI outcome using Day 3 embryo transfer.

Cumulus cells surrounding oocytes with high developmental competence exhibit down-regulation of phosphoinositol 1,3 kinase/protein kinase B (PI3K/AKT) signalling genes involved in proliferation and survival

STUDY QUESTION

Is the phosphoinositol 1,3-kinase/protein kinase B (PI3K/AKT) pathway expression profile in cumulus cells (CCs) a potential marker of oocyte competence and predictive of pregnancy outcome?

SUMMARY ANSWER

Eleven genes (AKT1, ARHGEF7, BCL2L1, CCND1, E2F1, HRAS, KCNH2, PIK3C2A, SHC1, SOS1 and SPP1) in the PI3K/AKT pathway were significantly down-regulated in CCs from oocytes that went on to produce a pregnancy compared to CCs associated with a negative outcome.

WHAT IS KNOWN ALREADY

The PI3K/AKT pathway plays a pivotal role in the interdependence and continuous feedback between the oocyte and CCs.

STUDY DESIGN SIZE, DURATION

The expression analysis of 92 transcripts in the PI3K/AKT pathway in CCs from patients with negative or positive pregnancy outcome, after single embryo transfer, was performed. Mouse CCs target gene expression was conducted to associate the expression profile of PI3K/AKT pathway to oocyte developmental profile.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Fifty-five good prognosis IVF patients who had been referred to IVF or intracytoplasmic sperm injection treatment for male-factor infertility or tubal disease were enroled. CCs from single cumulus-oocyte complexes (COCs) from 16 patients who underwent a single embryo transfer were analyzed. Twenty-five CD-1 mice were used to assess gene expression in CCs associated with oocytes with different competence in relation to hCG priming. A total 220 human COCs were collected. The RNA extracted from CCs of 16 selected patients was used to analyze PI3K/AKT pathway gene expression employing a 96-well custom TaqMan Array. Expression data of CCs associated to positive IVF outcome were compared to data from negative outcome samples. Mice were sacrificed after 9, 12, 15, 21 and 24 h post-hCG administration to obtain CCs from MII oocytes with different developmental competence. Akt1, Bcl2l2 and Shc1 expression were tested in the collected mouse CCs. In addition, the expression of upstream regulator ESR1, the gene encoding for the oestrogen receptor ERβ, and the downstream effectors of the pathway FOXO1, FOXO3 and FOXO4 was evaluated in human and mouse samples.

MAIN RESULTS AND THE ROLE OF CHANCE

Transcripts involved in the PI3K Signaling Pathway were selectively modulated according to the IVF/ICSI outcome of the oocyte. Eleven transcripts in this pathway were significantly down-regulated in all samples of CCs from oocytes with positive when compared those with a negative outcome. These outcomes were confirmed in mouse CCs associated with oocytes at different maturation stages. Expression data revealed that the down-regulation of ESR1 could be related to oocyte competence and is likely to be the driver of expression changes highlighted in the PI3K/AKT pathway.

LIMITATIONS REASONS FOR CAUTION

Small sample size and retrospective design.

WIDER IMPLICATIONS OF THE FINDINGS

The CCs expression profile of PI3K/AKT signaling genes, disclosed a specific CCs gene signature related to oocyte competence. It could be speculated that CCs associated with competent oocytes have completed their role in sustaining oocyte development and are influencing their fate in response to metabolic and hormonal changes by de-activating anti-apoptotic signals.

STUDY FUNDING/COMPETING INTEREST(S)

Supported by Merck Serono an affiliate of Merck KGaA, Darmstadt, Germany (research grant for the laboratory session; Merck KGaA reviewed the manuscript for medical accuracy only before journal submission. The authors are fully responsible for the content of this manuscript, and the views and opinions described in the publication reflect solely those of the authors). The authors declare no conflict of interest.

The activated DNA double-strand break repair pathway in cumulus cells from aging patients may be used as a convincing predictor of poor outcomes after in vitro fertilization-embryo transfer treatment

Women with advanced maternal age exhibit low anti-Müllerian hormone (AMH) levels and an altered follicular environment, which is associated with poor oocyte quality and embryonic developmental potential. However, the underlying mechanism is poorly understood. The present study aimed to assesswhether aging patients exhibit an activated DNA double-strandbreak (DSB) repair pathway in cumulus cells and thus, an association with poor outcomes after in vitro fertilization-embryo transfer (IVF-ET) treatment. Cumulus cells from young (≤29 y) and aging (≥37 y) human female patients were collected after oocyte retrieval. Our results indicated that aging patients showed a higher rate of γ-H2AX-positive cells than in young patients (24.33±4.55 vs.12.40±2.31, P<0.05). We also found that the mRNA expression levels of BRCA1, ATM, MRE11 and RAD51 were significantly elevated in aging cumulus cells. Accordingly, significantly increased protein levels of phospho-H2AX, BRCA1, ATM, MRE11 and RAD51 could be observed in aging cumulus cells. Moreover, aging cumulus cells showed a more frequent occurrence of early apoptosis than young cumulus cells. This study found that increases in DSBs and the activation of the repair pathway are potential indicators that may be used to predictoutcomes after IVF-ET treatment.

Maturation promoting factor destabilization mediates human chorionic gonadotropin induced meiotic resumption in rat oocytes

Human chorionic gonadotropin (hCG) mimics the action of luteinizing hormone (LH) and triggers meiotic maturation and ovulation in mammals. The mechanism by which hCG triggers meiotic resumption in mammalian oocytes remains poorly understood. We aimed to find out the impact of hCG surge on morphological changes, adenosine 3',5'-cyclic monophosphate (cAMP), guanosine 3',5'-cyclic monophosphate (cGMP), cell division cycle 25B (Cdc25B), Wee1, early mitotic inhibitor 2 (Emi2), anaphase-promoting complex/cyclosome (APC/C), meiotic arrest deficient protein 2 (MAD2), phosphorylation status of cyclin-dependent kinase 1 (Cdk1), its activity and cyclin B1 expression levels during meiotic resumption from diplotene as well as metaphase-II (M-II) arrest in cumulus oocyte complexes (COCs). Our data suggest that hCG surge increased cyclic nucleotides level in encircling granulosa cells but decreased their level in oocyte. The reduced intraoocyte cyclic nucleotides level is associated with the decrease of Cdc25B, Thr161 phosphorylated Cdk1 and Emi2 expression levels. On the other hand, hCG surge increased Wee1, Thr14/Tyr15 phosphorylated Cdk1, APC/C as well as MAD2 expression levels. The elevated APC/C activity reduced cyclin B1 level. The changes in phosphorylation status of Cdk1 and reduced cyclin B1 level might have resulted in maturation promoting factor (MPF) destabilization. The destabilized MPF finally triggered resumption of meiosis from diplotene as well as M-II arrest in rat oocytes.

Inside the granulosa transcriptome

The somatic component of follicular structure is a mixture of different cell types, represented by Granulosa cells (GCs) that are the paracrine regulators of the oocyte growth. GCs finely support this process by a continuous bidirectional talk with oocyte, which ensure oocyte quality and competence. Specific pathways are involved in the cross-talk and in both GCs and oocyte development. This review summarizes data from GCs gene expression analysis concerning both their physiological role and their interaction with oocyte. We also explore the CGs transcriptome modifications induced by controlled ovarian stimulation (COS) or pathological conditions and their impact in reproduction. The transcriptome analysis of GCs could be a powerful tool to improve our knowledge about the pathways involved in oocyte development. This approach, associated with new technologies as RNA-seq could allow the identifications of new noninvasive biological markers of oocyte quality to increase the efficiency of clinical IVF. Moreover, GCs expression analysis could be useful to shed light on new therapeutic targets by providing new options for the treatment of infertility.

Granulosa cell biomarkers to predict pregnancy in ART: pieces to solve the puzzle

Cumulus and mural granulosa cells of the ovarian follicle surround and interact with the developing oocyte. These follicular cells reflect the oocyte's overall health and may indicate subsequent developmental competence of embryos. Biomarkers of granulosa cells associated with individual oocytes could potentially be used in assisted reproduction to indicate which embryos have the best chance of implanting in the uterus and completing gestation. In this review, we have performed a comprehensive assessment of the recent literature for human cumulus and mural granulosa cell mRNA biomarkers as they relate to pregnancy and live birth. A critical discussion of variables affecting granulosa gene expression profiles for in vitro fertilization patients, including patient demographics and ovarian stimulation regimens, is presented. Although studies with microarray data were evaluated, this synopsis focuses on expressed genes that have been validated by quantitative RT-PCR. Furthermore, we summarize the current published data that support or refute identified granulosa expressed genes as potential biomarkers of embryos that give rise to ongoing pregnancy and live birth. Finally, we review studies that offer predictive models for embryo selection for uterine transfer based on biomarkers that show differential gene expression.