Machine-learning algorithm incorporating capacitated sperm intracellular pH predicts conventional in vitro fertilization success in normospermic patients

Cytosolic and Acrosomal pH Regulation in Mammalian Sperm

As in most cells, intracellular pH regulation is fundamental for sperm physiology. Key sperm functions like swimming, maturation, and a unique exocytotic process, the acrosome reaction, necessary for gamete fusion, are deeply influenced by pH. Sperm pH regulation, both intracellularly and within organelles such as the acrosome, requires a coordinated interplay of various transporters and channels, ensuring that this cell is primed for fertilization. Consistent with the pivotal importance of pH regulation in mammalian sperm physiology, several of its unique transporters are dependent on cytosolic pH. Examples include the Ca2+ channel CatSper and the K+ channel Slo3. The absence of these channels leads to male infertility. This review outlines the main transport elements involved in pH regulation, including cytosolic and acrosomal pH, that participate in these complex functions. We present a glimpse of how these transporters are regulated and how distinct sets of them are orchestrated to allow sperm to fertilize the egg. Much research is needed to begin to envision the complete set of players and the choreography of how cytosolic and organellar pH are regulated in each sperm function.

Predicting Male Infertility Using Artificial Neural Networks: A Review of the Literature

Male infertility is a relevant public health problem, but there is no systematic review of the different machine learning (ML) models and their accuracy so far. The present review aims to comprehensively investigate the use of ML algorithms in predicting male infertility, thus reporting the accuracy of the used models in the prediction of male infertility as a primary outcome. Particular attention will be paid to the use of artificial neural networks (ANNs). A comprehensive literature search was conducted in PubMed, Scopus, and Science Direct between 15 July and 23 October 2023, conducted under the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We performed a quality assessment of the included studies using the recommended tools suggested for the type of study design adopted. We also made a screening of the Risk of Bias (RoB) associated with the included studies. Thus, 43 relevant publications were included in this review, for a total of 40 different ML models detected. The studies included reported a good quality, even if RoB was not always good for all the types of studies. The included studies reported a median accuracy of 88% in predicting male infertility using ML models. We found only seven studies using ANN models for male infertility prediction, reporting a median accuracy of 84%.

Artificial Intelligence in Andrology: From Semen Analysis to Image Diagnostics

Artificial intelligence (AI) in medicine has gained a lot of momentum in the last decades and has been applied to various fields of medicine. Advances in computer science, medical informatics, robotics, and the need for personalized medicine have facilitated the role of AI in modern healthcare. Similarly, as in other fields, AI applications, such as machine learning, artificial neural networks, and deep learning, have shown great potential in andrology and reproductive medicine. AI-based tools are poised to become valuable assets with abilities to support and aid in diagnosing and treating male infertility, and in improving the accuracy of patient care. These automated, AI-based predictions may offer consistency and efficiency in terms of time and cost in infertility research and clinical management. In andrology and reproductive medicine, AI has been used for objective sperm, oocyte, and embryo selection, prediction of surgical outcomes, cost-effective assessment, development of robotic surgery, and clinical decision-making systems. In the future, better integration and implementation of AI into medicine will undoubtedly lead to pioneering evidence-based breakthroughs and the reshaping of andrology and reproductive medicine.

Na+/H+ Exchangers (NHEs) in Mammalian Sperm: Essential Contributors to Male Fertility

Na+/H+ exchangers (NHEs) are known to be important regulators of pH in multiple intracellular compartments of eukaryotic cells. Sperm function is especially dependent on changes in pH and thus it has been postulated that NHEs play important roles in regulating the intracellular pH of these cells. For example, in order to achieve fertilization, mature sperm must maintain a basal pH in the male reproductive tract and then alkalize in response to specific signals in the female reproductive tract during the capacitation process. Eight NHE isoforms are expressed in mammalian testis/sperm: NHE1, NHE3, NHE5, NHE8, NHA1, NHA2, NHE10, and NHE11. These NHE isoforms are expressed at varying times during spermatogenesis and localize to different subcellular structures in developing and mature sperm where they contribute to multiple aspects of sperm physiology and male fertility including proper sperm development/morphogenesis, motility, capacitation, and the acrosome reaction. Previous work has provided evidence for NHE3, NHE8, NHA1, NHA2, and NHE10 being critical for male fertility in mice and NHE10 has recently been shown to be essential for male fertility in humans. In this article we review what is known about each NHE isoform expressed in mammalian sperm and discuss the physiological significance of each NHE isoform with respect to male fertility.

Common as well as unique methylation-sensitive DNA regulatory elements in three mammalian SLC9C1 genes

The SLC9C1 gene (which encodes the NHE10 protein) is essential for male fertility in both mice and humans, however the epigenetic mechanisms regulating its testis/sperm-specific gene expression have yet to be studied. Here we identify and characterize DNA regulatory elements of the SLC9C1 gene across three mammalian species: mouse, rat, and human. First, in silico analysis of these mammalian SLC9C1 genes identified a CpG island located upstream of the transcription start site in the same relative position in all three genes. Further analysis reveals that this CpG island behaves differently, with respect to gene regulatory activity, in the mouse SLC9C1 gene than it does in the rat and human SLC9C1 gene. The mouse SLC9C1 CpG island displays strong promoter activity by itself and seems to have a stronger gene regulatory effect than either the rat or human SLC9C1 CpG islands. While the function of the upstream SLC9C1 CpG island may be divergent across the three studied species, it appears that the promoters of these three mammalian SLC9C1 genes share similar DNA methylation-sensitive regulatory mechanisms. All three SLC9C1 promoter regions are differentially methylated in lung and testis, being more hypermethylated in lung relative to the testis, and DNA sequence alignments provide strong evidence of primary sequence conservation. Luciferase assays reveal that in vitro methylation of constructs containing different elements of the three SLC9C1 genes largely exhibit methylation-sensitive promoter activity (reduced promoter activity when methylated) in both HEK 293 and GC-1spg cells. In total, our data suggest that the DNA methylation-sensitive elements of the mouse, rat, and human SLC9C1 promoters are largely conserved, while the upstream SLC9C1 CpG island common to all three species seems to perform a different function in mouse than it does in rat and human. This work provides evidence that while homologous genes can all be regulated by DNA methylation-dependent epigenetic mechanisms, the location of the specific cis-regulatory elements responsible for this regulation can differ across species.

SLO3: A Conserved Regulator of Sperm Membrane Potential

Sperm cells must undergo a complex maturation process after ejaculation to be able to fertilize an egg. One component of this maturation is hyperpolarization of the membrane potential to a more negative value. The ion channel responsible for this hyperpolarization, SLO3, was first cloned in 1998, and since then much progress has been made to determine how the channel is regulated and how its function intertwines with various signaling pathways involved in sperm maturation. Although Slo3 was originally thought to be present only in the sperm of mammals, recent evidence suggests that a primordial form of the gene is more widely expressed in some fish species. Slo3, like many reproductive genes, is rapidly evolving with low conservation between closely related species and different regulatory and pharmacological profiles. Despite these differences, SLO3 appears to have a conserved role in regulating sperm membrane potential and driving large changes in response to stimuli. The effect of this hyperpolarization of the membrane potential may vary among mammalian species just as the regulation of the channel does. Recent discoveries have elucidated the role of SLO3 in these processes in human sperm and provided tools to target the channel to affect human fertility.

Which sperm parameter limits could really guide the clinical decision in assisted reproduction?

BACKGROUND

The predictive role of sperm motility and morphology was recently detected in a large sample of more than 20000 assisted reproductive technology (ART) fresh cycles. However, the complete ART procedure consisted of both fresh and frozen-embryos transfers and only a comprehensive evaluation of the entire process could really confirm if these parameters really predict the ART success. The aim of the study was to evaluate which sperm parameter could predict the success of ART.

METHODS

A retrospective, real-world data analysis was performed, enrolling all couples attending ART from 2008 to 2021, including both fresh and frozen cycles and both in vitro fertilization (IVF) and intra-cytoplasmic sperm injection (ICSI) procedures.

RESULTS

Fresh cycles success (considering live birth rate) was predicted by female age (1.04 [1.02-1.06]), injected oocytes (0.96 [0.93-0.99]), embryo number (0.79 [0.75-0.83]) and progressive sperm motility (0.98 [0.97-0.99]). On the contrary, frozen cycle outcomes were predicted only by sperm motility (0.97 [0.95-0.99]). This prediction was confirmed in IVF but not in ICSI cycles.

CONCLUSION

Both female and male parameters predicted the ART success considering the entire path. However, frozen cycle success was predicted only by progressive sperm motility in IVF cycles, suggesting that the potential amelioration of this male parameter is relevant to improve ART success. Those couples expected to obtain the highest embryos after fertilization (low female age and better semen parameters) will have more attempts with frozen cycles and thus would benefit from a potential treatment focused to improve sperm parameters.

What advances may the future bring to the diagnosis, treatment, and care of male sexual and reproductive health?

Over the past 40 years, since the publication of the original WHO Laboratory Manual for the Examination and Processing of Human Semen, the laboratory methods used to evaluate semen markedly changed and benefited from improved precision and accuracy, as well as the development of new tests and improved, standardized methodologies. Herein, we present the impact of the changes put forth in the sixth edition together with our views of evolving technologies that may change the methods used for the routine semen analysis, up-and-coming areas for the development of new procedures, and diagnostic approaches that will help to extend the often-descriptive interpretations of several commonly performed semen tests that promise to provide etiologies for the abnormal semen parameters observed. As we look toward the publication of the seventh edition of the manual in approximately 10 years, we describe potential advances that could markedly impact the field of andrology in the future.

Live-Birth Prediction of Natural-Cycle In Vitro Fertilization Using 57,558 Linked Cycle Records: A Machine Learning Perspective

BACKGROUND

Natural-cycle in vitro fertilization (NC-IVF) is an in vitro fertilization (IVF) cycle without gonadotropins or any other stimulation of follicular growth. Previous studies on live-birth prediction of NC-IVF were very few; the sample size was very limited. This study aims to construct a machine learning model to predict live-birth occurrence of NC-IVF using 57,558 linked cycle records and help clinicians develop treatment strategies.

DESIGN AND METHODS

The dataset contained 57,558 anonymized register patient records undergoing NC-IVF cycles from 2005 to 2016 filtered from 7bsp;60,732 records in the Human Fertilisation and Embryology Authority (HFEA) data. We selected matching records and features through data filtering and feature selection methods. Two groups of twelve machine learning models were trained and tested. Eight metrics, e.g., F1 score, Matthews correlation coefficient (MCC), the area under the receiver operating characteristic curve (AUC), etc., were computed to evaluate the performance of each model.

RESULTS

Two groups of twelve models were trained and tested. The artificial neural network (ANN) model performed the best in the machine learning group (F1 score, 70.87%; MCC, 50.37%; and AUC score, 0.7939). The LogitBoost model obtained the best scores in the ensemble learning group (F1 score, 70.57%; MCC, 50.75%; and AUC score, 0.7907). After the comparison between the two models, the LogitBoost model was recognized as an optimal one.

CONCLUSION

In this study, NC-IVF-related datasets were extracted from the HFEA data, and a machine learning-based prediction model was successfully constructed through this largest NC-IVF dataset currently. This model is universal and stable, which can help clinicians predict the live-birth success rate of NC-IVF in advance before developing IVF treatment strategies and then choose the best benefit treatment strategy according to the patients' wishes. As "use less stimulation and back to natural condition" becomes more and more popular, this model is more meaningful in the decision-making assistance system for IVF.

Are sperm parameters able to predict the success of assisted reproductive technology? A retrospective analysis of over 22,000 assisted reproductive technology cycles

Background

An explosive increase in couples attending assisted reproductive technology has been recently observed, despite an overall success rate of about 20%–30%. Considering the assisted reproductive technology‐related economic and psycho‐social costs, the improvement of these percentages is extremely relevant. However, in the identification of predictive markers of assisted reproductive technology success, male parameters are largely underestimated so far.

Study design

Retrospective, observational study.

Objectives

To evaluate whether conventional semen parameters could predict assisted reproductive technology success.

Materials and methods

All couples attending a single third‐level fertility center from 1992 to 2020 were retrospectively enrolled, collecting all semen and assisted reproductive technology parameters of fresh cycles. Fertilization rate was the primary end‐point, representing a parameter immediately dependent on male contribution. Pregnancy and live birth rates were considered in relation to semen variables. Statistical analyses were performed using the parameters obtained according to the World Health Organization manual editions used for semen analysis.

Results

Note that, 22,013 in vitro fertilization and intracytoplasmic sperm injection cycles were considered. Overall, fertilization rate was significantly lower in patients with abnormal semen parameters compared to normozoospermic men, irrespective of the World Health Organization manual edition. In the in vitro fertilization setting, both progressive motility (p = 0.012) and motility after capacitation (p = 0.002) significantly predicted the fertilization rate (statistical accuracy = 71.1%). Sperm motilities also predicted pregnancy (p < 0.001) and live birth (p = 0.001) rates. In intracytoplasmic sperm injection cycles, sperm morphology predicted fertilization rate (p = 0.001, statistical accuracy = 90.3%). Sperm morphology significantly predicted both pregnancy (p < 0.001) and live birth (p < 0.001) rates and a cut‐off of 5.5% was identified as a threshold to predict clinical pregnancy (area under the curve = 0.811, p < 0.001).

Discussion

Interestingly, sperm motility plays a role in predicting in vitro fertilization success, while sperm morphology is the relevant parameter in intracytoplasmic sperm injection cycles. These parameters may be considered reliable tools to measure the male role on ART outcomes, potentially impacting the clinical management of infertile couples.

Conserved Mechanism of Bicarbonate-Induced Sensitization of CatSper Channels in Human and Mouse Sperm

To fertilize an egg, mammalian sperm must undergo capacitation in the female genital tract. A key contributor to capacitation is the calcium (Ca²⁺) channel CatSper, which is activated by membrane depolarization and intracellular alkalinization. In mouse epididymal sperm, membrane depolarization by exposure to high KCl triggers Ca²⁺ entry through CatSper only in alkaline conditions (pH 8.6) or after in vitro incubation with bicarbonate (HCO₃^–) and bovine serum albumin (capacitating conditions). However, in ejaculated human sperm, membrane depolarization triggers Ca²⁺ entry through CatSper in non-capacitating conditions and at lower pH (< pH 7.4) than is required in mouse sperm. Here, we aimed to determine the mechanism(s) by which CatSper is activated in mouse and human sperm. We exposed ejaculated mouse and human sperm to high KCl to depolarize the membrane and found that intracellular Ca²⁺ concentration increased at pH 7.4 in sperm from both species. Conversely, intracellular Ca²⁺ concentration did not increase under these conditions in mouse epididymal or human epididymal sperm. Furthermore, pre-incubation with HCO₃^– triggered an intracellular Ca²⁺ concentration increase in response to KCl in human epididymal sperm. Treatment with protein kinase A (PKA) inhibitors during exposure to HCO₃^– inhibited Ca²⁺ concentration increases in mouse epididymal sperm and in both mouse and human ejaculated sperm. Finally, we show that soluble adenylyl cyclase and increased intracellular pH are required for the intracellular Ca²⁺ concentration increase in both human and mouse sperm. In summary, our results suggest that a conserved mechanism of activation of CatSper channels is present in both human and mouse sperm. In this mechanism, HCO₃^– in semen activates the soluble adenylyl cyclase/protein kinase A pathway, which leads to increased intracellular pH and sensitizes CatSper channels to respond to membrane depolarization to allow Ca²⁺ influx. This indirect mechanism of CatSper sensitization might be an early event capacitation that occurs as soon as the sperm contact the semen.

Soluble adenylyl cyclase inhibition prevents human sperm functions essential for fertilization

Soluble adenylyl cyclase (sAC: ADCY10) has been genetically confirmed to be essential for male fertility in mice and humans. In mice, ex vivo studies of dormant, caudal epididymal sperm demonstrated that sAC is required for initiating capacitation and activating motility. We now use an improved sAC inhibitor, TDI-10229, for a comprehensive analysis of sAC function in mouse and human sperm. In contrast to caudal epididymal mouse sperm, human sperm are collected post-ejaculation, after sAC activity has already been stimulated. In addition to preventing the capacitation-induced stimulation of sAC and protein kinase A activities, tyrosine phosphorylation, alkalinization, beat frequency and acrosome reaction in dormant mouse sperm, sAC inhibitors interrupt each of these capacitation-induced changes in ejaculated human sperm. Furthermore, we show for the first time that sAC is required during acrosomal exocytosis in mouse and human sperm. These data define sAC inhibitors as candidates for non-hormonal, on-demand contraceptives suitable for delivery via intravaginal devices in women.

Artificial intelligence in the fertility clinic: status, pitfalls and possibilities

In recent years, the amount of data produced in the field of ART has increased exponentially. The diversity of data is large, ranging from videos to tabular data. At the same time, artificial intelligence (AI) is progressively used in medical practice and may become a promising tool to improve success rates with ART. AI models may compensate for the lack of objectivity in several critical procedures in fertility clinics, especially embryo and sperm assessments. Various models have been developed, and even though several of them show promising performance, there are still many challenges to overcome. In this review, we present recent research on AI in the context of ART. We discuss the strengths and weaknesses of the presented methods, especially regarding clinical relevance. We also address the pitfalls hampering successful use of AI in the clinic and discuss future possibilities and important aspects to make AI truly useful for ART.