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.

Anti-Müllerian hormone in seminal plasma and serum: association with sperm count and sperm motility.

STUDY QUESTION: Is anti-Müllerian hormone (AMH) in seminal plasma and serum associated with sperm count and sperm motility? SUMMARY ANSWER: AMH in seminal plasma is positively associated with sperm concentration, total sperm count, and progressive sperm motility, while no association was found between serum AMH levels and semen characteristics. WHAT IS KNOWN ALREADY: AMH is secreted by the Sertoli cells and is detectable in both serum and seminal plasma in adult men. It has been suggested as a marker of spermatogenesis, however, its function in the adult male is largely unknown. STUDY DESIGN, SIZE, DURATION: Participants were recruited in between 2008 and 2013, from the general population (n = 94) and from couples with female factor infertility in a fertility clinic (n = 32). AMH data were available for 126 participants. PARTICIPANTS/MATERIALS, SETTING, METHODS: Mean age of the participants was 36 years, and BMI was between 19 and 39 kg/m(2). Semen quality was evaluated by semen analysis according to the World Health Organization, and AMH levels were measured in seminal plasma. Blood samples were analyzed for AMH, total testosterone, FSH, LH, and inhibin B. AMH analysis was performed using the improved Beckman Coulter method. MAIN RESULTS AND THE ROLE OF CHANCE: The central 95% intervals of AMH concentrations were 2-2812 pmol/l in seminal plasma and 15-134 pmol/l in serum. Total AMH (pmol/ejaculate) in seminal plasma was positively associated with sperm concentration (B = 0.177, P< 0.001) and total sperm count (B = 0.212, P< 0.001) when adjusted for age, BMI, time of abstinence, and positively associated with progressive sperm motility (B = 6.762, P = 0.001) when adjusted for age, BMI, time of abstinence, and site of sample collection. No association was found between serum AMH and semen characteristics. Serum levels of inhibin B were positively correlated with total AMH in seminal plasma (B = 18.52, P< 0.001) and concentration of AMH in serum (B = 0.507, P< 0.001). LIMITATIONS, REASONS FOR CAUTION: Participants were recruited both from the general population and from a fertility clinic. This may limit the applicability to men in the general population. WIDER IMPLICATIONS OF THE FINDINGS: The AMH levels found in this study show large inter-individual variation, especially in seminal plasma. AMH in seminal plasma may serve as a marker of sperm production, however, in the lower range the predictive value is low. STUDY FUNDING/COMPETING INTERESTS: All funding for this study was received from Oslo and Akershus University College of Applied Sciences. The authors have no conflicts of interest to declare.

Genetic variations associated with the effect of testicular cancer treatment on gonadal hormones.

STUDY QUESTION: Do genetic variations in the testosterone pathway genes modify the effect of treatment on the levels of testosterone and LH in long-term testicular cancer (TC) survivors (TCSs)? SUMMARY ANSWER: Variations in LH receptor (LHR) and in 5α-reductase II (SRD5A2) genes may modify the effect of TC treatment on testosterone levels, whereas genetic variations in the androgen receptor (AR) may modify the effect on LH levels. WHAT IS KNOWN ALREADY: TCSs experience variable degrees of long-term reduction in gonadal function after treatment. This variability can in part be explained by treatment intensity, but may also be due to individual variations in genes involved in the function and metabolism of reproductive hormones. STUDY DESIGN, SIZE, DURATION: Cross-sectional study on testosterone and LH levels in 637 Norwegian TCSs in relation to genetic variants and TC treatment. PARTICIPANTS/MATERIALS, SETTING, METHODS: The single nucleotide polymorphisms LHR Asn291Ser (rs12470652) and Ser312Asn (rs2293275), as well as SRD5A2 Ala49Thr (rs9282858) and Val89Leu (rs523349) were analyzed by allele-specific PCR. The insertion polymorphism LHR InsLQ (rs4539842) was analyzed by sequencing. The numbers of AR CAG and GGN repeats were determined by capillary electrophoresis. Blood samples were collected 5-21 years after diagnosis (median 11 years) and serum total testosterone and LH were analyzed by commercial immunoassays. The TCSs were divided into four groups according to their treatment; surgery only, radiotherapy and chemotherapy with ≤850 or >850 mg of cisplatin. Polymorphisms presenting P < 0.1 for the interaction term with treatment in an initial two-way analysis of covariance (ANCOVA) were investigated further in two consecutive one-way ANCOVA analyses to elucidate the interaction between treatment and genotype. MAIN RESULTS AND THE ROLE OF CHANCE: For the whole group of TCSs, there were no significant differences between the hormone levels in homozygotes for the wild type and carriers of at least one polymorphic allele for the investigated polymorphisms. Three of the polymorphisms showed signs of interaction with treatment, i.e. LHR InsLQ, SRD5A2 A49T and the AR CAG repeat. Follow-up analyses revealed three situations where only one of the genotypes of the polymorphism where associated with significantly different hormone levels after surgery compared with after additional cytotoxic treatment: For LHR InsLQ, only the wild-type allele was associated with lower testosterone levels after cisplatin > 850 mg compared with after surgery (24% lower, P < 0.001). For SRD5A2 A49T, testosterone levels were lower after radiotherapy compared with after surgery, but only for the heterozygotes for the polymorphism (39% lower, P = 0.001). In comparison, the testosterone levels were just slightly lower after radiotherapy (6% lower, P = 0.039) or cisplatin ≤ 850 mg (7% lower, P = 0.041), compared with surgery, independent of genotypes. For AR CAG, only the reference length of CAG = 21-22 had significantly higher LH levels after cisplatin ≤ 850 mg compared with after surgery (70% higher, P < 0.001). Independent of genotypes, however, LH levels after cisplatin ≤ 850 mg were only 26% higher than after surgery (P = 0.005). LIMITATIONS, REASONS FOR CAUTION: Unadjusted P-values are presented. For analysis involving genotypes, the level of statistical significance was adjusted for the total number of polymorphisms tested, n = 7, i.e. to P < 0.007 (0.5/7). The rather weak associations indicate that additional polymorphisms are involved in the modulation. WIDER IMPLICATIONS OF THE FINDINGS: To our knowledge, this is the first study supporting the notion that polymorphisms may explain at least some of the inter-individual differences in endocrine response to TC treatment. Our findings suggest that individuals with certain genotypes may be more vulnerable to certain treatments. Knowledge on genetic predisposition concerning treatment-related endocrine gonadotoxicity to different treatment regimens may help tailoring TC therapy when possible. STUDY FUNDING/COMPETING INTERESTS: This study was supported by the Research Council of Norway (Grant No. 160619). There were no competing interests.

Variations in testosterone pathway genes and susceptibility to testicular cancer in Norwegian men.

Imbalance between the oestrogen and androgen levels in utero is hypothesized to influence testicular cancer (TC) risk. Thus, variation in genes involved in the action of sex hormones may contribute to variability of an individual's susceptibility to TC. Mutations in testosterone pathway genes may alter the level of testosterone in vivo and hypothetically the risk of developing TC. Luteinizing hormone receptor (LHR), 5α-reductase II (SRD5A2) and androgen receptor (AR) are key elements in androgen action. A case-control study comprising 651 TC cases and 313 controls in a Norwegian population was conducted for investigation of polymorphisms in the LHR, SRD5A and AR genes and their possible association with TC. A statistical significant difference was observed in patients being heterozygous for the LHR Asn312Ser polymorphism when comparing genotypes between all TC cases and controls (OR = 0.66, 95% CI = 0.48-0.89, p(adj) = 0.049). No statistically significant difference between the histological subtypes seminoma and non-seminoma was observed. Our results may suggest a possible association between genetic variation in the LHR gene and the risk of developing TC.

CYP1A1, CYP3A5 and CYP3A7 polymorphisms and testicular cancer susceptibility.

Testicular cancer (TC) incidence is increasing worldwide, but the aetiology remains largely unknown. An unbalanced level of oestrogens and androgens in utero is hypothesized to influence TC risk. Polymorphisms in genes encoding cytochrome P450 (CYP) enzymes involved in metabolism of reproductive hormones, such as CYP1A1, CYP3A5 and CYP3A7, may contribute to variability of an individual's susceptibility to TC. The aim of this case-control study was to investigate possible associations between different CYP genotypes and TC, as well as histological type of TC. The study comprised 652 TC cases and 199 controls of Norwegian Caucasian origin. Genotyping of the CYP1A1*2A (MspI), CYP1A1*2C (I462V), CYP1A1*4 (T461N), CYP3A5*3C (A6986G) and CYP3A7*2 (T409R) polymorphisms was performed using TaqMan allelic discrimination or sequencing. The CYP1A1*2A allele was associated with 44% reduced risk of TC with each polymorphic allele [odds ratio (OR) = 0.56, 95% confidence interval (CI) = 0.40-0.78, p(trend) = 0.001], whereas the CYP1A1*2C allele was associated with 56% reduced risk of TC with each polymorphic allele (OR = 0.44, 95% CI = 0.25-0.75, p(trend) = 0.003). The decreased risk per allele was significant for seminomas (OR = 0.46, 95% CI, 0.31-0.70, p(trend) < 0.001 and OR = 0.31, 95% CI = 0.14-0.66, p(trend) = 0.002, respectively), but only borderline significant for non-seminomas (OR = 0.65, 95% CI = 0.45-0.95, p(trend) = 0.027 and OR = 0.55, 95% CI = 0.30-1.01, p(trend) = 0.052, respectively). There were no statistically significant differences in the distribution of the CYP3A5*3C and CYP3A7*2 polymorphic alleles between TC cases and controls. This study suggests that polymorphisms in the CYP1A1 gene may contribute to variability of individual susceptibility to TC.

Fatty acid composition of spermatozoa is associated with BMI and with semen quality.

High body mass index (BMI) is negatively associated with semen quality. In addition, the composition of fatty acids of spermatozoa has been shown to be important for their function. The aim of the study was to examine the association between BMI and the composition of spermatozoa fatty acids in men spanning a broad BMI range. We also analysed the relation between fatty acid composition of spermatozoa and semen characteristics, and the relationship between serum fatty acids and spermatozoa fatty acids. One hundred forty-four men with unknown fertility status were recruited from the general population, from couples with identified female infertility and from morbid obesity centres. Standard semen analysis (WHO) and sperm DNA integrity (DFI) analysis were performed. Fatty acid compositions were assessed by gas chromatography. When adjusted for possible confounders, BMI was negatively associated with levels of sperm docosahexaenoic acid (DHA) (p < 0.001) and palmitic acid (p < 0.001). The amount of sperm DHA correlated positively with total sperm count (r = 0.482), sperm concentration (r = 0.469), sperm vitality (r = 0.354), progressive sperm motility (r = 0.431) and normal sperm morphology (r = 0.265). A negative association was seen between DHA levels and DNA fragmentation index (r = -0.247). Levels of spermatozoa palmitic acid correlated positively with total sperm count (r = 0.227), while levels of linoleic acid correlated negatively (r = -0.254). When adjusted for possible confounders, only the levels of arachidonic acid showed positive correlation between spermatozoa and serum phospholipids (r = 0.262). Changes in the fatty acid composition of spermatozoa could be one of the mechanisms underlying the negative association between BMI and semen quality. The relationship between fatty acids of spermatozoa and serum phospholipids was minor, which indicates that BMI affects fatty acid composition of spermatozoa through regulation of fatty acid metabolism in the testis. The role of dietary intake of fatty acids on the spermatozoa fatty acid composition remains to be elucidated.

Phosphodiesterase 4D and protein kinase a type II constitute a signaling unit in the centrosomal area.

The mediation of cAMP effects by specific pools of protein kinase A (PKA) targeted to distinct subcellular domains raises the question of how inactivation of the cAMP signal is achieved locally and whether similar targeting of phosphodiesterases (PDEs) to sites of cAMP/PKA action could be observed. Here, we demonstrate that Sertoli cells of the testis contain an insoluble PDE4D3 isoform, which is shown by immunofluorescence to target to centrosomes. Staining of PDE4D and PKA shows co-localization of PDE4D with PKA-RIIalpha and RIIbeta in the centrosomal region. Co-precipitation of RII subunits and PDE4D3 from cytoskeletal extracts indicates a physical association of the two proteins. Distribution of PDE4D overlaps with that of the centrosomal PKA-anchoring protein, AKAP450, and AKAP450, PDE4D3, and PKA-RIIalpha co-immunoprecipitate. Finally, both PDE4D3 and PKA co-precipitate with a soluble fragment of AKAP450 encompassing amino acids 1710 to 2872 when co-expressed in 293T cells. Thus, a centrosomal complex that includes PDE4D and PKA constitutes a novel signaling unit that may provide accurate spatio-temporal modulation of cAMP signals.

Characterization of the gene encoding the human type II cGMP-dependent protein kinase.

The type II cGMP-dependent protein kinase (cGK) plays a pivotal role in the regulation of intestinal fluid balance in man. Furthermore, mice carrying a null mutation for the gene encoding the type II cGK develop as dwarfs indicating that this enzyme has other less characterized roles. The present report describes the isolation and characterization of bacterial artificial chromosome (BAC)- and P1-derived artificial chromosome (PAC)-clones containing the gene encoding the human type II cGK. The gene was estimated to cover at least 125 kb and consisted of 19 exons separated by introns of various lengths. The splice junctions of the type II cGK gene corresponded well with the structure of the gene encoding human type I cGK and with the splice junctions observed in the Drosophila melanogaster DG2 gene. 5'-rapid amplification of cDNA-ends established the presence of a non-translated exon.

Cloning and characterization of a cDNA encoding an A-kinase anchoring protein located in the centrosome, AKAP450.

A combination of protein kinase A type II (RII) overlay screening, database searches and PCR was used to identify a centrosomal A-kinase anchoring protein. A cDNA with an 11.7 kb open reading frame was characterized and found to correspond to 50 exons of genomic sequence on human chromosome 7q21-22. This cDNA clone encoded a 3908 amino acid protein of 453 kDa, that was designated AKAP450 (DDBJ/EMBL/GenBank accession No. AJ131693). Sequence comparison demonstrated that the open reading frame contained a previously characterized cDNA encoding Yotiao, as well as the human homologue of AKAP120. Numerous coiled-coil structures were predicted from AKAP450, and weak homology to pericentrin, giantin and other structural proteins was observed. A putative RII-binding site was identified involving amino acid 2556 of AKAP450 by mutation analysis combined with RII overlay and an amphipatic helix was predicted in this region. Immunoprecipitation of RII from RIPA-buffer extracts of HeLa cells demonstrated co-precipitation of AKAP450. By immunofluorecent labeling with specific antibodies it was demonstrated that AKAP450 localized to centrosomes. Furthermore, AKAP450 was shown to co-purify in centrosomal preparations. The observation of two mRNAs and several splice products suggests additional functions for the AKAP450 gene.

CDK1-mediated phosphorylation of the RIIalpha regulatory subunit of PKA works as a molecular switch that promotes dissociation of RIIalpha from centrosomes at mitosis.

Protein kinase A regulatory subunit RIIalpha is tightly bound to centrosomal structures during interphase through interaction with the A-kinase anchoring protein AKAP450, but dissociates and redistributes from centrosomes at mitosis. The cyclin B-p34(cdc2) kinase (CDK1) has been shown to phosphorylate RIIalpha on T54 and this has been proposed to alter the subcellular localization of RIIalpha. We have made stable transfectants from an RIIalpha-deficient leukemia cell line (Reh) that expresses either wild-type or mutant RIIalpha (RIIalpha(T54E)). When expressed, RIIalpha detaches from centrosomes at mitosis and dissociates from its centrosomal location in purified nucleus-centrosome complexes by incubation with CDK1 in vitro. By contrast, centrosomal RIIalpha(T54E) is not redistributed at mitosis, remains mostly associated with centrosomes during all phases of the cell cycle and cannot be solubilized by CDK1 in vitro. Furthermore, RIIalpha is solubilized from particular cell fractions and changes affinity for AKAP450 in the presence of CDK1. D and V mutations of T54 also reduce affinity for the N-terminal RII-binding domain of AKAP450, whereas small neutral residues do not change affinity detected by surface plasmon resonance. In addition, only RIIalpha(T54E) interacts with AKAP450 in a RIPA-soluble extract from mitotic cells. Finally, microtubule repolymerization from mitotic centrosomes of the RIIalpha(T54E) transfectant is poorer and occurs at a lower frequency than that of RIIalpha transfectants. Our results suggest that T54 phosphorylation of RIIalpha by CDK1 might serve to regulate the centrosomal association of PKA during the cell cycle.