Sperm aminopeptidase N identifies the potential for high-quality blastocysts and viable embryos in oocyte-donation cycles

Abstract

STUDY QUESTION

Is there a relationship between human sperm aminopeptidase N (APN) and embryo development in humans?

SUMMARY ANSWER

Human sperm APN could possibly become a new molecular biomarker for identifying the potential for high-quality and usable embryos.

WHAT IS KNOWN ALREADY

The diagnosis of male fertility is one of the major concerns of reproductive medicine. Approximately 30–40% of men with otherwise normal fertility parameters are still unable to achieve pregnancy. The predictive clinical value of semen analysis to identify fertile or infertile males is limited; therefore, new diagnostic methodologies for sperm are urgently required. Sperm APN may be a relevant molecular marker due to its high concentration in sperm cells and its important roles in sperm physiology, such as its functions in motility, acrosome reaction and embryo development.

STUDY DESIGN, SIZE, DURATION

This study included 81 couples who underwent oocyte-donation cycles at Clínica IVI Bilbao (Spain), yielding 611 embryos, between September 2014 and July 2015.

PARTICIPANTS/MATERIALS, SETTING, METHODS

This study was conducted in an assisted reproduction unit and an academic research laboratory. All the semen samples were examined and classified following World Health Organization guidelines. Spermatozoa were isolated from semen using the discontinuous colloidal silica gradient (45–90%) technique. Embryo quality and development were determined according to the Spanish Association of Reproduction Biology Studies (ASEBIR) criteria. Human sperm APN levels were analyzed by quantitative and semiquantitative flow cytometry.

MAIN RESULTS AND THE ROLE OF CHANCE

The most well-developed and usable blastocysts were associated with low sperm APN levels. Semen samples that had lower APN levels generated more expanded, hatched and usable blastocysts and fewer early, arrested and non-usable blastocysts. The cumulative probability of having well-developed blastocysts increased by 1.38-fold at Day 5 and 1.90-fold at Day 6 of embryo development, and the likelihood of having usable embryos increased by 1.48-fold, when semen samples with low APN levels were used during the ICSI technique.

LIMITATIONS, REASONS FOR CAUTION

The data were obtained from a single fertility clinic. A multicentre study will be required to confirm the results.

WIDER IMPLICATIONS OF THE FINDINGS

Human sperm APN has the potential to become a new molecular biomarker to help identify the potential for high-quality embryos and diagnose male infertility, especially when seminal parameters are close to the threshold values. It could be a crucial tool for couples for whom the number of usable blastocysts is critical for ART success.

STUDY FUNDING/COMPETING INTEREST(S)

This study was supported by grants from the Basque Government (GIC15/165) and the University of the Basque Country (UPV/EHU) (EHUA14/17). The authors declare that they have no conflicts of interest.

TRIAL REGISTRATION NUMBER

N/A.

aminopeptidase N, human sperm, embryo, assisted reproduction techniques, infertility, male

Introduction

Worldwide, over 186 million people have fertility problems, and male infertility is a factor in 30–50% of cases of infertility (Inhorn and Patrizio, 2015). Abnormal semen parameters, such as a low sperm concentration and poor sperm motility, contribute to infertility and are commonly used to determine fertility status in men (Lewis, 2007; Samplaski et al., 2010). The diagnosis of male infertility is primarily based on seminogram analyses, which mostly rely on the microscopic analysis of sperm morphology, motility and concentration (WHO, 2010). However, 30–40% of men with otherwise normal fertility parameters are still unable to achieve pregnancy (Liu and Baker, 2003). In addition, semen analysis performed without adequate quality control is of almost no clinical value. Even when using appropriate methods, the predictive value of semen analysis for identifying fertile or infertile males is far from acceptable, and there is a considerable overlap between the values exhibited by males whose partners conceive and subfertile males, suggesting that male infertility can be caused by deficiencies not yet described (Lewis, 2007; Garrido et al., 2008; Samplaski et al., 2010). Therefore, the development of new diagnostic or prognostic methodologies for sperm that focus more on specific molecular characteristics is urgently required.

Sperm molecular features, such as RNA and proteins, are proposed to be involved in fertilization and embryo development (Janny and Menezo, 1994; Barroso et al., 2009; Govindaraju et al., 2012; Gianzo et al., 2018). Among these features, sperm aminopeptidase N (EC.34.11.2; APN) may affect early embryonic development (Khatun et al., 2017, 2018). APN, also known as CD13, is a surface membrane-bound enzyme belonging to the M1 family of zinc metallopeptidases that is widely expressed on the cellular surfaces of human tissues (Hooper, 1994; Mina-Osorio, 2008). This widespread enzyme plays a critical role in many physiological and pathological processes, such as cellular differentiation, chemotaxis and motility, viral receptivity, signaling pathways associated with pain, blood pressure regulation, phagocytosis and fertility, among others (Mina-Osorio, 2008; Chen et al., 2012). APN is present in both sperm and seminal fractions (Fernandez et al., 2002; Subiran et al., 2008), and functional studies have reported a role of APN in human sperm motility (Subiran et al., 2008, 2010; Khatun et al., 2017) and acrosome reaction regulation (Togo and Morisawa, 1997, 2004; Viudes de Castro et al., 2015; Khatun et al., 2017; 2018). Interestingly, APN activity levels were found to be altered in semen from subfertile patients (Irazusta et al., 2004), and changes in APN activity in sperm culture medium affects early embryonic development, reinforcing the idea that APN may play a key role in the early stages of embryo development. Because embryo quality is one of the major predictors of ART success, our aim was to evaluate whether there is a relationship between human sperm APN and embryo development in humans, in order to evaluate the possible use of human sperm APN as a molecular biomarker for identifying the potential for high-quality embryos and, therefore, to improve reproductive success during ARTs.

Materials and methods

Ethical approval

This prospective study was approved by the Ethics Committee of the University of the Basque Country (CEISH/61/2011). Semen samples were obtained from male partners of couples who underwent oocyte-donation cycles at the Clínica IVI Bilbao, Basque Country, Spain. Study participation and sperm samples for the research were obtained after written consent was provided by the patients. All the samples and data were kept anonymous. All the experiments were performed in accordance with the relevant guidelines and regulations.

Patients and semen analysis and preparation

A total of 81 patients were included in this prospective study between September 2014 and July 2015. Normal or pathological semen samples were collected on the day of oocyte retrieval by masturbation on site after a 2- to 5-day period of sexual abstinence. The samples were collected into a sterile container and allowed to liquefy at 37°C and 5% (v/v) CO₂ for 10 min before processing by the density gradient centrifugation method. In brief, the sperm samples were centrifuged at 300g for 20 min through a discontinuous colloidal silica density gradient of PureSperm (Nidacon, Gothenberg, Sweden). The pellets were collected and washed (at 400g for 5 min) in 2 ml of Global^® fertilization medium (LifeGlobal Group, Brussels, Belgium). Finally, the sperm cells were diluted in 0.2–1 ml of medium for the ICSI technique.

Semen volume, as well as sperm concentration and motility, were measured for each sample, in raw ejaculated samples and after capacitation (processes or prepared samples). The evaluation of the sample was performed 30–60 min after collection. The semen samples were analyzed according to the 2010 World Health Organization (WHO) criteria (WHO, 2010). The ejaculate volume (ml) was directly estimated by weighing the sample in the container in which it was collected. Sperm concentration (×10⁶ sperm/ml) and sperm motility (progressive, non-progressive and non-motile (%)) were determined in duplicate on a heated microscope stage, and at least 200 spermatozoa per replicate were counted. The mean of homogenous replicates was used in the analysis.

The remaining spermatozoa after the ICSI procedures were collected for molecular analysis by flow cytometry. Later, the molecular data were assessed to determine the basic sperm parameters and embryo morphology and quality parameters.

Donor and recipient stimulation in assisted reproduction cycles

The use of an oocyte-donation model allowed us to analyze the relationship between sperm APN levels and fertilization rates and embryo quality and development, while controlling for biases due to female factors.

All donors were from the clinic’s oocyte-donation program. Donor selection followed the criteria established by the Spanish Assisted Reproduction Law. All the donors underwent a psychological evaluation. The donors ranged between 18 and 35 years old. Their complete medical history was recorded, including current or past exposure to radiation or dangerous chemicals, intravenous drug use and reproductive history. The donors had normal physical and gynaecological examinations (normal uterus and ovaries, as confirmed by pelvic ultrasound), normal body weight (BMI 18–28 kg/m²), and no family history of hereditary or chromosomal diseases. Furthermore, all the donors had normal karyotypes and negative test results following screening for sexually transmitted diseases. Professional confidentiality regarding the identities of the gamete donors and recipients was maintained in accordance with the dictates of the aforementioned Assisted Reproduction Law. Donor oocytes were assigned to each recipient following a routine clinical procedure, which consists of matching and considering blood type, phenotypical characteristics and special requirements, such as screening for a specific disease.

The protocols for ovarian stimulation, oocyte recruitment and management and steroid replacement in the oocyte recipients have been described previously (Meseguer et al., 2006).

Oocyte insemination techniques

Recovered oocytes were fertilized by the ICSI technique. All the oocytes retrieved from a single donor were donated to a single compatible recipient. Oocytes were denuded prior to sperm injection by enzymatic treatment (in 40 IU/ml of hyaluronidase) (Hyaluronidase, LifeGlobal Group Guilford, EE. UU.) and mechanical methods. The oocytes were then placed in 20 µl drops in a pre-equilibrated culture dish. Subsequently, ICSI was performed under an inverted microscope at 400× magnification (Nikon Eclipse, Barcelona, Spain). Only mature metaphase II oocytes were selected for sperm injection. Finally, the injected oocytes were cultured at 37°C in a 5% (v/v) CO₂-controlled atmosphere.

Fertilization rates and embryo quality and development

Fertilization was assessed 16–19 h after microinjection by confirmation of two polar bodies and two pronuclei. A total of 843 oocytes were evaluated under an inverted microscope at 400× magnification.

Embryo quality was assessed at both cleavage stages (Days 2 and 3), at the early embryo stage, and in later phases of in vitro development (Day 5) at the blastocyst stage. Early embryo quality was evaluated by considering the number of blastomeres, type and percentage of fragmentation, blastomere symmetry (equal, similar, different), the presence and number of vacuoles (either absent, scarce and diameter <5 mm, or abundant), the zona pellucida (normal/abnormal), and the presence of multinucleated cells. According to the Spanish Association of Reproduction Biology Studies (ASEBIR) criteria, embryos were classified into four grades (A, B, C, D) based on implantation potential and in combination with the aforementioned morphological parameters (Supplementary Figs S1 and S2) (ASEBIR, 2008). Specifically, Grade A yielded the best prognosis, and Grade D yielded the worst prognosis for implantation (ASEBIR, 2008; Alpha Scientists in Reproductive Medicine and ESHRE Special Interest Group of Embryology, 2011). Human blastocysts were scored on Day 5 of embryo development (112–118 h post-microinjection). Embryos were grouped according to blastocoel expansion degree as early (BT), expanding (BC), expanded (BE) or hatching/hatched (BHi) blastocysts as previously described (Alpha Scientists in Reproductive Medicine and ESHRE Special Interest Group of Embryology, 2011). In addition, blocked or degenerated embryos (BD), as well as those that presented with slower development with a compact morula stage (CM), were also taken into account. To study the relationship between sperm APN and blastocyst viability, blastocysts were classified as usable (U) or non-usable (NU). They were determined to be usable if they were selected for transfer or freezing for storage or non-usable if they were arrested or of poor quality and were therefore discarded.

Flow cytometry

To measure the APN levels in sperm samples, we performed semi-quantitative and quantitative flow cytometry assays. To perform the quantitative studies, we used the QuantiBRITE™ PE* kit (BD Biosciences, San Jose, CA, USA). The same semen samples were simultaneously used for both analyses.

Surplus sperm samples were fixed in suspension with 4% (w/v) paraformaldehyde (Sigma-Aldrich, St. Louis, MO, USA), centrifuged at 3500g for 6 min, and then washed in phosphate buffered saline (PBS). The samples were incubated in blocking medium (PBS with bovine fetal serum 10% (w/v); Biochrom, Cambridge, UK) for 30 min and then with a primary antibody. The human APN monoclonal phycoerythrin (PE)-labeled primary antibody, at a ratio of 1:1 (PE: Antibody) (PE Mouse Anti-Human CD13, BioLegend, San Diego, CA, USA), was diluted 1:200 in PBS and incubated overnight at 4°C. The nuclei were stained with 0.5 μg/ml Hoechst 33258 (Molecular Proves, Eugene, OR, USA). Primary antibody specificity was confirmed by using an isotype control antibody (PE Mouse IgG1κ Isotype Control Antibody, BioLegend) at the same concentration as the primary antibody.

For quantitative flow cytometry analysis, we plotted a calibration curve with the mean fluorescence intensity values obtained from four different populations of PE-conjugated beads with a known number of PE molecules per bead as provided with the QuantiBRITE™ PE* kit. Briefly, QuantiBRITE™ PE* beads were diluted in 500 μl of 1× PBS with azide plus 0.5% (w/v) bovine serum albumin (Sigma-Aldrich, MO, USA) and analyzed by flow cytometry. The fluorescence intensity values of the semen samples and the different populations of the QuantiBRITE™ PE* beads were obtained at the same time using the same settings for fluorescence and compensation.

The fluorescence data from at least 10 000 events were analyzed in a flow cytometer (Gallios™, Becton Dickinson, San Jose, USA). Blue fluorescence (Hoechst 33258) and red florescence (PE) were collected with the FL9 and FL2 sensors, respectively. To ensure that the fluorescence data were from live spermatozoa, we used a discrimination gate around the sperm population on the forward (FSC) and side (SSC) scatter plots and selected Hoechst 33258-positive events. Then, the percentage of PE-positive sperm and the mean fluorescence of the sperm samples were determined by subtraction of the background fluorescence from each histogram using the controls as references. The PE and Hoechst 33258 fluorescence results were analyzed with Summit v4.3 software.

While the quantitative flow cytometry assay determined the average number of APN molecules per spermatozoon (nM-APN), the percentage of sperm cells positive for APN (%APN) measured in each semen sample was determined by semi-quantitative flow cytometry analysis. The average number of molecules can be extrapolated from a calibration curve obtained from the populations of PE-conjugated beads, given that the PE:Ab ratio of our primary antibody was 1:1.

Statistical analysis

The %APN and nM-APN were compared with the basic sperm and embryo quality parameters. Normal data distribution was evaluated by the Kolmogorov–Smirnov test. Spearman’s rank correlation analysis was performed to analyze the relationship between sperm APN levels and basic sperm parameters, and group comparison tests and Kruskal–Wallis and Wilcoxon non-parametric tests were used to analyze differences in embryo quality and viability. A cumulative link mixed model (CLMM) analysis was carried out to evaluate the cumulative likelihood during embryo development, taking into account that all of the embryos for each patient were not independent and belonged to the same embryonic cohort. Statistical analyses were performed using IBM SPSS Statistics 22 software and the R statistical environment, and significance were considered at P-values <0.05 and <0.01.

Results

The mean age of the men included in this study was 39.54 ± 0.48 years with a range between 32 and 52 years. Ejaculated semen samples, were characterized as follows (mean ± SEM): volume, 2.97 ± 0.17 ml; sperm concentration, 78.31 ± 4.63 × 10⁶ sperm/ml; total sperm count, 207.67 ± 12.59 × 10⁶ sperm/ml; progressive motility (PR), 54.46 ± 1.93%; immotility (IM), 37.51 ± 1.67%. After sperm preparation for the ART, characteristics were as follows: volume, 0.77 ± 0.02 ml; sperm concentration, 13.55 ± 1.17 × 10⁶ sperm/ml; total sperm count, 11.67 ± 1.03 × 10⁶ sperm/ml progressive motility, 94.58 ± 0.57%; immotile spermatozoa, 4.62 ± 0.56%. Measured by flow cytometry, the %APN was 88.07 ± 0.87% and the average nM-APN was 4346.74 ± 309.14.

Because abnormal embryo and blastocyst development have been linked to poor sperm quality (Janny and Menezo, 1994), we first evaluated the relationships among the basic seminal parameters (Table I). Both nM-APN and %APN were positively correlated with the percentage of sperm with PR (P < 0.05). Consequently, a negative correlation was reported between the percentage of sperm with non-progressive motility (NP) and nM-APN (P < 0.05) and between the percentage sperm with IM and %APN (P < 0.01). As we expected, our results showed a relationship between human sperm motility and APN levels.

Table I

Relationship between human sperm APN and basic sperm parameters.

		nM APN		%APN
		R	P value	R	P value
Volume		−0.091	0.419	−0.276	0.013*
Concentration		−0.049	0.667	0.217	0.052
Motility	Progressive motility (PR)	0.253	0.023*	0.258	0.020*
	Non-progressive motility (NP)	−0.262	0.018*	0.190	0.864
	Total motility (PR + NP)	0.182	0.101	0.286	0.010*
	Immotility (IM)	−0.174	0.120	−0.286	0.010*

*

Spearman rank correlation coefficients. APN, aminopeptidase; N %APN, percentage of APN positive sperm cells; nM APN, mean number of APN molecules per spermatozoa.

Next, we aimed to evaluate the correlation between human sperm APN levels and fertilization and embryo quality after ICSI technique. The total number of recovered oocytes was 953 and the total number of mature oocytes that were donated was 843, with a mean of 10.41 ± 0.29 oocytes received per patient. The overall fertilization rate of all cycles was 72.48% and we did not find any significant correlation between either %APN or nM-APN (R = −0.042, P = 0.685 and R = 0.005, P = 0.962, respectively). After the fertilization assessment, a total of 611 embryos were analyzed, with a mean of 7.57 ± 0.28 embryos per couple. The percentage of good quality embryos (Grade A + Grade B) on Day 2 was 71.97%, while on Day 3 this percentage was reduced to 48.50%. The mean characteristics of our embryo cohort were as follows (mean ± SEM): number of blastomeres on Day 2, 3.58 ± 0.05; embryo fragmentation on Day 2, 2.32% ± 0.18%; number of blastomeres on Day 3, 6.45 ± 0.09; and finally, embryo fragmentation on Day 3, 2.77% ± 0.20%. However, our results did not show any correlation between APN levels and early embryo quality (Supplementary Fig. S3).

Taking into account that blastocysts have higher implantation, pregnancy and live birth rates (Van den Abbeel et al., 2013), we aimed to analyze the relationship between sperm APN levels and blastocyst development. Our results showed that the most well-developed blastocysts were associated with low nM-APN levels (Table II, Fig. 1), although no association was noted regarding %APN and blastocyst development (Table II, Supplementary Fig. S4). Expanding (BC), expanded (BE) and hatching/hatched (BH) blastocysts were generated with semen samples that showed lower nM-APN levels than early blastocysts (BT) and blocked or degenerated blastocysts (BD), reaching statistical significance at Day 6 of embryo development (P < 0.01) (Table II, Fig. 1B and 1C). Consistent with this finding, we also observed the same pattern when we evaluated embryo usability (Table III, Fig. 1D) by classifying embryos into two groups: (i) embryos that reached the blastocyst stage (U: usable); and (ii) embryos that were arrested, degenerated, of insufficient quality, and/or at a low development stage (NU: non-usable). We found that usable human blastocysts at either Days 5 and 6 of embryo development were generated with semen samples with low nM-APN levels (P < 0.05). As in the previous analysis, we did not detect any relationship between blastocyst usability and %APN (Table III, Supplementary Fig. S4D).

Figure 1.

Relationship between human sperm APN levels and blastocyst embryo development embryo viability. (A) Representative images of blastocyst developmental stages. Violin plots of the nM-APN distributions at blastocyst developmental stages (B) at Day 5 of embryo development and (C) at Day 6 of embryo development. (D) Violin plots of the nM-APN distributions on blastocyst usability. APN, aminopeptidase N; BT, early blastocyst; BC, expanding blastocyst; BE, expanded blastocyst; BH, hatching/hatched blastocyst; BD, blocked or degenerated embryos; nM APN, mean number of APN molecules per spermatozoa; U, usable blastocysts; NU, non-usable blastocysts.

Table II

Relationship between human sperm APN and blastocyst stage at Days 5 and 6 of embryo development.

		Embryo stage	Median	Min	Max	P-value
Blastocyst development Day 5	%APN	BD	89.62	67.08	98.88	0.768
		BT	89.70	74.99	98.2
		BC	89.35	67.08	98.88
		BE	89.62	72.15	98.2
		BH	90.59	72.15	98.88
	nM APN	BD	3963	226	10 339	0.121
		BT	4659	226	10 238
		BC	3547	226	10 884
		BE	2648	226	10 238
		BH	2063	769	10 238
Blastocyst development Day 6	%APN	BD	90.24	74.99	98.2	0.983
		BT	89.62	75.19	93.5
		BC	88.73	75.19	94.75
		BE	89.47	78.41	94.95
		BH	89.87	74.99	94.75
	nM APN	BD	5232	888	10 884	0.005*
		BT	5629	1437	9226
		BC	1951	226	10 339
		BE	2630	226	10 238
		BH	1950	888	8037

*

Kruskal–Wallis test.

APN, aminopeptidase N; %APN, percentage of APN positive sperm cells; nM APN, mean of APN molecules per spermatozoa; BT, early blastocyst; BC, expanding blastocyst; BE, expanded blastocyst; BH, hatching/hatched blastocyst; BD, blocked or degenerated embryos.

Table III

Relationship between human sperm APN and embryo usability.

		Usability	Median	Min	Max	P-value
Embryo usability Day 5	%APN	U	89.70	67.08	98.88	0.964
		NU	90.13	74.99	98.88
	nM APN	U	2648	226	10 884	0.019*
		NU	4393	226	9226
Embryo usability Day 6	%APN	U	89.75	74.99	95.95	0.947
		NU	89.31	75.19	95.26
	nM APN	U	1950	226	10 238	0.029*
		NU	4235	226	10 339

*

Wilcoxon test.

APN, aminopeptidase N; %APN, percentage of APN positive sperm cells; nM APN, mean of APN molecules per spermatozoa; U, usable blastocyts; UN, non-usable blastocyts.

In addition, our results showed a relationship between the probability of having high-quality and usable embryos in an embryonic cohort and the presence of APN in the sperm samples used to generate these embryos (Table IV). In analysis with CLMM, this mixed model allowed us to measure the cumulative probability, taking into account the intragroup effect of each embryonic cohort. The odds of having more well-developed blastocysts increased 1.38-fold (P < 0.05) at Day 5 and 1.90-fold (P < 0.001) at Day 6 of embryo development using semen samples with low nM-APN levels during the ICSI technique. In the same way, the cumulative probability of having usable embryos increased 1.48-fold when sperm samples presented lower nM-APN levels. However, there was no relationship between %APN and the likelihood of reaching the blastocyst stage.

Table IV

Probability of have high-quality and usable embryos in an embryonic cohort and the presence APN in the sperm samples.

		Odds ratio	95% CI	P-values
%APN	Embryo quality Day 2	0.81	0.61–1.08	0.15
	Embryo quality Day 3	0.96	0.74–1.23	0.76
	Blastocyst development Day 5	0.98	0.73–1.30	0.89
	Blastocyst development Day 6	1.34	0.87–2.06	0.19
	Embryo usability	1.14	0.84–1.54	0.39
nM APN	Embryo quality Day 2	1.08	0.83–1.41	0.54
	Embryo quality Day 3	1.11	0.87–1.40	0.38
	Blastocyst development Day 5	1.38	1.04–1.83	0.02*
	Blastocyst development Day 6	1.90	1.37–2.64	0.0005*
	Embryo usability	1.48	1.11–1.98	0.008*

*

CLMM: Cummultative Link Mix Models.

APN, aminopeptidase N; nM APN, mean number of APN molecules per spermatozoa; %APN: Percentage of APN positive sperm cells.

Discussion

Approximately half of fertility problems arise because of male factors, and 40–30% of men with normal parameters have an inability to achieve pregnancy. The diagnosis of male fertility is one of the major concerns in reproductive medicine. As basic semen analysis is descriptive and is a surrogate measure of male fertility with limited prognostic significance (Guzick et al., 2001), the development of new methods to evaluate semen is necessary to ensure a more accurate prognosis and diagnosis of male fertility. Our results provide, for the first time, insight into the use of human sperm APN as a potential biomarker for identifying the potential for high-quality embryos, which could add value to assessment of male fertility prognosis.

APN is a widely distributed enzyme present in seminal fluid, prostasomes and human spermatozoa. APN constitutes 0.5–1% of the seminal plasma proteins (Khatun et al., 2017), and its activity is 10–20 times higher in this fluid than in other human tissues (Subiran et al., 2008; Khatun et al., 2017). As measured by flow cytometry, 88% of human spermatozoa are positive for APN, and this enzyme is described to be involved in several processes important for the acquisition of human sperm fertilization ability, such as motility and acrosome reaction, and has a role in several seminal pathologies (Togo and Morisawa, 1997; Irazusta et al., 2004; Togo and Morisawa, 2004; Subiran et al., 2008, 2010; Viudes de Castro et al., 2015; Khatun et al., 2017, 2018). Our results confirm that APN levels are positively correlated with sperm motility, although this parameter is not the most reliable predictor of male infertility (Bosler et al., 2014).

Molecular features of sperm are suggested to be involved in fertilization and in early embryo development. In addition, the development of additional molecular tools for assessing sperm malfunctioning related to infertility may complement current diagnostic methods. Clinical evidence derived from the use of ART indicates that defective sperm contributions may extend beyond fertilization, emphasizing that early and late paternal effects may be determinants for normal embryo development (Janny and Menezo, 1994; Tesarik, 2005; Barroso et al., 2009; Govindaraju et al., 2012; Khatun et al., 2017). The early paternal effect refers to different sperm cytoplasmic deficiencies, such as centrosome dysfunction or a deficiency of oocyte-activating factors, which can be detected both in the zygote and throughout preimplantation development, before the major activation of embryonic genome expression, which begins at the 4‐cell stage in humans. However, it has been shown that the sperm-derived genome is not completely silent in the period between fertilization and the first cleavage division; limited RNA synthesis can be detected in human pronuclei, and this early transcription of paternal genes has been shown to be required for the proper assembly of nucleolar precursor bodies.

The late paternal effect refers to nuclear deficiencies, such as sperm aneuploidy, DNA damage or abnormal chromatin packaging, which are usually detected in the 8-cell stage of embryo development, when the expression of the majority of sperm-derived genes begins (Tesarik, 2005). Furthermore, numerous studies have described a large number of sperm molecules that also affect embryonic development (Govindaraju et al., 2012). However, no single parameter or battery of assays definitively describes the fertilizing competence of spermatozoa either in vitro or in vivo.

Hormone levels analyzed to date represent only poor diagnostic markers for the success of both testicular sperm extraction and ICSI (Steger et al., 2011). On the other hand, although the protamine ratio has been considered a suitable diagnostic marker for the fertilizing capacity of testicular spermatids and ejaculated spermatozoa (Balhorn et al., 1988; Steger et al., 2001; Mengual et al., 2003; Steger et al., 2003; Nasr-Esfahani et al., 2004; Mitchell et al., 2005; Steger et al., 2008; Hammoud et al., 2009), its clinical value may be limited, especially for ICSI, as protamine levels have been reported to vary among individual spermatozoa within one ejaculate (Aoki et al., 2006).

To circumvent the variability among different ejaculates, the molecular analysis in our study was performed in the same sperm sample used for ART, in contrast to other approaches. This descriptive study was also performed with an oocyte-donation program to avoid introducing bias concerning the oocytes, strengthening our results and identifying relevant male factors with greater confidence. Based on this experimental design, our results indicate that blastocyst development is related to sperm APN levels. Low levels of APN are associated with the well-developed blastocysts, which are known to have higher implantation, pregnancy and live birth rates (Van den Abbeel et al., 2013). These findings are consistent with the fact that high APN activity was previously reported to have a significant inhibitory effect on embryonic development, possibly due to the induction of oxidative stress and DNA damage in spermatozoa (Khatun et al., 2017; 2018). In silico analysis revels that APN interacts with proteins and small molecules involved in several biological processes, such as cell adhesion, oxidative stress response, apoptosis and DNA integrity, some of which are extremely important in the regulation of normal sperm functions (Khatun et al., 2017). Among these, ADAM17, a putative inducer of germ cell apoptosis, and tumor necrosis factor, stand out. The inhibition or stimulation of APN in male germ cells could alter both molecules, which could induce the apoptosis process (Khatun et al., 2018) or damage sperm DNA (Meseguer et al., 2006; Garrido et al., 2008). Hence, these APN-derived abnormalities could be included within the late paternal effect.

In addition, altered APN activity affects protein tyrosine phosphorylation and markedly increases reactive oxygen species generation in sperm. This APN-derived cellular toxicity causes persistent damage in spermatozoa, and although they are still capable of correctly fertilizing oocytes, damaged sperm can lead to several adverse consequences that affect the quality and viability of the embryo (Khatun et al., 2018). In this sense, spermatozoa that carry a lower number of APN molecules on their surface have an increased chance of generating usable blastocysts with the greatest implantation potential (fully expanded and hatching blastocysts) than spermatozoa with a high number of APN molecules. Therefore, our findings indicate that high levels of sperm APN may affect the quality and viability of blastocysts.

Nowadays, clinical pregnancy rates following ICSI treatments is ∼30% per embryo transfer (Nyboe Andersen et al., 2009; Bühler et al., 2010). In this regard, the selection of unsuccessful spermatozoa contributes to failure rates (around 70%), leading to emotional, medical and financial consequences for the couple. Any molecular parameter able to estimate the chances for successful fertility treatment, therefore, could help embryologist and patients in counseling and treatment. By using an unprecedented approach that involves the use of the cumulative link mix model test, our results show that the cumulative probability of having more well-developed blastocysts as well as usable blastocysts increases by 1.5-fold when semen samples with lower APN levels are used during the ICSI technique. This method provides a highly robust test to evaluate the male contribution to embryo development, as it measures the cumulative probability taking into account the intragroup effect of each embryonic cohort.

Conclusion

Therefore, the evaluation of human sperm APN may provide useful complementary information to embryologists for identifying of high-quality embryos during ICSI treatments. By using flow cytometry assays, the measurement of APN levels as an alternate tool constitutes an affordable approach in terms of time and cost-effectiveness, as it possible to evaluate the APN levels in the same sperm sample fractions used for ART in <1 h. This would be critical for the final diagnosis when seminal parameters are close to the threshold values. In the case of suboptimal APN scores, we suggest obtaining a larger cohort of oocytes in these patients by accumulating vitrified oocytes over several stimulation cycles, as the number of usable generated embryos will be probably reduced. Further clinical studies will be required to confirm the use of APN as a prognostic factor, as this molecular approach could be a useful tool for couples for whom the number of usable blastocyst is critical for ART success.

Data Availability

The data underlying this article are available in the article and in its online supplementary material.

Authors’ roles

M.G. designed and conducted the experiments, analyzed the results and wrote the paper; I.M.-H. and I.U.-A. conducted the experiments; G.L. collaborated in the analysis of the results; Z.L. provided the samples and images, and participated in critical discussion; N.G. participated in critical discussion; J.I. participated in manuscript drafting and critical discussion and secured funding; N.S. designed the experiments, analyzed the results, wrote the paper and participated in critical discussion.

Funding

This research was funded by grants from the Basque Government (GIC12/173; to M.G. and I.M.-H.) and University of the Basque Country (UPV/EHU; to M.G. and I.U.-A.) (EHUA14/17).

Conflict of interest

The authors declare no conflict of interest.