Differential proMMP-2 and proMMP-9 Secretion in Human Pre-implantation Embryos at Day 5 of Development.

Background: The most commonly used non-invasive criterion for evaluating the probable success of transferring in vitro human embryos for implantation is their morphological development. With this criterion, however, embryos in cellular arrests go unnoticed. Extracellular matrix metalloproteases type 2 (MMP-2) and MMP-9 are key markers of embryonic development and the implantation process, according to various animal studies. The current study investigated the proMMP-2 and proMMP-9 expression in the culture media developing human embryos that were transferred for implantation. Methods: Forty-two patients were accepted in the Department of Reproductive Biology of a Hospital in México City, based on the Institutional inclusion criteria for in vitro fertilization. On day 5 of development, embryos were transferred to women, and the culture medium was stored at -70 to await assessment of the activity of proMMP-2 and proMMP-9 in substrate gel zymography. Results: The patients showing embryo sac development were assigned to the pregnant group (n =17) or non-pregnant (n =25). In both groups, the activity of proMMP-2 and proMMP-9 was evaluated in substrate gel zymography. Our results indicate for all 17 women able to achieve a full-term pregnancy, the activity band of proMMP-2 was found in the corresponding culture medium. For 11 of them, the band of proMMP-9. Regarding the other 25 patients, the expression band for proMMP-2 detected in 3 and that proMMP-9 in 11 individuals. Conclusions: On day 5 of embryo development, the evaluation of proMMP-2 and proMMP-9 in the embryo culture medium is a reliable indicator of embryo quality and capacity to establish pregnancy.

Regarding the evaluation of embryo quality prior to implantation, there is as yet no quantitative method.
Embryonic morphology is the criterion employed in clinical practice as a qualitative marker of the viability of embryos to be transferred to patients [32,33]. However, this method has not shown an acceptable clinical correlation, since even aneuploid embryos can have normal morphology, and some euploid embryos can have aberrant morphology [34]. Thus, the aim of present study, was to analyze the culture medium of embryos with good morphological development in order to determine whether signi cant differences exist in the expression of proMMP-2 and proMMP-9 between the groups of embryos with successful and implantation.

Ethics statements
The current protocol was reviewed and approved by the Ethics and Research Committees of the Instituto Nacional de Perinatología in México City (212250 − 22661). The purpose of the study was explained to all patients, and informed consent was signed by those who decided to participate.

Patients
Forty-two patients diagnosed with infertility were admitted to the Department of Reproductive Biology (Instituto Nacional de Perinatología) for in vitro fertilization. The following constituted the inclusion criteria: ≤37 years of age, a regular menstrual cycle, a normal uterine cavity con rmed by hysteroscopy, an absence of intrauterine adhesion or in ammation, an endometrial thickness in the late follicular phase of ≥ 7 mm measured by ultrasonography, a normal ovarian reserve (follicle-stimulating hormone < 9.0mU/mL), a normal ovarian response to the stimulation protocols (> 8 oocytes retrieved in a controlled ovary hyperstimulation cycle), and the lack of hormone (estradiol/progesterone) treatment during the endometrial cycle.

Hormonal stimulation of patients
The patients received controlled ovarian stimulation with FSH/LH. When the follicular diameter measured 18 mm, oocyte maturation was stimulated with human chorionic gonadotropin. Thirty-six hours later, follicular oocytes were obtained using ultrasound guidance.

In vitro fertilization and developing embryos
Oocytes were fertilized in vitro by exposing them to 1x10 6 capacitated spermatozoa/mL for 18 hours. This process was carried out in HTF/HEPES medium supplemented with 5% albumin under controlled conditions (37°C, 54% CO 2 and 95% air). Fertilization was assessed by the presence of a second polar corpuscle body. Oocytes were each maintained in 50 µL of G-1 PLUS culture medium (Sweden). Morphological development was observed daily up to the 36-cell stage (5 day), at which time the culture medium was retained for matrix metalloproteinase analysis.
On day 5 of embryo development two embryos (with quality type I, II, or III quality) were transferred to the uterine cavity with Flexible Pass intrauterine transfer cannula based on the Soft Cook technique. The process was assisted by abdominal ultrasound guidance and a real-time, 5-MHz sector electronic array endovaginal test (Philips Epiq CVx; MO, USA).

Blood samples
Five milliliters of peripheral blood were obtained from the patients by puncture of the cephalic vein, placed in EDTA-K2 tubes (BD Vacutainer), and centrifuged at 14,000 rpm for 10 minutes. Serum was collected in Eppendorf tubes and stored at -70°C until to quanti cation hormones assay.
Assay to determine the levels of sex hormones All evaluations of hormones were performed in the central laboratory of the Instituto Nacional de Perinatología on the Modular Analytical apparatus cobas e 411 (Roche, USA). A commercially available assay kit was used to measure the serum levels of P4, E2, T4, FSH, LH, AMH, and hCG (Roche system, USA), with according to the manufacturer´s recommendations and as previously described [35][36][37]. The lower limit of detection for these hormones was 0.4 ng/mL, 5.0 pg/mL, 0.025 pg/mL, 0.100mIU/mL, 0.100 mIU/mL, and 0.2 ng/mL, 0.1 mIU/mL, respectively. The intra-assay coe cient of variation was 3%, 5%, 5%, 3%, 2%, 3%, and 5% respectively.

Determination of pregnancy outcome
The serum concentration of P4, E2, T4, FSH, LH, AMH, and hCG was assessed 14 days after embryo transfer. The apparent successful implantation of the embryo and endometrial receptivity were con rmed by ultrasound, embryo sac development. In accordance with the nding of this analysis, the patients were assigned to one of two groups: with and without implanted embryos (n = 17, pregnant and n = 25, nonpregnant patients, respectively).

Protein quanti cation and matrix metalloproteinase activity
The concentration of proMMP-2 and proMMP-9 proteins in the culture media, indicating their secretion into the media by the embryos, was measured in the two groups by the colorimetric Bradford method [38] and activity using SDS-polyacrylamide gels with porcine gelatin (1 mg/mL), as described previously [39].
The internal control of electrophoretic mobility was a used a culture medium from U937 promyelocyte cells (ATCC, Manassas, VA, USA). Each sample was loaded with 0.75 µg of protein and the activity band was captured with the EpiChemi Darkroom gel documentation system (UVP, CA, USA). Optical densitometry was quanti ed on the NIH ImageJ program.

Statistical analysis
Difference between the two groups of embryos (implanted vs. non-implanted) base on the mean optical density values of the corresponding culture media, were examined with the Student´s t-test. All values are expressed as the mean ± SD, and statistical signi cance was considered at p ≤ 0.05. Statistical analysis was performed on GraphPad Prism version 8.0 (GraphPad Software, San Diego, CA, USA). STATA software was used to plot the receiver operating characteristic (ROC) curves and calculate the area under the curve (AUC-ROC) (StataCorp LLC; v16, TX, USA). The cutoff value selected (based on the ROC curve) was that at which the sensitivity and speci city were best and the distance to the top-left corner of the ROC curve was the least. Sensitivity, speci city, and positive and negative predictive values were determined on the STATA software.

Results
Patient characteristics Table 1 shows the characteristics of the 42 patients were compared between the two groups: pregnant (n = 17); and 2) non-pregnant (n = 25) patients. No signi cant difference existed in regard to any of these parameters: age (p = 0.23), body mass index (p = 0.43), and years of infertility (p = 0.39).  Data are presented as the mean ± standard deviation.
Expression of proMMP-2 and proMMP-9 in the culture medium of the embryos.
On day 5 of development, the embryo was transferred to the patients and the secretion pro le of proMMP-2 and proMMP-9 was determined in the culture medium (Fig. 1).
The presence of proMMP-2 was detected in the culture media corresponding to 100% of the pregnant patients (17 of 17; lane 1 to 14) and in 11% (3 of 25; lane 15 to 57) of the non-pregnant patients (Fig. 1).
The latter group included three women who became pregnant and underwent a spontaneous abortion (P52, P57, and P58; Fig. 1). The optical density of proMMP-2 quanti ed for each of the activity bands (Fig. 2). Was signi cantly (1.4-fold) lower in the three non-pregnant patients with proMMP-2 expression than in the pregnant women ( Fig. 2; p = 0.045).
On the other hand, proMMP-9 expression was found in the culture media corresponding to 11 of 14 pregnant and 39% of non-pregnant patients (11 of 25). The optical density of the bands of proMMP-9 showed a signi cantly (1.2-fold; Fig. 2) lower value in the culture media corresponding to the 11 nonpregnant patients with proMMP-9 expression versus the culture media corresponding to the 11 pregnant women with proMMP-9 expression ( Fig. 2; p = 0.002).
Predictive values of proMMP-2 and proMMP-9 in the culture medium of developing embryos from pregnant patients.
The ROC curve was used to evaluate the sensitivity and speci city of proMMP-2 and proMMP-9 for determining the capacity of transferred embryos to result in pregnancy (Fig. 3). For proMMP-2, the optical density of 423 was taken as the cutoff value and showed statistical signi cance (p = 0.02) which exhibited a sensitivity of 100% and a speci city of 100%. For proMMP-9, the optical density of 550 was adopted as the cutoff value and showed statistical signi cance (p = 0.0035) having a sensitivity of 81.8% and a speci city of 72.7%.

Discussion
MMPs play an important role in the remodeling of different structural and support components during ovulation [40,41] decidualization [42,43], and implantation [44][45][46]. The main ndings of the present study can be summarized in ve points in relation to the activity of proMMP-2 and proMMP-9 in the culture media of the embryos on day 5 of their development 1) the activity of proMMP-2 was found in 14 of 14 culture media corresponding to the patients who achieved a full-term pregnancy and in 3 of 25 culture media associated with women who did not have this outcome. It was 1.4-fold greater in the former group; 2) The activity of proMMP-2 was observed in 11 of 14 culture media corresponding to the patients who carried their pregnancy to term and in 11 of 25 culture media associated with women who did not have this outcome. It was 1.2-fold greater in the former group (Fig. 1); 3) In the three patients who did not carry their pregnancy to term, only proMMP-9 activity was detected; 4) the activity of both proMMP-2 and proMMP-9 was identi ed in three non-pregnant patients who had some pre-pregnancy complications (Fig. 1); and 5) There were no signi cant differences in the concentration of the hormones HCG, E2, or P4 between pregnant and non-pregnant patients (Table 1). Gu et al. (2015) reported the concentration of the active form of MMP-9 at 0.698 ± 0.022 ng/mL in the culture medium of developing human embryos, which resulted in pregnancy for 77.0% of the participating patients [47][48][49]. According to the present study, the activity of proMMP-2 and proMMP-9 (Fig. 2) was signi cantly higher in the culture media of the embryos yielding a full-term pregnancy than in the culture media of the other cases ( Table 1). The culture media exhibited a 1.4-fold greater proMMP-2 activity (p = 0.045) for the embryos generating a full term pregnancy compared the other cases (3/25) showing proMMP-2 activity. Similarly, the culture medium displayed a 1.2-fold greater proMMP-9 activity (p = 0.002) for the embryos engendering a full term pregnancy (11/14) compared to the other cases (11/28), considering the media with proMMP-9 activity.
A conceptual model is herein provided (Fig. 4) how MMPs are activated by epidermal growth factors [47], interleukin (IL) -1β, and tumor necrosis factor (TNF)-α [50,51]. Sequeira et al. (2015) reported a signi cantly (15.4-fold) greater value for the level of IL-1β in the culture media of developing human embryos that became implanted in patients versus the culture media of non-implanted embryos (0.55 ± 0.25 pg/mL), nding implantation in 42.0% of the total participants [52].
In the rst and second trimesters of pregnancy, syncytiotrophoblast cells have been described to secrete a 2.4-and 3.8-fold higher amount of IL-1β, respectively, than the pre-pregnancy level. The secretion of IL-1β is associated with an increase in MMP-9 expression [53]. After interacting with its receptor, IL-1β regulates the signaling pathway that involves the activation of the mitogen-activated protein kinase (MAPK), p38 MAPK, c-Jun N-terminal kinase (JNK), and the extracellular regulatory kinase (ERK) [54,55]. As a consequence, IL-1β promotes the activation of nuclear factor kappa-beta (NFkβ) and the expression of The collagenolytic activity of MMPs is regulated by the speci c tissue inhibitors of these proteinases [49,57]. Cytotrophoblast cells, treated with 50 nM of their tissue inhibitor, known as tissue inhibitor of metalloproteinase-2 (TIMP-2) exhibit an up to 40% reduction in invasiveness according to Librach (1991) [51,58]. The present results showed a 1.4-fold and 1.2-fold decrease in the activity of proMMP-2 and proMMP-9 respectively (Fig. 2), in the culture media corresponding to the cases of patients who were pregnant but did not carry to term. However, the expression of TIMPs was not herein evaluated in the culture medium of developing embryos. It would be interesting to determine the MMP/TIMP relationship is involved in the mechanism responsible for regulating the progress of implantation and pregnancy.
Recently, localized polymorphisms identi ed in the promoter region of MMP-2 (-1306 C/T; rs 243865) and MMP-9 (-1562 C/T; rs 3918242) were found to induce changes in the levels of transcription and or expression of the protein. These mutations have been proposed as a risk factor for spontaneous abortion [50,59]. Regarding the three patients of the present study with implanted embryos that spontaneously aborted (P52, P57, and P58), the corresponding culture media displayed proMMP-2 and proMMP-9 ( Fig. 2) activity. Future research should explore the possible relation of such pregnancy complications to mutations.
MMP-2 has the ability to degrade bronectin, elastin, and collagen type IV, V, VII, and. In contrast MMP-9 degrades proteoglycans, elastin, and collagen I, IV, V, and XI [51,[60][61][62] which allows the cytotrophoblast cells to invade the endometrium to prepare the way for implantation) [63] (Fig. 5). During the implantation window, according to in vivo (in animal models [64] and in vitro studies [65], the epithelial cadherin-like binding protein (E-cadherin) enables the embryo to adhere to the endometrial epithelium, which is degraded by MMP-9 [66]. Hence, previous reports evidence a key role by MMP-2 and MMP-9 in embryonic development. The present analysis based on the cutoff points of the ROC curve for both proteinases suggests that their evaluation in the culture medium of developing embryos has plausible predictive power in relation to the success of implantation.

Conclusions
These results of the current contribution demonstrate the feasibility of detecting proMMP-2 and proMMP-9 activity in the culture medium of embryos on day 5 of development by using in gel zymography. Additionally, such activity is associated with the capacity of embryos for implantation. Thus this evaluation could serve as a non-invasive method for determining the viability of human embryos developed in vitro conditions. Abbreviations BMI: Body mass index; IVF: in vitro fertilization; MMP: matrix metalloproteases.

Declarations Ethics approval and informed consent
Each patient was informed that, by agreeing to participate in the study, the culture medium of the embryo would be used to perform MMPs assays after embryo transfer, and that this procedure would not affect the development of the embryos. All candidates signed the consent form after the explication. The present protocol was reviewed and approved by the Ethics and Research Committees of the Instituto Nacional de Perinatología (212250-22661).

Consent for publication
All authors carefully read the nal version of the manuscript and gave their permission to submit to the journal of Reproductive Biology and Endocrinology.

Availability of data and materials
All of the relevant information from the study is described in the manuscript.

Competing Interest
All authors declare that they have no competing interests with respect to the research, or the authorship publication of this article.

Funding
The current study was supported by a grant (212250-22661 assigned to HFH) from the Instituto Nacional de Perinatología "Isidro Espinosa de lo Reyes" of Ciudad de México, México. The institute was not involved in any stage, and therefore has no con ict of interest with the content of the manuscript.
Authors´ contributions MOV performed the quanti cation of hormones in the serum of the patients. MOV and RJAG carry out the evaluating of MMP activity. RJAG and EIC performed in vitro fertilization, obtained the culture medium for the development embryos of the embryos. RJAG monitoring pregnancy and gestational sac formation. PVC performed the statistical analysis of the ROC curve, and participated in the discussion of results and writing of manuscript. MOV, RJAG, JSLC and JLC participated in the analysis and discussion of the results. MOC performed the analysis of MMP activity and participated in the writing of the manuscript. HFH participated in the design of the study, analysis of results, and writing the manuscript, as well as obtaining nancial and material support for the study.    Predictive values of the ROC curves. The cutoff value selected (based on the ROC curve) was that at which the sensitivity and speci city were best for proMMP-2 and proMMP-9 as biomarkers of the success of implantation. Figure 4