Natural fluids collected from the reproductive tract of women can be used as additives for human in vitro embryo culture media

doi:10.21203/rs.3.rs-53612/v1

Reproductive fluids have been recently proposed as the most adequate natural additive to complement the chemically defined in vitro culture media in different animal models, since they contain the full range of molecules necessary for the fertilization and development of preimplantation embryos. However, reproductive fluids potentially are problematic in terms of biosecurity and reproducibility of results, and their availability is scarce. Therefore, we have optimized the protocols to characterize the physico-chemical and functional properties of human reproductive fluids to establish the security ranges that could allow their use as natural additives and using the least possible volume of fluids collected. The results of this study offer, for the first time, a complete quality control guideline for human reproductive fluids, including the security levels (referred from physiological values) that should be taken into account for further use – i.e., osmolality, pH, total protein concentrations, presence of endotoxins and sterility conditions. In addition, a functional test has been developed evaluating the in vitro growth of bovine embryos in medium with human reproductive fluids. Finally, as a proof of concept, four embryos cultured with uterine fluid of their own mothers were transferred and two resulted in successful pregnancy and delivery of healthy babies. In conclusion, this study paves the way towards the future use of more natural culture media in human ART.

Sexual & Reproductive Medicine

Human reproductive fluids

culture media

quality control sheet

osmolality

pH

protein concentration.

Several epidemiological studies have shown an increase in the incidence of disorders in both female and male reproductive functions that prevents a successful conception (Dyer et al., 2016). Over the last 20 years, the Assisted Reproductive Technologies (ART) have developed worldwide to assist the growing number of patients who request these methods to conceive (Adamson et al., 2011). With these techniques, it is possible today to fertilize oocytes and culture the resulting embryos until day 5, when the respective transfer to a receptive womb is performed. However, in parallel, compelling evidence is suggesting the possibility that ART conceived offspring have a greater risk of short and long term health problems as a consequence of the stress on the gametes or embryos along the different steps of the culture protocols (el Hajj and Haaf, 2013), and one of the major stressors proposed is the culture medium composition (Fernández- Gonzalez et al., 2004; Chen et al., 2015). Indeed, the impact of the culture media on embryo development has been widely studied and has resulted in different strategies to achieve the most adequate formulation (Sunde et al., 2016).

One of these strategies consists of adding reproductive fluids (RF) to the embryo culture media, to mimic as much as possible the environment of the female reproductive tract. Although not yet tested in humans, reproductive fluids have been used successfully in two major livestock species, the pig and the cow, providing the birth of healthy offspring (Lopes et al., n.d.; Oller., 2020). Thus, on the one hand, the addition of reproductive fluids in these mammalian models even in small volumes (1% of the total volume of the culture media) promoted improvements in terms of number of cells per blastocyst, developmental kinetics of the embryos, cryotolerance and DNA methylation and gene expression patterns (Lopera-Vasquez et al., 2015; Canovas et al., 2017). On the other hand, healthy live offspring have been produced in the pig (74 animals) and cow (18 animals) and remain as a study colony in the Veterinary Farm of the University of Murcia, Spain, with their corresponding controls, to investigate their mid and long term phenotypic traits in terms of metabolism, cardiovascular health, reproductive capacity, as well as their possible imprinting and epigenetic disorders(Ivanova et al., 2020).

In humans, before using reproductive fluids (HRF) as additives in future clinical trials, it is necessary to establish a biobank of these fluids, while at the same time fulfilling the sanitary and legal requirements. Furthermore, to guarantee that their addition is made under strict conditions of biosecurity and that they are beneficial for the embryo, it is important to characterize their functional and physico-chemical properties. It is known that several factors such as oxygen, temperature, pH, osmolality or molecular composition of the embryo culture media may influence the morphology, developmental kinetics, epigenetic profile, physiology and metabolism of human embryos growing in an in vitro fertilization (IVF) laboratory (Gardner and Kelley, 2017). Therefore, it is crucial to establish the security ranges for all these parameters in HRFs, especially oviductal (HOF) and uterine fluids (HUF) in which the gametes and embryos develop naturally. The collection of these fluids has been reported previously from healthy patients by ex vivo aspiration of their fallopian tubes, as well as by in vivo aspiration of the uterine cavity (Canha- Gouveia et al., 2019).

However, it is known that there is a loss of fertility with aging mainly due to the decrease in the gamete’s quantity and quality (Baird et al., 2005). This decrease is more noticeable from the mid to late 30s and more profound afterwards, until menopause, which occurs at a median age of 50 years (Steiner and Jukic, 2016). Therefore, since the age of women in Canha-Gouveia´s study was above 30 years old, besides studying the physico-chemical and functional properties of HOF in these women it also is important to compare them with younger women such as oocyte donors attending fertility clinics, and under in vivo conditions. Unfortunately, for HOF, collection in vivo it is not possible in women due to the risk of irreversible damage during manipulation of the fallopian tubes, so only ex vivo samples can be studied.

It is also necessary to assure that HRF does not contain pathogenic agents that could represent a risk of disease transmission. To this end, the protocols to quantify the endotoxin levels in the HRF and to evaluate the environmental microbiological contamination during their collection and storage must be developed. Culture media are usually sold as “endotoxin-free” or certified with endotoxin levels ≤ 0.1 EU/ml, and the wider accepted protocol to assess endotoxin levels is the Limulus amebocyte lysate (LAL) test, originated from the work of Bang and Levin (Bang, 1956; Levin and Bang, 1964, 1968). Since there is no information about the endotoxin ranges of HRF, it is imperative to establish suitable methods to perform this quantification. Similarly, exhaustive controls must be performed on the process of HRF collection and storage to guarantee the lack of environmental contamination and, once again, the protocols remain to be defined.

With these objectives in mind, the present study aimed to establish and optimize the protocols, using the least possible volume, to compare the physico-chemical properties (osmolality, pH and total protein concentrations) in HUF collected in different conditions (HUF in vivo vs. HUF ex vivo) and HOF collected ex vivo. To evaluate the functional properties (embryo assay) of the HRF, only sterile and endotoxin free HUF from in vivo samples were used. Finally, a preliminary proof of concept was developed by transferring four embryos cultured with a small proportion of HUF from their own mothers, two of which developed to term and gave birth to healthy children. With the information provide here, a new natural additive has become available for use in a growing industry that involves millions of patients, and thousands of fertility clinics, donors and donor banks, as well as fertility drugs and fertility doctors.

Ethics statement

This study was approved by the Ethics Review Committee of CEIC University Clinical Hospital Virgen de la Arrixaca (HCUVA), Murcia, Spain (Approval No. EST: 04/16). All the women who accepted to participate in the study provided their written informed consent. The ex vivo collections were performed in women who underwent a planned abdominal hysterectomy plus bilateral salpingectomy at the Service of Obstetrics and Gynaecology of the HCUVA from January 2016 until June 2018, fulfilling the following inclusion criteria: premenopausal women with no hormonal treatment during the three months before surgery, and with a previous diagnosis of uterine fibroids leiomyomas who needed surgical treatment. The in vivo collection was performed in women under 35 years old participating in the oocyte donation program at the IVI-Murcia clinic, between January 2017 and January 2018, who accepted to participate in the study. All the donors were subjected to the preceptive serological controls applied routinely in these cases, including Hepatitis B HBs Ag (surface antigen), Hepatitis C HCV Ac (total antibodies), HIV Ac / Ag (antigen and antibodies), Treponema pallidum Ac (antibodies) and Hepatitis B HBc (core) Ac (core antibody). In addition, 8 more samples of HUF were collected in vivo from patients attending the clinic accepting to participate in the proof of concept “Validation of Addition of Uterine Fluid to Human Embryo Culture Medium” -1607-MUR-055-JL/ approved on 08/21/2019 and registered on ClinicalTrials.gov of the U.S. National Library of Medicine with the identifier NCT03436758. The HRF were collected following standard operating procedures in accordance with Directive 2004/ 23 /EC of the European Parliament and of the Council of March 31, 2004 concerning human blood and its components, Law 14/2007, of July 3, of Biomedical Research, and Royal Decree of Biobanks 1716/2011, of November 18.

HUF collection

The non-diluted HUF was obtained with an adapted Mucat device (CDD Laboratorie) to be easily introduced into the uterus under ex vivo conditions from the patients who underwent a planned abdominal hysterectomy (HUF ex vivo) and under in vivo conditions from oocyte donors after transvaginal oocyte retrieval. Once introduced, the aspiration of the fluid was performed with the integrated plunger as previously described (Canha-Gouveia et al., 2019). For the women acting as embryo recipients, their own HUF in vivo was obtained by uterine flushes (being therefore diluted fluid), during the secretory phase of the menstrual cycle, by introducing an ‘embryo-transfer’ cannula (Delphin cannula, Gynetics Medical Products, NV) connected to a 10 mL syringe into the uterine cavity with 400 µl of embryo culture media. Uterine flushes were performed by manual vacuum application with the syringe. Only those samples showing no presence of blood were considered for analysis. Samples were expelled into standard cryogenic tubes and immediately frozen at -80 °C until processed.

HOF collection

The oviductal fluid (HOF ex vivo) was collected from fallopian tubes after the internal genitalia were removed by abdominal surgery. Once dissected, fallopian tubes were clamped in both extremities to avoid sample waste. With an ascendant manual mechanical pressure between the extremities, the oviductal fluid that accumulated at the upper portion of the ampulla was aspirated through the Mucat device.

Once the fluids were collected in EDTA K2 (1.8 mg/mL) tubes, they were immediately centrifuged at 7000 g for 15 min at 4 °C to remove cell debris, and the supernatant was aliquoted and frozen at −80 °C until analysis. All the collected samples were stored at the reproductive fluid collection of Biobank-Mur (Biobanco en Red de la Región de Murcia, PT13/0010/0018; PT17/0015/0038, integrated in the Spanish National Biobanks Network, B.000859). All the sample collections, either in vivo or ex vivo, only allowed the collection of a small volume (13,9µl – 96,3µl for HUF and 8,9µl – 38,1µl for HOF) of fluid (table 1). Therefore, it became necessary to optimize the protocols to establish the corresponding security ranges using the minimum effective volume.

Osmolality

To evaluate the osmolality, the minimum volume needed was set at 10 μl. The quantification was performed using the Vapro® 5520 (Wescor®) osmometer. This type of osmometer allowed the determination of osmolality in mOsm / kg by means of ebullioscopy, which is suitable for microsamples. Osmolality was determined in 17 samples of HUF from oocyte donors (in vivo group), and in 9 and 10 samples, respectively, of HOF and HUF from patients (ex vivo group). In addition, the osmolality of the 8 samples of diluted HUF from the embryo recipients was also determined.

pH

Most of the equipment used to measure the pH requires a minimum volume of 100 µl, which is not usually reached during the collection of reproductive fluids. to reduce the volume of sample used and obtain a more accurate measurement, a specific electrode for microsamples was used ("pH OxyMini" from PreSens, Germany). To evaluate the pH of the collected samples, 20-30 μl of each thawed specimen in Eppendorf tubes were immersed in ice. The probe was introduced into the tube, the pH was measured and a graph with the values was made, so that if the pH value that remained stable after 2-4 minutes was taken as valid. The pH was determined in 17 samples of HUF from oocyte donors (in vivo group), and in 12 samples of HOF and HUF from patients (ex vivo group). In addition, the pH of the 8 samples of diluted HUF from the embryo recipients was determined to confirm if it was under the established security levels.

Sterility

Samples of reproductive fluids were diluted to 1% in sterile saline and analyzed by the BACT / ALERT® 3D microbial detection system after 7 days of incubation. A total of 15 samples of HUF from oocyte donors (in vivo group) and 8 samples of diluted HUF from the embryo recepients were assessed for sterility conditions.

Endotoxin Quantification

The endotoxin quantification was performed using the Endosafe® kit (Charles Rivers®), with a maximum sensitivity of 0.005 EU/ml. This test is routinely used in human plasma samples but not in HRF and it was necessary to calibrate the apparatus with different dilutions and to ask for the company’s advice (Charles River ®, France) before establishing the final protocols. A total of 15 samples of HUF from oocyte donors (in vivo group) and 8 samples of diluted HUF from the embryo recipients were assessed.

Total protein concentration

For the quantification of total proteins, the commercial kit Thermo Scientific Coomassie (Bradford) Protein Assay Kit (Cat. No 23200) was used. After thawing the samples in Eppendorf tubes, 30μl aliquots of each diluted sample were prepared in 1.5 ml of the Coomassie Reagent (Coomassie dye G-250, methanol, phosphoric acid and water). After gently mixing, the solution was incubated at room temperature for 5 min. Then, the samples were transferred to cuvettes that were introduced in a spectrophotometer capable of measuring the absorbance at 595nm. The curves obtained were compared with previously prepared standard curves using the standard BSA protein. A total of 9 samples of HOF and 15 samples of HUF from oocyte donors (in vivo group) and 8 samples of diluted HUF from the embryo recipients were assessed by this method

Hemoglobin concentration

For the measurement of this parameter, the HemoCue® Plasma / Low Hb photometer was used. The first step was the thawing of the samples, after which controls were used to calibrate the apparatus. Once the samples were thawed, 10 µl of each were transferred into microcuvettes. Each microcuvette was introduced inside the photometer for reading, the result appearing on the screen after min. A total number of 10 HUF and 9 HOF samples were assessed by this method.

In vitro culture of bovine embryos (2-cell Bovine Embryo Assay, BEA)

To characterize, at a functional level, the HRF as supplements for embryo culture media, a bovine embryo assay was established. In vitro maturation of bovine oocytes from ovaries obtained in a slaughterhouse was done following the protocol described by Lopes et al., (Lopes et al., 2019). The COCs previously collected were washed twice in maturation medium, consisting of TCM-199 (with Earle's salts), supplemented with 26.2 mM sodium bicarbonate, 0.2 mM sodium pyruvate, 2 mM glutamine, 50 mg/mL gentamycin, 10 IU/mL equine chorionic gonadotropin (Foligon; Intervet International BV, Netherlands) and 10 IU/ mL human chorionic gonadotropin (Veterin Corion; Divasa Farmavic, Spain), divided in groups of 30-50 and transferred to four-well dishes (Nunc®, cell culture tested) in a total volume of 500 µl, with 10% (v/v) fetal bovine serum. Thirty minutes before IVF, COCs were rinsed twice and transferred to Fert-TALP medium supplemented with 2 mg/mL heparin and 50mg/mL gentamycin.

Frozen semen from two bulls with known fertility was used. Thawing was performed in a water bath at 38ºC for 30sec, and the semen was transferred to a 15 mL Falcon tube with 10 mL Sperm- TALP medium supplemented with 50 mg/mL gentamycin, and incubated at 38C for 10 min. Sperm was centrifuged at room temperature at 700 X g for 3 min and the supernatant was discarded. Sperm motility and concentration were assessed, and sperm was diluted in Fert-TALP prior to insemination at a final concentration of 1 million spermatozoa per milliliter. IVF took place at 38.5ºC in a humidified atmosphere with 5% CO2. Eighteen to 20 h post-insemination (hpi), potential zygotes were put in a 15mL Falcon tube with 2mL of DPBS and vortexed for 3 min. Presumptive zygotes were washed twice in DPBS and three times in culture medium (synthetic oviduct fluid Holms [HSOF]) supplemented with BSA before transferring them to 25 ml microdrops of embryo culture medium containing (experimental group) or not (control group) 1- 5% of HUF, covered with paraffin oil (NidOil™, Nidacon, Sweden), with a maximum load of 25 embryos/microdrop. Incubation took place at 38.5C, 5% CO2, 5% O2 for 48 hours. After in vitro culture, the proportion of presumed zygotes that reached the 2-cell stage was quantified. The percentage of bovine embryos at the 2-cell stage on day 2 of embryo culture was considered acceptable to approve the sample. A total of 14 samples of HUF from oocyte donors (in vivo group) and 8 samples of diluted HUF from the embryo recipients were assessed by BEA.

Proof-of- concept

Eight patients were accepted to participate in the proof of concept “Validation of Addition of Uterine Fluid to Human Embryo Culture Medium” -1607-MUR-055-JL/ NCT03436758. The HUF was collected as previously described one month before the scheduled IVF cycle. Women were always at day 2 of their cycle. After oocyte collection and ICSI, half of the zygotes from the same patient were cultured in medium Gems without HUF (control group) and the other half with 10% mother's diluted uterine fluid (autologous culture). The percentage of 10% was established after 5 previous quantifications of the albumin concentration in pure vs. diluted HUF: in all of them it was shown that the proportion of albumin was reduced approximately 10 times, so that 10% of diluted HUF should, theoretically, be equivalent to 1% of pure HUF. This proportion of 1% is what has been used in the experiments performed with animal models (Canovas et al., 2017; Lopera-Vasquez et al., 2017). All patients underwent ICSI to reduce the biases derived by the technique used as well as to obtain the largest number of embryos for the study. The embryonic quality and kinetics of each embryo was blindly evaluated on day 5 of culture. Only the best quality embryos according to morphological criteria were transferred. The number of embryos successfully implanted, and the number of infants born were registered.

Statistical Analysis

Normality of the data was tested by using the Shapiro-Wilk test. According to this, 99% or 95% confidence interval for the mean was chosen. To compare parameters between the three groups (HUF in vivo, HUF ex vivo and HOF) Student’s t-tests were applied due to the normality and independence of the data For the samples HUF ex vivo and HOF ex vivo collected from the same donors paired-sample t-test was used. For the samples collected from different women, independent sample t-test was used. Differences were considered significant when p<0.05 and p<0.01.

Characterization of non-diluted oviductal and uterine fluids in patients and donors

The ex vivo collections (HUF and HOF) were successfully performed in 25 patients with average age of 45 years and the in vivo collection (HUF) was successfully performed in 31 oocyte donors with average age of 25 years (Table 1). Due to the variability in the volume of fluid collected per woman (Table 1), it was not possible to check all of the parameters in all samples (Table 1). The volume of fluid collected in the participants of the proof of concept is not included in Table 1 because it was collected by flushing instead of being pure.

Table 1. Demographic characteristics of study population and collected volume (mean±SD) of reproductive fluids used to stablish the characterized biobank collection.

	*ex vivo* collection Patients	*in vivo collection* Oocyte donors
Demographic characteristics
Total N	25	31
Age (years)	45 ± 4	25 ± 4
Uterine fluid Collection
Total (n)	20	31
Volume (µL)	62.8 ± 33.5	40.7 ± 26.8
Oviductal fluid collection
Total (n)	25
Volume (µL)	23.5 ± 14.6

Osmolality

The osmolality values of HUF samples from women under infertility treatment (HUF, in vivo group) ranged between 278.80 and 294.00 mOsm/kg (Figure1). The mean value was 285.21 and the standard deviation (SD) was 0.77. The values for HUF from women in the ex vivo group ranged between 265.00 and 360.00 mOsm/kg and the mean±SD was 304.67 ± 9.79 (Figure1). Finally, the values for HOF ranged between 274 and 377, with a mean±SD of 316.56±12.33 (Figure1).

According to these data, the values within the 95% confidence interval for the mean were chosen as the criterion to select the physiological ranges, so it was decided to set the acceptable osmolality for UF between 284-287 and 282-327 mOsm/kg for in vivo and ex vivo groups, respectively. As for OF, and following the same criterion, the range was established between and 288-344 mOsm/kg. All analyzed samples found within that range were considered suitable for use as a future supplement in embryo culture.

pH

The values for pH in HUF from the in vivo group ranged from 7.12 to 7.92, with a mean±SD value of 7.70±0.23 (Figure2). Due to normality properties of the data, values within the 99% confidence interval for the mean were chosen to be included as physiological ranges, so that samples between

7.46 and 7.94 were considered suitable for ART use. The values for HUF from the ex vivo group ranged between 7.51 and 8.16 (Figure 2), being the mean±SD 7.84±0.32. Following the same criterion, the physiological ranges decided to be included were those between 7.6 and 8.2. Finally, the values for HOF were found between 6.3 and 8,36, being the physiological range established between 6.62 and 8.12 (Figure2).

Total protein concentration

The total protein concentration, evaluated by Bradford assay, of HUF samples from women under infertility treatment (HUF, in vivo group) ranged between 26.97 and 53.10µg/µL (Figure3).

The mean value was 41.9 µg/µL and the SD was 1.47 µg/µL. The values for HUF the ex vivo group ranged between 12.68 µg/µL and 87.95 µg/µL and the mean±SD was 47.5± 4.5 (Figure3). Finally, the values for HOF ranged between 14.18 and 60.90, with a mean±SD of 32.21± 3.11 (Figure3).

According to the values within the 95% confidence interval for the mean, the values for our QC certificates were established between 38.9 and 44.9 µg/µL for HUF in vivo, 37.8 and 57.1 µg/µL for HUF ex vivo, and 25.6 and 38.8 µg/µL for HOF ex vivo (Figure 3).

Hemoglobin concentration

Regarding the differences between the total protein concentration between the three types of fluids, they were significant between HOF and HUF, both in vivo and ex vivo (P<0.05).

By contrast, as we can see in the figure 4, there were not significant differences in terms of hemoglobin concentration between HUF and HOF.

To evaluate the functional properties (embryo assay) of the HRF, only the sterile and endotoxin free samples were studied because any biological reagent used in humans should not contain pathogenic agents that could represent a risk of disease transmission. Consequently, the endotoxins and sterility were first evaluated. Sterility controls were done only in the HUF samples collected in vivo and the result in all the samples was “no growth”. Obviously, the criterion for this parameter had no ranges nor options other than the absence of any microorganism´s growth to be considered suitable for biobanking or use in ART.

Similarly, two samples were discarded because they didn’t fall in the range considered suitable for embryo culture media, according to the European Pharmacopoeia and could not consequently, be considered as “endotoxin-free” or certificated with endotoxin levels below ≤ 0.1 EU/ml (Figure 5).

Both sterility and endotoxin levels in HUF and HOF could be certificated, with the standardized and officially accepted protocols, within the security range for their use as additives.

Bovine embryo assay (BEA)

The percentage of cleavage in bovine embryos in vitro fertilized and further cultured with 1% of HUF added to the medium was above 70% for 13 of the 14 HUF samples tested (in vivo group) (table 2). At seventh day after IVF, an average of 28% ± 8,1% of the cleaved bovine embryos had reached the blastocyst stage (table 2). The sample (number 5) with less than 70% of cleavage (47%), and the samples (7 and 10 – 15) with less 30% embryos expanded to blastocyst (Lonergan and Fair, 2014), were considered not suitable for biobanking or use in ART but the remaining 7 were considered suitable for both purposes.

Table 2. Percentage of cleaved embryos and embryos reaching the blastocyst stage after culture with 1% of the respective HUF. The samples (asterisk) with less than 70% of cleavage and with less 30% embryos expanded to blastocyst were considered not suitable for biobanking or use in ART but the remaining 7 were considered suitable for both purposes.

HUF sample no.	% Cleavage	% Embryos expanded to blastocyst
1	80.75	38
2	83.87	34.74
3	82.14	32.28
4	75.9	31.8
5	47*	-
6	76.7	31.2
7	77.5	29.5*
8	77.14	33.5
9	78.75	37.1
10	71.42	24.15*
11	73.3	18.75*
12	71.42	13.6*
13	71.1	15.25*
14	77.15	23.66*

Proof-of-concept

The collection of HUF from the eight women undergoing an IVF cycle as infertility treatment that had accepted to participate in the proof of concept NCT03436758 was performed by uterine flushes to guarantee that enough volume of sample per woman could be collected, checked for QC and used for embryo transfer. All the collected samples were under the physiological ranges previously observed in in vivo and ex vivo groups (table 3) and sterility and functional properties (by BEA) were assessed as adequate. A total of 13 transfers were performed in the 8 patients since in 3 of them their frozen embryos were subsequently transferred after negative pregnancy diagnoses. According to their morphological assessment under the stereomicroscope, 69% of the best quality embryos were derived from those cultured with albumin, whereas the remaining 31% were derived from the group cultured with autologous HUF. However, 33% of the transfers from the albumin group generated newborns (3 babies) whereas 50% of the transfers from embryos cultured with autologous HUF produced offspring (2 babies).

Table 3. Physico-chemical properties of human uterine fluid samples collected from eight women undergoing 13 IVF cycles as infertility treatment that accepted to participate in the proof of concept NCT03436758QC.

Patient no.	Osmolality (mOsm/kg)	pH	Total protein (µg/µl)	Sterility	Endotoxin	BEA
1	283	7.5	39	No growth	0.03	93
2	289	7.3	45	No growth	0.03	94
3	294	7.6	41	No growth	0.04	82
4	312	7.8	44	No growth	0.06	76
5	288	7.7	42	No growth	0.06	77
6	262	7.5	37	No growth	0.02	73
7	283	7.4	42	No growth	0.03	72
8	262	7.6	42	No growth	0.02	71

For years, different strategies have been developed to improve the composition of the embryo culture media, mimicking as much as possible the physiological fluids of the female reproductive tract. For instance, some of the proteins identified in these fluids have been added to the media, trying to characterize their possible benefits on the embryos, such as GMC-SF in humans (Sjöblom et al., 1999) or OVGP1 in livestock species (Coy et al., 2008). However, the high cost of their synthesis in the laboratory, the realization that the whole list of proteins in the fluids is enormous and therefore unattainable (Canha-Gouveia et al., 2019), and the observed inefficacy of their individualized addition in biological or clinical assays for GMC-SF (Ziebe et al., 2013) and for OVGP1 (Algarra et al., 2018) have led to the conclusion that adding the complete reproductive fluids as a supplements to the culture media should be a more efficient approach to palliate short and long-term undesirable effects of ART (Coy and Yanagimachi, 2015).

In livestock species, reproductive fluids have been successfully used already (Lopes et al., n.d.; Canovas et al., 2017; Lopera-Vasquez et al., 2017; Oller., 2020) but the same is still not a reality in humans. In the present study, we aimed to test and solve the difficulties of i) collecting human reproductive fluids in vivo and ex vivo; ii) developing protocols to guarantee the biosecurity of such fluids, allowing at the same time the storage of the maximum possible volume at a biobank, and iii) enabling the use of HRF in the routine work of a fertility clinic.

The first conclusion we reached was that there are not suitable devices in the market to collect uterine fluid trans-vaginally in oocyte donors without damaging the endometrium, since the negative pressure exerted by the aspiration plungers in commercial devices always produces tissue injury. For this reason, in parallel with this study, a new device has been designed to allow the collection of fluid by capillarity, which it is currently being tested (Canha-Gouveia and Latorre, 2019). The preliminary data show that the fluid collected with this new prototype contains less blood and cellular debris than any other commercial device tested before, therefore offering a promising solution for the future collection of fluid in donors or patients.

As for the protocols to establish the physiological ranges for the safe use of the fluids, we show here that a volume of 30 ml is enough to get reliable data for pH, osmolality, protein and albumin concentration, and BEA. This is of enormous importance for the future use of tested batches of HRF stored at a biobank. Although we have checked here fluids from women of different ages and physiological status, the final objective in the future would be to use only in vivo collected HUF from young donors (approximately 20-30 years old), with proven fertility from previous oocyte donations, pooled in homogeneous batches, fully QC tested, and stored at a biobank. Regarding the HOF, and due to the impossibility of realizing in vivo collections, our proposal is to use only those collected from healthy women of proven fertility attending hospitals to undergo a planned bilateral salpingectomy by laparoscopy as a contraceptive method. By introducing the collection of these fluids in a particular hospital of any geographical region as a routine, the amount that can be stored at a biobank would be more than enough to supplement all the culture media used by the patients under IVF cycles in the population covered by that hospital, bearing in mind that only 0.1-0.5 ml of fluid is necessary per embryo (culture drops are usually between 10 and 50 ml). Therefore, every 100 patients, properly scheduled at the appropriate phase of the cycle, would provide enough HRF to culture approximately 11500- 23000 embryos.

Another crucial point of discussion is the fact that reproductive fluids are a source of variability and that their addition to the culture media represents the loss of the so-called chemical definition. In our opinion, any biological process is, by definition, subject to individual variability. Thus, standardized cycles in ART will be never reached just because each patient responds differently to the hormone stimulation, possesses gametes with different genetics, and the embryologists and physicians participating in the whole process are susceptible to errors. Needless to say, commercial media are also subjected to the human intervention during their preparation and human recombinant albumin, hyaluronan and some other molecules included in the media cannot be considered as chemically defined. Finally, and despite the global tendency to use protein-free media, the benefit that provides to the embryo versus the future health repercussions of growing in a medium lacking hundreds of natural components should be considered since differences between in vivo and in vitro grown embryos have become evident (Munné et al., 2019). As a useful comparison, one could try to imagine the world if the use of blood for transfusions was prohibited for the sake of “standardization” and chemically defined blood was manufactured and used instead.

Most of the physico-chemical parameters assessed in the different types of HRF in our study were within the physiological ranges. For osmolality, less variability was observed between in vivo samples, and the higher variability was observed in HOF. Osmolality has been established around 250–300 mOsmol for most of the commercially available ART culture media (Gruber and Klein, 2011). However, these values may not be the physiological ones, since osmotic pressure of oviduct fluid in mouse was reported to be higher than 360 mOsmol (Borland et al., 1980). Other studies have shown similar osmolality levels between HUF and serum and the range was below our data for HOF (Casslen and Nilsson, 1984). This gave us the confidence to suggest that HOF could be frequently of higher osmolality than HUF and that this fact should be considered in the elaboration of the commercial media.

Regarding pH, again less variability was observed between in vivo samples. According to Phillips et al. (Phillips et al., 2000), the pH range for embryo culture media may be set between 7.4 and 7.2, but most laboratories usually decrease these figures during the culture of cleavage stage embryos and increase them during the morula and blastocyst stages (Swain, 2012). From the data of the present study, it seems clear that the limits could be established higher (7.4-8.0) for embryo culture media, and that IVF media should be set at a lower pH. Although the pH in the female reproductive tract is graduated, with lowest pH in the vagina (~pH 4.42) increasing toward the Fallopian tubes (~pH 7.94) (Ng et al., 2000), it is also known that the variation of these values could reflect the variation in the site-specific microbiome and acid–base buffering at the tissue/cellular level and also the responses to menstrual cycle (Chen et al., 2017). Consequently, it is logical that the shortest interval of values, not only for pH but also for osmolality, were found in HUF in vivo, whose collection was performed in oocyte donors at the same phase of the menstrual cycle (preovulatory), while the other reproductive fluids were collected from women in different phases of the menstrual cycle.

One of the most standardized and informative parameters that can be assessed in any biological fluid is the total protein concentration and, particularly, in the embryo culture medium, albumin is of striking importance (Morbeck et al., 2014). Albumin has a major role as a carrier for lipids, hormones, and minerals (Lane and Gardner, 2007). However, albumin can also be associated with globulins and there are studies suggesting that supplementation of human embryo culture media with 16% globulins in addition to 84% albumin results in an overall increase in implantation and live birth rates (Meintjes et al., 2009). This observation supports our and others idea about the need of proteins in the in vitro embryo culture media (Itze-Mayrhofer and Brem, 2020).

For the total protein concentration, we found significant differences between HOF and HUF, both in vivo and ex vivo (P<0.05), and the blood contamination was not the reason. Previous work has also shown a low total protein concentration in oviductal fluid (Canha-Gouveia et al., 2019), so this should be considered for the different formulation of IVF media versus embryo culture media. As explained above, and although the chemically defined media are considered safe, the lack of proteins could have an impact on the epigenetic reprogramming of the embryo and even on fetal and offspring development (Blake et al., 2002; Chronopoulou and Harper, 2015). For that reason, once again, we consider it important to include proteins in the culture media.

The mouse embryonic assay (MEA) (Ackerman et al., 1984, 1985; Byers et al., 2006) is commonly used by companies manufacturing culture media. Although mice are still the most popular species used to this end, we consider its use unnecessary, and contrary to the 3 Rs recommendations from Russell and Burch´s principle (Kirk, 2018); in other words, the sacrifice of animals with the only purpose of testing the toxicity of the fluids in a biological assay should be avoided. Instead, livestock species offer an alternative option as animal models to check the safety of any reagent added to the in vitro embryo culture media, since the availability of gametes in slaughterhouses is practically unlimited.

The BEA results in our study showed that this can be a suitable test to detect altered functionality of HRF that could be toxic for the embryos. Even with fluids falling in the ranges we have determined for osmolality, pH, endotoxins, etc. and apparently suitable, the BEA test can detect other unrecognized factors in the fluids affecting the gamete interactions or the embryo development, as it was the case. Therefore, BEA is proposed as a suitable test to replace MEA for this type of QC to avoid unnecessary animal sacrifices and because cow, at the reproductive level, is more similar to humans than are mice.

Finally, we examined the option of introducing the HRF in the routine of a fertility clinic, starting by a small assay adding fluid of patients receiving their own embryos, as a preliminary proof of concept before starting clinical studies with a larger sample size or use of heterologous HRF. The limiting factor in this case was the volume of fluid collected, since only the fluid of the own patient could be used for the QC test and for embryo culture. Our results have demonstrated that the strategy of supplementing the culture media with autologous HUF is feasible, allowing the birth of two newborns. Therefore, the next step is to test the feasibility of the use of HRF in batches of proven quality from heterologous donors to examine their effect on embryo quality, embryo implantation rate and short and long-term health of the newborns.

In conclusion, a complete QC sheet (table 4) for new natural additives, HOF and HUF, was obtained, including the security levels that should be taken into account for osmolality, pH, endotoxins, sterility, total protein and albumin concentrations and also for a functional assay using bovine gametes and embryos. A fully characterized biobank collection of HRF following the quality control protocols described in the present paper has been created at Biobank-Mur and will be crucial for pertinent future clinical trials to demonstrate the benefits of HRF with a suitable sample size. The potential use of HRF was corroborated with the birth of the first two children derived from embryos cultured with HUF of their own mothers as a supplement to the medium.

Table 4. QC sheet for human reproductive fluids collected in vivo and ex vivo to be used as new natural products in culture media.

	HUF IN VIVO	HUF EX VIVO	HOF EX VIVO
Osmolality (mOsm/kg)	284-287	282-327	288-345
pH	7.5 - 7.9	7.5 - 8.2	6.6 - 8.1
Total protein (µg/µl)	39 – 45 38 - 57		26 -39
Sterility		“no growth”
Endotoxin	“endotoxin-free” ≤ 0.1 EU/ml
BEA (2 cell bovine embryo assay)		> 70%

Acknowledgements

The authors are grateful to Matadero Orihuela, S.A., Murcia – Spain, for the kindness in providing biological material and wish to thank Dr. Jordana Lopes for scientific and technical advice regarding Bovine Embryo Assay. The authors also want to thank to Centro Regional de Hemodonación, Murcia for providing Hemocue technologies to quantify the level of hemoglobin in the collected samples. The authors want to thank to Alejandro Torrecillas (Area Cientifica y Técnica de Investigación (ACTI) – University of Murcia) for his advice in quantification of total protein concentration, and Carmen Algueró (Unidad de Producción Celular (UPC) IMIB- Arrixaca) for her help with endotoxin level quantification. The authors are grateful to Dr. Larry Reynolds for critical reading and editing of the manuscript.

Authors’ roles

PC conceived the idea. PC and AC-G designed the study. PC and AC-G wrote the first manuscript draft. AC-G performed most of the analyses and data collection. PC and AC-G analyzed the data. MT P-S and ML S-F provided the ex vivo HUF and HOF samples at the Hospital. JL, JCMS and MM provided the in vivo HUF samples at the infertility clinic. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Funding

This study was funded by the European Union, Horizon 2020 Marie Sklodowska-Curie Action, REPBIOTECH 675526, the Spanish Ministry of Economy and Competitiveness and the European Regional Development Fund (AGL 2015-66341-R) and Fundación Séneca 20040/GERM/1.

Conflict of interest

The authors declare no competing financial interest.

Ackerman SB, Stokes GL, Swanson RJ, Taylor SP, Fenwick L. Toxicity testing for human in vitro fertilization programs. Journal of in vitro fertilization and embryo transfer : IVF 1985;2:132–137. United States.

Ackerman SB, Taylor SP, Swanson RJ, Laurell LH. Mouse embryo culture for screening in human IVF. Archives of andrology 1984;12 Suppl:129–136. England.

Adamson GD, Mouzon J de, Chambers GM, Zegers-Hochschild F, Mansour R, Ishihara O, Banker M, Dyer S. International Committee for Monitoring Assisted Reproductive Technology: world report on assisted reproductive technology, 2011. Fertility and Sterility 2011;110:1067–1080. Elsevier.

Algarra B, Maillo V, Avilés M, Gutiérrez-Adán A, Rizos D, Jiménez-Movilla M. Effects of recombinant OVGP1 protein on in vitro bovine embryo development. The Journal of reproduction and development 2018;64:433–443. The Society for Reproduction and Development.

Baird DT, Collins J, Egozcue J, Evers LH, Gianaroli L, Leridon H, Sunde A, Templeton A, Steirteghem A Van, Cohen J, et al. Fertility and ageing. Human Reproduction Update 2005;11:261–276.

Bang FB. A bacterial disease of Limulus polyphemus. Bulletin of the Johns Hopkins Hospital 1956;98:325–351. United States.

Blake D, Svalander P, Jin M, Silversand C, Hamberger L. Protein supplementation of human IVF culture media. Journal of Assisted Reproduction and Genetics 2002;19:137–143.

Borland RM, Biggers JD, Lechene CP, Taymor ML. Elemental composition of fluid in the human Fallopian tube. Journal of reproduction and fertility 1980;58:479–482. England.

Byers SL, Payson SJ, Taft RA. Performance of ten inbred mouse strains following assisted reproductive technologies (ARTs). Theriogenology 2006;65:1716–1726.

Canha-Gouveia A, Latorre R. Uterine Fluid in vivo Collection Device. 2019; MIA PATENTS. Canha-Gouveia A, Paradela A, Ramos-Fernández A, Prieto-Sánchez TM, Sánchez-Ferrer LM,

Corrales F, Coy P. Which Low-Abundance Proteins are Present in the Human Milieu of Gamete/Embryo Maternal Interaction? International Journal of Molecular Sciences 2019;20:.

Canovas S, Ivanova E, Romar R, García-Martínez S, Soriano-Úbeda C, García-Vázquez FA, Saadeh H, Andrews S, Kelsey G, Coy P. DNA methylation and gene expression changes derived from assisted reproductive technologies can be decreased by reproductive fluids. In Tyler JK, editor. eLife 2017;6:e23670. eLife Sciences Publications, Ltd.

Casslen B, Nilsson B. Human uterine fluid, examined in undiluted samples for osmolarity and the concentrations of inorganic ions, albumin, glucose, and urea. American journal of obstetrics and gynecology 1984;150:877–881. United States.

Chen C, Song X, Wei W, Zhong H, Dai J, Lan Z, Li F, Yu X, Feng Q, Wang Z, et al. The microbiota continuum along the female reproductive tract and its relation to uterine-related diseases. Nature Communications 2017;8:875.

Chen Z, Hagen DE, Elsik CG, Ji T, Morris CJ, Moon LE, Rivera RM. Characterization of global loss of imprinting in fetal overgrowth syndrome induced by assisted reproduction. Proceedings of the National Academy of Sciences of the United States of America 2015;112:4618–4623. National Academy of Sciences.

Chronopoulou E, Harper JC. IVF culture media: Past, present and future. Human Reproduction Update 2015;21:39–55.

Coy P, Cánovas S, Mondéjar I, Saavedra MD, Romar R, Grullón L, Matás C, Avilés M. Oviduct- specific glycoprotein and heparin modulate sperm-zona pellucida interaction during fertilization and contribute to the control of polyspermy. Proceedings of the National Academy of Sciences of the United States of America 2008;105:15809–15814. National Academy of Sciences.

Coy P, Yanagimachi R. The Common and Species-Specific Roles of Oviductal Proteins in Mammalian Fertilization and Embryo Development. BioScience 2015;65:973–984.

Dyer S, Chambers GM, Mouzon J de, Nygren KG, Zegers-Hochschild F, Mansour R, Ishihara O, Banker M, Adamson GD. International Committee for Monitoring Assisted Reproductive Technologies world report: Assisted Reproductive Technology 2008, 2009 and 2010†. Human Reproduction 2016;31:1588–1609.

Fernández-Gonzalez R, Moreira P, Bilbao A, Jiménez A, Pérez-Crespo M, Ramírez MA, Fonseca FR De, Pintado B, Gutiérrez-Adán A. Long-term effect of in vitro culture of mouse embryos with serum on mRNA expression of imprinting genes, development, and behavior. Proceedings of the National Academy of Sciences of the United States of America 2004;101:5880 LP – 5885.

Gardner DK, Kelley RL. Impact of the IVF laboratory environment on human preimplantation embryo phenotype. Journal of Developmental Origins of Health and Disease 2017;8:418-435. Cambridge University Press.

Gruber I, Klein M. Embryo culture media for human IVF: which possibilities exist? Journal of the Turkish German Gynecological Association 2011;12:110–117.

Hajj N el, Haaf T. Epigenetic disturbances in in vitro cultured gametes and embryos: implications for human assisted reproduction. Fertility and Sterility 2013;99:632–641. Elsevier.

Itze-Mayrhofer C, Brem G. Quantitative proteomic strategies to study reproduction in farm animals: Female reproductive fluids. Journal of Proteomics 2020;103884.

Ivanova E, Canovas S, Garcia-Martínez S, Romar R, Lopes JS, Rizos D, Sanchez-Calabuig MJ, Krueger F, Andrews S, Perez-Sanz F, et al. DNA methylation changes during preimplantation development reveal inter-species differences and reprogramming events at imprinted genes. Clinical epigenetics 2020;12:64.

Kirk RGW. Recovering The Principles of Humane Experimental Technique: The 3Rs and the Human Essence of Animal Research. Science, technology & human values 2018;43:622–648. SAGE Publications.

Lane M, Gardner DK. Embryo culture medium: which is the best? Best Practice & Research Clinical Obstetrics & Gynaecology 2007;21:83–100.

Levin J, Bang FB. The role of endotoxin in the extracellular coagulation of limulus blood. Bulletin of the Johns Hopkins Hospital 1964;115:265–274. United States.

Levin J, Bang FB. Clottable protein in Limulus; its localization and kinetics of its coagulation by endotoxin. Thrombosis et diathesis haemorrhagica 1968;19:186–197. Germany.

Lonergan P, Fair T. The ART of studying early embryo development: progress and challenges in ruminant embryo culture. Theriogenology 2014;81:49–55. United States.

Lopera-Vasquez R, Hamdi M, Maillo V, Lloreda V, Coy P, Gutierrez-Adan A, Bermejo-Alvarez P, Rizos D. Effect of bovine oviductal fluid on development and quality of bovine embryos produced in vitro. Reproduction, Fertility and Development 2015;

Lopera-Vasquez R, Hamdi M, Maillo V, Lloreda V, Coy P, Gutierrez-Adan A, Bermejo-Alvarez P, Rizos D. Effect of bovine oviductal fluid on development and quality of bovine embryos produced in vitro. Reproduction, fertility, and development 2017;29:621–629. Australia.

Lopes J, Alcazar-Triviño E, Soriano-Úbeda C, Hamdi M, Cánovas S, Rizos D, Coy P. Reproductive outcomes and endocrine profile in artificially inseminated versus embryo transferred cows. Animal Reproduction Science (under review).

Lopes JS, Canha-Gouveia A, Paris-Oller E, Coy P. Supplementation of bovine follicular fluid during in vitro maturation increases oocyte cumulus expansion, blastocyst developmental kinetics, and blastocyst cell number. Theriogenology 2019;126:222–229. United States.

Meintjes M, Chantilis SJ, Ward DC, Douglas JD, Rodriguez AJ, Guerami AR, Bookout DM, Barnett BD, Madden JD. A randomized controlled study of human serum albumin and serum substitute supplement as protein supplements for IVF culture and the effect on live birth rates†. Human Reproduction 2009;24:782–789.

Morbeck DE, Krisher RL, Herrick JR, Baumann NA, Matern D, Moyer T. Composition of commercial media used for human embryo culture. Fertility and Sterility 2014;102:759- 766.e9. Elsevier Inc.

Munné S, Nakajima ST, Najmabadi S, Sauer M V, Angle MJ, Rivas JL, Mendieta L V, Macaso TM, Sawarkar S, Nadal A, et al. First PGT-A using human in vivo blastocysts recovered by uterine lavage: comparison with matched IVF embryo controls†. Human Reproduction 2019;35:70–80.

Ng EH, Ajonuma LC, Lau EY, Yeung WS, Ho PC. Adverse effects of hydrosalpinx fluid on sperm motility and survival. Human reproduction (Oxford, England) 2000;15:772–777. England.

Oller. E de las MP. Impact of embryo transfer and In vitro production of embryos on phenotypical and molecular traits of offspring in the porcine species (Sus scrofa). 2020;

Phillips KP, Leveille MC, Claman P, Baltz JM. Intracellular pH regulation in human preimplantation embryos. Human reproduction (Oxford, England) 2000;15:896–904. England.

Sjöblom C, Wikland M, Robertson SA. Granulocyte-macrophage colony-stimulating factor promotes human blastocyst development in vitro. Human reproduction (Oxford, England) 1999;14:3069–3076. England.

Steiner AZ, Jukic AMZ. Impact of female age and nulligravidity on fecundity in an older reproductive age cohort. Fertility and Sterility 2016;105:1584-1588.e1.

Sunde A, Brison D, Dumoulin J, Harper J, Lundin K, Magli MC, Abbeel E Van Den, Veiga A. Time to take human embryo culture seriously. Human Reproduction 2016;

Swain JE. Is there an optimal pH for culture media used in clinical IVF? Human Reproduction Update 2012;18:333–339.

Ziebe S, Loft A, Povlsen BB, Erb K, Agerholm I, Aasted M, Gabrielsen A, Hnida C, Zobel DP, Munding B, et al. A randomized clinical trial to evaluate the effect of granulocyte- macrophage colony-stimulating factor (GM-CSF) in embryo culture medium for in vitro fertilization. Fertility and sterility 2013;99:1600–1609. United States.

Natural fluids collected from the reproductive tract of women can be used as additives for human in vitro embryo culture media

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Results

Discussion

Declarations

References

Status:

Version 1