Ovarian function after COVID-19: long-term effects and vaccine safety in ART patients

Abstract

Purpose

This study aimed to evaluate the long-term impact of mild COVID-19 infection and COVID-19 vaccination on ovarian function in patients undergoing assisted reproductive technology (ART). Specifically, we assessed ovarian outcomes between 9 and 18 months post-infection and investigated the effects of COVID-19 vaccines (inactivated virus and adenovirus) on reproductive parameters.

Methods

The study included two objectives: (a) examining ovarian function in post-COVID-19 patients (9–18 months post-infection) compared to a control group and (b) comparing reproductive outcomes in vaccinated versus unvaccinated patients. According to the study objectives, ART patients were divided into the following groups: a control group (n = 30), a post-COVID-19 group (n = 55), an unvaccinated group (n = 70), and a vaccinated group (n = 55). Findings revealed a reduction in the number of retrieved and mature oocytes in patients over 36 years in the post-COVID-19 group. Lower IL-1β levels were found in follicular fluid (FF) of post-COVID-19 patients, while VEGF levels were reestablished between 9 and 18 months post-infection. Although cell migration was reduced in endothelial cells incubated with post-COVID-19 FF, angiogenic factor levels and DNA integrity remained stable. No significant differences in retrieved or mature oocytes were observed between vaccinated and unvaccinated patients.

Conclusions

VEGF levels and DNA integrity in FF from post-COVID-19 patients were normalized between 9 and 18 months post-infection. Additionally, COVID-19 vaccination did not negatively impact ovarian response in ART patients, supporting vaccine safety in reproductive contexts.

Introduction

The most frequently observed symptoms in patients with COVID-19 include fever, cough, and pneumonia. However, thrombosis, pulmonary embolism, and high blood pressure have also been reported, suggesting that the virus targets the endothelium [1].

While the virus primarily affects the respiratory system, it has been shown to potentially infect other tissues and organs in the body [2, 3]. Organs such as the heart, intestine, testicles, and ovaries are potential targets for SARS-CoV-2 infection. The extent of disease and the effects on these organs can vary among individuals and may depend on various factors, including the person’s overall health and immune response.

Viral entry of SARS-CoV-2 requires its binding to the target cell receptor, angiotensin-converting enzyme 2 (ACE-2), through the spike glycoprotein (S). Subsequently, the ligand-receptor complex is processed by the transmembrane serine protease 2 (TMPRSS2), allowing for the fusion of the viral cell membrane and the host cell [4].

In the female reproductive system, ACE-2 is expressed in the uterus, vagina, placenta, and ovary. ACE-2 has been detected in the ovary from both reproductive age and postmenopausal women. In the human ovary, stromal as well as granulosa cells have been shown to express the ACE-2 receptor [5]. ACE-2 regulates follicular development, ovulation, ovarian angiogenesis, and embryo development [6]. Based on these considerations, ACE-2 plays a key role in fertility regulation, for which SARS-CoV-2 could damage the functionality of ovarian cells [7]. Furthermore, since the viral infection severely affects the immune system, the function of the hypothalamic-pituitary–gonadal axis may also be altered [7].

Previously, we have investigated the effect of SARS-CoV-2 infection on ovarian function post-COVID-19. To this end, we used follicular fluids (FF) from healthy patients and from patients who had recovered from COVID-19 (mild symptoms, 2–9 months post-infection) in assisted reproduction techniques (ART) and studied their effects on a human endothelial cell line and a human granulosa cell line [8]. Our results described for the first time that SARS-CoV-2 infection negatively affects the follicular microenvironment, altering ovarian function. Our findings are consistent with many other studies, emphasizing the implications of COVID-19 infection on female reproductive health, specifically about menstrual irregularities, ovarian dysfunction, and compromised fertility [9–11].

The female gamete is exposed to a microenvironment that includes the FF and somatic cells (granulosa and theca cells) within the follicle. The composition of FF differs from that of serum, containing a complex mixture of hormones, cytokines, growth factors, metabolites, and other proteins secreted primarily by granulosa cells [12, 13]. It is worth mentioning that the composition of FF reflects the oocyte’s development stage and its quality [14, 15]. Therefore, an altered FF composition is associated with reduced reproductive function. In line with these considerations, the vascular network also plays a significant role in oocyte development. Angiogenesis, the formation of new blood vessels, is a vital process in the reproductive system of healthy adult animals. In the ovary, developing microvasculature is essential for delivering nutrients and hormones to support both folliculogenesis and luteogenesis. The ovarian follicle secretes various angiogenic factors, with vascular endothelial growth factor A (VEGFA) thought to be central to ovarian angiogenesis regulation [16]. Although VEGFA is the main initiator of angiogenesis, establishing a structurally and functionally mature vascular network requires a coordinated interplay of additional factors. Among these factors are angiopoietins, ANGPT1 and ANGPT2, which signal through the tyrosine kinase receptor. Unlike VEGF, ANGPT1 does not promote endothelial cell proliferation but is critical for recruiting perivascular cells to stabilize newly formed capillaries. In contrast, ANGPT2 acts as a natural antagonist to ANGPT1, counteracting its stabilizing effect and enhancing endothelial cell migration and neovascularization [17–19].

It is important to mention when it comes to female reproductive health that an increase in the incidence of premature births and cesarean sections has been observed. Even older studies concluded that the SARS virus can cause higher rates of miscarriages, premature births, and intrauterine growth disorders during pregnancy [20]. Furthermore, the psychological effects of COVID-19 are increasingly recognized as a component of post-infection recovery, and evidence suggests that mental health impacts may indirectly influence reproductive health by altering hormonal regulation and ovarian function [11]. Stress and anxiety resulting from the infection or recovery period may affect the hypothalamic-pituitary–gonadal axis, which plays a key role in fertility. These psychological stressors could exacerbate the effects of COVID-19 on ovarian function, as seen in alterations to the follicular microenvironment.

On the other hand, since 2020, around 11 billion doses of vaccines have been administered all over the world. Thus far, there has been an important body of literature that has proven the short- to medium-term safety profiles and efficacy of these vaccines in preventing COVID-19 hospitalization and deaths [21].

The apprehensions surrounding the potential repercussions of COVID-19 vaccines on fertility, compounded by the dissemination of misinformation, significantly influence public health discourse [22]. Currently, there are very few studies that evaluate the potential impact of the different types of vaccines against COVID-19 on female fertility. In particular, several reproductive medicine centers in the USA and Israel have carried out studies that demonstrated the safety of mRNA-based vaccines on the reproductive system [23–27].

Thus far, in the female reproductive system, it has been observed that COVID-19 vaccines cause alterations in the menstrual cycle up to 3 months after administration [28–30]. A study conducted in England by Male et al. (2022) showed that 30,000 women experienced changes in their menstrual cycle or vaginal bleeding outside of their usual periods after vaccination. [28]. In Argentina, a study conducted by the Argentine Society for Fertility Preservation (SAPREF) in 2022 through virtual surveys revealed that 40% of women of reproductive age reported changes in bleeding frequency after vaccination [31]. In Argentina, until June 2023, 115.712.194 doses of vaccines for COVID-19 have been applied. In total, 41.494.317 people were vaccinated with 2 doses, among which 34.880.863 people were further vaccinated with a booster (www.argentina.gob.ar). The vaccines applied in our country are based on different manufacturing platforms: adenovirus (Sputnik V, Covishield, AstraZeneca), inactivated virus (Sinopharm), and mRNA (Moderna and Pfizer). Considering the dynamic and cyclical nature of menstrual bleeding, it serves as a distinct indicator of overall health and fertility. To the best of our knowledge, this is the first study in Argentina to assess the effect of adenovirus-based vaccines (Sputnik V and AstraZeneca) and inactivated viruses (Sinopharm) on the number of recovered and mature oocytes from patients undergoing assisted reproduction treatment.

Based on these considerations and in conjunction with our previous results, we hypothesized that (a) patients who have recovered from COVID-19 (mild symptoms) restore normal ovarian function after an extended period (9–18 months post-infection) and (b) COVID-19 vaccines (inactivated virus and adenovirus) affect the success of female reproductive outcomes in patients undergoing ART.

Materials and methods

Ethical approval

All the experiments were approved by the Ethics Committee of the Instituto de Biología y Medicina Experimental (IByME), Buenos Aires, Argentina (protocol No.2850). Informed consent was obtained from all patients before their recruitment.

Study population and FF collection

For this study, we enrolled a total of 125 patients (between 21 and 41 years old) who underwent assisted reproductive therapies from PREGNA Medicina Reproductiva (Buenos Aires, Argentina), Fertilis Buenos Aires (Buenos Aires, Argentina), We FIV (Buenos Aires, Argentina), and In Vitro (Buenos Aires, Argentina).

Patients with pathologies such as premature ovarian failure, PCOS, endometriosis, and uterine fibroids were excluded from this study. In addition, patients with low ovarian response (less than 3 antral follicles) were excluded.

For the first part of the study, the patients were separated into a control group of patients not infected with the SARS-CoV-2 virus (n = 30), and a group of patients recovered from COVID-19, up to 18 months post-infection (mild symptoms) (n = 55). Both groups of patients had not received the COVID-19 vaccine. The patients with mild symptoms of COVID-19 include individuals who have been infected with the SARS-CoV-2 virus and exhibit symptoms that are not severe or do not require hospitalization. These mild symptoms may include low-grade fever, mild cough, fatigue, sore throat, nasal congestion, and loss of taste or smell, among others.

For the second part of the study, the patients were divided according to whether they had been vaccinated (two doses) (n = 55) or not (n = 70) and to the type of COVID-19 vaccine manufacturing platform that had been applied (viral vector or inactivated virus). All FF were recruited between November 2020 and March 2022.

The ovarian stimulation protocol for all female patients included the administration of recombinant follicle-stimulating hormone (FSH) (Gonal-F, Merck-Serono, Germany) in combination with human menopausal gonadotropin (hMG) (Menopur, Ferring, Sweden). An initial daily dose of 150–300 IU of gonadotropins was prescribed for 5 days, with dosage adjustments based on individual ovarian responses. Ovulation was triggered with a single injection of 10,000 IU of human chorionic gonadotropin (hCG) (Gonacor 5000, Ferring Pharmaceuticals, Switzerland) or gonadotropin-releasing hormone agonist (GnRH agonist) (Gonapeptyl, Ferring, Sweden) 34–36 h before follicular aspiration. Oocyte retrieval was performed under vaginal ultrasound guidance 34–36 h after ovulation induction. Human follicular fluid (FF) was acquired from follicles between 16 and 20 mm in each patient. FF aspirated from each patient was collected into 15-ml conical-bottom polypropylene tubes. Only clean fluids, free from blood contamination, were used in this study. After oocyte retrieval, the FF was centrifuged for 10 min at 2000 g to remove cellular components and debris, transferred to sterile tubes, and the supernatant was stored at − 20 °C before performing the assays. To ensure a population as representative as possible, all FF obtained from patients across the four fertility clinics where the two standard stimulation protocols were implemented, were included. The control group and the post-COVID-19 group were composed of FF samples from different patients who underwent stimulation using the two most commonly employed protocols, with ovulation triggering achieved through either a GnRH agonist or hCG. For in vitro experiments on endothelial cells, 20 patients per group were randomly selected. Serum samples for estradiol determination were obtained on the day of the ovulation trigger. Basal hormone levels before ovarian stimulation (estradiol, progesterone, and prolactin) were obtained from the patient’s medical studies, when available. Several parameters were used to assess the efficiency of ovarian stimulation, including the number of cumulus-oocyte complexes obtained and the number of mature oocytes (those in metaphase 2).

Immunoassays

The levels of IL-10 and IL-1β in FF were calculated using a commercial ELISA Kit (IL-1β Catalog# 557,953; IL-10 Catalog# 555,157; BD Biosciences, CA, United States), as previously described by Gori et al. [32].

Endothelial cell culture

EA.hy926 cells have numerous characteristics of vascular endothelial cells [33]. It is a helpful in vitro model to study angiogenic processes in the ovary [34–37]. Cells were donated by Dr Martinez (IFIBIO-CONICET, Bs.As., Argentina) and were maintained in Iscove’s Modified Dulbecco’s Medium (IMDM, Invitrogen, NY, USA) with 10% FBS in the presence of 100 U/ml penicillin G and 100 mg/ ml streptomycin sulfate at 37 °C with 5% CO₂. The number of passages used for the experiments did not exceed 20.

Western blot

For protein analysis in a cell extract, EA.hy926 cells were seeded into 24-well cell culture plates at a density of 0.5*10⁶ cells/well, allowed to adhere to the surface, and grown to confluence. Then, cells were incubated with FF (25% FF in control media) from either control or post-COVID-19 patients for 24 h at 37 °C. After treatment with FF of both groups, EA.hy926 cells were lysed in lysis buffer (20 mM Tris–HCl pH 8, 137 mM NaCl, 1% Nonidet P-40, and 10% glycerol) supplemented with protease inhibitors (0.5 mM PMSF, 0.025 mMN-CBZ-l-phenylalanine chloromethyl ketone, 0.025 mMN-p-tosyl-lysine chloromethyl ketone, and 0.025 mM l-1-tosylamide-2-phenyl–ethyl chloromethyl ketone). The cell lysates were centrifuged at 10,000 g for 10 min at 4 °C. In the case of protein measurement in FF, this was diluted 1/10.

Both for the protein extract of cells and for the follicular fluid, protein concentration was measured by the Bradford assay. After boiling for 5 min, 20 μg of protein was applied to a 12% SDS–polyacrylamide gel, and electrophoresis was performed at 25 mA for 1.5 h. Subsequently, the resolved proteins were transferred for 2 h onto nitrocellulose membranes. The blot was preincubated in blocking buffer (5% nonfat milk, 0.05% Tween 20 in 20 mM TBS pH 8.0) for 1 h at room temperature and incubated overnight in blocking buffer at 4 °C with diluted primary antibodies as follows: β-actin 1:3000 purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, USA); VEGF 1:1000 (ab46154), γH2AX 1:1000 (ab26350), ANGPT-1 1:1000 (ab133425), and ANGPT-2 1:1000 (ab180820) purchased from Abcam (Cambridge, USA). The immunoblots were then incubated with HRP-conjugated secondary antibodies, namely anti-rabbit 1:1000 (A4914) (Sigma Aldrich), anti-mouse 1:1000 (HAF007) from R&D Systems (MN, USA), or anti-goat 1:2000 (#1,721,034), as required. The signal was detected by chemiluminescence. Protein levels were analyzed by densitometry using Scion Image for Windows (Scion Corporation, Worman’s Mill, CT, USA). Proper loading was assessed by staining the membranes with Ponceau-S and this stain was used to relativize in the case of protein detection in FF.

The density of each band for proteins from the cell line was normalized to the density of β-actin or ponceau, which was used as an internal expression control. Optical density data are expressed as arbitrary units ± SEM. All blots shown were representative of at least three independent experiments.

Endothelial cell migration

A wound healing assay was performed using the EA.hy926 endothelial cell line to study the effect of FF on endothelial cell migration as previously described by Scotti et al. (2013, 2014, 2016) [34, 35, 37]. EA.hy926 cells were detached by trypsinization, resuspended in IMDM, plated at a density of 3*105 cells per well in 24-well plates, and grown to confluence. Cell monolayers were wounded by a 1000-μl micropipette tip in one direction. After the injury, the cell culture was washed with PBS to remove cellular debris. The wounded cells were incubated with FF (25%) either from control (n = 20) or post-COVID-19 patients (n = 20). Serum-free DMEM/F12 was used as a negative control (n = 16). Cells were then incubated for 12 h at 37 °C. Cell migration was monitored at initial wounding (t 0 h) and 12 h (t 12 h) under a phase-contrast microscope and pictures were acquired at the same magnification and location every time. The resulting cell migration was calculated as the cell-free area at t 0 h – cell-free area at t 12 h and was expressed as a percentage of the mean migration of negative control wells (without FF). Endothelial cell migration in negative control wells (media without FF) is presented as 100%. We quantified the cell-free wounded areas using ImageJ software (National Institutes of Health, Bethesda, MD). All the experiments were carried out in duplicate.

Statistical analysis

Statistical analyses were performed using the statistical software Prism v8.0 (GraphPad Software, San Diego, CA, USA). All the patients were incorporated in the statistical analysis since there were no differences between the parameters studied in either group receiving GnRHa or hCG trigger for ovulation. Data are expressed as the mean ± SEM. Differences between groups were tested for significance using Student’s t-test for parametric variables. To estimate the endothelial cell migration, normally distributed data were analyzed using one-way ANOVA followed by Tukey’s test for statistical comparison of the groups. Statistical significance was defined as p < 0.05.

Results

The characteristics of the patients are shown in Table 1. Before starting the IVF procedure, the patients underwent a general clinical examination. Multiple indicators were recorded, including BMI, AMH levels, estradiol, progesterone, and prolactin. No significant differences were found in the parameters of the patients.

Table 1.

Characteristics of control patients and post-COVID-19 patients (9–18 months post-infection)

Baseline characteristics of patients	Control patients (n = 30)			Post-COVID-19 patients (9–18 months) (n = 55)			P value
	Mean	Min–Máx	SEM	Mean	Min-Máx	SEM
Age (years)	34.94	27–43	0.62	33.33	21–42	1.33	n.s
BMI (kg/m2)	23.02	19.38–29.41	1.35	23.65	16.20–33–12	1.02	n.s
AMH (ng/ml)	2.01	0.79–3.56	0.45	1.73	0.50–2.60	0.27	n.s
Basal serum estradiol (pg/ml)	25.81	10.70–40.00	2.60	22.77	16.20–33.12	1.23	n.s
Basal serum progesterone (ng/ml)	0.45	0.20–0.80	0.15	0.69	0.40–1.25	0.14	n.s
Basal serum prolactin (ng/ml)	15.20	8.00–22.70	2.18	16.81	9.20–32.00	2.65	n.s

Data are expressed as the mean ± standard error of the mean. Student's t-test was used for comparisons between groups. Statistical significance was defined as < 0.05

We previously observed that the number of retrieved and mature oocytes was reduced in the FF of patients 2–9 months post-infection compared to control patients [8]. Based on these results, we decided to analyze these parameters in FF of patients between 9 and 18 months after COVID-19 infection until the moment of follicular aspiration. We observed that both the number of oocytes retrieved and the number of oocytes in MII (Fig. 1) were significantly lower in post-COVID-19 patients compared to control patients (p < 0.05).

Fig. 1.

Effect of COVID-19 infection on number of recovered oocytes and number of mature oocytes (M2) between 9 and 18 months post-infection. Data are represented as the mean ± SEM. Different letters represent statistically significant differences between groups (a vs b p < 0.05)

Taking into consideration that fertility in women decreases sharply from 35 to 36 years of age, it was decided to evaluate the number of oocytes retrieved and the number of mature oocytes in patients after 9–18 months post-infection by COVID-19 and based on age. The results elucidate that the number of retrieved and mature oocytes decreased in patients older than 36 years of age who had suffered from COVID-19 between 9 and 18 months from infection to follicular aspiration (Fig. 2B). In contrast, no significant changes were observed in patients under or equal 35 years of age between the different groups (Fig. 2A).

Fig. 2.

Effect of COVID-19 infection on the number of oocytes retrieved and the number of mature oocytes (M2) between 9 and 18 months post-infection according to age. A Number of oocytes retrieved in patients ≤ 35 years of age and number of mature oocytes (M2) in patients ≤ 35 years. B Number of oocytes retrieved in patients > 36 years of age and number of mature oocytes (M2) in patients > 36 years. Different letters represent statistically significant differences between groups (a vs b p < 0.05)

Previously, we observed that IL-1β levels decreased in the FF of patients 2–9 months post-infection compared to control patients, while IL-10 levels remained unchanged in both experimental groups [8]. Based on these results, we aimed to evaluate the levels of these cytokines in FF of patients between 9 and 18 months after COVID-19 infection until the moment of follicular aspiration. The results demonstrate that IL-1β levels were not restored to control group levels in FF from patients who had suffered from COVID-19 up to 18 months before follicle aspiration (p < 0.001). IL-10 levels did not show significant differences between the groups analyzed (Fig. 3A). In the previous study, we observed lower VEGF levels in the FF of patients who recovered from COVID-19 up to 9 months post-infection. Based on these observations, we evaluated whether the levels were restored 18 months post-infection. As seen in Fig. 3B, VEGF levels are restored to control values from 9 to 18 months after COVID-19 infection.

Fig. 3.

Effect of COVID-19 infection on IL-10, IL-1β, and VEGF concentration in control and post-COVID-19 FF. A IL-1β and IL-10 concentrations in FF from post-COVID-19 and control patients determined by ELISA. B VEGF levels measured by Western blot in FF from post-COVID-19 and control patients. Different letters represent statistically significant differences between groups (a vs b p < 0.05)

We have previously observed that incubation of endothelial cells (EA.hy926) with FF from patients after COVID-19, decreased endothelial cell migration compared to control FF. In consequence, we decided to evaluate whether this alteration continues up to 18 months after infection by SARS-CoV-2. We observed that this alteration is maintained when endothelial cells are incubated with FF from patients between 9 and 18 months post-infection with SARS-CoV-2 (p < 0.01) (Fig. 4A and B).

Fig. 4.

Effect of FF from recovered COVID-19 patients on endothelial cells (EA.hy926) (9–18 months). A Quantification of the wound healing assay (angiogenesis assay). Columns show the percentage of endothelial cell migration normalized to the negative control, which is reported as 100% migration. Data are expressed as means ± SEM. B Representative images were obtained immediately after wounding (t 0) and after 12 h (t 12). The black lines represent the migratory fronts. C Effects of stimulation with control or post-COVID-19 FF on the expression of ANGPT-1; ANGPT-2, and ANGPT-1/ANGPT-2. D Effects of stimulation with control or post-COVID-19 FF on the expression of on DNA damage (γH2AX) in EA.hy926 cells. The graphs show the densitometric analysis of protein levels. The density of each band was normalized to the density of the β-actin bands. The lower panels show representative blots for each protein analyzed. Different letters represent statistically significant differences between groups (a vs b p < 0.05)

We analyzed ANGPT-1 and ANGPT-2 protein expression in endothelial cells and no changes were found between groups, as well as in VEGF protein levels when cells were stimulated with either control or post-COVID-19 FF (Fig. 4C).

Additionally, we studied the effect of FF from recovered COVID-19 patients on DNA damage, as determined by endothelial cell expression of γH2AX. Protein levels of γH2AX in EA.hy926 cells incubated with FF from post-COVID-19 patients and no changes were observed between groups (Fig. 4D).

To study the effect of the vaccines on ovarian function, the patients were divided into two groups: unvaccinated and vaccinated patients, including infected and uninfected patients with COVID-19 in the two groups. The group of vaccinated patients was subdivided based on the vaccine manufacturing platform: inactivated virus (Sinopharm) or viral vector (Sputnik V and AstraZeneca).

The characteristics of the patients are shown in Table 2. No significant differences were found in the age of the patients. Before starting the IVF procedure, the patients underwent a general clinical examination. Multiple indicators were recorded, including BMI, AMH levels, estradiol, progesterone, and prolactin. There were no significant differences in these parameters when comparing unvaccinated and vaccinated patients.

Table 2.

Characteristics of the unvaccinated and vaccinated patients

	Unvaccinated patients (n = 70)			Vaccinated patients (inactivated virus) (n = 25)			Vaccinated patients (viral vector) (n = 30)			p
	Mean	Min–Máx	SEM	Mean	Min–Max	SEM	Mean	Min-Máx	SEM
Age (years)	34.04	21–42	0.65	32.13	21–43	2.34	33.00	23–42	1.25	n.s
BMI (kg/m2)	23.02	19.38–29.1	1.34	23.79	18.81–29.06	1.34	24.41	18.42–35.86	1.91	n.s
AMH (ng/ml)	0.89	0.50–1.26	0.21	1.02	0.89–1.16	0.13	0.93	0.5–1.58	0.26	n.s
Estradiol (pg/ml)	43.38	13–63	3.49	43.20	29.10–65.50	11.28	42.23	19.01–54.20	4.24	n.s
Progesterone (ng/ml)	0.64	0.30–1.23	0.29	0.7256	0.20–1.25	0.25	0.65	0.40–0.90	0.09	n.s
Prolactin (ng/ml)	14.93	1–24	1.50	12.17	8.90–16.30	2.18	14.73	7.70–25.00	1.91	n.s

In the subsequent phase of the investigation, the objective was to assess the potential impact of COVID-19 vaccines administered in Argentina on female fertility. In particular, it was evaluated whether vaccination affected ovarian function by analyzing the number of oocytes retrieved and mature oocytes (M2). There were no significant differences between unvaccinated and vaccinated patients (Fig. 5A).

Fig. 5.

Effect of COVID-19 vaccines on: A the number of retrieved and mature oocytes in ART patients, B the number of oocytes retrieved and mature oocytes in unvaccinated patients vaccinated with inactivated virus or viral vector-based vaccines. In both groups, uninfected and infected patients were included. Data are expressed as mean ± SEM

We also set out to analyze whether the type of vaccine (inactivated virus or viral vector) affected ovarian function. Both the number of oocytes retrieved and the number of oocytes in M2 were not affected by the type of vaccine when compared to unvaccinated patients (Fig. 5B).

To assess the direct effect of vaccines on ovarian function, unvaccinated and vaccinated patients were compared, but only uninfected patients were included in both groups. No significant differences were found in the number of oocytes retrieved and the number of oocytes in M2 between both groups of patients (Fig. 6A). Finally, ovarian function was analyzed in unvaccinated and infected patients compared to vaccinated and infected patients, without finding significant differences in the number of oocytes retrieved and the number of oocytes in M2 between both groups of patients (Fig. 6B).

Fig. 6.

Effect of vaccines against COVID-19 on: A the number of oocytes retrieved and the number of mature oocytes in unvaccinated and vaccinated uninfected patients, B the number of oocytes retrieved and the number of mature oocytes in unvaccinated and vaccinated post-COVID-19 patients. Data are expressed as mean ± SEM

Discussion

In our previous research, we investigated the impact of SARS-CoV-2 infection on ovarian function, utilizing an endothelial cell line and a human granulosa cell line, as well as follicular fluids (FF) from unvaccinated control patients and individuals recovered from COVID-19 (post-COVID-19 FF) with mild symptoms. These patients were 2–9 months post-infection, unvaccinated against COVID-19, and undergoing ART [8]. These earlier findings indicated that SARS-CoV-2 infection negatively impacts the follicular microenvironment, disrupting ovarian function up to 9 months post-infection.

Building upon these results, our recent study aimed to assess whether ovarian function was restored after longer intervals post-infection (from 9 to 18 months post-infection). Here we showed for the first time that individuals with prior COVID-19 infection (mild symptoms), had a reduced number of recovered and mature oocytes up to 18 months post-infection. Moreover, our findings revealed that post-COVID-19 patients over 36 had a reduced number of recovered and mature oocytes up to 18 months post-infection, unlike those under 36.

In the female reproductive system, immune mediators—particularly macrophages and cytokines—are essential to the regulation of folliculogenesis, ovulation, and luteogenesis. Interleukin-1β (IL-1β), a multifunctional cytokine, plays a pivotal role in initiating and modulating processes related to ovulation, follicle development, and follicular atresia. Notably, IL-1β is not exclusively produced by intraovarian macrophages; it is also synthesized and secreted by oocytes, granulosa cells, theca cells, and cumulus cells within the human ovary.

Furthermore, IL-1β is subject to hormonal regulation, and its concentration rises during the periovulatory phase. Serum levels of IL-1β positively correlate with estradiol levels on the day of human chorionic gonadotropin (hCG) administration. Interestingly, no discernible correlation was established between IL-1β levels and the number of recovered or matured oocytes. Our study revealed that levels of IL-1β in follicular fluid (FF) continued to remain low up to 18 months post-infection regardless of vaccination. In contrast, levels of interleukin-10 (IL-10) remained unaltered across the patient groups. These results are consistent with our previous observations where IL-1β levels remained lower than in healthy patients up to 9 months post-infection [8].

Previously, we have shown reduced VEGF levels in FF from patients who had recovered from COVID-19 up to 9 months post-infection [8]. Nonetheless, the present study demonstrated the restoration of VEGF levels from 9 to 18 months post-infection compared to those observed in our previous study in patients up to 9 months post-COVID-19 infection. These results reveal a notable recuperation in VEGF levels in FF, indicating a return to baseline values comparable to those observed in the control group. The results contribute to a better understanding of the temporal dynamics of VEGF alterations post-COVID-19 infection, highlighting an encouraging trend toward normalization over an extended timeframe.

It is known that FF is a complex mixture of hormones, cytokines, metabolites, and other proteins primarily synthesized by granulosa cells [13]. Additionally, the composition of FF indirectly reflects the stage of oocyte development and its quality [15]. Therefore, if the composition of FF is altered, it can be associated with reduced reproductive function. Although we observed a restoration in VEGF levels, we observed that stimulation with FF from COVID-19-recovered patients caused reduced migration in endothelial cells compared to FF from control patients. Considering that blood vessels in the ovary are crucial for supplying oxygen and nutrients for oocyte growth, these results indirectly suggest that ovarian microvasculature would be affected in patients who experienced COVID-19 up to 18 months post-infection independently that VEGF levels are normalized. The altered endothelial migration could be due to the fact that there are other factors that are altered in the FF of post-COVID-19 patients, such as FGF or Notch/Dll4 systems that also modulate the ovarian microvasculature. All these results together suggest that the ovarian alterations caused by systemic and/or local SARS-CoV-2 infection extend up to 18 months post-infection. Further studies are needed over more extended post-infection periods to analyze the restoration of ovarian function.

H2AX, specifically phosphorylated at serine 139 (referred to as γH2AX), is a histone variant that plays a crucial role in the cellular response to DNA double-strand breaks (DSBs) [38, 39]. Upon the occurrence of DSBs, the serine 139 residue on the H2AX histone is phosphorylated by various DNA damage response kinases, such as ATM (ataxia-telangiectasia mutated). This phosphorylation event results in the formation of γH2AX foci at the sites of DNA damage. In particular, γH2AX is used to predict chronic inflammatory conditions that precede cancer as well as cardiovascular and nervous system disorders. Additionally, the entry of viral antigens can induce inflammation, making γH2AX a promising marker of viral infection (31). In our previous study, we showed that FF from recovered COVID-19 patients induced γH2AX levels compared with FF from control patients in endothelial cells [8]. In the present study, we observed that γH2AX levels did not vary in EA.hy926 cells incubated with FF from patients who experienced COVID-19 from 9 to 18 months post-infection. The observed results may potentially be attributed to a diminished synthesis of proinflammatory cytokines, thereby resulting in a reduction in the release of reactive oxygen species (ROS). This possible cascade effect could contribute to enhanced genomic stability within endothelial cells.

Systemic infection and even potential infection of the reproductive system require more attention because it not only affects the current generation but can also extend to future progeny through impaired gametes. To date, several studies have observed the effect of SARS-CoV-2 on both the male and female reproductive systems, including vertical transmission of the virus [40, 41], but a clear and conclusive report is still lacking, due to differences in study samples (post-infection period, type of COVID-19 symptoms, etc.), collection techniques, ethnicity, and, above all, the continued presence of SARS-CoV-2 infection despite vaccination. More research is required to elucidate in detail the timeframe required for the female gonad to restore its homeostasis after SARS-CoV-2 infection.

Several reproductive medicine centers in the USA and Israel have conducted studies that have affirmed the safety of mRNA-based vaccines concerning the reproductive system [23–27]. To date, in Argentina, no study has evaluated the impact of adenovirus-based vaccines (Sputnik V and AstraZeneca) and inactivated virus vaccines (Sinopharm) on the number of retrieved and mature oocytes from patients undergoing assisted reproductive treatments. In our study, we analyzed whether COVID-19 vaccines affect ovarian function. The results showed that the number of recovered oocytes and mature oocytes did not change in vaccinated patients compared to unvaccinated ones. Furthermore, no differences were observed in these parameters based on the type of vaccine, either based on inactivated virus or adenovirus. Taking into account whether patients had previously experienced COVID-19, it was also observed that the number of recovered oocytes and mature oocytes did not change in vaccinated patients compared to unvaccinated ones. These results suggest that inactivated virus-based vaccines (Sinopharm) and adenovirus-based vaccines (Sputnik V and AstraZeneca) do not affect ovarian function regardless of SARS-CoV-2 infection. More studies are needed in the future to assess the effect of these types of vaccines in a larger population. It is worth mentioning that Orvieto et al. (2021) have demonstrated that mRNA-based COVID-19 vaccines also do not affect embryo quality or reproductive success in patients undergoing ART [27, 42]. Bentov et al. (2021) have shown in ART patients that ovarian follicular function is not altered by mRNA-based vaccines [42].

Therefore, providing solid scientific evidence on the risks and benefits of COVID-19 vaccination is crucial for making informed decisions regarding vaccination. So far, results obtained in animal models [43], patients undergoing ART [25, 27, 42] clinical trials [44–46], along with our findings, provide robust insights, confirming that COVID-19 vaccines do not affect female fertility.

Conclusion

Both VEGF levels in follicular fluid (FF) and DNA integrity in endothelial cells incubated with FF from post-COVID-19 patients were restored to the levels observed in our previous study, which included patients up to 9 months post-infection, between 9 and 18 months post-infection. However, IL-1β levels in FF and endothelial cell migration in the presence of FF remained altered up to 18 months post-infection.

Moreover, the long-term psychological impact of COVID-19, including stress, anxiety, and depression, significantly influences reproductive health by affecting hormonal regulation and the hypothalamic-pituitary–gonadal axis, potentially worsening ovarian function in post-COVID patients. Therefore, our findings provide hope for this patient population.

Regarding the safety of inactivated virus vaccines (Sinopharm) and adenovirus-based vaccines (Sputnik V and AstraZeneca), it is conclusive that these vaccines exhibit no discernible impact on the quantity of retrieved and mature oocytes in patients undergoing ART. This observation applies both to patients without previous COVID-19 history and to those who have experienced mild COVID-19 symptoms, spanning from 2 to 18 months post-infection.

In conclusion, our results provide valuable insights into the field of reproductive healthcare, potentially leading to the enhancement of protocols for individuals who have recovered from COVID-19. These insights will be particularly beneficial for those seeking natural conception or requiring assistance through low or high-complexity assisted reproductive procedures.

Authors’ contributions

Conceptualization: D.A., F.P.; data curation: Y.H., L.S., C.V., V.H., I.Z., M.L., F.V., F.N., and G.M.; formal analysis: Y.H., L.S., C.V., V.H., F.V., M.L., and G.M.; investigation: Y.H., L.S., and V.H.; methodology: K.Z., A.K., P.W., and M.Z.; project administration: D.A. and F.P.; supervision: N.P.; validation: K.Z., A.K., and M.K.; writing—original draft: Y.H., L.S., and F.P.; writing—review and editing: Y.H., L.S., N.P., D.A., and F.P. All authors have read and approved the final manuscript.

Funding

This research received financing from Ferring Covid-19 Investigational Grant (2020); National Agency for Scientific and Technological Promotion (ANPCyT) (PICT 1603–2017); Ministry of Science, Technology, and Innovation; and the Baron and Williams Foundations, Argentina.

Data availability

Anonymized raw data will be provided upon request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Dr. Enrique T. Segura (IByME-CONICET) (protocol no. 2850). A written informed consent was obtained from all patients.

Conflict of interest

The authors declare no competing interests.