Seminal fluid extracellular vesicles restore human sperm osmoregulation after cystic fibrosis transmembrane conductance regulator inhibition

Abstract

Cystic fibrosis transmembrane conductance regulator (CFTR) is an anion channel that promotes the exchange of chloride (Cl⁻) and bicarbonate (HCO₃⁻), modulating the balance of ions and water in various tissues. CFTR is also expressed in the equatorial region of the spermatozoa’s head. However, the modulation of ionic and water transport via CFTR in spermatozoa needs further investigation. To better understand how CFTR participates in the molecular mechanisms behind spermatozoa modulation of water transport, and its impact on sperm physiology, CFTR function of human spermatozoa was inhibited with 20 µM CFTR_Inh-172. The inhibition of CFTR promoted a decrease in intracellular [Cl⁻] and membrane glycerol/water permeability, without any impact on sperm vitality or off-target effect in Ca²⁺ channels (assessed through the evaluation of intracellular [Ca²⁺]). To reverse the effects of CFTR inhibition, we incubated spermatozoa with CFTR-carrying seminal fluid extracellular vesicles (SF-EVs). We reported that the SF-EVs were able to restore the CFTR activity of spermatozoa previously exposed to the inhibitor, characterized by a restoration of Cl⁻, glycerol/water permeability, and a recovery of sperm osmoregulation capacity. Our results provide evidence for the important role of CFTR function in sperm osmoregulation and suggest that the use of CFTR-containing EVs - already employed in the treatment of cystic fibrosis - could be explored to enhance sperm quality.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-27409-8.

Introduction

Cystic fibrosis transmembrane conductance regulator (CFTR) belongs to the superfamily of ATP-binding cassette (ABC) transporters^1,2. The opening of the CFTR channel depends on its phosphorylation by the protein kinase A (PKA), an event dependent on the free concentration of cyclic adenosine monophosphate (cAMP)³. The CFTR function varies between tissues, but generally, CFTR mediates the efflux of chloride (Cl⁻) along with bicarbonate (HCO₃⁻)⁴. Through cooperation with neighboring transporters, CFTR controls the flux of water and other electrolytes, heading osmotic regulation through fluid transport to the luminal region of epithelia^4–7. Furthermore, thanks to its involvement in water and ion fluxes, CFTR participates in the regulation of cell volume^4,7, maintenance of the membrane potential, hormone secretion, muscle contraction, neuronal signaling, and calcium homeostasis^6,8,9.

In the male reproductive tract, more specifically within the seminiferous tubules, CFTR was detected in Sertoli cells¹⁰, spermatogonia, leptotene, and zygotene primary spermatocytes, pachytene primary spermatocytes, secondary spermatocytes, round spermatids, elongating and elongated spermatids¹¹. Its prominent presence in Sertoli cells and germ cells proposes the involvement of this channel in spermatogenesis. Although the role of CFTR in spermatogenesis remains largely elusive, CFTR knockout mice present a decreased sperm count and testicular weight¹². Concurrently, the inhibition of CFTR in a primary culture of rat Sertoli cells promoted a decrease in soluble adenylyl cyclase (sAC)-dependent cAMP production, and consequent phosphorylation of the cAMP-responsive element binding protein (CREB), two major modulators of spermatogenesis¹². CFTR activity was also found to be correlated with osmoregulation and glycerol permeability in mouse Sertoli cells. Furthermore, CFTR was detected to be close to aquaporins (AQPs) 3, 7, and 9, suggesting that a possible physical interaction between CFTR and aquaglyceroporins could modulate mouse Sertoli cell glycerol permeability¹³.

In human spermatozoa, CFTR was identified in the midpiece¹⁴ and equatorial head region¹⁵. CFTR was supposed to participate in sperm alkalinization during sperm capacitation in the uterine cavity. In this scheme, the CFTR channel would operate oppositely, driving HCO₃⁻ influx¹⁶. However, in a new proposed mechanism, HCO₃⁻ is not pumped into sperm through CFTR or other bicarbonate transporters, but is intracellularly produced through the reaction between CO₂ and water¹⁷. Nonetheless, the accumulation of HCO₃⁻ is the main pathway of stimulation for cAMP production, which in turn promotes protein phosphorylation and subsequent sperm hyperactive motility. In this new mechanism, CFTR is suggested to participate in the maintenance of the HCO₃⁻ cytoplasmic equilibrium¹⁷.

CFTR is also proposed to be involved in sperm volume regulation through its collaboration with AQPs. Several authors have correlated the function of Cl⁻ channels to human sperm volume regulation^18,19. Concurrently, AQPs were found to be key regulators of human sperm volume^20–23. In human sperm, the AQP3 and AQP7 were shown to participate in osmoregulation during epididymal transit and along the uterine hypotonic environment²⁴. Aquaglyceroporins promote entry of glycerol into spermatozoa, which is thought to act as an alternative non-conventional metabolic pathway²⁵, with a study reporting sperm motility linked to AQP7-mediated glycerol permeability²³. The inhibition of CFTR was found to impact mouse sperm volume regulation²⁶, and a correlation between CFTR function and AQPs function has already been reported in other cells^13,27. In 2024, it was reported that CFTR inhibition (via CFTR_Inh-172) promoted the loss of human sperm total motility in normozoospermia samples; the loss of sperm motility as follow by a disruption of glycerol diffusion and osmoregulation capacity; using a proximity ligation assay, it was reported that a potential functional interaction between CFTR-AQP7 in human spermatozoa, which could be another pathway in which CFTR is implied in sperm motility and osmoregulation²⁸.

The association of CFTR mutations with sperm quality (structure and function) has elicited controversial findings. Several reports evidenced an association of CFTR mutations with diminished sperm production in oligozoospermia²⁹, in severe oligozoospermia^30–33, in oligoasthenoteratozoospermia (OAT)^34,35, in obstructive azoospermia (OAZ)^30,36, and in nonobstructive azoospermia (NOA)^29–34,36. Nonetheless, it is also important to highlight that other studies did not find any correlation between CFTR mutations with severe oligozoospermia³⁷, OAT^37–39, or NOA^37,40.

From a different perspective, in vitro tests have shown that extracellular vesicles (EVs) can be used as a vehicle for the delivery of an exogenous CFTR glycoprotein and its encoding mRNA. The transfection of Calu-3 cells, a lung adenocarcinoma cell line, with an adenoviral vector overexpressing a green fluorescent protein (GFP)-tagged CFTR (GFP-CFTR) allowed for the isolation of EVs loaded with GFP-CFTR and CFTR mRNA⁴¹. When CF15 cells, a nasal epithelial cell line homozygous for the ΔF508 CFTR mutation, were incubated with the EVs produced by the Calu-3 cells, a correction of CFTR function was observed in CF15 cells in a dose-dependent manner⁴¹. Another study also reported that CFTR could be delivered via EVs to primary cultures of human airway epithelial cells and functionally restore the Cl⁻ channel⁴². In humans, EVs have been implicated in spermatogenesis⁴³, sperm maturation⁴⁴, sperm capacitation⁴⁵, oviduct gamete and embryo interaction^46,47, fertilization⁴⁸, embryo implantation and fetal development⁴⁹. As EVs from human seminal fluid (SF-EVs) were shown to be able to deliver compounds to the sperm⁵⁰, we wonder if we could use SF-EVs to treat spermatozoa with a defective function of CFTR. For this, we isolated SF-EVs from the human seminal fluid of normozoospermic men and analyzed whether these were able to restore the CFTR function of human spermatozoa with impaired (via inhibition) CFTR channel function.

Materials and methods

Ethical considerations

Ethical guidelines were followed for conducting the research, with written informed consent obtained before experiments. This work did not involve experimentation on humans or animals, and thus, the approval of the Ethics Committee and the Helsinki Declaration, revised in Tokyo 2004, on human experimentation does not apply to this work. Surplus semen samples were donated for research and came from assisted reproductive treatments at the Centro de Estudos de Infertilidade e Esterilidade (CEIE), Porto, Portugal. According to the determinations of the National Law of Medically Assisted Procreation (Law of 2006) and the guidelines of the National Council for Medically Assisted Procreation (CNPMA-2021), the use of clinical databases and biological material for research may be used without additional ethical approval, as long as the clinical data is used under strict individual anonymity and after informed and written consent from the patients. The study was further approved by the Joint Ethics Committee ICBAS/ULSSA under the code [2024/CE/P21(P432/2024/CETI)].

Isolation of seminal fluid extracellular vesicles

Surplus human semen samples were collected from 27 normozoospermic men attending fertility treatments. Semen samples were collected after a period of 2–7 days of abstinence, and semen parameters were assessed according to the World Health Organization (WHO) criteria⁵¹. Patient ages and semen characteristics are presented in Table 1. Semen samples were centrifuged at 500 g for 5 min to separate spermatozoa from the seminal fluid (SF). SF samples were then kept at -80 °C until further use. For later work, SF from normozoospermia samples were pooled together (SF pool, N = 27) and submitted to a differential centrifugation protocol to isolate EVs⁵².

Table 1.

Patient age and sperm-quality parameters of the samples that compose the seminal fluid (SF) pool.

Age (Years)	37 ± 5
Volume (mL)	3.9 ± 1.5
Concentration (106/mL)	118 ± 77
Total spermatozoa (106)	416 ± 230
Normal morphology (%)	8 ± 4
Total motility (%)	66 ± 7
Progressive motility (%)	56 ± 8
Vitality (%)	77 ± 7

Data is represented as mean ± standard deviation (SD), N = 27.

To start, the SF pool was centrifuged at 1000 g for 10 min at 4 °C, and the supernatant was then centrifuged at 20,000 g for 60 min at 4 °C to remove cellular debris and large cellular contaminants. Then, the SF pool was diluted with phosphate-buffered saline (PBS) in a 1:1 ratio and centrifuged at 20,000 g for 60 min at 4 °C. The dilution with PBS had the main objective of reducing the viscosity, allowing for the precipitation of smaller cellular debris, which could interfere with the results.

This purified SF pool was further centrifuged, first at 100,000 g for 3 h at 4° C (Beckman Coulter, California, USA). The resulting supernatant was further centrifuged at 200,000 g for 3 h at 4 °C. The resulting pellet of SF-EVs was resuspended in PBS.

Western blot

Pellets of SF-EVs and spermatozoa were resuspended in 200 µL of a 1% sodium dodecyl sulfate (SDS) solution and homogenized by vortexing. Protein lysates of SF-EVs and sperm were left to rest for 20 min at room temperature. Protein quantification of the lysates was performed with the Pierce Bicinchoninic acid protein assay kit (Thermo Fisher Scientific, Massachusetts, USA). A HeLa cell commercial lysate (Abcam, Cambridge, UK) was used as a positive control for calnexin and GM130 and processed equally as the remaining samples for western blot immunodetection. For validation of the SF-EVs isolation protocol, SF-EVs lysates (25 µg) were prepared for electrophoresis with a sample buffer (0.125 M Tris-HCl, pH 6.8; 4% SDS; 20% glycerol; 10% 2-mercaptoethanol).

Protein samples were loaded into a 10% TGX Stain-Free Acrylamide gel (Bio-Rad, Hercules, California, USA) and separated by size through an SDS-PAGE system (90 mV, for 90 min at RT). Through a Trans-Blot Turbo Transfer System (Bio‐Rad; 1.3 A, up to 25 V, 7 min), protein samples were transferred to polyvinylidene difluoride membranes. The membranes were blocked with 5% bovine serum albumin (BSA), 0.05% Tween 20 in Tris‐buffered saline solution (TBS) for 1 h at RT. Mouse anti-CD63 (10628D, Sigma-Aldrich, 1:1000) and mouse anti-CD9 (10626D, Sigma-Aldrich 1:500), specific for EVs membrane (positive markers), rabbit anti-calnexin (MA3-027, Thermo Fisher Scientific, 1:1000), specific for the smooth endoplasmic reticulum (negative marker) and rabbit anti-GM130 (ab52649, Abcam, 1:1500) specific for dictyosomes (negative marker), were diluted in 3% BSA in 0.05% Tween 20 TBS, overnight at 4 °C. On the next day, the membranes were washed with 0.05% Tween 20 in TBS at RT, and incubated with the correspondent peroxidase-conjugated secondary antibody [goat anti-mouse (AP308P, Millipore, Sigma-Aldrich); mouse anti-rabbit (AP307P, Millipore, Sigma-Aldrich)], diluted with 3% BSA, 0.05% Tween 20 in TBS, for 1 h at RT. Finally, membranes were washed with 0.05% Tween 20 in TBS at RT, and results were visualized in a ChemiDoc MP Imaging system (Bio‐Rad, Hercules, California, USA).

A similar protocol was performed to verify the presence of CFTR in SF-EVs and spermatozoa. For this, protein samples (60 µg) were prepared in the same way as described previously. Then, protein lysates were loaded into a 7.5% TGX Stain-Free Acrylamide gel and separated by size through an SDS-PAGE system (90 mV for 115 min, at RT). The transference protocol used on the Trans-Blot Turbo Transfer System was the following: 1.3 A, up to 25 V, 13 min. The mouse anti-CFTR (sc-376683, Santa Cruz, California, USA) was diluted with 3% BSA, and 0.05% Tween 20 in TBS to a proportion of 1:100. Protein loading of CD9 and CD63 membranes was assessed through TGX free staining (a fluorescent protein marker developed by BioRad). Protein loading of calnexin, GM130, and CFTR membranes was assessed through Ponceau S-staining. Uncut membrane versions are provided in the supplementary data (Figs. S1 and S2).

Extracellular vesicles size characterization and zeta potential assessment

SF-EVs pellets were resuspended in 100 µL of PBS and then diluted to a proportion of 1:50,000 with vesicle-free distilled water. Nanoparticle tracking analysis (NTA) was performed in a NanoSight NS300 (Malvern Instruments, Worcestershire, UK), equipped with a 488 nm laser and a high-sensitivity scientific CMOS camera. The samples were loaded into the viewing chamber with sterile syringes (1 mL) under a constant flow and infusing rate of 40 (equipment arbitrary units), at RT. For detailed nanoparticle size distribution, 5 videos of 20 s each were captured per sample (camera level of 15 and a detection threshold of 5). The obtained SF-EVs size distribution data was calculated from three independent assays (N = 3). Data were analyzed using the NTA 3.4. software (Malvern Instruments, Worcestershire, UK). The zeta potential of SF-EVs was assessed using Zeta-Pals (Brookhaven Instruments, Holtsville, New York, USA), operating at a scattering angle of 90° and room temperature (25 °C). The SF-EVs were diluted in vesicle-free distilled water (pH 7.4) to a proportion of 1:5000. The zeta potential values obtained were calculated from three independent assays (N = 3), with 6 measurements for each replica being performed. The results are represented as millivolts (mV).

Fluorescence staining of extracellular vesicles and CFTR immunolabelling

SF-EVs were probed with PHK67 green-fluorescent membrane linker using the PKH67 Fluorescent Cell Linker kit (PKH67GL, Sigma-Aldrich) according to the manufacturer’s instructions. Briefly, SF-EVs pellets were resuspended in 500 µL of diluent C. To this suspension, 2 µL of the PHK67 linker (which intercalates between membrane phospholipids) was added and continuously mixed for 30 s at RT. Then, suspensions were left to incubate for 15 min at 37 °C. To stop the labeling reaction, 2 volumes (1 ml) of 1% BSA in PBS were added. Then, SF-EVs were isolated with the Total Exosome Isolation reagent (Invitrogen, Thermo Fisher Scientific) according to the manufacturer’s instructions. Briefly, 0.5 volumes (750 µl) of the reagent were added to the suspension, which was left to incubate under rotation for 1 h at 4° C. Then, the samples were centrifuged at 10,000 g for 1 h at 4° C, and the pellet of SF-EVs was resuspended in PBS.

We exposed 30 × 10⁶ spermatozoa to labeled SF-EVs for 1 h at 37 °C in PBS. For the negative control, 30 × 10⁶ spermatozoa were exposed to unmarked SF-EVs. Afterward, spermatozoa were centrifuged at 500 g for 5 min and resuspended in 50 µl of PBS. A smear was prepared using 10 µL of the sperm suspension. The microscope slide was left to incubate in a 10% formalin solution for 20 min at RT and then washed with PBS. For cell permeabilization, microscope slides were left to incubation with a 5% BSA 0.1% Triton X-100 solution for 1 h at room temperature. After washing the slides with PBS, the cells were incubated with rabbit anti-CFTR primary antibody (1:50 dilution in 5% BSA, ab131553, Abcam) overnight at 4° C. For the negative control, cells were incubated overnight with 5% BSA solution. Afterwards, the slides were washed with PBS, and spermatozoa were exposed to Alexa Fluor 555 anti-rabbit IgG (1:100 dilution in 5% BSA, #4413, Cell Signaling Technology, Massachusetts, USA) for 1 h at room temperature. After washing in PBS, the microscope slides were submerged in a 5 µg/mL Hoechst 33342 solution to stain the nuclei. A drop of VectaShield mounting medium (Vector Laboratories, Burlingame, USA) was used to mount the coverslip on the microscope slide, and the edges were sealed with varnish to prevent drying.

The results were visualized in a Nikon Eclipse E400 microscope equipped with a CI-FL-2 Epi-Fluorescence Attachment, and a pE-300 Lite Multi-Band Power Supply (Nikon, Shinagawa, Tokyo, Japan), coupled with Nikon NIS Elements Image Software.

Inhibition of CFTR activity of human spermatozoa

After spermatozoa separation from the SF, the spermatozoa pellet was washed with PBS by centrifuging at 500 g for 5 min. Then, spermatozoa were resuspended in PBS (adjusted to 295 mOSm). For CFTR inhibition, 20 µM CFTR_Inh-172 [final dimethyl sulfoxide (DMSO) dilution of 0.2%] was added to the sperm suspension and left to incubate for 10 min at 37 °C. More specifically, to every 10 × 10⁶ cells resuspended in 250 µL of PBS, 0.5 µL of inhibitor was added, ensuring a final concentration of 20 µM CFTR_Inh-172 and 0.2% DMSO. This proportion was adjusted to the number of cells required for each technique, ensuring a consistent inhibition protocol throughout the entire work. Then, spermatozoa were centrifuged at 500 g for 5 min for inhibitor washout. Spermatozoa with compromised CFTR function (via inhibition) were resuspended in PBS for 60 min at 37 °C. Sperm vitality (N = 10) was evaluated before (control group – CTR), immediately after CFTR function inhibition (0 min after inhibitor washout), and at the end of incubation in PBS (60 min after inhibitor washout). The intracellular [Cl⁻] variation (N = 10), and glycerol and water permeability (N = 10) were also evaluated through the fluorescence quantification of a Cl⁻ -specific fluorescent probe, and Stopped-flow light scattering (SFLS), respectively. Additionally, we also explored the intracellular [Ca²⁺] variation (N = 10) of sperm with compromised CFTR function (via inhibition), to rule out possible non-specific effects of the inhibitor.

In the second experiment, spermatozoa with compromised CFTR function (via inhibition) were resuspended in PBS or SF-EVs suspension for 60 min at 37 °C. Sperm vitality (N = 10), intracellular [Cl⁻] levels variation (N = 10), and glycerol and water permeability (N = 10) were evaluated.

Sperm vitality

Sperm vitality was assessed with eosin-nigrosine staining. A total of 200 spermatozoa were counted per slide in continuous fields and evaluated blindly. A bright-field optical microscope (Nikon Eclipse Ci, Shinagawa, Tokyo, Japan) was used to assess sperm vitality. Results are represented as fold-variation to sperm with preserved CFTR function, used as control (CTR).

Intracellular [Cl⁻] measurement in human spermatozoa

The intracellular [Cl⁻] levels in sperm were measured using a Cl⁻ specific fluorescence probe, N-(ethoxycarbonylmethyl)-6-methoxyquinolinium bromide (MQAE, 46123, Sigma-Aldrich, Missouri, USA). We exposed 30 × 10⁶ spermatozoa to 5 mM of MQAE and 0.02% Pluronic acid in PBS for 30 min at 37° C in the dark. Then, 10 × 10⁶ MQAE-loaded spermatozoa were exposed to 0.2% DMSO (vehicle), while the remaining 20 × 10⁶ MQAE-loaded spermatozoa were exposed to 20 µM CFTR_Inh-172 for 10 min at 37° C. After incubation, spermatozoa were washed 2 times (500 g, 5 min centrifugation, followed by pellet resuspension in PBS) to ensure both MQAE and CFTR_Inh-172 washed out. The control (10 × 10⁶ MQAE-loaded spermatozoa treated only with the vehicle) was resuspended in PBS to a final volume of 200 µL. Meanwhile, the 20 × 10⁶ MQAE-loaded spermatozoa with inhibited CFTR function by CFTR_Inh-172 were divided into pellets of 10 × 10⁶ cells and resuspended in 200 µL of PBS and SF-EVs suspension, respectively. The samples were loaded into a 96-well dark microplate with a clear bottom, and the fluorescence intensity at 350 nm excitation wavelength and 460 nm emission wavelength was obtained with a microplate reader (SpectraMax iD3, Molecular Devices, California, USA) at 37° C. The data was acquired at the beginning of incubation (0 min) and at the end (60 min). As the fluorescence intensity of MQAE is inversely proportional to intracellular Cl⁻ levels, the variation of the intracellular Cl⁻ levels was calculated by subtracting MQAE fluorescence at the end of incubation (F) from MQAE fluorescence at the beginning of incubation (F0):

Results are represented as fold-variation to sperm with preserved CFTR function, used as control (CTR).

Stopped-flow light scattering (SFLS)

SFLS was performed to measure the membrane permeability of human sperm to glycerol, using a Stopped-Flow SX20 apparatus (Applied Photophysics, Leatherhead, Surrey, UK). The SFLS evaluates the alterations in cell volume. As aquaglyceroporins transport water and glycerol in a symport way, changes in cell volume reflect glycerol permeability. The osmolarity of all solutions was tested and adjusted accordingly with an Osmometer Basic (Löser, Berlin, Germany), using standards of 300 and 900 mOsm. (C2992, Sigma-Aldrich, Missouri, USA). In every experiment, a sperm suspension of 30 × 10⁶/mL was used to register the initial exponential rate coefficient of the volume change before CFTR inhibition. In the first experiment, 60 × 10⁶ spermatozoa were exposed to 20 µM CFTR_Inh-172 (0.2% DMSO) for 10 min at 37° C. After the inhibitor washout (centrifugation at 500 g for 5 min), spermatozoa were kept in PBS for 60 min. The exponential rate coefficient of the volume change was measured immediately after CFTR inhibition and 60 min after. In the second experiment, spermatozoa with inhibited CFTR function (60 × 10⁶) were resuspended in PBS or SF-EVs suspension, 30 × 10⁶ for each condition, and left to incubate for 60 min at 37 °C. After incubation, spermatozoa were centrifuged at 500 g for 5 min for SF-EVs washout, and were resuspended in 30 × 10⁶/mL PBS. The exponential rate coefficient of the volume change after CFTR inhibition was then measured for each condition.

In every run, 90° scattered light intensity at 450 nm was measured at every 1 ms, for a total of 1 s-time, where we met a stable light scatter signal. Ten runs were performed for each experimental condition, and a minimum of seven runs were used for further analysis. The osmotic shock was performed with a hyperosmotic glycerol solution (250 mM glycerol, 500 mOsm). In each run, 100 µL of sperm suspension (30 × 10⁶/mL) was mixed with an equal amount of the hyperosmotic glycerol solution. The exponential rate coefficient of the volume change was calculated through the SX Pro-Data software, which was assumed to be proportional to sperm membrane glycerol permeability. A logarithm transformation [Y = Log10(X)] was applied to the raw data to reduce the skewness of the distribution. Results are represented as fold-variation to the initial exponential rate coefficient of sperm membrane glycerol permeability in isosmotic (295 mOsm) PBS, used as control (CTR).

Intracellular [Ca²⁺] measurement in human spermatozoa

The intracellular [Ca²⁺] levels in sperm were measured using a Ca²⁺-specific fluorescence probe, Fluo-4 AM (F14217, ThermoFisher Scientific, Massachusetts, USA). We exposed 20 × 10⁶ spermatozoa to 1 µM of Fluo-4 AM and 0.02% Pluronic acid in PBS for 30 min at 37° C in the dark. Then, 10 × 10⁶ Fluo-4 AM loaded-spermatozoa were exposed to 0.2% DMSO (vehicle), while the remaining 10 × 10⁶ Fluo-4 AM loaded-spermatozoa were exposed to 20 µM CFTR_Inh-172 for 10 min at 37° C. After incubation, spermatozoa were washed 2 times (500 g, 5 min centrifugation, followed by pellet resuspension in PBS) to ensure both Fluo-4 AM and CFTR_Inh-172 washed out. The control (10 × 10⁶ Fluo-4 AM-loaded spermatozoa treated only with vehicle) and the inhibition group (10 × 10⁶ Fluo-4 AM-loaded spermatozoa treated only with 20 µM CFTR_Inh-172) were resuspended in PBS with Ca²⁺ and Mg²⁺ (0.90 mM calcium chloride and 0.50 mM magnesium chloride) to a final volume of 200 µL. The samples were loaded into a 96-well dark microplate with a clear bottom, and the fluorescence intensity at 488 nm excitation wavelength and 520 nm emission wavelength were obtained with a microplate reader (SpectraMax iD3) at 37° C. The data was acquired at the beginning of incubation (0 min) and at the end (60 min). Variation of the intracellular Ca²⁺ levels was calculated by subtracting Fluo-4 AM fluorescence at the beginning of incubation (F0) from Fluo-4 AM fluorescence at the end of incubation (F):

Results are represented as fold-variation to sperm with preserved CFTR function, used as control (CTR).

Statistical analysis

Sperm vitality passed the Shapiro–Wilk test and Kolmogorov-Smirnov test and were treated as normal distributions. Differences between the CTR and the inhibition of CFTR through time (0 and 60 min), and CTR and SF-EVs for reversing the effects of CFTR inhibition were tested using repeated measures (RM) One-way ANOVA for paired analysis (Geisser-Greenhouse correction), and Dunnett test for multiple comparisons. Sperm vitality data is expressed as fold-variation (FV) to control (CTR), mean ± standard deviation (SD).

Intracellular [Cl⁻] variation and intracellular [Ca²⁺] variation data passed the Shapiro–Wilk test and Kolmogorov-Smirnov test and were treated as normal distributions. Results are represented as FV to CTR, as mean ± SD. The difference between the CTR and CFTR inhibition group (20 µM Inh) was tested by Paired T-test. Differences regarding intracellular [Cl⁻] variation between CTR and SF-EVs for reversing the effects of CFTR inhibition were tested using RM One-way ANOVA for paired analysis (Geisser-Greenhouse correction) and Dunnett test for multiple comparisons.

Membrane glycerol permeability data passed the Shapiro–Wilk test and Kolmogorov-Smirnov test and was treated as normal distributions. Results are represented as expressed as FV to CTR, mean ± SD. Differences between the CTR and the inhibition of CFTR through time (0 and 60 min), and CTR and SF-EVs for reversing the effects of CFTR inhibition were tested using RM One-way ANOVA for paired analysis (Geisser-Greenhouse correction) and Dunnett test for multiple comparisons.

All statistical tests were conducted with 95% confidence (α = 0.05). GraphPad Prism 8.0.1 (GraphPad Software Inc., San Diego, CA, USA) was employed for the statistical analysis in this study.

Results

Purification and characterization of seminal fluid extracellular vesicles

To confirm that the pellet isolated from SF was, in fact, EVs, Western Blot was used to immunodetect two EVs membrane markers and two EVs negative markers, as proposed by the Minimal Information for Studies of Extracellular Vesicles (“MISEV”) guidelines⁵³. As positive markers, CD63 and CD9 were used. CD63 belongs to the tetraspanin subfamily and is known to participate in intracellular trafficking, being present in late endosomes and lysosomes^54,55. CD9 has a variety of biological functions, from motility⁵⁶ to cell adhesion and migration⁵⁷. CD9 was also found to be involved in sperm-egg fusion through interaction with the pregnancy-specific glycoprotein 17⁵⁸ and beta 1 integrins⁵⁹. For negative markers, calnexin and GM130 were selected. GM130 is a cis-Golgi matrix protein⁶⁰, and calnexin is a type-I integral membrane protein from the endoplasmic reticulum^61,62. These proteins are commonly used as negative markers for EVs since they are intimately associated with specific cellular organelles that may end up precipitating at the same rate as EVs. Results confirmed the presence and purity of SF-EVs (Fig. 1A and B).

Fig. 1.

Seminal fluid extracellular vesicles (SF-EVs) characterization. (A and B) Validation of the SF-EVs isolation method by immunoblotting; (A) CD63 and CD9 were used as positive biomarkers for extracellular vesicles, (B) calnexin and GM130 were used as negative biomarkers for extracellular vesicles; a commercial HeLa protein lysate was used as a positive control for calnexin and GM130; uncut western-blot membranes and loading controls are provided at the supplementary data (Fig. S1). EVs - seminal fluid extracellular vesicles protein lysate; EVs1 (isolated from a different seminal fluid pool); EVs2 (seminal fluid pool described in Table 1); HeLa –commercial HeLa cells protein lysate. (C) Nanoparticle size distribution of seminal fluid extracellular vesicles (isolated from the seminal fluid pool described in Table 1), results are represented as mean (N = 3).

Thereafter, the size, concentration, and electrical charge of SF-EVs were determined. The SF-EVs’ size and concentration were evaluated by NTA. The concentration/size profile for the SF-EVs is represented in Fig. 1C. The mean concentration of particles in the SF-EVs suspension was 2.35 × 10¹² ± 9.38 × 10¹⁰ particles/mL (Table 2). The vast majority of SF-EVs fall within the 20–300 nm range of hydrodynamic diameter (HD) (Fig. 1C), and the HD mean was 169 ± 94 nm. The most common particle of the suspension had an HD of 92 ± 23 nm (Table 2). Regarding the zeta potential, SF-EVs presented a neutral (zwitterionic) charge (between − 30 mV − 30 mV)⁶³, with a mean charge of -5.85 ± 0.86 mV (Table 2). This result was expected as EVs derive from the cytoplasmic membrane and thus have a typical range charge.

Table 2.

Characterization of seminal fluid extracellular vesicles.

Concentration (particle/mL)	2.35 × 1012 ± 9.38 × 1010
Mean HD (nm)	169 ± 94
Mode (nm)	92 ± 23
Zeta potential (mV)	-5.85 ± 0.86

Data is represented as mean ± standard deviation (SD), N = 3.

CFTR is present in the extracellular vesicles of the seminal fluid and in spermatozoa

The former proteomic analysis confirmed the presence of protein CFTR in SF-EVs by two-dimensional liquid chromatography-tandem mass spectroscopy (2D LC-MS/MS)⁶⁴. Based on these observations, we thereafter evaluated the presence of CFTR in protein lysates from SF-EVs and spermatozoa, through Western blot using a monoclonal antibody for CFTR. The SF-EVs presented at least two protein bands of CFTR, a 130 kDa protein band that corresponds to the native form of CFTR with no glycosylation⁶⁵, and a 150 kDa protein band that corresponds to a partially glycosylated protein band of CFTR⁶⁵. The fully glycosylated form of CFTR was found in human spermatozoa lysates, represented as a band of 180 kDa (Fig. 2). We also confirmed that SF-EVs could interact with spermatozoa in our experimental conditions. For that, we used the PHK67 green-fluorescent membrane linker (membrane intercalating dye) to mark the SF-EVs membrane. Spermatozoa were then exposed to fluorescent SF-EVs, and, after washing, samples were mounted on microscope slides and observed under a fluorescent microscope. Results showed an overlap of fluorescent SF-EVs with the sperm head region. CFTR is also naturally present in the human spermatozoa head, more specifically in the equatorial region (Fig. 3A). Ultimately, our results suggest SF-EVs integration in the sperm membrane or membrane interaction between SF-EVs and sperm.

Fig. 2.

Identification of CFTR in seminal fluid extracellular vesicles (SF-EVs). Identification of CFTR in SF-EVs and CFTR by Western-blot; Total protein was assessed by Ponceau S staining and used as loading control; Uncut Western-blot membranes and loading controls are provided in the supplementary data (Fig. S2). EVs - seminal fluid extracellular vesicles protein lysate; EVs1 (isolated from a different seminal fluid pool); EVs2 (seminal fluid pool described in Table 1); SPZ – human spermatozoa protein lysate.

Fig. 3.

Fluorescence staining of human spermatozoa interacting with SF-EVs; (A) Human spermatozoa incubated with PHK67-marked SF-EVs; nuclei is stain with Hoechst 33342 (blue), SF-EVs PHK67 green-fluorescent cell membrane intercalating labeling dye (green), and CFTR was immunolabeled with Alexa Fluor 555 (red). (B) Negative control – human spermatozoa were incubated with unstained SF-EVs, immunolabeling was performed only with the secondary antibody (Alexa Fluor 555). Scale Bar is equal to 100 μm.

The inhibition of CFTR activity does not impact sperm vitality but decreases intracellular chloride and membrane glycerol permeability

The absence or malfunction of CFTR results in impaired fluid and ionic homeostasis in the whole organism. To mimic a CFTR malfunction in spermatozoa, sperm were incubated with a CFTR inhibitor, 20 µM CFTR_Inh-172, for 10 min at 37 °C. After the inhibitor was washed out, sperm were kept in PBS at 37 °C (0 and 60 min). This CFTR inhibition protocol was confirmed not to promote a cytotoxic effect, as no differences were detected regarding sperm vitality before (CTR: 1.00 ± 0.08 FV) and after CFTR inhibition (0 min after inhibitor washout: 0.93 ± 0.09 FV, Fig. 4A). Sperm vitality remained unaltered 60 min after the inhibitor washout (0.90 ± 0.08 FV, Fig. 4A).

Fig. 4.

The inhibition of CFTR activity does not impact sperm vitality but decreases intracellular chloride and membrane glycerol permeability. Effects on (A) Sperm Vitality (N = 10) were tested by using RM One-way ANOVA for paired analysis (Geisser-Greenhouse correction), and Dunnett test for multiple comparisons; results are represented as fold-variation to CTR, mean ± SD. (B) Effects on intracellular [Cl⁻] variation, and (C) intracellular [Ca²⁺] variation were tested by Paired T-test (N = 10); results are represented as fold-variation to CTR, mean ± SD. (D) The mean value of light scattering variations associated with sperm volume change in response to a glycerol hyperosmotic solution (N = 10); (E) Effects on membrane glycerol permeability (exponential rate coefficient) of spermatozoa (N = 10) were tested by RM One-way ANOVA for paired analysis (Geisser-Greenhouse correction), and Dunnett test for multiple comparisons; results are represented as fold-variation to CTR, mean ± SD; Individual values are represented as ●. All differences compared to CTR; p < 0.05 was considered significant.

The effect of CFTR inhibition on the intracellular [Cl⁻] was evaluated by measuring the fluorescence intensity of a Cl⁻ specific fluorescence probe, MQAE. MQAE-loaded sperm with preserved (CTR) and inhibited (20 µM Inh) CFTR function were incubated at 37 °C in PBS for 60 min; fluorescence was measured at the beginning and the end of incubation. The variation of MQAE fluorescence was considered proportional to the variation of intracellular [Cl⁻], which was significantly decreased in the CFTR-inhibition group (CTR: 1.00 ± 0.30 vs. 20 µM Inh: 0.72 ± 0.24, p = 0.0100, Fig. 4B) when compared to the control.

To rule out the hypothesis that 20 µM CFTR_Inh-172 could impose off-target effects in Ca²⁺ channels, the intracellular variation of [Ca²⁺] was evaluated by measuring the fluorescence intensity of a Ca²⁺ specific fluorescence probe, Fluo-4 AM. As previously, Flou-4 AM -loaded sperm with preserved (CTR) and inhibited (20 µM Inh) CFTR function were incubated at 37 °C in PBS for 60 min; fluorescence was measured at the beginning and the end of incubation. The variation of Fluo-4 AM fluorescence was considered proportional to the variation of intracellular [Ca²⁺], and no difference was detected between the control and CFTR-inhibition group (CTR: 1.00 ± 0.26 vs. 20 µM Inh: 1.13 ± 0.52, Fig. 4C).

The effect of CFTR sperm inhibition on glycerol permeability was then tested using SFLS. The control follows a typical curve of response to a hyperosmotic glycerol shock; at first, water moves out of the cell through aquaporins, flowing from an area of lower osmolarity (inside the cell) to higher osmolarity (outside). As a result, sperm lose water and shrink. In a secondary response, aquaglyceroporins promote the entry of glycerol into the cell according to the concentration gradient; this process increases intracellular osmolarity, promoting the re-entering of water and restoration of cell volume (Fig. 4D). However, the inhibition of CFTR function promotes a significant loss of sperm membrane glycerol permeability, severely impacting the re-entry of water (Fig. 4D). The exponential coefficient of the volume change was assumed to be proportional to sperm membrane glycerol permeability. CFTR inhibition caused a significant decline of membrane glycerol permeability [CTR: 1.00 ± 0.06 FV vs. 0 min: 0.28 ± 0.22 FV, p = 0.0007, Fig. 4E). The effects of CFTR inhibition in sperm membrane glycerol permeability were still present 60 min after the inhibitor washout (60 min: 0.27 ± 0.14 FV, p = 0.0035 compared to CTR, Fig. 4E).

These results lead us to conclude that CFTR function inhibition does not promote cytotoxic effects in spermatozoa, as represented by the preservation of sperm vitality 60 min after inhibitor washout. Nonetheless, the inhibition of CFTR function leads to a decrease in intracellular [Cl⁻] and membrane glycerol permeability, with no off-target effects regarding Ca²⁺ channels.

Extracellular vesicles from seminal fluid can restore intracellular [Cl⁻] and membrane glycerol permeability after CFTR inhibition

We evaluated whether the CFTR-containing SF-EVs were able to restore the channel activity of spermatozoa with compromised CFTR function via inhibition. For that, spermatozoa were incubated with 20 µM CFTR_Inh-172 for 10 min at 37 °C, followed by centrifugation at 500 g for 5 min to wash out the inhibitor. Spermatozoa were resuspended in PBS or SF-EVs suspension and kept at 37 °C for 60 min. As sperm vitality remained unaltered, we confirmed that sperm incubation with SF-EVs did not impose any cytotoxic effects (CTR: 1.00 ± 0.09 FV, PBS: 0.92 ± 0.12 FV, EVs: 1.01 ± 0.06 FV, Fig. 5A).

Fig. 5.

The inhibition of CFTR in spermatozoa is reversed by seminal fluid extracellular vesicles (SF-EVs). (A) Effects of SF-EVs on sperm vitality (N = 10) were tested by RM One-way ANOVA for paired analysis (Geisser-Greenhouse correction), and Dunnett test for multiple comparisons; results are represented as fold-variation to CTR, mean ± SD. (B) Effects on intracellular [Cl⁻] variation (N = 10) were tested using RM One-way ANOVA for paired analysis (Geisser-Greenhouse correction), and Dunnett test for multiple comparisons, results are represented as fold-variation to CTR, mean ± SD. (C) Mean value of light scattering variations associated with volume change in response to a glycerol hyperosmotic solution of spermatozoa before CFTR inhibition (CTR) and after CFTR inhibition and incubation with SF-EVs (N = 10); (D) Effects on membrane glycerol permeability (exponential rate coefficient) were tested RM One-way ANOVA for paired analysis (Geisser-Greenhouse correction), and Dunnett test for multiple comparisons; results are represented as fold-variation to CTR, mean ± SD. Individual values are represented as ●. All differences are compared to CTR; p < 0.05 was considered significant.

As previously described, we observed a significant decrease in intracellular [Cl⁻] in sperm with compromised (via inhibition) CFTR function (PBS: 0.72 ± 0.24 FV) in comparison to sperm with preserved CFTR function, used as control (CTR: 1.00 ± 0.30 FV, p = 0.0199, Fig. 4B). Interestingly, sperm with compromised (via inhibition) CFTR function treated with SF-EVs presented a restoration of intracellular [Cl⁻] to levels similar to the control (EVs: 1.01 ± 0.43 FV, Fig. 5B). Similar results were found concerning sperm osmoregulation capacity and glycerol permeability. We observed that, when exposed to a hyperosmotic glycerol solution, control spermatozoa and sperm with compromised CFTR function treated with SF-EVs presented typical response curves for volume changes evaluated by SFLS (Fig. 5C). Meanwhile, sperm with compromised CFTR function in PBS presented a significant loss of sperm membrane glycerol permeability, which severely impacted the re-entry of water into the cells (Fig. 5C). This is better represented by the exponential rate coefficient of the volume change, which is assumed to be proportional to the sperm membrane glycerol permeability. A significant loss of sperm membrane glycerol permeability was observed after CFTR inhibition (CTR: 1.00 ± 0.08 FV vs. PBS: 0.25 ± 0.23 FV, p < 0.0001, Fig. 5D). Meanwhile, the incubation of spermatozoa with compromised CFTR function with SF-EVs significantly restored sperm membrane glycerol permeability to control levels (EVs: 1.11 ± 0.23 FV, Fig. 5D).

Discussion

In this work, we incubated human spermatozoa with the CFTR_Inh-172, intending to mimic spermatozoa from an individual with impaired channel function. CFTR_Inh-172 is a thiazolidinone compound that binds to the extracellular side of CFTR, more specifically near transmembrane helix 8, a structural element that links to the adenosine triphosphate hydrolysis with channel gating⁶⁶. This inhibitor is considered a better option to perform CFTR focus assays, since it presents less off-target effect than other popular options, such as glibenclamide⁶⁷ and the glycine hydrazide (GlyH-101)⁶⁸. CFTR_Inh-172 is an extracellular inhibitor that affects channel gating. CFTR_Inh-172 binds inside the channel, interacting with the transmembrane segments 1, 6, 8, 9, and 12 mostly through hydrophobic interactions⁶⁹. Additionally, the interaction with the transmembrane segment 8 of TMD2 promotes a conformational shift of the NBD domains and collapse of the chloride selectivity filter, inhibiting ATP hydrolysis through an allosteric mechanism⁶⁶.

Still, CFTR_Inh-172 is confirmed to have an off-target effect on the interference with chloride channel 2 (ClC-2), although the molecular mechanism by which this inhibition occurs has never been published⁷⁰. It was previously confirmed that human spermatozoa do not express ClC-2, which means that this off-target effect of CFTR_Inh-172 has no impact on our model⁷¹.

A 2010 study proposed that CFTR_Inh-172 could also present an off-target effect in mitochondrial functions⁷², but later reports have refuted this hypothesis^66,73. The CFTR_Inh-172 was also reported not to exert effects on the Ca²⁺-dependent Cl⁻ conductance (CaCC)⁷⁴.

A 2012 study exploring the response of different species CFTR orthologs (human, killifish, pig, and shark) to specific inhibitors of the channel reported that the human CFTR presented the better responsiveness to CFTR_Inh-172, with an observed 65.5% inhibition at a concentration of 20 µM; the IC50 for human CFTR was calculated to be 10.4 µM and a maximal percentage of inhibition (73.3%) was detected at 25 µM of CFTR_Inh-172⁷³. When applied to human spermatozoa studies, CFTR_Inh-172 concentration varies considerably (in µM): 0.1⁷⁵, 5⁷⁶, 25⁷⁷, and 60⁷⁸. Ultimately, we settle on the concentration of 20 µM for two main reasons: 20 µM CFTR_Inh-172 is known to induce a significant inhibition percentage of 65.5% in human CFTR⁷³, and similar concentrations (specifically, 24 µM and 25 µM) had already been used in human spermatozoa studies^77,78 with no reported cytotoxic effects.

Herein, it was crucial to ensure that the CFTR function remains inhibited to assess the potential of SF-EVs to recover the channel function. For this, we incubated spermatozoa with 20 µM CFTR_Inh-172 for 10 min (CFTR_Inh-172 is a rapid-onset inhibitor, with full inhibition seen in 3–5 min⁷³), followed by the inhibitor washout (via centrifugation) and resuspension of the spermatozoa pellet in an inhibitor-free environment. CFTR function was indirectly assessed by two hallmarks of CFTR inhibition in human spermatozoa: first, assessing the variation of sperm intracellular [Cl⁻] via a fluorescent probe, and then evaluating osmoregulation upon a glycerol hyperosmotic shock²⁸.

Lower variation of intracellular [Cl⁻] was found in sperm with inhibited CFTR function as compared to sperm with preserved channel function, suggesting that 20 µM CFTR_Inh-172 treatment was successful. Having confirmed that sperm Cl⁻ permeability was affected by the CFTR inhibition protocol, we assessed its consequences on sperm membrane glycerol permeability, which was used in this study as a parameter to evaluate sperm osmoregulation capacity^13,79. In human sperm, CFTR is supposed to regulate the osmotic equilibrium through interaction with AQP7, which simultaneously drives the flux of glycerol and water²³. Treatment with CFTR_Inh-172 resulted in a significant decrease in sperm membrane glycerol permeability, and this reduction persisted for at least 60 min after inhibitor washout. Our results are in accordance with a previous 2024 study that reported that CFTR inhibition (via CFTR_Inh-172) promoted the disruption of glycerol diffusion and osmoregulation capacity of human spermatozoa; using a proximity ligation assay, it was reported that a potential functional interaction between CFTR-AQP7²⁸. We believe this is the first report demonstrating the persistence of these effects over time. This prolonged effect of CFTR_Inh-172 is likely associated with its strong hydrophobic nature, thanks to its three aromatic rings and trifluoromethyl (CF₃) group, which will ultimately favor its bound to the channel.

Recent evidence reported that CFTR_Inh-172 exerts off-target effects on calcium channels, promoting a significant inhibition of store-operated calcium entry (SOCE) that was time-dependent, poorly reversible in epithelial cells (Calu-3, a human lung cancer epithelial cell line)⁸⁰. In the airway epithelium, CFTR is the main mediator of Cl⁻ efflux at the apical membrane of cells, and the movements of Cl⁻ have a strong influence on the movement of other ion species, including Ca^{2+ 5}. To understand if the inhibition protocol implemented could impact the function of Ca²⁺ channels in human spermatozoa, we evaluated the variation of intracellular [Ca²⁺] in spermatozoa with inhibited CFTR function in comparison to spermatozoa with preserved sperm function. As no variation of intracellular [Ca²⁺] was observed, we assumed that the CFTR inhibition protocol applied had no impact on Ca²⁺channels and associated pathways.

As sperm vitality was found to be unaltered by CFTR function inhibition, nor its maintenance for 1 h, we demonstrate that our protocol does not impose any cytotoxic effects on sperm.

Having demonstrated that the inhibition of spermatozoa’s CFTR function did not impose any cytotoxic effects but induced a prolonged deleterious effect on intracellular [Cl⁻] and sperm osmoregulation capacity, we sought a way to reverse these effects.

There has been some success in treating cystic fibrosis patients by using liposomes to deliver CFTR^cDNA and CFTR^mRNA81–83. In vitro studies have also demonstrated that CFTR channel function could be promptly restored through the delivery of a functional CFTR glycoprotein via extracellular vesicles to affected cells^41,42. Nonetheless, to our knowledge, these treatments have never been applied to improve the fertility of individuals impacted by CFTR dysfunction.

SF fulfills major reproductive roles contributing to pH control, spermatozoa nourishment, and protection against the female immune system⁸⁴. The proteomic analysis of SF-EVs had already confirmed the presence of CFTR in these nanostructures. Herein, we further confirmed the presence of this channel in SF-EVs isolated through differential centrifugation; furthermore, we could identify two different glycosylation states at 130 and 150 kDa of CFTR. The use of SF-EVs in this work presented many technical advantages for the pursuit of our goals, one of them being the easy accessibility to SF samples, and the simplicity of SF-EVs isolation protocol. More importantly, SF-EVs are known to interact with spermatozoa, as we further confirmed by labeling SF-EVs with PHK67, a membrane-intercalating fluorescent dye. These lipid bilayer membrane-bound vesicles are known to carry important biomolecules, which include lipids and other energetic substrates, RNAs, and proteins, and play an important role in the modulation of sperm physiology^64,85–87. Thanks to these properties, we hypothesized that CFTR-carrying SF-EVs would be able to restore the channel function that had been previously compromised by CFTR_Inh-172.

We started by confirming that the incubation of spermatozoa with compromised (via inhibition) CFTR function with the SF-EVs did not affect sperm vitality. Interestingly, EVs were able to restore both the intracellular [Cl⁻] and the sperm membrane’s glycerol permeability to the levels observed before CFTR function was inhibited.

Regardless, our results have an undisputable limitation; although we were able to demonstrate that SF-EVs carry CFTR, confirming the results provided by 2D LC-MS/MS⁶⁴, and confirmed that SF-EVs interact mainly with the head of spermatozoa, which is also the normal location of CFTR, our results do not prove protein transfer from SF-EVs to the target cells. Nonetheless, we present two strong evidence that support this hypothesis: the restoration of intracellular [Cl⁻], and the restoration of sperm osmoregulation capacity and membrane glycerol permeability, events that are associated with CFTR function^88,89. We can also roll out the hypothesis that the rise of glycerol permeability observed in spermatozoa with compromised (via inhibition) CFTR function treated with SF-EVs could result from the integration of new glycerol channels. In human spermatozoa, two glycerol channels have been identified: AQP7, expressed in the equatorial region of the head and midpiece (curiously, the same location as CFTR²⁰), and AQP3 is expressed in the tail^21,22. AQP9 and AQP10, other aquaglyceroporins, have not been identified in human spermatozoa. According to the published 2D LC-MS/MS results regarding SF-EVs, only two AQPs were identified in SF-EVs, AQP2 and AQP5, neither of which transport glycerol or have been described in human spermatozoa⁶⁴. Similarly, other Cl⁻ channels expressed in the human spermatozoa, which include the chloride channels family (primarily, ClC-3¹⁸), and Ca²⁺-activated Cl⁻ channels (CaCCs, primarily TMEM16A⁹⁰, have not been identified in SF-EVs⁶⁴. This excludes the premise that other Cl⁻ channels beyond CFTR could be incorporated in the human spermatozoa instead. Although our premise undoubtedly needs further exploration, we have observed compelling evidence that human spermatozoa are capable of acquiring CFTR, a transmembrane channel, via EVs, as already described in other cells^41,42.

Conclusion

In summary, through our human sperm model to study CFTR (dys)function, we observed that CFTR has an important role in the modulation of Cl⁻, glycerol, and water permeability. Its dysfunction (via inhibition) impaired sperm osmoregulation capacity. We demonstrate that a cohort of small and neutrally charged SF-EVs was able to promote recovery of intracellular [Cl⁻] and sperm osmoregulation capacity after CFTR inhibition. We have also shown evidence that the usage of CFTR-carrying SF-EVs, as already used to treat symptoms of cystic fibrosis, could be applied to improve the sperm quality of individuals affected by CFTR function disturbances.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

SCP and OF were responsible for experimental procedures, statistical analysis, and data analysis. SCP was responsible for the draft of the manuscript. ID, SO, and VA were responsible for the biological material supply. MS and MPM were responsible for data analysis, critical review, and manuscript editing. RLB was responsible for project administration, conceptualization, study design, supervision, and data analysis. All authors read critically and approved the final manuscript.

Funding

This work was funded by “Fundação para a Ciência e a Tecnologia”—FCT to UMIB (UID/215/2025), Sara C. Pereira (10.54499/2021.05487.BD), ITR—Laboratory for Integrative and Translational Research in Population Health (LA/P/0064/2020).

Data availability

The data underlying this article are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.