Letter to the Editor The presence of glutathione S-transferase in recombinant S100A9 alters its effect on human sperm function

Dear Editor,

In a recent study, we isolated a protein from human oviductal secretion, which could bind to spermatozoa^[1]. This protein was identified through chromatography and tandem mass spectrometry as human S100A9 and was detected in the tubal epithelium and human oviductal secretions.

S100A9 belongs to the S100 protein family^[2], which has been found in various body fluids and tissues, and has roles in extracellular functions, such as the enhancement of neutrophil extravasation, induction of proinflammatory cytokine release, antimicrobial properties through divalent ion sequestration, and modulation of cellular proliferation, differentiation and apoptosis, as well as a chemotactic factor^[3–4]. Because S100A9 is involved in various pathologies and the physiology of inflammation, the number of studies on the effects of S100A9 is growing every day. Recently, we have shown the presence of binding sites for S100A9 in human spermatozoa and also found that S100A9 modulated certain sperm capacitation parameters in vitro, such as induced acrosome reaction (AR)^[1]. To continue our studies on sperm function parameters, the current study aimed to express and purify human recombinant S100A9 and to assess its effect on a sperm capacitation parameter, specifically the AR.

The human S100A9 cDNA was inserted into the PGEX-2T plasmid (Cat. #28-9546-53, GE Healthcare Life Sciences, UK), alongside a built-in sequence of glutathione S-transferase (GST), and was cloned in E. coli. The expressed GST-S100A9 was purified using glutathione-agarose (GSH-A; Cat. #16100, Pierce, Thermo Scientific, Waltham, MA, USA). Subsequently, the fusion protein was treated with agarose-thrombin beads (Cat. #RECOMT-1KT, Thrombin-A, Thrombin CleanCleave Kit, Sigma-Aldrich, St. Louis, MO, USA) to remove the GST moiety (pS100A9). The recombinant proteins were analyzed by using Western blotting. Because the protein does not have glycosylated amino acid residues, the cloning and expression of human S100A9 were carried out in an expression vector in bacteria.

The results indicated that the "in batch" purification method allowed the recovery of approximately 10 mg/mL of fusion protein per milliliter of GSH-A (Fig. 1A). The fusion protein was treated with the enzyme thrombin to separate the S100A9 from the GST. The gel analysis showed that, despite purification steps, the sample still contained small amounts of residual GST and GST fusion protein (GST-S100A9) that did not bind to GSH-A (Fig. 1B). It is suggested that the target protein purity is usually greater than 90% at this stage^[5]. After purification, the expressed recombinant S100A9 was identified with high specificity and sensitivity through Western blot analysis using a rabbit anti-human S100A9 antibody (Cat. #sc-58706, Santa Cruz Biotechnology, Santa Cruz, CA, USA). The results revealed a strong band for the pS100A9 protein, demonstrating a molecular weight similar to that of the control human recombinant S100A9 (hrS100A9; Cat. #ab95909, Abcam Inc., Cambridge, MA, USA) and the native S100A9 from human oviduct fluid (Fig. 1C). Human spermatozoa were obtained from normozoospermic donors (n = 5) through masturbation after three to five days of sexual abstinence^[6]. Motile sperm (swim up) were incubated in the presence of increasing concentrations (0, 0.1, 1.0, and 10.0 µg/mL) of each of the following proteins: hrS100A9, pS100A9, or GST-S100A9 in Ham's F10 medium (Cat. #11550043, Gibco, Grand Island, NY, USA) supplemented with 0.5% BSA for 6 h at 37 ℃ with 5% CO₂. In addition, spermatozoa were incubated in the presence of GST protein (20 μg/mL) or a lysate of untransformed bacteria (10 µg/mL lipopolysaccharides [LPS]) for use as controls. Every experiment was performed in duplicate.

A: The SDS-PAGE (12%) analysis. Lane 1: Cell lysate obtained after isopropil ß-D-1-thiogalactopyranoside induction. Lane 2: Eluate after agarose-glutathione (A-GSH) purification, showing the band of the fusion protein (arrow). Both lanes exhibit a band corresponding to GST-S100A9. B: The SDS-PAGE (15%; Coomassie blue staining) analysis of the purified fusion protein treated with Thrombin-A, showing a GST-S100A9 band, a GST band (approximately 23 kDa), and a pS100A9 band (approximately 13 kDa). Lane 2: The sample from lane 1 was further treated with GSH-A, showing that the fusion protein and GST were almost completely removed with the treatment. C: Detection of the expressed S100A9 performed by Western blot. Lane 1: commercial hrS100A9 (0.5 μg). Lane 2: human oviductal fluid (10 μg total protein). Lane 3: pS100A9 (1 μg). D: Fluorescence images of representative results for the acrosome staining: a) Spermatozoon with intact acrosome; b) Spermatozoon with reacted acrosome. E: Sperm viability under different treatments. F: Sperm motility under different treatments. G: Effect of different recombinant S100A9 on acrosome reaction (AR), with the percentages of the inducible population (IP) obtained in different treatment groups. After a 6-hour incubation under capacitating conditions, two aliquots of sperm suspension from each treatment or control were taken: one was supplemented with progesterone (induced AR), and the other was supplemented with only the culture medium (spontaneous AR). The % of IP for each treatment was then calculated as follows: % induced AR – % spontaneous AR. Three different controls were performed: basal control with only culture medium (Control), bacterial lysate control (LPS), and GST control (GST). Data are indicated as mean ± SD (n = 5 experiments in duplicate). GST-S100A9: fusion protein; pS100A9: purified protein; hrS100A9: commercial human recombinant S100A9. Data were analyzed with ANOVA and the Kruskal-Wallis multiple comparison test. ^***P < 0.001 vs. control.

DownLoad: Full-Size Img PowerPoint

Aliquots of each sperm suspension were incubated in the absence (basal or spontaneous AR) or the presence (induced-AR) of 20 µmol/L progesterone to induce the AR for 30 min at 37 ℃ and 5% CO₂. The AR was detected using fluorescein isothiocyanate-Pisum sativum agglutinin labeling (Fig. 1D)^[1]. The results were presented as the percentage of the inducible population (% IP): the difference between % induced-AR and % basal AR. The analysis of variance (ANOVA) and Kruskal-Wallis multiple comparison test were used to compare the means of % IP, viability, and motility. P < 0.05 was considered statistically significant.

Results indicated that sperm incubation in the presence of either GST-S100A9, pS100A9, rhS100A9, GST, or bacterial lysate controls did not affect sperm viability or motility, which were always higher than 85% and 70%, respectively (Fig. 1E and 1F). These data suggest that the used treatments did not have cytotoxic effects on human spermatozoa under the experimental conditions and concentrations tested.

Neither GST-S100A9 nor pS100A9 affected the spontaneous AR of the treated sperm, as their values were similar to those of the control cells (data not shown). Spontaneous AR reflects the spermatozoa that undergo AR without the presence of an inducer, such as progesterone. This indicated that the proteins in the current study did not directly affect acrosome stability. The presence of GST or LPS did not affect the spontaneous or induced AR. The pS100A9 caused a similar effect on AR, compared with hrS100A9, wherein the lowest concentration of the proteins significantly increased the IP (Fig. 1G). The dose-response effect of hrS100A9 and pS100A9 on the induced AR exhibited an inverted U-shape behavior, where the highest assay did not affect the parameter, which are consistent with the potential mechanisms discussed by Massa et al. (2019) ^[1]. Briefly, there may be changes in receptor availability or the presence of S100A9 oligomerization, which may reduce ligand-receptor binding and, thus, affect the AR. In contrast, the fusion protein, even at the maximum concentration used, was unable to affect the sperm-induced AR with values similar to those in the controls.

Regarding the use of GST-tag recombinant protein, different studies in other cell models have successfully utilized the GST-S100A9 to assess its effects on various cell functions^[7-8]. The fact that GST-S100A9 did not affect the AR may be attributed to several factors. It must be noticed that the commercial hrS100A9 used was a recombinant protein with a His-tag at the C-terminus lacking the GST moiety. Additionally, GST is typically a homodimeric protein, and the oligomer formed by GST and the target protein might influence the properties of the latter^[5], perhaps impeding its action. It appears that, when the GST moiety was removed, pS100A9 was able to interact with human sperm cells, leading to an increase in the induced AR. Thus, both pS100A9 and hrS100A9 further enhanced the effect of progesterone on AR induction. Progesterone has been demonstrated to stimulate calcium influx, the phosphorylation of sperm proteins, and an increase in cAMP, ultimately activating sperm AR. In our previous study, we found that S100A9 was bound to the head, principal piece, and midpiece of human spermatozoa and also stimulated the sperm protein tyrosine phosphorylation^[1]. These binding sites overlap, at least partially, with the reported localization of toll-like receptor 4 (TLR-4) and the receptor for advanced glycation end products (RAGE) in the spermatozoon. It can be hypothesized, therefore, that the interaction of the protein with these receptors may mediate the effects on AR. Studies in various cell models have reported that S100A9 interacts with TLR-4 and RAGE^[9]. Both receptors have been detected in sperm cells, with TLR-4 specifically located in the acrosome, middle piece, and tail of human sperm. RAGE has also been detected in the head and equatorial segment as well as a small population in the tail of human sperm^[10].

It should be noted that in the current study, the AR was evaluated after six hours of incubation under capacitating conditions with the different treatments. This approach reflects the effects on the sperm capacitation process through the ability to undergo the acrosome reaction. At the end of incubation, the AR-inducer progesterone was added in the presence of the different treatments for 30 min. In other words, the proteins under study were not simply evaluated as AR-inducing agents in such a short incubation time of 30 min.

In conclusion, while some studies in other cell models have utilized the GST-tagged protein to assess its effects on cell functions, the present results indicated that only pS100A9 (without the GST moiety) was able to affect the induced AR, whereas the fusion protein did not. The presence of GST in the fusion protein may impede the activation mechanism by which S100A9 modulates the induced AR.

Yours sincerely,Estefania Massa, Gastón Prez, Sergio Ghersevich^✉, Area of Clinical Biochemistry, Facultad de Ciencias Bioquímicasy Farmacéuticas, Universidad Nacional de Rosario, Suipacha 531, (2000) Rosario, Santa Fe, Argentina ^✉Corresponding author: Sergio Ghersevich. E-mail: sghersevich@gmail.com.