17BIPHE2, an engineered cathelicidin antimicrobial peptide with low susceptibility to proteases, is an effective spermicide and microbicide against Neisseria gonorrhoeae

Abstract

STUDY QUESTION

Is 17BIPHE2, an engineered cathelicidin antimicrobial peptide with low susceptibility to proteases, a better spermicide in cervicovaginal fluid (CVF) than its parental peptides, LL-37 and GF-17?

SUMMARY ANSWER

At the same mass concentration, 17BIPHE2 exhibited the highest spermicidal activity on human sperm resuspended in CVF-containing medium.

WHAT IS KNOWN ALREADY

LL-37 and its truncated peptide GF-17 exert both spermicidal and microbicidal activities, although they are prone to proteolytic degradation in body fluids.

STUDY DESIGN, SIZE, DURATION

Spermicidal activities of 17BIPHE2 were evaluated in vitro in mouse and human sperm, both resuspended in medium, and then on human sperm incubated in CVF-containing medium; in the latter condition, the spermicidal activity and peptide stability in CVF of 17BIPHE2 were compared with that of LL-37 and GF-17. The in vivo contraceptive effects of 17BIPHE2 and the reversibility thereof were then assessed in mice. Finally, in vitro microbicidal effects of 17BIPHE2 on Neisseria gonorrhoeae were determined.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Sperm motility and plasma membrane integrity were assessed by videomicroscopy and exclusion of Sytox Green, a membrane-impermeable fluorescent dye, respectively. Successful in vitro fertilization (IVF) was determined by the presence of two pronuclei in oocytes following their coincubation with capacitated untreated or 17BIPHE2-treated sperm. Sperm alone or with 17BIPHE2 were transcervically injected into female mice and successful in vivo fertilization was indicated by the formation of two-cell embryos 42-h postinjection, and by pregnancy through pup delivery 21–25 days afterwards. Peptide intactness was assessed by immunoblotting and HPLC. Reversibility of the contraceptive effects of 17BIPHE2 was evaluated by resumption of pregnancy of the female mice, pretranscervically injected with 17BIPHE2, following natural mating with fertile males. Minimum inhibitory/bactericidal concentrations of 17BIPHE2 on N. gonorrhoeae were obtained through microdilution broth assay.

MAIN RESULTS AND THE ROLE OF CHANCE

At the same mass concentration, 17BIPHE2 was a more effective spermicide than LL-37 or GF-17 on human sperm resuspended in CVF-containing medium, with the spermicidal concentration of 32.4 µM. This was mainly due to lower susceptibility of 17BIPHE2 to CVF proteases. Importantly, the reproductive tract of mouse females treated three times with 32.4 µM 17BIPHE2 remained normal and their fecundity resumed after stopping 17BIPHE2 treatment.

LIMITATIONS, REASONS FOR CAUTION

For ethical reasons, the inhibitory effects of 17BIPHE2 on fertilization and pregnancy cannot presently be performed in women. Also, while our study has proven the effectiveness of 17BIPHE2 as a spermicide for mouse and human sperm in vitro, dosage formulation (e.g. in hydrogel) of 17BIPHE2 still needs to be developed to allow 17BIPHE2 to remain in the vagina/uterine cavity with controlled release for its spermicidal action.

WIDER IMPLICATIONS OF THE FINDINGS

Since 17BIPHE2 also exerted bactericidal activity against N. gonorrhoeae at its spermicidal concentration, it is a promising candidate to be developed into a vaginal multipurpose prevention technology agent, thus empowering women against unplanned pregnancies and sexually transmitted infections.

STUDY FUNDING/COMPETING INTEREST(S)

This work was supported by the Canadian Institutes of Health Research (PJT 173268 to N.T.). There are no competing interests to declare.

TRIAL REGISTRATION NUMBER

N/A.

Introduction

The ongoing high rates of unplanned pregnancies and sexually transmitted infections (STIs) have launched the World Health Organization toward the development of multipurpose prevention technology (MPT), which involves intravaginal administration of a single application/device or a single agent for both contraception and anti-STI purposes (Hynes et al., 2018; Hemmerling et al., 2020). Most MPT agents consist of two compounds, one acting as an anti-STI and the other as a contraceptive. Our approach, however, is to use a single compound as an MPT agent. We have described that LL-37, a human antimicrobial peptide in the cathelicidin family (Tanphaichitr et al., 2016), and its truncated peptides (GI-20 and GF-17; Fig. 1; Wang et al., 2014a) have spermicidal activity on both human and mouse sperm (Srakaew et al., 2014; Kiattiburut et al., 2018). The microbicidal action of LL-37 is well studied with known activity on more than 50 bacteria and 10 viruses (Tanphaichitr et al., 2016; Wang et al., 2019). GI-20 and GF-17 likewise exert microbicidal activity on a number of microbes (Li et al., 2006; Wang et al., 2008; Epand et al., 2009; Wang et al., 2012, 2014a; Mishra et al., 2013; Tripathi et al., 2015; Zhang et al., 2021). Notably, LL-37, GI-20 and GF-17 all exert microbicidal activity at a low minimal bactericidal concentration (MBC, <10 µM) on Neisseria gonorrhoeae, a bacterium, which causes STI and which is becoming antibiotic resistant (Unemo and Shafer, 2014; Kiattiburut et al., 2018; Unemo et al., 2021).

Figure 1.

Amino acid sequences of LL-37, GI-20, GF-17, GF-17d3 and 17BIPHE2. LL-37 is a cathelicidin antimicrobial peptide naturally expressed by various cells (e.g. epididymal epithelial cells, neutrophils). GI-20 and GF-17 are truncated peptides of LL-37, which are microbicidally and spermicidally active like LL-37. 17BIPHE2 is an engineered cathelicidin antimicrobial peptide with low susceptibility to proteases. GF-17 sequence is used as a template to engineer GF-17d3 and 17BIPHE2. The two L-isoleucines and the L-leucine (red font) in GF-17 are replaced with corresponding D-amino acids (underlined red font) in GF-17d3. Finally, in 17BIPHE2, all three D-amino acids are D-leucines (underlined red font). The presence of D-amino acids in 17BIPHE2 renders the peptide less susceptible to protease degradation. The two phenylalanines in GF-17 and GF-17d3 are also replaced by two biphenylanines in 17BIPHE2 (blue font) in order to decrease the hydrophobic defect in 17BIPHE2.

Like other antimicrobial peptides, LL-37 (MW:4493) is a small cationic amphipathic peptide (Gudmundsson et al., 1996; Tanphaichitr et al., 2016; Wang et al., 2019), which adopts an alpha helix structure upon interaction with membranes (Gudmundsson et al., 1996; Wang 2008; Wang et al., 2017; Kiattiburut et al., 2018). This secondary structure accelerates initial interaction between cationic amino acids prevalently existing on one helical face with anionic lipids selectively present on the bacterial cell membrane. The hydrophobic residues localized on the other helical face then interact with hydrocarbon chains of the bacterial cell membrane as well as each other on adjacent LL-37 peptides, leading to pore formation and loss of homeostasis, and finally death of the microbes (Wang 2007, 2008). GF-17 (MW:2102) and GI-20 (MW:2473), derived from the central region of LL-37, are also helical in complex with membrane-mimetic micelles, and their microbicidal activity mechanism is the same as LL-37 (Wang 2008; Wang et al., 2012, 2017, Zhang et al., 2021). The mechanism of the spermicidal action of LL-37, GI-20 and GF-17 may also be analogous to that of their microbicidal action, since the sperm plasma membrane selectively contains anionic sulfogalactosylglycerolipid, which is a primary binding ligand of LL-37 and its derivatives. Surface membrane perforation and shedding of LL-37/GI-20/GF-17-treated sperm have also been verified (Srakaew et al., 2014; Kiattiburut et al., 2018). Notably, most antimicrobial peptides with a spermicidal property have an alpha helix conformation (Tanphaichitr et al., 2016), suggesting a close link between this peptide structure and the spermicidal activity.

While LL-37, GI-20 and GF-17, having both microbicidal and spermicidal activities, are promising candidates to be developed into vaginal MPT agents, their high susceptibility to degradation by proteases in body fluids would hinder this development. To circumvent this problem, a D-amino acid incorporation (Wade 2010; Wang 2012) approach was used to engineer GF-17d3 based on the GF-17 sequence (Wang et al., 2014b). Two isoleucines and one leucine in GF-17 are replaced by corresponding D-amino acids to generate GF-17d3 (Fig. 1). The presence of D-amino acids decreases the degradation rate of the peptide by proteases in mammalian body fluid (Wade 2010; Wang et al., 2014b), although this leads to a 75% loss of the alpha helicity in GF-17d3 and also a hydrophobic defect, which decreases the ability of GF-17d3 to bind to the bacterial cell membrane (Wang et al., 2014b, 2017, 2018). To correct the hydrophobic defect in GF-17d3, the two phenylalanines in GF-17d3 are replaced by two biphenylalanines (more hydrophobic). The two D-isoleucines in GF-17d3 are also substituted with D-leucines. The resultant peptide, 17BIPHE2 (Fig. 1), not only is highly resistant to various proteases (Wang et al., 2014b; Narayana et al., 2019) but also gains full microbicidal activity, in particular, against multidrug-resistant bacteria (Wang et al., 2014b). Its microbicidal activity on Klebsiella pneumoniae, Pseudomonas aeruginosa, Staphylococcus aureus as well as Escherichia coli is even higher than that of LL-37, GI-20 or GF-17 (Narayana et al., 2019; Zhang et al., 2021). Furthermore, 17BIPHE2 is noted for its high antibiofilm activity of both Gram-negative and Gram-positive bacteria (Wang et al., 2014b; Narayana et al., 2019; Zhang et al., 2021).

The two unique properties of 17BIPHE2 (improved stability against protease degradation and high microbicidal activity) already make 17BIPHE2 a suitable candidate to be developed into a vaginal anti-infective. It would be more promising, however, if 17BIPHE2 possesses spermicidal activity, as it can be developed into a vaginal MPT agent. The first aim of our study was to determine the spermicidal activity of 17BIPHE2 in medium (without proteases). This was followed by an investigation of whether 17BIPHE2 was more resistant than LL-37, GI-20 and GF-17 to proteases in cervicovaginal fluid (CVF) and seminal plasma, thus serving as an effective spermicide in this environment. In the second aim of our study, the safety of repeated administrations of 17BIPHE2 in the mouse female reproductive tract was evaluated.

Materials and methods

Peptides and anti-LL-37

All peptides, LL-37, GI-20, GF-17 and 17BIPHE2, were chemically synthesized with >95% purity, by CPC Scientific (San Jose, CA, USA) for LL-37, and by Genemed Synthesis Inc. (San Antonio, TX, USA) for GI-20, GF-17 and 17BIPHE2. Rabbit anti-LL-37 IgG antibody was produced against the full length of LL-37 by Excel Biopharm, LLC (Burlingame, CA, USA); as expected anti-LL-37 recognized LL-37 as well as GI-20, GF-17 and 17BIPHE2.

Human samples and ethical approval

Cervicovaginal fluid

Participants were non-pregnant women of reproductive age, uninfected by HIV or classical STIs, and not actively menstruating. Cervicovaginal secretions were self-collected using an Instead Softcup (Evofem, San Diego, CA, USA). CVFs were then weighed and diluted 10-fold by weight in sterile PBS, centrifuged at 1730g for 10 min at 4°C, and the supernatant with pH of 7.4 was cryopreserved at −80°C for subsequent use. Five of these PBS-diluted CVF samples were collected for our study. The protocol (33381) was approved by the HIV Research Ethics Board at the University of Toronto and all participants signed the consent form prior to CVF collection for our research work.

Semen

Semen samples were donated by healthy men. The collection was by masturbation following 2–3 days of abstinence. All samples used in our study had normal semen parameters as described by the World Health Organization (WHO, 2010). Recruitment of donors and handling of semen samples for experiments were approved by the Ottawa Health Network Research Ethics Board (protocol 2005256-01 H) and all participants signed the consent form for semen donation for our research study.

Use and boarding of mice with ethical approval

CD-1 male (8–10 weeks old) and CF-1 female (6–8 weeks old) mice were obtained from Charles River Laboratories (Senneville, QC, Canada) and boarded in a temperature-controlled (22.5°C) room with a photoperiod of 14-h light/10-h dark. Use and handling of mice were approved by the University of Ottawa Animal Care Committee (protocol 2567), following ARRIVE checklists and guidelines.

Preparation of human seminal plasma, and human and mouse sperm

Ejaculated human semen was allowed to liquefy at 37°C for 30 min. Mouse sperm were collected from the cauda epididymis and vas deferens following our described protocol (Tanphaichitr et al., 1990). Both human and mouse sperm of best quality with close to 100% motility were prepared by subjecting the semen and the mouse sperm suspension to Percoll-gradient (45%/90%) centrifugation, as previously described (Tanphaichitr et al., 1988, 1990). Percoll-gradient pelleted sperm were incubated (37°C, 5% CO₂) in medium (Human Tubal Fluid (HTF) for human sperm and Krebs-Ringer Bicarbonate Buffer (KRB) for mouse sperm) without bovine serum albumin (BSA), and with 0.3% BSA for 1 h, and were referred to as ‘partially capacitated’ and ‘fully capacitated’, respectively (Srakaew et al., 2014). In addition, motile swim-up human sperm carrying residual amounts of seminal plasma were prepared from ejaculated liquefied semen as previously described (Srakaew et al., 2014).

After the completion of Percoll-gradient centrifugation, human seminal plasma, remaining in the top layer above the Percoll gradient, was collected for incubation with LL-37, GI-20, GF-17 and 17BIPHE2, as described below.

Effects of CVF and seminal plasma on peptide degradation

LL-37, GI-20, GF-17 and 17BIPHE2 (20 µM each) were separately incubated (37°C) in 50 µl of each of the five CVF samples for various times. At the end of each incubation time point, 6 µl of the mixture was taken for immunoblotting.

The CVF containing each peptide was subjected to Tricine- sodium dodecyl sulfate—polyacrylamide gel electrophoresis (16% polyacrylamide gel (Schagger 2006)) followed by western blotting (Towbin and Gordon, 1984). Immunoblotting using anti-LL-37 antibody was then performed as described (Srakaew et al., 2014). Peptide signals from immunoblotting were densitometrically analyzed using National Institutes of Health (NIH) ImageJ software (https://imagej.nih.gov/ij/docs/guide/). The peptide band intensity at various incubation times was compared to that at the zero hour, which was designated as 100% intactness.

17BIPHE2 or GF-17 remaining after incubation in CVF samples was also analyzed by HPLC. The CVF-peptide incubate was added with nine volumes of prechilled 100% ethanol. This treatment led to the precipitation of CVF proteins (with MW > 10 000). 17BIPHE2 or GF-17 and other small CVF peptides remaining in the ethanol supernatant were dried under a nitrogen stream and resolubilized in water (250 µl), an aliquot of which was injected into a symmetry C18 Waters column (4.6 × 150 mm) of a Waters HPLC system and peptides were eluted with a linear gradient of acetonitrile (with 0.1% trifluoroacetic acid) from 10% to 100% at a flow rate of 1 ml/min at 25°C. Peptide peaks were detected by absorption at A220. The retention time of 17BIPHE2 and GF-17 was determined by running the pure peptide standards.

In an alternative experiment, CVF samples were pooled and the pH was adjusted to 4.5 with 1 N HCl. This acidic CVF sample was used for incubation with LL-37, GI-20, GF-17 and 17BIPHE2. In another alternative experiment, seminal plasma, diluted 5X in HTF, was used for incubation with the four peptides. The intactness of the peptides at various incubation times in these two alternative experiments was analyzed in a similar manner to that described above for CVF at neutral pH.

Effects of 17BIPHE2 on sperm fertilizing parameters

Treatment of sperm with 17BIPHE2

Partially capacitated (P-CAP) and fully capacitated human and mouse sperm, as well as swim-up human sperm, were resuspended in the medium at the concentration of 10⁷ sperm/ml. The sperm suspension (100 µl) was either untreated or treated (37°C, 5% CO₂, 30 min) with various concentrations of 17BIPHE2. In experiments to determine the temporal effects of the peptide on sperm motility, the peptide treatment was performed for shorter time periods.

Sperm motility and membrane permeabilization

Motility and membrane permeabilization of untreated and 17BIPHE2-treated human and mouse sperm were concurrently assessed. Motile sperm were determined from videomicroscopic records of the sperm suspension placed in a CellVision CV 1020-2cv slide (CellVision Technologies, Heerhugowaard, The Netherlands), using a 10X objective with a Celestron HD Digital Microscope Imager (Torrance, CA, USA) containing a 30X magnifier lens attached to the eyepiece and expressed as percentages of the total sperm population (Saewu et al., 2017). Sperm membrane permeabilization was determined by nuclear incorporation of Sytox Green by flow cytometry, as previously described (Srakaew et al., 2014). Temporal effects of 17BIPHE2 on sperm motility were also determined by videomicroscopy in a microdrop (10 µl) of the sperm suspension placed on a glass slide.

Ability of sperm to fertilize eggs in vitro

Fully capacitated mouse sperm (10⁷/ml in KSOM-0.3% BSA) were untreated or pretreated (37°C, 5% CO₂, 30 min) with various concentrations of 17BIPHE2. After the removal of excess peptide by centrifugation, sperm were coincubated with ovulated mature eggs for another 6 h, as previously described (Kiattiburut et al., 2018). Eggs were microscopically scored for two pronuclei as evidence of fertilization in vitro (Tantibhedhyangkul et al., 2002).

Effects of sperm treatment with 17BIPHE2 on in vivo fertilization and pregnancy

Fully capacitated mouse sperm (10⁷/ml) were untreated or treated with 10.8 μM 17BIPHE2 in KRB-0.3% BSA (37°C, 5% CO₂). The sperm concentration was then adjusted, either immediately or 30 min after the peptide treatment, to 10⁸/ml by centrifugation (350g, 28°C, 5 min) in KRB-0.3% BSA. The concentrated sperm (50 µl) with or without 17BIPHE2 were transcervically injected into each female mouse naturally cycling to the estrus phase at 1 h prior to or at the expected ovulation time, following our described method (Srakaew et al., 2014). The estrous cycle of these mice was tracked daily by cytological assessment of the vaginal lavage (Byers et al., 2012). In one set of experiments, female mice were sacrificed 42-h postinsemination and two-cell embryos and unfertilized eggs were flushed out from the oviduct with KFHM (HEPES-buffered KSOM). Percentages of two-cell embryos over the sum of embryos and unfertilized eggs indicated the in vivo fertilization rates. In another set of experiments, pregnancy outcomes were determined, first by the abdominal enlargement by Days 11–14 postinsemination. Some females were sacrificed on Day 16 postinsemination, and their uteri were scored for implantation fossae that contained intact fetuses. Alternatively, pregnant mice were boarded to term to allow pup delivery. The litter size was obtained from the number of implantation fossae with intact fetuses or the number of pups delivered.

Comparison of the spermicidal property among 17BIPHE2, GF-17 and LL-37 in CVF-containing medium

To determine the effects of 17BIPHE2, GF-17 and LL-37 on motility of human sperm that were incubated in CVF, HTF components were added to PBS-diluted CVF samples, so that control (untreated with peptide) sperm still remained motile for at least 3 h under this incubation. The modified mixture, referred to as CVF-HTF, contained the original components from PBS (138.6 mM NaCl, 0.9522 mM KH₂PO₄ and 5.031 mM Na₂HPO₄) and the added HTF components (25 mM NaHCO₃, 2.04 mM CaCl₂.2H₂O, 0.33 mM sodium pyruvate, 24.35 mM sodium lactate, 2.78 mM glucose and 0.141 mM Phenol Red).

Motile human sperm prepared by Percoll-gradient centrifugation were added to CVF-HTF to the final concentration of 10⁷ motile sperm/ml. Sperm in this CVF-HTF were treated (37°C, 5% CO₂, 30 min) with various concentrations of LL-37 or 17BIPHE2 in the first set of experiments and then with GF-17 and 17BIPHE2 in the second set for comparative effects of the two sets of peptides on sperm motility (determined as described above).

Aliquots of the sperm-peptide suspension in CVF-HTF were also taken at 0- and 30-min time points for immunoblotting in order to analyze for peptide intactness after a 30-min incubation.

Histological assessment of the mouse female reproductive tract

Female mice were transcervically injected with 50 µl of 32.4 µM 17BIPHE2 or with 50 µl PBS ∼1 h before the expected ovulation time in the estrus. The injections were repeated in the two consecutive estrous cycles. Mice were sacrificed about 18 h after the last injection for collection of the tissues (vagina, cervix and uterus), which were fixed in Bouin’s solution and embedded in paraffin for tissue sectioning. Sections were stained with hematoxylin and eosin and digitally imaged on an Aperio CS2-Digital Pathology Scanner (Leica Biosystem, Buffalo Grove, IL, USA). Images obtained from each slide were analyzed by ImageScope software (Leica Biosystems Inc., Concord, ON, Canada).

Resumption of fecundity of females transcervically injected three times with 17BIPHE2

Female mice were injected with 17BIPHE2 (10.8 or 32.4 µM) or with PBS in three consecutive estrous cycles as described above. One week after the last injection, females were individually cocaged with fertile males for natural mating. Pregnancies of these females were assessed by pup delivery (21–25 days after cocaging), and the litter size was determined. These experiments were performed twice with four mice in each treatment group in each experiment (i.e., total of eight mice in each group in the replicate experiments).

Bactericidal effects of 17BIPHE2

Commensal Lactobacillus crispatus SJ-3C was purchased from ATCC (Manassas, VA, USA), while a clinical isolate strain of N. gonorrhoeae NRCC6879 was provided by Dr Wayne Conlan, National Research Council of Canada (NRCC), Ottawa, ON. Culture conditions of the two bacteria were as previously described (Kiattiburut et al., 2018), and microbicidal activities of 17BIPHE2 against them were determined using the minimal inhibitory concentration (MIC) assay, and then the MBC assay, following the standard broth microdilution method (Wiegand et al., 2008).

Sample size and statistical analyses

Experiments were repeated three times, unless mentioned otherwise, using biological samples collected from different individuals or animals. GraphPad Prism Software 8.0 (San Diego, CA, USA), one-way ANOVA or two-way ANOVA, both with Tukey’s multiple comparison, as well as Student’s t-test, were used for determining significant differences between samples. A value of P < 0.05 was considered significant.

Results

Spermicidal activity of 17BIPHE2 on human and mouse sperm

The degree of alpha helicity of 17BIPHE2 is only 25% of the parental peptide, GF-17, due to the replacement of three natural L-amino acids in GF-17 by D-amino acids in 17BIPHE2 (Fig. 1) (Wang et al., 2014b, 2017, 2018). Since alpha helicity of an antimicrobial peptide appears relevant to its spermicidal activity, it was important that we demonstrated that 17BIPHE2 could still act as a spermicide, like LL-37, GI-20 and GF-17.

Inhibitory effects of 17BIPHE2 on human and mouse sperm motility

Figure 2 demonstrates the spermicidal effects of 17BIPHE2 on human and mouse Percoll-gradient centrifuged (PGC) sperm, both partially capacitated and fully capacitated. Within 30 min of treatment, 17BIPHE2 decreased motility of these sperm and concurrently increased sperm membrane permeabilization in a concentration-dependent manner. Human sperm, both partially and fully capacitated, were completely immotile upon treatment with 17BIPHE2 at 21.6 µM (Fig. 2, top panel). The spermicidal action of 17BIPHE2 was also observed on PGC mouse partially capacitated and fully capacitated sperm (Fig. 2, bottom panel) with a complete loss of sperm motility at 10.8 µM of 17BIPHE2. The complete inhibitory effect on sperm motility together with the proportional induction of sperm membrane permeabilization was also observed in swim-up human sperm, which carry residual amounts of seminal plasma, at 21.6 µM 17BIPHE2 (Fig. 2, middle panel). Notably, the spermicidal concentrations of 17BIPHE2 on human and mouse sperm resuspended in medium (21.6 and 10.8 µM, respectively) were 2X and 3X that of GF-17/GI-20/LL-37, respectively (Srakaew et al., 2014; Kiattiburut et al., 2018). It was noted, however, that 17BIPHE2 exerted inhibitory effects on both human and mouse sperm motility immediately. Within 1 min of treatment with 21.6 µM 17BIPHE2, motility of human PGC sperm was significantly decreased and loss of forward motility was observed in >99% sperm within 5 min (Video 1). Similarly, the inhibitory effects of 10.8 µM 17BIPHE2 on mouse PGC sperm motility were observed within 1 min of treatment and by 5 min, all sperm did not move forward (Video 2).

Figure 2.

Effects of 17BIPHE2 on human and mouse sperm motility and membrane permeabilization. Partially and fully capacitated Percoll-gradient centrifuged human (top panel) and mouse (bottom panel) sperm, as well as human swim-up sperm (middle panel), were used for the study. Sperm motility was assessed by videomicroscopy, whereas sperm surface membrane permeabilization of the same sample was evaluated by nuclear Sytox Green staining. Percentages of motile sperm (closed circle) and sperm with permeabilized membranes (open circle) were expressed as means ± SD from three experiments using sperm samples from different donors. Statistical analyses were performed using one-way ANOVA followed by Tukey’s multiple comparison. The symbols *, **, *** and **** denote P < 0.05, 0.01, 0.001 and 0.0001, respectively, for the significant differences between the 17BIPHE2 treated sperm and control untreated sperm. 17BIPHE2 is an engineered cathelicidin antimicrobial peptide with low susceptibility to proteases.

Contraceptive effects of 17BIPHE2 in the mouse model

Capacitated mouse sperm pretreated with 17BIPHE2 had a decreased ability to fertilize eggs in vitro in a concentration dependent manner, and when the 17BIPHE2 treatment was at the spermicidal concentration (10.8 µM), the sperm fertilizing ability was completely abolished (Fig. 3A). The in vivo fertilization rate was also zero when sperm were treated with 10.8 µM 17BIPHE2 (Fig. 3B). Corroborating these results, none of female mice (n = 11) transcervically injected with sperm + 10.8 µM 17BIPHE2 became pregnant, whereas pregnancies were observed in 16 of 18 (89%) females transcervically injected with sperm alone with an average normal litter size (10 ± 2; Table I). In both in vivo experiments, which involved transcervical injection of sperm + 17BIPHE2 (10.8 µM), the zero rates of in vivo fertilization and pregnancy were observed when the injection was carried out either immediately after incubation of sperm with the peptide or after a 30-min incubation.

Figure 3.

Inhibitory effects of 17BIPHE2 on mouse sperm fertilizing ability. Assessment was performed in vitro (A) and in vivo (B). IVF was determined by counting eggs with two pronuclei following 6-h coincubation of capacitated sperm (either untreated control or 17-BIPHE2-treated) with mature eggs. In vivo fertilization was determined by scoring two-cell embryos in the oviduct 42 h after transcervical injection of sperm alone (control) or 17BIPHE2-treated sperm into females that were naturally cycling in estrus. Results were expressed as mean ± SD from three replicate experiments. n = total eggs assessed in (A) and = total two-cell embryos + unfertilized eggs in (B). Statistical analyses were performed using one-way ANOVA followed by Tukey’s multiple comparison in (A) and Student’s t-test in (B). *, **** denote P < 0.05 and 0.0001, respectively, for the significant differences between the results from the 17BIPHE2 treated sperm and control untreated sperm. 17BIPHE2 is an engineered cathelicidin antimicrobial peptide with low susceptibility to proteases.

Table I.

Female mice transcervically injected with sperm plus 17BIPHE2 (10.8 µM) fail to become pregnant.

Treatment	No. pregnant/total females 1	Litter size
Control2	16/18 (89%)	10 ± 2
17BIPHE23	0/11 (0%)	0

¹ Pregancy was assessed by pup delivery or formation of implantation fossae.

²Control females were transcervically injected with sperm alone.

³Females were transcervically injected with sperm + 17BIPHE2 (an engineered antimicrobial peptide).

17BIPHE2 was less susceptible than LL-37, GI-20 and GF-17 to degradation by proteases in CVF

When 20 µM each of LL-37, GI-20, GF-17 or 17BIPHE2 was incubated with CVF samples at pH 7.4, 17BIPHE2 consistently showed the highest percentages of peptide intactness within 24 h of incubation time (Fig. 4A and B). Among the five CVF samples used, CVFs 1, 3, 4 appeared to show a slower LL-37 degradation rate than CVFs 2, 5. Within 5 h of incubation, none of LL-37 remained in CVFs 2 and 5, but about 40–70% of LL-37 was still present in CVFs 1, 3, 4. The degradation rate of GI-20 and GF-17 was faster with less differential effects among the five CVF samples. By the end of 5 h, GI-20 was totally degraded by CVFs 1, 4, 5 and remained at 10% and 30% in the CVF 2 and CVF 3, respectively. GF-17 was completely degraded in CVFs 2, 3, 4 at the end of 5-h incubation, and remained at 35% and 3% in CVF 1 and CVF 5, respectively. In contrast, the intactness of 17BIPHE2 at the same 5-h incubation point appeared to be high, about 55–75% in the CVF 1, 3, 4 group and about 65–95% in the CVF 2, 5 group (Fig. 4B). In the scatterogram plot of all five CVF samples, the average percent intactness of 17BIPHE2 after a 5-h incubation was 70%, which was significantly higher than that of LL-37 (34.6%), GI-20 (7.4%) and GF-17 (7.4%) (Fig. 4B) with the P-value of the comparison pair, LL-37 versus 17BIPHE2, GI-20 versus 17BIPHE2 and GF-17 versus 17BIPHE2 of <0.001, <0.0001 and <0.0001, respectively (Fig. 4C). However, after 24 h of incubation in all five CVFs, 17BIPHE2 became markedly degraded with an average peptide intactness of ∼25% (ranging from 6% to 34%). Nonetheless, these peptide intactness levels of 17BIPHE2 were still significantly higher than those of LL-37, GI-20 and GF-17, all of which became zero at this time point (P < 0.05 for all three pairs: LL-37 versus 17BIPHE2, GI-20 versus 17BIPHE2 and GF-17 versus 17BIPHE2). At the 48-h incubation time point, 17BIPHE2 became completely degraded in four of five CVFs, but about 26% of it still remained in CVF 2. However, there were no significant differences in the percentages of the remaining amounts among all four peptides (Fig. 4C).

Figure 4.

17BIPHE2 is most resistant to degradation by proteases in human cervicovaginal fluid, as compared with LL-37, GI-20 and GF-17. (A) Immunoblotting of LL-37, GI-20, GF-17 and 17BIPHE2 following incubation in cervicovaginal fluid (CVF) 1 and CVF 5 for various times. 17BIPHE2 is an engineered cathelicidin antimicrobial peptide with low susceptibility to proteases. LL-37 is a cathelicidin antimicrobial peptide naturally expressed by various cells (e.g. epididymal epithelial cells, neutrophils). GI-20 and GF-17 are truncated peptides of LL-37, which are microbicidally and spermicidally active, like LL-37. (B) Percentage of intactness of each peptide (LL-37, GI-20, GF-17 or 17BIPHE2) following incubation in the five CVF samples (labeled as 1, 2, 3, 4, 5) for various times. The amount of each peptide at 0 h was designated as 100%. (C) Significant differences in the higher intactness levels of 17BIPHE2 as compared with LL-37, GI-20 and GF-17 at the 5  and 24 h time points following incubation in the five CVFs. Statistical analyses were performed using two-way ANOVA followed by Tukey’s multiple comparison. (D) HPLC analysis of 17BIPHE2 and GF-17 in the ethanol supernatant of the CVF-17BIPHE2 incubate and CVF-GF-17 incubate at various incubation times, respectively. HPLC was performed on a symmetry C18 Waters column (4.6 × 150 mm) of a Waters HPLC system. Peptides were eluted with a linear gradient of acetonitrile (with 0.1% trifluoroacetic acid) from 10% to 100% at a flow rate of 1 ml/min at 25°C, and were detected by their absorbance at A220. Arrows indicate the retention time of 17BIPHE2 and GF-17 as determined by using the peptide standards. Results shown are representative of three replicate experiments.

In an alternate experiment, CVF-peptide incubates were treated with 90% ethanol to precipitate CVF proteins and any peptides bound to them. Free peptides (MW <10 000), remaining in the ethanol supernatant, were subjected to HPLC (Materials and Methods section). 17BIPHE2 with the retention time of 11.017 min (Supplementary Fig. S1) was present in the ethanol supernatant of the CVF-17BIPHE2 incubate at all three incubation times (0, 1 and 5 h) with a temporal decrease in its amount. In contrast, the GF-17 peak (retention time of 11.354 min; Supplementary Fig. S1) was not observed in the ethanol supernatant of the CVF-GF-17 incubate of all incubation times (Fig. 4D). The absence of the GF-17 peak in the HPLC run at 0 h was in contrast to the presence of the GF-17 immunoblot band in the whole CVF-GF-17 extract (Fig. 4A). A twofold explanation could be given to this result. First, GF-17 was present in the ethanol supernatant, but its level was under the limit of detection in our HPLC analysis (which was 799 ng). Second, GF-17 interacted with CVF proteins and became precipitated in 90% ethanol. Through immunoblotting (which could detect GF-17 at 0.5 ng), our unpublished results indicated that GF-17 was present in the ethanol supernatant of the CVF-GF-17 incubate at 0  and 1 h (at ∼35% of the 0 h time point), but absent at 5 h. In addition, GF-17 was found in the ethanol precipitate in all three incubation times, indicating its interaction with the CVF proteins. A parallel immunoblotting analysis of CVF-17BIPHE2 incubates indicated that the levels of 17BIPHE2 in the ethanol precipitates were much less than those of GF-17 in all three incubation times. Overall, our results indicated that 17BIPHE2 was not only less susceptible than GF-17 to CVF protease degradation but also more available in a free form to exert its spermicidal activity.

A similar experiment on peptide intactness was also performed using the pooled CVF samples, with their pH adjusted with 1 N HCl to 4.5. Notably, LL-37 showed the highest degradation rate followed by GI-20. At 5 h of peptide incubation with the acidic CVF, about 43% of LL-37 and 62% of GI-20 remained intact, and at 24 h of incubation <1% of LL-37 and only 19% of GI-20 were present. In contrast, 86% of GF-17 and 92% of 17BIPHE2 remained intact at 5 h, and even at 24 h of incubation, the majority of these peptides (72% for GF-17, and 86% of 17BIPHE2) still existed (Supplementary Fig. S2). These results indicated that 17BIPHE2 was the least susceptible to protease degradation, even in CVF with acidic pH.

The intactness of LL-37, GI-20, GF-17 and 17BIPHE2 in seminal plasma was also evaluated. Interestingly, all the peptides remained intact throughout the 24-h incubation in seminal plasma (Supplementary Fig. S3).

Comparison of the spermicidal activity of 17BIPHE2, LL-37 and GF-17 on human sperm resuspended in CVF-HTF

Figure 5 shows comparative spermicidal effects between 17BIPHE2 and LL-37 and between 17BIPHE2 and GF-17 on motile P-CAP human sperm resuspended in CVF-HTF. In the presence of CVF, none of the three peptides at the spermicidal concentration in medium (SCM), i.e., 10.8, 10.8 and 21.6 µM for LL-37, GF-17 and 17BIPHE2, respectively, could totally inhibit human sperm motility. Sperm motility was still at 85–90% following treatment with 1X SCM of either LL-37 or GF-17. However, only 20–25% of sperm treated with 17BIPHE2 at 1X SCM remained motile. At 2X SCM (21.6 µM of both LL-37 and GF-17, equivalent to 97.0 µg/ml LL-37 and 45.4 µg/ml GF-17), motility of sperm treated with LL-37 and GF-17 was 50% and 60%, respectively. However, total inhibition of sperm motility occurred at 1.5X SCM of 17BIPHE2 (32.4 µM or 73.0 µg/ml). In contrast, 23% of sperm treated with GF-17 at 3X SCM (32.4 µM or 68.1 µg/ml) still remained motile. Therefore, at a similar mass concentration, 17BIPHE2 had the highest spermicidal activity in CVF-HTF, with the spermicidal concentration of 32.4 µM.

Figure 5.

17BIPHE2 is the most effective spermicide on human sperm resuspended in cervicovaginal fluid—Human Tubal Fluid. Comparative effects between 17BIPHE2 and LL-37 are shown in (A) and between 17BIPHE2 and GF-17 in (B). 17BIPHE2 is an engineered cathelicidin antimicrobial peptide with low susceptibility to proteases. LL-37 is a cathelicidin antimicrobial peptide naturally expressed by various cells (e.g., epididymal epithelial cells, neutrophils). GF-17 is a truncated peptide of LL-37, which is microbicidally and spermicidally active, like LL-37. Sperm in cervicovaginal fluid—Human Tubal Fluid (CVF-HTF) were treated (37°C, 5% CO₂, 30 min) with LL-37 (MW: 4493) at 1X and 2X spermicidal concentration in medium (SCM) (i.e., 10.8 and 21.6 µM), with GF-17 (MW: 2102) at 1X, 2X and 3X SCM (i.e., 10.8, 21.6 and 32.4 µM), and with 17BIPHE2 (MW: 2254) at 1X and 1.5X SCM (i.e., 21.6 and 32.4 µM). Motility was assessed by videomicroscopy and presented as % control (untreated sperm at the 30-min incubation point designated as 100%). Data are expressed as mean ± SD from three replicate experiments using sperm from different donors. Statistical analyses were performed using one-way ANOVA followed by Tukey’s multiple comparison. *, **** denote P < 0.05 and 0.0001, respectively, for the significant differences between the results from the peptide-treated sperm and control untreated sperm. ns = not significant.

The highest spermicidal efficiency of 17BIPHE2 was likely due to its low susceptibility to degradation by proteases in CVF (Fig. 4). Results shown in Fig. 6 confirm this postulation. About 80% of 17BIPHE2 (32.4 µM or 73.0 µg/ml) remained intact in the sperm suspension in CVF-HTF after 30 min of incubation. In contrast, the intactness of GF-17 (32.4 µM or 68.1 µg/ml) and LL-37 (21.6 µM or 97.0 µg/ml) was only 53% and 59%.

Figure 6.

More 17BIPHE2 remains intact than GF-17 and LL-37 in cervicovaginal fluid—Human Tubal Fluid. The highest concentration of 17BIPHE2 (32.4 µM), GF-17 (32.4 µM) and LL-37 (21.6 µM) used for treatment of sperm in cervicovaginal fluid—Human Tubal Fluid (CVF-HTF) for the motility study was assessed for % intactness by immunoblotting. The densitometry of the immunoblot band of the peptide at 30 min was expressed as % of the corresponding value at 0 min (designated as 100%). Results shown are representative of two replicate experiments. 17BIPHE2 is an engineered cathelicidin antimicrobial peptide with low susceptibility to proteases. LL-37 is a cathelicidin antimicrobial peptide naturally expressed by various cells. GF-17 is a truncated peptide of LL-37, which is microbicidally and spermicidally active, like LL-37.

Multiple uterine administrations of 17BIPHE2 did not cause damage to the uterus, cervix and vagina

Figure 7 demonstrates that the uterus, cervix and vagina of female mice transcervically injected with 32.4 µM 17BIPHE2 in the estrus phase for three consecutive estrous cycles were histologically normal with all cell layers being intact, similar to those in the corresponding tissues of PBS-injected mice. In an alternative set of experiments, where females that were thrice preinjected with 17BIPHE2 (10.8 or 32.4 µM for each injection) or PBS were naturally mated with fertile males, 75% and 100% of females preinjected with 10.8 and 32.4 µM 17BIPHE2, respectively, became pregnant. These pregnancy rates were comparable to that of PBS-injected females (87%; Fig. 8A). The average litter sizes from the three pregnant mouse groups (PBS-injected, 10.8 µM-17BIPHE2-injected, and 32.4 µM-17BIPHE2 injected) were 12, 11.7 and 11.2, respectively, with no statistically significant differences among them (Fig. 8B). All pups were anatomically normal. The normal pregnancy rates and the normal litter size in females thrice transcervically injected with 17BIPHE2 indicated that their uteri, as well as their vaginas and cervices, were functionally normal and able to support copulation, gestation and parturition.

Figure 7.

The lower mouse female reproductive tract tissues still show normal histology after multiple exposures to 17BIPHE2. Mice were transcervically injected during the night of estrus in three consecutive estrous cycles with 32.4 µM 17BIPHE2 or PBS. 17BIPHE2, an engineered cathelicidin antimicrobial peptide with low susceptibility to proteases. Histology of the sections stained with hematoxylin/eosin of the fixed vagina, cervix and uterus of mice injected with 17BIPHE2 is shown in comparison with that of females injected with PBS (controls). Results shown are representative of three 17BIPHE2-injected and three PBS-injected females.

Figure 8.

Resumption of fecundity of female mice previously transcervically injected with 17BIPHE2 (10.8 or 32.4 µM) in three consecutive estrous cycles. 17BIPHE2 is an engineered cathelicidin antimicrobial peptide with low susceptibility to proteases. Females injected with PBS in parallel served as controls. After the last injection, the females were boarded in normal conditions for two estrous cycles before being individually caged with fertile males in the next proestrus. Pregnancy was then monitored by abdomen enlargement followed by pup delivery. (A) Pregnancy rate. Data are expressed as an average percentage of pregnancy of total mice studied (n = 8 for each treatment or control group). (B) Litter size of the delivered pups. Data are expressed as mean ± SD from all pregnant mice.

Bactericidal effects of 17BIPHE2

17BIPHE2 had bactericidal effects on N. gonorrhoeae with an MIC and MBC both at 1.8 µM (Table II). In contrast, the bactericidal activity of 17BIPHE2 on the commensal bacterium L. crispatus (MBC = 10.8 µM) was 6X lower than that on the pathogenic N. gonorrhoeae (Table II). However, 17BIPHE2 had a milder bacteriostatic effect on L. crispatus (MIC = 5.4 µM; Table II).

Table II.

High bactericidal activity of 17BIPHE2 on Neisseria gonorrhoeae but not on Lactobacillus crispatus.

Bacterial species	Minimal inhibitory concentration* (µM)	Minimal bactericidal concentration* (µM)
N. gonorrhoeae	1.8	1.8
L. crispatus	5.4	10.8

^*Minimal inhibitory concentration (MIC) and minimal bactericidal concentration (MBC) were determined by the standard broth microdilution method (Wiegand et al., 2008).

Discussion

In this study, we demonstrated that 17BIPHE2, a GF-17-derived engineered peptide (Wang et al., 2014b, 2019), possessed spermicidal activity through induction of sperm membrane permeabilization in both human and mouse sperm. Notably, the spermicidal concentration of 17BIPHE2 on human sperm (21.6 µM) was higher than that on mouse sperm (10.8 µM), likely because of the more mature status of ejaculated human sperm compared with epididymal/vas deferens mouse sperm. However, the spermicidal concentrations of 17BIPHE2 on human and mouse sperm were 2X and 3X, respectively, higher than those of GF-17 and its parental peptide, LL-37 (Srakaew et al., 2014; Kiattiburut et al., 2018). A number of spermicidal antimicrobial peptides, including LL-37, GI-20 and GF-17, are in an alpha helix form in solution when bound to membranes, suggesting that this secondary structure is essential for their spermicidal activity (Tanphaichitr et al., 2016). In 17BIPHE2, the incorporation of three D-amino acids leads to a 75% decrease in its alpha helicity (Wang et al., 2014b, 2017). A higher concentration of 17BIPHE2 was therefore required for its spermicidal activity. In contrast, the microbicidal activity of 17BIPHE2 on N. gonorrhoeae was the same as that of GF-17 and LL-37, and even more effective than GI-20 (Kiattiburut et al., 2018). These results suggested that the reduced alpha helicity of 17BIPHE2 was not relevant to its microbicidal action. Rather, the correction of the ‘hydrophobic defects’ in bacterial membrane interaction/permeabilization of 17BIPHE2 by replacing Phe¹⁷ and Phe²⁷ with biphenylalanines was likely relevant in endowing the microbicidal property to this engineered peptide (Wang et al., 2014b, 2018, 2019). Biphenylalanines are more hydrophobic, thus likely interdigitating membranes better than phenylalanines (Wang et al., 2018) and this may explain the higher bactericidal activity of 17BIPHE2 on a number of microbes.

The presence of unnatural amino acids, the three D-leucines, and perhaps also the two biphenylalanines results in low susceptibility of 17BIPHE2 to degradation by proteases (Wang et al., 2014b; Narayana et al., 2019). Herein, we demonstrated that 17BIPHE2 was much more resistant than LL-37, GI-20 and GF-17 to degradation by proteases in human CVF at pH 7.4 in the first 5 h, and within 30 min of incubation in the sperm-CVF-HTF suspension. While the human vaginal lumen is acidic with a pH around 4.5 (Moller, 1991), seminal plasma ejaculated into the vagina has a high buffering capacity (Tevi-Benissan et al., 1997), thus neutralizing the pH of the vaginal lumen. With the lowest susceptibility to degradation by CVF proteases, 17BIPHE2 therefore had the highest spermicidal activity, as compared with LL-37 and GF-17, at similar mass concentrations. Even at the normal vaginal pH of 4.5, 17BIPHE2 had the lowest rate of degradation by CVF proteases, as compared with LL-37 and GI-20. All of these findings make 17BIPHE2 a promising candidate in terms of being intravaginally administered for a prolonged time period as a spermicide. However, all four peptides, LL-37, GI-20, GF-17 and 17BIPHE2, were very resistant to protease degradation when incubated in human seminal plasma for 24 h. While this observation is still unexplained, it should not be surprising since the precursor form of LL-37, hCAP-18, is present in seminal plasma at a high concentration (2–10 µM) without any detectable trace of degradation (Sorensen et al., 2003).

In order to be developed into an MPT agent, 17BIPHE2 has to be non-toxic to normal cells. Hemolysis of human red blood cells induced by the peptide occurs only at a high concentration of 17BIPHE2 (50% hemolytic concentration >200 µM; Wang et al., 2014b; Zhang et al., 2021). Our recent viability study on peripheral blood mononuclear cells (PBMCs) gave similar results. All PBMCs treated with 236 µM 17BIPHE2 lost their viability, while ∼95% of PBMCs exposed to 32.4 µM 17BIPHE2 (the spermicidal concentration on human sperm incubated in CVF) were viable (Supplementary Fig. S4). We also showed herein that multiple transcervical injections of 32.4 µM 17BIPHE2 in female mice did not cause any histological damage to the uterus, cervix and vagina, and these mice could resume fecundity upon mating with fertile males in the same manner as PBS-injected females. In addition, the microbicidal effect of 17BIPHE2 on the pathogenic bacterium in the vagina, N. gonorrhoeae, which is becoming antibiotic resistant (Unemo and Shafer, 2014; Unemo et al., 2021), affirms its potential to be developed into a vaginal MPT agent. 17BIPHE2 also has high microbicidal activity against bacteria causing urinary tract infections (UTIs; E. coli, K. pneumoniae, S. aureus; Narayana et al., 2019; Zhang et al., 2021). Therefore, its intravaginal administration should also be beneficial in UTI prevention, since the bacteria causing UTI harbor around the vaginal opening and the proximal periurethral area (Sihra et al., 2018). Importantly, formulation of 17BIPHE2 into various forms (hydrogel, nanoparticles, film strips, vaginal rings; Sánchez-López et al., 2021) for delivery into the female reproductive tract is urgently needed, since our unpublished results indicate that <2% of LL-37/17BIHE2 in a free peptide form remains in the mouse vagina and uterine cavity following intravaginal and transcervical injection, respectively (see example in Supplementary Fig. S5). The antibiofilm property of 17BIPHE2 will also be beneficial for its inclusion into a solid intravaginal device, which remains in the vagina for a prolonged time period. The use of 17BIPHE2 as a vaginal MPT agent would provide empowerment to women to protect themselves against unplanned pregnancies, STIs and possibly also UTIs.

Data Availability

The data underlying this article are available in the article and its online supplementary material. Additional details related to these published data will be shared on reasonable request to the corresponding author.