OUP user menu

Steroid hormones as bactericidal agents to Helicobacter pylori

Kouichi Hosoda , Hirofumi Shimomura , Shunji Hayashi , Kenji Yokota , Yoshikazu Hirai
DOI: http://dx.doi.org/10.1111/j.1574-6968.2011.02239.x 68-75 First published online: 1 May 2011

Abstract

Helicobacter pylori is a unique bacterial species that assimilates various steroids as membrane lipid components. Our group has recently found, however, that certain steroids may impair the viability of H. pylori. In this study, we go on to reveal that estradiol, androstenedione, and progesterone (PS) all have the potential to inhibit the growth of H. pylori. Of these three steroid hormones, progesterone demonstrated the most effective anti-H. pylori action. 17α-hydroxyprogesterone caproate (17αPSCE), a synthetic progesterone derivative, had a much stronger anti-H. pylori action than progesterone, whereas 17α-hydroxyprogesterone, a natural progesterone derivative, completely failed to inhibit the growth of the organism. Progesterone and 17αPSCE were both found to kill H. pylori through their bacteriolytic action. Among five bacterial species investigated, H. pylori was the only species susceptible to the bactericidal action of progesterone and 17αPSCE. The other four species, Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, and Staphylococcus epiderimidis, all resisted this action. Progesterone and free-cholesterol (FC) obstructed each other's effects against the H. pylori cell. Taken in sum, these results suggest that progesterone and FC may bind to the identical region on the H. pylori cell surface. We expect these findings to contribute to the development of a novel anti-H. pylori steroidal agent.

Keywords
  • Helicobacter pylori
  • progesterone
  • 17α-hydroxyprogesterone
  • 17α-hydroxyprogesterone caproate
  • 2,6-di-O-methyl-β-cyclodextrin

Introduction

Helicobacter pylori colonizes the human gastric epithelium and causes chronic gastritis and peptic ulcers (Marshall & Warren, 1983; Wyatt & Dixon, 1988; Graham, 1991). Over longer periods, it also contributes to the development of gastric cancer and gastric mucosa-associated lymphoid tissue lymphoma (Wotherspoon et al, 1991; Forman, the Eurogast Study Group, 1993). This bacterium possesses the unique biological feature of steroid assimilation. A recent study by our group demonstrated that H. pylori selectively absorbs 3β-OH and 3-OH steroids, glucosylates only the former, and uses both steroids, with or without glucosylation, as membrane lipid components (Hosoda et al, 2009). A number of investigations, including our own, have revealed the physiological significance of steroid assimilation in H. pylori. Wunder . (2006) demonstrated that H. pylori evades the host immune systems by glucosylating the absorbed free-cholesterol (FC). Our own study found that H. pylori retains the steroid (FC or estrone) in order to reinforce the membrane lipid barrier and thereby resists the bacteriolytic action of the phosphatidylcholines (Shimomura et al, 2009). This confirms that certain steroids are beneficial to the survival of H. pylori. Conversely, other steroids have been found to impair the viability of H. pylori. After examining the anabolic use of 10 steroid hormones in H. pylori, our group proposed that three hormones, namely, estradiol, androstenedione, and progesterone, may have the potential to inhibit the growth of H. pylori (Hosoda et al, 2009). These findings led to our interest in the development of antibacterial steroidal agents for H. pylori. To explore the potential for this, we must first precisely clarify the inhibitory effects of those steroids on the growth of H. pylori. In this study, we do so by analyzing the anti-H. pylori actions of the steroid hormones.

Materials and methods

Bacterial strains and culture

Four strains of H. pylori were investigated: NCTC 11638, ATCC 43504, A-13, and A-19. The A-13 and A-19 strains were clinical isolates from a patient with a gastric ulcer and a patient with a duodenal ulcer, respectively. The cultures were all grown in an atmosphere of 5% O2, 10% CO2, and 85% N2 at 37°C (Concept Plus: Ruskinn Technology, Leeds, UK). Helicobacter pylori colonies grown on agar plates of pleuropneumonia-like organisms (PPLO: Difco Laboratories, Detroit, MI) lacking both 2,6-di-O-methyl-β-cyclodextrin (dMβCD) and serum (simple-PPLO: see Supporting Information, Appendices S1 and S2) were suspended in a simple-PPLO broth (10mL) and cultured for 24h with continuous shaking under microaerobic conditions. The H. pylori cell suspensions (200μL) were transferred into a fresh simple-PPLO broth (10mL) and cultured for another 24h under the same conditions. This subculture procedure was repeated three times to yield a preculture H. pylori cell suspension.

Steroids

The following steroids, all from Wako Pure Chemical Industries Ltd. (Tokyo, Japan), were investigated: estradiol, androstenedione, progesterone (PS), 17α-hydroxyprogesterone (17αPS), and 17α-hydroxyprogesterone caproate (17αPSCE). Each steroid was dissolved in dimethyl sulfoxide (DMSO) at a concentration of 100mM and stored at room temperature in the dark until the day of the experiments. The 0.1% concentration of the DMSO used in this study did not affect the viability of the H. pylori.

CFU

The CFUs were determined using the following procedure. Helicobacter pylori cell suspensions were serially diluted 10-fold with a simple-PPLO broth, spread on plates of brain–heart infusion agar (Difco Laboratories) containing 5% horse serum (Gibco, Auckland, NZ), and cultured for 1 week under microaerobic conditions. The CFUs were calculated based on the colony counts and dilution factors of the bacterial cell suspension.

OD in cell suspension

The H. pylori cells were recovered from the simple-PPLO precultures (3mL) via centrifugation (8600g, 5min) and incubated for 24h with or without the steroid (progesterone: 100μM or 17αPSCE: 100μM) in a fresh simple-PPLO broth (3mL) with continuous shaking under microaerobic conditions in the dark. After the incubation, the H. pylori cells were harvested via centrifugation (8600g, 5min), resuspended in sterile saline (3mL), and examined using a spectrophotometer (Versa max microplate reader: Molecular Devices Coet al, CA) to measure the OD660nm of the cell suspensions (200μL).

Bacterial morphology

Helicobacter pylori cells were microscopically observed with differential interference using an AX80T microscope (Olympus Coet al, Ltdet al, Tokyo, Japan).

Proliferation assay

The H. pylori cell suspensions (60μL) were transferred into a simple-PPLO broth (3mL) containing the steroid at various concentrations and cultured for 24h with continuous shaking under microaerobic conditions in the dark. The CFUs were determined after the cultures.

Detection of proteins in the cell supernatant

Helicobacter pylori cells (approximately 108.3CFUmL−1) were suspended in phosphate-buffered saline (PBS: 15mL) containing progesterone (100μM) or 17αPSCE (100μM) and incubated for 5h with continuous shaking under microaerobic conditions in the dark. After the CFU of each cell suspension was measured, the cell supernatant (10mL) was filtrated using a syringe filter (a 0.45μm pore size; Whatman Japan KK, Tokyo Japan), concentrated using a centrifugal filter device (Centriprep YM-3: Millipore Coet al, Bedford, MA), and subjected to 80% acetone precipitation. The precipitates were then dissolved in a sample buffer [50mM Tris (pH 6.8), 5% glycerol, 0.002% bromophenol blue, 2% sodium dodecyl sulfate (SDS), and 40mM dithiothreitol] for SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and developed in the 10–20% polyacrylamide gradient gel (Tefco Coet al, Tokyo Japan) (Laemmli, 1970). The purified flavodoxin (FldA) protein (Shimomura et al, 2007) was also electrophoresed as an authentic sample. After the SDS-PAGE, the proteins in the gel were stained with Coomassie brilliant blue.

FC beads

After FC (50mg; Wako Pure Chemical Industries Ltd.) was dissolved in chloroform (5mL), beads (Iatrobeads 6RS-8060, 1g; Mitsubishi Kagaku Iatron Incet al, Tokyo, Japan) were added to the FC–chloroform solution and gently stirred for 5min at room temperature. Next, the chloroform was completely vaporized at 70°C with a rotary evaporator (Buchi Rotavapor R 114: Shibata Scientific Technology Ltdet al, Saitama, Japan) and the FC was tightly fixed to the beads by heating for 15min at 150°C. The FC beads were then cooled, suspended in distilled water (10mL), and stored at 4°C until use in the experiments. Control beads without FC fixation (100mgmL−1) were also prepared. The only difference between the procedures to prepare the FC beads and FC-free beads was the omission of the FC dissolution in chloroform in the procedure to prepare the latter.

Purification of lipids

Helicobacter pylori membrane lipids were purified using the Folch method (Folch et al, 1957). After the cell pellets were suspended and sonicated in a chloroform–methanol solvent (2:1), the supernatant (800μL) was recovered via centrifugation (10000g, 5min), treated with a 0.9% KCl solution (160μL), stirred vigorously, and centrifuged for 5min at 10000g to separate the water phase from the chloroform phase. The solvent of the recovered chloroform phase was vaporized using a centrifugal concentrator (Tomy Seiko Co. Ltdet al, Tokyo, Japan) to obtain the purified membrane lipids. The membrane lipids were analyzed by thin-layer chromatography (TLC) using a 60% sulfuric acid solution.

Quantification of cholesterol

The FC absorbed into the H. pylori cells was quantified by the following method. After the H. pylori cell suspension (1mL) was cultured for 24h in a simple-PPLO broth (30mL) containing progesterone (5 or 10μM) with continuous shaking under microaerobic conditions in the dark, cell pellets precultured with the progesterone were recovered via centrifugation (8600g, 5min) from the cultures, resuspended in a fresh simple-PPLO broth (30mL) containing FC beads (FC concentration: 250μM), and incubated for 4h with continuous shaking under microaerobic conditions. After the incubation, the FC beads were removed via centrifugation (10g, 1min) to obtained a supernatant (28mL) containing the H. pylori cells. Cell pellets were recovered via centrifugation (8600g, 5min) and purified into membrane lipids. The purified membrane lipids were dissolved in acetic acid (600μL), mixed with a ferrous chloride reagent [phosphoric acid–sulfuric acid (2:25) solution containing 0.2% FeCl2·6H2O: 400μL], stirred vigorously, and incubated for 15min at room temperature. After color reaction and cooling, the absorbance of the lipid solution (200μL) containing ferrous chloride reagent was measured using a Versa max microplate reader (Molecular Devices Co.) at a wavelength of A550nm. The amount of FC absorbed into the H. pylori cells was quantified based on the FC-standard curve and calculated per the CFU.

Proliferation assay to investigate the presence of FC beads

Helicobacter pylori cell suspension (600μL) was cultured for 24h with various volumes of FC beads (FC concentration: 30–90μM) in a simple-PPLO broth (15mL) containing progesterone (30μM), with continuous shaking under microaerobic conditions in the dark, and the CFUs were then measured. Next, an H. pylori cell suspension (600μL) was cultured for 24h with various concentrations of progesterone (from 10 to 30μM) in a simple-PPLO broth (15mL) containing FC beads (FC concentration: 500μM) or the FC-free beads (in a similar volume), with continuous shaking under microaerobic conditions in the dark, and the CFUs were then measured.

Results

Effects of steroid hormones in inhibiting the growth of H. pylori

In our first experiments, we investigated the effects of the steroid hormones estradiol, androstenedione, and progesterone in inhibiting the growth of H. pylori. When H. pylori (approximately 105.5CFUmL−1) was cultured for 24h in different simple-PPLO broths (3mL) containing single steroids at concentrations ranging from 10 to 100μM, every steroid hormone examined exhibited inhibitory effects on the growth of H. pylori at concentrations >50μM (Fig. 1). Estradiol appeared to act bacteriostatically on H. pylori, as the CFUs of H. pylori cultured in the presence of estradiol at the 50 and 100μM concentrations were entirely unaltered from the baseline CFU (105.5CFUmL−1) before the cultures (Fig. 1a). In contrast, androstenedione and progesterone exhibited growth-inhibitory effects that were dependent on the dose against H. pylori (Fig. 1b and c). Androstenedione, however, was less potent than progesterone in inhibiting the growth of H. pylori. The CFUs of H. pylori cultured for 24h with androstenedione at the 100μM concentration were slightly lower than the baseline CFU (105.5CFUmL−1), whereas the CFUs of the organisms cultured for 24h with progesterone at the 100μM concentration were below the limits of detection. Thus, progesterone demonstrated the most effective anti-H. pylori action of the three steroid hormones, and it appeared that this action was bactericidal to H. pylori. This led us to investigate the antibacterial effect of progesterone on H. pylori in more detail.

1

Effect of steroid hormones in inhibiting the growth of Helicobacter pylori. The H. pylori strain NCTC 11638 was cultured for 24h with various steroid hormones at the indicated concentrations in a simple-PPLO broth, and the CFUs were determined. The results are shown as mean CFU±SD obtained from three independent experiments. (a), (b), and (c) show the chemical structures and growth-inhibitory effects of estradiol (E2), androstenedione (ASD), and progesterone (PS), respectively, on H. pylori. The gray belts in the graphs indicate the baseline CFUs before the cultures.

Anti-H. pylori action of progesterone derivatives

Progesterone has two derivatives: 17α-hydroxyprogesterone (17αPS) and 17α-hydroxyprogesterone caproate (17αPSCE). The derivatives 17αPS and 17αPSCE are modified by a hydroxyl group and an acyl group (caproic acid), respectively, at the carbon 17 position of the progesterone framework (Fig. 2). Noting this, we next examined the anti-H. pylori action of 17αPS and 17αPSCE using the simple-PPLO broth. Surprisingly, 17αPS, a natural progesterone derivative, had no influence on the growth of H. pylori. Even in the presence of 17αPS (100μM) for 24h, the CFUs were comparable to the control CFU (108CFUmL−1) of the organisms cultured for 24h without steroid (Fig. 2a). In contrast, 17αPSCE, a synthetic progesterone derivative, had a stronger anti-H. pylori action than progesterone, and the CFUs were below the limits of detection when the organisms were cultured for 24h with 17αPSCE at a 10μM concentration (Fig. 2b). Incidentally, caproic acid, a constituent of 17αPSCE, did not affect the viability of H. pylori even when added to the cell suspension at a 100μM concentration (data not shown).

2

The chemical structures and anti-Helicobacter pylori action of progesterone (PS) derivatives. The H. pylori strain NCTC 11638 was cultured for 24h with 17αPS (a) or 17αPSCE (b) at the indicated concentrations in a simple-PPLO broth, and the CFUs were determined. The results are shown as mean CFU±SD obtained from three independent experiments. The gray belts in the graphs indicate the baseline CFUs before the cultures.

Bacteriolytic activity of progesterone and 17αPSCE on H. pylori

Next, we measured the OD660nm in the cell suspensions after the H. pylori (108CFUmL−1) was incubated for 24h with progesterone (100μM) or 17αPSCE (100μM) in a simple-PPLO broth (3mL). As it turned out, the OD660nm of the cell suspension incubated with progesterone or 17αPSCE declined to less than half of that in the control cell suspension of the H. pylori incubated in the absence of steroid (data not shown). These results suggest that H. pylori cells are lysed by the action of progesterone and 17αPSCE.

Next, we carried out a series of experiments to examine whether progesterone and 17αPSCE induce the cell lysis of H. pylori via membrane injury. When PBS was used in place of the simple-PPLO broth, the CFUs of H. pylori incubated for 5h with progesterone (100μM) were conspicuously reduced in comparison with the baseline CFU before the incubation (Fig. 3a). The control CFUs of H. pylori incubated for 5h without steroids were also reduced in comparison with the baseline CFU, but the magnitude of reduction was smaller in the control CFUs than in the CFUs observed in the H. pylori incubated with progesterone. When the H. pylori was incubated for 5h with 17αPSCE (100μM) in PBS, the CFUs declined sharply, nearly reaching the limits of detection.

3

Bacteriolytic activity of progesterone (PS) and 17αPSCE on Helicobacter pylori. (a) The H. pylori strain NCTC 11638 was incubated for 5h with PS (100μM) or 17αPSCE (100μM) in PBS (15mL) under microaerobic conditions, and the CFUs were determined. The results are shown as mean CFU±SD obtained from three independent experiments. Lane i shows the baseline CFU before the incubation. (b) The H. pylori strain NCTC 11638 was incubated for 5h with PS (100μM) or 17αPSCE (100μM) in PBS (15mL) under microaerobic conditions, and the cell supernatant was recovered via filtration. The proteins in the cell supernatant (10mL) were analyzed by SDS-PAGE. The results shown are representative of the findings obtained from two independent experiments. The left lane shows the band of purified FldA protein.

The proteins in the cell supernatant (PBS: 10mL) obtained from the H. pylori incubated for 5h with progesterone (100μM) or 17αPSCE (100μM) were analyzed by SDS-PAGE (Fig. 3b). The protein bands detected in the cell supernatant of H. pylori incubated with progesterone or 17αPSCE were considerably denser than the protein bands detected in the control cell supernatant of H. pylori incubated without steroid. A band for flavodoxin (FldA) was found among the other protein bands. The amounts of FldA protein detected in the cell supernatant correlated closely with the decreases of CFU: the FldA protein band became more noticeable when the CFU decreased by a greater magnitude. As FldA is an electron acceptor of the oxidoreductase that catalyzes acetyl-CoA synthesis in H. pylori cell (Hughes et al, 1995), we can assume that FldA is the intracellular protein. These results, thus, suggest that progesterone and 17αPSCE exert deleterious effects on the cell membrane of H. pylori and induce cell lysis more promptly than autolysis, resulting in abundant leakage of intracellular proteins (especially FldA protein) outside of the cells.

To confirm the cell lysis of H. pylori, we examined the bacterial morphologies using a differential interference microscope (data not shown). When H. pylori (107CFUmL−1) was incubated for 24h in a simple-PPLO broth (3mL) in the presence or absence of the steroids, the control cell suspension of H. pylori incubated without the steroids harbored the organisms in both mixed rod and coccoid forms. In contrast, the cell suspension of the H. pylori incubated with progesterone (100μM) or 17αPSCE (100μM) harbored hardly any organisms, although objects such as cellular debris were observed.

Inhibitory effect of progesterone on the absorption of FC in H. pylori

Helicobacter pylori is known to aggressively absorb any FC present in a medium, although the FC-binding site on the H. pylori cell surface has yet to be identified. In light of this, we hypothesized that progesterone acts on FC-binding sites on the H. pylori cell surface when inducing cell lysis. To verify this hypothesis, we carried out the following experiments using FC beads. After a 24-h preculture of H. pylori (106.3CFUmL−1) with progesterone (5 or 10μM) in a simple-PPLO broth (30mL), the H. pylori cells (108.3CFUmL−1) recovered were incubated for 4h in a simple-PPLO broth (30mL) containing FC beads (FC concentration: 250μM). Thereafter, the amount of FC absorbed into the H. pylori cells was quantified. The amount of FC per CFU obviously tended to reduce by preculturing H. pylori with progesterone (Fig. 4a). These results suggest that progesterone strongly binds to the H. pylori cell surface and thereby obstructs the FC absorption of H. pylori by inhibiting the cell surface binding of FC. Incidentally, progesterone had no influence on the growth of H. pylori at the 5 and 10μM concentrations: the CFUs of the H. pylori cultured with progesterone were similar to the control CFU of the H. pylori cultured without progesterone (data not shown).

4

Inhibitory effect of progesterone (PS) on the FC absorption of Helicobacter pylori, and the inhibitory effect of FC on the anti-H. pylori action of PS. After the H. pylori strain NCTC 11638 (106.3CFUmL−1) was cultured for 24h with PS (5 or 10μM) in a simple-PPLO broth (30mL), the H. pylori cells (108.3CFUmL−1) were recovered, resuspended in a fresh simple-PPLO broth (30mL) containing the FC beads (FC concentration: 250μM), and incubated for a further 4h under microaerobic conditions. (a) The FC amount absorbed into the H. pylori cells was quantified after the incubation. The results are shown as mean FC amount±SD obtained from three independent experiments. (b) The membrane lipids were purified and analyzed by TLC after the incubation. The H. pylori cells were pretreated with 10μM concentration of PS. The amount of lipid subjected to the TLC analysis was 250μg per lane. The results shown are representative of the findings obtained from two independent experiments. CGL, cholesteryl-α-d-glucopyranoside; CAG, cholesteryl-6-O-tetradecanoyl-α-d-glucopyranoside; CPG, cholesteryl-6-O-phpsphatidyl-α-d-glucopyranoside; PE, phosphatidylethanolamine; PG-CL, phosphatidylglycerol-cardiolipin. (c) After dispensing FC beads into a simple-PPLO broth (15mL) to attain the indicated FC concentrations, the H. pylori strain NCTC 11638 was cultured for 24h with the FC beads in the presence of PS (30μM). The CFUs were determined after the cultures. The results are shown as mean CFU±SD obtained from three independent experiments. The gray belt indicates the baseline CFU before the cultures. (d) The CFUs were determined after the H. pylori strain NCTC 11638 was cultured for 24h with PS at the indicated concentrations in a simple-PPLO broth (15mL) containing the FC beads (FC concentration: 500μM) (closed circles) or the FC-free beads (approximately the same volume) (open circles). The results are shown as mean CFU±SD obtained from three independent experiments. The gray belt indicates the baseline CFU before the cultures.

Helicobacter pylori glucosylates the absorbed FC and synthesizes cholesteryl glucosides (CGs). With this in mind, we decided to examine the influence of progesterone on the glucosylation of FC. After the 24-h preculture of H. pylori (106.3CFUmL−1) in the presence or absence of progesterone (10μM) in a simple-PPLO broth (30mL), the H. pylori (108.3CFUmL−1) recovered was incubated for 4h with FC beads (FC concentration: 250μM) in a simple-PPLO broth (30mL), and the membrane lipids were purified. The TLC analysis detected the CGs (CGL, CAG, and CPG) in the membrane lipids of H. pylori precultured with progesterone (Fig. 4b), although no FC was found to have accumulated within the lipids. Meanwhile, the CG levels detected in the membrane lipids of H. pylori precultured with progesterone were similar to the CG levels detected in the membrane lipids of H. pylori precultured without progesterone. These results indicate that progesterone exerts no inhibitory effects on the enzymes involved in the CG synthesis.

Inhibitory effect of FC on the anti-H. pylori action of progesterone

Next, we examined whether FC conversely inhibits the anti-H. pylori action of progesterone. When the H. pylori (106.6CFUmL−1) was cultured for 24h with FC beads at various volumes (FC concentration: 30–90μM) in a simple-PPLO broth (15mL) containing progesterone (30μM), the FC did not inhibit the anti-H. pylori action of progesterone: the CFU increase was not observed in any concentrations of FC (Fig. 4c). The 60 and 90μM concentrations of FC seemed to decrease the CFU of H. pylori cultured with progesterone (30μM), but the magnitude of CFU decreases was negligible. These results, at least, indicate that FC does not competitively inhibit the anti-H. pylori action of progesterone. This compelled us, in turn, to examine the inhibitory effect of a high concentration of FC on the anti-H. pylori action of progesterone. When the H. pylori (106.3CFUmL−1) was cultured for 24h with progesterone at concentrations ranging from 10 to 30μM in a simple-PPLO broth (15mL) containing FC beads (FC concentration: 500μM) or FC-free beads (approximately the same volume), FC at the highest concentration (500μM) had a noticeable influence on the anti-H. pylori action of the progesterone: the growth-inhibitory curve of H. pylori cultured with progesterone in the presence of FC-beads shifted from the control growth-inhibitory curve of H. pylori cultured with progesterone in the presence of FC-free beads to the right side (Fig. 4d). These results indicate that FC noncompetitively inhibits the anti-H. pylori action of progesterone. And taken in combination with the results shown in Fig. 4a and b, they also strongly suggest that progesterone nonreversibly binds to the H. pylori cells and thereby induces the cell lysis and/or inhibits the FC absorption of H. pylori.

Discussion

Earlier investigations (including our own) have shown that H. pylori morphologically converts from a bacillary form to a coccoid form when the organism is exposed to various stresses such as excessive oxygen, alkaline pH, or long-term culture (Catrenich & Makin, 1991; Benaïssa et al, 1996; Donelli et al, 1998; Shimomura et al, 2004). Cells that change to a coccoid form lack the ability to form colonies on an agar plate, which makes it very difficult to accurately determine the CFU in coccoid-converted H. pylori. The present study revealed that estradiol has the potential to inhibit the growth of H. pylori. We also confirmed that coccoid cells are microscopically unobserved in H. pylori cultured with estradiol (data not shown). Taken together, these results show that estradiol acts bacteriostatically on H. pylori without inducing the coccoid cell conversion. Another recent study demonstrated that estradiol somehow protects against the development of H. pylori-induced gastric cancer in a mouse model (Ohtani et al, 2007). The bacteriostatic action of estradiol may play some role in mechanisms preventing the development of H. pylori-induced gastric cancer. Further investigations will be necessary to elucidate the relationship between estradiol and H. pylori.

Among the five bacterial species investigated in this study, H. pylori was the only species susceptible to the bactericidal action of progesterone and 17αPSCE (see Appendix S3). The other species, Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, and Staphylococcus epiderimidis, all resisted this action. Given that H. pylori is a unique bacterial species that aggressively assimilates exogenous steroids, we can assume that progesterone and 17αPSCE attacked only one of the five bacterial species, H. pylori.

Our present results show that progesterone inhibits the FC absorption of H. pylori, and conversely, that a relatively high concentration of FC (500μM) inhibits the anti-H. pylori action of progesterone (especially 20μM). Progesterone and FC seem to bind to identical sites on the H. pylori cell surfaces and thereby inhibit each other's effects. This suggests that H. pylori may express a certain component, such as a steroid-binding protein, on the cell surface. Further investigations will be required to elucidate whether such a steroid-binding protein does indeed exist in H. pylori.

In addition to demonstrating the anti-H. pylori action of progesterone, our findings indicate that it may be possible to design a novel anti-H. pylori steroidal agent using progesterone as a fundamental structure. We now know that we can augment the bactericidal capability of progesterone on H. pylori by modifying progesterone with the short-chain fatty acid, caproic acid, at the carbon 17 position, and conversely, that we can abolish this effect by attaching a hydroxyl group to the same position (see Fig. 2 and Appendix S4). Thus, the acylation at the carbon 17 position of the progesterone molecule appears to play an important role in reinforcing the anti-H. pylori action. Further investigations will be essential for the development of new progesterone derivatives as adjuvants to the conventional treatments for H. pylori.

Supporting Information

Appendix S1. Acclimatization of Helicobacter pylori to a medium prepared without 2,6-di-O-methyl-β-cyclodextrin (dMβCD) or serum.

Appendix S2. Effect of 2,6-di-O-methyl-β-cyclodextrin (dMβCD) on the anti-Helicobacter pylori action of pro-gesterone (PS) and 17α-hydroxyprogesterone caproate (17αPSCE).

Appendix S3. The minimum inhibitory concentrations (MICs) of progesterone (PS) and 17α-hydroxyprogesterone caproate (17αPSCE) for Helicobacter pylori and other representative Gram-negative and Gram-positive bacteria.

Appendix S4. The time-dependent antibacterial effects of progesterone (PS) and 17α-hydroxyprogesterone caproate (17αPSCE) on Helicobacter pylori.

Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

Acknowledgements

This publication was subsidized by JKA through its promotion funds from KEIRIN RACE.

Footnotes

  • Editor: Arnoud van Vliet

References

View Abstract