1. Introduction
Chlorhexidine is one of the most widely used biguanides for antiseptic purposes, such as skin surface preparation and intravascular catheter maintenance. Chlorhexidine inhibits bacterial growth by disrupting the structure of the cell membrane, leading to the leakage of cellular contents [1]. Chlorhexidine gluconate (CHG) or chlorhexidine acetate is generally used because of the low solubility of chlorhexidine base. Although the antiseptic activity of CHG has been validated in gram-positive and gram-negative bacteria and enveloped viruses, its efficacy against mycobacteria, nonenveloped viruses, fungi, and bacterial endospores is limited [2]. In spore-forming bacteria, chlorhexidine acts as a sporestatic agent rather than a sporicidal agent [3] [4] [5] [6]. Because CHG is generally ineffective against bacterial endospores, less attention has been paid to the differences in the action of CHG in the genus Bacillus. The efficacy of CHG against bacterial endospores was mainly examined using B. subtilis, the representative species of the genus. The bactericidal or inhibitory concentration of chlorhexidine in Bacillus spp. had not been studied extensively, because most antiseptic agents were generally ineffective against bacterial spores. Bacillus consists of spore-forming, facultative anaerobic or aerobic gram-positive bacilli, and Bacillus spp. is ubiquitously isolated from the environment, including the skin surface. The genus Bacillus comprises many species ranging from pathogenicity to animals, including Bacillus cereus, to harmless species, including Bacillus subtilis. Among them, the harmful species are classified into the cereus group, which comprises B. cereus sensu stricto, Bacillus thuringiensis, Bacillus anthracis, and other species Bacillus weihenstephanensis, Bacillus mycoides, and Bacillus pseudomycoides are also included in this group [7]. Several studies examined the activity of chlorhexidine against B. subtilis, whereas studies on the sensitivity of harmful species in the B. cereus group, which is the main target of disinfection, to CHG have been limited.
This study aimed to evaluate the susceptibility of the B. cereus group to CHG and compare it between other genus Bacillus species. In this study, we divided bacterial strains into two groups, the cereus group or subtilis group. The cereus group consists of representative strains of B. cereus, B. thuringiensis, and B. weihenstephanensis, and clinical isolates of B. cereus. The subtilis group consists of the B. subtilis, Bacillus amyloliquefaciens, and Bacillus liqueniformis. The bacteriostatic and bactericidal concentrations of CHG were compared among Bacillus spp. The decrease in bacterial counts under bacteriostatic concentrations was assessed by a time-kill assay using representative strains.
2. Materials and Methods
2.1. Bacterial Strains and Culture Conditions
The Bacillus strains employed in this study are listed in Table 1. The strains included 3 representative or genome strains of B. cereus sensu stricto (ATCC14579, type strain, ATCC10987, NC7401), and 10 clinical strains of B.
![]()
Table 1. Bacterial strains used in this study and MIC, MBC and disk diffusion test.
In the table, descriptive statistics are provided for the minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of CHG, and the disk diffusion test was performed using polymyxin B and colistin. “T” in the “strain” column denotes the type strain. The sensitivity to polymyxin B or colistin was indicated by the zone of inhibition (mm), and “-” indicated no observation of growth inhibition around the antibiotic disk. The column “year” indicates the year that the strain was first described in the literature (a), separated from a source (b), or deposited in a bacterial culture collection (c).
cereus (BL6459, BL6460, TH119, TH120, STKT, 669601, 669602, SUMK, NC1241, and H27-5) [8] - [13]. For representative strains of other cereus groups, B. thuringiensis NBRC101235 (type strain) and B. weihenstephanensis NBRC101238 (type strain), were chosen [14]. B. thuringiensis NBRC101235 is listed as ATCC10792 in the American Type Culture Collection. B. weihenstephanensis NBRC101238 is the synonym of B. mycoides, and is listed as DSM11821 in Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures. B. subtilis sensu stricto B. subtilis subsp. subtilis NBRC13719 and PCI219, B. amyloliquefaciens IFO3007 and IFO3025, and B. licheniformis NBRC14206 [15] [16] [17] [18] [19]. E. coli ATCC25922 and S. aureus ATCC27664 were used as references for gram-negative and gram-positive strains, respectively. All isolates were grown in MH broth (Becton, Dickinson and Company, Sparks, MD, USA), MH agar, or standard nutrient agar (Atect Corp., Shiga, Japan) at 37˚C under atmospheric conditions. In this study, the counts of living bacterial cells were measured as the average colony-forming units (CFUs) on three MH agar plates.
2.2. Chemical Materials
CHG solution (20% w/v HIBITANE®, Sumitomo Dainippon Pharma Co., Ltd., Osaka, Japan) was used as a formulation of chlorhexidine. CHG was serially diluted with sterilized water in each concentration and used in sensitivity studies. For minimum bactericidal concentration (MBC) experiments, inactivation solution containing 3% Tween-80 (Sigma-Aldrich Co., St. Louis, MO, USA) and 0.3% lecithin (Fujifilm Wako Pure Chemical Co., Osaka, Japan) was prepared [20].
2.3. Determination of the MIC of CHG
The minimum inhibitory concentration (MIC) of CHG was assessed by the agar dilution method, referring to the method for dilution antimicrobial susceptibility tests of clinical and laboratory standard institute [21]. All tested bacterial strains were cultured overnight in MH broth, and suspensions were diluted in Dulbecco’s phosphate-buffered saline (PBS, without calcium chloride and magnesium chloride, Sigma-Aldrich Co.) to McFarland 0.5 standard, equivalent to approximately 1 × 108 CFUs/mL. The diluted broths were added to MH agar plates containing CHG at a concentration of 0.5, 1, 1.5, 2, 2.5, 3, 4, or 5 mg/L. After incubation at 37˚C for 48 h, the minimum concentration at which bacterial growth was inhibited was determined as the MIC.
2.4. Determination of the MBC of CHG in Spores
Each strain was seeded onto a standard nutrient agar plate. After incubation overnight at 37˚C followed by 3 days at 25˚C days, the formation of endospores was confirmed via microscopic observation using the modified Wirtz-Conklin staining method referring to rapid staining techniques [22] [23]. The bacterial culture containing endospores was suspended in PBS to McFarland 0.5 standard. CHG was added at a final concentration of 50, 100, 250, 500, 1000, 2500, 5000, or 10,000 mg/L. After incubation at room temperature for 30 min, nine volumes of inactivation solution were added to the mixture to inactivate CHG. Then, 10 μL of bacterial suspension were placed on MH agar plates without CHG. After incubation at 37˚C for 48 h, the minimum concentration at which bacteria were killed was determined as the MBC. These experiments were repeated at least twice to confirm the results.
2.5. Time-Kill Test Using a Sub-Bactericidal Concentration of Chlorhexidine
The modified time-kill test was performed using eight Bacillus isolates to evaluate the changes of bacterial cell populations under bacteriostatic concentrations of chlorhexidine. Initially, 100 μL of an overnight bacterial culture was inoculated into 100 mL of fresh MH broth, and chlorhexidine solution was added at a final concentration of 10 mg/L. The mixture in glass flasks was incubated at 37˚C with shaking at 160 rpm and recovered after 15 s, 10 min, 30 min, 1 h, 3 h, 6 h, 12 h, 24 h, or 48 h. The recovered mixture was seeded onto MH agar plates. After incubation at 37˚C overnight, the surviving bacterial population was calculated by counting colonies on the agar plate. The survival of bacterial cells was evaluated in comparison to the initial cell number. The experiments were repeated at least twice to confirm the results for each tested strain.
2.6. Microscopic Observation with CHG
Microscopic imaging was performed during incubation with several concentrations of CHG. Overnight cultures of B. cereus ATCC10987 and B. amyloliquefaciens IFO3007 were diluted in PBS to McFarland 0.5 standard and then incubated with 0, 10, 25, 50, or 100 mg/L CHG for 30 min at room temperature. One loop of the incubation mixture was placed on a slide glass, which was stained using the standard Gram staining method. After staining, the slides were examined using the oil immersion objective of an optical microscope (BX51 with DP73, Olympus, Tokyo, Japan).
2.7. Disk Diffusion Test Using Colistin and Polymyxin B
All isolates were subjected to a disk diffusion test using colistin (10 μg) and polymyxin B (300 μg), because the two antibiotics target bacterial cell membrane. A 0.5 McFarland standard suspension of the isolate was prepared and spread on an MH agar plate, and each antibiotic disk was placed on the plate. Plates were incubated at 37˚C for 20 h, and zone diameters (mm) were measured. The disk diffusion test was repeated at least two times, and the average diameter was calculated.
3. Results
3.1. Minimum Inhibitory Concentrations (MICs) of CHG
The MICs of CHG were higher in the B. cereus group than in the B. subtilis group (Table 1). The MICs of CHG ranged from 3 to 4 mg/L in the B. cereus group, which consisted of 13 strains of B. cereus, B. thuringiensis NBRC101235, and B. weihenstephanensis NBRC101238. Although chlorhexidine is generally considered more effective against gram-positive bacteria, the MICs of CHG were higher for the B. cereus group than for the gram-positive coccus Staphylococcus aureus ATCC27664. The MICs were also higher than that of the gram-negative strain Escherichia coli ATCC25922. It should be noted that the MICs of S. aureus and E. coli were both 1.5 mg/L in the present study.
In comparison, the MIC of CHG in the B. subtilis group was 1.5 mg/L, equivalent to that of the controls.
3.2. Minimum Bactericidal Concentrations (MBCs) of CHG for Spores of the Genus Bacillus
The MBCs of CHG were slightly higher for the B. cereus group than for the B. subtilis group (Table 1). The MBCs of CHG ranged from 2,500 - 10,000 mg/L for the B. cereus group, versus 1,000 - 2,500 for the B. subtilis group. In both groups, the sporicidal concentration was nearly identical to the practically used concentration.
3.3. Time-Kill Assay of Chlorhexidine
The time-kill test was used to examine the survival ratio of Bacillus spp. under sub-bactericidal concentrations (Figure 1). In this study, all strain counts were decreased to below detectable levels after 48 h of incubation with 10 mg/L CHG in Mueller-Hinton (MH) broth. As shown in Figure 1, the counts of all tested strains decreased to 10% of the control level (MBC90) within 30 min. There were no significant differences in effects between the B. cereus and B. subtilis groups.
![]()
![]()
![]()
![]()
Figure 1. Bar graph of the time-kill test using a sub-bactericidal concentration (10 mg/L) of chlorhexidine. The time before the addition of chlorhexidine was set as time zero. Bar height represents mean survival cell number with colony forming unit per mL (CFU/mL). The bracket on the bars indicates the standard error (S.E.) of triplicate determinations in three experiments. In total, eight bacterial strains were examined; B. cereus ATCC14579 (A), B. cereus ATCC10987 (B), B. cereus ATCC10987 NC7401 (C), B. thuringiensis NBRC101235 (D), B. weihenstephanensis NBRC101238 (E), B. subtilis NBRC13719 (F), B. amyloliquefaciens IFO3007 (G), and B. amyloliquefaciens IFO3025 (H). Each Bacillus strain was incubated at 37˚C with shaking at 160 rpm with 10 mg/L chlorhexidine and recovered after 15 s, 10 min, 30 min, 1 h, 3 h, 6 h, 12 h, 24 h, and 48 h. The horizontal dot line in the graphs indicated the 10% survival rate (90% minimum bactericidal concentration; MBC90) as a time zero for standard. ND indicated “not detected,” meaning that no living bacteria were isolated the mixture.
3.4. Microscopic Observation
Cellular damage induced by CHG was assessed in both B. amyloliquefaciens IFO3025 and B. cereus ATCC10987 (Figure 2). Damaged cells and debris were observed in B. amyloliquefaciens IFO3025 incubated with 25 mg/L CHG (Figure 2(A)), whereas no cellular damage or debris were observed in B. cereus ATCC10987 at this concentration (Figure 2(B)). Meanwhile, cellular damage and debris were observed in B. cereus ATCC10987 exposed to 50 mg/L CHG (Figure 2(B)). This morphological observation supported the differences in the susceptibility to CHG between the B. cereus and B. subtilis groups. In a preliminary study, other strains belonging to the cereus group, B. cereus ATCC14579, NC7401, B. thuringiensis NBRC101235, and B. weihenstephanensis NBRC101238, were also observed to have cell damage similar to ATCC10987. Other strains belonging to the subtilis group, B. subtilis ATCC13719 and B. amyloliquefaciens IFO3007 in the subtilis group, were also observed to have similar cell damage B. amyloliquefaciens IFO3025 (data not shown).
![]()
(A) ![]()
(B)
Figure 2. Microscopic observation of B. amyloliquefaciens IFO3025 (A) and B. cereus ATCC10987 (B). The pre-culture of the bacterial mixture was incubated with 10, 25, 50, or 100 mg/L chlorhexidine for 10 min and observed under the oil immersion objective of an optical microscope (×1000).
![]()
Figure 3. Disk diffusion test of colistin and polymixin B. The colistin and polymyxin disks were placed on which B. amyloliquefaciens IFO3025 (A) or B. cereus ATCC10987 (B) were grown. In both images, colistin disk was on the right side, and polymyxin B disk left side.
3.5. Disk Diffusion Test Using Colistin and Polymyxin B
All strains of Bacillus were tested with colistin and polymyxin disk on Muller-Hinton agar plate. Typical plate images of the growth inhibition zone around colistin and polymyxin disks were shown in Figure 3. Since no criteria have been set for drug resistance of polymyxins by the disk method, the diameters of inhibition zone were listed in Table 1. In the sensitivity for polymyxin B, the cereus group showed smaller inhibition zone (average 10.1 mm, S.D. ± 0.9) compared to that of subtilis group (average 15.3 mm, S.D. ± 1.0). In the colistin’s sensitivity, the cereus group showed no inhibition zone, whereas the subtilis group showed an inhibition zone (average 10.1 mm, S.D. ± 0.8).
4. Discussion
In this study, the MICs of CHG were higher for the B. cereus group than for the B. subtilis group despite the use of concentrations below the practical preparation level. The Centers for Disease Control and Prevention found that 4% w/v chlorhexidine exhibits effective antiseptic activity [24]. In Japan, chlorhexidine formulations are usually used at concentrations of 0.02% - 0.5% (w/v) for skin surfaces, whereas these concentrations are contraindicated for exposure in the bladder, vagina, and ear mucosa. For these reasons, even though chlorhexidine does not sufficiently kill bacterial cells, including spores, the drug can be expected effectively to inhibit the growth of Bacillus spp. at practical concentrations. The MBC results supported prior findings that the endospores of Bacillus spp. are highly resistant to CHG [5] [6]. CHG functioned as a sporestatic agent rather than a sporicide for the endospores of Bacillus spp. One of the advantages of chlorhexidine is the long-term duration of its antiseptic effects. Although its short-term effects are inferior to those of other disinfectants such as glutaraldehyde, chlorhexidine is expected to be useful for disinfecting surfaces contaminated by Bacillus spp. because it disinfects while inhibiting growth over a long period.
Our result suggests the existence of intrinsic differences in the susceptibility to CHG between the B. cereus and B. subtilis groups. The MIC may be dependent on the bacterial species, strains, determination method, and chemical formulation. Shaker et al. reported that the inhibitory concentration of chlorhexidine acetate for B. subtilis NCTC8236 ranged from 0.4 to 4.75 mg/L depending on the methods and inoculum size [3] [5]. Cheung et al. reported the MIC of chlorhexidine, which was dissolved in MH broth containing 1% ethanol, for B. subtilis 60015 as 0.75 mg/L using the procedures recommended by the National Committee of Clinical Laboratory Standard [25]. However, few reports described the MICs of CHG in Bacillus spp. other than B. subtilis.
One hypothesis is that the differences in the susceptibility to CHG between bacterial groups were associated with adaptation. Increased usage of antiseptics clinically has promoted the acquisition of resistance to CHG in Klebsiella pneumoniae and S. aureus [26] [27]. Chlorhexidine was first developed in the 1950s, and its use has widely spread since the 1970s [28] Because the strains belonging to the B. subtilis group in this study were isolated in the pre-chlorhexidine era, it is necessary to survey the susceptibility to chlorhexidine in modern strains of the B. subtilis group. However, the standard strains, i.e., B. cereus ATCC14579, ATCC10987, B. thuringiensis NBRC101235, and B. weihenstephanensis NBRC-101238, were also isolated in the pre-chlorhexidine era, and they were less sensitive to chlorhexidine than the B. subtilis group. Several strains of B. cereus isolated in the post-chlorhexidine era displayed lower susceptibility to CHG. For these reasons, we suggest that the differences in the susceptibility to CHG reflect intrinsic characteristics in the genus Bacillus rather than a chronological adaptation to chlorhexidine usage in humans.
The molecular mechanisms of the differences in chlorhexidine susceptibility between the examined Bacillus groups remain unclear. Previous studies identified the genes associated with chlorhexidine resistance. In K. pneumoniae, point mutations in phoPQ and smvR, which encode efflux pump regulators, have been linked to chlorhexidine resistance [27] [29]. In S. aureus, mutations in norA/B, which encodes an efflux pump, were suggested to be involved in resistance to chlorhexidine [26]. Interestingly, no gene homologous to norA/B was identified in Bacillus via a BLAST search [30]. However, many genes have been confirmed or hypothesized to encode multidrug efflux pumps in the B. cereus group [31]. Some efflux pumps may be involved in the sensitivity to chlorhexidine.
In general, chlorhexidine is more effective against gram-positive bacteria than against gram-negative bacteria. The negatively charged components of the cell wall and membrane, such as lipopolysaccharide, may act as permeability barriers, and they may be associated with the resistance to cationic antimicrobial agents in gram-negative bacteria [32]. The differential action of chlorhexidine was found in E. coli and B. subtilis [25]. The dented spots caused by chlorhexidine were localized to hemispherical caps in B. subtilis, whereas these spots were dispersed throughout the cell in E. coli. Several negatively charged phospholipids, such as cardiolipin and phosphatidylethanolamine, are localized on hemispherical caps in B. subtilis. These phospholipids control membrane integrity during the division of vegetative cells or sporulation [33] [34]. It may be necessary to compare the state of localization and protection of negatively charged phospholipids to clarify the differences in the susceptibility to CHG between the B. cereus and B. subtilis groups.
Differences in sensitivity between the B. cereus and B. subtilis groups were also found in other polypeptide antibiotics that target the bacterial cell membrane. We evaluated the sensitivity to the polymyxins polymyxin B and colistin, also known as polymyxin E, via the disk diffusion method. Polymyxin B and colistin are synthesized by Paenibacillus polymyxa, previously known as Bacillus polymyxa. Polymyxin B is used as a component of selective media for Bacillus species, such as mannitol egg yolk polymyxin agar. Polymyxins are generally considered more useful for controlling gram-negative bacteria rather than gram-positive bacteria. Many species of Bacillus and related genera produce polypeptide antibiotics [35]. Drug resistance mechanisms may have developed during evolution in a niche to prevent self-poisoning by self-produced polypeptide antibiotics that target the cell membrane.
5. Conclusions
The susceptibility of genus Bacillus, mainly focused on the cereus group, to chlorhexidine was compared with that of the subtilis group. The results showed that the growth of the cereus group was inhibited at practical concentrations in vitro. However, the cereus group showed a low sensitivity to chlorhexidine rather than that of the subtilis group.
The limitation of this study is that the number of strains is quite a few, especially for harmless subtlis groups. Hence, it requires attention must be adapted in generalizing the results of this study to the entire genus Bacillus. However, despite the different eras, geographical locations, and species of the isolates showed the clear-cut differences between the cereus and subtilis groups in this study, especially in terms of MIC, suggest that there may be intrinsic differences between species in their susceptibility to chlorhexidine.
The effects of chlorhexidine on spore-forming bacteria have been described mainly for the B. subtilis. The main target of disinfection should not be harmless B. subtilis, but a group of harmful cereus group. We argue to the need for future evaluations of the effectiveness of disinfectants to be conducted on the cereus group.
Acknowledgements
This study was supported by a Grant-in-Aid for Research from Aichi University of Education. We would like to thank Enago (https://www.enago.jp/) for the English language review and editing.