Antimicrobial activity of sulfur nanoparticles: Effect of preparation methods

2.1 Materials

Sodium thiosulfate pentahydrate, sodium sulfide, and elemental sulfur were purchased from Sigma-Aldrich (St. Louis, MO, USA). Chitosan (CS-001, molecular weight: 180–200 kDa; viscosity: 110 cp in 1% acetic acid solution at 25 °C; the degree of deacetylation: 90%) was procured from Fine Agar Co., Ltd. (Damyang, Jeonnnam, Korea). Tryptic soy broth (TSB), brain heart infusion broth (BHI), and agar powder were purchased from Duksan Pure Chemicals Co., Ltd (Ansan, Gyeonggi-do, Korea). Yeast malt broth (YMB), and yeast malt agar (YMA) were obtained from KisanBio Co., Ltd. (Seoul, Korea). Potato dextrose broth (PDB), potato dextrose agar (PDA), and peptone water were purchased from Difco (Becton Dickinson, Sparks, MD, USA). YM 3 M Petri film was obtained from 3 M Corporation (St. Paul, MN, USA). Test microbial strains, Staphylococcus aureus (ATCC 13565, enterotoxin A), Listeria monocytogenes ATCC 15313, Escherichia coli O157: H7 (ATCC 11234), Candida albicans (ATCC 18804), and Aspergillus flavus (ATCC 22546), were obtained from the Korean Culture Center of Microorganisms (KCCM, Seoul, Korea).

2.2 Preparation of sulfur nanoparticles

Sulfur nanoparticles (SNPs) were prepared using elemental sulfur (S⁰) and sodium sulfide following the procedure shown in Fig. 1. For this, 20 mL of 1 M sodium sulfide was heated to 100 °C and followed by the addition of 1.6 g of elemental sulfur reacted for 1 h with vigorous stirring. The yellow color of elemental sulfur turned reddish-orange, indicating the formation of polysulfide and designated as SNP_ES (Guo et al., 2006; Paralikar and Rai, 2017; Soleimani et al., 2013). Then 40 mL of a 0.5 wt% chitosan solution (dissolved in 1% acetic acid) was added to the polysulfide solution, and an 18 wt% hydrochloric acid solution was added dropwise with stirring until the color of the solution turned into milky white. The formed nanoparticles were collected by precipitation using a centrifuge and washed three times with distilled water until the pH reached 7. The final product was dried at 50 °C and designated SNP_ES. For comparison, spherical SNPs were prepared using the acidification method of sodium thiosulfate according to the method described previously (Shankar et al., 2018; Chaudhuri and Paria, 2011). Briefly, 2.482 g of sodium thiosulfate pentahydrate was dissolved in 900 mL distilled water and added 100 mL of dilute HCl (0.2 M) with stirring. As described in previous reports (Shankar et al., 2018; Shankar and Rhim, 2018), for coating with chitosan, the precipitated SNPs were washed and dispersed in 50 mL of 0.5 wt% chitosan solution with stirring for 1 h. The SNPs were collected, washed until the pH reached 7, and dried as described above and designated SNP_STS (see Fig. 2).

Close

Fig. 1

The schematic diagram for the preparation of sulfur nanoparticles from elemental sulfur.

Export to PPT

Close

Fig. 2

UV–vis spectra of sulfur nanoparticles.

Export to PPT

2.3 Characterization

The UV–vis light absorption patterns of the SNPs were examined in the range of 200–700 nm using a UV–visible spectrophotometer (Mecasys Optizen POP Series UV/Vis, Seoul, Korea) to confirm the formation of SNPs.

The morphology such size and shape of the SNPs were evaluated using a transmission electron microscopy (TEM). For this, 10 μL of SNPs suspension in ethanol was dropped on the carbon-coated copper grid and allowed to dry. The morphological image was observed using the JEOL-1010 TEM instrument (Tokyo, Japan) operated at an accelerating voltage of 200 kV. The particle size distribution of the SNPs was performed using Image J software (ImageJ 1.46r, National Institute of Health, USA).

Elemental analysis of the SNPs was performed using energy-dispersive X-ray spectroscopy (EDX) equipped on SEM instrument (FE-SEM, S-4800, Hitachi Co., Ltd., Matsuda, Japan).

X-ray diffraction patterns of SNPs were investigated using an XRD diffractometer (PANalytical X’pert Pro MRD Diffractometer, Amsterdam, Netherlands). The spectra were recorded using Cu Kα radiation (wavelength of 0.1541 nm) and a nickel monochromator operated at the voltage of 40 kV and a current of 40 mA at diffraction angles in the range of 2θ = 5-80° with a scanning speed of 0.4°/min.

The thermal stability of the SNPs was evaluated using a thermogravimetric analyzer (Hi-Res TGA 2950, TA Instrument, New Castle, DE, USA). About 10 mg of each sample in a standard aluminum pan was heated from room temperature to 600 °C at a heating rate of 10 °C/min under a nitrogen flow rate of 50 cm³/min. Derivatives of TGA (DTG) were calculated using the central finite difference method.

The surface charge of NPs was evaluated by zeta potential and particle size using the dynamic light scattering (DLS) method using a Zeta PALS-zeta potential analyzer (Brookhaven Instruments Corporation, NY, USA).

2.4 Antibacterial activity of SNPs

The antibacterial activity of the SNPs for E. coli, S. aureus, L. monocytogenes, and C. albicans was investigated using a disk diffusion method. For this, 100 µL of microbial suspensions (∼10⁶ CFU/mL) were inoculated by spreading on agar plates. Then, added 30 μL of each SNPs suspension (512 µg/mL) onto a sterile paper disc surface (Advantec 8 mm, Toyo Roshi Kaisha, Ltd., Tokyo, Japan), placed on the agar surface and incubated at 36 °C for 24 h. The zone of inhibition around the paper disc was measured using a caliper. For A. flavus, agar disc diffusion assay was performed by the same procedure as described above except for incubation temperature and time, which was incubated at 25 °C for 4 days to determine the zone of inhibition.

In addition, the minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of the SNPs for E. coli and L. monocytogenes were determined using a modified broth microdilution method (Shankar et al., 2018). For this, 100 μL of two-fold serially diluted SNPs were prepared in a 96-well microtiter plate to obtain final concentration ranged from 2 to 1024 μg/mL and added 50 μL of TSB and BHI media for E. coli and L. monocytogenes, respectively. 100 μL of the bacterial culture broth (∼10⁶ CFU/mL) was added to each well and incubated the bacteria at 37 °C for 15 h, then added 10 µL of resazurin (0.4 mg/mL solution) and incubated for 3 h. The growth was assessed using a microtiter plate reader by measuring the optical density (OD) at 595 nm. The MIC was determined as the lowest concentration of the SNPs to inhibit the bacterial growth completely. The MBC was determined with the SNPs that gave MIC by subculturing on fresh TSB and BHI agar plates. The MIC and minimum fungicidal concentrations (MFC) of A. flavus were determined using the same procedure after incubation at 25 °C for 4 days.

3.1 Characterization of SNPs

In the process of preparing SNPs, the yellow of elemental sulfur turned reddish-orange as polysulfide was formed, and then turned milky white as SNPs were formed. The formation of SNPs was also confirmed using a UV–visible spectrophotometer. The UV–visible spectra of both types of SNPs (SNP_ES and SNP_STS) are shown in Fig. 1. Both SNPs exhibited similar absorption patterns showing two maximum absorption peaks around 240 and 280 nm without light absorption of visible light above 400 nm. The absorption of UV light was due to the n → σ* transition of nonbonding electrons, and this property can be used as preliminary detection of the SNPs (Xie et al., 2012).

Shankar and Rhim (2018) also found two absorbance peaks of SNPs in chitosan-based nanocomposite films. Xie et al. (2012) observed two light absorption peaks of nano-sulfur sol prepared by dissolving sublimed sulfur in a solvent of PEG-200. They found a sharp peak centered at 218 nm and a weak shoulder at 260 nm. Shankar et al. (2018) reported a clear light absorption peaks of SNPs around 280 nm, but they missed the other absorbance peak at a lower wavelength since they scanned from 250 nm. The light absorption peaks around 240 and 280 nm confirms the formation of sulfur nanoparticle (Paralikar & Rai, 2017; Xie et al., 2012).

The morphology of SNPs prepared with different sulfur compound sources was observed using TEM, and the TEM image and size distribution of SNPs are shown in Fig. 3. The shape and size of the SNPs depended on the sulfur source and manufacturing method. The SNPs prepared by the acidification of sodium thiosulfate (SNP_STS) have a spherical shape with an average particle size of 17 nm, while the SNPs made with elemental sulfur (SNP_ES) have a rod-shape with an average length and diameter of 87 and 22 nm, respectively.

Close

Fig. 3

TEM images and size distribution of sulfur nanoparticles.

Export to PPT

Also, the average particle size of both SNPs was determined by the dynamic light scattering (DLS), and the results are presented in Table 1. These values are significantly higher than the actual size determined by TEM. This is due to the fact that the DLS method determines the hydrodynamic radius of particles in a solution (Shankar et al., 2018). It has been reported that the shape of the SNPs varies depending on the manufacturing method. Spherical SNPs were obtained when the SNPs were prepared using sodium polysulfide in a microemulsion system (Guo et al., 2006), using sodium polysulfide in the presence of leaf extract (Paralikar & Rai, 2017), and using sodium polysulfide and ammonium polysulfide with polyethylene glycol-400 (PEG-400) as a surface stabilizing agents (Choudhury et al., 2011). On the contrary, Xie et al. (2012) obtained rod-shaped SNPs having a typical diameter of about 80 nm and an average aspect ratio of 6–8 by dissolving sublimed sulfur in a PEG-200 solvent. Choudhury et al. (2013b) also obtained rod-shaped SNPs having a diameter of around 50 nm using a water-in-oil microemulsion technique. In contrast, Massalimov et al. (2014) reported that the size of the SNPs obtained by acidification of polysulfide and sodium thiosulfate increased with increasing acid/precursor concentration ratio. The zeta potential values ​​of SNPs are presented in Table 1. The positive value of SNP is due to the chitosan used for capping, which is cationic at pH = 7. They also reported that the type of acid affects the size of synthesized SNPs, and the size of the SNPs prepared using calcium polysulfide was>100 nm. Chaudhuri et al. (2011) reported that the presence of a surfactant decreased the size of SNPs effectively.

However, the SNPs synthesized without chitosan capping were very unstable and precipitated immediately, while the SNPs capped with chitosan formed stable suspension of SNPs. The chitosan adsorbed on the surface of SNPs contributes to the creation of a stabilized suspension of SNPs by inducing a net positive charge on the surface of SNPs and inducing electrostatic repulsion between synthesized particles. Thus, chitosan coating results in smaller and more stable SNP nanoparticles. It has been reported that spherical SNPs were prepared when sodium polysulfide and other stabilizers were used (Choudhury et al., 2011; Guo et al., 2006; Paralikar and Rai, 2017), but in this study, rod-shaped SNPs were obtained using chitosan as a stabilizing agent. Though the mechanism for the formation of rod-shaped SNPs in the presence of chitosan is not clear, capping the synthesized SNPs using chitosan formed smaller particle sizes with narrower size distribution (Shankar et al., 2018).

With regard to the various molecular structures and allotropes of sulfur, the allotropic form of the synthesized SNPs should be determined. XRD analysis was performed to determine the allotropic form of the SNPs, and the XRD patterns of elemental sulfur and the synthesized SNPs are shown in Fig. 4. The position and intensity of the diffraction peaks of the SNPs are in good agreement with the XRD patterns of SNPs having an orthorhombic (α) S₈ structure of sulfur (Choudhury et al., 2013c; Xie et al., 2012). However, the allotropic forms of SNPs may vary depending on the sulfur source and preparation method of SNPs. Choudhury et al. (2013c) synthesized SNPs with different allotropic forms of sulfur. They reported that PEG-stabilized SNPs had a rhombic (α), whereas Span-80/Tween-80 coated SNPs showed a monoclinic (β) allotrope. SNPs from line broadening of the most intense diffraction peaks corroborates their TEM size distribution. The crystallite size of elemental sulfur, SNP_STS, and SNP_ES from line broadening of the most intense diffraction peak was determined using the Scherrer equation were 11.64, 12.88, and 17.33 nm, respectively. Soleimani et al. (2013) reported a simple preparation of SNPs with a crystallite size of 22 nm, while Deshpande et al. (2008) reported the preparation of SNPs with the crystallite size of 10 nm. Khairan et al. (2019) reported that the presence of garlic extract decreased the crystallite size of synthesized SNPs. The crystallite size of SNP_STS and SNP_ES determined by the Scherrer equation is in good agreement with the average diameter of those determined from the TEM results. The difference in the intensity of diffraction peaks at 2θ = 23.7° results from the difference in (2 2 0) crystal plane of sulfur (Falentin-Daudré et al., 2017).

Close

Fig. 4

XRD patterns and EDX spectra of elemental sulfur and sulfur and sulfur nanoparticles.

Export to PPT

The results of the elemental analysis of the SNPs and elemental sulfur determined using EDS are also shown in Fig. 4. Elemental sulfur (ES) showed the typical peak of sulfur at around 2.4 keV. In addition to the sulfur peak, both SNPs showed additional oxygen peaks due to the presence of chitosan on the surface of the prepared SNPs (Shankar et al., 2018). Also, an intense carbon peak in the EDS spectrum of SNP_ES implies that a higher amount of chitosan was bound at the SNP surface. The higher amount of chitosan on the SNP_ES surface was due to the presence of the chitosan molecule used as a stabilizer during SNP_ES formation. Chitosan not only prevents aggregation by stabilizing SNPs but also provides hydrophilicity to SNPs, helping to create a more stable SNP suspension.

The thermal stability of the SNP was evaluated using thermogravimetric analysis (TGA), and the results of the TGA and DTG thermograms are shown in Fig. 5. Previously, Shankar et al. (2018) found that the thermal stability of SNPs was not affected by the chitosan coating, so the TGA test of the uncoated SNP_STS was not performed. Both SNPs (SNP_STS and SNP_ES) showed similar thermal degradation behavior of thermal degradation, but the thermal degradation temperature was slightly different. The initial weight loss of SNP_STS occurred at 80–100 °C due to the evaporation of moisture (Shankar et al., 2018). However, SNP_ES did not show weight loss below 100 °C, possibly due to the complete dehydration of moisture during the drying process. Perhaps the tubular form of SNP_ES promoted the easier removal of water vapor during the drying process. The main thermal degradation of SNP_STS took place in the range of 210–302 °C with the maximum thermal degradation at 282 °C, and that of SNP_ES occurred in the range of 210–324 °C with the maximum thermal degradation at 306 °C. The main thermal degradation of the SNPs is presumably due to the partial degradation of sulfur (Shankar et al., 2018). The TGA results indicated that SNP_ES was slightly more thermostable than SNP_STS.

Close

Fig. 5

TGA and DTG thermograms of sulfur nanoparticles.

Export to PPT

3.2 Antimicrobial activity

The results of the antimicrobial activity of the SNPs against E. coli, L. monocytogenes, S. aureus, A. flavus, and C. albicans determined by the disk diffusion method and their MIC/MBC(MFC) values are shown in Table 2. The disk diffusion test results for E. coli and L. monocytogenes showed that both types of SNPs did not show growth inhibition for E. coli, but showed significant growth inhibition of L. monocytogenes (Fig. 6). Shankar et al. (2018) compared the antibacterial activity of chitosan-capped SNPs (SNP_STS) with the SNPs without chitosan capping, and they found that the SNP_STS showed more potent antibacterial activity than the SNPs without capping. So, the antibacterial activity of chitosan-capped SNPs was compared with the SNP_ES in this study. Between the types of SNPs, SNP_ES showed a more significant inhibition zone of growth of L. monocytogenes. Consistently, the MIC/MBC test results showed similar results of the antibacterial activity of the SNPs as the disk diffusion test. All the test results of the disk diffusion and MIC/MBC analysis indicated that both types of SNPs did not show a significant antibacterial effect against the Gram-negative (E. coli) bacteria.

Close

Fig. 6

Antibacterial activity of SNPs.

Export to PPT

On the contrary, they showed a substantial effect on the Gram-positive bacteria. They also indicated that SNP_ES exhibited more potent antibacterial activity against L. monocytogenes than SNP_STS. It has been reported so far that there is controversy over the antibacterial effect of SNPs against Gram-positive and Gram-negative bacteria.

Shankar et al. (2018) found that the SNPs produced by acidification of sodium thiosulfate and capping with chitosan exhibited strong antibacterial activity against E. coli (Gram-negative) and Staphylococcus aureus (Gram-positive) with MIC/MBC values of 16/64 and 16/32 μg/mL, respectively. Paralikar et al. (2017) also found that SNPs synthesized using leaf extracts from medicinal plants showed a similar degree of antibacterial activity against E. coli and S. aureus with a MIC of 200 µg/mL. Choudhury et al. (2013b) reported that rod-shaped monoclinic sulfur nanoparticles (β-SNPs) inhibited the growth of S. aureus at 576 µg/mL, while inhibited the growth of E. coli and K. pneumoniae at 2304 µg/mL. On the other hand, Libenson et al. (1953) reported that the micro-sized elemental sulfur did not show any antibacterial activity against E.ocli. Suleiman et al. (2015) reported that SNPs showed significant antibacterial activity against Gram-positive bacteria (S. aureus), but they did not show any activity against Gram-negative bacteria (E. coli and P. aeruginosa) even at high concentration of 800 µg/mL. They explained that Gram-negative bacteria are less vulnerable to SNPs due to the protection by the outer membrane of Gram-negative bacteria. The change in morphology and membrane integrity induced by the SNP_ES was observed using FE-SEM, as shown in Fig. 7. By treatment with SNP_ES, changes in the morphological structure of the microorganisms appeared in different ways. After treatment with SNP_ES, the cell wall of S. aureus was destroyed entirely, but the cell wall of E. coli was shrinking without destruction. This may explain why Gram-negative bacteria (E. coli) are more resistant to SNPs. The SNP treatment destroyed spores and mycelium of A. flavus and intact spherical cell buds of C. albicans.

Close

Fig. 7

FE-SEM images of microorganisms before and after treatment with SNP_ES.

Export to PPT

Though the exact mechanism of antibacterial action of SNPs has not been yet elucidated, several probable antimicrobial mechanisms of SNPs have been proposed, such as interactions with biological molecules on the surface of bacteria resulting in the membrane rupture and lysis of the cell membrane, the production of toxic H₂S gas that reacts with thiol groups of proteins and lipids, and interaction with DNA resulting in denaturation of DNA and cell death (Rai et al., 2016). Previously, Libenson et al. (1953) explained that the antimicrobial activity of elemental sulfur in micro-size was due to the binding of elemental sulfur to an enzyme containing a -SH functional group, but they did not observe the micro-sized elemental sulfur attached to the enzyme. However, Shankar et al. (2018) demonstrated that the stabilized SNPs capped with chitosan (SNP_STS) exhibited more potent antibacterial activity than the non-stabilized SNPs. In this study, a rod-shaped orthorhombic α-SNP (SNP_ES) showed stronger antibacterial activity than the spherical orthorhombic α-SNP one (SNP_STS). On the other hand, Choudhury et al. (2013c) reported that a spherical α-SNPs hadd stronger antimicrobial activity than tetrapod-like β-SNPs. They reported that the MIC values of α-SNPs were lower than those of β-SNPs. In the case of fungi, the interaction of SNPs with cellular lipids may reduce the total lipid contents to alter the cellular structure, leading to the destruction of the fungal cell (Choudhury et al., 2012; Roy Choudhury et al., 2012). Since the antimicrobial activity of SNPs varies depending not only on the properties of the SNPs such as size, shape, source of sulfur, and preparation method but also on the type of target microorganisms and solvent for dissolving SNPs, further research is needed to draw conclusions on this issue.