Combating pathogens with Cs_2.5H_0.5PW₁₂O₄₀ nanoparticles: a new proton-regulated antimicrobial agent

Abstract

The transfer of pathogens from contaminated surfaces to patients is one of the main causes of health care-associated infections (HCAIs). Cases of HCAIs due to multidrug-resistant organisms have been growing worldwide, whereas inorganic nano-antimicrobials are valuable today for the prevention and control of HCAIs. Here, we present a cesium salt of phosphotungstic heteropolyacid (Cs_2.5H_0.5PW₁₂O₄₀) as a promising nanomaterial for use in antimicrobial product technologies. This water-insoluble Keggin salt exhibits a broad biocide spectrum against Gram-positive and Gram-negative bacteria, yeasts, and filamentous fungi even under dark conditions. The Cs_2.5H_0.5PW₁₂O₄₀ nanoparticles (NPs) act as a proton-regulated antimicrobial whose activity is mediated on the release of hydronium ions (H₃O⁺), yielding an in situ acidic pH several units below those tolerable by most of the fungal and bacterial nosocomial pathogens.

---

Introduction

Nosocomial or health care-associated infections (HCAIs) are infections that patients acquire while receiving treatment at a health care facility that were not present at the time of admission.¹ HCAIs are associated with increased health care costs – an additional cost between US$ 7453 and 15 155 per infected patient – and an average of 8 days of extra-length of stay per patient in hospitals.² HCAIs not only delay patient recovery but also significantly contribute to patient mortality.¹ Nowadays, the risks of HCAIs have become greater due to the fact that the available drugs to combat pathogens are losing their efficacy caused by the emergence of antimicrobial resistance.³ The incidence of HCAIs caused by multidrug-resistant organisms is increasing worldwide and finding a feasible drug against non-treatable HCAIs has become a challenge.⁴ It is estimated that the cumulative premature deaths caused by resistant microorganisms will reach 300 million by 2050 at a financial cost of up to $100 trillion to the global economy.⁵ The World Health Organization (WHO) and several researchers foresee the emergence of resistant microorganisms as one of the biggest public health threats of the 21st century,^6,7 a century entitled as the Antibiotic-Resistance Era.⁸

It is well known that there are pathogens able to survive for months on surfaces at health care facilities.⁹ Due to this survival capability, a large number of HCAIs are eventually caused by the transference of pathogens from contaminated surfaces to patients.¹⁰ To address this issue, there are a huge number of studies looking for antimicrobial surfaces and devices containing nanosized inorganic antimicrobials.^11,12 However, the approaches based on nano-antimicrobials to prevent HCAIs present some drawbacks: (i) the action of some inorganic antimicrobials is mediated by the release of heavy metal ions or nanoparticles (NPs), which are known to be potentially hazardous to humans and non-target cells,¹³ (ii) nanomaterials with antimicrobial activity based on the generation of reactive oxygen species (ROS) exhibit reduced efficacy under dark conditions,¹⁴ and (iii) some nanomaterials exhibit antimicrobial activity against a limited spectrum of pathogens.¹⁵ Indeed, the last concern is one important limitation for antimicrobial surfaces due to the high diversity of pathogens that occur at health care facilities.

Some of the above issues about the current inorganic antimicrobials have been addressed with an innovative concept based on acidic surfaces idealized by Guggenbichler, Zollfrank, and co-workers.^16–18 The fundamental basis of this concept is the fact that several clinically relevant human pathogens are not able to grow in acidic environments. From a simplistic point of view, acidic surfaces work by lowering the interfacial pH between the functionalized surface and the microorganism, thus creating an inhospitable acidic environment for pathogens. This concept mimics the natural acidic nature of the human skin. A healthy human skin is an acidic environment,¹⁹ with pH ranging from 4 to 6,²⁰ which is believed to be an evolutionary mechanism of defense against invading microorganisms.²¹ It also introduces an extremely relevant characteristic of acidic surfaces, that is; their intrinsic broad-spectrum antimicrobial activity. There are several clinically relevant human pathogens that are not able to grow in environments with pH less than 3.5.¹⁸ To build up acidic surfaces, an inorganic compound which has a large number of acidic sites able to release hydronium ions (H₃O⁺) is necessary. Until now, molybdenum oxide-based compounds have been the unique reported materials suitable for this technology.^22,23

This work presents a cesium salt of phosphotungstic heteropolyacid (Cs_2.5H_0.5PW₁₂O₄₀) as a new nano-antimicrobial with its action regulated by H₃O⁺ release for the fabrication of antimicrobial acidic surfaces and devices. The Cs_2.5H_0.5PW₁₂O₄₀ NPs are a water-insoluble polyoxometalate (POM)²⁴ with a Keggin structure that contains maximized density of surface acid sites,²⁵ high surface area,²⁶ and nanosized particles.²⁷ Besides, its chemical structural features allow expressing high rates of H₃O⁺ generation and proton mobility,²⁸ being, as will be discussed here, a potential nanomaterial against HCAIs and overcoming the multidrug-resistant mechanisms.

Experimental

Synthesis of Cs_2.5H_0.5PW₁₂O₄₀ NPs

Cs_2.5H_0.5PW₁₂O₄₀ NPs were synthesized by precipitation according to the procedure of Ito et al.,²⁹ with some modifications. 12-Phosphotungstic acid hydrate (H₃PW₁₂O₄₀·xH₂O, reagent grade – Sigma) and cesium chloride (CsCl, 99.9% – Sigma) were used as precursors without further purification. First, H₃PW₁₂O₄₀·xH₂O was dissolved in 55 mL of water to give an aqueous solution of 0.025 mol L⁻¹, and, the resultant solution was cooled to 2 °C in an ice bath. In another flask, CsCl was dissolved in 63.1 mL of water to give an aqueous solution of 0.055 mol L⁻¹, which was added dropwise at a rate of 0.7 mL min⁻¹ to the first solution under stirring. The whole synthesis procedure lasted for 120 min, consisting of 90 min for complete dropwise addition of the cesium chloride solution and, further 30 min of final stirring at 2 °C. The white precipitates were separated by centrifugation, dried at 60 °C and then ground in an agate mortar.

Determination of the crystal structure

X-ray diffraction (XRD) patterns were obtained using a Siemens D5005 diffractometer in the Bragg–Brentano geometry with Cu Kα radiation (λ = 0.15418 nm) operating at a voltage of 40 kV and a current of 30 mA. X-ray diffractograms were collected at 2θ intervals between 5 and 60° with a step size of 0.03° and a scan time of 2 s per step. Raman spectra were recorded using a CRAIC 20/30 PV micro-spectrometer from 190 to 1900 cm⁻¹ at 25 °C, under excitation at 830 nm and 49.8 mW.

Chemical composition

Structural water molecules (·nH₂O) in Cs_2.5H_0.5PW₁₂O₄₀ were determined by thermogravimetric analysis (TG) using a TA Instruments SDTQ 600 thermal analyzer. Mass loss associated with the water molecules was measured after heating 15 mg of the as-synthesized powder from 25 to 300 °C at a heating rate of 10 °C min⁻¹ and a nitrogen gas flow rate of 100 mL min⁻¹. The real concentration of the Cs atoms in the Cs_2.5H_0.5PW₁₂O₄₀ Keggin structure was determined by standardless X-ray fluorescence using a Shimadzu EDX-720 spectrometer.

Concentration of surface acid sites

The density of surface acid sites of the Cs_2.5H_0.5PW₁₂O₄₀ NPs (μmol m⁻²) was estimated dividing the surface acidity (mmol g⁻¹) by the surface area (m² g⁻¹). The surface acidity was determined by temperature-programmed desorption (TPD) of NH₃ using a Micromeritics AutoChem II 2920 chemisorption analyzer. Initially, the samples were treated at 60 °C for 4 h under a stream of pure He. Subsequently, the temperature was cooled down to 50 °C and the samples were exposed to NH₃ (10 vol% NH₃ in He) for 30 min and immediately flushed with He (30 mL min⁻¹) at the same temperature for 1 h to remove physically adsorbed NH₃. Finally, the samples were heated up to 450 °C at a heating rate of 10 °C min⁻¹ under a He stream (30 mL min⁻¹). The specific surface area was determined by N₂ physisorption at −196 °C via the Brunauer–Emmett–Teller method using a Micromeritics Model VII Gemini surface area analyzer.

Tested strains

The antimicrobial activity of Cs_2.5H_0.5PW₁₂O₄₀ NPs was evaluated against ten clinically relevant pathogens: Gram-negative bacteria Escherichia coli ATCC 25922 and Pseudomonas aeruginosa ATCC 14502, Gram-positive bacteria Staphylococcus aureus ATCC 25923 and multidrug-resistant Enterococcus faecium VRE16,³⁰ filamentous fungus Aspergillus fumigatus akuB^KU80,³¹ and yeasts Cryptococcus neoformans ATCC 90112, Cryptococcus gattii ATCC 32269, Candida albicans ATCC 64548, Candida glabrata ATCC 90030, and Candida parapsilosis ATCC 22019. All the antimicrobial tests with the Cs_2.5H_0.5PW₁₂O₄₀ NPs were carried out under dark conditions.

Electron microscopy

Scanning electron microscopy (SEM) images of the Cs_2.5H_0.5PW₁₂O₄₀ NPs were acquired using an FEI Magellan 400 L microscope at an accelerating voltage of 25 kV. Transmission electron microscopy (TEM) images of the Cs_2.5H_0.5PW₁₂O₄₀ NPs were obtained using a FEI Tecnai G² F20 microscope operating at 200 kV. Experimental electron diffraction patterns were simulated using the JEMS software version 3.6907U2011.³² For the simulations, we used the crystallographic information file (CIF) of the parent Keggin salt, K₃PMo₁₂O₄₀ (ICSD No. 150445), with structural parameters adjusted for the Cs_2.5H_0.5PW₁₂O₄₀ NPs after XRD analysis.

Electron microscopy images of C. parapsilosis were acquired with and without exposition to Cs_2.5H_0.5PW₁₂O₄₀ NPs. 50 mL of Sabouraud broth was inoculated with 200 μL of a preculture of C. parapsilosis (4 × 10⁹ CFU mL⁻¹) for previous growth at 37 °C in a reciprocal shaker (180 rpm) for 9 h. Subsequently, a flask containing the C. parapsilosis suspension was treated with ½ MIC of the Cs_2.5H_0.5PW₁₂O₄₀ NPs. Another flask without Cs_2.5H_0.5PW₁₂O₄₀ NPs was used as the control. Further, they were kept overnight at 37 °C in a reciprocal shaker (180 rpm). A pellet of C. parapsilosis cell suspension was collected by centrifugation at 3200g for 5 min and fixed at 4 °C for 24 h in 0.1 mol L⁻¹ sodium phosphate buffer, pH 7.4, containing 2.5% glutaraldehyde and 2% paraformaldehyde. The fixed cells were stained with 1% OsO₄. The final step of dehydration of the samples was performed using a series of increasing neat ethanol contents (25, 50, 75, 95, and 3 × 100%).³³ Finally, the samples were dried under vacuum at 25 °C and observed using a FEI Magellan 400 L microscope at an acceleration voltage of 2 kV. Sample preparation for TEM images was performed as reported elsewhere,³⁴ however using Spurr's resin. TEM images of C. parapsilosis cells were acquired using a FEI Tecnai G² F20 microscope operating at 200 kV.

Minimum inhibitory/biocidal concentration

Minimum inhibitory concentration (MIC) of the Cs_2.5H_0.5PW₁₂O₄₀ NPs was determined by the broth microdilution method recommended by CLSI M38-A2,³⁵ M27-A2,³⁶ and M7-A6,³⁷ respectively for fungi, yeasts, and bacteria. The experiments were performed in 96-well plates using Sabouraud broth for the filamentous fungus and yeasts, and Mueller Hinton broth for bacteria. Each well was filled with 175 μL of serial broth dilutions containing the Cs_2.5H_0.5PW₁₂O₄₀ NPs and inoculated with 5 μL of a fresh culture of each organism (10⁵ CFU or conidia mL⁻¹). 20 μL of alamarBlue (Thermo Scientific) was added into the above wells as a colorimetric indicator of metabolic activity. The plates were incubated for 24 h at 35 °C for bacteria and at 37 °C for the filamentous fungus and yeasts. The MIC corresponds to the lowest concentration of the compound that prevents the microbial growth. After the MIC evaluation, 100 μL of culture from the wells, in which no visible microbial growth was detected, were inoculated on agar plates under the same conditions as those of the MIC test. From these readings, the minimum biocidal concentration (MBC) was assumed to be the lowest concentration of the Cs_2.5H_0.5PW₁₂O₄₀ NPs that kills 100% of the initial population of microorganisms.

Microbial growth curves for yeasts and bacteria

The inhibitory effect of the Cs⁺ and [PW₁₂O₄₀]³⁻ ions on the microbial growth was assessed by measuring optical density at 600 nm (OD_600nm) using a Biochrom Libra S50 spectrophotometer. CsCl and H₃PW₁₂O₄₀·xH₂O were used as the sources of Cs⁺ and [PW₁₂O₄₀]³⁻ ions, respectively. Initially, the flasks containing 50 mL of Sabouraud or Mueller Hinton broth were inoculated with 200 μL of a fresh suspension of each microorganism at 4 × 10⁹ CFU mL⁻¹. The growth inhibition by the Cs_2.5H_0.5PW₁₂O₄₀ NPs was done at the MIC value of each strain. The concentrations of Cs⁺ and [PW₁₂O₄₀]³⁻ were evaluated based on the 6.4% water solubility reported for Cs_2.5H_0.5PW₁₂O₄₀.²⁷ Then, the respective concentration of Cs⁺ and [PW₁₂O₄₀]³⁻ was calculated considering the stoichiometric amount dissociated from the MIC of each strain. Experiments were performed at 35 °C for bacteria and 37 °C for yeasts in an orbital shaker (200 rpm). For OD₆₀₀ readings, aliquots of the suspensions were taken hourly up to 14 h. The turbidity generated by the broth, Cs⁺, [PW₁₂O₄₀]³⁻, or Cs_2.5H_0.5PW₁₂O₄₀ was considered by taking blank readings. The tests were conducted in triplicate, and the average values are reported.

Phenotypic characterization of A. fumigatus

A. fumigatus conidia were aseptically collected from three day cultures in solid complete medium [YG; glucose 2% (w/w), 0.5% yeast extract (w/w), 1X trace elements, pH 6.5]. The composition of trace elements was described previously.³⁸ Conidia were dispersed in autoclaved Milli-Q water and the concentration of the suspension was adjusted to 10⁵ conidia mL⁻¹ employing a Neubauer chamber. Photographs of radial growth were recorded after the inoculation of 5 μL of the conidia suspension in the center of Petri dishes containing solid YG for 72 h at 37 °C. Following the image acquisition, the same plates were used for conidia count. Conidia were counted after dispersion of agar fragments (0.8 cm²), taken from the radial zone, in an aqueous solution with 0.01% Tween 20. The reported values are the average of triplicate tests. The morphology of A. fumigatus was observed by optical microscopy using an Olympus Bx50 microscope. Prior to imaging, the fungi were harvested from the radial growth plates and stained with lactophenol blue solution (Sigma).

Concentration of H₃O⁺

The concentration of H₃O⁺ ions was assessed after dispersion of the Cs_2.5H_0.5PW₁₂O₄₀ NPs in distilled water. In the suspension, the powder was ultrasonically dispersed and the pH was measured using a PG 1800 (Gehaka) digital pH meter. The concentration of the hydronium ions was estimated using the following relation: [H₃O⁺] = 10^−pH.

Results

Nanostructured Cs_2.5H_0.5PW₁₂O₄₀

According to the XRD pattern shown in Fig. 1a, precipitation is a synthetic approach suitable for producing single-phase Cs_2.5H_0.5PW₁₂O₄₀ nanopowders. The cesium salt precipitates as cubic crystals with Pn3̄m symmetry and a measured lattice parameter of 1.3641 nm. It is known that the formation of three-dimensional Cs_2.5H_0.5PW₁₂O₄₀ crystals occurs when the primary§ Keggin heteropolyanion [PW₁₂O₄₀]³⁻ has part of its structural proton sites substituted by the counter-cation Cs⁺.^25,39 The Raman spectrum (Fig. 1b) confirms the formation of Cs_2.5H_0.5PW₁₂O₄₀ with [PW₁₂O₄₀]³⁻ as its fundamental building blocks. The presence of intense bands at 1008 and 996 cm⁻¹ matches with those of the Raman vibrational frequencies of the W–O_d bond predicted at 1011 and 996 cm⁻¹ for [PW₁₂O₄₀]³⁻.⁴⁰

Fig. 1 (a) XRD pattern and (b) Raman spectrum of Cs_2.5H_0.5PW₁₂O₄₀ NPs. The photograph of the synthesized nanopowder is the inset in (a). The vertical red lines show positions of Bragg peaks (powder diffraction file – PDF) for the crystal structure of the cesium salt of phosphotungstic heteropolyacid.

The powder characteristics of the Cs_2.5H_0.5PW₁₂O₄₀ NPs are summarized in Table 1. The cesium content of 2.4 mol indicates partial replacement of protons close to the expected quantity of 2.5 mol. This appropriate replacement of the protons by the large-sized Cs⁺ ensures the formation of a water-insoluble Keggin salt.⁴⁰ Its insolubility in water is a key feature for the prospect of this compound into the group of inorganic antimicrobials of industrial interest. Moreover, the secondary structure occurs associated with eight structural water molecules, yielding the real chemical formula Cs_2.5H_0.5PW₁₂O₄₀·8H₂O, although, for clarity, water molecules will be omitted in this paper. The morphological investigation by electron microscopy (Fig. 2) shows that the Cs_2.5H_0.5PW₁₂O₄₀ nanopowder contains primary nanoparticles with an average size of 10 nm. In addition, there is a self-arrangement of the nanosized primary particles to produce tertiary structures with a spherical morphology of 100 nm (Fig. 2a and b). The tendency of Cs_xH_1−xPW₁₂O₄₀ compounds to form tertiary Keggin structures with a spherical morphology has also been reported by other research groups.^25,41 Furthermore, the presence of microporosity among the primary particles in this tertiary spherical morphology is responsible for the relatively high surface area of 116 m² g⁻¹ measured for the Cs_2.5H_0.5PW₁₂O₄₀ nanopowder.⁴²

Table 1 Morphological and chemical characteristics of the Cs_2.5H_0.5PW₁₂O₄₀ NPs

Cesium salt of phosphotungstic acid
Tertiary morphology	Spherical
Size of tertiary structure	100 nm
Primary particle size	10 nm
Surface area	116 m^2 g^−1
Atomic content of Cs^+	2.4
·nH2O (mol)	8

---

Fig. 2 (a) SEM and (b) TEM images of Cs_2.5H_0.5PW₁₂O₄₀ NPs. Scale bars correspond to 100 nm. (c) The electron diffraction patterns from a polycrystalline region and a single particle oriented along the [111] zone axis reveals the cubic symmetry of the primary particles.

A broad spectrum of action

The Cs_2.5H_0.5PW₁₂O₄₀ NPs behave as a broad-spectrum antimicrobial agent according to the MIC values shown in Table 2. The table demonstrates the concentration where the growth is completely inhibited in MIC analysis. Interestingly, for most of the microorganisms tested, at this concentration the Cs_2.5H_0.5PW₁₂O₄₀ NPs also behave as a biocide, therefore leading to complete loss of viability of all the tested strains, except for the bacteria E. coli and P. aeruginosa. This result is evidence for the potential use of the Cs_2.5H_0.5PW₁₂O₄₀ NPs as a biocide against pathogens. Noteworthily, the Gram-negative bacteria also lost viability in the presence of the compound; however, for the two strains, the MBC was found to be higher than the MIC (3.00 mg mL⁻¹). The low susceptibility of Gram-negative bacteria to Keggin salts was also observed previously.⁴³ Comparing the antibacterial activity of Na₇PMo₁₁CuO₄₀ against S. aureus and E. coli, the authors found the MBC to be two times higher for E. coli. Moreover, the resistance of Gram-negative bacteria is also common for other nano-antimicrobials, such as ZnO NPs.⁴⁴ Importantly, the MBC of the Cs_2.5H_0.5PW₁₂O₄₀ NPs described herein is significantly lower than those reported for similar Keggin salts. For instance, Na₃PW₁₂O₄₀ exhibits a MBC of 11.03 mg mL⁻¹ against S. aureus, P. aeruginosa, and E. coli⁴⁵ while neutral silver-based POMs exhibited a MBC higher than 20.00 mg mL⁻¹ against S. aureus and E. coli.⁴⁶ POMs can behave as photocatalysts under UV irradiation and produce reactive oxygen species (ROS),⁴⁷ therefore, it is worth mentioning that our MIC/MBC experiments were performed under dark conditions, indicating that the Cs_2.5H_0.5PW₁₂O₄₀ NPs are an active nano-antimicrobial even in the dark.

Table 2 MIC and MBC values for the Cs_2.5H_0.5PW₁₂O₄₀ NPs against all tested strains

Microorganism	MIC (mg mL^−1)	MBC (mg mL^−1)
Escherichia coli	1.75	3.00
Pseudomonas aeruginosa	2.00	3.00
Staphylococcus aureus	1.50	1.50
Enterococcus faecium	1.25	1.25
Aspergillus fumigatus	3.00	3.00
Cryptococcus neoformans	3.00	3.00
Cryptococcus gattii	3.00	3.00
Candida albicans	3.00	3.00
Candida glabrata	3.00	3.00
Candida parapsilosis	2.50	2.50

---

Another interesting finding in Table 2 is the small MIC and MBC value against the multidrug-resistant E. faecium. In this case, the strain with documented resistance to most of the current antibiotics depicted the highest sensitivity to the Cs_2.5H_0.5PW₁₂O₄₀ NPs. Other authors reported that the resistance to drugs does not attenuate the susceptibility of bacterial strains to Keggin-structural POMs.^48,49 For instance, methicillin-resistant S. aureus (MRSA), multidrug resistant P. aeruginosa, and extended spectrum β-lactamase producing Klebsiella pneumonia exhibit a lower MIC value than the reference bacterial strains when exposed to Na₃PW₁₂O₄₀.⁴⁵

To gain further insight into the extension of the antimicrobial activity of the Cs_2.5H_0.5PW₁₂O₄₀ NPs, in the analysis of Table 2, we also included yeasts and a filamentous fungus with a notorious ability to infect immunocompromised patients with HCAIs.⁵⁰ All the yeasts and filamentous fungus exhibited sensitivity to the Cs_2.5H_0.5PW₁₂O₄₀ NPs, with the MIC values equal to the MBC. For comparison, the MBC value of 3.00 mg mL⁻¹ for the Cs_2.5H_0.5PW₁₂O₄₀ NPs in the fungal species is strikingly smaller than the values reported for the silver-based POMs, i.e. > 20.00 mg mL⁻¹.⁴⁶

A. fumigatus produces massive amounts of conidia of small size (2–3 μm) which can be highly dispersive in air. Conidiation is therefore one of the main virulence determinants of this pathogen. It is thought that, under normal conditions, every individual inhales about 100 conidia of this fungus per day due to its ubiquitous distribution worldwide coupled with its efficient conidiation process.⁵¹ Therefore, we further investigated whether the Cs_2.5H_0.5PW₁₂O₄₀ NPs could impair the ability of A. fumigatus to produce conidia. The phenotypic investigation revealed that the Cs_2.5H_0.5PW₁₂O₄₀ NPs induce the reduction of conidiation, which is noticed by the absence of the greenish color of the A. fumigatus colony and by the quantification of the number of conidia (Fig. 3a). More precisely, we found a reduction in conidiation on the order of 99% for A. fumigatus exposed to the Cs_2.5H_0.5PW₁₂O₄₀ NPs, which can be seen by comparing the optical microscopy images shown in Fig. 3b. Considering that invasive pulmonary aspergillosis is acquired after accidental inhalation of conidia by immunocompromised patients,⁵² this reduction of fungus conidiation by the Cs_2.5H_0.5PW₁₂O₄₀ NPs further emphasizes the relevant role of this compound in hospital environments to mitigate the dispersion of these infectious propagules. Altogether, these results reinforce the potential use of Cs_2.5H_0.5PW₁₂O₄₀ NPs as a biocide nano-antimicrobial against fungal and bacterial pathogens of high incidence in nosocomial infections worldwide.

Fig. 3 Morphological alterations in A. fumigatus induced by the Cs_2.5H_0.5PW₁₂O₄₀ NPs. (a) Photographs of radial growth with quantification of the level of conidiation. (b) Optical images of A. fumigatus without and with exposition to Cs_2.5H_0.5PW₁₂O₄₀ NPs. Scale bars correspond to 200 μm.

The mechanism of action of Cs_2.5H_0.5PW₁₂O₄₀ NPs is not regulated by the [PW₁₂O₄₀]³⁻ and Cs⁺ ions

Metal ion release is one of the dominant antimicrobial mechanisms of metal oxide nanoparticles,⁵³ therefore, we performed microbial growth assays to clarify the correlation between the [PW₁₂O₄₀]³⁻ and Cs⁺ ions and the antimicrobial phenomenon of the Cs_2.5H_0.5PW₁₂O₄₀ NPs. The growth curves of some selected pathogens are shown in Fig. 4. For all the investigated strains, the presence of [PW₁₂O₄₀]³⁻ in the liquid medium did not affect the growth curves, exhibiting the same growth profile as the controls. In the broth containing the Cs_2.5H_0.5PW₁₂O₄₀ NPs, S. aureus did not grow and the growth of P. aeruginosa corresponded to 67%, that of C. neoformans corresponded to 46%, and that of C. albicans corresponded to 51% surviving cells in comparison to the growth in the presence of [PW₁₂O₄₀]³⁻. Complete growth inhibition of P. aeruginosa, C. neoformans, and C. albicans in the presence of Cs_2.5H_0.5PW₁₂O₄₀ NPs occurs only at concentrations three times that of the MIC for each strain. Note that these concentrations differ from the MIC results due to the different inoculation and growth conditions used for the microbial growth curves. Therefore, there was a clear inhibition of the growth of microorganisms in the presence of the Cs_2.5H_0.5PW₁₂O₄₀ NPs. This suggests that no connection exists between the reduced growth rates of the strains in the presence of the Cs_2.5H_0.5PW₁₂O₄₀ NPs and the presence of the Keggin heteropolyanion in solution. This result is in agreement with the findings of Lu et al.⁵⁴ and Wang et al.⁵⁵ Both the authors observed that the antibacterial activity of silver-based POMs is not mediated by the release of [PW₁₂O₄₀]³⁻. Furthermore, according to Hu et al.,⁵⁶ even a concentration of [PW₁₂O₄₀]³⁻ as high as 1.06 mg mL⁻¹ does not inhibit the growth of S. aureus and E. coli. Another front of studies, however, have suggested that the release of counter-cations is a regulatory pathway for the antimicrobial phenomenon in Keggin salts. These studies have associated the origin of the antimicrobial behaviour with the release of NH₄⁺ in (NH₄)₃PMo₁₂O₄₀,⁵⁷ Ag⁺ in silver-based POMs,^46,55 and, recently, the tetraalkylammonium cations in exchanged silicotungstates.⁵⁸ However, as for the [PW₁₂O₄₀]³⁻, our experimental growth curves did not indicate any appreciable inhibition effect by the counter-cation Cs⁺. Hence, the antimicrobial activity of the Cs_2.5H_0.5PW₁₂O₄₀ NPs does not seem to be a consequence of the ion release, which is expected due to the low tendency of this Keggin salt to release heteropoly species into water.²⁷

Fig. 4 Microbial growth curves for: (a) P. aeruginosa, (b) S. aureus, (c) C. neoformans, and (d) C. albicans. Symbols represent the nutrient broth containing: (●) [PW₁₂O₄₀]³⁻, (▲) Cs⁺, () Cs_2.5H_0.5PW₁₂O₄₀ NPs, and (■) control without compounds. The symbol (◓) indicates experiments carried out at concentrations of Cs_2.5H_0.5PW₁₂O₄₀ NPs equal to three times the MIC value for each strain. For simplicity, we consider the [PW₁₂O₄₀]³⁻ as the stable heteropolyanion in the nutrient broth solution, although ³¹P and ¹⁸³W NMR studies indicate that the heteropolyanion can exist as the lacunary [PW₁₁O₃₉]⁷⁻ in the pH of the broth.⁷⁹

Internalization of nanoparticles is not necessary for the action of the Cs_2.5H_0.5PW₁₂O₄₀ NPs

Another antimicrobial mechanism often associated with nanosized materials is the physical interaction between the nanoparticles and the cells.⁵³ In particular in fungi and bacteria, this interaction is mediated by the cell wall which is a dynamic rigid, but permeable tridimensional structure composed mainly of carbohydrates that protects the cell against osmotic lysis and also provides cell shape.⁵⁹ In order to determine whether the Cs_2.5H_0.5PW₁₂O₄₀ NPs act by cell wall damage or internalization of nanoparticles, we performed a careful electron microscopy investigation of C. parapsilosis cells. This yeast was chosen because it was the most sensitive fungal organism to our compound (Table 2). In addition, C. parapsilosis is the main non-albicans Candida species isolated from patients in Latin America.⁶⁰ In a scenario of cell death induced by nanoparticles, visibly damaged cells with loss of the cell wall integrity are expected.⁶¹ We found that this scenario did not occur in C. parapsilosis cells exposed to the Cs_2.5H_0.5PW₁₂O₄₀ NPs (Fig. 5). Conversely, C. parapsilosis exposed to the nanoparticles did not show any morphological alteration in comparison to the control. Only an attachment of the tertiary structures of the Keggin salt to the cell wall was detected (Fig. 5e). Furthermore, we did not find internalization of the spherical tertiary structures or the 10 nm-sized primary particles. The TEM images in Fig. 5c–e demonstrate that the nanoparticles are not able to cross or weaken the cell wall structure of C. parapsilosis. This means that the Cs_2.5H_0.5PW₁₂O₄₀ NPs are unable to reach the cytosol, hence it cannot directly induce any internal physical damage to the cell. We propose that this result can be a common observation for all the fungal species that eventually interact with Cs_2.5H_0.5PW₁₂O₄₀ NPs. This hypothesis is further sustained by the susceptibility assays using A. fumigatus mutants for regulatory genes of the cell wall integrity (CWI) pathway (Fig. S1, ESI†). A common feature of these mutants is the defective maintenance of cell wall integrity, which ultimately makes the cells more susceptible to lysis or cell wall damaging agents.^34,62 We observed similar levels of tolerance of the A. fumigatus wild-type and the CWI mutants towards the Cs_2.5H_0.5PW₁₂O₄₀ NPs supporting the previous finding in C. parapsilosis, i.e., the action of the Cs_2.5H_0.5PW₁₂O₄₀ NPs do not involve disruption of the cell wall integrity.

Fig. 5 Electron microscopy images of C. parapsilosis cells. Secondary electron SEM image of the cells in (a) control and (b) exposed to Cs_2.5H_0.5PW₁₂O₄₀ NPs. Scale bars correspond to 2 μm. Bright-field TEM image of the cells in (c) control and (d) exposed to Cs_2.5H_0.5PW₁₂O₄₀ NPs, with magnification in the dark-field mode shown in (e). Scale bars correspond to 500 nm. Symbols: (●) Cs_2.5H_0.5PW₁₂O₄₀ NPs and (■) cell wall.

An antimicrobial mechanism regulated by H₃O⁺

The high amount of acid sites in the Cs_2.5H_0.5PW₁₂O₄₀ NPs^25,41 has the potential to be the main antimicrobial mechanism of this compound. To establish whether acid sites are important for the antimicrobial response of the Cs_2.5H_0.5PW₁₂O₄₀ NPs, the TPD-NH₃ technique was employed to probe the relation between the surface acid sites and the antimicrobial response. First, we reduced the original concentration of the surface acid sites in the as-synthesized nanopowder through a heat treatment at 300 °C. This heat treatment decreases the number of acid sites by the loss of structural water molecules⁴⁰ (Fig. S2, ESI†), and also restrains the mobility of the structural protons.⁶³ The TPD-NH₃ results (Fig. 6a and b) show that the total acid site concentration in the as-synthesized Cs_2.5H_0.5PW₁₂O₄₀ NPs is almost two times higher than that in the heat-treated powder. Moreover, a small concentration of strong acid sites (2.6 × 10⁻³ mmol g⁻¹) disappear after the heat treatment. Both the results indicate that the as-synthesized Cs_2.5H_0.5PW₁₂O₄₀ NPs are more acidic than their heat-treated form. As a result, there is a considerable modification in the antimicrobial activity. As shown in Fig. 6c, the loss of surface acid sites is accompanied by a pronounced decrease in the antibacterial efficacy of the Cs_2.5H_0.5PW₁₂O₄₀ NPs. For all the bacterial strains, the MIC values increase between 0.75 and 1.00 mg mL⁻¹ for the sample with a lowered concentration of surface acid sites. We can therefore assume that surface acidity is an important regulator for the antimicrobial effect in this Keggin salt.

Fig. 6 TPD-NH₃ profiles of Cs_2.5H_0.5PW₁₂O₄₀ NPs, (a) as-synthesized and (b) heat treated at 300 °C for 2 h. (c) Dependence of the MIC values on the surface acid site concentration in Cs_2.5H_0.5PW₁₂O₄₀ NPs, where 2.3 and 1.6 μmol m⁻² represent, respectively, the as-synthesized nanopowder and the powder heat treated at 300 °C for 2 h. (d) Concentration of H₃O⁺ in aqueous suspensions containing several additions of the as-synthesized Cs_2.5H_0.5PW₁₂O₄₀ NPs. For this measurement, distilled water was selected as a standard medium, instead of the nutrient broth, because it is a more appropriate medium for accurately estimating the number of H₃O⁺ released from Cs_2.5H_0.5PW₁₂O₄₀ NPs without interference of other organic compounds. Due to the buffering action, the acidification observed in aqueous suspension cannot be totally extrapolated to nutrient broths.

The apparent complex interrelationship between the surface acid sites and the antimicrobial response can be explained via the contribution of these acid sites to the generation and release of H₃O⁺ ions into the media. As shown in Fig. 6d, we found an exponential growth of the H₃O⁺ species in water as the concentration of the Cs_2.5H_0.5PW₁₂O₄₀ NPs increases. Obviously, a direct effect of the high protonation rate is the acidification of the environment in the presence of Cs_2.5H_0.5PW₁₂O₄₀ NPs. For instance, by simulating the MBC values of the Cs_2.5H_0.5PW₁₂O₄₀ NPs in aqueous suspensions, we found pH 3.8 for Gram-negative bacteria, 4.3–4.4 for Gram-positive bacteria, and 3.8–4.0 for yeasts. Surprisingly, these experimental values are close to the expected pH range for acid-induced death in the tested strains.¹⁶ In other words, the H₃O⁺ ions released by the Cs_2.5H_0.5PW₁₂O₄₀ NPs can reduce the local pH to levels theoretically not supported for the tested strains and presumably for the majority of nosocomial pathogens.

Discussion

Our systematic description of the antimicrobial activity exerted by the Cs_2.5H_0.5PW₁₂O₄₀ NPs demonstrates that this compound offers some advantages over the current nano-antimicrobials. An exciting feature comes from its biocide effect with a broad spectrum of action against Gram-negative and Gram-positive bacteria, yeasts, and filamentous fungus. Although being a highly desired characteristic, a broad spectrum of action not always occurs in oxide-based nanomaterials. For instance, ZnO NPs are ineffective against Proteus vulgaris,¹⁵ in addition to the low efficacy of CuO NPs against Salmonella typhimurium, Aeromonas liquefaciens, P. aeruginosa, and Proteus spp.^64–66 Another important feature of the Cs_2.5H_0.5PW₁₂O₄₀ NPs described here is the manifestation of biocide effect even with the susceptibility assays being strictly carried out under dark conditions. This suggests that the Cs_2.5H_0.5PW₁₂O₄₀ NPs do not require light or UV activation. Comparatively, the TiO₂ NP-based antimicrobials regulated by ROS have been reported to be significantly less active under dark conditions.¹⁴ Also, the fact that the antimicrobial action of Cs_2.5H_0.5PW₁₂O₄₀ NPs has no dependence on the [PW₁₂O₄₀]³⁻, Cs⁺, and internalization of nanoparticles, is an important aspect that mitigates the toxicity of nano-antimicrobials towards mammalian cells and the environment.^13,67 Nowadays, there are still some safety concerns regarding nano-antimicrobials with a mode of action mediated by the release of heavy metal ions or nanoparticles, such as, Ag NPs.⁶⁸

One significant finding of our research is the relationship between the antimicrobial action of the Cs_2.5H_0.5PW₁₂O₄₀ NPs and the release of H₃O⁺. Protonation in Keggin salts is a well-known intrinsic property induced by structural acid sites.^69,70 Even a substantial replacement of the proton sites with Cs⁺ in the phosphotungstic heteropolyacid structure did not compromise the protonation ability of the resultant Cs_2.5H_0.5PW₁₂O₄₀ salt. On the contrary, the antimicrobial behaviour can take advantage of the marked increase in proton mobility when counter-cations are added to POMs.⁶³ Furthermore, Cs_xH_3−xPW₁₂O₄₀-based antimicrobials can be safely designed upon modulation of the extension of H₃O⁺ release. The produced acidic environment (pH) can be adequately regulated to a desired level by controlling the replacement of the structural protons by the counter-cation Cs⁺,⁴¹ or in a more simple approach, by controlling the amount of the Keggin salt used (Fig. 6d).

Although not yet entirely described, the current mechanistic insight into the proton release by Keggin salts reveals that external oxygen atoms of the heteropolyanion are the active centers for H₃O⁺ generation^28,63,71 (see eqn (1)). Consequently, the acidic pH produced by the Cs_2.5H_0.5PW₁₂O₄₀ NPs has its origin in the interaction of the nanoparticle surface with water, and the subsequent release of H₃O⁺ species to the environment.^41,71 This release of protons is also facilitated by the small electrostatic attraction between the H₃O⁺ and the big-sized heteropolyanion.²⁴ Therefore, our observation of the superior antimicrobial activity of the Cs_2.5H_0.5PW₁₂O₄₀ NPs with a higher amount of surface acid sites supports a mode of action mediated by H₃O⁺. Acid sites are fundamental structures for proton formation and motion.^25,63 Thus, the physical meaning of the MBC values found in this work is an amount of Cs_2.5H_0.5PW₁₂O₄₀ NPs in which the release of H₃O⁺ reaches an extent sufficient to decrease the local pH to levels not supported by the tested strains. Although the MBC values of the Cs_2.5H_0.5PW₁₂O₄₀ NPs are considerably higher than those reported for metal nanoparticles such as Ag (0.10 mg mL⁻¹),⁶⁴ the values shown in Table 2 are comparable to those found for metal oxide nanoparticles such as ZnO (2.50–5.00 mg mL⁻¹)⁶⁴ and TiO₂ (>2.50 mg mL⁻¹).⁷²

(–W–O–W–) + 3 H₂O ⇆ 2 (–W–O)⁻ + 2 H₃O⁺ (1)

Since the majority of clinically relevant pathogens are neutralophiles, that is, pathogens with optimum growth responses only at near-neutral pH,^73,74 the antimicrobial mechanism of Cs_2.5H_0.5PW₁₂O₄₀ NPs regulated by the release of H₃O⁺ ions appears to be of great importance to combat HCAIs. As a rule, neutralophiles require a circumneutral cytoplasmic pH; however, they do not possess effective pH homeostasis mechanisms in extremely acidic environments.⁷⁵ Targeting this specific weakness, the H₃O⁺ ions released by the Cs_2.5H_0.5PW₁₂O₄₀ NPs can lead to an acid-induced death by lowering the cytoplasmic pH.⁷⁶ Considering Fig. 5e, the Cs_2.5H_0.5PW₁₂O₄₀ NPs behave as a proton carrier anchored on the cell surface, releasing protons into the cytoplasm and other cell structures affecting the functionality of cell molecules. Thus, the antimicrobial susceptibility of the strains towards the Cs_2.5H_0.5PW₁₂O₄₀ NPs in the order: Gram-positive bacteria > Gram-negative bacteria > yeasts/filamentous fungus, seems to be associated with the effectiveness of the cell walls/cell membranes in pH homeostasis in each of these different species.⁷³ For example, the superior resistance of yeasts to the action of Cs_2.5H_0.5PW₁₂O₄₀ NPs can be explained by the proton pump out mechanism.⁷⁷ Yeasts can maintain the cytosolic pH neutral even when exposed to acidic environments due to a combined response of Pma1p and V-ATPase. Pma1p is a single 100 kDa polypeptide localized in the plasma membrane that pumps protons out of the cell. V-ATPase is responsible for pH homeostasis in organelles.⁷⁸ However, the improbable adaptation of neutralophiles to an environment rich in H₃O⁺ makes the Cs_2.5H_0.5PW₁₂O₄₀ NPs a low risk nano-antimicrobial agent of eliciting resistance. Even in the absence of aqueous medium, condensation of moisture present in the environment should be sufficient to create an aqueous interface for H₃O⁺ diffusion from a surface or device containing Cs_2.5H_0.5PW₁₂O₄₀ NPs to the target pathogen.²² Furthermore, not only limited to the Cs_2.5H_0.5PW₁₂O₄₀ NPs, an antimicrobial mechanism regulated by H₃O⁺ can be expected for other members of the vast family of water-insoluble POM compounds.

Conclusion

In summary, we demonstrated that Cs_2.5H_0.5PW₁₂O₄₀ NPs hold potential as a proton-regulated inorganic antimicrobial. The key phenomenon behind their antimicrobial performance is the formation of an inhospitable acid environment through the generation and release of H₃O⁺. One important characteristic of the antimicrobial mechanism mediated by H₃O⁺ is the biocide effect against neutralophiles. It supports the prospect of this nano-antimicrobial to combat several clinically relevant human pathogens, and, with potential for utilization against multidrug-resistant organisms. Overall, the Cs_2.5H_0.5PW₁₂O₄₀ NPs expand the possibilities for surface or medical device functionalization as an attempt to reduce the incidence of HCAIs.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the São Paulo Research Foundation (FAPESP Grant no. 2016/16098-8 to R. H. P. and 2015/17541-0 to IM). R. H. P. thanks the postdoctoral program of the São Carlos Institute of Chemistry (IQSC) at the University of São Paulo (USP).

Notes and references

1. B. J. Hensley and J. R. T. Monson, Surgery, 2015, 33, 528–533.
2. H. Arefian, S. Hagel, S. Heublein, F. Rissner, A. Scherag, F. M. Brunkhorst, R. J. Baldessarini and M. Hartmann, Am. J. Infect. Control, 2016, 44, 160–166.
3. R. Sugden, R. Kelly and S. Davies, Nat. Microbiol., 2016, 1, 16187.
4. J. I. Barrasa-Villar, C. Aibar-Remón, P. Prieto-Andrés, R. Mareca-Doñate and J. Moliner-Lahoz, Clin. Infect. Dis., 2017, 65, 644–652.
5. J. O’Neill, Antimicrobial resistance: tackling a crisis for the health and wealth of nations, 2014.
6. S. Garneau-Tsodikova and G. D. Wright, Med. Chem. Commun., 2016, 7, 10.
7. World Health Organization, Worldwide country situation analysis: response to antimicrobial resistance, 2015.
8. E. D. Brown and G. D. Wright, Nature, 2016, 529, 336–343.
9. A. C. Abreu, R. R. Tavares, A. Borges, F. Mergulhao and M. Simoes, J. Antimicrob. Chemother., 2013, 68, 2718–2732.
10. B. Cohen, C. C. Cohen, B. Løyland and E. Larson, Clin. Epidemiol., 2017, 9, 297–310.
11. M. C. Sportelli, R. A. Picca and N. Cioffi, TrAC, Trends Anal. Chem., 2016, 84, 131–138.
12. L. Wang, C. Hu and L. Shao, Int. J. Nanomed., 2017, 12, 1227–1249.
13. O. Bondarenko, K. Juganson, A. Ivask, K. Kasemets, M. Mortimer and A. Kahru, Arch. Toxicol., 2013, 87, 1181–1200.
14. T. P. Dasari, K. Pathakoti and H. M. Hwang, J. Environ. Sci., 2013, 25, 882–888.
15. M. Ramani, S. Ponnusamy, C. Muthamizhchelvan, J. Cullen, S. Krishnamurthy and E. Marsili, Colloids Surf., B, 2013, 105, 24–30.
16. M. Lackner, S. Maninger and J.-P. Guggenbichler, Nachr. Chem., 2013, 61, 112–115.
17. S. Shafaei, M. Lackner, R. Voloshchuk, I. Voloshchuk, J. Guggenbichler and C. Zollfrank, Recent Pat. Mater. Sci., 2014, 7, 26–36.
18. C. Zollfrank, K. Gutbrod, P. Wechsler and J. P. Guggenbichler, Mater. Sci. Eng., C, 2012, 32, 47–54.
19. H. Schade and A. Marchionini, Klin. Wochenschr., 1928, 7, 12–14.
20. A. Zlotogorski, Arch. Dermatol. Res., 1987, 279, 398–401.
21. S. Ali and G. Yosipovitch, Acta Derm.-Venereol., 2013, 93, 261–267.
22. S. Shafaei, D. Van Opdenbosch, T. Fey, M. Koch, T. Kraus, J. P. Guggenbichler and C. Zollfrank, Mater. Sci. Eng., C, 2016, 58, 1064–1070.
23. S. Shafaei, M. Lackner, M. Meier, J. Plank, J. P. Guggenbichler and C. Zollfrank, Surf. Innovations, 2013, 1, 202–208.
24. T. Okuhara, Chem. Rev., 2002, 102, 3641–3666.
25. M. Misono, Chem. Commun., 2001, 1141–1152.
26. A. Corma, A. Martínez and C. Martínez, J. Catal., 1996, 164, 422–432.
27. T. Nakato, M. Kimura, S. Nakata and T. Okuhara, Langmuir, 1998, 14, 319–325.
28. I. V. Kozhevnikov, Chem. Rev., 1998, 98, 171–198.
29. T. Ito, K. Inumaru and M. Misono, J. Phys. Chem. B, 1997, 101, 9958–9963.
30. S. S. de Mello, D. Van Tyne, A. N. G. Dabul, M. S. Gilmore and I. L. B. C. Camargo, Genome Announc., 2016, 4, e00992.
31. M. E. da Silva Ferreira, M. R. V. Z. Kress, M. Savoldi, M. H. S. Goldman, A. Hartl, T. Heinekamp, A. A. Brakhage and G. H. Goldman, Eukaryotic Cell, 2006, 5, 207–211.
32. P. A. Stadelmann, Ultramicroscopy, 1987, 21, 131–145.
33. R. Wright, Microsc. Res. Tech., 2000, 51, 496–510.
34. M. C. Rocha, J. H. T. M. Fabri, K. Franco de Godoy, P. Alves de Castro, J. I. Hori, A. Ferreira da Cunha, M. Arentshorst, A. F. J. Ram, C. A. M. J. J. van den Hondel, G. H. Goldman and I. Malavazi, G3: Genes, Genomes, Genet., 2016, 6, 2983–3002.
35. NCCLS, M38-A2|Reference Method for Broth Dilution Antifungal Susceptibility Testing of Filamentous Fungi, 2008.
36. NCCLS, M27-A2: Reference Method for Broth Dilution Antifungal Susceptibility Testing of Yeasts, 2002.
37. NCCLS, M7-A6: Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically, 2003.
38. E. Käfer, Advances in Genetics, 1977, vol. 19, pp. 33–131.
39. M. Sun, J. Zhang, P. Putaj, V. Caps, F. Lefebvre, J. Pelletier and J.-M. Basset, Chem. Rev., 2014, 114, 981–1019.
40. T. Okuhara, N. Mizuno and M. Misono, Advances in Catalysis, 1996, pp. 113–252.
41. L. Matachowski, A. Zięba, M. Zembala and A. Drelinkiewicz, Catal. Lett., 2009, 133, 49–62.
42. N. R. Shiju, H. M. Williams and D. R. Brown, Appl. Catal., B, 2009, 90, 451–457.
43. R. Xing, F. Wang, L. Dong, A.-P. Zheng, L. Wang, W.-J. Su and T. Lin, Food Chem., 2016, 197, 205–211.
44. D. H. Piva, R. H. Piva, M. C. Rocha, J. A. Dias, O. R. K. Montedo, I. Malavazi and M. R. Morelli, Adv. Powder Technol., 2017, 28, 463–472.
45. L. Grama, A. Man, D.-L. Muntean, Şerban A. Gâz Florea, F. Boda and A. Curticăpean, Rom. Rev. Lab. Med., 2014, 22, 111–118.
46. T. Hu, Y.-H. Li, X.-F. Kuang and C.-Z. Lu, CrystEngComm, 2017, 19, 3445–3454.
47. D. A. Friesen, L. Morello, J. V. Headley and C. H. Langford, J. Photochem. Photobiol., A, 2000, 133, 213–220.
48. T. Yamase, N. Fukuda and Y. Tajima, Biol. Pharm. Bull., 1996, 19, 459–465.
49. S. Balici, M. Niculae, E. Pall, M. Rusu, D. Rusu and H. Matei, Rev. Chim., 2016, 67, 485–490.
50. Editorial, Nat. Microbiol., 2017, 2, 17120.
51. K. J. Kwon-Chung and J. A. Sugui, PLoS Pathog., 2013, 9, e1003743.
52. T. Heinekamp, H. Schmidt, K. Lapp, V. Pähtz, I. Shopova, N. Köster-Eiserfunke, T. Krüger, O. Kniemeyer and A. A. Brakhage, Semin. Immunopathol., 2015, 37, 141–152.
53. N. B. Saleh, D. J. Milliron, N. Aich, L. E. Katz, H. M. Liljestrand and M. J. Kirisits, Sci. Total Environ, 2016, 568, 926–932.
54. X.-X. Lu, Y.-H. Luo, C. Lu, X. Chen and H. Zhang, J. Solid State Chem., 2015, 232, 123–130.
55. L. Wang, W. Yang, W. Zhu, X. Guan, Z. Xie and Z. Sun, Inorg. Chem., 2014, 53, 11584–11588.
56. S. Hu, C. Ma, F. Zhan, Y. Cao, P. Hu and Q. Zhen, Chem. Pap., 2017, 71, 1323–1329.
57. J. He, H. Pang, W. Wang, Y. Zhang, B. Yan, X. Li, S. Li and J. Chen, Dalton Trans., 2013, 42, 15637.
58. A.-L. Kubo, L. Kremer, S. Herrmann, S. G. Mitchell, O. M. Bondarenko, A. Kahru and C. Streb, ChemPlusChem, 2017, 82, 867–871.
59. I. Malavazi, G. H. Goldman and N. A. Brown, Briefings Funct. Genomics, 2014, 13, 456–470.
60. A. C. R. Souza, B. B. Fuchs, H. M. S. Pinhati, R. A. Siqueira, F. Hagen, J. F. Meis, E. Mylonakis and A. L. Colombo, Antimicrob. Agents Chemother., 2015, 59, 6581–6587.
61. H. K. Daima, P. R. Selvakannan, A. E. Kandjani, R. Shukla, S. K. Bhargava and V. Bansal, Nanoscale, 2014, 6, 758–765.
62. M. C. Rocha, K. F. de Godoy, P. A. de Castro, J. I. Hori, V. L. P. Bom, N. A. Brown, A. F. da Cunha, G. H. Goldman and I. Malavazi, PLoS One, 2015, 10, e0135195.
63. E. A. Ukshe, L. S. Leonova and A. I. Korosteleva, Solid State Ionics, 1989, 36, 219–223.
64. G. Ren, D. Hu, E. W. C. Cheng, M. a. Vargas-Reus, P. Reip and R. P. Allaker, Int. J. Antimicrob. Agents, 2009, 33, 587–590.
65. D. Das, B. C. Nath, P. Phukon and S. K. Dolui, Colloids Surf., B, 2013, 101, 430–433.
66. T. Pandiyarajan, R. Udayabhaskar, S. Vignesh, R. A. James and B. Karthikeyan, Mater. Sci. Eng., C, 2013, 33, 2020–2024.
67. M. Heinlaan, A. Ivask, I. Blinova, H. C. Dubourguier and A. Kahru, Chemosphere, 2008, 71, 1308–1316.
68. M. Moritz and M. Geszke-Moritz, Chem. Eng. J., 2013, 228, 596–613.
69. I. V. Kozhevnikov, J. Mol. Catal. A: Chem., 2007, 262, 86–92.
70. I. V. Kozhevnikov, J. Mol. Catal. A: Chem., 2009, 305, 104–111.
71. A. I. Chikin, A. V. Chernyak, Z. Jin, Y. S. Naumova, A. E. Ukshe, N. V. Smirnova, V. I. Volkov and Y. A. Dobrovolsky, J. Solid State Electrochem., 2012, 16, 2767–2775.
72. M. A. Vargas-Reus, K. Memarzadeh, J. Huang, G. G. Ren and R. P. Allaker, Int. J. Antimicrob. Agents, 2012, 40, 135–139.
73. J. L. Slonczewski, M. Fujisawa, M. Dopson and T. A. Krulwich, in Advances in Microbial Physiology, Elsevier, 2009, vol. 55, pp. 1–317.
74. T. A. Krulwich, G. Sachs and E. Padan, Nat. Rev. Microbiol., 2011, 9, 330–343.
75. C. Baker-Austin and M. Dopson, Trends Microbiol., 2007, 15, 165–171.
76. J. W. Foster and H. K. Hall, J. Bacteriol., 1991, 173, 5129–5135.
77. R. Orij, S. Brul and G. J. Smits, Biochim. Biophys. Acta, Gen. Subj., 2011, 1810, 933–944.
78. U. Nehls and S. Dietz, Appl. Microbiol. Biotechnol., 2014, 98, 8835–8851.
79. J. B. Moffat, Metal-Oxygen Clusters – The Surface and Catalytic Properties of Heteropoly Oxometalates, 1st edn, Springer, US, Boston, 2001.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7tb02763j

‡ These authors contributed equally to this work.

§ In the solid state, POM compounds present hierarchical formation divided into primary, secondary, and tertiary structures. More information about this concept can be found in the paper of Misono.²⁵

---