Programmed Cell Death: The Primary Bactericidal Mechanism Induced by Copper Nanoparticles

Abstract

Copper, a reddish and ubiquitous material in the world, possesses malleable and conductive properties that render copper and its alloys indispensable in vertical integration manufacturing. With advancements in nanotechnology and nanomaterials in recent decades, copper and its related nanoparticles have been engineered. Their applications include engineering, material science, photo−/electro-catalysis, biomedical drug delivery, agriculture, and antipathogen microbicides. Here, we studied the differing toxicity effects of two sizes of copper nanoparticles (CuNPs), recognized for their potent bactericidal properties. Concentration-dependent effects of both 20 and 60 nm CuNPs were significant in Escherichia coli (E. coli), Acinetobacter baumannii (A. baumannii), and Staphylococcus aureus (S. aureus). Sodium dodecyl sulfate, the dispersant of nanoparticles, caused the synergy effects with CuNPs in A. baumannii and S. aureus but not in E. coli. Four modulators were added to CuNP-treated bacteria. By these modulator treatments, programmed cell death was found in E. coli, A. baumannii, and S. aureus. By the BLAST search, caspase-related proteins were commonly identified in gut bacteria and A. baumannii but not in S. aureus. Furthermore, many proteins from E. coli, A. baumannii, and S. aureus were found to harbor the ULK1-catalytic domain. In short, CuNPs can be potent therapeutic agents against bacterial infections.

1. Introduction

Copper, an earth-abundant and inexpensive material, has broad applications in manufacturing and catalysis due to its malleable and conductive properties [1]. Bulk copper is reddish and can be found in its pure form without chemical or physically controlled refining [2]. However, size-dependent color changes from brown to black show the different properties of copper at the nano-scales [3]. Bulk copper and its alloys are applied in material science, engineering, photo-electrical industries, and manufacturing. In contrast, nano-scaled copper is used in agriculture, biomedicine, environmental protection, photo-image, electronics, and superconductors [4]. Fabrication of copper nanoparticles (CuNPs) is not an emerging technology. Medieval red window glasses detected by scanning electron microscopy (SEM) and transmission electron microscopy (TEM) revealed the existence of CuNPs, which might be formed in a series process of heat and redox reactions [5]. In recent decades, various methods have been utilized for the production of copper and copper-related nanoparticles, such as copper nanoclusters and copper oxide nanoparticles. These methods include microemulsion, sol-gel techniques, green synthesis, physical smashing, chemical reduction precipitation, electro−/sono-chemical approaches, and microwave irradiation [6, 7, 8]. These copper and copper-related nanoparticles are categorized as metallic nanoparticles, containing high catalytic activities and easily interacting with organic materials like lipids and proteins [6, 7, 8]. Therefore, CuNPs are considered as potent antimicrobial agents against hazardous pathogens [9].

The increasing prevalence of new and untreatable infections has aroused concerns in recent years. Traditional antibiotics struggle with problems of invalid treatments for multidrug-resistant pathogens, and developing new strategies for these pathogen extirpations becomes an important issue [10, 11]. Gram-negative bacteria, such as Escherichia coli (E. coli) and Acinetobacter baumannii (A. baumannii), and gram-positive bacteria, Staphylococcus aureus (S. aureus), are common pathogens of opportunistic infections, which usually cause severe bacteremia and in-hospital death among hospitalized patients [12, 13, 14]. Metallic nanoparticles, containing antimicrobial properties, are reported as the workable bactericides for these three bacteria and are regarded as the last-line defense against multidrug-resistant bugs [10, 15, 16]. Various metallic nanoparticles have been studied, such as silver nanoparticles (AgNPs), CuNPs, copper oxide nanoparticles (CuONPs), and zinc oxide nanoparticles (ZnONPs), with their outstanding antimicrobial contributions and biomedical applications make them approved by the U.S. Food and Drug Administration (FDA) or undergoing clinical trials [15, 16]. Although AgNPs and CuNPs are the most commonly applied in biomedical fields, CuNPs display advantages in lower cost and higher biocompatibility [15]. There are lower risks of accumulations for CuNPs in human bodies because copper-transporting adenosine triphosphatases (Cu-ATPases) help control copper homeostasis and export excess copper out of the body [7, 17]. Our previous study also demonstrated that CuNPs were harmless to human normal skin fibroblast in vitro [15].

Recently, the mechanisms of bactericide in copper-based nanoparticles have been intensely discussed. Take most metallic nanoparticles, for example. Free ions released from metal, which lead to the production of reactive oxygen species (ROS), are the primary factor for killing bacteria [18]. Copper-based nanoparticles, a type of metallic nanoparticles encompassing CuO/Cu₂O, CuS, Cu⁰, and other core/shell nanoparticles formed from various materials, are known to generate ROS [9, 16]. This is particularly reasonable for them as they exhibit +1 and + 2 oxidation states which enable them to interact with cell membranes through electrostatic attraction [16, 19]. In contrast, CuNPs exist in Cu⁰ reduction states without charges, which should theoretically make it difficult to generate ROS. Nevertheless, Lia et al. demonstrated that the antimicrobial properties of CuNPs (Cu⁰) were from ROS [15]. Additionally, Chatterjee et al. indicated that CuNPs released nascent ions via redox reactions when CuNPs were exposed to aqueous conditions, such as cell medium and serum [20]. CuNPs may employ “dry” or “wet” methods for accumulation and dissociation to reduce the proton motive force across cell membranes and increase permeability [9]. Once CuNPs entered cells through damaged cell membranes, ROS-induced oxidative stress led to lipid peroxidation, protein denaturation, and DNA degradation [9, 15, 20]. These series of attacks increased the efficacy of antimicrobial abilities in CuNPs.

Our previous study on CuNPs corroborates these findings regarding ROS-induced results [15]. We specifically investigated CuNP toxicities in E. coli and elucidated the correlation among CuNP sizes, concentrations, and antimicrobial activities. We found that different sizes and concentrations of CuNPs induced various mechanisms of bacterial cell death in E. coli [15]. E. coli, A. baumannii, and S. aureus are three major pathogens associated with nosocomial infections [12, 13, 14]. Studies on the antimicrobial activities of CuNPs against these three pathogens hold clinical significance. In this study, we used two sizes of CuNPs, 20 nm, and 60 nm, to investigate the bactericidal abilities and the programmed cell death mechanisms across the three pathogens.

2. Materials and methods

2.1 Nanoparticle preparation

Two sizes of CuNPs, 25 nm (marked as 20 nm) and 60–80 nm (marked as 60 nm), were purchased from Sigma-Aldrich Company (Sigma-Aldrich, St. Louis, MO, USA). As a dispersant, 1.0 mM of sodium dodecyl sulfate (SDS) (Sigma-Aldrich) was applied. The aggregation of CuNPs was prevented by capping with SDS and ultrasonic bath at 40°C for 30 min. The aqueous CuNPs were freshly prepared and used immediately. To ensure the sizes and the physicochemical properties of CuNPs, different equipment, such as transmission electron microscope (TEM; H-7500; Hitachi, Tokyo, Japan), liquid particle attractor (FlowVIEW, Hsinchu, Taiwan), Flow AOI (FlowVIEW), and Zetasizer Nano ZS (MalvernInstruments, Worcestershire, UK), were used in this study [15].

2.2 Bacterial cell culture

E. coli (Migula) Castellani and Chalmers strain 25,922 and A. baumannii Bouvet and Grimont strain were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). S. aureus, initially derived from ATCC 13567, was purchased from the Bioresource Collection and Research Center (BCRC, Hsinchu, Taiwan). Three bacteria were cultured in Luria–Bertani (LB) broth (Becton, Dickinson and Company, Sparks, MD, USA) or LB agar (FocusBio, Miaoli, Taiwan). Bacteria were grown until the value of optical density (O.D.) at an absorbance of 600 nm reached 0.5 under aerobic conditions at 37°C with shaking at 200 rpm. The O.D. value was measured using a Bio-Rad iMark microplate reader (Bio-Rad Laboratories, California, USA).

2.3 Viability determination

To understand the antimicrobial activities of 20 and 60 nm CuNPs, three bacteria were treated with PBS (the negative control), 0 (mock), 1, 5, 10, 50, and 100 μg/mL of 20 or 60 nm CuNPs, respectively, at 37°C for 24 h. The difference between the group of negative control and mock is that mock contains dispersant while negative control does not. Bacteria were treated with 70% alcohol for 24 h as the positive control. For viability measurement, PrestoBlue® Cell Viability Reagent (Invitrogen, Carlsbad, CA, USA) was added to each treated bacterium. After 2 h of incubation at 37°C and 200 rpm, the fluorescence at 590 nm was measured by a Varioskan LUX multimode microplate reader (ThermoFisher Scientific, MA, USA). The value of fluorescence represented the survival of cells.

To detect the minimum bactericidal concentration (MBC), three bacteria were treated with either 20 nm or 60 nm of CuNPs at the concentrations of 0, 1, 5, 10, 50, and 100 μg/mL for 24 h. After overnight treatment, each group of bacteria was transferred to the LB agar plates and incubated for another overnight. The standard colony counting method was used for colony number calculation, and MBC was determined by the colony number.

2.4 Cell death mechanism measurement

To comprehend the bactericidal mechanism of 20 and 60 nm CuNPs, we chose four modulators of programmed cell death to apply in CuNP-treated bacteria. Z-VAD-FMK (marked as Z-VAD; Sigma-Aldrich, Saint Louis, MO, USA) is an apoptosis inhibitor that binds to caspase-family proteins [21]. SBI-0206965 (marked as SBI; BioVision, Milpitas, CA, USA) and wortmannin (marked as Wort; Abcam, MA, USA) are autophagy inhibitors, which work on serine/threonine kinase ULK1 and the class III phosphatidylinositol 3-kinase (PI3K), respectively [22, 23]. Necrosulfonamide (NSA; Sigma-Aldrich, Saint Louis, MO, USA) blocking mixed lineage kinase domain-like pseudokinase (MLKL) serves as the necroptosis suppressor [24]. Before the CuNP treatments, bacteria were pretreated with 100 nM Z-VAD for 30 min, 5 μM SBI for 2 h, 100 nM Wort for 30 min, or 0.5 μM NSA for 1 h. The pretreated bacteria were then incubated with 20 nm or 60 nm of CuNPs at 0, 1, 5, 10, 50, and 100 μg/mL for 24 h, containing (or not) the same modulator at the same concentration in the pre-treatment stage. The cell viabilities of bacteria rescued by modulators were detected with PrestoBlue® Cell Viability Reagent. The protocol of the PrestoBlue assay was the same as previously described.

2.5 Statistical analysis

Data are expressed as mean ± standard deviation (SD). Mean values and SDs were calculated from at least three independent experiments carried out in triplicates for each treatment group. Statistical comparisons were performed by one-way ANOVA and the Student’s t-test, using statistical significance at P < 0.05 with stars and symbols (*, #, $, and ‡) (Figure 1).

Figure 1.

The process steps in this study.

3. Results

3.1 Bactericidal activities in CuNPs

In our previous study, we disclosed the effects of different sizes and concentrations of CuNPs in E.coli [15]. Here, we further examined whether the same effects would be observed in other bacteria. We selected three clinically important pathogens, E. coli, A. baumannii, and S. aureus, which cover both Gram-negative and gram-positive bacteria, to investigate the bactericidal effects of 20 and 60 nm CuNPs. The bacteria were treated with different concentrations of the two sizes of CuNPs. Concentration-dependent bacterial cell viabilities and colony numbers were shown in both 20 and 60 nm CuNP treatments (Figure 2). Similar percentages of viabilities compared to the positive control (70% EtOH) were displayed in 50 and 100 μg/mL of 20 and 60 nm CuNPs (Figure 2A and B). Significant reductions in viability were observed in the three bacteria at low concentrations (1 and 5 μg/mL) of 20 nm CuNPs. Notably, E. coli exhibited greater sensitivity to 5 μg/mL of 20 nm CuNPs compared to A. baumannii and S. aureus (Figure 2A). In contrast, mild reductions of cell viabilities were observed at 1 and 5 μg/mL of 60 nm CuNP treatments in E. coli. (Figure 2B). Significant effects of 60 nm CuNPs were evident when the concentration reached 10 μg/mL or above. Unlike E. coli, remarkable effects of 60 nm CuNPs in A. baumannii and S. aureus (Figure 2B) were observed starting from the concentration of 1 μg/mL. The most obvious difference in cell viability among the three bacteria was seen at the concentration of 10 μg/mL in 60 nm CuNP treatments.

Figure 2.

The bactericidal abilities of CuNPs in three pathogens. E. coli, A. baumannii, and S. aureus were treated with different concentrations of CuNPs in two sizes. Bacteria treated with PBS or 70% alcohol served as the negative and positive controls, respectively. The relative cell viabilities of 20 nm (A) or 60 nm CuNPs (B) in E. coli, A. baumannii, and S. aureus were determined using the PrestoBlue assay. The colony numbers were calculated for E. coli (C), A. baumannii (D), and S. aureus (E) treated with either 20 or 60 nm of CuNPs. Data are presented as mean ± SD from three independent experiments with triplicates in each treatment group. Statistical comparisons were performed by ANOVA and student t-test; *, #, and $ indicate P<0.05 for 20 or 60 nm CuNP treatments relative to the negative control, respectively, and ‡P<0.05 for comparisons among the three pathogens in the same concentration treatments.

To determine the minimum bactericidal concentration (MBC) in CuNP treatments, three bacteria were incubated with either 20 or 60 nm CuNPs, and a standard colony counting assay was performed (Figure 2C-E). In E. coli treatments, 20 nm CuNPs exhibited an excellent inhibition of colony formation at concentrations of 1, 5, 10, 50, and 100 μg/mL, while 60 nm CuNPs displayed significant inhibition at concentrations of 5, 10, 50, and 100 μg/mL (Figure 2C). The MBC was determined to be 10 and 50 μg/mL in 20 and 60 nm CuNP treatments, respectively (Table 1). In A. baumannii treatments (Figure 2D), higher concentrations of 20 and 60 nm CuNPs were required for inhibition of colony formation. The MBCs reached 50 and 100 μg/mL in 20 and 60 nm CuNP treatments, respectively (Table 1). S. aureus displayed the highest sensitivity to both sizes of CuNPs (Figure 2E and Table 1). Notably, the results in mock groups (0 μg/mL) suggested a synergy effect between the dispersant, SDS, and CuNPs in S. aureus treatments (Figure 2A, B, and E).

	MBC (μg/mL)
	20 nm CuNPs	60 nm CuNPs
E. coli	10	50
A. baumannii	50	100
S. aureus	10	10

Table 1.

The MBC in two sizes of CuNPs for three bacteria.

3.2 Bactericidal mechanism studies in CuNPs

To understand the bactericidal mechanisms triggered by CuNPs, four modulators, SBI, Z-VAD, NSA, and Wort, were chosen for subsequent experiments. These modulators are frequently employed in research on mammalian cell death mechanisms. SBI binds to Unc-51-like kinase (ULK) and interrupts the initiation of autophagy [22], while Z-VAD binds to caspase-like family proteins and inhibits apoptotic programmed cell death [21]. NSA interacts with MLKL to prevent necroptic cell death [24], and Wort increases survival rates by depleting autophagosome formation [23]. In our experiments, E. coli, A. baumannii, and S. aureus were pretreated with the four modulators before co-treatment with the modulators and two sizes of CuNPs, respectively (Figure 3).

Figure 3.

Effects of CuNPs and programmed cell death modulators on bacterial cell viability.E. coli (A and B), A. baumannii (C and D), and S. aureus (E and F) were treated with PBS (negative controls) or varying concentrations (0, 1, 5, 10, 50, 100 μg/ml) of 20 or 60 nm CuNPs, either alone or in combination with programmed cell death modulators, including SBI, Z-VAD, NSA, and Wort, respectively. Cell viability was assessed by PrestoBlue assay. Statistically significant differences at P<0.05 (* vs. negative control) were determined using Student’s t-test. ANOVA was performed for statistical comparisons between the mock and other modulators at each concentration, with significance indicated by ‡P<0.05.

In E. coli, Z-VAD- and NSA rescued the survival rate in 20 nm CuNP-treated bacteria, indicating that the cell death pathways rescued by these two modulators were the major mechanisms at low (1 and 5 μg/mL) and high concentrations (10 μg/mL), respectively (Figure 3A). However, the death pathway that could be rescued by Z-VAD played a vital role in all concentration treatments of 60 nm CuNPs (Figure 3B). Similar results to those observed in E. coli could also be seen in A. baumannii treated with 20 and 60 nm CuNPs (Figure 3C and D), with the exception of the possible involvement of the cell death pathways rescued by SBI at low concentrations (1 and 5 μg/mL) of 60 nm CuNPs (Figure 3D).

In contrast, S. aureus displayed a possible different bactericidal mechanism in CuNP treatments (Figure 3E and F). Z-VAD affected bacterial viability only at 1 μg/mL of 20 nm CuNP group, while NSA increased viability at 10 and 50 μg/mL of 20 nm CuNP (Figure 3E). None of the four modulators were able to raise the viabilities of S. aureus in all concentrations of 60 nm-CuNP treatments (Figure 3F).

Interestingly, the mechanisms of bactericidal effects from the dispersant of nanoparticles, SDS, were distinct in A. baumannii and S. aureus. Modulators SBI, Wort, as well as NSA rescued bacterial viabilities in S. aureus incubated with SDS in the mock group (0 μg/mL) of 60 nm CuNPs (Figure 3F). Conversely, only NSA contributed to viability elevation when A. baumannii was incubated with SDS in the mock group (0 μg/mL) of 60 nm CuNPs (Figure 3D). However, SBI decreased viabilities in the mock group when E. coli was treated with 60 nm CuNPs (Figure 3B). No significant differences were seen in the effects of SBI, Z-VAD, NSA, and Wort treatments on E. coli, A. baumannii, and S. aureus when bacteria were incubated with the mock of 20 nm CuNP groups (Figure 3A,C, and E).

3.3 NCBI protein database alignment of programmed cell death modulators

To verify the effects of modulators on programmed cell death in E. coli, A. baumannii, and S. aureus, specific domain sequences were searched on the NCBI Protein database using BLAST (Figure 4). Z-VAD, a pan-caspase inhibitor, acts by irreversible binding to the catalytic site of caspase proteases. In the NCBI protein database, proteins such as WP_000709408 and WP_317300176 from E. coli and A. baumannii, respectively, were annotated as caspases. These caspases contain a Peptidase_C14 domain (pfam00656, caspase domain), which has been identified in some human proteins, such as mucosa-associated lymphoid tissue lymphoma translocation protein 1 isoform b (NP_776216). However, no significant similarity was found for S. aureus (Figure 4A).

Figure 4.

Sequence alignment of the possible Z-VAD- and SBI-responsive domains in E. coli, A. baumannii, and S. aureus. (A) Alignment of the caspase domain among the caspase-like proteins from human, E. coli, and A. baumannii. (B) Alignment between the human catalytic domain of threonine-protein kinase ULK1 (NP_003556) and proteins from E. coli, A. baumannii, and S. aureus. Hash signs (#) denote possible active and/or ATP binding sites annotated in the STKc_ULK1 domain. The alignment was generated by Jalview 2.

SBI is a potent inhibitor of the kinase ULK1 [25]. Through our analysis, many proteins from E. coli, A. baumannii, or S. aureus were identified to share similarity to the catalytic domain of human threonine-protein kinase ULK1 (NP_003556) (Figure 4B). Relative to the length of human ULK1 (1050 amino acids), only 260 amino acids were found to be matched among the proteins in these three bacteria.

Despite searching for the bacterial homologs of the NSA activation domain, 4HB domain of MLKL (NP_689862.1), and PI3K-like catalytic domains for Wort in the NCBI protein database using BLAST, no hits were found in E. coli, A. baumannii, or S. aureus (data not shown). Nevertheless, our sequencing alignment results and the bactericidal effects shown in Figure 3 displayed consistency.

4. Discussion

In this study, we demonstrated that both 20 and 60 nm CuNPs were effective bactericides in E. coli, A. baumannii, and S. aureus (Figure 2, Table 1). These nanoparticles displayed concentration-dependent antimicrobial activities. The non-toxic nature in normal human skin cells and the broad spectrum of antimicrobial activities suggest that CuNPs could be utilized in hydrocolloid dressing to defend against infections. Our investigation into the bactericidal mechanisms revealed the potential involvement of programmed cell death in these bacteria, as evidenced by the effects of different modulators and protein-sequence alignments of catalytic domains. These findings shed light on bacterial stress response and offer a promising new strategy against drug-resistant bacteria.

Metallic and metallic oxide nanoparticles have gathered interest in recent years due to their biomedical applications. They are capable of generating ROS and attacking pathogens. Our previous study confirmed this phenomenon [15], demonstrating that both 20 and 60 nm CuNPs destroyed the cell membranes of E. coli through ROS production. It is believed that the release of metals and electrochemical reactions from metallic-related nanoparticles, especially smaller particles with a higher surface-area-to-volume ratio, contribute to their heightened ROS production [26]. ROS can cause various cellular problems, including lipid peroxidation, protein damage, DNA breakage, blocking of carbohydrate metabolism, and inhibition of the TCA cycle [27]. As a consequence of ROS-induced damage, cellular growth, energy yields, and cell survival are reduced [28]. Some cellular organelles, such as cell membranes, mitochondria, DNA or RNA, ER systems, and some proteins or enzymes, are targets for ROS attacks [29]. Damage to cellular components can lead to cell death through necrosis, apoptosis, necroptosis, and autophagy [29]. In our previous studies, breakage of chromosomal DNA in E. coli treated with 20 nm CuNPs was observed by electrophoresis [15]. Furthermore, as the concentration increased, so did the damage to DNA. However, surprisingly, condensed chromosomal DNA was observed in E. coli treated with 60 nm CuNPs [15]. Boa and Dwyer suggested that bacteria used a specific signal transduction pathway to induce DNA condensation in the process of apoptosis when faced with ROS stress [30, 31]. Additionally, Jia et al. indicated that multiple programmed cell death pathways, such as necroptosis, ferroptosis, pyroptosis, oxeiptosis, autophagy, and apoptosis were related to ROS [32]. Therefore, it is plausible that bacterial programmed cell death is induced by ROS generated by metallic and metallic oxide nanoparticles.

Programmed cell death triggered by different stressors in E. coli was first discovered in 2004 [33]. In eukaryotes, apoptotic programmed cell death is characterized by DNA fragmentation, phosphatidylserine externalization, and membrane depolarization. In E. coli, two programmed cell death systems have been identified: the EDF-mazEF-mediated death pathway and apoptotic-like death (ALD) pathway. These pathways involve a series of signal transduction routes that control bacterial survival, ultimately leading to DNA breakage, phosphatidylserine externalization, and membrane depolarization [34]. Notably, there is a crosstalk between the EDF-mazEF-mediated death and ALD pathways. Comparing E. coli to eukaryotes, it emerges that multiple programmed cell death pathways found in mammalian cells, such as necroptosis, ferroptosis, pyroptosis, oxeiptosis, autophagy, and apoptosis, are highly related to ROS [35]. These programmed cell death pathways also exhibit crosstalk in their signal transduction pathways, regulated by various checkpoints [35, 36, 37, 38]. Some signal transduction pathways involved in their survival-or-death processes share common ancestor proteins [36, 37, 38]. Different modulators or inhibitors have demonstrated to halt these progresses at specific steps [39, 40]. In this study, we used four modulators that show critical roles in necroptosis, apoptosis, and autophagy in mammalian cell studies. Co-treatment of E. coli and A. baumannii with Z-VAD and both sizes of CuNPs (Figure 3A-D) resulted in increased cell survival. Additionally, we observed changes in cell viabilities in A. baumannii co-treated with SBI and low concentrations of 60 nm CuNPs, as well as S. aureus co-treated with SBI and dispersant (Figure 3D and F). These results suggest that bacteria possess active domains similar to caspase-related and ULK1 proteins and respond to these modulators. Amino-acid sequence alignments support these findings (Figure 4). Although cell viabilities of bacteria also increased in CuNP-NSA co-treated groups, no similarity in amino-acid sequences of 4HB domain of MLKL was found in E. coli, A. baumannii, or S. aureus.

There are many similarities between prokaryotes and eukaryotes, yet certain properties have long been accepted as essential eukaryotic characteristics that distinguish them from prokaryotes. For instance, endocytosis was traditionally considered a primary survival behavior in eukaryotes. However, discoveries such as cyanobacteria’s ability to engage in energy-dependent endocytosis challenged this notion [41]. Subsequent research revealed that diverse gram-positive bacteria like Arthrobacter ilicis D-50, gram-negative bacteria such as E. coli, and archaea like Thermus aquaticus possess the ability to uptake materials through macropinocytosis [42]. This process was shown to be triggered by cell-penetrating peptide-mediated protein transduction [42]. Similarly, while cytoskeletons were once believed to be exclusive to eukaryotes, proteins like FtsZ in Bacillus and MreB in Caulobacter crescentus exhibit similarities to actins in eukaryotes [43, 44, 45]. Additionally, the CreS protein in C. crescentus serves analogous functions to intermediate filaments in eukaryotes [43, 45]. Programmed cell death in bacteria, as discussed earlier, is another example of shared biological processes. In this study, we unveiled the potential for multiple programmed cell death pathways in E. coli, A. baumannii, and S. aureus induced by different sizes and concentrations of CuNP treatments. While our understanding is currently limited by the responses observed in modulator treatments and information available in the protein databank, the possibility of necroptosis, autophagy, or other programmed cell death pathways cannot be discounted. Understanding the regulation of cell survival-death and signal transduction pathways may hold the key to developing strategies targeting bacterial stress responses, thereby overcoming antibiotic resistance. This deeper insight into bactericidal mechanisms could pave the way for novel approaches to combat multiple drug-resistant bacteria.

5. Conclusion

Metallic and metallic oxide nanoparticles, including CuNPs, have been shown to reduce viabilities and colony formation abilities in E. coli, A. baumannii, and S. aureus. Nanoparticle dispersants and nanoparticles exhibit a synergic effect in killing bacteria, particularly S. aureus. These bacteria displayed size- and concentration-dependent sensitivities to CuNPs. Furthermore, four programmed cell death modulators elicited different responses at varying concentrations of 20 and 60 nm CuNPs. Through a search of the catalytic domains responding to the modulators in the protein database, some similarities between prokaryotes and eukaryotes were found. This study suggests the possibility of a programmed-cell death system in prokaryotes, though further research is needed to validate this concept.

Acknowledgments

We are grateful for the support from the Core Facility Center, Tzu Chi University, Taiwan. This work was supported by the Grant No. MOST 106-2813-C-320-005-B (to L.-I. T. and B. R. L.) from the National Science and Technology Council, Taiwan.

Conflict of interest

The authors declare no conflict of interest.

Appendices and nomenclature

A. baumannii	Acinetobacter baumannii
AgNPs	silver nanoparticles
ALD	apoptotic-like death
C. crescentus	Caulobacter crescentus
Cu-ATPases	copper-transporting adenosine triphosphatases
CuNPs	copper nanoparticles
CuONPs	copper oxide nanoparticles
E. coli	Escherichia coli
EtOH	alcohol
FDA	U.S. Food and Drug Administration
LB	Luria–Bertani
MBC	minimum bactericidal concentration
MLKL	mixed lineage kinase domain-like pseudokinase
NSA	necrosulfonamide
O.D.	optical densities
PI3K	phosphatidylinositol 3-kinase
ROS	reactive oxygen species
S. aureus	Staphylococcus aureus
SBI	SBI-0206965
SD	standard deviation
SDS	sodium dodecyl sulfate
SEM	scanning electron microscopy
TEM	transmission electron microscopy
ULK	Unc-51-like kinase
Wort	wortmannin
ZnONPs	zinc oxide nanoparticles
Z-VAD	Z-VAD-FMK