Cold Plasmas for Biofilm Control: Opportunities and Challenges

Abstract

Bacterial biofilm infections account for a major proportion of chronic and medical device associated infections in humans, yet our ability to control them is compromised by their inherent tolerance to antimicrobial agents. Cold atmospheric plasma (CAP) represents a promising therapeutic option. CAP treatment of microbial biofilms represents the convergence of two complex phenomena: the production of a chemically diverse mixture of reactive species and intermediates, and their interaction with a heterogeneous 3D interface created by the biofilm extracellular polymeric matrix. Therefore, understanding these interactions and physiological responses to CAP exposure are central to effective management of infectious biofilms. We review the unique opportunities and challenges for translating CAP to the management of biofilms.

Challenges in Biofilm Control

Biofilms (see Glossary) are complex consortia of microbial cells (bacteria, archaea, and fungi) that are attached to a substratum (biotic/abiotic surfaces and/or each other) and embedded within a matrix of self-produced or acquired extracellular polymeric substance (EPS), including polysaccharides, DNA, protein, and lipids [1-5]. Biofilms are characterized by significantly elevated tolerance to antimicrobial agents, attributed to restricted penetration of antimicrobials, altered (decreased) growth rates and transcription within the biofilm, formation of persister cells, and quorum sensing-controlled tolerance/protective mechanisms [6,7] (Figure 1). Consequently, the effectiveness of conventional antibiotics, biocides, and normal immune clearance is severely limited and biofilms are often implicated in chronic, persistent, and recurrent infections [1-4]. Furthermore, the structure of polymicrobial biofilms facilitates transfer of antimicrobial resistance, compounding the problem. Thus biofilm formation is now widely recognized as a major virulence determinant in a wide range of chronic infections [2]. The importance and ubiquitous distribution of biofilms (Box 1) as a distinct and predominant microbial phenotype in chronic and recurrent infections (including those on medical devices) and in environmental niches and engineered systems has prompted a rethink in how we discover and develop new antibiofilm therapeutic approaches.

Figure 1. The Plasma–Biofilm Interface.

The plasma-derived reactive species that diffuse into the biofilm encounter a hydrated, cationic extracellular polymeric matrix which may sequester RONS and attenuate plasma cidal efficacy and maintains a 3D architecture supporting heterogeneous microenvironments that in turn support multispecies microcolonies. Growth rate may be reduced due to nutrient and O₂ limitations within the biofilm, leading to elevated tolerance and persister formation. Quorum sensing, leading to alterations in microbial physiology may also affect microbial tolerance to plasma-derived RONS. Finally, RONS-mediated dispersal of microbes from the biofilm may reverse plasma tolerance. Adapted from [7,87]. Abbreviations: eDNA, extracellular DNA; RONS, reactive oxygen and nitrogen species.

Box 1. Biofilms.

A biofilm can be defined as a microbially derived sessile community of cells irreversibly attached to a substratum (biotic or abiotic), embedded within a matrix of EPS that they have produced and exhibiting an altered phenotype with respect to growth rate and gene transcription. Biofilms represent the predominant mode of growth of bacteria in both natural and engineered environments and in the colonization of plants, animals, and humans. In fact, the majority of chronic human bacterial infections (estimated at 60–80%) are caused by bacterial biofilms. Biofilms provide and maintain a unique, privileged environment that allows nutrient sequestration, facilitates cooperative activities (often controlled by cell-to-cell communication, or quorum sensing), and permits the bacteria within to withstand and tolerate antimicrobial challenges (by retarding penetration of antimicrobial agents), host immune clearance, and mechanical removal from a surface. The unique environment created within the biofilm gives rise to metabolically heterogenous populations, with a spectrum of metabolic activity and dormancy. This contributes not only to elevated tolerance to antimicrobial agents and antibiotics, but also drives the emergence of persister cells, phenotypic variants capable of withstanding lethal stresses. The biofilm matrix plays a central role in providing architectural and structural integrity to the biofilm, but also maintains pH and electrochemical gradients, which ensure the heterogeneity of the microbial population within. In human health, biofilms are frequently implicated in a diverse array of chronic, persistent infections and infections of indwelling medical devices; from urinary catheters to prostheses. Bacteria such as Enterococcus faecium, S., epidermidis, Klebsiella pneumoniae, Acinetobacter baumannii, P. aeruginosa, and Enterobacter spp. (collectively known as the ESKAPE pathogens) alongside Enterococcus faecalis, E. coli, methicillin-resistant S. aureus (MRSA) and Streptococcus spp. represent the most common and life-threatening causative pathogens of biofilm-associated infections.

While antibiotic resistance remains a major global issue, in the context of biofilm control, an important distinction between antimicrobial (antibiotic or biocide) resistance and antimicrobial tolerance must be recognized. Antimicrobial resistance arises from irreversible, heritable changes to the microbial genome (and is therefore genotypic); antimicrobial tolerance is a phenotypic and reversible, and enables the microorganism to withstand and survive antimicrobial treatments [7,8]. One of the greatest threats to our ability to control infectious diseases is the lack of an effective developmental platform for antibiotic discovery [9]. In the absence of a sustainable pipeline of antibiotics, the emergence of antimicrobial resistance continues unchecked, undermining the efficacy of the remaining armamentarium of clinically useful drugs.

The growing prospect of a postantibiotic era, general pessimism surrounding new antibiotic discovery, and a growing urgency as annually, increasing numbers of patients succumb to infections caused by resistant microbial pathogens, has prompted renewed interest in alternatives to conventional antibiotics [7,10].

One such nonantibiotic option is the use of nonthermal or CAP for the eradication and control of both acute and chronic biofilm infections. Plasma medicine is as nascent research field that may be uniquely positioned to address the particular challenges associated with controlling infectious biofilms. Is it possible to translate the increased understanding of biofilms, gained over the past four decades (in particular the mechanisms governing tolerance to antimicrobial challenge), to optimize cold plasmas to control these complex microbial communities?

Plasma Medicine: A New Frontier in Biofilm Control

The efficacy of CAP in controlling bacterial biofilms has been widely reported in the plasma medicine literature, (for recent reviews, see [11,12]) and is the subject of intensive research. Topical applications have been developed using a range of plasma sources for chronic wounds, mucous membranes, and the oral cavity. While a thorough discussion of the plasma sources used within the wider plasma medicine field is beyond the scope of this review, a recent perspective describes them in more detail [13]. The 2012 Plasma Roadmap outlined the key challenges within the field of low-temperature plasma physics and technology, and its various subfields including plasma medicine [14]. The updated 2017 Plasma Roadmap redefines the key challenges for plasma medicine; particularly the potential barriers to translation of cold plasma to clinical applications which includes: (i) the effects of gas phase interactions with hydrated biological matrices, leading to plasma activated media; (ii) defining plasma dose; (iii) tissue-specific effects on the flux of species delivered; and (iv) modulation of the immune response [15].

A wide range of nonthermal atmospheric pressure plasma devices, based mainly on plasma jets or dielectric barrier discharge (DBD) configurations, have emerged over the past two decades, which offer the possibility of either direct or indirect plasma biofilm exposures (Figure 2). Direct exposure has been defined as either exposure of the sample directly under the plasma plume discharge [16,17] or where the substrate (e.g., tissue) acts as an electrode that participates in creation of the plasma discharge [18]. In contrast, indirect exposure describes sample exposure whereby the sample placement is outside the plasma discharging area [16-18]. Therefore, in the case of direct plasma exposure, the biofilm is exposed to high electric fields (especially in the case of DBD), charged particles, and UV [16-20]. While atmospheric pressure plasmas are unlikely to produce UV doses capable of contributing significantly to bacterial inactivation [19,20], high electric fields may contribute to bacterial inactivation [18,19]. The bioelectric effect, the electrical enhancement of antimicrobials against biofilm bacteria is well established, especially in the medical devices field [21]. Intense nanosecond-pulsed electric fields have been shown to alter the physiology of mammalian and bacterial cells [22]. In addition, Stoodley and colleagues have demonstrated that contraction and expansion of biofilms occurs under the influence of electric fields (voltage applied with oscillating polarity), leading to increased susceptibility of the biofilm to antibiotics [23]. The role of charged particles in direct exposure inactivation of bacteria has also been described, whereby charge accumulation on the bacterial membrane and the resulting electrostatic force overcomes the tensile strength of the membrane, leading to rupture [24,25]. However, Lu and colleagues have demonstrated that excited N₂, N₂⁺, and He* are not expected to play a significant role in bacterial inactivation [19]. The combined effect of UV, electric fields, and charged particles may prove synergistic in inactivation of bacteria during direct plasma exposures [26].

Figure 2. Schematics of (A) a DBD CAP Source and (B) a CAP Jet Source for Biofilm Eradication and Control.

In A (direct plasma exposure), bacterial biofilm cells are exposed to plasma reactive species, charged particles, high electric fields/current and UV. In B (indirect plasma exposure), bacterial biofilm cells are exposed primarily to reactive oxygen and nitrogen species generated in the plasma afterglow and not the electric fields/current or charged particles. Adapted from [12,26]. Abbreviations: CAP, cold atmospheric plasma; DBD, dielectric barrier discharge.

In indirect exposure scenarios (Figure 1) CAP generates a dynamic and diverse array of reactive oxygen species (ROS), which include atomic oxygen (O), superoxide (O₂^•−), hydroxyl radicals (^•OH), singlet oxygen (¹O₂⁻), ozone (O₃), and hydrogen peroxide (H₂O₂); and reactive nitrogen species (RNS), which include nitric oxide (NO^•), nitrogen dioxide (NO_2^•), nitrite (NO₂⁻), nitrate (NO₃⁻), and peroxynitrite (ONOO⁻) [27-29]. While CAP-induced biological effects are thought to be due to the predominant reactive oxygen and nitrogen species (including organic radicals) which are generated in, and transferred to, the hydrated matrix of the biofilm [30], the overall composition of reactive species produced can vary significantly depending on how CAP is generated (the source configuration, feed gas, and gaseous environment) and nature of the liquid or biological interface with which the plasma interacts. Biofilms represent a complex, hydrated biological interface containing water, inorganic ions, and organic molecules [polysaccharides, extracellular DNA (eDNA), surfactants, lipids, and proteins] [4], dispersed throughout with bacterial microcolonies. The precise composition of plasma-derived reactive species that diffuse into the biofilm are therefore dependent on the interaction of reactive oxygen and nitrogen species (RONS) with matrix components, interaction with dissolved solutes in the hydrated matrix, and affected by alterations in pH and oxygen concentration within the biofilm.

A number of studies have examined the depth of penetration of plasma-derived active species into biofilm matrices. Xiong and colleagues have demonstrated that a He/O₂ plasma jet completely inactivated Porphyromonas gingivalis biofilms to a depth of 15 μm [31], while plasma penetrated Enterococcus faecalis biofilms 25.5 μm in depth [32]. The efficacy of an argon plasma was shown to be reduced in proportion to Pseudomonas aeruginosa PA103 biofilm thickness [33], while Alkawareek and colleagues reported (He/O₂) plasma penetration into P. aeruginosa PAO1 biofilms between 40 and 80 μm in depth [34]. However, these studies were limited by the final thickness of the cultured biofilms, rather than demonstrating definitive plasma penetration limits within the biofilm, which are likely to be influenced significantly by not only the extent of the matrix (biomass) but also its molecular composition (influencing the nature of the interaction between matrix and reactive species), which may vary between species and strains of bacteria. In a recent study, around 5% of ROS and 80% RNS produced by a He/O₂ plasma jet penetrated through 500 μm of biological tissue, with some RONS capable of penetrating up to 1.25 mm. This indicates that absolute penetration depths of RONS into biofilm matrices are likely to be underestimated by current studies [35].

Additionally, significant challenges exist to translating data from in vitro, ex vivo, or in vivo animal studies of CAP biofilm control to clinical applications. There exists an overall general lack of standardized tests to evaluate new antibiofilm technologies, antibiotics, biocides, and antibiofilm agents despite the development of a small number of standardized methods [including ASTM International (formerly ‘American Society for Testing and Materials’) methods for biofilm biocide susceptibility testing] [36], and a lack of agreement as to the clinical benefit of biofilm susceptibility testing [37]. Recently, the biofilm research community has developed guidelines, through the ‘minimum information about a biofilm experiment’ (MIABiE) initiative [38]. These guidelines aim to simplify the exchange, interpretation, and comparison of biofilm experimental data, by encouraging researchers in the field (partly by enforcing the use of controlled vocabularies and adherence to minimum information standards) to document their experiments comprehensively and unambiguously. The benefits of such guidelines within the biofilm field, where high-throughput screening and omics data sets are commonplace, are clear. Given the complexity, scale, and variability of experimental protocols and data generated within the highly interdisciplinary plasma medicine field, the research community should consider generating similar guidelines (for example, minimum information about a plasma medicine experiment – MIAPME or minimum information about a CAP experiment – MICAPE) for reporting on experiments on plasma–biological interactions and effects.

Biofilm Microenvironment in Chronic Infection: Preconditioning Pathogens to RONS?

Microorganisms account for less than 10% of the dry weight of the biofilm, with extracellular polymeric components largely accounting for the remainder [4,39]. These extracellular polymeric substances buffer RONS while facilitating microbial adhesion and aggregation, water retention, confer biofilm spatial arrangement/architecture, and importantly, provide a protective barrier against host defenses and antimicrobial agents [4]. The biofilm also contains multiple microenvironments (hypoxic and anoxic) in which a combination of bacterial aerobic, anaerobic, and microaerobic respiration and metabolic pathways operate [40]. Thus to develop a successful strategy for CAP-mediated control of biofilms in chronic infections, the bacteria type, metabolic state and oxygen microenvironment (as well as potential reaction products generated) must all be taken into consideration when controlling efficacy (Figure 1). Our understanding of the biofilm–plasma interaction requires fundamental experimental studies both in CAP physics, microbiology and meta-transcriptomics to understand the plasma-active species driving biological phenomena within the biofilm.

An important question regarding the potential efficacy of CAP is the propensity of certain pathogens to survive adverse conditions. For example, P. aeruginosa can withstand oxygen limitation within the biofilm by adjusting respiration from aerobic (occurring only in the presence of oxygen) to anaerobic (occurring in the absence of oxygen) and/or microaerobic (occurring only in oxygen tensions lower than that of air). In particular, NO₃⁻ respiration occurs under both hypoxic and anoxic conditions, where NO₃^™ is reduced to yield, for example, NO, N₂O and N₂ gases [40]. In vivo, hypoxia may be driven by polymorphonuclear (PMN) leucocyte response to a pathogen, where significant O₂ consumption occurs during the respiratory or oxidative burst (production of ROS) response and in the production of NO [40-42]. This consumption of O₂ can lead to hypoxic conditions in the cystic fibrosis (CF) lung, where the PMN-driven reduction in O₂ leads to reduced growth rates and increased tolerance to antibiotics [40]. Chronic exposure to PMN-derived ROS, superoxide and NO in vivo could predispose certain pathogens to display inherent tolerance towards CAP treatment. This may explain observations of significantly elevated tolerance to CAP treatment in clinical isolates from chronic infection, when compared with laboratory or type strains [43]. However, the observed elevated tolerance to plasma exposure in biofilms is likely to be multifactorial. Here, nonthermal plasma could have a significant advantage over conventional antibiotics or biocides, since the chemical diversity and flexibility created by nonthermal plasmas may be refined and tailored for such applications, creating a dynamic and responsive therapeutic approach.

Biofilm Matrix Interaction with Plasma-Derived RONS

Biofilm matrix components such as eDNA and polysaccharides may also attenuate the antimicrobial activity of CAP exposure via interaction and sequestration of plasma-derived RONS. Polysaccharides, such as alginate, have been shown to interact with RONS and therefore may contribute to the attenuation of the biocidal activity of CAP, similar to the effect exogenous organic matter on biocidal disinfectant activity when evaluated under ‘dirty’ conditions in standard efficacy tests. Alginate, an anionic polysaccharide, is a major component of mucoid biofilms of P. aeruginosa (and other pseudomonads), where it performs a range of vital functions including cell–cell adhesion, maintenance of biofilm fluidity and architecture, and resistance to desiccation [44].

Importantly, alginate can also scavenge free radicals, primarily ROS, released by neutrophils and macrophages as part of the respiratory or oxidative burst in response to pathogens [45]. Furthermore, alginate binds, limits diffusion of, and inhibits killing by cationic antimicrobial agents including antimicrobial peptides [46] and antibiotics such as tobramycin [47,48] and other aminoglycosides [49], leading to reduced EPS diffusion. The important role played by the biofilm EPS matrix in conferring elevated antimicrobial tolerance, both by providing a physiological shield preventing efficient ingress of toxic compounds such as antibiotics and biocidal agents, and by maintaining hydration, architecture and microenvironments within the biofilm, is increasingly recognized [2,3].

Another major component of many biofilms, including those of human pathogens such as Staphylococcus aureus and P. aeruginosa, is eDNA [3,4]. eDNA forms an integral component of the EPS, where it contributes to biofilm structure and architecture and to antimicrobial tolerance towards the aminoglycosides and cationic antimicrobial peptides. It has also been shown to induce tolerance to aminoglycoside antibiotics via acidification of the biofilm matrix [50]. Interestingly, exposure of Staphylococcus epidermidis biofilms to low (subinhibitory) doses of vancomycin increased the concentration of eDNA in those biofilms, impeding the penetration of vancomycin through the biofilm and increasing antibiotic tolerance [51]. Numerous studies have examined the effect of CAP on cellular DNA (from various biological sources). In each case, rapid degradation of DNA has been demonstrated, whereby DNA undergoes single- and double-strand breakage, the latter being regarded as a lethal event, with production of ROS recognized as an important causative factor [52-54]. The rapid accumulation of single- and double-strand breaks in DNA results in rapid fragmentation of the DNA into smaller oligonucleotides [54]. CAP-activated phosphate-buffered saline also gives rise to oxidative DNA damage and biocidal effects in Escherichia coli [55]. Clearly, eDNA sequesters a proportion of the reactive species during these events, however, rapid degradation of eDNA in the biofilm matrix following CAP exposure could increase the penetration of plasma-active species into the biofilm matrix, leading to increased rates of inactivation. Used alongside conventional antibiotics and biocides, CAP-mediated increased penetration could lead to synergistic effects. The effect of plasma-mediated eDNA degradation on biofilm viscoelasticity, architecture, and antimicrobial penetration warrants further investigation.

A recent study examined the susceptibility of biofilms of clinical isolates of the CF-associated pathogen Burkholderia cepacia [43]. Clinical isolates that produced the greatest biomass (EPS) were least sensitive to the bactericidal effects of CAP. Strains producing greatest biomass in vitro also exhibited the highest catalase activity, suggesting the enzymatic inactivation of CAP-derived ROS may have contributed to survival [43]. In a supporting study, catalase production by P. aeruginosa prevented penetration of H₂O₂, and protected aggregated bacteria within the biofilm [56]. Such highly CAP-tolerant biofilms often exhibit biphasic time–kill curves, with a marked time-dependent tailing in the rate of killing (after an initial rapid decline in bioburden) and often no complete eradication of the population [43,57]. However, the exact mechanisms by which RONS either stimulate or circumvent bacterial tolerance remain poorly understood. Despite this, evidence to support CAP-mediated tolerance and development of a persister phenotype, is emerging.

Plasma-Induced Persistence and Tolerance

Persister cells (which may be within biofilms or in the planktonic phase) are transient phenotypic variants in a microbial population, which exhibit metabolic inactivity, dormancy, and transient, high-level tolerance to stress, such as antimicrobial insult and starvation [58], and present additional challenges in effective biofilm control. Unlike resistant mutants, persistence does not occur by mutation, but rather by phenotypic variation and is thus nonheritable. This phenotypic variation results in a metabolically inactive, quiescent state whereby the cell neither grows nor dies in when exposed to a lethal stress [59]. Previously, the two most well-characterized states of microbial dormancy, persister cells and viable but nonculturable (VBNC) cells, which allow bacteria to tolerate and withstand environmental stress, were believed to coexist stochastically in microbial populations [60]. This work supported the ‘dormancy continuum hypothesis’ whereby the VBNC and persister cells utilize similar mechanisms of dormancy but occupy different physiological positions in the dormancy range [61]. The key difference in these studies was the speed of resuscitation of a culturable phenotype upon the stressor being removed [60]. However, recent work by Kim and co-workers has suggested that VBNC cells and persister cells are in fact the same dormant phenotype [62].

CAP exposure can induce the VBNC phenotype in bacterial populations exposed to nonthermal plasma [57,63-65]. More recently, persister cell formation following plasma exposure in P. aeruginosa has been described, with the redox-active molecule phenazine reported to play a role in the bacterial response. The results of this study clearly indicate the central role, played by oxidative stress in CAP-mediated killing and phenotypic alteration towards persistence [65]. Therefore, the ability to induce oxidative stress in bacterial populations, specifically in biofilms, by low level, sublethal CAP exposure needs to be carefully considered since it may prove a driver of cellular dormancy, mutation, and resistance, and incomplete eradication of the target microbial population. However, CAP treatment regimens, which rely on repeated application may, by altering the CAP/RONS profile, obviate such issues. Indeed, the specific CAP-derived stimuli which trigger these phenotypic responses warrants further investigation, as do the CAP generated reactive species which could potentially resuscitate persister or VBNC cells in the heterogeneous environment of the biofilm and return them to a metabolically active, antimicrobial sensitive phenotype.

The ability of repeated CAP exposures to lead to plasma resistance has been investigated recently [66]. In this work, MRSA and Enterococcus mundtii (Gram positive) and E. coli (Gram negative) exhibited no primary (natural) or acquired resistance to CAP after repeated exposure. This is perhaps unsurprising given the potential range of cellular targets of plasma-derived reactive species [67] and the ability of different reactive species to target discreet cellular death pathways [68]. However, these studies have been conducted in planktonic cultures, and do not take into account the effect of the privileged environment and heterogeneity of the biofilm.

NO-Induced Biofilm Dispersal

NO has been proposed to be a highly conserved regulator of dispersal in eukaryotes [69]. In the past decade, NO has been identified as a dispersal signal in a number of diverse bacteria, including P. aeruginosa, S. epidermidis, Serratia marcescens, Vibrio cholera and Fusobacterium nucleatum [70,71], modulating dispersal and antibiotic tolerance. A further study demonstrated that nitrite-induced stress in S. aureus resulted in impairment of both polysaccharide intracellular adhesion (PIA) synthesis and biofilm formation. Nitrite induced stress led to upregulation of genes associated with oxidative and nitrosative stress (including genes for DNA repair, iron homeostasis, and detoxification of ROS). Nitrite addition also eradicated preformed biofilms of S. aureus and repressed biofilm formation in S. epidermidis. Interestingly, nitrite-induced biofilm effects are repressed by addition of NO scavengers (carboxy-PTIO and bovine hemoglobin), suggesting an important role for NO in these processes [72]. Furthermore, NO has been reported to potentiate killing of B. cepacia by ROS [73]. Recently, a randomized UK clinical trial of inhaled low-dose (10 ppm) NO as an adjunctive therapy to treat chronic P. aeruginosa infection in CF patients demonstrated that NO led to disruption of biofilms in CF sputum, decreased tolerance to tobramycin and tobramycin/ceftazidime, and significant reduction in P. aeruginosa biofilms aggregates in vivo [74].

NO generation in nonthermal plasma jets has been described previously [75,76], where it was found to be the dominant long-lived species in high concentrations (concentration was dependent on jet configuration and operating conditions) and positively correlated with microbial inactivation rates [76]. Plasma-derived NO and RNS are therefore likely to participate in both direct killing in the biofilm, as well dispersal or activation of cell stress responses when exposed to nonlethal concentrations of NO. For example, RNS (particularly NO) have been shown to activate the SOS DNA repair response in E. coli [77]. The effect of sublethal exposure to nonthermal plasma-derived RONS in biofilms during clinical application of this technology contributes to understanding the complex mechanisms of plasma interaction with biofilm communities and the bactericidal/survival mechanism at play.

CAP Interference with Bacterial Quorum Sensing

Quorum sensing (QS) is a population-density-dependent transcriptional regulation mechanism (bacterial cell-to-cell signaling mechanisms reliant on the production and accumulation of self-produced extracellular chemical signals, called autoinducers), which permits bacterial populations to respond collectively to external stimuli or stress, and coordinate group behaviors, such as cellular differentiation, production of specific metabolites and biofilm formation [78-80]. In such systems, as bacterial cell number increases, the accumulation of the bacterial signaling molecule also increases until a threshold concentration is reached, whereupon gene transcription or repression occurs. The best characterized of these signaling networks involves the broad class of N-acyl homoserine lactone signaling molecules. Based on observations that QS-deficient mutants are attenuated for virulence and antimicrobial tolerance [81], QS inhibition by exogenous inhibitors has been demonstrated to control (for example) biofilm formation [82] and modulation has been proposed to control bacterial virulence and infection. For an excellent recent review, the reader is directed to [80].

QS is an attractive therapeutic target in biofilm control, since QS inhibitors or agents which otherwise attenuate cell-to-cell signaling via modification of the autoinducer molecule (e.g. acylases or lactonases), have the potential to reduce virulence factor production, increase antimicrobial susceptibility, and potentially reduce biofilm formation without imposing selective pressure on the target pathogen, which could ultimately lead to emergence of resistance. Relevant to development of CAP biofilm control applications, QS regulates expression of both catalase and superoxide dismutase in P. aeruginosa and is necessary for biofilm tolerance to H₂O₂ [83]. The ability of CAP exposures to chemically modify and inactivate N-acyl homoserine lactone QS signaling molecules, to inactivate QS-regulated virulence factors elastase (lasB) and pyocyanin [84], and to inhibit QS-regulated production of virulence factors has recently been demonstrated [85]. In addition to inactivation of other proteolytic enzymes in a time-dependent fashion [67,68], the combined effect of reduction in biofilm viability, the destruction of enzymatic activity of virulence factors and modification/inactivation of QS molecules, could facilitate healing and infection resolution in scenarios such as chronic wound biofilms. The application of CAP may also be described as an antivirulence intervention in the context of chronic biofilm infections that may require multiple exposures.

Concluding Remarks

A number of significant gaps in our current understanding of mechanism of action, relative importance of individual plasma-derived reactive species, and in understanding the long-term role of CAP in management of bacterial biofilm infections remain to be fully addressed (see Outstanding Questions). The development of devices for biofilm control is an area of active research. However, considering the multiplicity of biomedical, environmental, and agricultural niches where biofilms thrive, it is unlikely that a single device will suit all applications. A key challenge for in vivo treatment of patients and tissues for wound healing or drug activation purposes relates to local temperature increases and toxicity to nontarget surrounding cells. Ensuring complete inactivation of cells within biofilms prior to removal or matrix degradation is critical, to mitigate the potential for dislodgement to an alternative location and emergence of infection or contamination at a different site. The cytotoxic and mutagenic potential of cold plasma requires further evaluation specific to device and application. There is a small range of approved commercially available systems for clinical use, and while cold plasma demonstrates enormous therapeutic potential, ensuring the biological safety will promote clinical adoption and broader regulatory approval.

Outstanding Questions.

What are the most important antibiofilm species produced by CAP? Are all CAP RONS beneficial in controlling biofilms?

Can tolerance to CAP be overcome by development of a responsive and flexible profile of plasma-generated reactive species, which may be optimized for maximum efficacy according to bacterial species and stage of infection?

Does CAP exposure induce a persister phenotype, particularly in biofilms where EPS-embedded microcolonies may be exposed to repeated subinhibitory concentrations of ROS and RNS?

Can repeated subinhibitory CAP exposures ultimately lead to the emergence of tolerance or resistance to CAP-generated reactive species?

How can omics technologies best be applied to glean important information regarding the response of biofilm bacteria to specific RONS, to enable optimization of CAP for biofilm applications?

What parameters should be considered for the development of MICAPE?

As a therapeutic approach to the management of microbial biofilms in clinical practice, nonthermal plasmas may have significant advantages over conventional antimicrobial approaches. Principally, the ability to tailor the plasma-generated reactive species to target the multiple targets within the biofilm, discussed herein, is particularly attractive. Recently, the tailored generation of aqueous ROS in plasma-activated water has been demonstrated with AC spark and glow discharge sources in direct contact with water [86]. However, not all plasma-generated RONS may prove beneficial in biofilm control. Challenges remain in understanding the fundamental plasma–biofilm interactions that may drive tolerance and persistence in microbial consortia. This will require fundamental studies in both plasma physics and diagnostic evaluation of the diverse range of plasma sources currently used, and fundamental biofilm investigations including mass transfer of reactive species into the biofilm matrix, matrix interactions with plasma-derived reactive species, and their effects on populations surviving within the biofilm, where tolerance and ultimately resistance may emerge.

Highlights.

Biofilms are implicated in around 65% of all chronic human infections, including those associated with indwelling medical devices such as catheters and prostheses. Biofilm infections are often asymptomatic between exacerbations and challenging to detect and effectively treat using conventional antibiotics and antimicrobial agents.

CAP provides an effective multimodal, multitarget approach for controlling microbial biofilms.

Biofilms express a complex extracellular matrix of polymeric substances that may attenuate the antimicrobial efficacy of CAP via interactions with CAP-generated RONS.

Biofilm tolerance to CAP is variable between species and between strains of the same species, which may be due to production of EPS, RONS-detoxifying enzymes, or acquired tolerance to physiological RONS during chronic infections.

Glossary

Aerobic respiration

generation of respiratory energy whereby oxidation (of, for example, glucose) occurs in the presence of oxygen and where O₂ acts as the terminal electron acceptor.

Anaerobic respiration

respiratory energy generation in which electron acceptors other than oxygen are utilized. Occurs in the absence of oxygen.

Antimicrobial resistance

genetically acquired ability of bacteria to resist the effects of antimicrobial/antibiotic drugs. Resistant microorganisms are not killed or inhibited in the presence of certain antibiotic/antimicrobial agents. Antibiotic resistance is encoded by several genes within the bacterial genome, which may be transferred between bacteria.

Antimicrobial tolerance

reversible phenotypic state that confers a general lack of susceptibility to antimicrobial challenge, enabling the microorganism to withstand and survive antimicrobial treatments.

Bioelectric effect

electrical enhancement of efficacy of antimicrobial agents against biofilm bacteria.

Biofilm

aggregate of microorganisms in which cells that are frequently embedded within a self-produced matrix of EPSs adhere to each other and/or to a surface (IUPAC definition). Biofilms exhibit an altered growth rate and transcription, and elevated antimicrobial tolerance compared to free-floating (planktonic) bacteria of the same species.

Extracellular DNA (eDNA)

DNA that originates from the bacteria within the biofilm and forms an important structural component of the biofilm matrix EPS. It may be important for biofilm stabilization and antimicrobial tolerance by increasing the anionic character of the biofilm matrix, thus binding cationic antimicrobial agents.

Extracellular polymeric substance (EPS)

the major structural and functional components of microbial biofilms; they comprise a variety of high-molecular-weight polymers (for example, polysaccharides and eDNA). EPS forms a 3D, viscoelastic hydrated and often charged matrix in which microorganisms are embedded. The EPS is central to determining the physicochemical and metabolic profile of the biofilm.

Microaerobic respiration

generation of respiratory energy only in the presence of oxygen at lower oxygen tensions than in air, due to the microorganisms limited capacity to respire or presence of oxygenlabile cellular biomolecules.

Pathogen

a specific causative agent, especially a microorganism, of infection or disease.

Persister cells

phenotypic variants within a microbial population, which are dormant and transiently tolerant to stress. Persister cells neither grow nor die in the presence of lethal concentrations of antimicrobial agents or in response to lethal stress that renders the majority of the population nonviable.

Quorum sensing (QS)

mechanism of bacterial cell-to-cell communication that involves the production, release, detection, and transcriptional response to small diffusible molecules called autoinducers. This cell-to-cell communication mechanism permits population density-dependent gene regulation and transcription, and population-wide response to changes in the bacterial population.

Viable but nonculturable (VBNC)

bacterial cells that exhibit metabolic activity but cannot be cultured on laboratory media on which they would normally grow and divide. Therefore, VBNC cells are not detected by routine laboratory culture methods but may retain pathogenicity. The VBNC state may be induced by adverse conditions, such as environmental stress or antimicrobial insult. VBNC is distinct from dormancy, which is a reversible state of metabolic shutdown.