Reduced Susceptibility and Increased Resistance of Bacteria against Disinfectants: A Systematic Review

Disinfectants are used to reduce the concentration of pathogenic microorganisms to a safe level and help to prevent the transmission of infectious diseases. However, bacteria have a tremendous ability to respond to chemical stress caused by biocides, where overuse and improper use of disinfectants can be reflected in a reduced susceptibility of microorganisms. This review aims to describe whether mutations and thus decreased susceptibility to disinfectants occur in bacteria during disinfectant exposure. A systematic literature review following PRISMA guidelines was conducted with the databases PubMed, Science Direct and Web of Science. For the final analysis, 28 sources that remained of interest were included. Articles describing reduced susceptibility or the resistance of bacteria against seven different disinfectants were identified. The important deviation of the minimum inhibitory concentration was observed in multiple studies for disinfectants based on triclosan and chlorhexidine. A reduced susceptibility to disinfectants and potentially related problems with antibiotic resistance in clinically important bacterial strains are increasing. Since the use of disinfectants in the community is rising, it is clear that reasonable use of available and effective disinfectants is needed. It is necessary to develop and adopt strategies to control disinfectant resistance.


Introduction
Disinfectants, defined as biocides "Main group 1" [1], are an essential tool in combatting the spread of infectious diseases. When used properly and according to the instructions, disinfectants can help prevent pathogens' transmission and spread, especially in nosocomial infections. With the rise of life-threatening infections with antibiotic-resistant bacteria and newly emerging viruses, the use of disinfectants and virucidal sanitizing agents has increased [2,3].
Disinfectants contain one or more biocidal active substances by which harmful organisms are chemically or biologically deterred, rendered harmless or destroyed [4]. In the healthcare sector, in addition to hand hygiene, the disinfection of surfaces is just as crucial to effectively protect patients, healthcare workers and visitors from the transmission of pathogens [5,6]. Biocidal products are also used in everyday hygiene, where consumers are offered a wide range of antibacterial cleaners, hygienic dishwashers, anti-sweat textiles, hygiene wipes and hand disinfectants. Most of them contain biocidal active substances in various amounts [7,8]. Although personal and household hygiene is often equated with antimicrobial products, regular handwashing without disinfectants is far more essential and sufficient [9]. However, overuse and improper use of disinfectants can accumulate to be reflected in disinfectant resistance, potentially changing our way of life, from compromising food security to threatening our healthcare systems [6].
Antimicrobial resistance has aroused great interest in the scientific and medical community in the case of antibiotics [10][11][12][13], but less interest has been paid to disinfectants,

Data Sources and Search Strategy
We performed a systematic review using the examination, analysis and synthesis of literature and the compilation method. We followed the PRISMA guidelines [36]. The search was performed using search terms in English: (susceptibility OR resistance) AND (disinfectants OR biocides) AND (bacteria OR microorganisms). A literature search was conducted with the databases PubMed, Science Direct and Web of Science. We used the following search limits: research papers published in English related to the research topic until June 2020. We used the same search terms, search limits, inclusion, and exclusion criteria in all the databases. Predetermined inclusion and exclusion criteria were applied as presented in Table 1.

Study Selection
After revision of the databases, the results (n = 11,308) were exported and compiled with Mendeley's reference management software. We also performed hand searching and included 88 articles. Mendeley's automated process removed duplicates (n = 3318), followed by a manual search to identify and remove additional duplicates. The authors screened all abstracts (n = 7990). The search focused on articles describing bacteria that developed a substantial decrease in disinfectant susceptibility with known biocide ingredients. There were multiple reasons for excluding studies, mostly for lacking MIC values or disinfectant concentrations. For the final analysis and the review, 28 sources that remained of interest were included and screened based on their full text by two independent reviewers (See Figure 1).

Study Selection
After revision of the databases, the results (n = 11,308) were exported and compiled with Mendeley's reference management software. We also performed hand searching and included 88 articles. Mendeley's automated process removed duplicates (n = 3318), followed by a manual search to identify and remove additional duplicates. The authors screened all abstracts (n = 7990). The search focused on articles describing bacteria that developed a substantial decrease in disinfectant susceptibility with known biocide ingredients. There were multiple reasons for excluding studies, mostly for lacking MIC values or disinfectant concentrations. For the final analysis and the review, 28 sources that remained of interest were included and screened based on their full text by two independent reviewers (See Figure 1).

Data Extraction and Analysis
The relevant data were first extracted by MP and checked by UR. The characteristics of the identified relevant sources were presented in a table, where we described the findings from the review and analysis of the relevant literature. The main findings from the identified sources were highlighted. The extracted data included: (1) disinfectant category; (2) publication author(s), year, country, journal; (3) study aim/purpose; (4) main results of the identified research study.

Results
Articles describing the resistance of bacteria against seven different disinfectants were identified, namely: triclosan (10 articles), peracetic acid (2 articles), hydrogen peroxide (3 articles), ethanol and isopropanol (1 article), formaldehyde and glutaraldehyde (2 articles), chlorhexidine (4 articles), benzalkonium chloride and didecyldimonium chloride (5 articles). For this review, the definition of bacterial resistance to a disinfectant is based on an importantly decreased susceptibility in different tests (e.g., disk diffusion test, minimum inhibitory concentration MIC) reported by the clinical microbiology or research laboratories described in the reviewed studies. All 28 selected publications that met the search criteria are classified in Table 2.  In the study, the authors compared the proteomic profile of the susceptible serovar S. enterica Typhimurium with its isogenic triclosan tolerant strain to decode cellular mechanisms that promote biocide tolerance.
Changes in the proteome of Salmonella were observed when exposed to sublethal concentrations of triclosan, which gave insights into mechanisms for the response and tolerance. Resistance to benzalkonium chloride is higher in single and mixed-species biofilms than in planktonic grown cells. After exposure for 15 min to 100 µg/mL, mixed-species biofilms are more resistant to benzalkonium chloride than single-species biofilms. The resistance against peracetic acid treatments (15 min to 100 µg/mL) is also higher in single and mixed-species biofilms than planktonic grown cells, but the differences are less pronounced.
Spoering and Lewis (2001), USA, Journal of Bacteriology Studying biofilms of P. aeruginosa wild-type strain PAO1 and comparing its resistance against biocide when comparing it to planktonic cells.
When comparing biofilms to logarithmic-phase planktonic cells, biofilms were considerably tolerant to the biocide. On the other hand, stationary-phase planktonic cells were more tolerant of peracetic acid than biofilms. The MBC for all three populations was 400 g/mL. Newer clinical isolates of E. faecium were more resistant to alcohol than their predecessors. Using a 70% isopropanol surface disinfectant, mutated E. faecium isolates were ten times more tolerant to disinfectant than isolates from decades ago. Strain ST796 had a reduced tolerance to isopropanol of 1.14 log 10 . Four hundred nucleotide positions mutated on two or more pairs of sequences. The study's main objective was to perform a preliminary examination to detect apparent differences between Salmonella serotypes and isolates, to link them with the resistance to disinfectants, for which there are extensive data regarding Danish fattening flocks.
In MICs of five disinfectants commonly used in the Danish or English poultry sector, few variations were observed. Most differences from the isolates having high MICs were determined when using formaldehyde, but only a few isolates differed from the high MIC isolates when using the other four disinfectants. The aim was to investigate the effects of sub-MIC concentrations of CHX on gram-negative bacteria, particularly the P. aeruginosa strain, which is known to have an intrinsic resistance to CHX, and the susceptibility of CHX-resistant strains to antibiotics.
After the fourth subculture, growth occurred within 24 h with a further increase in the MIC in P. aeruginosa strains NCIMB 10421; the MIC was significantly increased from the original 10 µg/mL to more than 70 µg/mL. The significance of these findings is still unclear, as the concentration of CHX in clinical use is much higher than that at which the authors obtained resistance. Multiple studies have shown an increased MIC from approximately 4-to 60-fold for specific bacteria for the disinfectant triclosan, making it epidemiologically relevant for increased bacterial adaptability and resistance [16,[37][38][39][40][41][42][43][44][45]. The mechanisms for the elevated MIC were various mutations at the genetic level. For E. coli these were: deletion of the ycjD gene [45], mutation at codon 93 of the fabI gene, and mutation of the MarR transcription activator within the marRAB operon c, which regulates the operation of efflux pumps [41,43]. For P. aeruginosa, there was deletion of the fabV gene, leading to a decreased fatty acid synthesis and consequent inhibition of the production of acyl-homoserine lactones and other virulence factors, such as LasA/LasB, alkaline proteases, phospholipases, lipases, exotoxin A, rhamnolipid and pyocyanin, and a reduced pathogenicity [46]. It also affects the MexCDOprJ gene, PAO1, which encodes 12 RND pumps [47,48]. For S. aureus, intracellular malonyl-CoA inhibits the activity of the transcriptional repressor FapR, which directly interacts with the fabI gene, physiologically regulating its expression. This results in the most common mutations, polymorphisms, within the coding regions of C34T and MO035 in the sa-FabI region [49,50]. For S. enterica there was: a mutation in the fabI gene, mutation of the AcrAB and TolC genes that regulate efflux pumps, and inactivation of the transcriptional regulators ramA and marA [51][52][53].
For the disinfectant peracetic acid, the MIC was raised four-fold in one of the three bacteria tested (P. aeruginosa), while in the others, there were no significant changes in susceptibility. According to the described example, it could be classified as relevant in the indication of resistance, although the results are currently deficient due to the lack of multiple studies and unequal conditions [54,55]. For E. coli, mutations in the genes erm (B), tet (M), and tet (L) were observed [56].
For the disinfectant hydrogen peroxide, the MIC was also relevantly elevated in only one of the three bacteria studied (A. baumannii) and can be treated as a possible indicator of resistance here as well. However, due to deficient studies and unequal conditions, no conclusion regarding resistance can be made [57][58][59]. The cause of the elevated MIC were gene mutations. In all bacteria, mutations were in genes that regulate catalase (Kat), alkyl hydroperoxide reductase, and DNA-binding proteins that allow the catalase-reversible mechanism's inhibitory effect on SpxB expression [60,61].
For the disinfectant benzalkonium chloride in multiple relevant studies reviewed, in three of the four bacteria, the MIC rises only 1-4-fold. This could make it epidemiologically relevant for increased bacterial adaptability and resistance, and a research/clinically relevant biocide [68,[78][79][80][81]. The cause of the elevated MIC were gene mutations. For P. aeruginosa, mutations of efflux pump genes such as MDR mexA-mexB-oprM and mexC-mexD-oprJ were observed [82,83]. For E. coli, a mutation in the sugE gene located in the 94 regions of a chromosome that phenotypically inhibits a groEL mutation were observed [84,85]. For S. aureus, mutations of six different genes (i.e., qacA / B, qacC (smr), qacG, qacH and qacJ) that contribute to the development of resistance to QAC were observed [86][87][88][89].
The MIC values for most commonly used biocides against clinically important bacteria are presented in Table 3. The bacteria considered resistant had an increased MIC at least two times the average MIC in the first column.

Criteria to Identify Resistant Strains
In order to understand resistance, there is an emphasis to distinguish between intrinsic and extrinsic resistance. Inherent resistance, known as natural resistance, is chromosomally encoded resistance, which determines the basic spectrum of effects of a disinfectant and the phenotypic resistance, e.g., biofilms. Extrinsic or acquired resistance develops through mutation by incorporating mobile genetic elements (horizontal gene transfer), transferable plasmids and other cell elements [114,115]. A clear distinction also needs to be made between phenotypic adaptation, which is reversible when exposure to the biocides ends, and acquired resistance, being genetically determined and usually stable [115]. When studying antibiotic resistance, the European Committee on Antimicrobial Susceptibility Testing has decided to: "define separate dividing points for the detection of bacteria with resistance mechanisms and the monitoring of resistance development using wild-type cut-off values (WCV) or epidemiological cut-off values (ECOFF or ECV) and the guidance of therapy via clinical breakpoints" [101,116]. As defined by the European Committee on Antimicrobial Susceptibility Testing, the ecological concept of antibiotic resistance states that ECOFFs are defined based on the normal distribution of MICs in a given bacterial species. Any isolate with a MIC above the epidemiological cut-off value (ECOFF), which is the upper limit of the normal distribution of the MIC for a given antimicrobial agent and a particular species, is considered resistant [43,96].
In the case of studying biocide resistance, however, no limits have been set so far, and there are no clear criteria to determine whether a microbe is susceptible to the biocide or not. Therefore, we can use the average MIC values obtained from individual laboratory studies conducted under relatively similar conditions. The observed relevant increase in the MIC value can indicate a decreased susceptibility or even resistance. When interpreting the results, the in-use concentration of the disinfectants used must be considered since the in-use concentration may also be higher than the actual measured MIC values. In this case, we cannot talk about the resistance but only about a decreased susceptibility.

Most Common Bacterial Mechanisms to Develop Resistance against Disinfectants
Bacteria control and overcome the effect of disinfectants in different ways (Table 4), such as restricted permeability of the cell wall, the expression of efflux systems, enzymatic degradation, changes in target sites, and the formation of biofilms [23,117]. Changes in cell surface hydrophobicity, ultrastructure, protein composition, and fatty acid modifications appear to occur [118,119]. For example, inactivation of the lipooligosaccharide biosynthesis genes causes resistance in A. baumannii [120]. Modifying the outer membrane proteins and an increased expression of cellular structures may increase the sensitivity to disinfectants [118,119,121]. Impermeability of the outer membrane occurs because of the lipopolysaccharide component, which increases the penetration of disinfectants and affects the size and expression of pores, thereby preventing entry and affecting sensitivity [122]. The hydrophilic porin channels on the outer membrane regulate the passage of solutes and are a significant barrier to hydrophilic substance penetration [123]. They also have a negative charge, which can cause the disinfectant molecules to bounce away from the bacterial cell.
Bacteria can also grow as biofilms, endospores, and within cellular macrophages. In most natural habitats, microorganisms grow and survive as associated biofilms [124]. Monocultures of several different species or mixed phenotypes of a particular species can form biofilms. It is a community of nonmobile microorganisms that are irreversibly attached to a surface and inserted into a polymeric extracellular matrix. The insensitivity of biofilms to disinfectants is due to altered microbial growth rates, which can be attributed to nutrient depletion in the biofilm, and disinfectant binding to the biofilm, which is neutralized or degraded [35]. Such an organization may moderate the concentration of antimicrobial disinfectants and antibiotics to which deeper biofilm cells are exposed. Such cells accidentally grow slowly, starve, and express stress phenotypes, including regulating efflux pumps and flushing out disinfectants [125].
Slightly less effective mechanisms involve the enzymatic degradation or inactivation of disinfectants when concentrations of agents, such as formaldehyde, chlorhexidine, and quaternary ammonium compounds, are lower than those used in clinical trials practice [126]. The exposure of bacteria to minimal inhibitory concentrations of disinfectants results in the induced expression of neutralizing enzymes, which is crucial for the biodegradation of disinfectants [127]. Examples of the neutralization of disinfectants have been given in several species of bacteria, for example, Pseudomonas fluorescens TN4 isolated from sludge was able to degrade DDAC, which belongs to the group of quaternary ammonium compounds. The isolate was also able to degrade other QACs by the N-dealkylation process [128].
One major cause of bacterial resistance is the active transport of substances to the cell exterior, the so-called efflux with proteins. Efflux pump mechanisms perform essential physiological functions [97]. Although existing in all living cells, those found in bacterial and mammalian cells are especially important for clinicians and pharmacologists since they constitute an important cause of antimicrobial resistance. Multidrug Resistance (MDR) efflux pumps present an ongoing research topic in antibiotic resistance and are also responsible for disinfectant resistance mechanisms [129]. One of the fundamental mechanisms of action is the efflux pump's influence and the modulation of its genes. These efflux systems existed in bacteria long before the use of disinfectants and antibiotics in humans to treat infections. The mechanism involves the secretion of toxic compounds through a bacterial cell wall with a membrane-bound protein composed of at least three components. The increased expression of these pumps can raise the minimum inhibitory concentration to a high level, resulting in resistance to disinfectants [130] and greater sensitivity and crossresistance to antibiotics [40]. Research data show that pump expression reduces the efficacy of various classes of disinfectants, including chlorhexidine digluconate, hydrogen peroxide, benzalkonium chloride, chloroxylenol, iodine compounds, triclosan, quaternary ammonium compounds, phenolic parabens and intercalates [131,132]. Among the best-studied systems of genes that regulate the secretion of biocides are mexAB-oprM, mexCD-oprJ and mexEF-oprN in P. aeruginosa [133], acrAB-tolC, acrEF-tolC and emrE in E. coli [134], smeDEF in bacteria Stenotrophomonas maltophilia [135], and norA and mepA in S. aureus. In the highly resistant nosocomial bacterium A. baumannii, the efflux activity is regulated by the quacA and quacB genes [57,77,136]. Bacteria use the same pumps to remove antibiotics and biocides. Thus, they can select antibiotic-resistant mutants that over-regulate such pumps [137].
Another important factor contributing to the development of disinfectant resistance is the mode of action of disinfectants. Biocides have a broader spectrum of activity and may have multiple targets, while antibiotics tend to have specific intracellular targets [29]. However, in the case of biocides with a particular antimicrobial mechanism (e.g., quaternary ammonium compounds-QAC's), the development of antimicrobial resistance against disinfectants, and cross-resistance to antibiotics, are especially well documented [138].

70-95% Ethanol solution
Denaturation of bacterial membrane proteins and dissolving lipid components such as antiparallel β and 3 10 helical turns of proteins, C-H deformations in lipids, inhibition of nutrient transport via membrane-bound ATPases, alteration of membrane pH and membrane potential.
Horizontal gene transfer, transformation and transduction and core genome mutations in the chromosome nucleotide position on the rpoB gene β subunit of RNA polymerase.

Conclusions
Antimicrobial resistance in healthcare facilities has been occurring and regularly increasing over the last ten years. Growing evidence from in vitro studies has shown that bacteria have a tremendous ability to respond to chemical stress caused by biocides by several different mechanisms [160]. The main reason for emerging resistance is attributed mainly to the overuse, abuse and misuse of disinfectants [160][161][162]. Relevant increases in MIC concentrations, changes at the genetic level, and clearly altered mechanisms were observed in studies of several bacterial species in the presence of disinfectants. Through the most relevant of the reviewed articles, we can define the results for disinfectants based on triclosan and chlorhexidine, where the critical deviation of the MIC was observed in multiple studies.
Given the ongoing problems with multiple antibiotic resistance in clinically important bacteria strains and the potential for increased resistance to disinfectants, the use of which is rising in the community, it is clear that the prudent use of available and effective antimicrobials is needed. It is essential to develop and adopt strategies to control disinfectant resistance, for which the following factors will make a significant contribution. To solve the disinfectant resistance problem, it is essential to comprehensively summarize the disinfectant resistance mechanisms and to understand the resistance influencing factors [163]. It is also necessary to establish ECOFF values for biocides, without which any research is challenging and, to some extent, inaccurate. Harmonized methods for biocide susceptibility testing need to be developed.
Further studies are needed to establish a link between disinfectant exposure and resistance development, as many studies in clinical or external settings are currently limited. The rotation of disinfectants, where one disinfectant should be replaced by another having a different mechanism of action, is recommended [164]. The same types of disinfectants are used both in healthcare institutions and among the general population; therefore, their prudent use and consumption, as we know in the case of antibiotics, are complicated to control. Since selective pressure caused by disinfectants is exerted on both commensal and pathogenic bacteria [165,166], monitoring for resistant genes in nonpathogenic or commensal bacteria would make sense. Health-related infections acquired in the community need to be researched annually. More attention should be paid to the correct use of disinfectants by the general public, although supervising the proper use of disinfectants among the general population is very difficult to implement.
The risks and benefits of using disinfectants in the environment need to be weighed to determine whether additional precautions are required to guide the development and use of disinfectants [167]. If bacterial resistance increases and develops against many regularly used disinfectants in clinical and industrial settings, overuse in reflection of the COVID-19 pandemic could place an additional burden on global public health [168].
Author Contributions: U.R.: substantial contributions to conception and design, data acquisition, analysis and interpretation; drafting the article and revising it critically for important intellectual content; final approval of the version to be published. M.P.: substantial contributions to conception and design, data acquisition, analysis and interpretation; drafting the article. S.K.: substantial contributions to conception and design, critically revising the article. D.D.: substantial contributions to conception and design, data acquisition, analysis and interpretation; drafting the article and revising it critically for important intellectual content; final approval of the version to be published. S.Š.T.: substantial contributions to conception and design; final approval of the version to be published. All authors have read and agreed to the published version of the manuscript.

Funding:
We are very grateful to Sanlas Holding GmbH, Austria who acted as a co-financer of the project. The research was also financially supported by the national research program (P2-0118).
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.