Antimicrobial resistance expansion in pathogens: a review of current mitigation strategies and advances towards innovative therapy

Abstract

The escalating problem of antimicrobial resistance (AMR) proliferation in clinically important pathogens has become one of the biggest threats to human health and the global economy. Previous studies have estimated AMR-associated deaths and disability-adjusted life-years (DALYs) in many countries with a view to presenting a clearer picture of the global burden of AMR-related diseases. Recently, several novel strategies have been advanced to combat resistance spread. These include efflux activity inhibition, closing of mutant selection window (MSW), biofilm disruption, lytic bacteriophage particles, nanoantibiotics, engineered antimicrobial peptides, and the CRISPR-Cas9 gene-editing technique. The single or integrated deployment of these strategies has shown potentialities towards mitigating resistance and contributing to valuable therapeutic outcomes. Correspondingly, the new paradigm of personalized medicine demands innovative interventions such as improved and accurate point-of-care diagnosis and treatment to curtail AMR. The CRISPR-Cas system is a novel and highly promising nucleic acid detection and manipulating technology with the potential for application in the control of AMR. This review thus considers the specifics of some of the AMR-mitigating strategies, while noting their drawbacks, and discusses the advances in the CRISPR-based technology as an important point-of-care tool for tracking and curbing AMR in our fight against a looming ‘post-antibiotic’ era.

Introduction

The extensive use of antibiotics has led to a discourse on their roles in promoting resistance among microorganisms. Antimicrobial resistance (AMR) has become a perpetual global health problem, with an estimation that by 2050, deaths related to AMR will exceed 10 million, resulting in societal and financial costs of over US$100 trillion.¹ In 2017, the WHO published the priority list of antibiotic-resistant pathogens, including 12 bacterial families categorized as critical, high- and medium-risk pathogens posing the greatest threat to global health (Table 1). Seven of the 12 groups were resistant to β-lactam antibiotics. Consequently, there is a reinvigorated effort to discover new therapeutic alternatives and mitigate the impact of resistance against current antibiotics, thus setting the goal for future research and development strategies.²

Table 1.

WHO priority list of increasing AMR pathogens

Classification	Resistant microorganism
Priority 1: Critical	Carbapenem-resistant A. baumannii
	Carbapenem-resistant P. aeruginosa
	ESBL-producing Enterobacteriaceae
Priority 2: High	Vancomycin-resistant E. faecium
	MRSA, vancomycin-intermediate S. aureus and vancomycin-resistant S. aureus
	Clarithromycin-resistant Helicobacter pylori
	Fluoroquinolone-resistant Campylobacter species
	Fluoroquinolone-resistant Salmonella species
	Cephalosporin-resistant Neisseria gonorrhoeae and fluoroquinolone-resistant N. gonorrhoeae
Priority 3: Medium	Penicillin-non-susceptible Streptococcus pneumoniae
	Ampicillin-resistant Haemophilus influenzae
	Fluoroquinolone-resistant Shigella species

Source: WHO².

There are concerns about the evolution of novel resistant strains in clinical and environmental settings. The aquatic environments, for instance, have been noted as important hotspots for the emergence and spread of resistance because of the constant interaction of pathogens from different sources with subinhibitory concentrations of antibiotics, heavy metals and other biocidal agents, causing increased resistance due to natural selection.^3,4 Several surveillance studies have documented the contribution of discharged municipal effluents to the proliferation of resistant bacteria and their associated resistance genes in the aquatic milieus.^5,6 Likewise, healthcare facilities are considered important epicentres for the rapid spread of MDR pathogens due to factors such as suboptimal infection control, novel mutations and resistant strain selection.⁷ Additionally, multiple risk factors, including surgery, endotracheal intubation, indwelling catheters, intensive antibiotic use and chronic diseases, predispose patients to MDR healthcare-associated infection (HAI).^8,9 HAIs have become one of the major problems in terms of fatalities and associated increased cost of care. Important MDR pathogens, including MRSA, VRE and carbapenemase-producing Enterobacteriaceae (CPE) have been associated with many HAIs. Moreover, about one-third of infections in Europe are linked to antimicrobial-resistant microorganisms.¹⁰

AMR can occur through different mechanisms depending on the acquired or selective gene alterations. Generally, resistance may involve restricting drug entry through porin conformations or down-regulating porin levels in response to antimicrobial exposure. Bacteria can also extrude antimicrobials by increasing efflux activity, while some modify the active binding sites for the specific antibiotic. Moreover, they may produce a wide range of drug-inactivating enzymes, including β-lactamases, carbapenemases and NDM-1, and a myriad of virulence factors.^11,12 These enzymes are encoded by genes borne on extrachromosomal genetic material, which may be easily exchanged between different strains via horizontal gene transfer (HGT).

Similarly, there is an increasing problem of human infections associated with antifungal resistance. For instance, the number of deaths linked to fungal infections in the USA increased from 4000 to about 5000 between the years 2013 and 2018.¹³ Fungal species including Aspergillus, Candida, Cryptococcus and Pneumocystis are among the most notable species with rising resistance.¹⁴ The widespread use of antifungal agents in agriculture for the treatment of invasive fungal infections is one factor noted to contribute to the escalating challenge of antifungal resistance. Likewise, viral-encoded resistance has increasingly been reported against established antiviral agents while the genetic basis of viral resistance remains unclear. For instance, there is increased resistance to different classes of antiretroviral (anti-HIV) drugs, including PIs, NRTIs and NNRTIs.¹⁵ Viral-associated respiratory infectious diseases are among the most common causes of human mortality.^16–18

Reported data on AMR expansion continue to be worrisome while the situation is compounded by the lack of significant novel compounds to control known and emerging drug-resistant pathogens in the short term. Frequently, antimicrobials are prescribed or administered empirically at many primary healthcare points without a definite guarantee of their efficacy for the particular pathogen, resulting in overprescription. The empirical prescription of antimicrobials has been noted as a significant contributor to AMR development against many established drugs.^19,20 The need for rapid and guided antimicrobial prescription has heightened the need for improved point-of-care testing for AMR-related infections to make quick therapeutic interventions to forestall resistance development in the medium and long term.

Essentially, challenges relating to the expansion of AMR demand urgent mitigating strategies. In recent years, antimicrobial stewardship has aided the optimization of benefits from current antimicrobials and curtailed resistance development for better therapeutic outcomes.²¹ However, dwindled antimicrobial development in the last 30 years necessitates adopting innovative strategies to combat AMR. Among the possible options, the combined or single deployment of efflux activity inhibition, mutant selection window (MSW) closing, quorum quenching, biofilm disruption, antimicrobial peptides, nanoantibiotics, lytic bacteriophage particles and clustered regularly interspaced short palindromic repeats (CRISPR)-Cas9 technology hold prospects in curbing resistance. This review thus highlights the potential and drawbacks of some of these strategies in our effort to overcome the challenges of microbial resistance proliferation while proposing the optimization of CRISPR-Cas technology as an important tool in the prevention and control of AMR.

Efflux pump inhibition

Efflux pumps are important regulators of the internal environment of microbes through the extrusion of toxic solutes, including antibiotics, thereby conferring resistance to a wide range of pathogens. Six major classes of efflux pumps have been described in bacteria to date and are categorized as either primary [the ATP-binding cassette (ABC) family] or secondary [the major facilitator superfamily (MFS), the multidrug and toxin extrusion (MATE) family, the small multidrug resistance (SMR) family, the resistance-nodulation-cell division (RND) superfamily and the proteobacterial antimicrobial compound efflux (PACE) family] pumps, based on the energy source used in pumping out solutes.⁴ Substrate redundancy among these various efflux pump types allows the extrusion of a particular type of antibiotic by different efflux pump systems. Similarly, a particular efflux pump is capable of exporting structurally and chemically diverse solutes.²² Since efflux pumps play a critical role in stress response for bacteria, they are an interesting phenomenon in the molecular biology of resistance and potential targets for developing novel inhibitors toward revitalizing existing antibiotics that have lost their potency, or potentiating weak antibiotics. A number of strategies have been employed to discover or develop efflux pump inhibitors (EPIs). Some of these are highlighted in Figure 1. Plant-derived secondary metabolites have been noted as major sources of novel and effective natural EPIs, and therefore there is a crucial need for targeted screening of widely available medicinal plants.²³ For instance, 5′-methoxyhydnocarpin-D, a flavonolignan produced by Berberis species has been identified as a potent EPI with efficacy greater than some clinically used antibiotics against MDR Staphylococcus aureus.²⁴ Silybin is another effective flavonolignan MDR EPI that has been used to treat MDR strains of S. aureus.²⁵ Other plant families such as Apocynaceae, Fabaceae, Zingiberaceae, Cucurbitaceae and Convolvulaceae, among others, have been identified as important sources of EPIs.²² Table 2 shows some reported efflux activity inhibitors with tremendous potential against AMR. One major drawback associated with plant-derived metabolites, hampering their efflux inhibitory ability, is the possession of inherent antimicrobial activity, which should be carefully considered when selecting candidate medicinal plants for exploration as EPI.²² The exhibition of antimicrobial potency by these secondary metabolites triggers the development of alternative resistance mechanisms by the MDR pathogen. Another challenge with the current screening strategy is that the leading compounds often show inhibitory activity against specific efflux pumps, whereas pathogens are frequently equipped with multiple classes of efflux pumps.³⁴ This results in a lack of diversity in terms of substrates targeted by the EPIs.

Figure 1.

Reported strategies of efflux activity inhibitors.

Table 2.

Some reported plant-derived, microbial and synthetic compounds with efflux activity inhibition potential

EPI compound	Source	Targeted efflux activity	Reference
Anthraquinones, stilbene, procyanidin, polydatin and resveratrol derivatives	Belowground part of Fallopia japonica	P-glycoprotein in MDR1-transfected Madin–Darby canine kidney II (MDR1-MDCK II) cells, Caco-2 cells, and the ABC super-family G member (ABCG2)	26
Safrole	Essential oil of Ocotea odorifera	Transporter NorA	27
Synthetic antigen-binding fragments (Fabs)	Synthetic antibody phage-display libraries	Quinolone transporter NorA	28
Limonene compound	Synthetic chemical	MepA structure	29
Diosmetin (3,5,7-trihydroxy-4-methoxyflavone)	Synthetic chemical	NorA and MsrA transporters and MRSA specific pyruvate kinase	30
Baicalein (5,6,7-trihydroflavone)	Roots of Scutellaria baicalensis and leaves of Thymus vulgaris	NorA, MRSA pyruvate kinase	31
Piperine and its derivative, piperidine	Isolated from Piper nigrum	NorA, uncharacterized efflux pump-Rv1258c, glycoprotein of ABC transporters	32,33
Flavonolignan (5′-methoxy-hydnocarpin)	Isolated from Berberis fremontii	TetK	34
Gallotannin (1,2,6-tri-O-galloyl-β-d-glucopyranose)	Terminalia chebula fruits	Efflux pump activity of uropathogenic E. coli	35
Skullcapflavone II (5,2′-dihydroxy-6,7,8,6′-tetramethoxyflavone)	Plant-derived compounds	Efflux pump Mycobacterium. aurum and Mycobacterium smegmatis	36
Sideroxylin and its derivatives	Extract of Hydrastis canadensis	NorA MDR pump	37
Reserpine	Roots of Rauwolfia serpentina	MFS and RND superfamily efflux pumps	38,39
Palmatine	Roots and rhizomes of Berberis vulgaris	MexAB-OprM	40
Nobiletin (5,6,7,8,3′,4′-hexamethoxyflavone)	Plant-derived compounds	Efflux pump M. smegmatis	36
Conessine	Holarrhena antidysenterica extract	MexAB-OprM	41
Zerumbone	Extract of Zingiber zerumbet (L.)	RND-type efflux pumps	42
Milbemycin		CDR1 efflux pumps	43
Coumarins	Extract of Mesua ferrea	NorA, MepA pumps	44,45
Galangin	Extract of Alpinia calcarata	NorA	46
2-(2-aminophenyl) indole (RP2)	Soil bacterial isolate IMTB 2501	NorA, TetK and MsrA transporters	47
Synthesized indole derivatives	Chemically synthesized	NorA	48
EA-371α and EA-371δ	Streptomyces species extracts	MexAB-OprM	39,43
Enniatins B, B1 and D	Fusarium species Y-53	ABC transporters, Pdr5p pump	49
Beauvericin (BEA)	Fungal metabolites	ABC transporters	43,50
Ethyl 4-bromopyrrole-2-carboxylate (RP1)	Microbial extracts	AcrAB-TolC, MexAB-OprM	51

Although information is relatively scarce on microbial-derived EPIs, a number of studies have reported EPIs produced by microorganisms as a survival strategy against competing microbe species.⁵² EA-371α and EA317δ are two reported microbial-derived EPIs with inhibitory activities against Pseudomonas aeruginosa PAM1032 MDR pumps. At the MICs of 2.5 and 1.25 mg/L, EA-371α and EA317δ have been reported with an 8-fold reduction of fluoroquinolone effect.³⁸ Other EPIs are of synthetic or semisynthetic origins, e.g. peptidomimetics, piperazines, pyridopyrimidines, and pyranopyridine derivatives and quinolone derivatives.⁵³ They act either by energy dissipation, involving decoupling energy and efflux activity, or by directly binding to functional efflux pumps.⁵⁴ In spite of their prospects, the success of EPIs has been limited due to challenges from scientific to economic, and no EPI has been commercialized so far.^55,56 Another critical challenge militating against the use of EPIs is the lack of sufficient preclinical and clinical data, prompting the need for more research efforts toward generating sufficient data and side effects for their use in treating infectious resistance strains. Continuous research and development, coupled with leveraging the rapidly expanding field of artificial intelligence/machine learning (AI/ML) could certainly assist in predicting molecular properties and target sites for novel EPIs. AI/ML has been applied in bioinformatics and various models for structural prediction of biologics and other chemical compounds. Advances in neural network algorithms present scientists with the opportunity to influence the traditional scheme of antimicrobial discovery.⁵⁷ Importantly, ML allows the avoidance of dereplication by filtering out already reported antimicrobial drugs, including EPIs.

Closing of MSW by dual targeting compounds

Combined therapy is considered an effective strategy to prevent drug resistance development. The theory behind this is that the emergence of resistance against antimicrobials with differing mechanisms of action would require the simultaneous development of specific gene mutations at every target, a phenomenon which is extremely rare.⁵⁶ In relation to the MSW theory, selection for mutant strains would be closed when the MIC of a drug is equal to the mutant prevention concentration (MPC), that is when MIC = MPC. In practice, exposure to high drug concentrations is associated with toxicity. Similarly, combined antimicrobial therapy faces numerous challenges, such as the pharmacodynamics differences between the combined agents. Conversely, dual-targeting compounds, such as fluoroquinolones targeting DNA topoisomerase and DNA gyrase, offer advantages over multiple combined therapy without the pharmacokinetic mismatches challenge. Thus, dual-targeting compounds have good potential in reducing bacterial resistance and have been adopted in the novel therapy combination of a β-lactam and a β-lactamase inhibitor such as meropenem/vaborbactam.^58,59 Nyerges et al.⁵⁶ designed a balanced dual-targeting antibiotic using a medicinal chemistry workflow. The leading compounds of the rationally designed drug, ULD1 and ULD2, displayed excellent antimicrobial potency against a broad range of Gram-positive bacteria. The designed antibiotic almost equipotently inhibits the DNA gyrase and topoisomerase IV complexes of the bacteria, while also interacting with several evolutionary conserved amino acids in the ATP-binding pockets of the protein being targeted, with a very rare chance of resistance development.⁵⁶ The study concluded that the rationally designed antibiotics could serve as a novel therapeutic agent against MDR bacterial infections. In another report, a synergistic combination of two antimicrobials, closing each other’s MSWs, was used to control AMR development. The study noted that different proportions of a dual-targeting antibiotic preparation displayed different MPCs and MSWs. Also, the combination shows that the smaller the FIC indexes (FICIs) of the compounds, the higher the probability that their MSWs were closed to each other. Theoretically, the experimental results showed that synergistic combinations (roxithromycin/doxycycline) closing each other’s MSWs had a great potency to prevent resistance in MRSA, in line with MSW and MPC hypotheses.⁶⁰ One major limitation of the MSW closing is the lack of sufficient in vivo data, such as animal models and clinical trials.⁶¹ Likewise, the concept has limited its assumption to mutants with two rare spontaneous resistance mutations and does not consider higher frequencies of mutational events such as those associated with HGT of resistance factors.

Inhibition of quorum sensing (QS) and biofilm disruption

High microbial density in biofilms can produce sufficient small signalling molecules for QS. This process is often accompanied by the activation of a number of downstream cellular functions including virulence expression, production of toxins, spore formation, disinfectant tolerance, biofilm formation, motility and AMR, among others. Because biofilms play a critical role in resistance, inhibition of QS, the vital communication mechanism by which biofilm microorganisms synchronize their activities, has been a subject of intensive investigation and a very promising strategy towards curbing biofilm-associated resistance.⁶² Attempts at disrupting biofilm activities have led to the identification of numerous macromolecules capable of inhibiting or quenching the QS communication system, a process referred to as quorum quenching (QQ). QQ molecules generally reduce or even completely block the expression of virulence factors, including those involved in biofilm formation.⁶³ Reported QQ mechanisms (Figure 2) include blocking the synthesis of signal molecules,⁶⁴ enzymatic inactivation or degradation of signal molecules,⁶⁵ inhibition of signal transduction cascades⁶⁶ and antagonizing signal molecules–receptor analogues.⁶³ In relation to their structure and function, some QQ compounds mimic QS signalling molecules, e.g. synthetic autoinducer peptides, which are similar to autoinducer peptides (AIPs), and halogenated furanones, which are homologous to N-acyl homoserine lactones (AHLs). Another group of QQ molecules includes enzyme inhibitors, e.g. triclosan and closantel.⁶⁷ A list of common QS inducers in Gram-positive and Gram-negative bacteria and some QQ molecules reported is presented in Table 3.

Figure 2.

Some QQ molecules and their structures.

Table 3.

Reported QS inducers and QQ molecules

QS inducer molecules	QSI (QQ) molecules
Gram-negative bacteria (AHL) examples  N-butanoyl-l-homoserine lactone, produced by P. aeruginosa  N-(3-hydroxybutyryl)-l-homoserine lactone, produced by Xenorhabdus nematophilus  N-(3-hydroxy-7-cis-tetradecenoyl)-l-homoserine lactone, produced by Rhodobacter leguminosarum  N-dodecanoyl-l-homoserine lactone, produced by K. pneumoniae, among others	AHL-lactonase AHL-acylase Paraoxonase l-Canavanine AHL-oxidoreductase AHL-oxidase Furanones Human hormones
Gram-positive bacteria (autoinducer peptides) examples  Competence stimulating peptide (CSP), produced by Streptococcus mitis  Extracellular death factor (EDF), produced by E. coli  Carnobacteriocin B2 (CbnB2), produced by Carnobacterium piscicola  Autoinducing peptide 2 (AIP2), produced by S. aureus  iAM373, produced by Enterococcus faecalis, among others

In general, the enzymes involved in the inactivation of QS signals are termed QQ enzymes, while chemicals used in disrupting the QS signalling pathway are called QS inhibitors. Both of these groups of compounds are collectively referred to as QSIs (QS inhibitors).⁶² One major advantage of QSIs as a control strategy for resistant pathogens is their ability not to exert pressure on the growth of microorganisms, and thus reduce or eliminate resistance development. In spite of the significant gains recorded in adopting the QQ strategy, some concerns and challenges have taunted the QS pathway as an ideal therapeutic target.^68,69 For instance, the AI-2 (autoinducer-2) QS signals are equally involved in microbial physiological processes such as cell division, morphogenesis and DNA repairs. Non-selective blocking of the AI-2-producing lux system may have a direct or indirect influence on human microflora adherence ability, resulting in the disruption of the microbiota homeostasis.⁵⁴ Similarly, some studies have suggested that the QQ strategy can lead to the development of bacterial strains with improved survivability and increased pathogenicity.^70,71

Besides QQ, biofilm disruption strategy may involve the prevention of surface adhesion by coating a surface with antimicrobial compounds such as metal-based nanomaterial and biosurfactants. These agents can also be applied as anti-biofilm or anti-corrosive products, such as paints, to prevent rust and degradation of pipes and other surfaces.⁶³ Moreover, a combination of different strategies such as mechanical (sonication) and chemical (silver-containing antimicrobial agents) could be an effective anti-biofilm approach.⁷²

Interference with the QS mechanics, as well as microbial biofilm disruption, have found important applications in numerous scientific fields, including therapeutics and industrial applications. For instance, a QSI, acylase PvdQ, has been reported to exhibit an irreversible hydrolysis of the AHL signal molecule with great therapeutic efficacy for the treatment of pulmonary infectious in a mouse model.⁷³ Overall, continued research on anti-biofilm strategies would improve their application for treatment purposes. Areas such as advanced biofilm detection and imaging technology, targeted and personalized therapies and more clinical and validation trials would enhance their application.

Engineered antimicrobial peptides (AMPs) to curtail resistance

AMPs are short peptides (approximately 10 to 50 amino acids) that are ubiquitous in almost all life forms. AMPs are structurally amphipathic within the host defence context, allowing the segregation into clusters of hydrophobic and hydrophilic amino acids. Structurally, the most recognized classes of AMPs are the α-helix and the β-sheet classes.⁷⁴ Due to their ubiquity, AMPs act as the first line of defence against invading microbial pathogens, as part of the innate immune response. Depending on the composition of the amino acid amphipathic moiety, they act significantly as antibacterial, antiparasitic, antifungal and antiviral agents.^75,76 They hold obvious advantages over contemporary antimicrobials due to their ability to modulate host immune response, possession of broad-spectrum activity against biofilm formation and slower chances of resistance emergence. Contrary to conventional antibiotics, the activity of which often depends on vital cellular processes, AMPs act mainly as membrane-active antimicrobials, rapidly targeting pre-existing cellular structures such as LPSs, a process described as membrane perturbation.⁷⁷ While the majority of AMPs’ activity involves interaction with membranes, other mechanisms of activity such as nucleic acid binding and cell wall inhibition have been identified.⁷⁸ The mechanisms of action of AMPs can be classified broadly as membrane models and intracellular models. Other advantages of AMPs include their ability to elicit anti-infective host immune response, anti-biofilm properties, anti-tumour properties and the potential to enhance the efficacy of traditional antibiotics through synergistic combination. AMPs are found in microorganisms (for example, nisin and gramicidin from bacteria,⁷⁸ echinocandins from fungi⁷⁹), lower animals (for example, magainin and cancrin from amphibians^80,81) and mammals (for example, cathelicidins and defensins).⁸² The major setbacks with natural AMPs include decreased activity in vivo due to their binding to plasma proteins, susceptibility to digestion by proteases, the exhibition of unclear pharmacokinetic properties and host toxicity.^83,84 Efforts at overcoming some of these problems have led to the development of some rationally highly enhanced helical peptide designs called engineered cationic AMPs (eCAPs) through the extension of the amphipathic helical structure by adding arginine, valine and tryptophan.^77,85 The structural modification in eCAPs altered the hydrophobicity, amphipathicity and charge with improved antimicrobial potential and specificity.⁸⁶ eCAPs have also shown potent antimicrobial activities against colistin-resistant ESKAPE (Enterococcus faecium, S. aureus, Klebsiella pneumoniae, Acinetobacter baumannii, P. aeruginosa and Enterobacter spp.) pathogens with a lesser propensity to a selection of resistant bacterial strains in vitro.⁷⁷

Whilst some successes have been recorded with the development of eCAPs, their applications have structural and functional limitations, coupled with a stricter regulatory environment, which has hindered their use as potential therapeutic agents.⁸⁷ Increased risk of toxicity has also been associated with attempts to increase their antimicrobial potencies while they are mostly susceptible to proteolytic degradation and have a short half-life. The adoption of new advances in genetic engineering and screening techniques, such as the CRISPR technology, transcription activator-like effector nucleases (TALENS) and zinc-finger nucleases (ZFNs), offer opportunities and enhanced synthesis options for diverse forms of AMPs, opening up newer and immense future potential in this area yet unravelled.

Bacteriophage and depolymerization of microbial biofilm

Phages or bacteriophages are one of the most promising tools used as treatment alternatives for resistant microbial pathogens. Phages infect bacteria and are mostly highly specific to their host, although some broad host-ranged phages have been identified. They can reproduce rapidly in suitable hosts and have been found in association with a wide range of bacterial species in many natural ecosystems.⁸⁸ Exploration of natural bacteria predators, such as phages, is highly regarded in phage therapy and has been used to target specific bacterial strains while preserving the host microbiota. Numerous reports exist on phage treatment of resistant strains.^89–91 In the report of Jasim and colleagues,⁸⁹ a therapeutic cocktail of bacteriophages was developed for the treatment of XDR and pandrug-resistant strains of A. baumannii from an Iranian hospital. All the isolates treated were sensitive to at least one of the lytic bacteriophages. They noted that using phage cocktails rather than individual phage therapy reduces the chance of bacterial resistance to lysis compared with lysis by individual phage. The cocktail further cured laboratory mice of bacteraemia when applied in vivo. The endolysin activity of the cocktail against the A. baumannii strains showed a potent antimicrobial activity of >1 log bacterial density reduction in just about 1 h of treatment. The study concluded that the application of the phage cocktail was efficient at solving the problem of superbugs both in vitro and in vivo.

In related studies, bacteriophages were employed to depolymerize microbial biofilms.^92,93 Phages can produce depolymerases that penetrate the inner layers of the biofilm exopolysaccharide matrix to degrade its components. In the study of Gabisoniya et al.,⁹³ the combination of bacteriophages vB-Pa$ and vBPa5, specific for P. aeruginosa, prevented biofilm development by six strains of MDR P. aeruginosa. Pretreatment of the strain with the bacteriophages also prevented biofilm formation in vivo. Likewise, Chhibber et al.⁹² observed the disruption of a mixed-species biofilm of K. pneumoniae B5055 and P. aeruginosa PAO and an enhanced bacterial count reduction by a combination of KPO1K2 and Pa29 bacteriophages. The authors noted that depolymerase-producing phage KPO1K2 caused the degradation of the biofilm matrix, thus allowing the non-depolymerase-producing phage Pa29 to infect and kill the P. aeruginosa bacteria. They noted that there was no significant reduction in the density of P. aeruginosa when the non-polymerase-producing Pa29 alone was applied to the biofilm and concluded that the phage combinations could be used as a topical application of coating for foreign bodies to treat infections caused by these MDR strains. Additionally, there are genetically engineered phages developed through recombination processes. The model presents the opportunity for the development of bacteriophages with phenotypes suitable for specific hosts, such as biofilm-associated MDR strains of P. aeruginosa.^94–96 Such modifications may also include the expansion of the host range with altered specificity. In a study by Li et al.,⁹⁷ WGqlae, a recombinant T4-like phage, was obtained from the genetic engineering of two parental phages, WG01 and QL01, which involve changes in the receptor specificity determinant region of gene 37. The recombinant phage possesses the ability to lyse four additional hosts compared with the parent phages, with significant inhibition of plankton Escherichia coli strains and biofilm formation. The combination of phages with other therapeutics, such as antibiotics and depolymerase enzymes, has also been successfully deployed as treatment alternatives for drug-resistant bacteria and biofilm prevention/disruption.^98–100 One major limitation of phage therapy is the lack of standard protocols. The improvement of the production process, quality assurance, validation of products, and carefully controlled clinical studies are necessary to establish its efficacy in treating MDR pathogens. Other setbacks are the possible degradation of therapeutic phages by environmental stressors, decreased activity due to immune responses, and risk associated with the evolution of strains resistant to bacteriophages, among others.^101,102 While the future of bacteriophage therapy focuses on the expansion of the scope of phage and their associated enzymes, a clearer understanding of phage ecology and evolution would contribute immensely to the establishment of phage therapy in combatting AMR and further exploration of resistance development mechanisms.

Chemically synthesized nanoantibiotics (nAbts)

Nanostructured materials have been used effectively to combat microbial resistance.¹⁰³ Some nanostructured materials have antimicrobial potential themselves, while they can also be used as delivery vehicles to convey antimicrobial agents to target points of action. The most commonly utilized nanomaterials are nanoparticles (NPs), which are ultrafine particles with sizes ranging between 1 and 100 nm. A number of them have antimicrobial potential and can inhibit biofilm formation and other virulence processes in pathogens. Their unique efficacy in drug delivery is primarily because of their ultra-small size and very large surface-to-volume ratio.³ Silver and gold NPs are the most explored materials due to their inertness and non-toxicity.

NP-conjugated drugs or nAbts enter the host cell by endocytosis or via interactions with membrane lipids, giving them an advantage over conventional antimicrobials the entry of which into the cell may be hampered by several factors. Also, conjugating drugs to NPs offers the possibility of polytherapy by loading more than one antibiotic on the carrier, presenting highly complex antimicrobial mechanisms against MDR bacteria, thus reducing the possibility of resistance development.¹⁰⁴ Other benefits of employing NPs as drug carriers include targeted drug delivery, sustained drug release, improved bioavailability and enhanced solubility.¹⁰⁵ Mechanisms of NPs’ antimicrobial activity include direct interaction with microbial membrane and cell wall, induction of the host’s innate and adaptive immune responses, reactive oxygen species generation, dispersion of microbial biofilms, QS inhibition, efflux inhibition and plasmid curing.^106,107 Numerous reports have been published on the efficacy of nAbts in the treatment of MDR pathogens.^108–111 For instance, Yang and colleagues¹¹² reported the synthesis of broad-spectrum biodegradable polycarbonates with the potential for targeting intercellular pathogens such as Mycobacteria tuberculosis, which causes TB.

Although nAbts hold great potential for combatting MDR pathogens, toxicity and large-scale production sophistication remain the two major challenges to their therapeutic applications.¹¹³ Moreover, there remains the problem of a lack of unified standards on the antibacterial mechanism of NPs and most available information on the efficacy of nAbts is from in vitro studies, which cannot fully simulate in vivo situations. Likewise, an optimum dosing regimen is lacking for most nAbts, while in vitro concentrations causing cell damage are unrealistically high for animal or human applications. Besides, the non-biodegradability of the NPs into benign substrates leads to their prolonged retention in tissues, causing amplified toxic effects.^{61,62,114,115}

Application of CRISPR gene-editing technology in AMR

CRISPR-Cas is an adaptive immune system by which prokaryotes protect themselves from viral invasion. The detection of the CRISPR system in recent decades has opened up startling biotechnological opportunities and revolutionized genome-editing biology due to its ability to make site-specific double-stranded cuts in the genome. In its basic form, the CRISPR system consists of a dual RNA-guided trans-activating CRISPR RNA-CRISPR RNA (tracrRNA-crRNA) DNA endonuclease. The Cas protein, also known as RNA-guided (gRNA) DNA endonuclease of the system, allows the prokaryotes to recognize and cut foreign DNA at specific points with high precision. Cleavage proceeds strictly by recognizing a flanking protospacer adjacent motif (PAM) region, formation of an R-loop, a complementary base pairing of the gRNA and the target DNA, Cas–DNA interactions and various conformational changes.¹¹⁶ The CRISPR array is found in approximately 90% and 50% of archaea and bacteria, respectively.¹¹⁷ CRISPR-Cas systems are categorized as either a Class 1 (types I, III, IV) or Class 2 (types II, V, VI) system, based on their Cas gene assortment and the nature of the interference complex.¹¹⁸ Each of the class types is further divided into subtypes with their various associated Cas endonucleases (Figure 3). Three distinct phases have been identified in the natural CRISPR defence system of bacteria. These include: (i) the adaptation phase, involving the acquisition of the spacer sequences; (ii) the expression phase, where the crRNA and the Cas proteins are generated; and (iii) the interference phase, involving the crRNA-guided nucleic acid cleavage.¹¹⁹

Figure 3.

A summary of the classification of CRISPR systems and their associated Cas endonucleases.

The ability to reprogramme the dual tracrRNA-crRNA guide of the CRISPR system by substituting it with a mimicking synthetic single-guide RNA (sgRNA) allows the manipulation of the system as a powerful genome-editing tool. The CRISPR-Cas9 mechanism has been adapted in genetic engineering by coupling the Cas9 protein with synthetic sgRNA, allowing the editing of cells’ genomes by removing or introducing new genes and silencing or activating genes (Figure 4). This procedure has found applications in basic biology research, treatment of genetic disorders, biotechnological product development, crop yield enhancement, production of drought-resistant crops and AMR studies.¹²⁰

Figure 4.

CRISPR-Cas working mechanism and delivery approaches. AAVs, adeno-associated viruses; LVs, lentiviruses; AVs, adenoviruses; HTVI, hydrodynamic tail vein injection.

In a study by Citorik and colleagues,¹²¹ cytotoxicity was successfully induced in quinolone-resistant E. coli strains by deploying the CRISPR-Cas technology. The procedure involved designing a CRISPR-Cas complex designated ΦRGNgyrAD87G, which was specifically cytotoxic to quinolone-resistant E. coli strains that harbour the mutation gyrAD87G but not the WT with the gyrA gene. Interestingly, the study also engineered a modified complex, ΦRGNeae, to target intimin, a virulence factor encoded by the eae gene in E. coli O157:H7. The versatility of the procedure was demonstrated by a 20-fold reduction in viable cells when enterohaemorrhagic E. coli O157:H7 was treated with the CRISPR complex. In a similar study by Yosef et al.,¹²² engineered phages were used to deliver sequence-specific CRISPR constructs in order to resensitize antimicrobial-resistant bacteria. The CRISPR complex targeted the conserved region involved in the synthesis of β-lactamases, NDM-1 and CTX-M-5. Their results showed that all bacterial strains harbouring the pNDM and pCTX plasmids became sensitive to streptomycin after the treatment. They concluded that the CRISPR-Cas construct, together with engineered lytic phage, can be used for the simultaneous reversal of MDR, decrease HGT, and enrich for bacterial populations that are sensitive to the respective antimicrobials.

The Class 2 subtype II CRISPR systems are the most commonly studied and applied for gene editing in the elimination of AMR, mainly because of their relatively simple structure allowing easy manipulations.¹²³ An implementation of this technology in AMR elimination or restoring drug sensitivity involves designing a genomic gRNA that targets the resistance genes of the resistant strain. The next stage involves the development of an efficient delivery vehicle for the system, which is currently a major challenge limiting the full exploration of the CRISPR-Cas system potential. A number of carriers have been studied as vehicles for the delivery of the designed CRISPR-Cas system to the target sites. These can be broadly divided into phage vectors and non-phage vectors. The non-phage vectors are further classified as physical (including electroporation, ultrasonic microbubbles, hydrodynamic tail vein injection, laser and microinjection) or chemical (lipofection, nanoparticle and cell-penetrating peptides) methods of delivery (Figure 4).^112,124 Each of the various delivery methods presents its own advantages and disadvantages. The major problem with the physical methods is the limitation of the application for in vivo gene-editing applications, which have largely been limited to laboratory studies. The phage vector carriers present higher efficiency of delivery but have potential oncogenic characteristics, and can sometimes cause insertional mutation in the genome. The non-phage carriers often have less efficiency for gene-editing applications.^112,125 Currently, the CRISPR-Cas system is very limited in clinical applications, and has mostly been used at ex vivo level, particularly in stem cell research, because of the safety and low efficiency of the methods of delivery. Continuous research and development of the delivery approaches will assist in the realization of the full potential of the CRISPR-Cas technology.

Conclusions and future considerations

The rapid emergence of resistance against established antimicrobials, including those considered ‘drugs of last resort’, currently outpaces their efficacies for therapeutic applications. As discussed in this review, the continuous development of innovative strategies to mitigate the challenges of AMR is crucial. An integrated approach coupled with expanded exploration of the various strategies (Figure 5) holds promise in revolutionizing the diagnosis, treatment and prevention of MDR-related diseases. Optimized clinical outcomes can be achieved particularly through the synergistic combinations of these strategies or in augmentation of current treatment procedures. Moreover, there is urgent need to overcome the challenges between diagnosis and treatment of infectious diseases associated with resistant pathogens.

Figure 5.

Integration of strategies to combat AMR proliferation.

To conclude, although the various strategies discussed have great prospects at curbing the menace of AMR proliferation, knowledge gaps still exist between available data, mostly from simulated laboratory experiments and actual field applications of the procedures. In light of this, continued research and development in the molecular biology of AMR is paramount to further understand, restructure, advance and make many of these strategies available commercially. While this is ongoing, antimicrobial stewardship remains crucial in our battle against the development and spread of MDR pathogens in the environment and clinical settings.

Funding

This review paper is self-initiated.

Transparency declarations

None to declare.

Author contributions

M.A. conceptualized the review, and did the writing, organizing, editing and referencing of the manuscript. A.O. supervised M.A. and edited the manuscript.