Antipathogenic properties and applications of low-dimensional materials

Abstract

A major health concern of the 21^st century is the rise of multi-drug resistant pathogenic microbial species. Recent technological advancements have led to considerable opportunities for low-dimensional materials (LDMs) as potential next-generation antimicrobials. LDMs have demonstrated antimicrobial behaviour towards a variety of pathogenic bacterial and fungal cells, due to their unique physicochemical properties. This review provides a critical assessment of current LDMs that have exhibited antimicrobial behaviour and their mechanism of action. Future design considerations and constraints in deploying LDMs for antimicrobial applications are discussed. It is envisioned that this review will guide future design parameters for LDM-based antimicrobial applications.

Antimicrobial resistance is a growing global problem and low dimensional materials have emerged as a potential solution. Here, the authors review the progress which has been made on low dimensional antimicrobials looking at the materials synthesis, modes of action and currently applications.

Introduction

Antimicrobial agents play a vital role in the prevention of pathogenic bacterial and fungal infections, accounting for a predicted 95% reduction in mortality rates since their introduction in the late 1940s^1–4. However, in recent times, over-prescription and misuse of antibiotics in medical, veterinary and agricultural industries, combined with the rapid evolution of microbial species, has contributed to the emergence of antimicrobial resistance (AMR) in pathogenic microbes^1,2. Globally, single and multi-drug resistant (MDR) microbes are responsible for ~700,000 deaths each year¹. The World Health Organisation (WHO), has named several concerning drug resistant, pathogenic bacterial strains prevalent in global healthcare settings, including MDR tuberculosis (MDR-TB), carbapenem-resistant Enterobacteriaceae (CRE), methicillin-resistant Staphylococcus aureus (MRSA) and MDR Neisseria gonorrhoeae^1,3,4. Resistant fungal strains, including MDR Candida auris, are also becoming increasingly resistant to all available antifungal treatments². Without the development of new antimicrobial agents, the WHO estimates that by 2050, approximately 10 million people will die annually from previously treatable infections⁵. The predicted economic impact of AMR suggests it could be worse than the 2008 global financial crisis⁴, costing over US$100 trillion in healthcare⁵.

In response, considerable research has been conducted in an attempt to combat these infections^6–9. Unfortunately, the development of new antimicrobial drugs has faltered in recent years, with only two new classes of antibiotics receiving approval from international regulatory agencies in the past 20 years⁴ with neither being effective against Gram-negative bacteria. This has driven research towards the development of alternative antimicrobial nanomaterials including polymers⁶, particles⁷ and surface coatings⁸. One of the first nanomaterials that demonstrated high antipathogenic abilities is silver nanoparticles and has subsequently received an extensive amount of attention in the research community^10,11. More recently, low-dimensional materials (LDMs) have gained significant interest as potential antimicrobial agents^12,13. These materials are categorised as zero-dimensional (0D)¹⁴, one-dimensional (1D)¹⁵ or two-dimensional (2D)¹⁶ nanomaterials based on their respective dimensionality. These materials have demonstrated favourable properties that have led to their use in electronics and sensing^17,18. A more comprehensive understanding of the underlying properties of LDMs, in conjunction with improved fabrication methods has led to the development of LDMs for new applications, including biomedical^19,20 and antipathogenic technologies^15,21. LDMs that have demonstrated antimicrobial activity across a range of dimensions include 0D black phosphorous (BP)²¹; 1D zinc oxide (ZnO)²² and 2D molybdenum disulphide (MoS₂)²³; among others^13,24,25. The antimicrobial mechanisms of LDMs have not been clearly elucidated, and there remain conflicting opinions in the field, in part due to varied experimental designs and multifaceted antimicrobial mechanisms^26,27. It is thought, however, that the antimicrobial activity can occur through both physical and chemical means, or in combination. Importantly, it has been suggested that LDMs can limit the potential development of microbial resistance, while also enabling control over the antipathogenic mechanisms through tuning the physico-chemical properties of the materials, such as size, shape and composition, including heterostructures and the addition of functional groups^13,28.

This review focuses on the current understanding of the antimicrobial mechanisms for LDMs and highlights areas which require further investigation. Initially, the antimicrobial mechanisms, including passive and stimuli-activated actions of LDMs, will be summarised. Following this, elemental and compound LDMs will be reviewed, focusing on graphene and graphene analogues, metal oxides (MOs), transition metal dichalcogenides (TMDs) and early transition metal carbides, nitrides and carbonitrides (MXenes) classes of materials. Finally, recent advancements in combined LDMs either as heterostructures or composite forms and their tunability in terms of antimicrobial activity will be analysed. The review concludes with a discussion of future design strategies for single and composite LDMs to optimise antimicrobial efficacy for next-generation treatments.

LDMs—classifications and types

LDMs can be classified as 0D, 1D, or 2D depending on their size and aspect ratios²⁹. A visual summary of LDMs and their associated antimicrobial mechanisms are shown in Fig. 1. Some properties of common 0D, 1D and 2D materials are shown in Table 1 and their atomic structure and morphology are shown in Fig. 2. Common LDMs include carbon, graphene (Gr), graphene oxide (GO), graphitic carbon nitride (g-C₃N₄), BP, Boron nitride (BN), 2D hexagonal BN (hBN), MoS₂, titanium carbide (Ti₃C₂T_x) MXene and MO such as ZnO, titanium oxide (TiO₂) and copper oxide (CuO).

Fig. 1. The contrast of 0D, 1D and 2D materials.

Common structures for each dimensional classification and their general antimicrobial mechanisms.

Table 1.

Properties of common LDMs.

		Stable in watera	Stable in aira	Bandgap (eV)	Approximate zeta potential (mV)	Antibacterial activity
0D	Carbon	×b77	✓157	1.577	−2032,77	✓32
	BP	×b126	×158	0.5–1.1158	−10159	✓21
	BN	✓82	✓157	6.5160	−50161	NA
	g-C3N4	✓162	✓163	2.7164	−40165	✓123
	MoS2	✓166	×167	3.9–4.2157	−20168	✓74
	Ti3C2Tx	×169	×169	>0.183	−60170	NA
	ZnO	✓79	✓171	3.413	−4079	✓78
	TiO2	✓172	✓172	3.2129	N/A	✓136
1D	Carbon	✓173	✓173	0.5174	−30175	✓56
	BP	NA	✓18	2.6176	NA	NA
	BN	✓177	✓177	4.5174	−10178	✓90
	g-C3N4	✓163	✓163	2.8179	−30180	NA
	MoS2	✓181	✓182	2.5183	NA	NA
	CuO	✓39	✓39	3.2184	−2019	✓39
	ZnO	✓22	✓135	3.313	−15185	✓109
	TiO2	✓186	✓187	2.536	−30188	✓12
2D	Gr	×189	✓190	0191	−40192	×[b154
	GO	✓190	✓190	2.2193	−3023	✓97
	BP	×130	×194	2.0130	−3045	✓45
	hBN	×195	✓20	5.733	−35192	✓16
	g-C3N4	✓163	✓163	2.1163	−45196	✓99
	MoS2	✓49	×197	1.8198	−35192	✓47
	Ti3C2Tx	×169	×169	0.1–2147	−4023	✓118
	ZnO	✓94	✓94	3.413	−20199	✓94

^aStable over 7 days.

^bRequires functionalization.

Fig. 2. Visual summary of common LDMs.

0D The chemical structure (top), high-resolution transmission electron microscope (HRTEM) image (inset is corresponding selected area electron diffraction (SAED) pattern where applicable) (middle) and atomic force microscopy (AFM) scan (bottom) for carbon dots^77,200, BP QDs²⁰¹, BN QDs²⁰², g-C3N4 QDs¹²³, MoS₂ NDs¹⁶⁸, Ti₃C₂T_x QDs⁸³, ZnO QDs^133,203 and TiO₂ QDs^129,204. 1D The chemical structure (top), HRTEM image (inset is corresponding SAED pattern) (middle) and AFM scan or scanning electron microscope (SEM) image (bottom) for carbon NWs²⁰⁵, BP NWs¹⁸, BN NFs²⁰⁶/NTs²⁰⁷, g-C₃N₄ NRs^179,208, MoS₂ NWs^91,209, CuO NRs⁸⁶, ZnO NWs^22,210 and TiO₂ NWs¹²/NTs²¹¹. 2D The chemical structure (top), HRTEM image (inset is corresponding SAED pattern) (middle) and AFM scan (bottom) for graphene (Gr) NSs¹⁹², GO NSs^212,213, BP NSs¹⁹⁴, hBN NSs¹⁹², g-C₃N₄ NSs²¹⁴, MoS₂ NSs¹⁹², Ti₃C₂T_x NSs^43,215 and ZnO NSs²¹⁶.

0D materials

Zero-dimensional materials (0D materials) are confined to the nanoscale in all three dimensions and are composed of only a few atoms, resulting in an average size of less than 10 nm and uniform (i.e. isotropic) properties³⁰. Quantum dots (QDs) are crystalline clusters of semiconductor materials³¹, while nanoclusters and nanodots (NDs) consist of a wider range of materials under 10 nm^30,32. 0D materials have gained popularity due to their versatile size-dependent optical and physicochemical properties^14,30, along with the possibility for biomedical applications, such as bioimaging^33,34. Several advantages include tuneable oxidation potential³⁵; light-induced free radical generation³⁰; and high photothermal efficiency¹⁴.

1D materials

One-dimensional (1D) materials are confined to the nanoscale in two dimensions while the remaining dimension is typically in the microscale. These materials form cylindrical structures with a diameter ranging from 10 to 100 nm and generally are less than 12 µm in length^12,15. Nanorods (NRo) are characterised as solid structures with a low length to width aspect ratio in comparison to nanowires (NWs), nanotubes (NTs), nanoribbons (NRi) and nanofibers (NFs), which have a high length to width aspect ratios^36,37.  As the terms NWs and NFs are often used interchangeably, in this review, we classify NWs as structures composed of inorganic or metallic materials, whereas we define NFs as composed of organic materials^36,38. NTs possess a hollow centre, typically with an internal diameter of under 100 nm and an external diameter up to 1 µm, which results in the highest surface area for 1D materials³⁷. It should be noted that these NT dimensions are not absolute, and examples of larger materials being referred to at NTs in the literature are common. Unlike 0D materials, 1D materials are anisotropic as many of their physicochemical properties are non-uniformly distributed. Some favourable properties of 1D materials, include a larger/higher relative surface area, and tuneable oxidative stress and catalytic ability^22,36,39. Within biological systems, 1D materials have been investigated for potential anticancer applications⁴⁰ and other biomedical application^41,42.

2D materials

Two-dimensional (2D) materials are only confined to the nanoscale in one dimension, and the other two dimensions typically span into the micro to centimetre scale^43,44. They form nanosheets (NSs), also called nanoflakes, which can have more random shapes⁴⁵. When materials have a more uniform and controlled configuration with straight edges, they are called nanoplatelets (NPs), or nanodisks if they have rounded edges²⁰. Both NSs and NPs have a thickness under 3 nm (1–2 atomic layers)^16,45. Like 1D materials, 2D materials are also anisotropic, with tuneable properties, such as photothermal behaviour⁴⁶, hydrophobicity⁴⁷ and stability in a range of chemical environments^48,49, allowing 2D materials to be widely investigated for biomedical applications. Few-layered materials are formed when multiple atomic layers are stacked, and generally confined to a thickness of 3–50 nm thick, while their lateral area is typically under 1 cm^2 27,49.

Mechanisms of antimicrobial activity for LDMs

The antimicrobial mechanism of LDMs is multifaceted, with the precise mechanism of action varying between materials, systems and applications. This has important implications for the design, application and deployment of antimicrobial LDMs, and their use as viable treatments. To this end, it is important to classify the various modes of action to unlock their full potential as antimicrobial agents. Further, LDMs can be non-specific, meaning that potential synergistic or antagonistic effects must be assessed.

Within existing literature, there is contention surrounding the antimicrobial mechanisms of LDMs. This is due to the complexity of the system as a whole, with interrelated and often simultaneous modes of action, the infancy of the field, and current experimental limitations often causing confusion²⁸. As such, several mechanisms have been proposed, supported experimentally and theoretically, with both positive and negative correlations often being described. For clarity, we have assessed the currently reported LDM antimicrobial mechanisms and separated them into seven broad categories (Fig. 3).

1. Basal plane effects: Possessing a large surface area and amenability to chemical functionalisation, the basal plane of LDMs is believed to contribute to their antimicrobial activity. 2D materials have been observed to “wrap” microbial cells^50–52, which is thought to inhibit nutrient/waste exchange⁵² and/or cause perturbations in the integrity of the cell membrane through hydrophobic interactions^51,53, or charge transfer effects⁵⁴; all cases resulting in cell lysis. The basal planes of 2D materials can also interact with intracellular lipids, proteins and nucleic acids, through π−π stacking, hydrogen bonding and electrostatic adsorption²⁸. 0D⁵⁵ and 1D^56,57 nanomaterials have also been observed to demonstrate antimicrobial activity resulting from interactions with the basal plane.

2. Edge effects: The edge of LDMs possess unique physicochemical properties when compared to their respective bulk materials, which contribute to their antimicrobial action. Both 1D^56,58 and 2D^23,50 materials have been shown to puncture cell walls of microbial species upon contact, resulting in the leakage of intracellular components, and in-turn cell lysis. Experimental and theoretical studies demonstrate that 2D materials can translocate through the membrane in an orthogonal mode of penetration^59,60, or form membrane-nanosheet sandwich structures^59,61, due to different physico-chemical properties such as size and the degree of oxidation. These effects are not targeted and therefore, can pose a threat to mammalian cells.

3. Extraction of phospholipids from the membrane: Following an initial interaction with the edge of 2D materials, “nanoscale de-wetting” can occur due to strong hydrophobic interactions between the basal plane and the phospholipids, resulting in eventual membrane collapse^16,62,63. Interestingly, 1D materials have been shown to exhibit similar phenomena on membrane models⁶⁴ however, studies on microbial cells have yet to be conducted.

4. Oxidative stress: Reactive oxygen species (ROS) can cause damage to important cellular components including proteins, nucleic acids and phospholipids, resulting in cell lysis⁶⁵. ROS-independent oxidative stress can occur naturally in microbes, and interaction with or internalisation of LDMs can dramatically increase the intracellular ROS⁶⁶. Conversely, LDMs can generate ROS in an aqueous media, which then interacts with the microbial cells^25,67. For LDMs, the typically generated ROS species include singlet oxygen (¹O₂), superoxide anion radicals (^•O₂^–), hydroxyl radicals (^•OH) and hydrogen peroxide (H₂O₂). A summary of redox potentials for the generation of these ROS at physiological pH are shown in Supplementary Table 1 and referenced in Fig. 3.

5. Light-driven mechanisms: This includes photocatalytic activity^31,68, in which light, typically at a wavelength from ultraviolet (UV) to visible (Vis), drives the production of ROS which causes cell lysis. Alternatively, photothermal activity can be generated^25,46, typically by light in the Vis-near infrared (NIR) wavelength region, which can cause a localised temperature increase which inactivates the microbial cells through membrane damage and the denaturation of enzymes⁶⁹.

6. Synergistic effect: Combing LDMs can alter properties of materials, such as improved stability or enhanced antimicrobial activity^45,70. One way the antimicrobial activity can be improved is in the form of releasing of ions or molecular species, also known as degradative effects⁷⁰. Altering the materials properties, such as zeta potential, can promote new antimicrobial mechanisms^21,45.

7. Other physical interactions: When 1D and 2D materials come into contact with microbial cells, they are able to physically damage the membrane^23,71. This can be in the form of wrapping around, encapsulation or entrapping the cell, inducing cell lysis⁵⁶. This is distinct from edge effects, which rely on either the electrical interaction of LDMs or physical damage, which occurs at the apex of an atomically sharp material.

   While these categories broadly cover the reported antimicrobial mechanisms of LDMs, these mechanisms often work simultaneously, and the relative importance of each is still debated. Broadly, the precise antimicrobial efficacy is likely to depend on: (1) biological factors, such as cell type, size and shape, (2) nanomaterial factors, such as size, shape and chemical functionality and (3) environmental factors, including, temperature, pH and components in the media such as proteins. Additionally, LDMs based nanocomposites have been developed which demonstrate multifaceted antimicrobial mechanisms and with enhanced treatment properties^6,70.

Fig. 3. Summary of the most accepted antimicrobial mechanisms of LDMs.

A more detailed explanation of the mechanisms provided in antimicrobial mechanism section.

Fabrication of LDMs

The fabrication methodology of LDMs can also be used to manipulate their edges and basal planes. In general, synthesis of LDMs can be placed into two main categories: (1) bottom-up or (2) top-down methodologies. Bottom-up approaches are characterised by growing/synthesis from chemical precursors using a range of procedures and are more commonly used for metal oxides and 1D structures^72,73. Alternatively, top-down synthesis occurs when a bulk material is broken down into the desired nanostructure and are typically used for 0D and 2D materials^25,74.

Suspension-based LDMs as antimicrobial agents

An effective and common method of utilising the antimicrobial properties of LDMs are as suspension-based approaches. LDMs can be suspended within a range of liquid medium and allow for the free movement of LDMs along with providing a possible environment for the generation of ROS. This section will highlight example LDMs in suspension with antimicrobial properties. Figure 4 shows common interactions of selected suspension-based LDMs and microbial cells to provide an overview for the reader.

Fig. 4. Summary of antimicrobial interactions of LDMs in suspensions.

a Interaction of suspended Copper sulphide (CuS) NDs with Escherichia coli (E. coli)⁷⁵, carbon NTs wrapping around Lactobacillus acidophillus (L. acidophillus)⁵⁶ and tungsten disulphide (WS₂) NSs with E. coli¹⁰⁶. b The extraction mechanism of membrane phospholipids by MoS₂ NSs⁶³. c Generation of ROS via degradation of BP NSs over 120 min (left)²¹⁷ and the interaction of Ti/BP NSs film with S. aureus (right) with cellular damage indicated with red arrows⁶. d Photocatalytic activation of MoS₂ QDs against Staphylococcus aureus (S. aureus)⁷⁴ and e the photothermal activation of indium(III) selenide (In₂Se₃) NSs against E. coli⁴⁶. f Carbon NWs wrapping around L. acidophillus⁵⁶.

0D materials

Suspension-based 0D materials have several reported antimicrobial mechanisms, including cellular uptake⁷⁵, radiation stimulus⁷⁴ and ROS generation³¹. Due to their innately small size, 0D materials can easily transverse microbial cell membranes, and enter the intracellular space (Fig. 4a). This provides a unique capability to 0D materials, which is often not observed for 1D and 2D materials due to size constraints. Once inside the pathogenic cell, 0D materials can disrupt internal cellular processes (metabolic process, deoxyribonucleic acid (DNA) transcription, respiration, etc.) and also damage internal cellular material (DNA, ribosomes, and plasmids, etc.)⁷⁶. The precise mechanism is dependent on the specific LDM, but 0D materials can rapidly lead to cell lysis following cellular uptake.

Further, 0D materials have also shown potential for photo-induced antibacterial treatments (Fig. 4d, e). Semiconductor QDs with a bandgap greater than 3.1 eV have demonstrated non-specific toxicity towards both mammalian and microbial cells when illuminated with UV light⁷². However, other semiconductor QDs are typically less toxic towards mammalian cells, hence more biocompatible, when excited under visible or NIR light²¹. Regarding the microbial toxicity, materials including carbon NDs (CDs)⁷⁷, BP QDs²¹, cadmium telluride (CdTe) QDs³⁵, MOs QDs⁷⁵ and MoS₂ QDs⁷⁴, can generate excess ROS under visible, NIR and UV light irradiation. Metal oxides, such as ZnO QDs^78,79 and vanadium oxide (Vox) QDs^80,81 can also release ions under UV and ambient light irradiation. Tailoring the dimensions tunes the electronic structure of these LDMs to manipulate the wavelengths of light that activate the particles. Bare copper sulphide (CuS) QDs have three reported mechanisms, photothermal effects, Cu²⁺ ion release and photodynamic generation of ROS under NIR⁷⁵. These examples have demonstrated the importance of light in the antimicrobial mechanism of 0D materials, indicating the potential use in photodynamic therapy. Conjugating 0D materials with other LDMs can synergistically enhance the antibacterial mechanisms, by enhanced ROS production⁷⁸ or light assisted drug delivery²¹.

Research into 0D materials as antimicrobial agents is relatively recent, and several materials have potentially favourable properties. BP QDs have recently been used for bioimaging and drug delivery applications, demonstrating biocompatiblility⁸². Larger TiO₂ NPs have demonstrated ROS generation, but research into QDs is lagging. While MXene QDs, including titanium nitride (Ti₂N), Niobium carbide (Nb₂C) and Ti₂C₃T_x, have shown photothermal properties, with possible uses in cancer treatments and bioimaging⁸³, to the best of our knowledge, no tests have been so far conducted into their potential use as antimicrobials. Further, bare WS₂ QDs had no antimicrobial impact, but when conjugated with antimicrobial peptides, it was effective against both Pseudomonas aeruginosa (P. aeruginosa) and C. albicans biofilms⁸⁴. This highlights the potential for more 0D materials to potentially have antimicrobial properties, or to have induced action through conjugation, but further exploration into these materials is needed.

1D materials

Similarly, research into antimicrobial 1D structures, including carbon NTs⁵⁶ and a range of MOs^13,29, is increasing. There are different antimicrobial mechanisms that have emerged in 1D materials (Fig. 4a), such as carbon NTs, deactivate microbial cells through physical damage, which is heavily influenced by size⁷¹. For thinner single-walled NTs (SWNTs), the main mechanism is to puncture the bacterial membrane, acting as a nanodart/nanoknife⁷¹. Thicker, multi-walled NTs (MWNTs) tend to be less effective⁷¹ and predominantly wrap around bacteria⁵⁶ (Fig. 4f). Both SWNTs and MWNTs are potentially toxic against a range of microbes found in the human gut, including E. coli, S. aureus and Enterococcus faecalis (E. faecalis)⁵⁶. Alternatively, MOs NRs and NWs mainly generate ROS to chemically induce cell lysis⁸⁵. Interest into MOs has mainly focussed on ZnO⁸⁵ and CuO⁸⁶, however recently other MOs, such as cerium(IV) oxide (CeO₂)²⁹, magnesium oxide (MgO)⁸⁷, maghemite (Fe₂O₃)⁸⁸ and nickel(II) oxide (NiO)⁴⁰, have demonstrated promising antimicrobial properties. ZnO materials can also release Zn²⁺ ions, along with ROS, to aid in the damage of microbial cells⁸⁵, indicating the possibility that LDMs with non-physical modes of action could have antimicrobial potential.

There are several materials that have been shown to possess antimicrobial properties in their 2D form, but there is little research into their corresponding 1D forms. One material with such potential is graphitic carbon nitride (g-C₃N₄), which generates ROS under light irradiation⁸⁹. Recently, polymer functionalised BN NTs were also shown to have an increased efficacy compared to pristine BN NTs⁹⁰. This further indicated how chemical modifications of 1D materials can enhance the antimicrobial activity to materials with little-to-no inherent efficiency. Other 1D materials have attracted interest for applications including gas sensing¹⁸, catalysis⁹¹ and biomedicine⁹². Recent advancements in fabrication techniques have led to the synthesis of materials like BP NWs and NRis^18,93, which to our knowledge have not been explored for their antimicrobial properties.

2D materials

Recently, 2D materials have received considerable attention due to their relative ease of fabrication and their amenability to tuning their properties to make them responsive to a wider range of stimuli^45,48,94. There are several common antimicrobial mechanisms observed in 2D materials, including physically slicing through the membrane⁴⁹ (Fig. 4a), phospholipid extraction⁶² (Fig. 4b) or generating reactive oxide species⁷⁰ (Fig. 4c).

Graphene analogues

Pure graphene has limited antimicrobial properties, while its oxide derivatives GO and reduced graphene oxide (rGO) have demonstrated broader antimicrobial capabilities^62,95. Both GO and rGO can deactivate bacterial cells via oxidative stress⁹⁶ and piercing the membrane⁵³, along with potential destructive extraction of the phospholipids comprising the membranes⁶². When GO was incorporated into an injectable hydrogel, its high photothermal properties and increased conductivity lead to improved antimicrobial activity against E. coli and MRSA⁹⁵. It should be noted that the antimicrobial activity of graphene, GO and rGO remains controversial. A recent paper suggested that pure GO possess no inherent antibacterial character, but rather the widely reported behaviour is due to impurities⁹⁷.

Other graphene analogues also display antimicrobial properties. BP and hBN NFs have demonstrated the previously mentioned common antimicrobial mechanisms against E. coli and S. aureus^25,45. Similar to GO, it has been speculated that hBN NSs cause membrane stress through lipid extraction¹⁶. BP NSs generate ROS under NIR light irradiation²⁵, which can be enhanced when combined with larger nanoparticles⁷⁰.

Graphitic carbon nitride

Another 2D carbon-based material with antimicrobial potential is g-C₃N₄. Pristine g-C₃N₄ can effectively be used to treat against E. coli, MRSA and Bacillus anthracis spores, under visible or UV light irradiation conditions⁹⁸. The functionalization of the g-C₃N₄ surface using nitrogen plasma treatment (N-g-C₃N₄) can increase the activity without requiring light activation⁹⁹. The primary antimicrobial mechanism of pristine g-C₃N₄ is via the photoactivated generation of free radicals⁹⁸, while N-g-C₃N₄ is thought to induce cell death mainly through the interaction with phospholipids within the cell membrane⁹⁹.

Metal oxides

2D MOs such as ZnO⁹⁴, TiO₂¹², CuO¹⁰⁰ and MgO²⁷ have gained attention for effectiveness against a wide range of pathogenic microbes. The main antimicrobial mechanism used by MOs is the photo-induced generation of ROS^100,101. Under visible light, MOs are effective treatments against S. aureus, E. coli and Staphylococcus epidermidis (S. epidermidis)^27,102, as well as under UV light⁹⁴. Some MOs including ZnO and CuO release cations that electrostatically disrupt the membrane¹⁰².

Transition metal dichalcogenides

TMDs also generate ROS and damage cellular membranes⁴⁹. Recent studies have focused primarily on MoS₂ and WS₂ as potential antimicrobials as they are non-toxic¹⁰³. The antimicrobial mechanism for MoS₂ is multifaceted, with the primary mechanism involving the generation of superoxide anions⁶⁷, combined with slicing the membrane⁴⁹ and binding to peptide backbones¹⁰⁴. Alternatively, WS₂ primarily deactivates cell through membrane damage, not through ROS^105,106. The potential antimicrobial properties of other TMDs remain to be explored.

Other 2D materials

Several other 2D materials have shown potential for antimicrobial applications in limited studies but require further research. Preliminary studies into MXenes NSs, mainly Ti₃C₂T_x, and indium(III) selenide (In₂Se₃) have demonstrated activity against a limited number of Gram-positive and Gram-negative bacteria^24,46,50. The sharp edges of the NSs play a key role in the degradation of the cellular membrane²⁴. The photothermal properties of In₂Se₃ can also be used to increase the antimicrobial efficacy⁴⁶.

As new LDMs are developed for use within other applications such as electronics or sensing, their potential as possible antimicrobial agents should be explored. Enhancing the antimicrobial capabilities of LDMs though heterostructuring will be discussed in a later section.

Surface coated LDMs as antimicrobials

Planar surfaces

LDMs can be used to either generate or boost antimicrobial activity of a surface. For 1D and 2D materials, the orientation of the LDMs can influence the antimicrobial properties and affect the action of the primary antimicrobial mechanism (Fig. 5). The main orientations are vertical arrays²³, horizontal coatings¹⁰⁷ or randomly oriented arrays²⁶. These arrays and coatings can be used for implants⁴¹, wound dressings¹⁵, and other medically relevant surfaces¹².

Fig. 5. Summary of antimicrobial interactions with surfaces modified with LDMs.

a Membrane damage of E. coli from TiO₂ NWs⁸ and Cryptococcus neoformans from ZnO NWs (top and side view)²². b Various orientations of GO NSs on a surface and their interaction with E. coli¹¹⁰. c A multi-functional CuO NWs mesh with a superhydrophobic surface (left) which prevent E. coli and S. aureus growth (middle) and microbial adhesion after 5 days of incubation in domestic sewage water (right)³⁹.

Vertical arrays of LDMs are often reported for antimicrobial capabilities, through chemical activity and improved membrane damage, derived from their sharp edges (Fig. 5a, b)^12,108. 1D MOs have gained substantial attention as surface modifiers recently, with studies highlighting ZnO²², TiO₂¹² and CuO¹⁰⁸ arrays as potential antibacterial and antibiofilm surfaces, as well as a possible antifungal treatment²². When MOs are grown in vertical arrays, there is a synergistic effect between the physical puncturing of the cellular membrane, the generation of ROS^22,108, and the electrostatic interaction of the positive metal ions and the negatively charged membrane¹⁰⁹. Combining 2D and 1D structures in an array can enhance photoactivated antibacterial properties by altering the electronic band structures of the materials. When rGO NSs are combined with CuO NWs, the CuO injects an electron into the rGO, allowing for enhanced ROS generation under visible light irradiation¹⁰⁸. A recent study found that a higher edge density of an array can improve the antimicrobial activity, whilst also supressing any mechanisms of the substrate, such as wrapping²³. The morphology of the LDMs in the vertical array influences the antimicrobial efficacy; however, more research is needed to identify they key parameters for optimisation.

Similarly, randomly orientated LDMs (Fig. 5b) are used to enhance the antimicrobial properties of a surface. These materials are still able to penetrate cellular membranes through edge effects; however, the number of edges in the preferential orientation is reduced^26,110. Randomly orientated GO NSs effectively reduced a population of E. coli by 25%, compared to 44% for vertically aligned NSs¹¹⁰. Although the main mechanism for GO NSs is still being investigated, the decreased efficacy of random NSs was attributed to the decreased penetration of the cellular membrane¹¹⁰. Randomly orientated 1D MOs NWs can influence the hydrophobicity of the surface, leading to a two-stage mechanism (Fig. 5c). The more tightly packed surfaces increase the hydrophobicity, preventing cellular adhesion. Once cells begin to settle, the NWs can puncture the membrane, whilst the chemical activity of the MOs NWs also provide longer-term antimicrobial action²⁶. Initial research has been conducted into the relationship between material angle, microbial adhesion and membrane penetration. As fabrication methods continue to advance to allow for greater control of the angle, this relationship can be explored in more detail to allow for greater optimisation.

Although still capable of antimicrobial action, horizontal or planar LDMs (Fig. 5b) are not as effective as the previously mentioned orientations. When deposited horizontally, LDMs are more reliant on chemical mechanisms, as the physical antimicrobial mechanisms derived from edge effects are limited^15,110. For 1D MOs deposited horizontally, electrostatic interactions of ROS are predominantly used to inactivate cells^15,41. In the case of CeO₂ NRos deposited onto medical-grade titanium, the material was able to reduce biofilm formation of common plaque-forming bacteria by 99%. The main antimicrobial mechanism is believed to be via electrostatic interactions between the CeO₂ NRos and the cell membrane⁴¹. Horizontally orientated TiO₂ NWs coatings rely on ROS, as the NWs are more efficient at photodegradation than larger nanoparticles¹⁵. Horizontally deposited NSs tend to show lower antimicrobial activity compared to vertically orientated NSs^107,110. For example, horizontal GO NSs resulted in a ~20% reduction of E. coli cells, in comparison to 44% for the vertical GO NSs¹¹⁰. MoS₂ NSs have been shown to reduce E. coli growth by roughly 50% after a 3 hour incubation¹⁰⁷.

Surfaces can also be modified with the addition of 0D materials to improve their antimicrobial properties. The antimicrobial activity of 0D material modified surfaces has been investigated across a broad range of medically relevant materials, such as wound dressings^111,112, resins¹¹³, polymers¹¹⁴ and drug delivery systems¹¹⁵. Due to the shape of 0D materials, chemical pathways such as ROS generation¹¹² or electrostatic interactions¹¹⁵ are the main antimicrobial mechanism. Depositing carbon and graphene QDs onto common fabric bandages prevent E. coli and S. aureus growth and promoted wound healing via photo-induced ROS¹¹². When carbon QDs were doped with nitrogen, they were capable of treating MRSA infections to the same degree as the antibacterial drug vancomycin¹¹¹, indicating its potential use in treating vancomycin-resistant Enterococcus infections. Biofilm formation of Streptococcus mutans, a common dental pathogen, had a 99% reduction after exposure to an adhesive dental resin coated with ZnO QDs¹¹³. Further, E. coli and S. aureus growth was reduced by ZnO QDs deposited onto bioactive glass nanoparticles⁷⁹. The antimicrobial action of 0D materials can be increased by depositing multiple materials that have a combined effect. The ROS generation of ZnO can be increased by also depositing cadmium sulphide (CdS) QDs onto the surface and activating using UV light. Combining both QDs with the polysaccharide chitosan prevented both E. coli and Bacillus subtilis (B. subtilis) growth¹¹⁵. Using 0D materials as additives in materials with inherent antimicrobial properties can also increase the effectiveness. Under ambient light, indium-based QDs were not effective against clinical and environmental strains of P. aeruginosa and E. coli. Combing the indium-based QDs with crystal violet increases light absorbance and increases ROS generation to a lethal amount¹¹⁴.

Three-dimensional filters

LDMs can be deployed to improve water filtration to prevent infections from waterborne pathogens. One method to increase the efficacy of water filtration is to increase the hydrophilicity of the filter. This can be achieved by depositing 2D materials such as WS₂⁴⁸ or g-C₂N₄^116,117, or growing 1D materials like Cu NWs³⁹ (Fig. 5c). WS₂ was able to reduce the amount of E. coli and S. aureus in solution by around 90%⁴⁸. Photocatalytic materials such as g-C₂N₄ NSs also generate ROS under visible light exposure and have been shown to cause an over 99.99% reduction E. coli from solution¹¹⁷. Cu NW mesh uses multiple antimicrobial mechanisms, including super hydrophilicity, the generation of Cu⁺ ions and photothermal interactions³⁹. MXene membranes have also demonstrated antifouling properties and were effective against both E. coli and B. subtilis via oxidation and photothermal reactions¹¹⁸.

Another application for LDMs is enhanced filters for use in air purification to reduce the spread of airborne pathogens. Fe₂O₃ NWs grown on an iron mesh to capture common indoor bioaerosols via the generation of hydroxyl radicals⁷³. A single filter was able to capture 52% of airborne S. epidermidis and E. coli, and 5 filters in tandem were able to reduce airborne cells by 98%. Although a promising application, the instability of several LDMs in ambient conditions has limited their potential to be used for air purification.

Hetero-LDMs: current research and perspectives

When dissimilar materials are stacked in layers, they can result in new properties which have typically been used to great effect in electronics and optoelectronics^119,120. The ability to stack materials of desired size/thickness and different phases (solid/liquid) in lateral or vertical heterostructures, allows the manipulation of physical and electronic properties of the material at the atomic scale via quantum engineering¹²¹. The use of LDMs can be considered a win–win solution in the quest for achieving a combination of antimicrobial properties. Combining materials can aid in passivating materials from ambient degradation that otherwise show outstanding antimicrobial properties. These hybrid materials can also be prepared in a solution phase which can be used as surface/spray coatings on the implants using the dispersion technology¹²². The materials used to form heterostructures are usually chosen from the periodic table among the III-V, IV-IVI, II-VI groups. The material systems are stacked, altering their band structures to benefit the desired applications¹¹⁹ (See supplementary document).

Design parameters for LD-based next-generation antimicrobials

Recent advancements in LDMs have highlighted their potential as antimicrobial treatments. These highly antimicrobial materials can reduce the number of microbial cells by over 80% within a few hours. Such materials include: (1) 0D⁷⁷ and 1D⁷¹ carbon, (2) 0D¹²³ and 2D¹²⁴ g-C3N4, (3) 0D CdTe³⁵, (4) 1D metal oxides^29,109, (5) 2D GO¹¹⁰, (6) 2D BP^41,70 and (7) 2D MXene^24,50, which have all exhibited high antimicrobial ability. Most of these materials employ a combination of mechanisms that are influenced by their chemical composition and morphology^27,117. There are many factors to consider when selecting LDMs for a specific antimicrobial treatment or scenario. For instance, is the desired application biological or abiotic in nature? What is the delivery method of the required treatment (injectable, solution, surface, physical, etc)? Does the treatment need to last for a prolonged period, or does it degrade rapidly in response to treatment? Is there a single target pathogen or more-general antimicrobial activity required? Is a chemical approach or physical methodology better for the scenario in question?

To assist with answering these questions, the following section has been designed to give an overview of the currently available antimicrobial LDMs and provide insight into the future directions of research in this area. Notably, the selection of LDMs is not trivial and is dependent on the system, application and specific microbes involved. Furthermore, the application must be adequately tailored to either prevention or treatment. Table 2 summarises the currently measured antimicrobial efficacy, the respective cytotoxicity, relative commercial costs, chemical stability, and the current status of the LDMs previously described for medical, industrial, and scientific-based antimicrobial-based exploitation.

Table 2.

Summary of LDMs for antimicrobial applications^218–263.

^aAt least one Gram-positive and one Gram-negative bacterial cell, and one fungal cell (no fungi tested+).

^bMore than three mammalian cell lines tested.

^cGreen is <5 h, yellow is 5–12 h, orange is 12–24 h and red is >24 h.

^dGreen is < $499, orange is $500–$999, red is >$1000 (estimated using Sigma-Aldrich).

^eNewly fabricated.

When considering using LDMs within a biological system, the biocompatibility with mammalian cells is an essential factor. Table 2 summarises which LDMs have been tested for their effect on mammalian cells, either using cell cultures (in vitro) for 48–72 h^41,125 or live mouse models (in vivo) for over a period of seven days^81,126. If the material has demonstrated non-selective toxicity towards both microbial and mammalian cells, their use in treatment is not desirable.

Utilising the chemical routes for antimicrobial activity

Many LDMs that have demonstrated high antimicrobial activity can generate ROS. The ability to generate ROS in solution is directly linked to LDMs bandgaps and therefore redox potentials, with the bandgaps of numerous LDMs discussed above falling into the range 1.5–3.5 eV (see Table 1). These bandgaps are sufficient to elicit the generation of ROS (Supplementary Table 1) in solution, which can damage the microbial cell wall. Therefore, if the antimicrobial action requires the production of ROS, the use of materials which fall outside of this range is not recommended. LDMs that have demonstrated antimicrobial activity and are capable of generating ROS species in solution include BP⁷⁰, MXene¹¹⁸ and WS₂¹⁰³. However, there are also LDMs that do not generate ROS, but is still possess antimicrobial activity such as 2D hBN¹²⁷, which has limited antibacterial efficiency^90,128. For such materials, forming composites or heterostructures can alter their intrinsic bandgaps, and facilitate the generation of ROS^115,129. In some systems, the generation of ROS can be promoted via the addition of ultra-low concentrations of hydrogen peroxide^81,112.

LDMs which readily degrade into fragments, ions, sub-species and/or ROS can be advantageous for antimicrobial treatments or surface-functionalisation. For example, BP is well known to degrade under ambient atmosphere and solution conditions, producing ROS and P_x ions^130,131. The ability for the material to both degrade and produce antimicrobial species is useful for applications which require the biocidal agent to disintegrate from the treatment zone without removal. Other LDMs degrade at a slower rate, such as In₂Se₃, and have also shown potential antimicrobial activity¹³². This degradation or generation of ROS can also be enhanced using light irradiation^35,124. LDMs that have demonstrated photocatalytic properties can potentially be used to assist with targeted treatments²¹.

Morphological and physical exploitation

The morphology, size, surface charge and flexibility of LDMs can positively and negatively correlate to biocidal enhancement in both suspension-based and surface-based technologies. Broadly, the degree of antimicrobial activity within a system is both species and treatment dependent and represents an interplay between contributing factors. However, the precise nature of the LDM-microbial interactions is still relatively poorly understood, often resulting in conflicting results, even amongst similar systems. The following section aims to collate the current level of understanding of the morphological and physicochemical interactions which facilitate LDM antimicrobial action.

For 0D materials, cellular uptake, electrostatic disruption, and specific cell–surface interactions are the primary physical modes of action. Here, the comparatively small size of the material facilitates cellular uptake, which is often not achieved by larger 1D and 2D materials. This means that the physical size and surface chemistry of the material is a key determining factor. If an application requires intracellular interactions, then 0D materials are prime antimicrobial candidates - smaller 0D LDMs can possess enhanced activity¹³³. Further, 0D materials are known to facilitate intra- and extracellular damage via unfavourable electrostatic interactions. However, we found no reports of 0D LDMs that cause physical-based membrane rupture.

For 1D and 2D materials, the aspect ratio can influence the antimicrobial activity²⁹. For CeO₂ NRs, a higher aspect ratio resulted in more active sites on the NR surface, generating a higher concentration of hypobromous acid (HOBr), increasing the antimicrobial activity²⁹. The aspect ratio can also be linked to the cytotoxicity of a material²⁹. For some materials, a higher aspect ratio increases the toxicity towards human cells²⁹. 1D CuO NW mesh was found to be superhydrophobic and prevented bacterial adhesion³⁹. For 2D sheets, the edges of the material are known to be “sharp” and induce microbial cells damage upon cell-on-edge adsorption.

Several forces are at play when LDMs and microbes interact, including electrostatic, van der Waals and hydrophobic forces. Together, these forces can lead to microbial membrane damage during LDM material interactions. It should be noted that LDM-pathogen interactions are complex, possessing contributions from both the material and the microbial cell. For instance, the surface chemistry, charge, hydrophobicity, inherent roughness, geometry and free energy are all contributing factors from the LDM material. For the pathogen, molecular composition, surface charge, hydrophobicity, extracellular polymeric substance (EPS) and cell appendage interactions all play a role. In general, it has been suggested that positively charged materials attach to the negatively charged microbial membrane to induce membrane damage^45,100. Some materials can deactivate membrane components, such as the thiol group, through the generation of ions^41,100 or by extracting of phospholipids^62,63. Ions are typically generated by MOs, which cause the leakage of intracellular components^39,79 or directly damage the intracellular components¹³⁴.

Pre-infection treatments

One application for LDMs is in preventing microbial infections (i.e. pre-infection treatment) via limiting microbial adhesion to surfaces^26,135 and decreasing microbial growth^32,79. For LDMs to be used as an effective pre-infection treatment, they will need to be in a portable form with long-term stability, such as a bandage or implant coating or LDMs suspended in hydrogel^95,112. External stimuli can be applied to prevent infections in a clinical setting but are not practical for consumer products. Similarly, the chemical stability of LDMs must be improved for commercial products as they will need to be stored for longer periods. Overall antimicrobial activity is another important property to consider when using LDMs for pre-infection treatments. Materials, such as BP NSs⁴⁵ and g-C₃N₄ QDs¹²³, can be used as fast-acting treatments, while other materials including 0D MOs^41,136 and MXene NSs¹³⁷ have shown antimicrobial activity over several days. This duration is important as it influences the frequency of treatments, and how often the wound is exposed to unsterile conditions, increasing the risk of re-infection.

Post-infection treatments

Once an infection has formed, it can become much harder to treat and prevent more entrenched infection^138,139. The use of stimuli-activated LDMs, such as photothermal⁷⁵ or photoactivation³⁵ could be utilised within clinical settings. Importantly, LDMs have the ability to treat established infections with large quantities of microbial cells^138,140. One method of achieving this higher microbial inactivation can be the use of external stimuli, such as light activation^123,141. Several materials, including 1D MOs^26,41,140 and graphene QDs¹¹² have been able to reduce pre-existing biofilms by over 90% following photoactivation. Following the initial treatment to reduce the infection, previously discussed pre-infection treatments can be used in tandem to help prevent another infection.

Computational modelling as a guide for future treatments

Computational modelling techniques have demonstrated great potential to aid in the understanding of existing antimicrobial mechanisms and guide development of new LDMs. Classical molecular dynamics (MD) simulations have been used to show in atomistic detail how GO, N-g-C₃N₄ and MoS₂ nanosheets can destructively extract lipids from bacterial membranes^62,63,99 (Fig. 6a). In addition, coarse-grained MD simulations allow for a direct and fast in silico screening of LDMs candidate materials¹⁴². MD simulations have also shown why some LDMs are effective in vitro but not in vivo. For example, Duan et al. demonstrated that the efficacy of GO as an antimicrobial agent was significantly reduced by the presence of a protein corona formed by serum proteins that reduced the available surface area and sterically hindered membrane penetration and disruption¹⁴³. On the other hand, MD simulations have also shown how the effects of a protein corona can be overcome, or even utilised advantageously, for cell penetration of functionalized nanoparticles¹⁴⁴. While MD simulations are useful for studying interactions that can occur between LDMs and microbial membranes or biofilms^145,146, quantum chemical (QC) methods can calculate bandgaps of candidate LDMs^147,148, or examine the reaction mechanisms involved in ROS generation. Taking BP as an example, while the full ROS production reaction mechanism has yet to be elucidated, studies have shown that initial reactions leading to ROS production and BP degradation are most likely to occur at edges and defects in BP^17,149 (Fig. 6b). For the rapid and efficient exploration of a large number of candidate LDM properties, machine learning (ML) is often the best approach. ML algorithms can predict properties ranging from bandgaps of MXenes and hybrid 2D materials^150,151 to biocompatibility of ZnO nanoparticles¹⁵².

Fig. 6. Atomistic modelling of antimicrobial LDM behaviour.

a Lipid extraction by graphene oxide nanosheets from the outer membrane surface⁶². b Top view and side view of the reaction of O₂ (red) with monovalent defect BP (purple) from QC calculations.

LDMs with little or no antimicrobial activity

Although many LDMs have demonstrated high antimicrobial efficiency (over 80% cell death within several hours), there are some materials with little or no antimicrobial action. These materials were only capable of killing less than 60% of the microbial cells in 3 h or required over 8 h to inactive over 70% of microbial cells. These materials include hBN NSs¹⁵³, WS₂ QDs⁸⁴, BP QDs²¹ and graphene NSs¹⁵⁴. Some morphologies of these materials have demonstrated higher antimicrobial efficiency. One example of decreased antimicrobial action is BP, with the 2D BP achieving over 80% cell death within an hour^25,70, compared to QDs, which reached 60% cell death after 2 h²¹. Another example of a material with little antimicrobial activity is 2D hBN, with a 30% cell reduction in 3 h¹⁵³. This reduced antimicrobial activity is likely due to hBN not generating ROS, which means it is more reliant on a physical rather than a chemical mechanism which limits the overall antimicrobial potential^127,153. A careful review of the literature reveals the several QDs have lower antimicrobial activity compared to their 1D or 2D counterparts. This may be due to their antimicrobial activity arising from chemical/ion interactions^74,79, while having diminished ability to induce membrane damage.

The medium of the LDMs can also influence the antimicrobial efficiency. MOs have an increased antimicrobial potential on a surface^12,140, where most 0D materials are more effective as a suspension^72,77. For LDMs that rely on chemical interaction, suspension-based approaches have a higher antimicrobial efficiency^72,140. In contrast, physical-based antimicrobial mechanisms are efficient as both surface- and suspension-based treatments, depending on the desired application^23,116. For water purification, membranes equipped with LDMs are more effective than LDMs freely suspended in solution, and do not have to be removed from the purified water^27,39. For wound treatments, however, depositing LDMs onto traditional wound dressing surfaces, such as bandages or adhesive resins, have shown to promote improved wound healing compared to untreated dressings^75,112.

Current challenges

Although LDMs have promising antimicrobial properties, there are still limitations in fabrication processes and scalability that prevents practical implementation, for instance;

1. Synthesis methods for LDMs often use toxic solvents^87,137, require prolonged synthesis at high temperatures^18,116 and result in low yields^25,111.

2. Many 1D nanostructures, aside from MOs and GO, have only recently been synthesised and currently lack exploration into possible antimicrobial activity^18,93.

3. Some LDMs are less stable in desired environments such as in air or in solutions with a neutral pH^25,136.

One of the significant concerns using LDMs within a medical and commercial setting are related to their fabrication. Although LDMs should be available for feasible consumer products, consistent, cost-efficient methods need to be developed to allow for batch production. Major challenges to resolve for LDMs to attain market growth include (1) scalability, especially roll-to-roll manufacturing, (2) repeatable and reliable fabrication methods, (3) low contact resistance and (4) LDM-based precise characterisation techniques. One key point to a favourable outcome is the capability to prepare 2D materials at the wafer level. In this way, it will be possible to create large amounts of 2D devices for fabrication and decrease product cost¹²⁰. The current use of either high temperatures or toxic solvents^18,116 has led to an increased interest in developing green synthesis routes using natural materials for LDMs fabrication^40,82, increasing the potential for wider biomedical applications. Several current fabrication processes can take several days^15,47, and long-term storage can be limited^24,49.

One method of overcoming the rapid degradation of LDMs is to suspend the LDMs in liquid stabilisers or through storage in controlled environments^35,136. For example, some materials require a specific pH for storage of more than a few weeks, which is not ideal for biomedical applications^35,72. Although these stabilising measures are effective within a controlled laboratory setting, implementation on a larger scale for practical use is limited. For clinical applications, stabilisation could be achieved by embedding LDMs in medically relevant materials currently being used as wound treatments, such as hydrogels¹⁵⁵. The scalability and long-term impacts of LDMs on biological systems still require more research. There have been some cytotoxicity studies for a range of LDMs but these are predominantly carried out using in vitro cell cultures or mouse models^81,125. However, the method of excretion of LDMs from vital organs and the potential risks posed by LDMs aggregating within the body still needs to be examined futher^109,125.

Within the published literature, most LDMs have only been tested against a few key bacterial strains. The most common models are S. aureus for Gram-positive and E. coli for Gram-negative bacteria, which are human pathogens capable of significant morbidity and have several documented drug-resistant strains^48,79. Often in biological studies, fungal cells are overlooked, even though they pose a similar health threat⁵. This is important as fungal cells are larger than bacteria cells and possess different membrane structures and hence can be impacted differently by the antimicrobial mechanisms generated by LDMs²². Another limitation of the current microbial testing is the lack of “real world” scenarios. For example, if biofilm prevention is tested, typically only single strain models are used with limited testing on biofilms containing multiple bacterial strains, which are common on implant-associated infections¹².

Future outlook

LDMs^1,35,45,156 utilise a combination of chemical and physical modes of action to kill pathogenic microbes with extremely high efficacy in a range of conditions. Combining this with the emerging capability to control the properties of LDMs offers an unprecedented opportunity for the research community to explore a plethora of potential antimicrobial applications. Furthermore, the synthesis of composites LDMs which can have synergistic effects provides the basis to create new paradigms in a field of antimicrobials, which has stagnated to a dangerous point^24,71.

Importantly, there is a lot more work that needs to be done. Many facets of the antimicrobial mechanisms of LDMs remain unclear, and the library of prospective materials should be expanded. Further, the clinical and commercial applications of these materials remain under researched. Such areas of research need to be further investigated for LDMs to be considered a serious alternative to current antimicrobial treatment strategies. It is hoped that this review will provide a foundation for informed decisions and design parameters of next-generation antimicrobial LDMs with antipathogenic activity and reveal an antipathogenic technology capable of combatting AMR pathogens.

Author contributions

Z.L.S., S.K., M.D.D., J.G., A.E. and S.W. conceived the project; Z.L.S., S.K., S.C., A.J.C., J.C., V.K.T., A.E. and S.W. reviewed the relevant literature.; Z.L.S., S.K., S.C., M.D.D., J.G., R.J.C., C.F.M., A.J.C., J.C., V.K.T., A.E. and S.W. were involved in the discussion, outlining, analysing and developing perspectives for the review. Z.L.S., S.K., S.C., A.J.C., V.K.T., A.E. and S.W. wrote the manuscript, with input from all other authors. Each author was involved in the discussion surrounding the production of the manuscript, data interpretation and scope of the review.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-021-23278-7.