Antibiotic resistance is referred to as one of the biggest threats to the modern healthcare system because of the scale and speed at which it is progressing, which could eventually lead to the large-scale spread of resistant infectious diseases with severe outcomes [1]. In 2017, WHO published a list of multi-drug resistant (MDR) bacteria to promote further research in developing new antibiotics to combat them. Reported among the list are what are known as ‘ESKAPE pathogens’ which are notably categorised under ‘critical and high priority’. The bacteria listed are Enterococcus faecium (E. faecium), Staphylococcus aureus (S. aureus), Klebsiella pneumoniae (K. pneumoniae), Acinetobacter baumannii (A. baumannii), Pseudomonas aeruginosa (P. aeruginosa) and the Enterobacter species [2]. A study estimated the trends in the prevalence of hospital-associated drug-resistant infections (HARIs) in 195 countries and realised Escherichia coli (E. coli) was the common cause of the burden of HARIs in all low, middle and high-income countries [3]. Overconsumption and improper use of antibiotics are proposed to be the leading causes of the predominance of MDR bacteria [4,5]. Multi-drug resistance is described as acquired non-susceptibility to at least one agent in three or more antibacterial drug classes [6]. Antibiotic resistance is the ability of a bacteria to reduce antibacterial's effects, which may be resulted from intrinsic or acquired mechanisms. Of concern is acquired bacterial resistance, whereby bacteria adapt and become resistant to an agent to which they were initially susceptible, due to vertical or horizontal evolutions [7]. Bacteria may demonstrate resistance towards antibacterial compounds through various biochemical pathways or mechanisms [[8], [9], [10]]. The first mechanism is via enzymatic inactivation of the drug resulting in the decrease of drug concentration inside bacteria cells. For instance, chloramphenicol acetyltransferases acetylate hydroxyl groups of chloramphenicol, causing them unable to bind to ribosomal 50S subunits [8,10]. Another examples of bacterial enzyme in K. pneumoniae, A. baumannii, P. aeruginosa, and Enterobacter species are the β-lactamases which hydrolyse the β-lactam rings of penicillins, carbapenems, cephalosporins, cephamycins and monobactams [11]. Secondly, bacteria can prevent intracellular accumulation of antimicrobials by decreasing uptake, modification of a specific target as well as increasing efflux from the cell [10]. For instance, β-lactams and fluoroquinolones, which are small hydrophilic molecules that enter Gram-negative bacteria through the porin channels located in the outer membrane of the bacteria [10]. Hence, a decrease in the number of porin channels can reduce the entry of drugs into the bacteria cell, causing resistance to these drugs [10]. Another example is the modification in penicillin-binding proteins (PBPs), which is a favoured mechanism of resistance for Gram-positive bacteria such as S. aureus compared to Gram-negative bacteria whose production of β-lactamases is more common [9,10]. The presence of a mutation in PBPs reduces affinity to β-lactam antibiotics, thereby causing bacterial resistance to the drugs. On the other hand, efflux pumps are membrane proteins that release antibacterial agents out from the bacteria cell to maintain a low intracellular concentration [10]. In other words, the drugs are pumped out before reaching their targets and exerting their effects. There are five major families of efflux pumps, which differ in terms of structure, energy source, range of substrates they can remove and in which bacteria type they are distributed. The five important families of chromosomally encoded bacterial efflux pumps are major facilitator superfamily (MFS), small multi-drug resistance family (SMR), resistance-nodulation-cell-division family (RND), ATP-binding cassette family (ABC) as well as multi-drug and toxic compound extrusion family (MATE) [12]. RND family systems are major efflux pumps that contribute to resistance issues in Gram-negative bacteria while in Gram-positive bacteria, multi-drug resistance is more often related to MFS efflux systems. Antibiotics of all families except polymyxins are susceptible to the efflux pump system [8,10]. The third possible resistant mechanism adapted by the bacteria is the formation of biofilm, whereby bacterial cells within the biofilm have a slow metabolism rate and cell division, resulting in ineffective antimicrobials which target growing and dividing cells [4,13]. Therefore, the dire need for new alternative treatments effective against these resistant pathogens has led researchers to explore different drug compounds and administration strategies [14].

To date, currently available antibiotics for the treatment for infections caused by Gram-negative MDR bacteria include polymyxin, tigecycline, carbapenems, aminoglycosides, fosfomycin, ceftazidime, meropenem and ceftolozane [15]. However, reports of drug resistance developed against carbapenems by Enterobacteriaceae have been noted [[16], [17], [18], [19], [20], [21]]. Although new antibiotics such as delafloxacin, cefiderocol targeting MDR Gram-negative bacteria and drugs treating Gram-positive bacteria, such as eravacycline have been approved by the U.S. Food and Drug Administration (FDA), it is only a matter of time before the bacteria mutate again and become resistant to the currently available drugs [22]. Therefore, while the search for new antibiotics is essential, other therapeutic alternatives for MDR bacteria must also be introduced. Other novel strategies attempted in the past decade besides the development of new small molecule antibiotics include the development of quorum sensing inhibitors [23], cationic peptides [[24], [25], [26], [27], [28]], pathogen-specific monoclonal antibodies [29], efflux pump inhibitors [30], photoactive antimicrobials [31], antimicrobial nanoparticles [32] and phage therapies [33]. Among these new strategies, photodynamic therapy (PDT) which is less susceptible to antimicrobial resistance mechanisms of the MDR bacteria due to its free radical generation properties [34] is of particular interest.

PDT is a non-invasive therapeutic process that uses photosensitisers (PSs) and a light source of a specified wavelength to generate reactive oxygen species (ROS), which destroy the targets such as pathogens, cancer or tumour cells inside the body by oxidation [35]. PDT is a form of treatment that is successfully being used in current clinical treatment for cancer and symptoms of age-related macular degeneration in recent years. With the growing MDR bacteria crisis, the application of antimicrobial photodynamic therapy (aPDT) started to gain interest among scientists and other related stakeholders [36]. The process of PDT goes as follows: upon light irradiation, a photosensitiser (PS) absorbs a photon and gains energy to move from the ground So state, to a higher energy state Sn. From this excited state, the excess energy could be either given off as a fluorescence light, heat energy or it could be used to undergo intersystem crossing to a longer-lasting triplet state [35,37,38]. From triplet state, the PS could undergo two types of pathways to create ROS, type 1: where the triplet PS molecule directly reacts with the surrounding biomolecules and dioxygen to produce free superoxide anions (O2•-), that generate hydroxyl radical (•HO) by breaking down hydrogen peroxide (H2O2), and type 2: the PS directly reacts with dioxygen molecules to form singlet oxygen as shown in Fig. 1 [35,37,38]. The generated ROS ultimately induces irreparable damages to the exterior polymeric materials, cellular components, DNA and proteins present in the bacteria and causes their death [38]. After the reaction, PS returns to the So state and can be reused for the regeneration of ROS, making PDT a very efficient process, as one PS molecule can generate several ROS, though the product yield differs from the types of PSs and their structures [39]. PSs used in aPDT are different from the ones used for regular PDT as the charge of PS molecules is a crucial factor that determines the physiological stability and penetration ability [40]. For instance, the bacterial envelope of Gram-negative bacteria has an abundance of lipopolysaccharides on its surface, thus attributing a negative charge to the bacteria. This makes the bacteria a viable target for cationic PSs. Interestingly, PDT indiscriminately oxidises various biomolecules by producing ROS that do not have a specific target and may bypass the resistance mechanisms adopted by MDR bacteria. This clearly may lower the development of tolerance of bacteria to this treatment [41,42]. However, PDT involves ultraviolet, visible or near-infrared wavelengths which cannot penetrate the deep tissues in the body [43]. Other limitations of PDT include photobleaching, the higher dose needed to achieve the desired therapeutic effect, long elimination half-life of the PS as well as low singlet oxygen quantum yield [44].