Overview of methods and considerations for the photodynamic inactivation of microorganisms for agricultural applications

Antimicrobial resistance in agriculture is a global concern and carries huge financial consequences. Despite that, practical solutions for growers that are sustainable, low cost and environmentally friendly have been sparse. This has created opportunities for the agrochemical industry to develop pesticides with novel modes of action. Recently the use of photodynamic inactivation (PDI), classically used in cancer treatments, has been explored in agriculture as an alternative to traditional chemistries, mainly as a promising new approach for the eradication of pesticide resistant strains. However, applications in the field pose unique challenges and call for new methods of evaluation to adequately address issues specific to PDI applications in plants and challenges faced in the field. The aim of this review is to summarize in vitro, ex vivo, and in vivo/in planta experimental strategies and methods used to test and evaluate photodynamic agents as photo-responsive pesticides for applications in agriculture. The review highlights some of the strategies that have been explored to overcome challenges in the field.


PDI principle
Photodynamic inactivation, also referred to as antimicrobial photodynamic inactivation (aPDI), uses light-responsive molecules (photosensitizers, PS) that when excited by light react with molecular oxygen to generate reactive oxygen species (ROS) [1].ROS can be highly cytotoxic if produced in sufficient amounts and in close proximity to the pathogens as they can cause irreversible damages to both the outer membranes as well as internal structural and cellular components (proteins, lipids, DNA, membranes, cytoskeleton), culminating in microbial death [2,3].Depending on the type of PS, generation of ROS can follow two different photochemical mechanisms.Upon absorption of a photon by the ground-state (inactive) PS, the singlet excited state 1 PS * is formed.Intersystem crossing converts a portion of this state into the triplet state ( 3 PS * ) which can then interact with molecular oxygen by either electron transfer (Type I photosensitisation), or energy transfer (Type II photosensitization) (Fig. 1).In the former, the interaction between the activated electron and a molecular oxygen leads to the formation of ROS, such as hydrogen peroxide (H 2 O 2 ), superoxide (O 2 − •) or hydroxyl radical (•HO); in the latter, singlet oxygen ( 1 O 2 ) is produced [4][5][6] (Fig. 1).While both Type I and II photosensitization can occur simultaneously upon excitation of a PS, in general the Type II reaction is preferred in aPDI [7,8].The high reactivity of 1 O 2 is partly due to its lifetime (3-50 ms in aqueous media and several tens of ms (2-1000 ms) in lipid environments such as membranes), which allows this species to diffuse over relatively long distances before being deactivated [9,10].If Wenzi Ckurshumova and Cristina Rosa contributed equally to this work.generated in the apoplast, singlet oxygen can diffuse across the cell membrane [11], where it can last long enough to interact with its targets.While ROS can be non-specific in their interactions with pathogens and host membranes, toxicity towards plant cells has been observed only at high concentrations whereas microbial inactivation is effective at micromolar concentrations [12][13][14].Several studies have shown the non-toxic properties of PSs against various plant species at photochemically active doses [15][16][17] as well as no toxicity upon repeated applications in developmentallyrelevant plant stages [18].Having said that, sensitivities to a PS can be species-specific and similar concentrations may have contrasting-impacts.This might be linked to structural differences in the cuticle/wax layer, or antioxidative capacity in cells.For instance, cationic porphyrins, were non-damaging to tomato plants whereas they completely eradicated the model plant Arabidopsis [19,20].
The wavelength of light required for activation depends on the absorption characteristics of the PS used, but for agricultural uses PSs that can absorb a substantial portion of photons in the photosynthetically active radiation of the solar irradiance.Generally, antimicrobial activity of PSs seems to be more efficient if natural sunlight is used, likely because of its higher intensity as well as its use of the full absorbance spectra of the molecules [21][22][23]; however, antimicrobial activity is also effective in greenhouse settings with commercially available LEDs [15,[21][22][23].Several types of PSs have been shown to be effective against plant pathogenic bacteria and fungi, as summarized in Table 1.Nonetheless, the efficacy of such PSs is typically assessed in traditional liquid culture, which differs significantly from killing pathogens on the surface or within the body of a plant leaf.In the following section, three key challenges associated with such predictions are outlined as well as descriptions of protocols that aim to address these challenges.

Photostability
While activation of PSs with high intensity sunlight leads to effective antibacterial activity, it can cause rapid photodegradation/bleaching of PS molecules resulting in loss of their absorption and emission.Destruction of the molecules on one hand reduces their effective lifetime [3,[24][25][26]; on the other hand, for example for chlorophyll derived photosensitizers with degradation products that can be metabolized by the plant [27], it has a chance to prevent the accumulation of compounds in the environment.This reduces or eliminates pesticide residue issues that plague conventional pesticides and limit when such pesticides can be used in the growing cycle.Classically, photodegradation studies are done in liquid through monitoring changes in the absorption spectrum of the photosensitizer (i.e., the appearance of new absorption bands or decrease of the maximum absorption peak) [28].
In agricultural applications however, foliar sprays typically dry on the leaf surface in minutes and remain dry until dew or rain rewets them.This may lead to changes in the aggregation state of photosensitizing molecules.Recently, Islam et al. studied the stability and efficacy of the water-soluble semisynthetic derivative of chlorophyllin (sodium magnesium chlorophyllin (Mg-chl) following dry-rewet cycles mimicking conditions similar to the field [15] (Fig. 2).They observed that, there was a gradual decrease of fluorescence and single oxygen production with time, indicating gradual degradation of the compound under light exposure.However, despite some degradation, the compound was still able to eradicate bacteria significantly (3log10 reduction) up to 5 days after of light exposure, suggesting that the activity of the PS was maintained for at least up to 5 days in the dry state.

Contact/bacterial proximity
Given that photo-pesticides are contact pesticides, the proximity of PS to the pathogen is one of the crucial factors determining antimicrobial efficacy since the probability of ROS reacting with target structures and molecules drastically decreases with distance [29].For example Gram ( −) bacteria are wrapped in a densely packed protective layer (outer membrane) comprising mainly glycolipid lipopolysaccharides and phosphoglycerides that serve as a permeability barrier which can effectively inhibit penetration of pesticides into the cell [30].While both anionic and cationic photosensitizers alone can kill efficiently Gram ( +) bacteria in vitro [31], for complete eradication of Gram ( −) bacteria a synergistic combination of PS combined with a membrane disrupting molecule is necessary [15,[31][32][33].Uptake of anionic PS by bacterial cells may be mediated through

Xanthomonas gardneri
In vivo (tomato seeds) Methylene blue, toluidine blue, and a combination of both dyes 652 nm, 12.21 mW/cm 2 , 20 min [77] both protein transporters and electrostatic charge interaction, while the uptake of cationic PS is presumed to be facilitated through self-promoted uptake in addition to interactions with membranes [28,34].Efficacy in vitro is typically tested in liquid culture (Fig. 3), with plates kept in the dark after PS addition to allow the PS to adhere to bacterial cell walls prior to light exposure [8,15,31].Although the assay is adequate for determining effective concentration ranges of PSs against pathogenic bacteria in vitro, it does not represent the environment on plant surfaces.The epidermal cells of most plant surfaces are covered with cuticle, a lipid-rich layer that prevents water loss and protects plants against multiple biotic and abiotic stresses [35,36].To reflect interactions of bacteria with the cuticle and test aPDI on plant surfaces, both ex vivo experiments using detached plant leaves [17,37] and in vivo/in planta experiments using intact plants have been developed [15].In ex vivo experiments, detached leaves are sprayed with bacteria and PS (Fig. 4a); the in vivo method follows the same procedure except that bacteria and PS are applied through a foliar spray directly on the plant (Fig. 4b) [15], representing conditions in the field.
In the case of fungi, both anionic and cationic PS can efficiently eradicate the pathogens (refer to Table 1 and references therein).Fungal mycelia typically grow in 3-D structures that often limit access of the PS to all mycelia.To simulate the 3-D structure of mycelia, Hamminger et al., carried out in vitro aPDI of fungi using mycelial spheres in 24-well plates (Fig. 5) [18] and showed that the PSs can efficiently kill fungal species.

Light intensity
Exposure to a sufficient amount of light for PS activation is an indispensable requirement for PDI.Particularly in foliar pathogens, the first step in pathogenesis is the colonization of aerial tissue surfaces.Flat surfaces such as cucumber, tomato, and lettuce leaves provide excellent light exposure, and good microbial reduction was observed when the cationic curcumin derivative SACUR-3 was used against E. coli O157:H7 [38].Bacterial pathogens can gain access to internal plant tissues either through wounds or through natural openings such as stomatal pores used for gas exchange; alternately, pathogens can be delivered into plants by insect vectors [39,40].Until recently it was unclear whether a sufficient amount of light is available to activate PS molecules inside the leaf.Islam et al. investigated this question by infiltrating a mix of bacterial suspension and photosensitizer in the intercellular leaf spaces of tomato and N. benthamiana plants, exposing the plants to light, and determining the number of viable bacteria in the leaves after treatment (Fig. 6) [15].They found a significant inhibition of bacterial growth in planta when plants were exposed to light which suggests that PS can be activated inside the leaves and is able to kill intracellular bacteria.Mycelial spheres and PS are added to 24-well plates and kept in the dark before exposure to light for several hours.Fungal balls are then transferred to media and checked for growth after a few days depending on the growth rate of the specific fungi

Conclusion and outlook
The photochemical efficiency of aPDI in lab test conditions depends critically on the properties of the PS molecule (e.g. the presence or absence of charge and the charge distribution), the efficiency of light absorption, and the longevity of the triplet excited state or singlet oxygen and free radical production [4].However, when targeting plant pathogens in vivo, consideration of an additional layer of requirements depending on the lifecycle of the pathogen and tissues it inhabits is important.The design of the photosensitizer must consider the presence of peripheral functionalizations that may confer specific localization in the plant (i.e.hydrophilicity vs. lipophilicity, ability to translocate and/or bind to various cellular structures [41], as aPDI activity may change depending on cell location and/or the oxidative state of the cell.Nanoencapsulation of PSs that can facilitate improved photostability, targeted tissue penetration, and tunable PS release kinetics offer a promising route to address these challenges [42].In addition, while progress has been made to better represent processes occurring in the field via ex vivo, or in planta assays, further development is necessary to address the various challenges often occurring at once in the fieldfrom water fluctuations to intense sun and wind.

Fig. 1
Fig. 1 Schematic representation of aPDI on a plant leaf; viewed in cross-section.PS molecules entering through stomatal openings.GC guard cells, PM palisade mesophyll, SM spongy mesophyll

Fig. 2 Fig. 3 A
Fig. 2 Schematic diagram of the dry drop method.A droplet of PS is dried in the dark before exposing it to light.After irradiation, the dry PS droplet is resuspended in water and used to measure absorbance, fluorescence, ROS, and antibacterial activity

Fig. 4 Fig. 5
Fig. 4 Schematic diagram of the ex vivo and in vivo aPDI assays.a Ex vivo: Bacterial cultures of known concentration (OD 600 ) and PS are sprayed on detached leaves and kept in dark for an hour before exposing them to light for several hours.The leaves are then placed in buffer and shaken to dissociate the bacteria, and an appropriate dilution of bacteria is spread on solid LB media for colony forming

Fig. 6
Fig. 6 PS and bacterial suspension are infiltrated in leaves.Plants are kept in dark for an hour before they are exposed to light for 16 h.Leaf discs are collected, ground and bacteria are plated onto media for CFUs count

Table 1
Summary of photosensitizers used against plant pathogens