Phytoalexin Phenalenone Derivatives Inactivate Mosquito Larvae and Root-knot Nematode as Type-II Photosensitizer

Phytoalexins phenalenones (PNs) are phytochemicals biosynthesized inside the plant in responsive to exterior threat. PNs are excellent type-II photosensitizers, which efficiently produce singlet oxygen upon light irradiation. Based on the core functional structure of PNs, novel PN derivatives were synthesized here and their singlet oxygen generating abilities and their phototoxicity were evaluated. At the presence of light, these PNs have photoinduced toxicity towards Aedes albopictus larvae and nematode Meloidogyne incognita, while the activity lost in the dark. The obvious tissue damage was observed on the treated mosquito larvae and nematode due to the generation of singlet oxygen. Our results revealed the potential of phenalenones as photoactivated agents for mosquito and root-knot nematode management together with light.

have much influence on the activity level. 9-phenyl-phenalenones (PPN) or 2-hydroxyl-9-phenyl-phenalenones (HPPN) have the rapid and long-lasting antinematode activity to burrowing nematode Radopholus similis 20 . Teresa Abad-Grillo er al. reported some 9-heterocyclic substituted phenalenone analogues with good antiplasmodial and antiprotozoal activity against Leishmania amazonensis, Trypanosoma cruzi and Plasmodium falciparum. The above studies indicate the importance of the substituents at 9-position 17 . Therefore, we firstly prepared the 9-substituted PNs analogues (PPNs) to study its photosensitized ability. With a view to utilize these photosensitizer in aqueous media to control mosquito larvae and root-knot nematode, the water soluble derivatives 2-pyridiniium-9-phenyl-phenalenone (PPPN) were synthesized by attaching a pyridinium salt fragment. Meanwhile, the atom fluorine was introduced into the structure due to its particular role in the bioactivity, such as high electronegativity, good hydrophobicity and metabolic stability 21 . Phenalenones Synthesis. The synthetic sequence for the required compounds are depicted in Fig. 1. The PN was synthesized according to the reported procedure by reaction of naphthalene and cinnamoyl chloride via Friedel-Crafts reaction and aromatization through elimination of benzene 22 . PPNs were prepared by slow adding of Grignard reagents into PN in tetrahydrofuran at − 40 °C and the followed refluxing in dichloromethane catalyzed by 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ). Epoxidation of PPNs and acid treatment of the intermediary epoxide facilitate the appending of 2-hydroxyl group generating HPPNs 18 . Finally, the pyridinium salt PPPNs was attached through Blanc chloromethylation of PPNs followed by reactions with pyridine 19 . Detection of singlet oxygen. The 1 O 2 photogenerating process and ability of the synthesized compounds were determined through a chemical method based on 1 O 2 sensor 9, 10-anthracenediylbis(methylene)dimalonic acid (ABDA) 23 . ABDA rapidly and quantitatively converts to a steady-state endoperoxide product by reacting with singlet oxygen in the presence of photosensitizer ( Fig. 2A). Reaction process was monitored by the decrease of absorbance intensity of ABDA upon increasing irradiation at 420 nm (Fig. 2B). The quantum yields of photosensitized 1 O 2 generation were calculated from the comparison of quantum yield of PN (Φ △ = 1) 24 . PN is an efficient singlet oxygen photosensitizer with ca. 100% quantum yield in a large variety of solvents. It has good photostability under the irradiation wavelength of light and low ability to deactivate singlet oxygen. The water solubility of PPN1-2 and HPPN1-2 was too poor to measure their quantum yields. The quantum yields of PPPN1 and PPPN2 are measured to be 0.12 and 0.09 in acetonitrile, respectively, indicating that 9-phenyl installation reduced the efficiency of 1 O 2 generation. This decrease is caused by the intramolecular charge-transfer from the phenyl to the electron-deficient PN fragment 10 . Introduction of hydroxyl group at 2-position of PPN led to the further decrease of quantum yields with values of about 3 orders of magnitude lower than that of parent PN. The quantum yields of PPPN1 and PPPN2 were about 26-fold and 31-fold lower than that of PN, respectively. This trend could also be reflected in the time-dependent absorbance change of ABDA, in which PN or PPPN3 induced much faster extinction of ABDA (Fig. 2C). Although the 1 O 2 generating efficiency of PPPN1 and PPPN2 is low, the relatively-slow generation of singlet oxygen has its value for practical application in plants because the balance between defensing outside threaten and autotoxicity should be taken into accounts 10 . The PPPNs is resistance to photobleaching, since no significant photo degradation of PPPN1 was detected after 3 h exposition to light (Fig. 2D).
Photoinduced activity to mosquito larvae. Mosquitos are vectors of some of humanity's deadly diseases causing one millions deaths worldwide annually and almost everyone has suffered from their bite allergy 25 . This aquatic and transparent insect is the potential target for action of photosensitizer. The intrinsic and light-dependent toxicity of these PN derivatives were evaluated against Aedes albopictus larvae and the results were summarized in Table 1. Taking the water solubility and 1 O 2 producing ability into account, the PPPN1-3 together with PN were tested for photoactivated activity. The light sources used for 1 O 2 production are blue-light (420 nm) or sunlight, although the maximum absorbance wavelength of PN derivatives are around 370 nm. The reason for such selection is to avoid the harmful irradiation of UV-light that might cause damage to insects. Blue light (420 nm) alone do not have any detrimental effects to the larvae. PN derivatives in the absence of light showed very low killing potency to mosquito larvae with large LC 50 values ( >140 mg/L). A trend emerged that the assembling of phenyl component at 9-position led to a 2-fold activity decrease (Fig. 3A, PPPN1 and PPPN2 vs. PPPN3). When exposed to blue light, a significant activity enhancement for all the tested compounds was observed with LC 50 ranging from 13.0-22.5 mg/L, indicating 7-19 folds activity increase. This phenomenon that PN derivatives needs work together with oxygen and light to exhibit their efficacy demonstrated a photosensitized mechanism. To our delight, the same activated effects were achieved through sunlight irradiation, implying a clean, energy-saving and easily available light source in practical application. The slight decrease of activity of pyridinium form PPPN1-3 in comparison with PN might in part attribute to their high water solubility which limit effective tissue distribution of the chemical 11,26,27 .
Biological dysfunction induced by 1 O 2 includes cell death, membrane damage and enzyme inactivation 11 . In order to understand the damage occurring to the larvae, the microscopy and fluorescence images of mosquito larvae were acquired (Fig. 4). The pictures were collected 24 h after application of PPPN1. The blue light (420 nm) was irradiated in the first four hour of incubation. No significant adverse effect after exposure to blue light was detected in the control treatment without photosensitizer (Fig. 4, Aa and Ab) and the fluorescein isothiocyanate (FITC) concentrated mainly in the digestive system ( Fig. 4Ac and Ad), that is, gastric caeca and guts lumen 26,28 . Incubation with PPPN1 but without light (dark control) can not cause any damage to the larvae ( Fig. 4Ba,b,c and d), proving the inactivity of the compound in the dark. Aedes larvae treated with PPPN1 plus blue light underwent obvious tissue damage and rupture ( Fig. 4a and b) accompanied with the fluorescence dye leaking to the whole body ( Fig. 4Cc and d). As PN has weak intrinsic yellow fluorescence, it was therefore used as both photosensitizer and staining agent for visualization study (Fig. 5Aa,b,a and b) to elucidate the partitioning of the photosensitizer among the different organs of mosquito larvae. When PN was administrated without irradiation, the PN mainly accumulated in the digestive system ( Fig. 5Aa and b) and afterward blue light stimuli made it penetrate into other body parts ( Fig. 5Ba and b). The above observed symptoms is different to that of larvaes poisoned by fipronil and stained by FITC ( Fig. 5Ca and b). This observation demonstrated that initial allocation of photosensitizer around gut would destroy gut cells and the subsequent photosensitizer diffusion caused further photo damage to the larvae, leading to final death of the insect.
Photoinduced activity to root-knot nematode. Plant parasitic nematodes cause great crop losses annually. In the nematode-resistant banana cultivar, the main isolated defensive phytoalexins were phenylphenalenones which have the nematostatic and nematicidal activity towards burrowing nematode Radopholus similis. Due to the potential photo toxicity of photosensitizer to nematodes 29 , the biological responses of phenalenones PN and PPPN1-3 were evaluated here towards root-knot nematode Meloidogyne incognita. All the tested compounds have low dark activity. Similar synergistic effects of light to the synthesized PN derivatives were observed with the increase in the nematicidal activity ( Fig. 3B and Table 2). Although nematode was sensitive to UV light 29 , no obvious photobiological responses were detected under the wavelength of light used here. The increased sensitivity of nematode correlated closely with the increased length of light exposure 30 . Complete photocidal effects could be achieved by 1 h light stimuli for PPPN3 at concentration of 100 mg/L, while the photo toxicity of low dosage PNs increased gradually with prolongation of irradiation (Fig. 3C).
Nematode microscopy images of PPPNs-treated nematodes exerted obvious morphological abnormalities ( Fig. 6B and C) and larger differences with the dark control (Fig. 6A) and blank control ( Fig. 6E and F, with or without irradiation, respectively). Cell destroy and merging were observed in the body of PPPN-loaded larvae after imposition of light, presenting a droplet-like appearance. Sole treatment of nematode with PPPN1 did not lead to the formation of such large droplets which is opposed to the previous observation that yellow droplets appeared after application of anigorufone in Radopholus similis 20 . From present observation, the formation of droplets might be partly attributed to the photosensitization, since no dark controls were conducted in anigorufone-treated R. similis 20 . There is also a possibility that poorly water-soluble anigorufone mainly accumulated in the lipid droplets which prevents the efficient photosensitization process 31 . The water soluble PPPNs used in our study are more efficient for the generation of 1 O 2 upon light irradiation, which led to destroy and merge of the tissue components generating a droplet-like shape. In contrast, the nematode poisoned by avermectin did not form such droplets, indicating the unique toxic features caused by singlet oxygen.
The level of the photoinduced activity is preferentially associated with the 1 O 2 generating ability, light intensity and distribution in the body 11,32 . The distribution is determined by hydrophobicity, moreover, it also relates closely with the type of the target insects 11 . Cell membranes are preferential location for hydrophobic photosensitizer, while passive and active diffusion processes are involved in the hydrophilic one 32 . Amphiphilic photosensitizer is particular efficient in crossing the lipid cell membrane and subsequent diffusing in water-based environment inside the cell 11 . Previously studies showed that the primary oxidative modification sites were the membranes of midgut wall 11 . Despite of large difference in the 1 O 2 generating efficiency, PN and PPPNs have the same level of photo killing activity. The low water solubility of PN limits its efficient ingestion by the insect, although it has high quantum yields. In case of PPPNs, amphiphilic feature make them easily arrive at the target site, offsetting the weakness of relatively low 1 O 2 generating ability. Another possible explanation for the uncorrelation of the quantum yields with the activity is the limited localized oxygen concentration inside the insect. The generation ability of singlet oxygen correlates closely with the localized oxygen concentration. Inside the body, the localized oxygen was consumed up quickly upon irradiation, which prevents the further generation of the singlet oxygen.

Conclusion
We described here the ability of phenalenones derivatives to inactivate the mosquito larvae and root-knot nematode through generation of singlet oxygen upon light irradiation. The utilization of photochemical mechanism for controlling harmful species processes several substantial advantages including dark inactiveness, no cross-resistance and controllable activity. The present study provided a basic understanding of PN derivatives as potential mosquito-or nematode-management chemicals. For practical pest control, novel application methods are needed to be developed to efficiently use them, such as immobilization of the photosensitizer to recover or reuse them, anchored in the wall paintings, combining with the bait and coated on the surface of small beads. Switching the absorption band to the red-light range through structural modification is another challenge for better penetration to biological tissues and efficient use of sunlight, since using sustainable natural light is the     Chemical oxidation detection of singlet oxygen. An ABDA-based oxidation method was used to assess the capability of PNs to generate 1 O 2 23 . A mixture of PN derivative (0.3 mmol in 3 mL in ultrapure water) and 50 μL ABDA solution (10 mmol/L in DMSO) was irradiated with a light-emitting diode lamp (blue light, 420 nm, 7 W). The change of ABDA absorption at 400 nm was collected using Varian Cary 100 UV-vis spectrophotometer as a function of irradiation time. The control experiment was conducted using 3 mL ultrapure water containing above-mentioned 50 μL ABDA solution upon the same irradiation but in the absence of PN derivative. The 1 O 2 quantum yield was calculated according to Equation (1) using PN as reference (Φ PN = 1), where Φ is the quantum yield of 1 O 2 and K is the slope of the bleaching curve, R denotes reference, and S the sample.
Photoactivated activity of Aedes albopictus larvae. Photoactivated larvicidal activity was determined using previously-described methods 33 . The tested insects were the 4th-instar larvae of A. albopictus which were obtained from National South Pesticide Innovation Center in Shanghai, China. PN was dissolved and serially diluted with acetone and PPPN1-3 was dissolved and serially diluted with dechlorinated water. Each serial solution (0.2 mL) was added to a beaker containing 10 mL of dechlorinated water, and then 15 larvae were transferred into the dactylethrae. Two sets of experiments were performed for each compounds, one of which was for light-treated trials, and another was cultivated in the dark throughout the trials. The light-treated groups were irradiated with blue light for 3 h after 3 h incubation in dark, and then returned to darkness for 24 h incubation. The light intensity of the testing solution surface is about 50 W/m 2 . The average mortality of three replications at each concentration was calculated, and the LC 50 value was determined. All the experiments were conducted at least three times with three replicates in each case. The microscopy images of A. albopictus larvae were taken by a polarizing microscope.

Photoactivated activity against root-knot nematode. M. incognita population were grown on
Lycopersicon esculentum plants in a greenhouse, and collected as described by Rosso and associates 34 . Then hatched J2 were collected as described in Petri Plate Technique 35 . PN and PPPN1-2 were dissolved with DMF and serially diluted with dechlorinated water. Each serial solution (0.05 mL) was added to three wells of 96-well platers and then 0.05 mL water containing root-knot nematode (J2) was transferred into the 96-well plate. Three sets of experiments were performed for each compound, one of which was for light-treated trials, and another was cultivated in the dark throughout the trials. The light intensity of the testing solution surface is about 50 W/m 2 . After 3 h incubation in dark, the light-treated groups were irradiated with blue light for 1 h, 2 h, 3 h and 4 h, and then returned to darkness for 24 h incubation. The average mortality of three replications at each concentration was calculated. All the experiments were conducted at least three times with three replicates in each case. The microscopy of M. incognita were taken by a polarizing microscope.