Tannins-lignin mixed nanoformulations for improving the potential of neem oil as fungicide agent

Sustainability and circular economy are increasingly pushing for the search of natural materials to foster antiparasitic treatments, especially in the case of economically relevant agricultural cultivations, such as grapevine. In this work, we propose to deliver neem oil, a natural biopesticide loaded into novel nanovectors (nanocapsules) which were fabricated using a scalable procedure starting from Kraft lignin and grapeseed tannins. The obtained formulations were characterized in terms of size and Zeta potential, showing that almost all the nanocapsules had size in the suitable range for delivery purposes (mean diameter 150–300 nm), with low polydispersity and sufficient stability to ensure long shelf life. The target microorganisms were three reference fungal pathogens of grapevine (Botrytis cinerea, Phaeoacremonium minimum, Phaeomoniella chlamydospora), responsible for recurrent diseases on this crop: grey mold or berry rot by B. cinerea and diseases of grapevine wood within the Esca complex of diseases. Results showed that grapeseed tannins did not promote inhibitory effects, either alone or in combination with Kraft lignin. On the contrary, the efficacy of neem oil against P. minimum was boosted by more than 1–2 orders of magnitude and the parasite growth inhibition was higher with respect to a widely used commercial pesticide, while no additional activity was detected against P. chlamydospora and B. cinerea.


Introduction
In recent years, nanotechnology has proven to be a valuable strategy for delivering agrochemicals such as pesticides (Rajput et al. 2021;Chhipa 2017). Among the advantages of nanotechnology, there is (i) increased bioavailability of hydrophobic compounds by enhanced water solubility; (ii) protection of labile cargo molecules from biological and chemical degradation; (iii) administration in ultra-finely dispersed form, which provides extended contact interface. Moreover, the encapsulation of active principles allows to avoid or reduce their loss in the environment (Lowry et al. 2019;Kah et al. 2018). This is useful, in particular, when volatile compounds are to be administered or when the active compounds can be toxic toward other species. Among possible carriers, polymer-based nanosystems are considered very promising due to their versatility and safety, this latter being especially warranted if natural macromolecules (de Oliveira et al. 2018;Niculescu and Grumezescu 2021;Yeguerman et al. 2022) are used to fabricate the Responsible Editor: Giovanni Benelli nanocarriers. Therefore, choosing compatible matrices (Colzi et al. 2015) for the preparation of nanosystems, or, possibly, re-using by-products from the plant itself, is a procedure that can improve the performance of bioactive molecules while pursuing the aims of sustainability and circular economy (Clemente et al. 2018;Clemente et al. 2019;Savy and Cozzolino 2022). All these factors should be always taken into consideration when largescale applications are involved . In this context, lignin and tannins are good candidates to build sustainable and eco-friendly nanoparticles for delivering bioactive substances to plants (Sipponen et al 2019;Yu et al. 2019;Lima et al. 2021). They are both phenolic compounds; the former is the main component of the xylem secondary wall (Vanholme et al. 2010), and one of the most abundant polymers on earth, while tannins are produced by plants mainly for protection against biotic stress (Pizzi 2008) and are considered good antibacterial (Glaive et al. 2017) and antifungal agents (Glazer et al. 2012). Polymer nanotechnology acquires added value when also the encapsulated molecule is a natural compound, generally more biocompatible with respect to the conventional pesticides produced by chemical synthesis (Cantrell et al. 2012). Among the natural products, neem oil (extracted from Azadirachta indica Juss. leaves or seeds) is an ideal pesticide (Hashmat et al. 2012;Benelli et al. 2017), environmentally friendly and not toxic for the users. It contains more than 300 bioactive molecules (Pascoli et al. 2019); the most important of which is azadirachtin, a secondary metabolite extracted from neem seeds (Chaudhary et al. 2017). Azadirachtin can act also as an insect repellent substance and has toxic properties since it can induce sterility by preventing oviposition and sperm production (Nisbet 2000). Recently, it has been reported that propyl disulfide contained in the volatile fraction of neem seed extracts can exert antifungal activity (Khan et al. 2021;Singh et al. 1980), though its use in agriculture is limited by poor bioavailability and low persistence (Kumar et al. 2018;Shah et al. 2017). These drawbacks can be overcome by encapsulation, as it has been verified for other poorly soluble or volatile compounds (Seibert et al. 2019;Menicucci et al. 2021).
The search for new strategies to deliver natural products is particularly relevant for grapevine protection since this crop requires a very high input of chemicals for disease control, both in Integrated Pest Management and in Organic production (Pertot et al. 2017;Seibert et al. 2019;Di Marco et al. 2022).
Here we report on the design, preparation and physicochemical characterization of a new nanoformulation based on mixed tannins-lignin particles loaded with neem oil (nanocapsules, NC). The starting polymers were selected by knowledge acquired from previous works, which have shown that lignin provides a suitable shell for building nanovectors (Pereira et al. 2022;Machado et al 2020;Falsini et al. 2019) and by supposed affinity of grapeseed tannins with the plants to be treated, i.e. mainly grapevine. The efficacy of these carriers to improve the antifungal activity of neem oil was evaluated by in vitro administration to three fungal pathogens of grapevine, Botrytis cinerea, Phaeoacremonium minimum and Phaeomoniella chlamydospora. The first of these fungi causes Botrytis bunch root, a major disease of grapevines in temperate climates worldwide, which can lead to extensive economic loss by grape rotting and biochemical changes in the final production (Jacometti et al 2010). P. minimum and P. chlamydospora (Baloyi et al. 2018) are ascomycete fungi, which are considered among the primary responsible for Esca, a widespread grapevine trunk disease (GTD) (Guerin-Dubrana et al. 2019;Bertsch et al. 2013). Both pathogens can induce in young vines discoloration of the wood, besides chlorosis, stunted leaves, short internodes, and lack of vigour in young grapevines (Díaz and Latorre 2014).
The aim of this paper is to show that lignin and tannins, either alone or mixed in different ratios, are able to form stable nanovectors with size in the suitable range for drug delivery, providing a new route to fabricate cost-effective and eco-friendly carriers for delivery active principles to plants. Moreover, we loaded these vectors with neem oil and investigated their efficacy against pathogens to understand if they can preserve antifungal action, possibly enhancing the effect of this essential oil in large scale administration.

Experimental materials
Kraft lignin and acetone were purchased by Sigma-Aldrich. Grape seed tannins and neem oil were kindly provided by Natural-mente srl (www. natur al-mente srl. it). Potato destrose agar and the agar powder were purchased by Liofilchem. Pindarus 25 WDG and Idrorame Flow were purchased by Chimiberg.

Preparation of NCs
Lignin (L) and grape seed tannins (T) were dissolved separately or in combination (as shown in Table 1), in KOH solutions, at two different pH values (13.5 and 11.5). 300 µL of mixed neem oil and acetone solution 1:1 (v/v) was added respectively to 3 mL of polymers aqueous solutions. The choice of these pH values was motivated by the low solubility of lignin in water at acidic or neutral pH ). High-power sonication was used to emulsify the oil phase in the solution. Lignin and tannins empty formulations were sonicated as well as NCs. For this purpose, a Branson 450 Digital Sonifer was used at 95% of power, for 3 min, 1 s of pulse on and 0.5 s of pulse off (5 cycles for each sample).
The acronyms of the samples investigated in this study are listed in Table 1, where composition and starting/final pH are reported.

pH measurement
The BioClass XS pH 8 + DHS pHmeter (www. biocl ass. it) was used to measure the starting and the final pH of nanoformulations.

Dynamic light scattering (DLS)
DLS measurements were performed on a Malvern Zetasizer Nano ZS (ZEN 1600 model, Malvern Instruments Southborough, MA, USA), equipped with He-Ne 633 with backscattering detection. DLS experiments were performed over 11 runs and in duplicate. Samples were diluted at 1:200 with Milli-Q water before measuring to adjust turbidity.

Zeta potential (ZP)
ZP measurements were performed on a Zatasizer (Zetasizer Pro, Malvern Panalytical Co., Ltd., Malvern, UK) in DTS1070 cells, at 25 °C. The ZP was measured using phase analysis light scattering (M3-PALS). Samples were diluted at 1:100 with Milli-Q water. The measurement was repeated 3 times, and values are presented as the mean ± SD.

Antifungal activity
Isolates of various fungal species were selected to carry out a first screening on the potentialities of the new formulation in in vitro tests. The fungal isolates, all deposited in the fungal collection of DAGRI-Plant pathology and entomology section, had been obtained from grapevine and associated with different symptoms (Table 2). Conidial germination assays for three different fungal species Botrytis cinerea, Phaeomoniella chlamydospora, Phaeoacremonium minimum were performed following the  Fig. 1. The fungal colonies were grown on potato dextrose agar (PDA). Plates with growing colonies were flooded with 2 mL of sterile deionized water and conidia were released using a sterile inoculation loop. The number of conidia in the stock suspension was counted with a Burker chamber and diluted to give 1 × 10 5 conidia per mL suspension.
Conidia suspensions were mixed with PDA in order to have a final conidia concentration in the growth medium of 1 × 10 6 conidia/mL, after lowering the temperature to 30 °C. Moreover, to avoid overheating of the essential oil, nanoformulations were applied to the surface in drops after medium solidification.
The solutions under evaluation included the natural compounds alone and nanocapsules, formulated with grape seed tannins and lignin, as polymer of the shell, and neem oil as the main component of the core.
The samples were tested at different dilutions, as shown in Table 3. For each sample, the assays were conducted in triplicate.
The toxicity test was conducted by maintaining the fungi at 25 °C until they covered all the plate surfaces, alternating 12 h in darkness, and 12 h under cool fluorescent light. The efficacy was evaluated as the size and intensity of the no-growth halo formed around the application area.  Every experiment was repeated twice to check reproducibility.

Statistical analysis
The experiment was conducted in triplicate for each samples used. One-way ANOVA was used to check the significance of differences (at least p < 0.05) among means, using GraphPad Prism 7 (GraphPad Software, San Diego, CA, USA). An Honestly Significant Differences (HSD) -Tukey test was run for post hoc comparisons to determine differences among treatments on the Inhibition Zone Diameter of the mycelium after conidial germination.

Samples description
The formulations used in this work consisted of lignin and grapeseed tannins (alone or in combination) as nanocapsule shell components and neem oil as cargo. Phenolic polymers are in general scarcely soluble in aqueous media; therefore, to facilitate their solubilisation, basic pH values are used. In particular, we found that Kraft lignin could be completely dissolved only at pH > 13. Thus, we chose to work with starting solutions of this polymer at two different pH values that were 13.5 and 11.5, in order to evidence the effect of lignin solubility on the final system. The results obtained at starting pH lower than 11.5 are not reported in this work, since the formulations obtained from these solutions were not stable enough. Tannins could be dissolved in water at the concentration used herein for the preparation of nanocapsules, i.e. 1-5% (w/w), though we chose to maintain the same starting pH as for lignin formulation to allow direct comparison. The pH measured for the solution of tannins in water was 5.0. Pure neem oil has neutral pH. Table 1 shows the composition of the samples used in this work together with their final pH. For the sake of clarity, samples obtained from polymers only, without incorporating neem oil were named "empty". It is worth noting that the final pH values for all NCs formulations (i.e. aggregates encapsulating neem oil) were close to neutrality, which is a good requisite for their applicability to plants (Sing et al. 2021).

Dynamic light scattering (DLS)
The mean diameter and polydispersity index of the lignin/ tannins nanoparticles prepared in this work are reported in Table 4. The size of all NCs prepared at starting pH 11.5 was between 280 and 320 nm while the size of the NCs at pH 13.5 was between 160 and 280 nm. In the formulations prepared at starting pH 13.5, the mean diameter decreased proportionally to the increase in amount of tannins in the NCs. This suggested that in this pH condition, i.e. when the lignin acidic groups were fully deprotonated, mixing tannins with lignin favoured the formation of more tightly packed nanocapsules with respect to the aggregates prepared at starting pH 11.5, for which the trend of the mean diameter was not monotone, indicating a less specific interaction between the polymers. The nanoaggregates obtained by 100% tannins at starting pH 11.5 (sample NCs 5T_11) were unusually large, indicating that their morphology did not correspond to the classical "capsule-like" geometry. Aggregates with worm-like shape could be formed in these conditions. The polydispersity index (PDI) of most NCs samples was in the range of 0.01-0.26, corresponding to low polydispersity. However, the aggregates formed without oil (i.e. those present in sample denominated "empty") were much larger and more polydisperse with PDI ≥ 0.78. As a general remark, it can be observed that the initial pH of the polymer solution did not affect strongly the size distribution, which could be ascribed to the fact that tannins do not change their protonation status and consequent solubilization, in the range of the high pH values used for our preparations.

Zeta potential (ZP)
The surface of all the NCs investigated in this work was negatively charged. Specifically, the ZP values for the samples at initial pH 11.5 were around − 50 mV except for NCs 5 T while the ZP of samples prepared at starting pH 13 was around − 80 mV, as shown in Table 5. To be noticed that NCs5T_13 was the most negative of all formulations with ZP of − 86 mV, concomitantly with the smallest mean size. On the other hand, the least negative ZP was measured for NCs 5T_11. This property was correlated with the largest average diameter (1070 nm ± 50), suggesting that in these aggregates the tannin polymer chains were loosely packed and possessed low surface charge density. The NCs 5T_11 formulation was also the sample with lowest final pH. We could thus infer that slightly acidic solution partially neutralized the surface charge and promoted particle growth.

Antifungal activity assay
The fungal sensitivity to antimicrobial agents was evaluated through the appearance and extension of a zone around the drop where no fungal growth was recorded. The diameter of this inhibition zone (IZD) grew larger with increasing antifungal activity. After 7 days, conidia germinated and the mycelium expanded uniformly on the dishes, depending on the sensitivity to commercial antimicrobial agents and nanoformulations. Table 6 reports the antifungal activity of reference (i) commercial fungicides (Pindarus 25 WDG and Idrorame Flow), (ii) the activity of the single tested natural compounds (pure neem oil and "empty" nanoaggregates), and (iii) the nanoformulations, on selected fungal pathogens: B. cinerea, P. minimum and P. chlamydospora. Pindarus 25 WDG is a broad-spectrum sterol inhibitor synthetic fungicide, tebuconazole, that was able to inhibit the growth of all the three selected fungi. Idrorame Flow is another wide spectrum copper-based fungicide (15.2% copper). Regarding the natural compounds, neem oil showed inhibitory activity only in its undiluted status that is in the form of water emulsion (78% w/v). This activity was recorded only on P. minimum as no effect on both B. cinerea and P. chlamydospora was visible. Grape seed tannins and lignin did not exert antifungal activity when administered as empty nanoparticles.
Concerning the different pathogens, the most sensitive fungus to the commercial products and the various nanoformulations tested was P. minimum (Fig. 2), compared to the negative control (i.e. water) ( Fig. 2A). Idrorame Flow surprisingly did not show any significant antifungal activity against the chosen pathogens (Fig. 2B) while Pindarus 25 WDG reduced the P. minimum growth whose IZD was 2.68 cm ± 0.2 (Fig. 2C). Among the nanoformulations, the effect of pure neem oil (Fig. 2D) was significantly enhanced  Table 6 In vitro antifungal activity of commercial products, pure neem oil, empty nanoaggregates and NCs (see Table 1  by both NCs_5L_11 and NCs_LT50%_11 producing a IZD of 1.8 cm ± 0.1 and 2.7 cm ± 0.2, respectively. In particular, NCs 4L_1T_11 significantly increased the neem oil antifungal activity enlarging the inhibition of the cargo alone (IZD = 3.4 ± 0.1) at concentration ≥ 1.5% w/v. NCs 5L_13 (Fig. 2F) produced a significantly larger inhibitory zone at 1.13% w/v, i.e. at a concentration eighty times lower compared to the treatment with the neem oil alone. The inhibition area induced by NCs 5L_13 in these conditions had a diameter of 3 cm ± 0.1, which is more than double with respect to the inhibition area induced by the pure neem oil (IZD: 1.43 cm ± 0.2). Moreover, at the two concentrations of 0.45%w/v and 0.23%w/v, the obtained IZDs are 3 cm ± 0.1 and 1.9 ± 0.6, respectively. This indicated that the efficacy of neem oil as antifungal agent was boosted by ~ 200 and ~ 400 times, respectively, by administration in the form of nanocapsules (Fig. 3).

Discussion
The results obtained in DLS experiments show that both lignin and tannins are suitable polymers for neem oil encapsulation and delivery. The sizes of neem-loaded particles were mostly in the range of 200-300 nm, which is an appropriate size for nanocarriers to be used for delivery of bioactive compounds ). As mentioned above, the nanocapsules dissolved at higher initial pH (13.5) were smaller than those dissolved at lower initial pH (11.5). Regarding the size distribution, almost all samples had low PDI, that is equal to or below 0.26, suggesting that the obtained nanocapsules were monodisperse in size, with the exception of NCs 5T_11. These results were reproducible over three different preparations, confirming that the procedure led to well defined nanostructures, in spite of their complex polymeric and compartmentalized nature. The markedly negative surface charge measured for all lignin and tannin nanocapsules, besides helping to maintain the aggregates dispersed in solution should be considered as a favourable property for applications, since the presence of a positive surface charge has been associated with pronounced toxicity toward many living organisms (Weiss et al. 2021). Thus, nanoformulations were tested in antifungal assays in vitro that showed promising results against P. minimum after 7 days, inhibiting spores (conidia) germination and mycelium growth. In particular, among the various nanoformulations, the lignin-based vectors (i.e. fully composed of lignin and partially mixed with tannins) showed the highest delivery efficacy, implementing the antifungal effect of neem oil on P. minimum and reducing its effective dose up to 400 times compared to the control. This finding could be explained by the fact that lignin is able to mask the bioactive compound like a "Trojan horse", facilitating the uptake by fungal mycelium (Fischer et al. 2019;Peil et al. 2020).
Nanocapsules composed by tannins only did not exhibit any fungal growth inhibition capability, possibly due to their loosely packed structure and large size (700 nm-1 µm), but also to a pronounced "adaptive" compatibility developed by the fungi themselves toward tannins, which can invalidate their well-documented protective effects on plants (Ekambaram et al. 2016;Zhu et al. 2019). One of the most efficient formulations is NCs 5L_13 which had the smallest dimension: 150 nm. NCs 5L_13 induced an IZD even bigger than NCs5L_11, which showed, on the contrary, an average diameter of approximately 310 nm. Overall, the most successful nanoformulations were those prepared at starting pH 11.5, which had a final pH of approximately 6, similar to the xylem environment where P. minimum usually resides and proliferate producing conidia (Feliciano and Gubler 2001). In view of possible applications in agriculture, the lack of pH variation from the laboratory preparation to the infected tissue is not a limiting factor since cargo release depends on the activity of the enzymes (i.e. laccases) produced by these fungi themselves, Fig. 3 Inhibition zone diameter (cm) of P. minimum exposed to neem oil (7 g/L w/v) compared to the other formulations. Letters close to the histogram indicate the significant differences among the treatments according to the Tukey's test (at least p < 0.05) meanings that NCs, once injected into the trunk of the infected plants can be delivered upwards via xylem and downward by phloem (Karny et al. 2018;Tripathi et al. 2017) but drug release occurs precisely where the fungus is located. This allows to control the drug release by exploiting one of the main pathogen features (i.e. laccase production) maintaining plant biocompatibility and security without introducing further toxic substances, thus opening the way to new disease control strategies.

Conclusions
In this work, we showed that nanoformulations prepared from natural polyphenols of plant origin (lignin and tannins), are able to form well-defined nanoparticles, either alone or in mixture. These nanoparticles can be used as carriers for the bioactive neem oil, proving the advantages of administration through finely dispersed formulations, which are mainly enhanced surface/volume ratio and sustained release. More in detail, we found that while the oil-free systems (i.e. those obtained by using only polyphenols) showed large and polydisperse aggregates, the oil-loaded particles adopted a globular structure where the hydrophobic component (neem) was incorporated. This resulted in a typical emulsionlike configuration with polymers in the shell and oil in the core. Both empty and neem-loaded nanoaggregates were negatively charged, as evidenced by Zeta potential measurements.
In conclusion, lignin-based nanocapsules were shown to be efficient delivery agents for neem oil, reducing the effective dose by ~ 400 times on P. minimum. This is a very interesting result, but not surprising, since the interfaces available for the contact between neem oil and the pathogens are enhanced by orders of magnitude as a consequence of nanodelivery and if a biocidal potentiality exists, it can be magnified accordingly.
The activity showed on P. minimum can open perspectives and deserves to be further explored by in planta studies with this or other pathogens.