3.1 Preparation and characterization of PEI-AuNPs
To synthesize PEI-AuNPs, two sizes of PEI (10 kD and 25 kD) were mixed up with HAuCl4·3H2O with the additional ammonium hydroxide. PEI ligands provide a high density of positively charged amino functional groups to improve stability and dispersion in gold nanoparticle synthesis [30, 31]. The pH of the gold precursor solution has a strong influence on the final morphology of gold nanoparticles, and alkaline conditions favour the formation of uniformly sized Au nanoparticles [32]. Ammonium hydroxide enhances the reduction reaction and provides an alkaline environment for the synthesis of PEI-AuNPs, resulting in the fluorescence of PEI-AuNPs. Therefore, we obtained two types of uniformly dispersed spherical PEI-AuNPs (10 kD and 25 kD) (Fig. 1a). The original color of 10 kD PEI-AuNPs was magenta, whereas 25 kD PEI-AuNPs displayed darker magenta color (Fig. S1a). Next, the characteristics of PEI-AuNPs (10 kD and 25 kD) were measured by UV-vis absorption spectra, transmission electron microscopy (TEM), dynamic light scattering (DLS), and zeta potentials (Fig. 1, Fig. S1b and Fig. S2). Interestingly, the two types of PEI-AuNPs displayed almost the similar UV absorption pattern (Fig. S1b), by which the absorbance peaks of 10 kD PEI-AuNPs were at 346 nm and 551 nm and that of 25 kD PEI-AuNPs were at 353 nm and 523 nm. Both PEI-AuNPs were uniformly sized and dispersed, while the diameter of 25 kD PEI-AuNPs was significantly larger than that of 10 kD PEI-AuNPs (8.08 ± 0.24 nm vs 7.33 ± 0.25 nm) (Fig. 1b and Fig S2a-S2b). Although both PEI-AuNPs were positively charged and the hydrodynamic diameters of two types of PEI-AuNPs were not statistically different (Fig. S2c), the zeta potential value of 25 kD PEI-AuNPs was 49.87% higher than that of 10 kD PEI-AuNPs (Fig. S2d) with the value 15.53 ± 2.63 mV and 28.86 ± 3.55 mV for 10 kD and 25 kD PEI-AuNPs, respectively.
Considering PEI ligands that contain positively charged amine groups bind negatively charged nucleic acid through electrostatic adsorption and the greatest positive zeta potential of PEI-functionalized gold nanoclusters allows the highest loading capacity of nucleic acid carriers [21, 33, 34], 25 kD PEI-AuNPs (represented by PEI-AuNPs) was used for siRNA loading and downstream internalization experiments. Before siRNA loading, the core structure of the PEI-AuNPs was observed by the high-resolution transmission electron microscopy (HRTEM) image, which showed that PEI-AuNPs had a lattice spacing of 0.23 nm (Fig. S3). The X-ray diffraction (XRD) pattern of the PEI-AuNPs was also recorded in the range of 10–80 °. The lattice plane (111) was dominant and had a much lower ratio between the intensities of the diffraction peaks above the lattice planes (200), (220), and (311) (Fig. S4). Next, PEI-AuNPs were infiltrated into Arabidopsis leaves to monitor their biocompatibility. After five days of continuous observation, plants infected with PEI-AuNPs grew as healthy as the control groups (Fig. S5), suggesting that PEI-AuNPs exhibit an extremely low level of toxicity and are relatively safe for plants.
To verify whether PEI-AuNPs and nucleic acid molecules, such as siRNA, could form a complex and examine their optimal binding ratio, we assessed the loading efficiency of siRNA by PEI-AuNPs. The binding of siRNA on PEI-AuNPs did not affect the morphology, dispersion, and homogeneity of PEI-AuNPs. The average diameter of PEI-AuNPs-siRNA (8.97 ± 0.24 nm) was marginally statistically larger than that of PEI-AuNPs (Fig. 1c) and the hydrodynamic diameter of PEI-AuNPs-siRNA (12.06 ± 1.18 nm) was 4 times larger than that of PEI-AuNPs (61.17 ± 8.68 nm) (Fig. 1d), indicating loading of RNA was successful and the formation of a supramolecular complex [21]. Consistently, siRNA loading on PEI-AuNPs decreased the zeta potential of PEI-AuNPs-siRNA, but PEI-AuNPs-siRNA remained positively charged (Fig. 1e). However, siRNA loading had no influence on the optical properties of PEI-AuNPs, as reflected by the fluorescence spectra and UV-vis spectra (Fig. 1f, S6). Moreover, agarose gel electrophoresis showed that, in contrast to siRNA alone, PEI-AuNPs and siRNA were able to form a complex so that siRNA did not move in the positive direction of the electrophoresis tank (Fig. 1g). The binding ratio between PEI-AuNPs and siRNA was further measured, by which the ratio of 3: 7 (v: v) allowed the maximal loading of siRNA and was therefore used for subsequent experiments. Overall, though the fluorescence spectra, zeta potential, and TEM images of PEI-AuNPs before and after siRNA loading, siRNA could tightly bind to PEI-AuNPs without altering the morphology, fluorescence property, and colloidal stability of PEI-AuNPs.
3.2 Cy3-ssDNA-loaded PEI-AuNPs internalized into mature plant cells
To evaluate the efficiency of nucleic acid molecules internalization into the mature plant cells through PEI-AuNPs, Cy3 labeled ssDNA oligos were first loaded onto PEI-AuNPs to detect their penetration into epidermal and mesophyll cells. Numerous studies have been conducted on nanomaterials carrying nucleic acids, primarily examining their entry into epidermal cells [35], except for their penetration into mesophyll cells, the crucial functional units in plants. Specifically, 10 mM of Cy3-ssDNA was bound to PEI-AuNPs by a typic 7: 3 (v: v) ratio. After infiltration into the Arabidopsis leaves, the infected plants were placed in the dark for further incubation for about 3 h to avoid the fluorescence quenching of either PEI-AuNPs or Cy3-ssDNA [36, 37]. Next, the confocal microscopy observation showed that both free Cy3-ssDNA and PEI-AuNPs trapped Cy3-ssDNA (AuNPs-Cy3-ssDNA) were localized at leaf epidermal cells, whereas the fluorescence of AuNPs-Cy3-ssDNA was uniformly distributed across the entire leaf surface rather than those free Cy3-ssDNA (Fig. 2a). This is consistent with the fact that PEI-AuNPs enter into tobacco leaves [21, 38]. In addition to the epidermal cells, the fluorescence signal was observed in the mesophyll cells as well, where the Cy3 signal was found to be closely associated with chlorophyll (Fig. S7). This finding suggests that PEI-AuNPs may facilitate the entry of nucleic acids into the functional cells of plants.
In addition, the interactions between nanoparticles, Cy3-ssDNA, and plant cells were further explored utilizing the fluorescence properties of PEI-AuNPs. Indeed, PEI-AuNPs exhibit enhanced penetration capability, therefore allowing PEI-AuNPs-Cy3-ssDNA to easily enter mesophyll cells compared with the control group of Cy3-ssDNA (Fig. 2b). This finding further supports the notion that PEI-AuNPs significantly enhance the efficiency of nucleic acid delivery into plants. Next, the amount of PEI-AuNPs present in Arabidopsis leaves was quantified and a calibration curve was established. By measuring the fluorescence intensity of various concentrations of PEI-AuNPs in Arabidopsis leaf extracts, a linear equation y = 95914x + 4355.9 with an R2 value of 0.9978 was established (Fig. 2c). The accuracy of this equation was also assessed with the fact that the actual internalization of PEI-AuNPs per leaf was 0.02 g, while the theoretical internalization per leaf was calculated to be 0.0324 g. Thus, this results in an internalization efficiency of 61.73% for the Arabidopsis leaves.
3.3 PEI-AuNPs-siAtWRKY1 silenced AtWRKY1 to improve plant disease resistance
Since nuclear acids easily entered plant cells with the help of PEI-AuNPs, we next sought to evaluate whether these molecules could be functional by loading AtWRKY1 siRNA. The AtWRKY1 gene has been discovered to have a negative impact on plant resistance to Pst DC3000. AtWRKY1 loss-of-function mutants have impeded the pathogen's propagation, suggesting targeting the suppression of the AtWRKY1 gene appears to be an effective strategy for enhancing plant disease resistance. To dissect the relationship between gene silencing and plant pathogen resistance mediated by PEI-AuNPs, AtWRKY1 siRNA was loaded on PEI-AuNPs by the similar binding ratio as before. Next, Arabidopsis leaves were treated with control buffer, PEI-AuNPs, PEI-AuNPs-siAtWRKY1, and siAtWRKY1, then sampled for 0, 1, 2, and 3 days after growth, respectively, and the total RNA was extracted from the leaves to determine the gene expression of AtWRKY1. The results indicated that there was no significant change in AtWRKY1 expression at 0 and 1 day after treatments (Fig. 3a and 3b). However, after 2 days of treatments, the PEI-AuNPs-siAtWRKY1 group showed a significant reduction in AtWRKY1 expression compared to the control groups (Fig. 3c). The siAtWRKY1 treatment led to a slight decrease, but it was not statistically significant (Fig. 3c). The maximum efficiency of AtWRKY1 silencing was observed on the third day, resulting in an impressive 80% decrease in AtWRKY1 gene expression. In contrast, siRNA alone displayed no effect at all (Fig. 3d).
The maximum efficiency of AtWRKY1 silencing implies that this phase is optimal for testing plant resistance against Pst DC3000. To investigate the impact of silenced AtWRKY1 on disease susceptibility in Arabidopsis, we conducted an experiment where Arabidopsis leaves were infected with Pst DC3000. Prior to the infection, the leaves were pre-treated for three days with a control buffer, PEI-AuNPs, PEI-AuNPs-siAtWRKY1, and siAtWRKY1. Subsequently, the infected area was inoculated with Pst DC3000 with an OD600 of 0.001. After three days, the morphological, physiological, and disease susceptibility of the infected leaves were evaluated. Our results showed that Arabidopsis leaves treated with PEI-AuNPs-siAtWRKY1 exhibited minimal disease area (Fig. 4a) and unaffected leave growth (Fig. 4b), with a significantly increased content of total chlorophyll (Fig. 4c). The chlorophyll index was monitored for three consecutive days following Pst DC3000 infection. It was observed that the chlorophyll index continued to decrease as the infection time prolonged (Fig. S8), while treatment with PEI-AuNPs-siAtWRKY1 significantly helped maintain the chlorophyll index (Fig. S8). Furthermore, the number of colonies in infested leaves was quantified by extracting Pst DC3000. Initially, the extract solution was incubated at LB solid medium containing rifampicin to count the number of colonies right after Pst DC3000 infestation. No difference was observed between the treatments, ruling out the bacteria's effect on disease resistance (Fig. 4d and S9a). The colonies were confirmed to be Pst DC3000 using specific 16 s primers (Fig. S9b). Interestingly, after three days of infection, the log10 of bacterial growth number was significantly lower in the PEI-AuNPs-siAtWRKY1 treatment (Fig. 4e), as observed in the serial dilution (Fig. 4f). Collectively, these results indicate that pre-treatment with PEI-AuNPs-siAtWRKY1 enhances plant resistance to Pst DC3000.
3.5 PEI-AuNPs-siAtWRKY1 leads to a reduction of H2O2, changes of antioxidant enzyme activity system
Pathogen infections like Pst DC3000 cause a burst of reactive oxygen species (ROS) in plants, resulting in severe growth damage of infected leaves [39]. Extensive studies have shown that delivering ROS-scavenging nanomaterials to plants are able to alleviate abiotic stress [40–42], such as salt and heat stress. These materials are designed as nanozymes possessing ROS removing enzymatic activity. PEI-AuNPs were confirmed to contain ability to scavenge H2O2 in vitro with a scavenging rate of 24%, and siAtWRKY1 did not affect the scavenging ability of PEI-AuNPs (Fig. S2). To further investigate whether PEI-AuNPs delivery system contributes plant defense response by regulating ROS system, the level of hydrogen peroxide (H2O2), superoxide anion (˙O2−) and ROS scavenging system by antioxidant enzyme were measured. H2O2 level was visualized by 3,3'-diaminobenzidine (DAB) staining. DAB is a stain that reacts with H2O2 in the presence of certain proteins, such as peroxidases, producing a dark brown precipitate. It is thus used to detect the presence and distribution of H2O2 in plant cells. Interestingly, only treatment with PEI-AuNPs-siAtWRKY1, but not PEI-AuNPs, resulted in a decrease in DAB staining (Fig. 5a and S10) and total content of H2O2 (Fig. 5b). However, the measurement of superoxide anion (˙O2−) by nitro blue tetrazolium (NBT) straining and the content analysis revealed no significant difference among all treatments (Fig. 5c and 5d). These results suggest that the impact of PEI-AuNPs-siAtWRKY1 primarily lies in its effect on the H2O2 level, and PEI-AuNPs mediated AtWRKY1 gene silencing leads to reduced H2O2.
Besides, ROS scavenging system by antioxidant enzyme activity is also important for plants survival [43]. In our study, we measured the activities of SOD, POD, and CAT enzymes, and found that only CAT activity (Fig. 5e) increased by PEI-AuNPs-siAtWRKY1 treatment, while SOD (Fig. 5f) and POD activities (Fig. 5g) decreased. Consistently, the total protein content was significantly higher by PEI-AuNPs-siAtWRKY1 treatment (Fig. 5h). These findings suggest that PEI-AuNPs-siAtWRKY1 probably specifically targets the CAT enzyme system.
3.5 AtWRKY1 gene silencing improves plant photosynthetic efficiency
Pst DC3000 infections frequently lead to leaf necrosis and hinder plant growth by affecting the production of photosynthetic assimilates [1]. However, the Arabidopsis leaves treated with PEI-AuNPs-siAtWRKY1 exhibited notably increased total protein content (Fig. 5h), which was likely due to the enhanced photosynthetic assimilates.
Chlorophyll fluorescence imaging is a powerful tool for discriminating photosynthetic function in leaves [44]. To assess the impact on photosynthesis, several chlorophyll fluorescence indices, such as fluorescence origin (Fo), fluorescence maximum (Fm), photochemical efficiency of PSⅡin the light (Y(II)), Nonphotochemical quenching (NPQ), photochemical quenching coefficient (qP), Non-photochemical quenching coefficient (qN), and (electron transport rate) ETR were monitored by imaging and quantification in vivo (Fig. 6a). Fo and Fm represent the minimum and maximum fluorescence intensities of plants after a sufficiently long period of dark acclimation, respectively, and PEI-AuNPs-siAtWRKY1 significantly increased the Fo and Fm of the plant (Fig. 6b and 6c). Y(II) is the actual quantum efficiency of PSII in plants under the action of light [45]. NPQ dissipates the excess light energy absorbed as heat and is an important photoprotective process [46]. PEI-AuNPs-siAtWRKY1 improved Y(II) and NPQ and the differences were statistically significant (Fig. 6d and 6e). qP indicates the proportion of open reaction centers in all PSII reaction centers [47, 48]. qN correlates with non-photochemical burst energy dissipation [49]. PEI-AuNPs-siAtWRKY1 significantly increased qP and qN (Fig. 6f and 6g), suggesting that it can enhance the ability of light energy captured by PSII in plants to be used for photochemical reactions and dissipated for non-photochemical reactions. PEI-AuNPs-siAtWRKY1 was able to increase the level of ETR (Fig. S11), which reflects the actual electron transfer rate [50]. Taken together, The PEI-AuNPs-siAtWRKY1 treated Arabidopsis leaves had better chlorophyll fluorescence compared to other groups (Fig. 6, Fig. S11), indicating AtWRKY1 gene silencing by PEI-AuNPs-siAtWRKY1 could enhance disease resistance by promoting photosynthetic performance.