The biosynthesis of silver nanoparticles (AgNPs) through the use of endophytic bacteria is a safe replacement for the chemical method. The study aimed to synthesize AgNPs using endophytic bacterium Bacillus siamensis strain C1, which was isolated from the medicinal plant Coriandrum sativum. The synthesized AgNPs with the size of 25 to 50 nm were further confirmed and characterized by UV-visible spectroscopy, Fourier transform infrared spectroscopy, X-ray diffraction, transmission electron microscopy and scanning electron microscopy with EDS profile. The synthesized AgNPs at 20 μg mL⁻¹ showed a strong antibacterial effect against the pathogen of rice bacterial leaf blight and bacterial brown stripe, while an inhibition zone of 17.3 and 16.0 mm was observed for Xanthomonas oryzae pv. oryzae (Xoo) strain LND0005 and Acidovorax oryzae (Ao) strain RS-1, respectively. Furthermore, the synthesized AgNPs significantly inhibited bacterial growth, biofilm formation and swimming motility of Xoo strain LND0005 and Ao strain RS-1. In addition, the synthesized AgNPs significantly increased root length, shoot length, fresh weight and dry weight of rice seedlings compared to the control. Overall, this study suggests that AgNPs have the potential to protect rice plants from bacterial infection and plant growth promotion.

Nanoparticles are expected to play a vital role in the management of future plant diseases, and they are expected to provide an environmentally friendly alternative to traditional synthetic fungicides. In the present study, silver nanoparticles (AgNPs) were green synthesized through the mediation by using the endophytic bacterium Pseudomonas poae strain CO, which was isolated from garlic plants (Allium sativum). Following a confirmation analysis that used UV–Vis, we examined the in vitro antifungal activity of the biosynthesized AgNPs with the size of 19.8–44.9 nm, which showed strong inhibition in the mycelium growth, spore germination, the length of the germ tubes, and the mycotoxin production of the wheat Fusarium head blight pathogen Fusarium graminearum. Furthermore, the microscopic examination showed that the morphological of mycelia had deformities and collapsed when treated with AgNPs, causing DNA and proteins to leak outside cells. The biosynthesized AgNPs with strong antifungal activity were further characterized based on analyses of X-ray diffraction, transmission electron microscopy, scanning electron microscopy, EDS profiles, and Fourier transform infrared spectroscopy. Overall, the results from this study clearly indicate that the biosynthesized AgNPs may have a great potential in protecting wheat from fungal infection.

Amongst serious biotic factors deteriorating crop yield, the most destructive pathogen of rice is Xanthomonas oryzae pv. oryzae (Xoo), which causes bacterial leaf blight (BLB) disease. This study involved targeted use of biogenic silver nanoparticles (AgNPs) to control BLB in order to cope with the disadvantages of chemical disease control. AgNPs were biologically synthesized from natively isolated Bacillus cereus strain SZT1, which was identified through 16S rRNA gene sequence analysis. Synthesis of AgNPs in bacterial culture supernatant was confirmed through UV-VIS spectroscopy. Fourier transform infrared spectroscopy (FTIR) confirmed that the existence of AgNPs was stabilized with proteins and alcoholic groups. X-ray diffraction (XRD) data revealed the crystalline nature and imaging with scanning electron microscopy (SEM) and transmission electron microscopy (TEM), showing the spherical shape of AgNPs with particle sizes ranging from 18 to 39 nm. The silver presence in AgNPs was further confirmed by energy dispersive spectra. Biogenic AgNPs showed substantial antibacterial activity (24.21 ± 1.01 mm) for Xoo. In a pot experiment, AgNPs were found to be effective weapons for BLB by significantly increasing the plant biomass with a decreased cellular concentration of reactive oxygen species and increased concentration of antioxidant enzyme activity.

Most endolysins of dsDNA phages are exported by a holin-dependent mechanism, while in some cases endolysins are exported via a holin-independent mechanism. However, it is still unclear whether the same endolysins can be exported by both holin-dependent and holin-independent mechanisms. This study investigated the lysis system of OP2-like phage X2 infecting Xanthomonas oryzae pv. oryzae, causing devastating bacterial leaf blight disease in rice. Based on bioinformatics and protein biochemistry methods, we show that phage X2 employs the classic "holin-endolysin" lysis system. The endolysin acts on the cell envelope and exhibits antibacterial effects in vitro, while the holin facilitates the release of the protein into the periplasm. We also characterized the role of the transmembrane domain (TMD) in the translocation of the endolysin across the inner membrane. We found that the TMD facilitated the translocation of the endolysin via the Sec secretion system. The holin increases the efficiency of protein release, leading to faster and more efficient lysis. Interestingly, in E. coli, the expression of either holin or endolysin with TMDs resulted in the formation of long rod shaped cells. We conclude that the TMD of X2-Lys plays a dual role: One is the transmembrane transport while the other is the inhibition of cell division, resulting in larger cells and thus in a higher number of released viruses per cell.

Silver nanoparticles (AgNPs), synthesized with plant materials, are considered to be an emerging field of agriculture for their eco-friendly and outstanding antibacterial attributes. In this study, we synthesized AgNPs using pomelo (Citrus maxima) fruit extract as a biological capping and reducing material. The particle size was determined as 11.3–12.8 nm by using UV–vis spectrophotometer, TEM and x-ray diffraction analysis. UV–vis spectrophotometer analysis also confirmed the formation of AgNPs in colloidal solution and showed a maximum absorption at 426 nm. Fourier transform infrared spectra was used to analysed the involvement of biological molecule in AgNPs synthesis. The minimum inhibitory concentration of AgNPs against Acidovorax oryzae strain RS-2 was determined as 25 μg ml⁻¹ by agar well diffusion and bacterial growth assay. In addition, bacterial viability and swarming motility were significantly inhibited by AgNPs. Compared with the control, 25 μg ml⁻¹ of AgNPs lower bacterial biofilm formation up to 68.24%. The bacterial cell wall damaged by AgNPs was observed t TEM. Furthermore, AgNPs treatment resulted into the down regulation of expression of many type VI secretion system related genes, suggesting that AgNPs also have an effect on the virulence of bacteria. The overall conclusion of this study suggests that AgNPs can play an important role in controlling A. oryzae.

Neural networks are becoming a popular tool for solving many real-world problems such as object recognition and machine translation, thanks to its exceptional performance as an end-to-end solution. However, neural networks are complex black-box models, which hinders humans from interpreting and consequently trusting them in making critical decisions. Towards interpreting neural networks, several approaches have been proposed to extract simple deterministic models from neural networks. The results are not encouraging (e.g., low accuracy and limited scalability), fundamentally due to the limited expressiveness of such simple models. In this work, we propose an approach to extract probabilistic automata for interpreting an important class of neural networks, i.e., recurrent neural networks. Our work distinguishes itself from existing approaches in two important ways. One is that probability is used to compensate for the loss of expressiveness. This is inspired by the observation that human reasoning is often 'probabilistic'. The other is that we adaptively identify the right level of abstraction so that a simple model is extracted in a request-specific way. We conduct experiments on several real-world datasets using state-of-the-art architectures including GRU and LSTM. The result shows that our approach significantly improves existing approaches in terms of accuracy or scalability. Lastly, we demonstrate the usefulness of the extracted models through detecting adversarial texts.