A comprehensive study of the hormetic influence of biosynthesized AgNPs on regenerating rice calli of indica cv. IR64

Rice is one of the most widely cultivated crops worldwide; however, it is not amenable to genetic manipulations, owing to its poor response to tissue culture and regeneration in vitro. To improve its response to tissue culture, we evaluated the influence of biosynthesized silver nanoparticles on callus induction, regeneration and rhizogenesis in Indica rice cv. IR64. Silver nanoparticles were biosynthesized by using silver nitrate and Parthenium hysterophorus plant extract, and were characterized by UV-visible spectroscopy, Fourier-transform infrared spectroscopy, Transmission electron microscopy and X-ray diffraction. The biosynthesized silver nanoparticles (PHAgNPs), when supplemented in tissue culture medium, promoted callus induction frequency, callus regeneration and rhizogenesis at concentrations of 10 mg l −1, 5 mg l−1 and 10 mg l−1, respectively. Further examination of the endogenous hormonal levels in regenerating calli revealed that AgNPs enhanced regeneration by alleviating abscisic acid and ethylene levels in the plant tissue. The stimulatory influence eliciting the regeneration response was found to be optimal with the supplementation of 5 mg l−1 PHAgNPs in the regeneration medium; the malondialdehyde, proline and hydrogen peroxide levels were also lower than those in the control, thus suggesting improved antioxidant status. Our results indicated that biosynthesized PHAgNPs may have the potential to positively influence tissue culture of recalcitrant varieties.

flowering, and it is part of a complex regulatory network that includes auxins and cytokinins and is imperative for embryogenesis and shoot regeneration. Another important PGR is ethylene, a gaseous hormone that primarily influences ripening of fruits and senescence in plants 5,6 . Studies have demonstrated that in vitro tissue culture results in accumulation of ethylene, and supplementation with silver nanoparticles (AgNPs) in plant tissue culture medium makes explants healthier and improves their growth vigour and regeneration frequency, effects attributable to the ability of silver ions to inhibit ethylene synthesis [7][8][9][10][11] . However, the influence of AgNPs on other PGRs during regeneration remains unclear.
Another study has reported that plant tissues, even when propagated under optimal conditions, produce reactive oxygen species (ROS), an unavoidable by-product of general plant metabolism that is detrimental for plant growth and development. Even though the harmful free radicals are reduced by an internal antioxidant system, the process consumes vital resources in the cells, thus hindering growth and development 12 . Silver ions, beyond functioning as ethylene inhibitors, serve as electron acceptors and donors in redox reactions, especially in supporting exchange of electrons with CO 3 + and Fe 2 + 13 , thereby reducing the ROS and alleviating the strain on the plant antioxidant system. However, compared with silver ions, AgNPs are more efficient in chemical reactions and interact better with the surrounding environment because of their higher surface area to mass ratio 7 . Hence, in the present study, we attempted to correlate the influence of exogenously supplemented AgNPs on endogenous ROS as well as PGR levels in regenerating calli, to better understand the influence of AgNPs on plant development. AgNPs, because of their small size (1-100 nm) possess unique optical and physiochemical properties, and thus are used in various fields for conduction, biological detection, catalysis, wound healing, anti-microbial activity and phytostimulation 14 .
Synthesis of AgNPs through chemical and physical methods requires toxic chemicals and complex purification steps. However, AgNPs can be synthesized by utilizing plant extracts, a process that is simple, economical and eco-friendly. Parthenium hysterophorus (PH), used for bio-fabrication of AgNPs in the present study, is one of the most difficult weeds to control worldwide. It is toxic to animals and harmful to biodiversity, and it is responsible for economic losses in agriculture. Control measures such as burning the weeds, spraying chemical herbicides, and introducing pests, mycoherbicides, and competitive crops each have their own constraints. However, utilizing PH plants for biosynthesis of nanoparticles (NPs) is a way to put these weeds to good use 15,16 .
In the present study, we used a bio friendly route to synthesize NPs by using PH plant extract, an environmentally benign process resulting in more stable and biocompatible NPs, as compared with the chemical reduction process, wherein harsh chemicals adsorb onto the surfaces of NPs and render them toxic 17 . The biosynthesized silver NPs positively influenced callogenesis, regeneration and rhizogenesis in commonly grown Indica rice cv. IR64 plants in vitro. Because plant hormones play a crucial role in the sensing ability and adaptability of plants 18 , we correlated the hormonal profile with the regeneration efficiency of the calli exposed to PHAgNPs. The PHAgNPs were found to enhance regeneration efficiency by decreasing ethylene, ABA and ROS levels in the regenerating calli.

Results
Synthesis and characterization of PHAgNPs. The biosynthesis of PHAgNPs was initially confirmed visually, as the reaction mixture of 1 mM AgNO 3 and plant extract changed colour from pale yellow to deep brown following incubation (Fig. 1). The colour change was attributed to surface plasmon resonance caused by collective oscillation of free conduction electrons 14 . When subjected to UV-vis spectroscopic analysis, the PHAgNPs produced an absorption peak at 420 nm, thereby confirming the stability of synthesized PHAgNPs (Fig. 2a).
Transmission electron microscopy (TEM) indicated that the PHAgNPs were mostly spherical in shape and poly dispersed, with sizes ranging from 1 to 50 nm (Fig. 2b). Fourier-transform infrared spectroscopy (FTIR) spectroscopy was carried out to identify the functional groups from PH involved in the reduction of Ag + and capping of silver NPs. The spectrum bands were observed at 3434.94 cm −1 , 2850.78 cm −1 , 2076.86 cm −1 , 1403.54 cm −1 , 1124.20 cm −1 , 832.08 cm −1 , 764.73 cm −1 and 566.97 cm −1 , corresponding to O-H stretching in an alcohol, C-H stretching in an alkane, N=C=S stretching in isothiocyanate, O-H bending in carboxylic acid, C-O stretching in an aliphatic ether, C-C bending in an alkene, C-H stretching and C-I stretching in a halo compound, respectively. The results confirmed the involvement of plant extract in the bio-reduction of Ag + ions and encapsulation of Ag NPs (Fig. 2d).
X-ray diffraction (XRD) was used to characterize the crystalline nature of the PHAgNPs. After reduction, well resolved Bragg reflections for PHAgNPs were obtained at 2θ = 38.2°, 44 (Fig. 2c). The average size of NPs was determined to be ~25 nm with Debye-Scherrer's equation.

Effects of AgNPs on callus induction, regeneration and rooting in vitro.
The embryogenic calli generated from mature seeds displayed a variable response towards different concentrations of PHAgNPs in the callus induction medium (Fig. 3a).
The callus induction frequencies of rice with respect to different concentrations of PHAgNPs are given in Table 1. The callus induction response was highest with PHAgNP concentration of 10 mg l −1 with callus induction frequency of 82%, followed by 69% with 5 mg l −1 PHAgNPs, 37% with 15 mg l −1 PHAgNPs and 16% with 20 mg l −1 PHAgNPs, relative to control medium devoid of PHAgNPs, which displayed a callus induction frequency of 62% (Table 1). At lower concentrations of PHAgNPs, the calli had a friable texture and creamy appearance; however, with increasing concentrations of PHAgNPs, the calli turned brown.
The induced calli from IR64 seed explants were shifted to regeneration medium with different concentrations of PHAgNPs and control medium. Green spots formed on the calli within 14 days (Fig. 3b). The green calli were further sub-cultured onto fresh medium, and the regeneration potential was documented ( Table 1). The regeneration efficiency was observed to be highest (61%) on culture medium supplemented with 5 mg l −1 PHAgNPs, in contrast to the optimal PHAgNP concentration required for efficient callus induction. The increase in PHAgNP www.nature.com/scientificreports www.nature.com/scientificreports/ concentration decreased the regeneration frequency. At PHAgNP concentrations of 10 mg l −1 , 15 mg l −1 and 20 mg l −1 , the regeneration frequency was 40%, 18% and 4%, respectively, relative to the regeneration frequency (41%) observed on control medium (Table 1).   www.nature.com/scientificreports www.nature.com/scientificreports/ The individual regenerated plants were transferred to rooting medium. After 2 weeks, application of NPs up to only a certain concentration increased the number of roots and the root length. The greatest rooting response with healthy secondary roots and root hair was observed at a concentration of 10 mg l −1 , followed by 5 mg l −1 and 15 mg l −1 . At a PHAgNP concentration of 20 mg l −1 , the root length and number were drastically lower than those observed with control medium (Fig. 3d, Table 1).

Changes in expression of pGR genes.
Exposure of calli to PHAgNPs significantly influenced the expression of PGR responsive genes. The basic PGR responsive genes were downregulated upon exposure to PHAgNPs (5 mg l −1 and 10 mg l −1 ) as compared with control treatment, and the lowest expression was at 10 mg l −1 . The transcript levels of ethylene, ABA, auxin, cytokinin and gibberellic acid responsive genes decreased by a maximum of 0.3, 0.35, 0.2, 0.3 and 0.4 fold, respectively, after culture in medium with 10 mg l −1 PHAgNPs. However above 10 mg l −1 , the expression of all the genes markedly increased, probably because of the stress induced by increased exposure to heavy metal (Fig. 4).
Proline, H 2 o 2 and malondialdehyde levels. Under control conditions, the proline, H 2 O 2 and malondialdehyde (MDA) levels were 1.94 µM g −1 FW, 0.63 µM g −1 FW and 125.36 µg g −1 FW, respectively, and the values significantly decreased when the explants were cultured on medium supplemented with 5 mg l −1 and 10 mg l −1 PHAgNPs. The lowest levels were recorded on medium fortified with 10 mg l −1 PHAgNPs, with 0.62 µM g −1 FW, 0.36 µM g −1 FW and 72 µg g −1 FW for proline, H 2 O 2 and MDA respectively. However, the H 2 O 2 , proline and MDA levels sharply increased when the explants were exposed to 15 mg l −1 and 20 mg l −1 PHAgNP concentrations (Fig. 5).

Discussion
In the present study, the biosynthesized AgNPs were validated with FTIR, which analysed the displacement in the absorbance rate of biomolecules in PHAgNPs compared with that of biomolecules in PH plant extract. FTIR analysis confirmed the involvement of alcohol, alkane, isothiocyanate, carboxylic acid, aliphatic ether and alkene groups in the biosynthesis of AgNPs. Similarly, earlier reports have suggested that the biosynthesis involves the reduction and capping of AgNPs by saponin, free amino acids, galactose and other compounds present in the PH leaf extract [19][20][21] . The XRD pattern of PHAgNPs was determined by using JCPDS intensities (JCPDS files No. 893722), which suggested that the biosynthesized NPs have a face centred cubic structure, and the biosynthesized NPs have an average size of ~25 nm and a mostly spherical shape, as evidenced by TEM and calculations with Debye-Scherrer's equation. The results confirmed the synthesis of the PHAgNPs and were consistent with the previously reported results for phase centred cubic AgNPs 22,23 .
Plant tissue culture has become an indispensable tool for commercial micropropagation of plants, genetic transformation for crop improvement and functional genomics studies. In the present study, the application of biosynthesized PHAgNPs increased callus induction in a dose dependent manner, and the greatest response was found with MS medium supplemented with 2.5 mg l −1 2,4-D and 10 mg l −1 PHAgNPs. A 20% increase in callus induction frequency was observed with medium containing 10 mg l −1 PHAgNPs compared with control MS medium. However, further increases in PHAgNP concentration decreased the rate of callogenesis, and 15 mg l −1 was a sub-optimal concentration. At a 20 mg l −1 concentration of NPs, the induction response decreased further, by 46% as compared with control medium. These results contrast with those from previous experiments 24,25,26 in which treatment with AgNPs increased the callus induction frequency and fresh weight of S. nigrum in a dose and time dependent fashion. However, ZnO and TiO2 NPs at concentrations of 5 mg l −1 and 20 mg l −1 , respectively, resulted in the highest induction of calli from indica rice cv. RD49 27 . After transfer of the induced calli to regeneration medium, the effects of PHAgNPs differed. Control medium resulted in a regeneration frequency of 41%, whereas the highest greening of calli and regeneration frequency (61%) was observed with CIM supplemented with PHAgNPs at a concentration of 5 mg l −1 . However, greening and organogenesis drastically declined with increasing concentrations of NPs in the culture medium. Similarly, a previous study has reported that TiO 2 NPs (20 mg l −1 ) resulted in the highest regeneration percentage of rice calli; however, any further increase in concentration of NPs resulted in a decline in the regeneration frequency 27 . A further investigation into the influence of PHAgNPs on the expression of PGR responsive genes encoding five important plant growth hormones-auxin, cytokinin, gibberellic acid, ethylene and ABA 28 -revealed that at 5 mg l −1 PHAgNP concentration, the auxin and cytokinin expression levels were not significantly altered, whereas the ethylene and ABA responsive gene expression decreased, and the regeneration frequency in calli was enhanced. When 10 mg l −1 PHAgNPs were supplemented in the regeneration medium, a further decline in ABA and ethylene hormonal levels was observed, which in turn should have increased the regeneration frequency. However, cytokinin responsive gene expression levels also further declined relative to auxin levels, thus disturbing the cytokinin:auxin ratio, which probably was responsible for the decrease in regeneration frequency. The gibberellic acid responsive gene expression levels remained in concurrence with cytokinin and auxin responsive gene levels, thus suggesting little to no influence on enhancing the regeneration of induced rice calli. In agreement with our results, previous reports have suggested that a balance in cytokinin and auxin ratio is crucial for regeneration response in undifferentiated tissues 29 . In rice, the auxin and cytokinin ratio has been demonstrated to determine regeneration from calli, and an increase in the supplemented auxin to cytokinin ratio suppresses plant regeneration frequency in calli 30 . Furthermore, at a PHAgNP concentration of 20 mg l −1 a substantial (>6 fold) increase in ethylene and ABA concentrations was observed, which was detrimental for regeneration from calli (Fig. 4). An increased ABA level in plants is an indicator of environmental stress and is detrimental to growth and development 6 . Another mechanism through which AgNPs influence regeneration is that when explants are cultured in closed containers (in vitro) their development is severely influenced by plant-produced gases such as ethylene, which even in trace amounts can regulate senescence in plants 7 . Silver ions or nano silver suppresses the influence of ethylene www.nature.com/scientificreports www.nature.com/scientificreports/ by acting as an ethylene perception inhibitor, by down regulating the genes involved in ethylene perception, as reported previously in Arabidopsis thaliana 31 . Finally, when the regenerated shoots were subjected to rooting in vitro, they displayed more roots per plant and longer root length when grown on culture medium augmented with 10 mg l −1 PHAgNPs. At concentrations above and below 10 mg l −1 , a steady decline in root generation was observed, thus suggesting a hormetic influence of PHAgNPs. Similarly, in Arabidopsis, supplementation with AgNPs in the rooting medium positively influences root growth 31 . A previous study has also reported a positive influence of AgNPs on plant root growth through upregulation of auxin synthesis 32 .
Further analysis of the antioxidant status of the regenerating calli revealed that MDA, proline and H 2 O 2 levels decreased in calli cultured on medium supplemented with 5 mg l −1 and 10 mg l −1 PHAgNPs (Fig. 5), and an  13 . ROS damages proteins, lipids and nucleic acids, thereby leading to the development of MDA, which is toxic to plants 33 . Hence, a decrease in MDA, H 2 O 2 and proline levels implies a decrease in ROS production and consequently stress levels in the plant tissues, as regulated by silver ions 13 .Thus, we conclude that AgNPs influence organogenesis, tissue growth and development through multiple mechanisms.
Supporting our data, earlier studies have also reported a hormetic influence of AgNPs, CuNPs and ZnONPs on regeneration in Vanilla planifolia, rice and banana, respectively 34,17,35 . In Tecomella undulata, supplementation with AgNPs (10 mg l −1 ) positively influences callus formation, shoot induction and the number of generated shoots from explants 11 .
We report that PH biosynthesized AgNPs, when supplemented in tissue culture medium, prominently enhance callus induction frequency, regeneration frequency and rhizogenesis at concentrations of 10 mg l −1 , 5 mg l −1 and 10 mg l −1 , respectively, in the recalcitrant indica rice cv. IR64. The above study sheds light on the endogenous hormonal and ROS levels corresponding to the hormetic influence of AgNPs on shoot regeneration from induced rice calli. We therefore conclude that biosynthesized AgNPs have enormous potential in refining tissue culture of recalcitrant crops. In addition, elucidating the activity of AgNPs on endogenous ROS and phytohormone levels may provide a lucrative platform for crop manipulation to enhance growth and yield.

Materials and Methods
The phytofabrication and characterization of AgNPs was carried out as previously described with several modifications 38 . www.nature.com/scientificreports www.nature.com/scientificreports/ preparation of Parthenium hysterophorus aqueous extract. Fresh disease-free leaves of locally collected PH were rinsed with distilled water and shadow dried for 5-7 days to remove residual moisture. The dry leaves were crushed to a fine powder with a kitchen blender. The crushed leaf powder (10 g) was mixed in 100 ml of distilled water and boiled in a water bath for 30 min at 60 °C. The aqueous mixture was filtered through Whatman filter paper, grade 1, and the filtrate was stored at 4 °C.

synthesis of Parthenium hysterophorus AgNPs (PHAgNPs).
PHAgNPs were synthesized by the addition of 1 ml of PH aqueous extract to 9 ml of AgNO 3 (1 mM). The solution pH was adjusted to pH 8 with 1 N NaOH and incubated in a water bath for 30 min at 60 °C. The AgNPs were recovered by centrifugation at 15000 rpm for 20 min, and the resulting pellet was air dried and stored in ambient conditions until further use.
Characterization of PHAgNPs. The biosynthesized PHAgNPs were initially confirmed by UV-visible spectroscopy (Cyber lab-100 spectrophotometer), by measuring the absorbance spectra between wavelengths of 300 and 700 nm at regular intervals.
FTIR analysis was carried out to identify the reduced biomolecules from PH leaf extract in PHAgNPs. The PHAgNPs were pelletized by mixing with KBr and subjected to FTIR analysis with a JASCO 460 PLUS FTIR spectrophotometer, and the spectra were recorded between 4000 cm −1 and 400 cm −1 wavelengths.
TEM images were obtained to visualize the morphological features of PHAgNPs, with a JEOL 3010 transmission electron microscope at an acceleration voltage of 100 kV.
XRD analysis was performed to determine the structure and dimensions of PHAgNPs. The PHAgNP samples were lyophilized and subjected to XRD analysis with an XRD 600 instrument, Shimadzu, Japan instruments (voltage 40 kV; current 30 mA; CuKα radiation).
The mean particle size (L) (PAN analytical X-pert PRO Model) of the AgNPs was calculated with the following formula: L = 0.9 λ/β cos θ, per the Debye-Scherrer equation.
Where λ is the wavelength of the X-ray; θ is Bragg's angle; and β is the full width at half maximum.
Medium preparation. MS medium 39 supplemented with casein hydrolysate 500 mg l −1 , proline 500 mg l −1 , Gamborg B5 vitamins (HiMedia, India) and sucrose 3%, was used as the rice culture medium. PHAgNPs at different concentrations (5 mg l −1 , 10 mg l −1 , 15 mg l −1 and 20 mg l −1 ) suspended in water were supplemented into the culture medium, and the pH was adjusted to pH 5.8, this was followed by sonication of the medium for 30 min. The medium was then appropriately distributed into conical flasks (50 ml medium per150 ml conical flask), and finally 4 g l −1 CleriGel (HiMedia, India) was added before autoclaving. After autoclaving, the medium was allowed to cool to ~45 °C. The flasks were then swirled manually and kept at 4 °C to solidify quickly, avoiding agglomeration of AgNPs.
Influence of PHAgNPs on callus induction. Oryza sativa cv. IR64 seeds were obtained from TNAU, Tamil Nadu, India. The seeds were sterilized with 0.1% HgCl 2 and 70% ethanol as previously described 40  qRt-pCR. The primers ( Malondialdehyde content. MDA content was determined as previously described 48 with a few modifications. One hundred milligrams of regenerating callus tissue was homogenized in 5 ml of 10% trichloroacetic acid (TCA). The homogenate was centrifuged at 10000 rpm for 30 min at room temperature. Two millilitres of supernatant was collected, and 4 µl of 0.6% thiobarbituric acid prepared in 10% TCA was added. The mixture was then incubated in a water bath at 80 °C for 40 min and immediately cooled in an ice bath and centrifuged at 10000 rpm for 30 min. The absorbance was read at 450, 532 and 600 nm, and the MDA content was calculated with the formula: Hydrogen peroxide estimation. Hydrogen peroxide was quantified as previously described 13 . One gram of callus tissue was homogenized in 10 ml of aqueous TCA (0.1%, w/v). The mixture was centrifuged at 10000 rpm for 30 min at 4 °C, and the supernatant was retrieved. One millilitre of supernatant was mixed with 4 ml of 1 M potassium iodide reagent and 1 ml of potassium phosphate buffer. The solution was incubated in the dark for 1 h at room temperature. The absorbance was read at 390 nm and plotted against a standard H 2 O 2 curve. The results are expressed as µM g −1 FW. total proline estimation. Proline content estimation was carried out as previously described 49 . One gram of callus tissue was homogenized in 10 ml of 5-sulphosalicylic acid (3%). The mixture was centrifuged at 10000 rpm for 15 min, and the supernatant (2 ml) was mixed well with 5 ml of glacial acetic acid and 5 ml of 140 mM acid ninhydrin, and heated for 1 h at 100 °C. The mixture was cooled and then extracted in a separating funnel with 10 ml toluene. The absorbance of the separated toluene containing the chromophore was read at 520 nm and plotted against the proline standard curve. The results are expressed as µg g −1 FW. statistical analysis. All the experiments were carried out in triplicate (n = 3). The data were analysed in Windows Microsoft Excel 2007 software. The differences in mean values were determined with Duncan's multiple range test, with a significance level of p ≤ 0.05.

Data Availability
All data generated or analysed during this study are included in this published article.