A non-classical route of efficient plant uptake verified with fluorescent nanoparticles and root adhesion forces investigated using AFM

Classical plant uptake is limited to hydrophilic or water-dispersible material. Therefore, in order to test the uptake behaviour of hydrophobic particles, here, we tested the fate of hydrophobic particles (oleylamine coated Cu2-xSe NPs (CS@OA)) in comparison to hydrophilic particles (chitosan-coated Cu2-xSe NPs (CS@CH)) by treatment on the plant roots. Surprisingly, hydrophobic CS@OA NPs have been found to be ~ 1.3 times more efficient than hydrophilic CS@CH NPs in tomato plant root penetration. An atomic force microscopy (AFM) adhesion force experiment confirms that hydrophobic NPs experience non-spontaneous yet energetically favorable root trapping and penetration. Further, a relative difference in the hydrophobic vs. hydrophilic NPs movement from roots to shoots has been observed and found related to the change in protein corona as identified by two dimensional-polyacrylamide gel electrophoresis (2D-PAGE) analysis. Finally, the toxicity assays at the give concentration showed that Cu2-xSe NPs lead to non-significant toxicity as compared to control. This technology may find an advantage in fertilizer application.

Plants have evolved slowly through natural selection processes, which have been rapidly increased by biotechnology for human requirements 1 . Recently, in phytonanotechnology, plants have been tuned positively by the intrinsic properties of nanoparticles (NPs), such as electron conductivity (improved the electron transport rate of photosystem 1 by 8.8%) 2 , ROS scavenging 3 , water/nutrient retention/supply 4 , and genetic manipulation 5,6 . Furthermore, NPs loaded with the chemical active ingredients have also extended the scope of microsurgery in plants through triggered release [7][8][9][10] . To understand the physiological consequences [11][12][13][14] , a comparison of the vascular uptake of NPs was carried out. Interestingly, some aquatic plants showed greater uptake of NPs than ions 15 .
Classical plant uptake mechanism includes passive spontaneous diffusion, mass flow, ion exchange, and active energy-intensive carrier-assisted method 16 . In NPs uptake also, similar passive mechanisms were identified in metals (M) 17,18 , metal oxides (MO) 19,20 , chalcogenides (MS) 21 , and carbon materials 2,5,22 . Even carrier-mediated transport of NPs within the plant cell to different organelles were also documented 23 . These penetrating NPs were found to be transported both by symplastic and apoplastic modes and have shown xylem and phloem transport 24 . Furthermore, the role of NPs coating, such as with/without citrate 25 , and surface charges 26,27 on plant uptake behavior was also studied.
For efficient plant-gene manipulation with NPs, forced-injection strategies were developed by us and others, but such pressure-assisted delivery systems are not easy to adopt in large-scale field applications 28,29 . Hence, we envisage that a similar forced injection method in a large-scale may be feasible by having hydrophobic surface modification. Both theoretically and experimentally, hydrophobicity in combination with mildly hydrophilic groups was found to show excellent adhesion of water drop even with a tilt of 180°3 0-34 . Similarly, hydrophobic NPs were found to be more easily penetrate lipid membranes in aqueous media 35 . This enhanced adhesion

Results and discussion
Synthesis and characterization. To study the effects of NPs surface polarity on plant uptake, as-prepared hydrophobic Cu 2-x Se NPs with oleylamine coating and hydrophilic Cu 2-x Se NPs with chitosan coating were tested against the model plant viz., tomato. The X-ray diffraction (XRD) pattern of the as-prepared oleylamine coated Cu 2-x Se NPs (Fig. 1A) shows diffraction peaks at 26.77, 44.72, 53.03, and 65.27° that match the (111), (220), (311), and (400) planes of face-centred-cubic Cu 2-x Se (JCPDS 06-0680) (for brevity, this material will be denoted as CS@OA).
Following the XRD confirmation of the as-prepared Cu 2-x Se NPs, to develop hydrophilic substitutes of the same, Cu 2-x Se NPs were coated with chitosan (for brevity, this material will be denoted as CS@CH). The chitosan coating was difficult to confirm by Fourier transform infrared (FT-IR) spectrophotometer, as the oleylamine signals overlap with the chitosan signals w.r.t N-H bending at 1381 cm -1 and 1625 cm −1 and C-N bonding/N-H stretching at 3400 cm −1 (Fig. S1) [47][48][49] . However, the visual observation of the CS@CH NPs dispersed well in distilled water compared to the complete precipitation of the CS@OA NPs in distilled water confirms the coating (Fig. S2). Furthermore, to quantify the change in the NPs surface polarity, the material before and after chitosan coating was drop cast onto glass, and the contact angle was measured. The contact angle of the CS@OA NPs has been found to be 142° (this is close to the value of superhydrophobicity (150°) 50 , which after chitosan coating in CS@CH NPs decreased to 123° (inset in Fig. 1C,D). This chitosan coating has not been found to affect the absorbance intensity (Fig. S3). The intense fluorescence of the Cu 2-x Se NPs needed to be retained for the tracking of the NPs through confocal imaging. Hence, the photoluminescence (PL) spectra of CS@OA NPs and CS@CH NPs were measured in 1:1 ratio of ethanol: water mixture at 370 nm excitation wavelength. The spectra show no major compromises in signal intensity after coating, which ensures its insignificance on the imaging (Fig. 1B). The transmission electron microscope (TEM) image of the CS@OA NPs shows monodisperse NPs with a size distribution of approximately 15 ± 8 nm (Fig. 1C). The high resolution (HR) TEM (HR-TEM) (inset in Fig. 1C) shows a lattice spacing of 0.33 nm, which corresponds to the (111) plane of Cu 2-x Se NPs. The TEM image after chitosan coating shows that the size and shape of the NPs is stable (Fig. 1D). The size of CS@CH NPs has been found in the range from 15 to 30 nm. In agreement with the higher molecular weight of chitosan, the hydrodynamic peak size of the CS@CH NPs (PDI = 0.123) has been found to be 9 nm greater than that of the CS@OA (PDI = 0.087) NPs (Fig. S4). The calculated number of particles were 2.5 × 10 13 /mL. Uptake study. Following the material characterization, both CS@OA and CS@CH NPs were sprayed onto the roots of 30-day-old plants. One group was sprayed with CS@OA NPs and the other group was sprayed with CS@CH NPs. After brief air drying, the plants were incubated in the hydroponic medium. The NPs that had not landed on the roots were collected on a glass backspot and estimated to be ~ 200 µg. The root and shoot samples were collected at 1.5, 3, 6, 12, and 24 h intervals, oven-dried and then quantified with inductively coupled plasma mass spectrometry (ICP-MS) after microwave acid digestion ( Fig. 2A,B). Before oven drying and digestion steps, the root biomass of samples collected after every time intervals were washed with 0.1 M HNO 3 , to remove the NPs just adhered without uptake by the root 51 . The scanning electron microscope (SEM) images of the unwashed roots (Fig. S5) and after washing with 0.1 M HNO 3 (Fig. S6) confirms that the adhered particles were removed. The uptake study ( Fig. 2A) clearly shows that the hydrophobic CS@OA NPs have the ability to quickly enter into the roots and are taken up at a rate that is ~ 1.3 times the uptake rate of the hydrophilic CS@CH NPs after the initial 1.5 h incubation with the root.
The uptake concentration started to decrease gradually in the root, unlike previous studies where a continuous increase in the NPs content in the root was observed [52][53][54] . This is obviously because of the (1) lack of continuous NPs supply from the medium, unlike in previous studies, (2) transport to the shoot from the root (Fig. 2B), and (3) possible restriction of any particles adhered to the root. Despite the unavailability of additional NPs in the incubation medium, the uptake in the roots has been found to be more than that in the shoot at any given time point in the 24 h observation period. This may be due to an inhibition of the NPs movement, which is in agreement with the previously reported long observation times for different hydrophilic NPs compared to that of ions 20,55,56 . Restriction of movement across the tissue in hydrophilic CeO 2 particles resulted in their dissolution into ions after 4 days 19 . Interestingly, the ratio of the NPs movement to shoots from the roots have www.nature.com/scientificreports/ been found higher in the treatment sprayed with hydrophilic NPs (0.72) than in the hydrophobic NPs (0.48) (Fig. 2C). To negate the role of ethanol in uptake, we test the uptake of CS@CH NPs at 1.5 h, after pure ethanol spraying followed by CS@CH NPs spraying in water. The uptake of CS@CH NPs (212 ± 8 mg/kg) found to be a little lesser or almost equal to the treatment without ethanol pre-spraying, which clearly negates any role of ethanol. We also quantified the leaching of Cu ions from CS@OA and CS@CH NPs in the medium by ICP-MS after 24 h of incubation and found ~ 1.5% copper in CS@OA and < 1.5% in CS@CH NPs, which is insignificant to influence the treatment. The intense fluorescence from the Cu 2-X Se NPs allowed the NPs to be tracked in the plant tissue; here, an image taken from a root after 1 h of incubation using confocal laser scanning microscopy (CLSM) is given in Fig. 3A-D. This study confirms that the uptake occurred as intact NPs in the plant tissue rather than by dissolution or oxidation of the NPs to ions. The CLSM image of the untreated root sample reveals the absence of particles (Fig. S7). The 3D CLSM images of the tissues treated with CS@OA and CS@CH NPs are given in videos 1 and 2, respectively. This was further confirmed by the TEM micrographs of the microtome sections of the roots after 3 h of incubation (vide infra).
Following the CLSM study, the microtome sections of the roots were stained and observed by TEM. In the TEM images, the darkly contrasted CS@CH and CS@OA NPs shows differences in their patterns of particle aggregation, distribution, and transport in the tissue. The hydrophobic CS@OA NPs have been predominantly present in the intercellular region and showed similarly restricted movement as that observed for the less hydrophilic pristine carbon nanotubes in plant cells; however, in that study the polarity wasn't discussed 57 (Fig. 3E). www.nature.com/scientificreports/ CS@OA NPs aggregation in the intercellular space may be due to cell wall lipid assisted liposome formation and may eventually restrict its intracellular movement. Within the intercellular area, a major portion of the hydrophobic CS@OA NPs have been found to be aligned in the bilayer cell membranes unexposed to the polar heads (encircled in black gradient lines). The element composition of these particles was confirmed with point EDX elemental analysis (Fig. S8); the nickel signal observed is from the grid, since nickel grid was used in place of copper grid to avoid copper signal overlap and lead signal observed is from the staining agent. In the hydrophobic CS@OA NPs-treated plant, many endocytosis-like bodies have been observed (encircled in red), which once again supports lipid-covered body formation in the hydrophobic treatment. Supporting this claim, hydrophobic NPs were reported to easily form liposomes through bilayer disruption 35 .
In contrast, the hydrophilic CS@CH NPs have been found predominantly distributed in the intracellular region, which may be due to the ability of the hydrophilic NPs to interact with the polar head groups of the cell membrane (Fig. 3F). However, this does not negate the possibility of CS@CH NPs movement in the intercellular region. A closer look at the gradient circle shows NPs-aligned movement without aggregation. Both CS@OA and CS@CH NPs did occupy the intercellular gas space. The predominant intracellular distribution of the CS@ CH NPs and their well distributed (without aggregation) intercellular presence may be due to their compatibility with an aqueous medium. Unlike hydrophobic NPs, in CS@CH NPs treatment, endocytotic bodies have not been observed which may be due to their ability to directly enter the cell. Similar direct entry without endocytosis was observed in plant protoplast cells 58 . The particle size distribution in the intercellular space is given in Fig. S9 (the microtome location of the NPs is given just above the graph), which confirms the stability of the NPs in the plant tissue. However, over time, they may not be stable due to the enzymatic action of the plant, especially in the leaves, and over time, the particles may dissociate into ions, as previously observed by radioactive signals 19 . The untreated root sample did not show the presence of any particles (Fig. S10).

Mechanism of uptake. AFM adhesion force measurements. The enhanced penetration of CS@OA NPs
into the root may be due to their binding, which motivated us to measure the force vs. distance between the root and the NPs in water using AFM. AFM silicon nitride tips were modified with CS@OA and CS@CH, separately, by incubation in the respective solutions for 5 h followed by sequential washing with water and ethanol. A fresh www.nature.com/scientificreports/ root tip was fixed onto a glass plate with a resin that could dry fast without any cross-reactions in order to keep the root alive 30,59,60 . Following this, water was added carefully with a syringe, and the AFM tip was placed in contact with the root with the assistance of the microscope. The maximum from the Gaussian fitting of the adhesion force experienced by the CS@OA NPs has been found to be ~ 500 pN more than the force experienced by the CS@CH NPs (Fig. 4A,B). This adhesion could be driven by three reasons: (1) water, which pushes the hydrophobic NPs towards a solid substrate (here the root) or the hydrophobic particles unwillingness to allow water into the root/particle interspaces as confirmed with contact angle, (2) manipulation of the contacted hydrophobic NPs surface to become sticky with the ions present on the root surface as confirmed with AFM study 35 , and (3) as mentioned above, the ability of the hydrophobic NPs to penetrate the lipid surface as confirmed with ICP-MS 35 . Thus, active uptake where the energy contribution comes from the NPs, unlike classical active transport where the energy is expended by the plants, is identified. Thus, the adhesion force expressed by the NPs corroborates to the root uptake proportionally, similar proportional uptake was also reported earlier in the animal cells 61 . The maximum adhesion force expressed by the hydrophobic NPs have led to the enhanced root uptake.
Protein corona study. The curiosity to understand the ability of the hydrophilic NPs to show a greater root to shoot transport ratio motivated us to study its protein corona in comparison to that of the hydrophobic NPs. Protein coronae, which changes with the surface often decide the fate of NPs in animals [62][63][64][65][66][67] ; which has been ignored in plants, is studied here by 2D-polyacrylamide gel electrophoresis (2D-PAGE). The intact NPs from the plant tissues were obtained by enzyme-assisted extraction following the protocol standardized by Dan et al 17 . In 2D gels, protein spots have been found to be distributed within a molecular mass range of 19-43 kDa and covering the pH range of 4.3-9.6. For the CS@OA NPs, 10 protein spots have been detected, while 14 protein spots have been detected for the CS@CH NPs. In the CS@OA sample, 2 proteins have been found down-regulated (numbers 4 and 10), and 4 protein spots have not been detected (numbers 11, 12, 13 and 14) compared to the proteins detected in the CS@CH sample (Fig. 4C,D). Interestingly, all of these protein spots (except 13) have been observed at acidic pH. To explain this pattern, close approximations of foreign body movements in roots viz., a mycorrhizal association report were compared. A similar downregulation of an acidic pI membrane protein was observed in tomato by mycorrhizal association 68 , which may be the adaptation strategy of the plant to restrict them in the root zone. Additionally, the presence of a few spots in the higher pI range for the CS@OA sample are poorly separated due to the poor protein solubility and contaminants in the buffer, which interfered with staining. Therefore, they may not be considered protein spots. www.nature.com/scientificreports/ Toxicity study. Finally, after elucidating the uptake mechanism, the toxicities of the hydrophobic and hydrophilic NPs were compared by using the ascorbate peroxidase (APOX) and catalase (CAT) activity assays. In roots, the APOX activity has been found to be increase after incubation with the NPs, especially at 1.5 h (Fig. 5A). The initial stress due to the plant uptake of foreign bodies may have resulted in the synthesis of H 2 O 2, whose enzymatic conversion may have raised the APOX activity 51 . Interestingly, with increasing incubation time, the activity has been found to be reduced, possibly because the plant adapted to the initial NPs load and because there is no further accumulation. In the shoots, the APOX activity has been found to be increase after 12 h of incubation, which may have been due to the time needed for a threshold amount of NPs accumulation to trigger APOX in the shoot (Fig. 5B). In case of the CAT activity, the reduction in the activity has been noted (Fig. 5C,D), which is contrary to the previous studies on metal NPs uptake 69 . There are few studies where a reduction in the CAT activity was also documented in the presence of an overexpression of the APOX activity, for instance here APOX is overexpressed, which can control oxidative stress 22,51 . Visual observation of the plants over 3 days surprisingly shows that the plants incubated after being sprayed with CS@OA NPs found to healthier than the plants incubated after being sprayed with CS@CH NPs. Further, the MTT assay was also performed to evaluate the toxicity caused by the CS@OA and CS@CH NPs. The assay shows the biocompatibility of the NPs because the percent root viability has been found to be > 90% after 24 h of treatment with CS@OA and CS@CH NPs (Fig. S11). Thus, at a 100 µg/plant concentration, no visual effect on plants incubated with hydrophobic NPs. Apart from the NPs concentration; there is a fair chance that trace amounts of selenium ions could have leached from the particles and enhanced the antioxidant/photosynthetic activity and photo-oxidative stress control [70][71][72][73][74] .

Conclusions
Previous experience regarding NPs uptake has confirmed that uptake is a genus, species, variety, material, age, concentration, and size-dependent. However, there were no reports on the effects of NPs with hydrophobic surfaces, which are explained here. The enhanced uptake of hydrophobic NPs by the roots proves that non-classical forced penetration is more efficient. Hence, this enhanced uptake and sedentary behavior of hydrophobic NPs in the root can be adopted for eco-friendly leach-proof fertilizer application. This report also serves as an early warning to avoid exposing undesired hydrophobic NPs to edible plants in the context of enhanced phytoaccumulation. Furthermore, TEM images taken at the early incubation period reveal a predominance of hydrophobic NPs in the membrane bilayer, which has the future potential for spatial targeting in plants.

Synthesis of Cu 2-x Se NPs (CS@OA).
The arrested precipitation method was adopted for the synthesis of Cu 2-x Se NPs with minor modifications 75 . In brief, a nitrogen-filled glove box (< 0.5 ppm oxygen) was employed for the preparation of the precursor mixtures. First, copper and selenium reactant mixtures were individually prepared and allowed to react simultaneously by hot injection. The copper reactant was prepared by the addition of 10 mL of oleylamine to 0.198 g of cuprous chloride in a round-bottom flask. Then, the mixture was heated under a nitrogen environment to 130 °C for 15 min along with stirring using a stir bar. The solution was cooled to 100 °C before the injection of selenium. The selenium reactant was prepared by the addition of 1 mL of oleylamine to 0.123 g of selenourea in a round-bottom flask. Then, the mixture was heated under nitrogen conditions to 200 °C for 15 min with stirring. Selenourea and oleylamine were injected into a flask containing cuprous chloride and oleylamine after being cooled to 160 °C. After injection, the solution appeared black and was further heated to 240 °C for half an hour. Then, the solution was allowed to cool to room temperature. The www.nature.com/scientificreports/ synthesized CS@OA NPs were precipitated by adding 10 mL of ethanol and washed 5 times with a chloroformethanol mixture with a ratio of 1:2 and finally dried under reduced pressure of 65 cm Hg.

Synthesis of chitosan-coated Cu 2-x Se NPs (CS@CH). The synthesized CS@OA NPs powder (10 mg)
was dispersed in 10 mL of a chitosan solution by bath sonication for 30 min. The chitosan solution was prepared by adding 50 mg of chitosan powder to 10 mL of distilled water at pH 4.5 under stirring. The pH of the distilled water was adjusted by glacial acetic acid 76 . After dispersion, the solution was centrifuged at 12,000 rpm for 10 min to remove unbound free polymer, followed by washing with distilled water. Finally, the chitosan-coated CS@OA NPs were dispersed in distilled water.
Characterization of CS@OA and CS@CH NPs. A Bruker D8 Advance Diffractometer with a Cu Kα 1 radiation source (λ = 1.5406 Å) was used for XRD pattern analysis at 40 kV and 25 mA. A Cary 600 Series FT-IR spectrophotometer (Agilent Technologies) was used to record the FT-IR spectra. The contact angle was measured by DIGIDROP modular contact angle technology and analyzed by Visio drop software. For this study, CS@ OA and CS@CH NPs samples were dispersed in ethanol and distilled water, respectively, and a film was prepared by coating the samples onto a glass substrate surface. The glass substrate and other surfaces were cleaned with acetone and ethanol using a cotton swab, dried with a clean cotton swab, and then placed onto the sample stage. The contact angle was measured by gently depositing a drop of deionized water onto the film on the substrate surface using a microsyringe at ambient temperature. The contact angle was measured instantly after contact of the water drop with the NPs film. Contact angle was calculated by the computer software in the goniometer without the operator intervention. UV-Visible absorption spectra were recorded by a UV-Visible spectrophotometer (UV-2600, Shimadzu). An Edinburgh Instruments was used to record the PL emission spectra at 370 nm excitation wavelength. For UV-Visible and PL analysis, CS@OA and CS@CH powders were dispersed in an ethanol-water mixture with a ratio of 1:1 at a concentration of 300 ppm. A JEOL JEM-2100 microscope was used for TEM and HR-TEM analysis at 200 kV. The samples were prepared by dropping 7 μL of a highly diluted and dispersed sample solution onto a carbon-coated copper grid and wicking off excess solution after 1 min with filter paper. The size distribution on TEM images was calculated by using Image-J software. A Malvern Zetasizer Nano ZSP Instrument was used to measure the hydrodynamic diameter of each sample at 25 °C. Well-dispersed samples of 50 ppm concentration was prepared for the hydrodynamic diameter measurements were added into a clear glass dynamic light scattering (DLS) cuvette. In parameters, total three runs were set for each sample, each run was for 2 min and the equilibration time was set 120 s.

Instrumentation in determining NPs and plant interaction. The copper concentrations in root tis-
sues were quantified by using ICP-MS (Agilent 7700 series). CLSM (Carl Zeiss microscope LSM 800) was used to locate the CS@OA and CS@CH NPs in the root cells. CLSM images were processed using ImageJ software. AFM experiments were carried out on a commercial AFM instrument (Force Robot 00574, JPK Instruments, Berlin, Germany). The force-distance curves were recorded by commercial software from JPK and analyzed by custom-written procedures in Igor Pro 6.2 (Wavemetrics, Inc.). For the experiment, CS@OA and CS@CH NPs were dispersed in ethanol and water, respectively, at a concentration of 300 ppm. The PROTEAN i12 IEF system (Bio-Rad) was used for the separation of proteins in the first dimension, and the Mini-PROTEAN system (Bio-Rad) was used for the separation of proteins in the second dimension. Gel pictures were captured by using a gel doc (Bio-Rad) with Image Lab software.
Uptake study of CS@OA and CS@CH NPs. Plant treatment. As characterized particles were applied aeroponically on the tomato roots following the standard procedure 77,78 . For this study, 30-day-old tomato plants were collected from the green house and sorted in such a way that they have approximately similar biometric parameters (9-10 cm, and the root lengths were 1.5-2 cm) and divided into 2 groups. The as-prepared hydrophobic CS@OA NPs were dispersed in ethanol, and the hydrophilic CS@CH NPs were dispersed in distilled water for spraying. First group of plants was sprayed with the CS@OA NPs, and the second group was sprayed with CS@CH NPs, dispersed in 1 mL respective solvent at 300 ppm concentration. After briefly air drying, the plants were incubated in the 20 mL hydroponic medium for 1.5, 3, 6, 12, and 24 h, following which the samples were oven-dried and then quantified with ICP-MS after microwave acid digestion. The NPs that had not landed on the roots were collected on a glass backstop from two samples per treatment, digested, and analyzed with the ICP-MS. Details of the ICP-MS protocol is given below.
ICP-MS analysis. After the specified time period, the plants were collected and washed with 0.01 M HNO 3 and Milli-Q water to remove any adhered particles that had not entered the roots. Then, the plants were sectioned into roots and stems (cut 5 cm from the shoot start point) and dried properly at 60 °C in an oven, and weighed. The dry weights of the 50 mg roots and 150 mg shoots were taken for the metal content analysis. The treated samples were then digested by using 1 mL of metal-grade HNO 3 and H 2 O 2 (1:4) as described previously 51 . Then, the samples were diluted 13-fold in distilled water and analyzed using ICP-MS. Further, to quantify the leaching of copper ions from CS@OA or CS@CH NPs, the medium in which the plant was sprayed and incubated for 24 h, was collected. This medium was centrifuged at 10,000 rpm for 30 min to evaluate the supernatant ion concentration with ICP-MS.
CLSM analysis. For this experiment, the plant root tissue was treated with a 1 mL solution of CS@OA or CS@ CH NPs at a concentration of 300 ppm by using a sprayer. After 1 h, the root tissue was excised from the treated www.nature.com/scientificreports/ plant and washed thoroughly with deionized water. Then, the root tissue was placed in an FAA fixative solution containing 37% formaldehyde (v/v), 5% glacial acetic acid (v/v), 45% ethanol (v/v) and 45% distilled water (v/v), followed by vacuum infiltration for 15 min. After that, the root tissue was transferred into a fresh FAA fixative solution overnight at 4 °C. Then, the root samples were washed three times with phosphate buffer for 10 min each. Tissues were then dehydrated through a 30, 50, 75, 90, and 100% graded series of ethanol for 15 min each. Prior to microscopy, sample clearance was performed to increase the tissue transparency. For this, tissues were passed through 25, 50, 75 and 100% concentrations of glycerol for 1 h each with two changes of each solution and then incubated overnight in pure glycerol 79 . Cleared tissues were then mounted on long coverslips in glycerol and covered by small coverslips. Images were acquired by CLSM in the blue region after excitation at 405 nm and emission was collected in the wavelength range of 410-480 nm.
Root tissue section preparation for TEM imaging. For this study, the plant roots were treated with 1 mL of a CS@ OA or CS@CH NPs solution at a concentration of 300 ppm by using a sprayer. After 3 h, the roots were excised from the treated plants and washed thoroughly with deionized water. Then, the root samples were fixed with a 5% glutaraldehyde solution followed by incubation for 2 h at 4 °C. Then, the root samples were washed three times with phosphate-buffered saline for 10 min each time. After that, several steps, such as osmication with 1% osmium tetroxide, en bloc staining with 2% uranyl acetate, and dehydration with a graded series of ethanol followed by infiltration with embedding medium, were performed as reported 80 . Then, a few drops of the Spurr resin were placed into moulds, followed by transfer of the root samples into the moulds. The remaining space in the moulds was filled with Spurr resin, and then the moulds were kept overnight at 60 °C for polymerization. Then, the samples were sectioned into 70-80 nm thick samples with a Leica EM UC6 ultra-microtome, followed by staining with uranyl acetate and lead citrate. The root tissue sections were then placed on nickel grids and analyzed by TEM.
Uptake mechanism of CS@OA and CS@CH NPs. AFM force measurements. The force was measured by modifying AFM silicon nitride tips with CS@OA and CS@CH NPs and fixation of fresh root tip onto a glass plate by using epoxy resin 31,59,60 . First, the glass coverslips were placed in a warm chromium acid solution for 3 h to remove residual organic matter and then rinsed with Milli-Q water followed by drying under a stream of nitrogen. AFM silicon nitride tips were modified with CS@OA and CS@CH separately, by incubation in the respective solutions for 5 h followed by sequential washing with water and ethanol. A fresh root tip was fixed onto a glass plate with an epoxy resin that could dry fast without any cross-reactions in order to keep the root alive. Following this, water was added carefully with a syringe, and the AFM tip was placed in contact with the root with the assistance of the microscope and the adhesion force was measured. The protocol we selected was to record single measurements at different points along the length of the root by taking several measurements at the root surface. AFM silicon nitride cantilevers with silicon nitride tips (type MLCT, from APP NANO) were used in all of the experiments. The spring constants of the tips were calibrated by the thermal fluctuation method and were all in the range of 0.040-0.075 N m −1 . All of the experiments were carried out at a pulling speed of 1000 nm s −1 . The AFM experiments were conducted after allowing the system to equilibrate for 30 min. All of the AFM force measurements were carried out at 25 ± 1 °C.
Protein corona study. For this study, 30-day-old plants of the same length and weight were used for each treatment. Afterwards, the plant roots were treated with 1 mL of CS@OA or CS@CH NP solutions at a concentration of 300 ppm by using a sprayer, and the plants were then transferred into a hydroponic medium for 3 h. Then, two grams of root from all the treated plants were collected and ground in liquid nitrogen with a mortar and pestle. Then, the ground tissue powder from each treatment was added to 10 mL of a 2 mM citrate buffer at pH 4.5, followed by homogenization of the samples in an ice bath. Then, a 0.5 mM phenylmethylsulphonyl fluoride (PMSF) cocktail was added to each treatment solution to inhibit protease. The optimum pH range should be 3.5-7.0 for activity of the macerozyme R-10 as provided by the manufacturer; as such, the pH of the citrate buffer was adjusted to 4.5 with citric acid. Then, 2 mL of citrate buffer containing 800 mg of the macerozyme R-10 was added to the above solution, and the samples were incubated for 24 h at 37 °C prior to being digested as reported 18 . After digestion, samples were filtered through 0.45 µm filter paper, and the filtrate was collected. Then, the samples were centrifuged at 12,000 rpm for 30 min at 4 °C to separate the NPs from unbound proteins. The obtained pellet for each treatment was washed several times by dispersing in 1.5 mL of 1 × PBS, followed by centrifugation to completely remove NPs from the unbound proteins 81 . All of the pellets (having NPs with bound proteins) had 100 µL of a 10% sodium dodecyl sulphate solution added to them and were kept at 95 °C in a water bath for 10 min, followed by centrifugation at 12,000 rpm. Then, the supernatant was collected, which had NP-free protein. After that, samples were cleaned by using a ReadyPrep 2D clean-up kit. Finally, the clean protein pellets for each treatment were used for 2D-PAGE documentation. The cleaned sample pellets for each treatment were dispersed in 200 µL of rehydration buffer. The entire 200 μL volume of the IPG gel rehydration buffer-containing protein samples were loaded onto a 7-cm long IPG strip with a pH range of 3-10, which allows the proteins to enter the IPG strip gel after an overnight rehydration and equilibration process. After that, the IPG strips were transferred to an isoelectric focusing tray, and the proteins were separated by using an isoelectric focusing electrophoresis unit. The IPG strips were run for a total of 10 kVh. After isoelectric focusing, the IPG strips were equilibrated in equilibration buffer-I and equilibration buffer-II for 10 min each. The equilibrated IPG strips were then used for a second dimension of electrophoresis. The proteins were separated by using a 12% polyacrylamide gel. Gel electrophoresis was carried out at 100 V for approximately 80 min until the proteins had separated to the end of the gel. Precision plus protein kaleidoscope www.nature.com/scientificreports/ prestained protein standard was used as a molecular weight marker. The gels were stained by silver staining, and images were captured by a gel doc and analyzed by the PDQuest software (Bio-Rad). After electrophoresis, gels were fixed in a methanol (50 mL), acetic acid (12 mL), formaldehyde (50 µL), and MQ water (38 mL) mixture for 30 min followed by washing in a methanol (20 mL) and water (80 mL) mixture for 10 min. Then, the gels were sensitized in an aqueous solution of sodium thiosulfate (prepared by adding 20 mg to 100 mL of MQ water) for 10 min followed by washing of the gels with excess MQ water. After that, the gels were stained with a chilled silver nitrate solution (prepared by adding 200 mg of silver nitrate and 76 µL of formaldehyde into 100 mL of MQ water) at 4 °C followed by washing of the gels with excess MQ water. Then, the gels were developed in a development solution (sodium thiosulfate (0.4 mg), sodium carbonate (6 g), and formaldehyde (50 µL) in 100 mL of MQ water) under a white light transilluminator. After the protein spots appeared, the development solution was replaced with the stop solution (10 mL acetic acid in 90 mL distilled water) to avoid excessive development.
Toxicity study. The toxicity of the hydrophobic and hydrophilic NPs was compared by using the ascorbate peroxidase (APOX) and catalase (CAT) assays as reported 51 . For this study, the tomato roots were treated with 1 mL of CS@OA or CS@CH NPs at a concentration of 300 ppm by using a sprayer as described above. For CAT activity, treated root and shoot tissues were excised after a specific time period and ground in liquid nitrogen with a mortar and pestle followed by homogenization in 1 mL of pH 7.4 ice-cold potassium phosphate buffer. Then, the tissue extracts were centrifuged at 4 °C for 5 min at 10,000 rpm, and the supernatant was collected. Then, 10 µL of the supernatant was added to 990 μL of 10 mM H 2 O 2 in a quartz cuvette and mixed. The CAT activity was determined by a UV-Vis spectrophotometer based on the decrease in the reaction mixture absorbance at 240 nm over 1 min.
For APOX activity, plant samples were prepared as described above. A 100 μL volume of the supernatant was added to a 1 mL quartz cuvette with 886 μL of 0.1 M potassium phosphate buffer at pH 7.4, 4 μL of 25 mM ascorbate, and 10 μL of 17 mM H 2 O 2 and then mixed. The APOX activity was determined with a UV-Vis spectrophotometer by measuring the decrease in the reaction mixture absorbance at 265 nm over one minute. The protein concentrations for both assays were quantified by the Bradford method using a bovine serum albumin standard curve.
Further, the root viability was studied by MTT assay as reported 82 . In brief, the plant roots were treated with 1 mL of CS@OA or CS@CH NPs as described above and incubated for 24 h. After that, 10 mg of the fresh root tissue was taken and transferred to the 2 mL centrifuge tube followed by the addition of MTT dye. After 4 h of incubation in the dark, the MTT solution was discarded and the root tissues were transferred to the fresh petriplates. The root tissue was cut with a sterile scalpel into 1-2 mm pieces followed by the addition of 0.5 mL of KOH solution to this. The cut root pieces along with the KOH solution was transferred to the 2 mL of centrifuge tubes and 0.5 mL of DMSO solution was added to each tube to make the total volume of 1 mL. Then, the tubes were centrifuged at room temperature at 500 × g for 5 min. The supernatant was transferred to fresh tubes, resultant absorbance was measured at 570 nm and cell viability was calculated. Statistical analysis. GraphPad Prism 8.0 software was used for all the statistical data analyses. All data were plotted as mean ± standard error. A nonparametric t-test was used to reveal the significant difference at 95% confidence level (P < 0.05), as denoted by *asterisks. The data not showing asterisks, reveal not significant. www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.