Engineering bioactive surfaces on nanoparticles and their biological interactions

The successful integration of nanoparticles into biomedical applications requires modulation of their surface properties so that the interaction with biological systems is regulated to minimize toxicity for biological function. In the present work, we have engineered bioactive surfaces on gold (Au) and silver (Ag) nanoparticles and subsequently evaluated their interaction with mouse skin fibroblasts and macrophages. The Au and Ag nanoparticles were synthesized using tyrosine, tryptophan, isonicotinylhydrazide, epigallocatechin gallate, and curcumin as reducing and stabilizing agents. The nanoparticles thus prepared showed surface corona and exhibited free radical scavenging and enzyme activities with limited cytotoxicity and genotoxicity. We have thus developed avenues for engineering the surface of nanoparticles for biological applications.

Synthesis of gold (Au) and silver (Ag) nanoparticles. The Au and Ag nanoparticles were synthesized using different reducing agents, as shown in Fig. 1A,B. Typically, 100 ml of aqueous solutions with 1 mM KOH containing 0.1 mM Tyr, Trp, INH, EGCG, or Cur, respectively, were allowed to heat at constant stirring conditions, followed by the addition of 0.1 mM [AuCl 4 ] − or Ag + ions to obtain Au and Ag nanoparticles. Further, all these nanoparticle solutions were heated to reduce their volume to ~ 50 ml, so that the concentrations of nanoparticles are increased. All the nanoparticle solutions were found to be stable even after concentrating, which represents that the nanoparticles are capped by the corresponding reducing agents. This has further been confirmed using several techniques and discussed in later sections. Throughout the discussion, the nanoparticles of Au and Ag are symbolized by Au Tyr , Au Trp , Au INH , Au EGCG , Au Cur , Ag Tyr , Ag Trp , Ag INH , Ag EGCG , and Ag Cur , respectively based on the respective reducing agent.
Physicochemical characterization of nanoparticles. All the nanoparticles are characterized using several techniques to understand their physicochemical nature prior to biological activities. The UV-Visible spectrophotometer (Shimadzu UV-1800) operated at a resolution of 2 nm and used to understand the SPR properties of nanoparticles. Fourier-transform infrared spectroscopy (FTIR) analysis of all the nanoparticles was performed using a Bruker spectrophotometer with a resolution of 4 cm −1 . Transmission electron microscopy (Hitachi, HT-7700, Tokyo, Japan) was used to obtain micrographs of nanoparticles. For TEM, 1 µl of nanoparti-  Cell viability and cytotoxicity assessment of nanoparticles. Multiplex cell viability/cytotoxicity assay kit was used to determine the viable cell population and dead cells. Briefly, 1 × 10 4 cells/ml, m5S mouse skin fibroblasts were cultured in α-MEM (Nacalai Tesque Inc. Kyoto, Japan) conditioned with 10% FBS (Bio-Source, Rockville, MD, USA) and 1% antibacterial and antimycotic cocktail solution (Sigma-Aldrich, St. Louis, USA). Cells were cultured in the humidified chamber (5% CO 2 , 37 °C) up to the confluent stage. The cells were treated with different concentrations (0.5, 1.0, and 5.0 µg/ml) of nanoparticles for 16 h. After exposure, the cells were washed with phosphate-buffered saline (PBS), and 10 μl of the CCK-8 solution was added to each well of the plate. The plates were incubated at 37 °C in the CO 2 incubator for 4 h, and the absorbance was measured at 490 nm (Multiskan FC, Thermo-Fisher Scientific, Inc., Pittsburg, PA, USA). The untreated sets were also run under similar experimental conditions, and the data represent the percentage (%) of viable cells in three different groups. For cytotoxicity assessment, the cell membrane integrity was evaluated by measuring LDH release using the multiplex assay kit. The m5S mouse skin fibroblasts were seeded into 96 well plates at a density of 1 × 10 4 cells/ well, and nanoparticles (0.5, 1.0, and 5.0 µg/ml) were added to the wells, followed incubation at 37 °C for 16 h and centrifuged at 25 G for 5 min. Cell-free culture media was collected, and 100 μl of the reaction mixture was added. The cells were incubated at room temperature for 30 min in the dark to measure the amount of LDH released into the medium using a UV-visible spectrophotometer at an absorbance of 490 nm. The control experiments were conducted under similar experimental conditions, and the data represent the percentage (%) of cytotoxicity in three different test groups.
The cytotoxicity endpoint was determined by MTT assay for RAW-264.7 cells, which were cultured in 96 well culture plates (True line, Biotechnology, Germany) at a density of 1 × 10 4 cells/ml up to sub confluence stage. The cells were then treated with various doses (0.5, 1.0, and 5.0 μg/ml) of nanoparticles, and incubated for 12 h.
The control experiments were also conducted under similar experimental conditions. The intracellular purple formazan was quantified with a UV-Vis spectrophotometer at an absorbance of 570 nm (Thermo Scientific, Multiskan FC, Pittsburg, USA).
Genotoxicity/DNA damage assessment of nanoparticles. The m5S cells were seeded in 24 well culture plates and treated with different nanoparticles of various concentrations (0.5, 1.0, and 5.0 µg/ml) and incubated for 24 h. The single-cell DNA damage assay was carried out as per the reported protocol 46 to estimate

Scientific Reports
| (2020) 10:19713 | https://doi.org/10.1038/s41598-020-75465-z www.nature.com/scientificreports/ the potential genotoxic effect of nanoparticles. After incubation, the cells were harvested with a 0.5% trypsin EDTA solution, and the total cell population of 1 × 10 6 cells/ml was processed for DNA damage assay. The cells were suspended in 0.25% low melting agarose, and the suspension was embedded in 1% normal melting agarose coated slide in an ice-cold tray. Then, the slides were immersed in lysis solution containing 100 mM EDTA, 2.5 M NaCl, 100 mM Trizma base, 1% sodium sarcosinate (pH 10.0), 1% Triton X-100, and 10% DMSO at 4 °C for 1 h. Later the slides were subjected to alkaline treatment (1 mM Na-EDTA, 300 mM NaOH, pH 13.0), and electrophoresis was carried out for 30 min at 4 °C. After electrophoresis, the slides were neutralized with neutralizing buffer (0.5 M Trizma base-HCl, pH 7.5) and were fixed with 100% methanol. The slides were dried and then stained with EtBr. 100 cells were counted for fragmented DNA tail using a Leica fluorescent microscope (Leica, SP 8000, Germany), and results were analyzed using Comet Analyzing Software (Version 7.1.0 rev 13. April 2016, Andor, Ltd, Shinagawa, Japan, https ://andor .oxins t.com/produ cts/komet -softw are/komet -7) and expressed as DNA % tail (DNA tail intensity).

Result and discussion
As shown in Fig. 1A, different types of Au and Ag nanoparticles were synthesized using Tyr, Trp, INH, EGCG, and Cur. The biomolecules behaved as reducing as well as stabilizing agents in the preparation of nanoparticles. Based on the reducing biomolecule, the nanoparticles are denoted as Au Tyr , Au Trp , Au INH , Au EGCG , Au Cur , Ag Tyr , Ag Trp , Ag INH , Ag EGCG , and Ag Cur , respectively. The chemical structure and presene of specific functional group in Tyr, Trp, INH, EGCG, and Cur is provided in Fig. 1B. The reducing functional group reduces metal ions, and post-synthesis confer biomolecular surfaces around the nanoparticles 11,26,28 . In alkaline conditions (pH 8.4-8.6), the phenol (Tyr) and indole (Trp) are responsible for reducing the metal ions of Au and Ag, leading to the formation of Au Tyr , Au Trp , Ag Tyr , and Ag Trp , respectively 11,28,47 . Recently, we have shown that the slow hydrolysis of INH at alkaline pH produces hydrazine (H 2 N-NH 2 ), which effectively reduces [AuCl 4 ] − and Ag + ions leading to the formation of Au INH and Ag INH nanoparticles, respectively 26,40 . On the other hand, EGCG contains bioactive polyphenol, which enabled the formation of Au EGCG and Ag EGCG 41 . In the case of Cur, at higher pH value (pH 9.4), the production of phenolate and enolate anions starts the reduction of [AuCl 4 ] − and Ag + ions to generate Au Cur and Ag Cur , respectively 42 . The representative reduction of [AuCl 4 ] − or Ag + ions from Tyr and INH moieties under alkaline reduction conditions is depicted in Fig. 1C. The molecular orientation of oxidised-Tyr, oxidised-INH, and oxidised-Cur on the surface of respective Au and Ag nanoparticles is illustrated in Fig. 1D 26,28,42,48 .
Previously, it has been reported that in the case of metal nanoparticles, specific solution color can advocate the formation of nanoparticles, and it can be confirmed using UV-visible spectroscopy. The development of red or wine-red color in the case of Au nanoparticles and yellow color for Ag nanoparticles solutions is characteristic of the formation of nanoparticles 18,26 . Figure 2 represents the UV-visible absorbance spectra and digital photographs of Au and Ag nanoparticles developed by employing Tyr, Trp, INH, EGCG, and Cur as reducing as well as stabilizing agents, respectively. The inset digital photographs ( Fig. 2A-E) are showing the red to winered color or yellow color for Au and Ag nanoparticles solutions, correspondingly. The UV-Visible spectroscopy is a fundamental technique to confirm the formation of stable metal nanoparticles in an aqueous medium. In general, metal nanoparticles exhibit a unique phenomenon known as surface plasmon resonance (SPR), where the conducting electrons of metals oscillate together in resonance with certain wavelengths upon interaction with an electromagnetic field 49,50 . Therefore, UV-visible spectroscopic analysis of Au and Ag nanoparticles was carried out that confirmed the formation of nanoparticles. [AuCl 4 ] − and Ag + ions are reduced using different reducing agents under the alkaline condition to form Tyr, Trp, INH, EGCG, and Cur-capped Au and Ag nanoparticles. Different Au nanoparticles, Au Tyr , Au Trp , Au INH , Au EGCG , and Au Cur , exhibited their respective SPR band maxima at ca. 520 nm ( Fig. 2A), 528 nm (Fig. 2B), 532 nm (Fig. 2C), 503 nm (Fig. 2D), and 531 nm (Fig. 2E), respectively. All these SPR band maxima values are the characteristic feature of Au nanoparticles that gives rise to an absorption band at 510-540 nm. Similarly, UV-visible spectral analysis confirmed the formation of Ag nanoparticles. Ag Tyr , Ag Trp , Ag INH , Ag EGCG and Ag Cur nanoparticles display their absorption maxima at ca. 406 nm ( Fig. 2A), 417 nm (Fig. 2B), 395 nm (Fig. 2C), 406 nm (Fig. 2D) and 412 nm (Fig. 2E), respectively, the distinctive feature of Ag nanoparticles 18,27,28,30 .
Moreover, a single SPR absorption maximum was detected in the spectra of Au and Ag nanoparticles, indicating that these nanoparticles may be spherical in shape without forming any other shaped structures [50][51][52] . It was further confirmed by microscopic imaging of nanoparticles solutions as represented in Fig. S1. The TEM micrographs of Au and Ag nanoparticles solutions confirm the spherical to quasi-spherical shape and nano-size of all the nanoparticles. In addition to being predominantly spherical or quasi-spherical shapes, some of the nanoparticles have irregular morphology, as seen in Fig. S1. Further, the size distribution profile of synthesized Au and Ag nanoparticles was determined using a DLS particle size analyzer. The DLS measurements display that all the nanoparticles are falling in the nanoscale range and are found to be sub 100 nm in size (   www.nature.com/scientificreports/ observed that these nanoparticles bear moderate surface charge, and still, they are stable. This stability is due to the presence of Tyr and Cur on the surface of the nanoparticles; they stabilize the Au Tyr and Ag Cur nanoparticles by forming a protective steric shield around the nanoparticles 45,53,54 . Additionally, it has been reported that the negative repulsion forces in the solution help in high stability and high dispersity of the nanoparticles. The high ζ potential values for all nanoparticles were further confirmed by the stability of Au and Ag nanoparticles prepared using different reducing agents. We next employed FTIR spectroscopy to identify the coordination of coated nanoparticles, and also to obtain direct evidence for the presence of functional groups in the surface corona. The distinguishing functional group vibrational frequencies of reducing molecules (Tyr, Trp, INH, EGCG, or Cur) on the surface of Au and Ag nanoparticles confirmed their presence (Fig. S2). For example, Au INH nanoparticles reveal characteristic INH functional vibrational frequencies at 3277 cm −1 and 1634 cm −1 for N-H stretching and carbonyl group C=O, whereas these were shown at 3263 cm −1 and 1634 cm −1 for Ag INH nanoparticles. Similarly, the FTIR spectra of Trp synthesized Au Trp , and Ag Trp nanoparticles show the carbonyl stretching frequency of oxidised-Trp at 1636 cm −1 and 1635 cm −1 , respectively. These spectral frequencies are attributed to the formation of carboxylate ions during reaction 55 . Likewise, Tyr capped-Au Tyr and Ag Tyr nanoparticles have a carbonyl stretching frequency around 1639 cm −1 and 1635 cm −1 , due to the carboxylate ion formation. The observed vibrational frequencies in all the cases are concomitant to the reported distinguishing functional group vibrational frequencies 26,40,45 . The presence of functional groups vibrational frequencies even after Au and Ag nanoparticles formation with changes in the wavenumber verified the existence of oxidized-reducing moieties on the surface of respective nanoparticles, as illustrated in Fig. 1D.
It is apparent from the above studies that the nanoparticles prepared with Tyr, Trp, INH, EGCG, or Cur produced a biomolecular corona supported on inorganic metal (Au and Ag). This observation led to our hypothesis that nanoparticles may possess specific biological effects. Therefore, we assayed for biological activities and cytotoxic properties of the nanoparticles.
It has been reported that aging, cancer, cataracts, brain dysfunction, and cardiovascular diseases are associated with an excess amount of reactive oxygen species (ROS) in the body. The excess production of ROS influences metabolic processes and plays an essential role in the development of diseases 56 . On the contrary, nanoparticles have the competency to scavenge the free radicals, and they can play an important role in preventing various diseases linked with the free radicals attack on cells and biological targets 57,58 . In this context, various nanoparticles, such as silica nanoparticles, iron-oxide nanoparticles, cerium oxide nanoparticles, metal nanoparticles, and organic (i.e., melanin, lignin) nanoparticles have been designed and assessed for their potential free radical scavenging capacity (RSC) 18,30,[59][60][61][62] . Here, we have investigated the in vitro RSC of different Au and Ag nanoparticles employing ABTS assay. In this test, the decolorization of ABTS radical cation (ABTS ·+ ) solution in the presence of nanoparticles suggestive of the in vitro RSC of the respective nanoparticle. As illustrated in Fig. 3A-E, the RSC of nanoparticles was dose-dependent, and with the increasing doses of nanoparticles, the RSC of both Au and Ag nanoparticles for all the ten nanoparticle systems increased till the evaluated doses. For example, Au Trp nanoparticles show 0.37%, 0.89%, 0.89% and 1.41% of RSC, and Ag Trp nanoparticles display   Fig. 3 also reveal that Ag nanoparticles show higher RSC compare to Au nanoparticles, which is complimentary to reported literature 30 . As discussed, at the constant 1.2 ppm dosages of Au Trp and Ag Trp nanoparticles, 1.41% and 6.20% RSC were observed, respectively. This further specifies that in addition to distinctive biomolecular surfaces, the metal core also plays an important role in determining the in vitro RSC competency of nanoparticles. For control experiments, and to confirm the role of the metallic core in RSC of nanoparticles 18 , the pristine reducing agents and precursor metal ions were also employed to validate the intrinsic nature of metal core in RSC of nanoparticles (Fig. S3).
It is further envisioned that due to the presence of bioactive surfaces on nanoparticles, the nanoparticles may exhibit nanozyme-mimic activity. It was confirmed employing TMB, which is a chromogenic substrate and, in general, used to evaluate the peroxidase-like activity of nanomaterials 18,26,30 . The oxidation of chromogenic TMB in the presence of H 2 O 2 can be catalyzed by metal nanoparticles, and it can be observed from the blue coloration of the reaction mixture. It has been reported that the Ag nanoparticles have higher peroxidase-like activity than Au nanoparticles 18,26 . Here, we report a similar trend for Tyr, Trp, INH and EGCG synthesized nanoparticles, as shown in Fig. 4 (for 1.2 ppm metal concentration) and Fig. S4 (for 0.6 ppm metal concentration). The observations support the reported literature and validate the role of the metal fraction of nanoparticles in their peroxidase nanozyme-mimic activity. However, it was interesting to detect that in Cur reduced nanoparticles at a given metal concentration (1.2 ppm, Fig. 4), Au Cur has higher peroxidase-mimic activity compare to Ag Cur nanoparticles. This observation indicates that in addition to the metal core of the individual nanoparticles, the www.nature.com/scientificreports/ bioactive surfaces of nanoparticles can also be influential in determining their behavior. In addition, when we compared the activity of different biomolecular surface's reliant nanozyme-like activity, it was revealed that for various Ag nanoparticles, the activity was changed, and it was EGCG ˃ Tyr ˃ INH ˃ Trp ˃ Cur, respectively. In this context, it has also been reported that the size of the nanozyme is important, and smaller nanoparticles show higher enzyme-like activity due to the high surface-to-volume ratio 63,64 . In one of the reports, a size-dependent catalytic activity of core-shell nanoparticles is also discussed, and the particles were found to have biosensing applications 65 . Additionally, the influential role of Au and Ag nanoparticles surface corona, and metallic core reliant RSC and in-vitro peroxidase enzyme-like behavior has been confirmed 30 . Therefore, the active role of the surface of the nanoparticles in determining their behavior must be assigned carefully, and all the other parameters must be controlled. After in vitro assessment, to verify the impact of metal nanoparticles in conjugation with bioactive surfaces on cell viability/cytotoxicity, we employed them against m5S mouse skin fibroblasts. The multiplex viability/cytotoxicity assay kit was used to determine the cell viability and cytotoxicity. The multiplex viability/cytotoxicity assay kit comprises cell counting kit-8 (CCK-8) along with the LDH examination kit. In this, the CCK-8 can measure the dehydrogenases activity (in living cells) by determining the number of viable cells, whereas the released LDH from damaged cell membranes can be determined by employing the LDH assay kit to confirm the number of dead cells (cytotoxicity). In a single point incubation period of 16 h with various doses of nanoparticles, there was a reduction in the viability of cells with the increasing doses, as illustrated in Fig. 5. Amino acid synthesized Au nanoparticles showed almost similar behavior, and there was no significant difference even at different concentrations. For example, Au Tyr nanoparticles showed 86.27%, 85.60% and 82.63% cell viability at 0.5, 1.0 and 5.0 µg/ml treatments, whereas, Au Trp nanoparticles showed 89.87%, 88.23% and 84.27% cell viability respectively. A similar trend was observed for Ag Tyr nanoparticles, whereas in the case of Ag Trp nanoparticles, 92.8-81.27% cell viability was detected at 0.5 and 5.0 µg/ml, respectively. Although the higher impact on cell viability was reported in the case of Ag Trp with the increasing doses at a specific concentration, there was no significant difference than other nanoparticles prepared using amino acids. For examples, at a constant 5.0 µg/ml concentration Au Tyr , Au Trp , Ag Tyr and Ag Trp nanoparticles reveal 82.63%, 83.57%, 84.27% and 81.27% cell viability respectively. On the other hand, in the case of other reducing agents, differential action has been reported. For instance, at the constant 5.0 µg/ ml concentration Au Tyr , Au Trp , Au INH , Au EGCG , and Au Cur nanoparticles show 82.63%, 83.57%, 68.10%, 79.10% and 61.37% cell viability respectively. The differential cell viability of similar core containing nanoparticles with a constant concentration reveals that the biomolecule surface of the nanoparticles is an essential property of any material to determine their biological capabilities. Similarly, there was no significant difference in the case of an www.nature.com/scientificreports/ amino acid (Tyr and Trp) synthesized nanoparticles to confirm the role of the metal core. However, for all the other reducing agents, the impact of the metal core was noticeable, confirming the importance of nanoparticle composition. For a case, Au Cur and Ag Cur nanoparticles show 61.37% and 80.50% cell viability, respectively. It indicates that the metal fraction is also an important factor in controlling the biological behavior of nanoparticles since Cur surfaces are constant on both the nanoparticles. A similar trend was observed for INH and EGCG reduced nanoparticles of Au and Ag. Subsequently, a decrease in the viable cell number may be the hallmark of cytotoxicity originating from nanoparticles. Therefore, in the present study, we have further performed a dosedependent cell cytotoxicity assessment of Au and Ag nanoparticles. Furthermore, it is important to state that the cell cytotoxicity may be dependent on the surface moieties in addition to the metal fraction of nanoparticles, as discussed above for cell viability. The cell cytotoxicity was complimentary to the cell viability analysis of various nanoparticles, as illustrated in Fig. 6. The cell viability and cytotoxicity in similar experimental conditions were also determined for all the pristine reducing agents (Tyr, Trp, INH, EGCG, and Cur) and free metal ions (AuCl 4 − and Ag + ions) as illustrated in Fig. S5. Furthermore, to validate the hypothesis, we have employed various Au and Ag nanoparticles against macrophages, which are the master cells of the innate immune system, and contribute to the immunity, repair, and tissue hemostasis (Fig. S6). The Tyr, EGCG, and Cur synthesized nanoparticles stimulated cells, and there was an increment in the cell numbers at 1.0 and 5.0 µg/ml. In the recent past, comparable observations have been reported in quercetin capped metal nanoparticles 18 . Here, the observations complement that the cell viability of the nanoparticles can be controlled by a suitable surface corona of the particles in addition to the metallic core of Au or Ag nanoparticles.
Nevertheless, many metal and metal oxide nanoparticles have been reported for their potential toxicity, but the genotoxicity of nanomaterials is relatively unexplored, and inconsistencies are found in the literature 66,67 . However, urgent attention is required towards the evaluation of potential genotoxic properties of metallic nanoparticles, which can be crucial in regulatory safety assessment. Therefore, we have evaluated the nanoparticlesinduced genotoxicity by employing single-cell DNA damage assay, and results are expressed in % DNA tail intensity with respect to treatments in Fig. 7. Here, different Au and Ag nanoparticles of various concentrations (0.5, 1.0, and 5.0 µg/ml) were incubated with m5S mouse skin fibroblasts for 24 h, followed by cell harvesting and DNA damage assessment using Comet Analyzing Software. As illustrated in Fig. 7A, in terms of the extent of magnitude, only Au Tyr of 5.0 µg/ml treatment induced greater DNA damage with around four-fold tail intensification compared to control. However, it was surprising to witness that all the Au and Ag nanoparticles synthesized www.nature.com/scientificreports/ with different reducing agents exhibited lower levels of DNA damage as measured by tail intensity. Moreover, it is interesting to observe that the nanoparticles of Au and Ag showed reduced genotoxicity when compared with the corresponding free metal ions (Fig. S7). Free metal ions are known to have a higher level of genotoxicity originating from various chemical mechanisms 68 . Furthermore, in the case of pristine INH, the antibiotic was found to induce DNA damage. However, when it is present on the surface of nanoparticles (Au INH and Ag INH ), the toxicity impact was controlled. It is interesting that in the free state, both the ions and INH showed significant damage to DNA (Fig. S7), but in the form of Au INH and Ag INH nanoparticles, the toxic effect was deactivated as illustrated in Fig. 7C. These results are concomitant to investigated cell viability and cytotoxicity assessments, as discussed above. However, to further confirm the results of genotoxicity (DNA tail assay), the DNA ladder (DNA fragmentation) assessment was performed. The DNA ladder assay is an essential method for the rapid screening of DNA break in cell populations. The specific ladder pattern is an indication of apoptosis by the activation of a nuclear endonuclease 69,70 . In this technique, DNA cleavage can be seen in the form of a ladder pattern due to damage caused by nanoparticles or by the activation of the nuclear endonuclease. Moreover, DNA fragments after treatment with nanoparticles can be visualized in standard agarose gel electrophoresis. Therefore, DNA ladder assay was performed. As illustrated in Fig. 8A, it was confirmed that various Au and Ag nanoparticles do not induce apoptosis in m5S skin fibroblasts. It was interesting to observe that the DNA was damaged by pristine INH (Fig. 8B, Lane 2) and free metal ions (Fig. 8B, Lane 7, and 8). However, when the INH was used as a reducing agent, it did not promote any fragmentation to DNA after treatment with Au and Ag nanoparticles, as shown in Fig. 8A, Lane 4, and 9, respectively. This observation specifies that toxicity induced by DNA fragmentation must be considered while it is being assigned to nanoparticles. The individual surface moieties and free metal ions can show DNA fragmentation, leading to apoptosis. However, this nature may be altered during the synthesis of nanoparticles, and carefully designed nanoparticles can behave differently, as shown in the case of INH-reduced Au INH and Ag INH (Fig. 8). On the contrary, other reducing agents, namely Tyr, Trp, EGCG, and Cur were found to have no apoptosis activity (Fig. 8B, Lane 3-6), and when they are used as reducing agents, the respective nanoparticles (both Au and Ag) were also inheriting the same nature. It is an interesting observation that during the nanoparticles synthesis, the surface moieties controlled the inherent nature of nanoparticles, and the characteristic metal ions nature was altered in the nanoparticulate form as illustrated in Fig. 8A,B. From the discussion, it can be concluded that the biological action of nanoparticles is derived from the complementary effect of the metallic core and biomolecular www.nature.com/scientificreports/ surfaces of nanoparticles. However, it is imperative to state that all the nanoparticles formulated in this study have a distinctive hydrodynamic radius, which may also contribute to their biological interactions in addition to their composition and surfaces. Therefore, we must be cautious while assigning the biological characteristic of any physicochemical property of nanoparticles.

Conclusions
In the present study, single-step approaches to prepare Au and Ag nanoparticles with a variety of surface corona have been reported (Fig. S8). We have achieved different and bioactive surface corona using amino acids (Tyr and Trp), antibiotics (INH), and plant-derived polyphenol (EGCG and Cur) molecules. The impact of the metal composition of nanoparticles and their respective surface corona have been established in vitro using m5S mouse skin fibroblasts and RAW 264.2 macrophages. We have shown that surface corona dictates the physicochemical properties and the biological interactions of the Ag and Au nanoparticles. Overall, the present study suggests novel approaches towards the engineering of surface properties and the composition of nanoparticles to enhance their biological applicability. Additional studies are required to examine the interaction of nanoparticles with biological fluids for predicting their activities in vivo.