Isodon rugosus seeds, acquired from the Plant Cell Culture Laboratory, Department of Biotechnology, Quaid-e-Azam University, Islamabad, were germinated under sterile conditions. Callus culture was established on MS medium (Murashige and Skoog, 1962[31]) and maintained in growth room at 25±2 °C for 16/8 hrs (light/dark) photoperiod. After 40 days, explants fragments were shifted to MS medium supplied with various concentration of TDZ (1, 2, 3, 4, 5 mg/l) for callus induction (Figure 1A(Fig. 1)). Extract of Callus (CE) and in vitro derived whole plant (WPE) was prepared by adding 10 g of freshly harvested callus and plant sample separately into 500ml flask. 100 ml of distilled water was added to each flask and boiled. After overnight incubation at 40 °C, the broth was filtered carefully. These extracts were further used for the bio-reduction of Zn⁺ to Zn^°.

Synthesis of NPs depends on type and concentration of plant extract, metallic salt, response time, pH and temperature (Dwivedi and Gopal, 2011[13]). Zinc Acetate Dihydrate (C₄H₆O₄Z_n) was used as a precursor solvent. 0.02 M solution of zinc acetate (50 ml) was mixed with 1 ml sample extract separately (1:50) and was stirred constantly. pH of the medium was maintained at 12 by adding 2M Sodium Hydroxide (NaOH) drop-wise. The reduction was attributed to phenolics, terpenoids, polysaccharides and flavones present in the extract. White and pale precipitates were seen for CE and WPE, respectively. The resulting precipitates were washed three times (first with ethanol and then twice with distilled water). Samples were centrifuged at 6000 rpm for 15 min and incubated overnight at 60 °C for drying. The powdered samples of ZnONPs obtained after drying were stored until further use.

The Folin-Ciocalteu reagent (FCR) method, also called the gallic acid equivalence method (GAE), was used to measure the total phenolic content in CE and WPE (Singleton et al., 1999[45]). This reagent does not only measure phenolic content but is also used to measure any reducing substance in the sample extract. Folin's reagent assist to detect amine and sulfur containing compounds, which are mainly involved in reduction and stabilization of NPs. In brief, 20 µl of sample extract (CE and WPE) was mixed with 90 µl of FC reagent. After 5 min of incubation, sodium carbonate (Na₂CO₃, 6 % w/v) was added. Absorbance was measured (in triplicates) at 725 nm (Huang et al., 2007[21]).

To measure the level of flavonoids in CE and WPE, we followed the Willet's method of estimation with some alteration. 20 µl aliquot of sample extract was mixed with 10 µl of potassium acetate (1 M) and 10 µl of 10 % Aluminium chloride (AlCl₃). Samples were incubated for 30 min, followed by addition of 160 µl distilled water. Absorbance was detected at 415 nm (Thermo Fisher Scientific, model 4001/4). Quercetin was used to make the calibration curve. Calculation of TFC was carried out in triplicate (Willet, 2002[54]).

Characterization of the synthesized NPs for their configuration, diameter and surface modification was performed by XRD, EDX and SEM analysis. Optical properties of ZnONPs were characterized through HALO DB-20 UV-absorption spectra with the wavelength range of 300-700 nm. Samples were sonicated for 30 min and the aqueous solution was subsequently evaluated at room temperature for the optical band gap. Crystalline nature of powdered samples was evaluated by X-ray diffraction analysis (Shimadzu-6000), operating at 30 kV and 40 mA. Elemental compositions and size of the synthesized NPs was characterized using SEM (SIGMA model, MIRA3 TESCAN) operated at an accelerating voltage of 10 kV and EDX detector. All the apprehensible functional groups involved in capping and reduction of ZnO NPs were identified using FTIR (Bruker V70).

Investigation of four multiple drug resistant bacterial strains i.e., Bacillus subtilis (ATCC 6051), Pseudomonas aeruginosa (ATCC 9027), Klebsiella pneumoniae (ATCC 70068) and Staphylococcus epidermidis (ATCC 12228) was carried out for biocidal activities of ZnONPs. Microbial susceptibility testing was performed by agar disc diffusion technique. Bacterial broth was evenly dispensed on nutrient agar plate at density of 1×10⁸ CFU/ml, previously refreshed in shaking incubator at 37 °C, 200 rpm in nutrient agar broth (Oxoid-CM0003). Zinc Acetate Dihydrate (2 mM) and Amoxicillin (standard antibiotic) (10 µg/ml) were added in the experiment as negative and positive controls respectively. Samples were dissolved in deionized water prior to experiment. Each sterile filter disc placed on culture plate, was loaded with 10 mg/ml of sample and incubated for 24 hrs at 37 °C. Zones of inhibition (mm) were measured afterwards. Antibacterial assay was performed in triplicates.

All statistical measurements were performed in triplicates and analysed by Pareto analysis of variance (ANOVA) with Duncans multiple range test (Duncan, 1955[12]). Values in text and in figures were statistically analysed as mean ± SD (standard deviation). All the graphical analyses were done by using OriginPro software. The probability was considered to be significantly different at the level of P˂ 0.05 (95 %).

A number of primary and secondary metabolites (phytochemicals) have been reported to perform primal role in the reduction of metal ions to NPs. The selection of plants for the purpose of NPs biosynthesis is mostly based on their capacity to produce important secondary metabolites that plays their respective role as capping and reducing agents. These capping and reducing agents stabilize the NPs, thus helping them to avoid agglomeration. Ponarulselvam et al. (2012[39]) selected Catharanthus roseus for the evaluation of vincristine and vinblastine as phyto-reducing agents. Both were selected due to their potential nematicidal activities. Similarly, Akhtar et al. (2016[6]), reported that plectranthoic acid inhibits proliferation of prostate cancer cells by the activation of 5′AMP-activated kinase (AMPK). Present study reveals the maximum production of TPC (11.3 mg g^-1 DW) and TFC (0.54 mg g^-1 DW) when treated with TDZ (2.0 mgL^-1) in case of CE. The TPC and TFC for WPE were 6.1 mg and 0.5 mg g^-1 DW, respectively (Figure 1B(Fig. 1)). Presence of sufficient amount of TPC and TFC shows that Isodon rugosus can prove efficient in ZnONPs biosynthesis.

Optical characterization of the ZnONPs was carried out using UV-Vis absorption spectrophotometer in the range of 300-700 nm. Figure 1C(Fig. 1) shows characteristic spectra of ZnONPs synthesized using CE and WPE of Isodon rugosus. Dispersed ZnONPs absorbed the radiations approximately at 362 nm in case of CE mediated ZnONPs (C-ZnONPs) and 382 nm in case of WPE mediated ZnONPs (W-ZnONPs). Narrowing of the peaks confirms the reduction of zinc acetate into ZnONPs. Indramahalakshmi (2017[23]) and Jamdagni et al. (2016[24]) also reported similar characteristic peaks for ZnONPs. The band gap energy of C-ZnONPs and W-ZnONPs was 3.692 keV and 3.691 keV respectively, representing characteristic absorption peaks of ZnONPs. This may be due to the presence of different fractions of secondary metabolites in both CE and WPE, which in turn resulted in shape variation in ZnONPs. Similar findings were previously reported by Manokari et al. (2016[30]).

X-ray diffraction determines phase purity (crystalline nature), structural properties, thickness of thin films and the atomic arrangement of amorphous materials. In XRD, X-rays penetration provide information about overall structure of the material under observation. Diffraction pattern given in Figure 2A and 2B(Fig. 2) shows several peaks in the range of 30-80^° which corresponds to 100, 002, 101, 102, 110, 103, 200, 112, 201, 004 and 202 crystal planes, respectively. The data revealed crystalline hexagonal structure of ZnONPs.

SEM images of the biosynthesized ZnONPs are shown in Figure 2 (C, D(Fig. 2)). It is observed that alkaline pH results in shape-dependent synthesis of ZnONPs. Higher pH results in availability of large number of functional groups for capping and stabilization of NPs. The exact mechanism, however, is difficult to grasp due to involvement of large amounts of diverse phyto-reducing metabolites. Current study revealed that C-ZnONPs were irregular hexagonal shaped (Figure 2C(Fig. 2)) while W-ZnONPs were roughly triangular (Figure 2D(Fig. 2)). The stability and morphology of ZnONPs were increased by the combined effect of pH and reducing agents (primary and secondary metabolites of samples extract). Previous findings suggested that the adjustment of pH was compulsory for the stability of ZnONPs. The current findings are well supported by the findings of Song et al. (2009[46]) and Amin et al. (2012[7]). Ghodake et al. (2010[15]) reported similar findings that alkaline pH is responsible for hexagonal and triangular shaped NPs. Altering pH results in modification of the surface charge upon the secondary metabolites which alternatively affects their binding capacity and reduction potential of metal ions during the synthesis of NPs (Kumar and Yadav, 2008[28]; Pereira et al., 2015[37]). Sathishkumar et al. (2010[43]) confirmed that alkaline pH plays a major role in shape and size variations of NPs. Additionally, alterations in shape and size of NPs may also occur due to other parameters like reaction time, salt concentration, localization of NPs, and change in functional molecules.

The chemical composition of ZnONPs was confirmed by EDX-ray spectrum. The EDX profile exhibits strong signals of zinc and oxygen in both C-ZnONPs and W-ZnONPs (Figure 3A, B(Fig. 3)), (Table 1(Tab. 1)).

The FTIR spectra of ZnONPs (CE/WPE) was observed in the range of wave-numbers 500-3500 cm^-1 to identify the functional groups involved in the reduction and capping/stabilization of the synthesized NPs (Figure 4(Fig. 4)). FTIR analysis confirmed the presence of Zn-O bond in both C-ZnONPs and W-ZnONPs due to the presence of peaks at 893.71 cm⁻¹ and 886.04 cm⁻¹, respectively (Pinto and Nazareth, 2016[38]). ZnONPs showed broad bands at 3375.54 cm⁻¹ (C-ZnONPs) and 3377.88 cm⁻¹ (W-ZnONPs) which was attributed to the stretching vibrations of hydrogen bond (O-H) as reported by Thirumavalavan et al. (2013[51]). Spectra at 1216.79 cm⁻¹ (C-ZnONPs) and 1216.81 cm⁻¹ (W-ZnONPs) confirm C-N stretch bond of aliphatic amines. Peaks at 2360.81 cm⁻¹ (C-ZnONPs) and 2359.57 cm⁻¹ (W-ZnONPs) represent asymmetric C=O stretching (Shaikh et al. 2017[44]).

These findings clearly demonstrate the attachment of primary and secondary metabolites to the ZnONPs. Electrostatic forces between the negatively charged groups of organic molecules and positively charged zinc ions are responsible for the binding of these functional groups. This interaction makes these NPs a perfect candidate for various applications such as drug delivery (Wang et al., 2013[53]) and study of biological interactions (Song et al., 2009[46]).

Suforhodamine B assay was used to screen in vitro cytotoxicity of extracts and NPs using HepG2 cell line. Ionic strength and pH of NPs depend upon formation of hydroxyl (OH⁻) ions which directly influence the repulsion and attractions of particles (Peng et al., 2017[36]). Cells were treated with 200 µg/ml of ZnONPs (C-ZnONPs and W-ZnONPs), CE and WPE for 24 hours. Our results suggested that CE showed slightly higher cytotoxicity (29.47 %) as compared to WPE (31 %). Morphological changes were also observed in both samples i.e., cells became long and fibrous. On the other hand, cell viability in C-ZnONPs was lower (23.92 %) as compared to W-ZnONPs (35 %). Morphologically, cells became small and round upon treatment of C-ZnONPs while long and fibrous cells were observed in the presence of W-ZnONPs. Najim et al. (2014[33]) also reported morphological changes induced by cytotoxic ZnONPs on different cell lines (U937, SH-SY5Y, differentiated SH-SY5Y, and Hs888Lu). Morphological changes in cells may affect their metastatic process including substrate attachment, migration and invasion (Brandhagen et al., 2013[11]). DMSO (1 %), used as a negative control showed slight cytotoxicity and doxorubicin (positive control) showed high cytotoxicity at 34 µM concentration (Figures 5(Fig. 5), 6(Fig. 6)).

Akhtar et al. (2012[5]) reported cytotoxicity of ZnONPs in HepG2 cells due to production of reactive oxygen species. In spite of unique physical and chemical properties of metallic NPs, quantum size and shape also assist in the field of therapeutics (Khan et al., 2017[26]). In present study, the surface charge ratio of ZnONPs was pH dependent to absorb various chemical clusters (-ZnOH²⁺, -ZnOH, -ZnOˉ). In this regard, it was hypothesized that surface orientation of ZnONPs allows them to conjugate with protein targets or chemical groups. Therefore, ZnONPs possess cytotoxicity towards cancer cells (Figures 5(Fig. 5), 6(Fig. 6)). Literature suggests formation of ROS as leading cause of cell death by apoptosis (Figure 7(Fig. 7)).

Our present findings suggest configurational role of ZnONPs towards cytotoxicity. It was established that callus derived hexagonal NPs (C-ZnONPs) showed high cell inhibition (% viability 23.92 %) as compared to whole plant derived triangular NPs (W-ZnONPs) in which cell viability was 35 % (Supplementary Table 2). From FTIR and SEM data, we concluded that this may be due to difference in surface charge of secondary metabolites which eventually control the bonding of NPs with cancer cells. Suresh et al. (2018[49]) supported similar results. They reported that biologically synthesized hexagonal ZnONPs exhibit good cell inhibition and viability against Daltons Lymphoma Ascites (DLA).

Also, in vitro derived callus extract (CE) showed cell viability of 29.47 % while whole plant extract (WPE) showed 31 % cell viability (Figure 6(Fig. 6)). The current findings are in agreement with the results of Kamalanathan and Natarajan (2018[25]). They reported that potency of callus was better than wild leaf extract sample against human breast cancer cell line MCF-7. Literature review proved presence of commercially important anticancer phenolics and triterpenoids i.e., plectranthoic acid, limonene, rosmarinic acid, caffeic acid (Abdel-Mogib et al., 2002[2]; Akhtar et al., 2016[6]; Hossan et al., 2014[20]; Hafidha et al., 2018[17]; Krishna et al., 2017[27]). To the best of our knowledge very little data is available that compares the cytotoxicity of CE, WPE, C-ZnONPs and W-ZnONPs against HepG2 cell line.

Literature reported the increased emergence of drug resistant bacteria in the clinical field directs the scientific community to look for alternatives of antibiotics to minimize the risk from bacterial infections. Essential oils produced from Isodon rugosus consist of complex mixture of volatile molecules such as sesquiterpene, amorphene, germacrene-D, triterpenoids, plectranthoic acid A and B, acetyl plectranthoic acid and plectranthadiol. These volatiles compounds potentially exhibit bactericidal property against various human pathogenic bacteria. Similarly Hussain et al. (2017[22]) and Faleiro (2011[14]) reported the effectiveness of essential oils against human pathogenic bacteria. The bactericidal action of essential oils is linked to its hydrophobicity, resulting in increased cell permeability and subsequent leakage of cell constituents. Permeability of ZnONPs and their capability to bind with the cell membrane generates reactive oxygen species (ROS), thus enhancing the efficacy against drug-resistant bacterial strains (Abinaya et al., 2017[4]). Our results (Figure 8A, B(Fig. 8)) were in accordance with the data reported by Premanathan et al. (2011[40]). According to them, the effects of ZnONPs were more potent against Gram-positive bacterial strains as compared to Gram-negative strains. Our evaluation of potency of ZnONPs also revealed more potent bactericidal activity against Gram-positive S. epidermidis and B. subtilis as compared to Gram-negative, K. pneumoniae and P. aeruginosa, respectively (Table 2(Tab. 2)).

It was hypothesized that Gram-positive bacteria possess specific cell wall components that help in embedding ZnONPs within the cellular wall. The presence of ZnONPs within the walls results in cellular disintegration and leakage of internal components. In case of Gram-negative bacteria, no such embedding is observed hence only a small amount of surface bound ZnONPs causes partial leakage of internal contents.

Our findings endorse the idea of biocidal effect of ZnONPs due to combination of various mechanisms such as ROS production in response to zinc ions within the bacteria. Minor sensitivity of S. epidermidis, B. subtilis compared to K. pneumoniae and P. aeruginosa was due to fact that the outer layer of Gram-positive bacteria is tightly packed due to assembly of lipopolysaccharide (LPS) and peptidoglycans. Furthermore, rational estimation of decline in sensitivity may be due to reduced stability of O₂^¯ materialization at the surface and the decrease of pH value in the medium (Tang and Lv, 2014[50]) (Figure 9(Fig. 9)).