Zinc oxide nanoparticles with size <100 nm, phosphate buffer saline (PBS), 10 X, and glutaraldehyde were obtained from Sigma-Aldrich, United States of America. The antibiotics chloramphenicol and ampicillin were purchased from Bio Basic Inc., Canada. The bacterial culture medium Luria Bertani (LB) was purchased from Laboratories CONDA, Spain.

Scanning electron microscope with energy dispersive X-ray (SEM- EDX) (JOEL JSM 6710F, Japan) was used to characterize the surface morphology and size of ZnO NPs through SEM operated at an acceleration voltage of 4 kV with a working distance of 4.7 nm, while the elemental composition of ZnO NPs was studied by EDX analysis. X-ray diffractometer (XRD) (Lab X, SHIMADZU, XRD-6000, Japan), operated at an angle of 2θ with 40 V and 30 mA current, was applied to confirm the crystalline structure and size of the nanomaterial. Further, Fourier transform infrared (FTIR) spectroscopy (Perkin-Elmer Spectrum RX1, United States) was used to confirm the chemical composition of ZnO NPs.

A stock solution of 320 μg/ml ZnO NPs was prepared by suspending ZnO nanopowder in LB broth, mixed homogenously by vortex, and diluted with LB broth to prepare the working concentrations of ZnO NPs.

The Gram-negative bacterium S. marcescens (ATCC 43862) and the Gram-positive bacterium E. faecalis (ATCC 29121) were obtained from the Faculty of Science, Universiti Tunku Abdul Rahman and then subcultured to the mid-log phase in LB broth for further utilization.

A 5 ml of mid-log phase suspension of S. marcescens with an optical density of 0.05 at 600 nm (OD600) was exposed to 5 ml of ZnO NPs to obtain the final concentrations of 5, 10, 20, 40, 80, and 160 μg/ml of ZnO NPs in 15 ml centrifuge tubes and incubated for 24 h at 37°C. Simlarly, the mid-log phase suspension of E. faecalis with the same OD600 of 0.05 was treated with various concentrations of ZnO NPs ranging from 5 to 160 μg/ml of ZnO NPs for 24 h, but incubated at 35°C. The bacterial suspension without ZnO NPs was used as the negative control. The suspension of S. marcescens treated with 0.8 mg/ml of chloramphenicol and E. faecalis exposed to 1.0 μg/ml of ampicillin was used as the positive control, respectively (Conceição et al., 2014; Ray et al., 2017).

Turbidity test was conducted to investigate the antibacterial effect of ZnO NPs on S. marcescens and E. faecalis by determining the optical density of bacterial suspensions treated with 5, 10, 20, 40, 80, and 160 μg/ml of ZnO NPs by spectrophotometer at 600 nm (Libra S4, Biochrom, UK) along with negative and positive controls. LB broth was used as the blank. The absorbance from the respective concentration of ZnO NPs was subtracted from the test readings to avoid the interference by the NPs. Then, the absorbance values were compared to calculate the percentage of bacterial growth inhibition using equation 1 (Eq. 1).

Percentage of growth inhibition = (OD600 negative control–OD600 test)/OD600 negative control × 100 (Eq.1).

The FTIR spectroscopy was used to investigate the involvement of functional groups from the bacterial cell wall for the interaction of ZnO NPs on the bacterial surface. A volume of 25 ml of the bacterial suspensions treated with 160 μg/ml of ZnO NPs for 24 h, along with the negative control, were centrifuged for 10 min at 6,000 g to obtain the pellets. The pellets were then washed with 1X PBS three times to remove the unbound NPs, freeze-dried, and subjected to FTIR spectrum analysis at 4,000 to 400 cm⁻¹.

The SEM images were obtained to investigate the morphological damage caused by ZnO NPs on treated bacteria, while EDX was performed to identify the accumulation of ZnO NPs in bacterial cells. A volume of 25 ml of the bacterial suspensions treated with 160 μg/ml of ZnO NPs for 24 h, along with the negative control, were centrifuged for 10 min at 6,000 g to obtain the pellets. The pellets were then washed with 1X PBS three times and treated with 2.5% of glutaraldehyde in 1X PBS for overnight fixing. The next day, the samples were centrifuged and washed three times with 1X PBS for 10 min at 6,000 g. This washing process was repeated using distilled water. Then, the samples were dehydrated by a series of different concentrations of ethanol (25%, 50%, 75%, 95%, and 100%). The dehydration process with 100% ethanol was repeated three times, followed by critical point drying and sputter coating. The samples were then analyzed under SEM-EDX (JOEL JSM 6710F, Japan).

Statistical analysis was conducted to identify the variances when different concentrations of ZnO NPs interacted with S. marcescens and E. faecalis. The tests were done in triplicates (n = 3), and the data are presented as mean ± standard deviation. All data were analyzed by one-way analysis of variance (ANOVA) and the statistical significance was reported at p < 0.05.

The surface morphology and elemental composition of ZnO NPs were analysed by SEM-EDX. Figure 1(a) shows that the ZnO NPs were spherical particles with a mixture of rods, while the particle size ranged from 42.8 to 79.6 nm with a mean size of 59.1 nm. The EDX spectrum depicted in Figure 1(b) confirmed the presence of ZnO NPs by the peaks of zinc and oxygen molecules. The presence of zinc, oxygen, and carbon elements from EDX analysis indicated that ZnO NPs used in the present study were free from impurities. The carbon that appeared in the EDX spectrum may be from the carbon tape used for sample preparation (Varadavenkatesan et al., 2019).

The strongest diffraction peaks at 31.7°, 34.36°, and 36.19° and also the peaks at 47.05°, 56.09°, 62.38°, 65.90°, 67.45°, and 68.60° displayed in the XRD spectrum of ZnO NPs (Figure 2) corresponded to the characteristic hexagonal wurtzite crystalline structure of ZnO NPs, according to the JCPDS database card number [01-070-2551] (Ramesh et al., 2015; Kumar and Gautam, 2019; Muhammad et al., 2019). The XRD spectrum of ZnO NPs was also utilized to calculate the crystalline size of ZnO NPs using Scherrer’s equation. The crystalline size was found to be at the range of 40–47 nm with a mean size of 43 nm. In addition, FITR spectrum of ZnO NPs was used to confirm the composition of ZnO NPs. As shown in Figure 3, FTIR spectrum of ZnO NPs obtained in this study illustrated the peak absorbance at 3332.45, 1636.59, and 671.88 cm⁻¹. Earlier studies have reported Zn-O stretching at 1634.00 cm⁻¹ (Kumar and Rani, 2013), stretching of ZnO NPs at the range of 400–800 cm⁻¹ (Ramesh et al., 2015), and stretching of H-O-H vibrations between 3400 cm⁻¹ and 3600 cm⁻¹ (Sukri et al., 2019).

The absorbance of the bacterial suspensions were measured at 600 nm upon treatment with different concentrations of ZnO NPs from 5 to 160 μg/ml for 24 h, along with the negative and positive controls for the respective bacteria. The treatment of ZnO NPs resulted in a dose-dependent bacterial growth inhibition, as the higher the concentrations of ZnO NPs, the greater the percentage of growth inhibition on S. marcescens (Figure 4) and E. faecalis (Figure 5).

The percentage of growth inhibition at 5, 10, 20, 40, 80, and 160 μg/ml of ZnO NPs against S. marcescens were 1.34 ± 0.32, 5.65 ± 0.04, 16.03 ± 1.54, 21.73 ± 3.6, 32.1 ± 0.34, and 51.27 ± 4.56% respectively. Similarly, ZnO NPs caused 5.42 ± 1.27, 16.69 ± 0.88, 23.74 ± 1.23, 32.07 ± 1.65, 54.73 ± 2.15, and 63.50 ± 2.50% of growth inhibition at 5, 10, 20, 40, 80, and 160 μg/ml of ZnO NPs respectively on E. faecalis. The percentage of growth inhibition in positive control was 82.78 ± 1.08% and 83.93 ± 0.18% for S. marcescens (with 0.8 mg/ml of chloramphenicol) and E. faecalis (with 1.0 μg/ml of ampicillin), respectively.

The results demonstrated a significant (p < 0.05) bacterial growth inhibition from 10 to 160 μg/ml of ZnO NPs, however, no significant growth inhibition was detected at 5 μg/ml of ZnO NPs for S. marcescens. Whereas, the treatment of ZnO NPs caused significant (p < 0.05) growth inhibition on E. faecalis for all the tested concentrations from 5 to 160 μg/ml.

Our results indicate a better antibacterial activity of ZnO NPs on Gram-positive bacterium compared with Gram-negative bacterium as the highest concentration of ZnO NPs (160 μg/ml) caused 63.50 ± 2.50% of growth inhibition on E. faecalis compared to 51.27 ± 4.56% on S. marcescens as the maximum growth inhibition. Supporting the present results, Azam et al. (2012) reported a higher tolerance of Gram-negative bacteria over Gram-positive bacteria to ZnO NPs as the treatment of ZnO NPs caused 16 and 24% lower inhibition zone sizes on Gram-negative bacterial strains of P. aeruginosa and E. coli compared to Gram-positive bacterial strains of S. aureus and B. subtilis. Further, Premanathan et al. (2012) evidenced a more profound effect of ZnO NPs on Gram-positive bacterial strains compared to Gram-negative bacterial strains. Earlier studies by Chandra et al. (2019) and Gunalan et al. (2012) reported the growth inhibition of S. marcescens by ZnO NPs with the minimum inhibitory concentration of 64 μg/ml and 1.6 mM, respectively, at 24 h through disc diffusion assay.

Similar to our present results, previous studies have demonstrated a dose-dependent growth inhibition on E. faecalis using agar diffusion by gelatin-coated ZnO NPs and ZnO NPs, respectively (Dowlatababdi et al., 2017; Divya et al., 2018; Sukri et al. (2019) confirmed a higher antibacterial potential of green synthesized ZnO NPs on E. faecalis by reporting 22.09 μg/ml of ZnO NPs as the minimum inhibitory concentration inhibiting 50% growth for E. faecalis compared to 64.53 μg/ml for Gram-negative bacterium E. coli. A study by Nilavukkarasi et al. (2020) demonstrated a more effective antibacterial action of ZnO NPs on Gram-positive bacterium S. epidermidis than S. paratyphi, which is a Gram-negative bacterium.

Simlar to the present results, our earlier studies also demonstrated a dose-dependent bacterial growth inhibition by ZnO NPs with a higher growth inhibition of 81.18% on Gram-positive bacterium S. pyogenes compared to 65.73% on Gram-negative bacterium P. aeruginosa, respectively, when treated with 100 μg/ml of ZnO NPs for 24 h (Liang et al., 2020; Dhanasegaran et al., 2022).

The structural and functional differences between the cell wall of Gram-positive and Gram-negative bacteria have been proposed as the major reason for the higher sensitivity of Gram-positive bacteria to ZnO NPs than Gram-negative bacteria (Aysa and Salman, 2016; Getie et al., 2017; Sukri et al., 2019; Demissie et al., 2020). The more complex cell wall of Gram-negative bacteria with the outer liposaccharide membrane makes the cell wall impermeable to antibacterial agents, whereas the peptidoglycan layer present in the Gram-positive bacteria is not an effective barrier against antibacterial agents, thus making them more susceptible to antimicrobials.

The FTIR spectrums displayed in Figure 6 and Figure 7 show the involvement of several functional groups for the interaction of ZnO NPs onto the cell wall of S. marcescens (Table 1) and E. faecalis (Table 2), respectively.

The ZnO NPs treated S. marcescens showed the peak shifts from 3282.52 to 3296.21 cm⁻¹, 1641.65 to 1644.84 cm⁻¹, 1540.4 to 1531.69 cm⁻¹, 1399.24 to 1384.71 cm⁻¹, 1023.06 to 1028.17 cm⁻¹, and 844.3 to 830.46 cm⁻¹ corresponding to the stretching of O―H, C=O, C―N, COO―, C―O, and α-glycosidic linkage vibrations respectively.

A broad peak shift between 3282.52 and 3296.21 cm⁻¹ indicated the O-H stretching due to the presence of the hydroxyl group (Nandiyanto et al., 2019). According to Ifeanyichukwu et al. (2020), the hydroxyl group on the bacterial cell surface can facilitate the binding of Zn ions to the bacterial surface. The peak shifts between 1641.65 to 1644.84 and 1531.69 to 1540.4 were associated with amide I and amide II from proteins, while the peak shift from 1384.71 to 1399.24 cm⁻¹ corresponed to the carboxyl group of fatty acids and aminoacids (Ojeda et al., 2008; Liyanage et al., 2017; Nandiyanto et al., 2019). The peak shift beween 1023.06 and 1028.17 cm⁻¹ exhibited the C-O stretching coupled with C-O bending of the C-OH of carbohydrates (Nandiyanto et al., 2019), especially β 1,4 glucans, and the shift from 844.3 to 830.46 cm⁻¹ was possibly due to the α-glycosidic linkage vibrations of carbohydrates (Burattini et al., 2008).

Meanwhile, the peaks that shifted in ZnO NPs treated E. faecalis at 24 h were 3284.05–3288.62 cm⁻¹ with O-H stretching, 2927.34 to 2924.12 cm⁻¹ with CH₂ stretching, 1634.38 to 1644.95 cm⁻¹ with C=C stretching, 1231.96 to 1234.66 cm⁻¹ with PO₂ ⁻ asymmetric stretching, and 1060.29 to 1041.55 cm⁻¹ due to COH deformation. The possible involvement of hydroxyl (3288.62 cm⁻¹), methylene (2924.12 cm⁻¹) group from lipids and proteins, amide I (1644.95 cm⁻¹) from proteins, phosphate (1234.66 cm⁻¹) from phospholipids and phosphated proteins, and mannans (1041.55 cm⁻¹) (Burattini et al., 2008; Liyanage et al., 2017; Nandiyanto et al., 2019) of bacterial cell wall in the interaction of ZnO NPs onto the bacterial cell surface were determined.

Carboxylic groups on the cell wall contribute to the negative charge on the surfaces of bacteria cells. The strong electrostatic force at the negatively charged bacterial cells attracts the positively charged Zn ions to produce oxidative stress (Kim et al., 2020). Previous studies have acknowledged the existence of several functional groups such as carboxyl, hydroxyl, and amine groups on the bacterial cell surface which are predicted to bind with Zn ions (Yusof et al., 2020), and the involvement of proteins, polysaccharides, and lipids from the bacterial cell wall in the interaction of hematite-coated germanium crystal on bacteria (Ojeda et al., 2008). In this study, the hydroxyl, carboxyl, amides, methylene, and phosphate groups from the biomolecules such as proteins, carbohydrates, lipids, and phospholipids of bacterial cell wall are identified to be possibly involved in the interaction of ZnO NPs on the bacterial cell surface.

The cellular accumulation of ZnO NPs on the treated bacteria was analyzed by EDX. The results demonstrated a higher accumulation of Zn in Gram-positive bacterium E. faecalis (Figure 8) compared to Gram-negative bacterium S. marcescens (Figure 9). The cellular accumulation of Zn corresponds to the bacterial growth inhibition reported in the present study. The higher percentage of growth inhibition reported on E. faecalis may be due to the higher accumulation of ZnO NPs in E. faecalis.

A study by Manda et al. (2016) reported a higher accumulation of silver NPs in E. faecalis compared to Gram-negative bacterium P. vulgaris. The less negative zeta potential charge on the cell wall of Gram-positive bacteria was reported to be responsible for the higher accumulation of silver NPs in Gram-positive bacteria compared to Gram-negative bacteria, which have more negative charge in the cell wall (Arakha et al., 2015; Manda et al. (2016). Similarly, ZnO NPs have negative zeta potential charge (Berg et al., 2009; El-Waseif, 2019), hence the occurence of less electrostatic repulsion between NPs and Gram-positive bacteria can cause higher uptake of NPs and subsequently higher growth inhibition in Gram-positive bacteria.

The SEM image from the negative control of S. marcescens showed it as a rod-shaped smooth cell with intact cell wall. On the contrary, S. marcescens treated with ZnO NPs (160 μg/ml) for 24 h displayed the attachment of NPs on bacterial surface, rupture of cell membrane, and distortion of bacterial cells (Figure 10). Likewise, the negative control of E. faecalis showed smooth spherical cells with complete membrane, while E. faecalis treated with ZnO NPs showed attachment of NPs on bacteria, corrugations of cell membrane, and cell distortion (Figure 11).

The SEM images confirm the attachment of ZnO NPs on the bacterial cell surface and the subsequent disturbance to the integrity of bacterial cell membrane, which resulted in cell wrinkling and distortion. Similar findings were reported in the treatment of ZnO NPs, which caused cell membrane damage in E. coli (Arakha et al., 2015), cell shrinkage in E. coli and S. enterica (Bakur et al., 2019), membrane distortion, membrane blebbing, and leakage of cellular contents of E. coli and P. mirabilis (Maruthupandy et al., 2018).

The underlying reason that contributed to the cell rupture might be the production of reactive oxygen species due to the interaction of ZnO NPs on the bacterial cell surface, which can cause the loss of bacterial cell wall integrity by oxidizing the glutathione of bacterial cells to suppress their immune defense towards oxygen radicals (Siddiqi et al., 2018; Djearamane et al., 2020; Sánchez-López et al., 2020). According to Kairyte et al. (2013), treatment of ZnO NPs on L. monocytogenes resulted in bleb formation, which subsequently caused shrinkage of bacterial cells and intracellular component leakage, and eventually the death of bacterial cells. Further, our earlier studies have reported that the disruption in the cell wall integrity by the treatment of ZnO NPs caused cell death in S. Pyogenes and P. aeruginosa, respectively (Liang et al., 2020, and Dhanasekran et al., 2022).

Researchers have proposed various feasible mechanisms for the antibacterial properties of the metal and metal oxide NPs as illustrated in Figure 12 (Shaikh et al., 2019). Firstly, an increase in the production of reactive oxygen species by the interaction of NPs on bacteria can cause suppression of enzyme activity and ribosome disassembly, and lead to inactivation of protein synthesis, structural modification of essential proteins, and DNA damage, which eventually can result in the destruction of cellular components and cell death. Secondly, the disturbance in the cell membrane integrity due to the attachment of NPs on bacteria can increase the internalization of NPs and cause subsequent celluar damage (Shaikh et al., 2019; Lallo da Silva et al., 2019; Jiang et al., 2020). Finally, the dissolution of Zn ions from ZnO NPs can decrease the aminoacid metabolism and disturb the enzyme system, and cause loss in cell viability (Li et al., 2011; da Silva et al., 2019). Hence, the destruction of the bacteria cell wall observed from the present study might be due to the interaction of ZnO NPs on the bacterial surface and the subsequent induction of oxidative stress which resulted in the death of bacterial cells.