Synthesis and Characterization of AgNPs
Two different-sized AgNPs were synthesized using a chemical reduction method. The formation of AgNPs was monitored by the UV‐Vis spectrum in a range of 300-900 nm (Fig. 1a). The production of AgNPs was observed by the reaction color that changed from transparent to yellow or yellow‐brown and the characteristic localized surface plasmon resonance (LSPR) peak of AgNPs. The narrow spectral curve with the LSPR peak at 410 nm suggested the low size distribution of the small-sized AgNPs (Janthima et al. 2018). The formation of large-sized AgNPs was determined by the LSPR peak at 430 nm but with a broader spectral curve, suggesting their wider size distribution (Alsammarraie et al. 2018). TEM micrographs revealed predominantly spherical shapes of both synthesized AgNPs (Fig. 1b). The average sizes of the synthesized AgNPs were 5.6 ± 1.4 and 70.9 ± 24.1 nm, which were referred to as small (5‐nm) and large (70‐nm) AgNPs, respectively.
The synthesized AgNPs were characterized by SAED‐TEM and XRD analyses. The SAED patterns of both synthesized AgNPs were similar (Fig. 2a), which contained bright circular fringes corresponding to the (111), (200), (220), and (311) lattice planes, suggesting the crystalline nature of AgNPs according to the JCPDS No. 04-0783 (Acharya et al. 2021). The XRD analysis is shown in Fig. 2b, in which both AgNPs showed similar results of Bragg reflections with the 2-theta values at 38.1°, 44.3°, 64.4°, and 77.3°. These values corresponded to the (111), (200), (220), and (311) lattice planes of the face‐centered cubic structure (fcc) of silver according to the JCPDS No. 03‐065‐2871 (Alahmad et al. 2021). The SAED‐TEM and XRD analyses suggested that the identities of the synthesized nanomaterials were AgNPs.
Mortality Effects of Different Sized AgNPs
The 3-h stage embryos were incubated in pond water containing different concentrations (1, 10, 50, and 100 mg/L) of 5-nm and 70-nm AgNPs at 28 ºC for 24 h. Their cumulative mortalities in response to AgNPs were measured at 6, 12, 18, and 24 h. The 5‐nm AgNPs caused 100% mortality of frog embryos when incubating at the concentrations of 10, 50, and 100 mg/L for 24, 18, and 12 h, respectively. The mortality effect of 5‐nm AgNPs depended on dose and exposure time, shown by the linear regression equations (Fig. 3). The higher mortality rates were detected when embryos were exposed to a greater concentration of AgNPs or longer exposure time. The 70‐nm AgNPs also induced mortality of the frog embryos, but at a much lower rate. The highest mortality was 22.4% in the treatment of 100 mg/L AgNPs.
Although the mortality effect of AgNPs on frog embryos is unclear, the toxicity mechanism may be similar to fish embryos. Instead of a protective chorion in fish embryos, frog embryos are protected by jelly layers and porous vitelline envelopes (Sato and Tokmakov 2020). The silver nanoparticles in a size range of 1-20 nm can diffuse through lipid bilayers. However, this diffusion process was inefficient for the nanoparticles at 100 nm and larger (Osborne et al. 2015). Due to their small size, AgNPs may penetrate jelly layers and vitelline envelopes more efficiently than the larger-sized AgNPs. Inside the cells, AgNPs may interrupt normal cellular activities and cause impacts on the viability, growth, and development of frog embryos (Qiang et al. 2020). AgNPs can induce lysosomal membrane destabilization, apoptosis machinery, mitochondrial dysfunction, and reactive oxygen species (ROS). Also, the cellular oxidative stress caused by AgNPs can activate the autophagy process, mitochondrial ROS production, and phospholipase-induced membrane degradation (Quevedo et al. 2021). In general, larger-sized AgNPs have lower efficacy to penetrate through jelly layers and vitelline envelopes and accumulate there. These AgNPs can block the vitelline envelope pores, disrupt membrane transport, and destroy the vitelline envelope. Therefore, embryonic development is disruptive, resulting in embryonic malformation (Wu and Zhou 2012). Also, these accumulated AgNPs around the embryos can damage embryos via a release of Ag+, which can feasibly penetrate the vitelline envelope (Quevedo et al. 2021). The acute toxic mechanism of Ag+ in freshwater fish is well established, in which silver ions can induce a disruption of the Na+/K+-ATPase (NKA) ion channel via the inhibition of Mg2+ binding. As a result, Na+ uptake is reduced, thus causing juvenile death if the concentration of Na+ is decreased by 30% (Boyle and Goss 2018).
Growth Effects of Different‐sized AgNPs
In this work, the embryonic developmental stages of Hoplobatrachus rugulosus were studied in a time course of 3 – 28 h after fertilization to hatched tadpoles. Their development stages were classified according to the Gosner method (1960). Fig. 4 is the photographs of embryonic development of Hoplobatrachus rugulosus, including late cleavage/blastula stage (GS 9; 3 h), early gastrula stage (GS 10; 5 h), mid gastrula stage (GS 11; 7 h), late gastrula stage (GS 12; 9 h), neural fold stage (GS 14; 11 h), elongation and rotation stage (GS 15; 12 h), neural tube stage (GS 16; 13 h), tailbud stage (GS 17; 14 h), muscular response stage (GS 18; 15 h), gill buds and heartbeat stage (GS 19; 16 h), gill circulation and tail elongation stage (GS 20; 19 h), cornea transparent and mouth opening stage (GS 21; 24 h), and tail fins transparent and fin circulation (GS 22; 28 h). These results suggested that the metamorphosis period of Hoplobatrachus rugulosus was faster than Hydrophylax leptoglossa, Microhyla ornate, and Xenopus laevis (Hill et al. 2017, Traijitt et al. 2021). These developmental stages were used to identify the frog development in this work.
The frog embryos were treated with different concentrations (1, 10, 50, and 100 mg/L) of 5-nm and 70-nm AgNPs in a time course of 24 h to study their growth and developmental responses. The growth effect was determined by total length, which measured the embryo length (6 and 12 h) or their body length from snout to tail tip (18 and 24 h) (Phung et al. 2020). Fig. 5a shows the effects of 5‐nm and 70‐nm AgNPs (50 mg/L) on the body length during embryonic development in a time course of 24 h. The results suggested that 5-nm AgNPs had a more negative effect on the body length than 70-nm AgNPs. At 24 h, the average body lengths of the developed embryos treated with 5-nm and 70-nm AgNPs were reduced for 2.6 and 1.8 folds, respectively. Fig. 5b shows the effects of 5‐nm and 70‐nm AgNPs at different concentrations on the total length of the developed embryos. The results suggested the impact of AgNPs on a significant reduction of the body length of the frog embryos in a dose-respondent manner. Both AgNPs significantly caused the reduction of body length as compared with the control embryos, but more negative effect by 5-nm AgNPs. Also, malformations of frog embryos treated with AgNPs were observed, including scoliosis (abnormal sideways curvature of the notochord), lordosis (bent tail caused by inward curving of the notochord), kyphosis (exaggerated curvature of the upper notochord), and yolk sac edema (Fig. 5c).
These results revealed that 5‐nm AgNPs caused more negative effects on the growth of the frog embryos. The growth reduction of the embryos treaed with AgNPs could possibly cause by the interaction of AgNPs or the release of Ag+ with DNA, resulting in disruption of DNA replication, destruction of DNA, and induction of cell apoptosis (Wang et al. 2020). Also, they could affect signaling cascades (such as mitogen‐activated protein kinases) that regulate cell proliferation and development, resulting in growth retardation (Zhang et al. 2021). The effects of AgNPs on growth retardation and developmental abnormality might be similar to those in fish embryos via the induction of membrane-pore blockage and reduction of the oxygen exchange rate (Qiang et al. 2020). Also, the delayed growth of frog embryos might result from the toxicity of the releases of Ag+ via a down-regulation of some genes relating to metabolic rates, ATP synthesis, and ribosome biogenesis (van Aerle et al. 2013). The detected malformation included scoliosis, lordosis, kyphosis, and yolk sac edema. A group of Bmp genes played crucial roles in the early development stage of vertebrate embryos (Suzuki et al. 2017). The interference of these gene expressions might be a cause of abnormal development in AgNP‐treated embryos. Although the growth retardation and malformation of frog embryos treated with 70‐nm AgNPs were lower than 5‐nm AgNPs, this effect might have an indirect impact on the frog survival due to the disadvantages of food and mating competition.
Oxidative Stress in Response to AgNPs
The lipid peroxidation was determined in the frog embryos treated with 5-nm and 70-nm AgNPs at 1, 10, 50, and 100 mg/L for 18 h to reveal the oxidative damage effect of AgNPs on the cellular membrane. The lipid peroxidation was expressed as nmol MDA per mg protein. Fig. 6 shows the MDA levels of the frog embryos treated with 5-nm and 70-nm AgNPs as compared with that of the control embryos. The results revealed that 5‐nm AgNPs caused the significantly increased level of MDA in dose-response, and higher than those in the embryos treated by 70-nm AgNPs. At 100 mg/L, for example, the MDA level of the embryos treated with 5‐nm AgNPs was approximately 2.7 folds greater than those treated with 70‐nm AgNPs. These results also suggested that 5-nm AgNPs caused more toxicity to the frog embryos than 70-nm AgNPs. It was likely that 5-nm AgNPs might penetrate into the cells more efficiently, thus causing greater cellular oxidative stress as seen by the significantly increased level of MDA (the end product of lipid peroxidation) (Mas-Bargues et al. 2021). One possible action of AgNPs to induce oxidative stress is proposed via the ROS‐induced peroxidation of the oxidized lipid cell membrane, which disturbs the membrane fluidity and disrupts cell membrane integrity (Yao et al. 2019).
Synchrotron FTIR to Analyze Cellular Molecules
The molecular changes of frog embryos treated with 5‐nm and 70‐nm AgNPs for 18 h were analyzed by synchrotron FTIR. The spectra data analyses of the head and body regions of the frog embryos were also compared. Fig. 7 illustrates the synchrotron‐FTIR spectral analysis of 5‐nm AgNP‐treated, 70‐nm AgNP‐treated, and control embryos in head and body regions. The absorption spectra in the wavenumber 3000-1000 cm−1 revealed the changed biochemical profiles of lipids, proteins, carbohydrates, and nucleic acids, resulting in distinct separation of the 5‐nm AgNP‐treated, 70‐nm AgNP‐treated, and control embryos (Fig. 7a). These results also supported that 5‐nm AgNPs exhibited a higher effect than 70-nm AgNPs at a cellular level. FTIR spectral data of both head and body regions were quite similar. From the FTIR spectra, the lipid profiles were indicated by stretching vibrations of asymmetry CH3 of lipids (2970–2916 cm−1), symmetric stretching vibration of CH2 of acyl chains (lipids) (2855–2850 cm−1) (Talari et al. 2017), and ester group (C=O) vibration of triglycerides (1750–1745 cm−1) (Nandiyanto et al. 2019). The a‐helical and b‐sheeted conformations of amide I were the spectral peaks at 1657–1650 cm−1 and 1635–1625 cm−1 (Nandiyanto et al. 2019). The indication of carbohydrates was via the O–H in-plane bend of primary or secondary alcohol (1350–1260 cm−1) (Nandiyanto et al. 2019), C–O asymmetry stretch of ketone (1150–1145 cm−1), C–O–C asymmetry stretch of ether (1120–1094 cm−1), and C–O stretch of primary alcohol (1070–1050 cm−1) (Talari et al. 2017). For nucleic acid analysis, the spectral peaks at 1265–1201 cm−1 and 1099–1066 cm−1 corresponded to the asymmetric and symmetric PO2- stretching vibrations (Talari et al. 2017).
Fig. 7b shows the principal component analysis (PCA) plots. The PCA analyses demonstrated that the PC-1, PC-2, and PC-3 provided the best visualization of separated clusters of biochemical changes among three groups of the samples; 5‐nm AgNP‐treated, 70‐nm AgNP‐treated, and control embryos. The PCA score plot of the head region was sufficient to separate these three sample groups, which was explained by 37% PC-1 and 9% PC-3. Similarly, the PCA score plot of the body region was separately clustered by 38% PC-1 and 13% PC-2. Fig. 7c shows the first principal component (PC‐1) loadings plots that give the main variation between these sample groups. According to the second derivative spectra (Fig. 7d), the separation of 5‐nm AgNP‐treated, 70‐nm AgNP‐treated, and control embryos dominantly attributed to lipids (2960, 2958, 2925, 2923, 2852, and 1745 cm−1). Also, the separation could assign by proteins (1656 and 1629 cm−1), carbohydrates (1344, 1147, 1114, 1060, and 1058 cm−1), and nucleic acids (1239, 1238, 1089, and 1087 cm−1). In the head region, the second derivative plot revealed the lipid reduction at a higher level in the frog embryos treated with 5‐nm AgNPs than 70‐nm AgNPs. These reduced spectral peaks (2960, 2923, 2852, and 1745 cm−1) attributed to the lipid reduction. The protein conformational changes, a‐helical (1656 cm−1) to b‐sheeted (1629 cm−1), were also detected in frog embryos treated with AgNPs of both sizes. The frog embryos treated with AgNPs revealed induced carbohydrate levels indicated by the spectral peaks at 1344, 1147, 1114, and 1060 cm−1. Similarly, 5-nm AgNPs exhibited more carbohydrate induction as compared with 70-nm AgNPs. The increase and decrease of nucleic acid levels in response to AgNP-treatment were noticed by the spectral peaks at 1238 and 1087 cm−1, respectively.
In the body region, synchrotron‐FTIR results were similar to those in the head region analysis. The lipid reduction was detected, as indicated by the spectral peaks at 2958, 2925, 2852, and 1745 cm−1. For protein analysis, the α‐helical and b‐sheet structures were present in the embryos treated with 5‐nm AgNPs as indicated by the spectral peaks at 1656 and 1629 cm−1. The carbohydrate induction in frog embryos treated with AgNPs was also detected at the spectral peaks of 1344, 1147, 1114, and 1058 cm−1. The modulation levels of nucleic acids in response to AgNP-treatment were noticed by the spectral peaks at 1239 and 1089 cm−1.
Accumulation of AgNPs in Frog Eggs and Some Organs
The localization of AgNPs in the frog digestive system and the excretory system was analyzed by atomic absorption spectroscopy analysis (Table 1). In this work, 5-nm AgNPs were orally given to the adult frogs (1,000 mg/kg) every five days for 30 days. The levels of silver in the AgNP‐treated samples were significantly higher than the control samples at all analyzed tissues. The highest silver level was detected in eggs, followed by the stomach, liver, kidney, and intestine. The detected amount of silver in eggs was as high as 195.55 ± 0.87 µg/g of wet weight tissue, making the concern about the effects of these silver on their embryonic development. These results suggested that AgNPs could absorb into the blood circulation and translocate to the other tissues and organs. The localization of AgNPs in the liver and kidney was likely due to these organ functions involving detoxification and excretion of toxic materials. As eggs could store a high level of AgNPs, it raised a concern about the potential effects of AgNPs on the reproductive system and the survival rate of the frog embryos. As a result, at this exposure dose, AgNPs might affect the sustainability of this frog species in nature.