Three Noble Metal Nanoparticles Demonstrated Distinct Biological Interactions and Cell Death Pathways in Human Hepatocarcinoma Cell Lines [Response to Letter]

Dear editor

In response to the recently published letter by Danfeng Shen and Hongxing Xu,¹ which raises important points concerning our article, “Three Noble Metal Nanoparticles Demonstrated Distinct Biological Interactions and Cell Death Pathways in Human Hepatocarcinoma Cell Lines,” published in Nanotechnology, Science and Applications,² we welcome this opportunity to clarify certain methodological aspects of the study and to discuss the interpretation of our findings further.

We thank the authors of the Letter for raising the apparent inconsistency between the particle diameters determined by transmission electron microscopy (TEM) and dynamic light scattering (DLS). The observed variations between DLS and TEM measurements likely arise from the distinct physical parameters assessed by each technique. Thus, TEM and DLS measurements should not be interpreted as directly comparable, as they are obtained under fundamentally different conditions and are based on distinct statistical weighting principles. TEM reports the projected size of the dried, electron-dense inorganic core. DLS measures the apparent hydrodynamic diameter of particles diffusing in suspension, which may include the core, the hydrated surface coating, and a fraction of larger scatterers that can disproportionately influence the measured size distribution. Consequently, although a substantial discrepancy between TEM and DLS measurements may suggest the presence of particle aggregation, it does not, on its own, constitute conclusive evidence of extensive or irreversible agglomeration in the original suspension.³

We agree that the measured zeta-potential values do not indicate strong electrostatic stabilization and that the polydispersity index (PDI) of Pd NPs (0.50 ± 0.20) reflects a broad hydrodynamic size distribution. However, neither zeta potential nor DLS-derived PDI should be regarded as direct, stand-alone evidence of irreversible agglomeration. Zeta potential depends strongly on the measurement medium, including pH, ionic strength, ion composition, and adsorbed surface species, and a low absolute zeta-potential value indicates limited electrostatic stabilization but does not preclude the contribution of steric or other non-electrostatic stabilization mechanisms.^4,5 Furthermore, DLS is disproportionately sensitive to even a minor fraction of larger particles because scattering intensity scales strongly with particle size. The observed combination of relatively high PDI values and low zeta potentials should be interpreted in the context of the nanoparticle surface chemistry used in this study. Specifically, the nanoparticles were coated with polyethylene glycol (PEG), which is known to confer colloidal stability primarily through steric and hydration-mediated repulsive forces rather than through strong electrostatic interactions.⁶ Therefore, a low absolute zeta-potential value does not necessarily imply instability of PEG-coated colloids. Indeed, PEG coatings may reduce the measured zeta potential while effectively preventing close particle-particle contact and aggregation through steric stabilization mechanisms.⁷ Regarding the possible aggregation of Pd-Au NPs, we wish to clarify that the UV-Vis spectra provide an independent characterization of the Au-containing nanosystems (Figure 1D in our study).² Interestingly, no pronounced spectral broadening, substantial red shift, or marked loss of the plasmonic feature was observed. Such spectral changes are commonly regarded as indicators of strong plasmonic coupling associated with nanoparticle aggregation. Therefore, the UV-Vis results provide no evidence of extensive aggregation of the Au-containing nanoparticles at the time of measurement. Considering the above arguments, we further clarify our interpretation of these findings as follows: the Pd nanoparticles possess nanometre-scale inorganic cores, whereas DLS measurements indicate a broad and heterogeneous apparent hydrodynamic size distribution in suspension. Furthermore, the measured zeta-potential values suggest limited electrostatic stabilization under the experimental conditions employed. However, the presence of the PEG coating is expected to provide an additional steric stabilization mechanism, which is not directly reflected in zeta-potential measurements.

We appreciate the mechanistic comments regarding the inactivation of glutathione peroxidase 4 (GPX4) during iron-dependent cell death. We agree that GPX4 is a central regulator of canonical ferroptosis; however, the primary objective of our study was to compare the cytotoxic and antiproliferative effects of Au NPs, Pd NPs, and Pd-Au NPs, while simultaneously investigating the cell-death pathways associated with nanoparticle-induced oxidative stress.

Our conclusion that ferroptosis contributes to the cytotoxic effects of Pd-Au NPs is based on several complementary lines of functional evidence, including increased lipid peroxidation, depletion of intracellular glutathione (GSH), attenuation of lipid peroxidation by N-acetylcysteine (NAC), and protection against nanoparticle-induced cell death by ferrostatin-1 and liproxstatin-1. Because these chemically distinct lipophilic radical-trapping antioxidants inhibit ferroptotic lipid oxidation, their protective effects support the involvement of a lipid-peroxidation-dependent, ferroptosis-associated cell-death process.^8,9

In addition, GPX1 mRNA expression was evaluated as an exploratory indicator of alterations in glutathione-related redox homeostasis. The approximately 0.75-fold reduction observed in Huh7D-12 cells following Pd-Au NP exposure (Figure 4D in our study)² was not interpreted as evidence of GPX4 inhibition, nor as proof of a specific GPX1-mediated mechanism or ferroptosis pathway. Rather, this observation was considered together with GSH depletion and lipid peroxidation as part of a broader pattern consistent with disrupted cellular redox balance. Furthermore, NAC reduced lipid peroxidation, consistent with the contribution of reactive oxygen species (ROS)-dependent processes. Importantly, these data were not used to infer the involvement of a specific GPX-dependent molecular mechanism.

Accordingly, the recently described GPX1-OSBPL8 axis cannot be evaluated on the basis of our dataset, because OSBPL8, GPX1 protein abundance and intracellular localization, and oxidized phosphatidic acid species were not assessed. We further note that the GPX1-OSBPL8 pathway was reported after the completion of both the experimental work and the peer-review process for our study and therefore was not included in the original, pre-specified mechanistic framework.^10,11

Taken together, our findings support the presence of a ferroptosis-associated phenotype based on complementary evidence derived from oxidative stress, lipid peroxidation, and pharmacological intervention studies. However, we do not claim that our data distinguish canonical GPX4-dependent ferroptosis from alternative pathways that may also contribute to lipid-peroxidation-driven cell death.

We hope that these clarifications help to place our findings in the appropriate context and provide a clearer understanding of the scope and interpretation of our results.

Disclosure

The author(s) report no conflicts of interest in this communication.