Probing Cytotoxic and Oxidative Stress Effects of Nanoplastics on Human Intestinal Caco-2 Cells: Insights from Raman Spectroscopy and Machine Learning

Nanoplastics and microplastics, typically defined as plastic particles of sizes <1µm and 1µm-1mm respectively. Both are a growing ecological and health concern due to their widespread prevalence. The effects of these in biological systems primarily relate to their potential to produce reactive oxygen species, which lead to major cellular damage and eventually, death [1, 2]. Of particular interest in this work is the effect on mammalian intestinal cells, due to the regular unintentional ingestion of nanoplastics [3]. In this study, one particular strain of human intestinal cells, Caco-2, was cultured and exposed to polystyrene particles in vitro as a simple model of the human digestive endothelium. To further gain insights into the molecular changes induced by the nanoplastic, Raman spectroscopy, as an emerging non-invasive cell-based analytical tool, was employed to detect the vibrational spectra of biochemical compositions of the cells. These spectral datasets were then subjected to machine learning (ML) analysis including clustering visualization to extract the treatment-dependent spectral features. By combining Raman spectroscopy with ML analysis, we aimed to identify distinct molecular signatures and patterns associated with the different treatment groups [4]. Cells were cultured in Eagle's Minimum Essential Medium with 10% v/v Fetal Bovine Serum and 1% v/v penicillin/streptomycin, followed by treatment with plain polystyrene nanoparticles at the concentration range of 50 µg/mL to 500 µg/mL for 24 hours. To visualize nanoplastics entering the cells 100nm Yellow-Green fluorescent nanospheres were added to the cells and the result is shown in Fig. 1. Fig.2 shows ROS and viability changes assessed by the flow cytometry. As shown in Fig. 2, the viability of cells is influenced by nanoparticles. Raman spectra were obtained by using a Renishaw inVia Raman spectrometer (Fig. 3). Spectrum data were processed via Python scripts which employ t-SNE to categorize the data to identify related components within a dataset [5]. The preliminary Raman measurements of Caco-2 cells treated with various concentrations of nanoparticles were performed. The Raman spectroscopy was also repeated for 500 nm particles with the same concentrations. The Raman spectral analysis of Caco-2 cells exposed to nanoplastics reveals distinct biochemical shifts associated with produced oxidative stress. We observed changes in peaks associated with lipids (1445 cm⁻¹) and proteins (1654 cm⁻¹). The Raman spectra data were visualized using the clustering method as shown in Fig. 4 A and B [6]. For each particle size, different concentrations of nanoparticles formed a separate cluster, indicating their unique Raman signature. This result demonstrates the potential of Raman spectroscopy to detect the impact of different concentrations of nanoplastic [7]. 200µg/mL PS100-Yellow/Green Fluorescent label conjugated to PS100 used to visualize nanoparticle aggregation in and around cells. ROS and viability Caco-2 cells treated with plain polystyrene particles at different concentrations: 50 µg/mL, 200 µg/mL, 500 µg/mL. Averaged Raman spectra of single cell treated with plain 50 nm polystyrene particles at concentrations of 50 µg/mL, including overlay of 30 individual spectrum. Score plot of t-SNE clustering.