3.2 Validation of the optimal quenching - extraction procedure on 3D MTSs
3D tumor spheroids are important model systems due to the capability of capturing in vivo tumor complexity. To our knowledge, few relevant studies have ever reported comprehensive evaluation of intracellular metabolite extraction of 3D MTSs. To date, only a few metabolomics studies have been dealt with 3D MTS models [60, 61]. Despite showing strong potential for combining 3D MTSs and metabolomics study, a detailed protocol for obtaining absolute metabolome was far from complete. For example, Rusz et al. used the HCT116 3D MTS model for the experimental design of a metabolomics workflow. The established protocol consisted of a quick wash of the spheroids on the plate followed by extraction with cold methanol [51]. However, it was uncertain whether this method was suitable for the extraction of intracellular metabolites from 3D MTS in a leakage-free manner. In addition, this protocol has not been quantitatively validated and also has not been applied to 2D adherent cells for broad applicability. In this study, sample preparation protocols developed with 2D monolayer cultures were extended to acquire single spheroid metabolomics. We aimed to investigate whether the best sample preparation protocol for 2D cells was also applicable to 3D MTSs. The preparation method of tumor spheroids followed the steps as described in Material and methods 2.2 . As the incubation time increased, the spheroid became rounder and bigger (Figure S2A ). On the 6th day, the MTS reached a plateau with a maximum diameter of 538.26 ± 7.19 μm, the roundness of the MTSs exceeded 0.9, and the number of viable cells of a single MTS reached the maximum, about 34000±1000 (Figure S2B ). The morphology and microstructure of the 6th day MTSs were observed with a scanning electron microscope (SEM), and it was found that the MTSs had a good 3D structure (Figure S2C ). After 6 days of culture, tumor spheroids were collected and the intracellular metabolites were obtained using the above-mentioned 12 combinations of quenching and extraction methods.
The experimental results demonstrated that in 3D MTSs, the leakage of intracellular metabolite with methanol as the quencher was more serious than that of normal saline, which was the same as that found in 2D cells (Figure 9 ). Also, the same conclusion was drawn with 3D tumor spheroid metabolomics analysis that the degree of metabolite leakage was associated with molecular weight (Figure S3 ). For the 12 quenching-extraction methods, N2-A method was also the best method applied to 3D MTSs in terms of the number and amount of intracellular metabolites extracted (Figure S4 ). Moreover, the percentage of intracellular metabolites with RSD less than 30% in 3D MTSs for the N2-A method can reach 48.7% (Table S2 ). However, we found the average RSD values of each metabolite from 3D MTSs were significantly higher (P<0.01) than that from 2D cells as a whole (Figure S5 ), which might be ascribed to the fact that some small cell clusters in single cell suspension (Figure S6 ) have a certain impact on the reproducibility of metabolite extraction. This might be solved by extending the enzymatic hydrolysis time to reduce cell clusters, but long-time exposure to Accutase™ cell digestive solution likely affects membrane integrity and permeability and causes the leakage of intracellular metabolites, which needed to be verified by subsequent experiments. Taken above, our study showed that the optimized sample preparation protocol (N2-A method) developed for the intracellular metabolite extraction of 2D adherent cancer cells can be well applied to 3D MTSs.