1H NMR studies distinguish the water soluble metabolomic profiles of untransformed and RAS-transformed cells

Metabolomic profiling is an increasingly important method for identifying potential biomarkers in cancer cells with a view towards improved diagnosis and treatment. Nuclear magnetic resonance (NMR) provides a potentially noninvasive means to accurately characterize differences in the metabolomic profiles of cells. In this work, we use 1H NMR to measure the metabolomic profiles of water soluble metabolites extracted from isogenic control and oncogenic HRAS-, KRAS-, and NRAS-transduced BEAS2B lung epithelial cells to determine the robustness of NMR metabolomic profiling in detecting differences between the transformed cells and their untransformed counterparts as well as differences among the RAS-transformed cells. Unique metabolomic signatures between control and RAS-transformed cell lines as well as among the three RAS isoform-transformed lines were found by applying principal component analysis to the NMR data. This study provides a proof of principle demonstration that NMR-based metabolomic profiling can robustly distinguish untransformed and RAS-transformed cells as well as cells transformed with different RAS oncogenic isoforms. Thus, our data may potentially provide new diagnostic signatures for RAS-transformed cells.

21 Abstract: 22 Metabolomic profiling is an increasingly important method for identifying potential biomarkers 23 in cancer cells with a view towards improved diagnosis and treatment. Nuclear 24 magnetic resonance (NMR) provides a potentially noninvasive means to accurately 25 characterize differences in the metabolomic profiles of cells. In this work, we use 1

. Introduction
The dried hydrophilic layer was resuspended in 400 l of deuterated PBS at pH = 7.6 that 126 was prepared as previously reported (Sambrook, Fritsch, & Maniatis, 1989). The pH of each 127 sample was adjusted to 7.6 by the addition of either dilute HCl or NaOH as needed to ensure that 128 each metabolite appeared at the same chemical shift in all samples. In each sample, 0.5 l of a 129 0.1M aqueous solution of DSS (Sigma Aldrich) was added for chemical shift referencing. After 130 vortexing, each sample was transferred into a 5mm NMR tube. 131 The 1 H NMR spectra were acquired on a 500MHz Bruker Avance spectrometer 132 (operating at 500.13 MHz for 1 H observation) equipped with a 5mm TCI 500S2 H-C/N-D-05 Z 157 that best fit the sample spectra with the effective concentration of the internal DSS standard 158 being set to mM, which was the actual DSS concentration in each sample. The = 0.1248 159   Manuscript to be reviewed 218 identified by ANOVA and post-hoc/multiple comparison testing, quantitative confidence 219 intervals for the ratio in Eq. (4) were calculated using Fieller's method for unpaired data 220 (Motulsky, 1995). In this case, the (100-)% confidence range for in Eq. 221 (4), which is denoted by , is given by (Motulsky, 1995):  The results of immunoblotting total protein lysates from the four cell types against the 233 various RAS isoforms are shown in Fig. 1 Certain key metabolites are labeled using the codes, 1-21, given in Table 1.

245
The loadings of PC1 (score of 75.1%) and PC2 (score of 16.3%) from a PCA of 246 are shown in Fig. 2(A), where some of the components of both PC1 and PC2 are 247 labeled using the codes given in Table 1. In Fig. 2(B), a score plot of PC1 vs. PC2, with the 248 corresponding 99% confidence ellipses (Hoover, 1984) drawn for convenience, shows non-  Manuscript to be reviewed 260 hoc/multiple comparison testing using the BY algorithm found that for 17 metabolites 261 was significantly different between the control cells and at least one of the RAS-transformed cell 262 types whereas the "effective" NMR cellular content for 6 metabolites significantly differed in at 263 least two of the three RAS-transformed cell lines (adjusted p-values ≤ 0.01, which are given in 264 Table S1 in Supplementary Material). It should be noted that while the ANOVA analysis   In fact, and were found to be statistically larger than [ Figure 4 and  Table 2].
308 Likewise, the cellular N-acetylcysteine, a thiolic antioxidant (Oikawa et al., 1999), to glutamate 309 content was also elevated in all RAS-transformed cells relative to control cells (Fig. 5) with 310 statistically significant differences occurring for the KRAS-and NRAS-transformed cells, where 311 was 4.11 and 5.34 times larger relative to control cells, respectively [ Table 2]. The

315
The metabolomic signatures of two cellular osmolytes, taurine and myo-inositol, also 316 showed significant differences between the RAS-transformed and control cells. The cellular 317 taurine to glutamate content and the cellular myo-inositol to glutamate content were between 50-318 66% and 32-45% smaller in all RAS-transformed cells relative to the control cells, respectively 319 [ Table 2]. As osmolytes regulate the apoptotic cell death pathway (Lang et al., 2005), the 320 functional relevance of the lower values of and observed in RAS--321 transformed cells may be related to their relative resistance to stress-induced programmed cell 322 death.

323
An unexpected result from our study was the cellular phosphocholine to glutamate levels.
324 Choline metabolism is an important component in lipid biogenesis (Glunde,Bhujwalla,& 326 82% smaller in the RAS-transformed cells relative to the control cells, and statistically 327 significant differences among the RAS-transformed cells were also observed [ Figure 5 and   Table 1. Manuscript to be reviewed   Table 1. Score plots of PC2 versus PC1 of centered data with the corresponding 99% confidence ellipses (Hoover, 1984)    Box plots of the glutamate normalized signals identified by ANOVA analysis.   1-21 are for those metabolites that exhibited a significant difference (adjusted p-values ≤ 0.01) between at least two cell types in either their "effective" NMR metabolite fraction in Eq.

Table 2(on next page)
99% confidence intervals for relative fold change in the ratio of actual cellular metabolite to glutamate content between cell types. Table 2: 99% confidence intervals (CIs) for the relative fold change in glutmate normalized signals between cell lines calculated using Fieller's method ( Motulsky, 1995 )