We report on an optimized protocol for the digestion of cellular RNA, which minimally affects the cell membrane integrity, maintaining substantially unaltered the vibrational contributions of the other cellular macromolecules. The design of this protocol allowed us to collect the first Fourier transform infrared (FTIR) spectra of intact hydrated B16 mouse melanoma cells deprived of RNA and to highlight the in-cell diagnostic spectral features of it. Complementing the cellular results with the FTIR analysis of extracted RNA, ds-DNA, ss-cDNA and isolated nuclei, we verified that the spectral component centered at ∼1220 cm is a good qualitative and semiquantitative marker of cellular DNA, since it is minimally affected by cellular RNA removal. Conversely, the band centered at ∼1240 cm, conventionally attributed to RNA, is only a qualitative marker of it, since its intensity is majorly influenced by other macromolecules containing diverse phosphate groups, such as phospholipids and phosphorylated proteins. On the other hand, we proved that the spectral contribution centered at ∼1120 cm is the most reliable indicator of variations in cellular RNA levels, that better correlates with cellular metabolic activity. The achievement of these results have been made possible also by the implementation of new methods for baseline correction and automated peak fitting, presented in this paper.

The quality of a sample has always to be considered with respect to its purposes and to the technique used for assessing it. The aim of this work is to point out the most common interfering agents that vibrational spectroscopists may find when analyzing nucleic acids extracted from cellular cultures. Fourier transformed infrared and ultraviolet‐resonant Raman measurements have been carried out on DNA and RNA samples extracted from B16 cell cultures by standard protocols used in molecular biology. The routinely adopted quality control procedure, based on ultraviolet spectrophotometry absorption measurements, could only indicate the absence of protein, phenol, or other contaminants that absorb strongly at or near 280 nm. However, DNA and RNA samples of good quality from a biological perspective clearly showed the presence of chemicals interfering with the vibrational measurements, mainly ethanol and guanidinium salts. Here, we propose fast and inexpensive strategies for tailoring the extraction protocols in light of the requirements of both vibrational spectroscopies that allowed obtaining nucleic acid samples with ethanol concentration below the detection limit of both techniques and with a significantly reduced spectral interference from guanidinium salts.

Cytosine plays a preeminent role in DNA methylation, an epigenetic mechanism that regulates gene expression, the misregulation of which can lead to severe diseases. Several methods are nowadays employed for assessing the global DNA methylation levels, but none of them combines simplicity, high sensitivity, and low operating costs to be translated into clinical applications. Ultraviolet (UV) resonant Raman measurements at excitation wavelengths of 272 nm, 260 nm, 250 nm, and 228 nm have been carried out on isolated deoxynucleoside triphosphates (dNTPs), on a dNTP mixture as well as on genomic DNA (gDNA) samples, commercial from salmon sperm and non-commercial from B16 murine melanoma cell line. The 228 nm excitation wavelength was identified as the most suitable energy for enhancing cytosine signals over the other DNA bases. The UV Raman measurements performed at this excitation wavelength on hyper-methylated and hypo-methylated DNA from Jurkat leukemic T-cell line have revealed significant spectral differences with respect to gDNA isolated from salmon sperm and mouse melanoma B16 cells. This demonstrates how the proper choice of the excitation wavelength, combined with optimized extraction protocols, makes UV Raman spectroscopy a suitable technique for highlighting the chemical modifications undergone by cytosine nucleotides in gDNA upon hyper-and hypo-methylation events.