Abstract
Importance of the redox status of nicotinamide adenine dinucleotide (NAD), including its oxidized (NAD(+)) and reduced (NADH) forms, has been shown in many biological processes. However, NAD(H) redox status assessment is traditionally limited to biochemical assays in vitro or optical redox imaging (ORI) for superficial tissues in vivo and for deep tissues ex vivo. In recent years, phosphorous-31 magnetic resonance spectroscopy ((31)P-MRS) was utilized to quantify NAD(+), NADH, and the redox ratio NAD(+)/NADH in normal tissues in vivo. The quantification is based on the spectral fitting of the upfield shoulder of the αATP peak that contains signals of NAD(+) (a quartet) and NADH (a singlet), assuming pH-independence of peak positions. To evaluate the feasibility of measuring tumour NAD(H) redox status in vivo, we fitted single voxel (31)P-MR spectra of subcutaneous mouse xenografts of human breast cancer cell lines acquired on a 9.4-T horizontal bore preclinical MR scanner. We found larger variations in the chemical shift offsets of NAD(+) and NADH from αATP in these tumours than the literature values of normal tissues. Furthermore, our (31)P-MR spectra of αATP, NAD(+)(,) and NADH solution phantoms indicated that the chemical shift of αATP and thus the offsets between NAD(H) and αATP were pH dependent. Therefore, whether tumour pH should be incorporated into the spectral fitting model should be further evaluated. Additionally, spectral resolution and signal-to-noise ratio should be improved by optimising (31)P-MRS protocols, increasing data acquisition time, and using a more sensitive coil for signal detection.