Introduction

The pH value is an important marker for many disease processes and pathologies, including cancer and stroke. These pathologies result in changes in the intracellular pH (pH_i) of the tissue. Thus, a non-invasive spatially and temporally highly resolved measurement of pH_i can significantly contribute to the differentiation of healthy and diseased tissue. A conventional way of determining pH_i in magnetic resonance spectroscopy (MRS) is measuring the chemical shift difference between a pH-dependent resonance and a reference ¹.Prior knowledge on the chemical shifts and the J-coupling constants of metabolites is also of central importance when quantifying spectra based on established quantification programs that use a model function for each metabolite to minimize the number of variables during the fitting procedure ^2-5. However, the accuracy of prior knowledge on the chemical shifts is essential for avoiding systematic quantification errors ⁶. In the course of a systematic investigation of the pH and temperature dependence of chemical shifts and coupling constants of important metabolites of brain metabolism, we have performed in vitro measurements on phantoms in a broad temperature and pH range. In particular, we have studied phantoms containing taurine (Tau) solutions with different pH values.

Methods

(1) Measurements were performed on a wide-bore 400 MHz NMR spectrometer (Avance III HD electronics, Bruker Biospin, Germany) using a TXI NMR probe with a z-gradient. ¹H spectra were acquired using a single-pulse sequence. The following sequence parameters were used: acquisition time 6.39 s, relaxation delay 15 s, spectral width SW = 8012 Hz, dummy scans 2 and number of accumulations NA = 16. Sample temperatures were adjusted to 40°C- 1°C.
(2) A feasibility study was performed on a 9.4 T animal scanner (BioSpec 94/30 USR, AVANCE III, Bruker BioSpin, Ettlingen, Germany) equipped with a BGA‐12S HP B₀ gradient system. RF excitations and signal detection were achieved with a quadrature birdcage coil with an inner diameter of 86 mm. Localized ¹H spectra were acquired using a point resolved spectroscopy sequence (PRESS) with the following sequence parameters: echo time TE = 16.5 ms, repetition time TR = 2 s, NA = 128, SW = 4006 Hz, voxel size 8 × 8 × 8 mm³, acquisition time 4.16 min, and eddy current compensation using the unsuppressed water signal. Sample temperatures were adjusted to 37°C.
For the sample solutions taurine (10 mM) was dissolved in phosphate buffered saline (PBS). Additionally, (1) 2,2-dimethyl-2-silapentane-5-sulfonate (DSS) (5 mM) or (2) N-acetylaspartate (NAA) (10 mM) were added as chemical shift references and the solutions were titrated to six pH values between 5.5-8.0.
Spectrum processing and determination of chemical shifts and coupling constants were performed using the software MestReNova 14.2.0 (Mestrelab Research S. L.). Data processing consisted of zero filling to 2048 K complex data points, Fourier transformation, and an automatic phase and baseline correction.

Results and Discussion

Figure 1 shows spectra with the resonances of Tau referenced to DSS (0 ppm) for 40°C as a function of pH. Both triplets show a downfield shift with more acidic pH. No changes in multiplet structure can be observed. The chemical shifts show a bijective dependence on pH in the physiological range. Additionally, it is important to note that the signals of the two proton groups shift differently (Figure 2, 3). The data recorded on the 400 MHz spectrometer has been supported by MRS data measured on the 9.4 T tomograph (Figure 4). Furthermore, the reproducibility is also shown (n=3), (line width of NAA 3 Hz). Assuming a linear regression in the physiologically interesting pH range of 7.1-8.0, a pH change of one-tenth would correspond to a chemical shift of 0.002 ppm (0.8 Hz, 9.4 T). The extent to which the accuracies can be achieved under in vivo conditions is under investigation.

Conclusion

Both triplets of taurine show a rather large pH dependence, especially in the physiologically interesting range. Thus, in combination with MRS or MR spectroscopic imaging, this dependence provides a potential approach to determine pH_i values absolutely and without the necessity of referencing to exogenous metabolites. Furthermore, the different pH dependence of the taurine triplets should be considered to avoid quantification errors if model functions of metabolite signals are used and pH may be reduced or increased.

Acknowledgements

Deutsche Forschungsgemeinschaft, Grant/ Award Numbers: DR298/13-1, DR298/13-2, BO2467/4-1, BO2467/4-2, WE6575/1-1