In the past decade, it has become possible to use the nuclear (proton, 1H) signal of the hydrogen atoms in water for noninvasive assessment of functional and physiological parameters with magnetic resonance imaging (MRI). Here we show that it is possible to produce pH-sensitive MRI contrast by exploiting the exchange between the hydrogen atoms of water and the amide hydrogen atoms of endogenous mobile cellular proteins and peptides. Although amide proton concentrations are in the millimolar range, we achieved a detection sensitivity of several percent on the water signal (molar concentration). The pH dependence of the signal was calibrated in situ, using phosphorus spectroscopy to determine pH, and proton exchange spectroscopy to measure the amide proton transfer rate. To show the potential of amide proton transfer (APT) contrast for detecting acute stroke, pH effects were noninvasively imaged in ischemic rat brain. This observation opens the possibility of using intrinsic pH contrast, as well as protein- and/or peptide-content contrast, as diagnostic tools in clinical imaging.

In this work we demonstrate that specific MR image contrast can be produced in the water signal that reflects endogenous cellular protein and peptide content in intracranial rat 9L gliosarcomas. Although the concentration of these mobile proteins and peptides is only in the millimolar range, a detection sensitivity of several percent on the water signal (molar concentration) was achieved. This was accomplished with detection sensitivity enhancement by selective radiofrequency (RF) labeling of the amide protons, and by utilizing the effective transfer of this label to water via hydrogen exchange. Brain tumors were also assessed by conventional T 1 -weighted, T 2 -weighted, and diffusion-weighted imaging. Whereas these commonly-used approaches yielded heterogeneous images, the new amide proton transfer (APT) technique showed a single well-defined region of hyperintensity that was assigned to brain tumor tissue. (6) observed the presence of a composite resonance at ϳ8.3 Ϯ 0.5 ppm in the in situ proton spectra of cancer cells and cat brain, the signal intensity of which was sensitive to pH changes. Using a rat brain water exchange (WEX) experiment, we recently verified (7) that the composite resonance at 8.3 ppm is from the amide protons of cellular proteins and peptides. We also observed the pH-dependent in situ exchange effect with bulk water protons following cardiac arrest. The exchange rates measured were in accordance with knowledge gained from in vitro protein high-resolution MRS studies. The 8.3 ppm resonance was also reported in human brain proton spectra at 4 T by Chen and Hu (8).To increase MR detection sensitivity, it would be useful to be able to detect protein and peptide signals indirectly via the water resonance, especially for imaging purposes. Because there is exchange between amide protons of intracellular mobile proteins and peptides and the water protons, this is indeed possible. In recent studies using small molecules in solution, Wolff and Balaban (9) and Ward et al. (10) proposed the use of a chemical exchange saturation transfer (CEST) enhancement scheme, and demonstrated that the process of saturation transfer from exchangeable protons to water could be used for metabolite sensitivity enhancement. Subsequently, large enhancements in sensitivity (up to 500000 (11)) were demonstrated for amide protons of cationic polymers (poly-lysine, dendrimers), which have a favorable exchange rate range (10 -300 s -1 ) for effective selective irradiation and chemical exchange transfer under physiological conditions. These results suggested that detection sensitivity enhancement through selective saturation transfer via water-exchangeable amide protons of mobile proteins and peptides in biological tissue may allow spatial assessments of amide proton content and properties via the water signal. We recently demonstrated such amide proton transfer (APT) effects in rat brain and applied APT imaging to map pH effects in rat brain during ischemia (12). In this work, we demonstrate that this new imaging modalit...