Investigators developing or applying new MRI methods for characterization of ischemic stroke.Nuclear Overhauser Enhencement (NOE) effects result from the transfer of nuclear spin polarization from one nuclear spin population to another via cross-relaxation. An NOE-mediated magnetization transfer (MT) signal at -3.5 ppm (NOE(-3.5)) has been demonstrated in brain in vivo, and has been used to detect tumor ¹ and ischemic stroke ². Most recently, we reported a new NOE-mediated MT signal at around -1.6 ppm ³, named NOE(-1.6) which is postulated to report on immobilized phosphocholine headgroups. In the present work, we evaluate the application of this novel NOE(-1.6) signal for the early detection of ischemic stroke in a rodent model, and compare its magnitude with several other MT signals including amide proton transfer (APT), amine-proton water transfer, and NOE(-3.5).

MT measurements were performed by applying 5 s continuous wave (CW) irradiation before single-shot spin-echo Echo Planar Imaging (EPI) acquisition on a 7T Varian small animal scanner. MT Z-spectra were acquired with RF offsets from -5 ppm to 5 ppm (-1500 Hz to 1500 Hz at 7 T) with steps of 0.167 ppm (50 Hz at 7T) and RF power of 1mT. Images were acquired with matrix size 64 × 64, field of view 32 × 32 mm and number of acquisitions = 1.

Ischemic strokes were induced via middle cerebral artery occlusion (MCAO) in the left hemispheres of six F344 rat brains. NOE(-1.6), APT, amine, and NOE(-3.5) were quantified by using a multiple Lorentzian fitting method. Regions of interest (ROIs) of lesion (left hemisphere) and contralateral normal tissue (right hemisphere) were manually outlined for pixel-by-pixel data fittings. The peak fitting algorithm was implemented by inverting the Z-spectra between -5 to 5 ppm and removing the remaining baseline caused by non-specific MT effects so that the points at 5 ppm were set to be 0. A non-linear optimization algorithm was applied to decompose the baseline-corrected signal into a set of overlapping components.

As shown in Fig. 1, in addition to the 4 peaks representing amide at 3.5 ppm, amine at 2 ppm, bulk water at 0 ppm, and NOE at -3.5 ppm, an additional coupled pool at around -1.6 ppm upfield from water can be found on the Z-spectrum and can be quantified by using a 5-pool Lorentzian fitting method. Fig. 2 shows the time courses of the average fitted amplitudes of these pools from ischemic stroke lesions and contralateral normal tissues. It was found that both NOE(-1.6) and APT signals from stroke lesion have significant changes after MCAO. Compared with APT, NOE(-1.6) shows stronger contrast differences between stroke and contralateral normal tissues (30.7% decrease vs 14.8% at 0.5 h after MCAO). This difference stayed consistent over time until 2 h after MCAO. In contrast, NOE(-3.5) had a small increase and amine had no significant change after MCAO. Fig. 3 shows the maps of the fitted amplitudes of these pools obtained at each time point before and after MCAO from a representative rat. The stroke lesion can be clearly found on NOE(-1.6) and APT maps.The APT contrast between an ischemic lesion and contralateral normal tissue is potentially caused by pH changes. The molecular origin of the NOE(-1.6) is still not clear. In one previous publication, an NOE effect between choline headgroups in membrane phospholipids and water protons has been reported using 2D NOESY ⁴. Thus, it is likely that the NOE-mediated MT signal at -1.6 ppm originates from phospholipid choline headgroups. The changes that occur in stroke may reflect a change in mobility of phospholipids, of membrane depolarization, or other causes that warrant further investigation.A new NOE(-1.6) signal in rat brain is reported as a new biomarker for assessment of acute ischemic stroke.