In conventional quantitative magnetization transfer (qMT) studies, the MTC pool is usually assumed to be a single pool with slow exchange rate (1-2). However, a diversity of proteins, lipids and metabolites are present in tissue. Assuming fast-exchanging protons to be a part of the overall slow-exchanging MTC pool not only ignores existing information about the molecular constituents of the semisolid matrix, but also induces errors when estimating the pool size and exchange parameters for qMT. Here, we propose an on-resonance variable delay multi-pulse (onVDMP) approach (3) to separate slowly exchanging MTC and total fast-exchanging protons (FTP, from metabolites + some MTC from fast exchanging protons) by fixing the pulse number and varying the mixing time, which is used as an exchange rate filter to separate these two pools based to their unique characteristic buildup patterns as a function of mixing time.

The basis of onVDMP is a label-transfer module (LTM) of a simple binomial pulse (pp) followed by a mixing time (t_mix) A train of LTMs was applied at the water resonance (Fig. 1A) and t_mix was varied to separate and quantify MTC and FTP contributions. A binomial pulse can achieve high label efficiency for both MTC (4) and fast-exchanging protons. The onVDMP curve as a function of t_mix is sensitive to the exchange rate of slow-exchanging protons, but almost identical for exchange rates above 1 kHz (Fig. 1B). In tissue, the slow-exchanging pool at high B₁ mainly originates from the conventional MTC. We describe the slowly transferring pool (slowMTC) with a fraction x_slowMTC and exchange rate k_slowMTC. The pool of total fast-exchanging protons has concentration x_TFP . Since the saturation efficiencies and concentrations cannot be separated out for the TFP pool, an αx_TFP map will be calculated instead. The onVDMP curves were fitted with three variable parameters (x_slowMTC, k_slowMTC, αx_TFP) using a three-pool model (slowMTC, TFP and water pool).

Glucose (Glc, 200 mM), cross-linked BSA (10%w/v) and hair conditioner (Suave) were selected to represent metabolites and MTC pools arise from proteins and lipids found in tissues. All MRI experiments were performed on a horizontal 11.7 T Bruker Biospec system (Bruker, Germany). Images were acquired using a RARE sequence with TR/TE= 8 s/ 4 ms, RARE factor= 8, slice thickness =1 mm, a matrix size of 64 × 64. 32 binomial pulses (B1=93.6 μT; 2 ms pulse width) with ten mixing times (0 to 150 ms ) were used for slowMTC and TFP quantification, respectively.

Figure 2A illustrates a simulation of the observed CEST/MT signal when taking DS (water Direct Saturation), slowMTC and TFP contributions into account. At zero mixing time, the observed CEST/MTC signals originate mainly from the DS and TFP pools. The contribution from the slowMTC pool increases with longer t_mix, while that of DS and FTP signals decreases with T_1w. The time for the MT signal of slowMTC pool to reach maximum is determined by the balance between k_slowMTC and T_1w. This observation was verified by the phantom studies shown in Fig. 2B. The fraction x_slowMTC and the rate k_slowMTC were extracted by fitting the onVDMP buildup curves, and were found to be x_slowMTC=10.7% / k_slowMTC=50.6 Hz (cross-linked BSA) and x_slowMTC=7.7% / k_slowMTC=24.8 Hz (hair conditioner), respectively. αx_TFP was determined for the Glc phantom. Due to a significant amount of fast-exchanging protons in cross-linked BSA, the CEST/MTC signal at zero mixing time is already around 60 %, while only a small amount of fast-exchanging protons was found in hair conditioner.

The parametric maps of x_slowMTC, k_slowMTC and αx_TFP maps in healthy mouse brain by fitting a three-pool model to onVDMP data acquired using 32 binomial pulses are shown in Figs. 3A-C, respectively. The x_slowMTC and k_slowMTC maps show similar contrast as conventional qMT. The mixing time dependencies of signals from the cortex (cx) and thalamus (th) (Fig. 3D) resemble the one obtained from cross-linked BSA. The αx_TFP map is presented in Fig. 3C, showing uniform signal across the brain with an average value of 3.0%.

The onVDMP technique proposed provides a simple and comprehensive way of separately mapping slow and fast-exchanging protons in tissues. Besides the quantification of the exchange rate together with macromolecule fraction, this technique can provide further information about the metabolites in tissues.