D-glucose can be used as a biodegradable contrast agent for cancer detection by employing hydroxyl protons as a natural label for chemical exchange saturation transfer (glucoCEST) [1, 2]. However the origins of the glucoCEST signal are not yet completely understood [3], since contributions from glucose can originate from three different compartments: the vascular space (v), extravascular extracellular space (EES) and intracellular space (c), with the latter also containing phosphorylated glucose. The purpose of this study is to identify the contributions to glucoCEST contrast from these compartments, which is a first step to establish a model for studying perfusion related properties and glucose metabolism. We inhibited cellular glucose uptake via the glucose transporter (GLUT1) to delineate the extracellular glucose contribution. We applied dynamic glucoCEST enhanced (DGE) imaging [4] to study the effect of glucose transporter inhibitor in a mouse brain tumor model.Animal preparation: Human U87-EGFRvIII glioma cells were implanted (0.5×105 cells/μl) by stereotaxic injection into the right caudate/putamen of female SCID mice. Mice in the inhibitor group received intraperitoneal injection of GLUT1 inhibitor (rapamycin, 3 mg/kg every other day for two days) prior to MRI [5]. CEST imaging: Mice were fasted overnight and anesthetized using isoflurane and positioned in an 11.7T horizontal bore Bruker Biospec scanner. DGE images were acquired at a single frequency of 1.2 ppm for the hydroxyl protons before, during and after glucose injection at a temporal resolution of 10 s for both the inhibitor (N=5) and non-inhibitor group (N=4). 0.15 mL 50% w/w glucose was given over 60 s through the tail vein. The DGE images were acquired over 20 minutes using a short-echo Rapid Acquisition with Relaxation Enhancement (RARE) sequence, TR/TE = 5.0 s/3.8 ms, RARE factor of 23, a matrix of 128 x 64 and a FOV of 1.6 cm² (All images were normalized to an image without saturation acquired prior to injection). A single slice with 0.12 x 0.25 mm³ in plane resolution and thickness of 1 mm was imaged with fat suppression. Saturation was achieved by a single continuous wave magnetization transfer (MT) pre-pulse of 3 s at B1 = 1.6 μT. Data analysis: Area-under-curve (AUC) was calculated over 300 s post bolus injection to study the uptake and retention of glucose.DGE contrast increased during glucose injection and highlighted the tumor in both the inhibitor and the non-inhibitor groups (Fig. 1). After injection, the DGE signal decreased and was comparable to that of pre-injection in the case of contralateral brain in the non-inhibitor group while the tumor contrast persisted longer in both the non-inhibitor and inhibitor group. For both the tumor and the contralateral brain region, glucose signal from the inhibitor group is greatly increased compared to the non-inhibitor group. When considering the three-compartments, i.e. v, EES and c, if one assumes the glucoCEST signal originates in large part from the intracellular compartment then the glucoCEST signal should be reduced when the transport of glucose is blocked into the cells using the inhibitor. Since the opposite is observed, we attribute the glucoCEST signal to be mainly from the v and EES with minimal contribution from the intracellular compartment. The signal difference between the two groups should be contributed mainly to the effect of tumor metabolism when transport is available. Since the glucose in the inhibitor group was not further metabolized in the cells, the observed glucoCEST signal was greater. This needs to be further confirmed with histology. Stronger contrast can also be seen in the AUC maps shown in Fig. 2. For aggressive tumors such as human U87 glioma, the observed high glucoCEST contrast in the tumor can be attributed to the acidic EES together with some contribution of increased vascular volume.The newly developed dynamic glucose-enhanced (DGE) imaging approach enables the study of uptake and retention of glucose in tumors. Comparing two groups of mice implanted with human brain tumors treated with and without glucose transporter inhibitor, a great difference in glucoCEST contrast is seen. The stronger signal observed in the inhibitor group provides evidence that glucoCEST signal originates mainly from the EES and blood vessels with minimal contribution from the intracellular compartment.