Diffusion-Weighted Hyperpolarized 13C-Urea in a Murine Model of Liver Fibrosis
Irene Marco-Rius1, Jeremy A Gordon1, Peder EZ Larson1, Romelyn delos Santos1, Robert A Bok1, Aras Mattis2,3, Jacquelyn Maher3,4, Daniel B Vigneron1,3, and Michael A Ohliger1,3

1Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, United States, 2Department of Pathology, University of California San Francisco, San Francisco, CA, United States, 3Liver Center, University of California, San Francisco, San Francisco, CA, United States, 4Department of Medicine, Division of Gastroenterology, University of California San Francisco, San Francisco, CA, United States


Diffusion weighted MRI has been widely used to measure the movement of water molecules and study tissue microstructure in order to characterize both diffuse and focal liver disease. In liver fibrosis, for instance, increased collagen formation is associated with restricted diffusion of water. However, the majority of water within the liver is either in the vascular or intracellular space, making the diffusion of water a potentially poor marker for fibrosis, which is an extracellular process. Here, we investigated applying diffusion-weighted MRI with an exogenously injected extracellular agent, hyperpolarized 13C-urea, as a potentially more sensitive probe of the extracellular space in the liver in a mouse model of liver fibrosis induced with CCl4.

Target Audience

Researchers interested in liver fibrosis, diffusion and hyperpolarized 13C MRI.


Diffusion weighted (DW) MRI is typically used to measure the molecular movement of water in order to study tissue microstructure. In liver fibrosis, collagen accumulates in the extracellular space and restricts the motion of water molecules, decreasing the apparent diffusion coefficient (ADC)1. However, the majority of water within the liver is either intravascular or intracellular, potentially limiting the sensitivity of proton ADC for liver fibrosis, which is an interstitial, extracellular process. We investigated the progression of fibrosis using DW MRI with an exogenously injected extracellular agent, hyperpolarized 13C-urea, as a potentially more sensitive probe of the extracellular space in the liver.


Protocol was approved by the local institutional animal care and use committee. The experimental protocol is illustrated in Figure 1. Liver fibrosis was induced in three CD1 mice by administering 1 μl/g of 1:7 mixture of CCl4 and olive oil via IP injection once every 4-5 days. Images were acquired at baseline and every 2 weeks using a 14T MRI scanner (Varian, Inc) with dedicated coils for 13C and 1H imaging. Imaging was performed on the day before the next treatment dose in order to minimize acute inflammatory effects.

1H diffusion weighted spin echo images were acquired using the following parameters: 128 x 128 matrix, 32 mm FOV, slice 2 mm, diffusion time 13 ms, b=93, 239, 538, 957, 1496 s/mm2. Combined cardiac and respiratory gating was employed.

For hyperpolarized experiments, 90 mg of [15N,13C]urea dissolved in glycerol were polarized for ~90 min using a 3T HyperSense DNP polarizer (Oxford Instruments), rapidly warmed, dissolved in PBS (final concentration 110 mM), and injected via tail vein catheter. The acquisition of 13C DW images started 20 s after injection, using an echo planar imaging sequence2 with a diffusion-compensated variable flip angle sequential, slice-select gradient correction to reduce bias3 and with the following parameters: 32 x 32 matrix, 32 mm FOV, slice 8 mm, diffusion time 20 ms, b=100, 200, 350, 400, 600, 750, 1000, 1500 s/mm2. Only respiratory triggering was used. Images with obvious cardiac motion artifact were excluded. For both 1H and 13C, ADC maps were calculated and a mean value determined over a region of interest (ROI) chosen in the thicker lobe of the liver (right lobe).

Two mice were treated for 12 weeks and one mouse expired at 8 weeks. The mice treated for 12 weeks were imaged 5 days after the last CCl4 injection and then again one week later. At the end of the study, each animal was sacrificed and the liver stained for collagen using trichrome and sirius red. Slides were reviewed by an experienced pathologist.

Results and Discussion

The baseline 1H ADC of water was (0.70 ± 0.07)×10-3 mm2/s, which agreed with previously reports4,5. Urea ADC was (0.76 ± 0.15)×10-3 mm2/s, not significantly different from that of water. This was surprising given the larger molecular weight of urea. For comparison the solution-state ADC of water and urea at 37ºC are 2.9×10-3 mm2/s 6 and 1.5×10-3 mm2/s 7, respectively. The larger reduction of water ADC in tissue relative to its solution value may reflect the fact that a large fraction of water is intracellular while urea is extracellular.

For all animals, the 1H ADC initially increased and then decreased over time (Figure 3, black circles). For 2/3 animals, 13C-urea ADC also initially increased and then decreased (Figure 3, red stars). The small number of animals used in this study did not permit statistical significance to be determined.

Figure 4 summarizes the ADC results for the baseline, 5 days and 12 days after last treatment, and pathology fibrosis grade. In the last post-treatment studies, all three animals demonstrated a reduction in urea ADC compared to baseline (29% average drop). Interestingly, for the two animals that had repeat imaging 12 days after the final CCl4 injection, the ADC values increased compared to the images obtained 5 days after the final CCl4 injection (25% average increase). This effect was also surprising and may be due to either liver recovery between the two scans or variability in the ADC measurement. Final pathologic images and fibrosis grade for each animal are shown in Figure 5. The animal sacrificed after 8 weeks developed severe fibrosis while the two animals sacrificed after 12 weeks developed moderate fibrosis.


We have demonstrated the feasibility of measuring the ADC of hyperpolarized [15N,13C]urea in a mouse model of liver fibrosis. This pilot study establishes physical insights and preliminary data to determine the timing and sample size for larger studies of hyperpolarized agents in fibrosis.


We thank Daniel Tesfasilassie for technical help and Dr. Peng Cao for fruitful discussions. This work was supported by RSNA Research and Education Foundation, UCSF Liver Center grant P30DK026743, and NIH grants P41EB013598 and R01EB016741. CVM was supported by NIH K01DK099451.


1. Annet, L. et al. Assessment of diffusion-weighted MR imaging in liver fibrosis. J Magn Reson Imaging 25, 122–8 (2007).

2. Koelsch, B. L. et al. Rapid in vivo apparent diffusion coefficient mapping of hyperpolarized 13C metabolites. Magn Reson Med 74, 622–633 (2015).

3. Gordon, J. W. et al. Mis-estimation and bias of hyperpolarized ADC measurements due to slice-profile effects. in 5th DNP symposium (2015).

4. Anderson, S. W. et al. Effect of disease progression on liver apparent diffusion coefficient values in a murine model of NASH at 11.7 Tesla MRI. J Magn Reson Imaging 33, 882–888 (2011).

5. Zhou, I. Y. et al. Effect of diffusion time on liver DWI: An experimental study of normal and fibrotic livers. Magn Reson Med 72, 1389–96 (2014).

6. Holz, M., Heil, S. R. & Sacco, A. Temperature-dependent self-diffusion coefficients of water and six selected molecular liquids for calibration in accurate 1H NMR PFG measurements. Phys Chem Chem Phys 2, 4740–4742 (2000).

7. Koelsch, B. L. et al. Diffusion MR of hyperpolarized 13C molecules in solution. Analyst 138, 1011–1014 (2013).


Figure 1. Timeline of the experimental design. MR images were acquired before the beginning of the CCl4 treatment and subsequently two weeks after the first day of treatment. A last set of images were acquired 12 days after the last CCl4 treatment, on the same day that mice were sacrificed.

Figure 2. Representative 1H and 13C DWI data sets (mouse#2, 10 weeks of CCl4 treatment). (a) 1H-DW images. (b) 13C-DW images and acquisition scheme: two images were acquired at every respiratory cycle. (c, d) 1H and 13C ADC maps and fits of the data excluding images corrupted by motion.

Figure 3. Progression of the ADC values of the right lobe of the liver with treatment time (includes data from Figure 4).

Figure 4. Mean ADC values of the right lobe of the liver and their corresponding standard deviation comparing baseline, ADC after last treatment and one week after the last imaging session. Scoring based on the Batts-Ludwig staging system.

Figure 5. Pathological images of the livers of the three mice in the study (magnification 4x). Based on the Batts-Ludwig staging system, mouse #1 developed fibrosis stage 3, mouse #2 stage 2, and mouse #3 stage 2-3 (see table in Figure 4).

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)