In vivo¹H MRS and³¹P MRSI of the response to cyclocreatine in transgenic mouse liver expressing creatine kinase

Abstract

Hepatocyte transplantation has been explored as a therapeutic alternative to liver transplantation, but a means to monitor the success of the procedure is lacking. Published findings support the use of in vivo (31)P MRSI of creatine kinase (CK)-expressing hepatocytes to monitor proliferation of implanted hepatocytes. Phosphocreatine tissue level depends upon creatine (Cr) input to the CK enzyme reaction, but Cr measurement by (1)H MRS suffers from low signal-to-noise ratio (SNR). We examine the possibility of using the Cr analog cyclocreatine (CCr, a substrate for CK), which is quickly phosphorylated to phosphocyclocreatine (PCCr), as a higher SNR alternative to Cr. (1)H MRS and (31)P MRSI were employed to measure the effect of incremental supplementation of CCr upon PCCr, γ-ATP, pH and Pi /ATP in the liver of transgenic mice expressing the BB isoform of CK (CKBB) in hepatocytes. Water supplementation with 0.1% CCr led to a peak total PCCr level of 17.15 ± 1.07 mmol/kg wet weight by 6 weeks, while adding 1.0% CCr led to a stable PCCr liver level of 18.12 ± 3.91 mmol/kg by the fourth day of feeding. PCCr was positively correlated with CCr, and ATP concentration and pH declined with increasing PCCr. Feeding with 1% CCr in water induced an apparent saturated level of PCCr, suggesting that CCr quantization may not be necessary for quantifying expression of CK in mice. These findings support the possibility of using (31)P MRS to noninvasively monitor hepatocyte transplant success with CK-expressing hepatocytes.

In vivo absolute quantification of hepatic γ‐ATP concentration in mice using 31P MRS at 11.7 T

“…The γ-ATP concentrations were also not different from the 2.52 mM found in the B 6 D 2 F 1 mouse model. 17 Second, this study examined non-diabetic normoglycemic NOD mice, which develop hepatic insulin resistance associated with increased inflammatory and lipid peroxidation before manifesting diabetes at a later age. 25 These non-diabetic normoglycemic NOD mice showed no differences in hepatic γ-ATP or P i , which is in line with the absence of changes in ex vivo mitochondrial oxidative capacity from various substrates.…”

“…8 Transferring quantitative 31 P MRS methods to specific mouse models would facilitate translational mechanistic studies but also help to evaluate novel targets for the prevention and treatment of diabetes mellitus and NAFLD. Absolute quantification of hepatic molar γ-ATP concentrations has been previously reported for one mouse 17 and a few rat models. [18][19][20] In the transgenic mice expressing the brain isoform of creatine kinase isoenzyme in hepatocytes, 31 P 3D MR spectroscopic imaging required an acquisition time of nearly 1 h to complete phase encoding steps and achieve sufficient signal-to-noise ratios (SNRs).…”

“…[18][19][20] In the transgenic mice expressing the brain isoform of creatine kinase isoenzyme in hepatocytes, 31 P 3D MR spectroscopic imaging required an acquisition time of nearly 1 h to complete phase encoding steps and achieve sufficient signal-to-noise ratios (SNRs). 17 Such acquisition times would not be suitable for dynamic time course experiments, which call for a faster method, for example single voxel image selected in vivo spectroscopy (ISIS). 7 Absolute quantification of 31 P MRS further needs replacement phantoms of known concentrations as an external reference, 7,9,17,20 which make correction factors necessary, such as T 1 and T 2 relaxation times depending on pulse sequence, and pulse profiles.…”

“…17 Such acquisition times would not be suitable for dynamic time course experiments, which call for a faster method, for example single voxel image selected in vivo spectroscopy (ISIS). 7 Absolute quantification of 31 P MRS further needs replacement phantoms of known concentrations as an external reference, 7,9,17,20 which make correction factors necessary, such as T 1 and T 2 relaxation times depending on pulse sequence, and pulse profiles. 7 Moreover, elevated hepatocellular lipid (HCL) content, typical of metabolic diseases, 21 will effectively reduce the viable liver tissue volume for a given voxel in MRS.…”

“…Finally, the reproducibility/repeatability of quantitative 31 P MRS as assessed by coefficients of variation (COVs) has been reported only for a few human studies, 7,9,22 and not for the previous rodent studies. [17][18][19][20] Thus, this study aimed to establish an in vivo 31 P MRS method to quantify molar concentrations of hepatic γ-ATP in mice at 11.7 T with a short acquisition time, to assess the necessary correction factors (HCL volume, T 1 and T 2 relaxation times), to evaluate its repeatability (COVs) and to apply this method in mouse models with different degrees of insulin resistance and NAFLD.…”