Measuring muscle lipids with 1H-MR spectroscopy

Abstract

Fat is a key element of normal skeletal muscle. It is stored inside muscle cells as droplets called intramyocellular lipids (IMCL), or in adipocytes scattered between muscle cells, called extramyocellular lipids (EMCL). Since the 1980s, IMCL droplets have been of particular interest because they were noted to be in contact with mitochondria when examined under electron microscopy. This observation led to the concept that IMCL served as nearby fuel during prolonged exercise [1]. The focus on IMCL, however, intensified after a link with the etiology of diabetes was uncovered. In 1997, 38 Pima Indians—a group of Native Americans living in central and southern Arizona—participated in a study on diabetes [2]. Pima Indians have a marked predisposition to developing Type 2 diabetes and are less sensitive to insulin action. Such a condition, called insulin resistance, generally precedes the clinical development of Type 2 diabetes by 10 to 20 years [3]. Biopsies from the vastus lateralis muscle of these participants showed increased amounts of lipids that strongly correlated with their insulin resistance. This finding pioneered further research that established a connection between IMCL accumulation and insulin resistance. Recent studies indicate that IMCL may act indirectly: increased sub-products of IMCL, such as fatty acyl-CoA and diacylglycerol, hamper signals from insulin that promote transport of glucose into the cell [4]. Interestingly, increased IMCL is not always a herald of disease. Elite athletes show high levels of IMCL and are very insulin-sensitive. This paradox is explained by the fact that muscles of highly trained persons have enhanced fatburning (oxidative) capacity, being very efficient at lipid utilization [4]. Sedentary and obese individuals, on the other hand, have decreased muscle oxidative capacity in which IMCL accumulation may be harmful. Such observations relied on several techniques to measure muscle lipids. Biochemical analysis, histochemistry, and electron microscopy represent methods that directly assess lipid content. Yet they require biopsy, which limits multiple measures of the same area and application in children. On the other hand, non-invasive methods such as computerized tomography and magnetic resonance (MR) imaging have been optimized for muscle lipid quantification. However, only proton magnetic resonance spectroscopy (1H-MRS) succeeded in discriminating the IMCL and EMCL compartments. An important principle behind 1H-MRS is that protons resonate at different frequencies depending on the molecules they are linked to. For example, methylene protons in lipids compared with methyl protons in creatine resonate differently by about 1.5 ppm (ppm is a measure of frequency independent of magnetic field strength). MR spectroscopy data is displayed as a graphic where frequency is along the x-axis, allowing identification of metabolites. Peak intensity is plotted on the y-axis. The area under a peak corresponds to its concentration, which is measured by software that compares one peak with another (relative quantitation), or compares peaks with in vitro data (absolute quantitation). Several important metabolites can be identified in the proton spectrum of muscle, including: IMCL (0.9 and 1.3 ppm), EMCL (1.1 and 1.5 ppm), creatine (2.8, 3.0, and 3.2 ppm), trimethylamines (3.2 ppm), and an assumed taurine signal (3.5 ppm; Fig. 1) [5]. A signal of acetylcarnitine can be seen at 2.13 ppm, immediately after strenuous exercise. These data can be gathered by most clinical MR scanners, using pulse sequences that measure metabolites from very specific areas while ignoring the surrounding tissues. The most popular pulse sequence for IMCL measurement is single-voxel short echo time point-resolved spectroscopy (PRESS). Using this technique, Schick et al. [6] and later Boesch et al. [5] demonstrated that IMCL and EMCL showed slightly different frequencies due to contrasting spatial arrangements in muscle. While IMCL droplets are roughly spherical, EMCL are within sheets of adipocytes, causing each compartment to experience the magnetic field differentSkeletal Radiol (2007) 36:607–608 DOI 10.1007/s00256-006-0252-8