- Email: [email protected]
In Vivo Lipid Profiling Using Proton Magnetic Resonance Spectroscopy in an Experimental Liver Fibrosis Model
In Vivo Lipid Profiling Using Proton Magnetic Resonance Spectroscopy in an Experimental Liver Fibrosis Model Jerry S. Cheung, PhD, Shu Juan Fan, MSc, Darwin S. Gao, BEng, April M. Chow, PhD, Jian Yang, MD, PhD, Kwan Man, PhD, Ed X. Wu, PhD Rationale and Objectives: The aim of this study was to characterize early hepatic lipid changes in an experimental model of liver fibrosis using proton (1H) magnetic resonance spectroscopy (MRS) at high magnetic field in vivo. Materials and Methods: Liver fibrosis was induced in 12 Sprague-Dawley rats by twice-weekly carbon tetrachloride (CCl4) administration up to 4 weeks. Eight normal rats were used as controls. Single-voxel 1H MRS experiments were performed at 7 Tesla to measure signal integrals of various lipid peaks including –CH3, (–CH2–)n, –CH2–C=C–CH2–, =C–CH2–C= and –CH=CH– at 0.9, 1.3, 2.0, 2.8, and 5.3 ppm, respectively, and peak from choline-containing compounds (CCC) at 3.2 ppm. Total lipid, total saturated fatty acid, total unsaturated fatty acid, total unsaturated bond, polyunsaturated bond, and CCC indices were quantified. Results: Significant increases (P < .01) in total lipid and total saturated fatty acid indices were found in animals with CCl4-induced fibrosis as compared with normal animals. In addition, total unsaturated bond and polyunsaturated bond indices of animals at 4 weeks after CCl4 insult were significantly higher than (P < .01 and P < .05, respectively) those of normal animals and animals at 2 weeks following insult; whereas there was only significant increase (P < .01) in total unsaturated fatty acid index in animals with 4-week CCl4 insult as compared with normal animals. Conclusion: The hepatic lipid changes in CCl4-induced experimental fibrosis model were documented in vivo and longitudinally using 1 H MRS at 7 Tesla. The experimental findings suggested that total saturated fatty acid increase contributed mainly to the total lipid increase in animals with CCl4 insult. This study also demonstrated the potential value of high field MRS to resolve lipid composition and alterations in liver fibrosis. Key Words: Proton magnetic resonance spectroscopy (1H MRS); liver fibrosis; lipid; saturated fatty acid; unsaturated fatty acid; carbon tetrachloride (CCl4). ªAUR, 2011
L
iver fibrosis associated with chronic liver injury can progress to cirrhosis and ultimately hepatocellular carcinoma (HCC) (1,2). Percutaneous liver biopsy has been considered the standard technique for diagnosis and staging of liver fibrosis. However, liver biopsy is highly invasive and associated with risk of complications, limiting its applicability in longitudinal monitoring of fibrosis Acad Radiol 2011; 18:377–383 From the Laboratory of Biomedical Imaging and Signal Processing (J.S.C., S.J.F., D.S.G., A.M.C., J.Y., E.X.W.), Departments of Electrical and Electronic Engineering (J.S.C., S.J.F., D.S.G., A.M.C., J.Y., E.X.W.), Surgery (K.M.), and Anatomy (E.X.W.), The University of Hong Kong, Pokfulam, Hong Kong SAR, China; Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital and Harvard Medical School, Charlestown, MA 02129, USA (J.S.C); Department of Diagnostic Radiology of the First Affiliated Hospital, School of Medicine of Xi’an Jiaotong University, Xi’an, Shannxi Province, China (J.Y.). Received September 12, 2010; accepted October 29, 2010. Supported by the Hong Kong Grant Council (GRF HKU7808/09M). Address correspondence to: E.X.W. e-mail: [email protected] ªAUR, 2011 doi:10.1016/j.acra.2010.10.012
progression or regression in response to treatment (3). Because disease progression to liver cirrhosis can be prevented by early interventions and treatments (4–6), there has been a great interest in the development of noninvasive techniques for early diagnosis and characterization of liver fibrosis. Proton magnetic resonance spectroscopy (1H MRS) allows the study of cellular biochemistry and metabolism, and provides a noninvasive means to determine disease abnormalities and progression in vivo and longitudinally. Liver lipid content, which has been suggested to play an important pathogenic role in the development of liver fibrosis and cirrhosis in patients with chronic hepatitis C (7,8) and nonalcoholic steatohepatitis (9–11), can be measured by 1H MRS noninvasively. Most importantly, specific lipid changes related to saturated and unsaturated fatty acids can be monitored in vivo by high resolution 1H MRS. Carbon tetrachloride (CCl4) intoxication is a wellcharacterized, reproducible and the most commonly used experimental animal model of liver fibrosis. It has been widely 377
CHEUNG ET AL
Academic Radiology, Vol 18, No 3, March 2011
animals induced with liver fibrosis is shown in Figure 1. Normal male adult Sprague-Dawley rats (220 to 260 g; n = 8) were used as controls. In Vivo Liver 1H MRS Experiments
Figure 1. Schedule of carbon tetrachloride (CCl4) twice-weekly administration for induction of liver fibrosis in adult SpragueDawley rats, 1H proton magnetic resonance spectroscopy (MRS) experiments, and liver histology.
studied with respect to the histological, biochemical, cellular, and molecular changes associated with development of fibrosis (12,13). By interfering hepatic energy metabolism and protein synthesis, CCl4-induced hepatotoxicity can lead to triglyceride accumulation, mitochondrial injury, and necrosis (14). With the increased availability of high-field ($3.0 Tesla) magnetic resonance (MR) systems for clinical and preclinical studies, both signal-to-noise ratio (SNR) and spectral resolution of metabolites in the MR spectra can be improved significantly (15), allowing more accurate metabolite identification and quantification and thus disease characterization. Although MRS can provide insights into liver metabolism noninvasively, detailed in vivo 1H MRS study of liver fibrosis with high spectral resolution has been limited (16–18). The aim of this study was to characterize early hepatic lipid changes in the experimental CCl4-induced liver fibrosis model by means of single-voxel 1H MRS at high magnetic field in vivo.
All 1H MRS experiments were performed on a 7 Tesla MR imaging scanner with a maximum gradient of 360 mT/m (70/16 PharmaScan, Bruker Biospin GmbH, Germany). A 60-mm inner diameter quadrature resonator was used for both radiofrequency (RF) transmission and receiving. During liver imaging, each animal was anesthetized with isoflurane/ air using 1.0 to 1.5% for maintenance via a nose cone with respiratory monitoring (23,24). Body temperature was maintained at about 36.5 C by circulating warm water in a heating pad. Scout images were first acquired in three orthogonal planes with a fast low angle shot sequence for localization of a voxel or volume of interest for MRS. A 5 5 5 mm3 voxel was chosen within a homogeneous liver parenchyma to avoid large blood vessels. First- and second-order localized automatic shimming was first performed within the voxel until a full width at half maximum <50 Hz was achieved in the water peak. The water signal was suppressed by variable power RF pulses with optimized relaxation delays with bandwidth of 200 Hz. Outer volume suppression combined with respiratory-triggered single-voxel point-resolved spectroscopic sequence was used for acquiring liver MR spectrum, with repetition time (TR) = two respiratory cycles (∼2.0 to 2.5 seconds), echo time (TE) = 15 ms, receiver bandwidth = 4 kHz, 2048 data points, 256 averages, and total scan time of ∼10 minutes. Note that respiratory triggering was used to minimize voxel misregistration in the presence of respiratory motion, while short TE was chosen to reduce signal loss due to T2 relaxation to improve SNR (25).
MATERIALS AND METHODS Animal Preparation
Data and Statistical Analysis
All animal experiments were approved by the institutional animal ethics committee. Liver fibrosis was induced in male adult Sprague-Dawley rats (220 to 260 g; n = 12) by subcutaneous injection of 1:1 volume mixture of CCl4 in olive oil at a dose of 0.2 mL/100 g of body weight twice a week for 4 weeks (13,19). Intermittent administration of CCl4 has been widely used to experimentally induce liver fibrosis in rodents by evoking a marked infiltration of inflammatory cells, thus mimicking the changes in chronic viral hepatitis‑associated fibrosis in many ways (20,21). The twice-weekly dosing can induce early stage of liver fibrosis and established fibrosis after 2 and 4 weeks of CCl4 administration, respectively, in rodents (13,22). This well-controlled CCl4-induced liver fibrosis model allows the study of a homogeneous population of liver fibrosis. 1H MRS was performed in the CCl4-insulted animals at 2 and 4 weeks after the start of CCl4 administration. The animals were examined 48 hours after last CCl4 administration to avoid acute inflammatory effects (18). The overall schedule of the experiment for
The MRS data were processed using the MR spectroscopic analysis package provided by the MR imaging vendor (26,27). MR spectra were zero-filled to 8192 data points, apodized with a 2-Hz exponential filter, Fourier transformed, 0th- and first-order phase corrected, and baseline corrected. Signal integrals of lipid methyl protons (–CH3; 0.9 ppm), methylene protons ((–CH2–)n; 1.3 ppm), allylic protons (–CH2–C=C–CH2–; 2.0 ppm), diallylic protons (=C–CH2– C=; 2.8 ppm), methene protons (–CH=CH–; 5.3 ppm), and protons from choline-containing compounds (CCC; 3.2 ppm) (11,28,29) were measured by integrating areas under peaks. Spectral noise was calculated from the standard deviation of the last portion of spectrum from 7.3 to 11.3 ppm, in which no metabolite was observable in liver (11,28,29). Total lipid and CCC indices were quantified by dividing peak area of (–CH2–)n and CCC by spectral noise, respectively, given the fact that SNR was similar among spectra because of identical voxel size and nearly identical hardware settings in different measurements. In addition, total
378
1
Academic Radiology, Vol 18, No 3, March 2011
H MRS LIPID PROFILING IN LIVER FIBROSIS
After the MR examination following 2 weeks of CCl4 administration, 4 of 12 animals were sacrificed for histological evaluation. Furthermore, four of the remaining eight animals were sacrificed after MR examination after 4 weeks of CCl4 insult as shown in Figure 1. One additional normal animal was sacrificed as a control. Liver specimens were fixed in formalin, embedded in paraffin, sectioned and examined by light microscopy after standard hematoxylin-eosin staining and Masson’s trichrome staining (31,32).
signals from CCC, as compared with the normal animals. Water (H2O) signal at 4.7 ppm was effectively suppressed in the spectra with varying degree. Figure 3 shows the total lipid, CCC, total saturated fatty acid, total unsaturated fatty acid, total unsaturated bond and polyunsaturated bond indices in normal animals (n = 8) and animals with 2-week (n = 12) and 4-week (n = 8) CCl4 twice-weekly administration. Total lipid indices of animals at 2 weeks (3.97 1.57 104) and 4 weeks (5.66 1.31 104) after fibrosis induction were found to be significantly higher than (P < .01) those of normal control animals (0.72 0.17 104), along with significant difference (P < .05) between 2 and 4 weeks. Meanwhile, a similar trend was observed in total saturated fatty acid index. Significant increase (P < .01) in total saturated fatty acid index was found in animals with CCl4-induced liver fibrosis (8.74 1.68 and 10.42 1.35 for 2-week and 4-week CCl4 administration, respectively), as compared with normal animals (5.91 1.23), whereas the difference between 2 and 4 weeks was also statistically significant (P < .05). On the other hand, there was only significant increase (P < .01) in total unsaturated fatty acid index in animals with 4-week CCl4 insult as compared with normal animals. Furthermore, total unsaturated bond index of animals at 4 weeks after CCl4 insult (1.75 0.30) were significantly higher than (P < .01 and P < .05, respectively) those of normal animals (1.10 0.32) and animals at 2 weeks after insult (1.32 0.30). Similarly, polyunsaturated bond index of animals after 4-week CCl4 administration (0.95 0.18) were significantly higher than (P < .01 and P < .05, respectively) those of normal animals (0.57 0.18) and animals with 2-week CCl4 insult (0.73 0.14). No significant differences were observed in CCC between normal and diseased animals. Figure 4 shows the typical hematoxylin-eosin and Masson’s trichrome staining of normal liver and livers at 2 weeks and 4 weeks after CCl4 insult. Collagen deposition was stained as blue by Masson’s trichrome staining in fibrotic livers. Compared with normal liver (Fig 4a), collagen deposition and intracellular fat vacuoles were consistently observed in livers with CCl4 insult (Fig 4b, 4c). Similar histological findings were observed in all liver samples collected, and they were largely consistent with those from the earlier studies of CCl4induced liver fibrosis in rodent models (18). The histological observations of collagen deposition in the liver samples collected confirmed the liver fibrogenesis in the animals studied.
RESULTS
DISCUSSION
TABLE 1. Peak Area Ratios of Various Metabolite Indices Measured by Proton Magnetic Resonance Spectroscopy (1H MRS) Index Total lipid Total saturated fatty acid Total unsaturated fatty acid Total unsaturated bond Polyunsaturated bond Choline-containing compound (CCC)
Peak Area Ratio
Frequency
(–CH2–)n/noise 3(–CH2–)/2(–CH3) 3(–CH2–C=C–CH2–)/4 (–CH3) 3(–CH=CH–)/2(–CH3) 3(=C–CH2–C=)/2(–CH3) CCC/noise
1.3 ppm 1.3/0.9 ppm 2.0/0.9 ppm 5.3/0.9 ppm 2.8/0.9 ppm 3.2 ppm
saturated fatty acid, total unsaturated fatty acid, total unsaturated bond, and polyunsaturated bond indices were estimated by –CH2–C= dividing peak area of (–CH2–)n, C–CH2–, –CH=CH– and =C–CH2–C= by peak area of –CH3, respectively, and scaled to the relative number of protons contributing to the resonance (11), as shown in Table 1. Note that the signal from (–CH2–)n and –CH2–C= C–CH2– increases with increasing number of saturated and unsaturated fatty acids, respectively, whereas the signal from –CH=CH– and =C–CH2–C= increases with the percentage of total unsaturated and polyunsaturated double bonds in the unsaturated fatty acids, respectively (11,30). Furthermore, the signal contribution of glutamine and glutamate at 2.2 ppm was assumed to be negligible (11). Results were expressed as mean standard deviation. One-way analysis of variance with Tukey’s multiple comparison test was employed to compare differences in ratios of peak areas between fibrosisinduced and control animals, with P values less than .05 considered statistically significant. Histology
1
Figure 2 shows the typical liver H MRS spectra from a normal control animal and an animal assessed at 2 and 4 weeks after start of CCl4 twice-weekly administration, with a typical voxel placement shown in the anatomical image. It was consistently observed that all animals with CCl4-induced liver fibrosis exhibited substantial increase in various lipid peaks including –CH3, (–CH2–)n, –CH2–C=C–CH2–, –COO–CH2–, =C–CH2–C= and –CH=CH–, with similar
As early as 2 weeks after the start of CCl4 administration, total lipid and total saturated fatty acid indices were found to increase substantially (P < .01) in animals with liver fibrosis as compared with control animals (Fig 3a, 3c). Meanwhile, no increases in total unsaturated fatty acid index were observed (Fig 3d), suggesting that the total saturated fatty acid increase contributed mainly to the total lipid increase in animals with CCl4 insult. The total lipid increase in animals 379
CHEUNG ET AL
Academic Radiology, Vol 18, No 3, March 2011
Figure 2. Typical liver 1H proton magnetic resonance spectroscopy (MRS) spectra from a normal control animal and an animal scanned at 2 and 4 weeks after start of carbon tetrachloride (CCl4) twice-weekly administration, with a typical voxel placement (white square) shown in the anatomical image. Animals with CCl4induced liver fibrosis consistently showed markedly increases in various lipid peaks including methyl protons (–CH3; 0.9 ppm), methylene protons ((–CH2–)n; 1.3 ppm), allylic protons (–CH2–C=C–CH2–; 2.0 ppm), a-methylene protons to carboxyl (–COO–CH2–; 2.2 ppm), diallylic protons (=C–CH2–C=; 2.8 ppm) and methene protons (–CH=CH–; 5.3 ppm), except the peak from choline-containing compounds (CCC; 3.2 ppm). The water (H2O; 4.7 ppm) signals were effectively suppressed in the spectra.
Figure 3. (a) Total lipid, (b) choline-containing compounds (CCC), (c) total saturated fatty acid, (d) total unsaturated fatty acid, (e) total unsaturated bond, and (f) polyunsaturated bond indices in normal animals and animals with 2-week and 4-week CCl4 twice-weekly administration. One-way analysis of variance (ANOVA) with Tukey’s multiple comparison test was performed with **P < 0.01, *P < 0.05 and n.s. for insignificance.
with CCl4-induced liver fibrosis was likely due to fatty infiltration/fatty changes in hepatocytes (Fig 4). During toxic CCl4 insult, hepatocytes are incapable of synthesizing 380
lipoproteins that are needed for removing triglycerides in the cytoplasm as a result of the destruction of microsomal proteins by lipid peroxidation (18,33,34), leading to
Academic Radiology, Vol 18, No 3, March 2011
1
H MRS LIPID PROFILING IN LIVER FIBROSIS
Figure 4. Typical hematoxylin-eosin (H&E) staining (400; left column) and Masson’s trichrome staining (200 and 40; middle and right column, respectively) of normal liver (a), and livers subjected to 2-week (b) and 4-week (c) carbon tetrachloride (CCl4) twice-weekly administration. Collagen deposition (green arrows), fat vacuoles (blue arrows), and cell necrosis/apoptosis (black arrows) were observed in the insulted livers.
increased triglyceride accumulation (35). It is noteworthy that 1 H MRS can provide valuable information on lipid composition, which cannot be revealed by histological analysis. The total saturated fatty acid increase in liver may reflect the lipid-induced cell toxicity, which has been suggested to be related with activated apoptosis induced by saturated fatty acids (9,10,36). In nonalcoholic fatty liver disease, endoplasmic reticulum stress associated with increased saturated fatty acids in the liver have been shown to promote liver injury and partly contribute to the disease progression from simple steatosis to steatohepatitis (9,10). Nonetheless, the potential toxic effect of increased saturated fatty acids in fibrotic livers has yet to be determined. Although there was no significant change in total unsaturated fatty acid index at week 4 as compared with week 2 after CCl4 administration, significant increase (P < .05) in total unsaturated bond index was observed (Fig 3e). These findings suggested that without a substantial increase in amount of unsaturated fatty acids, more unsaturated double bonds were formed within the unsaturated fatty acids at 4 weeks after
the start of fibrosis induction. As such, significant increase in polyunsaturated bond index (P < .05) was observed (Fig 3f) as expected. Therefore, the amount of polyunsaturated fatty acids was likely to increase at the expense of monounsaturated fatty acids in the animals with 4-week CCl4 insult. Such increase in degree of polyunsaturation has been observed and ascribed to increased necrosis/apoptosis in various experimental models (11,37,38), which was also observed in the current study (Fig 4). Because CCC has been considered to represent the important constituents in phospholipid metabolism of cell membranes (39), similar CCC levels observed in the normal and CCl4-insulted animals were largely expected in the early stage of liver fibrosis because of the similar rate of cell turnover of hepatocytes prior to the development of cirrhosis (40). In vivo phosphorus-31 (31P) MRS has been found promising in evaluating the cellular turnover and liver energy status (41–43), but it cannot measure the hepatic lipid content and is not readily available on most clinical MR imaging scanners. By contrast, in vivo 1H MRS is feasible on most standard MR 381
CHEUNG ET AL
imaging scanners, where high-resolution anatomical images and functional information such as perfusion can be acquired in the same imaging session. 1H MRS has been used to investigate focal liver lesions and found to be useful in characterizing HCC in humans (44–46) and animal models (37,47–49) by providing valuable metabolic information. However, only limited 1H MRS studies have been performed to investigate liver fibrosis in both humans and animals (16–18). Note that our study used the highest field strength to study liver fibrosis compared with previous studies (16–18). The advantages of performing 1H MRS at high magnetic field include better SNR and increased spectral resolution (15), thus providing high-quality spectra in acceptable scan times. In this regard, 1H MRS at 7 Tesla in this study can provide precise biochemical information in liver noninvasively that can be used for characterizing and monitoring of liver diseases including liver fibrosis. It is worthwhile to note that it is relatively difficult to accurately resolve and quantify several lipid signals other than (–CH2–)n peak because of limited spectral resolution and poor SNR at low field strengths in previous studies (16,18). In contrast, our current study provided more detailed information about lipid composition by resolving various lipid peaks (eg, –CH2–C=C–CH2–, =C–CH2–C= and –CH=CH–) in fibrotic livers at 7 Tesla. Note that the separation between methene (5.3 ppm) and water (4.7 ppm) peaks is 180 Hz at 7 Tesla, thus the methene proton peak was minimally affected by the water suppression RF pulses with bandwidth of 200 Hz. In addition, the effects of motion on liver MRS were largely reduced using respiratory gating in the current study, providing water-suppressed MR spectra with high quality (Fig 2). It is worth noting that although fat-selective MR imaging using Dixon methods can provide quantitative information about hepatic lipid content, these techniques are only sensitive to one specific lipid proton species (ie, methylene protons [3.4 ppm apart from water]). In contrast, 1H MRS can resolve and quantify various lipid proton species so as to provide insights into lipid composition. Ratio measurements using spectral noise as internal reference were employed for quantifying total lipid and CCC levels in the current study. Note that spectral noise could be affected by a number of factors such as local sensitivity of coil and acquisition protocol. Such factors were expected to be similar among measurements in this study because of the use of volume RF coil, identical hardware settings and acquisition protocol, yielding similar noise levels in the spectra of different measurements (Fig 2). The use of lipid peak as internal reference (16) was avoided, because the amount of lipid increased markedly in the CCl4-induced liver fibrosis model (Fig 2). Unsuppressed water signal was not chosen as internal reference either because the water content may be affected by pathological conditions (46). Previous human studies reported inconsistent observations on the lipid levels in chronic hepatitis (16,17), probably due to varying types of virus-induced hepatitis investigated. Our results showed an increase in total lipid levels, consistent with other MR 382
Academic Radiology, Vol 18, No 3, March 2011
studies in CCl4-induced fibrosis model (18,35). However, changes in saturated fatty acids, unsaturated fatty acids and unsaturated double bonds were not investigated in those studies. Our MRS data at high field allowed better discrimination and quantification of the –CH2–C=C–CH2–, –CH=CH– and =C–CH2–C= lipid peaks, thus total unsaturated fatty acid, total unsaturated bond and polyunsaturated bond indices can be measured. Moreover, the elevated total saturated fatty acid measured by 1H MRS in the current study may suggest its role in the development of liver fibrosis (9,10). One potential limitation of the present study is lack of sham controls receiving olive oil alone. As subcutaneous injection of CCl4 with olive oil is a well-established protocol to induce liver fibrosis in rodents (13), effects of olive oil were not examined in the current study. Nonetheless, previous studies showed that hepatic lipid changes are associated with metabolic alterations due to intoxication of CCl4 (50), whereas no evidences of steatosis induced by olive oil have been found. In conclusion, the lipid changes related to saturated and unsaturated fatty acids in CCl4-induced experimental fibrosis model were documented in vivo and longitudinally using 1H MRS at 7 Tesla. Our experimental results demonstrated that 1 H MRS at high-field was useful in detecting and characterizing various hepatic lipid alterations as early as 2 weeks from the start of induction of liver fibrosis in the animal model. 1H MRS may be valuable in detecting fatty changes at early phase in human liver fibrosis prior to the development of cirrhosis, and potentially useful in determining treatment strategies and evaluating therapeutic outcomes.
REFERENCES 1. Moradpour D, Wands JR. The molecular pathogenesis of hepatocellular carcinoma. J Viral Hepat 1994; 1:17–31. 2. Bataller R, Brenner DA. Liver fibrosis. J Clin Invest 2005; 115:209–218. 3. Manning DS, Afdhal NH. Diagnosis and quantitation of fibrosis. Gastroenterology 2008; 134:1670–1681. 4. Wanless IR, Nakashima E, Sherman M. Regression of human cirrhosis. Morphologic features and the genesis of incomplete septal cirrhosis. Arch Pathol Lab Med 2000; 124:1599–1607. 5. Arthur MJ. Reversibility of liver fibrosis and cirrhosis following treatment for hepatitis C. Gastroenterology 2002; 122:1525–1528. 6. Veldhuijzen IK, Toy M, Hahne SJ, et al. Screening and early treatment of migrants for chronic hepatitis B virus infection is cost-effective. Gastroenterology 2010; 138:522–530. 7. Hourigan LF, Macdonald GA, Purdie D, et al. Fibrosis in chronic hepatitis C correlates significantly with body mass index and steatosis. Hepatology 1999; 29:1215–1219. 8. Castera L, Hezode C, Roudot-Thoraval F, et al. Worsening of steatosis is an independent factor of fibrosis progression in untreated patients with chronic hepatitis C and paired liver biopsies. Gut 2003; 52:288–292. 9. Wang D, Wei Y, Pagliassotti MJ. Saturated fatty acids promote endoplasmic reticulum stress and liver injury in rats with hepatic steatosis. Endocrinology 2006; 147:943–951. 10. Gentile CL, Pagliassotti MJ. The role of fatty acids in the development and progression of nonalcoholic fatty liver disease. J Nutr Biochem 2008; 19: 567–576. 11. Corbin IR, Furth EE, Pickup S, et al. In vivo assessment of hepatic triglycerides in murine non-alcoholic fatty liver disease using magnetic resonance spectroscopy. Biochim Biophys Acta 2009; 1791:757–763. 12. Tsukamoto H, Matsuoka M, French SW. Experimental models of hepatic fibrosis: a review. Semin Liver Dis 1990; 10:56–65.
Academic Radiology, Vol 18, No 3, March 2011
13. Constandinou C, Henderson N, Iredale JP. Modeling liver fibrosis in rodents. Methods Mol Med 2005; 117:237–250. 14. Recknagel RO. Carbon tetrachloride hepatotoxicity. Pharmacol Rev 1967; 19:145–208. 15. Gruetter R, Weisdorf SA, Rajanayagan V, et al. Resolution improvements in in vivo 1H NMR spectra with increased magnetic field strength. J Magn Reson 1998; 135:260–264. 16. Cho SG, Kim MY, Kim HJ, et al. Chronic hepatitis: in vivo proton MR spectroscopic evaluation of the liver and correlation with histopathologic findings. Radiology 2001; 221:740–746. 17. Orlacchio A, Bolacchi F, Angelico M, et al. In vivo, high-field, 3-Tesla 1H MR spectroscopic assessment of liver fibrosis in HCV-correlated chronic liver disease. Radiol Med 2008; 113:289–299. 18. Hazle JD, Narayana PA, Dunsford HA. Chronic carbon tetrachloride and phospholipase D hepatotoxicity in rat: in vivo 1H magnetic resonance, total lipid analysis, and histology. Magn Reson Med 1990; 15:211–228. 19. Man K, Ng KT, Lee TK, et al. FTY720 attenuates hepatic ischemiareperfusion injury in normal and cirrhotic livers. Am J Transplant 2005; 5: 40–49. 20. Proctor E, Chatamra K. Controlled induction of cirrhosis in the rat. Br J Exp Pathol 1983; 64:320–330. 21. Armendariz-Borunda J, Seyer JM, Kang AH, et al. Regulation of TGF beta gene expression in rat liver intoxicated with carbon tetrachloride. FASEB J 1990; 4:215–221. 22. Zeisberg M, Yang C, Martino M, et al. Fibroblasts derive from hepatocytes in liver fibrosis via epithelial to mesenchymal transition. J Biol Chem 2007; 282:23337–23347. 23. Cheung JS, Guo H, Leung JC, et al. MRI visualization of rodent liver structure and peritoneal adhesion with dialyzate enhancement. Magn Reson Med 2008; 59:1170–1174. 24. Cheung JS, Fan SJ, Chow AM, et al. In vivo DTI assessment of hepatic ischemia reperfusion injury in an experimental rat model. J Magn Reson Imaging 2009; 30:890–895. 25. Zhong K, Ernst T. Localized in vivo human 1H MRS at very short echo times. Magn Reson Med 2004; 52:898–901. 26. Chan KC, Khong PL, Lau HF, et al. Late measures of microstructural alterations in severe neonatal hypoxic-ischemic encephalopathy by MR diffusion tensor imaging. Int J Dev Neurosci 2009; 27:607–615. 27. Chan KC, So KF, Wu EX. Proton magnetic resonance spectroscopy revealed choline reduction in the visual cortex in an experimental model of chronic glaucoma. Exp Eye Res 2009; 88:65–70. 28. Hayakawa Y, Yoshioka Y, Yasuda N. Effects of ligation and reperfusion of hepatic afferent vessels on the composition of liver cell membrane in the rat: 1H- and 31P-magnetic resonance spectroscopic analysis. NMR Biomed 1997; 10:257–262. 29. Lim AK, Hamilton G, Patel N, et al. 1H MR spectroscopy in the evaluation of the severity of chronic liver disease. Radiology 2003; 226:288–289. 30. Zancanaro C, Nano R, Marchioro C, et al. Magnetic resonance spectroscopy investigations of brown adipose tissue and isolated brown adipocytes. J Lipid Res 1994; 35:2191–2199. 31. Kim H, Booth CJ, Pinus AB, et al. Induced hepatic fibrosis in rats: hepatic steatosis, macromolecule content, perfusion parameters, and their correlations—preliminary MR imaging in rats. Radiology 2008; 247:696–705. 32. Wu EX, Wu Y, Nicholls JM, et al. MR diffusion tensor imaging study of postinfarct myocardium structural remodeling in a porcine model. Magn Reson Med 2007; 58:687–695.
1
H MRS LIPID PROFILING IN LIVER FIBROSIS
33. Bruckner JV, MacKenzie WF, Muralidhara S, et al. Oral toxicity of carbon tetrachloride: acute, subacute, and subchronic studies in rats. Fundam Appl Toxicol 1986; 6:16–34. 34. Hazle JD, Narayana PA, Dunsford HA. In vivo NMR, biochemical, and histologic evaluation of alcohol-induced fatty liver in rat and a comparison with CCl4 hepatotoxicity. Magn Reson Med 1991; 19:124–135. 35. Stark DD, Bass NM, Moss AA, et al. Nuclear magnetic resonance imaging of experimentally induced liver disease. Radiology 1983; 148:743–751. 36. de Vries JE, Vork MM, Roemen TH, et al. Saturated but not monounsaturated fatty acids induce apoptotic cell death in neonatal rat ventricular myocytes. J Lipid Res 1997; 38:1384–1394. 37. Foley LM, Towner RA, Painter DM. In vivo image-guided (1)H-magnetic resonance spectroscopy of the serial development of hepatocarcinogenesis in an experimental animal model. Biochim Biophys Acta 2001; 1526: 230–236. 38. Griffin JL, Lehtimaki KK, Valonen PK, et al. Assignment of 1H nuclear magnetic resonance visible polyunsaturated fatty acids in BT4C gliomas undergoing ganciclovir-thymidine kinase gene therapy-induced programmed cell death. Cancer Res 2003; 63:3195–3201. 39. Ruiz-Cabello J, Cohen JS. Phospholipid metabolites as indicators of cancer cell function. NMR Biomed 1992; 5:226–233. 40. Menon DK, Sargentoni J, Taylor-Robinson SD, et al. Effect of functional grade and etiology on in vivo hepatic phosphorus-31 magnetic resonance spectroscopy in cirrhosis: biochemical basis of spectral appearances. Hepatology 1995; 21:417–427. 41. Corbin IR, Buist R, Peeling J, et al. Hepatic 31P MRS in rat models of chronic liver disease: assessing the extent and progression of disease. Gut 2003; 52:1046–1053. 42. Lim AK, Patel N, Hamilton G, et al. The relationship of in vivo 31P MR spectroscopy to histology in chronic hepatitis C. Hepatology 2003; 37:788–794. 43. Dezortova M, Taimr P, Skoch A, et al. Etiology and functional status of liver cirrhosis by 31P MR spectroscopy. World J Gastroenterol 2005; 11: 6926–6931. 44. Soper R, Himmelreich U, Painter D, et al. Pathology of hepatocellular carcinoma and its precursors using proton magnetic resonance spectroscopy and a statistical classification strategy. Pathology 2002; 34:417–422. 45. Kuo YT, Li CW, Chen CY, et al. In vivo proton magnetic resonance spectroscopy of large focal hepatic lesions and metabolite change of hepatocellular carcinoma before and after transcatheter arterial chemoembolization using 3.0-T MR scanner. J Magn Reson Imaging 2004; 19:598–604. 46. Li CW, Kuo YC, Chen CY, et al. Quantification of choline compounds in human hepatic tumors by proton MR spectroscopy at 3 T. Magn Reson Med 2005; 53:770–776. 47. Towner RA, Foley LM, Painter DM. Hepatocarcinogenesis tumor grading correlated with in vivo image-guided 1H-NMR spectroscopy in a rat model. Toxicol Appl Pharmacol 2005; 207:237–244. 48. Zhao WD, Guan S, Zhou KR, et al. In vivo detection of metabolic changes by 1H-MRS in the DEN-induced hepatocellular carcinoma in Wistar rat. J Cancer Res Clin Oncol 2005; 131:597–602. 49. Xu H, Li X, Yang ZH, et al. In vivo 1H MR spectroscopy in the evaluation of the serial development of hepatocarcinogenesis in an experimental rat model. Acad Radiol 2006; 13:1532–1537. 50. Le Page RN, Cheeseman KH, Osman N, et al. Lipid peroxidation in purified plasma membrane fractions of rat liver in relation to the hepatoxicity of carbon tetrachloride. Cell Biochem Funct 1988; 6:87–99.
383
L
iver fibrosis associated with chronic liver injury can progress to cirrhosis and ultimately hepatocellular carcinoma (HCC) (1,2). Percutaneous liver biopsy has been considered the standard technique for diagnosis and staging of liver fibrosis. However, liver biopsy is highly invasive and associated with risk of complications, limiting its applicability in longitudinal monitoring of fibrosis Acad Radiol 2011; 18:377–383 From the Laboratory of Biomedical Imaging and Signal Processing (J.S.C., S.J.F., D.S.G., A.M.C., J.Y., E.X.W.), Departments of Electrical and Electronic Engineering (J.S.C., S.J.F., D.S.G., A.M.C., J.Y., E.X.W.), Surgery (K.M.), and Anatomy (E.X.W.), The University of Hong Kong, Pokfulam, Hong Kong SAR, China; Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital and Harvard Medical School, Charlestown, MA 02129, USA (J.S.C); Department of Diagnostic Radiology of the First Affiliated Hospital, School of Medicine of Xi’an Jiaotong University, Xi’an, Shannxi Province, China (J.Y.). Received September 12, 2010; accepted October 29, 2010. Supported by the Hong Kong Grant Council (GRF HKU7808/09M). Address correspondence to: E.X.W. e-mail: [email protected] ªAUR, 2011 doi:10.1016/j.acra.2010.10.012
progression or regression in response to treatment (3). Because disease progression to liver cirrhosis can be prevented by early interventions and treatments (4–6), there has been a great interest in the development of noninvasive techniques for early diagnosis and characterization of liver fibrosis. Proton magnetic resonance spectroscopy (1H MRS) allows the study of cellular biochemistry and metabolism, and provides a noninvasive means to determine disease abnormalities and progression in vivo and longitudinally. Liver lipid content, which has been suggested to play an important pathogenic role in the development of liver fibrosis and cirrhosis in patients with chronic hepatitis C (7,8) and nonalcoholic steatohepatitis (9–11), can be measured by 1H MRS noninvasively. Most importantly, specific lipid changes related to saturated and unsaturated fatty acids can be monitored in vivo by high resolution 1H MRS. Carbon tetrachloride (CCl4) intoxication is a wellcharacterized, reproducible and the most commonly used experimental animal model of liver fibrosis. It has been widely 377
CHEUNG ET AL
Academic Radiology, Vol 18, No 3, March 2011
animals induced with liver fibrosis is shown in Figure 1. Normal male adult Sprague-Dawley rats (220 to 260 g; n = 8) were used as controls. In Vivo Liver 1H MRS Experiments
Figure 1. Schedule of carbon tetrachloride (CCl4) twice-weekly administration for induction of liver fibrosis in adult SpragueDawley rats, 1H proton magnetic resonance spectroscopy (MRS) experiments, and liver histology.
studied with respect to the histological, biochemical, cellular, and molecular changes associated with development of fibrosis (12,13). By interfering hepatic energy metabolism and protein synthesis, CCl4-induced hepatotoxicity can lead to triglyceride accumulation, mitochondrial injury, and necrosis (14). With the increased availability of high-field ($3.0 Tesla) magnetic resonance (MR) systems for clinical and preclinical studies, both signal-to-noise ratio (SNR) and spectral resolution of metabolites in the MR spectra can be improved significantly (15), allowing more accurate metabolite identification and quantification and thus disease characterization. Although MRS can provide insights into liver metabolism noninvasively, detailed in vivo 1H MRS study of liver fibrosis with high spectral resolution has been limited (16–18). The aim of this study was to characterize early hepatic lipid changes in the experimental CCl4-induced liver fibrosis model by means of single-voxel 1H MRS at high magnetic field in vivo.
All 1H MRS experiments were performed on a 7 Tesla MR imaging scanner with a maximum gradient of 360 mT/m (70/16 PharmaScan, Bruker Biospin GmbH, Germany). A 60-mm inner diameter quadrature resonator was used for both radiofrequency (RF) transmission and receiving. During liver imaging, each animal was anesthetized with isoflurane/ air using 1.0 to 1.5% for maintenance via a nose cone with respiratory monitoring (23,24). Body temperature was maintained at about 36.5 C by circulating warm water in a heating pad. Scout images were first acquired in three orthogonal planes with a fast low angle shot sequence for localization of a voxel or volume of interest for MRS. A 5 5 5 mm3 voxel was chosen within a homogeneous liver parenchyma to avoid large blood vessels. First- and second-order localized automatic shimming was first performed within the voxel until a full width at half maximum <50 Hz was achieved in the water peak. The water signal was suppressed by variable power RF pulses with optimized relaxation delays with bandwidth of 200 Hz. Outer volume suppression combined with respiratory-triggered single-voxel point-resolved spectroscopic sequence was used for acquiring liver MR spectrum, with repetition time (TR) = two respiratory cycles (∼2.0 to 2.5 seconds), echo time (TE) = 15 ms, receiver bandwidth = 4 kHz, 2048 data points, 256 averages, and total scan time of ∼10 minutes. Note that respiratory triggering was used to minimize voxel misregistration in the presence of respiratory motion, while short TE was chosen to reduce signal loss due to T2 relaxation to improve SNR (25).
MATERIALS AND METHODS Animal Preparation
Data and Statistical Analysis
All animal experiments were approved by the institutional animal ethics committee. Liver fibrosis was induced in male adult Sprague-Dawley rats (220 to 260 g; n = 12) by subcutaneous injection of 1:1 volume mixture of CCl4 in olive oil at a dose of 0.2 mL/100 g of body weight twice a week for 4 weeks (13,19). Intermittent administration of CCl4 has been widely used to experimentally induce liver fibrosis in rodents by evoking a marked infiltration of inflammatory cells, thus mimicking the changes in chronic viral hepatitis‑associated fibrosis in many ways (20,21). The twice-weekly dosing can induce early stage of liver fibrosis and established fibrosis after 2 and 4 weeks of CCl4 administration, respectively, in rodents (13,22). This well-controlled CCl4-induced liver fibrosis model allows the study of a homogeneous population of liver fibrosis. 1H MRS was performed in the CCl4-insulted animals at 2 and 4 weeks after the start of CCl4 administration. The animals were examined 48 hours after last CCl4 administration to avoid acute inflammatory effects (18). The overall schedule of the experiment for
The MRS data were processed using the MR spectroscopic analysis package provided by the MR imaging vendor (26,27). MR spectra were zero-filled to 8192 data points, apodized with a 2-Hz exponential filter, Fourier transformed, 0th- and first-order phase corrected, and baseline corrected. Signal integrals of lipid methyl protons (–CH3; 0.9 ppm), methylene protons ((–CH2–)n; 1.3 ppm), allylic protons (–CH2–C=C–CH2–; 2.0 ppm), diallylic protons (=C–CH2– C=; 2.8 ppm), methene protons (–CH=CH–; 5.3 ppm), and protons from choline-containing compounds (CCC; 3.2 ppm) (11,28,29) were measured by integrating areas under peaks. Spectral noise was calculated from the standard deviation of the last portion of spectrum from 7.3 to 11.3 ppm, in which no metabolite was observable in liver (11,28,29). Total lipid and CCC indices were quantified by dividing peak area of (–CH2–)n and CCC by spectral noise, respectively, given the fact that SNR was similar among spectra because of identical voxel size and nearly identical hardware settings in different measurements. In addition, total
378
1
Academic Radiology, Vol 18, No 3, March 2011
H MRS LIPID PROFILING IN LIVER FIBROSIS
After the MR examination following 2 weeks of CCl4 administration, 4 of 12 animals were sacrificed for histological evaluation. Furthermore, four of the remaining eight animals were sacrificed after MR examination after 4 weeks of CCl4 insult as shown in Figure 1. One additional normal animal was sacrificed as a control. Liver specimens were fixed in formalin, embedded in paraffin, sectioned and examined by light microscopy after standard hematoxylin-eosin staining and Masson’s trichrome staining (31,32).
signals from CCC, as compared with the normal animals. Water (H2O) signal at 4.7 ppm was effectively suppressed in the spectra with varying degree. Figure 3 shows the total lipid, CCC, total saturated fatty acid, total unsaturated fatty acid, total unsaturated bond and polyunsaturated bond indices in normal animals (n = 8) and animals with 2-week (n = 12) and 4-week (n = 8) CCl4 twice-weekly administration. Total lipid indices of animals at 2 weeks (3.97 1.57 104) and 4 weeks (5.66 1.31 104) after fibrosis induction were found to be significantly higher than (P < .01) those of normal control animals (0.72 0.17 104), along with significant difference (P < .05) between 2 and 4 weeks. Meanwhile, a similar trend was observed in total saturated fatty acid index. Significant increase (P < .01) in total saturated fatty acid index was found in animals with CCl4-induced liver fibrosis (8.74 1.68 and 10.42 1.35 for 2-week and 4-week CCl4 administration, respectively), as compared with normal animals (5.91 1.23), whereas the difference between 2 and 4 weeks was also statistically significant (P < .05). On the other hand, there was only significant increase (P < .01) in total unsaturated fatty acid index in animals with 4-week CCl4 insult as compared with normal animals. Furthermore, total unsaturated bond index of animals at 4 weeks after CCl4 insult (1.75 0.30) were significantly higher than (P < .01 and P < .05, respectively) those of normal animals (1.10 0.32) and animals at 2 weeks after insult (1.32 0.30). Similarly, polyunsaturated bond index of animals after 4-week CCl4 administration (0.95 0.18) were significantly higher than (P < .01 and P < .05, respectively) those of normal animals (0.57 0.18) and animals with 2-week CCl4 insult (0.73 0.14). No significant differences were observed in CCC between normal and diseased animals. Figure 4 shows the typical hematoxylin-eosin and Masson’s trichrome staining of normal liver and livers at 2 weeks and 4 weeks after CCl4 insult. Collagen deposition was stained as blue by Masson’s trichrome staining in fibrotic livers. Compared with normal liver (Fig 4a), collagen deposition and intracellular fat vacuoles were consistently observed in livers with CCl4 insult (Fig 4b, 4c). Similar histological findings were observed in all liver samples collected, and they were largely consistent with those from the earlier studies of CCl4induced liver fibrosis in rodent models (18). The histological observations of collagen deposition in the liver samples collected confirmed the liver fibrogenesis in the animals studied.
RESULTS
DISCUSSION
TABLE 1. Peak Area Ratios of Various Metabolite Indices Measured by Proton Magnetic Resonance Spectroscopy (1H MRS) Index Total lipid Total saturated fatty acid Total unsaturated fatty acid Total unsaturated bond Polyunsaturated bond Choline-containing compound (CCC)
Peak Area Ratio
Frequency
(–CH2–)n/noise 3(–CH2–)/2(–CH3) 3(–CH2–C=C–CH2–)/4 (–CH3) 3(–CH=CH–)/2(–CH3) 3(=C–CH2–C=)/2(–CH3) CCC/noise
1.3 ppm 1.3/0.9 ppm 2.0/0.9 ppm 5.3/0.9 ppm 2.8/0.9 ppm 3.2 ppm
saturated fatty acid, total unsaturated fatty acid, total unsaturated bond, and polyunsaturated bond indices were estimated by –CH2–C= dividing peak area of (–CH2–)n, C–CH2–, –CH=CH– and =C–CH2–C= by peak area of –CH3, respectively, and scaled to the relative number of protons contributing to the resonance (11), as shown in Table 1. Note that the signal from (–CH2–)n and –CH2–C= C–CH2– increases with increasing number of saturated and unsaturated fatty acids, respectively, whereas the signal from –CH=CH– and =C–CH2–C= increases with the percentage of total unsaturated and polyunsaturated double bonds in the unsaturated fatty acids, respectively (11,30). Furthermore, the signal contribution of glutamine and glutamate at 2.2 ppm was assumed to be negligible (11). Results were expressed as mean standard deviation. One-way analysis of variance with Tukey’s multiple comparison test was employed to compare differences in ratios of peak areas between fibrosisinduced and control animals, with P values less than .05 considered statistically significant. Histology
1
Figure 2 shows the typical liver H MRS spectra from a normal control animal and an animal assessed at 2 and 4 weeks after start of CCl4 twice-weekly administration, with a typical voxel placement shown in the anatomical image. It was consistently observed that all animals with CCl4-induced liver fibrosis exhibited substantial increase in various lipid peaks including –CH3, (–CH2–)n, –CH2–C=C–CH2–, –COO–CH2–, =C–CH2–C= and –CH=CH–, with similar
As early as 2 weeks after the start of CCl4 administration, total lipid and total saturated fatty acid indices were found to increase substantially (P < .01) in animals with liver fibrosis as compared with control animals (Fig 3a, 3c). Meanwhile, no increases in total unsaturated fatty acid index were observed (Fig 3d), suggesting that the total saturated fatty acid increase contributed mainly to the total lipid increase in animals with CCl4 insult. The total lipid increase in animals 379
CHEUNG ET AL
Academic Radiology, Vol 18, No 3, March 2011
Figure 2. Typical liver 1H proton magnetic resonance spectroscopy (MRS) spectra from a normal control animal and an animal scanned at 2 and 4 weeks after start of carbon tetrachloride (CCl4) twice-weekly administration, with a typical voxel placement (white square) shown in the anatomical image. Animals with CCl4induced liver fibrosis consistently showed markedly increases in various lipid peaks including methyl protons (–CH3; 0.9 ppm), methylene protons ((–CH2–)n; 1.3 ppm), allylic protons (–CH2–C=C–CH2–; 2.0 ppm), a-methylene protons to carboxyl (–COO–CH2–; 2.2 ppm), diallylic protons (=C–CH2–C=; 2.8 ppm) and methene protons (–CH=CH–; 5.3 ppm), except the peak from choline-containing compounds (CCC; 3.2 ppm). The water (H2O; 4.7 ppm) signals were effectively suppressed in the spectra.
Figure 3. (a) Total lipid, (b) choline-containing compounds (CCC), (c) total saturated fatty acid, (d) total unsaturated fatty acid, (e) total unsaturated bond, and (f) polyunsaturated bond indices in normal animals and animals with 2-week and 4-week CCl4 twice-weekly administration. One-way analysis of variance (ANOVA) with Tukey’s multiple comparison test was performed with **P < 0.01, *P < 0.05 and n.s. for insignificance.
with CCl4-induced liver fibrosis was likely due to fatty infiltration/fatty changes in hepatocytes (Fig 4). During toxic CCl4 insult, hepatocytes are incapable of synthesizing 380
lipoproteins that are needed for removing triglycerides in the cytoplasm as a result of the destruction of microsomal proteins by lipid peroxidation (18,33,34), leading to
Academic Radiology, Vol 18, No 3, March 2011
1
H MRS LIPID PROFILING IN LIVER FIBROSIS
Figure 4. Typical hematoxylin-eosin (H&E) staining (400; left column) and Masson’s trichrome staining (200 and 40; middle and right column, respectively) of normal liver (a), and livers subjected to 2-week (b) and 4-week (c) carbon tetrachloride (CCl4) twice-weekly administration. Collagen deposition (green arrows), fat vacuoles (blue arrows), and cell necrosis/apoptosis (black arrows) were observed in the insulted livers.
increased triglyceride accumulation (35). It is noteworthy that 1 H MRS can provide valuable information on lipid composition, which cannot be revealed by histological analysis. The total saturated fatty acid increase in liver may reflect the lipid-induced cell toxicity, which has been suggested to be related with activated apoptosis induced by saturated fatty acids (9,10,36). In nonalcoholic fatty liver disease, endoplasmic reticulum stress associated with increased saturated fatty acids in the liver have been shown to promote liver injury and partly contribute to the disease progression from simple steatosis to steatohepatitis (9,10). Nonetheless, the potential toxic effect of increased saturated fatty acids in fibrotic livers has yet to be determined. Although there was no significant change in total unsaturated fatty acid index at week 4 as compared with week 2 after CCl4 administration, significant increase (P < .05) in total unsaturated bond index was observed (Fig 3e). These findings suggested that without a substantial increase in amount of unsaturated fatty acids, more unsaturated double bonds were formed within the unsaturated fatty acids at 4 weeks after
the start of fibrosis induction. As such, significant increase in polyunsaturated bond index (P < .05) was observed (Fig 3f) as expected. Therefore, the amount of polyunsaturated fatty acids was likely to increase at the expense of monounsaturated fatty acids in the animals with 4-week CCl4 insult. Such increase in degree of polyunsaturation has been observed and ascribed to increased necrosis/apoptosis in various experimental models (11,37,38), which was also observed in the current study (Fig 4). Because CCC has been considered to represent the important constituents in phospholipid metabolism of cell membranes (39), similar CCC levels observed in the normal and CCl4-insulted animals were largely expected in the early stage of liver fibrosis because of the similar rate of cell turnover of hepatocytes prior to the development of cirrhosis (40). In vivo phosphorus-31 (31P) MRS has been found promising in evaluating the cellular turnover and liver energy status (41–43), but it cannot measure the hepatic lipid content and is not readily available on most clinical MR imaging scanners. By contrast, in vivo 1H MRS is feasible on most standard MR 381
CHEUNG ET AL
imaging scanners, where high-resolution anatomical images and functional information such as perfusion can be acquired in the same imaging session. 1H MRS has been used to investigate focal liver lesions and found to be useful in characterizing HCC in humans (44–46) and animal models (37,47–49) by providing valuable metabolic information. However, only limited 1H MRS studies have been performed to investigate liver fibrosis in both humans and animals (16–18). Note that our study used the highest field strength to study liver fibrosis compared with previous studies (16–18). The advantages of performing 1H MRS at high magnetic field include better SNR and increased spectral resolution (15), thus providing high-quality spectra in acceptable scan times. In this regard, 1H MRS at 7 Tesla in this study can provide precise biochemical information in liver noninvasively that can be used for characterizing and monitoring of liver diseases including liver fibrosis. It is worthwhile to note that it is relatively difficult to accurately resolve and quantify several lipid signals other than (–CH2–)n peak because of limited spectral resolution and poor SNR at low field strengths in previous studies (16,18). In contrast, our current study provided more detailed information about lipid composition by resolving various lipid peaks (eg, –CH2–C=C–CH2–, =C–CH2–C= and –CH=CH–) in fibrotic livers at 7 Tesla. Note that the separation between methene (5.3 ppm) and water (4.7 ppm) peaks is 180 Hz at 7 Tesla, thus the methene proton peak was minimally affected by the water suppression RF pulses with bandwidth of 200 Hz. In addition, the effects of motion on liver MRS were largely reduced using respiratory gating in the current study, providing water-suppressed MR spectra with high quality (Fig 2). It is worth noting that although fat-selective MR imaging using Dixon methods can provide quantitative information about hepatic lipid content, these techniques are only sensitive to one specific lipid proton species (ie, methylene protons [3.4 ppm apart from water]). In contrast, 1H MRS can resolve and quantify various lipid proton species so as to provide insights into lipid composition. Ratio measurements using spectral noise as internal reference were employed for quantifying total lipid and CCC levels in the current study. Note that spectral noise could be affected by a number of factors such as local sensitivity of coil and acquisition protocol. Such factors were expected to be similar among measurements in this study because of the use of volume RF coil, identical hardware settings and acquisition protocol, yielding similar noise levels in the spectra of different measurements (Fig 2). The use of lipid peak as internal reference (16) was avoided, because the amount of lipid increased markedly in the CCl4-induced liver fibrosis model (Fig 2). Unsuppressed water signal was not chosen as internal reference either because the water content may be affected by pathological conditions (46). Previous human studies reported inconsistent observations on the lipid levels in chronic hepatitis (16,17), probably due to varying types of virus-induced hepatitis investigated. Our results showed an increase in total lipid levels, consistent with other MR 382
Academic Radiology, Vol 18, No 3, March 2011
studies in CCl4-induced fibrosis model (18,35). However, changes in saturated fatty acids, unsaturated fatty acids and unsaturated double bonds were not investigated in those studies. Our MRS data at high field allowed better discrimination and quantification of the –CH2–C=C–CH2–, –CH=CH– and =C–CH2–C= lipid peaks, thus total unsaturated fatty acid, total unsaturated bond and polyunsaturated bond indices can be measured. Moreover, the elevated total saturated fatty acid measured by 1H MRS in the current study may suggest its role in the development of liver fibrosis (9,10). One potential limitation of the present study is lack of sham controls receiving olive oil alone. As subcutaneous injection of CCl4 with olive oil is a well-established protocol to induce liver fibrosis in rodents (13), effects of olive oil were not examined in the current study. Nonetheless, previous studies showed that hepatic lipid changes are associated with metabolic alterations due to intoxication of CCl4 (50), whereas no evidences of steatosis induced by olive oil have been found. In conclusion, the lipid changes related to saturated and unsaturated fatty acids in CCl4-induced experimental fibrosis model were documented in vivo and longitudinally using 1H MRS at 7 Tesla. Our experimental results demonstrated that 1 H MRS at high-field was useful in detecting and characterizing various hepatic lipid alterations as early as 2 weeks from the start of induction of liver fibrosis in the animal model. 1H MRS may be valuable in detecting fatty changes at early phase in human liver fibrosis prior to the development of cirrhosis, and potentially useful in determining treatment strategies and evaluating therapeutic outcomes.
REFERENCES 1. Moradpour D, Wands JR. The molecular pathogenesis of hepatocellular carcinoma. J Viral Hepat 1994; 1:17–31. 2. Bataller R, Brenner DA. Liver fibrosis. J Clin Invest 2005; 115:209–218. 3. Manning DS, Afdhal NH. Diagnosis and quantitation of fibrosis. Gastroenterology 2008; 134:1670–1681. 4. Wanless IR, Nakashima E, Sherman M. Regression of human cirrhosis. Morphologic features and the genesis of incomplete septal cirrhosis. Arch Pathol Lab Med 2000; 124:1599–1607. 5. Arthur MJ. Reversibility of liver fibrosis and cirrhosis following treatment for hepatitis C. Gastroenterology 2002; 122:1525–1528. 6. Veldhuijzen IK, Toy M, Hahne SJ, et al. Screening and early treatment of migrants for chronic hepatitis B virus infection is cost-effective. Gastroenterology 2010; 138:522–530. 7. Hourigan LF, Macdonald GA, Purdie D, et al. Fibrosis in chronic hepatitis C correlates significantly with body mass index and steatosis. Hepatology 1999; 29:1215–1219. 8. Castera L, Hezode C, Roudot-Thoraval F, et al. Worsening of steatosis is an independent factor of fibrosis progression in untreated patients with chronic hepatitis C and paired liver biopsies. Gut 2003; 52:288–292. 9. Wang D, Wei Y, Pagliassotti MJ. Saturated fatty acids promote endoplasmic reticulum stress and liver injury in rats with hepatic steatosis. Endocrinology 2006; 147:943–951. 10. Gentile CL, Pagliassotti MJ. The role of fatty acids in the development and progression of nonalcoholic fatty liver disease. J Nutr Biochem 2008; 19: 567–576. 11. Corbin IR, Furth EE, Pickup S, et al. In vivo assessment of hepatic triglycerides in murine non-alcoholic fatty liver disease using magnetic resonance spectroscopy. Biochim Biophys Acta 2009; 1791:757–763. 12. Tsukamoto H, Matsuoka M, French SW. Experimental models of hepatic fibrosis: a review. Semin Liver Dis 1990; 10:56–65.
Academic Radiology, Vol 18, No 3, March 2011
13. Constandinou C, Henderson N, Iredale JP. Modeling liver fibrosis in rodents. Methods Mol Med 2005; 117:237–250. 14. Recknagel RO. Carbon tetrachloride hepatotoxicity. Pharmacol Rev 1967; 19:145–208. 15. Gruetter R, Weisdorf SA, Rajanayagan V, et al. Resolution improvements in in vivo 1H NMR spectra with increased magnetic field strength. J Magn Reson 1998; 135:260–264. 16. Cho SG, Kim MY, Kim HJ, et al. Chronic hepatitis: in vivo proton MR spectroscopic evaluation of the liver and correlation with histopathologic findings. Radiology 2001; 221:740–746. 17. Orlacchio A, Bolacchi F, Angelico M, et al. In vivo, high-field, 3-Tesla 1H MR spectroscopic assessment of liver fibrosis in HCV-correlated chronic liver disease. Radiol Med 2008; 113:289–299. 18. Hazle JD, Narayana PA, Dunsford HA. Chronic carbon tetrachloride and phospholipase D hepatotoxicity in rat: in vivo 1H magnetic resonance, total lipid analysis, and histology. Magn Reson Med 1990; 15:211–228. 19. Man K, Ng KT, Lee TK, et al. FTY720 attenuates hepatic ischemiareperfusion injury in normal and cirrhotic livers. Am J Transplant 2005; 5: 40–49. 20. Proctor E, Chatamra K. Controlled induction of cirrhosis in the rat. Br J Exp Pathol 1983; 64:320–330. 21. Armendariz-Borunda J, Seyer JM, Kang AH, et al. Regulation of TGF beta gene expression in rat liver intoxicated with carbon tetrachloride. FASEB J 1990; 4:215–221. 22. Zeisberg M, Yang C, Martino M, et al. Fibroblasts derive from hepatocytes in liver fibrosis via epithelial to mesenchymal transition. J Biol Chem 2007; 282:23337–23347. 23. Cheung JS, Guo H, Leung JC, et al. MRI visualization of rodent liver structure and peritoneal adhesion with dialyzate enhancement. Magn Reson Med 2008; 59:1170–1174. 24. Cheung JS, Fan SJ, Chow AM, et al. In vivo DTI assessment of hepatic ischemia reperfusion injury in an experimental rat model. J Magn Reson Imaging 2009; 30:890–895. 25. Zhong K, Ernst T. Localized in vivo human 1H MRS at very short echo times. Magn Reson Med 2004; 52:898–901. 26. Chan KC, Khong PL, Lau HF, et al. Late measures of microstructural alterations in severe neonatal hypoxic-ischemic encephalopathy by MR diffusion tensor imaging. Int J Dev Neurosci 2009; 27:607–615. 27. Chan KC, So KF, Wu EX. Proton magnetic resonance spectroscopy revealed choline reduction in the visual cortex in an experimental model of chronic glaucoma. Exp Eye Res 2009; 88:65–70. 28. Hayakawa Y, Yoshioka Y, Yasuda N. Effects of ligation and reperfusion of hepatic afferent vessels on the composition of liver cell membrane in the rat: 1H- and 31P-magnetic resonance spectroscopic analysis. NMR Biomed 1997; 10:257–262. 29. Lim AK, Hamilton G, Patel N, et al. 1H MR spectroscopy in the evaluation of the severity of chronic liver disease. Radiology 2003; 226:288–289. 30. Zancanaro C, Nano R, Marchioro C, et al. Magnetic resonance spectroscopy investigations of brown adipose tissue and isolated brown adipocytes. J Lipid Res 1994; 35:2191–2199. 31. Kim H, Booth CJ, Pinus AB, et al. Induced hepatic fibrosis in rats: hepatic steatosis, macromolecule content, perfusion parameters, and their correlations—preliminary MR imaging in rats. Radiology 2008; 247:696–705. 32. Wu EX, Wu Y, Nicholls JM, et al. MR diffusion tensor imaging study of postinfarct myocardium structural remodeling in a porcine model. Magn Reson Med 2007; 58:687–695.
1
H MRS LIPID PROFILING IN LIVER FIBROSIS
33. Bruckner JV, MacKenzie WF, Muralidhara S, et al. Oral toxicity of carbon tetrachloride: acute, subacute, and subchronic studies in rats. Fundam Appl Toxicol 1986; 6:16–34. 34. Hazle JD, Narayana PA, Dunsford HA. In vivo NMR, biochemical, and histologic evaluation of alcohol-induced fatty liver in rat and a comparison with CCl4 hepatotoxicity. Magn Reson Med 1991; 19:124–135. 35. Stark DD, Bass NM, Moss AA, et al. Nuclear magnetic resonance imaging of experimentally induced liver disease. Radiology 1983; 148:743–751. 36. de Vries JE, Vork MM, Roemen TH, et al. Saturated but not monounsaturated fatty acids induce apoptotic cell death in neonatal rat ventricular myocytes. J Lipid Res 1997; 38:1384–1394. 37. Foley LM, Towner RA, Painter DM. In vivo image-guided (1)H-magnetic resonance spectroscopy of the serial development of hepatocarcinogenesis in an experimental animal model. Biochim Biophys Acta 2001; 1526: 230–236. 38. Griffin JL, Lehtimaki KK, Valonen PK, et al. Assignment of 1H nuclear magnetic resonance visible polyunsaturated fatty acids in BT4C gliomas undergoing ganciclovir-thymidine kinase gene therapy-induced programmed cell death. Cancer Res 2003; 63:3195–3201. 39. Ruiz-Cabello J, Cohen JS. Phospholipid metabolites as indicators of cancer cell function. NMR Biomed 1992; 5:226–233. 40. Menon DK, Sargentoni J, Taylor-Robinson SD, et al. Effect of functional grade and etiology on in vivo hepatic phosphorus-31 magnetic resonance spectroscopy in cirrhosis: biochemical basis of spectral appearances. Hepatology 1995; 21:417–427. 41. Corbin IR, Buist R, Peeling J, et al. Hepatic 31P MRS in rat models of chronic liver disease: assessing the extent and progression of disease. Gut 2003; 52:1046–1053. 42. Lim AK, Patel N, Hamilton G, et al. The relationship of in vivo 31P MR spectroscopy to histology in chronic hepatitis C. Hepatology 2003; 37:788–794. 43. Dezortova M, Taimr P, Skoch A, et al. Etiology and functional status of liver cirrhosis by 31P MR spectroscopy. World J Gastroenterol 2005; 11: 6926–6931. 44. Soper R, Himmelreich U, Painter D, et al. Pathology of hepatocellular carcinoma and its precursors using proton magnetic resonance spectroscopy and a statistical classification strategy. Pathology 2002; 34:417–422. 45. Kuo YT, Li CW, Chen CY, et al. In vivo proton magnetic resonance spectroscopy of large focal hepatic lesions and metabolite change of hepatocellular carcinoma before and after transcatheter arterial chemoembolization using 3.0-T MR scanner. J Magn Reson Imaging 2004; 19:598–604. 46. Li CW, Kuo YC, Chen CY, et al. Quantification of choline compounds in human hepatic tumors by proton MR spectroscopy at 3 T. Magn Reson Med 2005; 53:770–776. 47. Towner RA, Foley LM, Painter DM. Hepatocarcinogenesis tumor grading correlated with in vivo image-guided 1H-NMR spectroscopy in a rat model. Toxicol Appl Pharmacol 2005; 207:237–244. 48. Zhao WD, Guan S, Zhou KR, et al. In vivo detection of metabolic changes by 1H-MRS in the DEN-induced hepatocellular carcinoma in Wistar rat. J Cancer Res Clin Oncol 2005; 131:597–602. 49. Xu H, Li X, Yang ZH, et al. In vivo 1H MR spectroscopy in the evaluation of the serial development of hepatocarcinogenesis in an experimental rat model. Acad Radiol 2006; 13:1532–1537. 50. Le Page RN, Cheeseman KH, Osman N, et al. Lipid peroxidation in purified plasma membrane fractions of rat liver in relation to the hepatoxicity of carbon tetrachloride. Cell Biochem Funct 1988; 6:87–99.
383