Radiol Phys Technol (2015) 8:146–152 DOI 10.1007/s12194-014-0303-0
Factors affecting the chemical exchange saturation transfer of Creatine as assessed by 11.7 T MRI Shigeyoshi Saito • Yuki Mori • Nobuyoshi Tanki Yoshichika Yoshioka • Kenya Murase
•
Received: 13 May 2014 / Revised: 26 November 2014 / Accepted: 27 November 2014 / Published online: 5 December 2014 Ó Japanese Society of Radiological Technology and Japan Society of Medical Physics 2014
Abstract Chemical exchange saturation transfer (CEST) is a new contrast enhancement approach for imaging exogenous or endogenous substances such as creatine (Cr), amide protons, and glutamate in the human body. An increase in field strength is beneficial for CEST imaging because of the increased chemical shift and longer longitudinal relaxation time (T1). In high-field magnetic resonance imaging (MRI), establishing and evaluating the CEST effect is important for optimizing the magnetization transfer (MT) saturation radio frequency (RF) pulses. In this study, the CEST effect on Cr was evaluated at different concentrations in pH phantoms by appropriately selecting MT saturation RF pulses using 11.7 T MRI. The results showed that the CEST efficiency increased gradually with increasing applied saturation RF pulse power and that it
S. Saito (&) K. Murase Department of Medical Physics and Engineering, Division of Medical Technology and Science, Faculty of Health Science, Graduate School of Medicine, Osaka University, 1-7 Yamadaoka, Suita, Osaka 565-0871, Japan e-mail: [email protected] S. Saito Y. Mori Y. Yoshioka K. Murase Center for Information and Neural Networks (CiNet), National Institute of Information and Communications Technology and Osaka University, 1-4 Yamadaoka, Suita, Osaka 565-0871, Japan Y. Mori Y. Yoshioka Biofunctional Imaging Laboratory, Immol/Lunology Frontier Research Center (WPI-IFReC), Osaka University, 3-1 Yamadaoka, Suita, Osaka 565-0871, Japan N. Tanki RIKEN Center for Life Science Technologies (CLST), 6-7-3Minatojima-minamimachi, Chuo-ku, Kobe, Hyogo 650-0047, Japan
was affected by the number of saturation RF pulses and their bandwidths. However, spillover effects were observed with higher saturation RF pulse powers. In conclusion, we successfully performed in vitro Cr CEST imaging under optimized conditions of MT saturation RF pulses. Keywords Chemical exchange saturation transfer Creatine Magnetization transfer Spillover effect Abbreviations CEST Chemical exchange saturation transfer Cr Creatine MRI Magnetic resonance imaging MRS Magnetic resonance spectroscopy MT Magnetization transfer MTR Magnetization transfer ratio RARE Rapid acquisition with relaxation enhancement RF Radio frequency T1 Longitudinal relaxation time T2W T2-weighted images
1 Introduction The development of magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS) over recent decades has established these approaches as powerful imaging techniques for investigating morphologies and bioenergetics in the biomedical field. In the human body, creatine (Cr) is produced primarily by amino acids in the kidney and liver. It is then transported to the muscles by the circulating blood. Cr has been shown to be localized primarily to skeletal muscle [1]. Cr is also used as a marker of normal energy metabolism in the brain, and Cr levels
Factors affecting the CEST of Creatine
147
remain relatively unchanged in most disease states. Although Cr is considered to be fairly stable under most conditions, some studies have questioned its stability in tumors, hypoxia, and other situations [2]. Chemical exchange saturation transfer (CEST) imaging is a new MRI contrast enhancement approach for obtaining images of exogenous or endogenous compounds containing either exchangeable protons or exchangeable molecules [3]. The CEST technique is similar to a traditional magnetization transfer (MT) pulse sequence, but it is longer and uses low-power pulses with minimal effect on MT. Several CEST-MRI approaches have recently been investigated to monitor levels of cellular metabolites such as Cr [4], amide protons [5], glutamate [6], glucose [7], and paramagnetic exogenous agents [8] in vivo. In addition, some exchange processes have been shown to be distinctly pH sensitive and may allow for the development of pHweighted MRI [5]. CEST requires two pools of protons to be sufficiently separated according to their frequency. The exchange rate must be fast enough to be detectable before the saturated magnetization recovers with a longitudinal relaxation time (T1) [9]. Therefore, an increase in field strength is beneficial for CEST because the chemical shift between the proton pools as well as the T1 of water will increase with increasing field strength. CEST imaging of Cr CEST at 9.4 T has been reported [4]. However, the saturation radio frequency (RF) pulses for a combination of 11.7 T MRI and Cr have not yet been optimized. Therefore, establishing and evaluating the CEST effect is important for optimizing the saturation RF pulse parameters in ultra-highfield MRI. The purpose of this study was to establish a protocol for in vitro Cr CEST imaging at 11.7 T by optimizing the methodology for saturation RF pulse irradiation.
with a volume RF coil (m2m Imaging Corp., Cleveland, OH, USA) having an inner diameter of 25 mm for transmission and reception, using the Paravision 5.1 console system (Bruker Biospin). For B0 shimming, first- and second-order shimming for the selected volume were conducted using the fast automatic shimming technique by mapping along projections [10]. Single-slice multiple-MT MRI was performed using rapid acquisition with relaxation enhancement (RARE) for CEST imaging with the following parameters: repetition time = 5,000 ms; echo time = 22 ms; matrix size = 128 9 128; field of view = 19.2 9 19.2 mm2; slice thickness = 2.0 mm; RARE factor = 4; and number of average = 1. Z spectra were acquired from CEST images with varying saturation frequencies from -4.0 ppm to ?4.0 ppm in 0.1-ppm steps. The nominal voxel resolution was 150 lm 9 150 lm 9 2000 lm for these images. S0 images (without saturation RF pulses) were acquired before CEST images. CEST imaging was performed in three steps for optimization of the saturation RF pulse conditions. First, the bandwidth and number of pulses were set to 100 Hz and 125 times, respectively, to examine different saturation RF pulse powers (2, 4, 6, and 8 lT). Second, the saturation RF pulse power and number of pulses were set to 4 lT and 125 times, respectively, to examine saturation RF pulses of different bandwidths (75, 100, 125, and 150 Hz).Finally, the saturation RF pulse power and bandwidth were set to 4 lT and 100 Hz, respectively, to examine different saturation RF pulse numbers (100, 125, 150, and 175 times). Gaussian-shaped off-resonant saturation RF pulses were used in all CEST experiments. Z spectra were acquired from 42 T2-weighted images (T2W) with no saturation (S0) and with saturation frequencies varying from -4.0 (-Dx) ppm to ?4.0 (?Dx) ppm, in 0.1-ppm steps. The total acquisition time for one set of CEST images was approximately 2 h.
2 Materials and methods
2.3 MTR calculation
2.1 Sample preparation
Maps of the asymmetric MTR {MTR asymmetry (%) = [S(-Dx) - S(?Dx)]/S0} were created. S(-Dx) and S(?Dx) were the acquired MR signals in Cr when the saturation was -1.8 ppm and ?1.8 ppm, respectively. The regions of interest were drawn on areas that covered 80 % of these samples. All calculations and analyses were performed using ImageJ (Ver. 1.40g, National Institutes of Health, Bethesda, MD, USA) and MATLAB (MathWorks, Natick, MA, USA). The MTR asymmetries (%) in all groups were expressed as mean ± standard deviation (SD). All statistical analyses were performed using Prism 5 (Version 5, GraphPad Software, California, USA). P \ 0.05 was considered to indicate a significant difference.
To explore the presence of exchangeable protons in Cr (Sigma-Aldrich, Missouri, USA), 10-, 50-, and 100-mmol/ L(mM) phantoms were prepared in phosphate-buffered saline (PBS, Sigma-Aldrich). Sample pH values were adjusted to 4.0 and 7.0 with HCl in 50-mmol/L Cr samples. All samples were maintained at room temperature (23 °C) during MRI scanning. 2.2 MRI experiments All MRI experiments were performed on an 11.7 T MRI scanner (Bruker Biospin, Ettlingen, Germany) equipped
148
3 Results Figure 1a shows a horizontal view of the positions of the Cr samples and PBS. Figure 1b and c shows the typical S0, two MT images, and asymmetry maps. The CEST effect on Cr gradually increased from 10 to 100 mmol/L (Fig. 1c). It also increased from pH 4.0 to pH 7.0 at the same 50-mmol/ LCr concentration (Fig. 1c). Figure 2a and b shows typical Z spectra at different Cr concentrations and pH values. The CEST effect on Cr increased from 10 mmol/L to 100 mmol/Lat pH 7.0 (Fig. 2a). MTR asymmetry was not observed in PBS (Fig. 2a) or 50-mmol/LCr samples at pH 4.0 (Fig. 2b). However, the CEST effect on the Z spectrum of Cr increased from pH 4.0 to pH 7.0 at the same 50-mmol/LCr concentration. The Cr Z spectrum showed a broad CEST peak centered *1.8 ppm downfield of the bulk water resonance (Fig. 2a). Figure 3 shows asymmetry color maps obtained under different saturation RF pulse conditions. The CEST effect increased with the RF pulse power from 2 to 8 lT (Fig. 3, top). The CEST effect was unaffected by the bandwidth and number of saturation RF pulses (Fig. 3, middle and
Fig. 1 Typical MT T2W and asymmetry images. a Horizontal view of the positions of the Cr samples. b Typical MT T2W images (S0, 1.8 ppm, -1.8 ppm, with and without saturation RF pulses). c Typical asymmetry maps of the 5 samples. The CEST effect from Cr increased from 10 to 100 mmol/L (mM). The CEST effect from Cr increased from pH 4.0 to pH 7.0 at the same Cr concentration. Saturation RF pulse power = 4 lT, number of saturation RF pulses = 125 times and bandwidth = 100 Hz
S. Saito et al.
bottom). Figure 4 shows the MTR asymmetry under various saturation RF pulse conditions. The MTR asymmetry increased as the power of the saturation RF pulses increased from 2 to 8 lT (Fig. 4a). In addition, CEST imaging was affected by the bandwidth of the pulse (Fig. 4b). However, CEST imaging was unaffected at saturation RF pulse numbers of 100–175 because the RF duration was sufficiently long (Fig. 4c). Figure 5 shows the spillover effect at various saturation RF pulse powers. Spillover effects were observed at higher saturation RF pulse powers from 2 to 8 lT (Fig. 5a). The RF pulse saturation effects at 8 lT covered a broader range than those at 2 lT (Fig. 5b).
4 Discussion This study evaluated the CEST effect in vitro on Cr at different concentrations in pH phantoms by appropriately selecting MT saturation RF pulses using ultra-high-field 11.7 T MRI. Cr contains amine protons that exhibit a CEST effect at approximately 1.8 ppm and physiological pH (7.0). Wellresolved amine proton resonances from Cr at 1.8 ppm were visible in the Z spectrum with 11.7 T MRI (Fig. 2a). The ionic form of Cr contained two groups of amine protons that showed single resonance at 1.8 ppm. Physiological pH was important in detecting the CEST effect. MTR asymmetry was not observed in PBS (Fig. 2a) or 50-mmol/L Cr samples at pH 4.0 (Fig. 2b). Therefore, it is also important to select the appropriate MT saturation pulse parameters in pH-dependent CEST-MRI. The measurable CEST contrast signal was previously shown to depend on experimental parameters such as the magnetic field strength, saturation RF pulse irradiation power, number of saturation RF pulses, saturation RF pulse duration, and irradiation scheme [9]. This CEST contrast may be apparent in ultra-high-field 11.7 T MRI imaging because of the increase in the signal-to-noise ratio (SNR) and the use of smaller voxels, which reduce partial volume effects. In addition, the CEST effect is expected to increase in higher field 11.7T MRI because the T1 values increase with increasing field strength, allowing prolonged storage of saturation in the water pool. The labile amine group was only 1.8 ppm from the bulk water resonance, which was equivalent to 116 Hz in 1.5 T MRI, 230 Hz in 3 T MRI, and 860 Hz in 11.7 T MRI in the Cr CEST study. Along with the SNR advantages of higher fields, adequate spectral dispersion is critical to achieve selective irradiation of the amide protons. Therefore, the main advantage at high fields is the frequency separation in terms of better adherence to the slow-exchange condition and reduced interference of direct water saturation.
Factors affecting the CEST of Creatine Fig. 2 Typical Z spectrum at various Cr conditions. a Typical Z spectrum of PBS, 10, 50, 100 mmol/L Cr samples at pH 7.0 (saturation RF pulse power = 4 lT, number of pulses = 125 times, bandwidth = 100 Hz). The Cr Z spectrum exhibited a broad CEST peak centered at *1.8 ppm. b Typical Z spectrum of Cr 50 mmol/L (mM) samples at pH 4.0 and 7.0. The CEST effect from Cr increased from pH 4.0 to pH 7.0. The left scale shows MTR (left arrow), and the right scale (right arrow) shows MTR asymmetry (%) in both the graphs
Fig. 3 Asymmetry maps (%) of various saturation RF pulse parameters. Top asymmetry maps of the four different saturation RF pulse powers (2, 4, 6, and 8 lT). Middle asymmetry maps of the four different saturation RF pulses bandwidths (75, 100, 125 and 150 Hz). Bottom asymmetry maps for the four different numbers of saturation RF pulses (100, 125, 150 and 175 times)
149
150
S. Saito et al.
Fig. 4 MTR asymmetry (%) of various saturation RF pulse parameters. a MTR asymmetry (%) of saturation RF pulse powers (2, 4, 6, and 8 lT). b MTR asymmetry (%) of the different saturation RF
pulses bandwidths (75, 100, 125 and 150 Hz). c MTR asymmetry (%) of the different numbers of saturation RF pulses (100, 125, 150 and 175 times)
Saturation RF pulse irradiation may not only saturate labile protons but also directly attenuate the bulk water signal in the so-called spillover effect, thereby reducing the sensitivity and specificity of CEST-MRI [11]. Endogenous CEST effects are diluted by competing effects such as the spillover effect and macromolecular MT. A higher saturation RF pulse can be used for more efficient saturation of exchangeable protons. The longer T1 at a higher field also contributed to the higher CEST contrast measured. This is not the only benefit observed when ultra-high-field MRI is used for CEST imaging. Dielectric resonance effects were previously shown to become pronounced and cause RF field inhomogeneity in ultra-high-field MRI applications. CEST imaging is susceptible to magnetic field inhomogeneities of B0 and B1, which may need to be corrected in images obtained by high-field scanners with unsatisfactory B0 and B1 homogeneity. B0 inhomogeneity arises from imperfections of the field-generating magnet and locally varying magnetic fields induced by an object placed in a static magnetic field, whereas B1 inhomogeneity can arise from RF nonuniformity [10] and the prescan procedure for adjusting the transmitter amplitude for each subject. In our study, we used small, homogeneous in vitro Cr samples to observe the CEST effect and map shimming and
manual shimming for B0 and B1 compensation before and after acquisition of CEST data. B0 and B1 correction might be very useful in compensating for field-inhomogeneityinduced measurement errors in CEST imaging [9, 11]. Recently, various methods have been proposed to correct B0 and B1 inhomogeneity [12, 13]. These shimming effects improve the quantitation of the obtained CEST spectrum. Further experiments are necessary to assess the stability and reproducibility of Cr CEST. The efficiency of MT increased gradually with increasing RF power applied in vitro (Fig. 4a). A prominent CEST peak was observed at all saturation RF pulse powers. The efficiency of RF pulse saturation was limited, and reductions in the water signal were small when the irradiation power was low (Fig. 4). Consistent with previous reports [11], the saturation transfer effect initially increased with increasing power of the irradiation pulses. However, the CEST contrast was shown to be significantly attenuated by concomitant spillover effects at high RF powers [11]. In our study, the spillover effect was observed at a higher saturation RF pulse power of 8 lT (Fig. 5a). Both spillover and MT effects were shown to be augmented with increasing saturation RF pulse irradiation power [14].
Factors affecting the CEST of Creatine
151
Fig. 5 Spillover effect at various saturation RF pulse powers. a Typical Z spectrum at approximately 1.8 ppm (2.8–0.8 ppm, 100 mmol/L Cr, MT saturation RF pulse powers = 2 and 8 lT). b Typical MT T2W images (2.2–1.4 ppm, number of saturation RF pulses = 125 times, saturation RF bandwidth = 100 Hz)
An adverse artifact occurred in CEST images because of direct saturation of the water pool, which strongly influenced the evaluation of the CEST data. Thus, CEST sequences are needed to optimize variations in MT to yield maximum contrast [6]; however, the signal at the optimal saturation RF pulse was highly sensitive to spillover and required a correction in this case. Therefore, although the use of a high-power saturation RF pulse is sufficient to obtain a higher CEST effect, as it reduces the spillover effect by optimizing B1, the saturation RF pulse parameters must also be considered. In conclusion, we successfully performed in vitro Cr CEST imaging under optimized MT saturation RF pulses using 11.7T MRI. Acknowledgments The authors would like to thank Mr. Junpei Ueda, Mr. Hisato Sasahara and Mr. Isamu Yabata (Graduate School of Medicine, Osaka University, Japan) for technical assistance. This work was partly supported by Grants-in-Aid for Scientific Research (Kakenhi, Nos. 24791299 and 24300167) by the Japan Society for the Promotion of Science (JSPS). Conflict of interest The authors declare that they have no conflicts of interest.
References 1. Velan SS, Durst C, Lemieux SK, Raylman RR, Sridhar R, Spencer RG, et al. Investigation of muscle lipid metabolism by localized one- and two-dimensional MRS techniques using a clinical 3T MRI/MRS scanner. J Magn Reson Imaging. 2007;25:192–9. doi:10.1002/jmri.20786. 2. Usenius T, Usenius JP, Tenhunen M, Vainio P, Johansson R, Soimakallio S, et al. Radiation-induced changes in human brain metabolites as studied by 1H nuclear magnetic resonance spectroscopy in vivo. Int J Radiat Oncol Biol Phys. 1995;33:719–24. 3. Ward KM, Aletras AH, Balaban RS. A new class of contrast agents for MRI based on proton chemical exchange dependent saturation transfer (CEST). J Magn Reson. 2000;143:79–87. doi:10.1006/jmre.1999.1956. 4. Haris M, Nanga RP, Singh A, Cai K, Kogan F, Hariharan H, et al. Exchange rates of creatine kinase metabolites: feasibility of imaging creatine by chemical exchange saturation transfer MRI. NMR Biomed. 2012;25:1305–9. doi:10.1002/nbm.2792. 5. Zhou J, Payen JF, Wilson DA, Traystman RJ, van Zijl PC. Using the amide proton signals of intracellular proteins and peptides to detect pH effects in MRI. Nat Med. 2003;9:1085–90. doi:10. 1038/nm907. 6. Cai K, Haris M, Singh A, Kogan F, Greenberg JH, Hariharan H, et al. Magnetic resonance imaging of glutamate. Nat Med. 2012;18:302–6. doi:10.1038/nm.2615. 7. Chan KW, McMahon MT, Kato Y, Liu G, Bulte JW, Bhujwalla ZM, et al. Natural D-glucose as a biodegradable MRI contrast
152 agent for detecting cancer. Magn Reson Med. 2012;68:1764–73. doi:10.1002/mrm.24520. 8. Terreno E, Castelli DD, Aime S. Encoding the frequency dependence in MRI contrast media: the emerging class of CEST agents. Contrast Media Mol Imaging. 2010;5:78–98. doi:10.1002/ cmmi.369. 9. Sun PZ, Benner T, Kumar A, Sorensen AG. Investigation of optimizing and translating pH-sensitive pulsed-chemical exchange saturation transfer (CEST) imaging to a 3T clinical scanner. Magn Reson Med. 2008;60:834–41. doi:10.1002/mrm. 21714. 10. Barker GJ, Simmons A, Arridge SR, Tofts PS. A simple method for investigating the effects of non-uniformity of radiofrequency transmission and radiofrequency reception in MRI. Br J Radiol. 1998;71:59–67. doi:10.1259/bjr.71.841.9534700.
S. Saito et al. 11. Sun PZ, van Zijl PC, Zhou J. Optimization of the irradiation power in chemical exchange dependent saturation transfer experiments. J Magn Reson. 2005;175:193–200. doi:10.1016/j. jmr.2005.04.005. 12. Volz S, Noth U, Rotarska-Jagiela A, Deichmann R. A fast B1mapping method for the correction and normalization of magnetization transfer ratio maps at 3 T. Neuroimage. 2010;49:3015–26. doi:10.1016/j.neuroimage.2009.11.054. 13. Singh A, Cai K, Haris M, Hariharan H, Reddy R. On B1 inhomogeneity correction of in vivo human brain glutamate chemical exchange saturation transfer contrast at 7T. Magn Reson Med. 2013;69:818–24. doi:10.1002/mrm.24290. 14. Desmond KL, Stanisz GJ. Understanding quantitative pulsed CEST in the presence of MT. Magn Reson Med. 2012;67:979–90. doi:10.1002/mrm.23074.
Factors affecting the chemical exchange saturation transfer of Creatine as assessed by 11.7 T MRI Shigeyoshi Saito • Yuki Mori • Nobuyoshi Tanki Yoshichika Yoshioka • Kenya Murase
•
Received: 13 May 2014 / Revised: 26 November 2014 / Accepted: 27 November 2014 / Published online: 5 December 2014 Ó Japanese Society of Radiological Technology and Japan Society of Medical Physics 2014
Abstract Chemical exchange saturation transfer (CEST) is a new contrast enhancement approach for imaging exogenous or endogenous substances such as creatine (Cr), amide protons, and glutamate in the human body. An increase in field strength is beneficial for CEST imaging because of the increased chemical shift and longer longitudinal relaxation time (T1). In high-field magnetic resonance imaging (MRI), establishing and evaluating the CEST effect is important for optimizing the magnetization transfer (MT) saturation radio frequency (RF) pulses. In this study, the CEST effect on Cr was evaluated at different concentrations in pH phantoms by appropriately selecting MT saturation RF pulses using 11.7 T MRI. The results showed that the CEST efficiency increased gradually with increasing applied saturation RF pulse power and that it
S. Saito (&) K. Murase Department of Medical Physics and Engineering, Division of Medical Technology and Science, Faculty of Health Science, Graduate School of Medicine, Osaka University, 1-7 Yamadaoka, Suita, Osaka 565-0871, Japan e-mail: [email protected] S. Saito Y. Mori Y. Yoshioka K. Murase Center for Information and Neural Networks (CiNet), National Institute of Information and Communications Technology and Osaka University, 1-4 Yamadaoka, Suita, Osaka 565-0871, Japan Y. Mori Y. Yoshioka Biofunctional Imaging Laboratory, Immol/Lunology Frontier Research Center (WPI-IFReC), Osaka University, 3-1 Yamadaoka, Suita, Osaka 565-0871, Japan N. Tanki RIKEN Center for Life Science Technologies (CLST), 6-7-3Minatojima-minamimachi, Chuo-ku, Kobe, Hyogo 650-0047, Japan
was affected by the number of saturation RF pulses and their bandwidths. However, spillover effects were observed with higher saturation RF pulse powers. In conclusion, we successfully performed in vitro Cr CEST imaging under optimized conditions of MT saturation RF pulses. Keywords Chemical exchange saturation transfer Creatine Magnetization transfer Spillover effect Abbreviations CEST Chemical exchange saturation transfer Cr Creatine MRI Magnetic resonance imaging MRS Magnetic resonance spectroscopy MT Magnetization transfer MTR Magnetization transfer ratio RARE Rapid acquisition with relaxation enhancement RF Radio frequency T1 Longitudinal relaxation time T2W T2-weighted images
1 Introduction The development of magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS) over recent decades has established these approaches as powerful imaging techniques for investigating morphologies and bioenergetics in the biomedical field. In the human body, creatine (Cr) is produced primarily by amino acids in the kidney and liver. It is then transported to the muscles by the circulating blood. Cr has been shown to be localized primarily to skeletal muscle [1]. Cr is also used as a marker of normal energy metabolism in the brain, and Cr levels
Factors affecting the CEST of Creatine
147
remain relatively unchanged in most disease states. Although Cr is considered to be fairly stable under most conditions, some studies have questioned its stability in tumors, hypoxia, and other situations [2]. Chemical exchange saturation transfer (CEST) imaging is a new MRI contrast enhancement approach for obtaining images of exogenous or endogenous compounds containing either exchangeable protons or exchangeable molecules [3]. The CEST technique is similar to a traditional magnetization transfer (MT) pulse sequence, but it is longer and uses low-power pulses with minimal effect on MT. Several CEST-MRI approaches have recently been investigated to monitor levels of cellular metabolites such as Cr [4], amide protons [5], glutamate [6], glucose [7], and paramagnetic exogenous agents [8] in vivo. In addition, some exchange processes have been shown to be distinctly pH sensitive and may allow for the development of pHweighted MRI [5]. CEST requires two pools of protons to be sufficiently separated according to their frequency. The exchange rate must be fast enough to be detectable before the saturated magnetization recovers with a longitudinal relaxation time (T1) [9]. Therefore, an increase in field strength is beneficial for CEST because the chemical shift between the proton pools as well as the T1 of water will increase with increasing field strength. CEST imaging of Cr CEST at 9.4 T has been reported [4]. However, the saturation radio frequency (RF) pulses for a combination of 11.7 T MRI and Cr have not yet been optimized. Therefore, establishing and evaluating the CEST effect is important for optimizing the saturation RF pulse parameters in ultra-highfield MRI. The purpose of this study was to establish a protocol for in vitro Cr CEST imaging at 11.7 T by optimizing the methodology for saturation RF pulse irradiation.
with a volume RF coil (m2m Imaging Corp., Cleveland, OH, USA) having an inner diameter of 25 mm for transmission and reception, using the Paravision 5.1 console system (Bruker Biospin). For B0 shimming, first- and second-order shimming for the selected volume were conducted using the fast automatic shimming technique by mapping along projections [10]. Single-slice multiple-MT MRI was performed using rapid acquisition with relaxation enhancement (RARE) for CEST imaging with the following parameters: repetition time = 5,000 ms; echo time = 22 ms; matrix size = 128 9 128; field of view = 19.2 9 19.2 mm2; slice thickness = 2.0 mm; RARE factor = 4; and number of average = 1. Z spectra were acquired from CEST images with varying saturation frequencies from -4.0 ppm to ?4.0 ppm in 0.1-ppm steps. The nominal voxel resolution was 150 lm 9 150 lm 9 2000 lm for these images. S0 images (without saturation RF pulses) were acquired before CEST images. CEST imaging was performed in three steps for optimization of the saturation RF pulse conditions. First, the bandwidth and number of pulses were set to 100 Hz and 125 times, respectively, to examine different saturation RF pulse powers (2, 4, 6, and 8 lT). Second, the saturation RF pulse power and number of pulses were set to 4 lT and 125 times, respectively, to examine saturation RF pulses of different bandwidths (75, 100, 125, and 150 Hz).Finally, the saturation RF pulse power and bandwidth were set to 4 lT and 100 Hz, respectively, to examine different saturation RF pulse numbers (100, 125, 150, and 175 times). Gaussian-shaped off-resonant saturation RF pulses were used in all CEST experiments. Z spectra were acquired from 42 T2-weighted images (T2W) with no saturation (S0) and with saturation frequencies varying from -4.0 (-Dx) ppm to ?4.0 (?Dx) ppm, in 0.1-ppm steps. The total acquisition time for one set of CEST images was approximately 2 h.
2 Materials and methods
2.3 MTR calculation
2.1 Sample preparation
Maps of the asymmetric MTR {MTR asymmetry (%) = [S(-Dx) - S(?Dx)]/S0} were created. S(-Dx) and S(?Dx) were the acquired MR signals in Cr when the saturation was -1.8 ppm and ?1.8 ppm, respectively. The regions of interest were drawn on areas that covered 80 % of these samples. All calculations and analyses were performed using ImageJ (Ver. 1.40g, National Institutes of Health, Bethesda, MD, USA) and MATLAB (MathWorks, Natick, MA, USA). The MTR asymmetries (%) in all groups were expressed as mean ± standard deviation (SD). All statistical analyses were performed using Prism 5 (Version 5, GraphPad Software, California, USA). P \ 0.05 was considered to indicate a significant difference.
To explore the presence of exchangeable protons in Cr (Sigma-Aldrich, Missouri, USA), 10-, 50-, and 100-mmol/ L(mM) phantoms were prepared in phosphate-buffered saline (PBS, Sigma-Aldrich). Sample pH values were adjusted to 4.0 and 7.0 with HCl in 50-mmol/L Cr samples. All samples were maintained at room temperature (23 °C) during MRI scanning. 2.2 MRI experiments All MRI experiments were performed on an 11.7 T MRI scanner (Bruker Biospin, Ettlingen, Germany) equipped
148
3 Results Figure 1a shows a horizontal view of the positions of the Cr samples and PBS. Figure 1b and c shows the typical S0, two MT images, and asymmetry maps. The CEST effect on Cr gradually increased from 10 to 100 mmol/L (Fig. 1c). It also increased from pH 4.0 to pH 7.0 at the same 50-mmol/ LCr concentration (Fig. 1c). Figure 2a and b shows typical Z spectra at different Cr concentrations and pH values. The CEST effect on Cr increased from 10 mmol/L to 100 mmol/Lat pH 7.0 (Fig. 2a). MTR asymmetry was not observed in PBS (Fig. 2a) or 50-mmol/LCr samples at pH 4.0 (Fig. 2b). However, the CEST effect on the Z spectrum of Cr increased from pH 4.0 to pH 7.0 at the same 50-mmol/LCr concentration. The Cr Z spectrum showed a broad CEST peak centered *1.8 ppm downfield of the bulk water resonance (Fig. 2a). Figure 3 shows asymmetry color maps obtained under different saturation RF pulse conditions. The CEST effect increased with the RF pulse power from 2 to 8 lT (Fig. 3, top). The CEST effect was unaffected by the bandwidth and number of saturation RF pulses (Fig. 3, middle and
Fig. 1 Typical MT T2W and asymmetry images. a Horizontal view of the positions of the Cr samples. b Typical MT T2W images (S0, 1.8 ppm, -1.8 ppm, with and without saturation RF pulses). c Typical asymmetry maps of the 5 samples. The CEST effect from Cr increased from 10 to 100 mmol/L (mM). The CEST effect from Cr increased from pH 4.0 to pH 7.0 at the same Cr concentration. Saturation RF pulse power = 4 lT, number of saturation RF pulses = 125 times and bandwidth = 100 Hz
S. Saito et al.
bottom). Figure 4 shows the MTR asymmetry under various saturation RF pulse conditions. The MTR asymmetry increased as the power of the saturation RF pulses increased from 2 to 8 lT (Fig. 4a). In addition, CEST imaging was affected by the bandwidth of the pulse (Fig. 4b). However, CEST imaging was unaffected at saturation RF pulse numbers of 100–175 because the RF duration was sufficiently long (Fig. 4c). Figure 5 shows the spillover effect at various saturation RF pulse powers. Spillover effects were observed at higher saturation RF pulse powers from 2 to 8 lT (Fig. 5a). The RF pulse saturation effects at 8 lT covered a broader range than those at 2 lT (Fig. 5b).
4 Discussion This study evaluated the CEST effect in vitro on Cr at different concentrations in pH phantoms by appropriately selecting MT saturation RF pulses using ultra-high-field 11.7 T MRI. Cr contains amine protons that exhibit a CEST effect at approximately 1.8 ppm and physiological pH (7.0). Wellresolved amine proton resonances from Cr at 1.8 ppm were visible in the Z spectrum with 11.7 T MRI (Fig. 2a). The ionic form of Cr contained two groups of amine protons that showed single resonance at 1.8 ppm. Physiological pH was important in detecting the CEST effect. MTR asymmetry was not observed in PBS (Fig. 2a) or 50-mmol/L Cr samples at pH 4.0 (Fig. 2b). Therefore, it is also important to select the appropriate MT saturation pulse parameters in pH-dependent CEST-MRI. The measurable CEST contrast signal was previously shown to depend on experimental parameters such as the magnetic field strength, saturation RF pulse irradiation power, number of saturation RF pulses, saturation RF pulse duration, and irradiation scheme [9]. This CEST contrast may be apparent in ultra-high-field 11.7 T MRI imaging because of the increase in the signal-to-noise ratio (SNR) and the use of smaller voxels, which reduce partial volume effects. In addition, the CEST effect is expected to increase in higher field 11.7T MRI because the T1 values increase with increasing field strength, allowing prolonged storage of saturation in the water pool. The labile amine group was only 1.8 ppm from the bulk water resonance, which was equivalent to 116 Hz in 1.5 T MRI, 230 Hz in 3 T MRI, and 860 Hz in 11.7 T MRI in the Cr CEST study. Along with the SNR advantages of higher fields, adequate spectral dispersion is critical to achieve selective irradiation of the amide protons. Therefore, the main advantage at high fields is the frequency separation in terms of better adherence to the slow-exchange condition and reduced interference of direct water saturation.
Factors affecting the CEST of Creatine Fig. 2 Typical Z spectrum at various Cr conditions. a Typical Z spectrum of PBS, 10, 50, 100 mmol/L Cr samples at pH 7.0 (saturation RF pulse power = 4 lT, number of pulses = 125 times, bandwidth = 100 Hz). The Cr Z spectrum exhibited a broad CEST peak centered at *1.8 ppm. b Typical Z spectrum of Cr 50 mmol/L (mM) samples at pH 4.0 and 7.0. The CEST effect from Cr increased from pH 4.0 to pH 7.0. The left scale shows MTR (left arrow), and the right scale (right arrow) shows MTR asymmetry (%) in both the graphs
Fig. 3 Asymmetry maps (%) of various saturation RF pulse parameters. Top asymmetry maps of the four different saturation RF pulse powers (2, 4, 6, and 8 lT). Middle asymmetry maps of the four different saturation RF pulses bandwidths (75, 100, 125 and 150 Hz). Bottom asymmetry maps for the four different numbers of saturation RF pulses (100, 125, 150 and 175 times)
149
150
S. Saito et al.
Fig. 4 MTR asymmetry (%) of various saturation RF pulse parameters. a MTR asymmetry (%) of saturation RF pulse powers (2, 4, 6, and 8 lT). b MTR asymmetry (%) of the different saturation RF
pulses bandwidths (75, 100, 125 and 150 Hz). c MTR asymmetry (%) of the different numbers of saturation RF pulses (100, 125, 150 and 175 times)
Saturation RF pulse irradiation may not only saturate labile protons but also directly attenuate the bulk water signal in the so-called spillover effect, thereby reducing the sensitivity and specificity of CEST-MRI [11]. Endogenous CEST effects are diluted by competing effects such as the spillover effect and macromolecular MT. A higher saturation RF pulse can be used for more efficient saturation of exchangeable protons. The longer T1 at a higher field also contributed to the higher CEST contrast measured. This is not the only benefit observed when ultra-high-field MRI is used for CEST imaging. Dielectric resonance effects were previously shown to become pronounced and cause RF field inhomogeneity in ultra-high-field MRI applications. CEST imaging is susceptible to magnetic field inhomogeneities of B0 and B1, which may need to be corrected in images obtained by high-field scanners with unsatisfactory B0 and B1 homogeneity. B0 inhomogeneity arises from imperfections of the field-generating magnet and locally varying magnetic fields induced by an object placed in a static magnetic field, whereas B1 inhomogeneity can arise from RF nonuniformity [10] and the prescan procedure for adjusting the transmitter amplitude for each subject. In our study, we used small, homogeneous in vitro Cr samples to observe the CEST effect and map shimming and
manual shimming for B0 and B1 compensation before and after acquisition of CEST data. B0 and B1 correction might be very useful in compensating for field-inhomogeneityinduced measurement errors in CEST imaging [9, 11]. Recently, various methods have been proposed to correct B0 and B1 inhomogeneity [12, 13]. These shimming effects improve the quantitation of the obtained CEST spectrum. Further experiments are necessary to assess the stability and reproducibility of Cr CEST. The efficiency of MT increased gradually with increasing RF power applied in vitro (Fig. 4a). A prominent CEST peak was observed at all saturation RF pulse powers. The efficiency of RF pulse saturation was limited, and reductions in the water signal were small when the irradiation power was low (Fig. 4). Consistent with previous reports [11], the saturation transfer effect initially increased with increasing power of the irradiation pulses. However, the CEST contrast was shown to be significantly attenuated by concomitant spillover effects at high RF powers [11]. In our study, the spillover effect was observed at a higher saturation RF pulse power of 8 lT (Fig. 5a). Both spillover and MT effects were shown to be augmented with increasing saturation RF pulse irradiation power [14].
Factors affecting the CEST of Creatine
151
Fig. 5 Spillover effect at various saturation RF pulse powers. a Typical Z spectrum at approximately 1.8 ppm (2.8–0.8 ppm, 100 mmol/L Cr, MT saturation RF pulse powers = 2 and 8 lT). b Typical MT T2W images (2.2–1.4 ppm, number of saturation RF pulses = 125 times, saturation RF bandwidth = 100 Hz)
An adverse artifact occurred in CEST images because of direct saturation of the water pool, which strongly influenced the evaluation of the CEST data. Thus, CEST sequences are needed to optimize variations in MT to yield maximum contrast [6]; however, the signal at the optimal saturation RF pulse was highly sensitive to spillover and required a correction in this case. Therefore, although the use of a high-power saturation RF pulse is sufficient to obtain a higher CEST effect, as it reduces the spillover effect by optimizing B1, the saturation RF pulse parameters must also be considered. In conclusion, we successfully performed in vitro Cr CEST imaging under optimized MT saturation RF pulses using 11.7T MRI. Acknowledgments The authors would like to thank Mr. Junpei Ueda, Mr. Hisato Sasahara and Mr. Isamu Yabata (Graduate School of Medicine, Osaka University, Japan) for technical assistance. This work was partly supported by Grants-in-Aid for Scientific Research (Kakenhi, Nos. 24791299 and 24300167) by the Japan Society for the Promotion of Science (JSPS). Conflict of interest The authors declare that they have no conflicts of interest.
References 1. Velan SS, Durst C, Lemieux SK, Raylman RR, Sridhar R, Spencer RG, et al. Investigation of muscle lipid metabolism by localized one- and two-dimensional MRS techniques using a clinical 3T MRI/MRS scanner. J Magn Reson Imaging. 2007;25:192–9. doi:10.1002/jmri.20786. 2. Usenius T, Usenius JP, Tenhunen M, Vainio P, Johansson R, Soimakallio S, et al. Radiation-induced changes in human brain metabolites as studied by 1H nuclear magnetic resonance spectroscopy in vivo. Int J Radiat Oncol Biol Phys. 1995;33:719–24. 3. Ward KM, Aletras AH, Balaban RS. A new class of contrast agents for MRI based on proton chemical exchange dependent saturation transfer (CEST). J Magn Reson. 2000;143:79–87. doi:10.1006/jmre.1999.1956. 4. Haris M, Nanga RP, Singh A, Cai K, Kogan F, Hariharan H, et al. Exchange rates of creatine kinase metabolites: feasibility of imaging creatine by chemical exchange saturation transfer MRI. NMR Biomed. 2012;25:1305–9. doi:10.1002/nbm.2792. 5. Zhou J, Payen JF, Wilson DA, Traystman RJ, van Zijl PC. Using the amide proton signals of intracellular proteins and peptides to detect pH effects in MRI. Nat Med. 2003;9:1085–90. doi:10. 1038/nm907. 6. Cai K, Haris M, Singh A, Kogan F, Greenberg JH, Hariharan H, et al. Magnetic resonance imaging of glutamate. Nat Med. 2012;18:302–6. doi:10.1038/nm.2615. 7. Chan KW, McMahon MT, Kato Y, Liu G, Bulte JW, Bhujwalla ZM, et al. Natural D-glucose as a biodegradable MRI contrast
152 agent for detecting cancer. Magn Reson Med. 2012;68:1764–73. doi:10.1002/mrm.24520. 8. Terreno E, Castelli DD, Aime S. Encoding the frequency dependence in MRI contrast media: the emerging class of CEST agents. Contrast Media Mol Imaging. 2010;5:78–98. doi:10.1002/ cmmi.369. 9. Sun PZ, Benner T, Kumar A, Sorensen AG. Investigation of optimizing and translating pH-sensitive pulsed-chemical exchange saturation transfer (CEST) imaging to a 3T clinical scanner. Magn Reson Med. 2008;60:834–41. doi:10.1002/mrm. 21714. 10. Barker GJ, Simmons A, Arridge SR, Tofts PS. A simple method for investigating the effects of non-uniformity of radiofrequency transmission and radiofrequency reception in MRI. Br J Radiol. 1998;71:59–67. doi:10.1259/bjr.71.841.9534700.
S. Saito et al. 11. Sun PZ, van Zijl PC, Zhou J. Optimization of the irradiation power in chemical exchange dependent saturation transfer experiments. J Magn Reson. 2005;175:193–200. doi:10.1016/j. jmr.2005.04.005. 12. Volz S, Noth U, Rotarska-Jagiela A, Deichmann R. A fast B1mapping method for the correction and normalization of magnetization transfer ratio maps at 3 T. Neuroimage. 2010;49:3015–26. doi:10.1016/j.neuroimage.2009.11.054. 13. Singh A, Cai K, Haris M, Hariharan H, Reddy R. On B1 inhomogeneity correction of in vivo human brain glutamate chemical exchange saturation transfer contrast at 7T. Magn Reson Med. 2013;69:818–24. doi:10.1002/mrm.24290. 14. Desmond KL, Stanisz GJ. Understanding quantitative pulsed CEST in the presence of MT. Magn Reson Med. 2012;67:979–90. doi:10.1002/mrm.23074.
