muscle, respectively, with (right) and without (left) selective inversion of the water ... 'H MRS was performed with a double spin-echo PRESS sequence with TE!
Magnetic coupling between water and metabolites in human tissues R. A. de Graaf, R. Lamerichs”, M. J. Kruiskamp, K. Nicolay Dept. in viva NMR, Utrecht University, Utrecht, The Netherlands * Philips Medical Systems, Best, The Netherlands Results Figs. 1A and B show ‘H NMR spectra from human brain and muscle, respectively, with (right) and without (left) selective inversion of the water resonance. The tCr resonance intensity clearly decreased upon inversion of the water, indicating a magnetic coupling between tCr and water. The other resonances were not affected by the water inversion. Fig. 2 shows the signal intensities of the metabolites as a function of the inversion time (TI) of the water. In both brain and muscle, the tCr shows a clear biexponential behaviour while the other resonances remain constant. The maximum signal decrease for tCr (-25 %) as observed here at 1.5 T is smaller than observed in rat brain at 4.7 T (- 35%). This may be explained (in part) by the shorter T, relaxation times of water at 1.5 T as compared to 4.7 T. Inversion of tCr did not lead to a signal reduction of the water resonance (data not shown).
&&g&&g! Magnetic coupling between mobile water protons and immobile protons has been described in a number of in vitro protein solutions and in uiuo systems (I) and forms the basis of MTC imaging. Recently, a similar coupling between metabolite protons and immobile protons has been reported for a number of tissues in tivo and in particular for total creatine and lactate (2-4). Since water and metabolites are both magnetically coupled to immobilized protons, they may also be mutually coupled. This was recently observed for lactate in BSA solutions in vifro (5) and for metabolites in rat brain in vivo (4). The aim of this study was to investigate the magnetic coupling between water and metabolites in human brain and muscle. The results will contribute to the understanding of the mechanisms underlying metabolite magnetization transfer (MT). Furthermore, the results will provide guidelines for the optimization of watersuppression techniques used in metabolite quantification studies, since an inappropriate method may affect the metabolite resonance intensity.
Conclusion The results presented here show that total creatine and water are magnetically coupled in human brain and muscle. This information is essential for quantitative ‘H NMR spectroscopy, because an perturbation of the water (as used for water suppression) may lead to an underestimation of the metabolite concentrations under certain conditions. These results on the endogenous creatine resonance are in agreement with observations of exogenous ethanol in human brain (6) and creatine and lactate in rat brain (4). Further experiments are required to unravel the exact mechanisms of the observed magnetic coupling. With this knowledge one can design more optimal NMR experiments and interpret NMR spectra more quantitatively.
Methods All experiments were performed on normal volunteers (n=4) with a 1.5 T whole-body scanner (Gyroscan, ACS-NT, Philips Medical Systems). ‘H MRS was performed with a double spin-echo PRESS sequence with TE! = 272 ms. Brain spectra were obtained from VOIs (4 x 3 x 3 cm3) that included gray and white matter, while ventricles were largely excluded. Muscle spectra were measured from VOIs (3.5 x 5 x 8 cm”) that included gastrocnemius and soleus, while bone and marrow were excluded. The magnetic coupling between water and metab&tes was studied by a selective inversion of water and acquiring spectra at inversion times of 250, 400,600,800,1000,1200,1800,2500 and 4000 ms.
References Reson. Med. 10. I35 (I 989) [2] W. Dreher el al., Mogn. et al. Mqn. Reson. Med. in press (1999) [4] R. A. de Graaf et al. Pmt. ht. Sot. h4ugn. Reson. Med. p. 329 (1998) [5] S. D. Swansson J. Mqn. Reson. 135,248 (1998) [6] D. J. Meyerhoff etal. Alcohol Clin. Exp RPS. 20, [I]
S. D. Wolff
et al. Magn.
Reson. Med. 31,Bl
(1994)
[3] Y. Luo
1263 (1996)
A Irnl ***.*
‘J’
2
1 PPM3
2
.
-2;“. J 60:
lipids
-I PPM 3
.
2
&4
.&yxl~” 1
l m. l y.
..
a-
. . .
.
B . NAA
lrn: ,..... 2 80 s M- .
.
.
7g&T **.... . .
. Cho
‘3 20-
.
..m
O*
TI (set)
lipids
. tCr
.
. tCr
.
l
n
l
TMA
TI (set)
1
Figure 1 : (A) ‘H NMR spectra from human brain, with (right. without (left) inversion of the water. (B) ‘H NMR spectra from (right, Tl = ICOO ms) and without (left) inversion of the water.
TI = ICI00 ms) and human muscle, with
Figure
2 : (A) Signal
intensity
of the three major
brain
metabolites.
NAA.
tCr and Cho.
as a function of the water inversion time flf). (B) Signal intensity of the three major resonances in muscle spectra, lipids, tCr and TMA as a function of TI.
