1636

MRSI DETECTABLE CHOLINE METABOLISM AS MARKER FOR GENDER DIFFERENCES IN AGING
Ulrich Pilatus1, Gunter P. Eckert2, Nasir Ludin1, Silke Matura1, Johannes Pantel1, Carmen Silaidos2, Lena Wachter2, and Elke Hattingen1
1Goethe-University Frankfurt, Frankfurt, Germany, 2Justus-Liebig-University of Giessen, Giessen, Germany

Synopsis

Combining quantitative (mmol/l) 1H MRSI and 31P MRSI data from the brain of healthy volunteers revealed a choline component (rCho=tCho-(PCho+GPC)) which is visible with 1H MRSI but missing in 31P MRSI. This component, which may account for a fraction of mobile phospholipids, is reduced in young female (mean age 26) subjects. Lower levels would indicate a higher integrity of the membrane phospholipids in the brain of young women. A possible reason is the higher estrogen levels in this group.

Background & Purpose

Phospholipid membranes of the brain are involved in important signaling functions and cell aging processes. It has been shown that estrogens might change the fluidity and function of membrane phospholipid layers. In vivo 1H/31P MR spectroscopic imaging (MRSI) of the brain allows to quantify different compounds of the membrane lipid metabolism to investigate sex- and/or age-related differences in healthy subjects.

Methods

130 healthy volunteers (33 young female, 35 old female, 32 young male and 30 old male; mean age young 26.5 y; mean age old 71 y) were included in the study. In addition to localizer scans the protocol included:
  • 3D T1w MRI for segmentation (MPRAGE)
  • B1 mapping as described in [1]
  • 2D 1H MRSI at 40 ms TE (sLASER, 12 mm slice, FOV 240x240 mm2, matrix 20x20, repetiton time (TR) 2 s)
  • 3D 31P MRSI (FID-CSI, FOV 240x240x200 mm3, matrix 8x8x8, flip angle 60°, TR 2 s )
  • 2D 1H MRSI for detection of water (FID-CSI, 25 mm slice, FOV 240x240 mm2, matrix 16x16, flip angle 2°, TR 0.54 s)
1H MRSI data were analyzed using LCModel. 1H detetectable metabolite concentrations were quantified in terms of mmol/l as described in [2]. It should be noted that this quantification refers to tissue water. 31P MRSI data were quantified based on a calibration curve obtained from a dedicated phantom experiment. Special emphasis was put on correction for the different in coil loadings, which is important when focussing on gender differences. The concentration per voxel of 31P MR detectable metabolites was divided by the tissue water content obtained from the GM and WM fractions providing final concentrations in mmol per tissue water (as for 1H MRSI). The tissue fractions were adjusted for the point spread function (PSF) of 31P MRSI Data were analyzed from the target regions shown in Figure 1. The left column shows the grid resolution for 1H MRSI, the right for 31P MRSI. Graphs in the lower row show the WM and GM fraction from the respective region. While differences are clearly visible for 1H MRSI, the poor point spread function for 31P MRSI levels these differences thus data do not allow for discrimination between GM and WM. Consequently, all comparisons which included 1H MRSI and 31P MRSI data, were performed on averages over all regions. The metabolites total choline (tCho) from 1H MRSI and phosphocholine (PCho), glycerophosphocholine (GPC), phosphoethanolamine (PE), and glycerophosphoethanolamine (GPE) from 31P MRSI were analysed. We introduced an addtional variable which we called residual choline (rCho) defined as the difference between tCho and the sum of PCho and GPE. To detect significant differences regarding age and gender a multivariate analysis of variance (ANOVA) was performed, using age and gender as factors.

Results

Results for 31P MRS lipid metabolites are summarized in Figure 2. There was a significant increase in GPE when comparing young to old subjects for both genders. For tCho, metabolite concentration could be analyzed in the 3 ROIs separately as shown in the left panel of Figure 3. The most prominent finding is the lowest tCho concentration for young females in all regions. As expected this finding is also valid for the averaged regions shown in the upper right panel of Figure 3. The lower right panel shows the residual choline value. In general, the sum of the phosphorylated compounds PCho and GPC was less than the tCho concentration hinting to a residual choline fraction (rCho) invisible with 31P MRS. This residual component accounts for the lower tCho concentration in young female subjects.

Discussion

31P MRS reveals an increased GPE concentration with age. A shift from lipid anabolites towards higher concentrations of catabolites was already observed by Forester et al who found a significant decrease in phosphomonoesters with age [3]. Our study shows no significance for choline containing 31P metabolites. This is in contrast to the 1H MRS data, which show decreased tCho for young females. The difference endures the combination of the two measurements, thus reduced levels of a 31P MRS non-visible Cho compound are typical for young females. A missing component in the tCho signal has been observed in previous MRS studies ([4], [5]). The MR visibility of membrane phospholipids, which resonate at the frequency of the tCho signal but are generally invisible due to immobilization, has been discussed ([6], [7]) as potential mechanism. Thus, rCho would be a marker for a fraction of mobile membrane lipids.

Conclusion

Increased rCho may be attributed to increased lipid mobility [6]. Thus, lower levels of rCho would indicate a higher integrity of the membrane phospholipids in the brain of young women. A possible reason is the higher estrogen levels in young women.

Acknowledgements

We would like to thank Nadine Brüche for helping us in performing the MR examinations and Ralf Deichmann for calculation of B1-maps

References

[1] Volz S, Nöth U, Rotarska-Jagiela A, Deichmann R. A fast B1-mapping method for the correction and normalization of magnetization transfer ratio maps at 3 T. NeuroImage 2010; 49(4):3015

[2] Silaidos, C., Pilatus, U., Grewal, R., Matura, S., Lienerth, B., Pantel, J., Eckert, G. P. Sex-associated differences in mitochondrial function in human peripheral blood mononuclear cells (PBMCs) and brain. Biology of sex differences 2018; 9:34

[3] Forester et al. Age-related changes in brain energetics and phospholipid metabolism. NMR Biomedicine 2010; 23:242

[4] Michaelis T, Merboldt K, Bruhn H, Hänicke W, Frahm J. Absolute concentrations of metabolites in the adult human brain in vivo: quantification of localized proton MR spectra. Radiology 1993; 187:219

[5] Barker PB, Blackband SJ, Chatham JC, Soher BJ, Samphilipo MA, Magee CA, Hilton JD, Strandberg JD, Anderson JH. Quantitative proton spectroscopy and histology of a canine brain tumor model. Magn. Reson. Med. 1993; 30:458

[6] Hattingen et al. Combined 1H and 31P spectroscopy provides new insights into the pathobiochemistry of brain damage in multiple sclerosis; NMR Biomed., 2011; 24:536

[7] Klein, J. Membrane breakdown in acute and chronic neurodegeneration: focus on choline-containing phospholipids. Journal of neural transmission 2000; 107:1027

Figures

Figure 1: ROIs and tissue fractions for 1H and 31P MRSI. The GM region is marked in orange. GM an WM fraction are calculated from segmented tissue images filtered with the appropriate point spread function (PSF). Note the coarse grid size for 31P MRSI. In combination with the poor PSF this hampers discrimination between GM and WM based on position of the ROI.

Figure 2: Concentration of lipid metabolites as obtained by 31P MRSI. Significant differences (p < 0.05) comparing male vs. female (+) or young vs. old (*) within in each subgroup (either age or gender) are marked. Whisker indicates SD, box indicates SE.

Figure 3: Concentration of Cho containing metabolites obtained from 1H MRSI or in combination with 31P MRSI. 1H MRSI based data, which can be discriminated between mainly GM or mainly WM based on the postion of the ROI, are shown in the left panel. To combine those data with 31P MRSI the regions were averaged (upper right panel). The lower right panel shows the missing component (rCho = tCho - (PCho + GPC)). Significant differences (p < 0.05) comparing male vs. female (+) or young vs. old (*) within in each subgroup (either age or gender) are marked. Whisker indicates SD, box indicates SE.

Proc. Intl. Soc. Mag. Reson. Med. 30 (2022)
1636
DOI: https://doi.org/10.58530/2022/1636