Emily Louise Baadsvik1, Markus Weiger1, Romain Froidevaux1, Wolfgang Faigle2, Benjamin Victor Ineichen2, and Klaas Paul Pruessmann1
1ETH Zurich and University of Zurich, Zurich, Switzerland, 2University Hospital Zurich and University of Zurich, Zurich, Switzerland
Synopsis
Ultrashort-T2
signal from white matter can be captured for imaging using dedicated short-T2
techniques. In this work, such methods were applied to D2O-exchanged
human brain tissue from four donors diagnosed with multiple sclerosis to obtain
0.39 mm isotropic resolution MR images of the ultrashort-T2 tissue components
and a lower-resolution multi-TE image series which, via a model fitting
procedure, was used to produce maps of the relative content of myelin lipid-protein
bilayer. As proof-of-principle, the contrast in these images was compared with corresponding
myelin-stained cryosections and sample photographs, yielding promising correlation
for white matter, grey matter and multiple sclerosis lesions.
Introduction
Myelin is vital to the
development and function of the central nervous system, and, consequently, demyelinating
disorders such as multiple sclerosis (MS) can have devastating effects. It would
be of great interest for the diagnosis, treatment and monitoring of such
diseases, as well as the study of demyelination and myelin repair, to
noninvasively track myelin content.
Development of MRI
techniques for myelin mapping has largely been focused on the unique relaxation
properties of myelin water or on magnetisation transfer1-3. However, with the advancement of short-T2
imaging technology, approaches targeting the extremely rapidly decaying MR
signal of the myelin lipid-protein bilayer have become feasible4-7. The aim of the present work is to apply the
method presented in Ref. (7) for mapping the myelin content of ex-vivo porcine brain to tissue from MS
patients and validate the image findings with histological myelin staining.Methods
Samples
Tissue slices (2-3 mm
thick and around 15x15 mm2) from four MS patients (UK brain bank) were
investigated. Demographic features of the patients are presented in Table 1, labelled
as tissue samples 1-4.
In order to produce
images depicting primarily non-aqueous proton content, H2O in the
samples was exchanged with D2O prior to imaging (similar procedure
to Refs. (4,7)).
Imaging
To capture signal from
the myelin bilayer, dedicated short-T2 techniques and hardware are
needed. This work relied on a high-performance gradient system capable of
reaching a strength of over 200 mT/m at 100 % duty cycle8, fast transmit/receive switches9 and a proton-free loop coil of 40 mm diameter.
Two different imaging sequences were used, the zero-TE variant HYFI10 and single-point imaging (SPI)11. See Table 2 for details on the protocols.
Analysis
The SPI protocol,
consisting of fourteen images acquired at different TE, captures signal decay
over time. The behaviour of this decay was analysed by means of a fitting
procedure, as described in Ref. (7). The underlying model contains three signal
components of Lorentzian or super-Lorentzian lineshape (with T2 or T2,min,
respectively), namely water (T2 of 50 ms, chemical shift of 4.7
ppm), myelin bilayer (denoted uT2-S, T2,min of ~10 μs) and residual ultrashort-T2 signal (denoted uT2-L,
T2,min of ~100 μs).
Specific T2 values
of the non-aqueous components were found by fitting the average signal from normal-appearing
white matter (WM) regions in samples 1 and 2. Maps of the amplitude
contributions of each component were generated through voxel-by-voxel fitting
where, to increase fit stability, all readily-determined parameters were fixed.
Tricubic interpolation
was performed on all SPI data, reducing voxel size by a factor of 2, and an
additional factor of 3 was applied for display purposes.
Histology
Subsequent to imaging,
the samples were cryosectioned (10 µm) and underwent immunohistochemistry for myelin oligodendrocyte
glycoprotein (MOG) with haematoxylin counterstain. MOG staining density is
closely related to tissue myelin density. Results
Figure 1 demonstrates
signal decay as a function of time for the SPI image series.
The T2,min
times of the non-aqueous components were found to be 6.5 μs and 110 μs, which is in good agreement with expected values.
An example
single-voxel fit is shown in Figure 2. The T2 times of all three
components were fixed, and the fit found the chemical shifts of the uT2-S
and uT2-L components at 1.26 ppm and 2.12 ppm, respectively. The
former corresponds well with the expected chemical shift of fat, supporting the
claim that the uT2-S component represents the myelin bilayer.
Figure 3 presents MR results, visual photographs and myelin-stained cryosections for all samples. Note that as a result of the
D2O exchange, practically all signal in the MR images is
non-aqueous. The HYFI images demonstrate the ability of dedicated short-T2
imaging systems to image the non-aqueous components of brain tissue at high
resolution, and the amplitude maps of the uT2-S component are interpreted
as depicting relative myelin content.
Both MR results show significant
correlation with the sample photographs. Grey matter (GM) and WM are clearly
distinguishable, with GM producing less MRI signal and giving lower uT2-S
amplitude than WM. MS lesions hardly produce signal, indicating a lack of
practically any detectable non-aqueous protons. Direct comparison of the MR
images with the corresponding myelin-stained sections shows a clear correlation
between MR signal intensity and MOG staining density.
It is worth noting
that GM signal levels are significant enough for potential analysis also of GM
myelin.
The amplitudes of the
water and uT2-L components are small compared to the uT2-S
component, constituting only around 2 % and 8 %, respectively.Discussion and Conclusion
While the analysis of
the SPI data gave conclusive results for the uT2-S component
representing the myelin bilayer, there is likely additional information present
in the observed signal evolutions which is yet to be extracted. Further work should
also include validation in samples not subjected to D2O exchange and
translation to in-vivo studies.
Overall, accessing the
myelin bilayer through short-T2 imaging methods was shown to be feasible
also in human tissue. Notably, the MR image contrast enables depiction of MS
lesions to a similar degree as histological MOG-staining.Acknowledgements
The authors would like
to thank Richard Reynolds from the UK MS Tissue Bank for providing the brain
samples.References
- Piredda
GF, Hilbert T, Thiran J-P, Kober T. Probing myelin content of the human brain
with MRI: A review. Magn Reson Med. 2021;85:627-652.
- MacKay
AL, Laule C. Magnetic Resonance of Myelin Water: An in vivo Marker for Myelin. Brain
Plast. 2016;2:71-91.
- Varma
G, Duhamel G, de Bazelaire C, Alsop DC. Magnetization transfer from inhomogeneously
broadened lines: A potential marker for myelin. Magn Reson Med.
2015;73:614-622.
- Wilhelm MJ, Ong HH, Wehrli SL et al. Direct magnetic resonance
detection of myelin and prospects for quantitative imaging of myelin density. Proc Natl Acad Sci. 2012;109:9605-9610.
- Du
J, Ma G, Li S et al. Ultrashort echo time (UTE) magnetic resonance imaging of the short T2
components in white matter of the brain using a clinical 3T scanner. NeuroImage. 2014;87:32-41.
- Seifert AC, Li C, Wilhelm MJ, Wehrli
SL, Wehrli FW. Towards quantification of myelin by
solid-state MRI of the lipid matrix protons. NeuroImage. 2017;163:358-367.
- Weiger M, Froidevaux R, Baadsvik EL,
Brunner DO, Rösler MB, Pruessmann KP. Advances in MRI
of the myelin bilayer. NeuroImage. 2020;217:116888.
- Weiger
M, Overweg J, Rösler MB et al. A high-performance gradient insert for rapid and
short-T2 imaging at full duty cycle. Magn Reson Med. 2018;79:3256-3266.
- Brunner
DO, Furrer L, Weiger M et al. Symmetrically biased T/R switches for NMR and MRI
with microsecond dead time. J Magn Reson. 2016;263:147-155.
- Froidevaux
R, Weiger M, Rösler MB, Brunner DO, Pruessmann KP. HYFI: Hybrid filling of the dead-time gap for
faster zero echo time imaging. medRxiv.
2020. doi: 10.1101/2020.06.16.20132704.
- Balcom
BJ, MacGregor RP, Beyea SD, Green DP, Armstrong RL, Bremner TW. Single-point
ramped imaging with T1 enhancement (SPRITE). J Magn Reson Ser A.
1996;123:131-134.