Glutamate-sensitive CEST and MEMRI as novel biomarkers for studying ALS pathophysiology
Amit Kumar Srivastava1,2, Jiadi Xu3, Peter C.M. van Zijl2,3, Nicholas J Maragakis4, and Jeff W.M. Bulte1,2,3

1Cellular Imaging Section, Institute for Cell Engineering, Johns Hopkins University, Baltimore, MD, United States, 2Russel H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, Baltimore, MD, United States, 3F.M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States, 4Neurology, Johns Hopkins University, Baltimore, MD, United States


Amyotrophic lateral sclerosis (ALS) is characterized by selective loss of motor neurons. ALS treatment is very difficult because disease manifestation and diagnosis often happen much later than when ALS pathology occurs in the patient. In this study, we developed two non-invasive MRI biomarkers for the early detection of disease pathology and its progression thereafter. A newly developed Glutamate-sensitive CEST showed higher signal intensity in the spinal cord level of ALS at pre-symptomatic stage, an indicator of initiation of ALS pathology. Manganese-enhanced MRI showed higher T1-weighted signal in the ALS spinal cord at post-symptomatic stage suggesting activation of astrocytes.


Amyotrophic lateral sclerosis (ALS) is an incurable, progressive and fatal neurological disorder. Early stage detection and prediction of disease progression is at present extremely difficult. Currently, one of the few available non-invasive imaging techniques to probe ALS pathophysiology is proton spectroscopic imaging1. The disease is manifested by motor neuron degeneration, but recent evidence suggests that glutamate excitotoxicity and glial cell activation play an important role in early development of ALS pathology2. In this study, we report on using a glutamate-sensitive CEST approach3 and manganese-enhanced MRI as novel biomarkers for a better understanding of ALS pathophysiology at different stages of the disease.


Five male SOD1G93A transgenic ALS mice (B6SJL-Tg(SOD1*G93A)1Gur/J) and 5 male normal wild type (WT) littermates (Jacksons Laboratory, Bar Harbor, ME) were used in this study. MRI was performed using a horizontal bore Bruker Biospin 11.7T scanner Bruker, Ettlingen, Germany). A 72 mm volume resonator and a 4x1 phased array spinal cord coil were used for transmission and image acquisition, respectively. For glutamate-sensitive CEST of the spinal cord, we used an on-resonance variable delay multi-pulse sequence4 that is sensitive to fast exchanging protons and therefore detects predominantly glutamate in neuronal tissue. The VDMP sequence used 32 pulse-exchange modules with a 2 ms pulse width and a B1=46.8 μT was used. Ten mixing times (0 to 150 ms) were used for the MTC and GluCEST quantification by fitting the onVDMP buildup curves. For manganese-enhanced MRI (MEMRI), a spin echo T1-weighted pulse sequence with a TR/TE=550/8 ms, number of averages=8, FOV=12.8x12.8 mm, a matrix=192x192, number of slices=20, and a slice thickness=1 mm was used (spatial resolution of 67 μm in both dimensions). Animals were imaged pre- and 24 hours post-MnCl2 i.p. injection (0.4 mmol/kg bw). All animals were imaged at the pre- and post-symptomatic stage of the disease (67 and 125 days, respectively). Lumbar spinal cord sections were stained with anti-GFAP antibody as marker for astrocyte activation. For in vitro MEMRI cell activation studies, glial restricted progenitor (GRP)-derived astrocytes, undifferentiated GRPs, and fibroblasts were treated with and without a cytokine activation cocktail and with and without 2 μM of the voltage-gated Ca2+ channel blocker verapamil. Cultures were then incubated with 0.4 mM MnCl2 for 30 min. A total of 5x105 cells (n=2) were imaged for each cell incubation paradigm. Data were analyzed using PROC MIXED (Statistical Analysis System), employed with Type III fixed effects to evaluate the significance and lowest mean squares for direct comparisons between groups.

Results and Discussion

Fast-exchange sensitized VDMP MRI of the lumbar spinal cord showed a significantly higher signal in the grey matter of ALS spinal cord compared to wild type controls (p< 0.05) at only at the pre-symptomatic stage of the disease (Fig. 1). Based on exchange rate considerations, we interpreted this to be due predominantly to an increased level of glutamate. MEMRI demonstrated a significantly higher contrast in both white and grey matter of ALS mice at the post-symptomatic stage, which corresponded to anti-GFAP immunostaining for activated astrocytes (Fig. 2). In vitro manganese cellular uptake studies showed a significant increase of T1-w signal only in activated astrocytes that was blocked by the addition of verapamil indicating calcium channels as mechanism of Mn2+ intracellular transport.


Increased fast-exchange sensitized VDMP signal intensity during the pre-symptomatic stage of the disease may be used to predict if, when, and where ALS pathology develops. Increased MEMRI signal during the post-symptomatic stage of ALS corresponds to the histological presence of activated astrocytes, which are known to exhibit changes in calcium metabolism5. This agreed with the in vitro MEMRI experiments where activated astrocytes without Ca2+ blocker showed the highest signal. The two MRI modalities open up new avenues to study glial cell activation and ALS pathophysiology, and may be further used in drug development aimed to reduce glutamate excitoxicity.


Supported by ALS Association 16-IIP-252, NIH 2R01 NS045062, and MSCRF P41EB015909.


1. Bowen BC, et al. AJNR 2000;21:647-658.2.

2. Rothstein JD. Ann Neurol. 2009;65 Suppl 1:S3-9.3.

3. Cai K, et al. Nat Med. 2012;18:302-306.4.

4. Xu J., et al. Magn Reson Med 2014;71(5):1798-1812.5.

5. Volterra Aet al. Nat Rev Neurosci. 2014;15(5):327-335.


(A) Fast-exchange sensitized VDMP MRI of ALS and age-matched WT control mice at different disease stages. (B) Quantification of total fast exchange protons (%) in gray matter (GM) and white matter (WM); *p < 0.05, **p< 0.01

(A) MEMRI of ALS and age-matched WT controls at different disease stages. (B) MEMRI contrast corresponds to anti-GFAP immunostaining. (C) Quantification of T1-weighted signal intensities. ***p<0.001

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)