Synopsis

The utility of CEST MRI and MRS as a noninvasive strategy for visualizing and tracking glial-restricted progenitor cells (GRPs) after transplantation was evaluated. GRPs increased signal in on-resonance variable delay multiple pulse (onVDMP) chemical exchange saturation transfer (CEST) MRI maps and ¹H MRS at the lactate peak in cell phantoms, in naïve mice, and in an experimental autoimmune encephalomyelitis mouse model of multiple sclerosis at and/or proximal to the initial location of cells.

Introduction

Few methods are approved clinically to visualize and track transplanted cells. Tracking cells labeled with superparamagnetic iron oxide nanoparticles (SPIOs) using T2*-weighted MRI has been explored for nearly 2 decades. This approach is highly sensitive with the potential for single cell detection, but is not specific due to other endogenous sources of hypointense contrast and the prevalence of blooming artifacts in the images that preclude the demarcation of cells^1,2. Tracking cells labeled with fluorine nanoparticles using 19F MRI has emerged as a more specific approach that also permits cell quantification due to the lack of background signal, but this approach is limited by its low sensitivity, requiring 105-106 cells for detection^2,3. Long-term, the signal from both labeling agents can be retained in bystander cells well after transplant cell death, which can complicate image interpretation³.

Complementing label-based cell tracking strategies with MRI methods that are able to monitor distinguishing features of transplanted cells can circumvent this problem by resolving the cellular source of the label’s signal. The lactate content of glial-restricted progenitors (GRPs) is >1.5x higher than their progeny^4-7 and >10x higher than neurons⁸, microglia⁹ and macrophages¹⁰. We hypothesized that CEST MRI has potential as a complementary cell tracking strategy for transplanted glial progenitors.

Methods

Experimental autoimmune encephalomyelitis (EAE) induction: EAE was induced in C57Bl/6 mice by injection of 250 ng of pertussis toxin (i.p.) and 300 μg of MOG35-55 in IFA supplemented with 4 mg/ml tuberculin (s.c.) on days 0 and 2. Naïve mice were the control.

MRI: Mice were imaged 3, 5, or 17 days after transplantation using a horizontal bore Biospec 11.7T scanner with a 72-mm volume resonator.

On-resonance variable delay multiple pulse (onVDMP) CEST MRI (2 mm image slice), a 2x2 phased array coil was used with a TE=5 ms, TR=5 s, Rare factor=10, NA=1, repetitions=9, a B1=46.8 μT, 32 pulse-exchange modules with a 2 ms pulse width, and eight delays (mixing times from 1.14 to 100 ms), and FOV=128x128.

¹H MRS (STEAM): We used TE=3 ms, TR=2.5 s, NA=128, and voxel=4x2x2 mm, with a total scan time of 320 s. Metabolite concentrations were assessed using TARQUIN with the standard 1H brain library.

Cell transplantation: Murine GRPs were isolated from luciferase-transgene expressing neonatal pups (E14.5) and cultured for two weeks. Up to two million cells were injected using a Hamilton 31G microinjection needle and a stereotactic device into immunocompetent (C57Bl/6), immunodeficient (Rag2^-/-) or EAE-induced mice.

Statistics: Significance was evaluated using an ANOVA with Tukey post-hoc or Student’s t-test, as appropriate.

Results

Detecting GRPs using onVDMP CEST MRI and ¹H MRS (lactate peak) was first evaluated in phantoms (Figure 1). GRPs enhanced both signals compared to the negative control (media alone). Detecting GRPs in vivo was confirmed initially in naïve mice (Figure 2). onVDMP CEST MRI signal correlated well with the bioluminescence (BLI) used as imaging biomarker for cell survival and the ¹H MRS signal on day 3. Changes in MR signal from day 3 to day 17 post-transplantation in C57Bl/6 (immunocompentant) and Rag2^-/- (immunodeficient) mice differed (p<0.05) in all imaging modalities.

GRP detection was then evaluated in an EAE model of multiple sclerosis (Figure 3), a disease characterized by oligodendrocytes loss, which transplanted GRPs can potentially replenish. onVDMP CEST MRI and 1H MRS signals differed in multiple regions of the cerebral brain of EAE mice compared to naïve mice. Transplantation of 1x10⁶ allogeneic GRPs into the motor cortex or lateral ventricles partially restored MR signatures 5 days later at or proximal to the transplantation site. Transplantation of GRPs at either site attenuated paralysis in EAE mice (Figure 4). Cells were short-lived; however, fewer cells were needed to significantly attenuate paralysis when transplanted into the motor cortex.

Discussion and Conclusion

Acknowledgements

References

1. Masthoff M et al. Scientific Reports. 2018;8:9563.

2. Gaudet JM et al. Magnetic Resonance in Medicine. 2017;78:713.

3. Fink C et al. Scientific Reports. 2018;8:590.

4. Griffin JL et al. NMR Biomed. 2002;15:375.

5. Bhakoo KK and Pierce D. J Neurochem 2000;74:254.

6. Ferraiuolo L et al. PNAS. 2016;113:E6496.

7. Liddell JR et al. J Neurosci Res. 2009;87:2696.

8. Hertz L and Dienel GA. JNR. 2005;79:11.

9. Voloboueva LA et al. FEBS Lett. 2013;587:756.

10. Tan Z et al. J Biol Chem. 2015;290:46.