In vivo imaging of CNS microglial activation/macrophage infiltration with combined [18F]DPA-714-PET and SPIO-MRI in a mouse model of relapsing remitting experimental autoimmune encephalomyelitis

Abstract

Purpose: To evaluate the feasibility and sensitivity of multimodality PET/CT and MRI imaging for non-invasive characterization of brain microglial/macrophage activation occurring during the acute phase in a mouse model of relapsing remitting multiple sclerosis (RR-MS) using [¹⁸F]DPA-714, a selective radioligand for the 18-kDa translocator protein (TSPO), superparamagnetic iron oxide particles (SPIO), and ex vivo immunohistochemistry.

Methods: Experimental autoimmune encephalomyelitis (EAE) was induced in female SJL/J mice by immunization with PLP_139-151. Seven symptomatic EAE mice and five controls underwent both PET/CT and MRI studies between 11 and 14 days post-immunization. SPIO was injected i.v. in the same animals immediately after [¹⁸F]DPA-714 and MRI acquisition was performed after 24 h. Regional brain volumes were defined according to a mouse brain atlas on co-registered PET and SPIO-MRI images. [¹⁸F]DPA-714 standardized uptake value (SUV) ratios (SUVR), with unaffected neocortex as reference, and SPIO fractional volumes (SPIO-Vol) were generated. Both SUVR and SPIO-Vol values were correlated with the clinical score (CS) and among them. Five EAE and four control mice underwent immunohistochemical analysis with the aim of identifying activated microglia/macrophage and TSPO expressions.

Results: SUVR and SPIO-Vol values were significantly increased in EAE compared with controls in the hippocampus (p < 0.01; p < 0.02, respectively), thalamus (p < 0.02; p < 0.05, respectively), and cerebellum and brainstem (p < 0.02), while only SPIO-Vol was significantly increased in the caudate/putamen (p < 0.05). Both SUVR and SPIO-Vol values were positively significantly correlated with CS and among them in the same regions. TSPO/Iba1 and F4/80/Prussian blue staining immunohistochemistry suggests that increased activated microglia/macrophages underlay TSPO expression and SPIO uptake in symptomatic EAE mice.

Conclusions: These preliminary results suggest that both activated microglia and infiltrated macrophages are present in vulnerable brain regions during the acute phase of PLP-EAE and contribute to disease severity. Both [¹⁸F]DPA-714-PET and SPIO-MRI appear suitable modalities for preclinical study of neuroinflammation in MS mice models.

Keywords: EAE; Mice; Multiple sclerosis; Neuroinflammation; SPIO-MRI; TSPO-PET.

Fig. 1

Sagittal, coronal, and transaxial [¹⁸F]DPA-714 PET, SPIO T2*MRI, and PET/MRI fusion images of a representative EAE mouse (CS 2.5) (a) and control (b). PET images represent summed scans (20–50 min post-injection) normalized to the left cerebral neocortex SUV values. Increased radiotracer uptake and loss of T2* signal can be observed in the cerebellum (red arrow), brainstem (white arrowheads), and, to a lesser extent, right cerebral cortex (white arrow) in the EAE mouse but not in the control. R, right; L, left

Fig. 2

[¹⁸F]DPA-714 SUVR (a) and SPIO-Vol values (b) (mean ± SD) show a significant increase in the different brain regions in EAE mice compared with controls (mean ± SD); *p < 0.05, **p < 0.02, ***p < 0.01. Scatterplots show a positive significant correlation between [¹⁸F]DPA-714 SUVR and SPIO-Vol in the cerebellum (rho 0.727, p < 0.01) (c) and the brainstem (rho 0.860, p < 0.001) (d)

Fig. 3

Confocal images of double-labelled immunofluorescence for TSPO (red), Iba1 (green), and merged TSPO + Iba1 (orange) in the cerebellum (a), medulla (b), cervical (c), and thoracolumbar spinal cord (d) in the EAE and control reported in Fig. 1. The histograms represent quantitative TSPO and Iba1 fluorescence intensity and TSPO/Iba1 co-localization (mean ± SD) in the different regions. *p ≤ 0.05 for EAE vs control. Scale bar 20 μm

Fig. 4

F4/80 (brown) and Prussian blue staining in brain sections at the level of the cerebellum (a), medulla (b), cervical (c), and thoracolumbar spinal cord (d) from control (e–h) and EAE mouse (i–r) reported in Fig. 1. Higher-magnification images of the frame depicted in i–l displaying double-labelled microglia/macrophages in perivascular regions at level of primary fissure (m), area indicated with red arrow in i; spinal vestibular nucleus (q), region indicated with double red arrows in i; parvicellular part of medial vestibular nucleus (n and r), area pointed with arrowhead in j; white matter of spinal cord (o and p) in correspondence of dashed arrow and large arrow in k and l respectively. Higher magnification of the frame depicted in q shows a single double-labelled cells. Arrows in panels r, o, and p point to double-labelled cells. Arrows indicating the different regions are also reported in the left schematic panels (a–d). Scale bars in e–l 200 μm, in m–r 20 μm, in q 50 μm

Fig. 5

Distribution of TSPO (brown) and Prussian blue staining in the cerebellum (a) (red arrow) and medulla (a) (arrowhead). Light microscopy images displaying TSPO, iron-positive cells, and TSPO/iron double-labelled cells in brain sections from control (b, d) and EAE mice (c, e). Arrows point to perivascular double-labelled TSPO/iron cells at the level of primary fissure in cerebellum (c) and of parvicellular part of medial vestibular nucleus in medulla (e) of the symptomatic EAE. Scale bars in b–d 10 μm, in e 20 μm