Elsevier

Experimental Neurology

Volume 230, Issue 2, August 2011, Pages 248-257
Experimental Neurology

Adapted focal experimental autoimmune encephalomyelitis to allow MRI exploration of multiple sclerosis features

https://doi.org/10.1016/j.expneurol.2011.04.023Get rights and content

Abstract

We aimed to determine an optimal protocol for inducing a focal inflammatory lesion within the rat brain that could be large enough for an easier MRI monitoring while still relevant as a multiple sclerosis (MS) like lesion. We adapted a two-hit model based on pre-sensitization of the Lewis rat with myelin oligodendrocyte protein (MOG) followed by stereotaxic injection of pro-inflammatory cytokines (TNFα + IFNγ) within the internal capsule. We compared the following two strategies to increase focal lesion development for an easier MR translation: (1) a higher sensitization step (MOG50) or (2) a higher cytokine step with lower sensitization (MOG25). Control animals were administered only cytokines without MOG pre-sensitization. Animals were followed with T2, diffusion and T1 post gadolinium weighted images at 1, 3 and 7 days following cytokine injection. Immunostaining was performed at the same time points for macrophages (ED1), myelin (MBP and Luxol Fast Blue) and blood brain barrier integrity (IgG). At day 1, the focal lesions depicted with T2-weighted images were very similar among groups and related to vasogenic edema (high apparent diffusion coefficient (ADC), gadolinium enhancement and IgG extravasation) induced by cytokines irrespective of the pre-sensitization step. Then, at day 3, MOG50 rats developed statistically larger T2 lesions than MOG25 and control rats that were correlated with inflammatory cell accumulation. At day 7, MOG50 rats also showed larger T2 lesions than MOG25 and control rats, together with loss of anisotropy that were correlated with demyelination. In contrast, MOG25 and control rats developed similar MR lesions decreasing over time and almost undetectable at day 7. We conclude that with a high pre-sensitization step, the focal lesion can be monitored by MRI whose signal reflects some features of a MS-like lesion, i.e. edema, inflammatory cell accumulation and later demyelination.

Research highlights

► Focal EAE can be adapted in such a way that it can be monitored with MRI. ► The MR protocol can be the same one as the standard protocol used for MS patients. ► Adaptation of the model requires balancing the lesion with high pre-sensitization. ► The MR signal of the adapted model reflects the spectrum of MS-like lesion.

Introduction

Magnetic resonance imaging (MRI) is playing an increasing role in the investigation and clinical management of multiple sclerosis (MS) (Bakshi et al., 2008); MRI has helped in diagnosis, provided in vivo parameters for better understanding of the pathophysiology and has allowed monitoring of treatment efficacy in clinical trials (Bar-Zohar et al., 2008).

Animal models of MS are needed to further explore mechanisms of disease initiation and progression and to test therapeutic and restorative approaches (Gold et al., 2006, Steinman and Zamvil, 2006). Experimental autoimmune encephalomyelitis (EAE) is the most widely used animal model of MS because it shares several clinical and histological similarities with MS (Denic et al., 2011). However, the multifocal and disseminated nature of the lesions in EAE models makes it challenging to investigate a lesion at an early stage before it becomes clinically significant. The nature of these EAE lesions also makes it difficult to characterize the evolution and resolution of a given lesion. One solution is to use a focal lesion model, targeting a single EAE lesion that reflects a prototypic MS lesion (Kerschensteiner et al., 2004b). This model consists of a subthreshold sensitization of the Lewis rat with myelin oligodendrocyte protein (MOG) followed 20 days later by the induction of focal inflammation by microinjection of pro-inflammatory cytokines (tumor necrosis factor α [TNFα] and interferon γ [INFγ]). Focal lesions share histological similarities with those seen in MS (Kerschensteiner et al., 2004b, Merkler et al., 2006) and this methodology could provide several advantages, including knowledge of the exact onset and location of the lesion.

As with humans, MR endpoints could be of great interest in rodent studies, allowing in vivo longitudinal monitoring that could be better than the five step clinical scale, which is relatively crude and does not correctly quantify and correlate with structural damage (Wujek et al., 2002). The following two conditions are required when using MRI for rodent inflammatory lesion exploration: (i) the MR signal should reflect the features of a MS lesion, with development of the entire lesion spectrum (including edema, blood brain barrier [BBB] damage and demyelination); (ii) the lesion should be large enough to fall within MRI resolution. MR exploration of EAE lesions is challenging because of their small size and selective location within small anatomic structures such as the brainstem, optic nerve or spinal cord (Sakuma et al., 2004). The focal model, induced into a well-defined anatomic structure within the brain, can facilitate MRI detection. However, the initial protocol has to be more severe to ensure that it is detectable with the lower resolution of MRI. Cytokine doses can be increased (Rausch et al., 2009, Serres et al., 2009a, Serres et al., 2009b) but come with the risk that lesions may appear more cytokine-like than MS-like. This distinction is a key point because the combination of MRI and focal lesion induction could be used as a platform for therapeutic tests, implying that this is a relevant model.

Consequently, considering the focal EAE model as a “two-hit” model (first sensitization and then cytokine injection), we compared two strategies to increase lesion severity for better MR application. In one strategy, we balanced toward a higher sensitization step; in another strategy, we balanced toward a higher cytokine step.

Section snippets

Materials and methods

To compare the high sensitization and high cytokine strategies, we followed two experimental groups over time: MOG50 (balanced toward sensitization) and MOG25 (balanced toward cytokines). We also followed two control groups: complete Freund's adjuvant (CFA) and incomplete Freund's adjuvant (IFA; Fig. 1).

Biological evaluation

MOG50 rats exhibited signs of generalized disease after a mean delay of 16 days post sensitization (mean score = 2.5) in relation with histologically-detected lesions in the brainstem and spinal cord. Cytokine injection did not exacerbate the symptoms in any animals. In contrast, MOG25 rats and control rats (CFA and IFA) did not show any signs of disseminated EAE either before or after cytokine injection.

No anti-MOG antibodies were detected at baseline (the day of sensitization). On the day of

Discussion

In this paper we compared different strategies to increase a focal, “two-hit” (1: pre-sensitization and 2: cytokine injection) inflammatory model to facilitate MR exploration while still ensuring an EAE effect. Because MR resolution and sensitivity is different from that of histological data, a strong pre-sensitization step was necessary to significantly balance the lesion towards an EAE-like focal lesion instead of reflecting the local effect of cytokines only. With this model the lesion could

Acknowledgments

We thank Dr. Nora Abrous (INSERM U862, Bordeaux, France) for surgery supervision.

Grant support:

TT is the recipient of a research fellowship from the French Society of Radiology (Société Française de Radiologie) and CNRS-CHU assistant. Experimental work was supported by Conseil Régional d'Aquitaine (KGP and VD).

References (24)

  • M. Kerschensteiner et al.

    Remodeling of axonal connections contributes to recovery in an animal model of multiple sclerosis

    J. Exp. Med.

    (2004)
  • K.O. Lovblad et al.

    MR imaging in multiple sclerosis: review and recommendations for current practice

    AJNR. Am. J. Neuroradiol.

    (2010)
  • Cited by (0)

    View full text