In Vivo MRI of Endogenous Remyelination in a Nonhuman Primate Model of Multiple Sclerosis

Remyelination is crucial for recovery from inflammatory demyelination in multiple sclerosis (MS). Investigating remyelination in vivo using magnetic resonance imaging (MRI) is difficult in MS, where collecting serial short-interval scans is challenging. Using experimental autoimmune encephalomyelitis (EAE) in common marmosets, a model of MS that recapitulates focal cerebral MS lesions, we investigated whether remyelination can be detected and characterized noninvasively. In 6 animals followed with multisequence 7-tesla MRI, 36 focal lesions, classified as demyelinated or remyelinated based on signal intensity on proton density-weighted images, were subsequently assessed with histopathology. Remyelination occurred in 5 of 6 marmosets and 51% of lesions. Radiological-pathological comparison showed high sensitivity (88%) and specificity (90%) for detecting remyelination by in vivo MRI. This study demonstrates the prevalence of spontaneous remyelination in marmoset EAE and the ability of in vivo MRI to detect it, with implications for preclinical testing of pro-remyelinating agents and translation to clinical practice.


INTRODUCTION
Multiple sclerosis (MS) is a debilitating inflammatory demyelinating disorder affecting millions 50 worldwide (1). MS causes dynamic changes to myelin in the central nervous system (CNS), in-51 cluding the quintessential focal inflammatory destruction of myelin, as well as the phenomenon 52 of remyelination that can follow the demyelination (2-5). Although remyelination is a crucial 53 aspect of tissue repair, and as such represents an important therapeutic target (6), most 54 knowledge about remyelination in MS derives from postmortem studies using histochemical and 55 electron microscopy studies. This is because investigating remyelination in vivo in real time is 56 limited by imperfect discrimination on neuroimaging modalities such as magnetic resonance 57 imaging (MRI). Furthermore, in human beings, where collecting serial short-interval scans is 58 highly challenging, it is difficult to detect track the dynamic occurrence of remyelination. There-59 fore, to investigate the pathobiology of remyelination in the context of focal inflammatory demy-60 elination, a reliable preclinical model is needed to develop techniques that can then be applied 61 clinically.

63
Rodent models have been widely used to investigate various aspects of the pathobiology of de-64 myelination. However, while toxin models in mice, including the lysolecithin and cuprizone 65 models, display demyelination and even remyelination, they do not require involvement of adap-66 tive immune cells (7,8). Conversely, rodent experimental autoimmune encephalomyelitis (EAE) 67 models, while mediated by an immune response, are often neither focal nor profoundly demye-68 linating. There is no known rodent model that is characterized by multifocal inflammatory de-69 myelination in the brain that is disseminated in both space and time. 70 EAE in the common marmoset (Callithrix jacchus) is a well-recognized translational model that 72 serves as a bridge between the rodent EAE and human MS (9-11). Not only does EAE recapitu-73 late MS at the lesion level both radiologically and pathologically (12, 13), but lesions spontane-74 ously remyelinate (14), as occurs in MS.

76
Prior studies demonstrated that certain signal changes in MRI, such as magnetization transfer 77 ratio (MTR), correlate with remyelination in MS lesions (15)(16)(17)(18)(19). This has also been investigated 78 in animal models, albeit mainly in the rodent model and in the cortex (20, 21) rather than white 79 matter. It has also been demonstrated that partial remyelination can occur and can be localized  Here, we studied focal white matter lesions in marmoset EAE. We utilized serial in vivo MRI, 84 mainly involving proton density-weighted (PDw) and MTR sequences, to age and characterize 85 lesions. We further analyzed the lesions using histopathology, focusing on myelin lipids and pro-86 teins, to compare and study the reliability of using various in vivo MRI sequences in predicting 87 remyelination and their histopathological correlates.

258
Using in vivo MRI only, 43 focal white matter lesions were detected in the 6 EAE marmosets 259 (Table 3). Of these 43 lesions, 22 (51%) were classified as remyelinated, 7 as acute demyelinat-     MTR was separately used to classify lesions, in order to compare its utility to predict myelin 299 status with PDw. The sensitivity and specificity for detecting remyelination, using histopatholo-300 gy as the standard, were 82% and 79%, respectively, both lower than for PDw.

303
Based on our frequent longitudinal MRI-based sampling and aging of EAE lesions and analysis 304 of signal intensity, with comparison to histopathology, we found that inflammation and demye-305 lination are the primary components of lesion pathophysiology in lesions <5 weeks old, corrobo-306 rating our previous work (14, 24). The earliest remyelination detected was 5 weeks after initial 307 lesion detection on in vivo MRI, although most remyelinated lesions were >7 weeks old. To es-308 timate the rate of remyelination, we performed an initial estimate in 4 focal lesions followed 309 closely by MRI, based on continuous decrease in signal intensity on PDw, and validated by his-310 topathology ( Figure 5). These data suggest that remyelination period typically takes 4-9 weeks. period of remyelination based on the downward slope of intensity measurement. Green titles indicate that 322 the lesion subtype was confirmed with histopathology counterpart. M# corresponds to animal number in 323 Table 1.

341
In this study, we investigated the remyelination process in focal brain lesions of marmoset EAE, 342 a relatively faithful MS model, using high-resolution serial in vivo MRI and histopathology. Our The time course of marmoset EAE lesion development and repair is relatively stereotyped, as 365 confirmed here. Previous work from our group has shown that once lesions form, the period of 366 demyelination lasts around 6 weeks (14, 24), consistent with data from this study (4-7 weeks; 367 see Figure 5). On the other hand, remyelination occurs over a period of 4-9 weeks after peak 368 lesion signal intensity and volume, without much delay. This finding is in parallel with previous 369 studies that show that remyelination is especially prominent at early stages of MS lesion devel-370 opment, but rarer after years of disease duration in chronic lesions (5).

372
Our finding that remyelination can be detected on serial in vivo MRI with high sensitivity and 373 specificity, especially on PDw MRI, corroborates and expands our previous finding that PDw 374 MRI is sensitive to myelin status changes (24, 34). Those prior studies primarily focused on de-375 myelination but also included limited data on remyelination, with electron microscopy showing 376 thinned myelin sheaths in remyelinated axons (14). Here, we added comparison of PDw and 377 MTR, since MTR has been used to detect remyelination in vivo (15,16,35), finding that PDw 378 was more sensitive and specific for identifying remyelination. Several groups have reported that 379 MTR may be low in remyelinated lesions compared to normal-appearing white matter because of 380 the presence of incomplete or morphologically different myelin sheaths in previously demyelin-381 ated regions (3, 35). Unlike MTR, which suffers from signal-to-noise reductions due to its calcu-382 lation as a voxel-wise division of random variables, PDw signal intensity is directly measured by 383 the MRI system, and thus PDw may prove more useful for simply discriminating the presence or 384 absence of remyelination, and for characterizing its time course (as was done here).

385
To determine whether corticosteroid treatment alters the course of lesion repair, half of the mar-387 mosets (for each twin pair, the one that showed lesions first) received 5-day treatment courses. 388 Our data show that the difference in the prevalence of remyelinated lesions, and the duration of 389 the event, did not differ by treatment group. One possible explanation for this result is that the 390 early initiation of corticosteroid treatment (as soon as the first lesion was detected), resulting in 391 treatment completion before lesions were even 1 week old, might have been premature: the ini-392 tial demyelination period typically lasts 4-7 weeks before remyelination ensues. It is also possi-393 ble that reduction of inflammation via corticosteroid treatment could have interfered with the 394 remyelination process by slowing clearance of myelin debris, which is a prerequisite for OPC 395 recruitment (36) and differentiation (37, 38), thereby negating any potential direct benefit of cor-396 ticosteroids on remyelination.

398
Sex effect on remyelination was also investigated, with results suggesting that males had more 399 remyelinating lesions than females. However, this result must be viewed with caution, since the 400 two female animals (M#5 and M#6), twin pair, had shorter and more aggressive disease course 401 compared to the other animals studied here. Furthermore, the lesions on these two animals were 402 mainly ≤6 weeks old, possibly too little for substantial repair to have occurred.

404
The main limitation of the study is the small sample size. Although a higher number of animals 405 would strengthen the study, as nonhuman primates are a scarce resource, we were able to max-406 imize their use by focusing on individual lesions, rather than on number of animals developed 407 remyelination. Another limitation is that different EAE immunization protocols were applied 408 among marmosets. However, we did not observe any difference in lesion outcome across the different immunization schemes. Notably, previous studies from our group with a similar variety 410 of EAE induction methods have not shown notable differences in disease course or lesion patho-411 biology (14, 24).

413
In conclusion, our study clearly demonstrates in vivo remyelination in the marmoset EAE model,