Remyelination in Multiple Sclerosis J. W. Prineas, MB, BS, MRCP, and F. Connell, BS
Chronic plaques in central nervous system tissue fixed by in situ perfusion for electron microscopy were examined for evidence of remyelination in 2 patients with multiple sclerosis (MS).Fibers with abnormal central myelin sheaths of several types were found a t the margins of most of the plaques studied. T h e most common of these were: (1) the presence of bare stretches of axon between contiguous internodes, (2) the presence of thin paranodes, (3) internodes which changed markedly in thickness along their length due to premature termination of superficial or deep myelin lamellae that ended as hypertrophic lateral loops, and (4) abnormally thin internodes which were of uniform thickness along their length, which were shorter than normal, and which terminated in the form of normal nodal complexes. T h e finding of internodes of t h e last type at the edges of many plaques indicates that remyelination by oligodendrocytes can occur in the adult human CNS and that it is common in some cases of IMS, although limited in its extent. Prineas JW, Connell F: Remyelination in multiple sclerosis. Ann Neurol 5:22-31, 1979
Until recently it was considered unlikely that remyelination by glial cells occurs in multiple sclerosis (ME) [ l , 22, 341 even though earlier workers, including Alzheimer, Volsch, and Dawson, had described fibers with unusually thin myelin sheaths at the edges of heavily gliosed plaques (reviewed in [ 161). Following the demonstration by Bunge, Bunge. and Ris in 1961 [ 131 and Bornstein, Appel, and Murray in 1962 [ 101 that central nervous system tissue in some species has the capacity to reform myelin following experimental primary demyelination, PPrier and GrPgoire [39] suggested that the thinly myelinated axons which they had observed in the first electron microscopic study of chronic MS lesions might represent remyelinating axons. Since Perier and GrCgoire's study, thinly myelinated axons have also been noted in fresh MS lesions 14,4 11. In such lesions and in two additional ones of undetermined age studied by Suzuki et a1 [47], it was observed that many of these internodes were not uniformly thin, as would be expected for remyelinating internodes [ 131, but that they varied in thickness along the course of a single internode. It was suggested that these might represent degenerating o r damaged internodes L3, 4, 41, 47, 481, a possibility supported by the observation that a nonuniform reduction in myelin sheath thickness is a common, nonspecific form of myelin damage found near experimentally produced focal demyelinating lesions 121, 23-25, 361. T h e present study was designed to determine whether or not the abnormally thin myelin internodes present at the edges of chronic
From the Neurology Service, Veterans Administration Hospital. East Orange, and the Department of Neuroxiences. College of MeJicinc and I k n t i s r r ~ of N e w Jersey-New Jersey M e d i d School. Newark, NJ.
22
plaques are also of this type. It was found that uniformly thin, short, central myelin internodes are not uncommon in this location, indicating that remyelination by glial cells does occur in MS.
Materials and Methods CNS tissue was obrained from 2 parients wirh longstanding MS in whom the brain and spinal cord were fixed for electron microscopy by in situ perfusion with 3% glutaraldehyde in 0.1 M cacodylate buffer within 20 minutes of death. The clinical features in each case and details o f the fixation method have been reported previously [4 1. 4 2 , 441. The tissue was postfixed i n Dalron's solution, embedded in Spurr's epoxy resin, and sampled for electron microscopy using 1 pm thick sections stained with tol-
uidine blue. The lesions selected for study comprised seventeen typical old plaques from the spinal cord and cerebral hemispheres. To study whole myelin internodes, selected plaques were serially sectioned at 1 to 2 pm intervals, and composite drawings of selected internodes were traced from photomicrographic montages of fibers serially sectianecl longitudinally at plaque margins. Measurements were done on tracings of internodes magnified x 1,000.
Results Nerve fibers with abnormally thin myelin sheaths (ratio of axon diameterMiameter of axon plus sheath B0.75 [40]) were commonly observed among fibers with sheaths of normal thickness at t h e margins of most of the cerebral and spinal plaques studied (Fig 1). In longitudinal sections these abnormally thin myelin sheaths were found to be of three main types,
Accepted for publication May 2 3 . 1978. Address reprint reqursts 1 0 Dr Princas, Neurology Service ( 1 271. Veterans Administration Hospital, East Orange,NJ "7019,
0364-5134/79/010022-10501.25@ 1978 by John W. Prineas
as shown in Figure 2: (1) uniformly thin internodes; (2) tapering internodes, that is: internodes that became thinner over their length; and (3) internodes in which the thin region was restricted to the paranode (that is, was within 30 p m of the nodal gap [27]). O n light microscopy the uniformly thin internodes were seen as short, isolated segments of myelin on stretches of bare axon (Fig 2 A ) or as thin, short internodes separated by normal nodal gaps from contiguous thin or normal internodes (Figs 2D, 3). Longer internodes judged to be of uniform thickness throughout their length were also common. The basis for this judgment was the numerous examples in which no change in sheath thickness could be detected by light microscopy over long stretches of an internode even though it was usually not possible to identify both ends of these internodes. In eleven uniformly thin internodes that could be measured in their entirety, the lengths ranged from 11 to 280 p m (eight measured l l to 5 0 p m and three measured 86, 140, and 280 p m , respectively), sheath thickness in different internodes ranged from 0.2 to 0.7 p m , and the diameter of the affected axons varied from 1.6 to 6.2 pm. Electron microscopy of the same areas revealed numerous thinly myelinated fibers which appeared, in both transverse (Fig 4) and longitudinal sections, to be typical remyelinating fibers that terminated in the form of normal nodal complexes composed of large, tubule-rich lateral loops. By light microscopy the nonuniformly thin internodes usually had the appearance of normally thick internodes that became progressively thinner along their length (see Fig 2B). Electron microscopy revealed that the change in thickness was due to premature termination of groups of superficial or, less often, inner myelin lamellae that ended in the form of symmetrically disposed arrays of large, pale lateral loops containing numerous microtubules (Fig SA, B). These loops were joined to each other by tight junctions (Fig SC), and they commonly formed desmosome-like junctions with neighboring astroglial processes. O f particular interest was the finding that the loops also formed attachments to nearby demyelinated axons, the junctions displaying periodic densities similar in appearance to the transverse bars present between myelin lateral loops and the underlying axon at normal nodes of Ranvier (Fig 6). The center-to-center distance between adjacent densities
Fig I. ( A ) Edge of a chronic plaque with a zone of thinly myelinutedfiben between normally myelinuted tissue and demyelinated tissue. ( R ) Enlarged i & u ~ of boxed area in A. (Toluidine blue; A ~ 6 3 0B, X I ,400, both before 5 reduct ion.)
Prineas
and
Connell: Remyelination in MS 23
measured 28 nm, which agrees with the figure of 25 to 30 nm reported for normal nodes [33]. Fibers in which contiguous myelin segments were separated by gaps measuring 22 to 180 p m wide were not uncommon at the edges of most of the lesions studied. Gaps wider than this were present, but these were difficult to measure. The myelin internodes on each side of these gaps usually appeared normal and terminated as ordinary heminodes without paranodal thinning. T h e diameter of the nonmyelinated regions of such axons tended to be less than the diameter of adjacent, myelinated regions of the same axon (see Fig 3 ) . In totally demyelinated regions it was usually difficult to identify any oligodendrocytes. When they werc observed i n demyelinated tissue, however, they appeared ultrastructurally normal except that they lacked processes and the nucleus was reduced in size (Fig 7A). Also, transverse bar formation between oligodendrocyte perikarya and demyelinated axons was seen o n a number of occasions (Fig 7B). Oligodendrocytes with rather abundant cytoplasm were not uncommon among myelinated fibers at plaque margins. Aberrant myelin formation by cells in this location was observed, but it was difficult to judge if this was more prominent here than in unaffected white matter. This abnormality took several forms, which included myelin formation around oligodendrocyte perikarya, the formation of myelin whorls in paranodal oligodendrocyte cytoplasm, and myelin formation within oligodendrocyte perikarya. Subpial spinal cord plaques in both patients contained occasional axons myelinated by Schwann cells, as noted previously in MS by Feigin and Popoff [ 181 and by Ghatak et a1 [ 191. Rare axons with abnormally thin peripheral myelin sheaths were observed close to the cord in some spinal nerve roots (Fig 8A); these were sometimes surrounded by typical onion-bulb configurations (Fig 8B) that were found o n electron microscopy to be composed of Schwann cell pro-
24 Annals of Ncurology Vol 5 No 1 January 1979
Pig 2. lypes of thin myelin sheathsfoundat the edges of chronic plaques. ( A ) Uniformly thin internode. (B) Tapering internode. (C)Thin paranod. (D) Intercalated internode. (Toluidine blue;AandB, X 1 , J O O ; C a n d D , x1,700.)
cesses and redundant basement membrane-findings suggestive of previous demyelination of peripheral axons. More impressive, however, was the overall sparing of peripheral myelin, which in almost every root examined was seen to extend intact right to the glial limiting membrane, even in the presence of extensive subpial demyelination.
Discussion Remyefination
Experimental studies have shown that when central myelin regenerates following demyelination, it does so chiefly in the form of new internodes rather than by extension from preexisting internodes. T h e newly formed internodes are inappropriately thin for the diameter of the axon [ 5 , 141, they terminate in the form of normal nodal complexes [ 13, 231. and, after the first few turns of myelin are laid down, their thickness appears to be uniform along the length of the internode [21, 2 3 t h one study the variation in thickness along the length of the internode was noted to be rarely more than one lamella [13]. Studies by Gledhill, Harrison, and McDonald [20, 21, 231 and by Blakemore et a1 [8]have shown that new internodes are also shorter than normal, ranging down to 6 p m in length. Uniformly thin internodes with these features were identified at the edges of most of the chronic plaques examined in the present study. This observation strongly suggests that remyelination by glial cells occurs in MS. Tapering Myefin Internoaks
T h e present study is the fourth in which nerve fibers with myelin sheaths that change in thickness along
F i g 4. Thinly myelinated nersefibers in tissue coniposed largely offibrous glialprocessesat the edge of a chronicplaque. (X6,200.)
F i g 3. The drawing on the left wus tracedfrom a series of photonzicrographs,some ofwhich are shown o n the right. o f n whole, uniformly thin niyelin irzternode (arrows) on a fiber at the edge of a chronicplaque. All oligodendrocytenuclei (crosshatched) seen near thefiber are also shouin. Note that the axoti tends to be wide where it is most heavily niyeliriated and narrow u6ere it is noiimyelinated. (Left x540; right: toluidiiie blue, X I , 725.)
Prineas and Connell: Remyelination in MS
25
F i g 5 . ( A )Electron micrograph ofa tapering myelin internode. The sheath changes in thickness in the boxedarea. (Bi An eniargeinent ofthe boxed area shown i n A reveals pale. tubule-rich lateral loops disposed symmetrically on each side of the fiber. (C) A higher magnification ofthe boxed area in B shou's tight junctions (arrows)between adjacent lateral loops. (A X6,300, B X18,300. C X97,OOO.)
26 Annals of Neurology Vol 5 No 1 January 1979
Fig 6. (A)Tapering internode in which a group of lateral loops formed by terminating szper-rial myelin lamellae are attached t o a neurby axon (arrow). ( B ) This attachment revealsjunrtions with periodic dertsities (small arrows) similur to the transverse bars at normal nodes. (A X7,500, B X78.000.)
F i g 7. (A)An oligodendrocyte lies against u large dee,n-yelinated axon in the demyelinated zone of u chronicplaque. There is n o evidence of new myelin formation. (B)Transzierse burs (arrow) can be seen between the perikaryon of an olzgodendroiyte (N = nucleus) anda demyelinated axon. ( A x5,100, 0 x89,500.)
Prineas and Connell: Remyelination in MS 27
F ig 9. The lower diagram depicts a normalcentral myelin internode a$ it would appear if unrolled (according to Hirano and Dembitzer 129). Internodes that taper due topremature termination of groups o/iuperficial myelin lamellae, if similarly unrolled, would appear as shown in the upper diagram.
F i g 8. ( A ) Thinly myelinatedfiberclose to the cord in a spinal nerve root. The myelin i.cperipheral in type. (x6,400.) ( B ) Typical onion-bulbformation around a thinly myelinated axon in a spinal root close to the cord. (Toluidine blue; x 1,200.)
the course of a single internode have been noted in MS plaques [4, 41, 471. In the previous studies, which dealt with fresh lesions o r lesions of undetermined age, the prematurely terminating myelin lamellae were observed to end in the form of small, cytoplasm-free loops, an appearance which contributed to the view that such internodes may be degenerating. In similar fibers observed in the present study, the truncated lamellae ended as large cytoplasmic pockets filled with microtubules, an appearance similar to that seen in the CNS at developing nodes during myelinogenesis [281 and one which suggests growth rather than degeneration of the internode. The facts that the abnormally located lateral loops formed several types of specialized junctions with surrounding structures and that these were similar to the junctions normally formed by lateral loops at nodes of Ranvier [33, 371 further suggest that these are regenerating rather than degenerating internodes. The possibility must therefore be considered that these are original internodes generating new myelin or new internodes of an unusual type, even though similar internodes have not been de-
28
Annals of Neurology Vol 5 No 1 January 1979
scribed before in studies of CNS remyelination [ 5 , 8, 13, 23, 31, 431 or myelinogenesis [281. If these tapering internodes are not new, they can only be original internodes that have changed in some fashion. It is known that myelin sheaths may be damaged short of total destruction and that the Atered sheaths can persist in a variety of forms. These include paranodal demyelination, truncated internodes, and "myelin bubble" formation (see the review by Schroder [46]),thin paranodes 12, 301, several types of reversible intramyelinic edema (reviewed in [ 9 ] ) ,irregularities in the contour of the whole fiber [IS], changes in thickness of the sheath due to an increase or decrease in axon volume [ 171, mechanically displaced sheaths [38], and, finally, a class of internodal changes in which the essential structural alteration is a deviation from the normal shovel shape of the (conceptually) unrolled myelin sheath (Fig 9), a change first described by Hirano and Dembitzer [29] in animals recovering from a variety of central white matter disorders. As shown in Figures 9 and 10, the myelin internodes in MS which change in thickness along their course may be regarded as a further example of this latter t y p e of myelin abnormality. Other central and peripheral nervous system conditions in which nonuniformly thin myelin internodes have been described, and which may or may not prove to have a configuration comparable to that shown in Figure 10, include two genetically determined diseases in mice [12, 49, 501, chronic focal demyelinating spinal cord lesions produced in cats by local injections of diphtheria toxin [24, 251 and by acute spinal cord compression [21, 233, and chronic spinal cord lesions induced by x-irradiation in rats 1351. In the spinal lesions just mentioned, evidence was found that the abnormal internodes represent
F i g 10. Three-dimensionalrepresentationofa tapering interno& of h e type commonly found at the margins of chronic MS plaques. (BaJedonadrawing from Bunge et a1 [ 1 3 1.)
damaged original internodes that persist at the edges of the lesions for at least 18 months following the acute injury, an observation which led McDonald and his co-workers to introduce the term partial-thickness demyelination to describe these internodes. While their fine structure has yet to be determined, the fact that they are a common nonspecific finding at the margins of chronic demyelinating lesions in experimental animals suggests a similar origin for the nonuniformly thin internodes present at plaque margins in MS. Functional Implications
The finding of a modest population of fibers undergoing remyelination at the margins of some chronic plaques lends weight to the suggestion that remyelination may contribute to the slow type of recovery seen following some exacerbations of MS [ 3 6 ] . If the internodes that change in thickness along their length result from previous damage, their early repair
might be associated with a more rapid recovery of function: it could be argued, for example, that a loss of superficial myelin lamellae occurring during active partial-thickness demyelination would open the intramyelinic extracellular space of Mugnaini and Schnapp [37] to the exterior of the sheath; reestablishment of tight junctions around the damaged edge of the sheath (seen as tight junctions between lateral loops o n prematurely terminating myelin lamellae, as illustrated in the present study) would resequester this space, presumably with some benefit to the function of the fiber. Paucity of Remyelination in MS
The major difference between MS lesions and remyelinating CNS lesions described in experimental animals seems to be that oligodendrocytes are found in the experimental lesions [7, 8, 13, 26, 321 but fail to reappear in demyelinated tissue in MS [34]. Reasons put forward to explain this include: (1) the larger size of the lesions in MS [43], (2) a species difference in the capacity of interfascicular oligodendrocytes to divide or to be supplied from some other source such as the subependymal plate in rodents [6], and (3) the presence of myelination-inhibiting antibodies in the
Prineas and Connell: Remyelination in MS 29
brain in MS [ 11, 45, 5 11. The presence of very thin remyelinating internodes in old plaques suggests that while some oligodendrocytes at plaque margins retain the capacity to form new myelin internodes, there may also be a failure in MS to generate myelin at a normal rate after contact has been reestablished between oligodendrocytes and demyelinated axons. Supported by the Kroc Foundation and the Research Service of the Veterans Administration. Presented in p a n at the 53rd Annual Meeting of the American Association of Neuropathologists, Chicago, IL,June 17-19. 1977
References 1. Adams RD, Kubik CS: The morbid anatomy of the de-
myelinative diseases. Am J Med 12~510-546, 1952 2. Allt G : Repair of segmental demyelination in peripheral nerves. Electron microscopic study. Brain 92:639-646. 1969 3. Andrews JM: The ultrastructural neuropathology of multiple sclerosis, in Wolfgram F, Ellison G W , Stevens JG. et al (eds): Multiple Sclerosis. immunology, Virology and Ultrastructure. New York and London. Academic, 1972. pp 23-52 4. Andrews JM. Cancilla PA: Subacute demyelinaring disease in an adult (diffuse-disseminated sclerosis). Electron microscopic findings. Bull Los Angeles Neurol SOC 38:49-59, 1973 5 . Blakemore WF: Remyelination of the superior cerebellar peduncle in the mouse following demyelination induced by feeding cuprizone. J Neurol Sci 20:63-72, 1973 6. Blakemore WF: Remyelination of the superior cerebellar peduncle in old mice following demyelination induced by cuprizone. J Neurol Sci 22: 12 1- 126. 1974 7. Blakemore WF: Invasion of Schwann cells into the spinal cord of the rat following local injections of lysolecithin. Neuropathol Appl Neurobiol 2:21-39. 1976 8. Blakemore WF. Eames RA. Smith KJ, et al: Remyelination in the spinal cord of the cat following intraspinal injections of lysolecithin. J Neurol Sci 33:31-43. 1977 9. Blank WF. Bunge MB, Bunge RP: The sensitivity of the myelin sheath. particularly the Schwann cell-axolemmal junction. to lowered calcium levels in cultured sensory ganglia. Brain Res 67:503-518. 1974 10. Bornstein MB, Appel S H , Murray MR: The application of tissue culture to the study of experimental "allergic" encephalomyelitis. Demyelination and remyelination. in Jacob H (ed): Proceedings of the lVth International Congress of Neuropathology. Stuttgart, Thieme, 1962. vol2. pp 279-282 11. Bornstein MB. Raine CS: Experimental allergic encephalomyelitis. Antiserum inhibition of myelination in virro. Lab Invest 23:536-542, 1970 12. Bradley WG, Jaros E. Jenkinson M: The nodes of Ranvier in the nerves of mice with muscular dystrophy. J Neuroparhol Exp Neurol 36:797-806, 1977 13. Bunge MB. Bunge RP, Ris H : Ultrastructural srudy of remyelination in an experimental lesion in adult c a spinal cord. J Biophys Biochem Cytol 10:67-94, 1961 14. Bunge RP: Glial cells and the central myelin sheath. Physiol Rev 48:197-251, 1968 15. Collins GH. Webster HF, Victor M: The ultrastructure of myelin and axonal alterations in sciatic nerves of thiamine deficient and chronically starved rats. Acta Neuropathol (Berl) 3:511-521, 1964 16. Dawson JW:The histology of multiple sclerosis. Trans R SOC Edinburgh vol L:517-740 with plates, 1916. Reproduced by the Montreal Neurological Institute, Montreal, 197 3
30 Annals of Neurology Vol 5 No 1 January 1979
17. Dyck PJ: Pathological alterations of the peripheral nervous system of man, in Dyck PJ, Thomas PK. Lamben EH (eds): Peripheral Neuroparhy. Philadelphia-London-Toronto, Saund e n , 1975, vol 1, p 308 18. Feigin I, Popoff N : Regeneration of myelin in multiple sclerosis: the role of mesenchymal cells in such regeneration and in myelin formation in the peripheral nervous system. Neurology (Minneap) 16:364-372. 1966 19. Ghatak NR, Hirano A, Doron Y, et al: Remyelination in multiple sclerosis with peripheral type myelin. Arch Neurol 29:262-267. 1973 20. Gledhill RF. Harrison BM. McDonald W1: Patterns of remyelination in the CNS. Nature 244~443-444. 1973 21. Gledhill RF, McDonald WI: Morphological characteristics of central demyelination and remyelination: a single-fiber study. Ann Neurol 1:552-560, 1977 22. Greenfield JG, Meyer A: General pathology of the nerve cell and neuroglia, in Blackwood W, McMenemey WH. Meyer A, et al (eds): Greenfield's Neuropathology. Second edition. London. Edward Arnold, 1962. p 18 23. Harrison BM, McDonald WI: Remyelination after transient experimental compression of the spinal cord. Ann Neurol 1:542-551, 1977 24. Harrison BM, McDonald WI, Ochoa J: Remyelination in ;he central diphtheria toxin lesion. J Neurol Sci 17:293-302. 1972 25. Harrison BM. McDonald WI. Ochoa J, et al: Paranodal demyelination in the central nervous system. J Neurol Sci 16x489-494, 1972 26. Herndon RM. Price DL, Weiner LP: Regeneration of oligodendroglia during recovery from demyelinating disease. Science 195:693-694. 1977 27. Hildebrand C: Ultrastructure and light-microscopic studies of the nodal region in large myelinated fibers of the adult feline spinal cord white matter. Acta Physiol Scand [Suppll 364~43-80. 1971 28. Hildebrand C: Ultrastructural and light-microscopic studies of the developing feline spinal cord white matter: 1. The nodes of Ranvier. Acra Physiol Scand [Supplj 364231-108, 1971 29. Hirano A, Dembitzer H M : A structural analysis of the myelin sheath in the central nervous system. J Cell Biol 34:555-557, 1967 30. King RHM. Pollard JD, Thomas PK: Aberrant remyelination in chronic relapsing experimental allergic neuritis. N e u o pathol Appl Neurobiol 1:367-378, 1975 31. Lampen PW: Demyelinarion and remyelination in exprrimental allergic encephalomyelitis. Further electron microscopic observations. J Neuropath Exp Neurol 24:371-385. 1965 32. Lampert P: Electron microscopic studies o n ordinary and hyperacute experimental allergic encephalomyelitis. Acta Neuropathol (Berl) 999- 126, 1967 33. Lvingston RB. Pfenninger K, Moor H. et al: Specialized paranodal and internodal glial-axonal junctions in the peripheral and central nervous system: a freezeetching study. Brain Res 58:l-24. 1973 34. Lumsden CE: The neuropathology of multiple sclerosis, in Vinken PJ, Bruyn G W (eds): Handbook of Clinical Neurology. Vol 9, Multiple Sclerosis and Other Demyelinating Diseases. Amsterdam. North-Holland. 1970, pp 21 7-309 35. Mastagha FL, McDonald WI, Watson JV, et al: Effects ot' xradiation o n the spinal cord: an experimental study of the morphological changes in central nerve fibers. Brain 99:101-122. 1976 36. McDonald WI: PathophysioloRy in multiple sclerosis. Brain 97~179-196, 1974 37. Mugnaini E, Schnapp B: Possible role of zonula occludens of
the myelin sheath in demyelinaung conditions. Nature
251:725-727, 1974 38. Ochoa J, Marotte L: The nature of the nerve lesion caused by chronic entrapment in the guinea-pig. J Neurol Sci 19:491495. 1973 39. Perier 0,Gr6goire A: Electron microscopic features of multiple sclerosis lesions. Brain 88937-952, 1965 40. Peters A. Palay SI., Webster H deF. The fine structure of the 41. 42.
43.
44.
nervous system. The neurons and supporting cells. Philadelphia-London-Toronto, Saunders. 1976, p 190 Prineas J: Pathology of the early lesion in multiple sclerosis. Hum Pathol 6531-554, 1975 Prineas J: Paramyxovirus-like panicles in acute lesions in multiple sclerosis, in Shiraki H , Yonezawa T. Kuroiwa Y (eds): The Aetiology and Pathogenesis of the Demyelinating Diseases. Proceedings of a Symposium Held in September. 1973. Tokyo, Japan Science Press, 1976, pp 311-319 Prineas J , Raine CS. Wiiniewski H: An ultrastructural study of experimental demyelination and remyelination: 111. Chronic experimental allergic encephalomyelitis in the central nervous system. Lab Invest 21:472-483. 1969 Prineas JW, Wright RG: Macrophages, lymphocyres and plasma cells in the perivascular compartment in chronic multiple sclerosis. Lab Invest 3 1:409-421. 1978
45. Raine CS. Bornstein
MR: Experimenral allergic encephalomyelitis: a light and electron microscope study of remyelination and “sclerosis” in vitro. J Neuropathol Exp Neurol 29552-574, 1970 46. Schriider JM: Degeneration and regeneration of myelinated nerve h k r s in experimental neuropathies. in Dyck PJ. Thomas PK. Lamhen EH (eds): Peripheral Neuropathy. Philadelphia-London-Toronto, Saunders. 1975. vol 1 , pp
337-362 47. Suzuki K, Anclrews JM. Waltz JM, et al: Ulrrastructural studies of multiple sclerosis. Lab Invest 20:444-454, 1969 48. Suzuki K. Grover WD: Ultrastructural and biochemical studies of Schilder’s disease: I. Ultrastructure. J Neuropathol Exp Neurol 29392-404. 1970 49. Suzuki K. ZaKoren JC: Quaking mouse: an ultrastructural study of the peripheral nerves. J Neurocytol 6:71-84, 1977 50. Wiiniewski H . Morel1 P: Quaking mouse: ultrastructural evi-
dence for arrest of tnyebnogenesis. Brain Res 29:63-73, 1973 5 1 . Yonezawa T, Saida T, Hasegawa M: Myelination inhibiting factor in experimental allergic encephalomyelitis and demyelinating diseases, in Shiraki H, Yonezawa T, Kuroiwa Y (eds): The AetioloRy and Pathogenesis of the Demyelinating Diseases. Proceedings of a Symposium Held in September. 1973. Tokyo. Japan Science Press, 1976, pp 255-273
P r i n e a s a n d Connell: R e m y e l i n a t i o n in MS
31
Chronic plaques in central nervous system tissue fixed by in situ perfusion for electron microscopy were examined for evidence of remyelination in 2 patients with multiple sclerosis (MS).Fibers with abnormal central myelin sheaths of several types were found a t the margins of most of the plaques studied. T h e most common of these were: (1) the presence of bare stretches of axon between contiguous internodes, (2) the presence of thin paranodes, (3) internodes which changed markedly in thickness along their length due to premature termination of superficial or deep myelin lamellae that ended as hypertrophic lateral loops, and (4) abnormally thin internodes which were of uniform thickness along their length, which were shorter than normal, and which terminated in the form of normal nodal complexes. T h e finding of internodes of t h e last type at the edges of many plaques indicates that remyelination by oligodendrocytes can occur in the adult human CNS and that it is common in some cases of IMS, although limited in its extent. Prineas JW, Connell F: Remyelination in multiple sclerosis. Ann Neurol 5:22-31, 1979
Until recently it was considered unlikely that remyelination by glial cells occurs in multiple sclerosis (ME) [ l , 22, 341 even though earlier workers, including Alzheimer, Volsch, and Dawson, had described fibers with unusually thin myelin sheaths at the edges of heavily gliosed plaques (reviewed in [ 161). Following the demonstration by Bunge, Bunge. and Ris in 1961 [ 131 and Bornstein, Appel, and Murray in 1962 [ 101 that central nervous system tissue in some species has the capacity to reform myelin following experimental primary demyelination, PPrier and GrPgoire [39] suggested that the thinly myelinated axons which they had observed in the first electron microscopic study of chronic MS lesions might represent remyelinating axons. Since Perier and GrCgoire's study, thinly myelinated axons have also been noted in fresh MS lesions 14,4 11. In such lesions and in two additional ones of undetermined age studied by Suzuki et a1 [47], it was observed that many of these internodes were not uniformly thin, as would be expected for remyelinating internodes [ 131, but that they varied in thickness along the course of a single internode. It was suggested that these might represent degenerating o r damaged internodes L3, 4, 41, 47, 481, a possibility supported by the observation that a nonuniform reduction in myelin sheath thickness is a common, nonspecific form of myelin damage found near experimentally produced focal demyelinating lesions 121, 23-25, 361. T h e present study was designed to determine whether or not the abnormally thin myelin internodes present at the edges of chronic
From the Neurology Service, Veterans Administration Hospital. East Orange, and the Department of Neuroxiences. College of MeJicinc and I k n t i s r r ~ of N e w Jersey-New Jersey M e d i d School. Newark, NJ.
22
plaques are also of this type. It was found that uniformly thin, short, central myelin internodes are not uncommon in this location, indicating that remyelination by glial cells does occur in MS.
Materials and Methods CNS tissue was obrained from 2 parients wirh longstanding MS in whom the brain and spinal cord were fixed for electron microscopy by in situ perfusion with 3% glutaraldehyde in 0.1 M cacodylate buffer within 20 minutes of death. The clinical features in each case and details o f the fixation method have been reported previously [4 1. 4 2 , 441. The tissue was postfixed i n Dalron's solution, embedded in Spurr's epoxy resin, and sampled for electron microscopy using 1 pm thick sections stained with tol-
uidine blue. The lesions selected for study comprised seventeen typical old plaques from the spinal cord and cerebral hemispheres. To study whole myelin internodes, selected plaques were serially sectioned at 1 to 2 pm intervals, and composite drawings of selected internodes were traced from photomicrographic montages of fibers serially sectianecl longitudinally at plaque margins. Measurements were done on tracings of internodes magnified x 1,000.
Results Nerve fibers with abnormally thin myelin sheaths (ratio of axon diameterMiameter of axon plus sheath B0.75 [40]) were commonly observed among fibers with sheaths of normal thickness at t h e margins of most of the cerebral and spinal plaques studied (Fig 1). In longitudinal sections these abnormally thin myelin sheaths were found to be of three main types,
Accepted for publication May 2 3 . 1978. Address reprint reqursts 1 0 Dr Princas, Neurology Service ( 1 271. Veterans Administration Hospital, East Orange,NJ "7019,
0364-5134/79/010022-10501.25@ 1978 by John W. Prineas
as shown in Figure 2: (1) uniformly thin internodes; (2) tapering internodes, that is: internodes that became thinner over their length; and (3) internodes in which the thin region was restricted to the paranode (that is, was within 30 p m of the nodal gap [27]). O n light microscopy the uniformly thin internodes were seen as short, isolated segments of myelin on stretches of bare axon (Fig 2 A ) or as thin, short internodes separated by normal nodal gaps from contiguous thin or normal internodes (Figs 2D, 3). Longer internodes judged to be of uniform thickness throughout their length were also common. The basis for this judgment was the numerous examples in which no change in sheath thickness could be detected by light microscopy over long stretches of an internode even though it was usually not possible to identify both ends of these internodes. In eleven uniformly thin internodes that could be measured in their entirety, the lengths ranged from 11 to 280 p m (eight measured l l to 5 0 p m and three measured 86, 140, and 280 p m , respectively), sheath thickness in different internodes ranged from 0.2 to 0.7 p m , and the diameter of the affected axons varied from 1.6 to 6.2 pm. Electron microscopy of the same areas revealed numerous thinly myelinated fibers which appeared, in both transverse (Fig 4) and longitudinal sections, to be typical remyelinating fibers that terminated in the form of normal nodal complexes composed of large, tubule-rich lateral loops. By light microscopy the nonuniformly thin internodes usually had the appearance of normally thick internodes that became progressively thinner along their length (see Fig 2B). Electron microscopy revealed that the change in thickness was due to premature termination of groups of superficial or, less often, inner myelin lamellae that ended in the form of symmetrically disposed arrays of large, pale lateral loops containing numerous microtubules (Fig SA, B). These loops were joined to each other by tight junctions (Fig SC), and they commonly formed desmosome-like junctions with neighboring astroglial processes. O f particular interest was the finding that the loops also formed attachments to nearby demyelinated axons, the junctions displaying periodic densities similar in appearance to the transverse bars present between myelin lateral loops and the underlying axon at normal nodes of Ranvier (Fig 6). The center-to-center distance between adjacent densities
Fig I. ( A ) Edge of a chronic plaque with a zone of thinly myelinutedfiben between normally myelinuted tissue and demyelinated tissue. ( R ) Enlarged i & u ~ of boxed area in A. (Toluidine blue; A ~ 6 3 0B, X I ,400, both before 5 reduct ion.)
Prineas
and
Connell: Remyelination in MS 23
measured 28 nm, which agrees with the figure of 25 to 30 nm reported for normal nodes [33]. Fibers in which contiguous myelin segments were separated by gaps measuring 22 to 180 p m wide were not uncommon at the edges of most of the lesions studied. Gaps wider than this were present, but these were difficult to measure. The myelin internodes on each side of these gaps usually appeared normal and terminated as ordinary heminodes without paranodal thinning. T h e diameter of the nonmyelinated regions of such axons tended to be less than the diameter of adjacent, myelinated regions of the same axon (see Fig 3 ) . In totally demyelinated regions it was usually difficult to identify any oligodendrocytes. When they werc observed i n demyelinated tissue, however, they appeared ultrastructurally normal except that they lacked processes and the nucleus was reduced in size (Fig 7A). Also, transverse bar formation between oligodendrocyte perikarya and demyelinated axons was seen o n a number of occasions (Fig 7B). Oligodendrocytes with rather abundant cytoplasm were not uncommon among myelinated fibers at plaque margins. Aberrant myelin formation by cells in this location was observed, but it was difficult to judge if this was more prominent here than in unaffected white matter. This abnormality took several forms, which included myelin formation around oligodendrocyte perikarya, the formation of myelin whorls in paranodal oligodendrocyte cytoplasm, and myelin formation within oligodendrocyte perikarya. Subpial spinal cord plaques in both patients contained occasional axons myelinated by Schwann cells, as noted previously in MS by Feigin and Popoff [ 181 and by Ghatak et a1 [ 191. Rare axons with abnormally thin peripheral myelin sheaths were observed close to the cord in some spinal nerve roots (Fig 8A); these were sometimes surrounded by typical onion-bulb configurations (Fig 8B) that were found o n electron microscopy to be composed of Schwann cell pro-
24 Annals of Ncurology Vol 5 No 1 January 1979
Pig 2. lypes of thin myelin sheathsfoundat the edges of chronic plaques. ( A ) Uniformly thin internode. (B) Tapering internode. (C)Thin paranod. (D) Intercalated internode. (Toluidine blue;AandB, X 1 , J O O ; C a n d D , x1,700.)
cesses and redundant basement membrane-findings suggestive of previous demyelination of peripheral axons. More impressive, however, was the overall sparing of peripheral myelin, which in almost every root examined was seen to extend intact right to the glial limiting membrane, even in the presence of extensive subpial demyelination.
Discussion Remyefination
Experimental studies have shown that when central myelin regenerates following demyelination, it does so chiefly in the form of new internodes rather than by extension from preexisting internodes. T h e newly formed internodes are inappropriately thin for the diameter of the axon [ 5 , 141, they terminate in the form of normal nodal complexes [ 13, 231. and, after the first few turns of myelin are laid down, their thickness appears to be uniform along the length of the internode [21, 2 3 t h one study the variation in thickness along the length of the internode was noted to be rarely more than one lamella [13]. Studies by Gledhill, Harrison, and McDonald [20, 21, 231 and by Blakemore et a1 [8]have shown that new internodes are also shorter than normal, ranging down to 6 p m in length. Uniformly thin internodes with these features were identified at the edges of most of the chronic plaques examined in the present study. This observation strongly suggests that remyelination by glial cells occurs in MS. Tapering Myefin Internoaks
T h e present study is the fourth in which nerve fibers with myelin sheaths that change in thickness along
F i g 4. Thinly myelinated nersefibers in tissue coniposed largely offibrous glialprocessesat the edge of a chronicplaque. (X6,200.)
F i g 3. The drawing on the left wus tracedfrom a series of photonzicrographs,some ofwhich are shown o n the right. o f n whole, uniformly thin niyelin irzternode (arrows) on a fiber at the edge of a chronicplaque. All oligodendrocytenuclei (crosshatched) seen near thefiber are also shouin. Note that the axoti tends to be wide where it is most heavily niyeliriated and narrow u6ere it is noiimyelinated. (Left x540; right: toluidiiie blue, X I , 725.)
Prineas and Connell: Remyelination in MS
25
F i g 5 . ( A )Electron micrograph ofa tapering myelin internode. The sheath changes in thickness in the boxedarea. (Bi An eniargeinent ofthe boxed area shown i n A reveals pale. tubule-rich lateral loops disposed symmetrically on each side of the fiber. (C) A higher magnification ofthe boxed area in B shou's tight junctions (arrows)between adjacent lateral loops. (A X6,300, B X18,300. C X97,OOO.)
26 Annals of Neurology Vol 5 No 1 January 1979
Fig 6. (A)Tapering internode in which a group of lateral loops formed by terminating szper-rial myelin lamellae are attached t o a neurby axon (arrow). ( B ) This attachment revealsjunrtions with periodic dertsities (small arrows) similur to the transverse bars at normal nodes. (A X7,500, B X78.000.)
F i g 7. (A)An oligodendrocyte lies against u large dee,n-yelinated axon in the demyelinated zone of u chronicplaque. There is n o evidence of new myelin formation. (B)Transzierse burs (arrow) can be seen between the perikaryon of an olzgodendroiyte (N = nucleus) anda demyelinated axon. ( A x5,100, 0 x89,500.)
Prineas and Connell: Remyelination in MS 27
F ig 9. The lower diagram depicts a normalcentral myelin internode a$ it would appear if unrolled (according to Hirano and Dembitzer 129). Internodes that taper due topremature termination of groups o/iuperficial myelin lamellae, if similarly unrolled, would appear as shown in the upper diagram.
F i g 8. ( A ) Thinly myelinatedfiberclose to the cord in a spinal nerve root. The myelin i.cperipheral in type. (x6,400.) ( B ) Typical onion-bulbformation around a thinly myelinated axon in a spinal root close to the cord. (Toluidine blue; x 1,200.)
the course of a single internode have been noted in MS plaques [4, 41, 471. In the previous studies, which dealt with fresh lesions o r lesions of undetermined age, the prematurely terminating myelin lamellae were observed to end in the form of small, cytoplasm-free loops, an appearance which contributed to the view that such internodes may be degenerating. In similar fibers observed in the present study, the truncated lamellae ended as large cytoplasmic pockets filled with microtubules, an appearance similar to that seen in the CNS at developing nodes during myelinogenesis [281 and one which suggests growth rather than degeneration of the internode. The facts that the abnormally located lateral loops formed several types of specialized junctions with surrounding structures and that these were similar to the junctions normally formed by lateral loops at nodes of Ranvier [33, 371 further suggest that these are regenerating rather than degenerating internodes. The possibility must therefore be considered that these are original internodes generating new myelin or new internodes of an unusual type, even though similar internodes have not been de-
28
Annals of Neurology Vol 5 No 1 January 1979
scribed before in studies of CNS remyelination [ 5 , 8, 13, 23, 31, 431 or myelinogenesis [281. If these tapering internodes are not new, they can only be original internodes that have changed in some fashion. It is known that myelin sheaths may be damaged short of total destruction and that the Atered sheaths can persist in a variety of forms. These include paranodal demyelination, truncated internodes, and "myelin bubble" formation (see the review by Schroder [46]),thin paranodes 12, 301, several types of reversible intramyelinic edema (reviewed in [ 9 ] ) ,irregularities in the contour of the whole fiber [IS], changes in thickness of the sheath due to an increase or decrease in axon volume [ 171, mechanically displaced sheaths [38], and, finally, a class of internodal changes in which the essential structural alteration is a deviation from the normal shovel shape of the (conceptually) unrolled myelin sheath (Fig 9), a change first described by Hirano and Dembitzer [29] in animals recovering from a variety of central white matter disorders. As shown in Figures 9 and 10, the myelin internodes in MS which change in thickness along their course may be regarded as a further example of this latter t y p e of myelin abnormality. Other central and peripheral nervous system conditions in which nonuniformly thin myelin internodes have been described, and which may or may not prove to have a configuration comparable to that shown in Figure 10, include two genetically determined diseases in mice [12, 49, 501, chronic focal demyelinating spinal cord lesions produced in cats by local injections of diphtheria toxin [24, 251 and by acute spinal cord compression [21, 233, and chronic spinal cord lesions induced by x-irradiation in rats 1351. In the spinal lesions just mentioned, evidence was found that the abnormal internodes represent
F i g 10. Three-dimensionalrepresentationofa tapering interno& of h e type commonly found at the margins of chronic MS plaques. (BaJedonadrawing from Bunge et a1 [ 1 3 1.)
damaged original internodes that persist at the edges of the lesions for at least 18 months following the acute injury, an observation which led McDonald and his co-workers to introduce the term partial-thickness demyelination to describe these internodes. While their fine structure has yet to be determined, the fact that they are a common nonspecific finding at the margins of chronic demyelinating lesions in experimental animals suggests a similar origin for the nonuniformly thin internodes present at plaque margins in MS. Functional Implications
The finding of a modest population of fibers undergoing remyelination at the margins of some chronic plaques lends weight to the suggestion that remyelination may contribute to the slow type of recovery seen following some exacerbations of MS [ 3 6 ] . If the internodes that change in thickness along their length result from previous damage, their early repair
might be associated with a more rapid recovery of function: it could be argued, for example, that a loss of superficial myelin lamellae occurring during active partial-thickness demyelination would open the intramyelinic extracellular space of Mugnaini and Schnapp [37] to the exterior of the sheath; reestablishment of tight junctions around the damaged edge of the sheath (seen as tight junctions between lateral loops o n prematurely terminating myelin lamellae, as illustrated in the present study) would resequester this space, presumably with some benefit to the function of the fiber. Paucity of Remyelination in MS
The major difference between MS lesions and remyelinating CNS lesions described in experimental animals seems to be that oligodendrocytes are found in the experimental lesions [7, 8, 13, 26, 321 but fail to reappear in demyelinated tissue in MS [34]. Reasons put forward to explain this include: (1) the larger size of the lesions in MS [43], (2) a species difference in the capacity of interfascicular oligodendrocytes to divide or to be supplied from some other source such as the subependymal plate in rodents [6], and (3) the presence of myelination-inhibiting antibodies in the
Prineas and Connell: Remyelination in MS 29
brain in MS [ 11, 45, 5 11. The presence of very thin remyelinating internodes in old plaques suggests that while some oligodendrocytes at plaque margins retain the capacity to form new myelin internodes, there may also be a failure in MS to generate myelin at a normal rate after contact has been reestablished between oligodendrocytes and demyelinated axons. Supported by the Kroc Foundation and the Research Service of the Veterans Administration. Presented in p a n at the 53rd Annual Meeting of the American Association of Neuropathologists, Chicago, IL,June 17-19. 1977
References 1. Adams RD, Kubik CS: The morbid anatomy of the de-
myelinative diseases. Am J Med 12~510-546, 1952 2. Allt G : Repair of segmental demyelination in peripheral nerves. Electron microscopic study. Brain 92:639-646. 1969 3. Andrews JM: The ultrastructural neuropathology of multiple sclerosis, in Wolfgram F, Ellison G W , Stevens JG. et al (eds): Multiple Sclerosis. immunology, Virology and Ultrastructure. New York and London. Academic, 1972. pp 23-52 4. Andrews JM. Cancilla PA: Subacute demyelinaring disease in an adult (diffuse-disseminated sclerosis). Electron microscopic findings. Bull Los Angeles Neurol SOC 38:49-59, 1973 5 . Blakemore WF: Remyelination of the superior cerebellar peduncle in the mouse following demyelination induced by feeding cuprizone. J Neurol Sci 20:63-72, 1973 6. Blakemore WF: Remyelination of the superior cerebellar peduncle in old mice following demyelination induced by cuprizone. J Neurol Sci 22: 12 1- 126. 1974 7. Blakemore WF: Invasion of Schwann cells into the spinal cord of the rat following local injections of lysolecithin. Neuropathol Appl Neurobiol 2:21-39. 1976 8. Blakemore WF. Eames RA. Smith KJ, et al: Remyelination in the spinal cord of the cat following intraspinal injections of lysolecithin. J Neurol Sci 33:31-43. 1977 9. Blank WF. Bunge MB, Bunge RP: The sensitivity of the myelin sheath. particularly the Schwann cell-axolemmal junction. to lowered calcium levels in cultured sensory ganglia. Brain Res 67:503-518. 1974 10. Bornstein MB, Appel S H , Murray MR: The application of tissue culture to the study of experimental "allergic" encephalomyelitis. Demyelination and remyelination. in Jacob H (ed): Proceedings of the lVth International Congress of Neuropathology. Stuttgart, Thieme, 1962. vol2. pp 279-282 11. Bornstein MB. Raine CS: Experimental allergic encephalomyelitis. Antiserum inhibition of myelination in virro. Lab Invest 23:536-542, 1970 12. Bradley WG, Jaros E. Jenkinson M: The nodes of Ranvier in the nerves of mice with muscular dystrophy. J Neuroparhol Exp Neurol 36:797-806, 1977 13. Bunge MB. Bunge RP, Ris H : Ultrastructural srudy of remyelination in an experimental lesion in adult c a spinal cord. J Biophys Biochem Cytol 10:67-94, 1961 14. Bunge RP: Glial cells and the central myelin sheath. Physiol Rev 48:197-251, 1968 15. Collins GH. Webster HF, Victor M: The ultrastructure of myelin and axonal alterations in sciatic nerves of thiamine deficient and chronically starved rats. Acta Neuropathol (Berl) 3:511-521, 1964 16. Dawson JW:The histology of multiple sclerosis. Trans R SOC Edinburgh vol L:517-740 with plates, 1916. Reproduced by the Montreal Neurological Institute, Montreal, 197 3
30 Annals of Neurology Vol 5 No 1 January 1979
17. Dyck PJ: Pathological alterations of the peripheral nervous system of man, in Dyck PJ, Thomas PK. Lamben EH (eds): Peripheral Neuroparhy. Philadelphia-London-Toronto, Saund e n , 1975, vol 1, p 308 18. Feigin I, Popoff N : Regeneration of myelin in multiple sclerosis: the role of mesenchymal cells in such regeneration and in myelin formation in the peripheral nervous system. Neurology (Minneap) 16:364-372. 1966 19. Ghatak NR, Hirano A, Doron Y, et al: Remyelination in multiple sclerosis with peripheral type myelin. Arch Neurol 29:262-267. 1973 20. Gledhill RF. Harrison BM. McDonald W1: Patterns of remyelination in the CNS. Nature 244~443-444. 1973 21. Gledhill RF, McDonald WI: Morphological characteristics of central demyelination and remyelination: a single-fiber study. Ann Neurol 1:552-560, 1977 22. Greenfield JG, Meyer A: General pathology of the nerve cell and neuroglia, in Blackwood W, McMenemey WH. Meyer A, et al (eds): Greenfield's Neuropathology. Second edition. London. Edward Arnold, 1962. p 18 23. Harrison BM, McDonald WI: Remyelination after transient experimental compression of the spinal cord. Ann Neurol 1:542-551, 1977 24. Harrison BM, McDonald WI, Ochoa J: Remyelination in ;he central diphtheria toxin lesion. J Neurol Sci 17:293-302. 1972 25. Harrison BM. McDonald WI. Ochoa J, et al: Paranodal demyelination in the central nervous system. J Neurol Sci 16x489-494, 1972 26. Herndon RM. Price DL, Weiner LP: Regeneration of oligodendroglia during recovery from demyelinating disease. Science 195:693-694. 1977 27. Hildebrand C: Ultrastructure and light-microscopic studies of the nodal region in large myelinated fibers of the adult feline spinal cord white matter. Acta Physiol Scand [Suppll 364~43-80. 1971 28. Hildebrand C: Ultrastructural and light-microscopic studies of the developing feline spinal cord white matter: 1. The nodes of Ranvier. Acra Physiol Scand [Supplj 364231-108, 1971 29. Hirano A, Dembitzer H M : A structural analysis of the myelin sheath in the central nervous system. J Cell Biol 34:555-557, 1967 30. King RHM. Pollard JD, Thomas PK: Aberrant remyelination in chronic relapsing experimental allergic neuritis. N e u o pathol Appl Neurobiol 1:367-378, 1975 31. Lampen PW: Demyelinarion and remyelination in exprrimental allergic encephalomyelitis. Further electron microscopic observations. J Neuropath Exp Neurol 24:371-385. 1965 32. Lampert P: Electron microscopic studies o n ordinary and hyperacute experimental allergic encephalomyelitis. Acta Neuropathol (Berl) 999- 126, 1967 33. Lvingston RB. Pfenninger K, Moor H. et al: Specialized paranodal and internodal glial-axonal junctions in the peripheral and central nervous system: a freezeetching study. Brain Res 58:l-24. 1973 34. Lumsden CE: The neuropathology of multiple sclerosis, in Vinken PJ, Bruyn G W (eds): Handbook of Clinical Neurology. Vol 9, Multiple Sclerosis and Other Demyelinating Diseases. Amsterdam. North-Holland. 1970, pp 21 7-309 35. Mastagha FL, McDonald WI, Watson JV, et al: Effects ot' xradiation o n the spinal cord: an experimental study of the morphological changes in central nerve fibers. Brain 99:101-122. 1976 36. McDonald WI: PathophysioloRy in multiple sclerosis. Brain 97~179-196, 1974 37. Mugnaini E, Schnapp B: Possible role of zonula occludens of
the myelin sheath in demyelinaung conditions. Nature
251:725-727, 1974 38. Ochoa J, Marotte L: The nature of the nerve lesion caused by chronic entrapment in the guinea-pig. J Neurol Sci 19:491495. 1973 39. Perier 0,Gr6goire A: Electron microscopic features of multiple sclerosis lesions. Brain 88937-952, 1965 40. Peters A. Palay SI., Webster H deF. The fine structure of the 41. 42.
43.
44.
nervous system. The neurons and supporting cells. Philadelphia-London-Toronto, Saunders. 1976, p 190 Prineas J: Pathology of the early lesion in multiple sclerosis. Hum Pathol 6531-554, 1975 Prineas J: Paramyxovirus-like panicles in acute lesions in multiple sclerosis, in Shiraki H , Yonezawa T. Kuroiwa Y (eds): The Aetiology and Pathogenesis of the Demyelinating Diseases. Proceedings of a Symposium Held in September. 1973. Tokyo, Japan Science Press, 1976, pp 311-319 Prineas J , Raine CS. Wiiniewski H: An ultrastructural study of experimental demyelination and remyelination: 111. Chronic experimental allergic encephalomyelitis in the central nervous system. Lab Invest 21:472-483. 1969 Prineas JW, Wright RG: Macrophages, lymphocyres and plasma cells in the perivascular compartment in chronic multiple sclerosis. Lab Invest 3 1:409-421. 1978
45. Raine CS. Bornstein
MR: Experimenral allergic encephalomyelitis: a light and electron microscope study of remyelination and “sclerosis” in vitro. J Neuropathol Exp Neurol 29552-574, 1970 46. Schriider JM: Degeneration and regeneration of myelinated nerve h k r s in experimental neuropathies. in Dyck PJ. Thomas PK. Lamhen EH (eds): Peripheral Neuropathy. Philadelphia-London-Toronto, Saunders. 1975. vol 1 , pp
337-362 47. Suzuki K, Anclrews JM. Waltz JM, et al: Ulrrastructural studies of multiple sclerosis. Lab Invest 20:444-454, 1969 48. Suzuki K. Grover WD: Ultrastructural and biochemical studies of Schilder’s disease: I. Ultrastructure. J Neuropathol Exp Neurol 29392-404. 1970 49. Suzuki K. ZaKoren JC: Quaking mouse: an ultrastructural study of the peripheral nerves. J Neurocytol 6:71-84, 1977 50. Wiiniewski H . Morel1 P: Quaking mouse: ultrastructural evi-
dence for arrest of tnyebnogenesis. Brain Res 29:63-73, 1973 5 1 . Yonezawa T, Saida T, Hasegawa M: Myelination inhibiting factor in experimental allergic encephalomyelitis and demyelinating diseases, in Shiraki H, Yonezawa T, Kuroiwa Y (eds): The AetioloRy and Pathogenesis of the Demyelinating Diseases. Proceedings of a Symposium Held in September. 1973. Tokyo. Japan Science Press, 1976, pp 255-273
P r i n e a s a n d Connell: R e m y e l i n a t i o n in MS
31
