Effects of Chronic Demyelination on Axonal Transport in Experimental

Reports

Effects of chronic demyelination on axonal nerves showed primary focal demyelination with transport in experimental allergic optic perivasculitis and gliosis. Ultrastructural analysis neuritis. NARSING A. RAO, JOHN GUY, AND of the demyelinated axons revealed swollen axoplasm with disarray of neurotubules and neu- PAMELAS. SHEFFIELD. rofilaments. In view of these morphologic altera- Axonal transport studies were undertaken to determine tions, the present study of axonal transport was the effect of chronic demyelination on axonal function in undertaken to detect any functional alterations in experimental allergic optic neuritis in the guinea pig, an chronic demyelinating (allergic) optic neuritis. animal model for multiple sclerosis. Fast and slow components of axonal transport over the prelaminar, lami- Materials and methods. Chronic demyelinating nar, and retrolaminar portions of the optic nerve head optic neuritis was produced in 12 strain 13 guinea 10 and at the foci of demyelination in the retrobulbar optic pigs as previously reported. One eye of each of nerve were evaluated by the autoradiographic grain- these experimental animals received an in- counting technique. At 6 hr there was a significant intravitreal injection of 50 /xCi of tritiated leucine crease in grain counts over the demyelinated foci and in (L-leucine-5-4; 3H(N), sp. act. 58.5 Ci/mmol; New the regions proximal to the demyelination, including the England Nuclear Corp., Boston, Mass.) in 50 /u,l swollen disc. At day 1 there was no significant difference of sterile normal saline solution. Topical pro- in the grain counts at the site of demyelination when paracaine (E. R. Squibb & Co., Princeton, NJ.) compared to the myelinated portion of the nerve. How- and 0.5 ml of intraperitoneal pentobarbital were ever, at days 3 and 7 there was a decrease in the number used for anesthesia during the injection. Animals of grains over the demyelinated areas. These results indi- were sacrificed by intracardiac injection of 1 ml of cate impairment of axonal function in chronic demyeli- pentobarbital at intervals of 6 hr, 1 day, 3 days, nation. Moreover, in this pathologic process, most of the and 7 days after the intravitreal injection. The eyes synthesized materials appear to move in the fast transport phase, unlike in the normal optic nerve where the were enucleated with a segment of optic nerve bulk of materials move by slow transport. (INVEST measuring 6 mm or longer and fixed in 4% formal- OPHTHALMOL VIS SCI 21:606-611, 1981.) dehyde, 1% glutaraldehyde in 200 mOS phos- phate buffer at 4° C. The brain with the cranial Axoplanal transport of macromolecules and cel- portion of the optic nerve was removed from the lular organelles from the neuronal perikaryon to calvarium and fixed in 4% formaldehyde solution. the axon terminal includes a rapid phase moving at The sections with the optic disc and nerve were several hundred millimeters per day and a slow rinsed with 0.2M s-collidine buffer, postfixed in phase moving at a few millimeters per day. With 1% osmium tetroxide in 0.067M s-collidine, dehy- the tritiated amino acid leucine used as a substrate drated through a graded series of ethyl alcohol, for incorporation into synthesis of newly synthe- cleared in acetone, and embedded in Epon. Thin sized proteins, it has been found that the bulk of sections (1 /am) were cut on an ultramicrotome, transported materials normally moves in the slow placed on glass slides, and coated with Kodak phase.' NTB-2 emulsion (Eastman Organic Chemicals, Rochester, N.Y.). After a 3-week exposure, they Many factors have been shown to impair axo- were developed in Kodak D19 for 3 min, fixed in plasmic transport. Fast transport is interrupted by Regular Fix, and stained with toluidine blue. anoxia, by chemical depletion of energy supplied by ATP, mechanical compression,2 and by chemi- Autoradiographic analysis was performed by cal agents that interrupt intracellular micro- grain counting under 400X magnification using a tubules.3- 4 However, temperature increases in 0.5 by 0.5 mm grid (with 25 squares, 0.1 by 0.1 cold-blooded vertebrates as well as biogenic mm) over the prelaminar retina (point A), the monoamines in rabbits5 have been shown to en- lamina choroidalis (point B), the lamina scleralis hance fast axoplasmic transport. In addition, (point C), the postlaminar optic nerve head (point axonal transport is impeded at the scleral lamina in D), and the foci of demyelination (points E to I), experimental models of acute6 and chronic ocular located in the retrobulbar optic nerve. For the hypertension7 as well as in intracranial hyperten- internal controls, grains were counted at compa- sion8 and ocular hypotony.9 rable nondemyelinated areas of the optic nerve. We have previously reported an experimental The second control consisted in counting the model of optic neuritis resembling human mul- grains at the same topographical areas of the optic tiple sclerosis.10 Histopathologically, these optic nerve from normal guinea pigs (external controls) 606

Downloaded from iovs.arvojournals.org on 10/02/2021 Volume 21 Number 4 Reports 607

Fig. 1. Autoradiographs of the guinea pig optic nerve showing increased numbers of grains at the site of demyelination at 6 hr (A) when contrasted against a topographically similar myelinated area at the same time interval (B). (Toluidine blue; X600.) at each of the four time intervals mentioned nificant increase (p < 0.01) in grain counts within above. Although grain counts were unbiased and the foci of demyelination when compared to simi- masked as to time interval, we could not mask the lar topographical areas along the nerve showing demyelinated areas within the same section. normal myelin staining (Fig. 1), as well as between Therefore we have built into our protocol an in- the demyelinated foci and normal control (p < ternal as well as an external control. 0.05). Analysis of variance with random block Results. At 6 hr there was a statistically sig- design was the statistical test used on points D to

Downloaded from iovs.arvojournals.org on 10/02/2021 Invest. Ophthalmol. Vis. Sci. 608 Reports October 1981

Fig. 2. At day 3 there were a decreased number of grains at the focus of demyelination (A) when compared to a similarly matched myelinated area (B). (Toluidine blue; x600.)

H, which represent the start of the myelinated the same at 1 day, and decreased at 3 and 7 days, optic nerve. Grain counts were also increased in no obvious blockade of axonal transport was ap- front of the demyelinated area (prelaminar retina, parent despite obvious disc edema, which was lamina choroidalis, and lamina scleralis) when com- noted in all experimental animals (Fig. 3). pared to areas in front of myelinated optic nerve Discussion. The demyelinating disease of the (Table I). optic nerve seen in chronic experimental optic At day 1 there was no statistically significant neuritis exhibits not only a loss of the myelin coat- difference in grain counts between demyelinated ing from the axon but also morphologic alterations areas and topographically similar myelinated areas within the axoplasm. Ultrastructurally, the de- within the experimental group or between the ex- myelinated axons were found to have swollen perimental group and the control group. axoplasm, with disarray of neurotubules and neu- At day 3 (Fig. 2) there was a statistically sig- rofilaments.l0 nificant decrease in grain counts in front of and Axonal transport is dependent on at least two within the demyelinated area when contrasted factors: (1) incorporation and synthesis within the against topographically similar myelinated areas. neuronal cell body and (2) transport of these syn- This was found both within the experimental thesized proteins from the cell body to the axonal group (p < 0.01) and between experimental and terminal. Any factor(s) that interferes with these control animals (p < 0.01). Statistical significance may affect axonal transport. was achieved from points A to H (Fig. 3.) In the experimental model of demyelinating At day 7 grain counts were slightly decreased optic neuritis, an increased rate of fast transport, within the demyelinated area (p < 0.10). Al- as noted at 6 hr, would give rise to heavier labeling though grain counts were increased at 6 hr, were in the demyelinated areas. However, if only the

Downloaded from iovs.arvojournals.org on 10/02/2021 Volume 21 Number 4 Reports 609

I C D £ f OllUnci

A 6 C 0 t f G H

Fig. 3. Mean grain counts of demyelinated areas compared to controls, at the lamina retinalis (A), lamina choroidalis (B), lamina scleralis (C), retrolaminar area (D), and the foci of demyelination and comparable myelinated areas (E to H),

fast phase were affected, slow transport might pro- slow transport phase. This hypothesis would ac- ceed normally, resulting in the same grain counts count for the heavier labeling seen at 6 hr. Assum- as in the comparable control group. This was not ing normal protein synthesis in ganglion cells, the case, since the number of grain counts de- with less labeled material left in the perikaryon to creased at days 3 and 7, which indicates that de- be carried by slow transport, one would expect a myelination causes an increase in the ratio of syn- decreased count at the later intervals because thesized materials moving in fast transport to most of the labeled material had already departed those moving in the slow phase, whereas normally with the fast phase. This explanation would ac- the bulk of synthesized materials moves in the count for the decrease in grain counts seen at 3

Downloaded from iovs.arvojournals.org on 10/02/2021 Invest. Ophthalmol. Vis. Sci. 610 Reports October 1981

Table I. Summary of grain counting data at various intervals at the sites of demyelination and myelinated parts of optic nerve Grains counted 6 hr after 3H* I clay after 3H*

Exp. 1 Exp. 2 Con. 1 Con. 2 Exp. 3 Exp>. 4 Con. 3 Con. 4 Site] D MJ D M M M D M D M M M A 218 172 703 603 171 174 905 790 352 339 446 239 B 287 284 616 522 203 158 606 597 208 175 308 159 C 143 142 262 138 102 93 180 78 102 192 373 109 D 279 112 308 171 123 41 137 250 65 53 186 78 E 144 44 194 124 73 22 152 130 77 61 102 93 F 80 41 77 68 47 14 128 111 37 75 90 61 G 54 27 51 30 42 0 89 99 45 27 70 40 H 72 95 23 46 65 49 I 73 97 61

D = demyelinated; M = myelinated; Exp. = experimental; Con. = external control. *Time of sacrifice. tSites: A = prelaminar retina; B = lamina choroidalis; C = lamina scleralis (sites A, B, and C are normally unmyelinated); D = postlaminar optic nerve head; E, F, G, H, and I are in areas located in the retrobulbar optic nerve (E representing the proximal and I the distal part of the nerve). (Internal control: counts taken within myelinated areas of experimental group.

and 7 days. Further work in identifying the areas From the Departments of Ophthalmology and Pathol- in the brain (lateral geniculate body, superior col- ogy, Georgetown University Medical Center and Armed liculus, and others) of the guinea pig to which Forces Institute of Pathology, Washington, D.C. This ganglion cells send these transported materials are project was supported in part by NIH grant EY02155-02. needed before proof of this hypothesis can be Presented in part at the ARVO meeting, May 1980, Or- lando, Fla. Submitted for publication Jan. 23, 1981. Re- achieved. print requests: Dr. NarsingA. Rao, Department of Oph- The above hypothesis has some precedent. thalmology, Georgetown University Medical Center, Mice with muscular dystrophy (Bar Harbor 129/ 3800 Reservoir Road, N.W., Washington, D.C. 20007. ReJ) show defective myelination of the ventral 11 12 roots. Bradley and Jaros, in a study of axonal Key words: allergic optic neuritis, demyelination, axo- transport in the sciatic nerve of these dystrophic plasmic transport mice, showed more labeled material moving in fast transport in the dystrophic animals than in REFERENCES controls. Moreover, in goldfish optic nerves it has 1. Ochs S: System of material transport in nerve fiber been shown that the total amount of fast-trans- (axoplasmic transport) related to nerve function and ported protein was elevated to three to five times trophic control. Ann NY Acad Sci 228:202, 1974. the normal value just before the regeneration of 2. Ochs S and Hollingsworth D: Dependence of fast the axons." In our model of chronic optic neuritis, axoplasmic transport in nerve on oxidative metabo- we have observed varying stages of axonal altera- lism. J Neurochem 18:107, 1971. tions with demyelination and remyelination.10 3. Karlsson JO and Sjostrand JO: The effect of col- One or several of these alterations might have al- chicine on the axonal transport of protein in the tered the fast axonal transport. optic nerve and tract of the rabbit. Brain Res 13:617, The discovery of functionally impaired de- 1969. 4. Bunt AH: Effects of vinblastine in microtubule myelinated axons in our model, which closely re- structure and axonal transport in ganglion cells of sembles human multiple sclerosis clinically as well the rabbit retina. INVEST OPHTHALMOL 12:579, 1973. as histopathologically, could provide further in- 5. Guy J, Quigley HA, and Anderson DR: The effect of sight into our understanding of ultimate visual re- biogenic monoamines on rapid axonal transport in covery or permanent visual loss in human optic the rabbit optic nerve. INVEST OPHTHALMOL VISUAL neuritis. SCI 17:296, 1978.

Downloaded from iovs.arvojournals.org on 10/02/2021 Volume 21 Number 4 Reports 611

Grains counted

3 days after 3H* 7 days after 3H*

Exp. 5 Exp. 6 Exp. 7 Con. 5 Con. 6 Exp. 8 Exp. 9 Con. 7

D M D M D M M M D M D M M

205 263 536 635 327 743 484 423 444 425 478 521 362 157 193 140 268 331 637 410 336 320 280 354 537 294 68 90 187 202 226 562 462 295 217 187 160 341 124 56 61 84 199 44 359 387 201 135 58 20 144 131 52 69 47 117 50 332 265 193 21 101 64 173 127 38 66 60 114 48 265 213 228 43 139 80 94 114 35 29 89 101 53 270 192 214 66 93 55 119 88 42 35 142 144 19 108 65 121 25 141

6. Anderson DR and Hendrickson A: Effect of in- Royal College of Surgeons (RCS) and normal Long- traocular pressure on rapid axoplasmic transport in Evans (LE) rat pigment epithelium. RCS pigment epi- monkey optic nerve. INVEST OPHTHALMOL 13:771, thelium explants phagocytized liposomes composed of 1974. phosphatidylserine (pS) plus phosphatidylinositol (pi). 7. Gaasterland D, Tanishimia T, and Kuwabara T: However, phagocytosis was not observed in RCS pig- Axoplasmic flow during chronic experimental glau- ment epithelium explants incubated with liposomes com- coma. I. Light and electron microscopic studies of posed of phosphatidylcholine (pC) plus phosphatidyl- the monkey optic nerve head during development of serine (pS), phosphatidylethanolamine (pE) plus phos- glaucomatous cupping. INVEST OPHTHALMOL VISUAL phatidylinositol (pi), or pC + pS + pi + pE. Normal SCI 17:838, 1978. LE pigment epithelium explants phagocytized all prep- 8. Tso MOM and Hayreh SS: Optic disc edema in arations of liposomes. The ability of RCS pigment epi- raised intracranial pressure. Arch Ophthalmol 95: thelium to differentially phagocytize liposome prepara- 1458, 1977. tions of different phospholipid composition supports the 9. Minckler DS, Tso MOM, and Zimmerman LE: A idea that phospholipids in the membrane may participate light microscopic autoradiographic study of axo- in the phagocytic mechanism. (INVEST OPHTHALMOL VIS plasmic transport in the optic nerve head during SCI 21:611-616, 1981.) ocular hypotony, increased intraocular pressure and papilledema. Am J Ophthalmol 82:741, 1976. Retinal dystrophy in the Royal College of Sur- 10. Rao NA: Chronic experimental allergic optic neuri- geons (RCS) rat is associated with a genetic defect tis. INVEST OPHTHALMOL VISUAL SCI 20:159, 1981. in the retinal pigment epithelium, which is ex- 11. Grafstein B and Forman DS: Intracellular transport pressed as a reduced capacity to phagocytize pho- in neurons. Physiol Rev 60:1167, 1980. toreceptor outer segments in vivo1"3 and in vitro.4 12. Bradley WG and Jaros E: Axoplasmic flow in axonal The defect in the phagocytic mechanism appears neuropathies. II. Axoplasmic flow in mice with to be specific for the rod outer segment mem- motor neuron disease and muscular dystrophy. branes, since RCS rat retinal pigment epithelial Brain 96:247, 1973. cells are capable of phagocytizing carbon parti- cles5' 6 and polystyrene spheres.4 The importance of phospholipids as structural Selective phagocytosis of liposomes by cul- and functional components of biological mem- tured RCS rat pigment epithelium. JAN T. branes has stimulated numerous studies using pu- EFFRON, R. BRUCE SZAMIER, AND ROSS B. rified phospholipid vesicles (liposomes) as sim- EDWARDS. plified models of biological membranes.7 Recent Liposomes 0.5 to 5 /xm in diameter and of various phos- studies have shown that liposomes are incorpo- pholipid compositions were used as a simplified model of rated as intact vesicles by a phagocytic mechanism rod outer segment membranes to examine the phagocytic into cultured cells.8 The present study investigates mechanism of cultured explants of retinal dystrophic the ability of explants of normal and RCS rat pig-

Downloaded from iovs.arvojournals.org on 10/02/2021