@Pergamon

Neurochem.Int.Vol.29,No. 6, pp. 597-605,1996Copyright~ 1996ElsevierScienceLtd

PII: S0197-0186(96)00061-7 Printedin Great Britain.All rightsreserved01974186/96$15,00+0.00

TIME COURSE OF BIOCHEMICAL ANDIMMUNOHISTOLOGICAL ALTERATIONS DURING

EXPERIMENTAL ALLERGIC ENCEPHALOMYELITIS

DANIELA A. SLAVIN, ANA E. BUCHER, ALICIA L. DEGANO,NESTOR W. SORIA and GERMAN A. ROTH*

Departamento de Quimica Biologica (CIQUIBIC–CONICET), Facultad de Ciencias Quimicas,Universidad Nacional de Cordoba, C.C. 61, C6rdoba 5016, Argentina

(Received 18 January 1996; accepted 29 May 1996)

Abstract—A comprehensive biochemical, immunological and histological study was undertaken duringdifferent stages of experimental allergic encephalomyelitis (EAE). Wistar rats with EAE induced bysensitization with bovine myelin showed a maximum decrease of body weight 1+16 days post-inoculation(dpi), coincident with the appearance of the paralysis symptom (acute period). Quantitation of some braincomponents indicated a temporal dissociation among the alterations observed. The higher diminution ofmyelin basic protein (MBP) occurred at 6 dpi and then increased to reach 21 dpi, a normal value. Also,the activity of 2’,3’-cyclic nucleotide 3’-phosphohydrolase was reduced by 40°4 with respect to controlanimals only at 6 dpi. The total lipid content was normal; however, among the individual lipids, sulfatideswere principally degraded during the acute stage but the amount of cerebrosides was decreased during therecovery period (29%40dpi). Free cholesterol was similar in both groups of animals, whereas cholesterolesters were detected in EAE animals from 14to 40 dpi. Central nervous system meningeal and parenchymalinfiltration with mononuclear cells was recognized principally at 14 dpi, but some of cells were still presentat 40 dpi. Deposits of immunoglobulins in the infiltrated regions as well as in spinal cord motor neuronswere observed among 14-29 dpi. Total circulating antibodies to MBP began to increase at 14 dpi, reachinga plateau at 21 dpi and then maintaining this value until 40 dpi. However, the population of anti-MBPantibodies that also recognizes the neuronal protein synapsin was only present at 14 dpi. The presentresults suggest that the neurological symptoms can be related to some early changes in the myelin membranefollowed by alterations involving neuronal structures. The existence of immunological factors against someepitopes in MBP that also recognize a synaptosomal protein might account, at least in part, for the axonaldamage and disruption of the normal interneuronal activity in EAE and lead together with the alterationsin some specific myelin constituents and the concomitant CNS inflammatory process to the observedhindlimbparalysis.Copyright~ 1996ElsevierScienceLtd

Experimental allergic encephalomyelitis (EAE) is aninflammatory and sometimes demyelinating disease ofthe central nervous system (CNS) with clinico-patho-logical and immunological similarities to multiplesclerosis (see, e.g. Yu et al., 1982;Raine, 1984;Zamviland Steinman, 1990; Wekerle, 1993). This exper-imental disease can be induced in susceptible animals

*To whom all correspondence should be addressed.,4bbreuiatimr: BSA, bovine serum albumin; CFA, complete

Freund’s adjuvant; CNPase, 2’,3’-cyclic nucleotide 3’-phosphohydrolase; CNS, central nervous system; dpi,days post-inoculation; EAE, experimental allergic ence-phalomyelitis; FITC, fluorescein isothiocyanate; HRP,horseradish peroxidase; MBP, myelin basic protein; PBS,phosphate-buffered saline; SDSPAGE, sodium dodecylsrdfatepolyacry lamide gel electrophoresis.

by a single injection of myelin or myelin basic protein(MBP) homogenized in an adequate adjuvant. Theacute stage starts 12–14 days after inoculation and ischaracterized by body-weight loss, hindlimb paralysisand incontinence. The appearance of the clinical signof paralysis is mediated by autoaggressive T-lym-phocytes specific for MBP (Hashim, 1985; Wekerle,1993). However, large scale destruction of myelin andoligodendrocytes is not typically observed in mostMBP-sensitized animals. Conversely, primary demy-elination may be seen in animals actively immunizedwith whole myelin or by cotransferring MBP-specificT-cells to naive animals along with antibodies againstmyelin constituents (Raine et al., 1981; Tabira, 1988;Linington et al., 1988; Wekerle, 1993). Another his-topathological feature of EAE is the inflammatory

597

598 D. A. Slavin et al.

changes and perivenular infiltration of mononuclearcells into the CNS. Also, we have previously shownthat some specific lipid alterations can be associatedwith EAE, although they do not necessarily occurexclusively when the paralysis symptom is present(Maggio and Cumar, 1975; Roth et al., 1978, 1982;Maggio et al., 1983). The poor correlation betweenthe severity of the demyelination process and theappearance of clinical symptoms suggests that otherpathological events may also contribute to the onsetof paralysis (Lassmann et al., 1980; Raine et al., 1981;Roth et al., 1982; Maggio et al., 1983; White et al.,1990). In this respect, many investigators have dem-onstrated that marked morphological and functionalchanges also occur in axons and synaptic terminalsduring EAE (Bieger and White, 1981; Kraft andSlimp, 1984; White et al., 1985, 1990). Deposits ofimmunoglobulins visualized by immunohistochemicalmethods that always accompany the CNS inflam-matory lesions (Traugott et al., 1982; Juhler et al.,1985; Roth and Obata, 1991) were also observed inspinal cord motor neurons (Roth and Obata, 1991).These observations suggest that an effect on the neu-ronal system could be also involved in the devel-opment of the clinical symptoms of EAE. Regardingthis, we have previously explored the possible immu-nological connection between myelin and syn-aptosomal proteins. We have shown that affinity-purified antibodies and T-lymphocytes specific forMBP can also recognize a neuronal protein, synapsin1(Pedraza et al., 1988;De Santis et al., 1992;De Santisand Roth, 1996).

In general, the above-described research in EAEhas focused on the acute period of the disease. In anattempt to better understand the expression of thisdisease, the present study extends previous work byinvestigating the biochemical, immunological and his-topathologic events occurring previously to the onsetof the clinical signs, during the acute stage and thesubsequent recovery period. For this purpose weevaluated some brain components, CNS histologicalterations and immunological response during thesedifferent stages. The obtained evidence indicates thatin EAE early changes occur in the myelin membranefollowed by immunological and histologicalprocesses, also involving neuronal structures, whichcorrelate with the appearance of the paralysis symp-tom.

EXPERIMENTAL PROCEDURES

Antigens and antibodies

Bovine myelin was purified from spinal cords (Autilio e~al., 1964), rat myelin and synaptosomal fractions, as pre-

viously described (Pedraza et al., 1988) and synapsin fromrat brain (Schiebler et al., 1986).Complete Freund’s adjuvant(CFA), bovine serum albumin (BSA), fluorescein iso-thiocyanate (FITC)-conjugated goat anti-rat immu-noglobulins, horseradish peroxidase (HRP)-conjugated goatanti-rat immunoglobulins, HRP-conjugated protein A andbovine MBP were from Sigma Chemical Co. (St Louis, MO,U.S.A.). Mouse anti-MBP monoclinal antibody (MAbBP1.3G9) was from Eurogenetics (Belgium), rabbit anti-MBP antiserum was prepared and formerly used (Roth etaL, 1983; Pedraza et al., 1988)and rabbit anti-synapsin anti-serum was raised by the procedure of De Camilli etal. (1983),

Animal treatmenl

Albino rats of both sexes, 41&50days old, from a Wistarstrain inbred in our laboratory for 30 years, were used. Underlight ether anesthesia, animals were given intradermal injec-tions in both hind feet with 0.5 ml of an emulsion preparedby sonication for 3 min at 20 kHz of 0.05 ml of water and0,45 ml of CFA containing 8 mg of bovine myelin (GroupEAE) or without any antigenic preparation as control(Group CFA). The animals were weighed and examineddaily for clinical signs of EAE, The scores of clinical severitywere: O, normal; 1, flaccid tail, S1OWto recover from thesupine position; 2, clumsy gait or mild ataxia; 3, definitehindquarter weakness; 4, severe hind leg paralysis and uri-nary incontinence; 5, severe quadriparesis, moribund stateor death. Animals from both groups were sacrificed at 6, 14,21, 29 and 40 days post-inoculation (dpi). A total of 40animals was used in each group, and 8 rats of both CFA andEAE groups were sacrificed each of the five studied time-points. First, the rats were anaesthetized with ether, bled bycardiac puncture, and then perfused with phosphate-bufferedsaline (PBS), pH 7,4. The brain was dissected, cut longi-tudinally into two equal sections, and one half was stored at–20”C for posterior use in biochemical analysis. The otherhalf of the brain, and the cerebella lobules and spinal cordwere fixed with 10°/0formalin in PBS for 24 h at 4“C.

Immunocytochemistry and histopathology

The fixed specimens were equilibrated in 30% sucrose/PBSat 4°C over 2 days for cryoprotection. Then, the tissues werefrozen embedded in O.C.T. compound (Miles Lab., Elkhart,IN, U. S.A.) and sections 12pm thick were cut on a cryostatmicrotome and taken upon gelatin-subbed coverglasses.Deposits of immunoglobulins were searched by immu-nofluorescence staining (Roth and Obata, 1991).The sectionswere first covered for 10 min at room temperature with 0.1M glycine/PBS, pH 7.4. Then, they were washed and furtherincubated for 1 h with FITC-conjugated goat anti-rat immu-noglobulins appropriately diluted in PBS containing 10/0BSA. After rinsing in PBS, the sections were dehydrated inincreasing concentrations of ethanol and mounted with 900/0glycerol/PBS onto slides and examined under a Zeiss Axi-oplan epifluorescence microscope. Adjacent sections werestained with hematoxylin<osin to asses the extent of cellularinflammatory lesions.

Biochemical analysis

The brain samples were homogenized in water and aliquotswere removed for determination of total protein content

Time course of biochemical and immunohistological alterations 599

(Bradford, 1976). Quantitation of MBP was done by animmunoblot analysis after sodium dodecyl sulfate–poly-acrylamide gel electrophoresis (SDS–PAGE) of the brainproteins and posterior transference to nitrocellulose by elec-troblotting (Roth and Obata, 1991).The nitrocelhdose sheetscontaining the transferred proteins were incubated in 10°/0skim milk powder, washed with PBS and then successivelyincubated with a rabbit anti-PBM antibody followed byHRP-conjugated protein A. The blots were developed using0.05Y0 4-chloro-l-naphtol and 0.01’7. hydrogen peroxideand the individual immunodetected proteins quantitated byscanning densitometry at 590 nm in a Shimadzu CS-930apparatus. The 2’,3’-cyclicnucleotide-3’- phosphohydrolase(CNPase) activity was assayed by the method of Prohaskaet al. (1973), using the conditions previously described (Rothet al., 1983). One unit of enzyme was defined as the amountneeded to produce 1 pmol of 2’-AMP from 2’,3’-cyclicAMPper minute under the experimental conditions. The specificactivity is expressed in units/mg protein.

For lipid quantitation, samples of the homogenates weretreated with 20 volumes of chloroform-methanol (1:1 v/v),and the extracted lipids fractionated into neutral and acidicfractions on a DEAE–Sephadex (A-25, acetate form)column. The acidic lipid fractions were then partitioned withwater and washed with the synthetic upper phase to removesalts, The lipids in both fractions were determined by anHPTLC–densitometry method (Macala et al., 1983; Roth etal., 1983).

Antibody titers

The total antibodies to MBP and the fraction crossreactingwith synapsin were examined after immunoblotting (Pedrazaer al., 1988; Roth and Obata, 1991). Purified rat myelin andsynaptosomal fractions were separated by SDS–PAGE. Thegels were electroblotted to nitrocellulose membranes, whichwere cut into 7 mm strips. The strips containing the trans-ferred proteins were soaked in 10% skim milk, washed withPBS and then individually incubated with the serum fromeach tested animal. After washing with PBS, the strips wereincubated with HRP-conjugated goat anti-rat immu-noglobulins. Then, the blots were developed using 4-chloro-l-naphtol and quantitated as indicated above for MBP deter-mination, To confirm that the synapsin labeling is due toanti-MBP antibodies present in the EAE rat sera, anti-MBPantibodies from two EAE animals sacrificed at 14 dpi werepurified by affinity chromatography on a MBP-conjugatedSepharose column (Pedraza et al., 1988). Then, the purifiedantibody fractions were tested by immunoblotting using arat synaptosomal fraction as indicate above for the wholesera.

Measurement of comparative antibody levels was also per-formed for cerebrosides and gangliosides using an ELISAtechnique (Mizutamari et al., 1994). The plates were coatedwith 1 pg/well of glycolipid in 100 pl of methanol. After airdrying, the wells were blocked with 1% BSA/PBS for 1 hand then 50 pl of plasma diluted in the same buffer wasadded and incubated overnight. After washing thoroughlywith PBS, immunoglobulin binding was detected followinga 2 h incubation with HRP-conjugated goat anti-rat immu-noglobulins. The color development was achieved after reac-tion with 15mM o-phenylenediamine and 0.0150/. hydrogenperoxide in phosphat=itrate buffer pH 5.0. The reactionwas stopped after 30 min by addition of 100Alof 4 N HJ304and the color analysed with a microplate reader at 450 nm.

RESULTS

Clinical course of EAE

All the animals injected with bovine myelindeveloped the predictable clinical signs of the disease,although males seemed to have an onset of the symp-toms retarded 1–2 days with respect to females. Theneurological signs were symmetrical and associatedwith weight loss (Fig. 1). The variation of body weightwas evident in all sick animals at 11–12 dpi before thedevelopment of the paralysis. During the acute phaseof EAE (1*16 dpi) the animals showed the mostsevere symptoms characterized by hindlimb paralysiswith fecal and urinary incontinence and a diminutionof 15–200/. of their body weight. Then, they begangradually to recover, regaining thetotalability to walkby 18–20 dpi. The EAE animals remained thin, somewith a slight tremor, not recovering to normal weightuntil about 40 dpi. At this point, the animals from theEAE group showed an apparent normal state clini-cally indistinguishable from the control CFA-injectedanimals. To study the possible relationship among thedifferent histopathological CNS alterations and thedevelopment of EAE, the animals were sacrificed on6 dpi when no differences were yet noted between theEAE and CFA groups, 14 dpi corresponding to the

O 7 14 21 28 35 42

DPIFig. 1. Com~arison of the bodv weight variation determinedas-the perc~ntage difference with ~espect to day O of (0)CFA- and (@) bovine myelin-injected animals. The resultsare the mean of all animals studied in each group. The degreeof the clinical symptoms was assessed as: O, normal; + 1,flaccid tail, slow to recover from the supine position; +2,clumsy gait or mild ataxia; +3, definite hindquarter weak-ness; +4, severe hind leg paralysis and urinary incontinence;

+5, severe quadriparesis, moribund state or death.

600 D. A. Slavin e~al.

acute period of the disease, 21 and 29 dpi as inter-mediate stages, and 40 dpi representing the total clini-cal recovery of the EAE animals.

Histopathological$ndings

Examination of histological sections of CNS tissuesfrom EAE animals stained by a conventional hem-atoxylin+osin method revealed meningeal and par-enchymal infiltration and perivascular cufiing withmononuclear cells throughout the CNS (Table 1).These changes were more profuse at 14 dpi when theanimals showed the most marked clinical symptoms,and then they gradually decreased in number. Never-theless, few infiltrates, especially in brainsteam andcerebella lobules, could still be seen as late as 40 dpiwhen the EAE animals were clinically similar to CFAcontrols. Although parenchymal infiltration was seenmost frequently in the lumbar spinal cord, the thoraxicand cervical sections also contained many areas ofinflammation. Parallel sections immunostained withFITC-anti-rat immunoglobulins revealed the presenceof immunoglobulins in regions where infiltration waspreviously observed with hematoxylin+osin.However, this deposit of immunoglobulins was notdetected after 21 dpi, although the parenchymal infil-tration was still apparent. Also, the presence of thepreviously reported deposits of immunoglobulins inthe spinal cord motor neurons and motor-controllingneurons in the lateral and spinal vestibular nucleus ofthe brainsteam (Roth and Obata, 1991)was searched.These cells fluorescently stained by incubation withFITC-anti-rat immunoglobulins were consistentlyobserved in the EAE animals since 14 to 29 dpi witha maximum staining at 21 dpi (Table 1, Fig. 2).

Biochemical alterations

Determination of the MBP amount was used asquantitative evaluation of the integrity of the myelin

membrane. This protein was significantly diminishedpreviously and during the acute period of EAE (6-14dpi), and then regained its normal value (Fig. 3(A)).Another method to determine the degree of myelinalteration is by measuring the activity of the oli-godendroglial enzyme, CNPase. This enzyme is con-centrated in heavily myelinated regions, increasing inparallel with myelination in viuo as well as in culturednerve tissue (Roth et al., 1985;Tsukada and Kurihara,1992). The CNPase activity was reduced about Qoyoat 6 dpi in the EAE animals, previously to the onsetof the clinical symptoms and maintained this valueuntil 40 dpi. However, since 14 dpi there was no sig-nificant difference between the experimental and con-trol groups since the CNPase activity correspondingto the CFA-injected rats also diminished to the level ofthe sick ones (Fig. 3(B)). These changes occur withoutsignificant difference in the amount of total proteinper gram of brain between both groups of animals atany time (data not shown).

During the process of EAE, selective alterations inthe content of some brain lipids were observed. Ingeneral, these changes did not occur at the same timeand concomitantly with the clinical sign of paralysis.The content of total phospholipids was significantlylower in EAE animals than in controls only at 6 dpi(–20%, p< O.05). This diminution corresponds to asimilar decrease in all of the individual phospholipidsquantified, rather than to one of them in particular(data not shown). The concentration of sulfatidesdecreased in brains of EAE animals only at 14 dpiduring the acute stage (Fig. 4(A)). Meanwhile, cere-brosides, which are also enriched in myelin, were sig-nificantly altered during the recovery period (2940dpi) (Fig. 4(B)). Free cholesterol level was similar inthe CFA and EAE groups (Fig. 4(C)). Cholesterol

Table 1.Centralnervoussystemimmunohistologicalalterationsin EAEanimals

ImmunocytochemicalalterationsHistological

dpi alterations’ Meninges Parenc.hyma Motor neurons

6 0 0 0 014 +++ +++ +++21

+++++ ++ +

29++++

+ + o40

++* o 0 0

‘Histologicalalterationswerevisualizedby staining 12pm sectionsof specimensfrom four animalsateachtime-pointwithhematoxylin%osinand scoredas follows:O,normal; +inflammatorycellcuffinglimited to the perivascular spaces or meninges; 2++, mild, moderate or marked infiltration, respec-tively, of inflammatory cells into CNS parenchyma.

bTo detect the presence of deposits of immunoglobrdins fixed in the tissues, adjacent sections wereimmunofluorescently stained with FITC-conjrrgated goat anti-rat Igs. The amount of bound immu-noglobulins was scored relatively from Oto 4+

Time course of biochemical and immunohistological alterations 601

Fig. 2. Binding of FITC-conjugated anti-rat immuno-globulins to spinal cord from the following groups of rats:(A) CFA-21 dpi; (B) EAE-14 dpi; (C) EAE-21 dpi; (D) EAE-29 dpi. Original magnification ( x 150). Similar results werefound in the samples from 4 rats analysed in each time point

from both groups studied.

esters, which were mostly absent in control rats, wereconsistently found in EAE animals since 14 dpi, witha maximum increase at 21 dpi and still present at 40dpi (Fig. 4(D)).

5 o~ oL—n———Ao 1 14 21 2S 35 42 0 7 14 21 28 35 42

DPI DPI

Fig. 3. Quantitation in (0) CFA- and (.) EAE-rat brains of(A) myelin basic protein (mg/g protein); (B) CNPase activity(units/mg protein). Each point represents the mean of dupli-cate determinations of 8 animals and bars are the S.E. ofthese means. The p value was calculated by Student’s t-testfor grouped samples. The probability of significance with

respect to the control group was: “p<O.Ol; *“’p<O.001.

Serum analysis

Total antibodies to MBP from EAE and CFA ani-mals were analysed by immunoblot. Taking intoaccount that antibodies against MBP are able to rec-ognize the neuronal protein synapsin (Pedraza et al.,1988), we also quantified the relative amount of anti-MBP antibodies that bind to synapsin. Immuno-staining of MBP present in the myelin proteins elec-troblotted to nitrocellulose membranes was seen withsera taken as early as 6 dpi. This staining increased inintensity and remained prominent up to 40 dpi (Fig.5(A)). Conversely, sera from EAE animals that alsorecognize synapsin present in the rat synaptosomalfraction were only seen at 14 dpi (Fig. 5(B)). To con-firm that this labeling of synapsin was due to anti-MBP and not to a different antibody present in theEAE rat sera, two affinity-purified anti-MBP anti-bodies were assayed. The obtained results showed thatthese preparations are still able to label synapsinwhereas the preadsorbed sera lost their activity (datanot shown). Although all the EAE animals during theacute stage of the disease showed the presence of anti-MBP antibodies that also recognize synapsin, no cor-relation between the severity of the disease and thetiter of these antibodies was observed. All sera fromCFA-injected animals were negative for both popu-

o 7 14 21 28 35 42 0 7 14 21 28 35 42

o 7 14 21 28 35 42 0 7 14 21 28 35 42DPI DPI

Fig. 4. Individual lipid determinations: mg/g protein of (A)sulfatides; (B) cerebrosides; (C) free cholesterol; (D) esterifiedcholesterol quantitated by HPTLC scanning densitometry inbrains from rats injected with (0) CFA or (.) bovine myelin(EAE group). Duplicate samples from six rats were analysedon each specific day for both groups. Bars indicate the S.E.of each mean. A p value for the ratio between experimentalvalues with respect to controls (CFA group) was calculatedby Student’s t-test for grouped samples, significant at

‘p<o,05; “p<o.ol; ““’p<o.ool.

602 D. A. Slavin et al,

o 7 14 21 28 35 42 0 7 14 21 28 35 42

o 7 14 21 28 35 42 0 7 14 21 28 35 42

DPI DPI

Fig. 5, Relative antibody titers ofl/4dihrted sera takenatdifferent dpifrom(o) controls (CFAgroup) or(o) myelin-injected animals (EAE group). (A) Anti-MBP and (B) anti-synapsin antibodies were quantitated by scanning densi-tometry of immunoblots and are indicated as the cor-responding area x 10–3. (C) anti-cerebrosides and(D) anti-gangliosides were relatively determined byan ELISA tech-nique and are indicatedas O.D. units x 10–2. The numberof determinations included six samples from CFA and sevenfrom EAE rats on each specific day. Bars represent the S.E.of the mean. A p value for the ratio between experimentalvalues with respect to controls was calculated by Student’st-test for grouped samples. The probability of significance

with respect to the CFA group was “’”p<0.001.

lations of anti-MBP antibodies. A number of otherbands corresponding to high molecular weight pro-teins also appeared when rat myelin was immu-nodetected with sera from EAE animals.

A relative measurement of the antibody titer tocerebrosides and gangliosides was obtained by anELISA technique (Fig. 5(C) and (D), respectively).Sera from normal, CFA- and myelin-injected animalsshowed immunoreactivity to these lipids. Althoughsmall differences were observed between CFA andEAE animals, both antibodies tend to rise in the sickrats at 14 dpi to reach a peak at 21 dpi and then fade.

DISCUSSION

There are different opinions regarding the onset ofthe clinical symptoms in EAE, as to whether theseare caused by demyelination or by an inflammatoryprocess. Among the described alterations, if a par-ticular one contributes to the clinical signs, it wouldpresumably precede or be concomitant with the acutestage of the disease. The obtained results show thatthe changes in several myelin constituents observed in

the EAE-affected animals occurred at different stagesof the disease, some of them previously to the appear-ance of any neurological symptoms. Since EAE isprincipally induced by T-lymphocytes specific forMBP (Hashim, 1985; Wekerle, 1993), the hereinobserved early decrease in the MBP amount andCNPase activity since 6 dpi (Fig. 3) could indicate thebeginning of a demyelinating process mediated bythese autoaggresive cells. At this respect, we have pre-viously demonstrated that MBP-specific T-cell linesare capable of mediating demyelination in culturesof CNS tissue (Lyman et al., 1986). The pattern ofdemyelination observed with these cells is consistentwith those induced by both total mononuclear cellsfrom animals with EAE (Roytt~ et al., 1985) or pat-ients with multiple sclerosis (Raine et al., 1973). Dur-ing the acute stage of the disease at 14dpi, the amountof MBP is still diminished and then it begins toincrease reaching a normal value at 21 dpi, suggestinga possible end to the myelin deterioration. Clemensand Fan (1979) have also shown that CNPase activitydecreases in brains of EAE-affected animals beforeany clinical or histopathological alterations appear,although other authors were unable to detect anydifference in this activity on the whole brain (Salvatiet al., 1986). The latter change in CNPase activitydetected in CFA-injected animals resembles thedecrease of this activity induced by specific T-cellsagainst purified protein derivative (PPD) of M.tuberculosis on myelinated organotypic tissue cultures(Lyman et al., 1988). Although some crossreactivitymay well exist between PPD and MBP (Vandenbarket al., 1975), both in vivo and in vitro results supportthe observation that some myelin damage may beinduced by non-CNS antigen-specific T-cells.

Even though there are no lipids exclusively specificto myelin, sulfatides and cerebrosides are the mosttypical myelin lipids. In many pathological conditionsthat cause demyelination, changes in these lipids arealways involved in the degenerative process and areoften used to indicate an alteration of myelin (Nortonand Cammer, 1984). However, sulfatides aredecreased principally at 14 dpi and cerebrosides arefound to be altered after 29 dpi during the recoveryperiod (Fig. 4). Since MBP shows preferential inter-action with negatively charged lipids such as sulfatides(e.g. Maggio et al., 1983), the disturbance of inter-molecular interactions at the membrane level broughtabout by an immunological response to MBP canaffect the amount of sulfatides by a bystander processto the MBP removal. Meanwhile, cerebrosides canbe altered later by a specific humoral immunologicalresponse (Roth et al., 1978, 1985). Although the rela-

Time course of biochemical and immunohistological alterations 603

tive amount of anti-cerebrosides with respect to con-trol sera is very low (Fig. 5(C)), these antibodies seemto increase at the same time that cerebrosides begin tobe degraded (Fig. 4(B)). However, the appearanceof antibodies to other myelin components does notcorrelate with the alteration observed in the amountof the corresponding antigen. Anti-MBP antibodiesshowed a maximum value among 2140 dpi, after theacute stage of the disease and when the MBP contentis normal ,again (Fig. 5). Similar dissociation betweenthe anti-myelin antibody titers and the clinical symp-toms have been previously observed (Roth and Obata,1991; Sadler et al., 1991). Taking into account thatthe blood–brain barrier (BBB) is impaired at the timeof onset of EAE in various species (e.g. Kristenssonand Wisniewski, 1977; Juhler et al., 1984; Kato andNakamura, 1989), could be expected that circulatingantibodies directed to myelin constituents during theacute period fade from peripheral blood and reach thebrain to contribute to CNS demyelination (Sadler etal., 1991). When the animals begin to recover fromthe clinical symptoms, the permeability of the BBBreturns to normal (Juhler et al., 1984), and then theseantibodies may no longer be allowed to pass the BBB,and staying in serum may have a role in the recoveryof the disease.

The amount of free cholesterol was not significantlyaltered, although this lipid, in addition to gly-cosphingolipids, is also enriched in myelin relativeto other membranes. Cholesterol esters are formedprincipally in later stages of the development of thedisease after the acute period occurred. This sud-anophilic material can be the product of phagocytosisof the myelin membrane by macrophages and residentmicroglia (Lampert, 1967; Shaw and Alvord, 1984;Sadler et al., 1991), process that it is facilitated byantimyelin antibodies (Sadler et al., 1991). Withrespect to the histopathologic alterations, the mostmarked changes were observed at 14 dpi, but some ofthem were still present until 40 dpi when the EAEanimals did not show any apparent difference withcontrols.

On the other hand, we have previously dem-onstrated that antibodies against MBP from animalswith EAE purified by affinity chromatography havethe property to recognize the neuronal protein syn-apsin I (Pedraza et al., 1988). Herein, these antibodiesto MBP that’ crossreact with synapsin account forabout 50°/0of the total anti-MBP antibodies and areonly present at 14 dpi coincidentally with the appear-ance of the paralysis symptom (Fig. 5(A) and (B)).Similar behavior was also observed when the T-cellpopulations that bind both antigens were quantitated

(De Santis et al., 1992; De Santis and Roth, 1996).These results indicate that after the acute period, thespecificity of the anti-MBP antibodies changes. Thecrossreacting population fade, and other kind of anti-bodies against MBP are raised. Neuronal involvementduring the development of EAE can be also inferredby searching the previously described deposits ofimmunoglobulins in spinal cord motor neuron cells(Roth and Obata, 1991): These deposits were seenduring and after the acute stage and then they dis-appear concomitantly with the total recovery of theanimal (Fig. 2, Table 1).

EAE is a complex phenomenon in which the clinicalmanifestations appear to be the consequence of vari-ous immunological factors. The new obtained evi-dences show CNS early changes compressingalterations of some specific myelin constituents. Dur-ing the following acute stage, besides these alterationsin myelin, changes involving neuronal structures arealso detected. At this respect, the axonal damage andfunctional changes of synapses previously described(Bieger and White, 1981; Kraft and Slimp, 1984;White et al., 1985, 1990), the herein observed depositsof immunoglobulins in the spinal cord motor neuronsand the antibodies and T-cells (De Santis et al., 1992;De Santis and Roth, 1996) that recognize MBP andsynapsincorrelate with the appearance of the paralysissymptom. In conclusion, we can postulate that earlychanges in myelin initiate a process of C.NS impair-ment. Then, the existence of immunological factorsagainst some epitopes in MBP that also recognize asynaptosomal protein might account, at 1east in part,for the axonal damage and disruption of the normalinterneuronal activity in EAE and lead, together withthe alterations in specific myelin constituents and theconcomitant CNS inflammatory process to theobserved hindIimb paralysis.

Acknowledgements—We thank Dr B. Maggio for his adviceand critical reading of the manuscript. This work was sup-ported in part by grants from Consejo Nacional de Invest-igaciones Cientificas y Tecnicas (CONICET), Consejo deInvestigaciones Cientificas y Tecnologicas de la Provincia deCordoba (CONICOR) and Secretarial de Ciencia y Tecnicade la Universidad Nacional de C6rdoba (SeCyT-UNC),Argentina. A, L. D. and N. W. S. are fellows from CON-ICOR and G. A. R. is a career investigator from CONICET.

REFERENCES

Autilio L. A., Norton W. T., Terry R. D. (1964) The prep-aration and some properties of purified myelin from thecentral nervous system, J. Neurochem. 11, 17–27.

Bieger D, and White S. R. (1981) Anatomical evidence for

604 D. A. Slavin et al

bulbospinal monoamine axon damage in rats with exper-imental allergic encephalomyelitis. Circ. Res. 18, 118–125.

Bradford M. M. (1976) A rapid and sensitive method for thequantitation of microgram quantities of protein utilizingthe principle of protein~ye binding. Anal. Biochem. 72,248-254.

Clemens C. S. and Fan K. (1979) 2’,3’-cyclic nucleotide 3’-phosphohydrolase activity in brain of acute experimentalallergic encephalomyelitis. Brain Res. 169, 621M24.

De Camilli P., Cameron R. and Greengard P. (1983) Syn-apsin I (Protein I), a nerve terminal-specific phos-phoprotein. I. Its general distribution in synapses of thecentral and peripheral nervous system demonstrated byimmunofluorescence in frozen and plastic sections. J. CellBiol. 96, 1337–1354.

De Santis M. L., Roth G. A. and Cumar F. A, (1992) Cellularimmune crossreactivity between myelin basic protein andsynapsin in rats with experimental allergic ence-phalomyelitis. J. Neurosci, Res. 31,4651.

De Santis M. L. and Roth G. A, (1996) Experimental allergicencephalomyelitis: characterization of T-lymphocytes thatbind myelin basic protein and synapsin. J. Neurosci. Res.43, 5G54.

Hashim G. A. (1985) Neural antigens and the developmentof autoimmunity. .1.Immunol. 135(Suppl), 838–842,

Juhler M., Barry D. I., Offner H., Konat G., Klinden L. andPaulson O. B. (1984) Blood–brain and blood–spinal cordbarrier permeability during the course of experimentalallergic encephalomyelitis in the rat. Brain Res. 302, 347–355.

Juhler M., Laursen H. and Barry D. I. (1985) The dis-tribution of immunoglobulins and albumin in the centralnervous system in acute experimental encephalomyelitis.Acta Neurol. Scand. 73, 119–124.

Kato S. and Nakamura H. (1989) Ukrastructural and ultra-cytochemical studies on the blood–brain barrier in chronicrelapsing experimental allergic encephalomyelitis. AcfaNeuropath. 77,455464.

Kraft, G. H. and Slimp, J. C. (1984) Electrophysiologicalmonitoring of experimental allergic neuritis and exper-imental allergic encephalomyelitis. In ExperimentalAllergic Encephalomyelitis: A Useful Model for MultipleSclerosis, ed. Alvord, E. C., Kies, M. W. and Suckling, A.J., pp. 449+53. Alan R. Liss, Inc., New York.

Kristensson K. and Wisniewski H. M. (1977) Chronic relaps-ing experimental allergic encephalomyelitis: studies in vas-cular permeability changes, Acta Neuropath. 39, 189–194.

Lampert P. (1967) Electron microscopic studies on ordinaryand hyperacute experimental allergic encephalomyelitis.Acta Neuropathol. 9,99-126.

Lassmann H,, Kitz K. and Wisniewski H. M. (1980) Struc-tural variability of demyelinating lesions in different mod-els of subacute and chronic experimental allergicencephalomyelitis. Acta Neuropathol. (Berl) 51, 191–201.

Linington C., Bradl M., Lassmann H., Brunner C. and VassK. (1988) Augmentation of demyelination in ratacute allergic encephalomyelitis by circulating mousemonoclinal antibodies directed against a mye-lin/oligodendrocyte glycoprotein. Am. J. Pathol. 130, 443–454.

Lyman W. D., Roth G. A,, Chiu F.-C., Brosnan C. F.,Bornstein M. B. and Raine C. S. (1986) Antigen-specificT-cells can mediate demyelination in organotypic centralnervous system cultures, Cell. Immunol. 102, 217–226.

Lyman W. D., Roth G. A., Brosnan C, F., Bornstein M. B.

and Raine C. S. (1988) Quantitation of antigen-specific Tcell-induced demyelination in vitro. J. Neuroimmunol. 17,175–180.

Macala L. J., Yu R. K. and Ando S. (1983) Analysis of brainlipids by high-performance thin-layer chromatographyand densitometry, J. Lipid Res. 24, 1143–1150.

Maggio B. and Cumar F. A. (1975) Experimental allergicencephalomyelitis: dissociation of neurological symptomsfrom lipid alterations in brain. Nature 253,364-365,

Maggio B., Cumar F. A., Roth G. A., Monferran C. G. andFidelio G. D. (1983) Neurochemical and model membranestudies in demyelinating diseases. Acta Neuropathol.(Berlin) 9(Suppl), 71-85.

Mizutamari R. K., Wiegandt H, and Nores G. A. (1994)Characterization of anti-ganglioside antibodies present innormal human plasma. J. Neuroimmunol, 50, 215–220.

Norton, W. T. and Cammer, W. (1984) Isolation and charac-terization of myelin. In Myelin, ed. Morel], P., pp. 147–195. Plenum Press, New York.

Pedraza L. T., Roth G. A., Cumar F, A. (1988) Identificationas synapsin of a synaptosomal protein immunoreactingwith anti-myelin basic protein antiserum. J. Neurochem,51, 413J120.

Prohaska J. R., Clark D. A. and Wells W. W. (1973)Improved rapidity and precision in the determination ofbrain 2’,3’-cyclic nucleotide 3’-phosphohydrolase. Anal.Biochem. 56, 275–282.

Raine C. S. (1984) Analysis of autoimmune demyelination:its impact upon multiple sclerosis. Lab. Invest. 50, 608–635.

Raine C. S., Hummelgard A., Swanson E. and Bornstein M.B. (1973) Multiple sclerosis: serum-induced demyelinationin vitro. A light and electron microscope study. J. Neurol.Sci. 20, 127–148.

Raine C. S., Traugott U., Farooq M., Bornstein M. B. andNorton W. T. (1981) Augmentation of immune-mediateddemyelination by lipid haptens. Lab. Invest. 45, 17*182.

Roth G, A., Maggio B., Monferran C. G. and Cumar F. A.(1978) Experimental allergic encephalomyelitis. Identi-fication of the myelin component responsible for theappearance of brain esterified cholesterol. FEBS Letts 86,29-32.

Roth G. A., Monferran C. G., Maggio B. and Cumar F. A.(1982) Central nervous system lipid alterations in rats withexperimental allergic encephalomyelitis and its sup-pression by immunosuppressive drugs. Lije Sci. 30, 859–866.

Roth G. A., Yu R. K. and Bornstein M. B. (1983) Chemicalanalysis of organotypic cultures of mouse spinal cord innormal, demyelinative, and nondemyelinative conditions.J. Neurochem. 41, 171t&1717.

Roth G. A., Roytta M., Yu R. K,, Raine C. S. and BornsteinM. B. (1985) Antisera to different glycoliipids induce mye-lin alterations in mouse spinal cord tissue cultures. BrainRes. 339, 9–18.

Roth G. A. and Obata K. (1991) Experimental allergic ence-phalomyelitis: dissociation of immunochemical and cli-nicopathological responses in two strains of rats.Neurachem. Int. 19, 213–220.

Roytta M., Lyman W. D., Roth G. A., Bornstein M. B. andRaine C. S. (1985) Preliminary analysis of cell and serum-induced demyelination in vitro using a syngenic system.Acta Neurol. Scand, 71,226-236.

Sadler R. H., Sommer M. A., Forno L. S. and Smith M. E.(1991) Induction of anti-myelin antibodies in EAE and

Time course of biochemical and immunohistological alterations 605

their possible role in demyelination. J. Neuro.rci.Res. 30,61G624.

Salvati S., D’Urso D., Conti Devirgilis L. and Serlupi Cre-scenzi G. (1986) Biochemical changes in central nervoussystem membranes in experimental allergic ence-phalomyelitis. J. Neurochem. 47,239-244.

Schiebler W., Jahn R., Doucet J. P., Rothlein J. and Green-gard P. (1986) Characterization of synapsin I binding tosmall synaptic vesicles. J. biol. Chem. 261, 8383–8390.

Shaw, C. M, and AIvord, E, C. (1984) A morphologic com-parison of three experimental models of experimentalallergic encephalomyelitis with multiple sclerosis. InExperimental allergic encephalomyelitis:A usefidmodelforMultiple sclerosis, ed, Alvord, E. C,, Kies, M. W. andSuckling, A.J., pp. 61–66. Alan R. Liss, Inc. and NewYork.

Tabira T. (1988) Autoimmune demyelination in the centralnervous system. Ann. N. Y. Acad. Sci. 540, 187–201.

Traugott U., Shevach E., Chiba J., Stone S. H. and RaineC. S. (1982) Acute experimental encephalomyelitis: T- andB-cell distribution within the target organ. Cell. Immunol.70,345-356.

Tsukada, Y. and Kurihara, T. (1992) 2’,3’-Cyclic nucleotide3’-phosphodiesterase: molecular characterization andpossible functional significance. In Myelin: Biology and

Chemistry, ed. Martenson, R. F., pp. 449%480.CRC Press,Boca Raton, FL.

Vandenbark A. A., Burger D. R. and Vetto R, M. (1975)Cell-mediated immunity in experimental allergic ence-phalomyelitis: crossreactivity between myelin basic proteinand mycobacteria antigens. Proc. Soc. Exp. Biol. Med.148, 1233-1236.

Wekerle H. (1993) T-cell autoimmunity in the central ner-vous system, Intervirology 35, 95–100.

White S. R., Vyas D. and Bieger D. (1985) Serotonin immu-noreactivity in spinal cord axons and terminals of rodentswith experimental allergic encephalomyelitis. Neuroscience16,701-709.

White S. R., Samathanam G. K., Bowker R. M. and Wes-sendorf M. W. (1990) Damage to bulbospinal serotonin-,tyrosine hydroxylase-, and TRH-containing axons occursearly in the development of experimental allergic ence-phalomyelitis in rats. J. Neurosci. Res. 27,89-98.

Yu R, K., Ueno K., Glasser G. H. and Tourtellotte W,(1982) Lipid and protein alterations of spinal cord andcord myelin of multiple sclerosis. J. Neurochem. 39, 46+477.

Zamvil S. S. and Steinman L. (1990) The T lymphocyte inexperimental allergic encephalomyelitis. Annu. Rev. Zmmu-nol. 8, 579–621,