Neuro~hem lnt Vol 20, No 3, pp 371-383, 1992 Pnnted m Great Britain All rights reserved

0197-0186/925500+000 Copyright CO1992 Pergamon Press plc

G A N G L I O S I D E S A N D R E G E N E R A T I O N OF THE G O L D F I S H OPTIC N E R V E IN VIVO A N D IN VITRO* H . RAHMANN. H . ROSNER. U . SONNENTAG a n d S. ESDERS Instltut fur Zoologle (220), Garbenstr 30, 7000 Stuttgart 70, Fed Rep Germany (Recewed8 July 1991. accepted I August 1991)

Abstract--One to forty days after optic nerve transection, goldfish received an i p injection of [-'H]prollne (proteins), 3HNAcGluc (ganghosldes) or [3H]thymldIne (DNA) After 1 or 2 days of incorporation, both optic systems were analyzed by biochemical and autoradiographical procedures In the regenerating retina an enhanced retinal mitotic activity, protein synthesis (up to 2-fold) and ganglloslde synthesis (up to 1 5-fold) was found Simultaneously, a transiently enhanced accumulation (up to 4 5-fold) of axonally transported protein- and ganghoside-bound radioactivity in the regenerating optic nerve stump occurred These regeneration-related proliferative and metabolic changes were found to be maximal at 6~8 days post lesion, but still measurable after 40 days Concerning the endogenous ganghoside metabolism, in the regenerating retina no obvious change in ganglioslde synthesis and composition could be observed, while in the regenerating optic nerve there was an enhanced accumulation of the ganglioslde GPlc Daily l.p apphcatlon ofa ganghos,de mixture from bovine brain (GMlx) or of the monoslaloganglioslde GM 1, did not alter significantly the degree and time course of the above regeneration induced metabolic changes or the regain of visual acuity Sprouting activity of goldfish retinal explants was found to strongly depend upon a conditioning lesion of the optic nerve, reaching a maximum 8 days after nerve transection This result strictly coincided with the profile ofmetabohc changes observed m VlVO Again, daily I p or I o Injection of exogenous ganghosldes d~d not influence the lesion induced increase of retinal sprouting actiwty However, in normal, not regenerating animals, a local , o injection of GMIx or GM 1 led to a significant enhancement of the "basal" sprouting activity, normally occurring after lesion of the retina after injection of 0 9% NaC1 This ganghoside related stimulation was maximal at low concentrations (3 #g/eye) and did not occur at high concentrations ( > 30 #g/eye) Injection of the phospholipld phosphatldylchollne or phosphatldylserlne had no or a slightly inhibitory effect, when compared to NaC1 controls These data suggest an involvement of ganghosldes in the complex process of induction of axonal sprouting

Tsujl et al., 1983, Leon et al., 1988), stimulation of n e u n t e o u t g r o w t h (Rolsen et a l , 1981, G o r l o et a l , 1983) or Induction of synapse f o r m a t i o n (Obata, 1977). In vn,o, exogenously administered ganghosmdes have been reported to aid n e u r o n a l repair by enhancing sprouting a n d relnnervatlon (Ceccarelh et a l , 1976, Sparrow a n d Grafsteln, 1982, G o n o et al., 1983, G r a f s t e m et al., 1983) or recovery from central injury (Agnatl et al., 1983, Toffano et al., 1983) However, there are also some investigations revealing no such ganghoslde m vwo effects (Beuttler et a l , 1980, Verghese et a l , 1982, Butler et al., 1987) Antibodies to ganghosldes have been s h o w n to Inhibit neurlte o u t g r o w t h from dorsal root ganglia (Schwartz a n d Spirman, 1982), from explants of gold*Parts of the results of this refereed paper were presented at the ESN Satellite Meeting Chmcal and Behavtoural fish retina ( S p l r m a n et a l , 1982), a n d the progress of Aspects o f Ganghostde Research, held in Magdeburg, 28 regeneration o f goldfish retinal ganglion cell axons 31 July 1990 The Satelhte Meeting was organized by (Sparrow et al., 1984) Dr H Schenk, The Medical Academy of Magdeburg, In spite o f this increasing n u m b e r o f reports o f Fed Rep Germany and Dr G Tettamanti, The Unin e u n t o g e m c a n d n e u r o n o t r o p h i c effects of exogenous versltv of Milan. Milan Italy 371

Gangllosldes are acidic glycosplngohplds of high molecular complexity a n d particularly a b u n d a n t in n e u r o n a l m e m b r a n e s (Yu a n d A n d o , 1980 ; Wlegandt, 1982; Ledeen, 1984). They undergo, b o t h qualitative a n d quantitative, changes d u r i n g d e v e l o p m e n t (R6sner, 1980, R 6 s n e r et a l , 1985, R 6 s n e r a n d R a h m a n n , 1987, Seybold a n d R a h m a n n , 1985a, b , Yates, 1986), suggesting a n i m p o r t a n t ( t h o u g h still undefined) role in n e u r o n a l differentiation, growth and synapse f o r m a t i o n There are n u m e r o u s m vttro studies d e m o n s t r a t i n g effects o f exogenously a d d e d ganghosldes as prolongatlon of cell survival ( M o r g a n a n d Seafert, 1979.

"~72

H

RAHMAN"~ ('I a/

ganghosldes, the "'behavior" of endoqenou,s ganghosides during regeneratmn has received little attention Yates and collaborators (Yates et a l , 19851 reJected trltlated glucosamlne into rat dorsal root gangha and found up to 3 times as much radmacnvlty m ganghosldes of regenerating sciatic nerve 4 days after crush In goldfish, lntraocular rejection of radmactlve precursors led to a comparable accumulation of radiolabeled proteins (Sbaschmg-Agler et a l , 1984, Larrlvee and Grafsteln, 1987) and glycosamlnoglycans /Coughhn and Elam, 1989) and to a change of the phosphorylatlon pattern of proteins (Larrlvee and Grafsteln, 1987, 1989) The aim of the present study was to further evaluate the role of ganghostdes dunng regeneration of the goldfibh optic system with respect to morphological, functional and biochemical events Furthermore, the influence of exogenously apphed ganghosldes on each of these parameters was investigated Thereby, unlike prewous studies (Grafsteln et a l , 1983, SbaschnIgAgler et a l , 1984, Sparrow et a l , 1984, LarrIvee and Grafsteln, 1987, 19891 lesion of the optic nerve was done by transectmn instead of crush, in order to ensure regeneraUon of all retinal ganghon cells Con-

cernlng the biochemical parameters both optic systems were equally (systematically) supplied with m tlated precursors in order to directly compare metabolic changes of normal and regenerating goldfish retina and optic nerve (compare Fig 1) In a further approach, the influence of e~ogenou~ ganghosldes on regeneration of the goldfish optic system was investigated under m vitro conditions Thereby, goldfish retina explants were cultured m a 3dimensional fibrIn-mamx, ensuring adhesion of all explants and allowing a simple quantification of single outgrowing neurons

EXPERIMENTAL PROCEDURES

Mater tal~

[~H]Prollne (sp act 322 GBq/mmol -= 8 7 Cl/mmoll, [~H]N-acetylglucosamme (~HNAcGluc, sp act 1658 GBq/mmol _--44 8 C~/mmol) and [~H]thymldme (sp act 247 9 GBq:mmol-= 6 7 Cl/mmol) were purchased from NEN, Boston, MA, Sephade,~ LH-20 from PharmacJa Free Chemicals, Plscataway, NY, Blo-Gel P2 from Blorad, Rlchmont, CA, trlcalne methanesulfonate (MS), fibrlnogen, Dulbecco's modified Eagle's medium (DMEM), pemcdlln, ~treptomycln and tetracychn from Sigma, Mumch, Ihrom-

Rc

Retlno I

Tr

i

/

Optic

Eye

\

nerve

>Te~um

\ I

.....[3H] Prohne, / 5HNAc

--

G[u cosomlne

Tc

~'~ ~

I Retmo Rr

Fig I The goldfish optic system control retina (R~), regenerating retina (R,), control nerve (N~), regenerating nerve (Nr), control tectum (T~), regenerating tectum (T,)

Ganghosldes and regeneration of the goldfish optic nerve bin from Hoffman La Roche, Grenzach, sterile petn dishes (q535 mm) from Greiner, Nurtlngen, tritium sensitive ultrofilm from Cambridge Instruments, Nul~loch, Soluene 100 and Hionic Fluor from Canberra Packard Company. Frankfurt and cryostat mounting medium from RelchertJung The ganglioslde mixture Cronassial (GMIx, containing 21% G M I , 40% GDIa, 16% GDIb, 19% GTIb, 2% GD3 2% GQ 1b) and the monoslaloganghoside GM 1 were a kind gift from the Fidla company All other chemicals were of analytical grade and purchased from Merck, Darmstadt, if not otherwise stated

Ammals Goldfish (Carasstus auratus, 8 12 cm) were purchased from Tropical Center Kohlhase, Neuklrchen and maintained under diurnal lighting at 28 + 2~C without antibiotics Surgical procedures Fish were anesthesized with 0 08% MS in cold water For transection of the optic nerve, the con lunctival membrane of the eye was dissected and the exposed nerve cut close to the back of the orbit (Fig 1) In some cases, both optic nerves were transected Ganghostde treatment Fish received a daily i p injection o f a ganghoslde mixture from bovine brain (GMIx, 50 mg/kg) or of the monosialoganghoslde G M I (30 mg/kg), starting one day before lesion Alternatwely, goldfish were injected lntraocularly (l o ) using 3 #g GMIx or G M l / e y e Control animals were either not injected or recewed a dally injection of 0 9% sterile NaC1 Determination of visual acuttv Binocularly deprived goldfish were tested each day in the "'optomotonc drum" according to Rahmann et al. (1979) black and white stripes of equal width, mounted vertically inside the drum were slowly rotated by an electric motor around the vessel containing the fish Becoming aware the rotating stripes, the fish tried to follow the pattern by swimming with or against the rotation direction (optomotoric reaction, OMR) or at least by eye nystagmus (optoklnetlc nystagmus, OKN) The visual acuity (minimum separable) could be determined (after successive narrowing of the width of stripes) as visual angle calculated from the width of stripes and the distance between the fish and the rotating stripes Determmatton of the proh[eratton rate o[ goldfish retma ganqhon cell~ Goldfish, whose left optic nerves had been transected 1, 3, 7 and 20 days before, received an intraocular (i o ) injection of 5/IC1 [3H]thymIdlne Into both eyes After 24 h of incorporation, fish were killed and Incorporation of the trltlated thymldlne into the D N A of mitotic retinal ganglion cells was visualized autoradlographically according to Rahmann (1968) Vtsuahzatton of the reqeneratton of the optic nerve by means of autoradtography One sided lesioned animals received an i o injection of 10 /~C1 [3H]proline into the regenerating eye Twenty-four hours later, fish were killed and tectal reinnervatlon was determined by means of autoradiography For quantitative evaluation, the tectal area m which radlolabehng could be found was

373

determined planimetrlcally in percentage of the whole relnnervated optic tectum In addition, the progress of radiolabeling in the optic fibres was determined densltometrIcally by measuring the sliver grain density in successive compartments of the corresponding autoradlograms The determinatlon of the optic layers within the optic tectum was performed according to Vanegas (1975)

Radtochemlcal labehn9 oj protems and 9anyhostdes For intense radioactive labeling of retinal proteins and 9angltostdes, 10 #CI of [3H]prolIne (proteins) and 20 #CI of 3HNAcGluc, respectively, were injected 1 o into the vitreous body of the left eye (optic system r, regenerating in the case ofleslonedanlmals, compareFIg 1) Forsystemlcallabehng, 15 /~CI [~H]prollne or 30 #CI 3HNAcGluc were injected lntraperltoneally (l.p.) After 24 h ([3H]proline) and 48 h (3HNAcGluc) of incorporation both optic systems were carefully dissected (compare Fig 1) and both retinae (Rr, Re), optic nerves (Nr, N~) and optic tecta (Tr, To) frozen separately at - 2 0 ° C In the case o f 3HNAcGIuc injection, corresponding samples from 3 up to 26 animals were pooled Quant~catzon of protem-bound radioactivity Single frozen samples (Rr. R~, Nr, N~, Tr, T~, Fig 1) were homogenized with 300 /d of ice-cold water Ahquots were taken for protein determination according to Lowry et al (1951) The residues were precipitated according to Rosner (1975), solubihzed with Soluene 100 and dispersed in 10 ml Hlonlc Fluor for radioactivity determination Ganqho 9tde extraction Pooled tissue samples (Rr, Re, Nr, N~, Tr, Tc from up to 40 animals) were homogemzed in 0.8 ml of ice-cold water Ahquots were taken for determination of protein content and total radioactivity (see above) The remainder were extracted according to Svennerholm and Fredmann (1980) To the crude ganghoside extracts concentrated ammonia or methanohc NaOH (0 1 M NaOH m methanol) were added In surplus The samples were kept closed for at least 20 h at room temperature in order to split alkali-labile bindings. They were then exhaustwely dialyzed or freed from low molecular contaminants by gelfiltrauon over Biogel P2 (Biorad) and further purified by chromatography on Sephadex LH 20 (Byrne et al, 1983) Ahquots of the ganghoslde extracts were taken for deterruination of total ganghoslde siahc acid by the method of Svennerholm (1957), modified by Mlettinen and TakklLuukkalnen (1959) and ganghoslde-bound radioactivity. Thin-layer chromatography ( TLC ) One-dimensional TLC was performed on silica gel high performance TLC (HPTLC) glass plates (Merck) by use of the dual solvent system of Rosner (1980) Ganghoslde spots were visualized with resorclnol reagent (Svennerholm, 1957) and quantified by densitometnc scanning (Rosner, 1980) For determining radioactivity of single ganghoside fractions, plates were developed as above and HPTLC glass plates were kept under a tritium sensitive film After 4 weeks of exposure, contact autoradiograms were developed and ganghoslde composition of autoradiograms was quantified by densltometrlc scanning as described above

H RAHMANN et al

374

180 ~-

Table 1 Explants were &wded into 5 "sprouting-classes". according to the number of axons/explant Each class of explants (% o f total explants) was multlphed w a h a factor The sum of the resulting 5 values was defined as "sprouting-index" Number of axons

Factor(J)

% explants :.]+xz+~3+~:a+% = 100%

0 1 5 6-20 25 50 > 50

0(I,) 3 (J2) 10 ( f 0 30 (/4) 75 ( f 0

x2 ~, x4 ~

~,

"'Sprouting-index"= y~ f , , ' ~ . ' 1 0 n/- %

I O0 50

-ran 4- •

'~\

+ not-lnlected control (n=4) • NclCI-,nject.ed ( n = 1 3 ) o GMIx- infected, (n=15)

~\\

__.~

\\\ ~o

-

>

1

05

]

I 8

I 10

•

I _, 12 14

Days

Cultwatton oJ goldfish retinal explants Goldfish retinal explants were prepared as described previously (Landreth and Agranoff, 1979) Explants were cultivated In a 3-dimensional fibrin matrix, consmtmg of 25 "mg/ml fibrmogen, 300 N I H - E thrombm/ml and supplemented with DMEM plus antibmtlcs Instead of serum, Ultroser G was used Explants were incubated at 3T C, 5% CO~

Quant~catwn of neunte outyrowth Quantification was performed by counting the number of fibres/explant, starting 15 h after explantation and scoring them every 24 h until day 4 For evaluation, explants were divided into 5 "sprouting-classes'" according to the following criteria explants without axons (0) I 5 axons (1), 6 20 axons (2), 21 50 axons (3), more than 50 axons/explant (4) To simplify the graphical representahon, the following sproutmg-mdex was defined eachclass ofexplants (% proportion of total explants) was multlphed with a dmtmct factor Sprouting-class 0 (without axons) with the factor 0, sproutmg-class 1 (1-5 axons) with factor 3, sprouting-class 2 with factor 10, sprouting-class 3 w~th factor 30 and sproutmg-class 4 with factor 75 The sum of these resulting 5 values was defined as sprouting-index (Table 1)

RESULTS AND DISCUSSION

Regam o f t,lsual acuity oJ the regeneratmg goldl~sh optic" system In the first a p p r o a c h we investigated whether or not exogenously applied ganghosldes m i g h t influence the functional recovery of wsual acuity in d o u b l e - b h n d e d goldfish The observation o f the o p t o k m e t , c nystagmus reflex ( O K N ) turned out to be a valid criterion for this survey Test antmals, prewously doubleb h n d e d were datly tested for their O K N as reaction to presented m o v m g s t n p e p a t t e r n (Fig 2) After 7 days o f regeneration, the first O K N s could be detected to very w~de stripes; wsual angles which could be resolved were a b o u t 4 0 - 6 0 U The day-by-day Improve-

I 16

efter

r 18

I 20

1 22

24

I 26

I 28

transectfon

Fig 2 Regain of visual acuity (minimum separable) of untreated, NaC1- or GMlx- 150 mg/kg) treated goldfish 8-28 days after transection of the optic nerves

ment was extremely fast m the following 4~5 days, leading to the resolution o f angles between 1 a n d 2 after only 12 days following transection of the optxc nerves This steep Increase of functional recovery comclded with the morphological remnervaUon of retinal fibres, beginning after 8 days a n d e n d u n n g for up to 12 days p c (compare Fig 4) Thereafter, the stepwlse i m p r o v e m e n t o f wsual acmty was heawly retarded, resultmg m the already very precise resolution of stripes creating visual angles between 0 7 0 9 after a b o u t 16 davs and 0 6 - 0 7 ' after 27 davs when testmg was fimshed. These angles of previously b h n d fish were very s~mdar to those measured in a collectwe o f unles~oned controls, achieving angles of a b o u t 0 4', which was taken as the basehne o f the & a g r a m . The slow i m p r o v e m e n t of visual acuity after the first 2 weeks of regeneration xs strong ewdence for the long lasting effort in refinement of retmotoplc projection This had been already shown by electrophyslological m e a s u r e m e n t s which stated early, b u t lmpreo s e synapttc contacts that were c o n t m u o u s l y t r a n s f o r m e d into precise patterns after several weeks ( M a t s u m o t o et al, 1987) or even m o n t h s (Emele and Schmldt, 1988) O u r investigations support the idea t h a t increasing visual acuity was a c c o m p a m e d by the retraction of a b u n d a n t or mmgmded optic axons between 16-30 days p c (compare Fig 4 ) Daily ] p lnjechon o f exogenous gangllosldes ( G M l x ) led to no further i m p r o v e m e n t of visual acmty (Fig 2) This ts in contrast to data from G r a f s t e m et al (1983), who reported a slgmficantly reduced time being necessary to estabhsh a so-called "startle reactton" which, as O K N , ts a behavtoral p a r a m e t e r to detect functional recovery from optic nerve lesion

375

Ganghosldes and regeneration of the goldfish optic nerve

ganghon cells was enhanced during regeneration and reached Its maximum about day 8 with a 2 5-fold + NoClcont increase of [3H]thymidme incorporation into the NoClexp DNA (Fig 3). 30 {3 GM~xcont Up to now, retinal mitosis were observed to occur • GMixexp t only in the periphery, being the matrix zone of retinal I proliferation (Maler and Wolburg, 1979 ; Easter et al., m /x\ ~ 20 1981 ; Wolburg, 1981). In this study, retinal mitosis occurred over the whole retina with special preference, however, in the periphery. Because the enhanced pro1o i % hferatlon occurred only between 4 and 18 days p.c., o ~0 Q; there were only a few additional fibers (about 1% of all retinal ganghon cells) expanding from these newborn 0 t I I I cells and supporting the reinnervatlon of the optic 2 4 8 21 tectum. Days a f t e r t r o n s e c t J o n Again, dally l.p. apphcation of exogenous ganFlg 3 Proliferation rate of goldfish retina ganglion cells gliosldes resulted in no further enhancement of the determined 24 h after 1.o injection of 5 #C1 [3H]thymldme observed regeneration related increase in mltogemc into both eyes actwlty (Fig. 3). o

contcont

•

contexp

ProhJeratton rate of normal and regeneratmg goldfish retma qanghon cells

Ttme-course of re-ingrowth of regenerating optzc nerve fibers into the optic tectum

The purpose of this invesUgatlon was to investigate if regeneraUon of the optic system ~s due to an enhanced sproutmg actwlty of already existing retina ganglion cells or ff an addmonal proliferation supports the recovery of the optic nerve Comparison of the [3H]thymldme incorporation into the retina of normal and regenerating goldfish clearly demonstrated that the mitotic actiwty of retina

In a further series of experiments the influence of exogenously apphed ganghosldes on the morphological regeneration of the optic tract of goldfish was investigated by means of autoradlographical procedures With regard to this, silver gram density was measured along the rostrcr-caudal axis in the optic tecta of normal and regenerating goldfish (Fig 4) In unop-

65 60 T

55 A

.


operated

control

n= 11

20

03

'15 10

~

5

B

t

0 o Rostrol

1

3

4

5

6

7

8 Coudol

TectoI

sUbdlvlslons

Fig. 4. Densltometric registration of regenerating optic nerve fibers re-entermg the 8 progressive subdivisions of the optic tectum, determined 24 h after 1o Injection of 10 #CI [~H]proline Dashed hne sdver grain density in the optic tectum of unoperated control animals

376

H RAHMANNet

crated goldfish, the radioactwlty was homogenously distributed over all optic layers of the contralateral tectum Unspecific (blood-derived) background staremg reached a value of about 27% (Fig 4, dashed line), reflecting the normal protein metabolism in the retina Lesioned goldfish achieved comparable denSitles after 30 days of regeneration, indicating similar metabohc actwmes like normals As can be followed from Fig 4, the silver gram density was s~gnlficantly enhanced (up to 2-fold), especially after 10 days p c. in the rostral tectum These data demonstrate that the bulk of regenerating opUc nerve fibers entered the rostral tectum between 8 and 10 days after nerve transection The elevated densmes above the level of unoperated controls can be regarded as a d~rect consequence of the transiently enhanced protein metabohsm in the regenerating retina (compare Fig 7) Since not all outgrowing fibers expanded across the enUre tectum, caudal values were lower After 16 days IFlg 4), a homogenous density occurred at all ~ltes

65 60

(a) 8

~

55

~

50

within the optic layers, although the intensity was lower than after 10 days An Intermediate level between normal and early-regeneraUon condmons was achieved, reflecting successful remnervatlon Daily apphcahon of exogenous ganghos]des prowded only minor effects on the above-mentioned pattern (Fig 5) Eight days after transecUon, all rejected animals (NaCI as well as ganghoslde treated fish) showed h]gher levels of rad]oactiwty than untreated controls, although m the &stal tectum differences became undlstmgmshable due to h~gh standard dewaUons Since there was no significant difference between the GM1- and NaCl-mjected ammals, the elevation ofganghoside rejected fish vs untreated controis (P < 0 005) cannot be regarded as a specific ganglioside effect, but must be interpreted as a regeneration stimulus due to the injections themselves The latter aspect became more obwous when results from the planImetnc mesurements, which gave insight into the number of regenerating axons, were added

65

days p c • not- injected control [] NoCl - r e j e c t e d c o n t r o l

45 ~ 4o ~ 35 -T n o t o p e r a t e d

30

~

25

control

~ 2o ®

> = o9

15 10 5

1 Rostral

2

3

5

Tectal

65

(b)

6

0

7 8 Caudal

65

10 days p c

50 45

4#

~ 35 3O

25

~'

20

%

5

I 20

L 30

Tectal

i 40

50

Subdivisions

_

3

I 60

]3 C

(dl

60 55 50 45

0 80 Caudal

50

I 4

J 5

I 6

I 7

8 Caudal

subdw~sJons

days p c

40 35 30 25

E~ 2 0 15 > 10 09 5 0

>

L O0 10 Rostral

I 2

Tectal

55

~

__ I 0 1 Rostrol

subdw~s~ons

6O

~

16 days p c

40 35 30 _c 2 5 o 20 25 > 10 bq 5

0 0

(c)

60 55 50 45

~

c

al

1

I 2

I 3

~ 4

; 5

Rostral

Tectal

6

I r 7 8 Caudal

subdw~s~ons

Fig 5 Influence of dally l p rejected GMlx (50 mg/kg) or GMI (30 mg/kg) on the silver grain density m 8 progresswe subdivisions of the optic tectum, determined 24 h after mjecuon of l0 #CI [~H]prohne Control ammals were not mlected or recewed a dally mlectlon of 0 9%, Na('l

Ganghosldes and regeneratmn of the goldfish optic nerve We could show, that with successtve regeneration times, the tectal area was increasingly occup]ed by radmlabeled fibers which entered the tectum from about 8 days p c. on (Fig. 6) They expanded across the tectum up to 16 days p c. and finally exceeded the "area" of normal tecta above 30-45% This expanstun, which ~s due to an excessive productmn of retinal fibers was also shown by Meyer (1980), Kagejama and Mayer (1988) Interestingly, fiber-overshoot achieved ~ts maximum after 16 days and kept this level nearly unaffected for up to 30 days post nerve cut, a t~mepoint at wh]ch the protem and ganghos~de metabohsm of the goldfish optic system already turned towards normal values (see next secUon) From these data ~t can be concluded, that material for regeneratlonal sprouting ~s excessively metabohzed during the first week after nerve transection, being subsequently accompamed by an overshootproducUon of retinal axons, which became retracted after longer regeneraUon periods (Meyer, 1980, Rankm and Cook, 1986 ; Stfirmer, 1988) The pharmacological treatment with ganghosades turned out to s~gnificantly increase the number of regenerating fibers which reached the tectum after 8 days of crush (Fig. 6) But again, this elevatton above the values obtained from untreated controls was also ach]eved after lnjectmn of NaCI alone, providing no specific effect of the ganghos~de treatment

CO 0D x

I~

n ~

--

5O

Not mlected control NoCI- mlected control G=M=x- mjectecl fish GM1- inlected fish Not operated controt

-- '~ 4.8

__

.~

9 g

o ® o

20

0

8

10

"]6

30

Days after optic nerve "transection o ®

Fig 6 Influence of dally 1p injected GMIx (50 mg/kg) or GMI (30 mg/kg) on the regeneration of the optic nerve, In&cared by the planimetrlc measurement of the radlolabeled optic layers in comparison to the whole optic tectum, determined 24 h after i o mjectmn of 10/~CI [~H]prolme Control animals were not injected or received a daily i p rejection of 09% NaC1 * P < 0 0 1 , * * P < 0 0 0 1

377

Synthesis and axonal transport oJ proteins and gan9hosldes in normal and regeneratm 9 9oldfish opttc system To investigate regeneration related changes m protem- and ganghoside metabolism goldfish, whose optic nerves had been cut 1-39 days before, received an i.p. mjectton of either [3H]prohne or 3HNAcGIuc After 24 h ([3H]prohne) or 48 h (3HNAcGluc) from each ammal the retinae and the optic nerves were removed (Fig 1) and analyzed separately for incorporated radmacUvlty (see Experimental Procedures) From these values for each compartment of the optic system IpSl- ( = injected, regenerating) to contralateral (=non-reJected, control) ratios were formed and demonstrated m Fig 7 As expected, control animals showed a raUo of about one, both for the retina and the optic nerve, after systemtcal mjecUon of the radioactive precursors, demonstrating an equal supply of both optic systems with the radioactive material. Looking at the retina of regeneratmg fish [Fro 7(a, c)], a transient increase of the raUos R / R e (up to roughly 50%) was observed between 2 and 10 days after nerve transection Th~s transient time course of regenerationrelated enhancement of radloacUvlty incorporatmn into retina ganghon cell proteins and ganghosides was more clearly indicated by the values obtained for the optic nerves [Fig. 7(b, d)]. In the regenerating optic nerves, up to 5-fold higher protein- and ganghoslde bound radmactlvlty was found as compared to the contralateral intact nerve. The enhancement was seen already 2 days after nerve transectmn, reached a maximum after 8 days of regeneratmn and was still measurable after 30 and 40 clays These data clearly show that there is a regeneration-related enhancement of synthesis and axonal accumulation of retinal ganghon cell proteins and ganghosldes. The degree and time course was s~mllar for both, proteins and ganghosldes and reached a maximum at day 8 after nerve transection Slmdar results were obtained by Sbaschnlg-Agler et al. (1984), who found 8 days after nerve crush a 4-fold enhanced incorporation of ra&oact~wty into proteins of the optic tract, but an 8-fold increase m ganghoslde-bound radloactlwty. We also found a higher accumulation of radioactivity m ganghosldes as compared to proteins when the precursors were reJected l o , but no &fference, when both eyes were equally supphed wath l p reJected radioactive precursors (Sonnentag et al., 1992) Since by i o. injectmn an equal supply of both retinae can hardly be achieved, we think, that a systemlcal labehng, for instance by an ~ p mlectmn ~s more suitable m order

~78

H RAHMANNet a/ Ret=na R~/R c [3HJProhne J p

3HNAcGluc i p

(a) 2

__.

(c) 2

F,

III1 1 c~ 2

4

6

8 I0

'~2 20 3 0 4 0

o111

8

c°

12'

20

30

40

O

~'~ =~ I~ o

Nerve N~IN c

(b)

c~

(d)

2

4

6

8

IO 12

20 30 40

c~

[ ' 1 Controls (c e)

8

~2

20

30

40

Days after transec~mn

Fig 7 Synthesis and axonal transport of proteins and ganghosides m the regenerating goldfish optic nerve and retina, following an i p rejection of [3H]prohne or 3HNAcGluc Ganghoslde treated animals received a daily ~p injection of 50 mg/kg, starting 1 day before nerve transection Control animals received no injection o r a daffy injection of 0 9% NaC1 Unoperated control animals (c*) showed in all experiments a ratio of about one (continuous hne) Data represent mean + SD from 10 27 ammals/expermlental group

to observe one-sided quantltauve changes m this type of experiment Dally Lp rejections of 0 9% NaC1, 50 mg/kg of a ganghoslde mixture (GMlx) or, m one case, 30 mg/kg of the monosmloganghoslde G M I &d not alter significantly the raUos R J R c and Nr/N~.

Ganghos Me ~omposztton o / t h e normal and reqeneratmq qoldfish retma and optw nerre

The transiently increased formation and axonal accumulation of gangliosides in regenerating goldfish retinal ganglion cells during regeneration raised the question, whether this metabolic enhancement includes all ganghosides or is restricted to distinct ganghoslde types To check this, unoperated controls as well as 6 and 28 days regenerating animals got an 1 o injection of 3HNAcGIuc. After 2 days of incorporation the retinae and optic nerves from the control, respectively regenerating s~de of 40 animals per series were pooled separately for ganghoside analysis and radloact~wty determination (see Experimental Procedures) Figure 8(a) shows that there were no remarkable differences of the ganghoside composition [2'o distribution of ganghosJde-smhc acid) between the

normal and regenerating retinae Likewise, the radioactivity d~trlbutlon did not differ significantly [Fig 8(b)] These data indicate that neither the retinal ganglloslde composition nor the synthesis rates of individual ganghosldes was altered after cutting the optic nerve In Fig 8(c, d) corresponding data obtained for the opUc nerve are given Again, no striking differences of the pattern between normal and regenerating nerves were obtained, except about an 80% higher proportion of G P l c in the regenerating nerve The finding that GPIc, although a minor ganghoslde of the goldfish optlce nerve, was elevated during regeneration by more than 80% is of interest G P l c possesses, like G Q I c , the major ganghoside of the goldfish brain (Tanaka et a l , 1989, Sonnentag et a l , 1~92), an epitope, which is recognized by the monoclonal antibody Q211 ( R o s n e r e t a l , 1988 : Greis and Rosner, 1990) Furthermore. G P l c was found in embryonic chicken brain (Rosner et al., 1985) and rat brain (Rosner et a l , 1988) to be transiently expressed by migrating and growing neurons In goldfish brain, the c-pathway polyslaloganghosldes G P l c and G Q l c are abundant throughout life and it may be speculated, whether this feature is one of many preposmons

3HNAcGlu¢lo

(o)

$

(C)

R e t =na

Nerve

30 -

g i o o w

20

10

10

w

o

Controls

]

8 days

]

3 0 days

d)

(b) 30

3O

Oo

20 > o

]

I0

g: 0 Plc

Qlb /c

Tdb /c

D2 Dlb

x DIo

D3 MI

M2

M3

Plc

Olb /c

T1b /c

[Ylb

x DIo D2

D5 MI

M2

M5

• P