The time course of retrograde transsynaptic transport of tetanus toxin fragment C in the oculomotor system of the rabbit after injection into extraocular eye muscles

A.K.E. Horn and J.A. Büttner-Ennever

Institute of Ncuropatbo!ogy, University of Munich, Thalkirchncrstr. 36, 0-8000 Munich 2, Federal Rcpublic of Gcrmany

Received Oetober 6, 1989 / Acceplcd March 19, 1990

Summary. Thc aim of this study was to dcterminc the optimal surviva l time for labelling those neurons that monosynaptically terminate on cXlraocular motooeu­rons, i.c. tbc prcmotor neurons, after an injcction of tetanus toxin fragment C, aretrograde Iranssynaplic tracer substance, into the cye musclc of the rabbit. Con­centrated fragment C was injected intc the inferior rectus cr inferior oblique muscle aod delecled imrounocyto­chemically in the brain after survival times of 8 h, 17 h, 2 d, 3 d, 4 d, 5 d , 6 d, 8 d and 12 d. lmmunorcaetivity was eonfined to granules within motoneuronal and pre­motor neuronal eell bodies, but beeamc assoeia tcd with punctate profiles outlining the somata with longer sur­vival times. The strongest and most eonsistent labelling of premotor cell bodies was seen after 4 days survival time. The transsynaptie labelling pattern was shown to vary for individual premotor palhways.

Abbreviations: III ocuJomotor nudeus, IV trochlea! nudeus, Vmes mesencephalic trigeminal nudeus, Vmt motor trigeminal nudeus, VI abduccns nudeus, VIacc accessory abducens nudeus, VII faeial nuclcus, BC brachium eonjunctivuffi, co cochlear nuc\cus, CR TCS­tiform body, d dentate nuc\eus, DAß diamino-benzidine-tetrahy­drochloride, HRP horseradish peroxidase, iC interstitital nudeus of Cajal, iv inferior vestibular nudeus, Ignd latcral geniculate nudeus dorsalis, Ign". lateral geniculate nucleus ventralis, Iv lateral vcs­tibular nudeus, mgn medial geniculate nudeus, MLF medial lon­gitudinal fasciculus, mvp medial vestibular nudeus pars par­vocel1ularis, mv., medial vestibular nuc\eus pars magnoccllularis ( = vcntral part of thc Iv), N III oculomotor nerve, NV trigeminal nerve, NVll faeial nerve, NVIII vestibular nerve, PC posterior com­missurc, pg periaquaeduclal geey, ppH nudeus praepositus hypo­glossi, riM LF rostral interstitial nudeus ofthe medial longitudinal fasciculus, m red nuc1e us, sc superior colliculus, sn substantia nigra, so superior olive, 5V s.uperior vestibular nuc1eus, sv< superior ves­libular nuc1cus contralateral, SV; superior vestibular nucleus ip-5ilateral, TR tractus retroflexus, Y Y -group, 7.i 7.0na incerta

Offprinl requests IQ' J.A. Büttner-Ennever,c/o Institute of Physiol­ogy, PcuenkoferSlr. 12, 8000 Munich 2

Key words: Tetanus toxin - Transsynaplic traeing - Sur­vival time - Oculomotor system - Rabbit

Introduction

After injtx:tion into a musc1e, tetanus toxin has bcen shown to move retrograde lranssynaptieally into presyn­aptic terminals or the moloneurons (Schwab and Thoenen 1976 ; Priee et a1. 1977 ; for review: Mellanby and Grecn 1981; Wellhörner 1982). Some rragments of tetanus toxin, such as BIIb (Bizzini et al. 1977), or C (Helting and Zwisler 1977), have the same transport properties as the intact toxin, but are themselves not toxie. These ean be used to label those neurons that monosynaplically tcnninatc onto motoneurons, called premotor neurons, without toxic side-effects in studies involving 10ng survival limes. Tbe main advantage of using a transsynaptic tracer sueh as tetanus toxin rrag­ments ralher than using simple retrograde tracers such as horseradish peroxidase (H RP) to loeale premotor neu­rons is that fragment C ean be injtx:ted into a muscle, ralher than centrally in the motor nucleus, wh ieh is less speeific. One disadvantage is that the interpretation of the pattern of retrograde labelling in tbe motoneurons and premotor neurons is difficuh, since the loeation or label changes rapidly with survival time (Evinger and Erichscn 1986; Fishman and Carrigan 1987).

Non-toxie tetanus toxin fragments have been shown 10 label neuronal eell bodies retrograde transsynaptically in the oculomotor (Büttner-Ennever cl al. 198 1; Evinger and Eriehsen 1986) and sympathetie system (Manning et al. 1987; eabot el al. 1987), but up to now there has been no study of the pattern of labeUing in premolor neurons and its ehanges with time. Sinee monosynaptic connec­tions in the oculomotor system rrom prcmotor areas to motoneurons have been fully dcseribed (Graybiel and

Hartwieg 1974; Steiger and Büttner-Ennever 1979; for review: Evinger 1988; Büuner-Ennever and Büttner 1988; Highstcin and McCrea 1988), we ha ve chosen it as a suitable system in which to investigate the time course of transsynaptic transport of tetanus toxin fragment C.

Thc aim of this study is to investigate the pattern of labelling in motoneurons and premotor neurons after an extraocular eye muscle injection with fragment C, and to determine the optimal survival time for the retrograde transsynaptie labelbng ofpremotor neurons in the oculo­molor system. We chose to inject the inferior rectus or inferior oblique eye muscle of the rabbit whieh move the eyes in the vertieal plane. The results of this work also provide data on the localization of premotor neurons controlling vertical eye movements in the rabbit.

Methods

Five pigmented (Chinchilla) and seven albino (New Zealand) rab· bits were used in this study. Prior to injcclion the fragment C (Cal­biochem) was eonccntrated by ultr'a.filtra.tion in a microconccntra­tor (Amicon; 30000 molecular weighl eutot1) to at most 15% as limited by thc 34° angle of the fixed-angle rotor of the ccntrifuge. The final concentration of the fragment C solution was ca1culated from the cnd-volumc. For eye muscle exposure thc rabbits were anaesthetilCd with Ketamin (35 mg/kg) and Rompun (5 mgJkg) intramuscularly. The eyc musclcs WCrt exposed by retcacting thc eye1ids, coUapsing the cyc ball, and making a conjunetivaJ ineision. All animals ceceived an injection of 15 111 concentrated fragment C into the right inferior rectus or inferior oblique eye musclc. To prevent intraorbital spread of fragment C, the muscle to bc injcctcd was enelosed by a small piece of plastie film prior 10 injoclion: in some cases the injectioD hole made by the syringe in thc muselc was locally eaulerizcd as an additional precautioD. After survival times of8 h, 17 h, 2 d, 3 d, 4d, 5 d, 6 d, 8 d and 12 d the rabbits were deeply anaesthetizcd and transcardially perfused with saline (37° C) followed by 4% para formaldehyde in 0.1 M phosphate bufTer (pH 7.4). Thc bra.ins wcre removcd from the skull_ poslfixed for another 5 h at 4° C and then storcd either in 0.1 M phosphate buffer for Vibralome sections or in 30% sucrose al 4° C for fr07.cn scctions. The brainstem was tral1sversely cut at 35-40 I-Im. The free-Hoating scetions wert immunocytochemieaJly treate<! according 10 the avidin-biotin peroxidase method (Hsu Cl al. 1981) using a mono­elonal antibody (now avaiLable from Boehringer Mannhcim) that had bcen raiscd against fragment C (Evinger and Erichsen 1986) al a dilution of I : 3000. The avidin-biolin peroxidase complex was visualized by a 15 min ineubation in 0.05% diamino-ocm;idine­tetrahydrochloride (DAB) and 0.01" H1ü l . Aller mounting and drying the scctions were osmified in a 0.05% osmium tetroxide solution for 20 seconds before dehydration. A modifioo DAB-rcae­tion in the presencc of 0.6% Ni l-> in acetale buffer (pH 6.0) (Han­oock 1984) was proved 10 intensify tbc immunocytoehemical slain­ing signifieantly and it was used in some eascs for the visualization of the weak transsynaptic label. A scries of every fifth scction was counterstained with 1 % Cresyl violet. The tissue was cxamincd and photographed under brightfield with a light mieroscopc. Labclled cells were plotted onto camera lucida drawiDg5 by using a drawing tube thaI was eonneeted to the microseope. Tbc nuelcar boundaries were drawn from Nissl-Slained sections and the nomenelature for the vestibular nuelei was taken from Epcma el al. (1988).

Results

Motoneurons

After an injection of fragment C into the inferior oblique or inferior rectus muscle tbe motoneuron pools of these

muscles in the ipsilateral oeulomotor nucleus (J ll) were strongly labelIed. Additional retrograde label was found in the motoneurons of the medial rectus musc1e in the ipsilateral III, in the ipsilateral abducens nucleus (VI) and aecessory abduccns nucleus (Vlacc) eontaining motoneurons of the lateral reetus and thc retraetor bulbi muscles, and in the motoneuron subgroup of the faeial nucleus (V I I) that supplies the orbicularis oculi muscle of the injeeted side (Fig. 3). The spread of fragment C proved very hard to avoid as judged from the motoneu­ronal labclling.

Tbe labelling pattern of all the motoneurons ehanged rapidly with survival time: after 8 h single motoneurons were distinctly labelled. The DAB-reaetion produet was located in the cytoplasm - associated with granules -while the nudeus was free of staining (Figs. la and 2a) . Thc amount of filling within the cell bodies and proximal dendrites inercascd over tbe next JO h, at wb ich time a diffuse label was observed in the surrounding neuropil, tbat could not be correlated with neuronal elements using light microscopy (Pigs. Ib and 2b). In preparations with survival times longer than 2 days tbc fragment C immu­noreactivity was no longer restricted to neuronal somata, butin addition was seen associatcd with punetate profiles outlining the cel! bodies and dendrites of the motoneu­rons (Figs. Ic-f and 2c, d). At this stage the staining of the oculomotor nudeus lOok on a rcticulated ap­pearance, whieh did not change even with the longest survival time in our study, 12 days, except that the inten­sity decreased slowly witb time. Motoneurons containiog no immunoreaetive granules eould be fouod within the oculomotor nucleus after survival times longer tha n 4 days (large arrows in Pigs. 2c and d).

Premotor neurons

After at least 2 days survival time a weak immunoreactiv· ity was fouod in several additional areas that are known to send monosynaptic alTerents to the motoneurons of vertical eye muscles: the contralateral y-group, the superior vestibular nuclei (sv) of both sides, the lateral bo rder region of the parvocellular medial vestibular nu­deus (mvp) and the magnoceHular part of the medial vestibular nudeus (mv",) mainly contralateral, the ip­silateral interstitial nucleus or Cajal (iC), and the rostral interstitial nucleus of the medial longitudinal fasciculus (riMLF) in tbc rostral mesencephalon. In addition trans­synaptie label was found within thc contralateral VI presumably associated with internuclear neurons. Figure 3 illustrates the loeation of alllabelIed premotor neurons after 4 days survival time fo llowing an inferior reClus muscle injection with fragment C.

1t was striking that tbe staining pattern diffcred wide­ly between premotor regions. For instance the most im­pressive transsynaptic labelling was always seen in tbe y-group, while tbc iC contained ooly very few weakly stained cells. For the detailed description of the t rans­synaptic labelling pattern 3 prcmotor structurcs have been chosen as examples: 1. tbe ipsi lateral sv, 2. tbc eontra1atera1 sv, 3. thc ipsilateral riMLF.

a 8h b 17h

.. . .

c 2d

tt~· .

", ' . . ,kU , i\. '" ~ ' .... ' . "

d 4d e

Fig. la- f. Photographs of transverse sections through thc ocuJo· motor nuclcus \0 show thc pattern of labeHing of the motoneurons and neuropil after survival times of a 8 h, b 11 h, C 2 days, d 4 days, e 6 days, f 12 days. a Distinctly labelle<! motoocurons afe visible

Ln tbe sv 01' both sidcs retrograde lranssynaptically labelIed neurons \'lIerc fi rst faund after 2 days. While immunoreactive neurons contralateral to the injected sidc were located within the dorsal part of thc sv, the labelJed cells ipsilaleral were distri buted throughout thc central parts of thc nucleus. After 2 and 3 days thc labelled neurons in thc sv of balh sides showed a similar but weak staining pattern, which rescmbled the pattern

6d f 12d

after 8 hours sUfvival time. b-d There is a gradually incrcase in labcUing of the neuropi! with survivaltimes up to 4 duys. e, f The slaining intcnsity ofmotoneurons und neuropil decreascs. Calibra­tion = 500l!m

of direct labclling of motoneurons after an 8 h time period (Fig. 4a and b compare with Fig. 2a). The cells contained immunoreactivc granules within thcir somata aod pro,umal dendrites. This type of staining pattern did not change in tbe conlralateraJ sv with longer survival limes. After 4 and 5 days their somata were still distinctly labelIed, aod ooly few immunoreactive puncta were fouod in the surrounding neuropil (Fig. 4d and g). A dif-

a 8h

Fig. 2a--d. High power magnification of the rnoloncuronallabelling in the Ql;:u1omotor ßucleus with fragment C after different survival limes. Il Survival time of 8 h, thc DAB-reaction producI is asso­ciatcd with granules wi thin thc cytoplasm, whi1c the nucleus is free. b Surviva! time of 17 h, a diffuse label is present in the ncuropil. c Survival time of 4 days, thc fragment C immunoreaclivity is 00

longer reslricted 10 somata (Iarge arrows), but associatcd with punetatc profiles (small arrows) oUllining the cell bodics aod den­dritcs. d Survival time of 6 days, "empty" motoncurons ([arge arrows) are associated with labelIed puncta (small arrows). Calibra­lion =30)J.m

70

= 57

"

o

12

8

mm

}'ig. 3. Drawings taken from transverse scctions of a rabbi! brain­stern 10 show thc distribution of labelIed motor aod prcmotor neurons after an injcclion of fragment C into thc right inferior rectus muscle; 4 days survival time. Thc distribution or small dots indicatcs thc punclate labdling ofprcsynaptic terminals. Fordarity,

63

(0 11..,, ' • . , v ...... ,?

,

rostral

..,(v;· ~)--~ ~" \

o

thc puncta around labcllcd cell bodies (c.g. Fig. 4e) are omitted. Each retrograde Lranssynaptically labclled premolor neuron is rcprc­scnted by a large dot. ConScculivc section numbers are 200 firn apart

SYc

a

2d

d

4d

9

~. ~\ • ,fi!:

5d

b

e

" . ~.... " ,. · .

0' .

h ,'t-I , , , .. · .

, '

"jf ., ~".

, ' · ; . "'-=' , .

SYj

.'

.. ...,- ':.. -,; '"

, , .

~ .-,", "" ., ,, ' . ,f' .. . ~" ... .... . ~ ..

f :" , . ,

··"t' '-~:. , ! ,

~ .. " ~ .. ,

.. .h.

" 1

riMLF

c

•

\

\' '

'"~

", , • 1'1 "',

l'~ J"" .,,'. !", ... ~ ... . -~

.. '. •

" ", '

" ) , • • •

I'ig. 4a- i. High power magnificalion of labcUcd premotor neurons in the superior vestibular nudeus contraJateral (sv.) a, d, g; in tbe sv ipsilateral (sv) b, c, h j rostral interstitial nudeus ofthc medial longitudinal fasciculus (riMLF) c, r, i , 10 show thc changes of the

,.-"

, .

pattern of labeHing after survival limes of 2 days in a--ci 4 days in d-f ; 5 days in g--i. Thc strongest overall Iranssynaptic cclilabclling is see!! after 4 days. Note the difference in thc palIern of labclling in SVI and sv. after 4 and 5 days survival time. CalibratioD = 30 Ilffi

fercnltimc course of the staining pattern was seen in the ipsilateral sv. The eell somata becamc more weakly stained, and after 4 and 5 days they were only faintly visible, whereas numerous immunoreaetive puneta in the surrounding neuropil gave thc nudeus a reticulated ap· pearanee (Fig. 4e and h). Evell after 12 days a few im­munorcaclive pune la were observed in the ipsilateral sv, whi le in the contralateral sv no immuoorcaetivity was de­tccted after more lhan 6 days.

Retrograde transsynaptically labclled cells were fouod in thc riMLF exelusively ipsila teral to the injected side. The immunoreaetivilY was wcak and confined to numerous graoules v.'ithin the somata and proximal den­dritcs after 2 days (arrows in Fig. 4c), but after 3 and 4 days immunoreactivity was stronger and already asso­ciated with labellcd punctatc profiles oUlliniog the eell bodies (Fig. 41). One day later (5 days) the labcllcd riMLF neurons could ooly be identified by the immuno­reactivc puneta ouUining thc somata ofthe "empty" cclls (Fig. 4i). This was dearly seen by foeussing through the depth of the section, but is nOl obvious from the photo­graph (compare to motooeuronal labeJling in 2c, d). After 12 days very few labelied puneta cou ld be detected.

We ncvcr found any cvidencc for double transsynap­lic tran sport ; that is 00 labclled cells wcre found in those regions that are known to send monosynaptie projections onto premotor neurons, e.g. OInnipause cells in the nu­deus raphe interposilus in [he para median pontine reti· cula r formation , or cells in the deep layers of the su perior colJiculus, bOlh of which are known to project to the riMLF (for review: Büttner·Enncver and Büttner 1988).

Only after 3-5 days survival time were all prcmotor populations labelled following a n eye muscle injection with fragmen t C. We found the strongest and most eon· sisten t labelling of cell bocHes in all premotor areas with 4 days survival time. These findi ngs were eritically depen· dent upon the injection of highly eoncentrated fragment C, i.e. 15%. Lower conccnlralions (approx. 10%) were used in scveral cases and werc found to be less suitable for this study. These eases required longer survival times (5-6 days) to see any transsynaptie label at an. There was also no survival time at which al1 premotor neuronal ccll bodies were labelIed at the same timc: while some regions eontained dear cell body labelling, only punetatc label­ling was seen in olbers. lt is im portant to emphasize that for this reason only those eases that receive an eye muscle injectio n of the same bateh of fragment C concentrate ean be qualitatively eompared to eaeh other as in Fig. 4.

Discussion

The precautions that were usually adequate to prevenl leakage into the orbit after HRP or wbeatgenn aggluti· nin (WGA) injections into an eye muscle (Büttner­Ennever and Akert, 1981), were oot effective in tbese experiments v.'itb fragment C, Uptake by neigbbouring eye musc\es always occurred, aod resulted in additional labeJling of eorresponding motoneurons in IlI , VI and VII (Courville 1966; Gray et al. 198 1; Satoda cl al. 1987; Evinger 1988). This probably reftects a more efficient uptake system for the tetanus toxin fragment eompared

to other macromolecules used as neuroaoatomical tracers (Wan el al. 1982 ; Trojanowski el al. 1982). 11 did not interfere with the results ofthis slUdy because: I) thc premotor connections of VII (Takada et al. 1984) and Vlacc (for review: Evinger 1988) were not labclled, 2) the premotor neurons of the medial reetus motoneurons were labelIed (i.e. internuc\ear neurons in thc contra lat­eral VI, Highstein and Baker 1978; Labandeira·Garicia et al. 1989b), but did not involve thc 3 premotor regions (sv ipsilateral, sv contralateral, riM LF) tha i we chose to study in detail (Evinger 1988).

The present study eonfirms the observation of other iovestigators (Stöckel et al. 1975; Fishman and Carrigan 1987) that tetanus toxin fragments undergo very fas t intraaxona l retrograde transport, si nce ocular motoneu­rons were directly labelIed within 8 h: demonstrating a minimum transport rateofS mm/h fo r fragment C. This is in agreement with thc rate of transport of tetanus toxin ca1culated at 5-10 mm/h (Habcrmann 1978; Carrol cl al. 1978) and 7-7.5 mm/h (Stöekel el al. 1975) in the moto­neurons of rodents.

Tbe immunoreaetivc granules we observed in the light microscope inside tbe cytoplasm of tbe motoncu­rons at 8 h to about 4 days and of premotor neurons probably represent the membrane eompartments con­laining tbe tracer that wcre described in cJectron miero· scopie studies (Sehwab and Thoenen 1978; Schwab et al. 1979; Cabot et al. 1987). Tbe rapid change of the label­ling pattern over the first 2 days indieates that aflcr reaebing thc motoneuronal somata there is only a short time period of tracer accumulation within Ihe cell bodies and dendri les, before additional transsynaptie transfer into presynaptie terminals starts. This transfer is in· dicated by tbe appearancc of punetate profiles, and is alrcady seen 2 days a fter thc cye muscle injeclion . Previous studies have shown that tetanus toxin injcctcd into the spinal cord aceumulated within synaplie termi­nals (Price el al. 1977 ; Price and G riffin 1981), where thc spccific receptor is loeated (van Heyningen 1974), rather than within eell bodies. Schwab and Thoenen (1976) showed for the first time that after a muscle injection tetanus toxin accurnulates not only within corresponding cell bodies, but also in presynaptic tenninals. In thcir elcctron mieroscopie studies the aUlbors did not observe any sign ifieant labelover the ccll membrane or in tbe cxtraceIJular space (Schwab and Thoenen 1977 ; 1978; Sehwab et al. 1979).

Thc olivary pretecta l nucleus sends monosynaptic projections onto the Ed illger-Westphal nueleus in the oculomotor eomplex but not onto motoneurons, and it remained unlabclIed in an our experiments, although il is labelIed in HRP-studies of the oculomolor nueleus (Graybiel and Hartwieg 1974; Steiger and Büttner­Enoevet 1979). Such results, as weH as previous studies (Büttner-Ennever et a l. 1981; Evinger and Eriehsen 1986), provide furtber cvidcncc for areal transsynaptic transfer o f tetanus toxin fragments into presynaptie ter­minals ratber than a non-specifie leakage out ofthe cells by the fact that transsynaptic labelIed cell bodies were exelusively found in those areas that are known to specif­icly target onto motoncurons.

Tbe patterns of lranssynaptie labelling were de~ scribed o n the basis of three pathways with weil known properties: the sv contains two sets of premotor neurons that synapse on oculomotor neurons as part of the ves­tibulo-ocular thrcc neuronal arc. The central part of the sv gives risc to an inhibitory projection that ascends via the ipsilateral MLF and terminates ipsilateral in the oculomotor and trochlear nucleus on vertical motoneu­rons. Whcrcas dorsally located sv neurons send an excit­atory projection via thc brachium eonjunetivum (BC) onto vertieal motoneurons of the conlralateraJ side (Va­mamoto el al. 1978; for review: Büuner-Ennever 198 1; Highstcin and McCrea 1988). Tbe positive labelling of both sv neuronal populations in our experiments em­phasizes for the first time the capability of tetanus toxin fragments to travel across inhibitory as weil as excitat­ory synapses (see also Büttner-Ennever et al. 1981; Evin­ger and Eriehsen 1986).

The riMLF is involved in the generation of vertical saccadie eye movemcnts (Büttner et al. 1977) and has bcen shown to send direct projections to vertical moto­neuron subgroups (Büttncr-Ennever and Büttner 1978; Graybiel 1977; Nakao and Shiraishi 1983, 1985) in cat and monkey. The present study confirms the rcccnt de­scription of the homologue in the rabbit (Labandeira­Garcia et al. 1989a), and it implies a purely ipsi lateral connection to the oculomotor nucleus in tbis species.

The premotor neuronal label was weaker and occur­red later than motoneuronallabelling. Tbecbanges in the pattern ofthe transsynaptic stain ing in premotor neurons resembled the changes in mOloneuronal labclling de­scribed above: a short period of label accumulation with­in the soma is followed bya longer lasting punctate label­ling around tbe premotor neurons. It is therefore difficult to distinguish labelIed motoneurons from premotor neurons from the pallcrn of labclli ng alone. Other critcria such as thc labelling intensity, the survival time, and the concentration of thc injectcd fragment C (see below) must be taken into account as weiL

Four days were chosen as the optimal survival time for labelling prcmotor neuro ns in the oculomotor system with fragment C, since this was the peak of tracer accu­mu1ation within ccll bodies (Fig. 4). However th.is opti­mal time for prcmotor labclling was found to depend on the high concentration of thc fragment C injecled. We had the impression that the usage of lower conccntra­lions could to some extent be compensated for by longer survival times, enabling sufficicnt tracer accumulation in prcmotor cells, or nlther in their afferent terminals, for visualization (see Evinger and Erichsen 1986). A quan­titative comparison of the binding and neuronal [rans~ port capabilities of tetan us toxin and its fragments showed that under physiological pH-conditions SO- 100 times more of tbe fragments C or BII " must bc injcctcd into a museIe in order 10 obtain neuronaJ transport simi­lar to that seen with the whole tetanus toxin (Weller cl • 1. 1986).

The punctate labelling in thc vicini ty ofimmunoreac-­tive premotor neurons seen in the light microscope (see Fig. 4g and h), laken together with the knowlcdge from elcctron microscopic stud ies of the motoneuronal label-

ling, irnplies that fragment C has passcd a third synapse, although confirmation by ultrastructal analysis remains to be done. The failure to fi nd any labelling in regions that are known to send afferents to premotor neurons ean be attributed to the dilution effect causcd by thc increasing divergence of afferent systems.

A comparison of the transsynaptic stai ning in 3 se­lected premotor regions describcd in detail here, shows that the time course of the changes in the pattern of labelling is different for individual pathways. For exam­pie we found that in thc ipsilateral sv the pauern of transsynaptic label changed quiekly from somatic to punctate, whereas in the contralateral sv the pattern remaincd almost exelusively confined to the cell bodies (compare Fig. 4g and h). Such variations must bc related to ccrtain individual propertics of the prcmotor path­ways: Olle possibility is the strength oftbe synaptic inpul from a given premotor sourcc onto the motoneurons in the oculomotor nuelei that is reflectcd by the axonal termination pattern. For the horizontal oculomot or sys­tem of cats it has been shown thaI the excitatory con· traJateral projecting vestibulo-ocular tibers bad a more widespread axonal arborization within the motonueleus with a large number of synaptic boutons, whereas the inhibitory ipsilateral projecting vestibulo-ocular fibers lend to tenninate within restricted areas with less bout­ons (lshizuka el al. 1980; Ohgalci et al. 1988). On the other hand higher numbers of inhibitory tenninals were found to contact lower numbcrs ofmotoneurons indicat· ing a stronger convergence for the inhibitory vestibular input onto motoneurons than for the excitatory input (Ohgalci et al. 1988). The distance from prcmotor neu­rons 10 motoneurons appeared not to be an importanl parameter, since both vestibular projections showi ng a different staining pattern are about tbe same Icngth, but they do d Hfcr in the location oftheir synaptic inputs o nto the oculomotor neurons. Electron microscopic degenera­tion studies dcmonstrate that tbe excitatory contralateral vestibular input terminates on the distal dendritcs of exlraocular motoneurons, whereas thc inhibitio n of the ipsilateral vestibular pathway is mainly mediated by axo­somatic synapses (Bak el al. 1976; Destombcs and Rou· viere 1981). Anotber possibility for variations in the transsynaptie staining patterns is thc firing rate ofcertain premotor neurons. The rate of retrograde transsynaptic transport of WGA has been shown to be enhanccd by electrical stimulation of afferent axons (Jankowska 1985). A simi lar correlation between neuronal activity and the rate of transport has been described for tetanus toxin: the "muscle pump" hypothesis (ponomarev 1928) was tested by Wellhörner et al. (1973), wbo showed that the rate of ascent of 1251_tctanus toxin ioto the spinal cord was incrcascd by electrical stimulation of the nerve of the curarizcd musele.

In conclusion we fou nd that the optimal transsynap­tie labelling of premotor neurons in the ocuJomOlor sys­tem with fragment C was seen after 4 days survival time . The faetors thaI were found to most influence the time course and pattern of labclling were: tbe concentration of the injected fragment C and the type ofpathway 10 bc studicd. The parameters that control its transport are not

yct completely clcar and this makcs fragment C a Icss atlractivc tracer fo r thc strict definition of neural path­ways. However its propcrties da make il a highly useful neuroana tomical tool for locating synaptically related slructures, whereby the nature of the connection must be investigated by atber methods.

Acknowledgemenrs. The aulhors are gratefnl to Profcssor Mehraein for his continuous support and encouragcmcnt, and to Lydia Sach­er for her e)[ccllent technical assistancc. The monoc!onal antibody to fragrnclll C was generously supplied by Ors.1.T. Eriehsen and C. Evinger. Tbis work was supportcd by the German Research Council SFB 220/08.

References

Bak IJ , Baker R, Choi WB, Precht W (1976) Elcctron microscopie investigation of the vestibular projections 10 the eat trochlear uuclei. Neurosei I: 477-4&2

Bizzini B, Stocckcl K, Sehwab M (1977) An antigenic polypeptide fragment isolated from tetanus toxin: ehcmical charaeteriza­tion, binding to gangliosides and retrograde axonal transport in various neuron systems. J Neurochem 28: 529--542

Büllner U, Büttner-Ennever 1A, Henn V (1977) Vertieal eye move­meßt related activity in the rostral mcscllccphalie rclicular formation ofthe alert monkey. Brain Res 130:239--252

Bültner-Enncvcr JA (1981) Vestibular-oculomotor organization. In : Fu(;hs AF, Becker W (eds) Progress in oculomotor research. Elsevier, North Hol1and, pp 36 1- 370

BÜllner-Enncvcr JA, Akcrt K ( 1981) Medial rectus subgroups of the oculomotor Ducleus and thcir abduccos internuclcar input in the monkey. J Comp Neurol 197: 17- 27

Büttner-Ennever JA, Bullner U ( 1978) A eell group associated with \'erticaJ eyc mo\'emcnts in thc rostra! rnesenccphalic fetien!ar formation of the monkey. Brain Res 151: 31--47

BÜllncr-Enncver JA, Büllner U (1988) Thc reticlilar fonnation. In: Büttner-Ennever JA (ed) Neuroanatomy of the oculomotor system. Elsevier, Amsterdam, pp 119-176

Bültner·Ennever 1A, Grob P, Akert K, Bizzini B (198 1) Trans­synaptic retrograde labc1ing in the oculomotor system of the monkcy with [1251] tetanus toxin BUb fragment. Neurosei Lell 26:233-238

CabotJ B, Mermone A, Bogan N (1987) Transneuronal and afferent labcliug of sympathetie preganglionie neurons with thc C-frag­ment of tetanus toxin. Soc Neurosci Abstr 13: 80.20

Carroll PT, Price DL, Griffin JW, Morris J (1978) Tetanus toxin: immunocytochcmical evidence for retrograde transport. Neuro­sei Leu 8:335--339

Courville J (1966) The nucleus of the faeial nerve : the relation bctween eellular groups and periphcral branches of the nerve. Rrain Res 1: 338- 354

Dcstombes J, Rouviere A (1981) Ultrastruetural study of vestibular and reucular projections to the abduccns nuelcus. Exp Brain Res 43: 253- 260

Epema AH, Gerrits NM, Voogd 1 (1988)Commissura! and intrinsie connections of the vestibular nuc1ei in thc rabbit: arctrograde labeling study. Exp Rrain Res 71: 129--146

Evinger C (1988) Extraocular motor nudei: location, morphology, and afferents. In: Büttner-Ennever JA (cd) Ncuroanatomy of the oculomotor system. Elsevier, Amsterdam, pp 81- 117

Evinger C, Eriehsen JT (1986) Transsynaplic retrograde transport of fragment C of tetanus toxin demonstrated by immunohis­tochenueal localiz.alion. Bmin Res 380: 383-388

Fishman PS, Careigan DR (1987) Retrograde transneuronalt rans­fer of the C-fragmenl of tetanus toxin. Brain Res 406: 275- 279

Gray TS, McMaster SE, Harvey JA, Gorme7.8.no T (1981) Localiza· tion of retraetor bulbi motoncurons in the rabbi!. Brain Res 226:93- 106

Graybiel AM (1977) Organization of oculomoto~ pathways in the

eat and rhesus monkey. In: Baker R, Berthoz A (eds) Contro! of ga7.e by brain stem neurons. Elsevier, Amsterdam, pp 79-88

Graybiel AM , Hartwieg EA (1974) Some afferent connections of the oculomotor complcx in the cat: an experimental study Wilh tracer tcchniques. Hrain Res 81: 543- 551

Habermann E (1978) Tetanus. [n: Vinker PJ. Bruyn GW (eds): Handbook or clinical ncurology, Vol I. North Holland 33 :491- 547

Hancoek MB (1984) Visualization of pepride-immunorcaetive processes of serotonin·immunoreaelive cclls using two color immunoperoxidase staining. 1 Histochem Cytochem 32:311 - 314

Helting TB, Zwisler 0 (1977) Strueture of tetanus toxin. 1. Break­down of the toxin molecule and diserimination bctween poly­peptide fragments. J Biol ehern 252: 187- 193

Heyningen WE van (1974) Gangliosides as membrane r<."CCptors for tetanus toxin, cholera toxin and serotonin. Nature (Lond) 249:415-417

Highstein SM, Haker R (1978) Exeitatory termination of abdueens internuclear neurons on medial rectus motoneurons: relation­ship to syndrome of internuc1car ophtalmoplegia_ J Neuro­physiol41: 1647- 1661

Highstein SM, Meerea RA (1988) The anatomy ofthe vestibular nuclei. In: Büttner- Ennever JA (cd) Neuroanatomy of the oculomolor system. Elsevier _ Arnsterdam, pp 177 202

Hsu SM , Raine L, Fanger H (1981) Use ofavidin-biotin-pcroxidase complcx (ABC) in immunopcro)[idase techniques: a comparison between ABC and unlabclcd antibody (PAP) procedurcs. J His­tochern Cytochern 29: 577- 580

Ishizuka N, Mannen J, Sasaki S, Shimazu I-I (1980) Axonal branehes and terminations in the cat abduccns nuc\eus of secon­dary vestibular neurons in the horizontal canal system. Neurosei Lelt 16: 143~ !48

Jankowska E (1985) Further indications for cnhancement ofrelro­grade transneuronal transporl ofWGA- I-JRP by synaptic aeliv­ity. Rrain Res 341 : 403---4{)8

Labandeira-Garcia JL, Guerra-Seijas M1, Labandeira-Gareia JA (1989a) Oeulomotor nuc\eus alTerents from the interstilial nu· cleus of Cajal and the region surrounding the fasciculus relro· flcltus in the rabbit. Neurosei lell 101: 11 - 16

Labaudeira-Gareia JL, Guerra-Seijas M J, Labandeira-Gareia JA (1989b) The abducens motor and internuc\car neurons in the rabbit: retrograde horseradish peroxidase and double fluores­ccnllabeling. Brain Res 497:305-314

Manning RA, Evinger C, Eriehsen JT (1987) Retrograde trans­neuronallabeling of sympathelic preganglionic neurons follow­ing injection of tetanus to)[in fragment C into Mueller's eye lid mu;cle. Soc Neurosei Abstr 13:49.18

Mcllanby J, Green 1 (1981) How does tetanus toxin ael? Neuro­sciencc 6:281- 300

Nakao S, Shiraishi Y (1983) Oircct projection of cat rnesodience­phalie neurons to the inferior rectus subdivision or the oculo­motor nuc\eus. Neurosei Letl 39: 243- 248

Nakao S, Shiraishi Y (1985) Oirect excitatory and inhibitory synap­tie inputs rrom the medial mesodieneephalic jUlletion to moto· neurons innervaling extraocular oblique rnuseles in t.he ca!. Exp Brain Res 61: 62- 72

Ohgaki T, CurthoysIS, Markham CH (1988) Morphology ofphysi· ologically identified sccond·order vestibular neurons in cal, with intraeellularly injected HRI'. 1 Comp Neurol 276:387-4 11

I'onomarcw AW (1928) Zur Frage der Pathogenese des Tetanus und des Fortbewegungsmcchanismus des Tetanustoxins längs den Nerven. Z Ges Exp Med 61 :93-106

Pricc Ol, Griffin JW, Peck K (1977) Tetanus Toxin: evidellce for binding at presynaptic endings. Beain Res 121 : 379- 384

Price OL, Griffin JW (198 1) Immunocytochernicallocalil'.8.tion of tetanus toxin to synapses of spinal cord. Neurosei Leu 23: 149- 155

Satoda T, Takahashi 0, Tashiro T, Matsushima R, Uemllra-Sumi M, Mizuno N (1981) Representation ofthe main branehes or the faeial nerve within the raeial nue1eus of the Japanese rnonkey (Maeaca ruseata). NellIosci Let! 78:283-287

Schwab ME, Thocncll H (1916) Electron micro!;Copie evidencc for a transsynaptic migration of tetanus toxin in spinal cord rnoto­neurons: an autoradiographie and morphometric sludy. Brain Res 105:213-227

Sehwab ME, Thoenen H (1977) Se\ective trans-synaptic migration of tetanus toxin after retfograde axonallransporl in peripberal sympathetic nerves: a comparison wilh nerve growth factor. Brain Res 122:45'J-474

Schwab M E, Thocnen H (1978) Sclective binding, uptakc, and retrograde transport of tetanus toxin by nerve terminals in the rat iris. 1 Cell ßiol 77: 1- 13

Schwab ME, Suda K, Tbocnen H (1979) Sclective retrograde tr<lns­synaptic transfer or a protein, tetanus toxin, subsequent to il.'l retrograde axonal transport. J Cel1llioI82:798-810

Steiger IU, nültncr-Ennever JA (1979) Oculomotor nucleus afTer­ents in the monkey demonstrated with horseradish peroxidase. Brain Res 160: 1- 15

Stöekel K, Schwab M, Thocnen H (1975) Cornparison Ix:tween tbe retrograde axonaltIansport of nerve growlh faetor and tetanus toxio in motor, sensory and adrenergic neurons. Brain Res 99: 1- 16

Takada M, h oh K, Yasui Y, Mitani A, Nomufa S, Mizuno N (1984) Distribution of premotor neurons for orbicularis oculi

motoneurons in the cat, with particular rcfercnce to possible pathways for blink reflcx. Neurosei Leu 50:251- 255

Trojanowski 1Q, Gonatas JO, Gonatas NK (1982) Horseradish peroxidase (HR]» conjugatcs of cholera toxin and lectins afe more sensitive retrogradely transported markers than frte HRP. Brain Res 231 ; 33-50

Wan XS, Trojanowski JQ, Gonatas JO (1982) Cholera toxin and wbeatgerm aggluti nin eonjugatcs as neuroanatomieal prohes: their uptake and elcarancc, transganglionic and retrograde transport and sensitivity. Ihain Res 243:215-224

Weller U, Taylor CF, Hahcrmann E (1986) Quantitative corn­parison bctween tetanus toxin, sorne fragments and toxoid for binding and axonal transport in the rat. Toxicon 24: 1055-1063

Wellhörner Hf-( (1982) Tetanus neurotoxin. Rev Physiol Biochern Pharmac 93: 1-68

Wellhörner HH, Seib UC, Hensel ß ( 1973) Local tetanus in tats: the influence of ncuromuscular activity on spinal distribution of I B] lalx:lled tetanus toxin. Naunyn-Schmiedebcrg's Areh PharmacoI216:387- 394

Yamamoto M, Shirnoyama I, Highstein (1978) Vestibular nucleus neurons relaying excitation from the anterior canal to the oculo­motor nucleus. Brain Res 148:31-42