The effect of intraocular injection of tetrodotoxin on fast axonal transport of [3H]proline- and [3H]fucose-labeled materials in the developing rat optic nerve

0306-4522/U

Neuro.~ien~ Vol. 16, No. 4, pp. 1027-1039,1985 Printed

in Great

$3.00 + 0.00

Pergamon Press Ltd 1’ 1985IBRO

Britain

THE EFFECT OF INTRAOCULAR INJECTION OF TETRODOTOXIN ON FAST AXONAL TRANSPORT OF [3H]PROLINE- AND [3H]FUCOSE-LABELED MATERIALS IN THE DEVELOPING RAT OPTIC NERVE R. V. RICCIO* and M. A. MATTHEWS? Department of Anatomy, Louisiana State University Medical Center, 1100 Florida Avenue, New Orleans, LA 70119, U.S.A. Abstract-The fast axonal transport of [SH]proline-labeled proteins and [3H]fucose-labeled glycoproteins delivered to the dorsal lateral geniculate nucleus in the developing rat optic nerve was investigated during tetrodotoxin-induced monocular impulse blockade. Repeated intraocular injections of various dosages of tetrodotoxin or citrate buffer vehicle were made every two days in rats aged 5-21 days postnatal, and the accumulation of rapidly transported radioactivity in the lateral geniculate nucleus measured between three and twelve hours post-injection at each age. The effectiveness of prolonged tetrodotoxin treatment was monitored by loss of the pupillary light reflex and the level of cytochrome oxidase activity in the contralateral superior colliculus and dorsal lateral geniculate nucleus. Numbers of optic axons proximal to the chiasm and the frequency of retinal ganglion cells per unit distance from the optic disc were examined for signs of tetrodotoxin-induced degeneration of the retinofugal pathway. Tetrodotoxintreatment reduced the amount of fucosyl glycoproteins, but not proline-labeled proteins, axonally transported to the lateral geniculate nucleus during the first three weeks of postnatal development. Other studies indicated that tetrodotoxin significantly reduced the incorporation of [3H]fucose into retinal proteins indicating that the reduction in transport was probably due to a decrease in precursor incorporation into retinal ganglion cells. Electron microscopy of ganglion cells at 21 days revealed dilated and vacuolated Golgi cisternae associated with tetrodotoxin treatment, suggesting that tetrodotoxin may alter fucose metabolism by secondarily disrupting Golgi organization. Other protein synthetic machinery in these cells, including ribosomes and rough endoplasmic reticulum, appeared normal throughout tetrodotoxin treatment. These data indicate that Na+-dependent optic impulse activity may be indirectly related to the axonal transport of glycoproteins during early postnatal development by mediating the incorporation of precursor into glycoproteins at the Golgi apparatus and their subsequent entrance into the fast transport system.

The axonal transport of cellular components from the soma to the axon terminals has been extensively

(TTX),

studied in the developing CNS.26 These studies indicated that much of this material consists of proteins and glycoproteins which are important constituents membrane of the presynaptic and synaptic vesicles’2,22.60,62 and that their rate of transport increased up to three-fold during certain stages of postnatal development. U,65.70In the visual system, much of this increase coincided with the onset of functional activity,42.62.67.70 suggesting that changes in impulse activity may be associated with the maturation of the fast transoort svstem. In order to explore this concept, we investigated the effect of intraocularly injected tetrodotoxin

*Present address: Department of Physiology, University of Medicine and Dentistry of New Jersey, 100 Bergen Street, Newark, NJ 07103, U.S.A. tAddress correspondence to: Dr M. A. Matthews, Department of Anatomy, Louisiana State University Medical Center, 1100 Florida Avenue, New Orleans, LA 70119, U.S.A. Abbreviations: dLGn, dorsal lateral geniculate nucleus; dpn, days postnatal; SC, superior colliculus; TTX, tetrodotoxin.

a specific

blocker

of the voltage-dependent

sodium channels responsible for generating action potentials49,50 on the incorporation of labeled precursor into proteins and glycoproteins in the developing rat retina, and the fast axonal transport of these substances in the optic nerve. An ultrastructural analysis of the retina and optic nerve following prolonged TTX treatment was then conducted in order to examine the effects of impulse blockade on ganglion cell and optic nerve morphology. We used both [‘Hlproline and [3H]fucose as labeled precursors in these experiments. Proline, an imino acid. is delivered to the nerve terminals bv fast and slow axonal transport8.44.6’.65and is considered to be an excellent precursor for studies of fast axonal transport.‘9.32 It is thought to be a constituent of several axonal components, including neurotransmitterqb3 the preterminal membrane,” synaptic vesicles@ and actin’ among others. Fucose, a hexose sugar, is incorporated almost exclusively into glycoproteins as fucose24,37.40.73 and is not significantly incorporated into fucolipids.40.SE,73Fucosylation octhe curs in the Golgi apparatus, 7.16.17.23.28.36 from which newly formed glycoproteins exclusively enter the fast to become associated with transport system 2’.23.3’.5’.69

1027

1028

R. V. Rrccro and M. A.

the axolemma, preterminal

membrane

and synaptic

vesicfesh EXPERIMENTAL

PROCEDURES

Animak Seventy fitters of albino rats (Holtzman) were used. All animals were housed in standard laboratory cages (20°C) and maintained at 12 h light-dark cycles. The first apperarante of rat pups was considered as 0 days postnatal (0 dpn). Injection procedure Etherized rats, aged 5-21 dpn, were intraocularly injected into the right eye with 0.5 ~1 TTX every 2 days at concentrations of 1 x 10e4 M (N = 121). 1 x 10d5 M (N = 68) and I x 10b6M (N = 135) beginning at 5 dpn. All injections were given at a rate of 0.25~1/min with 10~1 Hamilton syringe and a 31 gauge needle. Control animals (N = 151) received an identical volume of the corresponding citrate buffer vehicle. For injections given to animals prior to eye opening (5-14 dpn), the needle was passed through a 2 mm horizontal incision made in the skin above the palpebraf fissure. After eye opening, injections were made at the corneoscferal junction. Following complete recovery from the anesthetic (S-IO min), the animals were returned to the mother. Efecr of tetrodota,~i~lon fast axanal iranspari The effect of TTX on fast axonal transport in the optic nerve was determined by analyzing the amount of accumulated labeled proteins in the contrafateral dorsal lateral geniculate nucleus (dLGn) after precursor injection. The right eye of etherized rats aged 5, 9, f3, 18 and 21 dpn was injected with 0.5 ~1 of L-~2,3,4,5-3H]profine (specific activity, 73 Ci/mmof, Amersham) or L-[3-‘Hlfucose (specific activity, 26Ci/mmof, Amersham) at a concentration of fOpCi/rl. The animals were reanesthetized and sacrificed by transcardiac perfusion at 3, 6, 9 and 12 h post-injection. In some experiments the optic nerve and tract was excised from the sclera to the surface of the dLGn and its length measured to the nearest 0.5 mm. Fresh samples of the contrafateraf and ipsilateral dLGn were dissected free under a stereomicroscone. sofubilized overnight in 300~1 NCS Tissue Solubifizer ‘(Amersham), and c&nted in a toluene-based scintillation cocktail. Counting efficiency (42.5%) was determined by automatic external standard channel ratios using sealed quenched standards (Beckman). Since the retinofugaf pathway is 9&95% crossed at the chiasrn5’ values for each time point were corrected for systemic background labeling by subtracting the radioactivity of the ipsilateral nucleus. Plotted values were expressed as dpm/pCi injected x IO’ and each time point was the mean of 3-6 animals. Percent differences in accumulated radioactivity were calculated between control and TTX-injected animals, and statistically compared using the Student’st test (P < 0.05 and P < 0.01). Precursor incorporation The effect of long-term TTX treatment on precursor incorporation into retinal proteins was examined in rdts aged 9, 13 and 21 dpn. At each age, the right eye of 12 rats treated with I x 10-j M TTX or citrate buffer was injected with 0.5~1 [‘Hlproline or [‘Hlfucose. The animals were sacrificed at I, 2, 4 and 8 h post-injection with an overdose of pentobarbital and transcardially perfused with saline to remove any blood-borne label, The retinae were excised, dissected free from the choroid, homogenized in 1.0 ml 10% trichloracetic acid at 4°C and kept on;ce for one hour. The homogenates were centrifuged at 3000g for IOmin, the supematant decanted, and the centrifugation-washing procedure repeated three times. In the [‘Hlfucose injected animals, lipid extraction procedures were not performed since little or no fucose incorporates into fucofipid.40.” The precipitates were then dissolved in Protosof (New England

MATTWEWS

Nuclear) and the acid-insoluble radioactivity countetl in a toluene-based scintillation cocktail. Counting efficiency was determined as described above. Values were expressed as dpm/pCi injected x 10’ and each time point represented the mean of at least 3-5 samples. Statistical comparisons were made between the TTX and vehicle-injected retinae using the Student’s f-test (P < 0.05). .Effect of‘ tetrodotoxin an retinal ganglion celis and t/w opric, ntwe Light microscopy. The integrity of the ganglion cell popuiation after Iong-term TTX treatment was monitored at 21 dpn by counting the number of retinal ganglion ceils along a fixed distance from the optic disc (500 ym) from 2 /.m sections stained with toluidine blue. The animals were sacrificed with an overdose of pentobarbitaf and transcardiaffy perfused for electron microscopy with 1.0% paraformaldehyde in 0. I M cacodyiate buffer @H 7.35) containing 0.01% CaCI,, followed by 3% paraformaldehyde in the same buffer.s5 The retinae were excised, stored in fixative overnight, rinsed in buffer and postfixed in I % 0~0, for one hour. The tissues were dehydrated in a series of graded alcohols, followed by propylene oxide and embedded in Epon. Comparisons were made between 8 retinae, each of which received 1 x fOm4M TTX or the corresponding vehicle, and further compared with counts obtained from 4 non-injected retinae. Retinal ganglion cells were identified by somaf size (B--2Opm diameter), euchromatic nuclei. prominent nucleoli and little cytoplasm. They were distinguished from neuroglia or displaced amacrine cells. identified by a smaller soma (337 ,~m diameter), homogeneously stained nuclei, and comparativefy fess cytoplasm.” Mean values from each group were compared and analysed by single factor analysis of variance (ANOVA) at P < 0.05. Electron microscopy. At 21 dpn, the integrity of the retinal ganglion cell layer was qualitatively analyzed after longterm TTX treatment. Thin sections were cut from retinae processed above and stained with 5.0% uranyf acetate and 0.4% lead citrate. Vehicle-injected retinae were similarly processed and examined. To examine for TTX-induced optic tiber degeneration. quantitative estimates of the axon population at 21 dpn were determined from 4 TTX-iniected (I x lO-‘M) vehicfeinjected, or normal non-injecied animals. After processing for electron microscopy, the nerves were transversely sectioned 1mm rostra1 to the optic chiasm, placed on formvarcoated (0.5%) slotted grids, and stained with 5% aqueous uranyl acetate and 0.4% lead citrate. Thirty overlapping micrographs were taken from the central and peripherat portions of each section. ‘7 Final prints were made at a magnification of 5198 x and coded to prevent sampling bias. The number of axons were counted in each micrograph, and the area of the print was determined by planimetry using a MOP-3 digitized tablet. No less than 10,000 axons were counted from each nerve.59 The total crosssectional area of the nerve was determined from tow magnifi~tion micrographs of the same sections.” The axon population of each nerve was then determined by cafculating the number of axons within the total area of the sample micrographs and extrapolating this value to the total cross-sectional area of the nerve. Statistical comparisons employed single factor ANOVA at P < 0.05. Efsect qf tetradotoxin an aptic newe impulse actitiity The efficacy of TTX in eliminating retinal impulses was indirectly determined by 2 methods. In the first, the loss of the pupillary fight reflex in the TTX-injected eyes was compared with those receiving vehicle injection or no treatment. Monitoring of the reflex could only be done with animals beyond the age of eye opening (1421 dpn). For each dosaze of TTX and the corresponding vehicle, the time elapsed between reflex termination and- restoration was

TTX effects on axonal trans~rt recorded, and used as a guide for scheduling repetitive TTX injections.” The second procedure examined for the activity of cytochrome oxidase in the superior colliculus (SC) and dLGn contralateral to the Tl’X-injected eye.” The staining pattern was compared with the ipsilateral nucleus which served as an internal control. At 21 dpn, 5 rats treated with I x 10-‘M TTX were sacrificed with pentobarbitaf, perfused with 0.1 M paraformaldehyde in 0.1 M phosphate buffer containing 4.0% sucrose (pH 7.40), and decapitated. Whole brains were stored in fixative overnight (4”C), and transferred into fresh buffer, the sucrose concentration of which was gradually increased to 30% with repeated washing. The next day, the brains were frozen by immersion for I min in isopentane (-70°C) cooled by liquid nitrogen. Sections (IOOpm) were cut on an IEC-CTD cryostat at - 15°C and rinsed in 0.1 M phosphate buffer containing 4.0% sucrose prior to incubation. The sections were incubated in the dark at 37°C for I .5 h in 0. I M phosphate buffer (pH 7.40) containing 4.0% sucrose, 0.26 m&ml cytochrom C, and 0.5Smg~ml diamino~nzidine as a chromagen. Autoxidation of diam~no~n~dine by endogenous peroxidases was controlled by adding 200 rg]ml catalase to the incubation medium. Control sections from 2 TTX-treated rats were incubated in medium minus the cytochrome c substrate. The sections were mounted serially on glass slides, dehydrated irtg raded ethanols, equilibratkd with xylene, and coverslipl 3ei I. The intensity of staining between contra-

in developing optic nerve

lateral and ipsilateral nuclei was then qualitatively assessed and photographed. RESULTS

The incorporation of t3Hjproline into retinal proteins was examined at 9, 13 and 21 dpn following TTX or vehicle treatment (Fig. 1). Control retinae showed the maximum incorporation of precursor by 2 h post-injection at 9 and 13 dpn, and from Z-4 h post-injection at 21 dpn. Intraocular TTX had no significant (P < 0.05) effect on 13H]proline incorporation at each of the ages examined. The incorporation of [‘Hlfucose into retinal glycoproteins under control and TTX-injected conditions was also determined at 9, 13 and 21 dpn (Fig. I}. Tetrodotoxin trea~ent si~i~cantly reduced the amount of [3HJfucose incorporation at each age, and this effect was seen as early as 1 h post-injection. The reductions seen at 9 dpn (range: 23-38%) and 13 dpn (3@-37%) were somewhat greater than those at 21 dpn (19-25%). For ail ages, the maximum amount

3H-PAOLINE 9dpn

13dpn

lldpn

3H-FUCOSE 13dpn

Zldpn

I

;

40 30

0

E iii 2 -

1 9dpn

1029

HOURS POST-INJECTION Fig. I. The effect of repeated intraocular injection of tetrodotoxin on the incorporation of [3H)proline and rHJfucose into retina1 proteins or glycoproteins during postnatal development, At 9, I3 and 21 dpn, retinae were examined for the incorporation of a single injection of labeled precursor into a trichioroacetic acid-insoluble precipitate I, 2, 4 and 8 h post-injection. Tetrodotoxin lowered, but did not significantly reduce, the amount of [3H]proline incorporation during this period. The amount of [3H]fucose incorporation was significantly reduced at each age during TTX treatment. Each value represents the mean of 3 experiments. Vertical bars indicate standard error of the mean. (a-----a), I x 10m4M m-treated; ( x-x ), citrate-injected control.

R.

1030

V. Rtccro and M. A.

of [3H]fucose incorporation in control and TTXinjected retinae occurred between 2 and 4 h postinjection. Fast axonal transport of [‘HJproline

Rapidly transported [‘Hlproline was detected in the dLGn approximately 3 h after precursor injection regardless of the animal’s age, length of the optic AGE ATI~CTION

MATTHEWS

pathway, or whether or not the retina received TTX or control injections. Regardless of the dose. the amount of [‘Hlproline labeled proteins transported to the dLGn in the TTX-treated animals did not differ significantly from controls (Fig. 2). No immediate :ffect of TTX on fast transport was observed at 5 dpn after a single intraocular injection. No significant differences were observed one day prior to the onset

-5 DAYS POST~TAL

AGE AT INACTION -9 DAYS POSTNAIAL

20-

a-

15lSIOIOs-

5-

I 3 AGE A’f INJECTION-13 DAYS POSTNATAL

6

I

9

I

12

ACE AT INJECTION-18 DAYS POSTNAlAL

20

i

3

6

9

12

AGE AT INJKTION -21 DAYS POSlNARt

ia

3

HOURS

6

9

12

POST-INJECTION

Fig. 2. The effect of repeated intraocular injection of tetrodotoxin on the time course of appearance of radioactivity in the dorsal lateral ganiculate nucleus following iatraocutar injection of ~rS)prohne during postnatal development. No change in the amount of proline-labeled proteifrs transported to the dL& was found during TfX treatment compared to citrate-injected controls. E&Y value is the mean of 3-6 experiments. Vertical lines indicate standard error of the mean. Controk 5 x iOe4M citrate buffer (e-0); TTX i x 10.-“M T-TX (A---&; 1 x WfM TTX (O...O); 1 x IO-*M T-TX (a----&.

TTX effects on axonal transport in developing optic nerve Table 1. Effect of tetrodotoxin

1031

upon the incidence of retinal ganglion cells

Number of cells per Tissue Normal Control* TTXt

Age (dpn) 500 pm length of retina k SEM 21 21 21

31.8 k 3.3 32.2 f 5.6 30.3 IfI3.9

N

% Difference

4 4 4

-1.03 -4.75

*5 x 10-4M citrate buffer. tl x 10-4M TTX. trefers to no sienificant difference from normal retina (P < 0.05). k-Number ofwretinae examined. of functional activity at eye opening (13 dpn), or during the first week thereafter (18 and 21 dpn). However, the amount of accumulated radiolabel did fluctuate during this period. After 12 h, maximal accumulation was evident at 5, 13 and 21 dpn (Table 1 and Fig. 2). There was a steady decrease with age in the maximum amount of accumulated radioactivity, such that peak levels recorded at 5 dpn were much higher than at 21 dpn. Substantially less label was detected at 9 and 18 dpn, although no temporal pattern could be established. Fast uxonal transport of [3H]fueose

[3H]Fucose labeled glycoproteins were first detected 3 h after precursor injection. Significant differences during development were seen between animals receiving control injections and animals receiving injections of maxima1 (1 x 10e4 M) and minimal (1 x lo-” M) doses of TTX. At 5 dpn, no immediate reduction in accumulated radioactivity was evident 3 and 6 h post-injection, however, a significant decrease was observed in the TTX-treated animals after 9 and 12 h (Fig. 3). This reduction averaged 20% less than controls for animals receiving 1 x lO-$M TTX, and reached 28% for those receiving a 1 x 10e4 M dose. At 9 dpn, the reduction was apparent 6 h post-injection. At 13 dpn, animals which had received maximal doses of TTX showed a significant decrease as early as 3 h post-injection, an effect which was not seen in the group given the lower dose. However, by 9 h both groups accumulated less radioactivity than controls. At later ages, the TTXinjected side showed reductions in glycoprotein transport which averaged 27% for both doses at 18 dpn, and ranged from 1622% at 21 dpn. The difference from controls for both groups was significant (P < 0.01) 12 h post-injection. Two major trends were indicated by the data: (1) as the age of the animal increased, the reduced accumulation of labeled glycoprotein was seen at earlier survival periods, and (2) the percent difference between control and T-TX-treated groups decreased with longer survival times. Greater differences were seen at earlier survival periods, which appeared to recover when the survival time was increased. in the dLGn approximately

E#ect of tetrodotoxin on impulse activity Pupillary light rejex. A single bolus of 1 x 10e4 M

TTX abolished

the reflex, causing

a pronounced

pupillary dilation within 10 min. Lack of an immediate or prolonged systemic circulation of TTX was evident by normal movement of the extraocular muscles in both the TTX-injected and normal eyes. At this dose the reflex recovered after 2 days. Reflex activity terminated approximately 15 min postinjection at doses of 1 x 10msM and I x 10m6M TTX. Recovery times for these doses decreased to l-l.5 days post-injection. Control injections induced a similar but less pronounced pupillary dilation by 2min post-injection, but normal reflex activity quickly returned after g-10 min. ~yto~~rorne oxidase. Cytochrome oxidase activity was sharply reduced in the SC contralateral to the side of TTX injection, as indicated by a reduction in staining intensity (Fig. 4). The most marked change was seen in the stratum griseum superficiale, which extended throughout the entire rostrocaudal and mediolateral extent of the nucleus. No evidence of degeneration was detected. The contralateral dLGn showed a significant decrease in cytochrome oxidase activity as compared to the ipsilateral nucleus (Figs 5 and 6). Although the differences between nuclei varied, each experiment demonstrated a reduced activity in the contralateral nucleus. A moderate reduction in cytochrome oxidase activity was also seen in the contralateral lateral posterior nucleus, although the degree of reduction was not as severe as for the dLGn. The surrounding tissues showed an even distribution of stain between comparable structures (e.g. medial geniculate nuclei), indicating that the reduced cytochrome oxidase activity in the SC, lateral posterior nucleus and dLGn was not due to a difference in section thickness or uneven staining characteristics. Control sections incubated in media minus cytochrome c revealed no activation of diamino~nzidine by endogenous peroxidases or cytochromes. This indicated that the differences in reactivity between the nuclei was not due to a non-specific oxidation of the chromagen. E$ect of tetrodotoxin on retinal ganglion cells and the optic nerve Gang&on cell morphology. Quantitative evaluation of the effect of TTX or vehicle injection on the retinal ganglion cell population revealed no major differences in cell numbers between these and normal {non-injected) retinae. The mean number of cells

1032

MATTHEWS

R. V. Rrccro and M. A.

ACE AT INJECTION-9DAYS POSTNATAL

AGEATINJECTION-5DAYS POSTNATAL

ln

3

9

6

12

1

s

3

6

1

9

I

12

?

AGE

ACEATINJECTION-13DAYSPOSTNATAL x

AT INJECTION -18OAYS POSTNATAL

20-

I

3

6

4

15

AGE AT INJECTION-21DAYSPOSTNATAL 2D-

15-

10-

5-

i HOURS

POST-INJECTION

Fig. 3. The effect of repeated intraocular injection of tetrodotoxin on the time course of appearance of radioactivity in the dorsal lateral geniculate nucleus following intraocular injection of [‘Hlproline during postnatal development. Each value is the mean of 3-6 experiments. Vertical lines indicate standard error of the mean. Control: 5 x 10m4M citrate buffer (a-0); TTX: 1 x IO-” M TTX (O---O); I x 10W6M T’fX (A--.-A,.

calculated at a distance

of 500 pm from the optic disc

was not reduced at 21 dpn after repeated injection of 1 x 10e4 M TTX or the corresponding vehicle (Table 1). This indicated that there was no cytotoxic effect

of TTX on the majority retinofugal pathway.

of ceils responsible

for the

Normal ganglion cell ultrastructure was maintained following repeated control injections (Fig. 7).

Fig. 4. Effect of intraocular tetrodotoxin on cytochrome oxidase activity in the superior colliculus. Age: 21 dpn. TTX dose: 1 x 1O-4 M. The colliculus contralateral to the side of injection (right) shows significant reduction in activity compared to the ipsilateral nucleus (left). Double arrows indicate equivalent thickness of the superficial layers. csc, commissure of the superior colliculus; sgs, stratum griseum superhciale; so, stratum opticum. Section thickness: 100 pm. Mag. 100 x . Fig. 5. A cytochrome oxidase-stained section through the rostrocaudal midpoint of the dorsal lateral geniculate nucleus ipsilateral to side of tetrodotoxin injection (internal control). Age: 21 dpn. TTX dose: 1 x 10m4M. Strong reactivity is seen in the dLGn (arrow) and lateral posterior nucleus (lp) indicating high cytochrome oxidase activity. Mg, medial geniculate nucleus. Section thickness: 100 pm. Mag. 100 x Fig. 6. Section through the rostrocaudal midpoint of the dorsal lateral geniculate nucleus contralateral side of tetrodotoxin injection stained for cytochrome oxidase activity (TTX-deprived nucleus). Age: 21 don. TTX dose: 1 x 1O-4 M. A sinnificant reduction in cvtochrome oxidase activitv is seen in the dLGn (outhned), and a moderate decrease is seen in the lateral posterior nucleus (1~). Gig, medial geniculate nucleus. Section thickness: 100 pm. Mag. 100 x to the

1033

1031

1035

TTX effects on axonal transport in developing optic nerve Table 2. Effect of tetrodotoxin

upon axon population within the optic nerve Total axon number

Age (dpn) 2.1

Animal No.

I 2 3 4

Animal No.

Normal 110,495 116,348 112,671 109.964

5 6 7 8

112,369 + 2510

Mean + SEM F statistic % Difference from normal % Difference from control

Control 114,361 108,494 112,865 109.946 111,416+2316 (1.00)’ NS*

NS

Animal No. 9 IO II 12

TTX 117.539 114,106 111,139 112,758 113,885 + 2356 NS NS

*Not significant at P < 0.05 (single factor ANOVA).

profiles showed no chromatolytic changes such as dissolution of the Nissl substance, cellular hypertrophy, or condensation of nuclear material. The Golgi apparatus remained in a perinuclear position and displayed distinct cis and truns surfaces. In contrast, the maximum dose of TTX caused a pronounced change in ganglion cell morphology (Fig. 8). A consistent swelling of the Golgi cisternae occurred at the cis and [runs surfaces of the complex. The nuclear envelope developed an irregular profile which was marked by occasional indentations in the surface adjacent to the Golgi apparatus. No dissolution of the Nissl substance was apparent. The smallest dose caused similar changes. Swelling of the Golgi cisternae was largely restricted to the cis face (Fig. 9), but was not as severe as that seen with a higher dose. Although some irregularity of the nuclear membrane occurred, marked folding was not readily apparent. As with cells exposed to the higher dose, no change in the rough endoplasmic reticulum was seen. Number ofoptic axons. At 21 dpn, no significant differences were found between the population of optic axons from normal, control-injected, or TTXinjected animals (Table 2). No obvious differences in axonal ultrastructure were observed between the control and TTX-treated nerves and no evidence of significant degeneration was apparent in either tissue.

Nuclear

DISCUSSION

Effectiveness of tetrodotoxin in inhibiting optic impulses

The combination

of behavioral and histochemical

techniques used in this study were sufficient to allow us to conclude that the generation of optic action potentials was severely reduced, if not completely inhibited, by repeated intraocular TTX injections. Pupillary light reflex. For each dose of TTX up to 1 x 10m4M complete loss of the pupillary light reflex occurred within minutes and lasted for up to 2 days post-injection. Since normal extraocular movement was observed in both the TTX-injected and the contralateral (non-injected) eyes, and since contralateral pupillary activity responded normally to light stimuli, it was concluded that systemic circulation of TTX was negligible. Loss of the reflex has previously been used to determine the duration of optic nerve blockade by TTX in kittens.‘.72 Therefore, as in the kitten, the period of reflex inhibition served as a convenient indicator of the duration of optic nerve blockade. Higher doses (i.e. 5 x lo-)7 x lo--’ M), while also inhibiting the reflex, reduced the probability of survival of the youngest animals. Since TTX is eliminated from the eye with a half-life of approximately 7 hi these latter results are what would be expected at such high doses. Cytochrome oxidase histochemistry. It has been repeatedly suggested that changes in cytochrome oxidase activity reflect changes in the functional demand of a neuron. ” Monocular impulse blockade by TTX significantly depresses the cytochrome oxidase activity in the kitten dLGn.72 Our studies represent the first demonstration of an effect of intraocular TTX on cytochrome oxidase activity in the SC and dLGn in the rat. Since there is little evidence for binocular interaction or discrete laminar

Fig. 7. Electron micrograph of a 21 dpn retinal ganglion cell after repeated injection of 5 x lO-4 M citrate buffer vehicle. The nucleus (N) and ribosomal elements (r) appear normal. The Golgi apparatus (arrows) shows distinct flattened parallel cisternae. The nuclear membrane (arrowheads) shows no surface irregularities. Mag. 17,IO0 x . Fig. 8. Electron micrograph of a 21 dpn retinal ganglion cell after repeated intraocular injection of 1 x 10m4M tetrodotoxin. Golgi cisternae are swollen at the cis face (arrows) and some disorganization of the cisternae is seen. The nuclear membrane shows distinct infoldings (arrowheads). N, nucleus. Mag. 17,100x. Fig. 9. High magnification of the Golgi apparatus of a 21 dpn retinal ganglion cell after repeated injection of I x 10moM tetrodotoxin. Note the marked swelling of the Golgi cisternae (arrows) at the surface (C). appearing coated are apparent the trans (T). Mag. x

1036

R. V. Ricrro and M. A. MATTHEWS

patterns in the rat dLGn’-‘.20~4* it is not surprising that a uniform reduction in cytochrome oxidase activity was produced in this nucleus. This contrasts with recent studies in the kitten, whose dLGn displays a distinct laminar organization” and exhibits a differential reduction in cytochrome oxidase activity between the monocular and binocular segments after TTX treatment.“2 The precise locus for the reduction in cytochrome oxidase activity was not determined. One could argue that the reduction occurred solely within the retinal terminals, and that TTX had no effect on the activity of the postsynaptic neurons. This seems highly unlikely since intraocular TTX also reduces cytochrome oxidase activity in the visual cortex.‘? In addition, since the normal optic axon population was maintained during TTX treatment, the reduced activity in the SC and dLGn was not due to a loss of mitochondria within degenerating optic terminals. Incorporation and axonal transport cf [‘Hlproline

These results demonstrated that the amount of rapidly transported [3H]proline-IabeIed materials in developing optic nerves is independent of retinal impulse activity. No tolerable dose of TTX suppressed protein transport, either in the short-term or following a prolonged period of impulse inhibition. In addition, long-term removal of optic impulses by TTX had no effect on the incorporation of [‘Hlproline into retinal proteins. These data indicate that the amount of [3H]proline-la~led proteins synthesized in the ganglion cell soma, some of which are destined for the nerve terminals, was unaffected by TTX treatment. These findings support previous reports in adult animals indicating that the axonal transport of amino acid labeled proteins is activity independent.9~‘6~‘5~‘o~39~5’~54 It should be noted, however, that measuring the total amount of accumulated radioactivity does not suggest that the transport of all proline-labeled proteins is TTX-insensitive. Conceivably, the transport of some protein species may be altered by TTX and, if in low proportion to the total amount of transported radioactivity, may not be detected by this technique. Studies in progress are currently investigating this possibility.

decrease in glycoprotein transport as early as 5dpn in the present study indicates that retinal ganglion cells are TTX-sensitive in the early stages of postnatal development. Since the size of the decrease in transported material was greater at earlier ages, the possibility exists that retinal ganglion cefls became more resistant to TTX as development proceeded. This wBs supported by the incorporation data, in that the amount of reduction seen in the TTX-treated retina at 9 and 13 dpn was greater than at 21 dpn. A temporal sensitivity of retinal ganglion cells to cotchicine impairment of axonal transport has been reported4? and though no correlation between TTX and colchicine treatment can be made. the available studies suggest that axonal transport is differentially sensitive to exogenous influences during development. Possible rne~han~s~lsqf the action of tetrod~~t~~xin on axonal transport

Inhibition of optic impulses by intraocular TTX injection may occur by several mechanisms. In optic axons, the covalent binding of TTX to Na + channels would prevent the inward Na+ flux required for the rising phase of the action potential.29,43 Similarly, TTX may also block Nat entry into the retinal ganglion cell soma, thus preventing generation of impulses at the initial segment. In addition, Na+ mediated electrical input onto ganglion cells could also be inhibited by TTX.4” Since TTX may block optic impulses at any one or all of these sites. a number of mechanisms could account for the reduced amount of axonally-transported glycoproteins seen in the present experiments. First. TTX may interfere with the local incorporation of [3H]fucose into axonal proteins. Several lines of evidence argue against this possibility. Most. if not all, of the protein moieties of axonal glycoproteins are synthesized in the cell body4,4’ and the terminal addition of fucose onto proteins is thought to occur exclusively in the Golgi apparatus, which is confined to the cell body. In addition, fucosyltransferases are localized to microsomal fractions and not concentrated within synaptosomes.” Second, TTX may indirectly limit Ca’

Incorporation and uxonal transport of [‘H&~ose

Intraocular TTX decreases the amount of fucosylated glycoproteins axonally transported in the optic nerve between 5 and 21 dpn. This effect was probably due to the reduced incorporation of [3H]fucose into retinal and therefore ganglion cell glycoproteins, thus decreasing the amount of label accumulated at the dLGn. These data support recent studies in the regenerating retino-optic system of adult goldfish, which indicated a reduced incorporation of [3H]glucosamine into retinal glycoproteins and a decrease in the fast axonal transport of these materials in the optic nerve as a result of TTX treatmenti The fact that TTX caused an immediate

+ entry into the soma is inhibited.‘7.28.“4.” Recent evidence also suggests that restricting Ca’* influx into the soma limits the transfer of [‘Hlfucosyl glycoproteins from the Golgi apparatus to the fast transport system, 34 thus reducing the amount of glycoproteins delivered to the nerve terminals. Electron microscopic and autoradiographic data show

TTX effects on axonal transport in developing optic nerve that when Ca*+ entry into the soma is inhibited

by exogenously-applied Co *+, labeled glycoproteins become concentrated over an enlarged and vacuolated Golgi apparatus.34 Hence, the initiation of glycoprotein transport after Golgi processing may be mediated by intracellular calcium. Whatever the precise mechanism, the present experiments indicate that the axonal transport of glycoproteins is indirectly linked to the Na+-generated impulse activity of the developing neuron. Our electron microscopic data further suggest that the site of inhibition may be at the Golgi apparatus, since after TTX treatment the retinal ganglion cells exhibited swollen Golgi cisternae remarkably similar to those seen after restriction of CA*+ influx.34,38

1037

TTX selectively suppresses the transport of particular axonal constituents, including nucleosides, glycolipids and g1ycoproteins.‘8.25 In the case of glycoproteins, many have been implicated in terminal recognition, cellLcel1 adhesion and synaptogenesis.‘,’ Some of these also undergo developmental regulation.‘4.47.67Additionally, the transneuronal transfer of glycoproteins from retinal terminals to visual relay neurons has been offered as one mechanism for synchronizing the development of subcortical and cortical visual centers.‘5.43,6”,70Hence, a close affiliation between impulse activity and glycoprotein transport, may be important in the regulation of cell-cell interactions associated with axonal growth and ,development.

Second Study Program (ed. Schmidt F. O.), pp. 728-736. Rockefeller University Press, New York. Barondes S. H. (1974) Synaptic macromolecules: identification and metabolism. A. Rev. Biochem. 43, 147. Barondes S. H. and Rosen S. D. (1976) Cell surface carbohydrate-binding proteins: role in cell recognition. In Neuronal Recognition (ed. Barondes S. H.). pp. 331-356. Plenum Press, New York.

Bennett G., Di Giamberardino L., Koenig H. L. and Droz B. (1973) Axonal migration of protein and glycoprotein to nerve endings. II. Radioautographic analysis of the renewal of glycoproteins in nerve endings of chicken ciliary ganglion after intracerebral injection of [3H]fucose and [3H]glucosamine. Brain Res. 60, 129-146. Bennett G. and Leblond C. P. (1977) Biosynthesis of the glycoproteins found in plasma membrane and in secretory materials. Histochem. J. 9, 393417. Black M. M. and Lasek R. L. (1979) Axonal transport of actin: slow component b is the principal source of actin for the axon. Brain Res. 171, 401413. Boegman R. J. and Riopelle R. J. (1980) Batrachotoxin blocks slow and retrograde axonal transport in viva. Neurosci. Lefts 18, 143-147.

Bull R. J. and Trevor A. J. (1972) Saxitoxin and tetrodotoxin and cation fluxes in isolated cerebral tissues. J. Neurochem. 19, 999-1009.

Bull R. J. and Trevor A. J. (1972) Sodium and the flux of calcium ions in electrically-stimulated

tissue. J. Neurochem.

19, 101 I-1022.

Cuenod M. and Schonbach J. (1971) Synaptic proteins and axonal flow in the pigeon visual pathway. J. Neurochem. 18, 809-816.

13. Cunningham T. J. and Lund R. D. (1971) Laminar patterns in the dorsal division of the lateral geniculate nucleus of the rat. Brain Res. 34, 394. 14. DeSilva N. S., Gurd J. W. and Schwartz C. (1979) Developmental alteration of rat brain synaptic membranes. Reaction of glycoproteins with plant lectins. Brain Res. 165, 283-293. 15. Drager V. C. (1974) Autoradiography of tritiated proline and fucose transported transneuronally from the eye to the visual cortex in pigmented and albino mice. Bruin Res. 82, 284292. 16. Droz B. (1975) Synthetic machinery and axoplasmic transport: maintenance of neuronal connectivity. In The Nervous System Vol. 1. The Basic Neurosciences (ed. Tower D. B.), pp. I1 I-127. Raven Press, New York. 17. Droz B. and Koenig H. L. (1970) Localization of protein metabolism in neurons. In Protein Metabolism af the Nervous System (ed. Laitha A.). DD. 93-108. Plenum Press. New York. 18. Edwards D. L. and Grafstkin B. (1984) Intraocular injection of tetrodotoxin in goldfish decreases fast axonal transport of [‘Hlglucosamine-labeled materials in optic axons. Brain Res. 299, 190-194. 19. Elam J. S. and Agranoff B. W. (1971) Rapid transport of protein in the optic system of the goldfish. J. Neurochem. 18, 375-387.

20. Fitkova E. and Hassler R. (1969) Quantitative morphological changes in visual centers in rats after unilateral deprivation. J. camp. Neurol. 135, 1677178. 21. Forman D. S., Grafstein B. and McEwen B. S. (1972) Rapid axonal transport of [‘Hlfucosyl glycoproteins in the goldfish optic system. Bruin Res. 48, 327-342. 22. Fu S. C., Cruz T. F. and Curd J. W. (1981) Development of synaptic glycoprotein: effect of postnatal age on the synthesis and concentration of synaptic membrane and synaptic junctional fucosyl and sialyl glycoproteins. J. Neurochem. 36. 133881351.

1038

R.

MATTHEWS

23. Goodrum J. F. and Morel1 P. (1982) Axonal transport, deposition and metabolic turnover of glycoproteins in the rar optic pathway. J. Ncurochem. 38, 696704. 24. Goodrum J. F.. Toews A. D. and Morel1 P. (1979) Axonal transport and metabolism of [‘Hlfucose and [‘5S]sulfate-labelled macromolecules in the rat visual system. Bruin Res. 176, 255 272. 25. Grafstein B., Edwards D. L. and Albert R. M. (1981) Axonal transport and neuronal activity. In l}ltf/r~atio~o~CeI! Biology 198&1981 fed. Schweiger H. G.). pp. 728-736. 26. Grafstein B. and Forman D. S. (19801 Intracellular transnort in neurons. Phvsiol. Rer. 1167-1283 27. Hammerschlag R., Chiu A. Y. and Dravid A. R. (1976) inhibition of fast ax&al transport of [‘Hlprotein by cobalt ions. Brain Res. 114, 353.-358. 28. Hammerschlag R. and Lavoie P.-A. (1979) Initiation of fast axonal transport: involvement of calcium during transfer of proteins from Golgi apparatus to the transport system. Neuroscience 4, 1195. 1201. 29. Hille B. (1975) The receptor for tetrodotoxin and saxitoxin: a structural hypothesis. Biophvs. J. IS, 615-619. 30. Hughes A. (1977) The pigmented rat optic nerve: fibre count and libre diameter spectrum. .i. camp. Neurnl. 176, 263-268. 31. Karlsson J. 0. and Sjostrand J. (1971) Rapid intracellular transport of fucose-containing glycoproteins rn retinal ganglion cells. J. Neurochem. 18, 2209- 2216. 32. Karlsson J. 0. and Sjostrand J. (1972) Axonal transport of proteins in retinal ganglion cells. Amino acid incorporation into rapidly transported proteins and distribution of radioactivity to the lateral geniculate body and the superior colliculus. Brain Res. 37, 2799285. 33. Lam K., Sefton A. J. and Bennett M. R. (1982) Loss of axons from the optic nerve of the rat during early postnatal development. De01 Brain Res. 3, 457-491. 34. Lavoie P.-A. and Bennett G. (1983) Accumulation of [3H]fucose-la~lled gtycoproteins in the Golgi apparatus of dorsal root ganglion neurons during inhibition of fast axonal transport caused by exposure of the gangiion to Co?+-containing or Ca*+-free medium. Neuroscience 8, 351--362. 35. Lavoie P.-A., Hammerschlag R. and Tjan A. (1978) Cobalt ions inhibit fast axonal transport of [lH]glycoproteins but not glycosylation. Bruin Res. 149, 535-540. 36 Leblond C. P. and Bennett G. (1977) Role of the Golgi apparatus in terminal glycosylation. In frrternational CeN Bio/og~ (eds Brinkley B. R. and Porter K. R.), pp. 326336. Rockefeller University Press, New York. 37. Levin B. E. (1977) Axonal transport of f3H]fucosyl glycoproteins in noradrenergic neurons in the rat brain. Brain Res. 130,421432.

38, Lindsey J. D., Hammerschlag R. and Ellisman M. H. (1981) An increase in smooth endoplasmic reticulum and a decrease in Golgi apparatus occur with ionic conditions that block initiation of fast axonal transport. Brain Res. 205, 275-287. 39. Lux H. D., Schubert P., Kreutzberg G. W. and Globus A. (1970) Excitation and axonal flow: autoradiographic study on motorneurons intraceltularly injected with a (‘Hlamino acid. Expl Brain Res. IO, 1977204. 40. Margolis R. K. and Margofis R. V. (1972) Disposition of fucose in the brain. J. ~euru~hem. 19, 3023-1030. 41. Margolis R. V. and Margolis R. K. (1977) Metabolism and function of glycoproteins and glyco~minogiycans in nervous tissue. ht. J. B&hem. 8, 85-91. 42. Matthews M. A., Narayanan C. H., Narayanan Y. and Siegenthaler-Matthews D. J. (1982) Inhibition of axoplasmic transport in the developing visual system of the rat. III-Electron microscopy and Golgi studies of retino-fugal synapses and post-synaptic neurons in the dorsal lateral geniculate nucleus. Neuroscience 7, 4055422. 43 Matthews M. A., West L. C. and Clarkson D. B. (1982) Inhibition of axoplasmic transport in the developing visual system of the rat. II-Quantitative analysis of alterations in transport of tritiated proline or fucose. Neuroscience 7, 385-404.

44 McEwen B. S., Forman D. S. and Grafstein B. (1971) Components of fast and slow axonal transport in the goldfish optic nerve. J. Neurobiol. 2, 361-377. 45. Meves H. (1975) Calcium currents in squid giant axon. Phil. Trans. R. Sot. Set. B. 270, 377.-387. 46. Miller R. F. and Dacheux R. (1976) Dendritic and somatic spikes in mudpuppy amacrine cells: identification and tetrodotoxin sensitivity. Brain Res. 104, 157-162. changes during development of the chick neural retina. J. cfli 47. Mintz G. and Glaser L. (1978) Specific glycoprotein _~ Biol. 79, 132-137.

48. Montero V. M. and Guillery R. W. (1968) Degeneration in the dorsal lateral geniculate nucleus of the rat followjng interruption of the retinal or cortical connections. J. camp. Neural. 134, 21 l-242 49. Narahashi T., Haas G. T. and Therrien E. F. (1967) Saxitoxin and tetrodotoxin: comparison of nerve blocking mechanism. Science, N. Y. 157, 1441-1442. 50. Narahashi T., Moore J. W. and Scott W. R. (1964) Tetrodotoxin blockage of sodium conductance increase in lobster giant axons. J. Am. Physiol. 47, 965-974. 51. Ochs S. (1972) Fast transport of materials in mammalian _ _ nerve fibers. Science, N.Y. 176, 252-260. . . _. ‘ * 1. I 52. Ochs S. and Hollingsworth D. (1971) Dependence ot fast axonat transport m nerve on oxtaartve me~aool~srn. J. Neurochem. IS, 107-l 14. 53. Perry V.H. (1981) Evidence for an amacrine cell system in the ganglion cell layer of the rat retina. Neuroscrcnce 6, 93 l-944. 54. Pestronk A.. Drachman D. B. and Griffin J. W. (1976) Effect of muscle disuse on acetylcholine receptors. Nature, Lotzd. 260, 352-353. 55. Peters A. (1970) The fixation of central nervous tissue and the analysis of electron micrographs of the neuropil with special reference to the cerebral cortex. In Conre~porary Research Methods in Neuroanatomy (eds Nauta W. J. H. and Ebbesson S. 0. E.), pp. 5676. Springer, New York. 56. Peters A. and Palay S. (1966) The morphology of laminae A and Al of the dorsal nucieus of the lateral geniculate body of the cat. J. Anal. 109, 451-486. 57. Polyak S. (1957) The Vertebrate Visual System. University of Chicago Press, Chicago. 58. Quarles R. H. and Brady R. 0. (1971) Synthesis of glycoprotein and gangliosides in the developing rat brain. J. Neurochem. 18, 1809-I 820.

TTX effects on axonal transport in developing optic nerve

1039

59. Rager G. and Rager U. (1978) Systems-matching by degeneration. I-A quantitative electron microscopic study of generation and degeneration of retinal ganglion cells in the chicken. Expl Brain Res. 33, 6>78. 60. Rambourg A. and Leblond C. P. (1967) Electron microscope observations on the carbohydrate-rich cell coat present at the surface of cells in the rat. J. Cell Biol. 32, 27-53. 61. Schlichter D. J. and McClure W. 0. (1974) Dynamics of axoplasmic transport in the optic system of the rat. Expl Brain Res. 21, 83-95. 62. Schonbach J., Schonbach C. and Cuenod M. (1971) Rapid phase of axoplasmic flow and synaptic proteins: an electron microscopical autoradiographic study. J. camp. Neurol. 141, 485498. 63. Siegel G. .I., Albers R. W., Katzman R. and Agranoff B. W. (1976) Basic Neurochemi.swy. pp. 329-341. Little, Brown and Co. Boston. 64. Sjostrand J., Karlsson J.-O. and Marchismo P. C. (1973) Axonal transport in growing and maturing retinal ganglion cells. Brain Res. 62, 395-397. 65. Specht S. C. (1977) Axonal transport in the optic system of neonatal and adult hamsters. Expl Neural. 56, 252-264. 66. Specht S. C. (1980) Transneuronal transfer of radioactive substances during postnatal development of the hamster visual system. Expl Neural. 69, 213-225. 67. Specht S. C. (1982) Postnatal changes in [‘Hlfucosyl glycopeptides of hamster optic nerve synaptosomal membrane. Devl Brain Res. 4, 109-I 14. 68. Specht S. C. and Grafstein B. (1973) Accumulation [3H]fucose into the eye. Expl Neural. 41, 705-722.

of radioactive protein in mouse cerebral cortex after injection of

69. Specht S. C. and Grafstein B. (1977) Axonal transport and transneuronal transfer in mouse visual system following injection of [3H]fucose into the eye. Expl Neural. 54, 352-368. 70. Sturman J. A. (1979) Taurine in the developing rabbit visual system: changes in concentration and axonal transport including a comparison with axonally transported proteins. J. Neurobiol. 10, 221-237. 71. Wong-Riley M. (1979) Changes in the visual system of monocularly sutured or enucleated cats demonstrable with cytochrome oxidase histochemistry. Brain Rex 171, 1 l-28. 72. Wong-Riley M. and Riley D. A. (1983) The effect of impulse blockage on cytochrome oxidase activity in the visual system. Brain Res. 261, 185-193. 73. Zatz M. and Barondes S. H. (1970) Fucose incorporation into glycoproteins into mouse brain. J. Neurochem. 17, 157-163. 74. Zatz M. and Barondes S. H. (1971) Rapid transport Neurochem. 18, 1125-l 133.

of fucosyl glycoproteins to nerve endings in mouse brain. J.

(Accepted 24 May 1985)