Journal of Histochemistry & Cytochemistry

Journal of Histochemistry & Cytochemistry http://jhc.sagepub.com/

Co-injection of wheat germ agglutinin-HRP and choleragenoid-HRP into the sciatic nerve of the rat blocks transganglionic transport. H Liu, I J Llewellyn-Smith and A I Basbaum J Histochem Cytochem 1995 43: 489 DOI: 10.1177/43.5.7730587 The online version of this article can be found at: http://jhc.sagepub.com/content/43/5/489

Published by: http://www.sagepublications.com

On behalf of:

Official Journal of The Histochemical Society

Additional services and information for Journal of Histochemistry & Cytochemistry can be found at: Email Alerts: http://jhc.sagepub.com/cgi/alerts Subscriptions: http://jhc.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav

>> Version of Record - May 1, 1995 What is This?

Downloaded from jhc.sagepub.com by guest on October 11, 2013

0022-15I4/95/$3.30 The Joumal of Histochemistry and Cytochemistry Copyright 0 1995 by The Histochemical Society, Inc.

Vol. 43, No. I, pp. 489-495. 1991 Printed in US.A.

Original Article

I

CO-injection of Wheat Germ Agglutinin-HRP and Choleragenoid-HRP into the Sciatic Nerve of the Rat Blocks Transganglionic Transport' HANTAO LIU, IDA J. LLEWEUYN-SMITH, and ALLAN I. BASBAUM' Departments of Anatomy and Physiology and Keck Center for Integrative Neurosciences, University of Cal$ornah San Francisco, AIB), and Centre for Neumscience and Department of Medicine, Flinders University, Bedford Park, San Francisco, Cali~ornia(HL, South Australah, Australia (IJL). Received for publication September 19, 1994 and in revised form December 14, 1994; accepted December 22. 1994 (4A3492).

We report on the surprisingloss of transganglionic and retrograde labeling in the spinal cord of the rat after co-injection of the tracers wheat germ agglutinin-HRp (WGA-HRP) and choleragenoid toxin-HRP (CTEi-HRP) into the sciatic nerve. Injection of WGA-HRP alone produced a pattern of transganglionic label consistent with transport by small-diameter primary afferent fibers. Small cell bodies were labeled in the ipsilateral dorsal root ganglion (DRG) and there was dense terminal labeling in the superficial dorsal horn of the lumbar spinal cord. Injection of CTB-HRP alone produced a pattern of transganglioniclabeling consistent with transport by large-diameterprimary afferent fibers. Large cell bodies were labeled in the DRG and there was dense terminal labeling in the nucleus proprius (Laminae HI-V) in the spinal cord. CTBHRP also produced extensive retrograde labeling of ventral horn motor neurons. When the two tracers were CO-injected,we found few labeled cells in the ipsilateral DRG and there was almost complete loss of transganglionic terminal labeling in the lumbar spinal cord. Retro-

Introduction A number of studies have established the utility of studying the spinal cord termination of primary afferent fibers by transganglionic transport of either wheat germ agglutinin conjugated to horseradish peroxidase (WGA-HRP) or the B fragment of cholera toxin conjugated to horseradish peroxidase (CTB-HRP). Of particular interest is that there is a differential transport of WGA-HRP and CTB-HRP by small- and large-diameterprimary afferents, respectively. For example, application of WGA-HRP to the cut end of a peripheral nerve or injection into skin results in transganglionic

Supported by NIH grants NS 14627, NS 21445, and DA 08377. Correspondence to: Dr. Allan Basbaum, Dept. of Anatomy, UCSF, Box 0452, San Francisco, CA 94143.

grade labeling of motor neurons was also significantly reduced. Even when one of the tracers (e.g., WGA-HRP) was injected 24 hr after and up to 10 m m proximal to the site of the f i t tracer (e.g., CTB-HRP), an inhibitory interaction was detected. The labeling pattern was always characteristic of the first tracer injected. Since the loss of labeling was observed in the DRG and in the spinal cord, we conclude that the mutually inhibitory interactions m r during uptake of the tracers by the sciatic nerve and/or in the mechanism of transport to the dorsal root ganglion and beyond. The fact that redud labeling was observed even when the injections were separated temporally and spatially suggests that the underlying mechanism is more complicated than mere creation of a WGA-HRP/CTEi-HRP complex that cannot be taken up and transported by the sciatic nerve. (J Histochem Cytochem 43489495,1995) KEY WORDS: WGA-HRP; CTE-HRP; Dorsal root ganglion; Trans-

ganglionic transport; Primary afferents; Rat.

labeling almost exclusively in Laminae I and I1 (1,6,9,11,12,15,16,18), which is the predominant target of unmyelinated and small myelinated primary afferent fibers (7). When CTB-HRP was injected into peripheral nerve or applied to the cut end of dorsal roots, transganglionic labeling was most concentrated in the nucleus proprius of the dorsal horn (Laminae 111, IV, and V), in the intermediate gray, and in the ventral hom, i.e., in regions that receive largediameter afferent fiber projections. The substantia gelatinosa, Lamina 11, which receives dense projections from both low- and highthreshold unmyelinated primary afferent fibers (7,17), shows almost no labeling (6,13-16).The dissociation between the patterns of WGA-HRP and CTB-HRP labeling, however, is not complete; some anterograde terminal labeling is observed in Lamina I, albeit less than after injection of WGA-HRP. Finally, both WGA-HRP and CTB-HRP are transported retrogradely to ventral horn motor neurons; typically, CTB-HRP labels greater numbers of motor neurons and the dendritic arbor is more extensive. 489

490

Because large- and small-diameterdorsal root ganglion cells are and CTB-HRP, Robertson and Grant (16) concluded that the differential labeling pattern in the spinal cord probably resulted from differential uptake and/or transport by large- and small-diameter primary afferents. The underlying mechanism is hypothesized to reflect the fact that the plasma membrane of small-diameterafferents contains higher concentrationsof N-acetyl-D-glucosamine, the carbohydrate for which WGA has highest affinity, and largediameter fibers are enriched in GM1, the ganglioside to which cholera toxin binds (2,16). Following on the proposal of Robertson and Grant (16) and with a view to maximizing the amount of transganglionic labeling in the spinal cord, we have used mixtures of WGA-HRF' and CTB-HRP in our studies of the central projections of primary afferent fibers. In this report we demonstrate that after combined injections of WGA-HRP and CTB-HRP into the sciatic nerve of rat, there is almost complete loss of retrograde and anterograde transport of the tracers, as measured by labeling in the dorsal root ganglion (DRG) and spinal cord. A preliminary report of these observations has appeared (8). also respectively labeled by peripheral injection of WGA-HRP

Materials and Methods Experimentswere performed on adult male Sprague-Dawley rats (240-280 g). The first studies established that under the conditions of our experiments, there is indeed a differential pattern of labeling with WGA-HRP and CTB-HRP. Animals were anesthetized by IP injection of sodium pentobarbital (60 mglkg) and the sciatic nerves were exposed bilaterally by blunt dissection. Using a 10 pl Hamilton syringe, we injected 1.5 p1of 5 % WGA-HRP (Sigma; St Louis, MO) into the left sciatic nerve and 1.5 p1 of 1% CTB-HRP (List Biological Laboratories; Campbell, CA), into the right sciatic nerve (n = 4). Several vehicles were used in these experiments: 0.1 M phosphate buffer (PB) and 5% dimethylsulfoxide (DMSO), 0.1 M PB and 0.4% DMSO,or double-distilled water. In six rats we injected a combination of J % WGA-HRP and 1%CTB-HRP (total volume of either 1.5 or 3.0 pl) into the left sciaticnerve and either WGA-HRP or CTB-HRP into the right sciatic nerve (volume 1.5 or 3.0 PI). In another group of rats (n = 10). we first injected 5% WGA-HRP into the left sciaticnerve and 1% CTB-HRP into the right sciatic nerve. Twentyfour hr later the nerves were re-exposed, at which point CTB-HRP was injected into the left sciatic nerve, 5 mm (n = 8) or 10 mm (n = 2) proximal to the site of the WGA-HRP injection and WGA-HRP was injected into the right sciaticnerve, the same distance proximal to the site ofthe CTB-HRP injection. Figure 3 schematicallyillustrates this experiment, which evaluated whether temporal and spatial separation of the injections influences the labeling pattern produced by combined injections of the two tracers. To control for the possibility that mere manipulation of the nerve on Day 1 could interfere with uptake and/or transport of the tracer injected on Day 2, another group of rats (n = 4) received an injection of either 5Yo DMSO or saline on the first day, followed by a more proximal injection of either WGA-HRP or CTB-HRP 24 hr later. Forty-eight to 72 hr after the last injections, all animals were deeply anesthetized with pentobarbital (100 mg/kg) and perfused through the ascending aorta with tissue culture medium (DMEMIHamm'sF12; Sigma D-8900) and then with a fixativesolution containing 2.5 % glutaraldehyde and 0.5% formaldehyde in 0.1 M phosphate buffer, pH 7.4. The lumbar spinal cord (L&) and the corresponding DRG were removed and postfixed in the same solution for 1 hr. The lumbar spinal cord was cut transversely at 50 pm on a vibratome; the DRGs were cryoprotected in a phosphate-buffered 30% sucrose solution and then cut at 20 pm with a

LIU, LLEWELKN-SMITH, BASBAUM

cryostat. Both the vibratome and cryostat sections were processed for HRP histochemistry according to the tetramethylbenzidine (TMB)-tungstate method of Weinberg and Ven Eyck (21). After the reaction, the sections were mounted on slides and viewed with brightfield illumination.

Results Consistent with previous reports, we found that there was a dlfferential pattern of labeling produced by sciatic nerve injection of WGA-HRP or CTB-HRP. WGA-HRP injection resulted in dense labeling in the superficial dorsal horn, Laminae I and 11; very few fibers could be detected ventrally (Figure 1A, left). This pattern was observed at both 48 and 72 hr after injection of the tracer. We also noted retrograde labeling of motor neurons in the ventral horn. However, labeling of dendrites was limited (Figure 1A, left). In the L& DRG ipsilateral to the WGA-HRP injection, labeling of small-diameter cell bodies predominated. In fact, very few large ganglion cells were recorded (Figure 1B). The pattern of labeling produced by injection of CTB-HRP into the contralateralsciatic nerve, i.e., in the same rats, was completely different. The densest transganglionic labeling was in the neck of the dorsal horn (Laminae 111-VI); some labeled fibers could also be seen in Lamina I. The substantia gelatinosa, Lamina 11, was almost devoid of labeling (Figure 1A, right). In contrast to the pattern of cell labeling in the DRG &er WGA-HRP injection, CTB-HRF' injection resulted in the 16 ling of many large-diameter cell bodies; only a few small DRG cells were labeled after CTB-HRP injection (Figure IC). Consistentwith the labeling of large-diameter DRG cells, we also found dense labeling in the ipsilateral dorsal columns (Figure lA, right). Finally, sciatic nerve injection of CTB-HRP resulted in labeling of large numbers of ventral horn motor neurons; the dendritic arbors of these neurons were particularly well labeled (Figure lA, right). CO-injection of WGA-HRP and CTB-HRP into the sciatic nerve resulted in the transganglionic labeling of only a few terminals in the spinal cord, in the marginal layer, Lamina I (Figure 2A, left). This was found whether the total volume injected was 1.5 or 3.0 pl. In three of six rats, Lamina I contained label somewhat comparable to that observed when CTB-HRP was injected alone; in the others, the amount of label in Lamina I was also significantly reduced. The number of labeled DRG cells was also dramatically reduced when the mixture of the tracers was injected, again regardless of injection volume (Figure 2B). Neither small nor large cell bodies were labeled. Finally, after co-injectionof the tracers, retrograde labeling of motor neuron somata was also reduced (Figure 2A, left).

Control Experiments To control for the possibility that the absence of labeling resulted from some artifact of perfusion, these same rats received an injection of CTB-HRP alone into the contralateral sciatic nerve. The pattern of spinal cord and DRG labeling on the CTB-HRP-injected side was normal (Figure 2A, right, and Figure 2C). In another control experiment, we adjusted the volume of the single tracer injection so that it corresponded to that of the combined injection, i.e., 3.0 pl. In some animals, 3.0 pl of a 1% solution of CTB-HRP was injected by itself; in others, 3.0 pl of an 0.5% solution of CTB-HRP

Figure 1. Differential panern of labeling in (A) the spinal cord and in (B,C) the DRG produced by sciatic nerve injection of WGA-HRP or CTBHRP. (A) After injection of WGA-HRP (W. left) the transganglionic labeling of primary afferentterminals is concentrated in the ipsilateral superficial laminae of the dorsal horn. There is also moderate retrograde labeling of ventral horn motor neurons. (E)Labeled cell bodies in the ipsilateral DRG are predominantly of small diameter. (C) After injeclionof CTB-HRP (right) the transganglionic labeling of primary afferents is concentrated in the ipsilateral nucleus proprius (np) ventral to the substantia gelatinosa (sg). which itself is not labeled. A few labeled terminals are present in Lamina I. There is extensive retrograde labeling of ventral horn motor neuron cell bodies and dendrites. Labeled cell bodies in the ipsilateral DRG (C) are predominantlyof large diameter. consistent with the presence of considerable transganglionic labeling of axons in the dorsal columns (dc; arrow in A). cc. central canal. Bars = 100 pm.

Figure 2. Results of combined sciatic nerve injection of (A) WGA-HRP and CTB-HRP (W+C; left) with lhat produced by injection of (C) CTB-HRP alone (right). When the two tracers are combined. there is always an almost complete loss of transganglionic labeling in the dorsal hom. In this experiment, only a few labeled afferents were detected in Lamina I (A, arrow). The overall labeling of motor neurons was typically less than that o b s e d after injection of WGA-HRP alone, allhough individual cells were often densely labeled (compare with Figure l A , left). (E) In the DRG ipsilateral to the combined injection. almost no labeled neurons were detected. The fact that the pattern of labeling produced by CTB-HRP alone was comparable to that observed in other experiments (see Figure 1A. right) indicates that the loss of labeling after combined injection was not due to inadequate processing of the tissue. In this experiment CTB-HRP injection resulted in particularly dense labeling of axons in the ipsilateral dorsal columns (A, dc) and of large-diameter cell bodies in the ipsilateral DRG (C) Bars = 100 pm.

WGA-HRP AND CTBHRP TRANSGANGLIONICTRANSPORT

WGA-HRP CT&HRP

WGA-HRP CTEHRP

Figure 3. Schematic illustrationof the experimental protocolthat examined the

effect of sequential injectionsof the two tracers. On day 1 we injected WGA-HRP into the lef sciatic nerve and CTB-HRP into the right sciatic nerve. Twenty-four hr later we injected CTB-HRP into the left sciatic nerve and WGA-HRP into the right sciatic nerve. The second injection was made either 5 or 10 mm proximal to the first. In control experiments, the first injection consisted of an equivalent volume of saline or DMSO.

was injected by itself. In both cases, there was considerable transganglionic and retrograde labeling in the ipsilateral lumbar spinal cord, comparable to that produced by injection of 1.5 p1 of 1% CTl3-HRP. When the combination tracer injection was made with different vehicles (5% DMSO, 0.4% DMSO, or double-distilled water) we still found almost no evidence of transganglioniclabeling. Befause the reduction in labeling might have resulted from some unusual complexing of the two tracers, we also evaluated the effect of separating the injections of the two tracers, both in time (24 hr) and location (5- or 1o-mm separation on the sciatic nerve). In one sciatic nerve, the WGA-HRP injection preceded the CTB-HRP injection; the order was reversed in the contralateral nerve (Figure 3). The rats were perfused 48 hr after the second injection, i.e., 72 hr after the first. The results from these experiments were very clear. On both sides of the spinal cord, the pattern of labeling was characteristic of the first tracer that was injected, i.e., when WGA-HRP was injected first, small neurons were labeled in the ipsilateral DRG and transganglionic labeling was most intense in the superficial dorsal hom (Figure 4A, left, and Figure 4B). When CTB-HRP was injected first, large neurons were labeled in the DRG and transganglionic labeling was most intense in the nucleus proprius and in motor neurons (Figure 4A, right, and Figure 4C). To control for the sequentialoperative procedures,we also evaluated the effect of a sham tracer injection (5% DMSO or saline) 24 hr before and 5 or 10 mm distal to one of the two tracer injections. In these cases, the pattern of labeling was always characteristic of the tracer injected on the second day. Importantly, since this was 48 hr after injection, the absence of labeling in groups that received sequential tracer injections cannot be explained by insufficient time for uptake and transport.

Discussion These results indicate that when the two anterograde tracers WGA-HRP and CTI3-HRP are co-injected into the sciatic nerve there is a dramatic decrease in the numbers of labeled neurons in the b-4,DRG and in the amount of transganglionic terminal labeling in the lumbar spinal gray matter. There is also a very significant decrease in the retrograde labeling of motor neurons. De-

493

spite the very dramatic mutually inhibitory effects observed in the present experiments, La Motte and colleagues (6) reported very different results. They found that more widespread labeling was produced by co-injectionof the same two tracers, as might a priori have been predicted. Since the source of the tracers appears to be the same in our two studies, we have no explanation for the discrepancy. Although subtle differencesin the sciatic nerve injection methods may be relevant, these could not explain the results from experimentsin which the uacers were injected on two different days. Importantly, the control studies demonstrated that neither differences in volume of injection nor nature of the vehicle could account for the results with combined tracer injection. Although 5% DMSO could conceivably have resulted in loss of labeling because of leakage and therefore dilution of the tracers, the pattern of labeling when the vehicle was 0.4% DMSO, which was used in the study by La Motte et al., or double-distilled water, was not different. Since our results are unexpected, it was important to establish that the pattern of labeling we observed when one of the tracers was injected alone is comparable to that reported in the literature (1,6,15,16,22).This was indeed the case. WGA-HRP alone resulted in the labeling of small-diametercell bodies in the DRG and dense transganglionic terminal labeling in the superficial laminae of the cord. CTB-HRP resulted in labeling of large-diameter DRG cells, dense terminal labeling in the nucleus proprius and in the intermediate gray of the cord, and dense fiber staining in the ipsilateral dorsal columns. The latter almost certainly derives from largediameter primary afferent fibers. It is generally assumed that the different labeling by the two tracers reflects their selective binding to different cell surface molecules, a carbohydrate (N-acetyl-D-glucosamine) in the case of WGA-HRP and GM1 ganglioside in the case of Cm-HRP (2-5,10,16,19,20). For this reason, the loss of labeling when the two tracers were co-injected was surprising. We considered the possibility that co-injection of the tracers resulted in the formation of a complex that may not bind to the appropriate sites for the individual tracers. To test for this possibility, we performed experiments in which the tracers were injected at different times. Under these conditions, we found that the pattern of labeling in the spinal cord and DRG was always characteristic of that produced by the first tracer. This inhibitory interaction was true whether the first tracer injected was WGA-HRP or CTB-HRP. Since the first tracer was injected distal to the second tracer, we presume that the loss of labeling from the second tracer resulted from transport of the first tracer to the site of injection of the second. Although it is possible that the first tracer masked the appropriate cell surface marker recognized by the co-injected tracer, this hypothesis is unlikely for two reasons. First, the two surface markers are located on different neurons, so that interference should be minimal. Second, sufficient time (up to 48 hr) was allowed for each tracer to displace a tracer of much lower affinity. Finally, since a prior injection of 5% DMSO or saline did not interfere with labeling produced by the second injection, the absence of labeling from the second tracer cannot be explained by the injury that was made 24 hr earlier. It is significant that the second, i.e., proximally injected, tracer was present neither in the DRG nor in the spinal cord. This suggests that the problem lies in the uptake and/or transport of the second tracer by the sciatic nerve. In other words, the loss of trans-

C Figure 4. Results of an experiment that followed the protocol outlined in Figure 3.(A) WGA-HRP was injected first and this was followed by CTB-HRP (W-C) 24 hr later (left). On the right side, CTB-HRP was injected first and this was followed by WGA-HRP (C-W). On both sides of the spinal cord, the pattern of transganglionic and retrograde labeling was characteristic of that of the first tracer injected, i.e., WGA-HRP on the left and CTB-HRP on the right. The size of labeled cell bodies in the DRG was also characteristic of the first tracer injected. i.e.. small cells were recorded ipsilateral to the sciatic nerve that received a WGA-HRP injection first (E) and large neurons were recorded ipsilateral to the sciatic nerve that received a CTB-HRP injection first (C). Bars = 100 pm.

LIU, LLEWEHYN-SMITH, BASBAUM

ganglionic transport cannot be explained by some trapping of the second tracer in DRG cells that had already transported the first tracer. Furthermore, since we found a reduction in labeling of terminals in the dorsal horn and of motor neurons in the ventral horn, whatever underlies the effect must occur in both sciatic nerve afferent and efferent axons, i.e., both anterograde and retrograde uansport was affected. Particularly puzzling were the results from the experiments in which the tracer injections were made at different locations. Since there should be a separation in the populations of axons that take up the two tracers, we predicted that interference at the proximal injection site would be minimized. Perhaps the first tracer (e.g., WGA-HRP) is taken u p by all calibers of axon but is transported only a short distance, but far enough to interfere with the transport of the tracer that was injected on the second day and 5 or 10 mm proximally. To ensure that the inhibition of labeling did not reflect inappropriate choice of survival times after tracer injection, we evaluated the labeling patterns at two times, 48 and 72 hr, which previous studies had established as optimal times for DRG labeling (6,15) and for transganglionic labeling in the dorsal horn (6,9,12,16). We found that the inhibition of transport and labeling was independent of the survival time. In fact, one rat that received temporally separated injections survived for more than 80 hr after the second injection. Even in this rat, the pattern of labeling we observed was still characteristic of the first tracer injected. In summary, this study demonstrates that co-injection of the tracers WGA-HRP and CTBHRP into the sciatic nerve of rats results in a dramatic reduction in the amount of anterograde transganglionic labeling in the dorsal horn, in the number of retrogradely labeled DRG cells, and in the number and extent of the retrograde labeling of ventral horn motor neurons. These experiments indicate that, in addition to the contribution of cell surface markers to the uptake and transport of these tracers, as yet undefined factors may contribute to the recognition process through which these tracers are incorporated.

Acknowledgments We thank Mr Cannen Tam and Ms Hene Vddivia for tbeirphotograpbic expertire.

Literature Cited Arvidsson J, Pfaller K Central projections of C4-CS dorsal root ganglia in the rat studied by anterograde transport of WGA-HRP. J Comp Neurol 292:349, 1990 Cuatrecasas P Interaction ofVibrio cholerae enterotoxin with cell membranes. Biochemistry 12:3547, 1973

495

5 . King CA, van Heningen WE:Deactivation of cholera toxin by a sizlidaseresistant monosialoganglioside. J Infect Dis 127:630, 1973

6. LaMone CC, Kapadia SE, ShapiroCM: Central projections ofthe sciatic,

saphenous, median, and ulnar nerves ofthc rat demonstrated by transmdionic transport of choleragenoid-HRP (B-HRP)and wheat germ ;ggLtinin-HRP(WGA-HRP).J Comp Neurol 311:546. 1991 7. Light A, Perl E R Re-examination of the dorsal root projection to the spinal dorsal hom including observations on the differential tcrmination of coarse and fine fibers. J Comp Neurol 186:117, 1979

I, Basbaum AI: Complicationsfollowing coinjection ofwheatgerm agglutinin-HRPand choleragenoid-HRPinto the sciatic nerve of the rat. Soc Neurosci Abst 189333, 1992

8. Liu H, Llewellyn-Smith

9. Molander C, Grant G Cutaneous projections from the rat hindlimb

foot to the substantia gelatinosa of the spinal cord studied by transganglionictransport of WGA-HRF’ conjugate.J Comp Neurol237:476, 1985 10. Nagata Y,Burger MM: Wheat germ agglutinin:molecular chancteristics specifity for sugar binding. J Biol Chem 249:3116, 1974 11. Nyberg G, Blomqvia A The central projection of musde &rent fibers

to the lower medulla and upper cervical spinal cord an anatomical study in the cat with the uansganglionic method.J Comp Neurol23099, 1984 12. Nyberg G, Blomqvist A The termination of forelimb nerves in the

feline cuneate nucleus demonstrated by the transganglionic transport method. Brain Res 248:209, 1982 13. River+Melian C, Grant G Distribution of lumbar dorsal root fibers in the lower thoracic and lumbosacral spinal cord of the rat studied

with the choleragenoid horseradish peroxidase conjugate. J Comp Neurol 299:470, 1990 14. Rivero-Melian C, Grant G Somatotopicorganizationof hindlimb mus-

cles afferent projections to the column of Clarke of the rat studiedwith the choleragenoid horseradishperoxidase conjugate. Soc Neurosci Abst 14:693, 1988 15. Robertson B, Arvidsson J: Transganglionic transport of wheat germ agglutinin-HRPand choleragenoid-HRPin rat trigeminal primary sensory neurons. Brain Res 348:44, 1985 16. Robertson B, Grant G A comparison between wheat germ agglutinin and choleragenoid-horseradishperoxidase as anterogradely transported

markers in central branches of primary sensory neurons in the rat with some observations in the cat. Neuroscience 14985, 1985 17. Sugiura Y,Lee CL. Perl E R Central projections of identified, unmyelinated (C) afferent fiben innervatingmammalian skin. Science 234358, 1986 18 Sweet JE. Woolf CJ: The somatotropic organization of primary affer-

ent terminals in the superficial laminae of the dorsal horn of the rat spinal cord. J Comp Neurol 231:66, 1985 19. TrojanowskiJQ, GonatasJO, GonatasNK: Horseradish peroxidase (HRF’) conjugates of choleratoxin and lectins ue more sensitive retrograde transported markers than free HRP. Brain Res 231:33, 1982 20. Wan XS, TrojanowskiJQ, GonatasJO: Cholera toxin and wheat germ

1979

agglutinin conjugates as neuroanatomical probes: their uptake and dearance, transganglionic and retrograde transport and sensitivity. Brain Res 243:214, 1982 21. Weinberg RJ. Van Eyck SL A tetramethylbenzidineltungstatereaction for horseradish peroxidase histochemistry. J Histochem Cytochem

Holmgren J, Lonnroth I, Svennerholm L Tissue receptor for cholera exotoxin: postulated structurefrom studies with Gml ganglioside and related glycolipids. Infect Immun 8:208. 1973

22. WoolfCJ, ShortlandP, CoggeshallRE: Peripheral nerve injury triggers central sprouting of myelinated afferents. Nature 351:75, 1992

Gonatas NK,Harper C, Mitutani T, GonatasJO: Superior sensitivity of conjugates of horseradish peroxidase with wheat germ agglutinin for studies ofretrogradeaxonal transport.J Histochem Cytochem 27:728,

39:1143, 1991