- Email: [email protected]
Disruption of Axonal Transport From Olfactory Epithelium by 3-Methylindole
Physiology & Behavior, Vol. 65, No. 3, pp. 479–487, 1998 © 1998 Elsevier Science Inc. Printed in the USA. All rights reserved 0031-9384/98 $19.00 1 .00
PII S0031-9384(98)00185-1
Disruption of Axonal Transport From Olfactory Epithelium by 3-Methylindole ANGELA K. SETZER* AND BURTON SLOTNICK†1 *Pennsylvania State University, Hershey Medical Center, Hershey, PA 17033, and †American University, Washington, DC 20016 Received 29 December 1997; Accepted 25 June 1998 SETZER, A. K. AND B. SLOTNICK. Disruption of axonal transport from olfactory epithelium by 3-methylindole. PHYSIOL BEHAV 65(3) 479–487, 1998.—The effects of 150, 350, and 400 mg/kg intraperitoneal 3-methylindole (3-MI) on anterograde transport of horseradish peroxidase from the olfactory epithelium to the olfactory bulb were investigated. In 400 mg/kg 3-MI–treated rats sacrificed after 7 days only about 2% of all glomeruli had normal levels of the reaction product, and most glomeruli had no detectable reaction product. Lower doses of 3-MI produced correspondingly less disruption of axonal transport, with savings located primarily in the ventral to midlateral and the ventromedial region of the bulb. There was a gradual recovery of bulbar connections in 12-, 22-, and 92-day survival rats. In all cases, the increase in axonal transport was greatest in glomeruli on the lateral, ventral, and ventromedial areas of the bulb, and least evident or absent on the dorsal and dorsomedial areas. © 1998 Elsevier Science Inc. 3-Methylindole
Axonal transport
Olfactory bulb
controlled vivarium. To equate for the dietary restriction that would be imposed in future behavioral studies, the rats were maintained on 10 cc water/day except for 2 days prior and 3 days after treatment when supplementary water was given. Pelleted Purina lab chow was available ab lib.
RECENT studies have identified a number of toxins that produce degenerative changes in olfactory epithelial tissue (2). For example, intraperitoneal administration of 3-methylindole (3-MI) and dichlobenil produce long-lasting effects in specific regions of olfactory epithelium [e.g., (2,4,9,10,12)]. In general, these studies of olfactoxins have examined chemical and gross pathological changes induced by the toxins. However, there has been no systematic investigation of regional changes of the epithelium or connections of the epithelium to the olfactory bulb after administration of an olfactoxin. The present study provides a quantitative and regional analysis of axonal transport from the epithelium to the olfactory bulb after treatment with 3-MI. 3-MI was chosen as a potentially useful toxin for future behavioral studies of olfaction on the basis of pilot studies indicating that the effects of the toxin were both dose- and time-dependent, and from a study by Peele et al. (4), suggesting that a high dose of the agent (400 mg/kg) may produce anosmia.
3-Methylindole Administration 3-Methylindole (99% purity, Sigma) was dissolved in corn oil (Mazola) to a concentration of 15, 35, or 40 mg/mL, and used for administration of 150, 350, or 400 mg/kg, respectively. These solutions were made shortly before use. Controls received corn oil only. Doses were based on pilot studies and on the work of Peele et al. (4) and Turk, Flory and Henk (8–10). Injections were made intraperitoneally using a 20-ga needle. Because rats treated with 3-MI emit a strong stench for several days after treatment, immediately after being injected the rat was placed in a clean polycarbonate cage lined with paper towels and kept in a vented hood for 3 days. It was then housed on wood chips and returned to the vivarium.
METHODS
Subjects
Surgical and Histological Procedures
Fifty Long–Evans rats approximately 90 days old (300– 400 g) were housed in plastic cages in a temperature- and light-
Horseradish peroxidase conjugated with wheat germ agglutinin (WGA*HRP, lectin from Triticum vulgaris, Sigma,
1To whom requests for reprints should be addressed at Department of Psychology, American University, Washington, DC 20016. E-mail: slotnic @american.edu
479
480
SETZER AND SLOTNICK
St. Louis, MO), was diluted to 1% with an aqueous solution of sterile 1% Fast Green and stored at 2178C until shortly before use. Five, 10, 20, or 90 days after injection of 3-MI, the epithelium was treated with WGA*HRP using a modification of the method described by Shipley (6). Rats were anesthetized with 350 mg/kg of chloral hydrate (7% aqueous solution), clamped into a Kopf stereotaxic machine, and the dorsal surface of each olfactory sac was exposed and incised. After homeostasis was achieved, a 1 3 5-mm pledget of gelfoam was inserted into the posterior aspect of each sac and infiltrated with 3–4 mL of 1% WGA*HRP. The nasal bone flap was replaced and the scalp was closed with metal wound clips. The animal was placed on a heating pad until it recovered its righting reflex. HRP procedures followed those described by Mesulam (3). Rats were perfused 48 h after application of WGA*HRP and, using a freezing microtome, the olfactory bulbs were sectioned at 50 microns in the frontal plane. Every fourth section was saved, reacted with tetramethylbenzadine (TMB), and mounted on gelatin-coated glass slides. The slides were lightly counterstained with thionin, quickly dehydrated through cold 70, 95, and 100% ethanol, cleared in xylene, and covered using Permount. Because, as described in the Results section, treatment with 3-MI appeared to completely block transport of WGA*HRP to some glomeruli, it was important that we avoided false negatives. To ensure that we used optimal levels of TMB and hydrogen peroxide in the reaction process, the concentration of these agents were varied in processing a series of olfactory bulb sections in pilot studies. The high level
of anterograde transport resulting from application of the enzyme to the epithelium required somewhat lower levels of TMB and hydrogen peroxide than those established by Mesulam (3) be used to avoid excess artifact in the tissue. In processing tissue from experimental and control rats, extra sections were saved and processed using higher or lower levels of TMB and/ or hydrogen peroxide to ensure that an apparent absence of a reaction product in the tissue was not due to insensitivity of our procedure. Groups Experimental groups are identified by 3-MI dose (150, 350, or 400 mg/kg) and survival times (7, 12, 22, or 92 days). There were four or five rats at each dose level in each of the 7-, 12-, and 22-day survival groups, but only one 92-day survival rat at each dose. Eleven rats, treated with the corn oil vehicle, served as controls. These rats were treated with WGA*HRP 5 (n 5 4), 10 (n 5 4), 20 (n 5 3), or 90 (n 5 3) days after injection with the vehicle. To assess the effects of direct application of 3-MI to the olfactory bulb, one rat was anesthetized with 350 mg/kg of chloral hydrate and the dorsal surfaces of the olfactory bulbs were exposed. Twenty-five microliters of 400 mg/ ml 3-MI was applied topically to the dorsal surface of each olfactory bulb. Most of this volume ran off into the medial and lateral surfaces of the bulb but a thin layer remained on the dorsal surface. Finally, three rats served as controls for endogenous peroxidase levels. These rats were injected with the
FIG. 1. Diagrammatic representation of frontal sections of one olfactory bulb at each of the eight representative levels used for anatomical analyses. Each section is divided into quadrants by bisection of the dorsal–ventral and medial–lateral planes. The glomerular layer within each quadrant is divided into units, each of which represents a small cluster of glomeruli. In each section dorsal is toward the top and medial is to the right.
AXONAL TRANSPORT AND 3-METHYLINDOLE corn oil vehicle but were not treated with WGA*HRP. They were perfused, and sections through the olfactory bulbs were reacted as described above. Histological Analyses A standard set of drawings of olfactory bulb sections was generated using the Slotnick and Hersch (7) atlas of the rat olfactory system. Eight sections, beginning at approximately 0.4 mm from the rostral tip of the bulb and separated by 0.6–0.8 mm were used to illustrate levels through the olfactory bulb. Each of the eight diagrammatic representations of frontal sections through the bulb was divided into four quadrants (dorsolateral, dorsomedial, ventrolateral, and ventromedial) by bisecting the long and short axes of the section. For purposes of illustration, the glomerular layer of each drawing was further subdivided into small circular segments (Fig. 1). Two rats from each of the 7-, 12-, and 22-day survival groups and the one 92-day survival rat for each dose were used for detailed analysis. However, all sections of all rats were examined microscopically, and the results for cases selected for detailed analysis were representative of all animals in a group. Sections from one olfactory bulb in each case were inspected at 1003 and 2003 using bright-field and polarized light optics. Histological sections that corresponded to the frontal levels in Fig. 1 were selected, and the density of a reaction product within each identifiable glomerulus in these sections was graded on a six-point scale. Glomeruli that had no detectable reaction product were scored as 0. Glomeruli with
481 questionable reaction product (very light reaction product, perhaps not above background levels) were scored as 1. A score of 1.5 was used to indicate a very light scattering of reaction product that was just visible in polarized light but not visible in bright field. A score of 2 was used to indicate a light scattering of reaction product that was clearly detectable under polarized light and just detectable in bright field. A score of 3 was assigned to glomeruli that had either moderately dense reaction product that filled the glomerulus or dense reaction product that filled only part of the glomerulus. A score of 4 was used for glomeruli that were filled with dense reaction product and appeared similar to the reaction seen in control rats. The approximate location of these glomeruli was plotted by hand on the drawings, and the density of reaction product was noted. Although, in most cases, the pattern and density of reaction product was similar for the right and left olfactory bulb in each animal, the bulb with the clearest or most dense reaction product was selected for measurement. In general, even brief inspection of sections at low magnification revealed that differences among dose levels at any survival time were much greater than differences within groups. If the survival time was known, each of the authors was essentially 100% accurate in identifying the dose level for each of the experimental cases. The mean total number of glomeruli in the eight sections of the bulb selected for analysis for four control rats was 551 (range, 540–572). As essentially each of these glomeruli were filled with dense reaction product, the number of glomeruli with reaction product in an experimental subject was expressed
FIG. 2. Left graph, mean percent of control value for glomeruli filled with dense or moderate levels of reaction product (glomerular ratings of 3 and 4) at 7, 12, 22, and 92 days after treatment with different doses of 3-MI. The data points for the 7-, 12-, and 22-day groups are based on the mean of two rats. The 92-day points are from a single rat in each group. Right graph, percent of control value for glomeruli filled with dense or moderate levels of reaction product in different quadrants of the olfactory bulb (collapsed across frontal levels).
482
SETZER AND SLOTNICK bulbs in each of the 11 control rats were filled with dense reaction product (Figs. 7C, 7F, and 8A). There was also dense anterograde transport in the glomeruli of the accessory olfactory bulb in two control rats due, probably, to leakage of WGA*HRP into the vomeronasal organ. The rat that received a direct application of 3-MI to the olfactory bulbs had some damage to the dorsal surface of one bulb from the surgical procedures, but the histological results were essentially identical to that of the other control rats. The three control rats not treated with WGA*HRP had no reaction product in their olfactory bulb glomeruli or, in one case, a uniform scattering of light reaction product. Seven-Day Survival Cases
FIG. 3. Diagrammatic representation of olfactory bulb sections showing glomerular clusters with detectable reaction product from 7-day survival rats treated with 150, 350, or 400 mg/kg 3-MI. The numbers at the top of the figure represent mm from the most rostral aspect of the accessory olfactory bulb.
as a percent of control by dividing the count in the experimental rat by 551 and multiplying by 100. RESULTS
Except for two cases in which there was poor transport to one olfactory bulb (probably due to inadequate application of WGA*HRP), virtually all glomeruli of the main olfactory
All doses of 3-MI produced disruption in axonal transport to olfactory bulb glomeruli but the extent of this disruption varied with dose and survival time (Fig. 2). In low-dose 7-day survival rats, only 20% of all bulbar glomeruli had relatively dense reaction product (scores of 3 or 4, Figures 2, 3, and 7). In the rostral 1/3 of the bulb virtually all glomeruli with moderate or dense reaction product were found in the midlateral and ventromedial regions. Rostral to the accessory olfactory bulb, glomeruli in the dorsomedial and ventrolateral quadrants had only light or questionable levels of reaction product. At more posterior levels many glomeruli on the medial surface had dense anterograde reaction product, and approximately 40% of glomeruli in the posterior 1/3 of the bulb had a moderate to dense levels of reaction product (Fig. 3).
FIG. 4. Diagrammatic representation of olfactory bulb sections showing glomerular clusters with detectable reaction product from 12-day survival rats treated with 150, 350, or 400 mg/kg 3-MI. The numbers at the top of the figure represent mm from the most rostral aspect of the accessory olfactory bulb.
AXONAL TRANSPORT AND 3-METHYLINDOLE The 350-mg/kg dose produced more severe disruption of anterograde transport. In the cases used for analysis, only 10% of all glomeruli had moderate or dense reaction product 7 days after treatment, and almost all of these were on the midlateral wall and in the posterior 2/3 of the bulb (Fig. 3). Few glomeruli had dense reaction product in the rostral 1/3 of the bulb, and no reaction product could be detected in many of these glomeruli. The 400-mg/kg 3-MI dose produced disruption of anterograde transport to virtually all glomeruli in the bulb. Indeed, only a few scattered glomeruli (less than 2%) had even moderate reaction product, and virtually all of these were located in the posterior ventromedial area (Figs. 3–5). Other glomeruli in this same area had very light reaction product but glomeruli in other areas of the bulb were remarkably clear (Fig. 8D and E). Changes in Longer Survival Cases There was an increase in the number of glomeruli with reaction product in 12 (Figs. 2 and 4) and 22 day (Figs. 2 and 5) survival rats at each dose, but this was most marked in the low-dose cases. The percentage of glomeruli with reaction product increased by a factor of 2 in the 150-mg/kg 12-day survival rats, but there was little further increase in glomeruli with dense reaction product in the 22-day survival animals (Figs. 4 and 5). Almost all glomeruli in the 22-day survival low-dose rats had some reaction product, but fewer than 50% had moderate to dense reaction product (Fig. 2). The glomeruli
483 with moderate and dense reaction product were primarily on the lateral, ventral, and ventromedial surfaces, and reaction product in these glomerular zones extended more rostrally than in the 7-day cases . For both the 12- and 22-day low-dose cases, glomeruli with dense reaction product extended more dorsally on the lateral surface than in the 7-day survival rats. However, reaction product was much lighter in glomeruli on the midmedial wall of the bulb (Figs. 4 and 5). Almost all glomeruli on the dorsal and dorsomedial surfaces had light levels of reaction product, and a few had dense reaction product. One 150-mg/kg 3-MI rat was allowed a 92-day survival. This rat had reaction product in virtually all glomeruli (Fig. 6). Those in the dorsal and dorsomedial areas had a relatively light reaction product, but most other glomeruli had dense reaction product, similar to that in controls. A similar but more restricted increase in glomerular input was observed in the 350 mg/kg-dose rats. The number of glomeruli with dense and moderate reaction product increased from approximately 10% in the 7-day survival cases to 17% in the 12-day cases and to 40% in the 22-day cases (Figs. 4 and 5). Interestingly, the density of reaction product and distribution of glomeruli with reaction product in the 22-day survival 350-mg/kg rats was quite similar to that for the 12-day survival 150-mg/kg cases (Figs. 4 and 5). Two rats in the 22-day survival 350-mg/kg group also had dense or moderate levels of reaction product in almost all glomeruli in the accessory olfactory bulb (Fig. 8F). The density and extent of this reaction product was similar to that in the two control rats that had reaction product in the AOB. We assume that, in each of these
FIG. 5. Diagrammatic representation of olfactory bulb sections showing glomerular clusters with detectable reaction product from 22-day survival rats treated with 150, 350, or 400 mg/kg 3-MI. The numbers at the top of the figure represent mm from the most rostral aspect of the accessory olfactory bulb.
484
SETZER AND SLOTNICK
cases, WGA*HRP, applied to the olfactory epithelium, had leaked into the vomeronasal organ. In the one 350-mg/kg rat allowed a 92-day survival period (Fig. 6), virtually all glomeruli except those on the dorsal and dorsomedial surface of the olfactory bulb were filled with dense reaction product. Only a few glomeruli on the dorsal surface but none on the dorsomedial surface had dense reaction product. Increase in reaction product was considerably less in the longer survival 400 mg/kg-dose animals. There was only a marginal increase in 12-day survival rats, characterized by a few glomeruli with dense reaction product in the posterior ventromedial area and very light or questionable levels of reaction product on the lateral surface of the bulb (Fig. 4). There was a more marked increase in the 22-day survival rats but, still, fewer than 15% of the glomeruli had moderate to dense reaction product (Figs. 2 and 5). Most of these were located on the lateral surface in the rostral 2/3 of the bulb, but there was also a more modest increase in the ventromedial region. Glomeruli in most other areas had very light or only questionable reaction product and, even in the 22-day survival cases, many glomeruli on the dorsal surface had little or no reaction product (Fig. 5). The one 400-mg/kg 3-MI 92-day survival rat had many more glomeruli with dense reaction product but, still, glomeruli on the dorsal and dorsomedial areas were empty. Glomeruli in the ventromedial area, particularly in posterior sections, had only light reaction product (Fig. 6).
In general, the glomeruli that had an increase in reaction product in longer survival rats were in the same region of the bulb as those having the most dense reaction product in the shorter survival cases (Fig. 7A–F). For example, in 350-mg/kg 12-day survival rats, only about half of the glomeruli in the midlateral region were densely filled with reaction product. Intermingled with these glomeruli were ones that had little or no reaction product (Fig. 8B). However, in 22-day survival rats, virtually all glomeruli in this region were filled with dense reaction product (Fig. 8C). In most cases there was a sharp rather than a gradual transition between regions that had dense and moderate levels of reaction product and those that had no reaction product. This was especially evident in midlateral and midmedial areas where clusters of glomeruli with dense reaction product were juxtaposed to more dorsal glomeruli with no reaction product (e.g., Fig. 8B).
DISCUSSION
The present results demonstrate that intraperitoneal injection of 3-MI produces a selective and dose- and time-dependent disruption of axonal transport from the olfactory epithelium to the olfactory bulb. The highest dose (400 mg/kg) produced complete disruption of anterograde transport to almost all glomeruli. For all doses at all survival times the most marked reduction in HRP transport was observed in the rostral half of the bulb, with the dorsomedial sector being most
FIG. 6. Diagrammatic representation of olfactory bulb sections showing glomerular clusters with detectable reaction product from 92-day survival rats treated with 150, 350, or 400 mg/kg 3-MI. The numbers at the top of the figure represent mm from the most rostral aspect of the accessory olfactory bulb.
AXONAL TRANSPORT AND 3-METHYLINDOLE severely affected. Glomeruli in this sector had little or no reaction product in any of the 7-day survival animals. A gradual and dose-dependent increase in reaction product to most other bulbar glomeruli was observed 3 weeks after treatment. But, even after 3 weeks, rats given the lowest dose of 3-MI still had reaction product scores that were less than 50% of control values. Although the increase in number of glomeruli with reaction product was more complete in the low-dose 92-day survival rats, the reaction product in many glomerular areas was less dense than that in controls. It is unlikely that the complete absence of reaction product observed in many glomeruli in experimental rats was a result of insensitivity of our histological procedures. This is because the method, as used, has been optimized for the detection of an HRP reaction product (3). Indeed, in sections with a light reaction product individual axons or small clusters of axons could be observed in the nerve plexus layer and within glomeruli. In trials with one-tenth of the usual amount of the hydrogen peroxide catalyst, clear reaction product was still observed to fill the glomeruli of control rats. Further, the “empty” glomeruli in treated rats was identical in appearance to glomeruli of control rats not treated with WGA*HRP. The
485 absence of a reaction product in experimental rats probably indicates that these glomeruli did not have a functional connection to the olfactory epithelium. HRP is a small readily transported molecule, and agents that disrupt its axonal transport probably interfere generally with the axoplasmic transport mechanism of the neuron and, hence, the ability of the neuron to transmit information. The recovery observed at all doses of the toxin was characterized by both an increase in density of reaction product and the number of glomeruli with reaction product. This recovery of input could be based on reversible changes in affected areas of the epithelium, growth of neurons that were immature at the time of treatment or, possibly, sprouting from mature neurons that were spared by the toxin. The epithelium was not examined in this study but histopathological reports indicate that the toxin produces widespread degeneration and, in affected areas, the olfactory mucosa was replaced by columnar epithelium, fibrosis, and bony tissue (10). In general, increased bulbar input in 22- and 92-day survival cases occurred in glomerular areas that were within or contiguous with those that had sparing in shorter survival animals. To the extent that a topographically ordered relation-
FIG. 7. Photomicrographs of olfactory bulb sections. (A and D) Anterior and midfrontal levels from a 7-day survival, 150-mg/kg 3-MI rat. (B and E) Anterior and midfrontal levels from a 22-day survival, 150-mg/kg 3-MI rat. (C and F) Anterior and midfrontal levels from a control rat. In each section dorsal is toward the top and medial is to the left.
486
SETZER AND SLOTNICK
FIG. 8. Photomicrographs of glomerular regions in experimental and control rats. (A) Nerve layer and glomeruli on the medial surface of the olfactory bulb in a control rat. Note that the axons and glomeruli are filled with dense reaction product. (B) Similar region in a 12-day survival 350-mg/kg rat. Note that only about one-half of the glomeruli in the lower part of the section have dense or moderate levels of reaction product, and that the more dorsal glomeruli contain no reaction product. (C) Similar region from a 22-day survival 350-mg/kg rat. (D) Similar region in a 7-day survival 400-mg/kg rat. None of the glomeruli in this field contained detectable reaction product. (E) Ventromedial sector of the olfactory bulb in a 7-day survival 400-mg/kg rat. Note that a single glomerulus in this field contains dense reaction product. Surrounding glomeruli have no detectable reaction product. (F) Section through the accessory olfactory bulb in a 350-mg/kg 22-day survival rat. In sections A–E medial is to the left and dorsal is toward the top. In section E medial is towards the bottom and dorsal is to the left.
ship exists in the projections of the epithelium to the bulb, this pattern points to recovery of sensory epithelium at the margin of unaffected areas. However, in situ hybridization studies indicate that individual glomeruli receive input from convergence of sensory neurons that are widely scattered within a zonal area of the bulb (5,11). This chemotopic mapping scheme makes it unlikely that near neighboring glomeruli receive their input from contiguous areas of the epithelium. Hence, it seems unlikely that the pattern of recovery observed represents recovery at the margin of epithelial areas destroyed by the toxin. An alternative mechanism for the observed pattern of recovery is that of axonal sprouting from spared neurons to vacated synaptic sites in nearby glomeruli. This could account for the observation that the most marked recovery in longer survival rats occurred within or at the margins of areas that had some input in shorter survival animals.
The mechanism(s) by which the toxin produces damage to the epithelium is not known but the nasal lesions in mice resemble those produced by 3-methyl-furan, an air pollutant that is toxic to Clara cells and activated by mixed function oxidases (10). Bray and Kubow (1) suggested that, in the lung, 3-MI is metabolized by mixed-function oxidases to a reactive intermediate, which in itself acts as a free radical or induces the formation of free radicals, and these, in turn, are responsible for the tissue pathology. The basis for the differential effects of 3-MI within the olfactory epithelium, however, has not been determined. The present demonstration that 3-MI produces reliable, selective, and time-dependent effects within the epithelium suggests that the toxin may provide a useful tool for examining anatomical and behavioral correlates of the disruption and recovery of input to the olfactory bulb.
AXONAL TRANSPORT AND 3-METHYLINDOLE
487
REFERENCES 1. Bray, T. M.; Kubow, S.: Involvement of free radicals in the mechanism of 3-methylindole-induced pulmonary toxicity: An example of metabolic activation in chemically induced lung disease. Environ. Health Perspect. 64:61–67; 1985. 2. Gaskell, B. A.: Nonneoplastic changes in the olfactory epithelium: Experimental studies. Environ. Health Perspect. 85:275– 289; 1990. 3. Mesulam, M.: Tracing neural connections with horseradish peroxidase. New York: Wiley; 1982. 4. Peele, D. B.; Allison, S. D.; Bolon, B.; Prah, D.; Jensen, K. F.; Morgan, K. T.: Functional deficits produced by 3-methylindoleinduced olfactory mucosal damage revealed by a simple olfactory learning task. Toxicol. Appl. Pharmacol. 107:191–202; 1991. 5. Ressler, K. J.; Sullivan, S. L.; Buck, L. B.: Information coding in the olfactory system: Evidence for a stereotyped and highly organized epitope map in the olfactory bulb. Cell 79:1245–1255; 1994. 6. Shipley, M. T.: Transport of molecules from nose to brain: Transneuronal anterograde and retrograde labeling in the rat olfactory
7. 8. 9. 10. 11. 12.
system by wheat germ agglutinin-horseradish peroxidase applied to the nasal epithelium. Brain Res. Bull. 15:129–142; 1985. Slotnick, B. M.; Hersch, S.: A stereotaxic atlas of the rat olfactory system. Brain Res. Bull. 5(Suppl. 5):1–55; 1980. Turk, M. A. M.; Flory, W.; Henk, W. G.: Dose response in 3-methylindole-induced bronchiolar epithelial necrosis in mice. Res. Commun. Chem. Pathol. Pharmacol. 46:351–403; 1984. Turk, M. A. M.; Flory, W.; Henk, W. G.: Chemical modulation of 3-methlyindole toxicosis in mice: Effect on bronchiolar and olfactory mucosal injury. Vet. Pathol. 23:563–570; 1986. Turk, M. A. M.; Henk, W. G.; Flory, W.: 3-Methylindole-induced nasal mucosal damage in mice. Vet. Pathol. 24:400–403; 1987. Vassar, R.; Chao, S. K.; Sitcheran, R.; Nunez, J. M.; Vosshall, L. B.; Axel, R.: Topographic organization of sensory projections to the olfactory bulb. Cell 79:981–991; 1994. Walters, E.; Buchheit, K.; Maruniak, J. A.: Olfactory cytochrome P-450 immunoreactivity in mice is altered by dichlobinil but preserved by metyrapone. Toxicology 81:113–122; 1993.
PII S0031-9384(98)00185-1
Disruption of Axonal Transport From Olfactory Epithelium by 3-Methylindole ANGELA K. SETZER* AND BURTON SLOTNICK†1 *Pennsylvania State University, Hershey Medical Center, Hershey, PA 17033, and †American University, Washington, DC 20016 Received 29 December 1997; Accepted 25 June 1998 SETZER, A. K. AND B. SLOTNICK. Disruption of axonal transport from olfactory epithelium by 3-methylindole. PHYSIOL BEHAV 65(3) 479–487, 1998.—The effects of 150, 350, and 400 mg/kg intraperitoneal 3-methylindole (3-MI) on anterograde transport of horseradish peroxidase from the olfactory epithelium to the olfactory bulb were investigated. In 400 mg/kg 3-MI–treated rats sacrificed after 7 days only about 2% of all glomeruli had normal levels of the reaction product, and most glomeruli had no detectable reaction product. Lower doses of 3-MI produced correspondingly less disruption of axonal transport, with savings located primarily in the ventral to midlateral and the ventromedial region of the bulb. There was a gradual recovery of bulbar connections in 12-, 22-, and 92-day survival rats. In all cases, the increase in axonal transport was greatest in glomeruli on the lateral, ventral, and ventromedial areas of the bulb, and least evident or absent on the dorsal and dorsomedial areas. © 1998 Elsevier Science Inc. 3-Methylindole
Axonal transport
Olfactory bulb
controlled vivarium. To equate for the dietary restriction that would be imposed in future behavioral studies, the rats were maintained on 10 cc water/day except for 2 days prior and 3 days after treatment when supplementary water was given. Pelleted Purina lab chow was available ab lib.
RECENT studies have identified a number of toxins that produce degenerative changes in olfactory epithelial tissue (2). For example, intraperitoneal administration of 3-methylindole (3-MI) and dichlobenil produce long-lasting effects in specific regions of olfactory epithelium [e.g., (2,4,9,10,12)]. In general, these studies of olfactoxins have examined chemical and gross pathological changes induced by the toxins. However, there has been no systematic investigation of regional changes of the epithelium or connections of the epithelium to the olfactory bulb after administration of an olfactoxin. The present study provides a quantitative and regional analysis of axonal transport from the epithelium to the olfactory bulb after treatment with 3-MI. 3-MI was chosen as a potentially useful toxin for future behavioral studies of olfaction on the basis of pilot studies indicating that the effects of the toxin were both dose- and time-dependent, and from a study by Peele et al. (4), suggesting that a high dose of the agent (400 mg/kg) may produce anosmia.
3-Methylindole Administration 3-Methylindole (99% purity, Sigma) was dissolved in corn oil (Mazola) to a concentration of 15, 35, or 40 mg/mL, and used for administration of 150, 350, or 400 mg/kg, respectively. These solutions were made shortly before use. Controls received corn oil only. Doses were based on pilot studies and on the work of Peele et al. (4) and Turk, Flory and Henk (8–10). Injections were made intraperitoneally using a 20-ga needle. Because rats treated with 3-MI emit a strong stench for several days after treatment, immediately after being injected the rat was placed in a clean polycarbonate cage lined with paper towels and kept in a vented hood for 3 days. It was then housed on wood chips and returned to the vivarium.
METHODS
Subjects
Surgical and Histological Procedures
Fifty Long–Evans rats approximately 90 days old (300– 400 g) were housed in plastic cages in a temperature- and light-
Horseradish peroxidase conjugated with wheat germ agglutinin (WGA*HRP, lectin from Triticum vulgaris, Sigma,
1To whom requests for reprints should be addressed at Department of Psychology, American University, Washington, DC 20016. E-mail: slotnic @american.edu
479
480
SETZER AND SLOTNICK
St. Louis, MO), was diluted to 1% with an aqueous solution of sterile 1% Fast Green and stored at 2178C until shortly before use. Five, 10, 20, or 90 days after injection of 3-MI, the epithelium was treated with WGA*HRP using a modification of the method described by Shipley (6). Rats were anesthetized with 350 mg/kg of chloral hydrate (7% aqueous solution), clamped into a Kopf stereotaxic machine, and the dorsal surface of each olfactory sac was exposed and incised. After homeostasis was achieved, a 1 3 5-mm pledget of gelfoam was inserted into the posterior aspect of each sac and infiltrated with 3–4 mL of 1% WGA*HRP. The nasal bone flap was replaced and the scalp was closed with metal wound clips. The animal was placed on a heating pad until it recovered its righting reflex. HRP procedures followed those described by Mesulam (3). Rats were perfused 48 h after application of WGA*HRP and, using a freezing microtome, the olfactory bulbs were sectioned at 50 microns in the frontal plane. Every fourth section was saved, reacted with tetramethylbenzadine (TMB), and mounted on gelatin-coated glass slides. The slides were lightly counterstained with thionin, quickly dehydrated through cold 70, 95, and 100% ethanol, cleared in xylene, and covered using Permount. Because, as described in the Results section, treatment with 3-MI appeared to completely block transport of WGA*HRP to some glomeruli, it was important that we avoided false negatives. To ensure that we used optimal levels of TMB and hydrogen peroxide in the reaction process, the concentration of these agents were varied in processing a series of olfactory bulb sections in pilot studies. The high level
of anterograde transport resulting from application of the enzyme to the epithelium required somewhat lower levels of TMB and hydrogen peroxide than those established by Mesulam (3) be used to avoid excess artifact in the tissue. In processing tissue from experimental and control rats, extra sections were saved and processed using higher or lower levels of TMB and/ or hydrogen peroxide to ensure that an apparent absence of a reaction product in the tissue was not due to insensitivity of our procedure. Groups Experimental groups are identified by 3-MI dose (150, 350, or 400 mg/kg) and survival times (7, 12, 22, or 92 days). There were four or five rats at each dose level in each of the 7-, 12-, and 22-day survival groups, but only one 92-day survival rat at each dose. Eleven rats, treated with the corn oil vehicle, served as controls. These rats were treated with WGA*HRP 5 (n 5 4), 10 (n 5 4), 20 (n 5 3), or 90 (n 5 3) days after injection with the vehicle. To assess the effects of direct application of 3-MI to the olfactory bulb, one rat was anesthetized with 350 mg/kg of chloral hydrate and the dorsal surfaces of the olfactory bulbs were exposed. Twenty-five microliters of 400 mg/ ml 3-MI was applied topically to the dorsal surface of each olfactory bulb. Most of this volume ran off into the medial and lateral surfaces of the bulb but a thin layer remained on the dorsal surface. Finally, three rats served as controls for endogenous peroxidase levels. These rats were injected with the
FIG. 1. Diagrammatic representation of frontal sections of one olfactory bulb at each of the eight representative levels used for anatomical analyses. Each section is divided into quadrants by bisection of the dorsal–ventral and medial–lateral planes. The glomerular layer within each quadrant is divided into units, each of which represents a small cluster of glomeruli. In each section dorsal is toward the top and medial is to the right.
AXONAL TRANSPORT AND 3-METHYLINDOLE corn oil vehicle but were not treated with WGA*HRP. They were perfused, and sections through the olfactory bulbs were reacted as described above. Histological Analyses A standard set of drawings of olfactory bulb sections was generated using the Slotnick and Hersch (7) atlas of the rat olfactory system. Eight sections, beginning at approximately 0.4 mm from the rostral tip of the bulb and separated by 0.6–0.8 mm were used to illustrate levels through the olfactory bulb. Each of the eight diagrammatic representations of frontal sections through the bulb was divided into four quadrants (dorsolateral, dorsomedial, ventrolateral, and ventromedial) by bisecting the long and short axes of the section. For purposes of illustration, the glomerular layer of each drawing was further subdivided into small circular segments (Fig. 1). Two rats from each of the 7-, 12-, and 22-day survival groups and the one 92-day survival rat for each dose were used for detailed analysis. However, all sections of all rats were examined microscopically, and the results for cases selected for detailed analysis were representative of all animals in a group. Sections from one olfactory bulb in each case were inspected at 1003 and 2003 using bright-field and polarized light optics. Histological sections that corresponded to the frontal levels in Fig. 1 were selected, and the density of a reaction product within each identifiable glomerulus in these sections was graded on a six-point scale. Glomeruli that had no detectable reaction product were scored as 0. Glomeruli with
481 questionable reaction product (very light reaction product, perhaps not above background levels) were scored as 1. A score of 1.5 was used to indicate a very light scattering of reaction product that was just visible in polarized light but not visible in bright field. A score of 2 was used to indicate a light scattering of reaction product that was clearly detectable under polarized light and just detectable in bright field. A score of 3 was assigned to glomeruli that had either moderately dense reaction product that filled the glomerulus or dense reaction product that filled only part of the glomerulus. A score of 4 was used for glomeruli that were filled with dense reaction product and appeared similar to the reaction seen in control rats. The approximate location of these glomeruli was plotted by hand on the drawings, and the density of reaction product was noted. Although, in most cases, the pattern and density of reaction product was similar for the right and left olfactory bulb in each animal, the bulb with the clearest or most dense reaction product was selected for measurement. In general, even brief inspection of sections at low magnification revealed that differences among dose levels at any survival time were much greater than differences within groups. If the survival time was known, each of the authors was essentially 100% accurate in identifying the dose level for each of the experimental cases. The mean total number of glomeruli in the eight sections of the bulb selected for analysis for four control rats was 551 (range, 540–572). As essentially each of these glomeruli were filled with dense reaction product, the number of glomeruli with reaction product in an experimental subject was expressed
FIG. 2. Left graph, mean percent of control value for glomeruli filled with dense or moderate levels of reaction product (glomerular ratings of 3 and 4) at 7, 12, 22, and 92 days after treatment with different doses of 3-MI. The data points for the 7-, 12-, and 22-day groups are based on the mean of two rats. The 92-day points are from a single rat in each group. Right graph, percent of control value for glomeruli filled with dense or moderate levels of reaction product in different quadrants of the olfactory bulb (collapsed across frontal levels).
482
SETZER AND SLOTNICK bulbs in each of the 11 control rats were filled with dense reaction product (Figs. 7C, 7F, and 8A). There was also dense anterograde transport in the glomeruli of the accessory olfactory bulb in two control rats due, probably, to leakage of WGA*HRP into the vomeronasal organ. The rat that received a direct application of 3-MI to the olfactory bulbs had some damage to the dorsal surface of one bulb from the surgical procedures, but the histological results were essentially identical to that of the other control rats. The three control rats not treated with WGA*HRP had no reaction product in their olfactory bulb glomeruli or, in one case, a uniform scattering of light reaction product. Seven-Day Survival Cases
FIG. 3. Diagrammatic representation of olfactory bulb sections showing glomerular clusters with detectable reaction product from 7-day survival rats treated with 150, 350, or 400 mg/kg 3-MI. The numbers at the top of the figure represent mm from the most rostral aspect of the accessory olfactory bulb.
as a percent of control by dividing the count in the experimental rat by 551 and multiplying by 100. RESULTS
Except for two cases in which there was poor transport to one olfactory bulb (probably due to inadequate application of WGA*HRP), virtually all glomeruli of the main olfactory
All doses of 3-MI produced disruption in axonal transport to olfactory bulb glomeruli but the extent of this disruption varied with dose and survival time (Fig. 2). In low-dose 7-day survival rats, only 20% of all bulbar glomeruli had relatively dense reaction product (scores of 3 or 4, Figures 2, 3, and 7). In the rostral 1/3 of the bulb virtually all glomeruli with moderate or dense reaction product were found in the midlateral and ventromedial regions. Rostral to the accessory olfactory bulb, glomeruli in the dorsomedial and ventrolateral quadrants had only light or questionable levels of reaction product. At more posterior levels many glomeruli on the medial surface had dense anterograde reaction product, and approximately 40% of glomeruli in the posterior 1/3 of the bulb had a moderate to dense levels of reaction product (Fig. 3).
FIG. 4. Diagrammatic representation of olfactory bulb sections showing glomerular clusters with detectable reaction product from 12-day survival rats treated with 150, 350, or 400 mg/kg 3-MI. The numbers at the top of the figure represent mm from the most rostral aspect of the accessory olfactory bulb.
AXONAL TRANSPORT AND 3-METHYLINDOLE The 350-mg/kg dose produced more severe disruption of anterograde transport. In the cases used for analysis, only 10% of all glomeruli had moderate or dense reaction product 7 days after treatment, and almost all of these were on the midlateral wall and in the posterior 2/3 of the bulb (Fig. 3). Few glomeruli had dense reaction product in the rostral 1/3 of the bulb, and no reaction product could be detected in many of these glomeruli. The 400-mg/kg 3-MI dose produced disruption of anterograde transport to virtually all glomeruli in the bulb. Indeed, only a few scattered glomeruli (less than 2%) had even moderate reaction product, and virtually all of these were located in the posterior ventromedial area (Figs. 3–5). Other glomeruli in this same area had very light reaction product but glomeruli in other areas of the bulb were remarkably clear (Fig. 8D and E). Changes in Longer Survival Cases There was an increase in the number of glomeruli with reaction product in 12 (Figs. 2 and 4) and 22 day (Figs. 2 and 5) survival rats at each dose, but this was most marked in the low-dose cases. The percentage of glomeruli with reaction product increased by a factor of 2 in the 150-mg/kg 12-day survival rats, but there was little further increase in glomeruli with dense reaction product in the 22-day survival animals (Figs. 4 and 5). Almost all glomeruli in the 22-day survival low-dose rats had some reaction product, but fewer than 50% had moderate to dense reaction product (Fig. 2). The glomeruli
483 with moderate and dense reaction product were primarily on the lateral, ventral, and ventromedial surfaces, and reaction product in these glomerular zones extended more rostrally than in the 7-day cases . For both the 12- and 22-day low-dose cases, glomeruli with dense reaction product extended more dorsally on the lateral surface than in the 7-day survival rats. However, reaction product was much lighter in glomeruli on the midmedial wall of the bulb (Figs. 4 and 5). Almost all glomeruli on the dorsal and dorsomedial surfaces had light levels of reaction product, and a few had dense reaction product. One 150-mg/kg 3-MI rat was allowed a 92-day survival. This rat had reaction product in virtually all glomeruli (Fig. 6). Those in the dorsal and dorsomedial areas had a relatively light reaction product, but most other glomeruli had dense reaction product, similar to that in controls. A similar but more restricted increase in glomerular input was observed in the 350 mg/kg-dose rats. The number of glomeruli with dense and moderate reaction product increased from approximately 10% in the 7-day survival cases to 17% in the 12-day cases and to 40% in the 22-day cases (Figs. 4 and 5). Interestingly, the density of reaction product and distribution of glomeruli with reaction product in the 22-day survival 350-mg/kg rats was quite similar to that for the 12-day survival 150-mg/kg cases (Figs. 4 and 5). Two rats in the 22-day survival 350-mg/kg group also had dense or moderate levels of reaction product in almost all glomeruli in the accessory olfactory bulb (Fig. 8F). The density and extent of this reaction product was similar to that in the two control rats that had reaction product in the AOB. We assume that, in each of these
FIG. 5. Diagrammatic representation of olfactory bulb sections showing glomerular clusters with detectable reaction product from 22-day survival rats treated with 150, 350, or 400 mg/kg 3-MI. The numbers at the top of the figure represent mm from the most rostral aspect of the accessory olfactory bulb.
484
SETZER AND SLOTNICK
cases, WGA*HRP, applied to the olfactory epithelium, had leaked into the vomeronasal organ. In the one 350-mg/kg rat allowed a 92-day survival period (Fig. 6), virtually all glomeruli except those on the dorsal and dorsomedial surface of the olfactory bulb were filled with dense reaction product. Only a few glomeruli on the dorsal surface but none on the dorsomedial surface had dense reaction product. Increase in reaction product was considerably less in the longer survival 400 mg/kg-dose animals. There was only a marginal increase in 12-day survival rats, characterized by a few glomeruli with dense reaction product in the posterior ventromedial area and very light or questionable levels of reaction product on the lateral surface of the bulb (Fig. 4). There was a more marked increase in the 22-day survival rats but, still, fewer than 15% of the glomeruli had moderate to dense reaction product (Figs. 2 and 5). Most of these were located on the lateral surface in the rostral 2/3 of the bulb, but there was also a more modest increase in the ventromedial region. Glomeruli in most other areas had very light or only questionable reaction product and, even in the 22-day survival cases, many glomeruli on the dorsal surface had little or no reaction product (Fig. 5). The one 400-mg/kg 3-MI 92-day survival rat had many more glomeruli with dense reaction product but, still, glomeruli on the dorsal and dorsomedial areas were empty. Glomeruli in the ventromedial area, particularly in posterior sections, had only light reaction product (Fig. 6).
In general, the glomeruli that had an increase in reaction product in longer survival rats were in the same region of the bulb as those having the most dense reaction product in the shorter survival cases (Fig. 7A–F). For example, in 350-mg/kg 12-day survival rats, only about half of the glomeruli in the midlateral region were densely filled with reaction product. Intermingled with these glomeruli were ones that had little or no reaction product (Fig. 8B). However, in 22-day survival rats, virtually all glomeruli in this region were filled with dense reaction product (Fig. 8C). In most cases there was a sharp rather than a gradual transition between regions that had dense and moderate levels of reaction product and those that had no reaction product. This was especially evident in midlateral and midmedial areas where clusters of glomeruli with dense reaction product were juxtaposed to more dorsal glomeruli with no reaction product (e.g., Fig. 8B).
DISCUSSION
The present results demonstrate that intraperitoneal injection of 3-MI produces a selective and dose- and time-dependent disruption of axonal transport from the olfactory epithelium to the olfactory bulb. The highest dose (400 mg/kg) produced complete disruption of anterograde transport to almost all glomeruli. For all doses at all survival times the most marked reduction in HRP transport was observed in the rostral half of the bulb, with the dorsomedial sector being most
FIG. 6. Diagrammatic representation of olfactory bulb sections showing glomerular clusters with detectable reaction product from 92-day survival rats treated with 150, 350, or 400 mg/kg 3-MI. The numbers at the top of the figure represent mm from the most rostral aspect of the accessory olfactory bulb.
AXONAL TRANSPORT AND 3-METHYLINDOLE severely affected. Glomeruli in this sector had little or no reaction product in any of the 7-day survival animals. A gradual and dose-dependent increase in reaction product to most other bulbar glomeruli was observed 3 weeks after treatment. But, even after 3 weeks, rats given the lowest dose of 3-MI still had reaction product scores that were less than 50% of control values. Although the increase in number of glomeruli with reaction product was more complete in the low-dose 92-day survival rats, the reaction product in many glomerular areas was less dense than that in controls. It is unlikely that the complete absence of reaction product observed in many glomeruli in experimental rats was a result of insensitivity of our histological procedures. This is because the method, as used, has been optimized for the detection of an HRP reaction product (3). Indeed, in sections with a light reaction product individual axons or small clusters of axons could be observed in the nerve plexus layer and within glomeruli. In trials with one-tenth of the usual amount of the hydrogen peroxide catalyst, clear reaction product was still observed to fill the glomeruli of control rats. Further, the “empty” glomeruli in treated rats was identical in appearance to glomeruli of control rats not treated with WGA*HRP. The
485 absence of a reaction product in experimental rats probably indicates that these glomeruli did not have a functional connection to the olfactory epithelium. HRP is a small readily transported molecule, and agents that disrupt its axonal transport probably interfere generally with the axoplasmic transport mechanism of the neuron and, hence, the ability of the neuron to transmit information. The recovery observed at all doses of the toxin was characterized by both an increase in density of reaction product and the number of glomeruli with reaction product. This recovery of input could be based on reversible changes in affected areas of the epithelium, growth of neurons that were immature at the time of treatment or, possibly, sprouting from mature neurons that were spared by the toxin. The epithelium was not examined in this study but histopathological reports indicate that the toxin produces widespread degeneration and, in affected areas, the olfactory mucosa was replaced by columnar epithelium, fibrosis, and bony tissue (10). In general, increased bulbar input in 22- and 92-day survival cases occurred in glomerular areas that were within or contiguous with those that had sparing in shorter survival animals. To the extent that a topographically ordered relation-
FIG. 7. Photomicrographs of olfactory bulb sections. (A and D) Anterior and midfrontal levels from a 7-day survival, 150-mg/kg 3-MI rat. (B and E) Anterior and midfrontal levels from a 22-day survival, 150-mg/kg 3-MI rat. (C and F) Anterior and midfrontal levels from a control rat. In each section dorsal is toward the top and medial is to the left.
486
SETZER AND SLOTNICK
FIG. 8. Photomicrographs of glomerular regions in experimental and control rats. (A) Nerve layer and glomeruli on the medial surface of the olfactory bulb in a control rat. Note that the axons and glomeruli are filled with dense reaction product. (B) Similar region in a 12-day survival 350-mg/kg rat. Note that only about one-half of the glomeruli in the lower part of the section have dense or moderate levels of reaction product, and that the more dorsal glomeruli contain no reaction product. (C) Similar region from a 22-day survival 350-mg/kg rat. (D) Similar region in a 7-day survival 400-mg/kg rat. None of the glomeruli in this field contained detectable reaction product. (E) Ventromedial sector of the olfactory bulb in a 7-day survival 400-mg/kg rat. Note that a single glomerulus in this field contains dense reaction product. Surrounding glomeruli have no detectable reaction product. (F) Section through the accessory olfactory bulb in a 350-mg/kg 22-day survival rat. In sections A–E medial is to the left and dorsal is toward the top. In section E medial is towards the bottom and dorsal is to the left.
ship exists in the projections of the epithelium to the bulb, this pattern points to recovery of sensory epithelium at the margin of unaffected areas. However, in situ hybridization studies indicate that individual glomeruli receive input from convergence of sensory neurons that are widely scattered within a zonal area of the bulb (5,11). This chemotopic mapping scheme makes it unlikely that near neighboring glomeruli receive their input from contiguous areas of the epithelium. Hence, it seems unlikely that the pattern of recovery observed represents recovery at the margin of epithelial areas destroyed by the toxin. An alternative mechanism for the observed pattern of recovery is that of axonal sprouting from spared neurons to vacated synaptic sites in nearby glomeruli. This could account for the observation that the most marked recovery in longer survival rats occurred within or at the margins of areas that had some input in shorter survival animals.
The mechanism(s) by which the toxin produces damage to the epithelium is not known but the nasal lesions in mice resemble those produced by 3-methyl-furan, an air pollutant that is toxic to Clara cells and activated by mixed function oxidases (10). Bray and Kubow (1) suggested that, in the lung, 3-MI is metabolized by mixed-function oxidases to a reactive intermediate, which in itself acts as a free radical or induces the formation of free radicals, and these, in turn, are responsible for the tissue pathology. The basis for the differential effects of 3-MI within the olfactory epithelium, however, has not been determined. The present demonstration that 3-MI produces reliable, selective, and time-dependent effects within the epithelium suggests that the toxin may provide a useful tool for examining anatomical and behavioral correlates of the disruption and recovery of input to the olfactory bulb.
AXONAL TRANSPORT AND 3-METHYLINDOLE
487
REFERENCES 1. Bray, T. M.; Kubow, S.: Involvement of free radicals in the mechanism of 3-methylindole-induced pulmonary toxicity: An example of metabolic activation in chemically induced lung disease. Environ. Health Perspect. 64:61–67; 1985. 2. Gaskell, B. A.: Nonneoplastic changes in the olfactory epithelium: Experimental studies. Environ. Health Perspect. 85:275– 289; 1990. 3. Mesulam, M.: Tracing neural connections with horseradish peroxidase. New York: Wiley; 1982. 4. Peele, D. B.; Allison, S. D.; Bolon, B.; Prah, D.; Jensen, K. F.; Morgan, K. T.: Functional deficits produced by 3-methylindoleinduced olfactory mucosal damage revealed by a simple olfactory learning task. Toxicol. Appl. Pharmacol. 107:191–202; 1991. 5. Ressler, K. J.; Sullivan, S. L.; Buck, L. B.: Information coding in the olfactory system: Evidence for a stereotyped and highly organized epitope map in the olfactory bulb. Cell 79:1245–1255; 1994. 6. Shipley, M. T.: Transport of molecules from nose to brain: Transneuronal anterograde and retrograde labeling in the rat olfactory
7. 8. 9. 10. 11. 12.
system by wheat germ agglutinin-horseradish peroxidase applied to the nasal epithelium. Brain Res. Bull. 15:129–142; 1985. Slotnick, B. M.; Hersch, S.: A stereotaxic atlas of the rat olfactory system. Brain Res. Bull. 5(Suppl. 5):1–55; 1980. Turk, M. A. M.; Flory, W.; Henk, W. G.: Dose response in 3-methylindole-induced bronchiolar epithelial necrosis in mice. Res. Commun. Chem. Pathol. Pharmacol. 46:351–403; 1984. Turk, M. A. M.; Flory, W.; Henk, W. G.: Chemical modulation of 3-methlyindole toxicosis in mice: Effect on bronchiolar and olfactory mucosal injury. Vet. Pathol. 23:563–570; 1986. Turk, M. A. M.; Henk, W. G.; Flory, W.: 3-Methylindole-induced nasal mucosal damage in mice. Vet. Pathol. 24:400–403; 1987. Vassar, R.; Chao, S. K.; Sitcheran, R.; Nunez, J. M.; Vosshall, L. B.; Axel, R.: Topographic organization of sensory projections to the olfactory bulb. Cell 79:981–991; 1994. Walters, E.; Buchheit, K.; Maruniak, J. A.: Olfactory cytochrome P-450 immunoreactivity in mice is altered by dichlobinil but preserved by metyrapone. Toxicology 81:113–122; 1993.