This is a preprint of an article published in

The Journal of Comparative Neurology, 1996, 376:557-566.

Ó1996 Wiley-Liss

Spatial Distribution of [14C]2‑Deoxyglucose Uptake in the Glomerular Layer of the Rat Olfactory Bulb Following Early Odor Preference Learning

Brett A. Johnson and Michael Leon

Department of Psychobiology, University of California, Irvine, CA 92697-4550

Number of text pages = 25

Number of figures = 5

Abbreviated title: Early odor learning and glomerular 2-DG uptake

Associate Editor: Jon H. Kaas

Indexing terms: Chemical senses, early experience, mapping, metabolic activity, olfaction

Send proofs and reprint requests to: Brett A. Johnson

Department of Psychobiology

University of California, Irvine

2205 BioSci II

Irvine, CA 92697-4550

Telephone: (714) 824-7303

Fax: (714) 824-2447

ABSTRACT

Previous work has shown that odors induce focal uptake of [14C]2‑deoxyglucose (2‑DG) within the glomerular layer of the main olfactory bulb and that the amount of 2‑DG accumulated in these foci increases after early odor learning. To determine if learning-associated changes in 2-DG uptake occur across the entire glomerular layer, we have mapped uptake throughout that layer at fixed angles in coronal sections through the bulb. Resulting arrays for individual bulbs were corrected for differing bulb size and averaged across experimental groups to address the spatial distribution of uptake. The average arrays revealed at least three discrete fields of uptake in naive, peppermint-exposed rats at postnatal day 19 that were not seen in air-exposed littermates. In agreement with previous studies, early preference training with peppermint odor given on postnatal days 1-18 increased 2‑DG uptake at postnatal day 19 within odor-dependent patches of uptake in the posterior half of the midlateral bulb, while odor-dependent, ventrolateral patches of uptake did not increase to the same extent. In addition, early preference learning was associated with significantly increased 2‑DG uptake averaged over the entire analyzed glomerular layer. These increases were smaller than those within odor-dependent foci and were distributed widely across the glomerular layer, showing low overlap between trained and control rats in anterior regions where peppermint odor did not stimulate 2‑DG uptake. The widely distributed increases in 2‑DG uptake after learning may reflect changed activity of centrifugal projections that diffusely innervate the glomerular layer.

When rats are exposed to an odor following an injection of [14C]2‑deoxyglucose (2‑DG), foci of high 2-DG uptake are found in the glomerular layer of the main olfactory bulb (Sharp et al., 1975, 1977). The patterns of uptake are bilaterally symmetrical (Sharp et al., 1975; Jourdan et al., 1980; Astic and Saucier, 1982; Royet et al., 1987) and different for distinct odors (Jourdan et al., 1980; Astic and Saucier, 1982; Royet et al., 1987). Focal signals occur over glomerular neuropil (Jourdan et al., 1980: Lancet et al., 1982), where synapses are made between olfactory receptor axon terminals and dendrites of bulbar neurons (Pinching and Powell, 1971). Glomeruli showing high uptake may correspond to projections from sensory neurons expressing receptor proteins specific for the odorant (Ressler et al., 1994; Vassar et al., 1994).

Odor preference training on postnatal days (PND) 1-18 increases the levels of 2-DG accumulated within foci of high uptake when animals are exposed to the learned odor on PND 19 (Coopersmith and Leon, 1984). The increased uptake is independent of changes in respiration of the learned odor (Coopersmith and Leon, 1984; Coopersmith et al., 1986), and is specific for the learned odor when compared to a novel odor evoking 2‑DG uptake in a different part of the bulb (Coopersmith et al., 1986). The increased focal 2‑DG uptake caused by early learning persists into adulthood (Coopersmith and Leon, 1986). A variety of reinforcers that are presented with the odor can lead to both a behavior preference and increased focal 2‑DG uptake (Sullivan and Leon, 1986; Do et al., 1988; Sullivan et al., 1990). Repeated pairings of odor with tactile stimulation as the reinforcer result in increased uptake, whereas repeated presentations of odor or tactile stimulation alone do not affect levels of uptake (Sullivan and Leon, 1986; Sullivan et al., 1989).

Many of these studies of olfactory learning have used peppermint odor, which elicits patches of 2-DG uptake in midlateral and ventrolateral portions of the glomerular layer (Coopersmith and Leon, 1984; Johnson et al., 1995). Recently, we found that early preference learning increased the Fos-like response of glomerular-layer cells associated with midlateral, but not ventrolateral, patches of 2-DG uptake (Johnson et al., 1995). In the same study, midlateral/ventrolateral ratios of 2-DG uptake were increased following preference learning (Johnson et al., 1995). These findings suggested regional heterogeneity in the bulb's modification by olfactory experience, prompting us to explore the spatial distribution of experience-dependent changes in 2-DG uptake throughout the glomerular layer. To accomplish this task, we generated maps of 2‑DG uptake across most of the glomerular layer of the bulb and averaged the maps obtained for individual animals in each experimental group.

METHODS

Subjects

All procedures involving the use of animals were approved by the UC Irvine animal care committee. Litters of Wistar rats were culled to six males and two females on PND 1 (birth = PND 0). Odor preference training was given to all pups on PND 1‑18 by exposing them for 10 min to air (Controls) or to a 1:10 dilution of saturated peppermint vapor (Schilling) (Trained) at a flow rate of 8 liters/min while subjecting them to five equally spaced, 15‑s bouts of perineal stroking with a sable hair brush (Coopersmith and Leon, 1984). As described previously, littermates of the animals used for 2‑DG injections were tested for a behavioral preference in a Y‑maze at PND 20 (Johnson et al., 1995).

The study was divided into two experiments, one defining peppermint-responsive regions and the other testing effects of learning. For the first experiment, six control (air-trained) litters were used at PND 19. Following an injection of [14C]2‑DG, one pup from each litter was exposed to air, and its littermate was exposed to peppermint odor (see below). For the learning study, a total of 14 different litters were used (7 control and 7 trained litters), and one pup from each litter was exposed to peppermint odor on PND 19 following an injection of [14C]2‑DG. The remaining littermates were used in other studies (Johnson et al., 1995; Woo and Leon, 1995), or for behavioral testing.

2-DG injections, sectioning, and autoradiography

Male pups (PND 19) were given a subcutaneous injection of [14C]2-DG (0.2 mCi/kg) and then were exposed to either air or 1:10 peppermint odor for 45 min (Coopersmith and Leon, 1984). An additional 4 naive pups from separate litters were exposed in clean 1-liter beakers. Pups then were decapitated, and their brains were removed rapidly and frozen in -45 °C isopentane. Brains were stored at -70°C, then were equilibrated at -18°C before sectioning with a cryostat. Every third coronal, 20-µm section was collected on a 22 x 22 mm coverglass and dried immediately on a slide warmer at 60°C. All sections from a given brain, as well as a set of 14C-standards (ARC-146A), were exposed to a single sheet of Kodak SB5 autoradiography film for 10 days. For some animals, alternate sections were stained with cresyl violet.

Analyses

Films were coded prior to image analysis, which was accomplished using MCID/M1 software (Imaging Research Inc.) and a CCD video camera (Sony). Film densities were converted to units of nCi/g by calibration to the 14C‑standards. Analysis of each bulb began with the first autoradiographic section possessing both a distinct glomerular layer and a lighter core. This hallmark was detected ~100 µm caudal to the beginning of the internal plexiform layer as determined from alternate, Nissl-stained sections. Analysis continued caudally through all sections possessing a lateral glomerular layer. The relative locations of these anterior and posterior hallmarks are shown in Fig. 1. The medial glomerular layer continues past the posterior hallmark, as can be seen in Fig. 1. Hence, foci of 2-DG uptake that were present within the medial glomerular layer caudal to the posterior hallmark were not analyzed in this study. Every collected section between the hallmarks was analyzed for glomerular-layer 2‑DG uptake. A protractor was centered within the bulbar core of a pseudocolor image of the autoradiograph section on the computer monitor, and 2‑DG uptake was sampled using a 9 X 9 pixel square (12 X 12 µm) placed at 9° increments around the glomerular layer. Data for each section were imported into a Microsoft Excel spreadsheet to produce an array of 40 values (40 x 9° = 360°) X as many sections as existed between the two anterior-posterior hallmarks. Missing values resulting from tears in single sections were replaced with averages of values at the same angle in preceding and following sections. The average uptake across all values in the arrays for both bulbs of a given brain was then calculated for statistical comparisons. Because bulbs varied in size, arrays were expanded to contain the same number of sections prior to performing any between-bulb operations on the spatial distributions of values. Expansion was accomplished by inserting at regular intervals mock sections containing the averages of values in preceding and subsequent sections. For the learning study, the mean number of mock sections inserted was 4.0 ± 2.7 (s.d.); the range was 0-9. Prior to calculating an average array for any experimental group, arrays for the left and right bulbs of a given brain were first averaged to give a better representation of the pattern of uptake for the individual animal. In order to correct for slight variations in the amount of 2-DG delivered to different animals, the array then was divided by a bulbar core value as described below. Arrays were visualized by printing contour charts by using Microsoft Excel software.

Starting at the fourth autoradiograph section within the hallmarks and continuing for every fourth section thereafter, a measurement was taken from the core of the bulb. The core values decreased across the first sections analyzed and then stabilized for many sections before again decreasing in the posterior bulb (Fig. 2A). Therefore, we averaged values obtained within the stable region (sections 12-40) of both left and right bulbs to obtain a single value for each brain. We used this value to obtain a measure of relative uptake (glomerular layer/core). This measure has been used in many previous studies (Coopersmith and Leon, 1984; Coopersmith et al., 1986; Sullivan and Leon, 1986; Do et al., 1988; Sullivan et al., 1990; Sallaz and Jourdan, 1992) and is predicated on the consistently low, odor-independent uptake of the core's subependymal zone, which primarily contains immature cells. In air-exposed animals, core values were correlated with average values of uptake across the glomerular layer (Fig. 2B). Similar correlations were obtained within the three other experimental groups, which suggests that variations in the amount of 2‑DG injected affect levels of 2‑DG to the same extent within the bulbar core as in the glomerular layer. This provides further support for the use of relative uptake when comparing results between different animals.

RESULTS

Distribution of uptake in individual bulbs

The patterns of uptake observed in individual sections from peppermint-exposed pups were very similar to those reported previously (Coopersmith and Leon, 1984). Uptake over the glomerular layer was higher than over the adjacent nerve layer throughout the bulb. In sections of the caudal half of the lateral bulb, foci of higher uptake were detected in the glomerular layer (Fig. 3A). The smaller foci had dimensions expected of individual glomeruli, while the larger ones were aligned with clusters of 2-3 glomeruli within a single section. Measures of relative glomerular layer uptake at fixed angles around each section (Fig. 3A, center) gave values ranging from just above 1 to just above 5 (Fig. 3B). Individual foci of high uptake in a single section were represented by one or two discrete measurements (Fig. 3B).

To give a representation of the anterior-posterior distribution of uptake, arrays of angle X section number were constructed, adjusted to contain a fixed number of sections between anterior and posterior hallmarks, and visualized using contour charts (Fig. 3C). In these charts, anterior is on the left, posterior on the right, ventral (180°) is the horizontal midline, lateral (90°) is above this midline, medial (270°) is below the midline, and dorsal is found at both the top (0°) and the bottom of the charts (Fig. 3C). The values within the arrays are located at the intersections of the gridlines. Intervals between contour lines are coded by increments of gray scale such that higher uptake is represented by darker shading. As can be inferred from the left-to-right extent of the darkest patches in the contour charts, foci of high uptake generally were encountered at a similar angle in two or more consecutively analyzed sections (Fig. 3C). It should be noted that anterior bulb sections are somewhat smaller than posterior ones (Fig. 3A), whereas these contour charts are uniform in height from anterior to posterior. The contour charts therefore disproportionately amplify the apparent area of the anterior glomerular layer.

In every bulb from pups exposed to peppermint for the first time, the highest uptake patches were observed within the posterior half of the lateral bulb (upper right quadrant of the contour charts), although their exact locations were somewhat variable across animals (Fig. 3C; right panels of Fig. 4). In individual peppermint-exposed pups, there were other patches of uptake in areas outside of the posterolateral bulb, but these patches contained lower values of relative 2‑DG uptake than the posterolateral ones. Such patches also were more variable across different animals in number, location, and magnitude of uptake (Fig. 3C; Fig. 4, right). Air-exposed littermates also showed lower-level patches of uptake distributed across the glomerular layer (Fig. 4, left). The complexity, and to some extent the pattern, of these patches were similar for littermates regardless of odor exposure. For example, the simple pattern in the peppermint-exposed animal in Litter 1 of Figure 4 was matched by a simple pattern in its air-exposed littermate, while both patterns from Litter 3 were more complex. To address the possibility that some of the pattern in the air-exposed animals was due to contaminating odorants in the test apparatus or air line, we also exposed 4 pups from separate litters in a clean beaker without the introduction of air. The activity maps obtained for these pups gave the same impressions as those from air-exposed pups: individual animals possessed patches of uptake across the analyzed glomerular layer, and the locations of most of the patches were unique to each animal (not shown). These data suggest that not all of the pattern seen in an individual peppermint-exposed animal was due to the test odor. Some of the pattern might instead be induced either by odors brought to the test apparatus from the home cage, by other experiences, or by genetic factors shared by animals from the same litter.

Experiment 1: effect of peppermint odor on average maps

To factor out the variables not dependent on peppermint odor, we averaged arrays, first for the left and right bulbs of each pup, and then across 6 air-exposed pups and across 6 peppermint-exposed littermates. The contour charts then were color-coded such that warm colors indicate high uptake and cool colors, low uptake. The average map for the air-exposed animals revealed a single patch of high uptake in the posterior-most extremity of the midlateral glomerular layer (Fig. 5A, far right at 90°). This patch was clearly evident in each of the individual bulbs from air-exposed pups (Fig. 4, left). It also was present in each of the 4 animals exposed in a clean beaker (not shown). The remaining variable patches of uptake detected in individual bulbs were dampened considerably by the averaging procedure. In addition, a field of low uptake was revealed in the anterior, medial glomerular layer (Fig. 5A).

The average pattern for the peppermint-exposed animals (Fig. 5B) was similarly simplified relative to the patterns for individual bulbs (Fig. 3C; Fig. 4). At least three fields of relatively high uptake within the posterolateral bulb can be inferred from the average map: a midlateral patch centered about 3/5 along the anterior-posterior extent of the chart, another located midlaterally but more caudally, and a third situated ventrolaterally (Fig. 5B, arrows). The caudal, midlateral patch of peppermint-evoked uptake extended more rostrally and dorsally than the single patch of high uptake in air-exposed animals. The coherent field of below-average uptake in the anterior, medial glomerular layer that also was detected in the air-exposed animals was particularly prominent in the peppermint-exposed pups (Fig. 5B). Comparisons of the average pattern to patterns from individual bulbs of peppermint-exposed pups (e.g., Fig 3C and Fig 4, right) yielded the qualitative impression that the pattern was present to differing degrees in each animal. Similar impressions were derived from inspections of average and individual maps from control and trained animals in experiment 2.

Subtraction of the air array from the peppermint array further emphasized the increased uptake in the posterolateral glomerular layer (Fig. 5C, upper right quadrant). It also revealed that the average uptake over portions of the layer distant from foci of uptake was decreased in the peppermint-exposed animals (blue-colored regions in Fig. 5C). Indeed, when relative 2‑DG uptake was averaged over the entire analyzed glomerular layer to give a single value for each pup, the overall uptake was slightly, but nonsignificantly, higher in air-exposed animals (mean ± s.e.: air, 2.48 ± 0.11; peppermint, 2.38 ± 0.08).

Experiment 2: effects of early preference learning on average maps

To assess the effects of early odor experience on patterns of 2‑DG uptake, pups were trained to prefer peppermint by daily pairings of odor and perineal stimulation, while control pups received pairings of air and perineal stimulation. As reported in a previous paper (Johnson et al., 1995), littermates of trained pups spent a significantly greater proportion of time in the presence of peppermint odor (0.47 ± 0.02) than did control pups (0.21 ± 0.03) in a Y-maze test involving a choice between peppermint and air (p < 0.002, Mann-Whitney U test). When relative 2‑DG uptake following exposure to peppermint odor was averaged over the entire analyzed glomerular layer to give a single value for each pup, trained animals (n=7) had significantly higher 2‑DG uptake than controls (n=7). The mean value for trained pups was 1.93 ± 0.09 (s.e.), while that for controls was 1.74 ± 0.03 (two-tailed Mann-Whitney U test: u=9, p < 0.05).

Averaged arrays of uptake for the control and trained pups (Fig. 5D and 5E) revealed patterns very similar to that obtained in the previous experiment with peppermint-exposed control pups (Fig. 5B), although the anterior-most midlateral field of high 2‑DG uptake that was present in Figure 5B showed evidence of subdivision into two discrete units in Figures 5D and 5E to give the appearance of four distinct fields (arrows). The overall similarity between the maps for the trained and control animals (Fig. 5D and 5E) indicates that the mapping strategy reliably represents patterns of uptake within an experiment and that training did not change markedly the global pattern of uptake.

The effect of prior experience on the magnitude of uptake is evident in Figure 5F, which displays the differences between arrays for trained and control pups. The experience-induced increases (yellow and reds) extend over most of the analyzed glomerular layer. The largest differences (bright red and maroon) emerge as distinct fields in the difference map and are associated with the three midlateral fields of high uptake in the posterior bulb (arrows in Fig. 5F). The ventrolateral field of peppermint-evoked uptake did not increase to the same extent and does not appear as a distinct field in the difference map (Fig. 5F).

Whereas Figure 5F shows differences in mean values, it does not indicate the degree of overlap between the individual animals in the two groups. To express this overlap, a Mann-Whitney U test was performed at each location within the arrays. This analysis can not define the statistical significance of the change at any particular site (Oken and Chiappa, 1986), but it does have the descriptive power to indicate the most reliable locations for the previously established significant difference across the entire analyzed glomerular layer. Descriptive statistics of this nature are commonly employed in electroencephalographic studies to identify leads giving the most distinct signals between experimental groups (Duffy et al., 1980, 1981; Thau et al., 1988; Jähnig and Jobert, 1995). As shown in Figure 5G, coherent fields with little overlap were found in the anterior portions of both lateral and medial glomerular layers (left third of the contour chart). These areas did not increase in uptake in response to peppermint odor (Fig. 5C). In addition to these fields, a coherent region of low overlap was found to correspond to the dorsal-most field of odor-dependent increased 2‑DG uptake (Fig. 5G, arrow).

DISCUSSION

The method we have used to map responses over the entire glomerular layer is similar to those used by others for mapping bulbar (Royet et al., 1987) and epithelial (Youngentob and Kent, 1995) responses to odors. This method offers many advantages for the analysis of the effects of learning as well as for the study of olfactory coding in the main olfactory bulb. By averaging spatially standardized arrays across animals in each experimental group, common areas of response could be identified while suppressing spurious activity that was detected in individual bulbs and that appeared to be due to factors other than the test odor. However, it is possible that there are are animal-specific regions of response to peppermint odor that are obscured by our averaging procedure in favor of the responsive regions that are consistent in location across different animals. The systematic nature of the sampling removes the subjectivity inherent in the identification of "focal" responses, which often relies as heavily on the contrast with the background and the area of the response unit as it does on the true magnitude of the response and its dependence on the test odor.

Air-exposed pups and pups isolated in clean beakers showed certain areas of high 2-DG uptake that were consistent in location across different animals. The principal area of uptake was located in a restricted midlateral region that appeared to represent the caudal-most glomeruli of the lateral bulb and that was posterior to the peppermint-responsive fields. Uptake of 2-DG in a similar region has been seen in pups exposed to nest odors composed of feces and wood shavings (Astic and Saucier, 1982). Therefore, a possible explanation for this field of uptake in our experiments would be odors common to all home cages that are brought with the animals to the test apparati.

The averaged maps of uptake in peppermint-exposed animals revealed fields of high uptake that were separated from one another by intervening lower values. Within Experiment 2, a very similar constellation of fields arose from averages of two different sets of 7 animals, which suggests that these fields may represent discrete biological response units rather than a chance variation in values. Nevertheless, the pattern was somewhat different in animals from separate experiments (Fig. 5B vs Fig. 5D), and the current data do not provide a statistical defense of the "discreteness" of any particular field. The differential effect of experience on the ventrolateral field versus the midlateral fields (see below) does, however, suggest that spatially distinct fields might be functionally distinct. We have not determined the source of variability across experiments in the patterns and levels of average uptake evoked by peppermint odor.

By using the present mapping method, we have confirmed that early learning increases the uptake of 2-DG in odor-dependent foci without greatly changing the overall pattern of the response (Coopersmith and Leon, 1984; Coopersmith et al., 1986; Sullivan and Leon, 1986; Do et al., 1988; Sullivan et al., 1989, 1990; Carmi and Leon, 1991). In addition, this method has allowed us to detect widespread, learning-associated increases in glomerular layer uptake occurring distant from the odor-dependent foci. Previous studies reporting no change in non-focal 2‑DG uptake in trained animals (Coopersmith and Leon, 1984; Do et al., 1988) involved measurements that were not matched systematically in location across different animals. The variability that would be expected in such random samples of a heterogeneous background probably obscured the change reported here.

Following learning, the greatest increases in 2‑DG uptake were associated with parts of the midlateral glomerular layer that showed odor-dependent increases in uptake. Similar conclusions were reached when the change was expressed as a percentage of the controls, although certain small patches within the anterior bulb approached a similar value (not shown). These data suggest that the increases in 2‑DG uptake within midlateral foci of high uptake might represent a phenomenon distinct from the widespread increases that we have observed. Spatially confined changes in the activity of the sensory epithelium following discrimination learning (Youngentob and Kent, 1995) would be expected to affect the activity of their glomerular targets in a topographical manner. Experience-dependent changes in glomerular area (Woo et al., 1987; Woo and Leon, 1991), juxtaglomerular cell number (Woo and Leon, 1991), density of glial processes (Matsutani and Leon, 1993), Fos-like responses (Johnson et al., 1995), and ß‑adrenergic receptor density (Woo and Leon, 1995) all have been found within the glomerular layer in areas associated with foci of high 2‑DG uptake, and suggest the existence of a spatially restricted alteration of these glomeruli. Indeed, measurements of glomerular width in non-focal regions were taken systematically with respect to high-uptake foci and revealed that experience-dependent increases were confined to the regions of high uptake (Woo et al., 1987). Similarly, changes in the activity of mitral cells following odor preference learning occur in the lateral bulb, where foci of high 2‑DG uptake are observed, but are not found in a region of the bulb remote from these foci (Wilson and Leon, 1988).

In agreement with our previous study dealing with both 2‑DG uptake and Fos-like responses (Johnson et al., 1995), the present results suggest that ventrolateral regions of the glomerular layer that respond to peppermint odor do not change with learning to the same extent as midlateral response regions. Experience-dependent changes in ß‑adrenergic receptors of the midlateral glomerular-layer also are not found in the ventrolateral and dorsolateral bulb (Woo and Leon, 1995). Thus, midlateral and ventrolateral glomeruli appear to be differently modified by early odor experience, and a continued investigation into the differences between these glomeruli might help uncover the bases for glomerular plasticity in early postnatal life.

We have found experience-dependent increases in glomerular layer 2‑DG uptake in regions distant from odor-dependent foci, and the magnitude of these increases seems to be rather uniform across the bulb. The lower overlap between trained and control animals in the anterior bulb may indicate a special importance of this region in relation to the early learning of peppermint odor, despite the fact that high uptake foci do not appear there in response to the odor. However, there are several aspects of the sampling procedure that may have led to this result. Anterior sections are smaller, but were sampled an equal number of times using the same size sample area employed for the posterior sections. Thus, the sampling of anterior sections was more thorough. Furthermore, 2‑DG uptake in the anterior sections was more uniformly low than in later sections, where small differences between animals in the locations of sharply rising peaks of uptake would be expected to contribute a greater variability to the measurements. It should be reiterated here that the rectangular contour charts employed in our presentations expand the apparent area of these smaller, anterior sections.

The diffusely distributed, learning-induced increases in glomerular layer 2‑DG uptake that we have observed contrast with the topographically restricted projection into glomeruli that is seen for sensory neurons carrying the same individual olfactory receptor mRNA (Ressler et al., 1994; Vassar et al., 1994). Diffuse input to the glomerular layer is more characteristic of centrifugal, neuromodulatory projections that include serotonergic afferents from the raphe nucleus (McLean and Shipley, 1987) and noradrenergic afferents from the locus coeruleus (McLean and Shipley, 1991; Woo et al., 1996). These projections terminate both within and around glomeruli, and non-focal accumulation of 2‑DG in the glomerular layer also occurs both in glomerular neuropil and in neuropil between cells, as well as in individual juxtaglomerular interneurons and glial cells (Lancet et al., 1982; Benson et al., 1985). It is possible that the widely distributed changes in 2‑DG uptake we describe here are caused by changes in the activity of such pathways occurring as a conditioned response in the trained pups or as a response to the odor's novelty in the controls.

Another possible explanation for the widely distributed, learning-induced changes would be provided if many glomeruli scattered across the entire layer responded to some extent to the multiple volatile components in peppermint extract by virtue of the activation of broadly tuned olfactory receptors, and if prior exposure to peppermint enhanced these responses. However, peppermint-evoked 2-DG uptake was not observed in all these areas. In fact, 2-DG uptake decreased over most of the glomerular layer upon exposure to peppermint in comparison to air. This phenomenon itself might indicate the effect of a diffuse centrifugal projection whose activity reduces 2-DG uptake in response to any novel odor.

ACKNOWLEDGMENTS

We thank Edna Hingco, Hongcam Duong, and Vicki Nguyen for their assistance in training and testing the animals. We also are grateful to Dr. Cynthia Woo for her advice and assistance. This work was supported by grant HD24236 from NICHD, grant MH45353 from NIMH, and a grant from the Irving Harris Foundation.

LITERATURE CITED

Astic, L., and D. Saucier (1982) Ontogenesis of the functional activity of rat olfactory bulb: Autoradiographic study with the 2‑deoxyglucose method. Dev. Brain Res. 2:243-256.

Benson, T.E., G.D. Burd, C.A. Greer, D.M.D. Landis, and G.M. Shepherd (1985) High-resolution 2‑deoxyglucose autoradiography in quick-frozen slabs of neonatal rat olfactory bulb. Brain Res. 339:67-78.

Carmi, O., and M. Leon (1991) Neurobehavioral responses of neonatal rats to previously experienced odors of different concentrations. Dev. Brain Res. 64: 43-46.

Coopersmith, R., and M. Leon (1984) Enhanced neural response to familiar olfactory cues. Science 225:849-851.

Coopersmith, R., and M. Leon (1986) Enhanced neural response by adult rats to odors experienced early in life. Brain Res. 371:400-403.

Coopersmith, R., S.R. Henderson, and M. Leon (1986) Odor specificity of the enhanced neural response following early odor experience in rats. Dev. Brain Res. 27:191-197.

Do, J.T., R.M. Sullivan, and M. Leon (1988) Behavioral and neural correlates of postnatal olfactory conditioning: II. Respiration during conditioning. Dev. Psychobiol. 21:591-600.

Duffy, F.H., P.H. Bartels, and J.L. Burchfiel (1981) Significance probability mapping: Analysis of brain electrical activity. Electroenceph. Clin. Neurophysiol. 51:455-461.

Duffy, F.H., M.B. Denckla, P.H. Bartels, and G. Sandini (1980) Dyslexia: Regional differences in brain electrical activity by topographic mapping. Ann. Neurol. 7:412-420.

Jähnig, P., and M. Jobert (1995) Methodological considerations for the evaluation of EEG mapping data: A practical example based on a placebo/diazepam crossover trial. Neuropsychobiology 31:31-46.

Johnson, B.A., C.C. Woo, H. Duong, V. Nguyen, and M. Leon (1995) A learned odor evokes an enhanced Fos-like glomerular response in the olfactory bulb of young rats. Brain Res. 699:192-200.

Jourdan, F., A. Duveau, L. Astic, and A. Holley (1980) Spatial distribution of [14C]2‑deoxyglucose uptake in the olfactory bulbs of rats stimulated with two different odours. Brain Res. 188:139-154.

Lancet, D., C.A. Greer, J.S. Kauer, and G.M. Shepherd (1982) Mapping of odor-related neuronal activity in the olfactory bulb by high-resolution 2‑deoxyglucose autoradiography. Proc. Natl. Acad. Sci., USA 79:670-674.

Matsutani, S., and Leon, M. (1993) Elaboration of glial cell processes in the rat olfactory bulb associated with early learning. Brain Res. 613:317-320.

McLean, J.H., and M.T. Shipley (1987) Serotonergic afferents to the rat olfactory bulb: II. Changes in fiber distribution during development. J. Neurosci. 7:3029-3039.

McLean, J.H., and M.T. Shipley (1991) Postnatal development of the noradrenergic projection from locus coeruleus to the olfactory bulb in the rat. J. Comp. Neurol. 304:467-477.

Oken, B.S., and K.H. Chiappa (1986) Statistical issues concerning computerized analysis of brainwave topography. Ann. Neurol. 19:493-494.

Pinching, A.J., and T.P.S. Powell (1971) The neuropil of the glomeruli of the olfactory bulb. J. Cell Sci. 9:347-377.

Ressler, K.J., S.L. Sullivan, and L.B. Buck (1994) Information coding in the olfactory system: Evidence for a stereotyped and highly organized epitope map in the olfactory bulb. Cell 79:1245-1255.

Royet, J.P., G. Sicard, C. Souchier, and F. Jourdan (1987) Specificity of spatial patterns of glomerular activation in the mouse olfactory bulb: Computer-assisted image analysis of 2‑deoxyglucose autoradiograms. Brain Res. 417:1-11.

Sallaz, M., and F. Jourdan (1992) Apomorphine disrupts odour-induced patterns of glomerular activation in the olfactory bulb. NeuroReport 3:833-836.

Sharp, F.R., J.S. Kauer, and G.M. Shepherd (1975) Local sites of activity-related glucose metabolism in rat olfactory bulb during olfactory stimulation. Brain Res. 98:596-600.

Sharp, F.R., J.S. Kauer, and G.M. Shepherd (1977) Laminar analysis of 2‑deoxyglucose uptake in olfactory bulb and olfactory cortex of rabbit and rat. J. Neurophysiol. 40:800-813.

Sullivan, R.M., and M. Leon (1986) Early olfactory learning induces an enhanced olfactory bulb response in young rats. Dev. Brain Res. 27:278-282.

Sullivan, R.M., D.A. Wilson, and M. Leon (1989) Norepinephrine and learning-induced plasticity in infant rat olfactory system. J. Neurosci. 9:3998-4006.

Sullivan, R.M., D.A. Wilson, R. Wong, A. Correa, and M. Leon (1990) Modified behavioral and olfactory bulb responses to maternal odors in preweanling rats. Dev. Brain Res. 53:243-247.

Thau, K., P. Rappelsberger, A. Lovrek, H. Petsche, C. Simhandl, and A. Topitz (1988) Effect of lithium on the EEG of healthy males and females. A probability mapping study. Neuropsychobiology 20:158-163.

Vassar, R., S.K. Chao, R. Sitcheran, J.M. Nuñez, L.B. Vosshall, and R. Axel (1994) Topographic organization of sensory projections to the olfactory bulb. Cell 79:981-991.

Wilson, D.A., and M. Leon (1988) Spatial patterns of olfactory bulb single-unit responses to learned olfactory cues in young rats. J. Neurophysiol. 59:1770-1782.

Woo, C.C., and M. Leon (1991) Increase in a focal population of juxtaglomerular cells in the olfactory bulb associated with early learning. J. Comp. Neurol. 305:49-56.

Woo, C.C., and M. Leon (1995) Early olfactory enrichment and deprivation both decrease ß-adrenergic receptor density in the main olfactory bulb of the rat. J. Comp. Neurol. 360:634-642.

Woo, C.C., R. Coopersmith, and M. Leon (1987) Localized changes in olfactory bulb morphology associated with early olfactory learning. J. Comp. Neurol. 263:113-125.

Woo, C.C., D.A. Wilson, R.M. Sullivan, and M. Leon (1996) Early locus coeruleus lesions increase the density of beta adrenergic receptors in the main olfactory bulb of rats. Int. J. Dev. Neurosci., in press.

Youngentob, S.L., and Kent, P.F. (1995) Enhancement of odorant-induced mucosal activity patterns in rats trained on an odorant identification task. Brain Res. 670:82-88.

FIGURE LEGENDS

Fig. 1. Schematic diagram of a parasagittal section through the main olfactory bulb showing the locations of anterior and posterior hallmarks used in this study. The hallmarks are separated by approximately 3 mm. AOB, accessory olfactory bulb; FC, frontal cortex; GL, glomerular layer; MCL, mitral cell layer.

Fig. 2. Measurements of 2‑DG uptake within the core of the bulb. A: Bulb core uptake was sampled in every fourth section numbered from the first anterior hallmark. Results from individual bulbs of two air-exposed animals are shown. The arrow indicates the span of sections (12-40) that gave stable values. These values were averaged for each bulb to provide a standardization for varying amounts of 2‑DG injected. B: Average core uptake was correlated with glomerular layer uptake averaged over the entire analyzed bulb. Points represent means of both bulbs from six individual air-exposed animals. The line is a result of linear regression, which gave a correlation coefficient of 0.91.

Fig. 3. Construction of 2‑dimensional maps of glomerular layer 2‑DG uptake. A: Individual, coronal autoradiograph sections from a peppermint-exposed animal. Tick marks around the center section indicate the locations where discrete measurements of uptake were made. The computer-generated images were produced using Image software (NIH) and have been enhanced in contrast to represent the full range of uptake encountered across the glomerular layer. The degree of relative enhancement can be deduced from the values shown in B. B: Relative uptake at each sampled position within the sections in A. C: Array of section number X angle showing the locations of the individual sections in A (arrowheads). Section number refers to every third 20-µm section between anterior and posterior hallmarks. Values within the array are found at intersections of gridlines, and contour maps are coded by increments of gray scale at the intervals noted.

Fig. 4. Contour maps of glomerular layer 2‑DG uptake in individual bulbs from air-trained littermates exposed to either air (left) or peppermint odor (right) following an injection of 2‑DG. The assignment of gray scale levels to intervals of relative uptake is as indicated in Figure 3.

Fig. 5. Color-coded contour maps showing the effects of peppermint exposure (Experiment 1: A-C) and odor preference training (Experiment 2: D-G). Color assignments from lowest to highest are white, purple, blue, light blue, green, light yellow, yellow, pink, red, maroon, black. Not all colors are used in each map. A,B: Maps of arrays averaged across both bulbs from six air-exposed animals (A) and six peppermint-exposed littermates (B). Each color represents a span of 0.25 units of relative uptake; purple is lowest (1.75-2.00 units), and black is highest (4.00-4.25 units). Arrows in B indicate fields of high uptake that are separated from one another by intervening low values. C: Map of the differences between values in B and A. Warm colors indicate where values are higher in the peppermint-exposed group, cool colors where values are lower. Each color spans 0.3 units; blue is lowest (-0.6 to -0.3), and maroon is highest (0.9-1.2). D,E: Maps of arrays averaged across both bulbs from seven control animals (D) and seven trained animals (E). The bulbs of these animals were larger than those for the preceding experiment (A-C). Hence, the maps of uptake spanned 55 sections (instead of 50) between anterior and posterior hallmarks. Arrows indicate fields of high uptake that are separated from one another by intervening low values. Each color spans 0.25 units of relative uptake; white is lowest (1.25-1.50), and black is highest (3.75-4.00). F: Map of the differences between values in E and D. Warm colors indicate where values are higher in the trained group, cool colors where values are lower. Each color spans 0.2 units; blue is lowest (-0.4 to -0.2), and maroon is highest (0.6-0.8). Arrows indicate coherent fields of increased uptake that correspond to midlateral fields of high uptake in D and E. G: Map of the overlap between individual trained and control animals at each location within the array. Mann-Whitney U tests were performed at each location and colors were assigned based on individual two-tailed p values (trained > controls: 0.05 < pink ≤ 0.1, 0.01 < red ≤ 0.05, 0.005 < maroon ≤ 0.01, black ≤ 0.005; controls > trained: 0.05 < light blue ≤ 0.01; light yellow > 0.1). The arrow indicates a coherent field of low overlap corresponding to one of the lateral fields of high uptake in D and E.