| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Committee on Neurobiology and 2Department of Psychology, Institute for Mind and Biology, The University of Chicago, Chicago, Illinois
Submitted 3 February 2007; accepted in final form 17 April 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
, and since then they have been noted in many species of waking and anesthetized mammals. These fast oscillations are centered at 
70 Hz in the rodent olfactory system but can be lower or higher, depending on the species and presence or absence of anesthesia (Bressler and Freeman 1980). Gamma oscillations have been the focus of modeling efforts and theories of sensory binding in olfactory and other systems. However, recent studies have shown that beta oscillations (15–30 Hz) are robustly generated by the olfactory system in an olfactory go/no-go task (Martin et al. 2004; Ravel et al. 2003). These oscillations depend in the OB on an intact bidirectional connection between the OB and piriform cortex (PC), appear to have different sources of generation in the OB than gamma oscillations (Martin et al. 2006; Neville and Haberly 2003), and have been proposed to be transmitted to the OB from the PC (Bressler 1984).
Beta oscillations have also been seen in response to specific odorants without explicit associative learning, and it has been claimed that these odorants may be privileged in their representation in the olfactory system by virtue of the types of oscillations they induce. Several studies have proposed that beta oscillations are specific to a predator odor response as they are found to be enhanced in the OB, PC, and hippocampus in response to two putative predator odorants, 2-propylthietane, a component of weasel anal gland secretions, and 2,4,5-trimethylthiazoline (TMT), a component of fox feces (Heale and Vanderwolf 1994b; Heale et al. 1994; Vanderwolf and Zibrowski 2001; Zibrowski and Vanderwolf 1997; Zibrowski et al. 1998). However, these studies also found that several organic solvents, such as toluene and xylene, evoke beta oscillations. Computational studies have suggested that beta oscillations present in distributed neural systems may depend on the strength of input drive or feedback (Olufsen et al. 2003; Whittington et al. 2000).
To address innate beta oscillatory responses to odorants and the system properties associated with generation of these oscillations, we examine the physical properties of odorants which evoke enhanced beta responses in the olfactory system independent of associative learning. We show that the response is not specific to odorants generated by potential predator species; it is instead related to their airborne concentration. We also show that the beta oscillatory response is more robust in waking than in anesthetized rats and develops in waking rats over several presentations in a sensitization-like fashion.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
450 g; purchased from Harlan HSD), maintained in the colony room on a 14–10 h light/dark schedule (lights on at 8:00 CST). Rats were housed singly after electrode implantation, and all animals had access to unlimited food and water for the course of the experiments. All animal procedures were done with approval and oversight by the University of Chicago animal care and use committee with strict adherence to AAALAC standards. Electrode implants
Seven of the rats were implanted with chronic electrodes, and the remaining six were used acutely for study of the effects under urethan anesthesia. The surgical procedure was the same for the two groups except that no headstage was constructed for those studied acutely. Anesthesia protocols differed for the two groups. For chronic implants, each rat was given a presurgical induction dose of ketamine cocktail (35 mg/kg ketamine, 5 mg/kg xylazine, and 0.75 mg/kg acepromazine) and was maintained with hourly intraperitoneal injections of 25 mg/kg pentobarbital (Nembutal). Acute procedures were done under urethan anesthesia (1.2–1.3g/kg dissolved in physiological saline). Bipolar stainless steel (100 µm wire; 1 mm vertical tip separation) electrodes, formvar insulated, were placed unilaterally (left) in the OB (8.5 mm anterior to bregma, 1.5 mm lateral) and aPC (0.5 mm anterior to bregma, 3.0 mm lateral, 15° angle), guided by concurrent stimulation of the lateral olfactory tract (LOT; 2.7 mm anterior, 1.6 mm lateral, 7 mm deep, 14° angle). Electrodes were positioned across the mitral cell layer in the OB and across layer 2/3 pyramidal cell layer in the aPC by lowering the electrode perpendicular to the cell layer until the evoked potential from LOT stimulation was reversed across the two leads of the electrode. Reference and ground stainless steel screws were placed in the right anterior and posterior regions of the skull, and additional screws were used for securing the headstage to the skull. For chronic implants connector pins for each lead were inserted into a round plastic receptacle (Ginder Scientific, Ottawa, Canada), and the assembly was embedded in dental cement. Chronically implanted rats were allowed to recover for 
2 wk after surgery before beginning the experimental protocol.
Verification of electrode placements
After experiments were complete, rats were given an overdose of Nembutal, and electrode placements were marked by passing current through the tip of the stainless steel wires. Electrode tips were marked using the Prussian Blue reaction. Prior to sectioning, brains were extracted from the skull and the electrode array was removed. Both OB and aPC placements could often be confirmed by visual examination of the external or cut brain for the blue stain marking the electrode tip. Additional confirmation of placements was obtained by sectioning the brain coronally and examining electrode tracks under a microscope.
Experimental design: waking rats
Each rat was placed in a large clean polycarbonate cage with a headstage cable (Neuralynx HS-27) connected to the recording setup (Neuralynx Cheetah-32), and was allowed to move freely in the cage throughout the experiment. For each odor presentation, an odor-saturated cotton swab was held under the rat's nose for 10 s or until the rat turned away from the swab, whichever was first. Turning away was defined as either turning the head away from the swab or backing >1 in away from the swab and ceasing sniffing behavior after presentation under the snout. The amount of time spent sniffing the odorant, or the latency to turn away from the swab, on each trial was measured in seconds and recorded with a timer during the experiment. Each odorant was presented 12 consecutive times followed by a single dry swab presentation. Intertrial time was 20 s.
Each test session consisted of six different odorants in a block design (12 trials for each odorant, as described in the preceding text) with a block of 12 dry swab presentations before and after each set of six odorants. All animals were presented with a total of six sets of six odorant blocks (36 odors). Most animals were presented with two sets of six blocks (12 odorants) per day, and no rat was presented with more than three sets (18 odorants) in 1 day. At least 2 h separated the end of the first odorant test set and beginning of the second odorant test set for an animal in a day. All tests took place during the light phase (8:00- 22:00 CST).
Odorant blocks were presented in balanced order across subjects, subject to the following restrictions for choosing the six odorants for each session (1 session = 6 odorants). Only one odorant of each functional group (carboxylic acids, ketones, aldehydes, and alcohols) was presented within each session. Odorant mixtures and odorants determined by published studies to reliably elicit beta oscillations (toluene, xylene, TMT) were distributed across sessions so as to avoid giving more than one in each set of six odorant blocks. Within these parameters, odorants were assigned randomly to test sessions. Odorant vials were coded and the experimenter was blind to the identity of the odorants during testing.
Experimental design: anesthetized rats
The design was the same as that used for the waking rats with a few exceptions. Each odorant block consisted of 12 consecutive presentations followed by a dry swab presentation as in the awake condition, but the length of all odorant exposures was 10 s. There were 20 s between odorant trials, as in the awake condition. Odorants were presented in sets of six odorant blocks with a block of 12 dry swab presentations before and after each set. Most anesthetized animals were tested on three sets (18 odorant blocks). The number of tests varied slightly across subjects due to various technical problems and anesthetic depth.
Odorants
Test odorants spanned 6 log units of vapor pressure (VP). Table 1 lists the 26 monomolecular odorants tested in these experiments with their theoretical VPs (at 25°C). We used the chemical structure of each monomolecular odorant to estimate various chemical properties, including mass, molecular weight, boiling and melting points, solubility, etc. (ChemDraw Ultra 2001). We also determined theoretical VPs at 25°C (Advanced Chemistry Development (ACD/Labs 2003). Material Safety Data Sheets (MSDS) were used when available to confirm these data. Ten odor mixtures were tested: fox urine, male rat urine, fruit-flavored cereal (Froot Loops), Rat Chow, apple juice, formaldehyde, vanillin mixed in mineral oil, indole mixed in mineral oil, water, and mineral oil. Male rat urine was collected <3 days prior to testing and stored in airtight vials. This urine was collected in a nearby laboratory from rats to which the subjects had not had prior exposure. All monomolecular odorants were tested at 100% concentration, as were the urines and apple juice; vanillin, indole and the solid food odorants were mixed with mineral oil for presentation on a cotton swab. The subset of odorants used for anesthetized rats is noted in Table 1.
|
All data were recorded using Neuralynx Cheetah-32 hardware and software, and for waking rats, behavior was also recorded using a Logitech webcam. Local field potential (LFP) data were recorded via a multichannel headstage (Neuralynx HS-27), with a digital sampling rate of 2,003 or 2,016 Hz. Each lead was recorded with reference to a skull screw in the contralateral dorsal skull, posterior to the olfactory bulb. Analog filters were set at 1–475 Hz for waking rats and 0.1–475 Hz for anesthetized rats. The experimenter tapped a piezoelectric strip (output recorded on one data channel and strip visible in the frame of the webcam) immediately prior to and immediately after delivering the saturated swab under the rat's nose to make a clear record within the data of time of odorant delivery. The piezo and video records were used to verify times when necessary.
Data analysis
Data with obvious movement artifacts were discarded (10% of the data recorded from waking rats). Odorants without complete 12-trial datasets for the same six waking rats were discarded from the LFP analysis, resulting in a smaller set of odorants for the LFP analysis (6 mixtures, mineral oil, Froot Loops, male rat urine, indole, vanillin and water; 19 monomolecular odorants) than for the behavioral analysis (Fig. 1; 26 monomolecular odorants). For each rat, the quality of signals across the two leads from each brain area were assessed. The lead with the best quality signals was chosen from each pair and used for analysis across the entire set of experiments. Quality was assessed for both leads by examining the record for significant movement artifact and line noise, as we have reported previously (Kay 2005). For the OB, the lead with most prominent theta and gamma rhythms was chosen. Sometimes this was the lead near the pial surface and sometimes it was the lead deep to the mitral cell layer, assessed by the polarity of the theta rhythm relative to the gamma burst during preexperimental recordings (positive or negative going theta peaks). For the aPC, the presence of gamma bursts coherent with the OB bursts during exploratory behavior was used to assess the quality of signals in addition to the criteria described in the preceding text.
|
0.25 s), stepped by 256 samples, yielding seven half-overlapping windows per second. This gave a frequency resolution of the FFT of 3.9 Hz. For a continuous representation of power (Fig. 2), data windows of 4,096 data points (2 s) were used; the power spectra from seven half-overlapping 1,024 point windows were then averaged to estimate the spectrum of the larger 2-s window, yielding a frequency resolution of 2 Hz. The entire span was then stepped 1,024 points to create a continuous measure of beta power through the course of the odorant block.
|
; Kay and Freeman 1998). Briefly, the tapered FFTs from all the half-overlapping individual data windows (512 sample windows, 0.25 s) were used to produce averaged auto- and cross-spectra for the odor sampling periods. Coherence was estimated from the square of the averaged cross spectrum divided by the product of the averaged autospectra. All analyses were performed using StatView 5.0.1 (SAS Institute) and IgorPro 5.03 (WaveMetrics, Lake Oswego, OR). |
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
). Sniffing duration
Each waking rat was allowed to sniff the odorized swab for 
10 s. A repeated-measures ANOVA on odorant sniffing durations for all odorants across the 12 trials and seven rats revealed a significant effect of trial on sniffing duration [F(11,2343) = 40.53, P < 0.0001; Fig. 1A] due to the 1st, 2nd, and 12th trials. (The elevation of sampling time on the 12th trial may indicate anticipation of new odors as the rats became accustomed to the experimental design.) There was also a significant effect of odorant [F(26,179) = 3.086, P < 0.0001] driven by differences in the first trial. Sampling durations across the seven rats in the first presentation showed a significant negative linear correlation with the log10VP of the monomolecular odorants (r = –0.749, P < 0.0001; Fig. 1B). There was no correlation in any of the subsequent trials, and the behavior averaged over all trials showed no significant correlation with log10VP.
Beta oscillations
LFPs recorded from the OB and anterior piriform cortex (aPC) can show large changes in beta oscillatory power between odor sniffing and resting or waiting conditions (Kay 2003). Figure 2 shows an example of the power spectrum as it changes from the beginning to the end of a set of presentations for one odorant (trials 1 and 11). Although qualitative examination of these plots makes it clear that during odor presentation there is an increase in beta band activity in later trials, statistical analysis bears this out across subjects and odorants. Figure 3 illustrates the method for estimating average power by taking the area under the curve of the spectrum in a frequency band (by integration) divided by the power for a period equal in length ending 2 s before the trial. Thus, a value of one indicates no change from baseline, and values above and below one indicate increases and decreases in power, respectively.
|
|
Beta sensitization and suppression across trials
We found a significant difference in beta power across the 12 trials in both the OB [F(11,1639) = 3.785; P < 0.0001] and aPC [F(11,1639) = 2.877; P = 0.0009] of the waking rats (Fig. 5, A and B). As suggested by qualitative assessment in previous reports (Heale and Vanderwolf 1994a; Heale et al. 1994; Vanderwolf and Zibrowski 2001; Zibrowski and Vanderwolf 1997; Zibrowski et al. 1998), these differences were due to an increase of beta power in later over earlier trials. Beta power in the first trial was significantly lower than in trials 4–12 (OB) and 5–12 (aPC). These effects were driven by odorants in the 1- to 120-mmHg range (Fig. 5, C and D). There was a small but significant negative correlation between sniffing duration and beta power (R = –0.157, P < 0.0001).
|
Coherence analysis is a measure of cooperativity between signals, normalized for the power of each signal. The level of beta band coherence between the signals in the OB and aPC was elevated when beta oscillatory power was elevated (Fig. 6, A and B). The level of coherence for monomolecular odorants was linearly correlated with log10VP (Fig. 6C; r = 0.73, P = 0.0006), although the two odorants with the highest pressure (acetone, 348.5 mmHg; ammonia, 5990.5 mmHg) showed a general decrease in coherence. Beta band coherence also differed significantly across the 12 trials [F(11,1639) = 10.652; P < 0.0001], with trials 2–12 significantly higher than trial 1 and trials 5–12 higher than trials 2–4 (Fig. 6D). In anesthetized rats, from a two-way ANOVA across trials and odors there was no significant difference in beta band coherence across trials [F(11,204) = 1.275; P = 0.3404], but there was a significant difference across odorants [F(17,204) = 5.292; P < 0.0001] driven by increases in coherence during exposure to odorants both within and outside of the 1- to 120-mmHg range. There was no interaction between trials and odorants.
|
Recent studies have suggested a tradeoff between beta and gamma oscillations (Martin et al. 2004; Neville and Haberly 2003). We therefore analyzed gamma band activity to determine if such an effect was evident in our study. In waking rats, gamma band power (65–100 Hz, to avoid contamination by 60-Hz noise) varied significantly across odorants in the OB [F(24,1440) = 23.824; P < 0.0001; Fig. 7A] both within and outside of the 1- to 120-mmHg VP range, with no significant variation across trials [F(11,1440 = 0.882; P = 0.5619; Fig. 7C] and no interaction between odorants and trials (although pairwise comparisons across trials suggests a trend toward decreasing gamma power after the 1st 2 trials). Gamma power in the aPC also varied significantly across odorants [F(24,1440) = 22.195; P < 0.0001; Fig. 7B] with increases within and outside of the 1- to 120-mmHg VP range. Gamma power in the PC decreased significantly across trials [F(11,1440) = 7.653; P = 0.0130; Fig. 7C] with a significant interaction between odorants and trials [F(264,1440) = 1.245; P = 0.0083]. Examination of individual trials on a finer time scale than the trials shows that strong gamma and beta oscillations alternate as illustrated in Fig. 3.
|
Theta coupling of OB and aPC during odor sampling
We considered the possibility that differences in sniffing behavior beyond sampling duration might account for the differences in beta oscillations. In the OB, the theta band rhythm (2–12 Hz) is strongly correlated with respiratory activity associated with breathing and sniffing (Kay 2005; Klingberg and Pickenhain 1965; Macrides and Chorover 1972). Thus theta frequency represents the frequency of afferent drive to the system. Although the power can represent intensity of the afferent input, this is complicated by gain control mechanisms that exist in the glomerular layer and release of neuromodulators by the trigeminal nerve (Aroniadou-Anderjaska et al. 1997; Ennis et al. 2001; Schaefer et al. 2002). Therefore the power of the theta rhythm is more accurately the intrabulbar representation of the afferent drive.
To examine the role that effective afferent drive might play in eliciting beta oscillations, we examined the power of the theta rhythm. We also examined the coherence between the OB and aPC in this band as a measure of functional coupling strength associated with sniffing. Theta band power in the OB and aPC showed significant variation across odorants [OB F(24,1704) = 2.938, P < 0.0001; aPC F(24,1704) = 2.947, P < 0.0001], but power did not vary systematically with VP (Fig. 8, A and B) or across trials [F(11,1639) = 1.148, P = 0.3193; F(11,1639) = 1.488, P = 0.1292]. Coherence within the theta frequency band did vary significantly across odorants [F(24,1704) = 2.826, P < 0.0001; Fig. 8C], and post hoc comparisons of all odorants to the mineral oil response showed a significant increase only for ammonia, the most volatile odorant. However, the level of theta band coherence correlated linearly with log10VP (r = 0.70; P = 0.0006; Fig. 8C). Theta band coherence between OB and aPC also varied across trials [F(11,1639) = 7.468, P < 0.0001; combined analysis of all rats and odorants, Fig. 8D] with a significant increase for trials 2–12 over the first trial and trials 6–11 over the second trial.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
). Whereas humans increase the length of inhalation, rodents increase investigation time without changing sniffing frequency. The net result on this first exposure is likely to regulate the amount of odor stimulus available to bind to receptors during this first sniffing bout because odorant diffusion in the epithelial mucous is slow enough that a series of brief sniffs (125 ms each at 8 Hz) should produce a sustained level of odorant in the mucous (Scott et al. 2006).
|
2 log10VP) induce significantly enhanced beta oscillation power in olfactory structures (Figs. 4, A and B, and 6, A and B). This increase of power does not appear with the first presentation but instead appears after three to four trials in a sensitization-like fashion (Figs. 5 and 6). Our third finding shows that coupling of the OB and aPC in the theta frequency band associated with the sniff cycle is enhanced after the first trial and its magnitude positively correlated with log10VP, without any systematic increase in the amplitude of the theta rhythm (power) across odorants (Fig. 8) or trials. The link between odorant volatility and beta and theta response patterns is enhanced in waking rats, and is greatly diminished in urethan-anesthetized rats. Theta band coupling of olfactory bulb and cortex varies with airborne concentration
The interplay of sniff duration and theta band coupling are related because the theta rhythm in the OB represents the frequency and effective bulbar afferent drive from the olfactory nerve (Bressler 1987; Macrides and Chorover 1972). After the first trial in our study, sniffing durations were brief and not statistically different across odorants (Fig. 1A). However, the effective coupling strength in both the theta and beta frequency bands increased after the first trial, and the magnitude of OB-aPC theta band coupling in these later trials correlated with vapor phase concentration. Theta oscillation power (amplitude) in the olfactory bulb did not vary across presentations and did not vary systematically across odorants. We may therefore infer that the increase in coupling strength between the two systems is not driven by a systematic increase in effective intrabulbar afferent intensity and may involve a separate mechanism. However, this does not rule out differences not detectable at the level of the theta band or the LFP signal in general.
Because all of our monomolecular odorants were delivered at 100% concentration, and most odorants at high enough concentration affect the trigeminal nerve in the nasal epithelium (Brand 2006), we suggest the trigeminal system as a possible parallel mechanism to effect changes in connection strength. Recent reports have noted in anesthetized rodents that respiratory linking of olfactory bulb and cortex cycles between "up" and "down" states similar to thalamocortical coupling (Fontanini and Bower 2005, 2006; Kay and Sherman 2007), and there are "sentinel" trigeminal receptors in the nasal cavity that may serve to track respiration and irritation (Finger et al. 2003). It is therefore possible that activation of the trigeminal system could facilitate a change in coupling strength in waking animals either by direct neural activation or by neuromodulators released into the olfactory areas by brain stem nuclei, the basal forebrain or the trigeminal nerve itself (Brand 2006; Breer et al. 2006).
Beta oscillations in waking rats are restricted to highly volatile odorants
The beta oscillatory response has been proposed to be a specific predator response (Heale et al. 1994). Our data argue against this interpretation, indicating that odor-specific beta oscillations are not specific to predator odorants but rather to highly volatile odorants (1–120 mmHg). Given that vapor phase concentration can predict beta oscillation power, this suggests that beta oscillations may be produced simply in response to strong or pungent odorants. However, our results also show that odorants with the highest VPs (acetone and ammonia) did not elicit a significant beta response. The very high VP odorants are also very strong trigeminal stimulants, and sensory and cortical responses to presentations of odorants with extremely high vapor pressures can be modulated by the trigeminal system by several mechanisms, as described in the preceding text. The most obvious role of trigeminal activation would be to decrease sniffing (intensity or duration) due to the irritant nature of trigeminal stimulation, but theta band power, which is an approximation of effective afferent intensity, was not lower for the highest intensity odorants.
Variations in sniffing behavior not detectable by theta rhythm analysis could vary the placement of odorants across the olfactory epithelium or change the temporal dynamics of the incoming afferent signal, contributing to a change in the OB activation patterns. In particular, the negative correlation between sniff duration and beta power supports the conclusion of an earlier study, showing that lower concentration odorants induce beta oscillations in anesthetized rats (Neville and Haberly 2003). If for higher VP odorants, less odorant is delivered to the nasal mucosa, this would result in a lower concentration stimulus. However, the constancy of theta oscillation power across odorants suggests that the results of inhalation are not different at the intrabulbar level.
Relationship to odor learning
Beta oscillations can also be produced in response to odorants not in the range of highly volatile odorants and at much lower concentrations than those used in our study when rats reach criterion performance in a go/no-go associative odor-discrimination task, suggesting that beta oscillations may represent a form of population synchrony involved in learning over many trials and sessions (Martin et al. 2004). In our study, beta oscillations arose only after several presentations in waking rats and were specific to a given odorant, resetting when a new odorant was presented, so this could represent a form of short-term memory process similar to that seen in the analogous insect system in a similar familiarization paradigm (Stopfer and Laurent 1999).
Previous studies showed that bidirectional connections between the OB and PC are essential for the production of beta oscillations (Martin et al. 2004, 2006; Neville and Haberly 2003). We show here that beta oscillations are predicted when coherence between the olfactory bulb and piriform cortex is higher in the theta/respiratory frequency band (Figs. 5 and 7). This suggests that coupling strength may be a determining factor or threshold for beta oscillations, even when the effective input strength (OB theta oscillation power) is constant. That the same type of beta oscillations can occur with learning, suggests that the effective connection strength may also be increased during the learning process. This result is consistent with computational studies which show that beta, rather than gamma, oscillations occur in hippocampal networks dependent on Hebbian changes in pyramidal cell connections and input and feedback drive strength (Olufsen et al. 2003).
We found that beta oscillatory responses, measured by power in the beta frequency band, varied across odorants in the anesthetized condition as well. However, these responses were much weaker with power significantly decreased after the first trial. Interestingly, although the odorants in the aPC that evoked beta oscillations followed the general pattern of the waking condition, being mostly restricted to the 1- to 120-mmHg range (Fig. 4D), those that evoked such oscillations in the OB were distributed among several odorants both within and outside of this range (Fig. 4C). Furthermore, unlike the responses in waking rats, OB-aPC theta band coherence and power did not vary systematically with vapor pressure.
In previous studies, it was reported that gamma oscillations, as measured by average gamma band power over many trials, often disappear when beta oscillations are enhanced and vice versa (Martin et al. 2004). Our examination of the gamma oscillatory band shows that in waking rats, when results from all odorants are considered, odor-evoked gamma oscillations decrease in the aPC after the first trial and show a modest decrement in the OB as well (Fig. 7C). However, during odorant exposure, we find enhanced gamma power in both the OB and aPC for several of the same odorants that induce enhanced beta oscillations (Fig. 7, A and B). Further inspection of the data reveals that many trials show within a single investigation period alternating gamma and beta events (e.g., Figs. 2B and 3A). In anesthetized rats, enhanced gamma oscillations are seen only to a limited number of specific odorants, two in the OB and three in the aPC (Fig. 7, A and B). The two strongest responses in both structures were chemically related odorants (hexanal and nonanal), suggesting that the gamma response in anesthetized rats may be related to odorant input structure.
The relationship between beta oscillations and unit activity has not been extensively studied, but several published reports shed some light on which cells may be involved in supporting these oscillations in the OB. Studies in frogs, rabbits, and rats have shown that as concentration is increased, more mitral cells show significant responses to tested odorants with the strongest effects being increases in the number of cells showing inhibitory responses to odorants during inhalation and in the number of cells showing excitatory responses to odorants during exhalation (Chaput and Lankheet 1987; Duchamp Viret and Duchamp 1997). Another study showed that the beta oscillation response in urethan-anesthetized rats is restricted to the exhalation phase of respiration, dominated by firing of neurons in the internal plexiform and granule cell layers (Buonviso et al. 2003). This suggests that OB beta oscillations evoked in response to high concentration odorants may be supported by cells outside of the mitral cell layer. This is consistent with the necessity for an intact bidirectional pathway between OB and aPC in producing beta oscillations (Martin et al. 2006; Neville and Haberly 2003) because centrifugal input to the OB from the aPC is restricted to the granule cell layer.
Comparisons between waking and anesthetized results
The data show that neural responses to odorants in waking rats are very different from those seen under urethan anesthesia. These effects are consistent with previous studies in which we have shown effects due to learning and behavioral state, which are not seen in the anesthetized preparation (Kay 2003; Kay and Freeman 1998; Kay and Laurent 1999). The most dramatic difference in these data are in the specificity of beta oscillatory increases to volatility in the waking condition, which are not repeated in the anesthetized condition (Figs. 4 and 6). Furthermore, the increases seen in anesthetized rats are much smaller than those seen in waking rats. These results suggest that the increase in beta oscillations in waking animals is not directly evoked by the strength of the odorant input signal but rather by the rat's internal response to the highly volatile odorants. This hypothesis is supported by the consistency of theta oscillatory power across odorants (Fig. 8). A recent study has suggested that glomerular circuitry serves to normalize the strength of inputs to the olfactory bulb across wide ranges of concentration, consistent with our observations of the theta rhythm power (Cleland et al. 2007). If the trigeminal system mediates part of the beta oscillatory response, we would again expect a similar effect in the anesthetized rats. We speculate that if the trigeminal system is involved it is at the level of the attentional or perceptual circuitry, rather than a simple reflex-like circuit (Brand 2006).
In conclusion, these data show that the relative airborne concentrations of pure odorants and associated linear increases in coupling strength between the OB and aPC represent some of the principal parameters that may effect beta oscillations in the olfactory system of waking rats independent of explicit associative learning. These oscillations have the same frequency, duration, amplitude, and apparent system properties as those produced in response to some types of odor learning, suggesting that the effect in our study represents a type of short-term perceptual learning that may serve information transfer between cortical areas. Whether the beta oscillations reported here are produced by the same neural circuits associated with go/no-go odor learning remains to be determined and should be the object of future studies.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Present address for C. A. Lowry: Neuroscience Associates, 10915 Lake Ridge Dr., Knoxville, TN 37934
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: L. M. Kay, Institute for Mind and Biology, 940 E. 57th St., Chicago, IL 60637 (E-mail: LKay{at}uchicago.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Aroniadou-Anderjaska V, Ennis M, Shipley MT. Glomerular synaptic responses to olfactory nerve input in rat olfactory bulb slices. Neuroscience 79: 425–434, 1997.[CrossRef][Web of Science][Medline]
Brand G. Olfactory/trigeminal interactions in nasal chemoreception. Neurosci Biobehav Rev 30: 908–917, 2006.[CrossRef][Web of Science][Medline]
Breer H, Fleischer J, Strotmann J. The sense of smell: multiple olfactory subsystems. Cell Mol Life Sci 63: 1465–1475, 2006.[CrossRef][Web of Science][Medline]
Bressler SL. Spatial-organization of EEGs from olfactory-bulb and cortex. Electroencephalogr Clin Neurophysiol 57: 270–276, 1984.[CrossRef][Web of Science][Medline]
Bressler SL. Relation of olfactory bulb and cortex. I. Spatial variation of bulbocortical interdependence. Brain Res 409: 285–293, 1987.[CrossRef][Web of Science][Medline]
Bressler SL, Freeman WJ. Frequency analysis of olfactory system EEG in cat, rabbit, and rat. Electroencephalogr Clin Neurophysiol 50: 19–24, 1980.[CrossRef][Web of Science][Medline]
Buonviso N, Amat C, Litaudon P, Roux S, Royet JP, Farget V, Sicard G. Rhythm sequence through the olfactory bulb layers during the time window of a respiratory cycle. Eur J Neurosci 17: 1811–1819, 2003.[CrossRef][Web of Science][Medline]
Chaput MA, Lankheet MJ. Influence of stimulus intensity on the categories of single-unit responses recorded from olfactory bulb neurons in awake freely breathing rabbits. Physiol Behav 40: 453–462, 1987.[CrossRef][Medline]
Cleland TA, Johnson BA, Leon M, Linster C. Relational representation in the olfactory system. Proc Natl Acad Sci USA 104: 1953–1958, 2007.
DuchampViret P, Duchamp A. Odor processing in the frog olfactory system. Prog Neurobiol 53: 561–602, 1997.[CrossRef][Web of Science][Medline]
Ennis M, Zhou FM, Ciombor KJ, Aroniadou-Anderjaska V, Hayar A, Borrelli E, Zimmer LA, Margolis F, Shipley MT. Dopamine D2 receptor-mediated presynaptic inhibition of olfactory nerve terminals. J Neurophysiol 86: 2986–2997, 2001.
Finger TE, Bottger B, Hansen A, Anderson KT, Alimohammadi H, Silver WL. Solitary chemoreceptor cells in the nasal cavity serve as sentinels of respiration. Proc Natl Acad Sci USA 100: 8981–8986, 2003.
Fontanini A, Bower JM. Variable coupling between olfactory system activity and respiration in ketamine/xylazine anesthetized rats. J Neurophysiol 93: 3573–3581, 2005.
Fontanini A, Bower JM. Slow-waves in the olfactory system: an olfactory perspective on cortical rhythms. Trends Neurosci 29: 429–437, 2006.[CrossRef][Web of Science][Medline]
Heale VR, Vanderwolf CH. Dentate gyrus and olfactory bulb responses to olfactory and noxious stimulation in urethan anaesthetized rats. Brain Res 652: 235–242, 1994a.[CrossRef][Web of Science][Medline]
Heale VR, Vanderwolf CH. Toluene and weasel (2-propylthietane) odors suppress feeding in the rat. J Chem Ecol 20: 2953–2958, 1994b.[CrossRef][Web of Science]
Heale VR, Vanderwolf CH, Kavaliers M. Components of weasel and fox odors elicit fast wave bursts in the dentate gyrus of rats. Behav Brain Res 63: 159–165, 1994.[CrossRef][Web of Science][Medline]
Kay LM. Two species of gamma oscillations in the olfactory bulb: dependence on behavioral state and synaptic interactions. J Integr Neurosci 2: 31–44, 2003.[CrossRef][Medline]
Kay LM. Theta oscillations and sensorimotor performance. Proc Natl Acad Sci USA 102: 3863–3868, 2005.
Kay LM, Crk T, Thorngate J. A redefinition of odor mixture quality. Behav Neurosci 119: 726–733, 2005.[CrossRef][Web of Science][Medline]
Kay LM, Freeman WJ. Bidirectional processing in the olfactory-limbic axis during olfactory behavior. Behav Neurosci 112: 541–553, 1998.[CrossRef][Web of Science][Medline]
Kay LM, Laurent G. Odor- and context-dependent modulation of mitral cell activity in behaving rats. Nat Neurosci 2: 1003–1009, 1999.[CrossRef][Web of Science][Medline]
Kay LM, Sherman SM. An argument for an olfactory thalamus. Trends Neurosci 30: 47–53, 2007.[CrossRef][Web of Science][Medline]
Klingberg F, Pickenhain L. Uber langsame atemsynchrone Potentiale vom Bulbus olfactorius der Ratte. Acta Biol Med Ger 14: 593–595, 1965.[Web of Science][Medline]
Macrides F, Chorover SL. Olfactory bulb units: activity correlated with inhalation cycles and odor quality. Science 175: 84–87, 1972.
Martin C, Gervais R, Hugues E, Messaoudi B, Ravel N. Learning modulation of odor-induced oscillatory responses in the rat olfactory bulb: a correlate of odor recognition? J Neurosci 24: 389–397, 2004.
Martin C, Gervais R, Messaoudi B, Ravel N. Learning-induced oscillatory activities correlated to odor recognition: a network activity. Eur J Neurosci 23: 1801–1810, 2006.[CrossRef][Web of Science][Medline]
Neville KR, Haberly LB. Beta and gamma oscillations in the olfactory system of the urethan-anesthetized rat. J Neurophysiol 90: 3921–3930, 2003.
Olufsen MS, Whittington MA, Camperi M, Kopell N. New roles for the gamma rhythm: population tuning and preprocessing for the beta rhythm. J Comput Neurosci 14: 33–54, 2003.[CrossRef][Web of Science][Medline]
Ravel N, Chabaud P, Martin C, Gaveau V, Hugues E, Tallon-Baudry C, Bertrand O, Gervais R. Olfactory learning modifies the expression of odour-induced oscillatory responses in the gamma (60–90 Hz) and beta (15–40 Hz) bands in the rat olfactory bulb. Eur J Neurosci 17: 350–358, 2003.[CrossRef][Web of Science][Medline]
Schaefer ML, Bottger B, Silver WL, Finger TE. Trigeminal collaterals in the nasal epithelium and olfactory bulb: a potential route for direct modulation of olfactory information by trigeminal stimuli. J Comp Neurol 444: 221–226, 2002.[CrossRef][Web of Science][Medline]
Scott JW, Acevedo HP, Sherrill L. Effects of concentration and sniff flow rate on the rat electroolfactogram. Chem Senses 31: 581–593, 2006.
Sobel N, Khan RM, Hartley CA, Sullivan EV, Gabrieli JD. Sniffing longer rather than stronger to maintain olfactory detection threshold. Chem Senses 25: 1–8, 2000.
Stopfer M, Laurent G. Short-term memory in olfactory network dynamics. Nature 402: 664–668, 1999.[CrossRef][Medline]
Vanderwolf CH, Zibrowski EM. Pyriform cortex 
-waves: odor-specific sensitization following repeated olfactory stimulation. Brain Res 892: 301–308, 2001.[CrossRef][Web of Science][Medline]
Whittington MA, Traub RD, Kopell N, Ermentrout B, Buhl EH. Inhibition-based rhythms: experimental and mathematical observations on network dynamics. Int J Psychophysiol 38: 315–336, 2000.[CrossRef][Web of Science][Medline]
Zibrowski EM, Hoh TE, Vanderwolf CH. Fast wave activity in the rat rhinencephalon: elicitation by the odors of phytochemicals, organic solvents, and a rodent predator. Brain Res 800: 207–215, 1998.[CrossRef][Web of Science][Medline]
Zibrowski EM, Vanderwolf CH. Oscillatory fast wave activity in the rat pyriform cortex: relations to olfaction and behavior. Brain Res 766: 39–49, 1997.[CrossRef][Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  R. A. Fuentes, M. I. Aguilar, M. L. Aylwin, and P. E. Maldonado Neuronal Activity of Mitral-Tufted Cells in Awake Rats During Passive and Active Odorant Stimulation J Neurophysiol, July 1, 2008; 100(1): 422 - 430. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Martin, J. Beshel, and L. M. Kay An Olfacto-Hippocampal Network Is Dynamically Involved in Odor-Discrimination Learning J Neurophysiol, October 1, 2007; 98(4): 2196 - 2205. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
in the middle column (size proportional to log10VP for the 6 odorants; see 
represent average beta power for each of the 6 odorants (size proportional to average OB power;